28829 world development report A Better Investment Climate for Everyone (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank 2005 world development report A Better Investment Climate for Everyone (c) The International Bank for Reconstruction and Development / The World Bank world development report 2005 A Better Investment Climate for Everyone A Copublication of The World Bank and Oxford University Press (c) The International Bank for Reconstruction and Development / The World Bank © 2004 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved. 1 2 3 4 07 06 05 04 A copublication of the World Bank and Oxford University Press. Oxford University Press 198 Madison Avenue New York, NY 10016 This volume is a product of the staff of the World Bank. The �ndings, interpretations, and con- clusions expressed herein do not necessarily reflect the views of the Board of Executive Direc- tors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any terri- tory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The World Bank encourages dis- semination of its work and will normally grant permission promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Of�ce of the Publisher, World Bank, 1818 H Street, NW, Washington, DC 20433, fax 202- 522-2422, e-mail pubrights@worldbank.org. Cover and interior design: Susan Brown Schmidler Cover illustration commissioned by the WDR 2005 team; © Linda Frichtel ISBN 0-8213-5724-7 (clothbound) ISBN 0-8213-5682-8 (paperback) ISSN 0163-5085 Library of Congress Cataloging-in-Publication Data has been applied for. (c) The International Bank for Reconstruction and Development / The World Bank Contents Foreword xiii Acknowledgments xv Abbreviations and Data Notes xvi Overview 1 The investment climate is central to growth and poverty reduction 1 Tackling costs, risks, and barriers to competition 4 Progress requires more than changes in formal policies 5 A process, not an event 7 Focus on delivering the basics 9 Going beyond the basics involves additional challenges 12 The international community can lend a hand 14 I PA R T Improving the Investment Climate 17 1 The investment climate, growth, and poverty 19 Understanding the investment climate 20 How investment climate improvements drive growth and reduce poverty 24 Sharpening the focus on poverty reduction 31 Creating a better investment climate for everyone 35 2 Confronting the underlying challenges 36 The basic tension: Firm preferences or the public interest? 37 Restraining rent-seeking 40 Establishing credibility 45 Fostering public trust and legitimacy 50 Ensuring policy responses reflect a good institutional �t 53 Making progress 54 v (c) The International Bank for Reconstruction and Development / The World Bank vi CONTENTS 3 Tackling a broad agenda 56 The investment climate as a package 56 Setting priorities 58 Managing individual reforms 68 Maintaining momentum 71 Strengthening capabilities 74 II PA R T Delivering the Basics 77 4 Stability and security 79 Verifying rights to land and other property 80 Facilitating contract enforcement 84 Reducing crime 89 Ending the uncompensated expropriation of property 92 5 Regulation and taxation 95 Regulating �rms 95 Taxing �rms 106 Regulating and taxing at the border 111 6 Finance and infrastructure 115 Financial markets 115 Infrastructure—connecting �rms and expanding opportunities 124 7 Workers and labor markets 136 Fostering a skilled and healthy workforce 137 Crafting interventions to bene�t all workers 141 Helping workers cope with change 151 III PA R T Going Beyond the Basics? 157 8 Selective interventions 159 The allure—and traps—of selective interventions 159 Experience in speci�c areas 163 (c) The International Bank for Reconstruction and Development / The World Bank Contents vii 9 International rules and standards 175 International arrangements and the investment climate 175 Enhancing credibility 176 Fostering harmonization 180 Addressing international spillovers 182 Future challenges 184 IV PA R T How the International Community Can Help 187 10 How the international community can help 189 Removing distortions in developed countries 189 Providing more, and more effective, assistance 190 Tackling the substantial knowledge agenda 195 Bibliographical note 198 Endnotes 199 References 210 Background papers for the WDR 2005 241 Case studies commissioned by the U.K. Department for International Development for the World Development Report 2005 242 Selected Indicators 243 Measuring the investment climate 244 Challenges in measuring the investment climate 244 The World Bank’s new measures 245 Technical notes 250 Selected world development indicators 253 Data sources and methodology 253 Changes in the System of National Accounts 253 Classi�cation of economies and summary measures 254 Terminology and country coverage 254 Technical notes 265 (c) The International Bank for Reconstruction and Development / The World Bank viii CONTENTS Boxes 1 The investment climate perspective 2 2.15 Decentralization and the investment climate 53 2 How do �rms in developing countries rate various 2.16 E-government and the investment climate 54 investment climate constraints? 5 3.1 Improving the investment climate, China’s way 57 3 Tackling a broad agenda—lessons from China, India, 3.2 India’s path 58 and Uganda 7 3.3 Do small �rms play a special role in economic 4 Main messages from World Development Report growth? 64 2005 15 3.4 International integration is especially important for 1.1 What do we mean by the investment climate? 20 small states 65 1.2 New sources of investment climate data from the 3.5 Exporting and productivity—what is the link? 65 World Bank 21 3.6 Trade liberalization in India—recent evidence 66 1.3 Geography matters, but it is not destiny 22 3.7 Foreign locals—the role of emigrants and diaspora 67 1.4 The environment matters for well-being and 3.8 Expanding the zone of feasible and desirable policy productivity: Main messages from WDR 2003 25 improvements 68 1.5 Improving the investment climate and growth: the 3.9 The Bulldozer initiative in Bosnia and Herzegovina 70 cases of China, India, and Uganda 27 3.10 Consultative mechanisms in Latvia and Turkey 72 1.6 Measuring productivity 28 3.11 Shepherding investment climate improvements in 1.7 Growth with a poor investment climate—possible, Vietnam 73 but unlikely to be sustained 29 3.12 The evolution of a reform champion in Senegal 74 1.8 Developing a product is a learning process—as Hyundai shows 29 3.13 Networks of regulatory professionals in infrastructure 75 1.9 Firm dynamics 30 4.1 Macroeconomic stability and the investment 1.10 Showing potential returns to investment climate climate 80 improvements 31 4.2 Property rights reform in China: Even modest 1.11 How growth translates to rising incomes for poor progress can ignite a strong response 80 people 32 4.3 Secure property rights and environmental 1.12 Women and the investment climate 34 stewardship 81 2.1 Governance and the investment climate 37 4.4 The distribution of property rights 82 2.2 Firms in history 38 4.5 Thailand’s 20-year program to title rural land 83 2.3 Firms and social responsibility 38 4.6 De-monopolizing property transaction 2.4 How do �rm differences affect their policy professionals 84 preferences and priorities? 39 4.7 Intellectual property rights: The ongoing debate 85 2.5 The predation of Gécamines in Mobutu’s Zaire 41 4.8 Crime, poverty, and inequality 90 2.6 Natural resource endowments: Blessing or curse? 41 4.9 New York City’s police reforms—are they 2.7 Combating corruption in Botswana and Lithuania 42 exportable? 91 2.8 The form of intervention: How many cheers for 4.10 Property wrongs: Is there ever a statute of transparency? 44 limitations? 93 2.9 Business associations and the investment climate 45 5.1 Public ownership, regulation, and the investment 2.10 Reducing policy uncertainty to stimulate climate 97 investment 46 5.2 Regulating in Jamaica—from transplants to better 2.11 Entrepreneurship and uncertainty 47 institutional �t 98 2.12 The power of credibility 49 5.3 Environmental regulation and global integration 99 2.13 Building credibility through persistence 5.4 Easing business registration requirements in Vietnam in Uganda 50 and Uganda 101 2.14 Shining the light on government–�rm dealings in 5.5 One-stop shops—or one-more-stop shops? 101 natural resources and infrastructure 52 (c) The International Bank for Reconstruction and Development / The World Bank Contents ix 5.6 Balancing the tradeoffs between speci�city and 7.4 The core labor standards 141 discretion in regulation 103 7.5 The role and impact of unions 142 5.7 Contracting for certainty 103 7.6 Labor regulation and global integration 146 5.8 Competition laws in developing countries 106 7.7 Do �rms’ perceptions square with actual labor 5.9 Taxation and global integration: A race to the regulations? 148 bottom? 108 7.8 Reforming severance pay in Colombia and Chile 154 5.10 Who pays taxes levied on �rms? 108 8.1 Unforeseen successes in Bangladesh and Kenya 160 5.11 Tax receipts as lottery tickets? 110 8.2 Picking “winners� can be an expensive gamble— 5.12 Dealing with short-term international capital SOTEXKA in Senegal 160 flows 113 8.3 Integrating informal traders in Durban 164 5.13 Reducing customs delays in Singapore 8.4 Rural credit in Brazil 165 and Ghana 114 8.5 Staying small in India—by design 167 5.14 Contracting out customs in Mozambique 114 8.6 China’s special economic zones 167 6.1 Governments and �nance markets: A long and 8.7 Export processing zones in Mauritius and the dif�cult history 117 Dominican Republic 168 6.2 Expanding access to �nance in rural areas—new 8.8 The WTO and selective intervention 169 approaches in India 120 8.9 Rolling the dice in Indianapolis 170 6.3 Commercial micro�nanciers enter the market 120 8.10 Competing to attract investment within countries 171 6.4 Establishing a registry for movable collateral in Romania 121 8.11 Fixing the FDI strategy for Mexico’s computer industry 172 6.5 Improving corporate governance in Brazil and South Korea 122 8.12 Successful “linkage programs� in Singapore and Ireland 172 6.6 The political economy of electricity in India 125 8.13 Public-private partnerships for R&D 174 6.7 Improving the investment climate for small private providers of infrastructure 127 9.1 Evaluating rules and standards—compliance mechanisms and participation 176 6.8 Better government accounting, better government policy 128 9.2 BITs—enhancing credibility one bit at a time? 177 6.9 Expanding rural access to electricity and 9.3 NEPAD and its peer review mechanism 179 telecommunications 131 9.4 The evolving system of investor-state dispute 6.10 The power to improve productivity in Nigeria 132 settlement 181 6.11 Port reform in Colombia and India 134 9.5 Harmonizing business law in Africa—OHADA 181 6.12 The bene�ts of rural roads in Morocco 9.6 International cooperation to combat corruption 183 and elsewhere 134 9.7 Privatizing international cooperation on corporate 7.1 Malaria and HIV/AIDS cloud the investment social responsibility 184 climate 137 9.8 A multilateral agreement on investment? 185 7.2 Why Intel chose Costa Rica as the site of a 10.1 Multidonor technical assistance facilities and the multimillion dollar plant 139 investment climate 193 7.3 Tackling skill imbalances through public support for 10.2 Knowing when to stop: UNDP’s micro�nance training and retraining programs 140 activities in Bangladesh 194 Figures 1 Domestic private investment dominates foreign direct 4 Growth is closely associated with poverty reduction 3 investment 2 5 How 60,000 poor people rated jobs and 2 Productivity accounts for a signi�cant share of self-employment as paths out of poverty 4 growth 3 6 The informal economy is substantial in many 3 More competitive pressure, more innovation 3 developing countries 4 (c) The International Bank for Reconstruction and Development / The World Bank x CONTENTS 7 Costs vary widely in level and composition 4 1.18 Rising GDP is associated with rising tax revenues— 8 Small and informal �rms are often hit hardest by expanding the opportunities to fund services for the investment climate constraints 5 poor 35 9 Constraints reported by �rms—comparing Bulgaria, Formal and informal �rms have different perspectives 39 Georgia, and Ukraine 8 2.1 The main locus of bribe-taking can vary 41 10 Firms in many developing countries lack con�dence 2.2 More business start-up procedures increase both in the courts to uphold their property rights 9 delays and corruption 42 11 The inadequacies of �nance and infrastructure are 2.3 More influential �rms face fewer constraints 44 severe for many developing countries 11 2.4 More influential �rms innovate less 45 Firms rating �nance and infrastructure as a “major� 2.5 Cronyism is reduced by greater accountability—and or “severe� constraint 11 legislatures play an especially important role 46 12 Firms often rate skill shortage and labor regulations 2.6 Policy uncertainty dominates the investment climate as serious obstacles 11 concerns of �rms 46 13 Manufacturing value-added in a single country 2.7 Policy uncertainty is a concern for informal �rms as can far exceed net global of�cial development well 47 �nance 14 2.8 Firms are more likely to invest when the policies are 1.1 Institutions, broadly measured, clearly matter for perceived to be credible 48 growth 21 2.9 Improving policy predictability can increase the 1.2 Costs vary widely in level and composition 23 probability of new investment by over 30 percent 48 1.3 Regulatory unpredictability is a big concern 2.10 The power of restraint: governments with less for �rms 23 discretion present lower investment risk 49 1.4 Competitive pressure can vary signi�cantly between 2.11 Support for markets does not always track economic countries 23 growth—as in Latin America 51 1.5 Investment climate conditions vary within 2.12 Strong support for international trade and business— countries 24 but less con�dence in corporations 52 1.6 Investment climate conditions can affect �rms 3.1 Constraints reported by �rms—comparing Bulgaria, differently 24 Georgia, and Ukraine 59 1.7 Signi�cant economic growth is a modern 3.2 Informality is a matter of degree 62 phenomenon 25 3.3 Women’s participation is concentrated in the 1.8 Fast sustained growth in East Asia—declines in Sub- informal sector, among the smallest �rms 62 Saharan Africa 26 3.4 The contribution of SMEs to GDP does not vary too 1.9 The contribution of private investment to GDP has much by income—but the relative importance of grown 26 informal and formal �rms shifts dramatically 63 1.10 Private investment has grown faster in countries with 3.5 Gross exports and FDI in developing economies better investment climates 27 jumped in the 1990s 65 1.11 Differences in TFP account for the largest share of 3.6 Gaining access to technological innovations—key differences in GDP growth per worker 28 sources 67 1.12 More competitive pressure, more innovation 29 4.1 Not entitled? 81 1.13 The contribution of net entry to productivity is 4.2 Leasing activity is more prevalent in Tunisia than in higher when barriers to entry are lower 30 Egypt or Lebanon, thanks to laws facilitating 1.14 Poverty reduction is closely associated with growth 31 repossession 84 1.15 Self-employment and wage income are the ways out 4.3 Many �rms do not believe the courts will uphold of poverty 33 their property rights 87 1.16 Growing economies generate more jobs—particularly 4.4 Reforms speed up court business in República in developing countries 33 Bolivariana de Venezuela 87 1.17 The informal economy is substantial in many 4.5 Crime takes a heavy toll on many Latin American developing countries 34 economies 89 (c) The International Bank for Reconstruction and Development / The World Bank Contents xi 4.6 Crime is a signi�cant constraint on �rms in all 6.11 Competition spurs the spread of mobile phones in regions 89 Sub-Saharan Africa 130 4.7 Risky business 93 6.12 Competition in international calls is still limited or 5.1 Low-income countries tend to regulate more 96 prohibited in much of the developing world 131 5.2 Starting a new business takes longer and is more 6.13 Many days of power outages a year, and a higher share costly in developing countries 100 of �rms having their own generators 132 5.3 Larger �rms spend more time dealing with 6.14 The declining costs of transport and regulations and are inspected more often 100 telecommunications 133 5.4 Firms of all sizes report that of�cials’ interpretations 7.1 Firms rate skill shortages and labor regulations as of regulations are unpredictable 102 serious constraints in many countries 136 5.5 Despite strong laws, competition policy is seen to be 7.2 The share of the population with secondary or higher less effective in low-income countries 105 education is still very low in many developing countries 138 5.6 Corporate tax and VAT rates are similar in high- income and developing countries 107 7.3 Skill constraints and innovative �rms 139 5.7 Corporate tax revenues remained stable or increased 7.4 The minimum wage is very high in many developing during the 1990s, except in ECA 107 countries and, at high levels, leads to weak compliance 144 5.8 Caught in the middle: Taxing �rms in Uganda and Cameroon 109 7.5 Developing countries have more stringent regulations on worker hours and paid leave than many developed 5.9 Many �rms rate tax administration as a serious countries 145 obstacle 109 7.6 High job turnover in developed and developing 5.10 Most changes in national regulations governing FDI countries in the 1990s 146 reduced restrictions 112 7.7 Job turnover is high because of both the entry and exit 5.11 Restrictions on FDI have fallen in manufacturing, but of �rms, and the reallocation among existing �rms 147 persist in other sectors 112 7.8 Many developing countries have more stringent 5.12 Clearing customs for imports—from under 2 days regulations on hiring and �ring than developed to 18 113 countries 147 6.1 The inadequacies of �nance and infrastructure are Perceptions of the burden of labor regulation vary severe for many developing countries 115 across countries and �rms 148 6.2 Sources of �xed investment �nancing differ for small 7.9 Strict labor regulation is not associated with more and large �rms 116 equality in the labor market 149 6.3 State-owned banks are holding on, especially in India 7.10 Unsynchronized job creation and destruction can give and in the Middle East and North Africa 118 rise to unemployment or underemployment 150 6.4 Infrastructure concerns expressed by �rms vary by 7.11 Since the labor reform of 1990, there has been higher size and sector 124 job turnover in Colombia 150 6.5 More developing countries are involving the private 7.12 Developing countries, particularly low-income ones, sector in infrastructure provision 126 offer much weaker and less diverse protection against 6.6 Investment in infrastructure projects with private unemployment risks than developed countries 151 participation has recently fallen 126 8.1 Competition has increased with more countries 6.7 Teledensity increases with the quality of the investment exporting a larger range of goods 161 climate, even controlling for incomes 126 8.2 Incentives can be costly 170 6.8 Perceived fairness allows lower rates of return to be 8.3 Policy advocacy by investment promotion agencies promised for a given legal protection 127 receives a small share of budget 171 6.9 Long delays for phone connections are common, 8.4 Grants make up the lion’s share of public funding for especially without competition 130 private R&D in many developed countries 173 6.10 Liberalization and good regulation accelerate the 9.1 Participation in bilateral investment treaties (BITs) growth of phone connections 130 has shot up in recent years 177 (c) The International Bank for Reconstruction and Development / The World Bank xii CONTENTS 9.2 NAFTA and Mexico’s investment pro�le 178 Standards are influencing business 184 9.3 Regional economic cooperation agreements 10.1 Manufacturing value added in a single country can far proliferated in the 1990s 182 exceed net global of�cial development �nance 189 Tables 1.1 Government policies and behaviors and investment New investment climate measures from the World decisions—some examples 22 Bank 245 2.1 Bribes vary by �rm size, sector, and region 40 A1 Investment climate indicators: World Bank 3.1 Who innovates? 67 Investment Climate Surveys 246 3.2 Consultative forums dealing with investment climate A2 Investment climate indicators: expert polls and other issues—some illustrations 72 surveys 248 In some developing countries competition agencies Other sources of investment climate–related deal with very few cases 106 indicators—selected examples 251 5.1 Firms report that tax rates are one of their top Classi�cation of economies by region and income, concerns 107 FY2005 255 8.1 Export processing zones have proliferated into the 1 Key indicators of development 256 thousands 168 2 Poverty and income distribution 258 8.2 Effective reductions in corporate tax rates due to �scal 3 Economic activity 260 incentives 169 4 Trade, aid, and �nance 262 8.3 IPAs are not cheap 171 5 Key indicators for other economies 264 8.4 Fiscal incentives for R&D in selected developing countries 173 10.1 Support for investment climate reforms and to �rms and transactions 191 (c) The International Bank for Reconstruction and Development / The World Bank Foreword This World Development Report is about creating opportunities for people to escape from poverty and improve their living standards. It is about creating a climate in which �rms and entrepreneurs of all types—from farmers and microenterprises to local manufacturing con- cerns and multinationals—have opportunities and incentives to invest productively, create jobs, and expand, and thereby contribute to growth and poverty reduction. The Report thus deals with one of the central challenges of development. Expanding opportunities for people in developing countries is a pressing concern for gov- ernments and for the global community. Nearly half the world’s population lives on less than $2 a day, and 1.1 billion barely survive on less than $1 a day. Young people have more than double the average unemployment rate in all regions, and population growth will add nearly 2 billion more people to developing countries over the next 30 years. Improving the climate for invest- ment in developing countries is essential to provide jobs and opportunities for young people and to build a more inclusive, balanced, and peaceful world. There is good news. More governments are recognizing that their policies and behaviors play a critical role in shaping the investment climates of their societies, and they are making changes. China and India provide compelling examples: investment climate improvements in these countries have driven growth and the most dramatic reductions in poverty in history. Many other governments are also taking on the agenda, but progress remains slow and uneven. Governments still saddle �rms and entrepreneurs with unnecessary costs, create substantial uncertainty and risk, and erect unjusti�ed barriers to competition. This year’s World Development Report, the 27th in the World Bank’s flagship series, looks at what governments can do to create better investment climates for their societies. Drawing on new research, including surveys of nearly 30,000 �rms in 53 developing countries, other new data, and country case studies, it makes four main points. First, the Report emphasizes that the goal should be to create an investment climate that is better for everyone—in two dimensions. The investment climate should bene�t society as a whole, not only �rms. Well-designed regulation and taxation are thus an important part of a good investment climate. And the investment climate should embrace �rms of all types, not just large or influential �rms. Small and large �rms, local and foreign �rms, and low-tech and high-tech �rms each have important and complementary contributions to make to growth and poverty reduction. Second, the Report argues that efforts to improve the investment climate need to go beyond just reducing business costs. Those costs can indeed be extraordinary in many countries, amounting to several times what �rms pay in taxes. But policy-related risks dominate �rms’ concerns in developing countries and can cripple incentives to invest. And barriers to competi- tion remain pervasive, dulling incentives for �rms to innovate and increase productivity. Gov- ernments need to address all three aspects of a good investment climate. Third, the Report underscores that progress requires more than changes in formal policies. The gaps between policies and their implementation can be huge, with the vast informal economies in many developing countries providing the most palpable evidence. Governments xiii (c) The International Bank for Reconstruction and Development / The World Bank xiv WORLD DEVELOPMENT REPORT 2005 need to bridge these gaps and address deeper sources of policy failure that can undermine a sound investment climate. Governments need to tackle corruption and other forms of rent-seeking, to build credibility with �rms, to foster public trust and legitimacy, and to ensure their policy interven- tions are crafted to �t local conditions. Finally, the Report reviews strategies for tackling such a broad agenda. It emphasizes that perfec- tion is not required and that everything does not have to be done at once. But progress requires gov- ernments to address important constraints in ways that give �rms the con�dence to invest—and to sustain a process of ongoing improvements. Persistence pays off. These �ndings are supported by detailed analysis and the many examples discussed throughout the Report, which should provide practical insights for policymakers and for others concerned with growth and poverty reduction in developing countries. Improving the investment climate is the �rst pillar of the World Bank’s overall development strat- egy. The World Development Report 2005 complements last year’s WDR, which addressed key aspects of the second pillar of that strategy: investing in and empowering people to take advantage of oppor- tunities. Together, these two Reports offer sound advice and research that will help the World Bank and our partners realize our common dream—a world free of poverty. James D. Wolfensohn President The World Bank (c) The International Bank for Reconstruction and Development / The World Bank Acknowledgments This Report has been prepared by a team led by Warrick Smith and comprising Mary Hallward-Driemeier, Gaiv Tata, George Clarke, Raj Desai, Timothy Irwin, Richard Messick, Stefano Scarpetta, and Ekaterina Vostroknutova. Leora Klapper and Sunita Kikeri also con- tributed. The team was assisted by Yanni Chen, Alexandru Cojocaru, Zenaida Hernandez, Tewodaj Mengistu, Claudio E. Montenegro, and David Stewart. Bruce Ross-Larson was the developmental editor. The work was initiated under the direction of Nicholas Stern and carried out under the general direction of François Bourguignon. Many others inside and outside the World Bank provided helpful comments, including Daron Acemoglu, Erik Berglöf, Robin Burgess, Ha-Joon Chang, Shantayanan Devarajan, David Dollar, John Haltiwanger, Michael Klein, Howard Pack, and Lant Pritchett. The Development Data Group contributed to the data appendix and was responsible for the Selected World Development Indicators. Much of the background research was supported by generous trust fund grants from the U.K. Department for International Development and from the Swedish and Swiss Governments. The team undertook a wide range of consultations for this Report, which included work- shops in Berlin, Dar-es-Salaam, London, New Delhi, Shanghai, and Washington, D.C.; video- conferences with sites in Brazil, Egypt, Guatemala, Honduras, Japan, Lebanon, Nicaragua, Rus- sia, and Serbia and Montenegro; and an on-line discussion of the draft report. Participants in these workshops, videoconferences, and discussions included researchers, government of�cials, and staff of nongovernmental and private-sector organizations. Rebecca Sugui served as executive assistant to the team, Ofelia Valladolid as program assis- tant, and Madhur Arora and Jason Victor as team assistants. Evangeline Santo Domingo served as resource management assistant. Book design, editing, and production were coordinated by the World Bank’s Of�ce of the Publisher under the supervision of Susan Graham, Denise Bergeron, and Janet Sasser. xv (c) The International Bank for Reconstruction and Development / The World Bank Abbreviations and Data Notes Abbreviations APEC Asia-Paci�c Economic Cooperation NAFTA North American Free Trade Agreement BEEPS II Business Environment and Enterprise NEPAD New Partnership for Africa’s Development Performance Survey II NGOs Nongovernmental organizations BITs Bilateral investment treaties OECD Organisation for Economic Co-operation and DFIs Development �nance institutions Development EPZs Export processing zones PPP Purchasing power parity EU European Union R&D Research and development FDI Foreign direct investment SMEs Small and medium enterprises GATT General Agreement on Tariffs and Trade TFP Total factor productivity GDP Gross domestic product U.N. United Nations GNI Gross national income UNCITRAL United Nations Commission on International HIV/AIDs Human immunode�ciency virus/ Trade Law acquired immunode�ciency syndrome UNCTAD United Nations Conference on Trade and ICRG International Country Risk Guide Development ICS Investment Climate Surveys UNDP United Nations Development Programme ICSID International Centre for Settlement of USAID U.S. Agency for International Development Investment Disputes VAT Value added tax ILO International Labour Organisation WEF World Economic Forum IMF International Monetary Fund WTO World Trade Organization MERCOSUR Common Market of the South Data Notes The countries included in regional and income groupings in other status of a territory. The term developing countries this Report are listed in the Classi�cation of Economies table includes low- and middle-income economies and thus may at the beginning of the Selected World Development Indica- include economies in transition from central planning, as a tors. Income classi�cations are based on GNP per capita; matter of convenience. The term developed countries is used thresholds for income classi�cations in this edition may be to denote the high-income economies. found in the Introduction to Selected World Development Dollar �gures are current U.S. dollars, unless otherwise Indicators. Group averages reported in the �gures and tables speci�ed. Billion means 1,000 million; trillion means 1,000 are unweighted averages of the countries in the group unless billion. noted to the contrary. The use of the word countries to refer to economies implies no judgment by the World Bank about the legal or xvi (c) The International Bank for Reconstruction and Development / The World Bank Overview Everyday, �rms around the world face impor- that bene�ts society as a whole, not just �rms, tant decisions. A rural microentrepreneur and one that embraces all �rms, not just large considers whether to open a small business to or politically connected �rms. In short, a bet- complement her family's farm income. A ter investment climate for everyone. local manufacturing company ponders whether to expand its production line and The investment climate is central hire more workers. A multinational enter- to growth and poverty reduction prise evaluates alternative locations for its Private �rms—from farmers and microen- next global production facility. Their deci- trepreneurs to local manufacturing compa- sions have important implications for growth nies and multinational enterprises—are at and poverty in each location. And their deci- the heart of the development process. Driven sions will depend largely on the way govern- by the quest for pro�ts, they invest in new ment policies and behaviors shape the invest- ideas and new facilities that strengthen the ment climate in those locations. foundation of economic growth and pros- A good investment climate provides perity. They provide more than 90 percent of opportunities and incentives for �rms—from jobs, creating opportunities for people to microenterprises to multinationals—to invest apply their talents and improve their situa- productively, create jobs, and expand. It thus tions. They provide the goods and services plays a central role in growth and poverty needed to sustain life and improve living reduction. Improving the investment climates standards. They are also the main source of of their societies is critical for governments in tax revenues, contributing to public funding the developing world, where 1.2 billion people for health, education, and other services. survive on less than $1 a day, where youths Firms are thus critical actors in the quest for have more than double the average unem- growth and poverty reduction. ployment rate, and where populations are The contribution �rms make to society growing rapidly. Expanding jobs and other is mainly determined by the investment cli- opportunities for young people is essential to mate—the location-speci�c factors that create a more inclusive, balanced, and peace- shape the opportunities and incentives for ful world. �rms to invest productively, create jobs, and New data from the World Bank provide expand (box 1). Government policies and fresh insights into how investment climates behaviors play a key role in shaping the vary around the world and how they influ- investment climate. While governments ence growth and poverty. These include have limited influence on factors such as Investment Climate Surveys, which cover geography, they have more decisive influ- more than 26,000 �rms in 53 developing ence on the security of property rights, countries, and the Doing Business Project, approaches to regulation and taxation which benchmarks regulatory regimes in (both at and within the border), the provi- more than 130 countries.1 World Develop- sion of infrastructure, the functioning of ment Report 2005 draws on those data, other �nance and labor markets, and broader new evidence, and emerging lessons of inter- governance features such as corruption. national experience to look at what govern- Improving government policies and behav- ments at all levels can do to create a better iors that shape the investment climate dri- investment climate—an investment climate ves growth and reduces poverty. 1 (c) The International Bank for Reconstruction and Development / The World Bank 2 WORLD DEVELOPMENT REPORT 2005 BOX 1 The investment climate perspective The investment climate reflects the many loca- • It recognizes that �rms assess investment society, but the interests of �rms and society tion-speci�c factors that shape the opportuni- opportunities and related government poli- are not the same in all respects. Good public ties and incentives for �rms to invest cies and behaviors as part of a package.This policy is not about giving �rms everything productively, create jobs, and expand. A good reinforces the importance of looking at prop- they might ask for, but rather about balancing investment climate is not just about generating erty rights, regulation, taxes, �nance, a range of social interests. pro�ts for �rms—if that were the goal, the focus infrastructure, corruption, and other areas of A good investment climate provides oppor- could be limited to minimizing costs and risks. A government policy and behavior as part of an tunities for people to better themselves, and good investment climate improves outcomes integrated whole, rather than in isolation. improving the investment climate is the �rst pil- for society as a whole.That means that some • It highlights the forward-looking nature of lar of the World Bank’s overall development costs and risks are properly borne by �rms. And investment activity. Investment is based on strategy. A critical complementary agenda is to competition plays a key role in spurring innova- expectations about the future and not just on invest in and empower people so they can take tion and productivity and ensuring that the current conditions.This underlines the impor- advantage of those opportunities; this is the bene�ts of productivity improvements are tance of governments fostering stability and second pillar of the Bank’s strategy. World Devel- shared with workers and consumers. credibility, which are critical elements of a opment Report 2004: Making Services Work for Looking at growth and poverty reduction sound investment climate. Poor People focused on key aspects of that sec- through an investment climate lens offers sev- • It treats as fundamental the need for policy- ond pillar. eral insights: makers to balance the goal of encouraging • It puts �rms—the actors making investment and productive private investment with other hiring decisions—at the center of the discussion. social goals. Firms provide many bene�ts for Source: Authors and Stern (2002). Driving growth risks, and barriers to competition. As a result With rising populations, economic growth is of investment climate improvements in the the only sustainable mechanism for increas- 1980s and 1990s, private investment as a ing a society's standard of living. A good share of GDP nearly doubled in China and investment climate drives growth by encour- India; in Uganda it more than doubled.2 In aging investment and higher productivity. Poland, Romania, Russia, Slovakia, and Investment underpins economic growth Ukraine �rms that believe their property by bringing more inputs to the production rights are secure reinvest between 14 and 40 process. Foreign investment is becoming percent more of their pro�ts in their busi- more important in developing countries, nesses than those who don’t.3 Improving but the bulk of private investment remains policy predictability can increase the likeli- domestic (�gure 1). hood of new investment by more than 30 A good investment climate encourages percent. Reducing barriers to competition in �rms to invest by removing unjusti�ed costs, telecommunications in the 1990s unleashed a surge of new investment worldwide— Figure 1 Domestic private investment dominates foreign direct investment including investment by microentrepreneurs 20 in Bangladesh and Uganda. But it is not just the volume of invest- ment that matters for growth—it is the pro- ductivity gains that result (�gure 2).4 A 15 good investment climate encourages higher productivity by providing opportunities Percent of GDP Private gross fixed capital formation and incentives for �rms to develop, adapt, 10 and adopt better ways of doing things—not just innovations of the kind that might merit a patent but also better ways to orga- 5 nize a production process, distribute goods, FDI and respond to consumers. What is required? Low barriers to the dif- fusion of new ideas, including barriers to 0 1980 1990 2000 importing modern equipment and adjusting Note: Annual averages of 92 developing countries. the way work is organized. And an environ- Source: World Bank (2004k). ment that fosters the competitive processes (c) The International Bank for Reconstruction and Development / The World Bank Overview: A better investment climate—for everyone 3 that Joseph Schumpeter called “creative 5). The private sector accounts for more Figure 2 Productivity accounts for a signi�cant share of growth destruction�—an environment in which than 90 percent of jobs in developing coun- �rms have opportunities and incentives to tries.7 Better job opportunities also increase Education 14% test their ideas, strive for success, and pros- incentives for people to invest in their edu- per or fail.5 A good investment climate cation and skills, thus complementing TFP 41% makes it easier for �rms to enter and exit efforts to improve human development. markets in a process that contributes to Firms that are more productive can also pay higher productivity and faster growth. Net better wages and invest more in training.8 market entry can account for more than 30 Capital 45% percent of productivity growth.6 And �rms As entrepreneurs. Hundreds of millions of facing strong competitive pressure are at poor people in developing countries make Note: Sources of growth for 84 countries from 1960–2000. “TFP� is total factor productivity. least 50 percent more likely to innovate than their living as microentrepreneurs—as farm- Source: Bosworth and Collins (2003). those reporting no such pressure (�gure 3). ers, as street vendors, as homeworkers, and in a range of other occupations. They often Reducing poverty operate in the informal economy, which The critical role the investment climate accounts for more than half of economic plays in poverty reduction can be seen in activity in many developing countries (�gure two ways. First, at the aggregate level, eco- 6). Firms in the informal economy face many nomic growth is closely associated with of the same constraints as other �rms, reductions in poverty (�gure 4). Indeed, including insecure property rights, corrup- investment climate improvements in China tion, policy unpredictability, and limited drove the most dramatic poverty reduction access to �nance and public services. Reliev- in history, lifting 400 million people out of ing these constraints increases incomes for poverty over 20 years. Second, the contribu- entrepreneurs and allows them to expand tion can be seen in the way a good invest- their activities. A good investment climate ment climate enhances the lives of people also increases incentives to become part of directly, in their many capacities. the formal economy. As employees. The World Bank’s “Voices of As consumers. A good investment climate the Poor� study found that poor people expands the variety and reduces the costs of identi�ed getting a job—whether through goods and services, including those consumed self-employment or from wages—as their by poor people. Investment climate improve- most promising path out of poverty (�gure ments lowered food prices in countries Figure 3 More competitive pressure, Figure 4 Growth is closely associated with poverty reduction more innovation 10 75 Major probability of undertaking activity pressure 8 Poverty reduction Percentage increase in Average annual percent change Moderate 50 GDP per capita growth rate pressure Some 6 pressure 25 4 0 2 Introduce Upgrade new product product Note: Percentage increase is relative to �rms reporting no competitive pressure. 0 Source: World Bank Investment Climate Pakistan Bangladesh India Vietnam China Surveys/BEEPS II in 27 countries in Eastern Europe and Central Asia. Note: All �gures for 1992–98 except Bangladesh (1992–2000) and India (1993–99). Source: World Bank (2002d). (c) The International Bank for Reconstruction and Development / The World Bank 4 WORLD DEVELOPMENT REPORT 2005 Figure 5 How 60,000 poor people rated jobs and self-employment as paths out of poverty Some investment climate improvements deliver broad bene�ts across society—such Self-employment or business as better macroeconomic stability and less Income from wages or salaries Female corruption. Others have a more focused Benefit of family and kin Male impact on particular locations or activities, Income from agriculture, livestock, or fishing creating opportunities for governments to Access to agricultural land influence the distribution of bene�ts. Gov- Skill acquisition ernments can design those investment cli- Hard work, perseverance, or thrift mate improvements to be even more “pro- Access to credit poor� by targeting constraints where poor Education people live and constraints to activities Migration poor people bene�t from, including in their Savings capacities as employees, entrepreneurs, and Other consumers. This means that pro-poor approaches are not limited to efforts that 0 10 20 30 40 50 60 70 focus on constraints that face small �rms. Percent Source: Narayan and others (2000). Tackling costs, risks, and barriers Figure 6 The informal economy is including Ethiopia, Ghana, Kenya, Vietnam, to competition substantial in many developing countries and Zambia.9 Lowering barriers to market Governments influence the investment cli- entry by 10 percent has been estimated to mate through the impact of their policies reduce the average price markup by nearly 6 and behaviors on the costs, risks, and barri- Georgia percent.10 ers to competition facing �rms. Creating a Peru better investment climate requires govern- Tanzania As users of infrastructure, property, and ments to tackle all three. Big variations in �nance. Improving infrastructure, property investment climates around the world high- Nigeria rights, and �nance can deliver broad bene- light the potential for improvement. Thailand �ts across the community. Building rural roads helps �rms get their goods to market, Russia and in Morocco also increased primary Costs Sri Lanka school enrollment from 28 to 68 percent.11 Government policies and behaviors influ- Providing more secure rights to land ence the costs of doing business and hence Morocco the range of investment opportunities that encourages farmers and other �rms to Mexico invest and can ease their access to �nance; might be pro�table. Taxes are the most in Peru more secure rights also allowed obvious example. But governments also 0 25 50 75 urban slum dwellers to increase their have important roles in providing public Informal output as percent of GDP incomes by working more hours outside the goods, supporting the provision of infra- home.12 Improving the functioning of structure, and addressing market failures. Source: Schneider (2002). �nance markets helps �rms take advantage Weaknesses in government performance in of promising investment opportunities, and these roles can greatly increase the costs for also helps poor people weather family �rms and make many potential opportuni- emergencies, educate their children, and ties unpro�table. How greatly? The costs of improve their homes. contract enforcement dif�culties, inade- quate infrastructure, crime, corruption, and As recipients of tax-funded services or trans- regulation can amount to over 25 percent of fers. Firms and their activities are the prin- sales—or more than three times what �rms cipal sources of tax revenue for govern- typically pay in taxes. Both the level and the ments, and growing economies generate composition of these costs vary widely more taxes.13 A good investment climate across countries (�gure 7). can thus expand the resources governments Costs also have a time dimension. There have available to fund public services are big variations in the time taken to (including health and education) and trans- obtain a telephone line and to clear goods fers to disadvantaged members of society. through customs, as well as in the time (c) The International Bank for Reconstruction and Development / The World Bank Overview 5 managers need to spend dealing with of�- Figure 7 Costs vary widely in level and composition cials. The time it takes to register a new 30 business ranges from 2 days in Australia to Contract enforcement more than 200 days in Haiti.14 difficulties 25 Regulation Bribes Risks Crime Unreliable infrastructure Cost as percent of sales Because investment decisions are forward 20 looking, �rms’ judgments about the future are critical. Many risks for �rms, including 15 uncertain responses by customers and com- petitors, are a normal part of investment, and �rms should bear them. But govern- 10 ments have an important role to play in maintaining a stable and secure environ- 5 ment, including by protecting property rights. Policy uncertainty, macroeconomic instability, and arbitrary regulation can also 0 Poland China Brazil Algeria Tanzania cloud opportunities and chill incentives to Note: See �gure 1.2 notes for methodology used. invest. Indeed, policy-related risks are the Source: World Bank Investment Climate Surveys. Countries chosen to illustrate range. main concern of �rms in developing coun- tries (box 2). Barriers to competition BOX 2 How do �rms in developing countries rate various Firms prefer to face less competition, not investment climate constraints? more. But barriers to competition that ben- Early results of the World Bank’s program of even within countries, looking at the overall e�t some �rms deny opportunities and Investment Climate Surveys cover more results highlights the importance of policy- raise costs for other �rms and for con- than 26,000 �rms in 53 countries. While pri- related risks, including policy uncertainty sumers. They can also dull the incentives for ority constraints can vary widely across and and macroeconomic stability. protected �rms to innovate and increase their productivity. High costs and risks can Policy uncertainty act as barriers to entry. Governments also Macro instability influence barriers more directly through Tax rate their regulation of market entry and exit Corruption and their response to anticompetitive Cost & access to finance Crime behavior by �rms. Competitive pressure is Regulations & tax administration reported to be signi�cant by 90 percent of Skills �rms in Poland but only 40 percent of �rms Courts and legal system in Georgia.15 Electricity Labor regulations Variations within countries Severe obstacle Transportation and across �rms Access to land Major obstacle Moderate obstacle Early efforts to assess investment climates Telecommunications Minor obstacle focused on developing a single indicator for 0 20 40 60 80 100 each country. But investment climates vary Percent of firms reporting not only across countries but also within Note: Firms were asked to rank the list of issues as to whether they were an obstacle to the growth and oper- countries because of differences in the way ation of their business on a 5 point scale, from “no obstacle� to “severe obstacle.� Additional information on the indicators is available at the back of the book, table A1. national policies are administered and in Source: World Bank Investment Climate Surveys. the policies and behaviors of subnational governments. Even within a single location, the same conditions can affect �rms differ- ently depending on the activity they are engaged in and their size, often hitting small and informal �rms the hardest (�gure 8). (c) The International Bank for Reconstruction and Development / The World Bank 6 WORLD DEVELOPMENT REPORT 2005 Figure 8 Small and informal �rms are often hit hardest structure of taxation, or the priority given to by investment climate constraints infrastructure improvements in different 75 locations. There can also be differences in policy preferences within �rms, between Large owners and managers on matters of corpo- rate governance, or between owners and Percentage of firms 50 workers on labor market policies. All gov- Medium Small ernments must arbitrate those differences in an environment where �rms, of�cials, and 25 other stakeholders seek to tilt the outcome to their advantage. Informal Four resulting challenges 0 Have a loan Confident Believe Responding to this tension requires govern- from a that courts regulations ments to navigate four interrelated challenges formal will uphold will be financial property interpreted that cut across all areas of investment climate institution rights consistently policy. The way governments respond to Note: Based on 10 countries for which formal and informal surveys those challenges has a big impact on invest- were conducted, controlling for industry, country, ownership, and ment climates and thus on growth and �rm age. Source: World Bank Investment Climate Surveys and WDR Surveys poverty. And each involves going beyond of Micro and Informal Firms. changes in formal policies to confront deeper sources of policy failure. Progress requires more than Restraining rent-seeking. Investment cli- changes in formal policies mate policies are an enticing target for rent- Many investment climate improvements seeking by �rms, of�cials, and other groups. require changes to laws and policies. But Corruption can increase the costs of doing more is required. Over 90 percent of �rms in business—and when it extends to higher developing countries report gaps between echelons of government, it can lead to deep formal policies and what happens in prac- distortions in policies. Surveys show that the tice. And the content as well as the imple- majority of �rms in developing countries mentation of policies are vulnerable to a expect to pay bribes when dealing with of�- deeper set of policy failures. At the heart of cials, but with big variations across coun- the problem lies a basic tension: Societies tries.16 Capture and patron-clientelism bene�t greatly from the activities of �rms, (reflecting unequal information and influ- but the preferences of �rms don’t fully match ence in policymaking) can also create large those of society. This tension is most evident distortions, tilting policies in favor of some in taxation and regulation. Most �rms com- groups at the expense of others. Eliminating plain about taxes, but taxes �nance public unjusti�ed interventions in the economy, services that bene�t the investment climate curbing discretion, and improving the and other social goals. Many �rms would accountability of governments, particularly also prefer to comply with fewer regulations, through greater transparency, help to but sound regulation addresses market fail- restrain rent-seeking. ures and can therefore improve the invest- ment climate and protect other social inter- Establishing credibility. The con�dence �rms ests. Similar tensions can occur across most have in the future—including the credibility areas of investment climate policymaking. of government policies—determines Creating a good investment climate whether and how they invest. Policies that requires governments to balance these inter- lack credibility will fail to elicit the intended ests. Complicating this task are the differ- investment response. Policy credibility can ences in preferences and priorities between be undermined by many things, including �rms. Firms have common perspectives on the temptations governments face to com- many issues, but their views can diverge on promise sound long-term policies to meet others—whether on market restrictions, the shorter-term or narrower goals (such as (c) The International Bank for Reconstruction and Development / The World Bank Overview 7 extracting rents for policymakers or currying But no country has a perfect investment favor with some voters). Mechanisms that climate, and perfection on even one policy allow governments to commit to sound poli- dimension is not necessary for signi�cant cies, discipline, and persistence all play a role. growth and poverty reduction. Experience shows that progress can be made by address- Fostering public trust and legitimacy. Good ing important constraints in a way that gives investment climates are nurtured by broad �rms con�dence to invest, and by sustaining public support: a consensus in favor of a process of ongoing improvements (box 3). building a more productive society can facil- Early rounds of economic reform were itate policy improvements regardless of the sometimes seen as one-off events. But invest- political party or group in of�ce. Absence of ment climate improvements involve an ongo- such support can make policy reform more ing process of policy adjustment and �ne dif�cult and undermine the sustainability tuning across a wide domain. This is as true (and hence the credibility) of reforms. Open in today’s developed countries as it is in and participatory policymaking and efforts developing countries. Policies need regular to ensure that the bene�ts of a better invest- review to reflect changes in the conduct of ment climate extend widely in society can business and lessons from ongoing experi- help to build that support. ence. Michael Porter has suggested that Ensuring policy responses �t local condi- reforms in this area are a marathon, not a tions. To be effective, policy interventions sprint,17 but even that assessment may need to take into account sources of poten- understate the task. International experience tial government failure and differences in provides insights about the essential elements local conditions. Failure to do so can lead to of reform processes in this area: setting prior- poor or even perverse results. Approaches that demand enforcement capacity beyond that available will not only fail to meet their BOX 3 Tackling a broad agenda—lessons from China, intended objective but also contribute to India, and Uganda informality and corruption and undermine credibility. Approaches that involve high China, India, and Uganda illustrate some In both cases the reforms addressed impor- levels of discretion can expose �rms to con- simple lessons about strategies for making tant constraints, and were implemented in investment climate improvements. ways that gave �rms con�dence to invest. siderable uncertainty and risk when effec- China and India have both grown And the initial reforms have been followed by tive safeguards against the misuse of that impressively in recent years, greatly reduc- ongoing improvements that addressed con- discretion are not yet developed. While ing poverty. China’s growth is of�cially straints that were less binding initially, and approaches in today’s developed countries reported at an average of 8 percent a year also reinforced con�dence in the future path for the past 20 years, and the share of its of government policy. can provide a valuable source of inspira- population living on less than $1 a day fell Such strategies are not limited to large tion, care needs to be taken to adapt from 64 percent in 1981 to less than 17 per- countries. Uganda launched its program of approaches to local conditions. In some cent in 2001. India’s growth has increased investment climate improvements in the from an average of 2.9 percent a year in the early 1990s, after a period of civil conflict. cases this may involve the choice of simpler 1970s to 6.7 percent by the mid-1990s, and Reforms covering many areas of the invest- rules with less discretion and additional the share of its population living on less ment climate provided the basis for grow- measures to restrain arbitrary behavior. than $1 a day fell from 54 percent in 1980 to ing its economy by an average of more than 35 percent in 2000. 4 percent per year during 1993—2002 (or Yet neither country has an ideal invest- eight times the average in Sub-Saharan ment climate. China only recently gave con- Africa) and reducing the share of its popula- A process, not an event stitutional recognition to private property, tion living below the poverty line from 56 Government policies and behaviors shaping and its banking sector is dragged down by percent in 1992 to 35 percent in 2000.The the investment climate cover a wide �eld, nonperforming loans. Problems in India’s persistence of the government’s reform power sector are legendary. Both countries efforts enhanced its credibility, giving �rms from contract enforcement and regulation unleashed growth and reduced poverty the con�dence to invest. to the provision of infrastructure and labor through what appeared to be fairly modest market policy. Policies and behaviors in each initial reforms. China began with a rudimen- Source: China: Chen and Wang (2001) Qian tary system of property rights that created (2003), and Young (2000); India: Aghion and area can influence the opportunities and others (2003), Ahluwalia (2002), De Long new incentives for a substantial part of its incentives for �rms. And the policy areas economy. India began with early efforts to (2003), Rodrik and Subramanian (2004), Varsh- ney (1998), and Panagariya (2003); Uganda: often interact, with progress in one area pos- reduce trade barriers and other distortions Holmgren and others (2001) and World Bank sibly influenced by progress in others. This that covered a signi�cant part of its economy. (2001d). implies a broad agenda for government. (c) The International Bank for Reconstruction and Development / The World Bank 8 WORLD DEVELOPMENT REPORT 2005 ities, managing individual reforms, maintain- investment response to reforms in any partic- ing momentum, and strengthening govern- ular policy area. A key consideration will be ment capabilities. the impact of improvements on opportunities for poor people, including in their capacities Setting priorities as employees, entrepreneurs, and consumers. The goal is to identify important con- Governments also need to consider ben- straints that face �rms. There are no stan- e�ts that may spill over beyond the �rms dard formulas. Instead, it requires an assess- and activities affected most directly. These ment in each case of current conditions, the may include spillovers to other �rms (for potential bene�ts from improvements, links example, from foreign direct investment to with national or regional goals, and imple- local �rms), to other policy areas (for exam- mentation constraints. ple, from rights to land to access to �nance), or to broader social goals (for example, Current conditions. The most important con- infrastructure improvements bene�ting the straints can differ widely across countries, broader community). There can also be even within a single region (�gure 9). Gov- spillovers to government capabilities, credi- ernments can identify them by surveying and bility, or constituency building. consulting with �rms, but recognizing that existing �rms will not always reflect the per- Links with national or regional goals. Invest- spectives of future entrants. New sources of ment climate improvements can affect �rms data also allow the benchmarking of current and activities differently. Because of this, pri- policy performance against international ority-setting will often be influenced by the comparators in a growing number of areas— weight governments place on a subset of the highlighting the scope for improvement. goals a good investment climate can deliver. These often include integrating the informal Potential bene�ts. When the goal is to acceler- or rural economies, unleashing the growth ate growth, an improvement that affects a potential of smaller �rms, taking advantage large part of the economy will usually have a of international openness, or enabling �rms bigger impact than reforms that affect a to climb the technology ladder. smaller part. Progress in achieving a reason- able level of political and macroeconomic sta- Implementation constraints. At any point the bility is thus fundamental; without it reforms range of potential policy improvements will in other areas will gain little traction. Enhanc- usually be constrained by administrative and ing policy credibility can also leverage the political feasibility. Well-designed strategies address these constraints through effective Figure 9 Constraints reported by �rms—comparing Bulgaria, Georgia, and Ukraine management of reforms and ongoing strengthening of government capabilities. Ukraine Security & stability 1 Georgia Managing individual reforms Bulgaria There is often resistance to investment cli- Regulation Labor 0.5 mate reforms from those who bene�t from the status quo. This resistance may come from �rms or other interest groups bene�t- ing from market distortions or other special privileges; of�cials bene�ting from bribes Infrastructure Taxation or other perquisites of of�ce; or even the wider community when the implications of reform are not certain. Experience shows Finance that progress is possible when committed Note: Indices based on surveys of formal sector �rms. Values are governments communicate to build public normalized by regional maxima and minima for each indicator. Resulting indicators range from 0 (best) to 1 (worst). Countries cho- support, engage stakeholders construc- sen to highlight potential differences. See �gure 3.1 notes for more tively, and (when appropriate) provide details. Source: World Bank Investment Climate Surveys and BEEPS II. some form of compensation to those disad- (c) The International Bank for Reconstruction and Development / The World Bank Overview 9 vantaged by change. Special efforts to help political and macroeconomic stability is a vulnerable groups cope with change are also threshold requirement for other policy important, particularly when economywide improvements to gain much traction. Unsta- safety nets are not yet in place. ble or insecure environments have their most tangible affect on investment through their Maintaining momentum impact on property rights, which link effort Many countries are creating specialist insti- with reward. The better protected these tutions to help with speci�c tasks and to rights from government or third parties, the sustain progress even through changes in stronger the link between effort and reward, government. These institutions can per- and thus the greater the incentives to open form one or a combination of several roles: new businesses, to invest more in existing consultation with stakeholders, policy coor- ones, and simply to work harder. Studies in dination, and the more systematic review of many countries show that the more secure existing investment climate constraints. the rights, the faster the growth. Improving Latvia, Senegal, Turkey, and Vietnam illus- the security of property rights requires trate possible approaches. Governments are action in four main areas: verifying rights to also creating mechanisms to review new land and other property, facilitating contract policy and regulatory proposals more sys- enforcement, reducing crime, and ending the tematically so that they do not introduce uncompensated expropriation of property. unwarranted distortions. Verifying rights to land and other property. Strengthening government Providing more secure rights to land and capabilities other property encourages investment and Strengthening capabilities in regulation is can ease access to �nance. Experience in often a high priority. Traditional models for Peru, Thailand, and many other countries building capacity are being complemented highlights the bene�ts of clarifying owner- by approaches that facilitate peer-to-peer ship of land and maintaining an effective learning. Local capacity can also be aug- registration system. Registries for equip- mented by contracting out some specialist ment and other forms of moveable prop- functions—a common strategy even in erty also play an important role. developed countries. Governments need to improve their ability to monitor the perfor- Facilitating contract enforcement. In many mance of their private sectors so that they developing countries, �rms lack con�dence can identify trends and emerging issues and in the courts to uphold their property rights evaluate the impact of their policies. (�gure 10). Improving courts is thus a high Figure 10 Firms in many developing countries lack Focus on delivering the basics con�dence in the courts to uphold their property rights Industrial development is usually a process of discovery, making it dif�cult to predict Bangladesh what a country or region will be good at Moldova producing. This underscores the impor- Guatemala tance of creating a good investment climate Kyrgyzstan for all �rms in the economy and so focusing on improving the “basics.� International Kenya experience highlights promising approaches Czech Rep. in each of the four core areas of a sound Brazil investment climate: stability and security, regulation and taxation, �nance and infra- Zambia structure, and workers and labor markets. Algeria Stability and security Malaysia The outbreak of war or other widespread 0 20 40 60 80 100 violence spells the end of almost all produc- Percent tive investment, and a reasonable level of Source: Investment Climate Surveys. Countries chosen to illustrate range. (c) The International Bank for Reconstruction and Development / The World Bank 10 WORLD DEVELOPMENT REPORT 2005 priority. Facilitating the free flow of reputa- The key is to strike a better balance between tion information and removing unnecessary market failures and government failures, impediments to the use of alternative dis- including by ensuring that approaches are pute resolution mechanisms can also help. adapted to local conditions and by enhanc- ing transparency. Successful reforms remove Reducing crime. Crime imposes large costs on unjusti�ed burdens and streamline proce- societies—around a quarter of GDP in some dures. They reduce regulatory uncertainty countries in Latin America.18 Surveys show and risk by curbing discretion and expand- that crime is also a serious constraint for ing consultation. And they remove unjusti- many �rms in all regions. Promising strategies �ed barriers to competition by reducing reg- involve efforts to prevent and deter crime, as ulatory barriers to entry and exit and by well as to improve enforcement. Community tackling anticompetitive behavior by �rms. policing strategies along the lines of those applied in New York City are being pursued Improving domestic taxation. Tax rates in by more countries around the world. developing countries are similar to those in developed countries. But a high level of Ending the uncompensated expropriation of informality, coupled with poor administra- property. All governments reserve the right to tion and corruption, reduces revenue collec- expropriate private property in some circum- tion, places a disproportionate burden on stances. Reducing concerns about the arbi- those who do comply, and distorts competi- trary exercise of this power requires credible tion. Keeping the size of government in restraints on expropriation without prompt, check and spending public money ef�ciently adequate, and effective compensation. help ease the pressure on revenue collection. Beyond this, broadening the tax base and Regulation and taxation simplifying tax structures can help. Increas- The way governments regulate and tax �rms ing the autonomy of tax agencies has also and transactions, domestically and at the bor- improved performance in Peru and many der, plays a big role in shaping the investment other countries. climate. Sound regulation addresses market failures that inhibit productive investment Improving regulation and taxation at the and reconciles the interests of �rms with border. Most countries have reduced barri- wider social goals. Sound taxation generates ers to international trade and investment in the revenues to �nance the delivery of public recent years, but many barriers remain. services that improve the investment climate Improving customs administrations can and meet other social objectives. The chal- also offer big bene�ts, with successful lenge all governments struggle with is how to approaches exploiting information tech- meet these objectives without undermining nologies to reduce delays and corruption, as the opportunities and incentives for �rms to in Ghana, Morocco, and Singapore.19 invest productively, create jobs, and expand. While there can be tensions between �rms’ Finance and infrastructure preferences and social goals in this area, there Financial markets, when functioning well, is huge scope for improving approaches in connect �rms to lenders and investors will- most developing countries without compro- ing to fund their ventures and share some of mising broader social interests. the risks. Good infrastructure connects �rms to their customers and suppliers and Improving domestic regulation. Too often, helps them take advantage of modern pro- governments pursue regulatory approaches duction techniques. Conversely, inadequa- that fail to achieve the intended social objec- cies in �nance and infrastructure create bar- tives because of widespread informality, yet riers to opportunities and increase costs and harm the investment climate by imposing risks for microenterprises as well as multi- unnecessary costs and delays, inviting cor- nationals. By impeding new entry into mar- ruption, increasing uncertainty and risk, and kets, inadequacies also limit the competitive creating unjusti�ed barriers to competition. discipline facing incumbent �rms, dulling (c) The International Bank for Reconstruction and Development / The World Bank Overview 11 their incentives to innovate and improve Figure 11 The inadequacies of �nance and infrastructure are severe for many developing countries their productivity. Such inadequacies are large in developing countries (�gure 11). Latin America & the Caribbean Improving �nance. The underlying chal- lenge with �nance flows from information Middle East & North Africa problems, which are often exacerbated by weak protection of property rights. Govern- Sub-Saharan Africa ment intervention through state ownership, barriers to competition, directed or subsi- South Asia dized credit, and similar approaches can create deep distortions and retard �nancial Europe & Central Asia market development. Better approaches recognize that �nancial markets are not Finance East Asia & Pacific only part of the investment climate for Infrastructure �rms, but are also profoundly shaped by the 0 10 20 30 40 50 60 70 investment climate facing providers of Percent of firms �nancial services. More governments are Note: Figure shows the share of �rms that report access to �nance, and any of electricity, telecommu- thus reducing barriers to competition nications, or transportation, as “major� or “severe� obstacles to their business. Source: World Bank Investment Climate Surveys. (including paving the way for nonbank �nancial intermediaries and commercial micro�nance), strengthening creditor and workforce; crafting market interventions to shareholder rights, supporting the estab- bene�t all workers; and helping workers lishment of credit bureaus and other mech- cope with change. anisms to address information problems, and improving bank regulation. Fostering a skilled workforce. Improving the investment climate goes hand in hand with Improving infrastructure. The underlying enhancing human capital. A skilled work- challenge with infrastructure flows from force is essential for �rms to adopt new and market power associated with economies of more productive technologies, and a better scale. But responses focusing on provision investment climate raises the returns to by public sector monopolies have produced investing in education. Government sup- poor results in many developing countries. port for education and training affects the Recognizing this, governments are now prospects for individuals and the ability of focusing on creating a better investment cli- �rms to pursue new opportunities. Many mate for providers of infrastructure ser- �rms in developing countries rate inade- vices. Competition, improved regulation, quate skills of workers as a serious obstacle and private participation have transformed to their operations (�gure 12). Govern- telecommunications and are playing a big- ments need to take the lead in making edu- ger role in electricity supply and ports. For cation more inclusive and relevant to the roads, promising strategies include con- skill needs of �rms, strengthening quality tracting-out services and improving fund- assurance mechanisms, and creating a sound ing mechanisms. Governments are also investment climate for providers of educa- working to improve management of public tion and training services. resources—to get more for their money when they �nance or subsidize infrastruc- Crafting market interventions to bene�t all ture services. workers. Regulation of labor markets is usually intended to help workers. But ill- Workers and labor markets considered approaches discourage �rms Government intervention in labor markets from creating more jobs and contribute to a should help connect people to decent jobs. swelling of the informal workforce that Improving policy performance requires lacks statutory protection. When this is the progress on three fronts: fostering a skilled case, some workers may bene�t, but the (c) The International Bank for Reconstruction and Development / The World Bank 12 WORLD DEVELOPMENT REPORT 2005 Figure 12 Firms often rate skill shortage and labor bene�t society as a whole. While a narrow tax regulations as serious obstacles base reduces the feasibility of creating com- Brazil prehensive social safety nets in most develop- ing countries, there are opportunities for of available workers Skills and education Zambia improving the insurance component in China income support schemes and the pooling of Algeria risks among individuals. Innovative pro- grams can also reach out to poor and infor- Estonia mal workers who cannot be covered by Bangladesh broader insurance schemes. Brazil Going beyond the basics Poland involves additional challenges Labor regulations Many governments go beyond the basics Philippines just described by making selective interven- Kenya tions to bene�t particular �rms or activities, Pakistan or by drawing on the growing body of inter- national rules and standards that deal with Algeria investment climate issues. Both can play a 0 20 40 60 role but involve additional challenges. Percent Note: Percentage of �rms reporting that skills and education of Selective interventions— available workers or labor regulations were a major or severe obstacle to the operation and growth of their business. approach with care Source: World Bank Investment Climate Surveys. Broad improvements to the investment cli- mate expand the pool of bene�ciaries, reduce unemployed, the low-skilled, and those in concerns about rent-seeking, and avoid new the informal economy will not be among distortions. Given the breadth of the reform them. Interventions need to be crafted to agenda, some �rms or activities may bene�t reflect this wider range of interests. More from improvements earlier than others—as countries are reviewing labor market poli- with infrastructure in a particular location or cies to encourage wage adaptability, to with regulatory reforms affecting a particular ensure workplace regulations reflect a good activity. But beyond the sequencing of institutional �t, and to ensure a reasonable reforms, some governments confer special balance between workers’ preference for policy privileges on targeted �rms or activi- employment stability and �rms’ need to ties. Those privileges take many forms: mar- adjust the work force. ket restrictions, tax breaks, access to subsi- dized credit, and a range of other measures. Helping workers cope with change. A good Some selective interventions have an investment climate facilitates the allocation economic rationale, such as the possible of labor to its most productive use while spillovers from foreign direct investment helping workers cope with labor mobility. or research and development. Some may Technological progress that leads to higher be regarded as a form of “second-best� productivity and economic growth improves response, given slow progress in address- working conditions and wages, but it can also ing the basics. Yet others aim to accelerate involve faster changes to �rms and industries. growth by targeting particular industries. In modern economies, many �rms are cre- Whatever the rationale, all such schemes ated and destroyed each year—about 20 per- must navigate the heterogeneous and self- cent in many countries—involving 10 to 20 interested requests of �rms, rent-seeking percent of the workforce.20 Inadequate mech- pressures, and other sources of potential anisms to help workers cope with change policy failure. restrict entrepreneurship and the adaptability While governments have been experi- of workers. The inadequacies can also menting with selective interventions for cen- increase resistance to reforms that would turies, international experience reveals no (c) The International Bank for Reconstruction and Development / The World Bank Overview 13 sure-�re strategies. Some countries in East nizing rules and standards, and by address- Asia appear to have made selective interven- ing international spillovers. All three tions successfully, but recent work suggests involve tradeoffs. that the contribution may have been rela- tively modest. Experience also shows how Enhancing credibility. By increasing the dif�cult it is to replicate similar approaches costs of policy reversal, entering into inter- elsewhere and in what is now a very different national obligations can reinforce the credi- international environment. Overall, experi- bility of government policies and so ence with government efforts to “pick win- strengthen the investment responses of ners� is discouraging. Efforts to woo �rms. But by design the tradeoff is foregone investors through special inducements have policy flexibility, which means that com- also met with mixed success; even when mitments need to be considered carefully. investment expands in the targeted industry, Strategies that involve the strongest form of it is dif�cult to know whether the induce- commitment—allowing �rms to enforce ments were necessary or cost-effective. treaty commitments against governments Indeed, there are many examples of selective directly through binding international arbi- interventions going spectacularly wrong—at tration—can enhance credibility but would best wasting public resources, but sometimes bene�t from efforts to improve the trans- creating large distortions that harm the parency of the arbitration process. Strate- investment climate, and distracting attention gies that rest more on the reputation con- from broader improvements. cerns of governments can also contribute to Even in the best of circumstances, many policy credibility, but their impact will selective interventions seem to be a gamble. depend on whether participants insist on The more ambitious the goal and the weaker high levels of mutual compliance. the governance, the longer the odds of suc- cess. Selective interventions should thus be Harmonizing rules and standards. To approached with caution, and not viewed as reduce costs in international transactions, a substitute for broader investment climate many efforts focus on harmonizing particu- improvements. The hazards of such strate- lar rules or standards, with examples rang- gies can be reduced by ensuring that ing from the harmonization of business schemes have a clear objective and rationale, laws in West Africa to the development of focus on the sources of problems rather uniform accounting standards. There can than the symptoms, match the instrument be bene�ts for developing countries. But to the rationale, impose discipline on their there can also be tradeoffs with adapting bene�ciaries, are administered transpar- approaches to local conditions and with ently, and are reviewed regularly. allowing a degree of competition between approaches. There are also tradeoffs among multilateral, regional, and bilateral International rules and standards— approaches to harmonization. many tradeoffs The body of international rules and stan- Addressing international spillovers. Over dards dealing with investment climate mat- the past two decades, concerted global ters has grown exponentially in recent action has been promoted for a growing decades. There are now more than 2,200 number of matters where the effects of pol- bilateral investment treaties, over 200 icy actions by one country may spill over regional cooperation arrangements, and a onto others. Addressing international plethora of new and proposed multilateral spillovers in the environmental area is instruments covering most aspects of the important for sustainable development. investment climate. International arrange- When the suggested spillover is less tangible ments have a clear role in reducing barriers or the bene�ts less evenly shared, coopera- to international trade and investment. But tive action is more dif�cult. Proposals in they might also contribute to investment these and other areas need to give due climate improvements in three broader weight to the perspectives of developing ways: by enhancing credibility, by harmo- countries. (c) The International Bank for Reconstruction and Development / The World Bank 14 WORLD DEVELOPMENT REPORT 2005 The international community times the development assistance currently can lend a hand provided for investment climate improve- ments. Helping to improve investment climate conditions in developing countries can pro- Providing more, and more effective, vide huge development dividends. The manufacturing value-added unleashed by assistance investment climate improvements in even a The international community has long pro- single country can far exceed the develop- vided development assistance to support ment assistance provided worldwide (�gure the design and implementation of invest- 13). The international community can help ment climate improvements. Substantial developing countries in three main ways: by support is also provided directly to �rms. removing distortions in developed coun- There is room to do better in both areas. tries that harm the investment climates of developing countries; by providing more, Development assistance for investment cli- and more effective, assistance; and by tack- mate improvements. Around one quarter of ling the substantial knowledge agenda. of�cial development assistance, or around $21 billion per year, currently focuses on Removing distortions in developed support to investment climate improve- countries ments, with the bulk directed to infrastruc- ture development.22 Technical assistance Developing countries are not alone in grap- plays an important role, but represents just pling with investment climate improve- 13 percent of assistance for the investment ments. The trade and market distortions climate, and its effectiveness can suffer from created by policies in developed countries supply-driven approaches and from inade- impose large costs on their own economies. quate attention to ensuring recommended These distortions also undermine opportu- solutions reflect a good �t with local condi- nities and incentives for �rms to invest in tions. developing countries. It has been estimated that removing trade protection and related Support provided directly to �rms and distortions in developed countries could transactions. Well-designed support of this provide gains to developing countries of kind can complement investment climate $85 billion by 201521—or more than four improvements. Development assistance to support small �rms through credit lines Figure 13 Manufacturing value-added in a single country can far exceed net global and capacity building has a mixed track of�cial development �nance record, and would bene�t from the same 500 guidelines suggested for selective interven- China tions made by governments. Developed countries and international agencies also 400 provide around $26 billion per year in non- concessional loans or guarantees to support Billions of 1995 dollars 300 speci�c transactions. Increasing the empha- sis on the contribution these transactions South Korea make to the creation of more transparent 200 and competitive markets would expand the development impact of this support. Global net official 100 development finance India Tackling the substantial knowledge agenda 0 New sources of data of the kind drawn on 1970 1975 1980 1985 1990 1995 2000 in this Report add to our understanding of Source: OECD online database (www.oecd.org) and World Bank (2004k). the foundations of growth and poverty (c) The International Bank for Reconstruction and Development / The World Bank Overview 15 BOX 4 Main messages from World Development Report 2005 The investment climate is central to 25 percent of sales—or more than three • to restrain corruption and other forms of rent growth and poverty reduction times what �rms typically pay in taxes. seeking that increase costs and distort poli- Improving the opportunities and incentives for • Firms in developing countries rate policy cies; �rms of all types to invest productively, create uncertainty as their top concern.This and • to build policy credibility to give �rms the jobs, and expand should be a top priority for other sources of policy-related risk—such as con�dence to invest; governments. It is not just about increasing the insecure property rights, macroeconomic • to foster the public trust required to enable volume of investment but also spurring produc- instability, and arbitrary regulation—chill and sustain policy improvements; and tivity improvements that are the keys to sustain- incentives to invest. Improving policy able growth. • to ensure policy responses are crafted to �t predictability can increase the likelihood of local conditions. • The goal is to create a better investment cli- new investment by over 30 percent. mate for everyone. A good investment • Barriers to competition bene�t some �rms climate bene�ts society as a whole, not just but deny opportunities and increase costs to �rms. And it embraces all �rms, not just large other �rms and to consumers.They also Investment climate improvements are a or politically connected �rms. weaken incentives for protected �rms to process, not an event innovate and improve their productivity. Government policies and behaviors influencing • Expanding opportunities for young people is Increasing competitive pressure can increase the investment climate cover a wide �eld. But a pressing concern for developing countries, the probability of �rm innovation by more everything does not have to be �xed at once, where 53 percent of people live on less than than 50 percent. and perfection on even a single policy dimen- US$2 a day, youths have more than double sion is not required. Signi�cant progress can be the average unemployment rate, and popula- made by addressing important constraints fac- tions are growing rapidly. Progress requires more than changes to ing �rms in a way that gives them the formal policies con�dence to invest—and by sustaining a Reducing unjusti�ed costs is critical, but Over 90 percent of �rms claim gaps between for- policy-related risks and barriers to process of ongoing improvements. mal rules and what happens in practice, and the competition also need to be tackled informal economy accounts for more than half of • Because constraints differ widely across and All three matter for �rms and thus for growth output in many developing countries. Creating a even within countries, priorities need to be and poverty reduction. better investment climate requires governments assessed in each case. Reform processes ben- • Costs associated with weak contract enforce- to bridge these gaps and to tackle deeper e�t from effective public communication and ment, inadequate infrastructure, crime, cor- sources of policy failure that undermine a sound other measures to build consensus and main- ruption, and regulation can amount to over investment climate.This requires efforts: tain momentum. reduction. But a long agenda lies ahead in By working together on these themes, broadening and deepening this understand- the international community can do a lot to ing to provide guidance to policymakers. help create better investment climates in This includes expanding the development developing countries—and so contribute to of objective indicators of the investment cli- a more balanced, inclusive, and peaceful mate and the systematic analysis of country world. experiences to distill emerging lessons. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank Improving the Investment Climate I THE REPORT ARGUES THAT THE INVESTMENT CLIMATE plays a central role in growth and poverty reduction. This part shows why improving the investment climates of their societies should be a top priority for PA RT governments, and looks at how the necessary improvements can be made. Chapter 1—The investment climate, growth, and poverty shows how governments influence the investment climate and how improv- ing the investment climate drives growth and reduces poverty. Chapter 2—Confronting the underlying challenges looks at why improving the investment climate can be dif�cult, and the sources of potential policy failure that governments must face. Chapter 3—Tackling a broad agenda reviews international experi- ence in making investment climate improvements and suggests practi- cal strategies for accelerating and broadening progress. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 1 A good investment climate fosters produc- more are speci�c to each location, to the tive private investment—the engine for investment climate in its broadest sense. growth and poverty reduction. It creates Governments may have limited influence chapter opportunities and jobs for people. It over such factors as geography. But they expands the variety of goods and services have much more influence over the security available and reduces their cost, to the ben- of property rights, the approaches to regu- e�t of consumers. It supports a sustainable lation and taxation (both at and within the source of tax revenues to fund other impor- border), the adequacy of infrastructure, the tant social goals. And many features of a functioning of �nance and labor markets, good investment climate—including ef�- and broader features of governance such as cient infrastructure, courts, and �nance corruption. markets—improve the lives of people Earlier work looking at differences in directly, whether they work or engage in incomes across countries highlighted the entrepreneurial activities or not. role of “institutions�—the broad organi- Improving the investment climate— zational framework governing market the opportunities and incentives for �rms transactions. New sources of data drawn to invest productively, create jobs, and on in this Report allow us to go further expand—is the key to sustainable progress and provide fresh insights into how the in attacking poverty and improving living details of institutional arrangements vary standards (box 1.1). Varying enormously across and within countries and influence around the world, both across and within the level and productivity of private countries, the investment climate influ- investment. ences the decisions of �rms of all types: The chapter then looks at how variations the decision of the farmer to sow more in government policies and behaviors affect seed; the decision of the microentrepre- the investment climate—and thus growth neur to start a business; the decision of the and poverty. The key is to remove unjusti- local manufacturing company to expand �ed costs, risks, and barriers to competition its production line and hire more workers; faced by �rms of all types. An investment the decision of the multinational to locate climate that encourages growth creates sus- its next global production facility. tainable jobs and opportunities for This chapter looks at how improving microentrepreneurs—the key pathways out government policies and behaviors that of poverty for poor people, pathways that shape the investment climate matters not will become more crowded with coming only for �rms—it also drives growth and demographic changes. A good investment improves opportunities for everyone. The climate also helps to reduce the costs of chapter opens by looking at what we know goods consumed by poor people, and about the investment climate. Some of the improves the living conditions of poor peo- many factors influencing the decisions of ple directly. It also contributes to an �rms to invest productively, create jobs, and expanding tax base that allows governments expand are speci�c to each �rm—its ideas, to invest in the health, education, and wel- its capabilities, and its strategies. Many fare of its people. 19 (c) The International Bank for Reconstruction and Development / The World Bank 20 WORLD DEVELOPMENT REPORT 2005 The key message: for governments at all Understanding the investment levels, a top priority should be to improve the climate investment climates of their societies. To do so, they need to understand how their poli- Firms invest to make pro�ts. Their invest- cies and behaviors shape the opportunities ment decisions are affected by their own and incentives facing �rms of all types, ideas, capabilities, and strategies, and by their domestic and foreign, formal and informal, assessment of the opportunities and incen- small and large, urban and rural. The agenda tives in particular locations. Early efforts to is broad and challenging, but delivering on it understand how governments influence these is the key to reducing poverty, improving liv- location-speci�c factors focused on broad ing standards, and creating a more inclusive, indicators of country risk, often based on balanced, and stable world. surveys of international experts and usually BOX 1.1 What do we mean by the investment climate? goal, the focus could be narrowed to minimizing costs and risks. It is about improving outcomes for society. Many costs and risks are properly borne by �rms. And reducing barriers to compe- tition expands opportunities, spurs innovation, and ensures that the bene�ts of productivity improvements are shared with workers and consumers. A good investment climate is one that bene�ts everyone in two dimensions. First, it serves society as a whole, rather than just �rms, including through its impact on job cre- ation, lower prices, and broadening the tax base. Second, it embraces all �rms, not just large or influential �rms. The vertical plane in the �gure represents the investment climate. Some aspects of the invest- ment climate, including geography and market size, are dif�cult for governments to change. But governments have more decisive influence over a range of other factors.The speci�c influences addressed in the Report are policies closely tied to investment behavior.Thus, the forward- looking nature of investment points to the importance of stability and security, especially the security of property rights (chapter 4). Regu- lations and taxes qualify property rights and have �rst-order implications for costs, risks, and barriers to competition (chapter 5). Finance, infrastructure, and labor are the key inputs to The investment climate is the set of location- The horizontal plane in the diagram above investment activities (chapters 6 and 7). speci�c factors shaping the opportunities and represents their investment decisions.Firms But �rms do not respond to formal policies incentives for �rms to invest productively, create decide whether to incur costs today to change or alone.They make judgments about how those jobs, and expand. Government policies and augment production in the future, such as invest- policies will be implemented in practice. And behaviors exert a strong influence through their ing in machinery, facilities, and research and devel- �rms (like other stakeholders) will try to influence impact on costs, risks, and barriers to competi- opment.Firms come to the decision with different policies in ways favorable to them.Thus, issues of tion—and are the focus of this Report. capabilities and strategies.Their decision is moti- government behavior and governance, in the Firms are the starting point of the vated by the quest for pro�ts—and pro�tability is broadest sense, are paramount (chapter 2). It is framework.This Report uses that term to include influenced by the costs, risks, and barriers to com- the interaction of formal policies and governance the full range of economic agents ranging from petition associated with the opportunity.The vol- that �rms assess in making investment decisions. individual farmers and microentrepreneurs to ume and productivity of the resulting investment This has important implications for strategies to domestic manufacturing establishments and contribute to growth and poverty reduction. improve the investment climate (chapter 3). multinationals, regardless of their size, activity, A good investment climate is not just about or formal legal status. generating pro�ts for �rms—if that were the (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 21 resulting in a single score for each country.1 Figure 1.1 Institutions, broadly measured, clearly matter for growth Many studies focused on the narrower ques- tion of the constraints facing foreign �rms. 12 The last 20 years have seen a broadening and Log GDP per capita, 2001 deepening of efforts to understand how vari- ous location-speci�c factors influence differ- 10 ences in incomes across countries. Researchers began by looking at various aggregate indicators of a country’s institu- 8 tional and policy environment, such as the rule of law, corruption, openness to trade, legal origins, and �nancial sector depth.2 Their work generated useful insights—the 6 –2 –1 0 1 2 most important is that secure property rights Aggregate governance measure and good governance are central to eco- Note: The horizontal axis represents the average of “rule of law,� nomic growth (�gure 1.1).3 However, relying “government effectiveness,� “regulatory quality,� and “control of corruption� as de�ned by Kaufmann, Kraay, and Mastruzzi (2003). on aggregate indicators and cross-country The variables are normalized so that the average is at 0, and the regressions provides limited insights into the standard deviation equals 1. Source: Kaufman, Kraay, and Mastruzzi (2003). heterogeneity of institutional arrangements across and within countries—or the impact within countries—and impact on �rm per- of those arrangements on the investment formance, growth, and poverty. decisions of different types of �rms.4 It is The opportunities and incentives �rms also dif�cult to distinguish the effects of spe- have to invest productively, create jobs, and ci�c policy actions from the broader back- expand can be traced through their impact ground institutions that influence the con- on expected pro�tability. And pro�tability is tent and impact of those actions.5 influenced by the costs, risks, and barriers to These limits inspired the search for more competition associated with particular disaggregated evidence on the quality of a opportunities. Each factor matters indepen- location’s investment climate and for ways dently, and all three are interrelated. Some to trace the impact of that climate on the risks can be mitigated by incurring greater investment decisions and performance of costs. High costs or risks can be barriers to �rms. The World Bank is contributing to competition. Barriers to competition can this work in several ways, including Invest- reduce risks for some �rms but deny oppor- ment Climate Surveys and the Doing Busi- tunities and increase costs for others. ness Project (box 1.2). These and other new Many factors shape the costs, risks, and sources of data provide fresh insights about barriers to competition in a particular how investment climates vary across and location. Factors like geography are dif�cult BOX 1.2 New sources of investment climate data from the World Bank The World Bank recently launched two major launched in 2001, with about 20 new surveys selected experts (lawyers, accountants). initiatives to understand more about the deter- conducted each year since.This Report draws Underlying information includes the time and minants of growth and productivity. on early results from this work, which covers costs of complying with various areas of regu- more than 26,000 �rms in 53 countries, and lation—including business registration, con- • Investment Climate Surveys. Large random together employ some 4.8 million people.The tract enforcement, and labor regulation. A �rst samples of �rms have been interviewed to Investment Climate Surveys build on the report was published in 2003, with annual collect assessments of constraints facing World Business Environment Surveys, updates scheduled with additional topics. �rms including governance, regulation, taxa- launched in 1999, which covered smaller tion, �nance, infrastructure, and labor.The sur- samples of �rms and relied more heavily on Selected data from these sources appear at veys also collect objective data, which allow perception data. the back of this Report. investment climate indicators to be linked • Doing Business Project. Covering over 130 This Report complemented these initiatives with �rm performance to understand their countries, this project reports on the costs of by surveying 3,250 entrepreneurs in the infor- impact on productivity, investment decisions, doing business for a de�ned hypothetical �rm mal sector in 11 countries recently completing and employment decisions.The surveys were and transaction based on the views of Investment Climate Surveys. (c) The International Bank for Reconstruction and Development / The World Bank 22 WORLD DEVELOPMENT REPORT 2005 how government policies and behaviors BOX 1.3 Geography matters, but it is not destiny interact as part of a package to influence the costs, risks, and barriers to competition Some aspects of the investment climate are agriculture and tourism. And countries in more dif�cult for governments to change malaria-affected regions face special disad- associated with particular opportunities. than others.The most important of them is vantages. The new data show how costs, risks, and geography, which can have direct and indi- Large endowments of natural resources barriers to competition can affect �rms’ rect effects on the investment climate. were once thought to be a big advantage. investment behavior—and how they vary Countries with large domestic markets, But such concentrations of wealth have con- or near larger markets, may be more attrac- sumed some societies in rent-seeking, raising around the world. tive to investors than smaller or more the question of whether such endowments remote markets, though moves toward are always a blessing (chapter 2). Costs more open trade and advances in Whatever the weight of geography, it is transportation and communications are clear that efforts to improve aspects of the The costs of producing and distributing reducing the gap. Within countries, low investment climate more amenable to gov- products influences the range of opportuni- population densities and distances from ernment influence can provide large payoffs. ties that may be pro�table. Many costs to markets can also affect the attractiveness of Such efforts help a society make the most of �rms are a normal function of commercial rural areas, though investments in its innate resources—physical and human. infrastructure can reduce that gap as well. activity, while others flow directly or indi- Climatic variables can also influence the Source: Easterly and Levine (2003) and Gallup, rectly from government policies and behav- feasibility of some types of activity, such as Sachs, and Mellinger (1999). iors. The most obvious direct cost is taxation. But governments have important roles in providing public goods, supporting the pro- to influence (box 1.3). Governments have vision of infrastructure, and mitigating other more decisive influence over many other market failures. The ways they do this can aspects of the investment climate, such as have a big impact on the costs that �rms face. the security of property rights, approaches For example, the costs associated with crime, to regulation and taxation, the adequacy of corruption, regulation, unreliable infrastruc- infrastructure, and the functioning of ture, and poor contract enforcement can �nance and labor markets (table 1.1). Gov- amount to over 25 percent of sales—or more ernment policies on these subjects inter- than three times what is typically paid in act—for example, secure rights to land can taxes. The level and composition of these ease access to �nance. Moreover, the content costs vary widely (�gure 1.2). The time costs and impact of formal policies in these areas of complying with particular regulatory are determined by broader features of the requirements also vary widely. For example, governance environment, including corrup- registering a new business can take 2 days in tion and credibility (chapter 2). Firms assess Australia, but over 200 days in Haiti.6 Table 1.1 Government policies and behaviors and investment decisions—some examples Factors that shape opportunities and incentives for �rms to invest Government has strong influence Government has less influence Costs • Corruption (chapter 2) • Market-determined prices of inputs • Taxes (chapter 5) • Distance to input and output markets • Regulatory burdens, red tape (chapter 5) • Economies of scale and scope associated with particular technologies • Infrastructure and �nance costs (chapter 6) • Labor market regulation (chapter 7) Risks • Policy predictability and credibility (chapter 2) • Consumer and competitor responses • Macroeconomic stability (chapter 4) • External shocks • Rights to property (chapter 4) • Natural disasters • Contract enforcement (chapter 4) • Supplier reliability • Expropriation (chapter 4) Barriers to competition • Regulatory barriers to entry and exit (chapter 5) • Market size and distance to input and output markets • Competition law and policy (chapter 5) • Economies of scale and scope in particular activities • Functioning �nance markets (chapter 6) • Infrastructure (chapter 6) (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 23 Risks Figure 1.2 Costs vary widely in level and composition Investment decisions are forward looking, 30 allocating resources today in the hope of Contract enforcement difficulties future rewards. Many investment risks, like Regulation 25 costs, are a normal function of commercial Bribes ventures, including uncertain responses Crime Unreliable infrastructure Cost as percent of sales from consumers and competitors, so �rms 20 should bear them. Governments, however, have an important role in helping �rms 15 cope with risks associated with the security of their property rights. Governments can 10 also increase the risks and uncertainties that �rms face directly—policy uncertainty and macroeconomic instability rank consistently 5 as the leading investment climate concerns of �rms (chapter 2). Unpredictability in the 0 interpretation of regulations is often a big Poland China Brazil Algeria Tanzania concern (�gure 1.3). And almost 95 percent Note: The survey asked registered �rms to report values either in monetary terms, directly as a share of sales, or in of �rms report a gap between formal poli- terms of time. “Contract enforcement dif�culties� captures the share of inputs that are below agreed-upon quality (weighted by material inputs in total sales) and overdue payments (as a share of total payments, using an interest rate cies and their implementation. of 10 percent for the average length of overdue payments). “Regulation� captures management time spent dealing with of�cials (weighted by the cost of management compensation to total sales), and the gap in actual employment Assessing the impact of risks is compli- relative to desired levels due to regulatory costs associated with hiring and �ring workers (weighted by total labor cated by the different ways �rms respond— costs in sales). “Bribes� are the total costs of bribes as a share of sales. “Crime “ is the sum of losses due to theft, security costs, and protection payments (as a share of sales). “Unreliable infrastructure� includes sales lost due to demanding higher returns, adopting interruptions in power and telecommunications and due to the loss or damage of goods in transit. Countries selected shorter planning horizons, or not investing to illustrate range. Source: World Bank Investment Climate Surveys. at all. Firms operating in some high-risk countries require more than twice the rate Figure 1.3 Regulatory unpredictability Figure 1.4 Competitive pressure can of return they would in lower-risk countries is a big concern for �rms vary signi�cantly between countries to compensate for the extra risks.7 Firm- 100 100 level surveys show that improving policy predictability can increase the probability of Guatemala Percent of firms reporting significant competitive pressure Poland new investment by more than 30 percent Hungary (chapter 2).8 Peru Slovakia Percent of firms finding the interpretation 75 Russia 75 Tajikistan Uzbekistan Zambia of regulations unpredictable Barriers to competition Pakistan Kyrgyzstan Firms naturally prefer less competition Indonesia rather than more. But a barrier to competi- 50 50 tion bene�ting one �rm denies opportuni- Estonia ties and increases costs for other �rms and to Georgia consumers. And competitive pressure drives �rms to innovate, to improve productivity, and to share the bene�ts of productivity 25 25 gains with consumers and workers. Many factors, including economies of scale and market size, can influence the level of com- petition in a market. Governments also 0 0 influence competitive pressure through their Note: Countries selected to illustrate the range of Note: Countries selected to illustrate range of regulation of market entry and exit—and responses. responses; data limited to Europe and Central Source: World Bank Investment Climate Surveys. Asia. their responses to anticompetitive behavior Source: World Bank Investment Climate by �rms. Competition is dif�cult to measure Surveys/BEEPS II. at the aggregate level, but �rm-level evidence shows how much competitive pressure can vary between countries (�gure 1.4). (c) The International Bank for Reconstruction and Development / The World Bank 24 WORLD DEVELOPMENT REPORT 2005 Figure 1.5 Investment climate conditions vary within countries The new evidence shows large variations in investment climate conditions not only Number of days to obtain a mainline telephone connection between countries, but also within coun- Number of days to clear customs for imports tries, as illustrated by China (�gure 1.5). 25 Percent of production lost because of power outages 7 This will often be the case with infrastruc- ture provision or when subnational govern- 6 ments determine policies. But even a single 20 national law may be applied differently 5 within a country: for example, the time to transfer property title in Brazil varies from 15 4 15 days in Brasilia to 65 days in Salvador.9 Percent Days Even within a single location, the same 3 conditions can affect �rms differently. This 10 can be true across activities—farmers, man- 2 ufacturers, and barbers each have different perspectives. But a poor investment climate 5 1 often hits smaller and informal �rms the hardest (�gure 1.6). 0 0 Bejing Chendu Guangzhou Shanghai Tianjin How investment climate China improvements drive growth and Source: World Bank Investment Climate Surveys. reduce poverty With rising populations, economic growth is Improving the investment climate is not the only sustainable mechanism for increas- about reducing all costs, all risks, and all ing a society’s standard of living. Growth is barriers. Taxes and regulation support a associated not just with higher incomes, but sound investment climate and protect with better indicators of human develop- broader social interests. Managing the ten- ment, such as lower infant mortality, broader sion between creating a favorable invest- education, and longer life expectancy. It pro- ment climate for �rms and achieving other vides opportunities for �rms of all types, cre- social goals is a major challenge for govern- ating jobs and expanding the tax base avail- ments—and a key theme of this Report. able to fund public services. Households as Figure 1.6 Investment climate conditions can affect well as �rms bene�t from better property �rms differently rights, �nancial markets, and infrastructure services. It is also now widely understood 75 that growth must be sustainable, safeguard- Large ing the value of national assets—including environmental assets—and the potential for Percentage of firms 50 future growth (box 1.4). A growing body of Medium research shows how investment climate poli- Small cies contribute to economic growth, and how policy approaches might be tailored to 25 better target the needs of poor people. What has been learned? Informal 0 Signi�cant economic growth is a Have a loan Confident Believe modern phenomenon, not shared by all from a that courts regulations formal will uphold will be Some early economists were concerned that financial property interpreted institution rights consistently the potential for rising incomes was inher- ently limited, while mercantilists believed Note: Uses 10 countries for which formal and informal surveys were conducted, controlling for industry, country, ownership, and that growth was a zero-sum game, with �rm age. Source: World Bank Investment Climate Surveys and WDR Sur- gains by some countries coming only at the veys of Micro and Informal Firms. expense of others. For centuries the average (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 25 BOX 1.4 The environment matters for well-being and productivity: Main messages from WDR 2003 Growth in income and productivity is required poverty, but agricultural productivity has been Why are environmental assets particularly to eliminate poverty in developing countries, stagnant over the past four decades. Much of threatened and underprovided? Because of but it needs to be environmentally sustainable. the cropland is degraded, and hillside erosion spillovers.The actions of one person may The immediate gains of depleting or degrading clogs downslope waterways.The country’s per impose environmental costs, such as pollution, environmental assets can be outweighed by the capita GDP slid from $383 (in 1995 dollars) in on other people—costs that the responsible costs in productivity and lost options. Over the 1960 to $246 in 2002. party does not bear. Addressing these environ- longer run, economic growth is unlikely to be Environmental conditions will only worsen mental problems requires governments to take sustained unless attention is paid to assets such if present trends continue. People in hundreds a long-term view and manage a broad portfolio as fresh water and �sh stocks. of developing-country cities live with of assets that includes not only human and Even in the short to medium run, addressing unhealthy air, which causes premature deaths, physical capital but also environmental assets. the objectives for growth and the preservation preventable at a modest cost. Nearly 23 per- Policies that have proved successful in solving or restoration of environmental assets can be cent of all cropland, pasture, forest, and wood- these problems are those that align individual critical to raising production and incomes. Con- land worldwide has been degraded since the incentives with social incentives—including sider Madagascar, where the conversion of bio- 1950s. Local conflicts over water and the loss those for property rights, regulation, taxes, and diversity-rich forests to mostly unsustainable of freshwater ecosystems loom in some subsidies. Such measures form an important low-yield agriculture has been costly. With regions. Two-thirds of all �sheries are part of a sound investment climate. three-quarters of the country’s people in rural exploited at or beyond their sustainable limits. areas and three-quarters of them poor, produc- Every decade another 5 percent of tropical for- tivity growth in agriculture is critical to reducing est is cleared. Source: World Bank (2003o). level of income did not change. This led to 1900—and that in Botswana, Mexico, and Malthus’ observation in 1798 that any rise Uruguay was �ve times greater.12 in income was quickly offset by a rise in Some countries have experienced tremen- population, leaving per capita incomes con- dous success, sustaining high growth rates over stant.10 Over the next hundred years, how- many years and achieving signi�cant reduc- ever, the leading countries doubled their per tions in poverty. China is the most striking capita incomes, with the speed accelerating recent example. India is another. Among over the 20th century (�gure 1.7). The time regions, East Asia has had the fastest sustained to double incomes fell from a millennium, growth, with Latin America more disappoint- to centuries, to just 20 to 30 years. ing in recent years and Africa suffering from Today the world’s per capita GDP is esti- stagnant and declining growth (�gure 1.8). mated to be at least �ve times what it was at Many countries in Eastern Europe and Central the beginning of the 19th century,11 and the Asia, after sharp declines in the early 1990s, are comparison actually underestimates the recovering their growth. While some growth achieved. It is a matter of looking Figure 1.7 Signi�cant economic growth is a modern phenomenon not just at real incomes to judge whether more goods can be purchased now— 25,000 Western Europe & North America because the goods available have changed Eastern Europe & former USSR dramatically. Inventions in medicine (peni- Latin America cillin, vaccinations), transportation (cars, 20,000 GDP in 1990 international dollars Asia (excluding Japan) Africa airplanes), and communications (mobile phones, e-mail) are just some examples of 15,000 new products greatly enhancing the quality, and even the length, of life. Using exchange rates that equalize the purchasing power of 10,000 different currencies, about two-thirds of the world’s people now live in a country with an average income more than that of the 5,000 United States a century ago. Taking into account new products, the average material 0 prosperity in Thailand or Tunisia in 2000 1600 1700 1820 1870 1913 1973 2001 was three times that of the United States in Source: Maddison (2003). (c) The International Bank for Reconstruction and Development / The World Bank 26 WORLD DEVELOPMENT REPORT 2005 Figure 1.8 Fast sustained growth in East Asia—declines in Sub-Saharan Africa productivity contribute to growth—and 8 how the investment climate determines the size of both contributions. Average percent growth of GDP per capita 7 6 1990s 1980s Investment and productivity 5 1970s The role of private investment has grown in 4 the last 20 years. Foreign direct investment 1960s has increased signi�cantly, but the bulk of 3 2 investment is by domestic �rms, reinforcing the importance of looking at the full spec- 1 trum of �rms in analyzing the investment 0 climate and its contribution to growth and –1 poverty reduction (�gure 1.9). –2 The investment climate has an obvious East Asia Latin Middle South Asia Sub-Saharan Europe role in influencing the level of private & Pacific America East & Africa & investment. The evidence con�rms that & North Africa Central Caribbean Asia improving the opportunities and incentives Note: Data for the Europe and Central Asia Region begin in the 1980s. for �rms to invest by reducing unjusti�ed Source: World Bank (2004k). costs, risks, or barriers has the predicted effect. For example, farmers in Thailand with secure rights invested so much more in developing countries have converged on the their land that their output was 14–25 per- income levels of the richest countries, limited cent higher than those working untitled progress by the poorest countries means that land of the same quality (chapter 4). Dis- incomes between the richest and poorest mantling monopolies in telecommunica- have diverged.13 Too prevalent are the periods tions around the world unleashed a dra- of short-lived growth—and of continued matic rise in investment in the sector, decline. Igniting a growth spurt is clearly pos- including that by microentrepreneurs in sible. The challenge is to sustain it.14 Bangladesh (chapter 6). At the aggregate The search for a magic formula that level, improvements in the investment cli- would guarantee faster economic growth mates in countries as diverse as China, has been a long-standing but elusive India, and Uganda have been marked by quest.15 Recent research, however, provides strong growth in private investment (box important insights on how investment and 1.5). Cross-country evidence using broad proxies for investment climate quality con- Figure 1.9 The contribution of private investment to GDP has grown �rm the link between the investment cli- 20 mate and private investment (�gure 1.10). Investment rates by themselves are not the main driver of growth. Capital accumu- lation brings more inputs to the production 15 process, but there is a limit to how much this process can sustain growth because of Percent of GDP Private gross fixed capital formation the decreasing marginal impact of addi- 10 tional capital. So, the measure of success of an investment climate is not the quantity of investment—it is the quality of investment, 5 and quality is also influenced by the invest- ment climate. FDI Indeed, experience provides many exam- ples of investment projects that yielded few 0 or no bene�ts. This is most obvious with 1980 1990 2000 “white elephant� projects in the public sector, Note: Annual averages of 92 developing countries. Source: World Bank (2004k). such as the Tanzanian shoe factory that pro- (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 27 BOX 1.5 Improving the investment climate and growth: the cases of China, India, and Uganda China and India have grown impressively in with more extensive trade liberalization and a to improve the opportunities and incentives for recent years, greatly reducing poverty. In both further dismantling of the so-called licensing �rms to invest productively. cases the roots can be found in improving the Raj. The bene�ts of a better investment climate investment climate. Beginning in the early 1980s The results? Private investment as a share of are not limited to large countries.Take Uganda. China introduced rudimentary systems of prop- GDP nearly doubled in both countries. Per Many countries in Africa have experienced lim- erty rights and private enterprise, liberalized capita GDP in China rose tenfold from $440 in ited or negative growth, with investment trade and investment, and embraced a broad 1980 to $4,475 in 2002 (in international prices), climates often clouded by historical legacies, program of improvements across the and India’s almost quadrupled from $670 in political instability, excess government interfer- investment climate. India introduced reforms to 1980 to $2,570 in 2002. Both experienced dra- ence, and other factors that stifle opportunities reduce tariffs and loosen licensing requirements matic reductions in poverty (see �gure)—each and incentives for �rms to invest productively. in the mid-1980s, followed in the early 1990s on distinctive paths, but both sustaining efforts Beginning in the early 1990s, however, Uganda embarked on a program to improve its invest- ment climate. Macroeconomic stability was Share of population below the $1 a day poverty line 18 75 achieved. Expropriations by a previous govern- Private investment Poverty ment were reversed.Trade barriers were reduced.Tax and court systems were reformed. Private sector participation and competition were introduced in telecommunications. Now 12 50 efforts are under way to improve business regu- Percent of GDP lation. While many challenges remain, these efforts are reaping rewards.The share of private investment in GDP more than doubled between 1990 and 2000. Per capita GDP grew by over 4 6 25 percent from 1993 to 2002 (8 times the average in Sub-Saharan Africa).The percentage of the population living below the poverty line fell from 56 percent in 1992 to 35 percent in 2000. 0 0 Source: Ahluwalia (2002); Chen and Ravallion (2004); 1990 1999 1981 2000 1990 2002 1981 2001 1981 2001 1992 2000 De Long (2003); Chen and Wang (2001); Qian (2003); China India Uganda China India Uganda Rodrik and Subramanian (2004);Young (2003);Young (2000); Holmgren and others (2001);World Bank Note: Dates reflect the timing of signi�cant reform efforts and, particularly for China, the availability of data. Poverty (2002d);World Bank (2001d); and IMF and IDA (2003); data for Uganda based on national poverty line. World Bank (2004k); IMF (2004). duced few shoes, the nuclear power plant in mance across countries. Over 1960–2000 the the Philippines that was never commis- bulk of the differences in growth between sioned, and the numerous roads to countries (45–90 percent) is accounted for Figure 1.10 Private investment has nowhere.16 The former Soviet Union also had not by the accumulation of physical capital, grown faster in countries with better very high investment rates in the 1950s, but or of human capital, but by total factor pro- investment climates too often in projects that provided little eco- ductivity (TFP)—the productivity contribu- 12 Annual growth of private investment nomic or social return. tions above those made by physical and Reflecting this, cross-country studies �nd human capital (�gure 1.11 and box 1.6).18 As 8 little correlation between aggregate invest- Krugman said, “Productivity isn’t everything, ment and growth, particularly if no distinc- but in the long run it’s almost everything.�19 4 tion is made between public and private Aggregate-level studies differ in the 0 investment.17 This highlights the importance weight they attach to TFP and factor accu- of ensuring that investment is undertaken mulation in explaining economic growth.20 –4 with some discipline to improve the likeli- The debate is important because it has hood of it being productive. That discipline implications for the sustainability of –8 3 4 5 6 7 8 9 will most likely be forthcoming when private growth. If growth is due to factor accumu- Investment profile �rms put their own money at risk to invest in lation, the diminishing marginal contribu- Note: Data presented are averages for a competitive business environment, so that tion of capital implies that high growth 1984–2000. ICRG’s index of “investment pro�le� they bear the consequences of their invest- rates, such as those achieved in East Asia, is based on measures of contract enforceability, expropriation, pro�t repatriation, and payment ment decisions. will not be sustainable. However, the same delays. Higher numbers are associated with less The critical role of productivity is under- limitation is not true for gains in TFP. In risk and stronger investment climates. Source: World Bank (2004k) and International scored by studies of aggregate growth perfor- practice the distinction between invest- Country Risk Guide. (c) The International Bank for Reconstruction and Development / The World Bank 28 WORLD DEVELOPMENT REPORT 2005 Figure 1.11 Differences in TFP account for the largest This provides encouraging news for devel- share of differences in GDP growth per worker. oping countries—improving investment cli- 4 mate conditions can directly improve ef�- TFP ciency, encourage the adoption of better Education technology, and strengthen incentives for Percent growth of GDP per worker 3 Capital investment in physical and human capital. Early growth research emphasized tech- 2 nological progress in explaining TFP, suggest- ing that differences in growth rates were dri- 1 ven by differences in the technologies adopted.21 The dramatic acceleration in income levels among the fast growing coun- 0 tries over the last 200 years can be understood 1 2 3 4 by improvements in technology. “Technol- Growth quartile ogy� in this sense, however, is not limited to –1 Note: Decomposition of sources of growth per worker; 1960–2000 scienti�c breakthroughs of the kind that for 62 developing countries. might merit a patent. It can also include more Source: Data from Bosworth and Collins (2003). modest advances, as well as new and better ways to organize production processes, inter- ment and TFP is not always straightforward. act with consumers, or distribute goods. For example, technological improvements Importantly, �rms and countries do not can encourage investment, and investment have to invent everything afresh. Even in can help to improve technology. And invest- countries that make some of the biggest con- ment climate improvements can spur both. tributions to innovation, the ratio of adapta- tion to innovation is extremely high— Jovanovic estimates it at 20 or 30 to 1 in the BOX 1.6 Measuring productivity United States.22 This highlights the huge Productivity is the key to growth—for indi- those made by the number of workers, their potential for developing countries to catch up viduals, for �rms, and for the economy as a skill level, and the machinery they use. In the with richer countries by creating an environ- whole. Increasing productivity means pro- above example it would capture the third ment that facilitates the diffusion of ideas ducing more with the same amount of and fourth sources of growth in labor pro- inputs.Two common measures are labor ductivity. In the macroeconomic literature, developed elsewhere, as well as the develop- productivity and TFP. studies initially emphasized differences in ment of new ones. The potential for catching Labor productivity is the value-added pro- technology. More recent work has expanded up is real. It took some of the �rst industrializ- duced by each unit of labor.Increases in labor this to reflect differences in institutional set- ing countries 40 to 60 years to double their productivity simply mean that an individual is ting (often proxied with measures of prop- able to produce more.How? Take the example erty rights security) or “social infrastructure� incomes in real terms, but others have done of a worker in the informal economy produc- that influence the opportunities and incen- this much faster—Costa Rica in 19 years start- ing garments from home.One possibility is tives to adopt new technologies and oper- ing in 1961, Jordan in 15 years starting in 1965, that she has access to more machinery—such ate ef�ciently.The latter measures are largely Taiwan, China, in 10 years starting in 1965.23 as greater access to a shared sewing machine. synonymous with what this Report refers to A second is that she has more skills or training as the investment climate. Recent research has emphasized that TFP in sewing.A third is that she has access to new Rather than being measured directly,TFP can also be understood to encompass more technology—such as a newer sewing is the residual that is not explained by differ- than just differences in technology.24 The machine.A fourth is that she works in an envi- ences in factor inputs. Calculations of TFP ronment that enables and provides stronger often generate debate because of dif�culties broader environment in which �rms oper- incentives to work ef�ciently—such as fewer in measuring capital stocks, questions of how ate matters too, whether this is understood dif�culties accessing raw materials, fewer dis- to attribute changes in the quality of factor in terms of property rights, institutions, or tractions dealing with bureaucratic inputs, and the assumptions needed to esti- the investment climate. A better investment harassment and demands for bribes, or less mate the necessary coef�cients. Despite chal- exposure to theft.Progress in any area allows lenges in measurement, it is not disputed that climate can directly improve productivity her to increase the number (and quality) of TFP makes a critical contribution to growth. by reducing unjusti�ed costs and risks garments she produces—and thus boosts her flowing from government policies and income.Her improved productivity is a contri- behaviors. By making it more attractive to Source: Acemoglu (2001); Barro and Sala-i-Mar- bution to economic growth ultimately reflected in macroeconomic statistics. tin (2003); Bosworth and Collins (2003); Easterly develop and adopt new and better ways of and Levine (2001); Hall and Jones (1999); Par- Total factor productivity (TFP) attempts ente and Prescott (2000); Klenow and doing things, a better investment climate to measure contributions to output beyond Rodríguez-Clare (1997); Young (1995). will help productivity through its impact on technology as well. Thus, at least as (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 29 important as reducing costs and risks is eliminating unjusti�ed barriers to the BOX 1.7 Growth with a poor investment climate— development, adoption, or adaptation of possible, but unlikely to be sustained new processes—and fostering competition Growth with a poor investment climate is protected sectors. Public investment to sus- to encourage �rms to take up those oppor- possible, but unlikely to be sustained. For tain growth led to severe debt problems— tunities (box 1.7). example, in the 1960s and 1970s Brazil and ultimately to a macroeconomic crisis. experienced strong growth while closing Subsequent efforts to improve the domestic markets to international competi- investment climate initially met with cau- Productivity and competition tion and pursuing heavy public investment tious responses from �rms. Many attribute through state-owned enterprises.The initial this to questions about the credibility of the Firms do not innovate or improve their pro- results were impressive, but the growth government’s commitment to reforms, par- ductivity from any sense of philanthropy, proved unsustainable. Protected �rms ticularly in the wake of repeated episodes of because the processes can be demanding lacked the incentives to improve their pro- macroeconomic instability. ductivity and fell further behind and disruptive. Most �rms would prefer the international best practices. Other �rms had “quiet life�—which Hicks noted was the best less access to new technologies and had to Source: Castelar Pinheiro and others (2001) and of all monopoly pro�ts.25 Instead, �rms pay higher prices for inputs supplied by Schor (forthcoming). adopt and develop new and better ways of doing business in response to the pressures they face to survive and prosper in a com- petitive marketplace.26 A sound investment BOX 1.8 Developing a product is a learning process—as climate supports the dynamic processes that Hyundai shows Schumpeter called “creative destruction.�27 Hyundai’s efforts to produce a car began in The team had to scrap eleven more It encourages �rms to experiment and learn, the 1960s. It purchased foreign equipment, broken prototypes before one survived the it rewards success, and it punishes failure hired expatriate consultants, and signed test. There were 288 engine design (box 1.8). The �rm-level surveys con�rm licensing agreements with foreign �rms. But changes, 156 in 1986 alone. Ninety-seven the importance of competitive pressure for the process was not a simple matter of test engines were made before Hyundai adopting the technology. Despite the train- re�ned its natural aspiration and incentives to innovate (�gure 1.12) and ing and consulting services of a foreign con- turbocharger engines, 53 more engines increase productivity.28 sultant and three experts, Hyundai were produced for durability improvement, Healthy market economies exhibit engineers repeated trials and errors for 14 88 more for developing a car, 26 more for fairly high rates of opening and closing months before creating the �rst prototype. developing its transmission, and 6 more for The engine block broke into pieces at its other tests, totaling 324 test engines. In �rms (box 1.9). In Organisation for Eco- �rst test. New prototype engines appeared addition, more than 200 transmissions and nomic Co-operation and Development almost every week, only to break in testing. 150 test vehicles were created before (OECD) countries, 5–20 percent of �rms No one on the team could �gure out why Hyundai perfected them in 1992. In 2003, the prototypes kept breaking down—cast- Hyundai sold close to 2 million vehicles enter and exit the market every year.29 ing serious doubts, even among Hyundai around the world. Firms that leave the market are the least management, on the company’s ability to productive, and their departures con- develop a competitive engine. Source: Kim (1997). tribute more than 20 percent of the pro- ductivity gains. New �rms are more pro- ductive—though it can sometimes take Figure 1.12 More competitive pressure, more innovation them several years before their productiv- 75 ity reaches that of incumbents.30 The Major probability of undertaking activity combined effect of net entry is substantial, pressure particularly in countries with fewer barri- Percentage increase in Moderate ers to entry (�gure 1.13). 50 pressure The contribution of new entrants to pro- Some ductivity is particularly strong in higher pressure technology sectors. There is also evidence 25 that sectors with many new entrants push incumbents to increase their productivity. Why might entry rates be strongly corre- lated with productivity growth by incum- 0 Introduce Upgrade Introduce new bents? Perhaps because new entrants are new product product technology attracted to productive sectors, or because Note: Surveys cover 27 countries in Eastern Europe and Central Asia. the new entrants stimulate incumbents to Source: World Bank Investment Climate Surveys/BEEPS II. (c) The International Bank for Reconstruction and Development / The World Bank 30 WORLD DEVELOPMENT REPORT 2005 BOX 1.9 Firm dynamics The private sector is not static, nor are individual to reallocation across sectors, in fact, much of tunities to smaller �rms as suppliers—often �rms.There is a large and ongoing reallocation of the reallocation of resources is from low- to accompanied by technical assistance and access output and jobs across �rms. Such dynamism is a high-productivity �rms within the same sector. to credit. Particularly when �nancial markets are sign of a vibrant economy, and accounts for a sig- There are large differentials in the levels and less developed, large �rms can be an important ni�cant share of productivity growth.This is true rates of growth of productivity across �rms source of credit to smaller suppliers. across OECD countries and in developing coun- within a sector, and low productivity helps pre- Economies of scale speci�c to particular tries. Firms are forced to compete in their search dict exit. technologies help de�ne the minimum ef�cient for pro�ts.There are enticements, such as the lure The evidence underscores the importance size of a �rm, but in practice there is a large of larger pro�ts, even if they are short lived. And of the process of creative destruction to the range of �rm sizes within the same sector. Some �rms dare not be left behind.This is the secret to a growth process. Barriers to exit need to be of this can be due to concerns about contract- market economy’s success and what Schumpeter addressed to free up resources that can be used ing, with some �nding it optimal to keep activi- called “the essential fact of capitalism.� more productively in other activities. Barriers to ties in-house.The inability to access credit or The role of entry and exit. Every year between entry can be particularly harmful, not just by sti- other investment climate constraints can keep 5 and 20 percent of �rms enter or exit an econ- fling the pressure to innovate and leading to �rms small. Large �rms can face challenges in omy. Many of the entrants are small. Most of more “technological sclerosis,� but by organization and can be less agile in responding them will remain small. Some will grow, with a forestalling the creation of new jobs. However, to change. few becoming the large �rms of the future. Firms the churning process can be disruptive, and the It is not that countries should aim to have a also contract and some will go out of business. government has a role in helping workers cope particular size distribution of �rms. Rather, what This entry and exit of �rms is an inherent part of a with change (chapter 7). Improving the invest- is important is allowing the selection market economy and an important source of ment climate is central to ensuring the process mechanism to work free of political interference innovation. Reducing barriers to entry is impor- of creative destruction works well—to the bene- that favors influential �rms. Large �rms often tant because new entrants—and even the threat �t of workers and society as a whole. have more political influence and try to use this of new entry—spur existing �rms to improve Implications of �rm size. Beyond entry and to manipulate policies to their advantage— their productivity. Entering �rms also tend to use exit, these same pressures impact on �rm size often at the expense of smaller �rms. A good newer technologies and production methods. It and growth. Large �rms do not grow as fast as investment climate facilitates the allocation of is not that they are all more productive from their small ones, but they are more likely to survive. resources, fosters innovation, and encourages beginning—not even in comparison to exiting Large �rms tend to be more productive, pay the selection of �rms that increase productivity �rms. Experience in the marketplace will deter- higher wages, and offer greater job security.The and so contribute to growth and higher living mine which �rms will be successful.The highest causation, however, runs from productivity to standards. exit rates are among small and young �rms. If size; �rms that are more productive are the ones �rms have survived the �rst �ve years, however, that are likely to grow. they are much more likely to remain in business The interactions between �rms can have Source: Bartelsman and others (2004); Klein and and to contribute to productivity growth. important implications for how they develop. It Hadjimichael (2003); Haltiwanger (2000); Roberts While trade theory predicts that much of is not always cutthroat. Firms at the top of a sup- and Tybout (1996); Schumpeter (1942); Caballero the adjustment to greater openness would lead ply chain tend to be large.They provide oppor- and Hammour (2000); Baumol (2002). Figure 1.13 The contribution of net entry to productivity is higher increase their productivity to maintain their when barriers to entry are lower market shares. Census data from develop- ing countries con�rm the importance of the Contribution of net entry Days to register a firm to productivity growth second explanation.31 The greater microeconomic flexibility Latvia associated with a good investment climate Netherlands helps �rms take advantage of internal U.K. opportunities. It also helps the economy Chile weather external shocks. Countries in Latin Korea America and East Asia with more microeco- nomic flexibility experienced less dramatic Finland declines in output and recovered faster than France those with less flexible economies.32 Argentina 80 60 40 20 0 20 40 60 80 Showing the potential returns of Note: Data cover manufacturing �rms from 1988–2000, using census data. investment climate improvements Source: Bartelsman and others (2004) drawing on World Bank Doing Business Project. Research showing the links between invest- ment climate improvements and improved �rm performance typically focuses on a sin- gle dimension of the investment climate, (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 31 such as aspects of property rights security or regulatory reform. The Bank’s Invest- BOX 1.10 Showing potential returns to investment ment Climate Surveys make it possible to climate improvements see how broader packages of policy The Bank’s Investment Climate Surveys link would increase by 15 percent and sales improvements can influence �rm perfor- �rm performance to objective measures of growth by 20 percent. mance by use of counterfactual compar- costs and risks affected by policy.This • If the investment climate for �rms in isons (box 1.10). For example, �rms in makes it possible to simulate how changes Dhaka, Bangladesh, matched that of states in India and provinces in China with in investment climate conditions might Shanghai, Dhaka would reduce its pro- contribute to improved productivity, sales, ductivity gap by 40 percent, and wages better investment climates show much and wages: could rise by 18 percent. For Calcutta the stronger growth and productivity than their effect is even larger: 80 percent of the • In India, �rms in states with poor invest- peers in states or provinces with less favor- ment climates have 40 percent lower pro- productivity gap could be closed, and able investment climates. The effects are ductivity than those in states with good wages could rise by 38 percent. large—improving the investment climate investment climates. could account for up to 80 percent of the • If Tianjin, a large port city east of Beijing, Source: Dollar, Hallward-Driemeier, and Mengis- could achieve the same investment cli- tae (2003b); Hallward-Driemeier, Xu, and Wall- differences in productivity among these mate as Shanghai, �rm-level productivity sten (2003); and Dollar and others (2004). locations. Sharpening the focus on from 54 to 63 years, infant mortality fell 40 poverty reduction percent, and evidence of malnutrition The investment climate clearly matters for dropped, too. growth. Even more important is under- The incomes of poor people can increase standing how investment climate improve- in two basic ways—if average incomes rise ments can enhance the situation of the and the distribution of income stays the nearly half the world’s people living on less same, or if the distribution of income shifts than $2 a day, especially the 1.2 billion peo- to become more pro-poor. Clearly the ple who barely survive on less than $1 a day. biggest impact is if growth is combined The relationship between the investment with a shift to a more equal distribution of climate and poverty reduction can be seen income. If the feedback from greater equal- in two ways: by looking at the links between ity reinforces growth processes, the growth and poverty reduction at the aggre- dynamic can signi�cantly reduce poverty gate level, and by looking at the ways invest- over time.36 ment climate improvements affect the lives of people directly. Figure 1.14 Poverty reduction is closely associated with growth The links with economic growth 50 Decline in share of population below $1 a day, 1981–2001 China There are almost no examples of countries East Asia other than China experiencing signi�cant growth without 40 reducing poverty.33 Growth in average incomes associated with broadly based 30 growth has been found to account for up to 90 percent of the reductions in poverty (�g- South Asia other than India ure 1.14).34 20 Uganda India Investment climate improvements in China and India have driven the greatest 10 reductions in poverty the world has ever Middle East & North Africa seen, and in China alone lifted 400 million Latin America & the Caribbean 0 people out of poverty (box 1.5).35 The Eastern Europe & Central Asia increases in income were also matched by Sub-Saharan Africa gains in health outcomes. In China, life –10 expectancy rose by four years, from 66.8 to –2 0 2 4 6 8 10 70.7 years from 1980 to 2002, and infant Average per capita growth rate, 1981–2001 mortality fell from 49 to 32 per 1,000 live Note: Data for Uganda are from 1992–2000 and uses its national poverty level due to data availability. births. In India, life expectancy increased Source: Chen and Ravallion (2004); World Bank (2004k). (c) The International Bank for Reconstruction and Development / The World Bank 32 WORLD DEVELOPMENT REPORT 2005 With income distribution relatively sta- with less social cohesion, less secure property ble, growth is often said to be good for the rights, and greater risk of signi�cant political poor because the share of income going to upheaval.39 So inequality can have important the poor rises in tandem with average implications for the likelihood and nature of incomes.37 But there is evidence that the investment climate improvements, the credi- level of inequality in a society affects the way bility of policy changes, and thus the impact growth translates into rising incomes for the on decisions of �rms. This reinforces the poor (box 1.11). It is not just that poor peo- importance of governments being sensitive ple’s share of income is relatively smaller in a to the distribution of gains from growth. more unequal society—it also rises by less than one-for-one with average incomes.38 The investment climate and the lives Inequality can be of concern for other of poor people reasons too. Greater inequality is associated Governments committed to attacking poverty aggressively need to look beyond aggregate numbers and understand how investment BOX 1.11 How growth translates to rising incomes for climate improvements can enhance the lives poor people of poor people directly. In this context it is useful to distinguish the impacts on poor The extent of inequality in a society affects still worse off in an absolute sense; in Ghana, people in their various capacities: as how much average growth is shared by declining inequality and growth combined to poor people. Concerns about whether boost the incomes for poor people. employees; as entrepreneurs; as consumers; growth is “pro-poor� raise a debate about Brazil and Ghana had roughly equal as users of infrastructure, �nance, and whether absolute or relative rates of growth overall growth rates. But the incomes of property; and as potential recipients of tax- for poor people are what matters.The �gure poor people grew by 1.8 percent in Ghana compares overall growth rates and the and only 0.7 percent in Brazil. Indeed, the funded transfers or services. growth rates for the poor to illustrate these growth rate for the poor in Ghana is even points. slightly higher than that in Bangladesh, As employees. Studies looking at house- Clearly, there is a strong relationship whose overall growth rate was triple that of holds that have escaped poverty �nd that in between rising average incomes and the Ghana. incomes of poor people as illustrated by a Using the relative de�nition of pro-poor more than 80 percent of cases the decisive selection of country experiences spanning growth, Ghana’s performance is better than factor was the head of household’s getting a the mid-1980s to the end of the 1990s (see India’s—even though in India the absolute new job.40 The World Bank’s “Voices of the the �gure). But countries above the 45 income of the poor grew by 3.9 percent, Poor� study of more than 60,000 poor men degree line in the �gure are the ones where twice that of Ghana.While Ghana’s inequality the growth in incomes of the poor is higher was falling, its slower overall growth trans- and women in 60 countries identi�ed get- than average: in those cases, growth lated into less poverty reduction than in ting a job and self-employment as the best resulted not only in stronger absolute India, with its slight increase in inequality but way to escape poverty (�gure 1.15). growth in the incomes of poor people, but faster growth. Private enterprise is the engine for sus- also stronger growth relative to the Aggregate numbers of this kind mask average. Under the relative de�nition of changes in which households are poor. tainable job creation and the dominant pro-poor growth, inequality must fall.The Income mobility can be considerable. In source of jobs worldwide. In 2003 the pri- absolute de�nition, by contrast, looks only Indonesia, among those who were in the vate sector employed more than 90 percent at the income growth of poor people, poorest quintile in 1993, 59 percent moved of people in developing countries and 95 whether inequality changes or not. up at least one quintile in the income distri- Inequality declined in both Ghana and bution by 1997, with 4 percent moving all percent of people in countries such as El Zambia: in Zambia the poor suffered smaller the way to the richest quintile. In South Salvador, India, and Mexico.41 Growing declines in income than the average, but were Africa, 62 percent of those who were in the economies create more jobs, particularly in poorest quintile in 1993 had moved up at least one quintile by 1998, with developing countries (�gure 1.16). The 8 Decreasing impact of investment climate improve- Growth rate for the poor (percent) 10 percent making it to the richest inequality quintile. In Russia, 60 percent of the ments on employment growth can also be 6 Chile poorest quintile households moved seen by looking at experiences in individual up between 1995 and 1998, with 9 4 India percent attaining the highest income countries. For example, investment climate quintile. In Peru 55 percent moved up, improvements in China, India, and Uganda Ghana 2 with 5 percent making it to the top contributed to employment growth of Bangladesh quintile between 1991 and 2000. Brazil Increasing more than 2 percent a year between 1985 0 inequality and 2000. The garment sector in Cambodia Zambia –2 Source: United Kingdom—DFID (2004); also illustrates the potential impact of a Pritchett (2003); Graham and Pettinato –4 –2 0 2 4 6 8 (2001); Fields and Pfeffermann (2003); thriving private sector: exports grew from Overall growth rate (percent) and López (2003). $20 million in 1995 to more than $1 billion in 2002, employing an additional 200,000 (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 33 workers, many of them women and many Figure 1.15 Self-employment and wage income are the ways out of poverty previously poor.42 Self-employment or business A vibrant private sector also contributes to Income from wages or salaries higher wages. More productive �rms, nur- Female Benefit of family and kin Male tured by a good investment climate, can pay higher wages and invest more in training Income from agriculture, livestock, or fishing their workers.43 The expansion of �rms can Access to agricultural land also have knock-on effects, raising the wages Skill acquisition of those in smaller �rms as the pool of avail- Hard work, perseverance, or thrift able workers tightens. Similar patterns are Access to credit found in rural areas, with rising nonfarm Education employment lifting agricultural wages—with Migration signi�cant impacts on poverty reduction.44 Savings Improving the investment climate does Other more than create jobs and improve living 0 10 20 30 40 50 60 70 standards today. It also encourages people to Percent invest more in their own education and Note: Reports views of 60,000 poor people on how they saw their best prospects of escaping poverty. skills to take advantage of better jobs in the Source: Narayan and others (2000). future. There is thus a two-way link between skills and jobs, with an improved investment climate complementing efforts to improve human development (chapter 7). their opportunities and reducing the costs of Demographic trends underline the inputs they transform. The way microentre- imperative to create more and better jobs in preneurs have bene�ted from telecommuni- developing countries. Nearly 3 billion peo- cations liberalization in Bangladesh and ple are under the age of 25 today, 1.5 billion Uganda shows how (chapter 6). under 15. In the next 30 years the popula- tion in developing countries is expected to As consumers. Improving the investment increase by nearly 2 billion people, and 7 climate reduces the costs of producing and out of 8 billion of the world’s people will distributing goods, and stronger competi- live in developing countries. The popula- tion helps to ensure these bene�ts flow on tion of Sub-Saharan Africa, the region with to consumers. Poor people bene�t from the most poor people, will double by that lower prices for the goods they consume, time, even with today’s incidence of including staples. HIV/AIDS.45 In Vietnam, where up to 80 percent of the poor’s caloric intake comes from rice, lifting As entrepreneurs. Hundreds of millions of poor people in developing countries make their living as microentrepreneurs—as Figure 1.16 Growing economies generate more jobs—particularly in developing countries farmers, street vendors, and homeworkers, and in a range of other occupations, a large 5 OECD share of them women (box 1.12).46 They are Non-OECD a big part of the informal economy, which is 4 Growth in employment substantial in many developing countries (�gure 1.17).47 3 Individual entrepreneurs and microenter- prises can bene�t from the same measures 2 that improve the opportunities and incen- tives for larger �rms. They bene�t from lower 1 costs of doing business (including less red 0 tape and corruption), and from lower risks 0 2 4 6 8 (including more secure property rights and GDP per capita growth less policy uncertainty). Reducing barriers to Note: Data are average annualized rates for 1960–2000. competition also bene�ts them by expanding Source: World Bank (2004k). (c) The International Bank for Reconstruction and Development / The World Bank 34 WORLD DEVELOPMENT REPORT 2005 rights can deliver broad bene�ts across the BOX 1.12 Women and the investment climate community: During the 1990s, women’s share of the for- estimated that higher female participation • Improving access to electricity helps mal labor force increased almost rates in the Middle East and North Africa in �rms—but it also reduces the burden on everywhere—to close to 40 percent world- the 1990s increased per capita GDP growth wide and to at least a third in all regions rates by 0.7 percentage points. women collecting �rewood, reduces the except the Middle East and North Africa and Investment climate improvements can health concerns associated with burning Europe and Central Asia. It has been deliver many tangible bene�ts for women. dung, and helps children study at night. estimated that women own between a In Burkina Faso, where women have more In the Philippines members of electri�ed quarter and a third of �rms. Women run secure land rights than in many other many multimillion dollar �rms that employ African countries, female farmers’ productiv- households attain about two years more thousands of staff. ity is signi�cantly higher. Providing secure formal education than do members of But women predominate in the infor- rights to land in Peru allowed more women unelectri�ed households. That translates mal economy, particularly in microenter- to work outside the home. Removing barri- into higher wage earnings of between prises or as homeworkers. Some of this ers to competition expands opportunities reflects discrimination and the dif�culties for women and other groups that have tra- $37 and $47 a month for households women can face obtaining a formal job. ditionally suffered from discrimination. A with electricity.52 But it can also reflect how children and more competitive economy can also reduce • Improving roads helps �rms get their other family obligations can make the flex- discrimination in the workplace by increas- ibility of jobs in the informal economy ing the costs to �rms of discriminating on goods to market—but it also helps poor more attractive. More than 95 percent of noneconomic grounds. people obtain access to health, educa- the female nonagricultural labor force tion, and other services, and connects work in the informal sector in Benin, Chad, Source: Black (1999); Ellis (2003); Field (2002); Grameen Bank website: www.grameen- them to other communities (chapter 6). and Mali—and more than 80 percent in Guinea, Kenya, India, and Indonesia. info.org; Kabeer (2003); Klasen (1999); Klasen In Morocco the construction of rural and Lamanna (2003); Maloney (2004); Narayan Higher female labor force participation and others (2000); Rama (2002); United Nations roads was associated with an increase in tends to result in signi�cantly faster growth (2000); World Bank (2001g); and World Bank primary school enrollment from 28 per- in incomes. For example, it has been (2004f ). cent to 68 percent (see box 6.14). • Improving the functioning of �nance markets helps �rms—but it also allows fees, taxes, registration permits, and police poor people to weather family emergen- checkpoints on internal trade lowered the cies, improve their homes, and educate Figure 1.17 The informal economy price of rice considerably.48 Studies in their children (chapter 6). is substantial in many developing countries Ethiopia, Ghana, Kenya, Mali, South Africa, • Improving security of land rights helps and Zambia found that liberalizing food �rms—but it also empowers people and Georgia markets lowered the real prices of food, delivers more tangible bene�ts. In Peru Peru with bene�ts extending to the poor in both granting land titles to city slum dwellers urban and rural areas in these countries.49 boosted labor participation rates outside Tanzania Reducing restrictions on secondhand cloth- the home. No longer needing to have Nigeria ing markets, which account for 80 percent someone stay to guard the home gave of garment purchases in countries such as family members additional choices. Bet- Thailand Uganda, can also broaden access to afford- ter security of title also increased invest- Russia able clothing for poorer members of soci- ments in improving housing quality by Sri Lanka ety.50 While food and clothing represent the 17 percent.53 vast majority of poor people’s expenditures, Morocco the phenomenon is applicable more As potential recipients of tax-funded ser- Mexico broadly. Lowering barriers to entry by 10 vices or transfers. Attacking poverty percent has been estimated to reduce the involves more than just improving the 0 25 50 75 average price markup by 5.8 percent.51 investment climate. It also involves efforts Informal output as percent of GDP to invest in and empower people, including As users of infrastructure, �nance, and prop- public investment in education, health, and Source: Schneider (2002). erty. Many features of a better investment cli- other services. But these services need to be mate raise the living standards of people paid for, and the expansion in economic directly, whether they work or engage in activity from a better investment climate entrepreneurial activities or not. Lowering permits increases in the tax revenues to consumer prices is one example. But improv- fund those services and make transfers to ing infrastructure, �nance, and property the disadvantaged in society. About 80 per- (c) The International Bank for Reconstruction and Development / The World Bank The investment climate, growth, and poverty 35 cent of taxes in developing countries are Figure 1.18 Rising GDP is associated with rising tax revenues— expanding the opportunities to fund services for the poor collected from �rms as value added taxes, corporate taxes, and labor taxes.54 There is a GDP Total tax collection close relationship between per capita China growth and tax revenues (�gure 1.18). Uganda Of course, there are tradeoffs between raising tax revenues and providing incen- Malaysia tives for �rms to invest, create jobs, and expand. Widening the tax base, rather than Rep. of Korea increasing rates, minimizes the tradeoffs Chile (chapter 5). The extent to which the public spending from a stronger tax base is Vietnam directed to services for the poor will depend India on the government and its ability to spend resources wisely.55 But economic growth Latvia remains the only way to sustainably Brazil increase the public resources to fund such services and transfers. Morocco 0 5 10 15 20 Can investment climate improvements Annualized rate of growth, 1960–2000 be made more pro-poor? Source: World Bank (2004k). Improving the investment climate promises huge bene�ts for a society, including the poor. But can governments fashion their investment climate improvements in ways The two approaches can also be combined that deliver even deeper reductions in by focusing on particular activities in par- poverty? Much depends on the part of the ticular locations. While the choice of strat- investment climate that is improved. Some egy can vary from country to country, the improvements—such as improving macro- key point is that pro-poor approaches need economic stability, reducing corruption, not focus exclusively on addressing the and dismantling distortionary barriers to needs of the smallest �rms—they can growth—deliver broad bene�ts across soci- encompass a much broader set of �rms. ety. Other measures are more focused— such as addressing regulatory constraints Creating a better investment affecting particular activities or improving climate for everyone infrastructure in particular locations. In the This chapter showed how investment cli- latter case governments can influence the mate improvements are the driving force distribution of bene�ts. for growth and poverty reduction. A good As discussed in chapter 3, there are sev- investment climate is one that is better for eral options for making investment climate everyone in two dimensions. It bene�ts improvements more pro-poor. One society as a whole, not just �rms. And it approach is to focus on improving the expands opportunities for all �rms, not just investment climate where poor people live, large or influential �rms. which can deliver bene�ts to poor people in The rest of the Report looks at how gov- that location in all of the capacities dis- ernments can create a better investment cli- cussed above. A second approach is to focus mate. The next chapter begins by looking at on removing constraints to activities that the important question of why progress in poor people bene�t from—including as making investment climate improvements employees, entrepreneurs, or consumers. is often slow and dif�cult. (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 2 An investment climate that enhances the taining detailed recommendations on how opportunities and incentives for �rms of all policies might be improved. chapter types to invest productively, create jobs, and expand is the key to unleashing growth and Slow progress in improving the invest- ment climate is better explained by the chal- reducing poverty. That was the message of lenges that arise when governments deal chapter 1—a message now understood by with a basic tension. Firms are the primary more governments around the world. But if creators of wealth, and a good investment a sound investment climate is so bene�cial, climate must respond to their needs. But a and understood to be so by governments, sound investment climate serves society as a why are there such large variations in invest- whole, not just �rms, and the preferences of ment climates across and within countries? the two can diverge. There can also be dif- Why is progress often slow and dif�cult? ferences in the policy preferences and prior- The government’s role in shaping the ities between and even within �rms. investment climate is traditionally explained Responding to the resulting tension creates by market failures—or the failure of laissez- four practical challenges, and the way gov- faire conditions to achieve ef�cient social ernments respond to those challenges has a outcomes. This is the textbook rationale for big impact on investment climates and thus most government interventions in the on growth and poverty: economy—to provide public goods such as law and order, to support the provision of • Restraining rent-seeking. Investment cli- infrastructure, and to regulate �rms and mate policies are an enticing target for transactions to address information asym- rent-seeking by �rms, of�cials, and other metries, externalities, and monopoly power. interest groups. Corruption can increase But governments often fail to mitigate mar- the costs of doing business—and when it ket failures—and too often intervene in extends to higher echelons of govern- ways that make matters even worse. Why? ment can lead to deep distortions in poli- Clearly, failure to create a sound invest- cies. Capture, patronage, and clientelism ment climate is not merely due to lack of can also create large distortions, tilting money. Many investment climate improve- policies toward some groups at the ments place few demands on government expense of others. budgets, and the growth unleashed by • Establishing credibility. Uncertainty about reforms contributes to greater tax revenues. the future affects whether and how �rms Indeed, considerable oil and mineral wealth choose to invest. Governments need to is often associated with a worse rather than provide clear rules of the game, but a better investment climate. Nor are poor approaches that lack credibility will fail to investment climates simply a result of a lack elicit the intended investment response, of technical expertise. While the design of no matter how well crafted the rule or some reforms can require the expertise of how sincere the policy pronouncement. specialists, administering the resulting poli- • Fostering public trust and legitimacy. Firms cies typically demands far less. And the and governments do not interact in a vac- bookshelves of ministries in most develop- uum. Trust between market participants ing countries are lined with reports con- nurtures productive exchange and reduces 36 (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 37 the burden on regulation and contract enforcement. Public trust and con�dence BOX 2.1 Governance and the investment climate in markets and �rms affect not only the The opportunities and incentives that feasibility of reforms, but, through their �rms face to invest productively, create impact on policy sustainability, and hence jobs, and expand are shaped by the credibility, also influence the response of costs, risks, and barriers to competition associated with particular investment �rms. opportunities (chapter 1). Governments • Ensuring policy responses reflect a good influence those factors through a com- institutional �t. The design of invest- bination of their formal policies in par- ticular areas—stability and security, reg- ment climate policies needs to take into ulation and taxation, �nance and account sources of government failure infrastructure, and workers and labor and differences in local conditions. Inad- markets—and broader governance fea- equate consideration of questions of tures.The latter include control of rent- seeking, credibility, public trust and institutional �t can lead to poor or even legitimacy, and institutional �t. perverse results. Formal policies and broader gover- nance features interact to shape the These challenges cut across all areas of investment climate experienced by investment climate policymaking, from �rms (see �gure). Poor control of rent- contract enforcement and business regula- seeking can influence both the content and the implementation of formal poli- tion to infrastructure provision and labor cies. Weak credibility can undermine markets, and directly impact on the costs, the impact of any formal policy. risks, and barriers to competition faced by Concerns about public trust and legiti- �rms (box 2.1). This chapter looks at the macy can impede the implementation of reforms and undermine the sustain- implications for creating a better invest- ability (and hence credibility) of ment climate and practical strategies for policies. Policy interventions that are moving forward. The main message: not well adapted to local conditions Improvements are certainly possible. But can also have poor or even perverse results.Tackling these four broader accelerating and broadening progress sources of policy failure is fundamental requires governments to go beyond formal to efforts to create a better investment policies and tackle deeper sources of policy climate. failure. The basic tension: Firm preferences or the So creating a favorable investment climate public interest? must begin with understanding the perspec- A half-century ago Charles “Engine Charlie� tives and preferences of �rms. Firms exist to Wilson was famously misquoted as claim- make pro�ts for their owners—something ing, “What’s good for General Motors is they’ve done for thousands of years (box good for the country.�1 Wilson may have 2.2)—and their policy preferences are provided grist for a commonly held view of guided by that objective. In contrast, gov- the �rm ever since: as an entity that con- ernment policies need to balance the prefer- flates the public interest with its own, and ences of �rms with broader social objectives. only looks at the public interest—if at all— Governments thus have to understand through a narrow, self-serving lens. It may where the interests of �rms may diverge be a caricature, but it also highlights the fun- from those of the wider society, and must damental tension that governments must deal with the implications of differences in confront in creating a better investment cli- preferences between and within �rms. mate. Stable macroeconomic policy, secure Firms are the generators of wealth and property rights, reliable infrastructure, and jobs in society, and an investment climate ef�cient �nancial markets bene�t �rms and that is hostile to �rms cannot expect to pro- society. But there is potential for great diver- mote economic growth or reduce poverty. gence in some areas. Obviously, most �rms (c) The International Bank for Reconstruction and Development / The World Bank 38 WORLD DEVELOPMENT REPORT 2005 pressure to innovate and perform ef�- BOX 2.2 Firms in history ciently—whatever the consequences for con- sumers and broader society. Similar ten- Since ancient times people have been striv- sequent disenchantment led to a strong ing to increase their opportunities by mov- renaissance of private enterprise.Today the sions can arise in most areas of investment ing from subsistence to exchange and private sector accounts for the bulk of climate policy. investment. As far back as 3000 BCE, busi- investment and the overwhelming majority This is not to suggest that �rms are rogues ness arrangements in Mesopotamia went of jobs in developing countries. or bandits. Most individuals would also pre- beyond simple barter. Sumerian families Private trade and investment are not who traded along the Euphrates and Tigris only ancient—they are extremely hard to fer to pay less in taxes and welcome subsi- rivers developed contracts that tried to suppress. Some private investment contin- dized loans. Many �rms also voluntarily rationalize property ownership. A thousand ues even in Somalia’s war zones, and there is accept obligations well beyond those required years later the Assyrians developed an early recent acknowledgment of private enter- version of a venture capital fund. prise even in North Korea. In the meantime by law, whether through a sense of philan- Early predecessors of companies private activities are becoming more global: thropy, as a form of brand differentiation, to appeared in Rome by the second Punic War Trade as a share of global GDP rose from 25 protect their reputation, or to earn the sup- (218–202 BCE). For much of the Middle Ages percent in 1960 to 57 percent in 2001, and port of their workers and surrounding com- guilds were the most important form of world flows of foreign direct investment business organization. In the 16th and 17th reached $1.4 trillion in 2000. munities (box 2.3). International economic centuries governments and merchants integration is increasing pressures on �rms to Source: Micklethwait and Wooldridge (2003); IMF combined to create chartered companies to (2004);, Bates (2001); Bernstein (1996);Yergin and build and maintain good reputations, but it is exploit the riches of the New World. While Stanislaw (2002);World Bank (1996b); McMillan not a new phenomenon: even the infamous the mid-20th century saw widespread (2002); The Economist (2003a); Chinoy (1998); experiments with public enterprise, the sub- World Bank (2004k); and UNCTAD (2003i). United Fruit Company provided its workers in Guatemala with schools and hospitals.2 Nor are there always tradeoffs between the would prefer to pay less in taxes—including preferences of �rms and other social goals, taxes required to sustain the public services even in matters of regulation and taxation. they bene�t from and to fund other social Improving the design and administration of objectives. Many �rms would prefer to com- regulatory or tax systems can reduce the bur- ply with fewer regulations—including those dens on �rms, but can also contribute to bet- to safeguard the environment and promote ter regulatory compliance and higher tax rev- other important social interests. Most �rms enues. When regulatory regimes have not would also welcome access to subsidized been reviewed in decades, are only partially credit—whatever the policy justi�cation or enforced, and are used more to extract bribes implications for �nancial sector develop- than to protect broader social interests—all ment. And most �rms would welcome too common in many countries—the oppor- monopolies or other restrictions on competi- tunities for solutions that bene�t both �rms tion to increase their pro�ts and reduce the and broader society can be huge (chapter 5). BOX 2.3 Firms and social responsibility The debate on �rms’ responsibility to social con- ulatory requirements—one study shows that to improve working conditions in their plants in cerns has a long history. Part of it stems from dif- af�liates of U.S. multinationals pay a wage pre- Asia, following criticisms and protests from civil ferent conceptions of the objectives of �rms.The mium of 40 percent in high-income countries society. More �rms are also adopting codes of Anglo-American model focuses primarily on and 100 to 200 percent of the local average conduct on matters of corporate social responsi- maximizing shareholder value, though corporate wage in low-income countries. bility, often based on international norms pro- philanthropy has long been important. European It can be hard to distinguish the motives for moted by civil society groups or international and Japanese models put more weight on other these behaviors.At one level it might be perceived agencies (chapter 9). For example, about 20 stakeholders, especially workers.While there has to be in the best interests of the �rm, taking a banks worldwide have adopted the Equator been some convergence between models, there broad view of reputation and risk.Firms may do it Principles, a voluntary set of guidelines for man- are still debates about the extent to which �rms to protect their interests in a healthy workforce, as aging social and environmental issues related to can—or should—worry about matters other with �rms in Africa that are providing HIV/AIDS �nancing development projects, based on the than wealth creation. drugs to their workers.Others may consider it part policies and guidelines of the World Bank and Social obligations are imposed on �rms of a brand differentiation strategy, as with dolphin- International Finance Corporation. through taxation and regulation. Some �rms free tuna, no animal testing for The Body Shop, or voluntarily accept broader obligations. For socially conscious mutual funds. Source: Graham (2000); The Economist (1999, 2002a); example, multinational �rms operating in devel- Still other �rms are responding to concerns and the Equator Principles Web site (www.equator- oping countries often exceed minimal local reg- about reputation. Nike and Disney have worked principles.com). (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 39 The task of balancing the preferences of �rms relying on steel as an input two-and- �rms and broader social interests is com- a-half times the bene�ts to local steel pro- plicated by differences in preferences and ducers.3 Similarly, proposals to develop a priorities between and within �rms. Firms bond market may be resisted by banks that share common perspectives on many prefer less competition in debt markets, but issues, but their interests may diverge on be welcomed by industrial �rms.4 Conflicts speci�c policy questions. This is most can also arise over the structure of taxation, apparent when considering proposals to the detailed design of particular regulatory reduce barriers to competition. Proposals regimes, or the priority given to infrastruc- to lower barriers will typically be resisted ture development in different locations. by protected �rms, but would bene�t �rms Even when engaged in the same activity in (and others) that rely on products from the the same location, �rms of different types protected sector as inputs. For example, it can face different constraints, leading to has been estimated that restrictions on steel different policy preferences and priorities imports into the United States in 2002 cost (box 2.4). BOX 2.4 How do �rm differences affect their policy preferences and priorities? Investment climate policymaking is complicated options for dispute resolution. Foreign �rms tend to pay higher interest rates—survey data by differences in the preferences and priorities also often place more priority on show that small �rms are 50 percent more likely of �rms.Those differences can be seen along infrastructure—in part reflecting more sophisti- to see this as a major or severe constraint. Larger multiple dimensions: the extent to which the cated production methods and a greater �rms are more likely to have a bank loan, reflect- �rm’s activity is labor- or capital-intensive; the propensity to export. ing the advantages of having a track record and extent to which the �rm serves local or export Large and small �rms. Fixed costs tend to holding more assets that can be pledged as col- markets, or is otherwise exposed to impose a disproportionate burden on smaller lateral. So improving the operation of �nancial international competition; the �rm’s speci�c �rms, as with license or permit fees and even markets will often be a higher priority for small location within a country; and a range of other bribes. Evidence from the Investment Climate �rms. factors particular to each industry or �rm. Prefer- Surveys indicates that bribe payments as a share Formal and informal �rms. Informal activities ences and priorities can also differ along four of sales are 50 percent larger for small �rms. account for more than half of economic activity broader dimensions. Large �rms may make higher payments, but the in many developing countries. Although �rms in Foreign and local �rms. Foreign �rms still burden on them may be smaller. When unreli- the informal economy operate free of many tax face many regulatory barriers intended to pro- able power supply requires �rms to have their and regulatory requirements, they have less tect local �rms, and foreign �rms may be more own generators, this cost can also be greater for secure property rights and more dif�culty get- vulnerable to expropriation. Foreign �rms tend smaller �rms.This means that smaller �rms ting public services and obtaining �nancing at to be less constrained in their access to �nanc- stand to bene�t more from broadly based reasonable cost (see �gure). In Peru the nominal ing than local �rms, may be able to relocate investment climate improvements than larger borrowing rate for informal �rms was found to more easily in response to adverse changes in �rms. Smaller �rms also tend to have greater dif- be more than four times that of formal �rms of the investment climate, and may have more �culty getting �nance than larger �rms and similar size. Noncompliance with taxes and reg- ulations can also make them easy targets for Formal and informal �rms have different perspectives bribes or bureaucratic harassment. Rural and urban �rms. Remoteness and lower Informal Formal population densities increase the cost of provid- ing infrastructure and other public services in Policy uncertainty rural areas. Access to �nance is also often more of a constraint. Informal �rms in rural areas can Cost & access to finance face even more constraints than their peers in urban areas. For example, in Cambodia informal Electricity rural �rms reported greater concerns about infrastructure and �nance than informal urban Access to land �rms.They also had greater concerns about cor- ruption, crime, and policy uncertainty. Transportation 0 10 20 30 40 50 Source: World Bank Investment Climate Surveys; WDR Surveys of Micro and Informal Firms; Hallward- Percent of firms reporting constraints Driemeier and Stone (2004); Hallward-Driemeier Note: Share of �rms reporting issue as a major or severe constraint. and Stewart (2004); Schneider (2002); and de Soto Source: World Bank Investment Climate Surveys, WDR Surveys of Micro and Informal Firms. (2000). (c) The International Bank for Reconstruction and Development / The World Bank 40 WORLD DEVELOPMENT REPORT 2005 Within �rms, owners, managers, and • Restraining rent-seeking employees share some common interests • Establishing credibility but conflict on others. Recent scandals • Fostering public trust and legitimacy involving Enron and Parmalat highlight the • Ensuring that policy responses reflect a potential for conflicts between the interests good institutional �t. of management and other shareholders (chapter 6). There are also tensions between Restraining rent-seeking owners and workers over wages, bene�ts, When asked why he robbed banks, Willie and employment protection. For owners, Sutton was reported to have replied, “That’s lower labor costs and greater flexibility in where the money is.�6 In a similar way, hiring and �ring workers have many bene- investment climate policymaking can act as �ts. Workers, of course, prefer higher wages a magnet for rent-seeking by �rms, of�cials, and more job protection. While regulations and other interests. that make it harder to �re workers are often Firms, of�cials, and other groups have seen as favoring workers over employers, incentives to manipulate the design or the cost of meeting those regulations is implementation of investment climate poli- often passed on to existing workers cies to advance their private interests. Cor- (through lower wages) and to the unem- ruption and outright predation are the ployed. Some workers may bene�t, but most glaring examples, but rent-seeking can there are often subgroups with different also include more subtle forms that do not interests (chapter 7). involve the breaking of laws or the exchange These differences mean that there is no of cash. Capture and patron-clientelism can single vision of an ideal investment climate. also undermine the development of a Governments need to arbitrate between rival sound investment climate. claims. Like other interest groups, �rms are not passive in this process and are often pre- Corruption and predation pared to devote resources to obtain favorable Corruption—the exploitation of public policy treatment. Lobbying is an ancient art, of�ce for private gain—can harm the invest- and regulated �rms have a long history of ment climate in several ways.7 When it infects trying to win favorable treatment from their the highest levels of government, it can dis- regulators.5 tort policymaking on a grand scale and Managing the tension that can arise undermine the credibility of government. between �rm preferences and broader social Even when played out through of�cials at interests gives rise to four practical challenges lower echelons of government, corruption for investment climate improvement: can be a tax on entrepreneurial activity, divert resources from the public coffers, and create a Table 2.1 Bribes vary by �rm size, sector, and region constituency for erecting or maintaining Firms Bribes as unnecessary red tape. The Investment Cli- reporting bribes share of sales mate Surveys show that the majority of �rms % % in developing countries expect to pay bribes. Formal sector �rms 55.5 3.9 They also show how corruption can vary by Micro (<10 employees) 49.9 4.4 �rm size and by region (table 2.1), and how Small (10–19) 56.7 4.8 Medium (20–49) 57.6 4.0 the main locus of bribe-taking can vary Large (50–249) 58.5 3.4 between countries (�gure 2.1). Very large (250+) 55.7 3.0 Corruption manifests itself as a public Informal sector �rms 27.4 8.6 Small (<10 employees) 25.5 8.5 sector phenomenon. Typically, �rms, con- Large (10+) 49.1 9.3 sumers, or other groups make payments to Central and Eastern Europe 43.1 2.8 politicians or public of�cials in return for Sub-Saharan Africa 50.0 5.2 favorable decisions—whether a high-level Commonwealth of Independent States 51.0 3.4 East Asia and Paci�c 59.1 4.2 policy decision or a more mundane matter, Latin America and the Caribbean 68.8 7.0 such as getting a connection to utilities, clear- South Asia 74.2 3.2 ing goods through customs, or registering a Source: World Bank Investment Climate Surveys, and WDR Surveys of Micro and Informal Firms. business. Unlike most production, corrup- (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 41 Figure 2.1 The main locus of bribe-taking can vary 3 BOX 2.5 The predation of Gécamines in Mobutu’s Zaïre Utilities At independence the Democratic Republic of source of ready cash. Supported by a coterie 6 Inspections Congo’s main asset was nothing less than a of foreign bankers, he used diverse schemes horn of plenty—a 300-kilometer-long, 70- to strip the company, ranging from diverting Licenses kilometer-wide mining complex (Union foreign exchange receipts to presidential 5 Other Minière du Haut Katanga), renamed accounts, to forward selling of minerals with Gécamines after its nationalization in 1966. the proceeds going to the presidency. Not all Percent of sales 4 The Belgians had left behind a supporting the proceeds went solely to the president’s network of re�neries, hydroelectric installa- personal account. Gécamines also tions, employee housing, schools, and hospi- guaranteed state debts and covered personal 3 tals.The company provided 70 percent of the expenses of top executives and their families. country’s export receipts. According to one outside audit, of�cials were 2 The war in Katanga (formerly Shaba) stealing around $240 million a year, often province contributed to an initial collapse of listed in corporate reports under the category output, but by the late 1960s Gécamines had redressment exceptionnel dé�citaire—�excep- 1 recovered. So important was the mine to the tional de�cit recovery.� nation’s economy that then-President These practices starved the company of Mobutu had a power line connected from any earnings,led to the deterioration of its 0 the mine to electricity generators 1,800 kilo- �xed assets,and when copper prices collapsed Tanzania Pakistan China meters to the north in Kinshasa as a way of in 1974,sped the company’s demise.By 1990 Note: Countries selected to illustrate variations. “Other� includes forever tying the mines to the capital.The Zairean copper cost twice as much to produce construction permits and government contracts. Inga-Shaba line bypassed thousands of elec- as its foreign equivalent.In 1994 production Source: World Bank Investment Climate Surveys. tricity-starved villages, as well as local dams dropped to 30,600 metric tons of copper and that might have supplied power to the mine 3,000 metric tons of cobalt a year,with zero more easily. revenues.According to some estimates,in tion is subject to increasing returns: an In the early 1970s the complex was pro- order to restore annual production to 300,000 increase in rent-seeking activity may make ducing between 400,000 and 700,000 metric metric tons a new investor would need to tons of copper and between 10,000 and inject around $3 billion,including $2 billion corruption more attractive, not less.8 So high 18,000 metric tons of cobalt a year, securing just to absorb the company’s debts. levels of corruption can be sustainable, and annual revenues between $700 million and divert energy from more productive activity. $900 million. For Mobutu, Gécamines was a Source: Wrong (2001). No country can claim to be immune from the problem. In the extreme, a “predatory� state consumes the surpluses of the economy, as government of�ces come to be treated as BOX 2.6 Natural resource endowments: Blessing or curse? income-generating property (box 2.5). In principle an abundance of natural rely heavily on revenues from such Rent-seeking behavior can be especially resources such as minerals or oil should be resources there are also weak incentives to pronounced in countries that have a high- a valuable asset in creating a modern, pros- develop a broad tax base or consistent and perous economy. Certainly many of today’s non-arbitrary tax policies. Far from being a level of dependence on exports of minerals, successful economies have been able to bene�t to the state, relief from needing oil, or other natural resources. While many leverage these assets to their advantage. effective local tax laws and administration of today’s successful economies—including But in many developing countries substan- can lead to unaccountable, inef�cient, and Australia, Chile, and Norway—prospered in tial endowments of natural resources often uninformed government. seem more like a curse than a blessing. How have some countries been able to part through natural resource endowments, A wealth of natural resources can have capitalize on resource endowments without dependence on natural resources has been several adverse consequences.When the succumbing to the resource curse? Historical more of a curse than a blessing for many discovery of natural resources attracts signif- and contemporary evidence suggests sev- developing countries (box 2.6). icant capital inflows, the value of the eral possibilities. It helps if natural resources national currency can appreciate, making do not dominate the local economy, and if Corruption can be traced to a combina- non-resource exports less competitive—the resource extraction is not dominated by tion of three basic factors: monopoly so-called “Dutch disease.� Heavy reliance on monopolies. It also helps if governments are power, discretionary authority, and inade- resource exports can also expose an econ- held accountable for their behavior through quate accountability for the exercise of that omy to the vicissitudes of international com- political competition and an informed popu- modity price movements. But the impact on lation. Efforts to create a better investment authority. As Klitgaard put it governance can be far more harmful.The climate for �rms outside the resource sector [C]orruption is a crime of calculation, not potential to exploit natural resources can can also play an important role by helping to prompt more intense rent-seeking behavior diversify the economy and so reduce depen- passion. True, there are saints who resist all by politicians and others, diverting attention dence on natural resources. temptations, and honest of�cials who resist from more productive activities. In the Source: Stijns (2000); Tornell and Lane (1999); most. But when the size of the bribe is large, extreme, competition over access to the Levi (1988); Sachs and Warner (2001); Leite and the chance of being caught small, and the rents from natural resources can lead to, or Weidmann (1999); Ross (2001); Chaudhry penalty if caught meager, many of�cials will perpetuate, civil war.When governments (1997); and Moore (1998). succumb.9 (c) The International Bank for Reconstruction and Development / The World Bank 42 WORLD DEVELOPMENT REPORT 2005 Figure 2.2 More business start-up procedures con�rm that bribe payments are higher increase both delays and corruption when dealings with of�cials cannot be ≥18 avoided.10 Evidence suggests that countries 17 with more interventionist approaches to 16 business regulation also tend to have more 15 corruption (�gure 2.2). Procedures to start a business 14 Where intervention is justi�ed, the scope 13 12 for bureaucratic discretion can be limited by 11 reducing unnecessary ambiguity or vague- 10 ness in government policies and regulations, 9 by promptly publishing implementing regu- 8 lations, and by promoting adherence to 7 precedent by publishing administrative 6 decisions and rulings (chapter 5). 5 4 The third and complementary strategy is 3 to enhance accountability for the exercise of 2 public authority. Political competition can 100 80 60 40 20 0 20 40 60 80 100 play an important role in holding govern- Days required Corruption ments responsible for their results and for Note: 133 countries are grouped by average number of procedures their behaviors. But experience shows that needed to start a business in each country. Number of days more is required. Enhancing the trans- required to start a business and level of corruption are then aver- aged according to the number of procedures needed to start a parency of government-�rm transactions is business across those groups of countries. Corruption is a one of the most promising strategies, and weighted average of multiple indicators of corruption, taken from Kaufmann, Kraay, and Mastruzzi (2003), and normalized by sample has become an increasing focus of efforts to maximum and minimum. Source: World Bank (2004b) and Kaufmann, Kraay, and Mastruzzi address corruption worldwide. A free press (2003). also plays a critical role in monitoring gov- ernments and informing citizens, helping to Strategies for tackling corruption focus keep potential abuses in check.11 A growing on the same three points. The scope of number of countries are also creating spe- monopoly power can be reduced in several cialist bodies to investigate and prosecute ways. Competition can be facilitated wher- corruption and lead broader prevention ever possible, and government interven- strategies (box 2.7). tions that lack a compelling policy justi�ca- Developing clear standards of public tion can be eliminated. Firm surveys conduct and conflict-of-interest laws for the civil service can constrain discretion and influence social norms within an agency. BOX 2.7 Combating corruption in Botswana and Providing protections to whistleblowers can Lithuania reinforce those norms and complement In 1974 Hong Kong established a three- persons before the court and recovering other monitoring mechanisms.12 Low pronged anticorruption strategy focused on approximately $1 million in �nes, forfeitures, salaries in the civil service are often believed investigation, prevention, and education, seizures, and taxes. It has sustained an active to contribute to corruption, but the rela- implemented by the autonomous Commis- publicity campaign through seminars, tionship can be complex. Certainly civil ser- sion against Corruption. Drawing poster campaigns, displays at trade exhibi- inspiration from its success, similar tions, and cartoon strips, as part of the moral vice salaries are less likely to influence large- initiatives have been adopted in countries education of the young. scale corruption at higher echelons of as diverse as Botswana and Lithuania. Lithuania. In 1997 Lithuania established a government, which can be particularly Botswana. Following a series of high- Special Investigation Service that reports to destructive to the investment climate and to level corruption scandals, Botswana created the president and the parliament.The number a Directorate of Corruption and Economic of prosecutions for bribe-taking increased society generally. And while studies suggest Crime in 1994 with powers to investigate sevenfold between 1997 and 2002 (from 10 a that increasing salaries for lower-level of�- and prosecute suspects, prevent corruption, year to 73), and the cases of prosecution for cials might reduce the incidence of smaller- and educate the public.The directorate is an abuse of of�ce, from 2 in 1997 to 19 in 2002. autonomous agency under the Of�ce of the scale corruption, this will not always be a President. In its �rst two years of operation, it Source: Open Society Institute (2002); Fombad feasible or cost-effective strategy.13 So, while launched 828 investigations, bringing 141 (1999); and Doig and Riley (1998). improving civil service wages and condi- tions can be an important part of improv- (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 43 ing the quality and professionalism of the relationships with private interests. In soci- civil service, merely increasing salaries does eties with democratic forms of government, not substitute for broader efforts to limit elected representatives make policy in the monopoly power, curb discretion, and interests of their constituents in exchange enhance accountability. for political support. This is a necessary part of ensuring the accountability and Capture and patron-clientelism responsiveness of policymakers to their citi- Investment climate policies can be distorted zens. But representative government can by rent-seeking in forms that do not involve devolve into patron–clientelism when poli- breaking laws or direct exchanges of cash. cymakers distribute policy privileges to par- Industrial-�nancial elites, workers, con- ticular groups on the basis of ethnic or cul- sumers, and other groups influence policy- tural solidarity or political support, often at making to very different degrees in different the expense of society as a whole. The prob- settings. When one group has dispropor- lems can be even worse in dictatorships, tionate influence, the design or implementa- where leaders still need to curry favor with tion of policies can be skewed in their favor particular groups, but are subject to fewer at the expense of society as a whole in ways constraints.16 that establish long-lasting privileges for that Investment climate policymaking pre- group. There are two related phenomena: sents myriad opportunities for granting capture and patron-clientelism. bene�ts to, and redistributing resources toward, favored groups. Policies that would Capture. Firms and other groups can skew bene�t the investment climate may not be policies in their favor by formal or informal implemented because they cannot reward lobbying, controlling access to information, loyalty and strengthen ties between patrons or a variety of other strategies. It has long and clients.17 The result: property rights, tax, been recognized that regulatory agencies are and regulatory regimes are designed with vulnerable to becoming “captured� by the speci�c constituencies in mind. Govern- industries they are charged with regulating, ments suppress competition by conferring and so promote the interests of the industry monopolies, devising market restrictions, or rather than those of the broader public.14 tolerating cartels. Tax systems become rid- The concept of “state capture� has more dled with special exemptions—or are recently been used to describe how �rms enforced selectively. Financial markets are and other groups can shape the formation underdeveloped because governments help of laws and policies (as opposed to their middlemen maintain their stranglehold on implementation) through informal and the allocation of funds. Public investment in opaque channels of influence—by control- infrastructure and related tariff policies are ling the policy agenda or by changing the designed to reward favored groups.18 basic nature of representation and constitu- Patron-clientelism can be exacerbated in tional design.15 Firms or other groups most polarized and fragmented societies, where directly affected by particular laws or poli- politicians use their authority to bene�t their cies will have stronger incentives to invest in particular constituencies. Governments with influencing policy than consumers and low credibility in the eyes of the public as a other groups, and usually also face fewer whole may also resort more to clientelistic logistical dif�culties in framing a coordi- approaches to buy support from particular nated view. These groups often also have groups.19 Unequal access to information can superior access to information and technical have an even more pervasive impact on expertise than legislators, regulators, or oth- clientelism. Citizens may want leaders who ers affected by the policy decision. will implement policies that bene�t society as a whole rather than favor particular Patron-clientelism. Under conditions of groups, but they cannot always tell the differ- capture, it is usually the private interest ence—particularly when governments use group that derives bene�ts. But politicians less transparent forms of intervention (box and of�cials also have incentives to exploit 2.8). Uninformed voters are more likely to (c) The International Bank for Reconstruction and Development / The World Bank 44 WORLD DEVELOPMENT REPORT 2005 BOX 2.8 The form of intervention: How many cheers for transparency? Governments wishing to confer bene�ts on a sumers), and those costs usually far exceed the make direct transfers.This is not always the case, particular group can choose between two main bene�ts received. For example, restrictions on however, and an explicit tax rebate could often strategies.They can make an explicit budgetary the import of steel in the United States in 2002 achieve similar results. A more common expla- transfer, or they can create a market restriction were estimated to deliver bene�ts to the nation is that the less ef�cient approach is more or provide other forms of less transparent sup- protected industry of $240 million, but impose appealing politically.The transfer is not port. costs on U.S. steel-using industries of nearly transparent. It is usually not exposed to the From an economic standpoint, the �rst $600 million. Market restrictions also create same level of scrutiny as a budgeted item. And approach is more ef�cient.The costs are borne additional costs for society by dulling the incen- consumers or others are often not in a position by taxpayers in general. And when the tax sys- tives of the protected group to innovate and to evaluate the magnitude of the costs being tem is reasonably ef�cient, those costs are usu- improve productivity. imposed on them. ally of the same order of magnitude as the ben- Why do governments so frequently choose e�t. Market restrictions, by contrast, impose the the less ef�cient option? One possible explana- Source: Tullock (1983); Acemoglu, Johnson, and costs on a subcategory of society (typically con- tion is that they lack the budget resources to Robinson (2001); Hufbauer and Goodrich (2003b). support or oppose policies based on crude, ment projects and targeted tax breaks tend to visible criteria—for example, whether the proliferate as elections approach.21 economy seems to be prospering or whether There is some evidence to suggest that new highways are being built.20 In many the more widespread the direct personal countries—rich and poor—public invest- connections between owners of �rms and politicians, the poorer the quality of a coun- Figure 2.3 More influential �rms face fewer constraints try’s investment climate.22 These connec- tions can yield substantial bene�ts to �rms Most influential Least influential firms firms and politicians alike, creating incentives for Telecommunications both parties to invest in such relationships. It has been estimated that as much as a Transportation quarter of the share value of Indonesian �rms before 1998 could be attributed to Access to land dependence on the Suharto family.23 The Bank’s surveys con�rm that �rms Crime that are part of the favored circle tend to Electricity face a more attractive policy environment than other �rms (�gure 2.3). The evidence Labor regulations also suggests that more influential �rms are likely to innovate less (�gure 2.4).24 One Licenses and permits interpretation is that a more challenging environment is more conducive to innova- Legal system tion. More likely, perhaps, the favored �rms Customs and trade regulations are more concerned with maintaining their influence and enjoying the resulting bene- Access to financing �ts than focusing on improving their pro- ductivity. Corruption Every society faces the challenge of creat- Tax administration ing governance arrangements that can accommodate a spectrum of interests while Policy uncertainty preventing the formation of undue or illicit influence by any particular group to the Tax rates detriment of others. Three complementary 0 10 20 30 strategies can help: Percent of firms reporting constraint to be a “severe� or “major� obstacle • Enhancing the transparency of govern- Note: Percentages based on simulations controlling for country, �rm size, and sector. “Influence� is measured as the dif- ment-�rm relations. Regulatory arrange- ference, as perceived by �rms, between their own ability to influence national policies and legislation and the ability of ments can be designed and administered other domestic �rms to do so. Figure shows responses for those �rms regarding themselves as most and least influential. Source: Desai (2004), drawing on World Bank Investment Climate Surveys. in ways that facilitate public scrutiny, (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 45 including through use of regulatory Figure 2.4 More influential �rms innovate less impact assessments (chapter 3). The dis- 0 closure of budgetary or quasi-budgetary support provided to �rms or industries – 10 Percentage decrease in probability can be mandated. Government procure- ment practices can be made open and of undertaking activity – 20 competitive. “Sunshine laws� can require certain government decisions to be pre- Moderately – 30 influential ceded by opportunities for public com- firms ment and for public access to certain – 40 records. No less important, the disclo- sure of funding for political parties can Most – 50 be mandated.25 influential firms • Broadening policy dialogues. Investment – 60 climate policymaking affects a broad Introduce Introduce Upgrade New range of interests—not just those of new new product facility technology product lines large or influential �rms. Creating an lines investment climate that bene�ts every- Note: Percentage decrease is relative to �rms’ regarding themselves as least influen- one requires processes to ensure this tial. The �ndings are based on simulations controlling for country, �rm size, and sec- tor. “Influence� is measured as the difference, as perceived by �rms, between their fuller set of interests is heard, including own ability to influence national policies and legislation and the ability of other representatives of consumers and domestic �rms to do so. Source: Desai (2004), drawing on World Bank Investment Climate Surveys. smaller �rms. Business associations can sometimes give smaller �rms more of a voice in policymaking (box 2.9). Many the life of their proposed investment. governments are also establishing dedi- Addressing �rms’ concerns about uncer- cated consultative mechanisms to tainty, and building policy credibility, are broaden the dialogue on investment cli- fundamental to creating a better investment mate issues (chapter 3). climate. • Strengthening accountability mechanisms. Strong and competitive legislatures can The central role of uncertainty permit disenfranchised groups to chal- Uncertainty plays a central role in invest- lenge the authority and privilege of ment decisions. Because those decisions are incumbents, and make it more dif�cult forward looking, with the bulk of costs for executive branch of�cials to deliver clientelistic policies (�gure 2.5).26 Expanding legislative authority over BOX 2.9 Business associations and the investment budgetary matters and strengthening climate oversight of regulators reduces the pref- erentialism in taxation and the preva- Business associations can lower the costs of Association represents the policy concerns of lence of regulatory capture.27 A free and information and help �rms seek opportunities more than 300,000 members working in the and make transactions in new markets.They informal economy. independent media can make the public can be economywide, or “peak�associations, Experience suggests that business asso- aware of the costs of clientelistic prac- such as confederations of industry, manufac- ciations are more likely to contribute to a tices and reinforce accountability turers’associations, and entrepreneurs’associ- sound investment climate when: ations.They can also be sectoral lobbies. • They are free of state influence and not through the ballot box. In some cases business associations con- reliant on governments for resources, solidate the influence of already powerful capital, or personnel. groups.The Thai Bankers’Association,for Establishing credibility example,represents 13 banks,four of which • They are unaffected by endemic sectar- ian divisions. Firms do not make decisions based on the control more than two-thirds of Thailand’s • They have a broad constituency. banking assets.But business associations can formal content of laws, regulations, or pol- • They exercise their influence through for- also help to broaden the dialogue on invest- icy statements alone. Because investment ment climate policy issues,giving voice to mal, transparent channels. decisions are forward looking, �rms need to �rms that might not otherwise be heard.In Source: Max�eld and Schneider (1997), and assess the likelihood of those policies actu- India,for example,the Self-Employed Women’s Recanatini and Ryterman (2001). ally being implemented and sustained over (c) The International Bank for Reconstruction and Development / The World Bank 46 WORLD DEVELOPMENT REPORT 2005 Figure 2.5 Cronyism is reduced by greater Figure 2.6 Policy uncertainty dominates the accountability—and legislatures play an especially investment climate concerns of �rms important role Crime 2% Skills 2% Effective legislature Policy 2.0 Electricity 2% Partially effective uncertainty legislature Finance 4% 28% Ineffective or Corruption 10% nonexistent 1.5 legislature Crony bias Regulation 1.0 10% Tax 19% 0.5 Macroeconomic instability 23% Note: Share of countries where �rms report issue as top constraint 0.0 in surveys of 48 countries. –2 –1 0 1 2 Source: World Bank Investment Climate Surveys. Voice and accountability Notes: “Crony bias� (vertical axis) is the difference between per- uncertainty as their dominant concern ceived influence of �rms with political ties and influence of busi- among investment climate constraints (�g- ness associations, based on WEF Executive Opinion Surveys. “Voice and accountability� (horizontal axis) reflects various mech- ure 2.6). anisms to hold governments accountable, based on Kaufmann, Concerns about policy uncertainty can Kraay, and Mastruzzi (2003). Measures of legislative effectiveness are based on Banks (2001). Crony bias based on predicted values. stem from vagueness or ambiguity in cur- Source: Kaufmann (2003); Banks (2001). rent policies and laws. But no matter how well-de�ned current policies may be on borne upfront and the potential bene�ts paper, there may still be concerns about spread over time, there is always uncertainty how they will be implemented in practice or about what the bene�ts will actually be— evolve over time. The latter concerns reflect because of uncertainties about the way con- on the credibility of governments and their sumers or competitors will respond, about policies, including the ability of govern- the broader economic outlook, and about ments to deliver what is promised. how government policies may evolve. The The impact of policy uncertainty on Investment Climate Surveys show that investment decisions varies along several �rms in developing countries rate policy dimensions. The nature of the investment obviously matters. While all investments involve up-front costs, some can be reversed BOX 2.10 Reducing policy uncertainty to stimulate more easily than others. The less reversible an investment investment, and the greater the �rm’s vulner- Lowering interest rates is often proposed as diminishes.True, lower interest rates give ability to uncertain future changes, the the best way to spur investment. Interest greater weight to the future and thus the greater the value in waiting to see if the rates affect investment decisions because expected stream of bene�ts, but they also uncertainty is resolved before investing.28 For they are a measure of the opportunity cost of increase the value of waiting.The overall the resources dedicated to the project—that effect is thus weak or even ambiguous. example, �rms in Ghana and Uganda were is, the return these resources could otherwise Research �nds that reducing the sources of more likely to increase their hurdle rate of have earned.They affect the cost of borrow- uncertainty about future pro�ts—or about return as uncertainty increased, and uncer- ing by �rms and the returns that equity the likely future path of interest rates—has tainty had a more negative effect on �rms investors look for. As interest rates fall, invest- more important effects on investment than ment should rise because the expected ben- does the current level of interest rates. with less reversible investments.29 Uncer- e�ts now need to clear a lower value. Reducing unnecessary uncertainty, includ- tainty and irreversible investments imply that But many empirical studies have failed ing that about government policy, is thus reductions in uncertainty, rather than to �nd a signi�cant relationship between likely to be the better approach to stimulat- changes in interest rates, may be more effec- interest rates and investment rates. Real ing investment. options theory helps explain why. With tive in influencing investment (box 2.10). uncertainty and irreversible costs, the Source: Blanchard (1986); Caballero (1999); and Beyond issues of reversibility, some importance of interest rates in investment Dixit and Pindyck (1994). investments are more sensitive to policy changes than others. Investments in heavily (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 47 regulated sectors such as infrastructure can Figure 2.7 Policy uncertainty is a concern for informal �rms as well be especially sensitive to policy uncertainty Informal sector firms Formal sector firms because the pro�tability of the venture is often determined directly by government Tanzania regulation. For example, Hungary’s initial Cambodia attempt to involve private investment in its energy sector—before de�ning the policy Pakistan and regulatory framework—attracted few bids and the tender was aborted in 1993. Bangladesh Two years later, with a clearer regulatory Indonesia framework in place, it attracted bids of nearly $2 billion.30 Guatemala Firms also differ in their ability to cope with risks. Larger �rms will typically have Kenya more opportunities to diversify risk than smaller �rms, and multinational �rms can Brazil diversify country-speci�c risks across several 0 20 40 60 80 countries. While �rms in the informal econ- Percent of firms reporting policy uncertainty omy are usually less constrained by regula- as a “severe� or “major� obstacle tion than their counterparts in the formal Note: Based on responses in 11 countries where surveys of informal �rms were undertaken. Source: World Bank Investment Climate Survey and WDR Surveys of Micro and Informal Firms. economy—and so may be less concerned about the risk of policy changes—they usu- ing to accept more risk to avoid a loss than to ally also have fewer opportunities to diver- realize a gain of the same size. There can also sify or manage such risks. Reflecting this, the be an endowment effect—placing greater Bank’s surveys show that policy uncertainty value on something already owned just is still a signi�cant concern to �rms in the because it is owned. Anchoring can also informal economy (�gure 2.7). interfere with judgment—people place dis- Access to information influences how proportionate weight on recent experiences, �rms respond to uncertainty. Constrained particularly their own, rather than on longer access to information can lead �rms to historical trends. Conservatism can have the herd—basing decisions on how other �rms same effect—slowing the response to changes are seen to be responding. Enhancing the in trends.33 These phenomena influence the transparency of government policies has also been found to increase the level of international investment.31 BOX 2.11 Entrepreneurship and uncertainty Uncertainty, credibility, and information go a long way toward explaining some of the Entrepreneurship—or attitudes toward authors have argued that some countries in innovation, pro-activity, and risk-taking— Africa may exhibit relatively low levels of apparent mysteries of �rm behavior—what influences the way individuals and �rms entrepreneurship. Keynes referred to as “animal spirits.�32 But respond to uncertainty, including policy If this is true, and has adverse implica- �rm responses can also be conditioned by uncertainty, when assessing investment tions for investment and growth, the ques- other factors. Ultimately, the way �rms opportunities. tion is whether such attributes are deeply Despite dif�culties in measurement, it is ingrained or are responsive to government respond to uncertainty is shaped by their generally accepted that the personal char- policies that shape the investment climate. con�dence in the future, and some �rms will acteristics that make up entrepreneurship The evidence supports the second view, be more optimistic than others. Attitudes are not distributed equally in any given indicating that the incentives provided by society—some individuals and �rms are government policies and behaviors can toward risk can also vary depending on the less daunted by risk and uncertainty than have a big impact on observed levels of entrepreneurial characteristics of individuals others.There may also be differences entrepreneurship in any society. and the �rms they own and manage—and between societies. Studies exploring this possibly across societies as well (box 2.11). question often focus on the incidence of Source: Covin and Slevin (1989); Etounga- new business registration or self-employ- Manguelle (2000); Hart (2003); Hofstede (1984); Recent work in behavioral economics and ment, which may not be reliable indicators Iyigun and Rodrik (2003); Lee and Peterson psychology provides some additional when applied to developing countries with (2000); Lumpkin and Dess (1996); McGrath, MacMillan, and Scheinberg (1992); Miller insights, suggesting that people are not as signi�cant informal economies and fewer (1983); Miller and Friesen (1982); Porter (2000); rational as traditional theories assumed. For alternatives to self-employment. But several Reynolds and others (2004); and Wild (1997). example, people tend to be loss-averse—will- (c) The International Bank for Reconstruction and Development / The World Bank 48 WORLD DEVELOPMENT REPORT 2005 Figure 2.8 Firms are more likely to invest when the percent (�gure 2.9). The impact of uncer- policies are perceived to be credible tainty can increase more than proportion- 0.9 ately, so large sources of uncertainty can be especially damaging.35 Likelihood of new investment (percent) The quest for policy credibility 0.8 Improving the clarity of existing policies and regulations, and managing changes to those policies and regulations in ways that minimize unnecessary uncertainty for �rms, 0.7 are relatively straightforward (chapter 5). Addressing concerns about how policies will be implemented or will evolve over time can have an even bigger impact (box 2.12)—but 0.6 is also more challenging. The credibility of –4 –2 0 2 4 investment climate policies can be under- Credibility perceptions score mined by many factors. A recent track Note: The �gure plots �rms’ predicted probabilities that they have record of political or macroeconomic insta- increased investment in the past year against a measure of credi- bility. Credibility perceptions score is derived from principal com- bility does not help—creating a special bur- ponents analysis of �rm responses to questions of policy predictability, consistency, and enforcement, with higher scores den for governments seeking to rehabilitate meaning greater credibility. Data points represent average proba- the reputations of their countries.36 The bilities for each credibility score. Probability of new investment is based on predicted probabilities generated from a logistic regres- credibility of a government’s policies may sion controlling for �rm size, industry, and region. also be in doubt if there are questions about Source: World Bank: World Business Environment Survey data- base. its willingness or ability to enforce its stated policies, or to sustain them over time. way �rms respond to government policies, To some degree the ability of government but do not undermine the fundamental roles to achieve greater policy credibility is of uncertainty, credibility, and information. bounded by the broader polity and social Uncertainty, including that stemming consensus. Normal, constitutionally based from credibility concerns, can impact turnover in government does not preclude a Figure 2.9 Improving policy investment decisions in various ways. Firms government from making credible commit- predictability can increase the probability of new investment by over may demand higher rates of return to com- ments. Indeed, even frequent changes in gov- 30 percent pensate for the extra risk involved—result- ernment may not undermine policy credibil- 40 ing in less investment at higher prices. They ity when there is a broad consensus for a may shorten their planning horizon, thus particular policy direction. For example, Significant influencing the level and form of invest- Estonia and Latvia have each aggressively improvement 30 ment, the choice of technology, and the pursued investment climate improvements Percentage increase willingness to train workers. They may pur- since independence in the early 1990s, sue various risk management strategies, notwithstanding having each had 12 changes 20 from buying insurance to cultivating per- in governments during that time. Replacing Some sonal relations with political leaders. They policymakers can even improve credibility 10 improvement may use an initial limited investment to when the new leaders are considered more elicit more information—about the oppor- likely to honor policy commitments. But Minor tunity, or about the reliability of govern- instability manifested through frequent shifts improvement 0 ment policies—before committing to a in policy direction can demolish credibility. Changes in unpredictability of laws and regulations larger or less reversible investment.34 Or All governments face the challenge of Note: Percentage increase is relative to �rms �rms may simply refuse to invest at all. committing today to policy actions in the reporting no improvement in predictability. Sim- Firm-level surveys con�rm that �rms are future, particularly when it is understood ulations based on �rm responses to survey, with 80 countries, controlling for region, �rm size, more likely to invest when policies are that circumstances and incentives can and sector. regarded as credible (�gure 2.8). The sur- change. Some policy flexibility is essential to Source: World Bank World Business Environ- ment Survey. veys also show that improving policy pre- adjust to changing circumstances. But unre- dictability can increase the probability of strained governments too often succumb to making new investments by more than 30 the appeal of short-run political goals that (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 49 BOX 2.12 The power of credibility Policy credibility plays a powerful role in the tion. At extremely high levels the distortions are restrictive barriers (point C in the �gure) can investment climate, influencing the level of such that there are no incentives for private pull policies in their preferred direction. Pub- �rms’ response to any given set of policies. investment. lic concerns about the role of �rms or markets One can think of the main dimensions of This view of an “optimal� investment may lead to lack of public support for more the investment climate influenced by govern- climate is captured in the �gure. It shows social desirable policy approaches. Or the chosen ment policies and behaviors—costs, risks, and bene�ts of the investment climate—higher policy design may represent a poor �t with barriers to competition—as ranging from zero productivity of investment or growth—as a local conditions for other reasons. Restraining to very high levels. At zero, costs and risks are function of barriers to competition (and could rent-seeking, building consensus, and improv- minimal and �rms face no barriers to competi- be similarly applied to costs and risks). The ing institutional �t can lead to policy socially optimal position is not zero— outcomes that increase social welfare (a move A′ some barriers might be justi�ed as part from point B to point A in the �gure). of a regulatory strategy for dealing with Improving the content of policies can make A High credibility pollution or other social concerns, just a big difference. But enhancing the credibility of Social benefits as some costs may be justi�ed through those policies provides additional bene�ts by B′ taxation, or some risks (and uncertainty) increasing the level of �rms’ investment can be justi�ed to preserve a degree of responses to any given set of policies. In the �g- B C′ policy flexibility. In the �gure the status ure, enhancing credibility shifts the frontier of Low credibility C quo is to the right of the optimum, indi- the curve outward (the status quo for a more cating the presence of undesirable bar- credible government would be at B′ rather than 0 riers to competition. B). Improving both the content of investment Social Status Monopolists’ Current policies may fall short of climate policies and the credibility of those poli- optimum quo preferences their optimum for several reasons. cies (the shift from B to A′) thus results in the Barriers Rent-seeking by �rms looking for more largest gain in social welfare. leave society as a whole worse off. Examples straints are associated with lower per- abound, from printing money to �nance ceptions of investment risk (�gure 2.10). profligate public spending to reneging on • Entrusting discretion on sensitive subjects speci�c commitments to investors and cred- to more autonomous agencies. Examples itors. To address these concerns, govern- include independent central banks and ments need mechanisms to commit credibly specialist regulatory agencies for infra- to sound long-term policies.37 Just as the tri- structure—areas where the temptation umph of the English Parliament over the to renege on commitments is particu- Crown in 1689, for example, limited the larly acute (chapter 6).40 ability of the monarchy to con�scate wealth, restraining the arbitrary behavior of govern- ment is considered a watershed in the cre- Figure 2.10 The power of restraint: governments with less discretion present lower investment risk ation of modern capital markets in devel- oped and developing economies.38 10 Governments can draw on a variety of mechanisms and strategies to enhance their 8 credibility. The main formal mechanisms High political constraint involve constitutions, institutions, con- Investment profile tracts, and international agreements: 6 • Establishing effective veto points on deci- Low political constraint 4 sionmaking and providing other guaran- tees through national constitutions. This can include formal checks and balances 2 among different branches of govern- ment, autonomous subnational govern- 0 ments, and constitutional prohibitions 1984 1986 1988 1990 1992 1994 1996 1998 2000 on the expropriation of property, cou- Note: The Henisz index of political constants measures restrictions on policymakers’ ability to make discretionary policy changes. The pled with independent judiciaries able to ICRG investment pro�le is an indicator of risk to investment. enforce those rules.39 Political con- Source: Henisz (2000) and International Country Risk Guide. (c) The International Bank for Reconstruction and Development / The World Bank 50 WORLD DEVELOPMENT REPORT 2005 • Providing speci�c contractual commit- was one rationale for mass privatization ments on particularly sensitive matters. programs. In Bolivia and Chile similar While clearly not feasible for all �rms or effects were obtained by including pension topics, this is a common strategy for funds among the investors in privatized major natural resource and infrastructure utilities. Improving the ability of �rms and projects, and increasingly common on consumers to monitor and evaluate policy matters of taxation for a broader range of actions can also enhance credibility,41 and activities (chapter 5). The credibility of so can create structures to sustain an ongo- contractual commitments can be further ing process of reforms, including effective enhanced by making them subject to consultation and policy review mechanisms international arbitration (chapter 4). (chapter 3). • Entering international agreements that Establishing credibility can be particu- commit governments to sound policies. larly challenging for governments building International agreements cover a growing on a legacy of political and economic range of investment climate policy areas. instability. But Uganda’s experience in the They can enhance credibility by increas- 1990s shows how persistence can pay off ing the costs of reneging on relevant pol- (box 2.13). icy commitments, whether through rep- Firms and governments can also come to utation effects or by the threat of more other arrangements that may allow invest- tangible sanctions (chapter 9). ment to proceed but that involve longer- term costs for society. For example, in the Formal mechanisms of these kinds are aftermath of the Mexican revolution of not the whole story. For example, privatiza- 1910–20 one might have expected private tion programs in sensitive areas often allo- investment to collapse as revolutions, civil cate at least some of the shares in the priva- wars, and coups took their toll. Yet invest- tized enterprise to a wide range of local ment was not disrupted. One explanation is people to raise the political costs of a policy that revolution-era Mexican governments reversal. In the transition economies, this offered credible protection to existing investors by incorporating them into ruling coalitions.42 The phenomenon of “crony BOX 2.13 Building credibility through persistence in capitalism� in Indonesia and other coun- Uganda tries in more recent history can be Many economies in Africa have stagnated cial court was established in 1996.The explained through the same lens: forging or shrunk in recent decades, largely reflect- telecommunications sector was close ties between selected �rms and politi- ing poor investment climates.Yet Uganda modernized through competition and pri- cians allowed investment to proceed in an climbed out of civil conflict and chaos in the vate sector participation, including the pri- late 1980s and severe macroeconomic vatization of Uganda Telecom Limited in environment with few formal checks on instability in the early 1990s to more than 2002.The power sector was opened to pri- government.43 But these arrangements can double the share of private investment in vate participation, and in 2002 a 20-year ossify to the detriment of the broader GDP between 1990 and 2000, and boosted concession was awarded for the country’s investment climate—and to the detriment its per capita GDP by over 4 percent a year main generating station. Efforts are under from 1993 to 2002—or 8 times the average way to improve business regulation. of more innovative entrepreneurs, smaller in Sub-Saharan Africa. How? Each reform had some impact on the �rms, and consumers. This underscores the Beginning in 1991–92 the government opportunities and incentives for �rms. Just importance of drawing on commitment launched reforms that eventually encom- as important, the determination of policy- mechanisms that embrace broader seg- passed most aspects of the investment cli- makers to stick with reforms—including mate. Macroeconomic stability was dealing with setbacks along the way— ments of society—not merely elites or the achieved, and the independence of the cen- enhanced the credibility of the largest �rms, but smaller �rms and other tral bank was strengthened. Monopolies in government’s commitment to create a groups as well. coffee, cotton, and tea were dismantled, and more productive society. For example, the trade barriers were reduced. A new invest- privatization of Uganda Telecom succeeded ment code providing protection against only on the third attempt.The Uganda Fostering public trust expropriation was introduced, and the Commercial Bank was privatized only in and legitimacy return of property expropriated by an ear- 2002, after an earlier unsuccessful attempt. lier government was accelerated. An Governments and �rms do not interact in a autonomous tax agency was created. Public Source: Holmgren and others (2001) and World vacuum. The broader social context can enterprises were privatized. A new commer- Bank (2001d). influence the investment climate in two main ways: in the level of social cohesion (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 51 and trust between market participants, and mate thus bene�ts from a social consensus in the level of trust and con�dence citizens in favor of creating a more productive soci- have in �rms and markets. Governments ety—and from widely held perceptions that influence, and are influenced by, both. processes and outcomes are legitimate in the sense that they are consistent with social Social cohesion and trust norms, values, and beliefs.48 Social cohesion and trust can reduce the Public attitudes toward �rms and mar- costs of regulation and contract enforce- kets can be deeply rooted in history, but ment—a plus for the investment climate. also reflect more contemporary experience. Trust and shared values and expectations They can also be complicated, not least (social capital) facilitate cooperative rela- because even a single individual often tionships and can encourage �rms to needs to reconcile divergent perspectives, lengthen their planning horizons as they including as a consumer, a worker, a tax- think about investing.44 Richer networks of payer, and often also as an investor.49 To trust also make it easier for participants to further complicate matters, support for exchange reliable information about each markets does not always track economic other, and to monitor the actions of policy- growth50 (�gure 2.11). makers. Recent opinion surveys suggest that atti- The potential positive economic effects tudes toward international economic inte- of social capital have been documented gration and �rms vary considerably around since Alexis de Tocqueville’s travels in the the world, but tend to be favorable. For United States in the early 19th century. But example, for more than 85 percent of coun- social capital can also have negative effects tries surveyed, between 77 percent and 98 given its tendency to foster closed, insular percent of respondents believed interna- relations among individuals of similar tional trade and business were positive backgrounds, to encourage conformity, and forces for their country (�gure 2.12).51 to ostracize innovators and individualists.45 Similar surveys often �nd that con�- Cronyism and corruption may also be tol- dence in major corporations is somewhat erated more in communities characterized less positive. Ambivalence toward markets by high levels of social capital.46 At the other extreme, societies that are Figure 2.11 Support for markets does not always track economic highly fragmented along ethnic or linguistic growth—as in Latin America lines can experience social conflict that 20 Percent change in respondents supporting a market economy undermines the investment climate. Cross- country studies show that ethnic and lin- Mexico guistic fractionalization is negatively associ- ated with economic growth.47 The negative 10 effects on the investment climate may range Nicaragua from open conflict and political instability to clientelist distortions in policymaking. 0 Creating a society that bridges these divides Paraguay Venezuela, RB de Honduras can take generations. Ensuring that the ben- Brazil Costa Rica e�ts of a better investment climate extend Uruguay Bolivia Panama –10 Guatemala to all members of society can help build Argentina Chile those bridges. Ecuador Peru Trust and con�dence in �rms and –20 Colombia markets El Salvador Public attitudes toward �rms and markets –20 –10 0 10 20 can affect the feasibility of policy improve- Percent change in real per capita income ments. They can also affect the sustainabil- Note: Change in support measured as change in those responding, “strongly agree� or “agree� to the statement “In general, a market economy is best for our country.� Responses ity of reforms and hence the credibility of cover years 1998–2002. government policies. The investment cli- Source: www.latinobarometro.org. (c) The International Bank for Reconstruction and Development / The World Bank 52 WORLD DEVELOPMENT REPORT 2005 Figure 2.12 Strong support for international trade and business—but less con�dence in corporations 4 100 Support for international trade and business 80 Confidence in major corporations 3 International trade and global business (right axis) 60 Confidence in corporations (left axis) 40 2 20 1 0 n . Sl Ru tina ak sia lg p. Ja ria r an p. P y Ko u M rea o ly Ch .K. Fr ina ra e e Ve an S. zu a Br la ng Ind il la ia Ph Po sh So lip land Af s Ni rica Gh ria a ge ep Re man az h e of er ic Uk nc in ne ad an Ita Bu Re e de C U. de ut pin Ge p U a ge ov s Ar h R ex a ec i Cz Ba RB Note: Survey questions were (right axis) “Are international trade and business ties good for your country?� (percent responding yes); (left axis) “How much con�dence do you have in major corporations?� (1 = none, 4 = a great deal). Source: The Pew Global Attitudes Project (2003) and Inglehart and others (2000). and �rms, particularly toward “big busi- Multinational �rms have long aroused ness,� has a long pedigree.52 Historically suspicion due to concerns about their loyal- these concerns have been heightened by ties and their possible economic power.54 corporate or corruption scandals, leading to This has recently led to mutual efforts to public backlashes against �rms and markets promote corporate social responsibility and to demands for more intrusive regula- through the elaboration of various codes of tion or even nationalization.53 These con- conduct (see box 2.2). Other concerns cerns also reflect responses to the way gov- about government–�rm relationships, ernments manage conflicts and protect their including those about corruption and other citizens. forms of rent-seeking, are also creating BOX 2.14 Shining the light on government–�rm dealings in natural resources and infrastructure Proposals to enhance the transparency of gov- publicly traded, private, and state-owned extrac- framework for revenue management from the ernment–�rm dealings are often seen as mainly tive enterprises; and others with an interest in pipeline, earmarking revenues for poverty reduc- addressing corruption or other forms of rent- the sector to work together to develop a frame- tion, and requiring private operators to conduct seeking. But reducing concerns about inappropri- work for reconciling payments by �rms to gov- business only with �rms that comply with trans- ate behavior can also contribute to broader pub- ernments and account for any missing amounts. parency and disclosure rules. lic support for �rms and markets, and so facilitate Nigeria took an initial lead in enhancing rev- The impetus for enhanced transparency is ongoing investment climate improvements. enue transparency. In 2003 the Nigerian govern- also extending to private infrastructure arrange- Two recent global initiatives focus on ment agreed to publish budgets and records of ments.Traditionally many countries treated con- improving the transparency of revenue arrange- oil revenue collection, as well as applicable cession contracts and licenses like commercial ments between international investors and host statutes and rules. It also encouraged oil compa- agreements, not publicly disclosed. Growing governments in the natural resources sector.The nies doing business in the country to make full recognition of the public character of these Publish What You Pay campaign, supported by a disclosure of their revenues and costs of opera- arrangements, and of the importance of foster- coalition of more than 200 nongovernmental tion.The accounts are then to be examined by ing broad public support for reforms, has led organizations (NGOs), proposes legislation an “aggregator�—an independent auditor—to Argentina, Brazil, Panama, and Peru to publish requiring publicly listed oil and mining compa- assess any discrepancies. these contracts by placing them on a public nies to disclose information about payments to Under the Extractive Industries Transparency Web site.Together, they have published more government as a condition of stock exchange Initiative a commission was also established in than 120 contracts covering a range of listing.The Extractive Industries Transparency Azerbaijan to publish revenues of the State Oil infrastructure sectors. Initiative, launched at the World Summit on Sus- Fund. In a similar vein the Chad-Cameroon tainable Development in 2002, encourages gov- Petroleum Development and Pipeline project, Source: World Bank (2000b); World Bank (2001e); ernments; international organizations; NGOs; supported by the World Bank, established a and World Bank staff. (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 53 impetus to enhance the transparency of This means governments need to weigh dealings between governments and �rms, carefully the costs and bene�ts of alterna- particularly in areas where relationships can tive approaches and take local conditions be especially troublesome (box 2.14). into account when designing particular pol- Because public support for markets does icy responses. Failure to give suf�cient not necessarily track economic growth, and weight to local conditions can leave impor- because the growth response from reforms tant market failures unchecked—or make is not always immediate, governments often matters worse. For example, approaches need to actively foster public support for that demand enforcement capacity beyond investment climate improvements. Building that available may not only fail to meet the a consensus in favor of a more productive intended social objective but can also con- society not only enhances the feasibility of tribute to informality and corruption and reform, but through its impact on sustain- undermine government credibility. Simi- ability and hence credibility can also have a larly, in the absence of effective safeguards, big influence on the size of the investment approaches that involve signi�cant discre- response. There are no simple formulas in tion may be misused to obtain bribes or this area, but experience underlines the expose �rms to unnecessary uncertainty importance of four key elements: and risk (box 5.2). The challenge of ensuring that policy • Ensuring the bene�ts of a better invest- responses �t with local institutional condi- ment climate are not con�ned to partic- tions has implications for policy design ular categories of �rms, but extend across the investment climate. It plays an widely across society especially important role in the design of • Promoting broad public understanding regulatory strategies but is also relevant to of the bene�ts of reform the distribution of responsibilities between • Enhancing the transparency of govern- tiers of government (box 2.15). ment-�rm dealings to reduce concerns Because conditions vary across coun- about rent-seeking tries, transplanting approaches uncritically • Protecting vulnerable groups that may from one country to another often leads to be disadvantaged during the transition. Ensuring policy responses BOX 2.15 Decentralization and the investment climate reflect a good institutional �t Market failure is the textbook rationale for Decentralization has been a theme in con- dictions.They may also face more severe stitutional design since at least the founda- capacity constraints and be unable to most government interventions intended to tion of the Swiss Confederation in 1291, and exploit economies of scale associated with improve the investment climate. But those remains a major theme to this day. How particular functions. And subnational gov- interventions can fail to achieve their does decentralization affect the investment ernments are not immune from governance climate? problems—and in some contexts may be intended result for myriad reasons, includ- Decentralization can contribute to a more vulnerable to them than national ing inadequate information, expertise, or sound investment climate in several ways. authorities. resources—or from rent-seeking, credibility Decentralization of regulatory responsibili- Reflecting these tradeoffs, the optimal gaps, and lack of public support. The suc- ties can help locales adapt approaches to location of particular policy and administra- their conditions and preferences and facili- tive responsibilities will depend on the cess of any policy intervention ultimately tate the involvement of stakeholders. Fiscal country and policy issue concerned. Small depends on the extent to which the chosen decentralization can assure local authorities countries present fewer opportunities for approach reflects a good �t with local insti- that taxes raised locally will not be appro- decentralization than larger ones. But even tutional conditions. priated by the central government, giving in large countries, some matters will be best local authorities incentives to develop their handled centrally, some subnationally, and Market failures may be more prevalent local tax base. Decentralization also permits others may require some form of shared in developing countries than in developed a degree of institutional competition responsibility. A clear delineation of respon- countries.55 But government failures can between centers of authority that can stim- sibility between tiers of governments also be more severe in countries with lim- ulate policy innovation and reduce the risk reduces uncertainty and risk for �rms and that governments will expropriate wealth. improves accountability. ited resources and expertise and less devel- But there are tradeoffs. Subnational oped checks on government behavior. Pol- authorities are not well placed to deal with Source: Brueckner (2000); Treisman (2000); Tanzi icy interventions make sense only when the issues that involve spillovers between juris- (1995); and Weingast (1995). expected bene�ts exceed the likely costs. (c) The International Bank for Reconstruction and Development / The World Bank 54 WORLD DEVELOPMENT REPORT 2005 poor results. Historically, many regulatory the information costs faced by foreign systems in developing countries were trans- investors and help signal the application of planted from colonial or occupying powers high standards to local stakeholders. But with little regard to how they might operate failure to adapt approaches to local realities in a very different environment. Because can lead to outcomes as poor as their more they were less relevant to local circum- ancient forebears. stances, they were often ignored or enforced Strategies for tailoring approaches to local selectively to solicit bribes. While the laws in conditions vary according to the area of pol- the source country went through a continu- icy intervention. They may involve develop- ing process of modernizing and upgrading, ing simpler rules with less discretion; relying the regimes left behind often did not. For more heavily on transparency, competition, example, company law regulating business and market monitoring; and reinforcing local entry dates back to 1884 in the Dominican institutional safeguards, including through Republic and to 1901 in Angola, while laws the use of appropriate international arrange- dealing with insolvency date back to 1916 in ments. These strategies need to be comple- Nicaragua. One result is a high level of mented by efforts to strengthen government informality, with regulations ostensibly capabilities (chapter 3). aimed at mitigating market failures or pro- Advances in information technology are moting other social objectives often com- also creating opportunities to reduce plied with by less than half the economy— demands on government capabilities, while yet placing a disproportionate burden on enhancing transparency and easing the bur- �rms that do comply. den on �rms.56 These approaches have been A tendency to transplant approaches applied to a wide range of investment cli- uncritically from one country to another mate areas, including business regulation continues to this day. Policy approaches in and land titles (box 2.16) as well as tax and today’s rich countries can provide a useful customs administration (chapter 5). source of inspiration. They may also reduce Making progress These four separate but related challenges BOX 2.16 E-government and the investment climate can produce vicious circles of worsening Advances in information technology, includ- India’s Karnataka state introduced an governance and stagnating investment cli- ing the Internet, are paving the way for electronic land-titling system, Bhoomi, in mates. Weak control over rent-seeking not investment climate improvements that the late 1990s.The online system is only directly leads to poor economic out- reduce demands on public administration, delivered through kiosks installed in all land comes, but also undermines government enhance transparency, and ease compliance of�ces of Karnataka.These kiosks provide burdens on �rms. Approaches to business copies of a Record of Rights, Tenancy, and credibility and can create or exacerbate �s- regulation in Singapore and land titling in Crops (RTC). Obtaining an RTC once sures in society, and erode public trust in India’s Karnataka state illustrate the potential. required up to 30 days, and typically a bribe �rms and markets. Low government credi- The e-government initiative launched of as much as Rs. 2,000 (about $43). Land bility can contribute to rent-seeking and a by Singapore in 2000 included business records could be deliberately “blurred� for registration and licensing procedures. It fees of Rs. 10,000 ($220).These records were lack of public trust in �rms and markets. provides an online application system for not open to the public, and it sometimes Lack of public con�dence in �rms and mar- business registration and licensing and a took two years for the records to be kets can undermine the credibility of policy one-stop online application system for cer- updated under the manual accounting sys- reforms. Policy interventions that are tain special licenses (for example, building tem maintained by 9,000 “village� accoun- and construction permits) that previously tants—state employees responsible for poorly adapted to local conditions can leave required separate submissions to as many three to four villages each.Today an RTC can important market failures unchecked, as 12 regulatory authorities.The integrated be obtained for a �xed fee of Rs. 15 ($0.32) encourage informality and rent-seeking, approach reduced the cost of incorporating in 5 to 30 minutes.The records are open for a new company from anywhere between public scrutiny. Citizens can now request undermine credibility, and also weaken S$1,200 and S$35,000 (around $700 to that land titles be updated quickly through public trust in �rms and markets. Con- $20,000) (depending on the capital of the the kiosks, a process that has increased the versely, the circles can be virtuous—with company) to a flat fee of S$300 ($175). What number of annual applications for updates progress in one area contributing to that in used to require two days now requires less by 50 percent. than two hours. Streamlining the submis- others. sion process for construction permits saves Source: Tan (2004); Bhatnagar and Chawla A common strategy for addressing all applicants more than S$450 ($260). (2004); and Lobo and Balakrishnan (2002). four challenges is to enhance the trans- parency of government-�rm dealings. This (c) The International Bank for Reconstruction and Development / The World Bank Confronting the underlying challenges 55 can play a critical role in restraining rent- taken not to encumber weak administrations seeking, in contributing to policy credibility, with some of the more elaborate procedures and in helping to build public support for adopted in some developed countries, more reforms. It can also be part of a strategy for pragmatic approaches, including those that complementing government capabilities and exploit the potential of new information thus helping to ensure policy interventions technologies, create opportunities to trans- reflect a good institutional �t. Governments form governments—and the investment cli- in both rich and poor countries have a long mates they produce. history of resisting calls for more openness, and some �rms bene�t from the resulting secrecy.57 But more governments are open- Improving the investment climate requires ing their policy processes to public scrutiny governments to address these challenges in and improving public access to information. the context of speci�c policy areas affecting Stakeholders are being consulted on regula- stability and security, regulation and taxa- tion in Bolivia and Ghana. Infrastructure tion, �nance and infrastructure, and work- contracts are being placed on public Web ers and labor markets. The agenda is broad sites in Argentina and Peru. Freedom of and demanding. Chapter 3 looks at what information legislation is being introduced has been learned about successful strategies in China and Mexico. While care needs to be for tackling such a broad agenda. (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 3 As chapter 2 highlighted, improving the • Managing individual reforms. Reforms investment climate requires governments to often need to overcome resistance from chapter navigate four sources of potential policy failure that play out across a broad range of those who bene�t from the status quo. This can require a high level of political policy areas, from property rights and busi- commitment, but also bene�ts from ness regulation to infrastructure and labor effective communication, consultation, markets. While the task may seem daunting, and when appropriate, compensation. more countries are making signi�cant • Maintaining momentum. Given the improvements—and are being rewarded breadth of the agenda, and the need to with faster growth and deeper poverty review policies regularly, reforms in this reductions. China, India, and Uganda, men- area can be characterized as a marathon tioned for their achievements in chapter 1, rather than a sprint. To help maintain are hardly alone. Many countries have momentum, many governments are cre- improved at least some areas of their invest- ating specialized supporting institutions, ment climates. Their experience provides including those that facilitate consulta- insights into possible strategies for broad- tion, coordination, the review of existing ening and accelerating progress. constraints, and the review of new policy This chapter opens by looking at the and regulatory proposals. implications of the investment climate’s • Strengthening government capabilities. breadth, encompassing a wide range of gov- Improving government capabilities is an ernment policies and behaviors, many of essential complement to any reform them interrelated, and all possibly influenc- process. This means building not only ing the opportunities and incentives facing more technical expertise, but also better �rms. The good news is that perfection is and more reliable sources of information. not needed in any given area to ignite signif- icant growth and poverty reduction. The key The investment climate is to address important constraints in a way as a package that gives �rms con�dence to invest—and to Government policies and behaviors shaping sustain a process of ongoing improvements. the investment climate play out over a broad The chapter then looks at lessons of domain, from contract enforcement, busi- experience in each of the four key require- ness regulation, and taxation—to �nance, ments for managing such a process: electricity supply, and labor markets. Govern- • Setting priorities. The key is to reduce ments typically administer each area in isola- unjusti�ed costs, risks, and barriers to tion, distributing responsibilities across a competition. But there are no simple range of ministries and agencies. In contrast, formulas for translating those principles �rms tend to view particular investment to speci�c reform areas. Priorities need opportunities as a package, with government to be determined in each case based on policies and behaviors that influence the an assessment of current conditions, the costs, risks, and barriers to competition as potential bene�ts from improvement, part of that package. Why might this matter? the links with national or regional goals, First, the impact of any policy improve- and implementation constraints. ment will depend on how it addresses a 56 (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 57 constraint that is actually binding on �rms. simultaneous and comprehensive reform is So expanding access to credit will not have necessary for signi�cant results. Indeed, much impact on �rms’ investment deci- efforts to tackle the full set of investment sions—an effort sometimes described as climate policies simultaneously, even if “pushing on a string�1—until more funda- technically feasible, could generate so much mental concerns about the security of their uncertainty for �rms that it might deter property rights have been addressed.2 Pro- rather than encourage investment, at least viding tax breaks may not be enough to temporarily.4 Deep and rapid institutional compensate for other weaknesses in the change can also be disruptive for society, investment climate in some situations— possibly undermining public support and but may be unnecessary in others.3 Simi- thus the sustainability of reform. So some larly, introducing a competition law may sequencing of reforms is inevitable in a �eld not have a big impact on the economy as broad as the investment climate. Fortu- when the main barriers to competition nately, experience shows that countries can stem from trade restrictions, government reap signi�cant bene�ts by addressing monopolies, or other regulatory barriers to important constraints in a way that gives entry and exit. �rms con�dence to invest—and sustaining Second, different areas of the investment a process to address other constraints as climate policy can interact. Clarifying rights they become more binding. to land can help ease access to credit by Take China, the country enjoying the �rms and households—but only when world’s fastest growth and poverty reduc- complementary aspects of �nancial infra- tion in recent years. The reform that structure are in place. Reducing barriers to ignited growth was the introduction of a trade will not deliver its full potential if rudimentary system of property rights, ini- weak bankruptcy laws slow the exit of less tially for township and village enterprises ef�cient �rms, or if labor market policies and then for individual farmers and entre- limit the ability of �rms to adjust produc- preneurs. Once of�cial targets were met, tion processes to respond to a more com- additional production could be sold for petitive environment. Similarly, efforts to personal gain. The improvements unleashed encourage local R&D can be hobbled by a strong response because of the size of the shortages of skilled workers, limited com- economy bene�ting from the change, and petition, or weak intellectual property because the changes were implemented in rights. ways that gave people the con�dence to So investment climate improvements invest (box 3.1). Subsequent improve- involve more than one-off, “stroke-of-the- ments—including those attracting foreign pen� reforms. But this does not mean that direct investment (FDI) and improving BOX 3.1 Improving the investment climate, China’s way Growth in China is of�cially reported at an average by the large size of the economy affected.No less the envy of many developing countries—and it of 8 percent a year for the last 20 years—giving it important,the reforms were interpreted by individ- is not just about wages or exchange rates.The the most impressive (if disputed) sustained growth uals and emerging enterprises as a decisive shift in surveys show that in �ve of the main industrial performance in history.Declines in poverty have government policy favoring private initiative,rein- centers, the costs of infrastructure disruptions, been equally dramatic—from 60 percent of the forced by a high level of policy stability,strengthen- crime, bribes, regulation, and contract enforce- population to 17 percent.Yet China only recently ing the con�dence to invest.The initial signal was ment dif�culties average less than 14 percent of gave constitutional protection to private property con�rmed by subsequent reforms that improved sales.This is well below the average in countries rights, inef�cient state-owned enterprises still clut- the environment for private business.These such as Brazil and Pakistan, and half the average ter the landscape, and the �nancial sector is included efforts to attract FDI,improvements to in Tanzania (see �gure 1.2). China still has a long dragged down with nonperforming loans.How business regulation and infrastructure,accession to way to go—especially in extending similar was such sustained growth possible? the World Trade Organization (WTO),and efforts to improvements across the country—but its Growth was ignited by introducing a rudimen- tackle corruption and improve transparency. strong performance is less of a riddle when tary system of property rights that gave farmers The Bank’s Investment Climate Surveys viewed in this light. and township and village enterprises incentives to show that China has created an investment cli- Source: Chen and Wang (2001); Qian (2003); and take risks and invest.The response was magni�ed mate in its main industrial centers that would be Young (2000). (c) The International Bank for Reconstruction and Development / The World Bank 58 WORLD DEVELOPMENT REPORT 2005 BOX 3.2 India’s path In India much attention is paid to the liberaliza- were further reduced. Over the 1990s the pace productivity signi�cantly but that the aggregate tion efforts of 1991. Growth actually began pick- slowed but reform continued. Licensing has numbers have been slow to respond. In many ing up in the 1980s.The early reforms were less been eliminated in all but seven industries. Pri- sectors the dispersion of productivity has dramatic, more ad hoc, but they signaled an vate �rms have been allowed to compete in increased, with the more advanced �rms realiz- important shift in government policy toward more and more sectors. A new competition law ing additional gains, and the least productive the private sector. replaced the former Monopolies and Restrictive �rms falling behind.The expected pattern In 1984 Rajiv Gandhi’s government initiated Trade Practices Act, which had required special would have been to see greater competitive reforms to encourage exports, facilitate foreign approval for any large investment. Long-stand- pressures reduce dispersion as less successful technology transfers, and rationalize the tax sys- ing problems in infrastructure are being tackled. �rms left the market.This highlights the signi�- tem. Quantitative controls on the import of capi- Anticorruption efforts are also being scaled up cance of continuing barriers to exit. According tal goods were eliminated.Tariffs were cut by 60 at the national and state levels. to the Bank’s Doing Business Project, it can take percent.Taxes on pro�ts from exports were cut The effects have been substantial. Private 10 years to complete bankruptcy procedures in by half. Fewer industries were subject to licens- investment as a share of GDP grew from less India. Firms may be taking advantage of ing.The policies were a major shift in approach than 9 percent in 1981 to more than 15 percent stronger incentives to invest, but there clearly is away from socialism and the primacy of redistri- in 2000. Growth increased from an average of scope for further improvement. bution over growth in production. 2.9 percent a year in the 1970s to 5.8 percent in In the early 1990s the reforms were more the 1980s, and to 6.7 percent in the mid-1990s. Source: Aghion and others (2003); Ahluwalia (2002); dramatic—the Rupee became convertible, More puzzling, however, has been the De Long (2003); Rodrik and Subramanian (2004); restrictions on foreign ownership were relaxed, impact on total factor productivity.The general Varshney (1998); and Panagariya (2003). additional quotas were abolished, and tariffs pattern is that many �rms have increased their business regulation and infrastructure— tor, while evident, has been reduced by bar- addressed constraints initially less binding. riers to exit that slow the pace of industrial A degree of autonomy between provinces restructuring. Similarly, labor market has also fostered experimentation and cre- restrictions have limited the productivity ated incentives for lagging provinces to improvements from trade reforms in many emulate the success of their faster moving countries in Latin America.6 Investment cli- counterparts.5 mate policies also require regular review to India’s experience highlights the same take into account changes in the conduct of basic point (box 3.2). Its current period of business, and ongoing lessons of experi- growth began with some trade, tax, and reg- ence. Both considerations underline the ulatory reforms in the 1980s. Firms importance of processes to support ongo- responded because the reforms addressed ing policy improvements. As Porter important constraints and because they observed, reforms in this area are a were seen as signaling a decisive policy shift marathon, not a sprint.7 toward private sector–led growth. Subse- quent reforms, including the dismantling of Setting priorities the “licensing Raj� and further trade liberal- Improving the investment climate involves ization in 1991, did more to reduce costs reducing unjusti�ed costs, risks, and barri- and increased competitive pressure in the ers to competition. In practice, costs, risks, economy. Just as in China, a degree of and barriers are a function of government autonomy between state governments cre- policies and behaviors that play out ated room for states to innovate. Competi- through a wide range of speci�c policy tion between states is creating incentives for areas. Where should governments begin? lagging states to follow the leaders, includ- The diversity of investment climate con- ing by addressing long-standing problems ditions across and within countries, and the in the power sector. potential for reforms to impact on �rms Even when a policy improvement and activities differently, mean that there addresses an important constraint, and is are no standard formulas. Governments implemented in a credible way, the extent of need to determine priorities by assessing the bene�ts often depends on going on to current conditions, the potential bene�ts address constraints that may have been less from improvement, the links with broader binding initially. For example, productivity national or regional goals, and implementa- improvement in India’s manufacturing sec- tion constraints. (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 59 Current conditions Figure 3.1 Constraints reported by �rms— comparing Bulgaria, Georgia, and Ukraine As chapter 1 highlighted, investment cli- mate conditions vary dramatically across Ukraine Security & stability 1 and within countries. A major impediment Georgia in one country may be much less important Bulgaria Regulation in another—as a simple comparison Labor 0.5 between Bulgaria, Georgia, and Ukraine illustrates (�gure 3.1). Assessing constraints on existing �rms is fairly straightforward—�rms can be asked directly through dialogues with representa- Infrastructure Taxation tives of the business community or through surveys. The World Bank’s Investment Cli- mate Surveys collect not only subjective Finance assessments of constraints, but also more Note: Resulting indicators range from 0 (best) to 1 (worst). Indices are based on surveys of formal �rms. Values are normalized by objective data on the impact of those con- regional maxima and minima for each indicator. Countries selected to highlight differences. straints. Engaging with �rms has the addi- Source: World Bank Investment Climate Surveys. tional bene�t of enhancing a government’s credibility with �rms, and also helping with possible implementation issues. But focus- ing on the views of existing �rms has one Progress in addressing broader gover- obvious drawback: those �rms cannot (or nance issues, particularly those affecting the will not) speak on behalf of �rms that have government’s credibility, also tend to pay not yet entered the market, and so may bigger dividends than reforms in any one place less emphasis on barriers to competi- policy area, because they can leverage the tion. Policy barriers to entry (and exit) thus impact of other policy improvements warrant particular scrutiny. (chapter 2). Efforts to build credibility and Comparing a country’s performance in a legitimacy are usually especially important given policy area with that of other coun- in weak or vulnerable states. In these cases tries also provides insights into the poten- emphasizing consultative processes and tial scope for improvement. For example, transparency can help to heal the social the Bank’s Doing Business Project shows wounds from conflict—or from distrust that it takes more than 200 days to register a about whose interests are being served. For business in Haiti but less than 20 in Latvia example, Uganda placed special emphasis and just 2 in Australia. Similarly, it takes on ensuring that the bene�ts from improve- 1,000 days to enforce a contract in Poland, ment were widely understood—and widely but less than 50 days in the Netherlands and shared. Similarly, the Bulldozer Initiative in Tunisia.8 New sources of data make bench- Bosnia-Herzegovina emphasizes grassroots marking of this kind feasible for a growing involvement and broad consultation (see range of policy parameters. box 3.9). Building credibility can be critical in stemming, and reversing, the capital flight Potential bene�ts and “brain drain� in states under stress.9 Addressing constraints that affect a large share When accelerating overall growth is the of economic activity will usually have a bigger priority, the share of GDP affected and the impact than those affecting only a smaller severity of the constraint will usually be share. War and major episodes of political important criteria. Targeting constraints that instability trump all other constraints on this unlock opportunities and improve incentives criterion, and progress on these issues is fun- for a large share of GDP—as China did with damental to creating a decent investment cli- its rural sector—can have a big impact on mate (chapter 4). Improving macroeconomic aggregate growth. stability also falls within this category, because without it changes in other areas will have Poverty impacts. When direct poverty limited traction. reduction is given priority, the key will be to (c) The International Bank for Reconstruction and Development / The World Bank 60 WORLD DEVELOPMENT REPORT 2005 understand how potential investment cli- • Constraints facing �rms that can deliver mate improvements impact the poorest other bene�ts to poor people. While self- members of society in their various capaci- employment and jobs have been identi- ties: as employees, as entrepreneurs, as con- �ed by poor people themselves as the sumers, as users of public services, and as most promising pathways out of poverty, recipients of tax-funded services or trans- investment climate improvements can fers (chapter 1). The breadth of these deliver additional bene�ts to poor peo- impacts means that there is no one best way ple. Improving conditions for �rms that to make investment climate improvements produce or distribute goods and services more pro-poor. Certainly, poverty reduc- consumed by poor people can have a big tion does not justify an exclusive focus on impact on their living standards. Improv- small or informal �rms. ing infrastructure in a particular location One approach is to focus on constraints can also enhance living conditions for in locations where poor people live, which poor people, whether or not they work or can bene�t poor people in all their various engage in entrepreneurial activities. capacities. Rural poverty is a major chal- Because larger �rms are more likely to lenge in many countries. Nonfarm employ- pay taxes, improving their conditions ment can contribute much to the incomes increases the potential for them to con- of the rural poor, and research in India sug- tribute to social objectives. gests that manufacturing jobs contribute twice as much as agricultural productivity Potential spillovers. When considering the in raising nonfarm income. There can also potential bene�ts from an improvement, it be opportunities to focus improvements on is also important to look at the possible urban or peri-urban areas with high con- spillovers beyond the �rms and activities centrations of poverty. most directly affected. Six are worth high- A second approach is to focus on con- lighting: straints to particular activities that bene�t • Spillovers to other �rms. Sometimes the poor people in their various capacities: bene�ts of an improvement spill over • Constraints facing microentrepreneurs. from the �rms that immediately bene�t Hundreds of millions of poor people earn from the reform to others. For example, their livings as microentrepeneurs in the one of the attractions of increasing FDI informal economy. Improving the invest- is that technology and expertise may ment climate they face can involve improv- spill over to local suppliers, customers, ing the security of their property rights, and competitors. reducing red tape in business registration, • Spillovers to other policy areas. Improve- and removing distortions that make access ments in some policy areas can make a to �nancing more dif�cult. Sometimes the positive contribution to others. For impact may not be fully anticipated: for example, increasing the security of rights example, liberalizing telecommunications to land can help ease access to �nancing in Bangladesh and Uganda created oppor- (chapter 4). tunities for microentrepreneurs to enter • Spillovers to government credibility. The the market, helping them and their broader way governments approach policy communities. improvements can help—or harm— • Constraints facing other �rms likely to cre- their credibility and resulting investor ate jobs for poor people. Improving con�dence. Efforts to engage �rms and investment climate conditions for �rms other stakeholders openly and transpar- likely to hire poor people can do much ently, with timely execution of reforms, for poverty reduction. This may mean can enhance �rms’ con�dence and so focusing on constraints faced by larger elicit a stronger investment response. �rms, which create jobs directly and also The corollary is that overly ambitious or create more opportunities for suppliers poorly executed reforms can undermine of a range of goods and services. credibility and con�dence. (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 61 • Spillovers to government capabilities. Some avoids the dif�culty of governments trying investment climate improvements can to “pick winners� where the track record has strengthen a government’s �scal position— been discouraging (chapter 8). It creates and so facilitate other improvements. For opportunities for unforeseen success stories example, Uganda gave early priority to bet- to emerge. It reduces concerns about rent- ter revenue collection, nearly doubling the seeking. And ensuring that opportunities ratio of tax revenue to GDP between 1991 for growth are shared widely in society and 1996. Privatizing state-owned enter- helps build social cohesion and support for prises can sometimes play a similar role. ongoing policy improvements. • Spillovers to broader social goals. Many Investment climate improvements can features of a good investment climate affect �rms and activities differently. deliver bene�ts that extend beyond Because of this, priority-setting may be �rms. For example, more effective courts influenced by the weight governments place can help defend civil and political rights, on a subset of the goals a good investment not just property rights (chapter 4). Bet- climate can deliver: ter infrastructure and �nancial systems • Integrating �rms in the informal and help all members of the community, rural economies whether engaged in entrepreneurial • Unleashing the growth potential of activities or not (chapter 6). smaller �rms • Spillovers to constituency building. The • Taking advantage of opportunities from choice of initial priorities can also influ- international openness ence the feasibility of later improvements. • Allowing �rms to climb the technology For example, reducing barriers to new ladder. business formation can increase the pool of �rms with an interest in broad-based What are the implications for priority-setting? policy improvements. Similarly, ensuring Integrating �rms in the informal economy. that improvements extend to �rms across Most developing countries have a dual society—rather than just to large or con- structure, with a modern economy operat- nected �rms—can contribute to the pub- ing alongside a more traditional economy lic support necessary to sustain progress. with high levels of informality. Estimates Priority-setting may also be influenced suggest that more than half the economy is by broader strategic considerations. For informal in many developing countries (�g- example, barriers to entry may be easier to ure 1.17)—and that informality is grow- address than labor market distortions—and ing.11 There are also degrees of informality. may facilitate subsequent labor market One criterion is whether �rms are registered reforms by reducing the rents available for with the government, another is compliance the participants to contest.10 with regulations and taxes. What is striking Some improvements—such as reducing is how few �rms are completely “formal� by barriers to entry—can deliver fairly quick the second de�nition (�gure 3.2). results. Others require a longer process of The informal economy is diverse, rang- institutional development to deliver their ing from subsistence farmers and those full potential—such as reforms to courts engaging in entrepreneurship out of neces- and the development of new regulatory sity,12 to more affluent �rms that �nd it fea- agencies. They promise large bene�ts but sible to evade tax and regulatory obliga- require patience and persistence. Of course, tions, and others in the middle. A large pool the sooner the longer-term projects begin, of individual workers also exists in the the sooner the bene�ts arrive. informal economy, sometimes working for formal �rms “off the books,� sometimes Link with national or regional goals working for enterprises that are themselves Creating an investment climate that allows informal. Women are disproportionately �rms of all types to grow and contribute to concentrated among the smallest of the poverty reduction has many advantages. It informal microenterprises (�gure 3.3).13 (c) The International Bank for Reconstruction and Development / The World Bank 62 WORLD DEVELOPMENT REPORT 2005 Figure 3.2 Informality is a matter of degree lowest rung of the ladder out of business and so exacerbate poverty. Recent work in Unregistered, 1–4 workers Registered, <50 workers Egypt suggests that society as a whole can Unregistered, 5+ workers Registered, 50–199 workers 100 Registered, 200+ workers be worse off if this were to happen, but be better off if formalization were encouraged in an environment with reformed regula- Percent of sales reported to tax authorities tions.14 Experience in Vietnam and Uganda 75 shows that reducing unjusti�ed regulatory burdens, including the costs of going for- mal, can do much to encourage formality 50 (chapter 5). Beyond encouraging formality, govern- ments can focus on addressing constraints faced by microentrepreneurs in the infor- 25 mal economy. The constraints they per- ceive can differ from those of formal �rms.15 Informal �rms can evade many regulatory and tax obligations, but face 0 Bangladesh Brazil Cambodia Guatemala Indonesia Tanzania other obstacles, including less secure prop- Source: World Bank Investment Climate Surveys and WDR Surveys of Micro and Informal Firms. erty rights and greater dif�culty obtaining access to �nance and public services. Entre- Governments have an interest in preneurs who do not have a �xed place of expanding the net of the formal economy to business, such as street vendors, are partic- broaden the tax base, extend the reach of ularly vulnerable.16 While constraints need regulations intended to meet important to be assessed in each context, surveys social objectives, and remove distortions in undertaken for this Report show that pri- competition between �rms in the formal ority areas will often include strengthening and informal economies. They also have an property rights, such as clarifying rights to interest in reducing obstacles to growth land (chapter 4);17 reforming regulations faced by �rms, and in expanding income- or taxes that encourage informality or con- earning opportunities for those on the low- tribute to harassment and corruption est rung of the economic ladder. Getting the (chapter 5); and improving access to credit, balance right can be dif�cult. Simply including though micro�nance schemes enforcing existing regulations and taxes (chapter 6). Reforming labor market regu- more strenuously may drive those on the lations can also encourage greater formal- ity in employment relationships, and so Figure 3.3 Women’s participation is concentrated in the informal extend the coverage of important protec- sector, among the smallest �rms tions for workers (chapter 7). 75 Unregistered, 1–4 workers Integrating �rms in the rural economy. Unregistered, 5+ workers Registered Many �rms operating in rural areas also Share of workers who are women tend to be part of the informal economy, but rural location can be a separate source 50 of disconnection from the modern econ- omy. Seventy percent of people in low- income countries live in rural areas, and 25 improving their opportunities can make a direct contribution to reducing poverty. Increasing the productivity of agricul- ture expands opportunities in rural areas— 0 not least because it increases the demand Bangladesh Brazil Cambodia Guatemala Pakistan Tanzania Source: World Bank Investment Climate Surveys and WDR Surveys of Micro and Informal for local services and provides an important Firms. means of diversifying risks.18 Improving (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 63 security of rights to land has been shown to Figure 3.4 The contribution of SMEs to GDP does not vary too much by income—but the relative have a big impact on agricultural produc- importance of informal and formal �rms shifts tivity (chapter 4), and breaking up agricul- dramatically tural monopolies can also expand opportu- 100 nities for poor farmers (chapter 5). But increasing rural nonfarm income is often identi�ed as the most important way to Informal combat rural poverty.19 75 activity Nonagricultural activities account for up Percent of GDP to 50 percent of rural employment and household income in many developing 50 SME countries, with the �gures highest in Africa, activity followed by Latin America and East Asia, and lowest in South Asia.20 Nonagricultural 25 salaried employment is associated with the Other richest quintiles in rural areas, agricultural wages with the lowest, and self-employment in the middle.21 Rural areas with lower agri- 0 cultural productivity can make substantial Low-income Middle-income High-income contributions to incomes through manu- countries countries countries facturing. Labor and land costs are typically Source: Ayyagari, Beck, and Demirgüç-Kunt (2003). lower than in urban areas, leading some manufacturing companies in India to relo- the majority of GDP across country groups cate to rural areas to serve urban markets (�gure 3.4). There is ongoing debate about and even to export.22 whether small �rms play a special role in Distance and low population density add economic development and so might merit to the challenges of �rms in rural areas. special policy privileges (box 3.3). But what- Lower concentration denies them the bene- ever the weight given to such claims, smaller �ts of agglomeration economies that �rms �rms do tend to face more burdens than in urban centers enjoy. It also makes it more larger �rms in a weak investment climate. costly to supply modern infrastructure and Investment climate constraints that rep- provide other services valued by �rms. Sub- resent a �xed cost hit small �rms harder— sidizing infrastructure and other services whether through regulatory compliance for rural communities is politically popu- costs,23 the costs of self-provision of elec- lar—but often poorly targeted and dif�cult tricity or security services, or bribes.24 Lim- to sustain. In some cases the patronage ited assets to pledge as collateral and threatens the viability of service provision shorter credit histories can also make it across the economy (see box 6.6 on India’s more dif�cult for smaller �rms to obtain power sector). access to �nance. This means that improve- Many governments are responding with ments to the broader investment climate more pragmatic approaches to the provi- will tend to provide disproportionate bene- sion of infrastructure and other services. �ts to smaller �rms. Creating a better investment climate for Removing policy and regulatory distor- small private providers, such as those deliv- tions will usually be the most effective strat- ering electricity in rural areas in Cambodia egy to help unleash the growth potential of and Yemen, can play an important role small �rms. If �rms remain small because of (chapter 6). policy-induced distortions or disproportion- ate burdens that inhibit their growth, remov- Unleashing the growth potential of smaller ing those distortions is an important step.25 �rms. Small and medium �rms (SMEs) Strengthening the protection of property account for the bulk of �rms and employ- rights and establishing credit bureaus and ment in the formal economy and, together asset registries can also help small �rms with informal microenterprises, account for obtain access to �nance (chapter 6).26 (c) The International Bank for Reconstruction and Development / The World Bank 64 WORLD DEVELOPMENT REPORT 2005 BOX 3.3 Do small �rms play a special role in economic growth? Microenterprises in the informal economy often production.They do tend to be nimbler than plausible, but imply that policy responses should receive particular attention due to their role in large �rms in responding to niche opportunities aim to remove barriers facing all �rms in the econ- poverty reduction. Small �rms in the formal econ- and changing market conditions. But while omy, rather than targeting a particular group for omy are also often targeted for special policy there are many anecdotes about small �rms pio- special treatment based solely on size. treatment in the belief that they play an especially neering particular technologies or ideas, �rms Recent macroeconomic evidence also casts powerful role in economic development, but that �t that pro�le seem to be the exception doubt on the claim that SMEs are especially impor- these claims are dif�cult to substantiate. rather than the rule. Indeed, most R&D in devel- tant for growth and poverty reduction.A cross- Some believe that SMEs warrant special oping countries is undertaken by larger �rms country study looking at the correlation between attention because of their high rate of job cre- (see table). SMEs also appear less likely to economic growth and SMEs’share of total employ- ation.True, SMEs as a group typically create more engage in activities that promote technology ment found that although the SME sector is larger jobs than larger �rms. But they also tend to shed transfers. For example, small �rms in Brazil, Cam- in countries where growth is faster, the size of the more workers, with a higher rate of “churn,� so bodia, and Pakistan are less likely than larger SME sector did not appear to cause faster growth. do not necessarily lead to greater net job �rms to license technologies from abroad and The study also found no correlation between creation. Large �rms (more than 100 employees) less likely to have technical assistance contracts. poverty reduction and SME development.One were estimated to account for a greater share of Studies in Colombia, Indonesia, Malaysia, Mex- interpretation is that policies that successfully pro- net job creation in Ghana (56 percent), Kenya (74 ico, and Zimbabwe show that small �rms are mote growth—such as those to improve the percent), and Zimbabwe (76 percent) in the early less likely to have formal training programs. investment climate—also promote SME develop- 1990s than small �rms in the formal economy Small �rms in developing countries are also less ment, but that policies that target SME develop- did. SMEs might, however, play a larger role in likely to export than larger �rms. ment do not necessarily result in faster growth. providing opportunities for low-skilled workers. Others believe that expanding opportunities Source: Biggs, Ramachandran, and Shah (1998); Some believe that SMEs are particularly for SMEs can play a special role in helping to Biggs (2003); Acs and Audretsch (1987); Biggs, Shah, innovative—adopting, designing, and produc- broaden public support for markets and in and Srivastava (1995); Batra and Tan (1995); and ing new technologies and new approaches to expanding domestic competition.These claims are Beck, Demirgüç-Kunt, and Levine (2003). Small Medium Large Very Large (< 20) (20–49) (50–249) (250 and up) R&D expenditures (% of sales) 0.9 1.4 1.5 1.4 Any R&D expenditures (% of �rms) 6.7 13.6 20.4 24.9 Formal training program (% of �rms) 27.2 41.6 56.7 63.4 Exports (% of sales) 5.7 10.1 21.0 34.0 Any exports (% of �rms) 12.6 20.9 39.6 56.8 Uses e-mail to communicate with suppliers and customers (% of �rms) 36.0 46.9 55.4 58.9 Source: World Bank Investment Climate Surveys. Taking advantage of international open- economies of scale. The higher productivity ness. Few countries have grown without of successful exporters (box 3.5) can also being open to trade.27 Expanding markets result in spillovers to other �rms in the local and lowering barriers to new products economy. Exporting �rms can contribute to and ideas creates opportunities for devel- raising other �rms’ productivity through oping countries to grow faster and catch demonstration effects, labor turnover, and up with richer countries. More developing connections to overseas markets: �rms in countries are taking advantage of oppor- Mexico in locations where multinational tunities to connect to the international �rms exports are higher are more likely to economy. Their exports increased from 12 export themselves.28 Removing regulatory percent of global GDP in 1970 to 29 per- and other policy-related barriers to export- cent in 2001, and FDI to developing coun- ing is usually a top priority.29 tries increased from 0.1 percent of global What then, about imports? Reducing GDP in 1970 to 3 percent in 2001 (�gure barriers to imported goods can be bene�cial 3.5). While all economies can bene�t, in three ways: international integration is crucial for smaller states (box 3.4). • Reducing the cost of imported inputs. Price Exporting expands access to foreign markups are lower in countries where exchange and allows �rms to exploit foreign competition is greater, however (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 65 BOX 3.4 International integration is especially important for small states Forty-�ve developing countries have fewer than gration can also reduce the cost of telecommu- nomic Cooperation countries. Concerned about 1.5 million people each.Their small local nications and energy infrastructure. the high costs of transportation in the region, markets and small pools of workers limit domes- In the Caribbean two main organizations the Forum’s main priority is shipping. tic competition and the diversity of economic deal with economic integration. The Caribbean Among the many African regional integra- activities. For them, greater integration with Community (CARICOM), with 15 members and tion initiatives, the Southern African Develop- international markets is crucial. It involves pro- a total population of 15 million people, is dis- ment Community (SADC) is one of the more viding adequate infrastructure to facilitate trade cussing a single market and economy to allow successful. It has enabled greater FDI from the and fostering regional cooperation. the free movement of goods, capital, and peo- more developed countries (South Africa and Regional integration enables �rms to ple. The Organization of Eastern Caribbean Mauritius) to the less developed countries, giv- achieve economies of scale by expanding mar- States, a smaller organization with nine mem- ing a new dynamism to the region. French- ket size. It can reduce transaction costs and ber states and 500,000 inhabitants, has already speaking countries in West Africa have created a investment risk, also encouraging more invest- established a common central bank, a common common central bank and have an active pro- ment. Increased opportunities for competition currency, and a common regulator for telecom- gram for harmonizing business regulation (see also strengthen incentives for �rms to innovate munications. It is working on an economic box 9.5 on OHADA). and improve their productivity. Where regional union. Source: Commonwealth Secretariat and World Bank integration involves a common currency or The South Paci�c Forum, a 16-member orga- Joint Task Force on Small States (2004), Brautigam common regulatory frameworks and agencies, nization (including Australia and New Zealand), and Woolcock (2001), Commonwealth Secretariat there can be big reductions in the transaction has adopted investment principles along the (2003), Harsch (2002), and Fairbairn and DeLisle and administrative costs for �rms. Regional inte- lines of those drawn up for the Asia Paci�c Eco- (1996). Figure 3.5 Gross exports and FDI in developing economies jumped in the 1990s competition is measured (by import penetration, effective protection rates, or 40 5 license coverage rates).30 The costs that FDI: Low- and middle-income countries FDI: High-income countries import restrictions impose on �rms and Exports: Low- and middle-income countries Exports as a share of GDP (percent) 4 consumers relying on inputs from the FDI as a share of GDP (percent) 30 Exports: High-income countries protected sector usually far outweigh the bene�ts to the protected �rms.31 3 • Facilitating the diffusion of knowledge and 20 modern technology. Imported machinery is an important source for new technolo- 2 gies. Productivity growth is faster in developing countries that import more 10 1 capital goods from developed economies. One study estimates that if developing countries expanded their trade by 5 per- 0 0 cent of GDP, their output would be about 1970 1975 1980 1985 1990 1995 2000 6.5 percent greater in the long term.32 Note: Gross exports are of goods and services; FDI refers to net inflows in reporting economy. Source: World Bank (2004b). BOX 3.5 Exporting and productivity—what is the link? Economists suggest two possible explanations for kets when protected from international competi- the learning-by-exporting hypothesis. Studies of exporters’higher productivity. One is that export- tion by natural barriers (high transportation costs) exporters in South Korea and Taiwan, China, ing directly improves the productivity of the �rms and policy barriers to trade (tariffs and quotas), found that export buyers were an important doing it (the learning-by-exporting hypothesis). they are unable to survive in international markets. source for new technologies, which they The discipline of competing in international mar- Thus, only ef�cient �rms end up exporting. provided in forms including blueprints, informa- kets encourages �rms to improve their productiv- The two hypotheses are not mutually exclu- tion about manufacturing processes and quality ity or exposes them to foreign technologies and sive. Even if ef�cient �rms are more likely to start control methods, technical advice and on-site modes of production. In addition, exporting exporting, this does not rule out the possibility plant inspections, and training for technical and allows �rms to achieve greater economies of scale that exporting will help them increase their pro- production staff. Some econometric studies also by expanding their potential market. ductivity further. support the learning-by-exporting hypothesis. The second explanation is that because �rms The evidence supports both hypotheses to have to be ef�cient to compete in international some degree. Several econometric studies have Source: Aw, Chung, and Roberts (2000); Bernard and Jensen (1999); Clerides, Lach, and Tybout (1998); markets, only �rms that are already ef�cient can found that productivity improvements precede Hallward-Driemeier, Iarossi, and Sokoloff (2002); export (the self-selectivity hypothesis).Although exporting, providing support for the self-selec- Kraay (1999); Liu, Tsou, and Hammitt (1999); and inef�cient �rms might prosper in domestic mar- tivity hypothesis. But case studies often support Westphal (2002). (c) The International Bank for Reconstruction and Development / The World Bank 66 WORLD DEVELOPMENT REPORT 2005 • Strengthening incentives for local �rms to Foreign �rms also put competitive pres- innovate and improve their productivity. sure on local �rms. This can bene�t �rms Firm-level studies �nd that trade liberal- and other consumers that depend on inputs ization improves productivity among from the industry gaining FDI. In principle �rms competing with imports.33 Episodes the rival �rms might also bene�t from tech- of trade liberalization in Brazil between nological spillovers as well as sharper incen- 1990 and 1995, Chile in the 1970s and tives to innovate and improve their produc- 1980s, India in the early 1990s, and tivity. However, the evidence of horizontal Colombia between 1977 and 1991, were all spillovers from FDI (to �rms that compete associated with higher �rm productivity in with the foreign-owned �rm) is more import-competing sectors.34 The effect of mixed than evidence for vertical spillovers liberalization can be large (box 3.6). In (to �rms that supply or use inputs pro- Colombia a 10 percent decline in tariffs duced by the foreign �rm).39 was associated with as much as a 3 percent Trade and foreign investment are often increase in productivity in �rms.35 The facilitated by informal contacts through emi- productivity gains reflect within-plant grants and diaspora (box 3.7). But the bene- gains and the exit of inef�cient �rms.36 �ts from international openness provide a strong rationale for giving priority to easing Foreign investment can also do much for relevant policy constraints. The agenda productivity—by providing access to new includes improving customs administration, investment capital, new technologies, man- liberalizing trade and foreign investment agement expertise, and export markets. The regimes (chapter 5), and improving trans- positive impact of foreign participation on port infrastructure (chapter 6). Adoption of productivity is demonstrated by studies from international rules and standards can also China, the República Bolivariana de help improve the environment for interna- Venezuela, and transition Europe.37 There can tional transactions (chapter 9). also be productivity spillovers to local suppli- ers and customers. Foreign multinationals Climbing the technology ladder. Techno- often help local suppliers by providing them logical progress is important for economic with new technologies and advice on how to growth. That does not mean every country improve quality and productivity so that they has to invent everything afresh—or that all can meet international standards. Studies in technological improvements have to be Indonesia and Latvia found that foreign entry cutting edge, pushing out the technological in downstream industries boosts the produc- frontier. For most countries adopting and tivity of local suppliers upstream.38 adapting available technologies is more fea- BOX 3.6 Trade liberalization in India—recent evidence India began reducing trade restrictions in the increase in total factor productivity. Firms that 1990s the Indian machine tool industry was pro- mid-1980s—eliminating quantitative were most ef�cient appear to have improved tected by tariffs of up to 100 percent and by restrictions on imports of industrial machinery their performance the most. Another study other restrictions. When tariffs were reduced to and reducing tariffs on capital goods by 60 per- found that investment and productivity around 15 percent in 1992, local �rms found cent. But its trade policies remained quite improved in industries close to the technologi- themselves unable to compete with more ef�- restrictive at the beginning of the 1990s. In 1991 cal frontier, but failed to improve in less techno- cient foreign producers. After several dif�cult the average tariff rate was about 83 percent, and logically advanced industries. years, some of the local �rms adapted to foreign only 13 percent of goods were importable with- Few �rms closed down following trade liber- competition by boosting their productivity. But out a license. By 1998 average tariffs had been alization.This might suggest that most �rms the �rm that led the recovery was not one of the reduced to 30 percent, and the range of goods managed to cope with the additional competi- �rms that had enjoyed protection for 40 years— importable without any restrictions was tive pressure, but it might also be because exit it was a fairly new producer, Ace Designers, that increased to 57 percent. was very dif�cult for �rms in India at that time. started operating only two years before the tar- Firm and industry studies that compare per- Although recent government reforms should iffs were reduced. formance in the 1980s with that in the 1990s speed up bankruptcy procedures, in 2003 they �nd that productivity increased for �rms took longer in India (11 years) than in any other exposed to competition from imports.The effect country with comparable data. Source: Aghion and others (2003); De Long (2003); was large.Topalova found that a 10 percent Looking at a speci�c industry brings out the Rodrik and Subramanian (2004); Sutton (2002); decrease in tariffs resulted in a 0.5 percent lessons clearly. From the 1950s until the early Topalova (2003); and World Bank (2004k). (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 67 BOX 3.7 Foreign locals—the role of emigrants and diaspora Emigrants, or diaspora, have been an important and white goods manufacturers. In Canada a dou- operate in both India and the United States.This source of investment and contacts for export bling of skilled immigrants from Asia was accom- has boosted the con�dence of overseas markets throughout history, with networks eas- panied by a 74 percent increase in Asian imports. investors in India’s potential. Several overseas ing some investment climate constraints and In the mid-1990s, when India started to open Indians who had reached high management building bridges between local and foreign �rms. its economy, it began to attract its 20 million positions in western multinationals helped to Overseas Chinese contributed 70 percent of compatriots living abroad.The Indian diaspora, convince their �rms to set up operations in India, China’s FDI over the past 15 years. By 1995, 59 per- second only to China’s, contributed 9 percent, or with Hewlett-Packard a prime example. cent of the accumulated FDI in China came from $4 billion, to the country’s FDI in 2002. Members Hong Kong, China, and Macao, with a further 9 of IndUS Entrepreneur, a networking group of Source: Biers and Dhume (2000);The Economist percent from Taiwan, China. Korean Americans Indian information technology entrepreneurs (2003c);The Economist (2001); Head and Reis (1998); were the bridgeheads for the successful penetra- and professionals, are funneling funds into star- Gillespie and others (1999); Kapur (2001); Li, Li, and tion into the U.S. market by Korean car, electronics, tups in India as well as hybrid companies that Zhang (1999); and Rauch and Trindade (2002). sible and can still improve productivity. Figure 3.6 Gaining access to technological innovations—key sources The Bank’s Investment Climate Surveys Low-income Middle-income con�rm the important role of competitive Purchasing machinery discipline in encouraging �rms to innovate (chapter 1). Developed or adapted with clients/suppliers For �rms a long way from the technolog- Hiring key ical frontier, the most cost-effective strategy personnel for technological upgrading is to tap tech- Developed or nologies developed elsewhere, through adapted internally trade and licensing.40 Several studies high- Licensing light the impact of machinery and equip- ment imports on productivity in develop- Universities and public institutions ing countries.41 Consistent with this, 33 percent of �rms in low-income countries Other and 49 percent of �rms in middle-income 0 20 40 60 countries reported that knowledge embed- Share of firms ded in new machinery was their most Note: “Other� includes transfers from parent companies, trade fairs, study tours, consultants, and busi- important source for technological innova- ness associations. Source: World Bank Investment Climate Surveys. tion (�gure 3.6).42 Another way to climb the technology Table 3.1 Who innovates? ladder is to encourage local R&D. Firms in High-income Developing developing countries perform only about countries countries 26 percent of the R&D (as a share of GDP) Patents granted by the U.S. Patent and Trademark Of�ce a 0.35 0 of those in developed economies (table Patents granted by the European Patent Of�ce a 0.15 0 3.1). This difference can be understood in part because high-income countries tend R&D personnel a 16.16 3.87 to have better intellectual property protec- R&D expenditure b 1.58 0.41 b tion, deeper credit markets, higher-quality R&D �nanced by the productive sector 0.74 0.13 research institutions, and more govern- R&D �nanced from abroad b 0.04 0.01 ment capacity to mobilize public R&D R&D performed by the productive sector b 0.96 0.25 expenditures.43 Low skill levels can also R&D performed by higher education b 0.34 0.12 hinder moves to more technology-inten- R&D performed by the public sector b 0.28 0.22 sive industries (chapter 7).44 a. Per 10,000 inhabitants. b. As a percent of GDP. Implementation constraints Source: Lederman and Saenz (2003). The priority-setting process is also influ- enced by implementation constraints— straints are discussed later in this chapter. both administrative and political (box 3.8). Political constraints often require both a Strategies for strengthening government high level of commitment as well as effec- capabilities to relieve administrative con- tive strategies for managing change. (c) The International Bank for Reconstruction and Development / The World Bank 68 WORLD DEVELOPMENT REPORT 2005 change? And how might such changes be BOX 3.8 Expanding the zone of feasible and desirable successfully managed? policy improvements Catalyzing change Proposed improvements to investment cli- market restrictions or distortions of vari- mate policies must meet three tests. Clearly, ous kinds provide examples. Options in Change tends to occur when something the proposed reform should be desirable, in zones B or C would be sound policy but shifts the incentives for maintaining the sta- the sense that it improves public welfare. It are not feasible in the short run, so reform tus quo. International experience illustrates should be administratively feasible, in the efforts in these areas would either be how a diverse range of factors can trigger sense that the government has the �nancial unsuccessful or, if implemented, would resources and technical expertise to imple- lack credibility. policy change even in the face of resistance ment the reform. And it must be politically Over time the goal is to expand the by bene�ciaries of the status quo. Those feasible, in the sense that the government is “sweet spot� by increasing the congruence triggers can include external shocks and able to secure suf�cient support to of the three elements.The sphere of desir- overcome resistance from those who prefer able policies can be expanded through pol- crises, technological change, new opportu- the status quo. icy innovation and learning. Administrative nities, new information and institutional At any point the menu of possible pol- feasibility can be enhanced by mobilizing competition, political change, and the ini- icy options that meet all three tests is lim- resources and expertise. Political feasibility tiative of policy entrepreneurs. ited—as shown in zone A in the �gure. can be enhanced by effective change man- Options in zone D are technically and agement, including strategies for building politically feasible but not desirable— public support. External shocks and crises. External shocks or crises can weaken the bargaining position of those who would normally oppose reform.46 Policy They can also create opportunities for desirability Policy reformers to exploit rapidly changing eco- desirability nomic or social conditions to justify or legit- B C B C imize reform. In Korea reducing cross-subsi- A A dies among chaebol subsidiaries, tried Administrative Political feasibility feasibility throughout the early 1990s without success, D Administrative Political feasibility feasibility was implemented only after the 1997–98 D �nancial crisis.47 In Slovakia a deteriorating �scal situation combined with high unem- Source: Adapted from Lax and Sebenius (1986). ployment led the government to pass a host of reforms in 2002, including collateral, tax, and labor reforms. Crises in a single sector can also prompt policy change. Power brownouts in Managing individual reforms the Philippines in the 1980s led to efforts to Land titling obviously differs from trade lib- engage the private sector in power delivery. In eralization, and improving the courts differs the U.S. coal industry, labor restrictions were from labor market reform. But a common reformed only when movements in oil prices issue across most areas of investment cli- put the future of mines in question.48 But mate reform is the need to deal with resis- crises do not always have this effect, and tance from those who have incentives to indeed the heightened social tensions associ- maintain the status quo. Resistance may ated with large-scale crises can overwhelm come from �rms or other interest groups policymakers. that bene�t from market restrictions or other special privileges. It may come from Technological change. Technological change of�cials who bene�t from informal pay- can threaten the interests of those commit- ments or other perquisites of of�ce. Even the ted to current technologies and provoke broader community may have a bias toward �erce resistance. Recall the Luddites in early the status quo when the implications of 19th century England who rioted against change are not certain,45 or where there are technological progress in the textile indus- other concerns about the reform process. try. But technological progress can also alter Overcoming this resistance is a key part the costs and bene�ts of policymakers main- of any strategy to broaden and accelerate taining current policies. For example, investment climate improvements. What advances in telecommunications technology has been learned about the catalysts for created new opportunities for introducing (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 69 competition, increased the costs of inertia public.53 In Peru the effort to reform land for those beholden to national monopolies, titles can be traced in part to the Institute for and so sparked a wave of telecommunica- Liberty and Democracy’s persuading the gov- tions reforms around the world in the 1990s. ernment and the wider community of the value of reform. Civil society groups are also New opportunities. New opportunities, such playing an active role in promoting improve- as access to new markets, can catalyze change. ments in investment climate policies and For example, the lure of EU accession altered behaviors. For example, Consumers Interna- the reform agendas of governments in Eastern tional and its national chapters champion the and Central Europe,49 and joining NAFTA did bene�ts of greater competition, and Trans- the same for Mexico. The prospect of joining parency International has emerged as an the WTO also had wide-ranging effects on the influential champion for greater trans- reform agenda in China. parency in government-�rm dealings.54 The level of resistance to any reform will New information and institutional competi- be influenced by what the bene�ciaries of tion. New information can shake assump- the status quo have at stake, and by their tions about the desirability of the status quo alternatives. Firms bene�ting from clien- and highlight the costs of inertia. Informa- telistic relationships with governments, tion that benchmarks a jurisdiction’s perfor- ineffective regulation, market restrictions, mance against other jurisdictions in terms of or other privileges that weaken the broader costs, productivity, or other measures can investment climate might be expected to spur change through its impact on local �ercely resist change. But this is not always prestige and concerns about future living the case. Concerns about corporate reputa- standards. Success from policy reforms in tions, about the long-term future of their neighboring jurisdictions can also have tan- businesses, or about the implications of gible effects. In China competition among more drastic government action can lead provinces for investment is spurring changes �rms to take a more enlightened view of across a range of policy areas,50 and similar their self-interest. This is evident in moves effects are evident in India. by �rms to burnish their reputations through corporate philanthropy, corporate Political change. Marked shifts in policy social responsibility initiatives, and forms of approaches can occur on a grand scale—as self-regulation. Similar considerations can with the collapse of central planning in the lead �rms to moderate their resistance to former Eastern bloc. They may also reflect a reform and even to cooperate with reform- changing social consensus, as when the ers to develop workable solutions. emergence of the merchant class in England drove the protection of property rights.51 A Communicating to build support growing middle class can also create a con- Communicating the costs and bene�ts of stituency against con�scatory, populist poli- alternative policy approaches is a central cies.52 Political transitions and changes of feature of successful reforms across most leadership also provide reformers with a areas of the investment climate. Indeed, a fresh mandate and an interest in differentiat- study of senior of�cials and civil society ing their policies from those of their prede- representatives from 60 developing and cessors. In Colombia, a second round of transition economies cited the public’s poor labor reforms, after having been defeated in understanding of economic reform as a key 2000, was implemented in 2002 under a new obstacle to success.55 government acting quickly to take advantage Gathering and disseminating information of political support. that benchmarks a country’s performance or that analyzes the costs and bene�ts of Policy entrepreneurs. Individuals identifying reform—including the costs of not reform- and promoting policy changes are often ing—can build public awareness and under- found within government—and in places standing of reform. It can also help mobilize a that have the ear of the government or the broader range of support, including citizens, (c) The International Bank for Reconstruction and Development / The World Bank 70 WORLD DEVELOPMENT REPORT 2005 consumers, and groups of smaller entrepre- Buro de Crédito undertook a campaign to neurs who would bene�t from change. Build- increase consumer awareness by placing the ing public awareness and support can also regulatory framework on their Web sites reduce the risk of later policy reversal and and listing the rights of consumers in a sim- thus enhance the credibility of the reform, ple and accessible way. As part of its judicial increasing the likely investment response reforms, Georgia launched a comprehen- (chapter 2). sive communication effort to educate the The most effective form of communica- public about newly acquired rights, increase tion depends on the issue, the society, and the trust in the system, and help users navigate groups that need to be reached. In Tanzania a the courts.56 song highlighting the case for privatization became a popular favorite. In Uganda radio Engaging stakeholders talk shows and plays in local dialects were Early consultation with key stakeholders, important. In Peru television commercials including potential winners and losers, on and public ceremonies at the delivery of land proposed changes can help validate assump- titles were the main channels. In Lesotho and tions behind the proposed improvement. It the República Bolivariana de Venezuela comic can garner suggestions on how proposals books reached a wide audience. In post-con- might be �ne-tuned to lead to better out- flict Bosnia and Herzegovina, the Bulldozer comes or easier implementation. It can also Initiative came up with a brand name and reduce the uncertainty �rms face when deal- used a range of communication devices, ing with changing policies and regulations— including the staging of symbolic events. and thus elicit a faster and stronger invest- Apart from building support, communi- ment response. Broad consultations can also cation campaigns can educate the public allay concerns that favored groups might about the reforms and help change public exercise disproportionate influence in policy- behavior. Educating �rms, consumers, and making processes, thus enhancing the trans- other groups about their rights and the parency and public acceptance of reforms. measures to uphold them is part of the The form and structure of consultation process. In reforming credit rating agencies can vary. In Vietnam reforms to simplify in Mexico, the �nancial authorities and the business registration involved consultations with private sector associations, domestic business groups, lawyers, the media, and BOX 3.9 The Bulldozer initiative in Bosnia and members of the National Assembly. In Pak- istan business registration reforms were Herzegovina designed and approved after a consultative Bosnia and Herzegovina launched the Bull- sive dialogue between the Bulldozer Com- process that involved circulating and dis- dozer Initiative in 2002 to involve the pri- mittee and the Council of Ministers and cussing draft rules with various chambers of vate sector in reforms. A reform coordina- Regional Governments. Once the reform is commerce, industry, professional bodies, and tion unit invited 30 local associations to designed, the Committee becomes an help in proposing, evaluating, and re�ning implementation watchdog. A biannual pub- the public. In Peru’s land reforms, urban set- reforms. Among them were regional busi- lication informs the public of progress, tlers were consulted through public assem- ness associations, municipal associations including scores for each reform. blies to inform them about the method and of entrepreneurs, the Employers’ Confeder- The initiative has helped to reduce sig- ation, the Women’s Business Network, the ni�cantly the burden of bureaucratic proce- schedule of land formalization programs and Micro-Credit Network, and the Association dures on �rms. It halved the number of to elicit their views. In Latvia reform priori- of Honey and Bee Production—all mem- steps to register FDI, expedited customs ties and an action plan were developed bers of the Bulldozer Plenary Committee. clearance procedures, bridged the through consultations with business associa- A group of lawyers and economists constituency gap by training and empower- evaluates proposals. Each proposal is sub- ing local advocacy groups, and established tions and a wide range of inspectorates. In jected to a cost-bene�t analysis, and indus- mechanisms for civic participation in gov- China, Hangzhou municipality recently try experts are invited to comment on ideas ernment. In June 2003 it established established a hearing system, inviting stake- before taking the reform to the next stage. regional Bulldozer committees, all voluntary holders and the public to express their views This way no single �rm can exploit the and self-�nanced. process to serve its own interests. on reform proposals.57 In Bosnia and Herzo- The proposed reforms are then submit- govina, the Bulldozer initiative includes ted to the government, opening an inten- Source: Herzberg (2004). grassroots involvement in identifying, evalu- ating, and monitoring reforms (box 3.9). (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 71 Engaging with prospective losers from compensation for utilities disadvantaged by reform—a group unlikely to remain silent changes in the regulatory environment in any event—is also important They can came from a levy imposed on consumer provide feedback on the details of the pro- tariffs. Reform programs can sometimes be posed reform, and engaging them construc- designed so that �rms disadvantaged by one tively may facilitate implementation. Partic- reform (liberalizing trade) bene�t from ularly if some workers stand to be others (improving business regulation). disadvantaged by a reform, early and con- When compensation is proposed, a com- structive engagement can mitigate any neg- mon concern is that governments might be ative social impacts (chapter 7). In South held hostage by the affected group, who use Africa the government provided funds and their resistance to reform to extract larger training programs to help trade unions payments. Mechanisms for arbitrating dis- become more effective interlocutors in the putes can reduce the incidence of strategic dialogue on privatization. behavior, as can benchmarks or principles derived from experience in other countries. Compensating when appropriate When state-owned enterprises are restruc- Maintaining momentum tured or privatized, it is common to give Investment climate improvements are a some of the shares to employees and to pro- process, not an event. Given the breadth of vide severance, pension, retraining, or other the agenda, and the need to review policies support to ease the adjustment to new regularly, many countries are creating sup- employment. Special mitigation measures for porting institutions to help with speci�c workers can also be adopted when particular tasks and to sustain progress through industries are undergoing signi�cant restruc- changes in government. Those institutions turing, particularly if effective economywide take many forms, but perform one or a safety nets are not yet in place (chapter 7). combination of four main functions: The case for compensating �rms affected • Facilitating consultation by policy changes tends to be different. If a proposed reform would violate property or • Facilitating coordination contractual rights, failing to compensate • Reviewing existing laws and policies can chill the investment climate—as recent • Reviewing new policy and regulatory expropriations in Zimbabwe show (chapter proposals. 4). When no speci�c rights are affected, arguments for compensation involve more Facilitating consultation judgment. Firms tend to be compensated Many governments have created special when they are a small group in society and structures to facilitate ongoing dialogues with the reform would disrupt their legitimate representatives of stakeholders. To be effec- expectations. For example, investors in Sin- tive, these structures should encourage the gapore’s privatized telecommunications free flow of information, build trust among company were compensated when the gov- participants, and assist in framing solutions. ernment shortened the promised period of It is particularly important that they reflect exclusivity.58 Power utilities in the United the diversity of interests affected by invest- States were compensated when the transi- ment climate reforms and not merely tion to a competitive market “stranded� entrench elites. A high level of transparency some of the assets built under a previous in their operation—such as the regular publi- regulatory regime.59 Compensation is less cation of reports—can also increase public common when all or most �rms in society con�dence in reform programs. are affected by a change seen as a normal The scope of representation varies widely risk of doing business—such as changes in (table 3.2), as do their mandates. Some look taxes or the introduction of a new competi- at policymaking economywide while others tion law. focus more sharply on private sector issues. Compensation need not always involve Many of the latter have a mandate that goes cash. In the United States, for example, beyond dialogue and includes identifying (c) The International Bank for Reconstruction and Development / The World Bank 72 WORLD DEVELOPMENT REPORT 2005 Table 3.2 Consultative forums dealing with investment climate issues—some illustrations Civil Government Business Unions Legislators society Donors Economywide focus Latvia—Tripartite Cooperation Council ✔ ✔ ✔ South Africa—National Economic Development and Labor Council ✔ ✔ ✔ ✔ Papua New Guinea—Consultative Implementation and Monitoring Council ✔ ✔ ✔ ✔ ✔ Private sector issues Vietnam—Private Sector Forum ✔ ✔ ✔ ✔ Uganda—Private Sector Foundation ✔ ✔ ✔ ✔ Pakistan—Workers and Employers Bilateral Council ✔ ✔ ✔ Singapore—Competitiveness Council ✔ ✔ Source: World Bank staff. bottlenecks, building consensus, recom- reforms and help overcome resistance from mending policy approaches, and monitor- agencies that may have a stake in maintain- ing progress of reforms. Latvia and Turkey ing the status quo. illustrate common approaches (box 3.10). Forums for consulting with external stake- holders can contribute to policy coherence Facilitating coordination when led by senior policymakers. But mecha- Responsibilities for investment climate pol- nisms are also often needed within the gov- icy issues are often distributed among sev- ernment. This may take the form of high-level eral government ministries and agencies, cabinet committees or even the establishment and often across tiers of governments as of a dedicated ministry. For example, coun- well. Fostering coordination between rele- tries acceding to the EU often created min- vant agencies can be important to deal istries for Europe to foster coordination of effectively with issues of common interest individual reform initiatives across ministries. and to promote policy coherence. Central In Poland that task was given to a Committee leadership can also help give impetus to for European Integration.60 More day-to-day coordination may be undertaken by the technical secretariat to the consultative forum or the coordination BOX 3.10 Consultative mechanisms in Latvia and Turkey committee. In 2000 Vietnam established an Many countries have created dedicated Both bodies have clearly de�ned objec- Inter-Ministerial Steering Group on Enter- structures to facilitate an ongoing dialogue tives and mandates.Their tasks cover a prise Law Implementation to support the with stakeholders on investment climate broad spectrum of issues with a view to ongoing implementation of its reform pro- improvements.The approaches in Latvia developing concrete proposals and strate- and Turkey illustrate some of the key gies for ongoing reform.They are usually gram (box 3.11). features. managed by technical committees.Turkey Fostering policy coordination between In Latvia the Steering Committee for has nine committees, and Latvia started national and subnational governments can Improvement of the Business Environment with four, but the number and focus change be tricky politically, but also raises other reports to the Minister of Economy. In Turkey with the needs and concerns of business. the Coordination Council for the Improve- Both bodies help to design and imple- issues. As China and India show, institutional ment of the Investment Climate reports to ment reforms.Turkey’s Council helped competition between subnational govern- the Undersecretariat of the Prime Ministry. design laws on recruitment of foreign per- ments can be a source of strength for the Both bodies comprise representatives from sonnel, FDI, company registration, and labor. investment climate by fostering policy inno- key ministries, as well as from associations of It is also engaged in reforms for customs, local �rms, exporters, and foreign investors. licensing, intellectual property rights, and vation and providing a check on arbitrary In both countries the bodies are served by a land acquisition. Latvia’s Committee government behavior (chapter 2). But some secretariat responsible for the daily work contributes to implementing ongoing leg- coordination may be desirable to address and for monitoring reforms—in Latvia, the islative and procedural reforms of inspec- spillovers across jurisdictional boundaries. In Business Environment Improvement Unit at tions, registration, taxes, customs, land the Latvian Development Agency; in Turkey, acquisition, and construction. Mexico, for example, procedures for state and the General Directorate for Foreign Invest- Source: Coolidge, Grava, and Putnina (2004) municipal governments to make regulations ment in the Treasury. and www.yased.org.tr. on road freight compatible and complemen- tary are being improved. (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 73 Reviewing existing laws and policies BOX 3.11 Shepherding investment climate improvements Most distortions in the investment climate in Vietnam stem from existing laws and policies. To sus- Vietnam began its transformation from a cen- lished an Inter-Ministerial Steering Group tain an ongoing process of policy review trally planned to a more market-oriented on Enterprise Law Implementation, chaired and reform, many governments are creating economy in the late 1980s. Despite many by the Minister for Planning and improvements, particularly in opening to FDI, Investment.The steering group, continuing institutions with a mandate to more sys- there was a cumbersome, overlapping, and to improve interagency coordination at the tematically review such arrangements and inconsistent regulatory environment for the center, recently exhorted state agencies to recommend reforms. domestic private sector. “change their management mindset and This role may be given to the technical To advance the needed reforms, of�cials put themselves in the shoes of enterprises.� worked with a broadly based business asso- Local authorities seem caught between secretariats of consultative or coordinating ciation (the Vietnam Chambers of regaining their discretionary powers over bodies. For example, Thailand’s National Commerce and Industry) and a team in the business registration (often for personal Competitiveness Committee and Singapore’s Central Institute for Economic Management gain) and streamlining procedures to Committee on Competitiveness have man- within the Ministry of Planning and Invest- attract new businesses to locate within their ment—the technical “champions� of the geographic areas. dates to study constraints on competitive- reform. In January 2000 a new Enterprise A recent survey of �rms noted a “return ness and to make speci�c recommendations. Law was passed to facilitate the entry of of troublesome and cumbersome unwritten Thailand’s committee is chaired by the prime new �rms, protect businesses from bureau- procedures among various local minister, with the National Economic and cratic interference in business operations, authorities.�Vietnam thus shows that con- increase flexibility to expand business oper- tinuing vigilance is often needed to ensure Social Development Board as its secretariat. ations, and improve corporate governance. that reforms take deep roots. It has undertaken assessments of several sec- Recognizing that passing the law was tors of the economy, including handicrafts, only the �rst step, the government estab- Source: Mallon (2004). tourism, and software, and brought several sector-speci�c and economywide issues to the attention of the government: one-stop are also often given a mandate to act as shopping for international investors, infor- champions of reform in their particular mation about laws and regulations, and the areas (chapter 5). skill levels of the workforce.61 Experience with dedicated reform champi- Sometimes the body has a broader man- ons in low-income countries remains limited, date. For example, Australia’s Productivity but there have been successes. For example, Commission focuses on providing detailed Senegal created a Growth and Competitive- analyses of particular areas of policy ness Review Group to identify policy and reg- referred to it by the government. A strong ulatory constraints to investment and compet- reputation for rigorous and independent itiveness and to formulate and implement work, coupled with effective consultation remedial measures (box 3.12). with stakeholders, has allowed it to exercise signi�cant influence. Japan’s Regulatory Reviewing new policy Reform Committee, reporting to the prime and regulatory proposals minister, has responsibility for coordinating Governments also need to ensure that new the implementation of a broad deregulation policy or regulatory proposals do not under- plan.62 In Mexico an Economic Deregula- mine the investment climate by introducing tion Unit was created in 1988 to oversee unjusti�ed burdens or other distortions. A improvements to business regulation. common response in Organization for Eco- Among other reforms, it proposed disman- nomic Co-operation and Development tling price controls, deregulating the trans- (OECD) countries has been to establish port sector, and streamlining the standard- processes for regulatory impact assessment. ization process. In 2000 it was transformed Proposed laws and regulations are subjected into the independent, nongovernmental to a quantitative assessment of their costs Regulatory Improvement Commission and bene�ts, with the information made (COFEMER), maintaining broad formal available to legislators and other policymak- oversight powers for the analysis of federal ers. These processes help to ensure proposals regulations and working with subnational reflect an economywide perspective. The governments to reduce red tape. Competi- additional scrutiny involved can also act as a tion and investment promotion agencies check on rent-seeking. (c) The International Bank for Reconstruction and Development / The World Bank 74 WORLD DEVELOPMENT REPORT 2005 In the United States some 60 percent of tise and promote consistent assessments, regulations are changed as a result of review but are often seen as too intrusive on the by the Of�ce of Information and Regula- prerogatives of line ministries. Delegating tory Affairs. Variations of these arrange- responsibility to line ministries can help to ments are in place in 22 OECD countries get their buy-in to the process, but doing so and in some upper-middle-income coun- without a clear framework can lead to dis- tries in Eastern Europe, Latin America, and appointing results. In Ghana, for example, Asia.63 In Mexico the review process is sup- no ministry was really in charge of policy ported by COFEMER, which reviewed and regulatory reviews. Instead, each pro- almost 1,500 regulations between 2000 and duced its own checklists, expressing differ- early 2003.64 In Korea a regulatory review ent preferences in what were not much committee reviewed nearly 3,000 regula- more than qualitative assessments.68 tions between 1998 and 2002, declining 387 Bulgaria’s review process had similar draft regulations and returning 1,157 to weaknesses until recently, with each agency sponsoring agencies for revision.65 The performing different types of evaluations, question is whether such impact assess- using different accounting methods and dif- ments can work in lower- income countries. ferent benchmarks, and publicly releasing Strong political commitment is essential, different amounts of information. The and without it schemes can disintegrate in reviews did not have a perceptible impact on any country. Technical capacity can be more legislation until uniform review criteria and of a constraint in low-income countries, methods were devised.69 In Lithuania, by although drawing on the expertise of local contrast, assessment for all draft legislation universities or other entities can often aug- was mandated under the leadership of the ment this.66 For example, Bulgaria’s regula- presidency. Reviews are undertaken by the tory review processes bene�ted from collab- sponsor of the legislation in consultation oration with a not-for-pro�t think tank.67 with those affected by the proposed policy Questions of institutional design can be changes. Summary assessments accompany thornier. There is a tension between creat- all draft legislation and are reviewed at inter- ing a central entity with the autonomy and ministerial, sectoral, and cabinet levels, any of expertise to take an objective view of regu- which can return the legislation to the spon- lations and creating a process that is ade- sor with a list of requested improvements.70 quately nested in the government’s day-to- Mechanisms and processes of the kind day policymaking and administrative discussed here can help to maintain structure. Independent central review units momentum, but they depend for their suc- can help to leverage scarce technical exper- cess on high levels of political commitment and on being credible to stakeholders. They also bene�t from ongoing processes to BOX 3.12 The evolution of a reform champion in Senegal strengthen capabilities within government. Senegal’s Growth and Competitiveness In 2000 the Group’s functions were inte- Review Group was created by presidential grated into a new Investment Promotion and Strengthening capabilities decree in 1993 to identify policy and regula- Major Projects Agency (APIX), directly Investment climate improvements differ in tory constraints to investment and competi- attached to the President’s Of�ce. APIX was tiveness and to formulate and implement directed to identify and support investors, their demands on resources, expertise, and remedial measures. facilitate the restructuring of the private sec- information. Many do not demand much Established as a coordinating body, the tor, simplify administrative procedures, and from the budget—and improving eco- Group also consults broadly with represen- implement strategies for the development of tatives of government, private sector orga- priority sectors such as tourism and building nomic growth can increase the tax revenues nizations, labor unions, universities, and the and civil engineering works. It established a to governments. All governments, however, media. It set up committees to review one-stop shop for processing all procedures have to improve the quality of their civil domestic competition issues, export and for the registration of change of status of a services and the quality of the information investment promotion, labor–management business, reducing the amount of time relations and labor regulation, and required for the registration to operate under available to guide and administer reforms. transportation costs. It took the lead in facil- the investment code from 60 days to 14. itating substantial improvements to the Expertise investment climate. Source: Diop (2003). See also www.apix.sn. Creating a skilled, professional, and account- able civil service can bene�t all areas of the (c) The International Bank for Reconstruction and Development / The World Bank Tackling a broad agenda 75 investment climate. In some areas of invest- response by �rms to particular policy ment climate policy there is also a need to changes, governments need access to reli- draw on more specialist expertise that remains able data on the operation of their private scarce in many countries. Examples include sectors. Consultation processes can be one areas of regulation and aspects of tax adminis- source of information, but there is no sub- tration. The skills, credibility, and effectiveness stitute for more objective and consistent of staff can have a big effect on the policy envi- sources of data. Data on even basic mea- ronment faced by �rms. sures, such as the level of private invest- To make it easier to recruit and retain staff ment, are lacking or inadequate in many with the requisite skills, many countries are developing countries. Similar de�ciencies establishing more autonomous administra- exist in data from of�cial business registers. tive structures for these functions (chapter Designed to meet various purposes—tax 5). There is also growing experience in con- and social security collections—these data tracting-in or contracting-out some speci�c can provide powerful insights into the functions to outside experts, even in devel- dynamism of �rms. Greater standardization oped countries. A recent survey of regulatory and proper updating of business registry agencies for infrastructure across the devel- data—as Eurostat is doing for EU coun- oping world found that three-quarters of tries—can help governments monitor the agencies engaged consultants or other exter- evolution of the private sector and alert nal parties in regulatory tasks. In more than them to emerging policy issues. Introducing 90 percent of these cases, contracting-out or improving enterprise surveys—a stan- was found to also improve the competence dard tool in developed countries—can also of the regulatory agency.71 When local help. The surveys provide information on capacity is weak, entire functions can be con- investment, job creation and destruction, tracted out—such as customs administra- and productivity and output growth at �ne tion in Mozambique (chapter 5). Capacity levels of disaggregation. While many devel- building strategies are also being adapted to oping countries have enterprise surveys, the particular needs of specialist agencies, there are opportunities to improve the rep- including the formation of international net- resentativeness of samples, the standardiza- works of regulatory professionals (box 3.13). tion of structures, and the regularity of con- ducting them. Learning and information The need to expand government capabilities BOX 3.13 Networks of regulatory professionals in extends beyond technical expertise. Govern- infrastructure ments need to improve their processes for ongoing learning—including that from pol- Beginning in the early 1990s governments more than 350 participants. Beyond formal icy experiments abroad as well as within worldwide began embracing a new model training, these initiatives build direct for delivering infrastructure services. It networks of regulators to facilitate ongoing their own countries. Decentralization and involved improving the government’s capa- information sharing and mutual support. institutional competition have been sources bilities as a regulator of services delivered pri- Complementary regional initiatives of policy innovation and learning in coun- marily by private �rms. As part of this process, have since been launched in South Asia, tries including China and India—states and more than 200 autonomous regulatory agen- Africa, and East Asia.The South Asian Forum cies for infrastructure have been set up in for Infrastructure Regulation, established in provinces experiment with alternative pol- developing countries. 1999, offers training programs and other icy approaches, and successful approaches The International Forum for Utility Regu- learning and knowledge-sharing support to tend to be quickly emulated by other regions lation, established by the World Bank in regulators.The African Forum for Utility 1996, is an umbrella structure for learning Regulation, launched in 2000, provides a and, in some cases, by the central govern- and networking initiatives. Its �rst major ini- mechanism for sharing experiences and ment. In Peru land reform pilot projects in tiative was a two-week training program information on particular regulatory issues, the 1990s paved the way for a bolder focusing on the needs of regulators in water, and meetings focus on speci�c themes, national program. In Uganda efforts to electricity, gas, and telecommunications. such as strategies for engaging consumers Since 1997 more than 1,000 regulators from and other stakeholders. A similar regional improve business registration processes are 115 countries have attended the twice-a- initiative for utility regulators in East Asia beginning with a demonstration project in year program. A complementary program and Paci�c was launched in 2003. Entebbe (chapter 5). for transport regulators, launched by the To take advantage of these experiments, World Bank Institute in 1998, has reached Source: World Bank staff. and to track trends and monitor the (c) The International Bank for Reconstruction and Development / The World Bank 76 WORLD DEVELOPMENT REPORT 2005 and to sustain a process of ongoing Part I argued that improving government improvements. It looked at issues associated policies and behaviors shaping the invest- with setting priorities, managing individual ment climate is critical to spurring growth reforms, maintaining momentum, and and reducing poverty—and so should be a strengthening government capabilities. top priority for governments. The remainder of the Report looks at more Chapter 1 argued that the key is to detailed issues associated with the design and improve the opportunities and incentives for implementation of effective strategies to cre- �rms of all types to invest productively, create ate a better investment climate. jobs, and expand. This in turn requires efforts to reduce unjusti�ed costs, risks, and barriers • Part II examines lessons of experience in to competition. Chapter 2 focused on the delivering the basics—the foundations of basic tension that governments need to a sound investment climate—stability and confront in investment climate policymak- security (chapter 4), regulation and taxa- ing: While �rms play a key role in improv- tion (chapter 5), �nance and infrastruc- ing living standards in society, their policy ture (chapter 6), and workers and labor preferences can diverge from those of soci- markets (chapter 7). It reviews a rich body ety as a whole. Arbitrating these differences of international experience to highlight successfully requires governments to navi- opportunities for policy improvement in gate four sources of potential policy failure: all areas. rent-seeking, credibility gaps, lack of public • Part III looks at the possible role of mea- trust, and poor �ts between policy sures that go beyond the basics—selec- responses and local conditions. It outlined tive interventions (chapter 8) and the use lessons of experience in addressing those of international rules and standards challenges, highlighting the powerful role of (chapter 9). These measures can play a transparency. This chapter looked at practi- supporting role, but also raise special cal strategies for tackling a broad agenda. It challenges that warrant careful attention. argued that the key to accelerating and • Part IV concludes by looking at how the broadening improvements is to address international community might help important constraints facing �rms in a way developing countries improve the invest- that gives �rms the con�dence to invest— ment climates of their societies. (c) The International Bank for Reconstruction and Development / The World Bank Delivering the Basics II THE REPORT ARGUES THAT GOVERNMENTS SHOULD STRIVE to create a better investment climate for everyone by tackling unjusti�ed costs, risks, and barriers to competition. This part of the Report highlights PA RT opportunities for governments to improve their performance in delivering the basic foundations of a good investment climate. Chapter 4—Stability and security suggests measures that governments can take to enhance the security of property rights in their societies. Chapter 5—Regulation and taxation highlights the huge opportunities for improving approaches in these areas without compromising other social goals. Chapter 6—Finance and infrastructure shows how governments are getting better results through new approaches to the provision of these services. Chapter 7—Workers and labor markets outlines a three-pronged agenda for strengthening the connection between people and decent jobs to create a more productive and equitable society. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank Stability and security 4 Nothing so undermines the investment limited impact (box 4.1). An unstable or climate as the outbreak of armed conflict. insecure environment has its most tangi- Capital of all kinds—human, physical, ble effect on property rights, so this chap- chapter and social—is destroyed, investment dis- ter concentrates on the impact of insecure rupted, and resources diverted from rights on the investment climate and what growth-enhancing activities. Civil war, the government can do to make them more predominant form of warfare over the secure. past half century, has a particularly devas- Secure property rights link effort with tating impact on poverty and growth. By reward, assuring all �rms—small and large, one estimate, over the past 50 years the informal and formal, rural and urban—that typical civil war lasted 7 years and cut 2.2 they will be able to reap the fruits of their percent off the projected annual growth investments. The better protected these rate—at the end of hostilities GDP was 15 rights, the stronger the link between effort percent lower than it would have other- and reward and hence the greater the incen- wise been. A particularly severe civil war tives to open new businesses, to invest more can, in the short run, also reduce income in existing ones, and simply to work harder per capita in neighboring states by as (box 4.2). much as a third.1 New evidence con�rms how important Civil war and low income go hand in secure property rights can be. Recent sur- hand. The odds that a civil war will erupt in veys from Poland, Romania, Russia, Slova- low income states are 15 times greater than kia, and Ukraine show that entrepreneurs in a developed country. The poorer the who believe their property rights are country, the greater the risk of a nation secure reinvest between 14 and 40 percent being trapped in a downward spiral of vio- more of their pro�ts in their businesses lence and economic decay. A doubling of than those who don’t.3 Farmers in Ghana per capita income can halve the risk of civil and Nicaragua invest up to 8 percent more war. Accordingly, the poorer the country, in their land when their rights to it are the stronger the imperative to improve its secure.4 By contrast, to compensate for the investment climate to reduce the likelihood greater insecurity of property rights, of falling into a conflict trap.2 equity investors in �rms in some low- While peace is essential to unleash pro- income countries can require returns ductive investment, �rms require more much higher than those in �rms in devel- than this. They require an environment oped countries.5 with a reasonable level of political and Studies across a broad range of countries economic stability, and one where person- �nd that the more secure the rights, the nel and property are reasonably secure. faster the growth. They also show that even Political instability can create considerable modest improvements in security can uncertainty and risk for �rms, undermin- increase annual economic growth rates by ing the credibility of current laws and as much as one percentage point.6 No mat- policies (chapter 2). Macroeconomic sta- ter what factors are included in the analyses bility also plays a critical role, because and what measures of property rights secu- without it changes in other areas will have rity are used, all report a close connection 79 (c) The International Bank for Reconstruction and Development / The World Bank 80 WORLD DEVELOPMENT REPORT 2005 BOX 4.1 Macroeconomic stability and the investment climate A sound investment climate requires suf�cient instruments, and credit from overseas. Medium ico, Uruguay, and Peru over 100 percent a year. macroeconomic stability before microeconomic and small �rms are likely to be hardest hit. By 2001 inflation in all countries in the region policies will gain much traction. Low inflation, sus- The costs of macroeconomic instability can was under 15 percent, with the exception of tainable budget de�cits, and realistic exchange be high. Several countries in Latin America, the Ecuador at 38 percent. Countries have also low- rates are all key. Instability deters investment by region that experienced tremendous ered budget de�cits signi�cantly. Brazil and making future rewards more uncertain. It can also fluctuations in the 1980s, suffered absolute Mexico, having run double-digit de�cits, have undermine the value of assets. declines in GDP per capita, leading the 1980s to since seen periods of mild de�cits and surplus. High inflation and volatile real exchange be dubbed the “lost decade� in Latin America. The crises in East Asia and Russia in the late rates are two examples.Their effects are particu- Achieving stable macroeconomic policy was 1990s, while sharp and painful, have been larly harmful for those with �xed incomes, local the focus of much policy attention in the wake followed by recovery—with countries with currency–denominated assets, and few means of the oil crises of the 1970s and the debt and more flexible microeconomic conditions and of protecting themselves from declining related �nancial crises of the 1980s.There has better investment climates recovering faster. purchasing power.They also weaken the posi- been good progress.The extremely high infla- tion of creditors, making access to credit more tion in Latin America has been brought down. In dif�cult. Large �rms are more likely to have tools the 1980s Bolivia and Nicaragua experienced Source: World Bank (2003h); Easterly (2001); at their disposal to cope with these risks, includ- inflation of over 10,000 percent a year; Brazil and Hnatkovska and Loayza (2004); Desai and Mitra ing better access to dollar accounts, �nancial Argentina over 3,000 percent a year; and Mex- (2004); and Caballero, Engel, and Micco (2004). between growth and property rights secu- • Verifying rights to land and other rity. Indeed, the large number of studies all property reaching the same conclusion led one • Facilitating contract enforcement commentator to observe that the link • Reducing crime between secure property rights and growth • Ending the uncompensated expropria- has “withstood an unusually large amount tion of property. of scrutiny.�7 This chapter focuses on four measures Verifying rights to land governments can take to improve the investment climates of their societies by and other property enhancing the security of property rights: Providing secure rights reduces the risks of fraud and mistake in property transactions, thus allowing buyers, renters, lenders, and others wanting to acquire an interest in land BOX 4.2 Property rights reform in China: Even modest or other property to do so with con�dence progress can ignite a strong response that they will get what they bargained for. The reduced risks are evident in the differ- After China’s Maoist revolution, households In some parts of China it now runs as long were allocated farmland that they could not as 30 years. ence in price between titled and untitled sell, rent, or otherwise transfer.All production Improvements in agricultural productiv- land. The value of rural land in Brazil, belonged to the government, and periodically ity depend in part on investments that take Indonesia, the Philippines, and Thailand the authorities would reassign land in time to pay off: increases in the amount of increases by anywhere from 43 percent to response to their assessment of “need.� fertilizer applied, the number of wells Whether a household worked the land hard, drilled, and so forth. As the length of farm- 81 percent after being titled.8 For urban or hardly at all, did not affect its well-being.All ers’ land tenure increased, so too have land, titling increases the value by 14 per- households received an equal share of the investments to make that land more cent in Manila,9 by almost 25 percent in community’s total production.Effort was productive. Besides altering rights to land to divorced from reward, resulting in the stagna- better link effort with reward, Chinese poli- both Guayaquil, Ecuador,10 and Lima, Peru, tion of agricultural production in the 1970s. cymakers liberalized prices and took other and by 58 percent in Davao, Philippines.11 Since 1982 China has been granting steps to boost production.The combined Providing more secure rights to natural farmers greater rights to land. Initially, they effect of these reforms was to boost agricul- resources also fosters environmental stew- were permitted to sell anything they pro- tural output by 42 percent over 1978–84. duced in excess of a �xed amount due the Almost half this increase has been ardship (box 4.3). government each year. This was followed attributed to changes in land rights. by a gradual lengthening of the time they were allowed to farm the land. At �rst land Rights to land could be taken away every 3 years, but that Source: McMillan (2002), Lin (1992), and World Secure rights to land also encourage period has been progressively lengthened. Bank (2003m). investment. Farmers in Thailand with title invested so much more in their land that (c) The International Bank for Reconstruction and Development / The World Bank Stability and security 81 their output was 14–25 percent higher than those working untitled land of the BOX 4.3 Secure property rights and environmental same quality.12 In Vietnam rural house- stewardship holds with a document assigning clear When rights to natural resources are uncer- Another example of the relationship rights of control and disposition commit tain, those in immediate control often feel between secure rights and environmental 7.5 percent more land to crops requiring a they should “use it or lose it.� After all, if they stewardship involves land. Ethiopian farm- greater initial outlay and yielding returns are not sure the resource will be theirs ers are less likely to plant trees and build after several years than households with- tomorrow, why not take as much as possible terraces to protect against erosion—and today? more likely to increase the use of fertilizer out documentation.13 In Peru almost half Fisheries are often depleted because of and herbicides—if their rights to land are those with title to their property in Lima’s this use-it-or-lose-it syndrome. Each �sher- insecure. When the rights to gather squatter settlements have invested in man catches as much as he can as fast as he �rewood, graze animals, and otherwise use improvements, compared with 13 percent can, knowing that others are doing the the resources of neighboring forests are rec- same.The result: the stock of �sh is soon ognized, Kenyan communities keep careful of those without title.14 exhausted. Iceland devised a way to use watch to ensure that they are not overused. Titling can improve access to credit property rights to overcome this problem. Cross-country studies con�rm the when product markets, lending institutions, Each �sherman was awarded a quota of �sh close link between secure rights and envi- and the other elements of a �nancial infra- based on how much he had historically ronmental stewardship. One recent analy- caught.The �sherman was free to catch up sis of 53 developing countries concluded structure are present. Land ownership is an to that number of �sh each period or to that a modest improvement in the protec- important indicator of creditworthiness, trade some or all of the quota to others.The tion of property rights could reduce the and a registered title allows lenders to easily quota prevented over�shing while giving rate of deforestation in these countries by each holder an interest in ensuring the as much as one-third. verify ownership. Titled land is also future health of the �shery. Since the adop- accepted more readily as collateral. Lenders tion of a quota for herring, stocks have can determine whether others have an Source: Deininger and others (2003); Samuel increased, as has the catch. Quota schemes and Pender (2002); Pender and others (2001); interest in the property and thus assess the have produced similar results in New Mwangi, Ongugo, and Njuguna (2000); Norton Zealand and Nova Scotia, and Peru is experi- (2002); Gissurarson (2000); and Newell, likelihood of seizing the land if the bor- menting with a quota system as well. Sanchirico, and Kerr (2002). rower refuses to repay the debt. Farmers with secure title in Costa Rica, Ecuador, Honduras, Jamaica, Paraguay, and Thailand obtain larger loans on better to land spend an average of nine weeks more terms than those without. In Thailand time working off the farm than those with- farmers with title borrowed anywhere from out secure rights. 50 percent to �ve times more from banks Improving the security of property rights and other institutional lenders than farmers can raise important distributional questions with land identical in quality but without in society (box 4.4). But even the landless title.15 The bene�ts extend beyond farmers. poor can bene�t when rights are secure. In Peru residents of urban areas in Lima Owners with insecure rights are often reluc- that received title to their land have used the tant to rent their land, fearing that a tenant titled land as collateral to buy microbuses, may try to assert a claim to the property. build small factories, and start other types of small businesses. Lack of secure title— Figure 4.1 Not entitled? common in many countries (�gure 4.1)—is thus one more obstacle smaller entrepre- 100 neurs face when trying to �nd �nancing for living in untitled houses (percent) Estimated proportion of citizens their operations. Rural 75 Urban Securing rights contributes to a better investment climate in ways besides boosting investment and easing access to credit. 50 Owners with secure rights do not have to waste time at home guarding their property. In Peru those with title to their land work 25 outside the home an average of 20 hours more per week than those in the same 0 neighborhood whose land is not titled.16 In Egypt Haiti Peru Philippines Vietnam farm households with secure rights Source: De Soto (2000). (c) The International Bank for Reconstruction and Development / The World Bank 82 WORLD DEVELOPMENT REPORT 2005 BOX 4.4 The distribution of property rights As economic activity intensi�es, the returns Why do some groups reach agreement and those who can most ef�ciently exploit them at from establishing rights over a resource increase others not? One explanation is social capital— prices that reflect their value in a market econ- accordingly.The paradigmatic case involves the trust, norms, and networks that facilitate omy. Some privatizations have favored a select rights to the lands of the Labrador Peninsula, coordinated action. Social capital tends to be few, however, as have several recent mass priva- now part of Canada.These lands were home to higher when the community is smaller and tizations in Eastern Europe and the former otters, raccoons, and other fur-bearing animals, more homogeneous and when information Soviet Union. Indeed, one poll shows that 80 and as demand for pelts grew in 18th century about the effect of different solutions circulates percent of the Russian public believes that the Europe, the risk of overhunting increased appre- freely.These factors have been critical to com- mass privatizations of Russian enterprises in the ciably. But because the land was held munity-wide agreements on the use of 1990s were unfair. collectively, individuals had no incentive to cur- resources. And as World Development Report As Hobbes recognized, and several modern tail hunting. 2002 explained, the larger the community, the authors have recently tried to model, there is a Parceling the land out to extended families more ethnically and culturally diverse, the more dynamic at work in enforcing property rights. If solved the problem, giving each an incentive open to trade, and the greater the spread in lev- enough citizens accept (or at least acquiesce to) to restrict the taking of animals living on their els of income and wealth, the less likely that an the current regime, the government’s land to sustainable levels. Allocating rights to accord will emerge. enforcement resources can be devoted to bring- natural resources to groups, families, or indi- In an ideal world government would act as ing those who refuse to accept existing arrange- viduals continues to protect against deforesta- an impartial arbiter among competing interests, ments into compliance. Greater compliance pro- tion, over�shing, and other practices that per- nudging them in the direction of a mutually duces a virtuous circle. As more people believe manently damage the environment, as advantageous agreement while curbing oppor- others will respect the current regime, their described in box 4.3. tunistic behavior and ensuring that norms of jus- incentives to respect it increase too.Those who Changes in property rights regimes can cre- tice are observed. Governments rarely achieve would undermine secure property rights also ate strains within a community. Although the this ideal, however. Instead, as an analysis of the face ever higher levels of deterrence as govern- transition to a system of better-de�ned rights creation of rights to farmland in 23 countries ment brings more resources to bear on fewer increases the wealth of the community, some over three millennia shows, those in control of individuals.The corollary is that when large num- members will gain more than others, and some the governmental machinery often use their bers of citizens believe the existing distribution of may even be worse off. Less skilled hunters real- power to favor certain interests—aristocrats, rights is unfair, the resulting tensions can under- ized more from curbs on hunting in Labrador colonists, and others with political influence—at mine secure rights. Government is not helpless in than the more skilled.The problem is political: the expense of the small peasant farmer. the face of widespread discontent with the distri- �nding acceptable mechanisms to allocate the A more recent example is the privatization bution of rights, however (see box 4.10). gains and losses while maintaining the advan- of state property. As with land rights, the sale of Source: Demsetz (1967); Libecap (1994); Putnam, tages from more clearly speci�ed rights.The state-owned property in an ideal world would Leonardi, and Nanetti (1993); Ostrom (2000); bargaining to reach a solution can be dif�cult, produce a solution advantageous to all, transfer- Binswanger, Deininger, and Feder (1995); Transition and an impasse is always possible. ring factories and other productive assets to (2003); and Hoff and Stiglitz (2004). Many will thus let their land lie fallow rather �ts, that price should not be inflated because than risk leasing it to a household wanting of red tape or demands for “unof�cial� pay- to work it, or they will rent it only to those ments by registry staff. Maintaining monop- they know well. In either case productivity olies over surveyors, notaries, and other pro- suffers because the rural poor—less likely to fessionals who prepare the necessary be a part of the owner’s social circle—are documentation can also boost costs and so denied access to land.17 In the Dominican deter registration. In Russia surveyor fees Republic the effect of securing owners’ equal to two years of the minimum wage rights increased the number of plots leased keep many from registering their property.19 out by 21 percent, with 17 percent more In Peru the key to titling urban land belong- households obtaining access to land. Poor ing to the poor and near-poor was breaking households realized the largest share of ben- the notaries’ monopoly over drafting deeds. e�ts. The percentage of poor tenants increased by 40 percent, and the actual area Maintain an effective titling program. Gov- rented to them grew by 67 percent.18 ernments can improve the security of land Securing rights bene�ts both individual title by maintaining an ef�cient land reg- landholders, through investment incentives istry, something becoming easier with and credit access, and the community at advances in computer technology (see box large, through its impact on growth and 2.16). Even this straightforward measure poverty reduction. While the price govern- can face challenges. ments charge for titling services should First, the cost of issuing initial titles can reflect this mix of public and private bene- be signi�cant, particularly when a large per- (c) The International Bank for Reconstruction and Development / The World Bank Stability and security 83 centage of land is untitled, as is the case in many developing countries. Conflicting BOX 4.5 Thailand’s 20-year program to title rural land claims may need to be resolved, boundaries In 1982 the Thai government began a 20- mitment by the government and key determined, and accurate maps drawn. In year project to title and register farmland stakeholders to project implementation. recent World Bank-supported projects the throughout the kingdom.The aim? Enhanc- • A strong policy, legal, and institutional unit cost of a title or �rst registration ing farmers’ access to institutional credit framework was in place for land adminis- and increasing their productivity by giving tration. ranged from $9.90 in Moldova to $24.40 in them an incentive to make long-term Indonesia to $1,354 in Latvia.20 investments. • The project built on earlier efforts to issue documents recognizing holders’ Second, it is typically much easier to pro- Just over 8.5 million titles were issued rights to their land. vide title to land where de facto ownership during the life of the project. Along with those issued outside the project, the num- • Registration procedures developed by rights are recognized in the community than ber of registered titles increased from 4.5 the Department of Lands were ef�cient where titling may encroach on the claims of million in 1984 to just over 18 million by and responsive to public demand. others. The programs in Peru and Thailand September 2001. Studies conducted during • The public had con�dence in the land the project show that it met both its objec- administration system and actively par- began by issuing titles to residents whose tives: titled farmers secured larger loans on ticipated in the reform process. rights were essentially uncontested, either by better terms than untitled farmers, and pro- • The interests that can complicate other individuals or by the government. As ductivity on titled parcels rose appreciably. projects in other countries—public support for the projects grew, and experi- The success in Thailand is attributed to notaries, private lawyers, and private ence in administering them increased, the several factors. surveyors—were not present. titling effort was extended to areas where the • There was a clear vision for the project, a long-term plan to achieve it, and a com- Source: Burns (2004). issues were more complex (box 4.5). Even when there are no rival claimants, titling reforms can run into resistance. Land registry personnel often oppose moderniza- have protected holders from eviction while tion, either from inertia or the loss of the government considers options to address opportunities to collect side-payments urban landlessness.25 Other examples where from registrants. In Russia, Ukraine, and secure rights were achieved without a full- other former socialist countries, opposition blown title come from India and Vietnam. to rural land titling has come from the managers of collective farms. Titling Foster competition among service providers. requires breaking these farms up into indi- Whatever the means chosen to enhance the vidual parcels, threatening managers’ jobs security of property rights, governments and income and weakening their power need to be sure that consumers are not over- over the farmers.21 charged for the required services. Experience in Australia, the Netherlands, and the United Consider alternatives to full-blown titling. Kingdom shows that transaction costs can be A large titling program is costly and reduced, without sacri�cing the quality of requires many trained professionals. Before service, by introducing competition into the initiating a program, governments should provision of services associated with land consider whether their policy objectives can transactions (box 4.6). be realized through measures short of pro- viding a full legal title. Indeed, experience Title to other property around the globe shows that a diversity of Titling automobiles, equipment, machinery, tenure options can facilitate access to land. and other valuable forms of “movable� In Niger, security of rights was realized property can provide bene�ts similar to through a simple, community-based regis- titling land. As with land the registration can tration scheme.22 In Honduras simple title facilitate access to credit. Lenders can verify documents that lenders can hold while the ownership and determine whether others loan is outstanding have been enough to have already either lent against the property improve the flow of formal credit to small or may have some other interest that would farmers.23 In urban areas, too, interim mea- make it dif�cult to foreclose on the property sures short of full titling can begin to meet in the event of default. Experience in residents’ needs for greater security.24 Indonesia and Romania demonstrates the Botswana has issued use certi�cates that value of movable property registries. Both (c) The International Bank for Reconstruction and Development / The World Bank 84 WORLD DEVELOPMENT REPORT 2005 or a �rm that specializes in leasing, a BOX 4.6 De-monopolizing property transaction monthly rental fee. professionals The extent to which �rms can take advantage of leasing depends largely on In early 1984 the British government initial quotes for fear of losing the business. how secure the lessor’s rights in the prop- announced it was considering ending the Despite the price reductions, consumers legal profession’s 180-year monopoly on buying or selling real estate after de- erty are. If, in the event of nonpayment, the providing the services required to buy or monopolization reported the same satisfac- lessor must go to great lengths and expense sell real estate. Within months reports tion with the services, if not more. to reclaim the property, the lessor will began circulating that prices for conveyanc- Similar deregulation initiatives have ing services were falling. Spurred by favor- brought the costs of land transactions require that �rms leasing the equipment able responses from consumer down in the Australian state of New South provide large cash advances or other forms organizations, and over the vigorous objec- Wales and in the Netherlands. Conveyanc- of guarantees.27 In Egypt it can take years to tions of the organized bar, the government ing fees in New South Wales fell an average repossess leased property, in Lebanon any- went through with its proposal to open the of 18 percent in the mid-1990s after the market to non-lawyers. Starting in October market was opened to non-lawyers, saving where from nine months to two-and-a-half 1987 anyone passing a rigorous licensing the community close to A$100 million in years. In Tunisia, by contrast, it takes three examination was permitted to offer fees. In the Netherlands abolishing the pro- months at most to obtain a court order per- conveyancing services. In accord with an fessional monopoly held by real estate mitting repossession.28 Partly as a result of earlier reform, both legal professionals and agents lowered costs and provided licensed conveyancers were allowed to consumers with a greater selection of ser- lessors’ ability to obtain these orders advertise rates and services. vices.These results are consistent with a quickly, estimates from 2000 show leasing Competition brought prices down with- broader study of professional regulation to be far more prevalent in Tunisia than out sacri�cing quality.The mean price conducted by the European Commission. It either Lebanon or Egypt (�gure 4.2). charged for conveyancing services fell found that less regulation of lawyers, almost 10 percent between 1983 and 1986, notaries, and other legal professionals One type of property of growing impor- a period when housing prices, and thus the enhances consumer welfare without com- tance is intellectual—the patents, copy- fees under the old schedule, rose promising other values. rights, trademarks, and other legally created signi�cantly. Consumers asking for an esti- Source: Domberger and Sherr (1989); Baker rights to enjoy the fruits of one’s intellectual mate of the cost of the services realized sig- (1996), Philipsen (2003); and European Com- ni�cant savings because providers cut their mission (2004b). efforts. Because intellectual property is intangible, and can be transported easily across national boundaries, international Figure 4.2 Leasing activity is more agreement on the enforcement of these countries created one in 2000. By the end of prevalent in Tunisia than in Egypt or rights is critical to its protection (box 4.7). Lebanon, thanks to laws facilitating 2003 Romania had recorded 200,000 repossession entries, while in Jakarta alone creditors sought in 2003 to register 12,000 interests in Facilitating contract enforcement 16 vehicles, machinery, and other items cov- Property rights are more secure, and more ered by the new law.26 According to the valuable, when the costs and risks of Bank’s Doing Business Project, the time exchanging them are low. Delays or uncer- Percent of gross fixed investment 12 Tunisia required to register property ranges from 3 tainties in the enforcement of exchange days in Lithuania, to 274 days in Nigeria, erode the value of property rights and in plant & equipment and nearly 1,000 days in Croatia. diminish the opportunities and incentives The easier it is for banks and other �nan- to invest. In an ideal world all contractual 8 cial institutions to recover the property in exchanges would occur without a hitch. the event of nonpayment, the more willing Neither party would ever fail to deliver the they are to lend (chapter 6). A common promised good or service or be short on the 4 method for lowering the costs of foreclo- quality or quantity promised. It is easy to sure is for the lender to retain title to the see why such a world would have an extra- Lebanon property. The lender simply leases it to the ordinarily favorable investment climate. Egypt 0 borrower for a �xed time at an agreed upon Firms could commit to long-term, complex Source: World Bank (2002b). price, often payable monthly. Not only can commercial relationships with perfect such leasing agreements simplify foreclo- strangers, con�dent that the other side sure proceedings, but they are also an would faithfully uphold its end of the bar- important means for broadening access to gain over as many years as the contract lasts. �nance. Small and medium �rms do not It is also easy to see why such a world need to accumulate the funds required to doesn’t exist. Anytime the parties do not pay buy machinery or other assets outright but with one hand and take with the other, there can obtain the equipment by paying a bank, is a risk that the party to perform later will (c) The International Bank for Reconstruction and Development / The World Bank Stability and security 85 BOX 4.7 Intellectual property rights: The ongoing debate Inventors and authors often require an incentive these products to strengthen their enforcement attest. Consumers usually suffer the most when to develop innovative products, as has been rec- of intellectual property rights to prevent the laws protecting trademarks and brand names ognized since at least the 4th century BCE.Today dilution of incentives to innovate. While the are not vigorously enforced. the incentive is provided by granting creators of debate continues, four recent developments are Fourth, there is a trend toward addressing new inventions, software programs, or other helping to change its terms. intellectual property issues one by one, helping products a patent, copyright, or other similar First, more �rms in more developing coun- to identify areas of agreement and �nd right to their creation. An idea of how powerful tries are now producing innovative products and common ground on points of difference. An this stimulus can be comes from a recent analy- thus have a direct stake in the protection of agreement at the WTO ministerial meeting in sis of spending on research and development by intellectual property rights. In Brazil and the November 2001 reflects developing countries’ American �rms. A modest increase in the value Philippines short-duration patents have helped need for access to medicine. Discussion is also managers expect to realize from patenting new domestic �rms adapt foreign technology to local under way on policies that would give manufac- products was found to boost R&D by anywhere conditions, while in Ghana, Kuwait, and Morocco turers of patented goods greater flexibility to from 11 percent in the biotech industry to 8 per- local software �rms are expanding into the inter- sell at lower prices in poor countries than in cent in the pharmaceutical industry to 7 percent national market. India’s vibrant music and �lm wealthier ones. in the chemical industry. industry is in part the result of copyright protec- How nations recognize intellectual property This stimulus comes at a price. Intellectual tion, while in Sri Lanka laws protecting designs rights can be as important as the decision to property rights give their holders the exclusive from pirates have allowed manufacturers of protect them. When the United States right to sell the product embodying their creation quality ceramics to increase exports. Software established a patent regime in the early 19th for a limited time. During this period, holders are and recording industry �rms in Indonesia report century, it modeled its laws after those of the free to charge whatever price they wish irrespec- that they would expand production if their copy- United Kingdom. But unlike the United tive of production costs. Intellectual property rights were better protected. Kingdom, the fees for registering a patent were rights thus need to strike a balance between soci- Second, a growing number of developing very low, innovators were free to license their ety’s interests in fostering innovation and in keep- countries are seeking to attract FDI, including in patents to others, and administrative ing prices to consumers low. industries where proprietary technologies are procedures ensured even-handed application of Finding the right balance is a challenge. important. But foreign �rms are reluctant to the law to all. Broadening access to intellectual Early in the 19th century, when the United transfer their most advanced technology, or to property rights spurred an enormous increase States had few authors but many readers, Eng- invest in production facilities, until they are con- in innovative activity, and shortly after a mid- lish writers complained that the American gov- �dent their rights will be protected. Innovators century exhibition in London, where British of�- ernment did not enforce their copyrights. But as in many knowledge-intensive industries simply cials were shocked by America’s technological the United States produced more authors, gov- will not invest where the protection of their achievement, they followed its example and ernment policy toward copyright changed until intellectual property is uncertain. opened up their patent regime. the United States became a leader in persuad- Third, there is growing recognition that con- ing other nations to honor copyrighted works. sumers in even the poorest countries can suffer Source: Braga, Fink, and Sepúlveda (2000); Maskus For the past two decades countries where the from the sale of counterfeit goods, as examples (2002); Arora, Ceccagnoli, and Cohen (2003); Nathan majority of innovative products are produced ranging from falsely branded pesticides in Associates Inc. (2003); Hoff (2003); and Luthria and have urged countries that mainly purchase Kenya to the sale of poisoned meat in China Maskus (2004). breach the agreement. Governments can help institutions, industrial companies, and oth- �rms cope with these risks by fostering the ers in the business community. Those who dissemination of accurate reputation infor- contemplate breaching their obligations mation and supporting effective dispute reso- know that if they do, all will soon know. lution and enforcement mechanisms. Government policy sometimes hinders the creation of �rms that market reputation Facilitate the flow of information information by restricting the flow of com- about reputation mercial or �nancial data. Free-rider problems, Reputation is central to ensuring contract highly concentrated �nancial systems, and performance in all societies.29 In deciding other market failures can also retard the emer- whether to contract with a new partner, gence of private organizations that gather and �rms are guided by what they know about disseminate reputation information.30 Gov- the potential partner’s history of complying ernments should �rst remove the impedi- with contractual obligations. A �rm is more ments to circulating accurate data on credit- likely to contract with those who have a worthiness.31 If private �rms still do not enter good reputation. Various entities have the market, government can. In Bangladesh, emerged to meet the demand for such Bolivia, Bulgaria, Nigeria, Romania, and Viet- information. They collect information on nam, government-owned reporting agencies the creditworthiness and reliability of indi- have been established, building on data col- viduals and �rms and provide it to �nancial lected by the central bank.32 (c) The International Bank for Reconstruction and Development / The World Bank 86 WORLD DEVELOPMENT REPORT 2005 Improve courts and other dispute customers and so turning arm’s length trans- resolution mechanisms actions into transactions within �rms.37 There are limits to the reach of reputation- New research underlines the importance based mechanisms. Firms without a history of well-performing courts for a sound of creditworthiness will have dif�culty investment climate. Studies from Argentina gaining a foothold in the market resulting, and Brazil show that �rms doing business in extreme cases, in the prevention of new in provinces with better-performing courts entry altogether.33 Reputation mechanisms enjoy greater access to credit.38 New work in also depend on participants being willing to Mexico shows that larger, more ef�cient collectively boycott anyone with a bad repu- �rms are found in states with better court tation. As economies expand, however, the systems. Better courts reduce the risks �rms dif�culties of enforcing a group boycott face, and so increase the �rms’ willingness increase. More information must be col- to invest more.39 lected and disseminated on more individu- • Firms in Brazil, Peru, and the Philippines als and �rms, and the temptation to cheat, report that they would be willing to or free-ride, on the agreement grows. Even- increase investment if they had more tually a centralized contract enforcement con�dence in their nation’s courts.40 mechanism operated by the state becomes a less costly alternative.34 Rather than incur- • Firms in Albania, Bulgaria, Croatia, ring substantial costs before entering into a Ecuador, Moldova, Peru, Poland, Roma- transaction, �rms �nd it less expensive to nia, Russia, Slovakia, Ukraine, and Viet- turn to a court after the fact to resolve dif- nam say they would be reluctant to ferences over performance. The importance switch suppliers, even if offered a lower of courts grows as the number of large and price, for fear they could not turn to the complex long-term transactions increases. courts to enforce the agreement.41 The impact of a well-functioning court • Firms with con�dence in the courts in system extends far beyond the number of Poland, Romania, Russia, Slovakia, and cases it resolves. The more timely and pre- Ukraine are more likely to extend trade dictable a court’s decisions, the better able credit and to enter new relations with �rms are to predict the outcome of any dis- local �rms. pute. As predictability and timeliness • In Bangladesh and Pakistan the World improve, the number of disputes �led may Bank’s Investment Climate Surveys show decline, because a credible threat of pursu- that while �rms with con�dence in the ing a remedy in court provides incentives courts make half their sales on credit, for the parties to honor their obligations. those with little con�dence extend credit Bargaining takes place in the shadow cast by on only one-fourth of their sales. the courts and the laws they enforce. The • In Burundi, Cameroon, Côte d’Ivoire, stronger the shadow they cast, the lower the Kenya, Madagascar, Zambia, and Zim- risk of transacting, the larger the number of babwe, where �rms have little con�dence transactions, and the lower their cost.35 in the courts, they are unwilling to Where the shadow is weak, a �rm’s costs expand trade by doing business with any- and risks increase. In India those whose con- one other than those they know well.42 tracts have been breached or who have suf- fered other injury must either accept a The Investment Climate Surveys show sharply discounted settlement or wait years, that in many countries, �rms have little if not decades, to have their case resolved in con�dence in courts (�gure 4.3). One rea- court.36 A weak shadow can also make some son may be the length of time and the cost transactions so risky that they never occur, required in many countries to resolve even for if there is no way to ensure performance, simple cases. The World Bank’s Doing Busi- the risk of going forward may simply be too ness Project shows that in 2003 the time great. Or �rms may circumvent the judicial required to enforce a contract range from system altogether, taking the costly but less under 50 days in the Netherlands, nearly risky route of purchasing their suppliers or 600 days in Bolivia, to nearly 1,500 days in (c) The International Bank for Reconstruction and Development / The World Bank Stability and security 87 Guatemala. Nor does the evidence show Figure 4.3 Many �rms do not believe the courts will uphold their property rights that slower, more costly courts deliver better Bangladesh results than less expensive, more expedi- tious ones.43 Moldova Strengthen court systems. As World Devel- Guatemala opment Report 2004 showed, agencies that Kyrgyzstan provide a public service perform better Kenya when they are accountable to users, when users have a say in the policies governing Czech Rep. the delivery of the service, and when those Brazil providing the service have a strong incen- Zambia tive to deliver quality services. These same Algeria principles apply to courts. A common result of giving users more Malaysia voice in the operation of the courts is pro- 0 20 40 60 80 100 cedural simpli�cation. Court procedures in Percent many developing countries are more com- Note: Countries selected to illustrate range of responses. plex and costlier than those in developed Source: World Bank Investment Climate Surveys. countries. Not only do these lengthier and more expensive procedures provide no off- before it, and although its �ling fees are setting bene�ts, they are often simply a fur- higher than the ordinary courts, to which ther drag on entrepreneurial activity.44 In litigants can also turn, its case load contin- Brazil complex court procedures retard ues to grow. credit markets and increase the cost of Efforts to create specialized commercial credit transactions.45 courts in Bangladesh, Indonesia, Cape Coupling procedural reform with Verde, Côte d’Ivoire, Pakistan, and Rwanda changes in the way courts are managed and have so far been less successful. The differ- combining both with the introduction of ence often lies in the political support information technology can dramatically courts enjoy. In Tanzania the court handles cut the time needed to decide a case. This cases �led by banks and other �nancial mix produced an average reduction in pro- institutions that constitute a powerful lobby cessing time of 85 percent in six pilot courts in support of the court. But progress is in Ecuador. Similar results were realized more dif�cult when the targets of court across a range of courts in República Boli- action hold signi�cant political influence.In variana de Venezuela as well. In Barques- Bangladesh, for example, the defendants imeto and Ciudad Bolivar, reforms intro- duced in 1999 trimmed the time required to dispose of leasing and debt collection cases Figure 4.4 Reforms speed up court business in República Bolivariana de Venezuela from anywhere between half to two-thirds 800 (�gure 4.4). Judges were relieved of routine 1999 Days from initiation to termination administrative tasks, clerical work was cen- tralized in a judicial support of�ce, while 600 the entire litigation process, from the �ling of a complaint to the scheduling of hearings 400 to the issuance of judgment, was auto- mated. 2001 One frequently considered option for 200 speeding up commercial cases is the cre- ation of either a separate court or a separate division or chamber within an existing 0 Barquesimeto Ciudad Barquesimeto Ciudad court to handle business disputes. Tanza- Bolivar Bolivar nia’s recently created commercial court Leasing Debt collection draws praise from lawyers who appear Source: Supreme Court of República Bolivariana de Venezuela. (c) The International Bank for Reconstruction and Development / The World Bank 88 WORLD DEVELOPMENT REPORT 2005 include influential citizens being asked to involves technical issues, the parties can select repay millions of dollars in loans from state- an engineer or other expert versed in the rele- run banks. Similarly, in Indonesia the defen- vant issues to decide the matter. dants include those being asked to accept Some governments discourage private dis- signi�cant losses in court-ordered reorgani- pute resolution through unnecessary restric- zation and liquidation proceedings. tions on procedures. In Bolivia and Tanzania Court performance depends on judges, various restrictions on alternative dispute res- lawyers, clerks, and other participants work- olution mechanisms prevent �rms from tak- ing to ensure the timely and accurate resolu- ing full advantage of them.48 By contrast, in tion of disputes. Differences in court perfor- Colombia and Peru—where government has mance are largely a function of different enacted legislation supporting the use of incentives.46 When participants have strong alternatives—the results have been promising. incentives to see that cases are decided expe- A commercial arbitration chamber run by the ditiously, accurately, and at a reasonable cost, Bogotá Chamber of Commerce handled 371 court performance improves dramatically cases in 2001 involving claims of Col$3.2 bil- Legal professionals who work in and lion. The Lima Chamber of Commerce around courts often fear that changing incen- resolved 182 commercial disputes in 2000 in tives will affect their incomes. In Tanzania an average time of less than six months.49 reformers overcame the lawyers’ opposition Where the parties to an arbitration or by persuading key members of the profession other alternative dispute resolution mecha- that they would bene�t from reform. As con- nism contemplate continued dealings, each �dence in the courts increased, reformers has an incentive to abide by the arbitrator’s argued, more cases would be �led, so the award. Each may also comply because of the demand for legal services would increase. In effect on its reputation if it refuses to do so. several countries, working groups of senior If a party refuses to honor an arbitrator’s judges, respected members of the bar, and decision, it runs the risk that other �rms will civil society have come together to develop a decline to do business with it in the future. consensus on the bene�ts of reform. Where the incentives of reputation or A special challenge in court reform is repeat dealing are not present, the courts that the judiciary is usually a separate and need to backstop arbitration by permitting independent branch of government. Of�- the prevailing party to bring an enforce- cials in the executive can urge judges to ment action. To be an effective backstop, the reform, and the legislature can pass laws to law must not give the loser in an arbitration streamline procedures, but implementation proceeding a long period or numerous ways depends on the courts. One step the execu- to challenge the award. The United Nations tive branch can take on its own is to review Commission on International Trade Law its use of the courts. Governments are often recommends that courts should be permit- the largest single user of the courts, and as a ted to set aside awards only in limited and study in the Indian state of Andhra Pradesh precisely de�ned situations. Otherwise, as shows, government often contributes to happened in India, litigation over the valid- delays by pursuing matters it has no chance ity of awards can spiral out of control as the of winning and lodging appeals it is sure to losing side seeks to win in court what it lost lose.47 Curbing such behavior can reduce at the arbitration table.50 the demands on the courts and allow them Access to arbitration in a neutral country to concentrate on genuine disputes. is often important to foreign investors, who may fear that the courts in the country of the Remove impediments to private dispute set- investment are biased against them, or too tlement. Fostering private resolution through slow, or too inexpert to hand down a timely arbitration, mediation, or conciliation will and accurate decision. International arbitra- also improve the contracting environment. tion has emerged as an important way for Not only are these methods often less expen- investors to reduce the risks of submitting sive than a lawsuit, they can produce more disputes to local courts.51 To improve the accurate decisions as well. Where the dispute investment climate, governments should (c) The International Bank for Reconstruction and Development / The World Bank Stability and security 89 remove obstacles to international arbitration Figure 4.5 Crime takes a heavy toll on many Latin American economies as well, by joining relevant international con- ventions and ensuring effective mechanisms exist to enforce the resulting awards. For El Salvador example, the Russian government recently Colombia clari�ed that awards by international Mexico arbitrators in disputes involving minority shareholders in Russian corporations are RB de Venezuela enforceable in domestic courts. The role of Brazil international dispute settlement mecha- Peru nisms is discussed further in chapter 9. 0 10 20 30 Cost of crime as a percentage of GDP Reducing crime Note: Data cover 1999–2000. Source: Londoño and Guerrero (2000). Robbery, fraud, and other crimes against property and against the person undermine cent of the small ones do. the investment climate. Rampant crime dis- A 2002 survey of 400 Jamaican �rms courages �rms from investing and increases offers further insights into the way crime the costs of business, whether through the can affect incentives to invest.52 Just under direct loss of goods or the costs of taking two-thirds of �rms surveyed reported being precautions such as hiring security guards, the victim of some kind of property crime building fences, or installing alarm systems. during 2001, with many �rms repeatedly In the extreme, foreign �rms will decline to victimized. More than one-fourth had prop- invest, and domestic ones will flee the coun- erty stolen once a quarter, with 9 percent try for a more peaceful locale. reporting theft once a week, and 22 percent Estimates compiled in 2000 show the saying they were defrauded at least once a devastating impact of violent crime and quarter. Firms of all sizes and locations were property crime on the economies of six victimized. Eight of 10 farmers reported Latin American nations. In Colombia and equipment or livestock stolen. Financial El Salvador almost one-quarter of national �rms were most vulnerable to fraud. Manu- GDP was lost to crime; only in Peru was the facturing, distribution, and construction cost of crime less than 10 percent of gross companies all reported signi�cant theft and domestic product (�gure 4.5). fraud. Smaller �rms were more likely to be The World Bank’s Investment Climate victims, and more often, than larger �rms. Surveys show that crime retards entrepre- Extortion, fraud, robbery, burglary, and neurial activity in every region. In Latin America more than 50 percent of �rms sur- Figure 4.6 Crime is a signi�cant constraint on �rms veyed judged crime to be a serious obstacle in all regions to conducting business. In Sub-Saharan 60 Africa and East Asia more than 25 percent Firms reporting crime a major or moderate or more said the same (�gure 4.6). Latin America constraint on doing business (percent) & the Caribbean The impact of crime varies by country. In Nigeria the Investment Climate Survey 40 shows 37 percent of respondents identify crime as a major or severe constraint on Sub-Saharan Africa their operations, in Zambia 50 percent, and in Kenya 70 percent. In Guatemala an extra- East Asia & Pacific ordinary 80 percent of surveyed �rms said 20 South Asia that crime is a major or severe constraint. Eastern Europe Crime tends to have a similar effect on �rms & Central Asia of all sizes. One exception is Bangladesh. Although 45 percent of medium and large 0 �rms say crime is a constraint, only 20 per- Source: World Bank Investment Climate Surveys. (c) The International Bank for Reconstruction and Development / The World Bank 90 WORLD DEVELOPMENT REPORT 2005 arson caused 116 of 400 �rms to suffer Firms can do much to reduce property losses amounting on average to J$665,000 crime—from installing burglar alarms to (around $11,000). The cost of crime as a posting security guards. But there are limits percentage of revenue was 9 percent for to what they can achieve without govern- small �rms and �rms with J$20-50 million ment assistance. Property crime is rarely a in annual revenues. Manufacturing �rms crime of passion, suddenly triggered by reported that crime cost them nearly 6 per- overwhelming feelings of jealousy, betrayal, cent of annual revenues. or rage. Instead, as analysts have recognized When government is not strong enough for more than two centuries, it is almost to protect property rights, private organiza- always motivated by a calculation, however tions selling “protection services� �ll the void. rough, of the bene�ts to be gained against Some private security services cooperate the consequence of apprehension and pun- closely with the police. Others are not nearly ishment (box 4.8).54 so law abiding. “Violent entrepreneurs,� as a Combating crime is a major challenge in recent analysis of organized crime in Russia all societies, rich and poor. Experience sug- labeled them, rely on force and intimidation gests that governments can change the and often end up demanding a share of the incentives for criminals through better law pro�ts of the �rms they “protect.�53 While enforcement, stronger deterrence, and more respondents to the World Business Environ- effective crime prevention programs. ment Survey said that organized crime has less impact on their business than street Better enforcement crime, the Jamaica survey suggests the data Apprehending and punishing criminals is a may understate its effect. Many businesses are classic government function, but one often reluctant to admit they are victims of extor- not performed with great ef�ciency. Reac- tion, either from shame or for fear of violent tive policing, where police simply receive reprisal. reports from victims and then attempt to apprehend the responsible party, is notori- ously ineffective. Modern policing is “prob- lem oriented.� It attempts to identify recur- BOX 4.8 Crime, poverty, and inequality ring crime problems and—with other government agencies and civil society— Evidence from a single country over time or there is always a residual, something left change the conditions leading to these from many at one point in time shows that over after all economic factors are included, problems. The approach emphasizes crime an increase in relative poverty or income that explains part of the crime rate. This equality leads to a rise in crime. One study residual becomes more signi�cant as the mapping, working with communities, and drawing on data from developing countries time frame lengthens. Long-run changes in investigative techniques, rather than a gen- �nds that a relatively modest increase in the crime rate appear to respond to soci- eralized “get tough on crime� approach. inequality would produce an average etal forces largely independent of Nor can policing be separated from human increase in robberies of 30–45 percent. economic ones. Another study suggests that a 5 percent Several policy implications follow from rights concerns: force may be required to drop in GDP would produce an immediate what is known. One is the importance of capture and detain suspects, and the police 50 percent jump in the robbery rate. reducing relative poverty and inequality, need to be accountable to multiple con- According to one view this relation- not only for reasons of social justice, but for stituencies when they use it. ship follows as a matter of economic the very practical reason that it is a sure logic. The decision to commit a crime way to cut crime rates. A second is that Stronger deterrence depends on whether the return, sharp increases in relative poverty or discounted by the likelihood of apprehen- inequality call for an immediate response. Government can also improve the deterrent sion and punishment, exceeds the gain Both crime prevention and crime effect of its criminal justice system. Are the from working. The more unequal the dis- deterrence programs need to be expanded penalties for theft, robbery, and other prop- tribution of income and wealth in a soci- to dampen the inevitable rise in crime ety, the larger the potential gains from likely to follow. Third, each society needs to erty crimes enough to alter a thief ’s cost- crime for those at the bottom of the scale. examine what accounts for the part of the bene�t calculus? Are they applied consis- Another view is that inequality is associ- crime rate not explained by economic fac- tently? How effective is the overall system at ated with discrimination and other social tors. In other words, what shapes the char- preventing and deterring crime? factors that affect character formation— acter of its citizens? and is thus the “real� cause of crime. No matter the penalties, criminal law is Study after study af�rms the great Source: Bourguignon (2000); Demombynes and only as effective as those who enforce it. The power of the economic explanation. But Özler (2002); and Wilson (1991). police are the frontline enforcement agency, and any crime reduction initiative must (c) The International Bank for Reconstruction and Development / The World Bank Stability and security 91 begin with ensuring their effectiveness. But Putting police into the community police reform is a challenge—for developed increases their accountability and provides and developing countries alike. It is hard to citizens with a greater voice in their opera- monitor the actions of individual of�cers on tions. Community policing contributes to patrol or to prescribe how they should handle more effective policing as well. Police solve the variety of often dangerous situations they very few crimes on their own, fewer than 10 confront on a daily basis. Dif�cult working percent in a recent U.S. study. Victims and conditions coupled with the sometimes hos- witnesses from the community have to come tile relations between police and citizens tend forward with information about the perpetra- to isolate the police, creating a strong sense of tors for police to improve on this �gure. As the loyalty among of�cers, and making external South African police have learned, by forging oversight and accountability dif�cult. stronger ties between the police and citizens, Despite these hurdles, some promising community policing has furthered coopera- approaches emerged in the 1990s under the tion between the two, leading to higher arrest rubric of community policing (box 4.9). rates and greater respect for human rights.56 Although the details differ across countries, and even within countries, such policing More effective prevention includes one or more of the following: Governments need to resist the temptation to look for answers only within the con�nes • Assigning of�cers to foot patrols so that of the criminal justice system. Several recent they can concentrate on those offenses studies show that well-designed crime pre- that are major annoyances to local resi- vention programs are more cost-effective dents and �rms and build relationships than criminal justice approaches.57 A classic with the community study in the United States found that for • Meeting regularly with individuals and every dollar invested in prevention pro- �rms to hear their views on police priorities grams, six to seven dollars could be saved in • Analyzing trends and focusing efforts criminal justice expenditures.58 Effective pre- rather than responding to each individ- vention strategies include early interventions ual crime as a report comes in.55 for at-risk teens, school-based initiatives to BOX 4.9 New York City’s police reforms—are they exportable? In the mid-1990s crime rates in New York City fell devolved to precinct commanders, the depart- aspects of the New York City reforms. Fortaleza, dramatically. Murders plunged 68 percent, bur- ment’s line managers, and a new career system a resort stop on the Brazilian coast, has created glaries 53 percent, and car thefts 61 percent.This rewarded commanders who reduced crime rates. its own version of Compstat and is striving to extraordinary turnaround in crime helped sup- Strategy. Department personnel developed improve police-citizen relations. Chile adopted port an economic renewal as employment, prop- a new computerized data management system several New York City–style reforms, including erty values, and the growth rate rose sharply. to rapidly compile crime statistics and plot the redeployment of police to high-crime areas, Much of the credit for this achievement is emerging trends and the locations of crime. more policing on foot, and better methods of attributed to police reforms introduced by “Compstat� turned out to be critical to the entire collecting and analyzing crime statistics. William Bratton during his tenure as head of reform process. By providing weekly totals of Bratton acknowledges that New York City’s New York City’s Police Department from 1994 to crime and arrests by precinct and comparing experience must be adapted to the very differ- 1996.The reforms were built around two princi- them to historical data, supervisors could evalu- ent cultures and crime environments in the ples: wholesale changes in management, to ate the performance of the precinct comman- developing world, where the police are often reward those who succeeded in combating ders, which they did at weekly strategy sessions. not yet fully subject to civilian control and crime while penalizing those who didn’t, and a The system also allowed the department to respect for citizen rights can be weak. Even so, proactive crime-�ghting strategy. adopt a new strategy. Rather than react to indi- the underlying principles—devolving power to Incentives. Bratton inherited a department vidual crime reports, managers could spot local commanders, holding them accountable where promotion depended not on arresting evolving patterns and redeploy personnel for results, building citizen con�dence in the criminals but on avoiding scandals, conflicts with accordingly. At the same time the police began police, and adopting proactive crime �ghting the community, and indeed any activity that concentrating on the infrastructure that strategies—are as applicable in Santiago or For- might make waves. After long consultations with supports individual crimes. Instead of targeting taleza as in New York’s toughest neighborhoods. of�cers and other stakeholders, some 400 individual car thieves, they went after those who changes were made in the way the department dealt in stolen automobiles, thus shrinking the operated. Recruiting standards were raised, train- thieves’ market. Source: Bratton and Andrews (1999); Lifsher (2001); ing was improved, and disciplinary procedures Exportable? Several Latin American cities Fundación Paz Ciudadana (2001, 2002); Webb-Vidal were modernized. Most important, power was have begun experimenting with different (2001); and Bratton and Andrews (2004). (c) The International Bank for Reconstruction and Development / The World Bank 92 WORLD DEVELOPMENT REPORT 2005 teach social competency skills and reduce vio- clear. Mass expropriations usually occur in lence in and around schools, and other pro- the wake of violent upheavals, as when the grams to build character and foster commu- post-1917 government in Russia or those in nity responsibility.59 Emerging evidence from power in Eastern and Central Europe after Colombia shows that handgun control and World War II seized private property. Since restrictions on the sale of liquor can reduce then, expropriation has been most com- violent crime signi�cantly.60 Situational crime monly associated with the nationalization prevention—in which physical space is modi- of foreign investments, though recent expe- �ed to make the commission of a crime more rience in Zimbabwe shows that local �rms risky or less lucrative for the potential are not immune (box 4.10). offender—is a promising prevention strategy Property need not be taken in its entirety, for housing ministries or local governments.61 or in a single stroke, to constitute a “taking� or Government can also take some of the “expropriation.� Taxes may be progressively pro�t out of organized crime by reducing the raised to con�scatory levels or regulations regulatory burden on �rms. Surveys of retail made so onerous that an owner is forced to stores in three Russian cities show that pro- sell all or part of the property at a depressed tection rackets and other forms of organized price. While the outright expropriation of for- crime flourish when the regulatory burden is eign investments has become less common in high.62 As the regulatory burden increases, recent times, these forms of indirect or “creep- store owners are less able to comply with the ing� expropriation have grown signi�cantly.65 rules and thus more reluctant to call on state Foreign investors are often particularly agencies to protect them from criminals or vulnerable, because it may be politically enforce their contracts. Organized criminals attractive for politicians to target foreigners, then step forward to meet the demand. and local courts may be reluctant to rule against the host government if a dispute arises. Large and immobile investments are Ending the uncompensated especially at risk. Because they cannot be expropriation of property moved to another location in response to The discussion so far has focused on how changing circumstances, they constitute governments can help �rms cope with threats what Vernon called an “obsolescing bargain,� to their property rights from third parties. As being exposed to host government efforts to chapter 2 showed, however, government can renegotiate unilaterally the terms of the orig- itself threaten the security of property rights. inal agreement.66 Foreign investments in pri- A government strong enough to protect vate infrastructure projects have both these property is also strong enough to take it.63 features—and are often in politically sensi- All governments reserve the right to take tive sectors with returns subject to regula- property in some circumstances.64 To combat tion, making them even more vulnerable.67 health emergencies, government must be able The threat of expropriation varies from to order the destruction of livestock or poul- project to project, even in a single country.68 try spreading disease. Without the power to While not amenable to precise measure- take land, those holding parcels needed to ment, the risk of expropriation is reflected in complete an expressway can “hold up� gov- measures of “country risk� or “political risk� ernment by demanding unreasonable prices prepared by various rating agencies (�gure to sell. The taking, or expropriation, of prop- 4.7). Some governments have credible mech- erty can also be a more ef�cient means of anisms to restrain these threats and, coupled obtaining it for public purposes than open with a history of treating investors fairly, market purchases. When an underground investments in these countries are perceived water pipe must traverse a large number of to involve only modest risks. Other govern- properties, the costs of reaching an agreement ments have not yet established the same on price with each individual owner will be record—or have not been able to credibly far greater than setting a price by decree. commit to restrain such risks. When this is Governments have also seized private the case, investors will decline to invest, avoid property where the public interest was less undertaking investments that are dif�cult to (c) The International Bank for Reconstruction and Development / The World Bank Stability and security 93 BOX 4.10 Property wrongs: Is there ever a statute of limitations? If one buys a watch from someone who found it Statutes of limitations and other the extreme, lead to civil war, as in Guatemala.But on the street, can the original owner later mechanisms embody the consensus societies ill-conceived attempts to redistribute property can recover it? If a farmer settles on apparently have reached to resolve the ownership-transac- also have disastrous consequences.Since unclaimed land and plants crops, can the land’s tion tradeoff between individuals. Reaching a Zimbabwe began seizing white-owned land in real owner later evict him? consensus is dif�cult when different communi- 2000, agricultural production has dropped precipi- Easy cases are when the watch purchaser or ties within a society are on opposite sides. In tously.Africa’s fastest growing economy in 1997 farmer knew, or had reason to know, that the Zimbabwe many argue that land taken during became its fastest shrinking in 2003. original owner was lurking out there the colonial wars of the 19th century should be Between inaction and ill-conceived action, pol- somewhere—the watch had the owner’s name returned to the descendants of the true owners. icymakers have many options for reaching a solu- engraved on it, the land was registered or Current holders reply that, in most cases, they tion.One is to purchase land for redistribution, a fenced. But when the watch seller genuinely bought the land after independence, decades policy Zimbabwe had pursued until 2000, albeit at appeared to own it or the land really did seem after the initial seizures, and preserving the a glacial pace, and one Brazil, Colombia, and South to be unclaimed, the answer requires trading off security of these transactions should trump the Africa are following with World Bank support.Poli- the right of ownership against the security of right of ownership. Similar arguments are a fea- cies to remedy the consequences of the existing transactions. ture of the political landscape in countries as distribution of property are also promising, from Allowing the watch purchaser to defeat the different as Australia and Guatemala.That the efforts to equalize educational opportunities to original owner’s claim to recover it makes trans- argument has in some cases gone on for many changes in tax policies.Addressing the needs of actions more secure. Permitting the original decades is a sign that no easy solutions exist to those disadvantaged by the current distribution of owner to reclaim it makes the right of claims of property wrongs, that there is no property by such “leveling up�measures requires ownership more secure. Societies have statute of limitations to invoke to extinguish signi�cant resources and is considerably easier developed a number of mechanisms for manag- claims, no matter how well-grounded. when the economy is growing.The relationship ing these tradeoffs, including statutes of limita- Policymakers confronting these situations face between a sound investment climate and prop- tions—laws setting a �xed period for challeng- a dilemma.To allow the argument over the fairness erty wrongs thus comes full circle. ing a transaction. Once that time has elapsed, of the current distribution of property to fester can the transaction can no longer be attacked. undermine the security of property rights and, in Source: Pound (1959). reverse, or require higher rates of return to some assurance to �rms that the loss will Figure 4.7 Risky business compensate for the extra risks. not be total if assets are expropriated. These 25 The signi�cance of the risk of expropria- provisions also help to deter governments tion is reflected in the diversity of strategies from “over taking� by ensuring that govern- �rms pursue to address it.69 Political risk ment must pay for what it takes. 20 by income level, on political risk Average country rating in 2003, Low-income countries insurance can be purchased, but it protects Setting the compensation price can be the investor only partly and can add 2 percent dif�cult. A requirement that governments Lower-middle-income 15 countries a year to the cost of the investment. Involving pay “fair market value� is dif�cult to apply a connected local �rm as an investment part- because by de�nition there is no willing Upper-middle-income countries ner can be another form of insurance, though seller. Various broad formulas have been 10 such strategies often back�re when a new adopted to determine compensation: “just� crowd takes power. Better strategies involve in the United States, “proper� in Spain, 5 efforts to ensure the power to expropriate “adequate� in Malaysia and Mauritius.70 In High-income countries property is subject to credible limits. This South Africa the amount must reflect the means devising ways to limit its reach and use, history, market value, and previous 0 establishing an effective mechanism to review state investment in the property as well as Note: Political risk is de�ned as the risk of non- payment or nonservicing of payment for goods its exercise, as well as addressing the incentives the purpose of the expropriation. or services, loans, trade-related �nance and governments may face to misuse the power. Although these policies could be set forth dividends, and the nonrepatriation of capital. Principal risks include war, civil disturbance, in a statute, at least the basic elements should nationalization, a change in rules on capital out- Limit the reach of the power to be in the instrument most dif�cult for gov- flow or currency convertibility, and lack of for- eign exchange. Higher values mean higher polit- expropriate ernment to change: the nation’s constitution. ical risk. Source: Euromoney website. Governments should be clear that property This is the approach of most developed will be expropriated only to serve a public countries and a growing number of develop- purpose—and that when it is expropriated ing countries. The same assurance is also the there is assurance of prompt, adequate, and cornerstone of most international agree- effective compensation. The public purpose ments dealing with investment (chapter 9). limitation reduces the ability of govern- De�ning the precise boundaries of an ments to use the power to favor private expropriation for which compensation is interests. Compensation provisions provide payable is also not straightforward. While (c) The International Bank for Reconstruction and Development / The World Bank 94 WORLD DEVELOPMENT REPORT 2005 creeping or indirect expropriation may take decades. It requires not only a well-trained many forms, governments cannot be and dedicated cadre of judges but vigorous expected to compensate �rms for every and sustained political competition as action that influences the value of their well.71 property in some way. Routine adjustments When domestic courts are weak, or their to tax policies and regulatory regimes may credibility is low, government can agree to make some �rms worse off but do not con- submit disputes involving expropriation to stitute an expropriation for which compen- an international tribunal. As discussed in sation should be paid. Detailed standards chapter 9, a growing number of interna- have emerged in the laws, regulations, and tional investment treaties provide for this court decisions of many countries, and option. Although these treaties deal with make it clear, for example, that a change to foreign investment, there are halo effects for an environmental regulation that has a dif- local �rms, and there is no reason why gov- fuse effect across a large number of �rms ernments cannot agree to similar arrange- does not rise to a compensatory taking. Sim- ments for local �rms. ilar standards are emerging in the arbitra- tion cases decided under international Create incentives against the misuse investment treaties (chapter 9). of the expropriation power In devising their policies in these areas, The incentives governments face to misuse governments must bear in mind that they the expropriation power are ultimately are not writing on a blank slate. Where their influenced by the broader social and politi- predecessors have recklessly expropriated cal context. As chapter 2 emphasized, tam- property, the current government may need ing the “grasping hand� of government can to overcome the effects of a reputation that involve a range of strategies, including it had no hand in making. Following a wave efforts to improve the accountability of of expropriations in the 1980s under a law governments, enhance the transparency of providing minimal compensation, Peruvian the government-business interface, and �rms in the early 1990s were reluctant to strengthen competition. No less important expand operations or invest in new ones. To is to foster a broad social consensus in favor help restore con�dence the 1993 constitu- of building a more productive society— tion requires government to reimburse including by ensuring that the opportuni- �rms for the actual value of any property ties of a better investment climate are taken and any “possible loss� as well. Where, shared widely across society. for example, the land on which a factory is located is taken for a highway, government must not only pay for the land but also Expropriation is the most direct way govern- reimburse the owner for the costs of mov- ments threaten the security of property ing the machinery and other equipment to rights and so dampen incentives to invest a new location. productively, but it is not the only way. Policy uncertainty and unpredictability also under- Establish a mechanism to review the mine the value of property rights by creating exercise of the expropriation power additional risk for �rms (chapter 2). Limitations on the government’s power to Governments also qualify property rights expropriate are credible only if means exist through the ways they regulate and tax �rms to ensure that the limits will be respected. and transactions. In these cases, the quali�ca- Ordinarily this will be a court separate from tion of property rights is deliberate, and the executive. While courts in many devel- intended to balance the bene�ts of more oping countries are gradually breaking free secure property rights with other social goals. of executive branch control, the history of Some of the special issues government must judicial independence in developed coun- grapple with in striking that balance are the tries teaches that this process can take subject of chapter 5. (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 5 The way governments regulate and tax tected �rms to innovate and boost their �rms and transactions—both within and at productivity. their borders—plays a big role in shaping Tax systems are plagued by similar prob- chapter the investment climate. Sound regulation lems. Tax structures often bene�t favored addresses market failures that inhibit pro- groups, distorting competition and foisting ductive investment and reconciles the inter- higher taxes on others. And tax administra- ests of �rms with those of society. Sound tion can be burdensome, increasing compli- taxation generates the revenues to �nance ance costs, reducing revenues, and opening public services that improve the investment the way to corruption. climate and meet other social goals. The That such problems exist is hardly news. challenge all governments struggle with is But new sources of evidence underline the how to meet these objectives without extent of the problems and their impact on undermining the opportunities and incen- productivity and growth. While the under- tives for �rms to invest productively, create lying problems do not always have simple jobs, and thereby contribute to growth and solutions, a growing body of international poverty reduction. experience points to some practical steps There is huge scope in most countries that governments can take to improve these for improving regulation and taxation areas of their investment climates. This without compromising broader social chapter takes a broad view and considers interests. Too often, governments pursue regulation and taxation behind and at a approaches that fail to meet the intended country’s borders. It shows that there is social objective, yet harm the investment great scope for improving performance. climate. How? By imposing unnecessary Later chapters look at speci�c challenges in costs, by increasing uncertainty and risks, regulating the �nancial system and infra- and by erecting unjusti�ed barriers to com- structure (chapter 6), regulating labor mar- petition. kets (chapter 7), as well as issues associated Examples of regulatory problems with selective interventions (chapter 8) and abound. Regulations to promote social the use of international rules and standards goals are often enforced only partially—as (chapter 9). is evident in the huge informal sectors in most developing countries. Yet they can Regulating �rms impose signi�cant burdens on �rms that do Governments regulate �rms in many comply—whether through the extraordi- ways—for many reasons. They regulate to nary requirements to set up a new business restrict who may participate in a market, or the long delays in getting goods through where �rms may locate, the production customs. The interpretation and applica- process used, the quality or other parame- tion of regulations can be unpredictable— ters of the goods and services produced, creating uncertainty and risk for �rms and and the way products are marketed and inviting corruption. Regulations also create distributed. Indeed, it is hard to �nd any monopolies or cartels for favored groups— aspect of a �rm’s business and investment imposing costs on consumers and other decisions that is not affected in some way �rms, and stifling incentives for the pro- by regulation. While it is dif�cult to �nd a 95 (c) The International Bank for Reconstruction and Development / The World Bank 96 WORLD DEVELOPMENT REPORT 2005 Figure 5.1 Low-income countries tend to regulate more (positive externalities) on others. Pollu- tion is a classic negative externality: a �rm More 66 Court powers in bankruptcy (index) regulation 63 that releases pollution into a river can Entry procedures (number) 56 Contract impose costs on its neighbors farther 30 procedures downstream. If the �rm fails to take 27 (number) Number of procedures 27 account of the effect of its pollution on 43 others, it will generate more than is socially optimal. Governments can recon- Index 18 cile the �rm’s incentives with those of the wider community by restricting pollu- 11 12 10 tion. They may do this through tradi- 7 tional command-and-control regulation, such as prohibiting certain activities or Less establishing standards for acceptable regulation effluent levels, or they might fully assign Low- Lower- Upper- High- income middle- middle- income property rights or tax the product that countries income income countries causes the negative externality.1 countries countries Note: Data from World Bank Doing Business Project. • Information problems arise when con- Source: World Bank (2004b). tracting parties have unequal access to information about the good or service in single indicator that captures the many question. For example, consumers may dimensions of regulation and the varia- lack reliable information about the qual- tions in its intensity, recent work suggests ity or safety of a product, or the quali�ca- that developing countries tend to regulate tions of a service provider. Regulation more than richer countries in many areas may address these concerns in several (�gure 5.1). ways. Over and above prohibiting fraudu- How, then, can governments make lent conduct, governments may require progress? The key is to strike a better bal- �rms to disclose certain information ance between market failures and govern- about their products (as through product ment failures, and to ensure a good �t with labeling), require the safety of products to local conditions. This requires efforts to be independently veri�ed (as with drugs address regulatory costs and informality, to in many countries), or simply ban the sale reduce regulatory uncertainty and risk, and of hazardous products. to tackle barriers to competition. • Monopoly arises when a �rm (or group of Balancing market and government �rms acting in concert) has enough mar- failures and achieving a good ket power to raise prices above the com- institutional �t petitive level and thereby extract higher pro�ts at the expense of consumers and Regulation improves social welfare—and economic ef�ciency. In assessing market the investment climate—when it responds power, competitive pressure is not limited to a market failure cost effectively. This to direct head-to-head competition requires an assessment of market failures between existing �rms offering identical and government failures, and the extent to products. It can also come from the threat which the proposed regulatory strategy of entry by new �rms, as well as from reflects a good �t with local conditions. products that may be effective substitutes (rice might compete with beans for some Market failures. The usual rationale for reg- uses). Governments can address monop- ulation is market failure, the three most oly by removing unjusti�ed regulatory common of which are externalities, infor- barriers to competition, by dealing with mation problems, and monopoly. anticompetitive behavior by �rms • Externalities arise when producing or through competition law, or in extreme consuming a product imposes costs (neg- cases by regulating the price and quality ative externalities) or confers bene�ts of the goods or services provided. Some (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 97 BOX 5.1 Public ownership, regulation, and the investment climate Modern notions of regulation involve a set of enterprises have weaker incentives to comply companies in South Asia have developed a explicit rules that de�ne acceptable conduct with regulations than private �rms. While the highly organized system to extract bribe pay- that are administered and enforced by an entity threat of being �ned can motivate private �rms, ments from customers.The result can be higher operating at arm’s length from regulated �rms. governments have only weak incentives to pros- costs for �rms and reduced revenues for the Some governments have also experimented ecute enterprises that they own, for both politi- public enterprise, reducing public investment with public ownership as a form of regulation. cal and �scal reasons.Third, public enterprises or increasing the burden on taxpayers. Combining production and regulatory roles that depend on budget support, or whose • When public enterprises are granted a monop- involves an inherent conflict of interest. Experi- prices are regulated with political criteria in oly, opportunities are denied to other �rms. ence shows that this conflict—coupled with mind, often lack the resources to meet environ- Even when competition is permitted between political interference, protection from competi- mental or other regulatory standards. public enterprises and private �rms, it is notori- tion, and weak accountability—often leads pub- Overall,public ownership has the potential to ously dif�cult to create a level playing �eld.The lic enterprises to have dismal productivity.The weaken the investment climate in three main ways: problems are especially acute when the public dramatic improvements unleashed through pri- enterprise has a regulatory role, because it will vatization have highlighted how signi�cant the • When public enterprises are responsible for face incentives to use that role to advance its costs can be. providing inputs relied on by private �rms interests over those of competitors—a No less important, public enterprises in (such as power, telecommunications, or phenomenon common in telecommunications. developing countries have a poor record in �nance), weaknesses in their productivity and Even when such obvious conflicts of interest meeting regulatory requirements. For example, incentives can contribute to higher costs and have been addressed by moving regulatory state-owned enterprises in Indonesia were less reliable service, to the detriment of �rms responsibility to a more independent body, found to emit more than �ve times as much pol- (and other consumers) dependent on those pressures to favor the interests of public enter- lution as similar private �rms. State-owned pulp inputs (chapter 6). prises can continue.Public enterprises often and paper plants in Bangladesh, India, Indone- • Public ownership can increase demands for also enjoy a range of exemptions (by law or by sia, and Thailand also controlled pollution less corrupt payments, because public managers practice) from taxes and other regulations that well than similar private �rms. usually have weaker incentives to reduce leak- can also distort competition. Several factors seem to be at work. First, dif- age and graft. For example, �rms in transition fuse objectives, political interference, and weak economies are more likely to have to pay Source: Clarke and Xu (2004); Djankov and Murrell accountability can conspire against good per- bribes to get telecommunications and electric- (2002); Hettige and others (1995); Lovei and McKech- formance. Second, even when regulation is ity services when they are provided by public nie (2000); Megginson and Netter (2001); Shirley and entrusted to a separate regulatory body, public enterprises. Employees of state-owned power Walsh (2000);Wheeler (2001); and World Bank (1995a). countries have also used public owner- technical expertise, the absence of which ship as a form of regulation, typically can undermine effectiveness. with poor results (box 5.1). • Rent-seeking. Regulation may be dis- torted by rent-seeking in its many forms Government failure. Regulation that (chapter 2). Firms or other groups may addresses a market failure can bene�t soci- seek regulation to protect them from ety and the investment climate. However, competition. Of�cials may use regulation even when a market failure exists, it makes to extract bribes in return for favorable sense to intervene only when the expected interpretations, quick decisions, or selec- bene�ts exceed the likely costs. This involves tive enforcement, and regulated �rms balancing market failures with potential have incentives to try to “capture� their government failures. There are three com- regulators through a range of strategies. mon sources of government failure: • Rigidity. Regulation tends to be rigid, • Information and capacity problems. In making it hard to keep up with changes designing and implementing interven- in technology or the way business is con- tions, governments often face severe ducted. Indeed, many regulations in information problems. Governments developing countries have not been will never have as much information as reviewed for many decades or longer. �rms about the impact of interventions Part of the problem lies in inertia, but on their costs or incentives. This is a par- �rms, of�cials, or other interest groups ticular challenge in utility regulation, but that bene�t from particular regulations can arise in other areas as well. And the can have strong incentives to resist implementation of some kinds of regu- reform, no matter how bene�cial it may lation demands a reasonable level of be to society. (c) The International Bank for Reconstruction and Development / The World Bank 98 WORLD DEVELOPMENT REPORT 2005 The challenge of “institutional �t.� As dis- of free incorporation until a century after cussed in chapter 2, interventions that work France and Spain did so.4 In some cases the well in one country may lead to very differ- transplanted laws remain in place today. For ent results in others. This means the costs example, the law regulating business entry in and bene�ts of intervention, and the choice the Dominican Republic dates back to 1884. of regulatory strategy, need to take account The tendency to transplant laws and regu- of local conditions. While there is ample latory systems from other countries contin- scope to learn from regulatory experience ues to this day.5 Regulatory systems in rich in other countries, too often regulatory sys- countries can seem a convenient way to mod- tems have been transplanted uncritically to ernize regulation by offering a proven system developing countries from elsewhere. that is familiar to foreign investors, or foreign Many developing countries inherited experts advising on these matters may simply their regulatory systems from former colo- be more familiar with the approach in their nial powers. Particularly when the colonizing home country. But in many cases adaptation power had little interest in establishing long- to local conditions is required, and without it term settlements, there was little incentive to transplanted approaches can lead to poor adapt approaches to the needs of the broader results.6 Regulatory standards may be set at community.2 Being largely irrelevant to con- unrealistic levels relative to local circum- ditions in the host society, the regulations stances, contributing to compliance prob- were often ignored, or used mainly as a lever lems, informality, and unjusti�ed costs. for of�cials or others to extract rents.3 Those Approaches may not �t easily with related bene�ting from the status quo have incen- parts of the policy and regulatory framework, tives to resist reform, no matter how dys- generating additional uncertainty and risk. functional the regulations may be for the Or regulatory systems may involve high levels investment climate. So the same laws and of discretion relative to the effectiveness of regulations often remain unchanged for local institutional safeguards. Experience in decades, even as laws in the source country Jamaica’s telecommunications sector illus- evolve. For example, Chile established a trates the hazards of the last phenomenon restrictive corporate law in 1854, based upon (box 5.2). Spanish and French law from that time. The Government failures and poor institu- restrictive law was maintained until 1981, tional �ts combine to create many distortions when the code underwent a major revision. in regulatory approaches that harm the As a result, Chile did not adopt the principle investment climate in developing countries. BOX 5.2 Regulating in Jamaica—from transplants to better institutional �t Regulatory systems for utilities need to reconcile In 1965 Jamaica adopted a regulatory system safeguards, the discretion of the regulatory the investor’s need to receive a reasonable rate modeled closely on those in the United States.The agency was reduced considerably.The license of return on an investment with the concern that Jamaica Public Utilities Commission was guaranteed the private operator a �xed rate of a �rm with monopoly power can misuse it to the authorized to determine a “fair�rate of return but return based on shareholder equity and allowed detriment of consumers (chapter 6). A variety of lacked the complementary institutional safeguards for arbitration when the government and the approaches to reconcile these interests have that developed over decades in the United States. investor could not agree on rates. In 1995 developed around the world. In the United The commission became politicized,and despite Jamaica undertook more wide-ranging changes States the system involves giving substantial dis- increased inflation and the need to expand to its regulatory system for utilities, replacing the cretion to an independent regulatory agency, services,the private phone company was not Public Utilities Commission with a new Of�ce of with legislative guidance on tariffs often de�ned granted a single rate increase between 1962 and Utility Regulation.While the new agency has only as “fair� or “just.� Discretion of this breadth 1971.The company’s pro�ts fell and after 1970 some discretion, the new law retained a mecha- on an issue as politically sensitive as tariffs is a failed to cover the real depreciation of its assets. nism for providing speci�c pricing and other source of considerable risk to investors in capital- Service deteriorated and disputes developed,lead- commitments to investors through contracts, intensive sectors with immobile assets.Those ing to the company’s nationalization in 1974. thus helping to mitigate the risks of a traditional risks have been mitigated in the United States, With poor service and a shortage of funds U.S.-style agency operating in a country with less however, by a series of Supreme Court decisions, for investment under public ownership, the gov- developed institutional safeguards. dating from the 1890s, that have interpreted the ernment reintroduced private participation in Constitution in ways that create safeguards for the telephone company in 1985.This time, to Source: Spiller and Sampson (1996); Phillips (1993); investors in regulated industries. compensate for the lack of broader institutional and Jamaica Of�ce of Utility Regulation Act. (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 99 Tackling those problems requires a three- �t, or a combination of these. Regulation that pronged approach: imposes costs beyond the expected social ben- e�ts is usually regarded as red tape. • Addressing regulatory costs and infor- A growing body of evidence highlights mality the toll of outdated or ill-considered regula- • Reducing regulatory uncertainty and tions on the investment climate. Recent risk studies looking at the effect of regulation in • Removing unjusti�ed barriers to compe- Organisation for Economic Co-operation tition. and Development (OECD) economies show that both investment and the productivity of Addressing regulatory costs that investment are lower in countries where and informality the regulatory burden is greater.7 The effect All regulations can impose costs on �rms, can be large. For example, it has been esti- whether in the need to adapt business mated that reducing the burden of transport processes to meet regulatory requirements, to regulation in Italy to the level in the U.S. pay licensing fees, to await delays in obtaining could increase the investment rate in that regulatory approval, or to spend management sector by 2.6 percentage points.8 time dealing with of�cials. A good investment Recent work focusing on objective mea- climate does not seek to eliminate those sures of the compliance costs for particular costs—instead, it seeks to ensure they are no regulations highlights the wide variations higher than necessary to meet social interests across countries. For example, the World (box 5.3). The goal is thus better regulation, Bank’s Doing Business Project shows that the not no regulation. Too often the costs are time to set up a new business ranges from 2 unnecessarily high as a result of rent-seeking, days in Australia and 9 days in Turkey to inef�cient administration, poor institutional more than 200 days in Haiti.9 The overall BOX 5.3 Environmental regulation and global integration As it became easier for goods and investments ment decisions between two locations that are one, may be to the top rather than the bottom to flow across borders in the 1990s, concern otherwise highly comparable, such as states in as countries become more prosperous. arose that a race to the bottom in environmen- the United States or countries in Europe. Incentives to comply with higher tal regulation might follow. For goods that can But developing countries tend to face disad- standards are already strong be transported between countries, �rms might vantages relative to developed countries on this Multinational �rms often have stronger incen- choose to produce in locations with low envi- broader set of criteria, so differences in environ- tives to comply with higher environmental stan- ronmental standards and then export to coun- mental regulation tend to carry less weight. dards than local regulations require, both tries with higher standards.The concern is that Indeed, a recent study of foreign direct invest- because of advantages in adopting common countries with high standards would �nd them- ment (FDI) in developing countries found no technologies and standards across the countries selves at a disadvantage and, as capital left their evidence that environmental standards signi�- in which they operate, and also to protect their economy, would feel under pressure to relax cantly affected investment decisions. corporate reputations. Indeed, the evidence sug- their own standards to stem the outflow. Coun- Society’s preferences for higher standards gests that multinational �rms tend to exceed tries with already low standards might reduce rise with income local regulatory requirements in many areas. them further to vie for footloose investment.So As societies prosper, the value they place on Concerns about a possible race to the bot- far, however, there is little evidence to support higher environmental standards tends to tom need to be distinguished from the possibil- such concerns.There seem to be three main increase. Environmental quality appears to have ity of low environmental standards in one explanations. improved, rather than deteriorated, in many country reducing the environmental quality of Environmental regulation is only one part countries over the past decade. For example, air other countries by producing effluents that of the investment decision pollution in industrial areas fell in the 1990s in flow across national boundaries. The The cost of complying with environmental regu- Brazil, China, and Mexico—three developing international community has been addressing lation can influence �rms’ investment decisions, countries that have received signi�cant FDI. As these concerns in recent decades, including but it is only one of many factors, and the countries improve their broader investment cli- through a host of new international rules and weight given to it will vary by �rm, by industry, mates and experience faster economic growth, standards (chapter 9). and by location. Polluting industries tend to be there is likely to be pressure for more environ- capital intensive, which means investors tend to mental regulation, not less.The preferences of Source: Copeland and Taylor (2004); Wheeler (2001); Becker and Henderson (2000); Dowell, Hart, and place a high premium on the broader policy citizens in high income countries for high stan- Yeung (2000); Frankel (2003); Greenstone (2002); environment, particularly political and regula- dards of environmental protection also show no Jaffe and others (1995); Keller and Levinson (2002); tory risk. Costs associated with environmental signs of abating, further reducing the risk of a Klein and Hadjimichael (2003); and List and others regulation might carry more weight in invest- collapse in standards. Indeed, the race, if there is (2003). (c) The International Bank for Reconstruction and Development / The World Bank 100 WORLD DEVELOPMENT REPORT 2005 Figure 5.2 Starting a new business takes longer and is more more costly for them to hire professionals costly in developing countries to help them complete bureaucratic proce- 160 80 dures. Large �rms in Peru are almost three times as likely as small �rms to hire lawyers Cost of starting a business (left axis) to help them complete application proce- Days to start a business dures for licenses and permits.11 Other costs Cost (percent of GNI per capita) 120 60 are greater for large �rms: managers of large �rms spend more time dealing with govern- ment regulations, and large �rms are also more likely to be inspected than small �rms Days 80 40 (�gure 5.3). When it is costly to comply with regula- tion, �rms have an incentive to evade these 40 20 costs through informality. By staying infor- mal, �rms can reduce—but not completely eliminate—compliance costs (�gure 5.3). Informality is widespread in many develop- 0 0 Low-income Middle-income High-income ing countries, often accounting for more countries countries countries than half of GDP.12 The fact that most of Note: Based on median cost as percent of gross national income (GNI) per capita the economy is not complying with regula- and median days reported in the World Bank Doing Business Project. Source: World Bank (2004b). tions raises fundamental questions about the effectiveness of the chosen regulatory pattern is that delays are greater and costs strategy. higher in low-income countries (�gure 5.2). The answer is not simply to apply greater When compliance costs are the same for efforts to enforce all existing regulations. �rms of different sizes, they impose a dis- Unless the regulations themselves are well proportionate burden on smaller �rms. In considered, this may just put a dispropor- Tanzania small formal �rms, on average, tionate burden on poor entrepreneurs in the pay an amount equal to about 0.4 percent of informal economy and lead to perverse their sales for an operating license—large results. Efforts are required to �rst see if the enterprises pay only about 0.01 percent.10 regulation is necessary to meet an important Other regulations can also be a greater bur- social objective and, if so, whether the den for small �rms because it is (relatively) expected social bene�ts outweigh the likely Figure 5.3 Larger �rms spend more time dealing with regulations and are inspected more often costs. A growing number of countries are now focusing on reducing requirements for 20 20 business registration in this light, with posi- tive results. For example, when the munici- Percent of management pal government of La Paz, Bolivia, reduced Percent of management time 15 time spent dealing with 15 the number of procedures required to regis- Number of inspections government regulation ter a business, the number of registered busi- Inspections nesses increased by 20 percent.13 Even larger 10 10 gains have been observed in Vietnam and Uganda (box 5.4). Governments are also making efforts to 5 5 streamline other regulatory approval processes. This may involve using informa- tion technology that allows on-line process- 0 0 ing of regulatory approvals as in the case of Micro Micro Small Medium Large Extra informal formal (10–20) (20–50) (50–250) large Singapore (box 2.15) or the creation of (1–10) (1–10) (250 & up) “one-stop shops� (box 5.5). To encourage Firms & number of employees agencies to act upon approvals quickly, Note: Data covering �rms in Bangladesh, Brazil, Cambodia, Guatemala, India, Indonesia, Kenya, Pakistan, Tanzania, more countries are also adopting “silence as and Uganda. Source: World Bank Investment Climate Surveys and WDR Surveys of Micro and Informal �rms. consent� rules for some licenses and per- (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 101 BOX 5.4 Easing business registration requirements in Vietnam and Uganda The high cost of business registration discour- ing machines). It took 6 to 12 months to ful�ll 30 minutes.This reduced the cost of registering ages new �rms from entering the formal econ- the legal requirements to establish a business at a business by 75 percent. Although business omy. Vietnam and Uganda illustrate successful a cost of $700 to $1,400. registration is only one of several steps to start a strategies for reducing these costs. The new law reduced the costs of establish- new business in Uganda (businesses have to ing a new business.The time to establish a new register for tax purposes and many need addi- Vietnam business came down to about two months— tional licenses), the cost can be signi�cant Before a new Enterprise Law was enacted in Jan- with business registration taking only 15 days— because registration needs to be repeated uary 2000, business registration and licensing and total start-up costs were reduced to about annually for most businesses. requirements were extremely burdensome in $350. Vietnamese entrepreneurs responded. The pilot program increased business regis- Vietnam. Entrepreneurs were required to submit Fewer than 6,000 new businesses had registered trations, with an estimated four times as many detailed business plans, curricula vitae, character in 1999, but the number shot up to more than businesses registering in Entebbe the year after references, medical certi�cates, and other docu- 14,000 in 2000 and to more than 21,000 in both the pilot. Despite the lower fees, the higher ments along with their applications for registra- 2001 and 2002. number of registrations meant that revenue col- tion. On average, registering a business took lections increased by 40 percent. With adminis- about three months, and required visits to 10 Uganda trative savings of 25 percent in staff time and 10 different agencies and submissions of about 20 A recent pilot program in Entebbe reduced the percent in �nancial resources, the program also different documents with of�cial seals. time and monetary costs to register a business. bene�ted the municipal authority. Additional licenses were often required before By streamlining licensing processes and reduc- �rms could start operating. Some of these ing the number of previously required licenses did not appear to serve vital public approvals and assessments, the time to register Source: Vietnam: Mallon (2004); and Uganda: Sander interests (such as those to operate photocopy- a business was reduced from two days to about (2004). BOX 5.5 One-stop shops—or one-more-stop shops? In many countries �rms have to receive authority to grant some approvals and provide led to poor results, requiring the government to approvals from a range of different agencies assistance on others. For approvals that remain reorganize the center in the late 1990s. before they can start operating: one to register the responsibility of other agencies, the one- When agencies lack authority to grant all the business, another to register for taxes, stop shops may house staff from the relevant necessary approvals, it is important that they another to get environmental approvals, agencies or simply pass the applications on to still add value to the process and do not just another for health and safety clearances, and so them. Even when the staff from other agencies constitute an additional regulatory burden. In on.To reduce this burden some governments that are housed at the one-stop shop are unable Thailand the Investment Services Center could have established “one-stop shops� where �rms to approve the application themselves, they can issue establishment licenses for nonpolluting can �nd all the information and complete all the often facilitate the approval process. activities, but factories still had to get regulatory procedures that they need to start The Tanzania Investment Center houses nine permission from the Ministry of Industry before operating a business in a given jurisdiction. senior of�cials from other ministries, and normally production could actually start.To avoid delays One approach would be to give a single manages to turn around applications within a few later in the process, many �rms preferred to agency the power to grant all licenses, permits, days.The rapid turnaround is due in part to a “no obtain the necessary licenses directly from the approvals, and clearances necessary for a new objection�provision written into the investment ministry from the outset. �rm to start operating. In practice this is dif�cult. code—unless a ministry objects within 14 days, One-stop shops with narrower mandates Existing ministries and agencies often resist sur- the Center is entitled to approve the application. have sometimes accelerated the process of rendering their powers to a new agency. More- This approach has been less successful gaining speci�c approvals. For example, by shift- over, to the extent that approvals are a response when the lines of authority are not clearly ing from a pre-auditing to a post-veri�cation to a valid policy concern, the one-stop shop drawn. After being set up in 1987, the One-Stop system, the One-Stop Service Center for Visas would need to duplicate expertise and facilities Action Center in the Philippines housed repre- and Work Permits in Thailand reduced the time elsewhere in the government. Of course, if the sentatives from seven agencies who were it took foreign �rms to get visas for foreign approvals do not meet valid policy objectives, responsible for providing information to appli- workers from about 45 days to just 3 hours. the procedures could simply be eliminated. cants and acting on some applications. Lack of Because of these considerations, most one- effective agency representatives—and the non- Source: Bannock Consulting (2001); Brimble (2002); stop shops have narrower mandates, with reporting of some representatives to the Center Miralles (2002); and Sader (2003). mits.14 If the licensing of�ce does not Reducing regulatory uncertainty respond within a set period of time, the and risk license is issued automatically. The Bank’s Regulations can increase the risks �rms face Doing Business Project shows that business when the regulations change frequently, are registration takes an average of 28 days less vaguely drafted, or are interpreted or when a time limit is combined with a silent enforced inconsistently. The result in each consent rule.15 case is greater uncertainty, which makes it (c) The International Bank for Reconstruction and Development / The World Bank 102 WORLD DEVELOPMENT REPORT 2005 hard for �rms to make long-term decisions regulations take effect to enable �rms to about entering markets, choosing production adjust to the new requirements. When the technologies, or hiring and training workers. regulatory change could have a big impact Uncertainty can also reduce the response to on major investments made on the basis of otherwise bene�cial reforms. Evidence from earlier regulations, it may also be appropri- �rm-level surveys shows that improving the ate to grandfather those investments, or predictability of regulation can increase the provide a longer transition period. probability of making a new investment by more than 30 percent (chapter 2). Promoting certainty in the interpretation and application of existing regulations. Managing regulatory change. Of course, Uncertainty about how existing rules will be concerns about regulatory uncertainty do interpreted or applied can also be a signi�- not mean that regulations should never cant source of risk, and can be especially change. Indeed, there is a huge agenda for burdensome for �rms in capital-intensive change in most developing countries, and and heavily regulated industries. effective regulation requires regular review Firm-level surveys con�rm that con- and �ne-tuning to ensure it keeps up to date cerns about the predictability of regulation with changes in the way business is con- loom large for �rms in developing coun- ducted and lessons from experience. The tries. In many countries the majority of key is to minimize the adverse impact of �rms report that of�cials’ interpretations uncertainty on �rms. The best way to do were unpredictable (�gure 5.4). In most this is to consult �rms and other stakehold- countries, small and medium �rms were ers early in the process about proposed more likely than larger �rms to report that changes that are likely to affect them. This interpretations were unpredictable. can reduce the concerns of �rms, elicit use- The simplest strategy for improving pre- ful suggestions, and facilitate later imple- dictability is to ensure laws and regulations mentation. Yet �rm surveys show that the are drafted with as much clarity and preci- majority of �rms in developing countries sion as possible. While there are tradeoffs are seldom or never consulted on proposed between speci�city and discretion (box changes. More countries are now improving 5.6), it is often far from clear that the degree consultation, however, including by placing of discretion reserved to of�cials meets any draft proposals on the Internet. socially useful purpose. Indeed, in some In some cases it may be appropriate to cases discretion appears to be used more to provide a transition period before the new expand opportunities for of�cials to collect informal payments. Figure 5.4 Firms of all sizes report that of�cials’ interpretations of regulations are Some uncertainty is inherent in any new unpredictable law or regulation, but governments can 100 reduce uncertainty by quickly promulgat- ing more detailed regulations or imple- Percent of firms reporting interpretation mentation guidelines. The timely publica- 75 tion of regulatory and administrative decisions can also help build a body of is unpredictable precedents that can curb administrative 50 discretion and foster predictability. Small & medium Improving the transparency of regulatory Large decisionmaking can also do much to pro- 25 mote consistency—and reduce concerns that discretion will be misused. On complex or sensitive matters, an 0 advisory opinion or preclearance process Peru Moldova Zambia India Pakistan Nigeria Tanzania Turkey might be instituted—common for competi- Note: Firms designated as small or medium if they have fewer than 50 employees; and large if they have 50 employees or more. tion laws in many countries and a growing Source: World Bank Investment Climate Surveys. practice with complex tax issues. In some (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 103 cases it may be feasible to promote certainty by entering speci�c contractual commit- BOX 5.6 Balancing the tradeoffs between speci�city and ments on particular issues of interpretation discretion in regulation (box 5.7). Firms have a strong interest in regulatory cer- returns from the investment; and where Removing barriers to competition tainty.Without such certainty—both for the political economy problems can create stability and interpretation of rules—there incentives for governments to renege on Regulation also affects the investment climate can be concerns about the extent of their commitments (chapter 6). Regulatory dis- through its impact on competition. While regulatory obligations and thus the potential cretion may have a less deleterious effect individual �rms typically prefer less competi- returns from an investment opportunity. on investments that are more easily Providing �rms with appropriate assur- reversed, where regulation plays a minor tion, not more, competition plays a critical ances on the stability of the regulatory role in influencing expected returns, and role in the investment climate by creating regime can reduce their risks and thus where there are no special political sensitiv- opportunities for new �rms and providing encourage investment. Reducing discretion ities about regulation. But regulatory discre- incentives for existing �rms to innovate and can also reduce concerns about corruption. tion can still create uncertainty for �rms and But there can be tradeoffs. Highly speci�ed be used as a source of bribes by of�cials in improve their productivity. regulatory regimes reduce the flexibility to any sector. Much early evidence on the bene�ts of �ne-tune applications to particular cases, Concerns about regulatory discretion competition came from experience in and to accommodate changing can also vary by country. In the United States, OECD countries. For example, a study of circumstances. legislative guidance on the regulation of The optimal balance between infrastructure involves considerable discre- the impact of pro-competitive regulatory speci�city and discretion will vary according tion—but broader institutional safeguards reform in several industries in the United to the issue, sector, and country. For exam- help provide assurance to investors. Coun- States found that annual welfare gains in ple, highly discretionary regimes can have a tries that have not yet established credible the part of GDP affected by reform were chilling effect on private investment in safeguards for investor interests need to pro- infrastructure—where investments are vide more speci�c regulatory assurances—or more than 7 percent, with 90 percent of the large, long-lived, and immobile; where regu- expect reduced investment at higher cost to bene�ts flowing to consumers.16 New work lation has a signi�cant impact on the reflect the risks (see box 5.2). in developing countries shows signi�cant gains as well.17 For example, the bene�ts of greater competition from trade reform have Regulation has a pervasive impact on com- been documented in countries such as petition. Regulatory approaches that increase Brazil, Chile, Colombia, and India.18 Firm costs or risks faced by �rms can deter �rm surveys also show that competition plays a entry and thus dull competitive pressure. But much larger role in encouraging �rms to be regulation can also influence competition ef�cient than do customers, shareholders, more directly, including by creating barriers or regulators. The surveys also show that to market entry or exit, and by addressing �rms reporting strong competitive pressure anticompetitive behavior by �rms. are at least 50 percent more likely to inno- vate than those feeling no such pressure Regulatory barriers to market entry. Regu- (chapter 1). latory barriers to entry can take many forms BOX 5.7 Contracting for certainty One strategy governments can adopt to promote applied to major resource projects and extended af�liates. Because these transfer prices can be regulatory certainty is to enter speci�c contrac- to private infrastructure projects (where they manipulated to shift tax liability from one coun- tual commitments with �rms.While it is obviously often include speci�c commitments on tariff reg- try to another, tax agencies usually reserve the not feasible to do this with every �rm in the econ- ulation) and to other major investments. right to determine whether the prices reflect omy, this approach can be useful in dealing with Besides such global efforts to deal with policy market conditions.The methods for making risks associated with major investments. certainty, �rms often seek advance rulings and these determinations involve a good deal of During the �rst wave of foreign investment other forms of before-the-fact signals on how judgment, thus introducing much uncertainty after World War II, many �rms entered contracts government will interpret various laws and regu- into the calculation of the taxes due.To make with host governments that included “stabiliza- lations. One example is the transfer pricing agree- �rms’ tax bills more predictable, governments tion clauses.� Covering everything from tax rates, ments that developing and developed countries have entered advance agreements on the appro- to the duties payable on capital goods imported often sign with domestic and foreign �rms. priate level of transfer prices. China, Colombia, to develop a project, to the rules governing for- A major factor in determining a and Mexico have entered into hundreds of such eign exchange and pro�t repatriation, these multinational �rm’s income tax is whether agreements. India and Thailand are considering clauses sought to freeze in place those host gov- national tax authorities in the countries where it similar programs. ernment policies that could affect the return on operates will agree with the prices it uses to the investment.These approaches have been transfer goods and services among its corporate Source: Waelde and Ndi (1996) and Tropin (2003). (c) The International Bank for Reconstruction and Development / The World Bank 104 WORLD DEVELOPMENT REPORT 2005 and have many rationales. Requirements to helping their communities while improving set up a new business are one obvious form their own livelihoods (chapter 6). When of entry barrier, but can be designed in ways Bangladesh introduced competition in cellu- that are not especially burdensome. But lar phone services, one of the new entrants unnecessarily high registration costs can encouraged female entrepreneurs to set up still have a negative impact on competition. and run phone shops in rural areas. By 2004 For example, estimates for a group of devel- these shops provided service to about 5,000 oping countries—none of them the worst villages and an estimated 12.5 million people offenders—suggest that reducing the cost of who previously had no access to this ser- registration procedures to the level in the vice.23 Barriers have been lifted even more in United States (0.6 percent of per capita Uganda, opening new opportunities for income) could increase the number of new small entrepreneurs across the country and entrants by more than 20 percent.19 expanding service in rural areas. Governments often erect more substan- tial regulatory barriers to entry in particular Regulatory barriers to market exit. Compe- industries. Some of these may be part of a tition is also affected by barriers to �rms strategy to address a market failure but are leaving the market. The most pervasive bar- vulnerable to being made more onerous rier to exit is bankruptcy regulation. When than necessary through rent-seeking by the those procedures are long and costly, dis- protected groups. Other restrictions lack tressed �rms and their creditors are less any clear economic rationale. Public enter- willing to use them, and markets become prises also often bene�t from legislated cluttered with failed �rms that block oppor- monopolies. tunities for new entrants. Firms will also be In India the manufacture of certain less likely to risk entering new markets, and products is reserved for small �rms, reduc- lenders will be less willing to lend to �rms ing opportunities for other �rms to partici- they do not already have a relationship pate—and reducing incentives for small with, further reducing competition.24 As a �rms to grow (box 8.5). Agricultural mar- result, long and costly bankruptcy proce- kets in many countries have been heavily dures have a negative impact on productiv- regulated, with parastatals granted monop- ity—over 20 percent of productivity gains olies over marketing or processing of export can be attributed to the least productive crops, and traders who purchase goods �rms exiting (chapter 1). from farmers required to be licensed. Bankruptcy procedures tend to be longer Recent efforts to liberalize agricultural mar- and more expensive in developing countries kets have, for the most part, bene�ted poor than in developed countries. A standard bank- rural producers of export crops by increas- ruptcy procedure takes an extraordinarily long ing producer prices relative to border time in some countries. According to the prices.20 While supply responses have some- Bank’s Doing Business Project, a procedure times been slower than expected, this seems that takes only �ve months in the fastest coun- to reflect continuing impediments in other try (Ireland) would take 10 years in Brazil, parts of the investment climate (including India, and Chad. The costs can also consume a insecure property rights and poor infra- large share of the estate. While taking only structure)21 or concerns about the credibil- about 1 percent of the estate value in several ity of the government’s commitment to lib- countries (Colombia, the Netherlands, Nor- eralization.22 way, and Singapore), they take up to 76 per- Removing unjusti�ed regulatory barriers cent in Chad and Lao PDR. Bankruptcy proce- to entry can have a big impact not only on dures also appear less likely to result in ef�cient competition but also on opportunities for outcomes (rehabilitating viable businesses and individual entrepreneurs. For example, liquidating unviable businesses) in developing reducing regulatory barriers to competition countries. A growing number of developing in telecommunications has created opportu- countries are recognizing the importance of nities for microentrepreneurs to enter the reform in this area, with recent examples market and provide services in rural areas, including Bulgaria, India, and Poland.25 (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 105 Addressing anticompetitive behavior by �rms. that markups were no different in countries Regulation is not the only source of barriers to with and without competition laws.30 While competition. Firms can curb competition by agencies in countries such as Brazil, Chile, colluding or forming cartels, by entering Korea, and Mexico have achieved some restrictive agreements with suppliers or cus- standing, implementation in many other tomers, by misusing their market power, or countries has so far been less impressive. simply by merging with competitors. Recent work suggests that while competi- To address these concerns, a growing tion laws in developing countries tend to be number of countries have introduced com- no weaker than in developed countries, petition (or antitrust) law.26 While the competition policy is perceived to be much details vary, most competition laws include less effective (�gure 5.5). Why? Limited provisions to do the following:27 resources and slow and inef�cient courts are part of the story. Perhaps more impor- • Prevent �rms from colluding or forming tant, however, are other policies that reduce cartels to limit competition. Prohibited competition (such as regulatory barriers to actions typically include agreements to �x entry and exit) and the politics of prosecut- prices, restrict output, allocate markets ing �rms that have close ties to the govern- and customers, and rig bids or tenders. ment, such as state-owned enterprises and • Prevent dominant �rms from abusing �rms owned by influential people (box 5.8). their market positions by engaging in predatory pricing, forcing �rms that buy Toward better regulation for the particular goods or services to also buy investment climate other goods or services, foreclosing mar- The challenge of regulatory improvement kets for inputs or distribution, or setting is large and ongoing. It requires continu- discriminatory prices or terms of service. ing efforts to review and modernize • Require proposed mergers to be approaches in line with changes in the way reviewed by a specialist agency to ensure business is conducted and lessons of expe- that any resulting reduction in competi- rience, but doing so in ways that provide tion has offsetting public bene�ts. as much predictability as possible for �rms. This is true in all countries, but it is Competition laws are usually enforced by specialist agencies. In addition to their Figure 5.5 Despite strong laws, competition policy is seen to be less roles in enforcing competition law, the effective in low-income countries agencies often act as advocates for competi- 6 20 tion by commenting on policy proposals by Perceptions about competition policy other government agencies and performing Strictness of competition law Perceived effectiveness of competition policy studies to make policy recommendations on competition-related issues (chapter 3). 15 Strictness of competition law According to a recent survey, 65 percent of 4 43 responding agencies participate early in the regulatory review and decision process, 10 while 28 percent were consulted through- out the process or at any stage.28 Indeed, some argue that competition advocacy 2 should be the �rst priority of competition 5 agencies—particularly in economies with a legacy of heavy-handed government inter- ventions.29 0 0 Competition laws are relatively new in Low-income Lower- Upper- High-income developing countries and early results pre- countries middle-income middle-income countries sent a mixed picture. A recent study that countries countries Note: For competition policy data, higher values mean more effective policy; for competition looked at price markups in a number of law, the higher values mean stricter laws. developed and developing countries found Source: World Economic Forum (2002) and Nicholson (2003). (c) The International Bank for Reconstruction and Development / The World Bank 106 WORLD DEVELOPMENT REPORT 2005 BOX 5.8 Competition laws in developing countries Given the importance of competition to a sound One explanation might be constrained tion agency in Tanzania forbade a local brewer investment climate, competition laws and agen- resources. For example, the competition from barring independent agents and mini- cies could be expected to play a key role. How- agency in Tanzania had only two economists wholesalers from stocking competitors’ ever, experience in developing countries remains and no lawyers in 2000, while the authority in products, the �rm, with support of government mixed.There are several possible explanations. Zambia had four economists and one lawyer. A of�cials, contravened the agency’s orders. When First, competition laws do not usually second explanation is that enforcement often of�cials intervene against agency decisions on address barriers to competition flowing from depends on effective courts. Unless the com- behalf of influential �rms, competition agencies government policy in other areas—including petition agency can rely upon the judiciary to will be hesitant to move against them in the �rst trade barriers, mandated monopolies, licensing support its decisions and protect it from politi- place. regimes, and other regulatory barriers to entry cal interference, the agency will �nd it dif�cult The main message? Well-designed competi- and exit. When those barriers are pervasive— to enforce its rulings. tion laws can be an important tool to improve still the case in many countries—competition A third explanation is that it can be dif�cult the investment climate. But they need to be laws and agencies will not be enough to to prosecute politically connected �rms, even seen as part of a broader strategy that includes unleash a competitive and productive economy. when the competition agency is independent, reducing regulatory barriers to competition, and The primary lever for governments is to address unless the law and the agency command a high helping to promote a more pro-competition the policy barriers directly. level of public support. For example, when the culture. And as elsewhere, a high level of politi- Second, competition laws are not always independent Monopoly Control Authority in cal commitment is key. enforced vigorously in developing countries. Pakistan tried to take action to reduce carteliza- Although agencies in some countries appear tion in the cement market in 1998–99, the gov- to be quite active, others appear to be less so ernment intervened, �xing prices at a “mutually Source: CUTS Center for Competition (2003) and (see table). Why is enforcement often weak? acceptable� level. Similarly, when the competi- Economic and Social Research Foundation (2002). In some developing countries competition agencies deal with very few cases India Kenya Pakistan South Africa Sri Lanka Zambia (1999) (1996–2000) (1996–2000) (1999) (1996–2000) (1998–2000) Total cases disposed of annually 206 30 166 273 6 50 Mergers and acquisitions 0 22 16 236 1 22 Anticompetitive practices 206 8 149 37 6 28 Cases per professional 9.0 1.3 33 7.4 0.9 24.8 Source: CUTS Center for Competition (2003). especially important in developing coun- jobs. All societies struggle with how best to tries where the existing body of regulation strike the balance in an ef�cient, equitable, too often bears little relationship to con- and sustainable way. This section reviews temporary circumstances, is only partially the nature of the challenge and highlights enforced, and if enforced more vigorously some promising areas for improvement. could lead to even more perverse results. As highlighted in chapter 3, tackling the Taxes and the investment climate regulatory reform agenda requires efforts Throughout history, governments have to systematically review existing regula- raised revenues in many ways. They have tions, as well as assessing new regulatory seized the assets of their enemies—and their proposals more carefully. Strengthening subjects. They have created monopolies to the skills and expertise of regulators and sell to the highest bidder. They have taxed those on the front line of government- land, production, transactions, income, and �rm relations also plays an important consumption—and in most cases still do. role. Indeed, income taxes are fairly recent. The �rst income tax, levied by the Dutch Batavian Taxing �rms Republic, dates from 1797,31 but the United Governments need revenue to cover the States did not have a corporate income tax costs of providing public services—includ- until 1909 or an individual income tax until ing those that improve the investment cli- 1913.32 The value added tax (VAT) is even mate—and of meeting other social goals. more recent—the �rst was levied in France in Yet taxes represent a cost to �rms and so 1948, and it did not become common until reduce their incentives to invest and create the 1970s and 1980s.33 (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 107 For as long as governments have levied Table 5.1 Firms report that tax rates are one of their top concerns taxes, those who pay them have com- Share of countries where �rms report tax rates as key obstacle plained. Firms in developing countries are no exception, and cite tax rates as a major Biggest Among top Among top obstacle three obstacles �ve obstacles constraint on their operations (table 5.1). Taxes affect the incentives for �rms to invest All countries 18 56 82 productively by weakening the link between Upper-middle-income 40 90 100 effort and reward, and by increasing the cost Lower-middle- income 12 35 71 Lower-income 11 56 83 of inputs used in the production process. Tax Eastern Europe & Central Asia 14 62 86 rates and compliance costs both matter. Sub-Saharan Africa 33 67 83 When levied or applied unevenly, taxes can Asia 14 29 71 also distort competition. Latin America 50 50 50 Note: Reports share of countries where �rms rank tax rates as a top constraint in a list of 18 possible obstacles. Source: World Bank Investment Climate Surveys. Tax rates. Tax rates are a function of the size of government and the way the burden is Figure 5.6 Corporate tax and VAT rates are similar in high-income and developing countries allocated among alternative sources. While VAT rate Corporate tax rate views on the appropriate size of government differ, government’s share of GDP in many High-income countries developing countries is much larger than in Upper-middle-income countries today’s developed countries when they were at similar stages of development.34 The share Lower-middle-income countries of the tax burden carried by �rms can be Low-income countries influenced by ef�ciency and equity consid- erations, as well as by more pragmatic con- Middle East & North Africa cerns about collecting revenue.35 Narrow tax bases and weak tax administrations lead South Asia governments in developing countries to col- Sub-Saharan Africa lect a larger share of their revenues from �rms and from commercial transactions Europe & Central Asia than is the case in developed countries. Indeed, corporate taxes, direct taxes on Latin America & the Caribbean goods and services, and trade taxes account East Asia for over 70 percent of government revenues 0% 10% 20% 30% 40% in low-income countries.36 Note: Data are for 1999–2000. While tax rates and structures differ Source: World Bank (2004k), and Ebrill and others (2001). across countries, corporate tax rates and value-added tax rates are broadly similar in Figure 5.7 Corporate tax revenues remained stable or increased during the 1990s, except in ECA developing and developed countries (�gure 5.6). Despite similar rates, revenues col- High-income countries lected from corporate taxes tend to be lower in developing countries than in developed South Asia countries due to the narrowness of the tax base and problems of tax administration Sub-Saharan Africa (�gure 5.7). Corporate tax revenues either increased slightly or remained stable during Europe & Central Asia the 1990s in all developing regions except Europe and Central Asia, where revenues Latin America & the Caribbean fell due to privatization and a general con- 1993–94 traction in the size of the state.37 This is Middle East & North Africa 2000–01 contrary to some of the dire predictions of 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 those concerned about the impact of tax Percent of GDP competition between countries as a result of Note: Averages are for 81 countries for which comparable data was available for both periods. increasing global integration (box 5.9). Source: IMF (2003); OECD (2002d); Dobrinsky (2002). (c) The International Bank for Reconstruction and Development / The World Bank 108 WORLD DEVELOPMENT REPORT 2005 BOX 5.9 Taxation and global integration: A race to the bottom? Concern is often expressed about whether com- from wasting public resources or becoming tion (�gure 5.7). Further, whether the decrease in petition for investment between countries is overly intrusive. marginal rates is a result of tax competition or leading to a race to the bottom in corporate tax Other theoretical models suggest that tax other factors is not clear—governments might rates. Competition might pressure governments competition might have some adverse conse- reduce rates in an attempt to stimulate private to cut corporate taxes to attract new investment quences. One concern is �scal externalities. investment by local �rms. or retain existing investment.The concern is When a government cuts its tax rates on capi- The dire predictions of some commentators greatest for investment by �rms that are the tal—and does not cut expenditures that owners may not be bearing out for two reasons: most footloose, such as multinational �rms pro- of capital care about (if it cuts only expenditures • Tax rates are not the only factor influencing ducing tradable goods. that bene�t immobile workers)—it might investment decisions. Infrastructure, law and attract capital from neighboring jurisdictions. If Do tax rates affect where �rms invest? order, and the education of the workforce can it does not take into account the effect of this The answer seems to be yes, but like other be even more influential, and it is hard for on taxes (and thus expenditures) in the neigh- aspects of the investment climate, the weight governments to sustain those services with a boring jurisdictions, it can set tax rates lower will likely vary between �rms, industries, and shrinking tax base. Location decisions are also than are globally optimal. A second concern is locations. A meta-analysis of 25 studies that influenced by agglomeration economies. that tax competition might have an undesirable looked at the effect of tax rates on FDI (mostly Together, these factors mean that investment impact on the distribution of taxes. In particular, using data on FDI into the United States or FDI is not as responsive to changing tax rates as if capital is mobile but workers are not, a greater by U.S. �rms) concluded that a one percentage some fear. part of the burden of corporate taxes will fall on point increase in tax rates reduces FDI by about • Corporate tax rates also affect the taxes paid workers rather than on capital. 3.3 percent. Other surveys and evidence by domestic �rms and �rms producing non- A host of other factors—such as other tax support a similar conclusion. tradable goods, and investment by these instruments available to the government—also Is tax competition harmful? affect whether tax competition improves, or �rms is likely to be far less responsive to dif- Because corporate taxes affect the decisions of reduces, public welfare in theoretical models of ferences in tax rates than investment by for- investors, countries might try to use tax rates to the economy.The broader point, however, is that eign �rms, especially those producing traded compete for foreign investment. International tax competition is not necessarily harmful. goods.This means that across-the-board cuts tax competition can have both positive and in corporate tax rates would be a costly way negative effects on welfare and ef�ciency, and it Have corporate taxes fallen as international to attract foreign investment. Rather than cut- is not immediately clear that it will make coun- economic integration increased? ting taxes across the board, governments tries worse off. Allowing countries or regions to If tax competition was resulting in signi�cant tend to offer tax incentives—or other advan- set taxes and expenditures based on local pref- �scal externalities and thus a race to the tages—targeted speci�cally to �rms thought erences for and costs of providing local public bottom, corporate taxes should have fallen in to be the most responsive (chapter 8). goods (ones that affect people only in that juris- the 1990s as international integration increased. diction) is generally more ef�cient than requir- Although marginal corporate tax rates have Source: Baldwin and Krugman (2004); Brennan and ing that governments mandate uniform taxes fallen over the past decade, bases have often Buchanan (1980); De Mooij and Ederveen (2001); De and expenditures across regions. Many been broadened. As a result, corporate tax rev- Mooij and Ederveen (2002); Devereux, Grif�th, and Klemm (2002); Glaeser, Johnson, and Shleifer (2001); commentators also argue that a degree of com- enues have increased or remained steady on Gordon and Hines (2002); Haufler (2001); Hines petition between governments on taxes and average, except in the European transition (1999); Mitra and Stern (2003); Oates (2001); Rodrik other policies can be a good thing, because it economies, where the decrease in revenues was (1997); Tiebout (1956); Wilson (1999); and Wunder disciplines governments and prevents them more from privatization than economic integra- (2001a). The burden that taxes impose on �rms BOX 5.10 Who pays taxes levied on �rms? can vary along several dimensions. First, When governments levy taxes on �rms, �rms gests that raising corporate taxes is neces- because �rms can partially pass the costs of will often pass the costs of the tax on to oth- sary to make �rms “pay their fair share,� taxation on to consumers or workers, the ers. For example, if government levies a pay- labor bears a large part of the burden of actual burden can differ from the statutory roll tax on �rms, increasing the cost of hiring corporate tax in the United States. Because workers, �rms will hire fewer workers. As labor’s share of the corporate tax burden is burden (box 5.10). Second, many �rms and unemployment increases, real wages will fall higher when capital is more mobile, labor activities bene�t from special tax exemp- (or increase more slowly than they would may bear a greater part of the burden in tions or privileges, whether as a result of have otherwise), passing the cost of the tax developing countries than it does in the government deliberately trying to promote on to workers. So workers ultimately bear United States. As capital becomes more some of the tax burden in the form of lower mobile—and multinational �rms become some kinds of activity—as is often the case wages, even though the tax is levied on the more sophisticated in their tax with foreign investment and research and �rm. Part of the burden might also be passed minimization strategies—the share of the development (chapter 8)—or as a reward to on to consumers through higher prices. corporate income tax falling on labor will favored constituencies. Third, a large pro- Incidence has been especially contro- likely increase. versial for corporate taxes. Although the portion of �rms in many developing coun- corporate income tax is often seen as a tax Source: Fuchs, Krueger, and Poterba (1998); tries are in the informal economy, where on capital, and the popular press often sug- Mulligan (2002); and Rosen (1995). they typically do not pay taxes. This includes microentrepreneurs, but weak (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 109 enforcement capacity means that even Figure 5.8 Caught in the middle: taxing �rms in Uganda and Cameroon larger �rms evade at least some taxes. Cor- ruption in tax administration contributes 10 Cameroon to informality, resulting in less revenue for Uganda government and a higher burden on those 8 that do pay. Taxes as a percent of sales Small �rms can often reduce their tax burden through informality and evasion. 6 Large �rms can also reduce taxes because of their ability to negotiate various tax privi- leges and to avoid taxes through sophisti- 4 cated legal means (hiring accountants to search for existing loopholes in the tax sys- 2 tem). This can lead to a disproportionate burden for medium �rms. For example, they pay a greater share of their revenues in 0 2–5 6–25 26–75 76–200 Over 200 taxes than either small or large �rms in Number of employees Cameroon and Uganda (�gure 5.8).38 Source: Gauthier and Reinikka (2001) and Gauthier and Gersovitz (1997). Tax administration. Firms rate tax adminis- tration as a separate and additional obstacle from tax levels. In countries including almost twice as high as in smaller �rms, Bangladesh, Brazil, and Ethiopia, more than small informal processors can undercut the 50 percent of �rms said that tax administra- prices of the large �rms by evading taxes and tion was a very severe or major problem not complying with all regulations.40 (�gure 5.9). Red tape and corruption in tax administrations are common, and weaken Better taxes for the investment the incentives to comply with taxes and climate contribute to leakages. Crafting better tax policies for the invest- ment climate requires governments to rec- Taxes and competition. Taxes can also affect ognize the tradeoffs between ef�ciency, the level of competition between �rms in equity, and pragmatic implementation con- two main ways. First, many developing cerns, and the impact of tax policies have on countries have traditionally relied heavily on the incentives of �rms to invest produc- trade taxes (tariffs and export taxes), in part tively, create jobs, and so contribute to a because of the ease of collection, which has reduced competitive pressure on local �rms. Figure 5.9 Many �rms rate tax administration as a To take advantage of global integration, gov- serious obstacle ernments have been reducing trade taxes with a positive impact on the competitive Brazil discipline facing local �rms—and reducing Ethiopia costs for �rms and consumers. They have Bangladesh typically made up for the lost revenues by introducing or increasing VAT.39 Kyrgyzstan The second way taxes influence competi- Guatemala tion is through differential treatment of local China �rms in the same market. As noted above, medium �rms may be disadvantaged relative Czech Republic to smaller and larger �rms. Firms in the 0 20 40 60 80 informal sector can have advantages over Percent those in the formal sector. In Argentina, for Note: Percentage of �rms reporting tax administration as a example, it has been suggested that although “major� or “severe� obstacle to the operation and growth of their establishment. labor productivity at large meat processors is Source: World Bank Investment Climate Surveys. (c) The International Bank for Reconstruction and Development / The World Bank 110 WORLD DEVELOPMENT REPORT 2005 growing tax base over time. A �rst step is to from businesses (box 5.11). In Mongolia ensure the tax burden is no higher than some local governments issue awards, includ- necessary, including by keeping the size of ing consumer goods, cash, and plaques to the state in check and striving for more ef�- �rms nominated as the best taxpayers. ciency in public spending. For example, World Development Report 2004 identi�ed Simplifying tax structures. Simplifying many opportunities for governments to complicated tax systems can be bene�cial better leverage public funding for public for three main reasons. First, tax systems services. Beyond this, the most promising riddled with exemptions are not transpar- strategies involve broadening the tax base ent and can act as magnets for rent-seeking (including by addressing informality), sim- behavior by �rms and other groups. While plifying tax structures, and improving tax this bene�ts the favored groups, it reduces administration in its various dimensions. revenues and puts a greater burden on oth- ers. Second, such systems can provide sig- Broadening the tax base. Reducing impedi- ni�cant opportunities for corruption.43 ments to the emergence of new �rms that Third, complicated systems increase the contribute to growth expands the tax base cost of administration. Large �rms can and creates the potential to reduce the tax devote resources to reducing their total tax BOX 5.11 Tax burden on other �rms. Addressing infor- burden. This in turn increases the burden of receipts as lottery mality of existing �rms can require a more administration for the agencies responsible tickets? nuanced approach. For larger �rms that for administering taxes and auditing evade tax obligations, more vigorous returns. Simplifying the tax system is espe- Shop owners sometimes have enforcement action is justi�ed, but compli- cially useful in countries where administra- problems with employees who pocket the customer’s cash rather ance can also be encouraged by simplifying tive capacity is limited or control of corrup- than putting it into the register. tax structures and tax administration. Sev- tion is weak. To discourage employees from eral countries in Eastern Europe are also doing this, some stores and fast experimenting with flat corporate and per- Increasing the autonomy of tax agencies. A food restaurants offer customers a small amount if the checker fails sonal taxes to encourage tax compliance, common strategy for improving revenue to issue them a receipt. By giving reduce distortions, and simplify adminis- collection and reducing compliance costs the customer an incentive to tration. Reducing impediments to �rms is to give tax agencies more autonomy. Since report employees who fail to joining the formal economy—including by autonomous tax agencies were introduced enter sales into the register, the owners effectively enlist the cus- simplifying business registration require- in Bolivia and Ghana in the 1980s, more tomer in their attempts to ments and relieving other unjusti�ed regu- than 15 countries have set them up.44 prevent employee theft. latory burdens—can also play a role. Autonomous tax agencies promise better In 2002, to boost tax collec- tions, the city government of Bei- performance than traditional ministries. jing, China, instituted a similar Confronting informality. Microenterprises They can bypass restrictive civil service rules program to encourage in the informal economy raise more dif�cult and pay better salaries to attract and retain enterprises to issue proper and sensitive issues (chapter 3). Some small well-quali�ed professionals.45 They are also receipts. Under this program, a small scratch box was added to �rms may not be viable if they have to com- better protected from political interference.46 of�cial receipts. When the ply with all taxes and regulations.41 Forcing Autonomy usually improves the perfor- customers scratch the box, they them to comply might simply result in them mance of revenue agencies.47 A recent study can win small prizes ranging closing down, with an adverse impact on of agencies in Latin America and Africa between 100 and 5,000 Yuan.To discourage forgery, a second poverty. And even a big increase in formality concluded that the agencies granted the scratch box with a code number among microenterprises may not lead to a most autonomy were the most successful in allows customers to check over signi�cant increase in revenues but would boosting revenue collection and ef�ciency, the Internet whether the greatly increase the cost of collecting taxes.42 increasing compliance, and improving ser- business gave them a valid receipt. In a pilot program out- Governments are experimenting with vice quality.48 After the reform of the Kenya side Beijing a small town novel schemes to improve tax morality. In Revenue Agency in 1995, revenue ef�ciency increased tax revenues by China, to encourage businesses to issue of�- and compliance improved and, despite an $732,000 while giving out cial receipts, some local governments have across-the-board reduction in tax rates, rev- $17,100 in prizes. experimented with a scheme that allows of�- enues declined by less than had been fore- Source: The Economist (2002b). cial receipts to double as lottery tickets, to cast.49 But sustaining autonomy requires a encourage customers to demand receipts high level of political commitment.50 (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 111 Autonomy also has to be balanced with after it became autonomous.54 But auton- accountability. Although an autonomous omy is not a universal salve: for example, agency needs to have control over its day- corruption remained a serious problem in to-day operations (deciding whom to hire Tanzania after the reform of its revenue and whom to audit), it is important that it agency.55 remains accountable for its overall perfor- mance, including its relationship with tax- Improving compliance through computeri- payers. In Mexico the autonomous agency zation. Increasing computerization in rev- has to present a report on its performance enue administration agencies can some- to the legislature three times a year. In times help.56 Singapore reduced tax arrears Kenya the head of the tax authority is and staff turnover, while public satisfaction required to present quarterly audit reports, with the tax service improved.57 But experi- conducted by the internal audit unit, to the ence suggests that increased computeriza- agency’s board, the minister of �nance, and tion is likely to be successful only when part the auditor general. The agency head is also of an overall strategy that takes into account required to present the agency’s �nancial civil service wage structures and human statements, performance indicators, and capital constraints.58 Computerization pro- annual report to both the board and the jects tend to be more successful when minister of �nance. The auditor general implemented with other reforms to also conducts an annual audit, which the improve tax administration.59 Using off- minister of �nance presents along with the the-shelf software and hardware can also annual report, to the National Assembly.51 reduce the risks of having to develop pro- prietary technologies.60 Tackling corruption in tax administrations. Corruption in the tax authority under- mines collection efforts. Corruption can be Regulating and taxing a persistent challenge because the problems at the border are rarely unique to tax administration. But In addition to regulating and taxing �rms governments can take several practical within their borders, governments regulate steps.52 One general principle is to mini- and tax goods at the border and impose mize direct contact between tax of�cials additional regulations and restrictions on and taxpayers—by automating and com- foreign-owned �rms. puterizing procedures, increasing the use of Although the regulation of domestic third-party data for assessments, and rely- transactions can often be justi�ed on ef�- ing on tax withholding.53 A second useful ciency grounds, such as addressing a market step is to organize the tax agency along failure, similar arguments rarely apply to functional lines (such as auditing, taxpayer restrictions on trade or FDI. Apart from rev- assistance, and processing tax returns) enue goals for import tariffs, policies in this rather than by tax type, because this makes area are often driven by the preferences of it harder for of�cials to develop relation- local �rms to face less competitive pressure. ships with taxpayers. Broader strategies for A growing appreciation of the bene�ts of addressing corruption in civil service orga- openness has resulted in both developed and nizations can also help, such as allowing developing countries signi�cantly reducing independent internal and external audits, barriers to trade and investment in recent protecting whistleblowers, and giving citi- years (chapter 3). However, many barriers zens a way of complaining about harass- that weaken the investment climate remain. ment (chapter 2). In some cases corruption also appears to Regulatory barriers to foreign have been reduced when agencies have investment become autonomous. In Peru, 85 percent Since 1995 at least 60 countries have made of taxpayers surveyed believed that there regulatory changes affecting foreign invest- was substantially less or much less corrup- ment every year, with the vast majority tion in SUNAT, the Peruvian tax agency, reducing restrictions (�gure 5.10). (c) The International Bank for Reconstruction and Development / The World Bank 112 WORLD DEVELOPMENT REPORT 2005 Figure 5.10 Most changes in national regulations governing FDI reduced restrictions of such arrangements is mixed at best (chapter 8). Regulations more favorable to FDI Second are those that seek to exclude or Regulations less favorable toward FDI Number of countries making changes otherwise more tightly control foreign par- 250 80 ticipation in sectors perceived to be espe- cially “sensitive�—such as infrastructure 200 and media services. For example, the United 60 States restricts foreign ownership of radio Number of regulations Number of countries licenses and prevents majority foreign– 150 owned companies from operating domestic 40 air services.61 Although many middle- income countries maintain few restrictions 100 on foreign ownership in manufacturing, they often impose greater restrictions on 20 50 foreign ownership in electricity, telecom- munications, transportation, and �nancial services (�gure 5.11). Given the bene�ts of 0 0 foreign ownership in improving productiv- 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 ity, and the fact that many domestic �rms Source: UNCTAD (2003e). rely on the services from the restricted sec- tors, restrictions can weaken the investment Figure 5.11 Restrictions on FDI have fallen in manufacturing, but persist in other sectors climate. A third objective may be to control the potentially destabilizing effects of large, Manufacturing short-term capital flows—with the empha- sis on short-term portfolio investment Electricity rather than FDI (box 5.12). Telecommunications Regulatory barriers to foreign trade Tariff and nontariff barriers to trade have Transportation been reduced over the past decade, but the remaining restrictions and weaknesses in Finance customs administration still have a big impact on the investment climate. Business services Mexico Trade protection. Average tariff rates remain Czech Republic moderately high in developing countries Poland Hotels Hungary (13 percent).62 It has been estimated that if developing countries reduced their average 0 0.2 0.4 0.6 0.8 1.0 tariffs to 10 percent on agricultural prod- Restrictions on FDI ucts and to 5 percent on manufacturing (higher values mean more restrictions) products, their gains would exceed $100 bil- Source: Golub (2003). lion by 2015. This is greater than the gains developing countries would get from devel- oped countries reducing the tariffs and Restrictions that discriminate against other restrictions they impose on goods foreign investors usually have one of three from developing countries (chapter 10).63 objectives. First are those that seek to encourage FDI but also to promote Improving customs administration. When spillovers to the local economy by impos- customs are administered poorly, signi�- ing requirements to enter joint ventures cant costs can be imposed on �rms with local �rms or to meet other require- engaged in importing or exporting—and ments. Experience with the effectiveness indirectly on �rms that supply exporters (c) The International Bank for Reconstruction and Development / The World Bank Regulation and taxation 113 BOX 5.12 Dealing with short-term international capital flows Although most countries now actively court flows into their economy. Governments in many percent reserve deposit in an unremunerated FDI, there is more debate about the merits of developing countries, including those in Latin account for up to a year for all portfolio inflows capital account liberalization, particularly for America, have run procyclical �scal policies, con- from abroad. It also required that FDI stay in the short-term capital flows. Recent crises in Asia, tributing to cycles of booms and busts. Avoiding country for at least three years—a restriction Latin America, and Russia have contributed to overspending and overborrowing during reduced to one year in 1992. the debate, with many observers questioning booms is thus important. Evidence on the effectiveness of capital con- whether it is wise to allow short-term invest- Strengthen oversight of the �nancial system. trols is mixed. Some studies have found that ment to flow freely in and out of developing One way to reduce problems associated with capital controls have altered the composition of countries. capital inflows is to improve management of capital inflows, increasing the share of FDI and Most of the debate has focused on short- �nancial sector risk. In addition to ensuring that decreasing the share of short-term and portfolio term portfolio investment. FDI—especially banks are adequately capitalized and have investment. Other studies found that capital green�eld investment—is dif�cult to reverse. appropriate levels of provisioning for bad loans, controls can have harmful side effects. Because Portfolio flows, in contrast, can change direction it is important to ensure they do not develop they impose costs on foreign investors whether very quickly, putting pressure on exchange rates portfolio mismatches in currencies or terms. they restrict inflows or outflows, controls gener- and fragile banking sectors and sometimes Banks might also have to be discouraged from ally increase the cost of borrowing in the coun- causing currency or banking crises. What can lending foreign currency to �rms with earnings try. Further, because controls can often be governments do to insulate themselves from primarily in domestic currency (those operating circumvented, especially in countries where cor- these reversals without deterring all foreign in nontraded sectors). Removing implicit or ruption is a problem, it is unclear whether they investment? Several proposals have been put explicit government deposit insurance might are an effective way of deterring crises. forward, some more controversial than others. also be valuable. Avoid overspending and overborrowing dur- Capital controls. Regulations aimed at pre- ing periods of rapid inflows. Although several venting sudden outflows of investment or dis- Source: Schmukler (2003); World Bank (2002d); recent crises have been the result of private bor- couraging short-term inflows are more contro- Ariyoshi and others (2000); de Ferranti and others rowing (Asia in 1997), governments often con- versial. Several countries have experimented (2000); Edwards (1999); Kaminsky, Reinhart, and tribute to crises by overborrowing from interna- with capital controls. In 1991 Chile imposed a Végh (2003); Montiel and Reinhart (1999); and tional capital markets as foreign investment requirement that foreign investors make a 20 World Bank (2001f ). or depend on imported goods. Delays in many areas in most countries, bribes were imports can also prevent �rms from common for �rms that reported applying adopting production processes that rely for licenses. Around 10 percent of �rms that on just-in-time deliveries and mean that applied for import licenses reported that �rms have to hold larger inventories than bribes were requested or expected when they would otherwise. Firms in Estonia reported that, on average, imports cleared Figure 5.12 Clearing customs for imports—from under 2 days to 18 customs in less than 2 days. By contrast, the average for �rms in Tanzania was 18 Exports Imports days and in Ecuador, 16 days (�gure 5.12). Tanzania These delays can impose real costs on Ecuador workers and �rms in developing coun- tries: on average �rms in the garment China industry grew more slowly, in both output and employment, and wages were lower in Russia countries where customs clearance took longer.64 India Corruption can also be a major problem Honduras in customs administration. Of�cials can impose large costs on importers—especially Slovenia for importers of perishable goods—by delaying the processing of imports. In Morocco Eastern Europe and Central Asia more than Estonia 20 percent of �rms that directly imported some inputs reported that bribes were 0 5 10 15 20 needed to deal with customs and imports. Days Although import licenses are not needed in Source: World Bank Investment Climate Surveys. (c) The International Bank for Reconstruction and Development / The World Bank 114 WORLD DEVELOPMENT REPORT 2005 applying for them, with the median pay- BOX 5.13 Reducing customs delays in Singapore ments exceeding $100 in several countries. and Ghana Improving customs administration promises large gains. Increasing the use of Firms in developing countries often face Singapore’s success, and a similar program information technology can help accelerate long delays when importing and exporting in Mauritius, inspired the government of goods. In recent years computerization has Ghana to adopt a similar program as part of its customs processing (box 5.13).65 Comput- demonstrated the potential to dramatically strategy to become a more attractive location erization is becoming less costly and less speed up parts of the process. One initia- for exporters.Before the program, importers demanding of human capital than before tive uses software and procedures based estimated that the fastest clearance time at on a program called TradeNet. Rather than seaports was four days, with an average clear- because of standardized software packages. submit multiple unique forms to multiple ance time of several weeks.After implement- In addition to reducing delays, computeri- agencies, a trader can electronically submit ing the program, about 14 percent of zation can increase transparency and so a single document that contains all the clearances took less than a day at Tema port reduce corruption.66 Importers in Morocco information required by the different and only 11 percent took more than �ve days. agencies. TradeNet then submits the infor- At the airport, average clearance times fell now �nd out in real time the progress of mation to the relevant agencies, which can from three days to four hours, with 18 percent customs operations and the status of their then respond with the necessary permits of clearances taking less than two hours. imports under special import regimes, or request additional information. By elimi- Although computerization can reduce monitoring payments of duties and taxes, nating overlapping requirements and mul- delays, it will not succeed unless procedures tiple forms, the process reduces are modi�ed to fully exploit its bene�ts. and even monitoring clearance times.67 transaction costs for �rms and minimizes Before implementing TradeNet, the Ghana- Customs can also be improved by con- direct contact between public of�cials and ian customs administration was already tracting out functions to private �rms as in the trader, reducing opportunities for side using a standard software package to help Mozambique (box 5.14). payments. process imports, but procedures were not Singapore used these methods in 1989 designed to take advantage of the package, to reduce processing time from two to four so the technology was underused. For days to a few minutes, and the number of example, customs declarations had to be Government approaches to regulation and required documents from between 3 and manually entered into the database, a taxation are not limited in their impact on 35 to a single document. Freight forwarders process that took up to 24 hours, rather estimate that the program has reduced than being submitted electronically. product markets. They also play a big part in their cost of handling trade documentation the quality of a country’s �nancial system and by between 20 and 35 percent. Source: De Wulf (2004), and World Bank (1998b). its infrastructure—the subject of chapter 6. BOX 5.14 Contracting out customs in Mozambique Before 1995 customs administration had been a • Introducing a new salary scale and compen- slight decline in 2001).The reform also helped serious problem in Mozambique.There was no sation package that was higher than for other the investment climate. By 2002 the median reliable system for detecting and punishing cor- civil servants and that compared well with number of days for imported goods to clear cus- rupt of�cials. More than three-quarters of staff private sector salaries toms was signi�cantly lower in Mozambique lacked a high school education.There was little • Adopting a new software package and new than in Tanzania or Kenya and similar to the use of information technology, and all goods were computer hardware number in China. physically inspected after arriving in the country. Some questions remain. It is not clear • Reducing the agency’s reliance on physical So revenue collection was poor.The inspection whether the improvements can be sustained inspections process was slow. Corruption was serious, with after Crown Agents leaves. In 1999 Crown importers and customs of�cials frequently collud- • Adopting anticorruption measures. Agents’ three-year contract was extended until ing to undervalue and misclassify imports. 2003 and then extended again until 2005. In 1995 the government initiated an ambi- In addition the government, with support Crown Agents’ responsibilities and number of tious program to improve customs operations. from the U.K. Department for International staff have declined since the �rst contract, but a The program included the following: Development (DFID), entered a contract with review by DFID and the Mozambique govern- Crown Agents, a private company, which took ment concluded that the improvements were • Issuing a new customs code to update the pre- over the management of customs in 1996. not yet sustainable in mid-2003. vious law, which dated from the colonial period Even with a reduction in nominal tariff rates, • Replacing many workers with better- better administration and reduced exemptions educated personnel, while boosting employ- increased the ratio of customs revenue to ment by 20 percent imports between 1996 and 2000 (there was a Source: Mwangi (2003). (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 6 Financial markets, when functioning well, ment interventions have made matters connect �rms to lenders and investors willing worse. Financial markets have been chapter to fund their ventures and share some of the risks. Good infrastructure connects them to repressed and distorted by state ownership, monopolies, directed or subsidized credit, their customers and suppliers and helps them and other policies appealing to the short- take advantage of modern production tech- term interests of politicians and favored niques. Conversely, inadequacies in �nance groups. Those measures undermine �nan- and infrastructure create barriers to opportu- cial sector development, �rm-level produc- nities and increase costs for rural microentre- tivity, and economic growth.1 Infrastruc- preneurs as well as multinational enterprises. ture provision has been undermined by By impeding new entry into markets, these governments using state ownership or regu- inadequacies also limit the competitive disci- lation to pursue objectives unrelated to ef�- pline facing incumbent �rms, dulling their cient service delivery—typically favoring incentives to innovate and improve their pro- some groups over broader interests and ductivity. Such inadequacies are large in introducing new sources of inef�ciency.2 developing countries (�gure 6.1). The problems in both areas usually hit The underlying problem with both smaller �rms the hardest. �nance and infrastructure can be traced to a Governments are confronting these speci�c market failure—for �nance it is issues, but progress is slow and uneven. information asymmetries, and for infra- They are pursuing new approaches that rec- structure, market power associated with ognize that �nance and infrastructure are economies of scale. But too often govern- not only part of the investment climate for other �rms, but are also profoundly shaped Figure 6.1 The inadequacies of �nance and infrastructure are severe for many developing countries by the investment climate for providers of �nancial and infrastructure services. That is Latin America & the Caribbean why many governments are taking steps to increase competition among providers of �nance and infrastructure, secure their Middle East & North Africa property rights, and regulate them in ways that recognize the tradeoff between market Sub-Saharan Africa failures and government failures. Govern- ments are also working to improve manage- South Asia ment of public resources—to get more for their money when they �nance or subsidize Europe & Central Asia infrastructure services. Finance East Asia & Pacific Infrastructure Financial markets Developed �nancial markets provide pay- 0 10 20 30 40 50 60 70 ment services, mobilize savings, and allo- Percent of firms cate �nancing to �rms wishing to invest. Note: Figure shows the share of �rms that report access to �nance, and any of electricity, telecom- munications, or transportation as "major" or "severe" obstacles to the operation and growth of their When these markets work well, they give business. Source: World Bank Investment Climate Surveys. �rms of all types the ability to seize promis- 115 (c) The International Bank for Reconstruction and Development / The World Bank 116 WORLD DEVELOPMENT REPORT 2005 Figure 6.2 Sources of �xed investment �nancing differ for small and large �rms Getting �nancial markets to work well, State sources however, runs into market failures and State sources 1% 5% Other problems of political economy.9 Market Other 13% Banks 5% 12% failure arises mainly from information Equity 5% asymmetries. Firms seeking to borrow promise to repay loans, but there is always a Family & chance they will not. If lenders could accu- friends 11% Internal funds 70% rately estimate the likelihood of default, Internal funds 52% Banks they could protect themselves by calibrating 22% interest rates to the risk of default. Lenders do charge more for riskier loans, but the fact that their knowledge of risk is imper- Family & friends 4% fect, and poorer than that of borrowers, Small firms Large firms means that increasing interest rates cannot Note: The data are for �rms in Bangladesh, Brazil, Cambodia, Guatemala, India, Indonesia, Kenya, Pakistan, Tanzania, and Uganda. Small �rms are de�ned as those employing fewer than 10 people, large �rms as those employing 50 or fully protect them: when lenders charge more. higher interest rates, they discourage bor- Source: World Bank Investment Climate Surveys, WDR Surveys of Micro and Informal Firms. rowers with low-risk, low-return ventures, ing investment opportunities. They reduce leaving them mainly with borrowers for �rms’ reliance on internally generated cash high-risk projects. By its nature, then, rais- flows and money from family and friends— ing interest rates increases the risks lenders giving them access to external equity and are exposed to. The problem is heightened debt, something that smaller �rms in par- by the possibility of dishonesty and weak ticular often lack (�gure 6.2). They allow contract enforcement—only honest bor- poor entrepreneurs to grow their busi- rowers are discouraged by high interest nesses, even though they have little money rates. themselves. Well-functioning �nancial mar- Providers of debt and equity also have kets also impose discipline on �rms to per- imperfect information about what the form, driving ef�ciency, both directly and recipients are doing with the capital. by facilitating new entry into product mar- Lenders cannot be sure that borrowers are kets. And they create opportunities for steering clear of risks that increase the �rms and households to manage risks. As a chance of default. Shareholders cannot be result, �nancial market development leads sure whether managers are investing wisely to faster growth in productivity and out- or merely enriching themselves. put.3 Doubling private credit as a share of These failures can make it hard for �rms GDP is associated with an increase in aver- to obtain �nancing unless they have collat- age long-term growth of almost two per- eral to secure a loan—or good connections. centage points.4 Failures also make it hard for people with Developed �nancial markets also reduce savings to �nd attractive opportunities to poverty—directly and through their role in invest or lend. The severity of the failures economic growth. They reduce income depends partly on factors outside govern- inequality by alleviating credit constraints ment’s immediate control, such as the effect and increasing access to investment oppor- of technology on the costs of getting better tunities for poor households.5 By facilitat- information, but it also depends on govern- ing competition between �rms that pur- ment policy. chase goods produced by poor households, Financial markets are also affected by they can help poor households escape political economy. Government policies exploitation by those �rms.6 They can also toward �nancial markets are influenced by stabilize the economy by reducing volatility: the wishes of powerful groups and the self- doubling private credit as a share of GDP interest of politicians. Competition often can reduce the volatility of growth from suffers from that influence. In the United four percent a year to three.7 There is also States, until the mid-1990s, state banks per- evidence that child labor is lower in coun- suaded governments to shelter them from tries with greater access to �nancing.8 competition by maintaining unwarranted 116 (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 117 restrictions on interstate banking. And in Japan until the mid-1980s established banks BOX 6.1 Governments and �nance markets: A long and persuaded the government to protect them dif�cult history from competition from bond markets by Throughout history governments in need of Not all governments solved their �nan- maintaining a rule that required would-be funds have found it convenient to expropri- cial dif�culties through taxation and wide- bond issuers to �rst get approval from a ate the �nancial assets of their citizens, spread protection of property rights. In committee that the banks controlled.10 often by repudiating debt. In England the Mexico in 1876, President Por�rio Díaz was Financial markets have a long history of cycle of expropriation was broken only confronted with the twin problems of politi- when the monarchy recognized that the cal disorder and economic stagnation. He similar problems (box 6.1). Overcoming the sums from taxing production on private needed resources to combat his political problems presents policymakers with a property outweighed those from periodic opponents immediately, but Mexico’s long challenge at least as dif�cult as that created expropriation.The Crown �rst seized and history of government defaults made bor- by information failures. sold vast lands owned by its rivals—the rowing from the private sector impossible. church and the nobles—thus creating a He could have forced loans and con�scated market for land. A dispersed landholding property, but that would hurt productivity Avoiding the pitfalls of traditional gentry then emerged, which used in the long run. government interventions parliament as a coordinating mechanism to Díaz opted instead to protect the rights protect their economic interests. of a select group of asset holders and use Responding to market failures and political Over time the economic might of the the rents generated to combat his political pressures, governments in the post–World gentry grew so much that they could openly opponents.The largest bank, Banamex, the War II period intervened heavily in �nancial defy the Crown and the nobles in government’s primary �nancier, enjoyed markets—directing credit to favored groups, parliament, in part because their wealth special protections, including reserve ensured that they could hire their own army requirements half those of other banks, guaranteeing loans by private banks, and pro- if necessary.The gentry thus used Parliament exemptions from taxes, and the sole right to viding many �nancial services themselves to ensure that the Crown honored its com- open branches. While these arrangements through state-owned banks and development mitment to respect property rights, the basis might have suited Díaz, the lack of contesta- for their economic prosperity, despite occa- bility in �nancial markets would dampen �nance institutions (DFIs). To protect domes- sional attempts to renege. A credible com- growth throughout the 20th century. tic banks, governments also restricted compe- mitment to respect and enforce property tition from foreign banks and other �nancial rights helped the government borrow vast Source: Rajan and Zingales (2003) and Haber, institutions. They often justi�ed state owner- sums to �nance the British Empire. Razo, and Maurer (2003). ship and other interventions in the �nancial sector as ways of ensuring that small and rural borrowers had access to funding. The overall �rms particularly.13 Although their impor- record of these interventions is discouraging. tance has been diminishing, state-owned banks remain signi�cant in many parts of State ownership of banks. State-owned the developing world (�gure 6.3). banks can be given broad mandates or the task of developing a speci�c industry, sec- Development �nance institutions. By subsi- tor, or region—often making loans at subsi- dizing credit to customers unable to borrow dized rates. Their performance in the devel- from traditional banks, DFIs can be justi- oping world has generally been poor. �ed if they overcome a market failure cost- Having a large proportion of state owner- effectively. A few have been able to lend ship in the banking sector has been found pro�tably and maintain high repayment to reduce overall access to �nancing, reduce rates without the use of traditional collat- competition, worsen the allocation of eral.14 More often, they have supported credit, and increase the likelihood of �nan- political projects with little economic value cial crises.11 Studies of bank privatization in or bene�ted favored constituencies. They Brazil, Egypt, and Nigeria �nd less govern- usually lack disciplining tools, such as active ment ownership is associated with better pro�t-motivated shareholders. Because bank performance.12 State-owned banks are they raise funds through the tax system or frequently associated with weak gover- government-guaranteed borrowing rather nance, corruption, and poor procedures for than through deposits, they often have a collecting debts from borrowers. As cross- weak sense of the cost of capital. country studies show, state ownership of Improvements in governance can begin banks, by impeding private competition, to change this. For example, the Thai Bank can also impede the development of the for Agriculture and Agricultural Coopera- �nancial system, hurting small and medium tives is an unusual case of a development (c) The International Bank for Reconstruction and Development / The World Bank 118 WORLD DEVELOPMENT REPORT 2005 Figure 6.3 State-owned banks are holding on, especially in India and in the where lending is discouraged by sparse pop- Middle East and North Africa ulations, high transaction costs, and a lack 100 of traditional collateral. Japan, Singapore, 90 and South Korea appear to have had some success with directed lending for manufac- 80 turing, but experience in most countries has at private commercial banks 2003 Percentage of bank assets 70 1995 been poor.18 60 Directing credit for some purposes 1985 means restricting it for others. In Colombia 50 in the 1980s, a subsidized credit scheme 40 required �rms to show that they needed 30 new capacity, so credit was directed to the building of new plants—and away from 20 improving the ef�ciency of existing plants. 10 Such direction works only when of�cials are better than bank managers—which is rarely 0 India Middle East Asia Europe & Latin Developed the case—at deciding whether new plants East & & Pacific Central America are more valuable than greater ef�ciency. North Asia & the Africa Caribbean Directing credit, even when it may help Note: “East Asia� includes South Korea, Malaysia, Philippines, and Thailand. “Europe and Central meet some social objective, is dif�cult in Asia� includes Bulgaria, Croatia, Czech Republic, Hungary, Poland, Romania, Russia, Slovakia, and practice because it pushes against the cur- Slovenia. “Latin America & the Caribbean� includes Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Panama, Peru, Trinidad and Tobago, rent of market forces. Lenders and borrow- Uruguay, and República Bolivariana de Venezuela. “Middle East & North Africa� includes Algeria, ers want to lend and invest where the Egypt, Jordan, Lebanon, Morocco, and Tunisia. “Developed� includes Australia, Austria, Belgium, Canada, Denmark, Finland, Germany, Greece, Italy, and Japan. returns are greatest, not in sectors deemed Source: Data for 2003 are from Clarke, Cull, and Shirley (2003); data for 1985 and 1995 are from La Porta, López-de Silanes, and Shleifer (2002). a priority by the government. So lenders reclassify loans to comply with the direc- tions, and borrowers surreptitiously use bank with mandated lending objectives that credit for unintended purposes. Both does not depend on subsidies, yet succeeds lenders and borrowers might bribe of�- in providing credit to farmers. In 1998 it cials to turn a blind eye. And as in South extended loans to more than 80 percent of Korea during the height of its enthusiasm Thailand’s farming households.15 Its gover- for directed credit, markets can develop for nance arrangements hold local managers borrowers with access to directed credit to accountable for their branch’s performance on-lend to those without it. In the and require managers to meet pro�t targets. extreme, directed-credit policies merely More often, however, DFIs make poor reallocate wealth and leave the ultimate quality loans and fail to ensure their repay- allocation of credit unchanged. For these ment. A study of 18 industrial DFIs found reasons, directed credit often fails to reach that almost 50 percent of their loans were in its intended bene�ciaries.19 arrears.16 Credit does not always reach dis- Directing credit has also slowed the advantaged borrowers, either. In Brazil the development of �nancial markets. Many rural �nance credit program provides more directed loans go to unpro�table projects than 57 percent of its loans to the largest 2 and are not repaid. Some borrowers simply percent of borrowers, only 6 percent to the refuse to repay their loans, hoping that smallest 75 percent of borrowers. Interest being in a favored sector will protect them rate subsidies and low repayment rates also from court action. And large, diverse �rms strain government budgets. Mexico injected can operate an internal credit market, almost $23 billion into agricultural DFIs bypassing the political direction of credit between 1983 and 1992.17 and cutting banks out of the picture. So banks suffer losses and �nancial markets Directed lending. Governments have often falter.20 Reflecting this experience, govern- directed banks to lend to speci�c regions ments are now backing away from directing and sectors, often singling out rural areas credit. (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 119 Credit Guarantees. Credit guarantees offered providers of �nance �nd it more pro�table by governments can encourage more lending to hold government securities than to make to riskier borrowers, including new and loans to �rms. For example, in Brazil, the small �rms. But shifting the risk of default to expansion of government borrowing taxpayers raises several practical challenges. between 1995 and 2003 was associated with Because guarantees encourage banks to a slowdown in expansion of private sector worry less about credit risk and monitoring credit. borrowers, default rates can be high, raising issues of sustainability.21 To better align Fostering competition. Restrictions on com- incentives, programs can be designed that petition between providers of �nance can require banks to share in the default risk, mean slower economic growth, reduced involve independent screening of loan appli- employment growth, and slower exit of cations, and impose fees that are high mature �rms in concentrated bank mar- enough to discourage banks from using the kets.24 Policies that impede competition— guarantee for loans that do not need it. These such as entry restrictions, restrictions on measures increase the costs facing borrow- foreign banks, and state ownership of ers, however, and so reduce participation by banks—hurt the �nancial system and eco- smaller �rms, which are often the intended nomic performance. Removing these barri- bene�ciaries.22 While many countries have ers to competition has been shown to experimented with these schemes, examples improve banking stability, reduce interest of unambiguous success stories in develop- margins, and expand access to �nance.25 ing countries remain scarce.23 One way to foster competition is to (prudently) issue new domestic banking Better approaches licenses. In the United States the wave of Governments are learning from the past mergers and acquisitions in the 1980s and and taking new approaches that involve �ve 1990s created large banks, which reduced key elements: lending to new and small �rms. Yet fairly liberal licensing policies allowed new banks • Ensuring macroeconomic stability to form to help offset the lack of supply and • Fostering competition keep interest margins low.26 Competition is • Securing the rights of borrowers, credi- also bene�ting from technological innova- tors, and shareholders tion, as in India’s rural areas (box 6.2). • Facilitating the flow of information Policymakers are sometimes concerned • Ensuring that banks do not take exces- that the competition from foreign banks sive risks. will weaken the banking system. However, evidence shows that foreign banks improve Ensuring macroeconomic stability. Macro- the ef�ciency and performance of domestic economic stability—more speci�cally, low banks and reduce interest rate margins.27 inflation, sustainable debt, and realistic This is what happened when the Philip- exchange rates—is fundamental to the pines allowed more foreign bank competi- effective functioning of �nance markets. tion—interest rate spreads fell and the ef�- Macroeconomic instability increases the ciency of domestic banks increased.28 volatility of interest rates, exchange rates, Foreign banks can also use their cross- and relative prices, imposing additional border experience to introduce innova- costs and risks on �nancial institutions and tions. Citibank responded to the scarcity of their clients. High inflation erodes the capi- good credit information on individual tal of �nancial institutions and makes it dif- �rms in many developing countries by �nd- �cult to mobilize savings or to expand ser- ing other ways to assess creditworthiness. vices. High �scal de�cits increase interest The company identi�es industry segments rates and spreads. The increase in holdings with the potential to grow quickly and then of government paper by banks, mutual seeks out borrowers in those segments. In funds, and investment funds crowds out India it has about 500 customers in 15 credit to the private sector, because these selected industrial segments. (c) The International Bank for Reconstruction and Development / The World Bank 120 WORLD DEVELOPMENT REPORT 2005 domestic banks did, and in Argentina and BOX 6.2 Expanding access to �nance in rural areas—new Chile, real growth in lending to small �rms approaches in India was higher for foreign banks.29 While bank-to-bank competition is Firms operating in rural areas often have a ers and lenders.They make it easier to get important, other sources of �nance can also hard time getting �nancing, but �nancial credit and renew loans, once the initial innovations and new technology are mak- screening has been done.They reduce the strengthen competition. For example, �rms ing a difference, as India shows. number of visits to branches, and they with access to public bond �nancing have The agricultural agency model uses a increase the operation of accounts at desig- 35 percent more debt (after controlling for third-party intermediary to coordinate the nated supply branches. �nancing of inputs, the delivery of produce to The increasing sophistication of �nan- other �rm characteristics).30 Nonbank the end buyer, and the repayment to the bank cial markets is helping farmers smooth their �nancial intermediaries can also broaden before the farmer receives the proceeds.The incomes in the face of fluctuating prices �nancial markets. For example, leasing intermediary improves information by advis- and harvests. Fledgling futures markets are companies and �nance companies often ing farmers on crop decisions that affect the allowing them to �x the prices they will quantity and quality of the produce.The inter- receive in advance. Innovations in insurance �nance start-up �rms unable to raise funds mediary can also negotiate better prices on are allowing them to protect themselves from banks. As nonbank �nancial interme- �nal goods than individual farmers can. from losses caused by poor weather.The diaries develop, they often securitize their The Kisan Credit Card, offered by com- payouts are based on an index measuring assets, further deepening securities mar- mercial, rural, and cooperative banks, is a local weather, which allows an objective technological innovation in providing credit determination of the payout and maintains kets.31 Pension funds and contractual sav- to the agriculture sector in India, including farmers’ incentives to maximize their output ings can also compete to supply funds, small farmers. Since its introduction in despite poor weather. increasing banking ef�ciency and lowering 1998–99, some 31.6 million cards had been the cost of capital.32 Finally, commercial issued by April 2003.Though not truly credit Source: Hess and Klapper (2003) and World cards, the cards have advantages for borrow- Bank (2004j). micro�nance is beginning to have an impact on �nancial services for microentre- preneurs and poor households (box 6.3). A second concern is that foreign entry How, then, to encourage the develop- might reduce access to �nancing by small ment of nonbank lenders? By not overregu- and medium �rms. But again, foreign banks lating lenders that do not take deposits, and have been found to improve access to credit by harmonizing the tax treatment of �nan- for those �rms. In Chile and Peru, foreign cial products. In Turkey, factoring compa- banks loaned more to small �rms than nies pay a 5 percent transaction tax while BOX 6.3 Commercial micro�nanciers enter the market Micro�nanciers provide thrift, credit, and other reduce their vulnerability to external shocks. It banks. In 1992 ProDem, a micro�nance �nancial services of very small amounts, mainly can also empower the poor, especially women. nongovernmental organization (NGO), became to the poor, in both rural and urban areas.They Subsidized micro�nance relying on donors, BancoSol, the �rst commercial bank in Latin offer an alternative to banks, which in most however, is unlikely to be big enough to reach America dedicated to micro�nance.The trans- developing countries serve only 5–20 percent of all potential borrowers.That will require formation enabled the expansion from 14,300 the population.They use noncollateralized loans commercial micro�nance that mobilizes the sav- clients to 70,000 within �ve years of to deliver short-term working capital to ings of the general public, raising questions commercialization, and by 1998 BancoSol was microentrepreneurs and households. about the appropriate role for governments. the most pro�table licensed bank in Bolivia. One of the key characteristics of micro�nance, Governments are sometimes tempted to As in other segments of the credit market, pioneered by Grameen Bank in Bangladesh and mandate below-market interest rates, but this allowing the sharing of credit information now replicated throughout the developing world, usually causes more problems than it solves.The among microlenders can foster micro�nance is substituting joint liability, access to future loans, removal of interest rate controls in Indonesia in lending, especially by commercial lenders that and frequent repayment periods for traditional 1983 allowed Bank Rakyat Indonesia to experi- may not have preexisting relationships with collateral.These alternatives to collateral are espe- ment with new �nancial products, most notably borrowers in rural areas. South Africa has two cially important for borrowers who do not have market-priced working capital and investment private credit bureaus operating in the micro�- assets to pledge—and for lenders who operate in capital loans. By 1986 its micro�nance business nance sector. Information can be obtained by countries with weak secured-lending laws and had turned from a chronic loss-maker to a prof- touch-tone phone, and the micro�nance enforcement. itable department. bureaus charge much lower fees than larger Micro�nance has demonstrated its success Governments can also eliminate unfair com- bureaus—making them affordable even for in reducing poverty. By 2002 more than 1,000 petition from public institutions and change small microlenders. micro�nance programs around the world had regulations to facilitate competition on a level reached about 30 million borrowers, lending playing �eld. In particular, they can allow micro- Source: Ghatak and Guinnane (1999); Morduch about $3.5 billion, with an average loan size of �nance institutions to transform themselves (1997); Morduch, Little�eld, and Hashemi (2003); $280. Micro�nance has helped the poor increase into licensed �nancial institutions and facilitate Hubka and Zaide (2004); CGAP (1997); Klapper and household income, build viable businesses, and the provision of micro�nance by commercial Kraus (2002); and www.mixmarket.org. (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 121 banks pay only 1 percent.33 Pension rules States show that small �rms are 25 percent can also be liberalized as capital markets more likely to be denied credit if they are in mature and regulatory systems develop. states that provide creditors with less pro- For instance, investment in more asset tection when the borrower is bankrupt.42 classes, such as equities, can be allowed.34 The effectiveness of creditor rights also Better insurance regulations can also depends on strong enforcement of the laws. encourage insurance providers to innovate Russia, for example, has “imported� strong and operate ef�ciently—and to create a laws protecting shareholder and creditor competitive market open to new �rms and rights, but the lack of an effective legal sys- the exit of insolvent �rms.35 Mutual funds tem to enforce these laws has been a big can be developed under strong accounting impediment.43 Laws and registries permit- and auditing rules and strict disclosure ting the collateralization of movable prop- requirements.36 erty can offer even greater bene�ts to smaller �rms that are less likely to have Securing the rights of borrowers, creditors, �xed assets (box 6.4). and shareholders. Governments can miti- The need for strong shareholder rights gate the problems for creditors and share- and good corporate governance has been holders—and increase their willingness to underscored by structural changes in most provide �nance—by ensuring that the par- developing countries—including privati- ties have clearly de�ned rights and can zation and the widespread listing of �rms enforce them.37 A strong legal environment on stock markets.44 Improvements in cor- and strong enforcement are important for porate governance are associated with access to external �nance and the develop- higher operational performance of �rms, ment of �nancial markets. When creditor through better management, better alloca- rights are weak, �nancial institutions will be tion of resources, and other ef�ciency less willing to extend credit to �rms that improvements.45 Governance is particu- have a high risk of default. When share- larly important for foreign investors, who holders’ rights are weak, investors will be may have informational disadvantages. A less willing to provide �rms with equity.38 global investor opinion survey by McKin- Securing borrowers’ property rights to sey suggests that good governance matters assets they can pledge as collateral (includ- most to investors (ranking higher than ing land) can increase access to �nancing �rm performance or growth prospects) and investment (chapter 4). Secure prop- and that institutional investors prefer to erty rights also allow �rms to borrow longer-term and encourage more foreign lending.39 The cost of external �nancing is BOX 6.4 Establishing a registry for movable collateral in also lower in countries with stronger prop- Romania erty rights protection and less corruption. A Legal impediments previously restricted all registered security interests.Ten opera- study of 37 countries found that if a coun- the use of movable property as collateral in tors and 366 agents are licensed to register try improved its property rights protection Romania and thereby limited the access to collateral in the electronic archive.The from the 25th to the 75th percentile, loan credit. First, the system did not allow supervisory authority provides guidelines lenders to access information on whether on the archive’s operation and clari�es rules spreads would decline by 87 basis points.40 other creditors or lenders had claims on and regulations. Strong creditor rights—stemming, say, the same goods. Second, the enforcement The archive functions ef�ciently, allow- from laws guaranteeing secured creditors’ of agreements and repossession of collater- ing �nancial intermediaries to access infor- priority in the case of default—allow alized goods was a long process (often mation about creditors, debtors, or assets exceeding the economic life of the securing a commercial or civil transaction in lenders to reduce their risk of future losses, movable good). the country.This information, accessible by therefore encouraging them to make more A new law, adopted in 1999, introduced people all over the world, presents huge loans. For example, one explanation a system for registering security interests. cost-saving and time-saving offered for the low level of private credit in The registration, valid for �ve years, is opportunities—improving the investment required to secure new collateral.The law climate. Mexico is that many social constituencies provides for both stronger enforcement must be repaid before secured creditors, and a new electronic archive of outstanding Source: Fleisig (1998) and Stoica and Stoica often leaving creditors with few assets to liens.This online collateral registry includes (2002). back their claims.41 Studies in the United (c) The International Bank for Reconstruction and Development / The World Bank 122 WORLD DEVELOPMENT REPORT 2005 invest in countries where legal rules and explain why the Czech Republic, whose gov- enforcement are both strong.46 ernment took a relatively hands-off In countries where laws do not guaran- approach to the enforcement of regulation of tee strong protection of shareholders, �rms the capital markets, had an inactive equity may be able to improve their access to exter- market—while Poland, which had stricter nal equity �nancing by voluntarily improv- enforcement of regulation and disclosure, ing their governance through greater trans- witnessed strong growth in its capital mar- parency, preparing �nancial reports ket.49 In countries with developed �nancial according to international accounting stan- intermediaries—such as brokers, accounting dards, and appointing independent direc- �rms, and investment advisers—exchanges tors. So governance standards need not be may be able to delegate some disclosure legislated for all corporations. Governments enforcement to these intermediaries and can still facilitate shareholder monitoring reduce the cost of enforcement. In emerging by requiring all large and listed �rms to dis- markets, however, government prosecution close �nancial and ownership information. may be necessary to protect investors and Stricter regulation (in the form of high dis- promote market development. Internation- closure requirements set by the stock ally agreed principles for corporate gover- exchange or government) and strong nance create opportunities for governments enforcement are associated with greater to signal the quality of their regulatory sys- market liquidity, lower costs of capital, and tems in this area (chapter 9). higher valuations of �rms (box 6.5).47 Transparency and disclosure require- Using credit bureaus to facilitate the flow of ments for listed �rms are generally set and information. One way lenders can address supervised by the local exchange, but the their information disadvantage is to collect government may need to enforce exchange information about their customers directly standards.48 Differences in enforcement help through costly screening and monitoring. Lenders in most developed countries—and more now in developing countries—can also rely on reports from credit information BOX 6.5 Improving corporate governance in Brazil and bureaus. These reports include loan payment South Korea histories that allow lenders to use information on how borrowers met past loan obligations South Korea is leading corporate governance exchanges usually loosen listing reforms in East Asia. Ceilings have been requirements. But Novo Mercado goes to better predict future loan performance. removed on foreign ownership.The against this trend, requiring corporate gov- Credit reporting also improves borrowers’ minimum shareholding required to under- ernance requirements far stricter than in incentives to repay loans promptly, because take class actions has been reduced, prompt- the old segment. At least 25 percent of the late payment with one lender can result in ing many instances of shareholder activism capital stock must be floating in the market (for instance, People’s Solidarity Participatory and listed companies must adopt interna- sanctions by many institutions.50 Democracy challenged Samsung Electronics tionally recognized accounting standards Credit bureaus can increase bank lend- and SK Telecom).The appointment of (U.S. generally accepted accounting princi- ing and reduce default rates. They also ben- outside directors on the boards of �nancial ples or International Financial Reporting institutions and major conglomerates is Standards). In a merger both controlling e�t small and new �rms by alleviating credit required. Some exchange listing and minority shareholders must be treated rationing based on the lack of a credit his- requirements were also added, which apply equally.The companies can issue only com- tory.51 In one survey more than half the to �rms with an asset size greater than W2 mon shares—something particularly impor- credit bureaus indicated that credit history trillion (about $2 billion).Those �rms must tant in Latin America, where the use of non- have an audit committee with at least two- voting preferred stock is commonplace and information reduced the processing time, thirds of the directors from outside the �rm allows certain shareholders to exert control costs, and default rates in their country by and an outside director as chairman.These disproportionate to their �nancial commit- more than 25 percent.52 On average, coun- reforms promise to ease the mobilization of ment.The migration to the Novo Mercado tries without credit registries have a private- investment capital. lifted the market value of companies In 2001 BOVESPA (the São Paulo Stock around the migration date. credit-to-GDP ratio of about 16 percent, Exchange) established a new market seg- those with publicly owned credit registries Source: McKinsey & Company (2002); Dyer ment, Novo Mercado, modeled on the (2001a, 2001b); Weiss (2002); BOVESPA Web about 40 percent, and those with private Neuer Market in Germany.To attract smaller site; Nova Mercado regulations 10.303; and de bureaus about 67 percent.53 enterprises, new market segments in other Carvalho (2003). Governments can create a supportive environment for credit bureaus by enacting (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 123 and enforcing data protection and credit �rms, or banks can “capture� their supervi- reporting laws that allow the sharing of sors, dissuading them from taking action credit information. The laws can safeguard when a regulation has been violated.55 consumer rights by allowing consumers to Because of such problems, several studies obtain data about themselves, requiring dis- have cast doubt on the effectiveness of pru- closure of information on who gets the dential regulation and supervision. On the credit report, and providing mechanisms one hand, indicators of its strength, such as for resolving disputes and correcting erro- supervisory power, the stringency of mini- neous information. Laws that allow the mum capital ratios, and the tenure of super- sharing of both positive and negative infor- visors, are not strongly linked to bank per- mation do more to improve lenders’ infor- formance and �nancial stability.56 On the mation and thus facilitate more lending. other, intensive of�cial supervision is associ- Credit reports that contain only negative ated with corruption, �nancing constraints, information (such as cases of late payment) and the need for political connections to get have less predictive power than reports with �nance.57 Effectively regulating risk-taking both positive and negative information.54 therefore calls for a cautious approach— Because credit reports are more important adapting it to �t the institutional features of for borrowers with limited collateral, limits the country at hand. Indeed, an alternative on data collection disproportionately harm school of thought stresses the ef�cacy of smaller borrowers. “sunshine� regulations that force informa- tion disclosure and so strengthen the ability Controlling risk-taking. Governments limit of depositors and other stakeholders to risk-taking by banks and other �nancial monitor banks directly.58 institutions for various reasons. Limited lia- Indeed, banking systems seem to work bility can cause banks to take excessive risks better when market discipline is encour- and, unlike in other industries, such prob- aged through market monitoring—not lems can lead to systemic crises—failure of strong supervisors.59 Possible private moni- one bank can lead to a run on all banks, toring agents include large depositors, sub- undermining the payments and credit sys- ordinated debt holders, shareholders, and tem. Deposit insurance can reduce the risk rating agencies. A study of banks in of bank runs. But the expectations of gov- Argentina found that those with a higher ernment bailouts from explicit or implicit share of nonperforming loans (seen as a deposit insurance can make the problem measure of risk) lose market share.60 In worse, by causing depositors and others to addition, Argentine banks were required monitor banks less carefully. (until the recent crisis) to issue subordi- Prudential regulation limits the �nancial nated debt for 2 percent of their deposits risks banks can take by requiring them to every year. After the introduction of subor- diversify and maintain at least a minimum dinated debt in 1998, complying banks paid ratio of capital to loans. It is administered lower deposit rates and had faster growth in by prudential supervisors who monitor deposits, lower capital ratios, and fewer banks on behalf of depositors and take nonperforming loans. Banks that failed to action to avert problems. Prudential regula- comply were penalized by having to tion can serve a useful purpose—reducing increase capital and liquidity.61 The market the risk of government bailouts and sys- also punished poorly performing banks in temic banking crises—but doesn’t always Thailand: equity prices of listed Thai banks work in practice. predicted their dif�culties in 1997—before As in other areas, choosing appropriate rating agency downgrades.62 regulations and administering them effec- The effectiveness of private monitoring tively requires �nancial resources and tech- depends on how well information disclo- nical capacity that are usually scarce. In sure regulations are enforced, whether rat- addition, good intentions may later be per- ing agencies compete with each other, the verted by corruption and clientelism. proportion of state ownership of banks, and Supervisors can direct loans to favored the nature of deposit insurance.63 Banks (c) The International Bank for Reconstruction and Development / The World Bank 124 WORLD DEVELOPMENT REPORT 2005 can be required to disclose standard �nan- ply, and ef�cient transport links stand out cial information and governance informa- from �rms without them. They invest tion, such as the compensation structure of more, and their investments are more pro- bank management (to better understand ductive. Yet in most developing countries, how risk-taking is rewarded). In addition, many �rms must cope with infrastructure the credibility and independence of rating that fails to meet their needs. The prob- agencies can be augmented by requiring the lems, as expressed by �rms, vary by region, disclosure of all business relationships and with Sub-Saharan Africa and South Asia track records, such as the number of times a having poorer infrastructure than Europe �rm receiving a favorable rating later devel- and Central Asia (see �gure 6.1). They also oped problems. tend to vary by infrastructure service and Information constraints in many devel- �rm size—electricity is often the biggest oping countries raise questions about how problem, and larger �rms express more well market monitoring can work.64 How- concerns than smaller �rms about all ser- ever, commercial rating companies now vices (�gure 6.4). provide some form of rating for 439 banks All types of infrastructure—including in 50 developing countries.65 There is also airports, railways, and distribution net- evidence that market discipline, de�ned as works for water and natural gas—matter market reactions to bank risk, can work well to some �rms. This Report looks at four in developing countries. Argentines pulled that matter to a very wide range of them: out their peso and dollar deposits in roads, ports, electricity, and telecommuni- response to increases in an individual cations. Although the Report focuses on bank’s exposure to a government default.66 the impact of infrastructure services on Better disclosure is also associated with �rms, improvements in the coverage and higher valuations of banks in emerging quality of these services also bene�t markets.67 households. Common challenges in infrastructure Infrastructure—connecting �rms Building and maintaining roads, ports, and expanding opportunities electricity grids, and telecommunications Firms with access to modern telecommu- networks is expensive, so it is no surprise nications services, reliable electricity sup- that poor countries in Africa, South Asia, and elsewhere have worse infrastructure Figure 6.4 Infrastructure concerns expressed by than rich countries. But the challenge of �rms vary by size and sector improving infrastructure is not just one of 30 �nding more money.68 Percentage of firms rating infrastructure Very large firms Market power, irreversible investments, as a “major� or “severe� constraint Large firms and politics. The problem of infrastructure 20 provision has its roots in the potential for Medium firms market power that results from economies Small firms of scale. It rarely makes sense to have two competing roads between two points—or competing electricity grids. Indeed, all 10 infrastructure activities were once thought to be “natural� monopolies, so that a par- ticular market could be served at least cost by a single supplier. However, the potential 0 abuse of market power in services that Telecom- Transport Electricity munications affect many consumers creates pressure for Note: Firms designated as small have fewer than 20 employees, governments to intervene, either through medium �rms have 20–49, large have 50–249 and very large have 250 or more. intensive regulation of private suppliers or Source: World Bank Investment Climate Surveys. through provision by the public sector. (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 125 Whether provision is public or private, gov- poor.70 Governments also used their infra- ernments tend to tightly control the prices structure agencies to channel assistance to that infrastructure providers charge and are particular regions and give jobs to favored often reluctant to allow prices to rise even groups, increasing the agencies’ costs and when costs have. frustrating attempts to hold them This reluctance can create problems accountable for the ef�cient delivery of because of another feature of many infra- services. With high costs and low prices, structure services—long-lived, immobile the agencies were unable to �nance invest- investments. Once built, a road or hydro- ment from their own cash flows or borrow electric dam cannot sensibly be disman- on their own credit (box 6.6). tled and moved elsewhere. Investors in As long as governments heavily subsi- infrastructure are often vulnerable, there- dized public infrastructure agencies, the fore, to changes in government regula- agencies could still operate and expand. tions, including those limiting prices. Fiscal pressures and mounting dissatisfac- Before they invest, the government may tion with public services, however, made promise them prices high enough to cover governments reluctant to go on providing the costs of investment, but afterward the large subsidies. That—combined with a government will be tempted to please cus- change in the prevailing views about mar- tomers and voters by keeping prices low. kets and private ownership—led many So long as prices cover operating costs, the governments to turn again to the private investors cannot credibly threaten to with- sector for at least some infrastructure ser- draw their services. vices. While public provision remains The underlying problem in the provi- important, private participation has now sion of much infrastructure is thus the spread throughout much of the developing combination of two reasonable concerns: world (�gure 6.5). customers fear that �rms will use their market power to overcharge, and �rms fear that governments will use their regulatory BOX 6.6 The political economy of electricity in India power to prevent them from covering their costs. Private �rms originally created Indian electricity utilities generally provide The subsidies became popular in the unsatisfactory service to their customers, late 1970s. In Andhra Pradesh the govern- much of the world’s infrastructure, but the whether �rms or households. In a recent bud- ment offered flat-rate tariffs to farmers as an playing out of these fears, combined with a get document the central government noted election promise. Soon after, in Tamil Nadu, prevailing skepticism about markets and that electricity shortages routinely lead to demonstrations by the Agriculturalists private ownership, led to widespread outages and voltage fluctuations that disrupt Association led to the provision of free elec- all aspects of economic life—and require sub- tricity to some farmers. Other states then nationalization of infrastructure after stantial investments in voltage stabilizers, followed with their own agricultural subsidy World War II.69 generators, and new motors. programs. Many of the recipients are fairly Under public provision, however, the Most electricity is generated and sup- well-off land-owning farmers. plied by state-owned electricity boards, Farmers are not the only bene�ciaries: problems reemerged in different guises which are experiencing severe �nancial dif- many customers steal their electricity, cost- and were joined by others. Infrastructure �culties and draining state budgets. Before ing suppliers an estimated $4 billion a year. services remained highly politicized, and privatizing its electric utility in 2002, for According to one report, utility employees governments frequently kept prices below example, the Delhi government provided it who conspire in the theft of electricity can with implicit subsidies of $200 to $300 mil- receive many times their annual salary in costs. The low prices were sometimes pre- lion a year, in loans unlikely to be repaid. bribes. sented as necessary to help the poor, but Even so, the company still faced �nancial Although some farmers, employees, the bene�ciaries tended to be those who problems and provided poor service: power and politicians bene�t, low prices discour- had access to services, so the poorest cuts were common in summer and winter. age both the conservation of power and The problems in Delhi, in other parts of further investment in increasing supply members of the community usually India, and indeed in much of the developing and improving its reliability. That is why missed out. To take just one example, a world are political.Under pressure from well- other users, including many �rms, have to study of the incidence of “lifeline� elec- organized groups of voters, governments pay more. tricity tariffs in Honduras, under which have kept average prices below average costs, allowing politically influential customers to the government subsidized the �rst block pay especially low prices.Farmers often Source: Agarwal, Alexander, and Tenenbaum of household electricity consumption, receive electricity for irrigation pumps at (2003); Dubash and Rajan (2001); India– found that about 80 percent of the subsi- prices well below costs. Ministry of Finance (2003); and Lal (2004). dies went to households that were not (c) The International Bank for Reconstruction and Development / The World Bank 126 WORLD DEVELOPMENT REPORT 2005 Figure 6.5 More developing countries are involving Although private provision has often the private sector in infrastructure provision lowered costs and improved services, the 100 problems of political economy remain. Telecommunications Many customers have opposed privatiza- 80 tion, believing it will do more to enrich big Percent of countries business and its political allies than improve 60 Electricity public services. At the same time, many infrastructure investors have been disap- 40 pointed by their returns in developing Ports countries, often believing that governments 20 have broken their promises on regulation Toll roads for fear of losing votes. Partly because of 0 these problems, the amount of investment 1990 1992 1994 1996 1998 2000 2002 in private infrastructure projects in devel- Note: The number of countries in the database varies over the period, starting at 128 in 1990 and ending at 151 in 2002. Private oping countries has declined in the last few participation includes management and lease contracts, conces- years (�gure 6.6). sions, and divestitures. Source: World Bank Private Participation in Infrastructure Project Database. Improving infrastructure by improving the climate for investment in infrastructure. Figure 6.6 Investment in infrastructure projects Addressing these problems requires recog- with private participation has recently fallen nition that the performance of infrastruc- 40 ture providers is shaped by their investment Telecommunications climate: a good investment climate helps 30 improve infrastructure (�gure 6.7).71 Electricity In some respects, the concerns of infra- Billions of dollars structure �rms—whether private or pub- 20 licly owned but commercially run—are no different from those of other �rms. All 10 Toll roads �rms worry about the security of their property rights and the burdens imposed by regulation, taxation, and corruption. Ports 0 They want to be able to hire good workers 1990 1992 1994 1996 1998 2000 2002 without having to keep them if business Note: Data show total investment in facilities with private partici- pation and exclude privatization revenues and similar payments. turns down. And they want access to Source: World Bank Private Participation in Infrastructure Data- �nancing.72 base. The problems arising speci�cally from Figure 6.7 Teledensity increases with the quality of the investment climate, even controlling market power and immobile investments in for incomes infrastructure highlight the central role of 1,000 400 secure property rights. Infrastructure �rms Not controlling for incomes Controlling for incomes are concerned not only about outright expropriation, but also about whether gov- 750 200 ernments will progressively undermine Teledensity Teledensity their pro�tability by imposing ever more 500 0 severe regulation. The problems affect small providers as well as multinationals (box 250 –200 6.7). Governments must therefore take care to craft rules and institutions that constrain 0 market power without unduly weakening –400 3 5 7 9 11 –4 –2 0 2 property rights. Investment profile index Investment profile index With this aim, governments often set out Note: Data are for 2001. The second graph controls for incomes by plotting the residuals of regressions of teledensity regulations and infrastructure investors’ and the investment pro�le index, respectively, on GDP per capita. Investment pro�le is a measure of risk to investment rights in contracts that cannot be changed (higher values mean lower risk). Source: International Country Risk Guide, International Telecommunication Union. unilaterally and allow disputes to be settled (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 127 BOX 6.7 Improving the investment climate for small private providers of infrastructure Much private investment in infrastructure comes �nd out the prices of commodities.This valuable where, uncertainty about license renewals cre- from multinationals from rich countries in Asia, service has been hampered by the state-owned ates a policy risk that can discourage investment Europe, and North America.When concerns are company BTTB, which has used its monopoly and increase electricity prices. (It also expressed about the investment climate for infra- over �xed lines to restrict interconnections encourages the substitution of easily moved structure providers, it is these �rms that most nat- between mobile phones and the �xed-line net- investments for those less costly but less easily urally come to mind. However, small (often infor- work. moved.) The providers do not know whether mal) infrastructure providers are also important their license will be renewed—or what bribe Small electricity suppliers in Cambodia for electricity and telecommunications, especially they might be asked to pay to ensure its In Cambodia the biggest electricity supplier is in rural areas, and the investment climate for renewal. Most of the small providers are, in fact, the state-owned Electricité du Cambodge, them matters, too. unlicensed.They thus face a different policy risk: which supplies Phnom Penh and a few towns. being prosecuted and closed down—or having Phone operations in Bangladesh But several hundred small private providers sup- to pay a bribe to avoid that. In many countries small entrepreneurs buy a ply electricity to more than 100,000 households All providers are also vulnerable to a change mobile phone and then run a small business and small �rms in rural areas, sometimes by in government policy that would give either charging others to use it. In Bangladesh, with recharging batteries and sometimes through Electricité du Cambodge or other providers one of the world’s lowest telephone densities metered connections to small electricity grids. exclusive rights to provide service. All are and waiting times of many years for a �xed con- Although charging fairly high prices, they sup- vulnerable to the possibility that, as they grow nection, village phone operators, most of them ply customers who would otherwise have to and become better established, the government women, provide mobile phone access to their supply themselves or go without. will come under pressure to regulate the prices rural neighbors. Bene�ting in many cases from By law these private providers require they charge in a way that undermines their prof- loans from the Grameen Bank, village phone licenses, which the government issues for a itability. operators are present in thousands of villages. renewable term of three years. Because the capi- At fairly low cost they enable villagers to com- tal invested in electricity grids can have a useful municate with people in markets in neighbor- life of more than three years and the assets can- Source: PPIAF and World Bank (2002); Burr (2000); ing towns—avoiding the need to walk there to not be costlessly dismantled and moved else- and Cohen (2001). by domestic or international arbitration about whether bribes or the public interest when investors do not trust the indepen- had motivated policy. Responding to these dence or reliability of local courts (chapter concerns, most countries have turned to 4). Decisionmaking about the implementa- transparent competitive bidding to award tion of rules is often delegated to indepen- contracts. Such countries as Brazil, Panama, dent regulatory agencies more insulated and Peru now publish many infrastructure than politicians from day-to-day political concession contracts on the Internet.74 In pressures (see box 5.2).73 2002 Mexico passed a freedom-of-informa- To work well, however, the government’s tion law that will require information about approach must not only secure investors’ such contracts to be made public. property rights on paper. To be credible to The creation of independent regulatory �rms, the arrangement must be sustainable, agencies can be viewed as an attempt to which means it must be perceived as rea- reconcile the partly competing demands sonably fair and legitimate by consumers for investor protection and public legiti- (chapter 2). Arrangements widely perceived macy. If legitimacy could be ignored, as legitimate and fair thus reduce risks faced investors’ property rights would be most Figure 6.8 Perceived fairness allows lower rates of return to be promised by providers, lower the returns that com- secure if contractual tariff adjustment rules for a given legal protection mercial investors must be promised, and so were interpreted by independent interna- Rate of return that must be promised lower the prices that customers must pay, tional experts and serious disputes resolved Rules perceived as unfair for any given degree of legal protection (�g- by international arbitration. Using national ure 6.8). regulatory agencies, courts, or arbitration One cause of popular resistance to pri- increases one type of risk for investors, vate participation in infrastructure in the because the national institutions are more 1990s was the opacity of some procedures susceptible to political pressures to keep used to privatize infrastructure businesses prices below costs—but decisions made by and adjust the tariffs the privatized business national institutions may be viewed as Rules perceived as fair could charge. In the absence of trans- more legitimate, enhancing the sustainabil- parency, suspicions were reasonably raised ity of the arrangements. Strength of legal protections (c) The International Bank for Reconstruction and Development / The World Bank 128 WORLD DEVELOPMENT REPORT 2005 BOX 6.8 Better government accounting, better government policy Traditional government accounting emphasizes accounts. For example, such a focus encourages borrowed money and repaying the loan in the cash de�cit as a measure of �scal performance a government to get a toll road privately monthly installments over 20 years: the govern- and the level of ordinary public debt as a measure �nanced, and to ensure its creditworthiness by ment’s obligations to make payments may be the of �scal position.The focus on these two guaranteeing the project company’s debt or same. Moreover, the arrangement does little to indicators—at the expense of measures that incor- providing a minimum revenue guarantee under address the problems of political economy porate noncash costs, assets, and traditionally “off- which the government tops up the toll revenue discussed earlier.Yet under traditional accounting balance-sheet�debt—encourages two biases in if it falls below a threshold. Although the guar- rules the “private�option spares the government infrastructure provision. antees are valuable to the project company and from disclosing new debt. First, it discourages pro�table public invest- costly to the government, they typically leave Government guarantees and long-term pay- ment and maintenance. Even when investment the cash de�cit and public debt unchanged— ment commitments can help get good projects or maintenance is expected to generate future unless and until the guarantee is called. under way, but as long as a government’s revenues for the government that outweigh the In another manifestation of the second bias accounting fails to pick up the effects on the gov- initial expenditure, the immediate effect is to the focus on ordinary public debt can encourage ernment’s �nancial performance and �nancial increase the cash de�cit and debt. Other biases, governments to prefer off-balance-sheet debt. position, doubts may reasonably remain about such as politicians’ desire for ribbon cutting and Instead of borrowing money to have a new power the government’s motivation for using them. In big bribes, may encourage public investment plant constructed, for example, a government can the long run the only way to remove the biases is projects, but there is evidence that ask a private company to �nance the plant, in for governments to adopt accounting rules that governments sometimes invest too little in return for the government’s signing a long-term take into account the value of the assets created infrastructure, especially when under pressure power-purchase agreement that commits it to or enhanced by public investment and mainte- to reduce cash de�cits and debt. making monthly payments to the private nance and the costs of guarantees and long-term Second, the focus on cash de�cits and debt company for, say, 20 years—with the monthly pay- payment commitments given to private investors. encourages governments to seek private �nanc- ments having a present value equal to the cost of ing for infrastructure projects, irrespective of the power plant. In substance the “privately their merits, and then subsidize the projects in �nanced�arrangement is similar to the govern- Source: Easterly and Servén (2003); Irwin (2004); ways that don’t show up in budgets and ment’s having the power plant constructed with and Tanzi and Davoodi (1997). Competition has the power to trans- when the government is no longer a form infrastructure industries by increas- provider of services, it can more easily ing legitimacy and strengthening investors’ allow genuine competition (see box 5.1). property rights. It pushes �rms to become So private participation can be part of a more ef�cient and cut prices. As a result, it strategy to help garner the bene�ts of com- helps assure customers that they are get- petition—reducing costs and the prop- ting a reasonable deal. This in turn reduces erty-rights problems of intensive regula- pressure on governments to regulate in tion. Second, to attract private investment, ways that weaken investors’ property a government needs to make a credible rights. Where competition works, it can commitment to allow prices to cover costs thus help infrastructure provision escape and not interfere in commercial opera- the problems that have traditionally tions—a commitment it cannot make afflicted it under both public and private under public provision, because it can provision. renege on commitments to public agencies Private participation is often advocated with impunity. If a government can credi- because it provides an alternative source bly make this commitment to investors by of �nancing to governments that have using the policies described above—and limited resources. Such reasoning is simultaneously persuade customers that flawed—and can encourage privatization their interests are being protected—it will with few real bene�ts (box 6.8). The big have gone much of the way toward creat- problem is paying for services, not �nanc- ing a good investment climate for infra- ing them, and though private investors structure providers, thereby doing much to may �nance services, they don’t pay for provide good infrastructure services to all them.75 �rms and to their broader societies. The real advantage of well-designed private participation is different and Improving public management. Although deeper: it lies in changing the political private participation plays a powerful role, economy of infrastructure provision. First, governments remain major �nanciers and (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 129 providers of much infrastructure, especially reports according to international account- roads. Even in sectors where a good deal of ing standards. Even when all these steps are investment is private, complementary public taken, however, it can be dif�cult for gov- investment in the parts of the sector owned ernments to resist political pressures to by the government can be important. When interfere in business decisions and keep governments do not provide or �nance infra- prices below costs. This is part of the reason structure, they often subsidize it—sometimes why many governments undertaking these directly, sometimes indirectly through guar- reforms have eventually turned to private antees and other instruments. Because gov- participation. ernment budgets are always more limited The challenges of improving infrastruc- than the plans of project proponents, govern- ture are similar in all sectors, but there are ments need ways of deciding how much to enough differences between sectors, espe- spend on infrastructure, how to allocate that cially in the opportunities for competition, spending, and how to administer it. to make it easier to discuss them one at a The questions are both technically dif�- time. cult and politically charged. For example, if the government can afford to construct and Telecommunications—competition maintain just one more road in the next makes the difference year, should it connect a poor rural area to Modern telecommunications services have the capital, or should it strengthen the net- become more important to �rms of all work around a congested and more pros- kinds—allowing them to communicate perous commercial center? Answering rapidly and cheaply with distant suppliers requires technical capability to undertake and customers. The services provide access cost-bene�t analyses, �nancial reporting to the Internet, underpin modern �nancial that reasonably reflects the true costs of markets, and help governments communi- different policies (box 6.8), and decision- cate with �rms and citizens. Modern making processes that give weight to the telecommunications are vital to the invest- results of those analyses while allowing a ment climate. In Bangladesh, China, socially acceptable balancing of competing Ethiopia, and India the Bank’s Investment interests. Climate Surveys found that garment man- When governments provide infrastruc- ufacturers are more productive, pay higher ture, they need to think about the best way to wages, and grow more quickly when organize themselves to do it. Traditionally, telecommunications services are better.76 governments provided services through min- Among developed countries, investments istries, but a desire to free service providers in telecommunications in the last 20 years from some of the constraints of bureaucratic appear not only to have followed growth, procedures, give them some managerial inde- but to have fueled it.77 In Latin America a pendence from ministers, and increase their 10 percent increase in the number of main accountability for results led many govern- phone lines per worker has been estimated ments to establish legally independent, to increase output per worker by about 1.5 though still wholly government-owned, percent.78 infrastructure agencies. The extent to which telecommunica- Some governments have taken extra tions services meet �rms’ needs varies steps, such as making the state-owned greatly from country to country, as well as agency subject to company law, appointing within countries. A three-minute call to the as directors people outside the government United States costs $0.17 from Finland, but with commercial experience, and requiring $9 from Chad, where the government effec- the agency to prepare audited �nancial tively taxes international calls to subsidize reports according to high-quality account- local calls and other services.79 Getting a ing standards. In South Africa, for example, new phone line takes only a couple of days the state-owned electricity agency, Eskom, in Lithuania, but most of a year in Algeria is now a company with mainly outside (�gure 6.9). In East Asia few �rms report directors with business experience, which having to pay a bribe to get a mainline (c) The International Bank for Reconstruction and Development / The World Bank 130 WORLD DEVELOPMENT REPORT 2005 Figure 6.9 Long delays for phone connections are Figure 6.10 Liberalization and good regulation accel- common, especially without competition erate the growth of phone connections 300 Cumulative annual growth of phone mainlines Algeria Eritrea in developing countries 200 Days of delay Bangladesh Honduras Ethiopia Ecuador 6.7 Nicaragua Kenya 100 Zambia India Belarus 5.2 Guatemala 0 5.4 Monopoly Partial Competition 4.6 competition Note: Data for waiting times are for �xed-line connections, and are from 2002 and 2003. Countries with delays longer than 40 days are shown. Competition status relates to competition in local calls. Source: International Telecommunication Union and World Bank Investment Climate Surveys. phone connection—in Africa, 20 percent or more do.80 Liberalized On average, however, telecommunications Not liberalized services have improved dramatically. Over No separate Separate regulator regulator the last 20 years prices have fallen at an aver- age of 7 percent a year, while the number of Note: Data cover years 1996–2001. Source: Qiang, Pitt, and Ayers (2004). phone subscribers per capita in low-income countries has quintupled.81 The changes have prices, shorter waiting times for connections, been driven by changes in technology and by and faster expansion of services (�gure 6.10 changes in policy. Most governments have at and �gure 6.11).82 least partly privatized their country’s main Although challenges remain, including phone company and allowed at least some the extension of access in rural areas (box competition. The policy changes mean lower 6.9), the combination of technological change and liberalization has transformed Figure 6.11 Competition spurs the spread of mobile phones in Sub-Saharan Africa telecommunications. Providers need no longer be monopolies, and with the advent Uganda of cellular telephony, investments are no 8 Côte d’Ivoire longer so immobile. Together these changes 7 greatly reduce the policy-related risks of Guinea 6 investment in the sector and go much of the Madagascar way toward solving the problems that have Index of teledensity 5 traditionally afflicted infrastructure. Ghana 4 Many governments have yet to take full Tanzania advantage of the opportunities of technolog- 3 ical change. By 2002 all developed and most Zambia 2 Latin American countries allowed full com- petition in international telephone calls, but 1 most other countries did not (�gure 6.12). 0 – 6 – 5 – 4 – 3 – 2 –1 0 1 2 3 4 5 6 Electricity—competition is possible, Year relative to start of mobile competition but not as easy Note: Teledensity refers to total number of �xed and mobile connections per capita. The data are rebased so that the index of teledensity equals 1 in the year the second Access to a reliable electricity supply at a rea- mobile operator enters the market. The countries selected are all those that liberal- sonable price is vital for most �rms—from ized before 1998, plus Uganda, which liberalized in 1998. Source: International Telecommunication Union and World Bank staff. small factories in rural areas to multina- (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 131 Figure 6.12 Competition in international calls is still limited or prohibited in much of the developing world Source: World Bank staff; created by the Map Design Unit of the World Bank. tional �rms. Most urban �rms are served by lose on average around 5 percent of their utilities, but �rms in small towns and rural annual sales.84 The problems are especially areas in developing countries may have to severe in Nigeria (box 6.10). Elsewhere in supply themselves.83 Firms with access to Africa, �rms report that it takes two or three grid electricity seldom get good service. months to get a new electricity connection Temporary losses of supply are frequent in and often requires a bribe.85 Limited access many countries, especially in Africa and in rural areas and poor quality in cities cause South Asia (�gure 6.13), as are fluctuations many �rms to rely on self-supply, which for in voltage that damage machinery. Firms most is more expensive than a regular sup- estimate that such outages cause them to ply from a utility. BOX 6.9 Expanding rural access to electricity and telecommunications For many years governments in developing coun- combination of liberal regulation and well-tar- yearly installments, conditional on the operator tries relied on state-owned monopolies to bring geted, output-based subsidies. Removing legal meeting its performance targets. Although the electricity and telecommunications services to barriers to entry by new providers of electricity operators are struggling �nancially even with rural areas.Typically they required the monopolies and telecommunications services helps ensure the subsidies, most results from the pilot project to charge the same price in rural and urban areas, that pro�table opportunities to extend service appear promising. For the scheme’s bene�cia- even though the costs were higher in the rural. in areas unserved by the incumbent are seized ries the average distance to the nearest pay Because that made the rural services unpro�table, quickly (as illustrated by Cambodia in box 6.7). phone fell by more 90 percent. And competitive governments gave the monopolies budgetary Liberal entry rules may not by themselves bidding led to a subsidy 41 percent lower than subsidies and allowed them to bene�t from cross- cause access to increase as fast as governments the government had budgeted for and 74 per- subsidies from low-cost, high-revenue customers. want. In such a case governments may �nd care- cent lower than the subsidy previously In many countries, however, the subsidies have fully targeted direct subsidies more effective requested by the incumbent. Similar schemes been too small to �nance rapid expansion. Even than cross-subsidies or subsidies aimed only at have been used for rural electri�cation in when expansion was affordable, the monopolies keeping providers afloat. Peru, for example, has Argentina, Chile, and Guatemala. had a �nancial incentive to go slow. used a least-subsidy approach to bring pay An alternative that some governments have phone service to targeted rural areas. Some of Source: Cannock (2001); Harris (2002); Tomkins used, especially in the last decade, is to rely on a the subsidy is paid up front, the rest in half- (2001); Wellenius (1997a); and Jadresic (2000). (c) The International Bank for Reconstruction and Development / The World Bank 132 WORLD DEVELOPMENT REPORT 2005 Figure 6.13 Many days of power outages a year, and a higher share of Poor electricity supply makes existing �rms having their own generators investments less productive and discour- 100 ages new investment. In Uganda �rms that experienced fewer problems of supply from the (generally poorly performing) Uganda Percentage of firms with generators 80 Electricity Board invested less in self-supply Kenya Bangladesh and more in their own productive capac- 60 ity.87 In Bangladesh, China, Ethiopia, and Tanzania Pakistan the Bank’s Investment Climate Pakistan Eritrea Surveys found that more reliable power 40 Zambia supply increases garment manufacturers’ Uganda Ecuador total factor productivity and the growth Algeria Bhutan rates of their output and employment.88 In 20 Guatemala Nicaragua Latin America a 10 percent increase in elec- Brazil tricity-generating capacity per worker has Poland 0 Uzbekistan been estimated to increase GDP per worker 0 90 180 270 360 by around 1.5 percent.89 Days of outages As in telecommunications, changes in Note: The �gure shows all countries for which data on both the days of outages and the technology, coupled with dissatisfaction share of �rms having their own generators were available. Data are for various years with monopoly provision by public enter- between 1999 and 2003. Source: World Bank Investment Climate Surveys. prises, have led many governments to liber- alize and introduce private participation. Economies of scale in generation declined Many �rms also pay higher than neces- in the 1980s, allowing more countries to sary prices for electricity, as governments have enough generating stations to make direct utilities to hold down prices for competition in the supply of electricity (often middle class) households and effec- workable.90 Countries that can trade elec- tively tax �rms to make up some of the dif- tricity with their neighbors have further ference. The largest industrial users some- opportunities. times have enough influence to avoid such Almost all countries in the developed levies, leaving small and medium �rms to world and most in Latin America now allow bear most of the burden. In the Indian state at least some �rms to choose their electric- of Kerala industrial users pay twice as much ity supplier. Elsewhere the picture is mixed. per kilowatt-hour as households, but com- Many countries have allowed a sort of com- mercial users—of�ces and shops—pay petition in generation under which a state- nearly twice as much again.86 owned utility contracts out the �nancing, construction, and operation of new power stations to privately owned independent BOX 6.10 The power to improve productivity in Nigeria power producers. The state-owned utility, however, usually retains a monopoly on Poor service from the government-owned as much themselves as they bought from selling electricity to customers, limiting the National Electric Power Authority (NEPA) NEPA.The average cost of self-generation causes severe problems for Nigerian manu- was high, however—$0.30 a kilowatt-hour, bene�ts of such competition. In addition, facturers. or about three times more than NEPA such projects can create disguised govern- In a 1998 survey 93 percent of respondents charges. Small �rms may be particularly vul- ment debt (see box 6.8). reported experiencing power outages more nerable because they are less able to bear Getting competition to work in electric- than �ve times a week.On average the outages the �xed costs of self-generation. caused them to lose 88 working days per year. Accordingly 16 percent of small �rms relied ity is harder than in telecommunications, The �rms also reported that poor supply led to only on NEPA service, while no medium or as high-pro�le problems in California the destruction of raw materials,restart costs, large �rms did. In addition, small �rms lost show.91 Many small countries have too few and equipment damage.They ranked poor 24 percent of their output to outages, while generators to allow real competition, while electricity supply as by far their most important medium �rms lost 14 percent and large obstacle in infrastructure. �rms 17 percent. in larger countries, individual electricity Many �rms invested in self-generation as companies may still have market power if a result. On average they generated almost Source: Adenikinju (2003). they own many generation plants. Even when electricity generators do not have (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 133 market power at most times of the day, they Figure 6.14 The declining costs of transport and telecommunications may have it when demand peaks, and like sellers in many markets, they may collude 120 to increase prices. Competition is fostered by separating generation from transmis- 100 sion, and distribution from retail supply, so that the owners of the transmission and 80 distribution lines cannot use their monop- Index oly in these industry segments to stifle 60 competition in generation. But such Sea freight unbundling makes it harder to coordinate 40 investments among these segments of the industry. 20 Air transport Overall the evidence suggests that com- Telecommunications petition (usually combined with commer- 0 cial provision and new forms of regulation) 1930 1940 1950 1960 1970 1980 1990 2000 has led to better service. Countries that Note: Index for all services set at 100 in 1930. Source: Busse (2003). early on introduced competition, private provision, and new forms of regulation— Transport costs depend on distance, so such as Argentina, Chile, and the United countries far from rich markets in Europe, Kingdom—have bene�ted from lower North America, and East Asia face a disad- prices and higher quality.92 In Chile whole- vantage they can do nothing about. Yet sale prices fell by 37 percent and retail prices poor infrastructure has been found to by 17 percent between 1986 and 1996. Pri- account for 40 percent of the cost of trans- vate companies were suf�ciently con�dent port in the average country and 60 percent in the market to invest in hydroelectric gen- in landlocked countries. So while distance eration, transmission, and distribution.93 accounts for much of transport costs, ship- More generally, competition in electricity ping goods from ef�cient ports, such as has been found to increase labor productiv- those in Hamburg and Rotterdam—or ity and generating capacity per capita.94 inland cities bene�ting from good infra- Competition also tends to lower prices for structure, such as Ankara and Vienna—is small and medium �rms because they need cheap for the distance.99 According to one no longer buy from a utility that over- study a country could lower its transport charges them. costs by an amount equivalent to moving several thousand kilometers closer to other Transport—overcoming the tyranny countries—considerably reducing the of distance “tyranny of distance�—if it could improve Transport infrastructure creates opportuni- its transport (and telecommunications) ties for �rms to buy and sell not only in infrastructure from the median to the 75th neighboring markets but in the entire percentile.100 world. As governments eliminate import Reducing transport costs requires paying quotas and reduce import tariffs, transport attention to particular transport modes, becomes more important as a source of fur- such as ports and roads. Yet governments ther gains in trade.95 Although global trans- should not lose sight of the links among dif- port costs have been falling over the long ferent modes: ports and airports, for exam- term (�gure 6.14), further progress is ple, become more valuable when served by important. For Chile and Ecuador trans- good roads and railways. Transport costs port costs to the United States are now 20 are also affected by factors other than trans- times larger than U.S. tariffs.96 If they could port infrastructure, such as whether reduce their transport costs by 10 percent, telecommunications systems allow compa- they could expect to increase their trade by nies to track their goods in transit and how 20 percent.97 Other evidence suggests that quickly goods are cleared through customs they would also grow faster.98 (see chapter 5). (c) The International Bank for Reconstruction and Development / The World Bank 134 WORLD DEVELOPMENT REPORT 2005 Ports—many types of competition. More BOX 6.11 Port reform in Colombia and India than 80 percent by weight of the trade of developing countries goes through ports.101 Colombia and India show two ways of con- administered by a Port Trust representing fronting the challenges posed by port reform. various interest groups. Port reform began The ef�ciency of those ports affects In Colombia port ef�ciency had become with the issuance of a new policy frame- exporters and importers directly and almost a major issue by the early 1990s. Early pro- work in 1994 and guidelines for private all �rms indirectly. Improving one measure posals involved the reorganization of participation in 1996. Private participation of port ef�ciency from the 25th to the 75th Colpuertos, the state-owned company, but was to start with the concessioning of the not private participation. President Gaviria, container terminal at Jawaharlal Nehru percentile—achievable in part by reducing however, favored a bolder approach and Port, established in 1989 as a satellite port the influence of organized crime—has been raised the issue in his inaugural address in to Mumbai. found to reduce shipping costs by more than 1990. His government drove the reform, The implementation of reforms was left with little involvement from labor groups. to the ports, and the Jawaharlal Nehru Port 12 percent.102 As with improvements in Legislation to allow private participation Trust (the majority of whose trustees repre- other transport infrastructure, the reduction in ports, including severance packages for sented the government or labor) chose to in costs is equivalent to moving thousands workers, passed within 60 days.The overall engage the main stakeholders in the reform of kilometers closer to trading partners.103 program—liquidating Colpuertos, establish- process and to protect the interests of labor ing new policymaking and regulatory bodies, by keeping the existing port under public Unlike the customers of electricity and concessioning the �ve major ports to private ownership. But they did allow a new private telecommunications utilities, port cus- �rms, introducing competition in stevedoring terminal to compete with it.The competi- tomers are mainly �rms, not households, in each port, and retrenching nearly 6,750 tion improved performance, with which makes tariff setting less politicized. workers—was completed in three years.The preberthing and turnaround time falling combination of competition and private par- from around 11 days in 1996 to less than 3 Ports, however, require immobile invest- ticipation led to impressive improvements in days in 2002. ments and often have market power, so they performance. face many of the challenges common to India approached the task differently. infrastructure services. Under public own- Each of the 12 major ports in India is Source: Navarrete (2004) and Ray (2004). ership and restrictions on competition within and sometimes between ports, they have tended to be overstaffed, have restric- tive labor practices, act as a magnet for cor- BOX 6.12 The bene�ts of rural roads in Morocco and ruption—and as a result offer slow and elsewhere expensive service to �rms.104 To improve the ef�ciency of ports, gov- When built in the right locations (and not ment.The estimated economic rate of return ernments have tried to expose them to “roads to nowhere�), good roads can create to the projects ranged from 16 to 30 percent. substantial new opportunities for entrepre- As is often the case, the improvement in more competition, often while introducing neurs in rural areas and small towns, as illus- infrastructure did not bene�t only �rms. It private participation (box 6.11). Colombia trated by a Moroccan government program made it easier for children to go to school and Argentina split their national state- to pave gravel roads and dirt tracks. and, by making the delivery of butane more Upgrading the roads meant they were affordable, reduced the need for women owned companies into several separate usable all year round, causing less damage to and girls to collect �rewood. After the road companies that compete with each other for the vehicles using them.The new roads improvements, primary school enrollment some services.105 Governments can also cre- allowed farms and other �rms to move their rose from 28 percent to 68 percent. ate competition within a single port in ser- goods more often and more cheaply. In some The Moroccan experience is not an iso- cases the time it took to get to rural markets lated case. Recent work by the International vices not inherently monopolistic: different fell by half.The cost of shipping a truckload of Food Policy Research Institute suggests that terminals in a port can sometimes compete merchandise also fell by half. In the areas ben- Uganda’s investment in rural feeder roads with each other, and different stevedoring e�ting from the road upgrading, the land is connecting farmers to otherwise remote companies can sometimes compete at the more productive, and the volume and value markets has high returns in agricultural of agricultural produce is higher. As it became growth and rural poverty reduction. In same terminal.106 easier to ship produce quickly without dam- China investment in rural roads is very The combination of private participa- aging it, farmers shifted from low-value cere- socially pro�table. In India such investment tion and increased competition has led to als to high-value fruit. As the price of bringing is the most socially productive form of pub- better services.107 In Colombia average ves- goods to the farms fell, farmers used more lic investment in reducing poverty. fertilizer. Improvements in the agricultural sel waiting time fell from 10 days before pri- economy spurred the growth of other busi- Source: World Bank (1996a); Fan, Hazell, and vatization and competition to a matter of ness. Off-farm employment grew twice as fast Thorat (1999); Fan, Zhang, and Rao (2004); Fan, hours afterward, throughput per hour as in areas not bene�ting from road improve- Zhang, and Zhang (2002). increased, and the ports moved to all-year, all-day operation.108 In Argentina the aver- age stay fell from 72 hours to 33, through- put per worker rose from 900 tons to 4,850, and capacity increased �vefold.109 (c) The International Bank for Reconstruction and Development / The World Bank Finance and infrastructure 135 Roads. Almost all goods are transported by designing a system that gives the managers road at some stage, making a country’s road of the road fund the information, incentives, network a critical part of its infrastructure and capability to make decisions aligned and the investment climate (box 6.12). Not with the public interest is crucial. surprisingly, the extent of the network has Developing countries often spend too lit- been found in many studies to be associated tle on maintenance compared with invest- with better economic performance. In Latin ment, perhaps because of donors’ traditional America a 10 percent increase in the length of preference for subsidizing capital rather than roads per worker has been estimated to outputs, and perhaps because large invest- increase GDP per worker by nearly 2 per- ment projects offer opportunities for politi- cent.110 Not all roads are equally valuable, of cians to cut more ribbons or for decision- course; in the United States the interstate makers to collect bigger bribes. Countries road building of the 1950s and 1960s seems afflicted with higher levels of corruption to have signi�cantly boosted productivity, seem to spend more on public investment in while recent spending on roads has had only roads and other infrastructure, but less on modest bene�ts.111 Even so, the evidence sug- maintenance, and seem accordingly to have gests that governments should pay close poorer quality roads.113 There is no simple attention to the extent and quality of their answer, but an emphasis on making decision- road networks. The challenges relate to plan- making more transparent can help reduce ning appropriate network expansion, execut- corruption and improve decisions. Govern- ing the required investment and mainte- ments can consult on, publish, and explain nance, and working out how best to pay for it. the principles for allocating funds and the All the typical challenges are more dif�- decisions implementing those principles, and cult because the transaction costs of impos- they can use open and transparent processes ing user fees (tolls) to fund roads are high, at for awarding contracts to do the work. least on city streets and rural roads. Even on Road agencies that decide on the alloca- intercity highways, where the transaction tion of funds need not build or maintain costs are lower, user fees remain uncom- roads themselves. More road agencies now mon.112 So prices rarely ration demand on contract out such work to private �rms, congested roads, cover the costs of mainte- under output-based contracts. In Argentina nance, or signal that new capacity is needed. the highway authority maintains many One avenue for tackling these problems is roads by letting long-term maintenance thus to increase the use of tolls. The advent contracts that require private �rms to main- of electronic tolls and related information tain roads to a de�ned standard. One review technology is making direct pricing feasible concludes that the program reduced the on more roads and, in the long term, it may proportion of roads in poor condition from make the road industry much more like 25 percent to less than 5 percent, reducing other utilities. In the near future, however, road users’ costs by more than 10 percent.114 only a small proportion of roads will have tolls. Therefore, many governments focus on using other sources of revenue linked to road Improving the provision of �nance and use to pay for roads, such as use-related infrastructure services in an economy can license fees and especially fuel taxes. have a big impact on the investment cli- Many governments are assigning funds mate—and ultimately depends on improv- from fuel taxes and other sources to a road ing the investment climate for providers of fund that operates with some autonomy those services. Similar links exist in the labor from ministers. The funds are allocated to market, where the quality of the investment investment and maintenance projects climate has important implications for the according to a set of principles established incentives of workers to invest in their own by political authorities. Road users may be skills. The effectiveness of the labor market represented on the agency, and the agency in connecting people with productive jobs is may consult with road users and others on critical to growth and poverty reduction. the allocation of funds. As in other areas, These issues are the subject of chapter 7. (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 7 Governments around the world share the Crafting an investment climate that pro- goal of having more and better jobs for their vides �rms with the opportunities and citizens. Jobs are the main source of income incentives to expand is fundamental to chapter for people—and the main pathway out of meeting this challenge. Government poli- poverty for the poor. Young people domi- cies affecting the labor market play a critical nate the ranks of the unemployed, with over role in this effort by helping to connect peo- double the average unemployment rate in ple to jobs. And there is room for improve- all regions.1 And in many developing coun- ment in most countries. tries more than half of the working popula- Government support for education and tion is in the informal economy, where training affects the prospects for individuals— working conditions can be poor.2 Demo- and the ability of �rms to enter new mar- graphic changes over the coming decades kets and adopt new technologies. Firm-level will add nearly 2 billion more people to surveys show that more than 20 percent of developing countries, compounding the �rms in many developing countries rate challenge of creating more and better jobs. inadequate skills and education of workers as a major or severe obstacle to their opera- tions (�gure 7.1 top). Regulation of labor markets is usually Figure 7.1 Firms rate skill shortages and labor intended to help workers, but can also be a regulations as serious constraints in many countries signi�cant constraint on �rms (�gure 7.1 Brazil bottom). Ill-considered regulations can dis- courage �rms from creating more jobs and of available workers Skills and education Zambia contribute to a swelling of the informal China economy. When this is the case, some work- Algeria ers may bene�t, but the unemployed, the low-skilled, and those in the informal econ- Estonia omy will not be among them. Bangladesh Public policy also needs to facilitate allo- cation of labor to its most productive use Brazil while helping workers cope with labor mobility. Technological progress that leads to Poland Labor regulations higher productivity and economic growth Philippines improves working conditions and wages, but Kenya it can also result in more rapid changes to �rms and industries. In modern economies, Pakistan many �rms are created and destroyed each Algeria year—about 20 percent in many countries— 0 20 40 60 involving 10–20 percent of the workforce. Percent This chapter looks at opportunities for Note: Percentage of �rms reporting that skills and education of governments to improve policies in all three available workers or labor regulations were a major or severe areas as part of the effort to create a better obstacle to the operation and growth of their business. Source: World Bank Investment Climate Surveys. investment climate: 136 (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 137 • Fostering a skilled and healthy workforce that can contribute to a productive and BOX 7.1 Malaria and HIV/AIDS cloud the investment climate prosperous society. Improving the invest- Malaria and HIV/AIDS have a debilitating HIV/AIDS erodes morale, lowers produc- ment climate goes hand in hand with impact on people—and growth.They can tivity, weakens con�dence in the future, and enhancing human capital. A skilled work- also be debilitating for the opportunities undermines the willingness to save and force is essential for �rms to adopt new and incentives facing �rms to invest invest. It affects the most economically productively, to create jobs, and to expand. active age groups and reduces the quantity and more productive technologies, and a Malaria-affected regions tend to have and quality of labor. Skilled professionals better investment climate raises the lower worker productivity and lower per are being lost, and shorter life expectancies returns to investing in education. Govern- capita incomes than other regions. are raising the cost of training and reducing ments need to take the lead in making HIV/AIDS is also having a pervasive impact, short-term returns. with an estimated 40 million people living HIV/AIDS not only destroys human education more inclusive and relevant to with HIV/AIDS worldwide, including 2.5 mil- capital—it also weakens the transmission of the skill needs of �rms, and create a sound lion children under 15. Sub-Saharan Africa knowledge and abilities from one genera- investment climate for providers of edu- had more than 80 percent of the new infec- tion to the next. cation and training services. tions and 75 percent of the deaths in 2003. Not surprisingly, almost 90 percent of �rms • Crafting labor market interventions to ben- there are concerned about HIV/AIDS. A sur- Source: Sachs (2003); McArthur and Sachs (2001); UNAIDS (2003); Bloom and others e�t all workers. In many developing coun- vey of African �rms has quanti�ed its (2003); United Nations Economic Commission tries labor regulation provides a high stan- impact on the region’s economic productiv- for Africa (2000); and Bell, Devarajan, and Gers- ity at around 1 percent of GDP. bach (2003). dard of protection to a few workers but limited or no protection for most of those in the informal economy. It can also dis- Health strengthens the incentives and ability courage �rms from creating new jobs. to invest in education. And apart from the Regulatory strategies need to be crafted to human gains, controlling diseases such as reflect this wider range of interests, and to malaria and HIV/AIDS increases the produc- ensure a good �t with local circumstances. tivity of workers, encouraging �rms to pur- • Helping workers cope with change in a more sue worthwhile opportunities in once- dynamic economy. Inadequate mecha- affected locations (box 7.1). nisms to help workers cope with change The links between education, health, and restrict entrepreneurship and the adapt- growth can create virtuous circles: good ability of workers. They can also increase education and health enable growth, which resistance to reforms that would bene�t in turn promotes further investment in society as a whole. While a narrow tax base them. The circles can also be vicious: poor reduces the feasibility of creating compre- education and health reduce incentives for hensive social safety nets in most develop- productive investment and entrepreneur- ing countries, there are opportunities for ship, which limits the resources for enhanc- improving the insurance component in ing education and health. income support schemes and the pooling Issues associated with the delivery of of risks across individuals. Innovative pro- health and education services were dis- grams can also reach out to poor and cussed extensively in World Development informal workers who cannot be covered Report 2004 and will not be revisited here. by broader insurance schemes. The focus instead is on the complementari- ties between the education and skills of Fostering a skilled and healthy workers and the investment decisions of workforce �rms—and on some of the ways education People’s skills and health affect their ability to policies need to evolve to equip individuals participate in society, escape poverty, cope with the skills required in a more produc- with economic and natural risks, and con- tive and dynamic economy. tribute to productivity increases and growth. The availability of skilled and healthy workers The skills of workers and the also shapes the decisions of �rms to adopt investment climate new technologies, expand, or enter new mar- Educational attainment has improved in all kets. Education improves health through developing regions, particularly in East Asia greater awareness and access to information. and Paci�c and in the Middle East and North (c) The International Bank for Reconstruction and Development / The World Bank 138 WORLD DEVELOPMENT REPORT 2005 Figure 7.2 The share of the population with secondary or higher education is still very receive higher wages because schooling signals low in many developing countries to employers positive individual characteris- 1980 2000 tics, such as ambition and motivation. But Developed countries these characteristics may have modest or no effects on actual productivity if there are no Europe & Central Asia opportunities to take advantage of them. Social returns to education can also be low East Asia & Pacific when the demand for educated workers is Latin America stagnant. If �rms are not subject to competi- & the Caribbean tive pressures that stimulate technical progress South Asia and the demand for more skilled workers, the effective demand for education will be weak.6 Middle East & North Africa Another problem is that human capital Sub-Saharan Africa may not be applied to its most productive uses. Bloated bureaucracies and overstaffed 0 20 40 60 80 state enterprises can crowd skilled workers Percent of population with secondary or higher education out of private sector activities. In some Note: Population aged 25 and over with secondary or higher education attainment as a share of total popula- cases their contribution to society can be tion age 25 and over. Source: Barro and Lee (2001). low or even negative.7 Africa but still remains low in many develop- ing countries. In Sub-Saharan Africa and Investment climate improvements interact South Asia more than 40 percent of those age strongly with education. The link between 25 and over in 2000 had not completed any investment in human capital and growth is formal education. And while there have been mediated by the way education services are signi�cant improvements in the proportion delivered and skills are allocated in the of adults who have completed secondary and economy. But investment climate improve- higher education in all regions, their share in ments almost always increase the demand the working-age population remains very low for human capital. As �rms have more in many countries (�gure 7.2). opportunities and better access to new tech- nologies, they demand more skilled workers Strengthening the impact of education on and have stronger incentives to engage in growth requires better incentives. There is a growth-enhancing activities, raising both strong link between education and living the private and social returns to education. standards across developed and developing Skilled workers are needed to adopt new countries, but the strength of that link technologies because they are better at deal- largely depends on the quality and delivery ing with changes.8 This is true for different of education and on the incentives �rms types of �rms and different levels of techno- face to hire more skilled workers. The link logical development. Technology transfers between education and living standards has by multinational �rms, and technology often been broken, prompting some to ask adoptions by local �rms, require a mini- “Where has all the education gone?�3 For mum of human capital and training (box example, some African countries with rapid 7.2). New technologies generally require sig- increases in human capital over the past two ni�cant organizational changes, which are decades have been growth disasters. also handled better by a skilled workforce.9 Having more schooling tends to raise indi- Even among self-employed farmers in low- vidual wages. Indeed, private returns from income countries, having at least primary schooling are high in many countries around education enables them to use more ef�- the world, even if the social returns from edu- cient production techniques.10 cation, in the form of higher output, are often Skill constraints are a common problem disappointing.4 The quality of education is for �rms in developing countries (�gure 7.1). essential: higher investment in schooling of The constraints are especially severe for �rms very low quality may not lead to higher pro- planning to innovate and expand. The World ductivity.5 More educated workers may still Bank’s Investment Climate Surveys show that (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 139 the �rms that consider a lack of skilled work- ers to be a “major� or “very severe� constraint BOX 7.2 Why Intel chose Costa Rica as the site of a are those upgrading their production multimillion dollar plant processes. Those �rms are also more inclined In 1996 Costa Rica beat out Brazil, Chile, Since 1948,when democracy was restored, to invest in training their workforce (�gure Indonesia, Mexico, the Philippines, and Thai- Costa Rica has placed a strong emphasis on 7.3). While large �rms have the capacity to land to become the site of Intel’s $300 mil- education.The government invested heavily in organize internal training for their work- lion semiconductor assembly and test plant. education and technology training,and force, smaller �rms often do not. Many factors made Costa Rica attractive to adopted a bilingual English as a Second Lan- Intel, as well as other U.S. companies: its sta- guage curriculum.Computers were introduced A sound investment climate strengthens ble economic and political system; its cen- in elementary schools as early as 1988,and by the incentives for individuals to obtain more tral location within the hemisphere; its 1996 many schools were equipped with them. education. This is best exempli�ed by the openness and liberalized economy, includ- In response to the large investment by Intel major surge in returns to education in the for- ing the absence of capital controls; and its and other U.S.companies,several education receptive investment environment. Another centers—providing technical skills in the elec- merly centrally planned economies during key factor was its educated labor force, and tric and electronic �elds—have emerged. their transition to market systems. Similar the government’s commitment to invest in patterns have emerged in other countries. In further training. Source: World Bank (2003e) and Spar (1998). Cambodia investment climate improvements, coupled with higher returns to well-trained Figure 7.3 Skill constraints and innovative �rms people, boosted the demand for vocational training, mostly provided by private �rms. Innovators face more Innovative firms with skill severe skill constraints constraints invest in training High levels of formal education are not 30 80 needed for all �rms or activities. Lack of More than 100 employees availability of workers with tertiary education Share of innovators that provide training Share of firms reporting skill constraints may be more of a constraint for �rms in 20–100 60 as an obstacle to business higher value-added manufacturing and ser- Innovators employees vices than for those in less complex industrial 20 processes. For some activities, language pro�- Less than 20 employees ciency may be important. For example, a Non-innovators 40 large English-speaking population has helped India attract “back-of�ce� services for foreign 10 �rms. In many cases education to provide 20 basic literacy and numeracy skills can be complemented by on-the-job or vocational training to enhance the productivity and 0 0 hence potential wages of workers. Firm size Note: “Innovators� are �rms that have upgraded an existing product line or developed a Creating a skilled workforce major new product line in the previous three years. “Skill constraints� refers to those �rms that considered lack of adequate skills a “major� or “very severe� obstacle to the World Development Report 2004 discussed operation and growth of their business. The left panel is based on a sample of 13,300 �rms in 33 developing countries. The right panel is based on a sample of 12,099 �rms in strategies for improving the delivery of 29 developing economies. basic education. Secondary and tertiary Source: World Bank Investment Climate Surveys. education and vocational training also mat- ter for a good investment climate. Govern- providers sharpening the incentives to be ef�- ments can help in a variety of ways. cient and responsive. Options for providing such support include income-contingent Public funding to expand access to educa- loans (as in Namibia)11 and voucher schemes tional opportunities. Public funding can of various kinds. For example, the Africa Edu- improve the equity of the education system by cational Trust provides educational vouchers opening opportunities to those who could not in Somalia to enable disadvantaged girls and otherwise afford it. Many traditional young ex-militiamen to attend special after- approaches focused on providing funding noon and evening classes.12 through public educational institutions. Newer approaches direct resources through Improving quality assurance mechanisms. individuals so that they have greater choice, Minimum quality requirements and quality with the resulting competitive pressure on assurance mechanisms through certi�cation (c) The International Bank for Reconstruction and Development / The World Bank 140 WORLD DEVELOPMENT REPORT 2005 or accreditation schemes can foster quality providers according to de�ned competency improvements at schools and universities. It levels (China, Mauritius, Mexico, can also boost demand for education by Uganda).13 students and increase demand for skills Facilitating private provision. The market from �rms. More than 20 developing coun- for private education has grown strongly in tries have introduced accreditation agencies recent years, augmenting public resources or national evaluation systems. Experience and providing a broader range of choices for suggests that quality assurance is best pro- students. In Brazil, for example, private vided by agencies that have authority over institutions accounted for more than 70 per- both public and private providers, rely on cent of higher education enrollments in explicit standards, and publicly report 2002. Strong increases have also occurred in results. Evaluation criteria are moving from most regions of the world, including Africa, the measurement of inputs (characteristics where the private sector is a signi�cant of the service provider) to a stronger focus source for secondary and tertiary education on outputs (student performance). Many in countries such as Côte d’Ivoire, Gambia, countries are also establishing national and Ghana.14 Expanding opportunities for quali�cations frameworks that allow com- private education involves improving the parison of quali�cations from different investment climate for private providers. While private providers of education face BOX 7.3 Tackling skill imbalances through public support many of the same constraints as other �rms, for training and retraining programs additional constraints can flow from poorly de�ned regulatory frameworks and policies Government support for the training and associated with displacement of other that discriminate in favor of public sector retraining of workers can take many forms, workers. providers. The private sector may also be depending on the target group, the funding A growing number of countries are source, the form of training, and the mode funding enterprise-based training and engaged through public-private partner- of delivery. retraining through compulsory levies on ships of various kinds. In Burkino Faso, for In Mexico the Job Training Program for �rms rather than relying on general tax example, the management of colleges of Unemployed Workers (PROBECAT) combines revenues. Brazil’s National Industrial Train- general education is being delegated to pri- short-term training for unemployed and dis- ing Service (SENAI) funds training from a placed workers with income support (at the compulsory contribution from industries vate education providers.15 minimum wage) and, more important, place- of 1 percent of payroll. SENAI has been ment services from the local employment associated with an increase in the provi- Supporting lifelong learning. Lifelong learn- of�ces. On-the-job training was found to be sion of training, especially among medium ing improves the adaptability and employa- more effective than classroom training, and and large �rms. Singapore’s Skills Develop- bility of workers as economies undergo eco- private training centers seem to outperform ment Fund relies on a levy of 1 percent on nomic and technological change. Worldwide, government-run centers. payroll for low-wage workers, and Training programs for youths, even reimburses the levies by the amount of annual spending on corporate training when well-targeted, tend to have a poor training that �rms provide. The number of reached $28 billion in 2002. By the end of the track record. Earlier interventions at the individuals trained has tripled since the 1990s almost half the workers age 35–54 in schooling stage are likely to be more effec- Fund’s inception in 1979. the United States were adult learners.16 tive than trying to later remedy education’s While these schemes can facilitate a failures. The experience of some Latin more systematic, structured approach to Although most workers are involved in some American countries offers interesting enterprise training, many �rms, especially on-the-job training, it is often not enough to insights, however. The “Jovenes� programs small ones, may not have the capacity to enable them to adjust to major changes in in Argentina, Chile, Peru, and Uruguay are provide training to their workers.Training targeted at disadvantaged youth—combin- funds are also dif�cult to manage in coun- technology or to move across different jobs. ing training and work experience with tries with weak administrative capacity and Firms themselves may have dif�culty inter- other services, including psychological where public provision tends to be supply- nalizing the returns to training investments development and vocational assessment. driven.To address these concerns, Kenya because workers may move to other �rms. At While effective in promoting employability has established a voucher scheme for train- of the targeted youths, they tend to be ing services that allows the trainee to the same time workers’ incentives to invest in costly. An evaluation in Argentina choose among providers and courses. training may be low if wages are compressed estimated that at least nine years of higher or if workers cannot �nance their training earnings due to the program would be because of credit market inef�ciencies. In all Source: Middleton, Ziderman, and Adams required to show a positive net present value for the groups with statistically signif- (1993); Calderon-Madrid and Belem (2001); these cases there is a role for government to Betcherman, Olivas, and Dar (2003); Aedo and icant results. Enhanced job opportunities Núñez (2001); and de Ferranti and others support training and retraining. Experience for the targeted group also tended to be (2003). with schemes to meet these goals remains mixed, however (box 7.3). (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 141 balance between these interests is struck Crafting interventions to bene�t will be influenced by social preferences in all workers each country. But as in other areas of gov- Governments intervene in worker–�rm ernment intervention, approaches can devi- relations on three main fronts. They inter- ate from the socially optimal level because vene in the wage-setting process, they regu- of factors such as rent-seeking by particular late working conditions, and they control interest groups and a failure to adapt the hiring and �ring of workers. These approaches to local circumstances (see interventions are theoretically justi�ed by chapter 2). Indeed, as in other areas of regu- the (perceived or effective) inability of lation, labor regulation in many developing laissez-faire conditions to deliver ef�cient countries mimics or exceeds that in devel- and equitable outcomes. Ef�ciency argu- oped countries,17 bene�ts only part of the ments stress information problems and a population because of widespread infor- need to improve the matching of labor mality, and imposes a disproportionate demand with supply. There may also be burden on those �rms that do comply equity arguments if there is unequal bar- (chapter 5). gaining power between employers and From an investment climate perspec- workers, discrimination against vulnerable tive, the question is how labor market groups, or incomplete or imperfect insur- interventions influence the opportunities ance of workers against risks. and incentives for �rms to invest produc- Beyond the core labor standards—the tively, create jobs, and expand. Firm-level minimum framework for a sound labor surveys show that labor regulations can be market (box 7.4)—government interven- a major or severe constraint on �rm oper- tions need to strike a balance between sev- ations in many developing countries (see eral interests. It has been common to por- �gure 7.1). Regulations can reduce incen- tray the tension as primarily between the tives to make new investments, adjust the interests of �rms and workers. But this organization of work to take advantage of ignores the broader range of interests new technologies or opportunities, or hire involved. Workers in the informal economy more workers. Some curtailment of those and the unemployed can have very different incentives can be justi�ed by social goals interests from those currently employed in beyond those reflected in the core labor the formal economy. And consumers and standards including, for example, the pro- potential recipients of tax-funded services motion of workplace safety. But ill- also have a stake in the outcome. Where the conceived approaches can exacerbate BOX 7.4 The core labor standards The international community, acting through can go a long way toward promoting labor mar- new technology that has higher productivity conventions elaborated through the ket ef�ciency and better economic performance. potential but requires more skilled workers. International Labour Organisation (ILO), has And there are obvious economic and social rea- Reforms that promote stronger economic identi�ed four core labor standards as the mini- sons for banning slavery and all forms of forced growth are fundamental to combating child mum for all countries, whatever their stage of labor. Unfortunately, child labor and different labor. In Vietnam strong economic growth in the development: eliminating all forms of forced or forms of explicit or implicit discrimination, while 1990s led to a signi�cant rise in poor families’ compulsory labor, abolishing child labor, provid- generally perceived as violations of human rights, wealth, reducing the number of children in the ing equal opportunity and nondiscrimination in are still widespread in many developing countries. workforce by 28 percent. Improving the delivery employment, and ensuring the freedom of asso- Child labor in particular still looms large in of education is generally more effective than ciation and the right to collective bargaining. the developing world, where one child in six banning child labor. Such bans are generally not The past decades have witnessed an accelera- between the ages of 5 and 17 is at work. Child enforced in many developing countries, and tion in the number of countries that have labor hinders human development, reducing where they are, can also force children into signed these conventions, particularly that ban- future earnings for the children and aggregate more dangerous, hidden forms of work (prosti- ning the worst forms of child labor. growth for the economy. For example, children in tution), especially where parents have no choice The economic effects of enforcing core labor India perform tasks that require no particular but to use child labor to survive. standards depend on the interventions and skills and develop no human capital. Cheap child Source: ILO (2003b); Burra (1995); Edmonds (2004); sociopolitical circumstances. Ensuring the labor, if combined with poor investment condi- Krueger (1996); Brown (2000); OECD (2000a); Martin freedom of association and collective bargaining tions, reduces the incentives for �rms to invest in and Maskus (2001); and Miles (2002). (c) The International Bank for Reconstruction and Development / The World Bank 142 WORLD DEVELOPMENT REPORT 2005 poverty by contributing to unemployment economy. This reinforces the importance of and swelling the size of the informal and looking at labor market policies in the con- unprotected economy. If a society’s goal is text of broader strategies, including efforts to advance the interests of all workers— to foster a more skilled and adaptable work- rather than just those who currently bene�t force and to help workers cope with change. from regulated employment—governments Governments can take three steps to need to confront these dif�cult and often ensure labor market interventions bene�t sensitive tradeoffs. all workers: Striking a balance between promoting • Encourage wage adaptability and ensure job creation by �rms and protecting existing workers are properly compensated for jobs or workers is particularly contentious their work during periods of economic reforms—when the long-term bene�ts of increased employ- • Ensure workplace regulations reflect a ment and wages are often clouded by short- good institutional �t term concerns for the job and wage security • Balance workers’ preference for employ- of those affected during the transition. Suc- ment stability with �rms’ need to adjust cessful reforms bring about higher wages the workforce. and better working conditions—as well as higher employment and lower unemploy- Encouraging wage adaptability ment and informality in the long run.18 Governments intervene in the wage-setting There are, however, short-term costs due to process by establishing rules for wage bar- changes in job characteristics and greater gaining and for industrial relations. These labor mobility in a modern, productive interventions can reduce negotiation costs if they do not reinforce the monopoly power of the parties or impose rigidities in wage BOX 7.5 The role and impact of unions adjustments. Many governments also set Trade unions can play an important role in greater in unionized �rms because unions wage floors in an attempt to reduce the num- representing the interests of workers.Their facilitated communication between man- ber of working poor, but setting the floors too impact on wages and economic conditions agement and workers. high can reduce the jobs available for low- varies greatly across countries and regions, Given the reductions in union member- skilled workers and the opportunities for however, and depends largely on the eco- ship in recent years, and the growing size of nomic and social context. Wage premiums the informal economy, unions in many low-tech �rms to emerge in the formal sector. for unionized work tend to be fairly small in developing countries have started to developed countries but quite high in coun- expand their engagement with the informal Wage bargaining bene�ts from a clear policy tries or sectors with weak competition in sector. A union in Argentina operates a framework. The dialogue between freely output markets and large rents. Available health insurance and unemployment fund estimates suggest high wage premiums in that also covers unregistered and elected (and representative) associations of countries such as Ghana (21–28 percent) unprotected agricultural workers. In the workers and employers can reduce uncer- and South Africa (10–24 percent) but much Philippines unions initiated loan schemes tainty and transaction costs and improve lower premiums in countries such as South for poor areas. In Ghana an agricultural information flows.19 Collective bargaining Korea (2 to 4 percent). workers’ union includes self-employed rural Union members also tend to enjoy workers as members; it supports them offers a platform for involving both longer job tenure and receive more training through revolving loans and facilitates their employers and workers in discussions with than their nonunionized counterparts. And in access to other forms of institutional credit. government about structural reforms. Con- a number of countries employers favor deal- In India a union helps unorganized and self- sider the tripartite negotiations promoting ing with unions, because highly representa- employed workers to obtain licenses. tive unions can reduce industrial unrest. Associations of informal workers have macroeconomic and structural reforms in The effect of unions on productivity is also been created, with some taking a high several western European countries in the less clear cut and depends on market condi- pro�le in defending informal workers’rights. past decade. Also consider the pivotal role tions and industrial relations. In Mexico Examples include the Ghana Private Road unions have attempted to protect low- Transport Union, the Cissin-Natanga Women’s of unions in promoting political openness skilled jobs at the expense of higher Association in Burkina Faso, and the Self and democracy in other countries, as with productivity. In Guatemala unionization is Employed Women’s Association in India. Solidarity in Poland and black labor unions associated with lower productivity of coffee in South Africa. But unions can sometimes farmers. However, greater participation of workers in some aspects of company man- Source: Aidt and Tzannatos (2002); Harrison and act as monopolists, improving wages and Leamer (1997); Maloney and Ribeiro (2001); agement in Brazil contributed to better pro- Urízar and Lee (2003); Menezes Filho and oth- conditions for their members at the expense ductivity and pro�tability.The effect was ers (2002), OECD (1997a); and Ratnam (1999). of nonunionized workers and broader soci- ety (box 7.5). (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 143 In industries where regulation shelters ing and transition economies have rein- �rms from competition, unions are likely to forced wage responsiveness by shifting bargain for a share of the rents. An unstable wage bargaining to the �rm. In the Baltic political environment also tends to reduce States, the Czech Republic, and Hungary, incentives for unions to “invest� in wage unionization is low in newly created pri- restraint in exchange for expected better eco- vate �rms, especially small ones, and wage nomic outcomes in the future.20 High union bargaining mostly takes place at the �rm wage premiums and bigger drags on produc- level.22 Along the same lines, the wage- tivity are indeed found in countries and sec- bargaining system in Peru was reformed tors lacking competitive pressure. Invest- in 1992, increasing direct negotiation by ment climate improvements that enhance relaxing the collective negotiation economic stability and competition in out- process, introducing voluntary arbitra- put markets are likely to lead unions to tion as an alternative to state administra- behave in ways more conducive to stronger tive decisions, and eliminating state economic growth and job creation.21 approval of agreements. The reform also increased collective autonomy by protect- Enhancing wage adaptability. Governments ing the unions’ right to registration, and can foster wage adaptability by promoting strengthened union pluralism by allowing pluralism of representation in wage bar- more than one union to exist in a �rm.23 gaining. They can also reinforce the links between wage agreements and �rm perfor- Reassessing minimum wages. The main goal mance either through improving coordina- of setting minimum wages is to promote tion among social partners or through decent jobs and reduce poverty among more decentralized negotiations. workers. But its effectiveness in many devel- oping countries is questionable. Minimum • Improving coordination. Some developed wages represent a high proportion of the countries with a tradition of collective average wages in these countries, and any bargaining have reinforced coordination further increase shifts the wage distribution among the different levels of wage nego- upward, punishing rather than helping the tiation (national, sectoral, �rm). In some workers intended to be supported—young, of them, such as Denmark, Italy, and low-skilled, and female workers. When Portugal, nationwide agreements now �x enforcement is weak, as is often the case, a only the basic wage increase, leaving to hike in the minimum wage encourages even the �rm-level negotiation further more underreporting of wages and increases consistent with a �rm’s perfor- strengthens incentives for �rms and jobs to mance. Unions have also been part of the remain in the informal economy. design and implementation of large The minimum wage cuts the lower end structural changes in many countries. In of the wage distribution and makes �rms Mexico and Israel, as well as in the and jobs with low productivity levels unvi- Netherlands, Ireland, and Italy, unions able, at least in the formal sector. The level have participated in the design of adjust- of the minimum wage affects �rms, jobs, ment programs, including actions in the and income distribution: labor market, and agreed on social pacts that facilitated macroeconomic stabiliza- • In developed countries minimum wages tion. In Kenya, following the abolition of tend to be relatively low (although in price controls in the mid-1990s, govern- some cases may approach 50 percent of ment guidelines on wages were removed, the median wage) with only a modest giving employers and workers greater impact on low-tech �rms and the employ- latitude in wage negotiations. ment of low-productivity workers.24 • Decentralizing negotiations. Following the • In several low-income countries mini- experience of other developed coun- mum wages are close to, if not higher tries—such as Australia, New Zealand, than, the average income per capita (�g- and the United Kingdom—some emerg- ure 7.4).25 At these levels many private (c) The International Bank for Reconstruction and Development / The World Bank 144 WORLD DEVELOPMENT REPORT 2005 Figure 7.4 The minimum wage is very high in many developing countries and, at high levels, leads to weak compliance Minimum wages in many low-income countries High levels of the minimum wage lead are high relative to income per capita to high evasion in Latin America 4 40 Compliance: workers that earn less than Nicaragua Ratio of yearly minimum wage the minimum wage (percent) 3 30 to GDP per capita Colombia 2 20 RB de Venezuela Costa Rica Panama 1 10 Peru Honduras Brazil Chile Bolivia Salvador Mexico Argentina 0 0 Uruguay 4 6 8 10 0 25 50 75 100 Log of GDP per capita Level of minimum wage relative (PPP constant dollars) to wage of median worker (percent) Note: In the left panel each dot represents one country/year observation. Data refer to the period 1980–2000. Per capita income is expressed in constant dol- lars adjusted for purchasing power parity (PPP). The wage used in the right panel is the median wage for workers between 26 and 40 years of age, who work for more than 30 hours per week during the reference period of the surveys. Source: Left panel: Rama and Artecona (2002). Right panel: IDB based on countries’ of�cial data. �rms, especially those in low-tech activi- the minimum wage can act as a strong pay ties, cannot afford to comply. The poor signal for the informal sector, implying that continue to work in informal activities hikes in the minimum wage can have distri- for only a fraction of the mandated min- butional implications that go beyond the imum wage. formal sector—the income of the low-paid • In middle-income countries, the mini- might increase in both segments of the mum wage is generally about half the economy, but their employment prospects median in the formal sector. Its coverage might decline.28 and enforcement tend to be low, but its Given these effects, a growing number of impact on low-productivity �rms and countries are reassessing minimum wages jobs can be large. In Latin America the to expand opportunities for low-skilled largest proportion of workers who earn workers and encourage formalization. They less than the minimum wage is found in have done so mainly by reducing indexation countries where it is comparatively high of the minimum wage and by having lower (�gure 7.4). Examples include Paraguay, subminima for some groups (young work- where the majority of workers earn less ers) or for subnational labor markets. For than two-thirds of the minimum wage; example, the erosion of the minimum wage Nicaragua (40 percent of workers below in Mexico in the 1990s is credited with the minimum); and Colombia (25 per- boosting female employment. Submini- cent).26 mum apprenticeship wages are estimated to have signi�cantly increased job opportuni- Noncompliance with the minimum ties for young graduates in Chile.29 wage is also concentrated among the most vulnerable workers. Youths and other work- ers lacking skills or work experience may Ensuring workplace regulations have little chance of being hired at the min- reflect a good institutional �t imum wage when it is set much higher than Promoting health and safety conditions in their productivity potential. In backward the workplace, regulating working time, areas the national minimum wage may be and encouraging paid leave have been close to the underlying local average wage, major achievements in all societies. As in severely affecting labor demand from small most other areas, improvements in working and medium �rms that rely largely on low- conditions in developed countries evolved skilled workers.27 Despite low compliance, gradually, hand in hand with more general (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 145 economic progress. Attempting to apply the Figure 7.5 Developing countries have more stringent regulations on worker hours and paid leave than many developed countries same or higher standards to countries at earlier stages of economic development and Europe & Central Asia with weaker enforcement capacity often Developed countries: others leads to poor or even perverse results. Improving workplace safety is an impor- Latin America & the Caribbean tant objective for all countries, and well- Middle East & North Africa designed regulation can help to achieve this Sub-Saharan Africa goal. But safety or other regulations will have limited impact if they or other features South Asia of labor regulation have the effect of keep- East Asia & Pacific ing �rms or workers in the informal econ- Developed countries: common law omy where workers usually lack any statu- tory protection. Stronger enforcement 0.0 0.2 0.4 0.6 0.8 efforts can help in some cases. When regu- Average index on annual leave and hours worked Note: Higher numbers mean more stringent regulations. lations are out of step with local realities, Source: Pierre and Scarpetta (2004). The indicator is based on the World Bank Doing Business Project. however, there will be tradeoffs between providing high levels of protection for workers that enjoy regulated employment • Paid annual leave. Some developing and expanding protection and opportuni- countries have mandated relatively gen- ties to a broader group of workers. erous annual leave—30 days in Burkina Regulations affecting working hours and Faso, 33 in Ethiopia, and 39 in Sierra paid leave can involve similar tradeoffs. Many Leone31—but in most other countries developing countries have adopted far-reach- paid annual leave is less than 30 days. ing regulations on these subjects—in some The United States leaves the decision on cases going beyond what is on the books in annual leave to individual or collective most developed countries (�gure 7.5).30 Even agreements. among countries at similar stages of devel- opment, the differences in regulations can These regulations can bene�t workers in be large, with signi�cant effects on labor the formal sector and, by promoting better costs and on the ability of �rms to accom- working conditions and motivation, can modate fluctuations in demand: contribute to productivity. Beyond any potential productivity effect, however, the • Workweek. Botswana, Chile, Costa Rica, impact on �rms’ incentives to create jobs Ireland, Malaysia, Morocco, the United depends on who bears the costs. The evi- Kingdom, and Vietnam all allow 48- dence suggests that wages do not fully hour workweeks. Most western Euro- adjust to compensate for the additional pean countries have 40-hour limits, with costs of these bene�ts. For example, in France recently moving to a 35-hour Latin America, �rms bear up to 50 percent workweek. In cyclical or seasonal indus- of the costs of nonwage bene�ts,32 thus tries, �rms often use overtime work to reducing �rms’ potential for expansion and accommodate demand. In Burkina Faso, job creation. These effects would not be a Cameroon, Hong Kong (China), Spain, source of concern if they reflected the and the United Kingdom, there are no rational choice of workers to trade off not regulatory requirements to pay a pre- only lower earnings, but also some unem- mium for overtime work. In Bangladesh, ployment, for better working conditions. Belarus, India, Nicaragua, Pakistan, and When this is not the case, workplace regu- Uzbekistan the mandated premium is up lations reduce wages below what poor to twice the regular pay. To promote workers would be willing or able to accept. employment, many developing coun- They can also encourage unregulated and tries are moving to liberalize restrictions unprotected employment. in these areas—examples include Hun- Indeed, workplace regulations have long gary, Latvia, Namibia, and Slovakia. suffered from poor compliance in many (c) The International Bank for Reconstruction and Development / The World Bank 146 WORLD DEVELOPMENT REPORT 2005 Balancing employment stability with BOX 7.6 Labor regulation and global integration �rms’ need to adjust the workforce Differences in labor regulations and their Indeed, a body of evidence suggests Probably the most contentious government enforcement might give a cost advantage in that multinational �rms tend to provide intervention in the labor market is the regu- internationally traded goods to countries with better working conditions and pay higher lation of the hiring and �ring of workers— weak regulations,and new technologies allow wages than alternative local employment. labor services to be directly subcontracted to The World Bank’s Investment Climate Sur- generally referred to as employment protec- workers in countries with less onerous regula- veys also suggests that foreign-owned �rms tion legislation. Regulatory intervention may tion.This has led to concerns that multinational tend to have a larger share of workers with be justi�ed to protect workers from arbitrary �rms may be exploiting weak labor regulation permanent contracts and tend to provide action and to provide some stability in or putting pressure on governments not to more training for their workers. enforce existing regulations. Multinational �rms concerned with employment, which can be particularly Evidence of noncompliance with labor maintaining their corporate reputations are important in the absence of effective social regulations abounds in developing countries, also increasingly adopting codes of conduct safety nets. To the extent job protection leads but there is no clear indication that this is that reflect global norms on a range of related to greater integration in the world issues, including labor practices (chapter 9). to long-lasting work relationships, it may market.This is true whether integration is Compliance with codes is monitored by also encourage �rms to provide training. measured by export market shares, revealed buyers or by independent auditors. But as elsewhere, governments need to comparative advantages, FDI, or trade prices. Poor working environment conditions are, balance these potential bene�ts against the Even in export processing zones—which are however,the reality for many workers at the often used by governments to attract invest- end of the supply chain.Only recently have likely costs. By affecting the cost of work- ment by providing �rms with a more favor- some multinational �rms revised their purchas- force reorganization, employment protec- able policy environment (chapter 8)—it is not ing practices and improved compliance with tion legislation can strongly influence the clear that enforcement of labor regulation is labor standards by local subcontractors. cost of doing business, especially the oppor- systematically lower than what is observed outside the zones. Of 73 zones reviewed in a tunities and incentives for �rms to adopt Source: OECD (2000a); Krumm and Kharas recent study, in only 6 was there any deliber- (2003); Basu (1999); Maskus (1997); Brown, new technologies and to expand. Modern ate attempt by government to restrict work- Deardorff, and Stern (2003); World Bank and IFC economies require a continuous process of ers’rights. (2003); OECD (2001); and Raworth (2004). �rms’ retooling and �rm turnover to chan- nel resources to their most productive uses. In countries for which data are available, developing countries. And while recent gross rates of job creation and destruction progress toward global integration is some- each range between 5 and 20 percent, times thought to result in a lowering of adding up to a total job turnover of up to 40 standards, experience suggests that this is percent (�gure 7.6). A signi�cant part of this not necessarily the case (box 7.6). job turnover (often 30–50 percent) is due to the entry and exit of �rms, an important Figure 7.6 High job turnover in developed and developing countries in the 1990s factor for output and productivity growth (�gure 7.7).33 Onerous employment protec- 20 8 tion legislation can discourage job creation Employment growth because �rms will be reluctant to hire work- ers if they face signi�cant costs in adjusting Employment growth (percent) Job creation 6 the workforce to changes in demand. As percent of total employment Job creation/destruction as Job destruction with other areas of labor regulation, onerous requirements in this area can also contribute 10 4 to the adoption of informal employment arrangements, where workers will receive no statutory protection. 2 Regulating hiring and �ring. The protection offered to regular workers and the condi- tions for temporary employment vary con- 0 0 siderably across countries (�gure 7.8). Countries in Latin America and in Eastern y ia il y a ce le S. a ia o ar az an bi tin ic on i tv U. Ch an om ex ng Br m La n Europe and Central Asia tend to offer the t Es Fr ge er M Hu l Co .G Ar strongest employment protection for regu- W Note: Data for Brazil, Colombia, and Chile refer to manufacturing. Data are from �rms that have 20 or more employees and cover different periods during the 1990s. lar workers.34 Common law developed Source: Bartelsman and others (2004). countries tend to have the least statutory (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 147 protection.35 Differences within regions are Figure 7.7 Job turnover is high because of both the entry and exit of �rms, and the reallocation among existing �rms also large. For example, most countries allow the termination of contracts under a Job destruction Job creation because list of “fair� causes, but the list can be very by incumbent firms of firm entry narrow, as in Bolivia, where redundancy is Job destruction Job creation by not considered a fair cause for dismissal. because of exit of firm incumbent firms Advance notice and severance payments W. Germany also range from a few days and a small pro- U.S. portion of the wage to several months and high compensation. In Sri Lanka dismissed Argentina workers receive 2–3 months salary for each Slovenia year of service, and severance payments in Colombia some cases exceed 25–30 months’ wages. Estonia Procedures for dismissal can also be Romania cumbersome and opaque. In Sri Lanka the government decides the amount of com- Hungary pensation for laid-off workers and has the Latvia authority to reject employer demands. The Chile time needed for processing the request for a Mexico layoff can be highly unpredictable, taking Brazil six months on average, but much more if the procedure involves hearings where –15 –10 –5 0 5 10 15 employers explain their �nancial perfor- Percent of total employment mance and business plans to the govern- Note: Data are for the manufacturing sector, �rms of 20 or more employees. Source: Bartelsman and others (2004). ment to justify the layoff. In Russia, before the reform of the labor code, trade unions had veto power over dismissals for staff Figure 7.8 Many developing countries have more stringent regulations on reductions or for employees not suited to hiring and �ring than developed countries the job.36 Before the 1999 reform in Brazil, repre- Developed countries sentatives of employers and workers sat on (common law) the jury of labor courts, a practice that often Employment Middle East & protection for those led to protracted procedures and dif�culties North Africa with permanent in reaching compromise. About 2 million contracts salaried workers (more than 6 percent of Sub-Saharan Africa Employment the total) usually �led a lawsuit every year protection for those with temporary and the average labor dispute took almost East Asia & Pacific contracts three years. The reform restricted the jury to professional lawyers and cut the time to Developed countries (other) resolve a dispute by half.37 Latin America & the Caribbean The impact on �rms. Firms in many develop- ing countries regard employment protection South Asia legislation as a signi�cant obstacle to their expansion. When asked to evaluate eight Europe & Central Asia areas of regulation for the burden imposed on the operation and growth potential of their businesses, �rm managers ranked labor 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Summary indicator of the stringency regulations as the major or secondmost of employment protection legislation important obstacle in many countries of Note: Higher numbers mean more stringent regulations. Data refer to employment protections Latin America, Central and Eastern Europe, afforded each type of contract. Source: Pierre and Scarpetta (2004) based on World Bank Doing Business Project. and South Asia. There is also a close correla- tion between managers’ perceptions of labor (c) The International Bank for Reconstruction and Development / The World Bank 148 WORLD DEVELOPMENT REPORT 2005 BOX 7.7 Do �rms’ perceptions square with actual labor regulations? The signi�cance of regulations in different mar- The Bank’s World Business Environment Sur- parison suggests that the more stringent the kets can be assessed in two main ways.The �rst vey asked managers in 73 developed and devel- regulations, the greater the likelihood that is based on international comparisons of formal oping countries how problematic they found �rms will report that labor regulations are a laws and regulations. When noncompliance regulations in different areas, including labor, for major obstacle. In other words, strict labor with regulations is high—as for labor the operation and growth of their �rms. Overall, regulations, even if not fully enforced, affect regulations in many developing countries— the data suggest that close to 70 percent of �rms’ performance by limiting their opportu- international comparisons of laws and regula- respondents reported some concern (minor, nities. Medium-sized �rms are most affected, tions may give rise to inaccurate assessments. moderate, or major) about labor market regula- while both small �rms and large tend to be Moreover, labor laws are often complex and tions. Around 15 percent reported that these less concerned. Downsizing �rms are more interact with laws in other areas.The second regulations were a major obstacle to the opera- likely than the average to report that labor approach is to ask those affected by speci�c tion and growth of their �rms. regulations are a major obstacle. Firms whose regulations, such as the employers.Their percep- These data can be combined with more business is expanding are on average less tions are subjective, however, and can be objective indicators of the strictness of concerned. affected by a range of factors. employment protection legislation. This com- Perceptions of the burden of labor regulation vary across countries and �rms By stringency of actual regulations By the performance of firms By the size of firms High Firms’ perception of the burden of labor regulation Average Low Low High Firms Firms Small Medium Large Stringency of employment expanding contracting protection legislation employment employment Note: The �gures are based on a sample of 9,000 �rms in 81 countries around the world. All estimations control for age and size of �rms, region, and public ownership. Small �rms are those with less than 20 employees; medium �rms have 20–100 employees; large �rms have over 100 employees. Source: Pierre and Scarpetta (2004), Bertola, Boeri, and Cazes (2000), and Batra, Kaufmann, and Stone (2002). regulations and the stringency of such regu- and development (R&D) expenditure lations from a more objective standpoint and tend to tilt specialization away from (box 7.7). high-tech industries. For example, one Onerous regulations can affect �rms’ cross-country study suggests that by spending on innovation, the entry of new reforming their labor rules to the OECD �rms, their average size, and the incidence average, developed countries with very of informality. strict employment regulations could reduce their productivity gap with the • Cost of doing business and exploiting tech- technological leader by about 20 nological opportunities. Onerous hiring percent.38 Similar reforms in developing and �ring regulations raise the cost of countries could yield even larger produc- workforce reorganizations required by tivity gains, given the greater potential new vintages of technology, reducing for catchup by adopting technologies incentives for �rms to innovate and available in international markets. adopt new technologies. Evidence from developed countries suggests that stricter • Creative destruction. Onerous regulations rules are associated with lower research also have repercussions on the turnover (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 149 of �rms in the market. Because new �rms Figure 7.9 Strict labor regulation is not associated with more equality in the labor market are often better at harnessing new tech- nologies than incumbent �rms, stringent 60 regulations reduce the potential for pro- ductivity gains. Data for 19 developed Gini index of income inequality and developing economies suggest that 50 countries with more flexible hiring and �ring rules experience signi�cantly higher entry rates of small �rms (but not 40 microenterprises, often exempt from such regulations or managing to avoid them). Stringent rules also tend to dis- 30 courage foreign direct investment (FDI), especially in countries where rules are opaque and enforcement is uncertain.39 0 • Self-employment and informality. Oner- 0.5 1.0 1.5 2.0 2.5 Employment laws index ous labor regulations are associated with larger proportions of self-employed, Note: Employment laws index is for year 2000; Gini data are for 1995–2000. Higher values of the employment laws index are associated with more stringent labor regulation, and a higher Gini informal �rms, and small �rms.40 Firms coef�cient indicates greater income inequality. Source: Authors’ calculations from World Development Indicators, and World Bank Doing Busi- facing high labor adjustment costs either ness Project. remain very small—and more or less informal and thus exempt from employ- ment regulations—or move to a higher experience, and those with low skills.43 The scale or to more capital-intensive tech- incidence of long-term unemployment nologies, in both cases reducing the inci- (more than 12 months without a job) is low dence of hiring and �ring costs on total in the United States (6 percent of total unem- expected adjustment costs. In Russia ployment) and other countries with moder- many large �rms have circumvented ate employment protection legislation, but it strict regulations by pushing workers to is more than 50 percent in many European leave the �rm voluntarily, through wage countries with more onerous regulations. arrears, prolonged administrative leaves, When compliance is weak, as it is in many reduced hours, and other forms of dete- developing countries, stringent regulations riorating working conditions. With no do not reduce the size of labor reallocation, future in the �rm and no source of but they do change its nature and reduce its income, many workers eventually quit.41 effectiveness. In Argentina—a country with fairly rigid labor regulation—job flows had a Onerous employment protection legislation negative contribution to aggregate produc- hurts vulnerable groups. To the extent strin- tivity growth in the 1990s, as many workers gent regulations reduce the potential for �rm transited from formal jobs to jobs in the expansion and job creation in the formal sec- informal economy.44 Similarly, in some of tor, they also reduce workers’ access to decent the transition countries lagging behind in jobs. More job stability for some workers market-oriented reforms, stringent labor often implies fewer job opportunities in the regulations have not prevented job destruc- formal sector. So it is not surprising that tion—but rather discouraged job creation in stricter employment laws are not associated the formal economy. This has led to job with a more equal labor market. If anything, destruction leading job creation (or unsyn- income disparities tend to be greater in coun- chronized job flows) and the buildup of a tries with stricter regulations (�gure 7.9).42 large pool of unemployed or informal work- Strict regulations in developed countries, ers (�gure 7.10). Women, young people, and where compliance is high, tend to promote the unskilled—facing greater dif�culties in job stability for prime-age males but reduce obtaining a job in the formal sector—are job opportunities and lengthen unemploy- more frequently unemployed or engaged in ment spells for youths, women lacking work informal activities. (c) The International Bank for Reconstruction and Development / The World Bank 150 WORLD DEVELOPMENT REPORT 2005 Figure 7.10 Unsynchronized job creation and destruction can give rise to unemployment or underemployment Synchronized job flows Less synchronized job flows Percent of total employment 16 Estonia 16 Russia Percent of total employment Job destruction 12 12 8 8 Job destruction 4 4 Job creation Job creation 0 0 1990 1992 1994 1996 1998 2000 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 16 Slovenia 16 Romania Percent of total employment Percent of total employment Job destruction Job destruction Net 12 12 employment loss 8 8 Job creation Job creation 4 4 0 0 1993 1994 1995 1996 1997 1998 2000 1994 1995 1996 1997 1998 1999 Source: Bartelsman and others (2004) and Brown and Earl (2004). Figure 7.11 Since the labor reform of 1990, there has norms. Colombia and Peru liberalized their been higher job turnover in Colombia employment protection in the 1990s, mov- 16 ing their legislation closer to the standards Labor market of the (still quite regulated) European reform developed countries. The reforms led to a 14 Percent of total employment Job creation higher response of employment to output growth, with speedier employment adjust- 12 ment (�gure 7.11) but also positive employ- ment effects. In Colombia the reform also 10 contributed to increased compliance with labor legislation by lowering the costs of formal production. A recent study on India 8 Job destruction suggests that amendments to the strict employment regulation in one state 0 (Andhra Pradesh) in the 1980s allowed 1.8 1988 1989 1990 1991 1992 1993 1994 1995 1996 million urban poor to �nd jobs in manufac- Source: Bartelsman and others (2004). turing and service companies in the next decade.45 Italy and Spain also experienced sizable positive effects on employment after Reducing labor adjustment costs and for- some easing of their restrictive �ring regu- malizing work relations. Reforming govern- lations in the past decade.46 Similarly, after ments have adopted two main strategies to more than a decade of debate, both Egypt reduce labor adjustment costs. The �rst and Morocco revised their labor codes eas- focuses on reducing the burden of adjust- ing contract termination for economic rea- ment for workers hired under regular sons. In Kenya since the mid-1990s employ- employment contracts by bringing stan- ers no longer have to seek permission from dards more in line with international the government to dismiss workers. (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 151 A second strategy focuses on liberalizing sectors in response to changes in technology, �xed-term or temporary contracts, an demand, and other conditions. While this approach pursued by several countries in reallocation of labor bene�ts society as a Western Europe, Latin America, and Central whole, workers may need to change jobs sev- and Eastern Europe. Surveys in many devel- eral times in the course of their working lives. oping countries show that �rms facing strict This has long been a feature of work in the regulation of regular contracts make greater informal economy, but can be painful for use of temporary employment to foster the workers who have grown accustomed to more adaptability of their workforce. In 1991 Peru stable employment in protected industries. revised its labor law by lengthening the max- Helping workers cope with these changes not imum duration of temporary contracts. only bene�ts the individuals concerned, but The number of workers on term contracts can also enhance economic ef�ciency insofar shot up, and young and informal workers as it enables better matches between worker bene�ted the most. Poland, Russia, and abilities and the requirements of new jobs. It Slovakia have also recently increased the can also reduce resistance to investment cli- duration of term contracts and expanded mate improvements. In many developing their applicability.47 countries, inadequate or nonexistent social But liberalizing temporary contracts, while insurance mechanisms mean that unem- leaving in place strict regulations on regular ployed workers cannot afford to remain with- contracts, reinforces the inequality in the out income and are forced to accept the �rst labor market. Firms will have stronger incen- job that comes their way, even if it is not a tives to hire more workers at the entry level good or productive one (�gure 7.12). and employ them for a limited period, with- Improving government policies in these out giving them a regular position thereafter. areas requires three interrelated actions: This increases job turnover but not necessar- • Helping workers affected by large-scale ily overall employment or productivity, restructurings because the additional hires will be accompa- nied by additional layoffs at the end of the • Reinforcing social insurance mecha- temporary contracts, and there will be little or nisms no development of internal human capital.48 • Reaching out to the large share of work- The effects of reforming labor regula- ers in the rural and informal economies. tions are likely to differ depending on initial conditions and on the sequencing of the Figure 7.12 Developing countries, particularly low-income ones, offer much weaker reforms in product and labor markets and less diverse protection against unemployment risks than developed countries (chapter 3). For example, stringent employ- 100 ment protection legislation can influence High-income 90 the outcomes of trade liberalization by Upper-middle-income shifting more jobs to the informal econ- 80 Share of countries in the group omy.49 Colombia’s trade liberalization was Lower-middle-income 70 associated with increased informal employ- Low-income ment in industries with the largest tariff 60 cuts, but once labor market reforms were 50 introduced, this pattern was reversed. Simi- larly, Indian states with less stringent labor 40 regulations experienced stronger growth in 30 the formal sector after trade liberalization 20 than those with stricter labor regulations.50 10 Helping workers cope with 0 change Unemployment Unemployment UI savings Severance Public Investment climate improvements that help insurance (UI) assistance accounts pay works create a modern, productive economy facili- Note: Based on the presence of the following programs: unemployment insurance, unemployment assistance, unemployment insurance savings accounts, mandated severance pay, and public works programs. tate the reallocation of labor across �rms and Source: Vodopivec (2004). (c) The International Bank for Reconstruction and Development / The World Bank 152 WORLD DEVELOPMENT REPORT 2005 Helping workers cope with Bangladesh provided for jute workers.53 To large-scale restructurings make schemes more effective, early interven- There is often strong pressure to compensate tion and effective targeting are essential, as groups directly threatened by structural are efforts to tailor approaches to local cir- reforms, such as workers in previously pro- cumstances. Particularly when labor demand tected industries. Typically not poor, these is weak, removing impediments to job cre- groups are very vocal and could represent ation through investment climate improve- concentrated opposition to reforms that ments plays a critical role. bene�t society as a whole. Providing one- time compensation to them may be a socially Reinforcing social insurance to ef�cient way to allow reforms to move ahead. promote labor mobility Workers affected by large-scale dismissals A variety of strategies can be adopted to can also face particular dif�culties. They may help workers cope with the income risks be specialized in activities that may not be associated with external or domestic shocks highly demanded in the broader economy, as well as the demands of a more flexible and may be concentrated in speci�c locations, labor market. Sound macroeconomic poli- making it more dif�cult to regain employ- cies and public support for education are ment locally. This was the case for most tran- the best risk prevention instruments. Social sition economies, where many one-company protection programs can also mitigate the towns and certain rural areas experienced a impact of risks while encouraging ef�cient collapse in labor demand and major surges in labor reallocation and entrepreneurship. unemployment and underemployment. Even when public resources to �nance these The traditional approach to dealing with schemes are limited, as in the case of most large dismissals is to promote voluntary developing countries, much can be done to departures with generous severance pay.51 improve their effectiveness by reinforcing This can reduce worker opposition and the insurance principles and better targeting. social impact of restructuring or downsizing. The policy mix best suited to each coun- The challenge is to set severance pay at a level try depends on the factors driving eco- that will be acceptable to workers yet be nomic insecurity and the cost-effectiveness �nancially feasible. Setting severance pay at of alternative options.54 However, interna- too high a level can lead to high short-term tional experience highlights the importance costs and the adverse selection of the best of four broader measures: employees leaving �rst. It can also slow or even stop the process of �rm restructuring. In • Reducing economic volatility. Many devel- Ghana downsizing was halted because the oping countries remain vulnerable to government could not afford the severance external shocks. When a negative aggre- payments. In the 1990s Pakistan made sever- gate shock hits the economy, capital— ance payments to workers affected by the pri- often the most mobile factor of produc- vatization of industrial units that included tion—tends to leave the country, while �ve months’ salary for each year of service— labor tends to bear the brunt of the much higher than international norms. The adjustment in either real wage cuts or agreement set a precedent for the later priva- unemployment and underemployment. tization of public utilities, delaying reforms.52 Export diversi�cation can reduce expo- Governments can also provide speci�c sure to large fluctuations in external retraining programs to help workers regain demand, and deeper capital markets and employment, but when these programs stronger �nancial systems can help miti- operate in a context of weak labor demand it gate the impact. The welfare bene�ts from is dif�cult to identify the best training curric- reducing macroeconomic volatility in ula and to motivate workers to participate. In developing countries can be substantial.55 many cases a small proportion of eligible • Moving away from procyclical �scal policy. workers takes these courses, which often The exposure of workers to shocks is come too late, after workers have already left, compounded by the fact that govern- as was the case with the retraining ments often lack the discipline to pro- (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 153 mote countercyclical �nancing for social with lower returns, reducing their income programs. Many governments tend to potential by an estimated 25 percent in rural adopt an expansionary �scal stance in Tanzania and 50 percent in a sample of rural good times and a contractionary stance villages in India.58 Similarly, uninsured risk in bad. Mounting budget de�cits in can lead to the use of outdated but less risky recessions thus create pressures to reduce production technologies, such as holding public spending on social protection livestock as a form of precautionary savings. (among other things) just when the need Second, uninsured shocks that reduce indi- for it is increasing. Greater �scal disci- vidual consumption below the threshold pline and better diversi�cation of the �s- needed to maintain productivity can give rise cal revenue base are essential to ensuring to “dynamic poverty traps.� This happens resources are available to cushion the when families are forced to sell productive necessary labor adjustment process. assets needed to support their microenterprises • Removing market distortions. Beyond or other ventures.59 Third, unemployment- macroeconomic policies, the most effec- related bene�ts can provide resources to tive strategy for risk prevention and mit- increase the effectiveness of the job search or igation is to develop a sound investment to enter self-employment.60 climate where �rms have opportunities Expanding and improving social insur- and incentives to invest productively and ance schemes can involve reinforcing self- create jobs. Investment climate improve- insurance among workers in the formal ments allow for stronger job creation in economy through severance pay arrange- the formal sector and expand tax ments and increasing the pooling of risks resources available for social programs. across workers. Improving the operation of �nancial markets also expands opportunities for Reinforcing self-insurance among formal �rms to insure themselves against tem- workers. Mandatory severance pay provisions porary shocks without resorting to wage are the main form of insurance against or employment cuts.56 unemployment for workers in the formal sec- • Supporting workers’ adaptability. In addi- tor in most developing countries. Generally tion to improving the coverage and qual- easy to administer, the provisions exchange ity of education, governments can resources in the event of unemployment for improve the ability and willingness of an “insurance premium.� Whether the sever- workers to move to more productive and ance pay premium is paid by the workers or rewarding jobs by supporting training, not has implications for the overall labor counseling, and placement services. While costs for �rms and hence their incentive for the effectiveness of these programs is hiring workers in the �rst place. Even when mixed, especially in countries with limited workers bear the costs, the schemes offer only administrative capacity, when well-targeted a limited pooling of unemployment risk they can complement skill enhancements because they are �rm-speci�c and because and income support measures. bene�ts generally evolve with job tenure rather than the risk of unemployment.61 These broader measures can be accompa- Severance pay provisions also suffer nied by social insurance schemes. Beyond from noncompliance in many countries, enhancing the welfare of the unemployed, increasing worker resistance to leaving a these schemes improve the investment cli- job. Required disbursements of severance mate by facilitating the allocation of labor to payments tend to increase when �nancial more productive uses and encouraging entre- resources are lacking because the �rm is preneurship. They do so in three main ways. experiencing dif�culties—and the resources First, they can stimulate riskier but more pro- may simply not be available if the �rm goes ductive jobs, industries, and portfolio bankrupt. Noncompliance looms particu- choices.57 For example, lack of access to larly large among small �rms and among insurance among poor rural households low-skilled workers who have few alterna- pushes them to take up low-risk activities tive instruments to smooth consumption.62 (c) The International Bank for Reconstruction and Development / The World Bank 154 WORLD DEVELOPMENT REPORT 2005 BOX 7.8 Reforming severance pay in Colombia and Chile In 1990 Colombia introduced fully funded sev- provided severance pay (for example, �rms To stimulate reemployment, bene�t recipi- erance-pay savings accounts, requiring about to go bankrupt could simply not pay sev- ents �rst draw resources from their own employers to deposit a percentage of wages erance or could negotiate a package substan- accounts, and upon their depletion, from the into guaranteed individual accounts available tially below what was owed in severance pay- solidarity account. Withdrawals from individual to workers in the event of job separation (lim- ments).The prefunding requirement increased accounts are triggered by separation from the ited access to funds while employed was also the likelihood that the legal entitlement to sev- employer, regardless of the reason. Insuf�cient foreseen). The reform reduced labor market erance pay would actually be carried out.The resources in individual accounts trigger distortions and promoted job creation. new severance-pay savings accounts also withdrawals from the solidarity fund if the Employers shifted most of the cost of reduce transfers from other government claimant meets the criteria for unemployment severance payments onto wages, but the total programs as well as from relatives. insurance (such as not working and being avail- compensation of workers (wages plus In 2002 Chile introduced a new unemploy- able and searching for job). Withdrawals are lim- deposits to their savings accounts) rose. In ment insurance system that combined social ited to two every �ve years. Bene�ts are linked addition, because the reform removed the dis- insurance with self-insurance. Employers and to past earnings, with a declining schedule. cretionary nature of severance payments, both employees both contribute to individual savings Workers can also move any unused savings from job separations and hiring increased. accounts, but an additional contribution from their individual accounts to their old-age pen- By transforming uncertain and conditional employers and a small public subsidy are allo- sion accounts on retirement. payments to unconditional payments cated to a solidarity fund.The new program is monitored by the government, the reform also effectively a funded system, with individual enhanced the insurance function of severance accounts managed by an administrator selected Source: Vodopivec (2004); Kugler (2002); and pay. Before the reform, few �rms actually through a competitive tender. Acevedo and Eskenazi (2003). To tackle these shortcomings, some coun- omy that offers many opportunities for tries have introduced pre-funding or undeclared paid work. In Argentina, for brought payments more in line with inter- example, the administration of unemploy- national norms. Colombia moved toward a ment bene�ts was found to involve signi�- funded system under individual savings cant leakage of bene�ts to those who have accounts in 1990, and Chile introduced a found jobs in the informal economy.66 social insurance component to its system in Even when countries have the required 2002 (box 7.8). administrative capacity, unemployment bene�ts should provide only a fraction of Increasing the pooling of risks across work- the previous wage—and they should be ers. Experience in developed countries sug- short-lived—to provide incentives for gests that unemployment insurance bene�ts recipients to seek a new job. Poland intro- are the next natural step to pooling unem- duced a generous and open-ended unem- ployment risks and facilitating ef�cient ployment insurance scheme in the early labor allocation.63 Following this model, 1990s, offering it to all job seekers irrespec- most transition countries have introduced tive of whether they had lost a job. Not sur- unemployment insurance schemes since the prisingly, the number of claimants soared, early 1990s. The schemes have been the making the system �nancially unviable and main source of income for workers affected contributing to the buildup of a large pool by labor reallocation during the transition.64 of long-term unemployed. The scheme, The clear welfare gains for workers affected later reformed to reduce disincentive by job losses need to be weighed against the effects, now provides a low flat bene�t for a costs of these schemes, including their limited duration. The Czech Republic, by impact on economic ef�ciency. Both the contrast, opted for less generous, short- costs and the impact depend largely on the lived bene�ts (only six months) and, partly ability to monitor eligibility requirements to because of this, had lower unemployment minimize moral hazard and make sure that in the early phases of the transition. workers have incentives to actively search for a new job.65 Effective enforcement is dif�- Reaching out to workers in the rural cult in developing countries, which gener- and informal economies ally have weak public employment services Most of the programs discussed so far fail to or none, coupled with a large informal econ- reach workers in the rural and informal (c) The International Bank for Reconstruction and Development / The World Bank Workers and labor markets 155 economies, which in many developing insurance—not emergency—programs for countries account for the majority of the informal and rural workers. The schemes population. They typically rely on support generally transfer income to poor house- from employers or private transfers to cope holds by providing unskilled manual work- with income losses. Rural employers often ers with short-term employment on pro- pay workers a �xed wage when they are jects such as road construction and employed, regardless of seasonal and other maintenance, irrigation infrastructure, fluctuations in demand, or provide loans to reforestation, and soil conservation. workers who face unexpected expenses.67 Workfare programs have often smoothed Given the informality of the employment consumption and kept poor people in con- arrangement, employers have a lot of dis- tact with the labor market.71 Well-designed cretion. Poor households also rely on their programs build much-needed infrastruc- own savings and private transfers to cope ture and so reduce the tradeoff between with shocks. In Indonesia, the Philippines, public spending on income transfers and on and Russia, private transfers account for development. The Maharashtra Employ- between 2 percent and 41 percent of ment Guarantee Scheme in India, operating income for net receivers and between 1 per- for more than three decades, has created cent and 8 percent of income for net considerable irrigation, infrastructure, and givers.68 A study in Kyrgyzstan found that rural roads in the state of Maharashtra.72 private transfers are provided to 12 percent Workfare programs have also helped many of households and account for more than small private contractors emerge and grow. one-third of the incomes of the households A key feature of successful workfare pro- who receive them.69 grams is the ability to target participants These forms of private risk-coping pro- through self-selection processes. In Argentina vide only limited help to poor and informal the Trabajar program kept the wage rate workers, and can force people to resort to below the minimum wage, encouraging the unproductive strategies including selling poor to self-select into the program. In the productive assets, withdrawing children Philippines, in contrast, the program wage from school, and cutting back medical was much higher than the agricultural market expenditures.70 The most promising strat- wage, attracting a substantial number of non- egy for improving their situation is through poor into the program. Kenya, Malawi, Mali, investment climate improvements that and Senegal also paid wages above the market expand job opportunities in the formal wage rates, undermining the self-targeting economy and contribute to greater tax rev- design and diverting jobs away from the very enues to fund the provision of education poor.73 Self-selection of participants can be and other services. But governments can accompanied by targeting to the poorest areas also complement private risk-sharing with to ensure that programs also promote local targeted public support. Three main strate- development. In South Africa a demand-dri- gies have been adopted in developing coun- ven approach to the allocation of funds for tries that can also contribute to better workfare programs in the mid-1990s was investment conditions: workfare programs, found to favor more developed and better social funds, and conditional cash transfers. connected communities at the expense of some of the neediest communities.74 Workfare programs as social protection schemes. In South Asia workfare programs Social funds to improve opportunities—and started as “food-for-work� schemes, in the investment climate—in poor areas. which workers were paid for their labor Social funds, introduced in Bolivia in the with food aid from donor countries. Work- late 1980s, have become one of the main fare programs have gradually moved to tools of community-led poverty reduction. “cash for work,� operated by a variety of They �nance small projects in poor com- agencies, including local and state govern- munities. Early programs focused on pro- ments and nongovernmental organizations viding temporary work opportunities while (NGOs). They are increasingly viewed as also �nancing better access to basic services. (c) The International Bank for Reconstruction and Development / The World Bank 156 WORLD DEVELOPMENT REPORT 2005 Recent programs give greater emphasis to response to economic crisis (Colombia) or a service delivery and connecting communi- natural disaster (the earthquake in Turkey). ties—which generally identify and partly In others they address long-term human �nance projects—with local governments. development goals, such as school enroll- Social funds in developing countries now ments in Nicaragua. absorb close to $10 billion per year in for- As with any transfer program, condi- eign and domestic �nancing. tional cash transfers can be problematic A recent review of social funds in Arme- when the increased demand for services is nia, Bolivia, Honduras, Nicaragua, Peru, and not met by increased supply (schools or Zambia offers a fairly positive assessment of clinics) or when the targeting is not suf�- their effectiveness in providing income sup- ciently robust. However, evaluations show port and promoting local development.75 that they can raise school enrollment and Evidence suggests that spending was highly attendance rates and improve child health progressive, with poor districts and poor and nutrition.77 The Mexican program households receiving more per capita sup- Oportunidades increased primary school port than wealthier districts or households.76 attendance by more than 2 percent and sec- Schools and health centers that received ondary enrollment by more than 8 percent, funds have enjoyed equal or greater access to while increasing health visits by some 20 staff and inputs and greater participation by percent. Likewise, Brazil’s Bolsa Escola local communities than other institutions. reduced school dropout rates from 5.6 per- The effects on poor households can also be cent to 0.4 percent.78 The programs also sizable. Investments in school infrastructure tend to be better targeted than general sub- were estimated to have increased primary sidies because of proxy means testing and enrollment rates, especially in Armenia, geographic targeting. They are also trans- Nicaragua, and Zambia. parent about who receives the transfers, and the level of bene�ts and the number of ben- Conditional cash transfers to preserve e�ciaries can easily be adjusted to take human capital and health. Conditional cash account of changing circumstances. transfers are another way to combine income support with local development. They belong to a family of transfer programs that Creating a better investment climate is fun- combine close targeting with capital accu- damental to improving the lives of people, mulation by making income support condi- including in their capacities as workers. An tional on either basic needs triggers, such as investment climate that bene�ts all mem- utility offset payments (in some transition bers of society looks beyond the protection economies), or behavioral changes, such as of existing jobs and confronts the challenge the continued school enrollment of children of creating opportunities for those in the or attendance at health clinics. They typically informal economy, the unemployed, and address chronic poverty rather than idiosyn- young people joining the workforce for the cratic risks of job loss. �rst time. Labor market policies that meet The focus of conditional cash transfers on this test play a critical role in the investment human capital formation makes them suit- climate by helping to connect people to able to address poverty and local develop- opportunities. ment at the same time. In Mexico Oportu- This and previous chapters in Part II nidades (formerly Progresa) reached 2.3 focused on delivering the basics of a sound million families in 1999. In Brazil (Bolsa investment climate. Part III considers Escola and PETI) and Jamaica (PATH), con- whether there is something extra that gov- ditional cash transfers are used largely to pro- ernments might do—beyond the basics—to mote the health and education of children. In improve the investment climates of their some countries the transfers are a quick societies. (c) The International Bank for Reconstruction and Development / The World Bank Going Beyond the Basics? III GOVERNMENTS CAN GO BEYOND THE BASICS of a sound investment climate by conferring special policy privileges on particular �rms or activities or by drawing on the growing body of international rules PA RT and standards that deal with investment climate issues. This part looks at the role these measures might play in creating a better investment climate. Chapter 8—Selective interventions reviews international experience with a variety of strategies and highlights the special challenges of each. Chapter 9—International rules and standards looks at how these measures might contribute to better investment climates, and the challenges they can present for developing countries. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank Selective interventions 8 The approaches to improving the investment target speci�c industries or activities. They climate discussed in Part II can bene�t all vary in their policy instruments, too, from �rms and activities in the economy. Given market restrictions, to special tax or regula- chapter the breadth of that agenda, some �rms or tory privileges, to information-based strate- activities may bene�t from improvements gies, to enclave approaches or “clusters,� to earlier than others—as with infrastructure in directed or subsidized credit, to public risk- a particular region, or regulatory reforms sharing. Some interventions have an eco- affecting a particular activity. As stressed in nomic rationale—externalities or other chapter 3, policy perfection isn’t needed to market failures.4 Some may be regarded as a ignite signi�cant growth and poverty reduc- form of “second best� response given slow tion. The key is to address important con- progress in addressing the basics.5 Yet oth- straints in a way that gives �rms con�dence ers seek to accelerate growth by fostering to invest—and to sustain a process of ongo- particular industries. Whatever the ratio- ing improvements. But beyond the sequenc- nale, all such schemes must navigate the het- ing of reforms, beyond delivering the basics erogeneous and self-interested requests of of a good investment climate, can govern- �rms, rent-seeking pressures, and other ments accelerate growth by providing special sources of potential policy failure. and more selective support to particular This chapter begins by examining some �rms or activities? Possibly. of the general lessons in undertaking selec- Governments have been experimenting tive interventions. It then looks at emerg- with such selective interventions for a long ing practices aimed at several common time. In the 14th and 15th centuries, English objectives of such interventions: integrat- monarchs encouraged further processing of ing �rms in the informal and rural the wool industry.1 After World War II many economies, unleashing the growth poten- developing countries pursued “infant indus- tial of smaller �rms, taking advantage of try� strategies to support local industries by international openness, and climbing the erecting import barriers—with nominal tar- technology ladder. iff rates for consumer goods exceeding 250 percent in Argentina, Brazil, and Chile.2 In The allure—and traps— the 1960s and 1970s several East Asian of selective interventions countries undertook selective interventions If speci�c activities or industries that are to support export-oriented industries— sure to deliver strong bene�ts could be prompting an ongoing and sometimes identi�ed and targeted cost effectively, heated debate on the desirability, ef�cacy, growth might be ignited or accelerated and replicability of such strategies.3 without addressing the often dif�cult chal- The experiments continue to this day, lenges in improving the basics of a good with governments pursuing a wide variety investment climate. Such strategies also of strategies and approaches. They vary in hold great political appeal. Governments their special efforts—to accelerate research often feel under pressure to be seen as pro- and development or regional development, moting economic development, and �rms to promote foreign direct investment (FDI) bene�ting from preferential treatment wel- or exports, to help small or rural �rms, to come their special privileges.6 That is why 159 (c) The International Bank for Reconstruction and Development / The World Bank 160 WORLD DEVELOPMENT REPORT 2005 governments explore the feasibility of vari- worthy of special treatment on this basis ous selective interventions. alone. Even within a country the goal of Experience suggests that such strategies expanding economic activity and employ- are far from straightforward—and can go ment in a given location may prompt special spectacularly wrong. There are three general efforts by local governments to attract invest- challenges: identifying candidates that merit ment. Particular types of �rms—such as special policy treatment, resisting rent- small and rural �rms—are also often believed seeking, and ensuring that any intervention is to suffer special disadvantages that justify cost effective. additional measures. In other cases governments seek to target Identifying candidates that merit particular industries through special policy special policy treatment treatment. Sometimes the choice of industry Some interventions are motivated by broad to target might appear fairly clear: for exam- notions of market failure. As discussed in ple, many countries that are natural resource chapter 3, research and development, FDI, exporters have an interest in increasing the and (possibly) exports can create positive level of processing in their economies, and a spillovers for the economy, and so may be country endowed with tourism assets may seek to leverage that advantage. Sometimes governments look beyond obvious areas of BOX 8.1 Unforeseen successes in Bangladesh comparative advantage in the hope of pro- and Kenya moting industries that promise even higher returns. While schemes of the latter kind may Bangladesh and Kenya show how tough it is exporter of horticultural products—fruits, veg- promise large bene�ts, experience shows they for a government to predict a winning sector. etables, and cut flowers.Among developing Garments in Bangladesh. Hoping to beat countries, Kenya is now the second-largest are also far more challenging. U.S. quotas and get rid of old textile exporter of fresh vegetables to the European Industrial development is usually a process machinery, South Korea’s Daewoo teamed Union and the second-largest exporter of cut of discovery, and it is dif�cult to predict what with a Bangladeshi entrepreneur in the joint flowers.Horticultural exports exceeded $350 venture Desh garments in 1979. Desh’s million in 2003, surpassing coffee exports, and a country or region will be good at produc- employees and managers spent some time the sector employs over 135,000 people, ing.7 There is no shortage of examples of gov- in Korea to learn new processes and man- many of them women.The sector emerged ernments missing what turned out to be win- agerial techniques. Nobody (not even Dae- from the entrepreneurial efforts of �rms, not ners—garments in Bangladesh, cut flowers in woo) had very high expectations for Desh, from government intervention.Smallholder but it turned out to be successful. Eventu- farmers, foreign investors, exporters from the Colombia, software in India, horticulture in ally, all but 5 of the 130 original workers left Kenyan Asian minority—all played important Kenya, and Honda and Mitsubishi in Japan’s Desh to create their own factories or join roles in developing contract farming arrange- automotive industry (box 8.1).8 And many other new businesses. Bangladesh became ments, introducing new technologies and interventions targeting speci�c industries a major player in the garment industry, with varieties, and connecting the horticulture sec- close to 1 million workers, most of them tor to global markets. have ended up producing losers (box 8.2). women, and exports in 2003 of $3.6 billion. Even where selective intervention seems Horticultural products in Kenya. Over the Source: Easterly (2001); Rhee (1990); and Eng- to have been successful, the contribution to last 10 years Kenya has become a major lish, Jaffee, and Okello (2004). growth has been debated. For example, recent work suggests that South Korea’s promotion of its heavy and chemical indus- BOX 8.2 Picking “winners� can be an expensive tries did not have a clear impact on growth.9 gamble—SOTEXKA in Senegal Measures that curb competition can be par- ticularly costly for the incentives �rms face SOTEXKA (Société Textile de Kaolack) was cre- when the Kaolack factory operated briefly to innovate and perform ef�ciently, retard- ated around 1980. It was intended to be an at 20 percent capacity. It was shut down ing rather than helping the long-term internationally competitive textiles and cloth- after a few months because of technical dif- ing conglomerate with a spinning, weaving, �culties and the inability to pay for cotton development of industries.10 knitting, dyeing, and printing factory in Kao- and electricity. In 1990 it operated for just a Identifying speci�c industries that might lack and a garment factory in Louga.The ini- few months, but then shut down again. emerge as winners outside a country’s obvi- tial $25 million investment was �nanced by Despite a series of efforts to revitalize the ous areas of comparative advantage is government-guaranteed loans and 28 per- initiative, success remains elusive. cent direct government participation. becoming even more dif�cult. The falling The factories, completed in the mid- cost of information, the greater mobility of 1980s, did not begin operating until 1989, Source: Golub and Mbaye (2002). capital, the emergence of global supply net- works, and ongoing advances in technology (c) The International Bank for Reconstruction and Development / The World Bank Selective interventions 161 mean that patterns of industrial develop- Figure 8.1 Competition has increased with more countries exporting a larger range of goods ment and areas of competitive advantage are shifting faster than ever before.11 Competi- 160 Countries reporting tion among countries is also intensifying. exports When East Asian countries experimented with 120 Number of countries selective interventions to support their export- Countries that oriented industries, few other developing coun- export this good tries were doing the same. Today it is dif�cult to 80 �nd a government without the same ambitions, yet heightened competition reduces the prospects for success. Since 1962 the number of 40 countries exporting electrical equipment has tripled, and the number exporting motor vehi- 0 cle parts has more than doubled (�gure 8.1). 1962 1980 2000 1962 1980 2000 So strategies that may have worked in earlier Road vehicles and parts Electrical equipment periods offer few insights into what might work Source: World Integrated Trade Solutions database (available at wits.worldbank.org). today. At best, identifying speci�c industries is a gamble. Individual �rms make such gambles as Subsidized or directed credit can also obscure a matter of course, but they are betting with the cost to taxpayers and other borrowers. their shareholders’ money, and their sharehold- Schemes that create rents for �rms are also ers capture the rewards—and take the risk of notoriously dif�cult to dismantle—even when losing their stake. When governments enter the the costs clearly exceed the bene�ts. Firms ben- casino, they are betting with taxpayer resources, e�ting from special privileges have strong which should mean something for the size of incentives to resist their removal and often treat the bet and the length of the odds they are will- them as entitlements. Those who bear the bur- ing to accept. den of the distortion are typically more dis- persed and have weaker incentives to organize. Resisting rent-seeking Successful interventions need to resist the Getting value for money inevitable rent-seeking by �rms. Most �rms Selective interventions would be less haz- regard their contribution to economic devel- ardous if governments could be reasonably opment as special in some way, and can be sure they would get value for money. Some- willing to invest considerable resources in times the results of intervention do meet making their case to policymakers. Selective expectations. For example, successfully interventions that transfer costs and risks to attracting Intel to Costa Rica created con- consumers, taxpayers, or others are enticing. siderable spillovers to the economy (see box Forms of intervention that obscure the extent 7.2).13 Recent work in the United States sug- of the transfer are particularly attractive. gests that at least some cities that success- Import barriers and other market restric- fully attract major investments through tions have been especially popular. They offer incentive schemes may also get value for �rms monopoly pro�ts and reduce pressure to money when the bene�ts are construed perform ef�ciently. The costs to consumers broadly, including increases in the local tax (including �rms dependent on inputs from the base due to higher land prices.14 protected sector) through higher prices typi- Unfortunately, good outcomes cannot be cally far exceed the bene�ts gained by the pro- taken for granted. For example, when offering tected industry, but can be hard for consumers special incentives to attract investment, gov- to evaluate. Transferring commercial risks to ernments face a severe information disadvan- taxpayers—whether through government tage. They can never know the “right� level of guarantees of speci�c risks or broader pooling incentive to induce the desired behavior. They of risks through public–private joint ventures can easily fall prey to opportunistic behavior of various kinds—also weakens �rms’ incen- by �rms to provide incentives when none tives to perform ef�ciently. The risks borne by were necessary—or they can simply pay too taxpayers are rarely accounted for explicitly.12 much.15 Particularly in a competitive setting, (c) The International Bank for Reconstruction and Development / The World Bank 162 WORLD DEVELOPMENT REPORT 2005 pressures on politicians to overbid may con- • Relationship with the basics of a sound tribute to the “winner’s curse� that can afflict investment climate. Given the many fac- bidders in any auction.16 Governments can tors that shape the incentives for �rms to also fail to get value for money when the invest productively, the more a scheme incentive is paid up front or takes the form of tries to substitute for inadequacies in the the provision of speci�c infrastructure and basics of a sound investment climate, the �rm does not deliver as expected—as the rather than build on them, the longer the U.S. city of Indianapolis recently discovered odds of success. (see box 8.9). • Quality of governance. Selective interven- Nor are the costs limited to forgone tax tions do not necessarily require more revenues or speci�c public investments. expertise or resources than more basic Schemes that involve market restrictions measures—indeed, many demand less. transfer costs to consumers, and those But selective interventions are more vul- involving directed credit transfer costs to nerable to rent-seeking by �rms and of�- other borrowers. Selective interventions can cials, and the weaker the restraints on such also create distortions that ripple through behavior, the longer the odds of success. product and factor markets. Indeed, many distortions in countries’ investment cli- When positive outcomes cannot be mates today are the legacies of earlier efforts assured, the size of the bet matters. Schemes to intervene selectively. involving large budget outlays, transferring substantial risks to taxpayers, or creating Overall lessons of experience serious market distortions involve greater stakes than measures focusing on the dis- Before looking at particular strategies, it is semination of information—although even useful to spell out some general lessons. In those can be costly. Beyond such calcula- theory, selective interventions can yield tions, what else might governments do to positive social outcomes. In practice, cases reduce the risks inherent with selective of unambiguous success are rare, and there interventions? International experience sug- are many examples of costly failures, even in gests six basic guidelines for the design and developed countries with abundant techni- implementation of any such scheme: cal expertise and well-established checks on rent-seeking. Selective interventions that 1. Have a clear objective and rationale. target speci�c industries outside a country’s Unless a clear objective is stated, it will be obvious areas of comparative advantage are impossible to judge whether a scheme is most clearly a gamble. But a review of inter- meeting its intended goal at all, let alone cost national experience reveals no sure-�re effectively. Often multiple (and sometimes strategies even for less ambitious schemes, conflicting) goals are pursued simultane- suggesting that the analogy applies more ously.17 Schemes with vague objectives or broadly. The potential size of the reward is rationales can also mask the conferral of ben- obviously one factor governments need to e�ts on politically influential groups without consider. But what determines the odds of broader social bene�t. success in realizing those bene�ts in a cost- effective way? Three factors stand out: 2. Focus on the sources of problems, not the • Breadth of focus of the intervention. symptoms. Many obstacles facing �rms stem Given the dynamic nature of industrial from government failures in other areas— development, the narrower the focus of weak protection of property rights, red tape, the intervention in terms of speci�c corruption, dysfunctional infrastructure poli- �rms or industries, the longer the odds cies, or government crowding out credit mar- of success. Thus measures to encourage kets. Progress in addressing the underlying FDI or technological progress in gen- causes promises a broader and more sustain- eral involve fewer risks than those able impact than targeted measures that may focusing narrowly on a speci�c �rm or introduce new distortions or simply distract industry. attention from dealing with those causes. (c) The International Bank for Reconstruction and Development / The World Bank Selective interventions 163 3. Match the instrument to the rationale. design of the scheme—including the level Different rationales call for different instru- and form of the support provided and the ments. Financial market interventions will bene�ciaries of that support—facilitates reg- rarely be the most effective way to address ular public scrutiny of program effectiveness. potential spillovers. Tax incentives do not Well-de�ned objectives, instruments, and address credit market constraints. The provi- performance measures all play a role. Being sion of public infrastructure has no clear transparent is easier with explicit subsidies impact on incentives to innovate, and the and tax incentives. It is harder to evaluate conferral of market restrictions weakens market restrictions, directed credit, or other those incentives. schemes where the level of private bene�ts Where a selective intervention is intended and social costs are opaque and thus more to address poverty alleviation or other social vulnerable to capture and misuse. Most objectives, policymakers need to consider a Organisation for Economic Co-operation range of alternative instruments. For exam- and Development (OECD) countries and a ple, direct transfers to individuals or the pro- growing number of other countries publish vision of education or training are usually estimates of the cost in tax revenue of prefer- more effective at helping poor people than ential treatment.21 providing support to �rms that employ poor people, because in the latter case owners and 6. Review schemes regularly. Even schemes managers will capture many of the bene�ts. that meet the �rst �ve criteria may fail to deliver intended results, create unanticipated 4. Maintain discipline. One of the key failings distortions, or not keep up with changing of traditional import replacement strategies conditions. Yet the bene�ciaries of such was that �rms faced little discipline to schemes have strong incentives to resist improve their performance. Instead, �rms efforts to dismantle them. That makes it typically grew complacent, dependent on important to review schemes at regular inter- ongoing public support. Many forms of vals. Botswana and Taiwan, China, elimi- �nancial and other support to �rms have also nated schemes following reviews that raised not been conditional on performance, result- concerns over their effectiveness.22 Policy- ing in weak discipline even in repayment, let makers can ensure that schemes have sunset alone delivering the intended social bene�ts. clauses making continuation or extension Where feasible, special policy treatment beyond a speci�ed date conditional on a should be conditional on demonstrated per- transparent evaluation of costs and bene�ts.23 formance against objective criteria.18 South The time between reviews needs to be long Korea’s interventions to promote export- enough to give �rms some predictability— oriented �rms bene�ted from performance- but not too long (in all but the most capital- related discipline.19 That discipline can take intensive industries). many forms. Rather than providing up- front payments or tax holidays, support Experience in speci�c areas might be based on accelerated depreciation Beyond attempts to pick winners, govern- and so accrue to �rms only if they make the ments often use selective interventions to intended investment.20 Making any special hasten progress toward a subset of the goals treatment time-bound can also impose that a good investment climate can deliver. As more discipline on �rms. discussed in chapter 3, these include: 5. Be transparent. Transparency is the key to • Integrating informal and rural �rms disciplining both governments and �rms. • Unleashing the growth potential of Rent-seeking is behind many demands for smaller �rms special treatment, and bene�ts can easily be • Taking advantage of international tied to corruption. Schemes that give of�cials openness signi�cant discretion in selecting eligible • Climbing the technology ladder. �rms create uncertainty for �rms—and opportunities for abuse. Transparency in the What has been learned? (c) The International Bank for Reconstruction and Development / The World Bank 164 WORLD DEVELOPMENT REPORT 2005 Integrating �rms in the informal other local groups to draft a national policy economy to give street vendors legal status and The informal economy comprises a diverse address crime and licensing.24 Durban, set of �rms and so calls for a multidimen- South Africa, shows other ways for govern- sional approach (chapter 3). Strategies for ments to expand the opportunities for strengthening incentives to become formal important sectors in the informal economy were discussed in chapter 5. Here the focus (box 8.3). is on the possible role of selective interven- tions to improve the conditions facing Improving access to credit. Micro�nance microentrepeneurs in the informal econ- offers an important source of external omy. Those �rms bene�t from the basics of credit for informal �rms without collateral, a sound investment climate—from more and can help microentrepreneurs build secure property rights, better approaches to viable �rms (chapter 6). While most micro- tax and regulation, more ef�cient �nance �nance programs have been funded by gov- markets and infrastructure, and well-func- ernments and donors, efforts are now shift- tioning labor markets. Some governments ing to fostering commercial micro�nance do more. institutions—by removing regulatory impediments, supporting credit informa- Expanding voice and access. A �rst step in tion bureaus, and ensuring that noncom- dealing with the concerns of microentre- mercial entities do not undermine market preneurs in the informal economy is to give development (see box 6.3). them more voice in policy circles. Many are not recognized by the government and not Fostering links with formal �rms. Promot- seen as constituents, but there are examples ing links with formal �rms, often seen as a of their voices being heard. In Ahmedabad, key way to bring informal �rms into the India, the Self-Employed Women’s Associa- formal economy, is seldom successful. Even tion helped organize 550,000 women to so, initiatives that facilitate information provide cooperative �nancial, health, and sharing can be low cost and help match childcare services. It has also worked with suppliers and buyers. PROMICRO in Cen- the Ministry of Urban Development and tral America provides an example: Inter- national organizations, nongovernmental organizations (NGOs), and local associa- tions of microenterprises have joined to use BOX 8.3 Integrating informal traders in Durban the Internet to link �rms across �ve coun- With South Africa’s transition from Durban charges less than other cities for tries and disseminate information on sector- apartheid to democracy in 1994, the status the use of inner-city space. Flat rates are still speci�c events of interest, economic data, of small business development rose in charged for sites, but a new policy recom- and links to related sites.25 national economic policy thinking. Under mends charging formal and informal �rms apartheid, many informal activities were different rents and rates for different levels disallowed. For example,“move-on laws� of service. Decentralized registrations and Integrating �rms in the rural dictated that street vendors had to move pay points reduce transaction costs for economy their sites of trading every half hour. poorer traders. An integrated information Integrating rural �rms can overlap with The Amended Businesses Act allowed system is being developed to link incentives local authorities to formulate bylaws over a (such as access to subsidized training) to addressing informality, because many wider range of activities. With only one in registration. �rms in rural areas are informal. However, three economically active people The program bene�ted from consulta- rural locations bring added challenges. employed in the formal sector, Durban tions. Durban engaged in a year-long con- Some of the main impediments for rural responded by establishing a Department of sultative policy development process about Informal Trade and Small Business Oppor- priority issues, eliciting the views of formal �rms are inadequate infrastructure and tunities, which came up with innovative and informal business associations, politi- public services, and dif�culty in getting approaches to support informal enterprises cians, civil society, and community organiza- credit (chapter 3). and expand their link to the formal sector. tions. Informal traders are now represented Treating informal activities as contributors as stakeholders in pilot initiatives in area- to the local economy is apparent in the based management. Expanding infrastructure and public ser- structure of levies, the system of registra- vices. Expanding infrastructure and public tion, and the provision of services. Source: Lund and Skinner (2004). services in rural areas is an important part of any strategy for integrating the rural (c) The International Bank for Reconstruction and Development / The World Bank Selective interventions 165 economy, but subsidizing services for rural communities is dif�cult to sustain for BOX 8.4 Rural credit in Brazil resource-constrained governments (chapter The Brazilian rural �nance credit program Rarely did directed credit programs 6). Some governments are responding by illustrates some of the problems in directed reach their targeted recipients: the largest 2 removing obstacles to the entry of small credit programs. percent of borrowers receive more than 57 commercial providers, which play a big role Although many rules for directed lend- percent of the loans; the smallest 75 percent ing have been relaxed recently, it remains of borrowers receive a mere 6 percent. in providing electricity services in rural an important source of credit (about 38 per- Wealthy farmers seem to have captured the areas in countries such as Cambodia.26 cent of lending in Brazil in March 2002). subsidies, pushing up rural land prices as These programs, along with below market subsidies were capitalized into land values. Improving access to credit. Thinking on how interest rates, segment markets and distort The cost of funding these subsidies, borne by prices, raising the overall cost of capital. mandated lending rather than the Treasury, to improve access to credit in rural areas is Loan recovery remains low, and public sec- widened interest rate spreads and increased evolving (chapter 6). The early emphasis on tor banks, with poor loan portfolios and the cost of �nance for nonpriority sectors. providing subsidized or directed credit operating inef�ciencies, required recapital- through public agencies often had disap- ization in June 2001. Source: Klapper and Zaidi (2004). pointing results (box 8.4). Schemes proved unsustainable and failed to reach the major- ity of farmers.27 They also discouraged the with agricultural research, and unsustain- entry of private �nancial intermediaries.28 able �nance. New approaches try to address The programs generated an unintended these problems, contracting service delivery “grant� in the form of negative on-lending to private providers, decentralizing program interest rates, captured by wealthy and design and management, and making pro- influential groups rather than the poor. grams more demand-driven. But �nancial Loan repayment rates of subsidized credit sustainability remains a challenge.34 Fee-for- often dropped well below 50 percent, and service arrangements improve sustainability the costs of subsidies ballooned.29 but reduce demand from poorer farmers. The traditional approach was based on Decentralization can enhance accountabil- misconceptions about the rural credit mar- ity, but it also increases the risk of political ket: rural communities were seen as too interference. poor to save, so efforts concentrated on credit. Financial institutions were discour- Providing tax incentives. Many countries aged from mobilizing rural savings, which offer tax breaks, particularly to larger �rms might have been available for lending to that locate in rural areas. Beyond appeals entrepreneurs and households. Yet the lack for creating jobs and diversifying activities of savings institutions is cited as a signi�- in areas with higher poverty, there can be a cant constraint in rural surveys.30 justi�cation given the more limited avail- The new emphasis is on improving the ability of public services.35 But reducing investment climate for commercial taxes also reduces the resources govern- providers of �nance, including stronger ments have to improve those services. property rights and better regulation. Improving the environment for micro�- Unleashing the growth potential nance can also extend more credit to the of smaller �rms rural poor.31 Approaches are being devel- Governments often give special attention to oped to adapt micro�nance to the needs of the needs of small formal �rms. While rural areas for seasonal borrowing and non- many of the bolder claims about the contri- farming activities.32 bution small �rms make to growth are dif�- cult to substantiate (chapter 3), they do Supporting rural extension services. Exten- tend to face disproportionate burdens in a sion services can help to improve agricul- poor investment climate and have more dif- tural productivity and increase rural �culty getting credit than larger �rms. incomes, and some studies have found high Improving the basics of a sound invest- rates of return.33 Public provision of these ment climate will provide disproportionate services, however, has often been plagued bene�ts to smaller �rms. This includes with poor accountability, poor coordination improving the security of property rights, (c) The International Bank for Reconstruction and Development / The World Bank 166 WORLD DEVELOPMENT REPORT 2005 reducing red tape, improving the ef�ciency grams.40 However, experience highlights the of tax administration, curbing corruption, possible conflicts in trying to achieve out- improving the functioning of �nance mar- reach and sustainability simultaneously,41 kets, and strengthening infrastructure. Some and the cost-effectiveness of the newer governments go beyond this by providing approaches has not yet been evaluated.42 special bene�ts to smaller �rms. Fostering industry clusters. Agglomeration Improving access to credit. The disadvan- economies associated with proximity to other tages smaller �rms face in getting credit �rms can stimulate productivity upgrading stem from information asymmetries, are and growth.43 Efforts to stimulate those exacerbated by weak property rights, and economies through industry clusters gained are further compounded when govern- momentum in the 1990s as a way of helping ments create other distortions in �nancial small �rms grow and upgrade through markets (chapter 6). Instead of addressing sharing complementarities.44 A recent study these problems, many governments come identi�ed more than 500 such initiatives, up with special schemes to provide directed mainly in developed and transition or subsidized credit to small �rms. These economies.45 But governments have dif�- schemes have a poor track record in devel- culty identifying sectors where clusters will oping countries. Loans tend to go to politi- succeed,46 and the heterogeneity of clusters cally connected �rms. Weak repayment dis- makes it dif�cult to come up with recipes for cipline jeopardizes sustainability. And successful intervention.47 In clusters of low- subsidized credit crowds out potential productivity �rms there is also a tradeoff providers of credit on a commercial basis.36 between strengthening individual �rms and Nor do subsidized loans help most �rms reinforcing their synergies, and opportunis- grow faster.37 A survey of small �rms in tic behavior by �rms can undermine collec- South Korea found that subsidized credit tive services.48 was no more valuable than commercial Experience shows that cluster initiatives credit, mainly because of narrow eligibility need to be private sector–driven and that criteria and delays in obtaining the funds.38 public support cannot substitute for lack of Efforts to expand access to �nance will also private commitment. A review of U.S. Agency have little impact when other investment for International Development (USAID) climate concerns reduce the incentives for experience in cluster development in 26 �rms to reinvest their own resources.39 countries concluded that large amounts of public funding weakened local ownership of Providing business development services. projects.49 The success of cluster initiatives Small �rms are often assumed to face spe- depends on �rms being able to work together cial dif�culty in obtaining access to business for their common interests. Overcoming ani- development services—training, consult- mosities among �rms can be challenging, as a ing, marketing, technology transfer, and donor-driven initiative in the Mongolian business links—tailored to their needs. Tra- cashmere sector discovered. In that case, ditionally, governments or donors created however, the realization of bene�ts from new public institutions, or arranged for NGOs markets built further trust in the process and to deliver these services to �rms for free or led to the sector’s expansion.50 at highly subsidized rates. The efforts were generally found to be ineffective, with low Providing market privileges. Some coun- take-up rates, cost overruns, and dif�culties tries erect regulatory barriers to shield in tailoring services to the needs of clients. smaller �rms from too much competition These efforts also deterred the emergence of from larger �rms. But regulatory barriers commercial providers of these services. also discourage �rms from growing. Con- More market-friendly approaches are now sider the reservation of market segments for being explored that aim to increase out- small �rms in India. In addition to limiting reach to currently underserved sectors with participation by larger and more ef�cient self-sustained and cost-effective pro- �rms—to the detriment of consumers—the (c) The International Bank for Reconstruction and Development / The World Bank Selective interventions 167 scheme kept �rms small, stunting overall productivity growth (box 8.5). BOX 8.5 Staying small in India—by design Since 1967 the manufacture of speci�ed lower capacity than those producing unre- Taking advantage of international product lines in India has been reserved for served items, are technologically less openness small �rms (with investments in plant and dynamic, and perform less well in productiv- FDI and exporting both have the potential machinery of up to about $200,000).The list ity and even in pro�tability. of reserved product lines has grown from As much as it intends to protect small to provide spillovers to the local economy 47 when the scheme was introduced to �rms, the reservation policy is self-defeat- (chapter 3). To capture these bene�ts, many some 675 items in 2004. Once a product line ing. Many reserved products are either governments pursue selective interventions is reserved, no new medium or large �rm is freely importable or local levels of produc- allowed, and those already producing the tion are low. A review in 1997 found that to attract FDI, promote exports, or both. product are restricted to the highest annual more than 550 items on the list of reserved level achieved in the three years preceding products could be freely imported, and as Enclaves and export processing zones. One the date of reservation. many as 90 were manufactured by just one way to begin improving the investment cli- Reservation tends to motivate many �rm. Sixty-eight items accounted for 81 per- mate in dif�cult environments is to create small �rms to “stay small.� If they do increase cent of the total value of production of operations, they do so by establishing more reserved products and 83 percent of the enclaves that provide participating �rms small units.The policy, encouraging stagna- �rms.The review recommended abolishing with better security and infrastructure and tion and incurring high costs for producers the reservation system. By the end of 2003, a less burdensome tax and regulatory envi- and consumers, has hampered growth in 165 items had been taken off the list. ronment. Enclaves allow governments to light engineering and food processing, as Source: Morris and others (2001); Hussain (1997); well as in textile and leather exports. Survey Gupta (1999); India–Ministry of Small Scale focus efforts on a speci�c geographic loca- results and empirical tests show that �rms Industries (2003); Harsh (2003); Katrak (1999); tion. They can also be used to test new pol- manufacturing reserved products operate at World Bank (2003c); and Deccan Herald (2003). icy approaches—as China did with its Spe- cial Economic Zones after 1980 (box 8.6). Export processing zones (EPZs) are a common example of enclave approaches. BOX 8.6 China’s special economic zones By the end of 2002 some 3,000 EPZs had been created in 116 countries, providing In 1980 China designated four Special Eco- of light manufacturing, real estate, and later jobs for some 43 million workers—most of nomic Zones: three in Guandong province �nancial services. In Shenzhen exports grew (Shenzhen, Zhuhai, and Shantou), and one in at an average of 75 percent. While most them women (table 8.1).51 the Fujian province (Xiamen), adjacent to inputs were imported initially, local content Despite their popularity, not all EPZs suc- Hong Kong, China, and Taiwan, China, respec- grew in the early 1990s, showing further ceed. Countries with poor protection of tively.The zones offered special incentives to integration of the zones into the domestic property rights, weak governance, or poor foreign investors, including tax breaks and economy. duty exemptions for exporters and flexible The zones soon expanded to other infrastructure can fail to attract investors to labor regulations. Infrastructure and the legal areas. In 1984 14 coastal cities and Hainan their EPZs.52 Even in successful cases closer framework for FDI were also improved. Island opened to foreign investment. In the analysis suggests the EPZ was often comple- Domestic �rms were encouraged to establish late 1980s more coastal areas opened to mented by other favorable factors (box 8.7).53 links with foreign investors. In fact, a thriving create a coastal belt, including the Yangtze domestic private sector developed in the River Delta, the Pearl River Delta, and other The bene�ts from enclave approaches zones, favored by learning from FDI and by areas in Fujian, Shandong, Liadong, Hebei, are inherently limited when they con�ne the better investment climate. and Guangxi provinces. In 1990 the Pudong investment climate improvements to one Two factors contributed to the success New Area was created in Shanghai along of the �rst zones. One was the proximity to with other cities in the Yangtze River valley. area—or confer special privileges that can- fast-growing Hong Kong, China, and Taiwan, Since 1992 border areas and the capital not be easily generalized to the broader China, whose investors were attracted by cities of all inland provinces have been economy. This is likely to be especially the low cost of land and labor in the zones. opened to foreign investment, as the Chi- problematic in small economies without a The other was the agreement between cen- nese authorities try to balance the previous tral and provincial authorities to share �scal concentration of foreign investment in developed industrial base. Without a broad revenue, an incentive to develop infrastruc- coastal areas.The eastern provinces along base of local suppliers, enclaves are less ture in the zones. the coast still account for 85 percent of the likely to develop linkages and channels for FDI in the zones shot up from $23.4 mil- accumulated stock of FDI. Fiscal incentives, spillovers to local �rms or to create con- lion in 1980 to $672 million in 1993 in the such as tax holidays, vary across zones— Shenzhen zone alone.The average annual and are generally more generous in export- stituencies for broader trade liberalization. growth rate exceeded 35 percent in oriented and high-tech sectors. They are most likely to generate bene�ts the 1980–95, three times China’s average.The more they are integrated into a broader growth was mainly driven by the expansion Source: OECD (2003b);Chen (2002);and Ge (1999). strategy to test and demonstrate the bene- �ts of reforms and to progressively improve the investment climate for the broader economy, as in China. (c) The International Bank for Reconstruction and Development / The World Bank 168 WORLD DEVELOPMENT REPORT 2005 Table 8.1 Export processing zones have proliferated into the thousands vide advice on trade logistics, and, in some 1975 1986 1995 1997 2002 cases, administer export incentives. With a Countries with EPZs 25 47 73 93 116 few exceptions (Australia, Finland, Ireland, EPZs 79 176 500 845 3,000 New Zealand, and Singapore), the results Employment (millions) .. .. .. 22.5 43 appear to have been modest. One clear les- China ... .. .. 18 30 Other countries with available �gures 0.8 1.9 .. 4.5 13 son is that export promotion activities can- not substitute for progress on more funda- Note: .. = not available. Source: ILO (2003a). mental obstacles to successful exporting, including a poor climate for �rms to develop world-class products and weak Promoting exports. To encourage exporting, transport infrastructure.55 governments often provide duty exemption and drawback systems, provide export Providing incentives to attract FDI. In the credit, and support trade promotion activi- mid-1990s more than 100 countries offered ties. Because bene�ts granted on the condi- �scal incentives to attract FDI, a trend that tion of meeting export targets can distort continues.56 A recent survey of 45 develop- international trade flows, they are being ing countries found that 85 percent offered phased out under World Trade Organiza- some kind of tax holiday or reduction of tion (WTO) rules (box 8.8). Duty drawback corporate income tax for foreign invest- systems and export subsidies helped expand ment.57 The incentives can be substantial East Asian exports, but many countries have (table 8.2). In Tunisia incentives for FDI embarked on similar strategies with little amounted to almost 20 percent of total pri- success.54 These programs often require vate investment.58 In Vietnam it was esti- burdensome procedures and paperwork mated that the revenue loss from incentives that increase costs and create opportunities reached 0.7 percent of GDP.59 The package for corruption. The problems can be espe- India offered Ford in 1997 was estimated to cially severe in countries with weak tax and cost $420,000 per job.60 Incentive packages customs administrations. often include tax incentives, special regula- Information asymmetries in interna- tory exemptions, subsidies, and public tional markets are sometimes used to justify funding of related infrastructure. government support for trade promotion Do these incentives actually influence activities. Many countries have created the decisions of �rms? The answer seems to trade promotion organizations to conduct be sometimes. Firms tend to assess invest- market research, organize trade fairs, pro- ment opportunities, including relevant gov- BOX 8.7 Export processing zones in Mauritius and the Dominican Republic Despite their popularity, not all EPZs meet What accounted for the impressive perfor- Linkages Program, developed in the late 1980s expectations. Experience in Mauritius and the mance? Certainly, the EPZs played a role. But and early 1990s, had the goal of developing Dominican Republic throw light on two several complementary factors also seem to backward linkages to 40 local manufacturers common issues. have been important. Mauritius enjoyed fairly and $80 million of local value-added. stable macroeconomic conditions and high lev- Progress has been disappointing. By 1993 Mauritius—More than just EPZs els of political stability, contributing to the secu- only 12 local suppliers participated in backward Mauritius used EPZs as part of a successful strat- rity of property rights. It also enjoyed preferen- linkages, with local value-added of just $4 million. egy to spur export-led growth and diversify its tial access to the apparel markets in the EU and Local value-added has remained low. In 2002, economy. EPZ status was granted to �rms inde- U.S. And the diversity of its population, with Chi- only 55 of 720 EPZ �rms purchased raw materials pendent of location. Manufactured exports nese and French minorities and an Indian major- from local �rms, a decline from 61 the previous grew at 5.9 percent a year between 1991 and ity, helped attract investments from Hong Kong year.Why? Local manufacturers, isolated from 2001, and accounted for 73 percent of merchan- and mediate investments in India. competitive pressures by import substitution dise exports in 2002. Employment in the EPZs policies, showed no interest in assuming new ranges between 80,000 and 90,000. Many work- Dominican Republic—The elusive quest risks to meet the standards of the EPZs. ers and managers trained in the foreign sector for backward linkages later created their own businesses. Economic Like many countries, the Dominican Republic Source: For Mauritius, Subramanian and Roy (2003); Moran (2002); Rodrik (1999); and World Bank growth in 1980–2002 averaged 5.5 percent, hoped to build backward linkages from its EPZs (2004k). For Dominican Republic, Schrank (2001) accompanied by substantial improvements in to its local industries, so that local �rms would and Consejo Nacional de Zonas Francas de human development indicators. become exporters themselves.The Industrial Exportación (2002). (c) The International Bank for Reconstruction and Development / The World Bank Selective interventions 169 BOX 8.8 The WTO and selective intervention Selective interventions to promote �rms or may be challenged at the Dispute Settlement Services. Under the General Agreement on activities may distort international trade and Body by other WTO members if they hurt their Trade in Services (GATS), countries commit ser- harm other countries.To address these concerns interests. vices to national treatment and market access international agreements impose restrictions on Trade-related investment measures. The according to their own schedule, leaving room trade-distorting policies. Restrictions on export Agreement on Trade-Related Investment Mea- to accommodate their policy goals. subsidies date from 1947 in Article 16 of the sures (TRIMs) imposes limits on measures aimed The Doha Round of multilateral trade negoti- General Agreement on Tariffs and Trade (GATT). at extracting bene�ts from FDI.The agreement ations, launched in 2001, includes proposals to The Uruguay Round of multilateral trade negoti- includes a list of measures inconsistent with the negotiate a tightening of disciplines in the use of ations, which led to the creation of the WTO in principles of national treatment and the GATT agriculture subsidies and antidumping measures. 1995, set new limits on what governments can prohibition of quantitative restrictions, includ- The above arrangements include special do to support domestic industries, promote ing local content and trade-balancing require- and differential treatment for developing coun- exports, or affect the consequences of foreign ments. tries. For example, the prohibition of export sub- investment: Intellectual property rights. The Agreement on sidies is waived for countries with a GDP per Subsidies. The Agreement on Subsidies and Trade-Related Aspects of Intellectual Property capita below $1,000. Countervailing Measures prohibits subsidies Rights (TRIPs) strengthens the rules and enforce- contingent on meeting certain export targets or ment of intellectual property rights. Practices Source: World Bank (2004d); Hoekman, Mattoo, and on using domestic rather than imported goods. such as compulsory licensing and reverse engi- English (2002); Hoekman, Michalopoulos, and Winters Other subsidies to speci�c �rms or industries neering are limited by the agreement. (2003); and GATT. ernment policies, as a package. The level of Table 8.2 Effective reductions in corporate tax rates due to �scal incentives (percent) tax and other obligations can influence that Philippines Malaysia Thailand package but rarely will be enough to cancel Effective tax rate (before incentives) 47 30 46 out other factors, including more funda- Reduction in effective rate due to: mental concerns about policy stability, the Tax holiday 19 0 28 quality of infrastructure, and the quality of Indirect tax concessions 7 8 11 a workforce. Indeed, the Bank’s Investment Effective tax rate (after incentives) 21 22 7 Climate Surveys show that unreliable power supply, weak contract enforcement, corrup- Source: Chalk (2001). tion, and crime can impose costs several times greater than taxes (chapter 1). The weight applied to any one factor Do governments get value for money varies between industries and even between when they offer special incentives? The �rms in a single industry. Incentives will costs and bene�ts need to be assessed in typically carry less weight when �rms are in each case. If the �rm would have made the extractive industries or intend to serve the same investment without the incentive, or local market. In such cases �rms will usually with a lower level of incentive, the answer have identi�ed the market for other reasons would be no.64 Certainly the cost per job and cannot pursue the same opportunity created can be high, as the examples illus- elsewhere. Investments in manufacturing, trate (�gure 8.2). However, governments especially in export-oriented sectors, might are rarely interested only in the jobs associ- be more responsive to tax incentives.61 But ated with the immediate investment; they tax holidays are only rarely the decisive fac- usually expect broader bene�ts in spillovers tor. A survey of 191 companies with plans to local �rms. Governments often also hope to expand operations found that only 18 that winning a major investment will signal percent in manufacturing and 9 percent in to the broader universe of investors that services considered grants and incentives to their country is a good place to do business. be influential in their choice of location.62 But experience suggests that these bene�ts Of 75 Fortune 500 companies surveyed, cannot be taken for granted. only four identi�ed them as influential.63 The design of the incentive package can When alternative locations are otherwise also influence the net return to the closely matched, however, differences in tax country.65 Incentive schemes that involve obligations can influence decisions at the up-front subsidies or the provision of margin. highly speci�c infrastructure are generally (c) The International Bank for Reconstruction and Development / The World Bank 170 WORLD DEVELOPMENT REPORT 2005 Figure 8.2 Incentives can be costly investment opportunities, bene�ts occur- ring in the future are of declining influence, 200 Total budgetary cost (right axis) and �rms tend to apply bigger discount rates to projects in countries they perceive to be Cost per job (left riskier. Often more important than the level Thousands of dollars per job created 2,000 150 axis) of tax rates is their predictability. Firms may prefer to pay a �xed rate for a de�nite period Percent of GDP per capita than pay no taxes now and an uncertain amount in the future—Chile and Colombia 100 offer this option to foreign investors.66 1,000 A better strategy is to improve the qual- ity of the overall investment climate, thus reducing the pressure to compete on taxes. 50 Tackling bottlenecks of particular concern to foreign investors (customs administra- tion, property rights security) will likely do 0 0 more to make a location attractive—and U.S. Germany Brazil will bene�t local �rms, too. The same prin- 7 plants VW 3 plants ciples apply not only to efforts to attract (1980s) (1997) (1995–6) Note: Percent of GDP per capita is PPP adjusted. foreign investment, but also to subnational Source: McKinsey Global Institute (2003). governments that compete for investment within a country (box 8.10). riskier than tax incentives, because if the �rm fails to deliver, the infrastructure may be of Promoting inward investment. Govern- less value to other �rms (box 8.9). Tax incen- ments also try to attract FDI through tives have the advantages of being reasonably investment promotion agencies (IPAs). transparent and conditional on income There are now at least 160 national and earned—if the investment does not proceed more than 250 subnational IPAs, compared or the �rm chooses to relocate the govern- with only a handful two decades ago.67 ment’s exposure will be limited. Providing These agencies play a variety of roles tax incentives based on accelerated deprecia- including the following:68 tion can strengthen the link between the • Information dissemination. Collation incentive and actual investment. and presentation of information on the It may not be necessary to offer tax holi- local economy. days of long duration. Because of the dis- • Image building. Promoting the percep- count rates �rms apply when evaluating tion that the country is an attractive loca- tion for investment through activities such as advertising and public relations. BOX 8.9 Rolling the dice in Indianapolis • Investment facilitation. Helping investors Governments often offer subsidy packages recession hit the industry and felt pressure through administrative procedures and to �rms that promise to create jobs and to cut costs. clearances needed to set up and operate bring new technology. Experience in the The result: high sunk costs for state and business establishments. In some cases U.S. city of Indianapolis shows that the local governments in highly speci�c expected bene�ts can remain elusive. infrastructure, resources that could have been IPAs serve as one-stop shops (chapter 5). Local and state governments granted used for other priorities. In all likelihood, new • Investment generation. Identifying and up-front subsidies worth over $300 million tenants for the facilities would come only if directly targeting �rms in sectors that to build an advanced aircraft maintenance new subsidies are offered. More than 80 �rms center for United Airlines. The deal was had been contacted to take over the mainte- might be attractive for foreign invest- negotiated during an economic slowdown nance center in the 18 months following its ment through direct mailings, telephone in the early 1990s, and the authorities con- closure.Yet the facility’s size and technologi- campaigns, and presentations to individ- sidered the subsidy was worth the promise cal sophistication imply high operating costs, ual investors. to create 5,000 high-paying jobs. That a hard sell in a distressed industry. number was never achieved, however, and • Investor monitoring and aftercare. Assist- the company walked away in 2003 after Source: O’Malley (2004) and Uchitelle (2003). ing �rms already established to continue and expand their operations. This is (c) The International Bank for Reconstruction and Development / The World Bank Selective interventions 171 BOX 8.10 Competing to attract investment within countries Without speci�c efforts to influence location mate is essential for success, including the secu- 1990s some Brazilian states also joined the com- choices, �rms tend to prefer to locate in areas rity of property rights, adequacy of infrastruc- petition for automobile plants, offering incen- with stronger investment climates and to con- ture, a skilled labor force, and the like. tive packages in the range of $54,000 to centrate to take advantage of product or factor Subnational governments also often extend $340,000 per job. markets. Agglomeration economies help explain special incentive schemes. At least 20 U.S. states Most of the issues associated with attracting the concentration of industrial activity in most were interested in the Mercedes-Benz plant that investment at the national level apply to subna- countries, with the effects reinforced by and �nally located in Vance, Alabama, with a $153 tional governments as well.This includes the dif- reinforcing the urbanization around the world. million incentive package in 1993. More than �culty in assessing whether any incentives To help spur agglomeration economies, build 250 European locations competed for a BMW offered are necessary or cost-effective. Similar their industrial base, or create jobs, many subna- plant that went to Leipzig with $224 million in design issues can arise as well. tional governments or cities compete for invest- incentives in 2001. A recent study found that ment in much the same way as their national revenue forgone by state and local Source: Yusuf (2003); Scott and Storper (2003); Charl- counterparts. As with competition for interna- governments in the United States due to �scal ton (2003); Christiansen, Oman, and Charlton (2003); tional investment, the broader investment cli- incentives was up to $50 billion. In the mid- and Peters and Fisher (2004). emerging as an important function in tions and local content or joint venture Figure 8.3 Policy advocacy by investment promotion agencies second-generation reforms. requirements to promote the likelihood of receives a small share of budget • Policy advocacy. Identifying issues that FDI spillovers. Dif�culties with those approaches have led more recent efforts to 40 inhibit investment and advocating pol- icy changes that might stimulate devel- focus on incentives to encourage the desired opment. IPAs often act as champions of behavior from foreign investors.71 Investment generation Investor services reform in lobbying other government Local content requirements have been 30 Percent of total budget agencies to correct observed problems. used to ensure that foreign investors use Image building This function, potentially the most inputs from local �rms. Because the evi- effective in attracting FDI, usually repre- dence suggests that local �rms bene�t from 20 sents only a small part of the budget supplying foreign-owned �rms (see chapter (�gure 8.3).69 5), this might seem to be a way of increasing the bene�ts from FDI. Unfortunately, such There is some evidence that IPAs can 10 restrictions also increase the costs of FDI, Policy advocacy help countries attract FDI. One study found reducing the foreign investors’ incentives to that FDI increases by about 0.25 percent for enter and expand production (box 8.11). every 1 percent increase in the IPA’s budget. Local content requirements in the automo- 0 IPAs appear to be more successful in coun- bile sectors in Chile and Australia also Source: Morisset and Andrews-Johnson (2003). tries where the investment climate is already resulted in large inef�ciencies.72 Local con- amenable to foreign investors: increases in tent requirements are also inconsistent with the budget of an IPA increased FDI nearly international trade rules and so are being twice as much in countries with the most phased out (see box 8.9). favorable investment climates than in coun- Another approach has been to require tries with the least favorable.70 Success sto- foreign investors to participate in joint ries in investment promotion have been costly in per capita terms, however, espe- Table 8.3 IPAs are not cheap cially at the image building stage (table 8.3). Annual FDI promotion budget Population Per capita budget Fostering spillovers from FDI. Beyond ($ million) (millions 1999) ($) attracting investment, governments often Singapore (EDB) 45.0 3.2 14.06 make special efforts to increase the likeli- Ireland (IDA 1999) 41.0 3.7 11.16 hood of positive spillovers to the broader Costa Rica (CINDE) 11.0 3.5 3.14 economy. Governments often look to FDI Mauritius (MEDIA 1996) 3.1 1.2 2.58 to help develop local industry and promote Dominican Republic (IPC) 8.8 8.4 1.05 technology transfer, but local suppliers and partners may not develop automatically. In Malaysia (MIDA) 15.0 22.7 0.66 the past governments used import restric- Source: Velde (2001). (c) The International Bank for Reconstruction and Development / The World Bank 172 WORLD DEVELOPMENT REPORT 2005 Because foreign investors in the automo- BOX 8.11 Fixing the FDI strategy for Mexico’s computer bile sector in China were required to have a industry local partner, major international �rms were reluctant to use up-to-date processes. As a In 1985 computer production in Mexico was between 100,000 and 180,000 computers a result, manufacturing methods lagged protected by import quotas. Local content year. requirements were set at 25 percent for The proposal triggered strong opposi- behind industry standards by about 10 minicomputers and 35 percent for micro- tion from domestic suppliers.Their years.73 Similarly, Kodak was required to computers for the �rst year, rising to 50 per- argument was that the large investment have local joint venture partners in its cent and 60 percent in the third and fourth would create a monopoly, crowding out years. Foreign ownership was allowed as a domestic players—but the prediction was investments in China but allowed to have minority share in joint ventures with local not ful�lled when the IBM proposal was one wholly owned subsidiary. It invested six �rms.The market was dominated by joint accepted. Indeed, competition increased as times more in the wholly owned �rm than it ventures involving two U.S. �rms, Apple (58 other foreign �rms, including Apple and HP, did in the average joint venture partner. Its percent) and Hewlett-Packard (18 percent). also invested in wholly owned large facili- High protection meant computer prices ties.The share of imports in the �nal prod- wholly owned subsidiary ended up produc- in Mexico were 74 percent higher for Apple uct decreased and the component industry ing its most advanced �lm and camera tech- and 61 percent higher for HP models than gained technological upgrading. With these nologies, while the joint ventures produced in the United States. Both �rms were assem- investments, computer exports surged from conventional �lm under the Kodak label.74 bling computers at volumes well below the $21 million in 1985 to $252 million in 1989, ef�cient scale of 20,000 units annually.The and $9.6 billion in 2001. Another strategy is to work with foreign perverse incentives of this policy surfaced af�liates and local �rms to overcome infor- when IBM presented the Mexican govern- mation and cultural barriers. These pro- ment with a proposal to invest in a wholly Source: Moran (1998) and OECD International grams are often combined with incentives owned export-oriented facility to produce Trade by Commodity Statistics Database. to help the domestic suppliers meet the production standards demanded by foreign ventures with local partners. In some cases investors. This approach has been followed these requirements have been used to bene- in economies such as Ireland, Malaysia, Sin- �t speci�c local �rms by allowing them to gapore, and Taiwan, China (box 8.12).75 participate in a lucrative foreign invest- ment, but they are also intended to increase Climbing the technology ladder technological spillovers. As with other Technological progress plays an important mandatory measures, however, they have role in economic growth, leading many costs. They may deter rather than encour- governments to encourage innovation age investment, and they can make foreign (chapter 3). But innovation is not limited to �rms wary about using advanced or sensi- activities that might merit a patent. It tive processes, reducing rather than enhanc- includes more modest advances and the ing spillovers. implementation of better business processes. BOX 8.12 Successful “linkage programs� in Singapore and Ireland Singapore and Ireland illustrate the potential ing 670 local businesses with 30 foreign ing increased from 9 to 19 percent during that impact of well-designed programs to foster af�liates and 11 large local businesses and gov- period. More than 200 foreign �rms and 83 spillovers from FDI. ernment agencies in 1999. domestic �rms participated. Suppliers saw sales rise by 83 percent, productivity by 36 percent, Singapore’s Local Industry Upgrading Ireland’s National Linkage Program and employment by 33 percent—and some Program Ireland’s Industrial Development Agency (IDA) became international subcontractors.The pur- To promote technology and skill transfers from led a consortium of agencies that identi�ed chase of Irish materials and services by foreign foreign �rms to local suppliers, Singapore’s Eco- potential linkages in a range of sectors, devel- af�liates supported by IDA in 2001 reached nomic Development Board (EDB) offered orga- oped a group of domestic suppliers, and offered €5.49 billion and €5.12 billion respectively. nizational and �nancial support. An engineer or buyer support and development services.The The programs in Singapore and Ireland share manager from the foreign �rm was paid by the program targeted “winner� companies in two characteristics. First, they are market-based, EDB for two to three years to select and assist selected sectors and worked with them to creating fewer distortions than imposed local local suppliers.Thirty-two partnerships were enter subcontracting arrangements with multi- content requirements. Second, they combine created between 1986 and 1994 involving 180 national �rms. Between 1985 and 1992, foreign policy advocacy, proximity to suppliers, and spe- domestic suppliers.The electronics industry was af�liates increased their local purchases of raw ci�c linkage opportunities.Their goal is to reduce the biggest sector, followed by services. Produc- materials by half (from 438 to 811 million Irish the risks perceived by suppliers and buyers. tivity of suppliers in the early stages rose by an pounds) and their purchases of services by one average 17 percent, and value added per worker third (from 980 million to 1.46 billion Irish Source: Battat, Frank, and Shen (1996); UNCTAD increased by 14 percent.The program was link- pounds). In the electronics industry, local sourc- (2001b); and Ireland–IDA (2002). (c) The International Bank for Reconstruction and Development / The World Bank Selective interventions 173 It also involves lots of adaptation and adop- Table 8.4 Fiscal incentives for R&D in selected developing countries tion—countries don’t need to invent every- R&D depreciation R&D capital depreciation Tax credit Country rate rate rate thing afresh. This underlines the impor- tance of reducing barriers to trade and Brazil 100% 100% None FDI—and to the competition that provides India 100% 100% None incentives for �rms to improve their pro- South Korea 100% 18–20% 10–25% ductivity. Mexico 100% 3 years’ straight-line None As countries move closer to the techno- depreciation logical frontier, governments often seek to South Africa 100% 25% None encourage original innovation in their Taiwan, China 100% Same as other investment 15–20% economies, including local R&D. To do so, Malaysia 200% Same as other investment None governments have experimented with a Note: Depreciation methods of 100 percent or more indicate full expensing of R&D. range of selective interventions. The cost- Source: Mani (2001a) and de Ferranti and others (2003). effectiveness of these schemes has not been evaluated in all cases, but their impact is countries have also used �nancial market likely to depend on the adequacy of other interventions to encourage �rms to pursue aspects of the investment climate critical to R&D, including directed credit schemes innovation, including a skilled labor force, (South Korea) and venture capital funds competitive pressure, and the protection of (Malaysia).80 intellectual property rights. Without those elements, it is not clear that government Other strategies for supporting local R&D. To interventions can do much to increase R&D. support innovation, the public sector can undertake R&D activities directly—on its Providing tax incentives, grants, and �nan- own or with private partners. The experience cial market interventions. Many govern- is mixed, however (box 8.13). The govern- ments provide tax deductions to encourage ment is seldom in a good position to judge private R&D. Some developed countries the types of research that would help �rms or offer tax credits, full expensing of R&D, and have market potential. There is also a debate even double deductions of some R&D about whether public R&D would crowd out spending (table 8.4). Although these or complement private efforts. A review of schemes are not too costly, they have their the econometric evidence �nds mixed results, weaknesses. Firms may claim R&D deduc- but concludes overall that well-designed tions for spending barely linked to any real efforts can be complementary.81 R&D. Firms also tend to choose projects Fostering high-tech industrial clusters has with the highest rates of private return, not also met with mixed results. Following the those with the largest spillover effects.76 In the United States almost 80 percent of tax returns claiming R&D credits are audited, Figure 8.4 Grants make up the lion’s share of public funding for private R&D in many developed countries with an average downward adjustment of 20 percent of the claimed credits.77 While Australia (1997) Tax credits Grants some studies of Pakistan and Canada found evidence that R&D incentives were cost- Canada (1995) effective, others are more skeptical.78 The use of R&D tax incentives, grants, or France (1997) a combination of both varies from country to country (�gure 8.4). Grants are preferred Japan (1997) by governments that want to influence the type of R&D, but this raises more dif�cul- Netherlands (1997) ties in governments “picking winners� than United States (1999) broadly based tax incentives. Interestingly, Sweden and Finland, two countries with 0 5 10 15 20 25 high levels of private R&D, do not offer Share of total industry R&D (percent) substantial direct or tax support.79 Some Source: OECD (2003f). (c) The International Bank for Reconstruction and Development / The World Bank 174 WORLD DEVELOPMENT REPORT 2005 high-tech clusters—both remain as isolated BOX 8.13 Public-private partnerships for R&D research centers.84 Recent work on national innovation sys- Many governments have established R&D with international technological advances. centers to promote the technological In India the network of publicly funded tems emphasizes the importance of collab- upgrading of �rms.The support of the research organizations under the Council of oration between industry and universities. Industrial Technology Research Institute in Scienti�c and Industrial Research had little Governments can foster links between uni- Taiwan, China, helped spin off the �rst inte- contact with industry. Latin America has its versities and �rms by strengthening prop- grated circuit manufacturer. However, own cautionary tales. Competing agendas attempts to create partnerships between between different government agencies in erty rights for universities and encouraging R&D centers and private �rms do not always Brazil and Argentina made public–private private contracts.85 meet expectations. partnerships in R&D ineffective. In the Philippines the Department of Science and Technology had little interac- tion with industry. Its staff did not have very Source: UNCTAD (2003c); de Ferranti and others So the possibility exists for governments to high quali�cations and were not in touch (2003); and Mani (2001b). intervene selectively in ways that contribute to growth and poverty reduction. Experi- ence shows, however, that such strategies are not straightforward, and that the likelihood success of the Hsinchu Science Park in Tai- of success is greater when they complement wan, China, and the Magnet Program in rather than attempt to substitute for broader Israel, some governments created science investment climate improvements. Schemes parks and business incubators.82 But innov- that meet the guidelines suggested at the ative clusters require a dynamic interplay of beginning of this chapter reduce the risk of entrepreneurship, R&D institutions, skilled selective interventions going astray. labor, capital, and infrastructure. Without Another strategy governments can adopt these factors, government-led initiatives are to complement the basics of a sound invest- unlikely to succeed.83 For example, top- ment climate is to draw on the growing quality infrastructure in such science cen- body of international rules and standards in ters as Tsukuba Science City (Japan) and this area. The strengths and weaknesses of Daeduck (South Korea) failed to turn into such strategies are discussed in chapter 9. (c) The International Bank for Reconstruction and Development / The World Bank International rules and standards 9 The approaches to delivering the basics of a The number of international arrangements sound investment climate discussed in Part dealing with investment climate issues has II of the Report rest mainly on domestic grown dramatically in recent decades. There chapter laws, policies, and institutions. In recent are now more than 2,200 bilateral investment decades the volume and range of interna- treaties, 200 regional cooperation arrange- tional rules and standards dealing with ments, and some 500 multilateral conventions investment climate issues has shot up dra- and instruments. These arrangements cover matically. Can these arrangements help most areas of the investment climate—from governments improve the investment cli- property rights protection, taxation, and mates of their societies? corruption, to regulation in areas as diverse Arrangements that reduce regulatory bar- as banking, shipping, telecommunications, riers to international trade and investment labor, and the environment. can improve investment climates in obvious When considering particular arrange- ways—such as by expanding market size, ments, the detail of the speci�c rule or stan- reducing costs, facilitating the diffusion of dard obviously matters. Some arrangements technology, and enhancing competition (or provisions within broader arrange- within an economy (chapter 5). Arrange- ments) focus on the process of international ments that foster closer regional integration cooperation—such as facilitating coopera- can be especially important for smaller tion between national regulatory agencies on economies (chapter 3). But this chapter takes enforcement issues. Many others deal with a broader view and considers the potential the substantive rules that form part of the advantages—and tradeoffs—in using inter- investment climate facing �rms directly, and national arrangements as part of a strategy so in principle could be implemented by gov- for improving the investment climate. It ernments acting unilaterally. For example, focuses on three possible contributions: governments can unilaterally provide guar- antees against expropriation, liberalize their • Enhancing the credibility of government trade and investment regimes, protect intel- policies and commitments to reduce lectual property rights, and regulate to safe- risks faced by �rms. guard their environment in the absence of • Harmonizing rules and standards to international commitments. When making reduce costs in international transactions. judgments on their domestic policies and • Addressing the spillover effects policies rules in each area, governments need to con- in one country can have on others. sider the costs and bene�ts of alternative approaches. International arrangements can International arrangements influence the calculation in several ways: and the investment climate International arrangements affecting the • Entering an international obligation on investment climate have a long history. In the a particular issue increases the costs of 12th century cities in northern Europe joined policy reversal and so enhances policy to form the Hanseatic League to protect com- credibility. This can improve the invest- merce.1 At least since the 1920s international ment climate by reducing the risks facing law has recognized limits on the ability of gov- �rms. But the tradeoff is forgone policy ernments to expropriate foreign property.2 flexibility on the issue in question. 175 (c) The International Bank for Reconstruction and Development / The World Bank 176 WORLD DEVELOPMENT REPORT 2005 • Adopting common or harmonized rules or or the North American Free Trade Agree- standards on some issues can reduce transac- ment (NAFTA), may lead governments to tion costs in international trade and invest- offer policy commitments on a range of ment, and so facilitate exports or inward matters that, considered alone, might be investment. It can also signal compliance less appealing. In these cases governments with high international standards. But there need to evaluate the package of rights and can be tradeoffs in adopting approaches obligations as a whole. that are less customized to local circum- • Entering international commitments can stances, and in foregoing the bene�ts from a be used as part of a strategy for pursuing degree of competition between approaches. or sustaining domestic policy reforms. • Pursuing collaborative approaches on Entering commitments to reduce the risk some policy issues may be necessary to of policy reversal is one manifestation of address spillover effects that national this, but governments can also use inter- policies can have on other countries. In national norms to help build consensus these cases there can be tensions between for new policy approaches.3 national sovereignty and international Given the many tradeoffs in this area, collaboration as well as over the most international arrangements vary not only in appropriate form of cooperation. their content, but also in the level of com- Beyond the substantive effect of particular mitment and in the scope of their participa- international obligations, calculations may be tion (box 9.1). These tradeoffs need to be influenced by two broader considerations: considered in the context of particular pro- posals. But it is useful to review some of the • Accepting international obligations on broader tensions and tradeoffs in the three some issues may be necessary to obtain areas of particular importance from an bene�ts in other areas as part of a broader investment climate perspective: enhancing negotiation. For example, the potential credibility; fostering harmonization; and bene�ts from joining an international addressing international spillovers. “club,� such as the World Trade Organiza- tion (WTO), the European Union (EU), Enhancing credibility The impact of particular government poli- BOX 9.1 Evaluating rules and standards—compliance cies, laws, and regulations in supporting pro- mechanisms and participation ductive investment is ultimately determined by their credibility (chapter 2). Can �rms rely The role and impact of any particular inter- allegations of noncompliance.The OECD on them with con�dence when making their national rule, norm, or standard is affected Corporate Governance Principles go further by the mechanisms for securing compliance by providing a mechanism for governments investment decisions? Credibility can be and by the scope of participation in the to voluntarily have their compliance undermined by many things, including the arrangement. assessed by an independent third party. pressures governments face to pursue short- Compliance mechanisms. At one end of Participation. Some arrangements are term political goals at the expense of longer- the spectrum norms may be expressed as bilateral—such as the more than 2,200 bilat- formal treaty obligations, and violating them eral investment treaties concluded since term bene�ts to society. Governments can may expose defaulting governments to 1959. Others are regional—examples enhance the credibility of their policy com- sanctions of various kinds. In some cases the include the EU, NAFTA, the Common Market mitments through domestic institutions, arrangement includes detailed mechanisms of the South (MERCOSUR), APEC, and New such as enshrining key protections in consti- for dealing with allegations of noncompli- Partnership for Africa’s Development ance (WTO Dispute Panels). At the other end (NEPAD). Still others are multilateral, and so tutions and creating independent judiciaries of the spectrum, norms may be no more could have global adherence—examples (chapter 2). When domestic institutions are than a statement of common intent or aspi- include various U.N.–sponsored at early stages of development their impact ration, influencing governments mainly arrangements and the WTO. Arrangements through reputation effects, such as Declara- with a large number of parties have the on credibility may be weak, however, increas- tions by the Asia-Paci�c Economic Coopera- potential for broader impact but can also ing uncertainty and risk for �rms. Entering tion (APEC). In between is a rich menu of involve arduous and protracted speci�c contractual commitments with �rms hybrid approaches that seek to leverage the negotiations. For example, the Uruguay may complement these efforts, but they need reputation concerns of governments. For Round of multilateral trade negotiations example, the OECD Guidelines for Multina- involved active negotiations over nearly to be negotiated �rm by �rm, limiting the tional Enterprises involve no formal obliga- eight years, and negotiations for the U.N. impact on the broader investment climate. tions but contain a mechanism for reporting Convention on the Law of the Sea took nine. Entering international arrangements on particular policy issues can enhance credi- (c) The International Bank for Reconstruction and Development / The World Bank International rules and standards 177 BOX 9.2 BITs—enhancing credibility one bit at a time? The �rst bilateral investment treaty (BIT) dates agencies.Despite this,empirical studies have not BIT is in place at the time of considering an invest- from 1959 (Germany–Pakistan), and the number found a strong link between the conclusion of a BIT ment, and indeed investors may remain oblivious has since proliferated. By the end of 2002 BITs and subsequent investment inflows.Why? until some issue arises when its provisions may be covered around 22 percent of the stock of foreign Several factors may be at work. First, as high- relevant. If so, promoting wider understanding of direct investment (FDI) in developing countries. lighted in chapter 2, �rms make their investment BITs might enhance investor responses. At the center of most BITs are obligations decisions based on an assessment of opportuni- For all these reasons, the impact of BITs on not to expropriate property without compensa- ties as a package, and treaty protections alone investment flows should not be over sold.Well- tion. BITs also typically include provisions gov- will rarely be decisive. A BIT addresses only one crafted treaties can nevertheless form a useful erning the repatriation of pro�ts and the trans- part of �rms’investment equation, and so by itself part of strategies to address policy risks than can fer of funds.They also include standards of is not enough to overcome problems with infra- stifle private investment.They can be particularly nondiscrimination on admission, establishment, structure or other parts of the investment climate. valuable for countries with weak domestic institu- and post-establishment phases of investment. In Indeed, given the costs and delays associated with tions—including the many countries where �rms addition, they provide mechanisms for settling enforcing treaty obligations, BITs are not a com- lack con�dence in the courts to uphold their disputes between the two contracting states, plete solution even to the issues they address. property rights (chapter 4). Indeed China signed and often also between an investor of one state Second, the negotiation of BITs is often driven by nearly 100 BITs in the 1980s and 1990s, at a time and the government of the host state. governments seeking to foster closer diplomatic when its constitution did not provide protections Assurances of this kind can contribute to the ties, rather than immediate interest from for private property rights. investment climate of the host country,and there is investors.To the extent this is so, there need be no some evidence that investors rely on those assur- direct connection between signing a treaty and Source: Dolzer and Stevens (1995); World Bank ances.Indeed,in some cases a BIT is a precondition subsequent investment activity.Third, there is evi- (2003b); Hallward-Driemeier (2003); UNCTAD for obtaining political risk insurance from bilateral dence that many investors are not aware that a (2003e); and UNCTAD (1998). bility by increasing the costs of reneging on at particular levels, with any subsequent tariff the commitment.4 The price of such credi- increase creating an obligation to provide bility is forgone policy flexibility. While few compensation. Dispute settlement mecha- governments today would claim the right to nisms under the WTO facilitate the enforce- expropriate private property without com- ment of these obligations and thus enhance pensation, the prudence of entering binding the credibility of government trade policy commitments on many other policy issues is commitments. Similarly, bilateral investment less straightforward. Reflecting these trade- treaties (BITs) include commitments not to offs, international instruments provide a expropriate property without compensation, menu of approaches to calibrate the form prohibit discrimination between investors, and extent of commitment to particular pol- and provide a range of other obligations (box icy issues. Traditional approaches focused on 9.2). The number of countries participating in government-to-government treaty obliga- BITs has grown steadily since 1960 (�gure 9.1). tions, but two other models are rising in prominence for investment climate issues. Figure 9.1 Participation in bilateral investment treaties (BITs) has shot up in recent years The �rst involves a lower level of commit- 50 ment, through voluntary compliance, and African countries Number of countries with BITs in each decade rests mainly on leveraging governments’ Developing countries in Asia & Pacific Latin America & the Caribbean concerns about their reputations. The sec- 40 Central & Eastern European countries ond involves a higher level of commitment Developed countries by allowing private �rms to enforce the obligations against the government directly 30 through binding international arbitration. Traditional government-to- 20 government treaty obligations Traditional approaches involve governments 10 entering reciprocal commitments, with default by one party creating the possibility of sanc- tions at the initiative of other government par- 0 ties. For example, the WTO provides a mecha- 1960s 1970s 1980s 1990s nism for governments to “bind� import tariffs Source: UNCTAD (2000b). (c) The International Bank for Reconstruction and Development / The World Bank 178 WORLD DEVELOPMENT REPORT 2005 Joining a regional economic cooperation and on the parties’ incentives to enforce the arrangement can also enhance policy credi- agreement. Agreements between parties that bility. For example, in return for access to a demand high levels of mutual compliance fairly liberal internal market, the EU will have a bigger impact on credibility than requires member states to comply with a agreements involving those with lower range of policy requirements. The prize of expectations. access to a larger market provides incentives for governments to improve their policies Arrangements with voluntary to meet EU requirements, and the desire to compliance mechanisms remain in good standing encourages gov- Given the tradeoffs between commitment ernments to sustain those policies. Similar and flexibility, international arrangements factors can be seen at work as NAFTA opens on some issues do not impose binding treaty to new members. obligations. These arrangements may never- In these cases it can be dif�cult to disen- theless enhance credibility if they leverage tangle several complementary effects. First, governments’ interest in improving or pre- access to a larger market can itself enhance serving their reputations. For example, the investment opportunities. Second, the pol- OECD Corporate Governance Principles do icy improvements undertaken as a condi- not impose binding obligations—govern- tion of joining the club can improve the ments can ignore them with impunity. They investment climate. Third, there is the do, however, include a mechanism that impact on credibility through reduced like- allows governments to submit their domes- lihood of reversing policy reforms in ways tic laws and policies to scrutiny by an inde- that might jeopardize continuing member- pendent third party. Governments inter- ship of the arrangement. Indicators of a ested in signaling to investors that they country’s “investment pro�le�—which apply high regulatory standards in this area focus on perceived risk to investment—sug- have incentives to submit their policies to gest that the impact on credibility may be scrutiny—and to attain high standards. signi�cant (see �gure 9.2). Countries including Brazil, Georgia, India, The impact of an international treaty on the Philippines, Poland, and Turkey have each party’s policy credibility will depend on subjected their policies to such assessments.5 the speci�c provisions of the agreement— A similar model is being adopted by the New Partnership for Africa’s Development Figure 9.2 NAFTA and Mexico’s investment pro�le (NEPAD; box 9.3). 12 As with arrangements resting on more tangible sanctions, the attitudes of other participants toward compliance make a dif- 10 ference—low standards of compliance will lower the impact on credibility. Arrange- Investment profile ments that maintain high membership standards will thus deliver stronger bene�ts 8 Mexico than more permissive schemes. When com- pliance depends on reputation alone, the transparency and integrity of the monitor- 6 ing mechanism is critical to success. Other Latin America Arrangements giving private �rms 4 direct recourse to governments 1992 1994 1996 1998 2000 2002 Traditionally the remedy for foreign Note: The NAFTA agreement between Canada, Mexico, and the United States went into effect in 1994. “Other Latin America� is an investors who believed they had been average of 18 other Latin American countries. The ICRG investment harmed by an action of the host govern- pro�le index reflects factors affecting risk to investment, including contract viability/expropriation, pro�ts repatriation, and payment ment was to pursue their claim against the delays. The maximum value of the index is 12. Source: Authors calculations based on the International Country government before local courts. Investors Risk Guide database. often felt this was inadequate, with con- (c) The International Bank for Reconstruction and Development / The World Bank International rules and standards 179 cerns that the local court might be biased in favor of the host government or otherwise BOX 9.3 NEPAD and its peer review mechanism not provide an effective remedy. The imme- As part of an effort to improve the quality of governments. Each participating country diate response was for investors to enlist the governance in Africa, the New Partnership submits to peer review and ongoing moni- support of their home government to pur- for Africa’s Development was created in toring.The country is evaluated on sue the �rm’s interests through diplomatic 2001 by regional governments. It puts economic and political grounds according enhancing government credibility front and to a set of standards that include democracy channels. This also had its limits and weak- center. An African Peer Review Mechanism and political governance, economic gover- nesses. The fate of the �rm’s claim often is the core instrument. nance and management, corporate gover- depended on diplomatic and political rela- NEPAD includes principles to improve nance, and socioeconomic development. tions between the two governments. In political governance and economic The review is to be undertaken by experts reform—and to promote competition, trade, appointed by an independent panel, with some cases claims might be ignored. In oth- investment, macroeconomic and political the results made public. ers what was essentially a commercial dis- stability, and sustainable development.The pute became politicized, sometimes culmi- peer review mechanism enhances the trans- Source: Funke and Nsouli (2003) and NEPAD nating in interminable negotiations—and parency and accountability of participating of�cial documents. sometimes in the use of armed force.6 When the rights and obligations of the investor and the host government are set out provisions are included in NAFTA. This has in contracts, one option is for the parties to expanded access to ICSID jurisdiction, and agree to submit any contractual disputes to the volume of cases submitted to ICSID has international arbitration by a neutral party. grown strongly in recent years—more than This approach has a long history in interna- half the 129 cases it has registered since its tional commerce, and is supported by a inception were �led in the last �ve years.9 range of international conventions and insti- The use of BITs and other agreements tutions.7 In 1966 the International Centre for that include prior consent to ICSID juris- Settlement of Investment Disputes (ICSID) diction creates a new source of discipline on was established by international convention host governments—and a potentially pow- to specialize in investment disputes between erful tool to enhance the credibility of their host governments and foreign investors.8 contractual and policy commitments. Gov- The convention has since been rati�ed by ernments and �rms can both bene�t. Gov- 140 countries. Under ICSID �rms from one ernments bene�t from a commitment member state can pursue their investment device that can address concerns from disputes against other member states investors, and thus help them attract more through binding international arbitration, investment at lower cost, and also reduce without the need to involve their home gov- the risk of any later dispute becoming ernment. The governments can pursue politicized. Firms bene�t from reduced investors directly as well. The parties are risks and a more reliable mechanism for responsible for appointing the arbitrators protecting their rights if the relationship and abiding by the decision. Typically the with the host government deteriorates. investor and the host state each choose an While ICSID is designed to encourage for- arbitrator, and the parties have to agree on a eign investment, domestic �rms can bene�t third arbitrator. Sitting in a neutral venue, from the halo effect provided by stronger the arbitrators hear evidence and render an constraints on arbitrary government action. award. ICSID provides the procedural rules As with effective courts (chapter 4), the and a small secretariat to support the arbi- bene�ts from an effective system of interna- trators and the parties. tional dispute settlement are not measured in As with other forms of arbitration, the number of cases heard, but in the incen- ICSID’s jurisdiction rests on the consent of tives it creates for the parties to adhere to the parties, often given through clauses their commitments. The threat of possible inserted in investment contracts. In the 1990s sanctions that might later be imposed by an it became common for BITs to include provi- arbitration panel can deter governments sions for governments to give their prior con- from reneging on their commitments and sent to ICSID jurisdiction, thus eliminating give the parties an incentive to come to a the need for case-by-case agreement. Similar negotiated solution. (c) The International Bank for Reconstruction and Development / The World Bank 180 WORLD DEVELOPMENT REPORT 2005 Despite the potential advantages, the (chapter 4). Arbitration panels have so far system of investor-state dispute settlement tended to interpret the treaty provisions has raised several debates. Does it impose equally cautiously,10 and can also deter too much discipline on governments? Does frivolous claims by the threat of sanctions. that discipline encroach on governments’ regulatory prerogatives? And is the process Suf�cient transparency? Investor–state dis- suf�ciently transparent? pute resolution involves agreement by the parties (including rati�cation by govern- Too much discipline? Some governments ments of relevant treaties), and both parties have recently been subjected to claims from are equally involved in determining the �rms for substantial damages as a result of composition of the arbitration panel. Arbi- alleged breaches of contractual or treaty tration evolved from diplomatic and com- commitments. The sums actually awarded mercial practice, where it was customary for by arbitration panels, if any, depend on proceedings to be con�dential. This has led �ndings of liability and on the losses experi- some observers to question whether the enced by �rms, but for large infrastructure arrangements are suf�ciently transparent, or resource investments the sums might be particularly when matters of broad public substantial. Is this too heavy a burden to interest are involved. While practice under place on governments? The main alterna- different arbitration regimes varies, ICSID tives would be to return to an approach that has always promoted transparency, and led to the politicization of investment dis- efforts are underway to further increase the putes, or to allow governments to ignore opportunities for public participation in their commitments with impunity. While dispute proceedings, making the procedure the second path might appear attractive for more analogous to a court hearing. ICSID governments in the short term, the conse- also has a procedure for challenging awards. quence would be that no �rm could rely on As the system evolves, there will likely be a government’s commitments, and this risk pressures for even greater transparency will be reflected in investment decisions (box 9.4). (chapter 2). Encroaching on regulatory prerogatives? Fostering harmonization Most BITs and similar agreements include In the normal course of events each country a prohibition against expropriation with- or jurisdiction tends to develop its own out compensation, and there is general rules and standards on particular issues to consensus that prohibitions against out- reflect local customs, conditions, and prior- right seizure of property are appropriate. ities. This adaptation is an important part There is concern, however, about how pro- of ensuring a good institutional �t—and hibitions against “indirect� expropriation one reason to be cautious in uncritically might affect a government’s regulatory transplanting regulatory systems from prerogatives. It is clear that some govern- other countries (chapter 2). A mixture of ments have used arbitrary regulation or adaptation and experimentation can also taxation to achieve a result equivalent to lead to the discovery of new and better ways expropriation, and most observers agree of achieving particular policy goals. Institu- that such behavior should be caught by the tional competition between jurisdictions prohibition. But concern has been can also encourage governments to attain expressed that the provisions might be higher standards.11 interpreted to restrict legitimate regulatory Divergent approaches to some regulatory action by host governments, or that even issues, however, can increase the costs of the potential for such claims might induce international trade and investment transac- a “regulatory chill.� Similar issues have tions. If goods or services need to meet dif- been debated under guarantees against ferent standards and regulatory require- expropriation contained in national con- ments in every country, customization can stitutions, where the result has been to pre- drive up the costs of production and distrib- serve legitimate regulatory prerogatives ution and reduce competition. Diverse (c) The International Bank for Reconstruction and Development / The World Bank International rules and standards 181 BOX 9.4 The evolving system of investor-state dispute settlement The recent rise in the number of investment disputes brought before The acceptability of investor-state arbitration also depends on the per- ICSID arbitration panels has put investor-state arbitration in the ceived fairness of the results.The state party prevailed in half of the 24 dis- spotlight. putes that went to �nal award between 1987 and 2003. Arbitration proceedings were traditionally con�dential, but ICSID’s Cases brought to ICSID,1987–2003 Under NAFTA Under BITs rules require making a dispute public and encourage parties to publish information about the dispute and its outcome. Concerns about the trans- Cases registered 10 87 parency of international arbitration between investors and states are also Cases concluded (including settlement) 6 31 leading to procedures that more closely resemble those of judicial Final awards rendered 6 18 proceedings. For example, in a recent case brought against the United Cases in which investor prevailed 2 10 States under NAFTA, the parties agreed to use an amicus curiae (friend of Cases in which state prevailed 4 8 the court) procedure allowing nondisputing parties to make submissions Average duration (from constitution of to the arbitration panel.The U.S. government has also modi�ed its model tribunal or ad hoc committee), months 29.5 28.2 BIT agreement, incorporating provisions for greater transparency in new agreements.The Chile–U.S. Free Trade Agreement contains a requirement Note: Data through February 2003. that arbitration panels conduct the hearings open to the public and dis- Source: ICSID Web site, World Bank staff, and of�cial texts of the mentioned close key documents. agreements. approaches can also increase the costs for- To be effective, common international eign �rms face when evaluating alternative standards do not always require binding investment locations, perhaps deterring treaty obligations. Countries, or even �rms, them from pursuing investments in coun- can voluntarily adopt common norms, with tries with unfamiliar arrangements. Beyond the incentives to comply driven by reputa- reducing transaction costs, adoption of tion. Some international agencies have also international standards can also facilitate developed “model laws� to encourage conver- domestic policy reform when local interest gence on common approaches, but leaving groups have conflicting preferences.12 Adop- countries the freedom to adapt approaches to tion of international standards can also sig- local circumstances; the UNCITRAL model nal to �rms, consumers, and other groups law on international commercial arbitration, the application of high regulatory standards. for example, has been adopted by more than The tensions between local customiza- 35 jurisdictions. tion and international harmonization play There can also be alternative strategies for out in proposals to develop common inter- achieving the same end. For example, rather national rules and standards on a wide range than adopting identical rules in each jurisdic- of issues relevant to the investment climate. tion, participating governments may agree, in Efforts to develop uniform standards to ease mutual recognition schemes, to accept in international commerce have long been a focus of private bodies such as the Interna- tional Chamber of Commerce.13 Comple- BOX 9.5 Harmonizing business law in Africa—OHADA mentary efforts at the intergovernmental The Organisation pour l’Harmonisation en �rm investing in Togo will thus be dealing level include those of the United Nations Afrique du Droit des Affaires (OHADA), with many of the same regulatory require- Commission on International Trade Law14 established in 1993, promotes the harmo- ments as in its own country, and a foreign nization of business law in Africa. It has 16 investor familiar with the laws in one coun- (UNCITRAL) and a variety of other inter- member states: Benin, Burkina Faso, try can apply the same understanding to national agencies. In francophone Africa, for Cameroon, Central African Republic, other OHADA countries.The result should example, harmonization of business law is Comoros, Congo, Côte d’Ivoire, Gabon, be lower transaction costs and reduced being facilitated by the Organisation pour Guinea, Guinea-Bissau, Equatorial Guinea, uncertainty. Mali, Niger, Senegal, Chad, and Togo. The OHADA Treaty also establishes a l’Harmonisation en Afrique du Droit des Under OHADA, the texts of “Uniform Common Court of Justice and Arbitration, Affaires (OHADA, box 9.5). The possible Acts� are endorsed by a Council of Ministers which acts as an advisory body to the Coun- areas for cooperative action range from and then made directly applicable in each cil of Ministers, serves as an appeal body to developing a common set of international member country. So far the harmonization foster common interpretations of the Uni- process has resulted in uniform acts in six form Acts, and supports the resolution of rules on contract law to harmonizing inter- areas: general commercial law, companies, commercial disputes. national accounting standards. Clearly the securities, debt recovery, bankruptcy and costs and bene�ts of each approach need to insolvency, and arbitration. A Senegalese Source: Ba (2000) and OHADA of�cial documents. be considered case by case. (c) The International Bank for Reconstruction and Development / The World Bank 182 WORLD DEVELOPMENT REPORT 2005 their jurisdiction goods or services that meet the largest bene�t, but increase the challenge the regulatory requirements of another par- of developing approaches that will meet the ticipating jurisdiction. This approach has interests of all participating governments. done much to facilitate commerce within the They can also involve protracted negotia- EU, between the EU and some nonmember tions. Reflecting these tradeoffs, the number states, and between Australia and New of regional economic cooperation arrange- Zealand. Similar approaches could have wide ments has grown strongly in recent years application across a range of investment cli- (�gure 9.3). mate issues. For the liberalization of trade and invest- A more ambitious form of harmoniza- ment, there is an ongoing debate over tion is to agree not only on common rules whether regional arrangements are building but also to delegate responsibility for blocks or stumbling blocks to a liberal multi- administering them to a common regula- lateral system.15 Proposals that focus on the tory body. This presents opportunities for harmonization of standards tend to pose greater consistency in interpretation, lower fewer concerns of this kind, although there administrative costs, and possibly enhanced can be other tradeoffs. For example, harmo- credibility for participating governments. In nizing standards at the regional level can practice supranational regulatory bodies are reduce transaction costs for intraregional more often proposed than implemented, in trade and investment, but harmonizing stan- part because of concerns over national sov- dards with major capital exporters or export ereignty. There are exceptions. For example, markets outside the region might offer even OHADA has a common court to foster con- greater bene�ts. sistent interpretations of harmonized busi- ness laws, and the Eastern Caribbean Addressing international Telecommunications Authority regulates spillovers telecommunications in �ve small countries Many international arrangements, existing in the Caribbean. Progress usually requires a and proposed, seek to address international governance framework that gives each par- spillovers of some kind—where actions in ticipating government effective voice—and one country can have effects on others. a high level of trust between participants. The clearest cases involve environmental The advantages and disadvantages of har- protection. For example, emissions or efflu- monization proposals also depend on the ents from industries in one country may number of countries participating in the harm the environment in other countries. arrangement. Multilateral approaches offer When this happens, international coopera- tion may be needed to mitigate the negative Figure 9.3 Regional economic cooperation agreements proliferated in the 1990s externality and achieve an ef�cient out- 200 come. Indeed, there has been a growing volume of international rules on various matters affecting the environment since the 1970s.16 Not all environmental issues have Cumulative number of agreements 150 an international dimension, however, and thus warrant international action. For example, when the adverse effects of pollu- 100 tion are contained within a country’s bor- ders, the case for overriding the sovereignty of that government is weak.17 50 Outside environmental protection, there are also many areas where the argument for international cooperation can be strong. This is the case with international efforts to 0 combat corruption, for example, which can 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 Note: Agreements reported to the WTO. seriously undermine investment climates Source: WTO Web site. (box 9.6). (c) The International Bank for Reconstruction and Development / The World Bank International rules and standards 183 When spillovers are less tangible, or the bene�ts less evenly shared, the case for BOX 9.6 International cooperation to combat international cooperation can be more corruption complex. Take competition policy. There is National antibribery laws date from at least ance.Transparency International growing understanding of the importance the Law of Moses in the 9th century BCE. complements of�cial monitoring with a of adopting cooperative approaches to the The �rst attempt to address bribery on an series of public reports on each country’s investigation and prosecution of interna- international level came in the 1976 OECD progress in stemming the bribery of foreign tional cartels, which can impose large costs Guidelines for Multinational Enterprises. of�cials. This foreshadowed the most signi�cant step An even more ambitious effort to foster on countries. In the 1990s about 40 interna- to date, the rati�cation of a multilateral con- international cooperation is the U.N. Con- tional cartels were prosecuted in the EU and vention committing parties to make the vention against Corruption, signed in 2003 United States alone. The average interna- bribery of a foreign of�cial by one of its citi- by 106 countries and entering into force in tional price increases due to those cartels zens a criminal offense. 2005. It stems from two previous U.N. The OECD Convention on Combating arrangements—the U.N. Declaration are estimated to have been around 20–40 Bribery of Foreign Public Of�cials in Interna- against Corruption and Bribery in Interna- percent. It was also found that many of tional Business Transactions, signed in 1997 tional Commercial Transactions and the U.N. these cartels speci�cally targeted developing by all 30 OECD member countries and 5 Convention on Transnational Organized countries without appropriate national leg- nonmember countries (Argentina, Brazil, Bul- Crime—and complements the OECD con- garia, Chile, and Slovenia), went into force in vention. It addresses cross-border issues islation in place. The imports of 12 1999.The Convention provides guidelines associated with recovering assets, freezing cartelized products by developing countries and a monitoring mechanism to improve accounts, and seizing foreign property of in 2000 alone exceeded $10 billion.18 Even domestic antibribery laws and outlines areas corrupt of�cials. where coordinated action to reduce corrup- when the argument for action is strong, tion should be taken.To ensure the parties Source: Of�cial texts of Conventions, however, there is room for debate about the live up to their agreement, the Convention Transparency International (2004), and Braith- best form of that action. Should it be lim- establishes procedures to monitor compli- waite and Drahos (2000). ited to coordination between national agen- cies? Should efforts focus on providing technical assistance to help national govern- rules on taxes with the goal of slowing the ments establish effective national regimes? movement of �rms to countries that prefer Or is a multilateral agreement on competi- lower taxes—but the latter countries have tion policy required?19 The last option no incentives to cooperate. Such differences could have signi�cant implications for in perspective have stymied progress in developing countries, most of which have reaching agreement on these matters, even not yet established competition agencies. between countries at similar levels of devel- Proposals to develop new international opment, such as in the EU.20 The prospects rules to address issues associated with com- of achieving a truly global accord on mini- petition for investment between countries mum tax rates that incorporates countries can be even more problematic. Competi- with even more divergent perspectives tion between governments to attract or seems a distant prospect at best. retain investment plays an important role in When these differences exist, the chal- driving investment climate improvements lenge extends beyond the feasibility of (chapter 3). But it has led to concerns that negotiating an agreement. Even if uniform there may be a “race to the bottom� in tax international tax rates could be agreed on rates, environmental regulation, or other and enforced, countries could simply shift matters. As discussed in chapter 5, the theo- competition for investment to other dimen- retical support for such races is mixed, and sions of their investment climate policies, so far the dire predictions of some com- such as the provision of infrastructure or mentators do not seem to be taking place. the enforcement of a host of other regula- Indeed, in some cases the race seems to be tions.21 Indeed, given the breadth of policy to the top rather than the bottom. But the areas that influence the investment deci- concern illustrates some of the tensions and sions of �rms, efforts to curb competition practical challenges for international coop- would need to cover a vast �eld—leaving eration on matters where countries can little scope for sovereign states to reflect dif- have divergent perspectives. ferences in social preferences or in levels of Take tax harmonization. Countries that development. Without evidence that such prefer high tax rates may favor international competition is leading to real welfare losses, (c) The International Bank for Reconstruction and Development / The World Bank 184 WORLD DEVELOPMENT REPORT 2005 BOX 9.7 Privatizing international cooperation on corporate social responsibility Efforts to promote international cooperation on Standards are influencing business matters related to the investment climate are not limited to arrangements between govern- ments.There has been a growing trend to Ethical Trading Initiative develop international norms applicable to �rms OECD Guidelines directly, without the intermediation of states. Particularly in the area of corporate social U.N. Global Compact responsibility, many of these initiatives also spring from the nongovernmental sector. ILO Core Conventions These codes of corporate conduct outline World Business Council for basic principles of behavior for �rms, including Sustainable Development corruption and respect for environmental and labor norms. Not legally binding, the codes typi- Global Reporting Initiative cally depend on the reputation concerns of major �rms that operate in more than one coun- ISO14000 try, with compliance often reinforced through 0 20 40 third-party inspection and transparency arrangements. Examples of such initiatives Percent of firms include the Global Reporting Initiative, the U.N. Note: Percentage of �rms that indicated standards influenced their business. Standards emanating from intergovern- Global Compact, the Equator Principles, the Pub- mental initiatives are in dark blue, those by nongovernmental organizations are in light green. The International Organi- lish What You Pay Initiative, and Transparency zation for Standardization (ISO), striped in this �gure, is a nongovernmental organization, but has members drawn from public and private sectors. ILO stands for International Labour Organisation. International’s Business Principles for Counter- ing Bribery. These mechanisms may help �rms adopting high standards to signal their compliance and to or maintaining their international reputations, nating from the nongovernmental sector were burnish their reputations, thus complementing they will also have less impact on the behavior often as influential as those developed by inter- national laws and policies.The proliferation of of other �rms. national agencies (see �gure). new codes and arrangements can, however, cre- A recent survey showed that many �rms ate confusion about acceptable standards. take standards of corporate social responsibility Source: Jorgensen and others (2003); Smith and Because these initiatives affect mainly multina- into account when making location and produc- Feldman (2003); UNCTAD (2001a); Berman and tional �rms that have an interest in enhancing tion decisions—and suggests that those ema- Webb (2003). the case for intruding on the prerogatives of important for countries with domestic national governments seems weak. institutions at an early stage of develop- An alternative strategy is to leverage the ment. Stronger commitment devices offer concerns �rms have for their reputations. greater bene�ts, but they also involve for- As discussed in chapter 2, a growing num- feiting more policy autonomy—and so ber of initiatives aim to address concerns need to be considered carefully. To be sus- about international economic integration tainable, measures that curb domestic pol- by targeting �rms directly, rather than gov- icy autonomy must also be accepted as ernments. Many of these initiatives emanate legitimate, reinforcing the importance of from the nongovernmental sector (box 9.7). efforts to enhance transparency. Measures to reduce costs through inter- Future challenges national harmonization offer many bene�ts International rules and standards can be but involve several tensions. There is the expected to do more in shaping investment tension between harmonization and cus- climates as the intensity of interactions tomization—taking local circumstances between governments and cross-border into account. There is the tension between trade and investment expand. As this brief harmonization and competition—where survey highlighted, progress in that direc- some degree of competition between stan- tion will need to grapple with several gen- dards can be an important part of the learn- eral tradeoffs. ing process. There is the tension between Measures to enhance the credibility of multilateral and other approaches, and in government commitments can be especially the latter case between harmonization with (c) The International Bank for Reconstruction and Development / The World Bank International rules and standards 185 neighbors and harmonization with major markets or sources of capital. Given the BOX 9.8 A multilateral agreement on investment? tradeoffs involved, the preferred approach Proposals to develop a multilateral agree- experience under NAFTA, however, suggests will often vary from issue to issue—there ment on investment have a long history.The that proposals in this area need to place will be no universal models. �rst attempt was in 1929 at the Paris Confer- special emphasis on clarifying the interac- Measures to address international ence on the Treatment of Foreigners.The tions between prohibitions on indirect experiment was repeated again in the 1948 expropriation and domestic regulation— spillovers also need to reflect the divergent Havana Charter. In 1959 two private initia- and enhancing the transparency of perspectives of countries at different levels tives were combined as the Abs-Shawcross investor-state dispute settlement mecha- of development. Care needs to be taken not Draft Convention on Investment Abroad. In nisms.The treatment of restrictions on for- to curtail the policy space of emerging 1967 the OECD produced a Draft Conven- eign capital flows may also be subject to tion on the Protection of Foreign Property. debate (chapter 5). In principle it should be nations without a compelling rationale. At a In 1995–98 the OECD attempted to develop possible to craft an agreement that meets minimum the voices of developing coun- a Multilateral Agreement on Investment. these interests, but the same agreement tries need to be heard when framing these Investment issues were proposed for inclu- would need to meet the interests of devel- initiatives. sion in the Doha Round of the WTO oped countries, which will typically place launched in 2001. In each case the proposal greater emphasis on market-opening mea- While the emerging network of interna- failed to �nd suf�cient support. sures, including between themselves. tional rules and standards can help govern- Looking back, each proposal had its A broad negotiating forum provides ments improve the investment climates of own features and encountered different opportunities to trade concessions across a their societies, a critical challenge is to obstacles. But there are basic challenges in range of subject areas, but it can also constructing an agreement that includes involve complex negotiations that can eas- ensure the arrangements reflect the inter- investment protection provisions (along the ily be derailed. Another option could be to ests of developing countries. Uniform lines of BITs) and market-opening develop or expand regional agreements global rules may be appropriate for some provisions, that meets the interests of capi- with effective investment provisions. NAFTA matters, but differences in priorities and tal exporters and importers, and that could be an example. However, this reflects the interests of both developed and approach offers little help to low-income capabilities need to be reflected in others developing countries. countries in other regions, which would (box 9.8). For a developing country, a multilateral stand to gain the most from effective com- The international community has a agreement that provides high standards of mitment devices. And creating a regional protection for investment should have investment agreement covering only devel- responsibility to help ensure that new inter- many attractions as a tool to reinforce the oping countries would likely offer only lim- national rules and standards reflect the per- credibility of government policies. A multi- ited bene�ts because it would exclude the spectives of developing countries. The best lateral agreement would also reduce the principal sources of investment capital. way to do so is to ensure that developing transaction costs associated with negotiat- Source: Ferrarini (forthcoming); Henderson ing scores of BITs, and reduce inconsisten- (2000);World Bank (2003b); Parra (2000); and countries have the opportunity to partici- cies between those agreements. Recent Warner (2000). pate fully in the development of those arrangements. Recognizing this, multilat- eral and bilateral donors mobilized more than $700 million in technical assistance to support developing country participation need to be mobilized across a range of new in the Doha Round of multilateral trade areas. Other ways that the international negotiations.22 Given the increasing role of community can help developing countries international arrangements in the invest- improve the investment climates of their ment climate area, similar support may societies are the subject of chapter 10. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank How the International Community Can Help IV IMPROVING INVESTMENT CLIMATES IN DEVELOPING COUNTRIES can deliver huge development dividends for the countries concerned, and contribute to a more inclusive, balanced, and peaceful world. PA RT Chapter 10—How the international community can help suggests three things that the international community can do to help developing countries improve their investment climates. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank How the international community can help 10 Improving the investment climates of their motives. Demographic changes over the societies is �rst and foremost the responsi- next 30 years will add nearly 2 billion more bility of host governments, both at the people to developing countries, which will chapter national and subnational levels. They hold become home to 7 billion of the world’s 8 the levers—through their policies and billion people.3 Improving the opportuni- behaviors—to make a huge difference in the ties for young people is fundamental to cre- opportunities and incentives that �rms ating a more peaceful and balanced world— have to contribute to growth and poverty to addressing the roots of political instability reduction. But the international commu- and conflict, and to addressing the pressures nity can lend a hand. for migration. The case for providing that help is com- The development payoffs from support- pelling. There is the imperative of improv- ing better investment climates can be particu- ing the conditions of nearly half the world’s larly strong. For example, the manufacturing people that live on less than $2 a day—and value added unleashed by investment climate the 1.2 billion that barely survive on less improvements in even a single country can than $1 a day.1 Indeed, recognizing the far exceed the development assistance pro- importance of growth in developing coun- vided worldwide (�gure 10.1). tries, the international community has com- This chapter highlights three ways the mitted to the Millennium Development international community can help improve Goals—the �rst of which is to halve by 2015 the investment climates in developing the proportion of people living on less than countries: $1 a day.2 There are also more pragmatic • By removing policy distortions in devel- oped countries that harm the investment Figure 10.1 Manufacturing value added in a single country can far exceed net global climates in developing countries of�cial development �nance • By providing more, and more effective, 500 assistance to the design and implementa- China tion of investment climate improve- 400 ments, and better leveraging support pro- vided directly to �rms and transactions Billions of 1995 dollars • By tackling the substantial knowledge 300 agenda to help policymakers broaden South Korea and accelerate investment climate improvements. 200 Removing distortions in Global net official 100 development finance India developed countries Developing countries are not alone in grap- pling with investment climate improve- 0 ments. Developed countries have distorted 1970 1975 1980 1985 1990 1995 2000 Note: Data for China, India, and South Korea show manufacturing value added. their own investment climates, imposing Source: OECD online database and World Bank (2004k). signi�cant costs on their societies but often 189 (c) The International Bank for Reconstruction and Development / The World Bank 190 WORLD DEVELOPMENT REPORT 2005 also harming the investment climates of equivalent to tariffs of 10 to 30 percent.7 developing countries. Why? Because of the The mutual gains from trade make a strong same clientelist politics that can plague case for removing these restrictions— developing countries. They maintain tariff rather than responding to protectionist and nontariff barriers to trade and provide urges that penalize developing countries support and export subsidies to their for progress. industries that distort incentives in their domestic markets and reduce opportuni- ties for productive investment in develop- Providing more, and more ing countries. effective, assistance The magnitude of these distortions can As highlighted throughout this Report, be staggering. While average import tariffs improving the investment climates of their have been declining as a result of successive societies involves many challenges for gov- rounds of multilateral trade negotiations, ernments. The international community tariffs on individual products can still can help by providing development assis- exceed 100 percent—and in some cases tance to help design and implement those reach 500 percent. Tariffs also tend to esca- improvements. Those efforts can be com- late on semi-processed and fully processed plemented by support provided directly to products, contributing to effective rates of �rms and transactions. The international protection that can far exceed the nominal community has long been active in both tariffs involved.4 Nontariff barriers and areas, but there are opportunities to do other distortions are also pervasive. In agri- better. culture, for example, OECD countries pro- Development assistance to support the vided $311 billion of subsidies to their design and implementation of investment farmers in 2001—nearly four-and-a-half climate improvements can take many times the amount allocated to of�cial devel- forms. According to estimates prepared for opment �nance.5 this Report, assistance provided by major The impact of these distortions on bilateral and multilateral donors for invest- developing countries is substantial. Tariff ment climate improvements averaged $21.1 escalation is particularly harmful because it billion per year between 1998 and 2002—or reduces opportunities for developing coun- about 26 percent of all development assis- tries to diversify away from commodities by tance.8 The bulk of that assistance went to expanding into higher-value-added prod- infrastructure development, followed by ucts. It has been estimated that removing policy-based support and technical assis- the various distortions imposed by devel- tance. Most of that support was provided in oped countries could deliver gains to devel- the form of loans (table 10.1). oping countries of $85 billion in 2015—or Support provided directly to �rms and more than four times the development transactions also has the potential to con- assistance currently provided for invest- tribute to or complement investment climate ment climate improvements.6 Expanding improvements. That support accounted for market access opportunities for products an average of $3.1 billion per year of devel- from developing countries would be espe- opment assistance between 1998 and 2002, cially bene�cial for poverty reduction and a further $26.4 billion of support in because agricultural and labor-intensive other forms. goods usually face import tariffs twice as high as those for other products. Supporting investment The bene�ts are not limited to trade in climate improvements goods. Service trade is a growing source of Assistance for the investment climate has opportunities for many people in develop- bene�ted from recent improvements in the ing countries—and is delivering bene�ts to planning and delivery of development assis- �rms and consumers in developed coun- tance generally. There is a growing empha- tries as well. Service industries in OECD sis on improving the effectiveness, not just countries already bene�t from protection the volume, of assistance. There is a sharper (c) The International Bank for Reconstruction and Development / The World Bank How the international community can help 191 Table 10.1 Support for investment climate reforms and to �rms and transactions: annual averages 1998–2002 (billions of 2001 dollars) Development assistance Other assistance Concessional Total grants Non-concessional Grants loans and loans loans Guarantees Support to investment climate reforms Policy-based support 1.5 5.5 7.0 n.a. n.a. Technical assistance 1.7 1.0 2.7 n.a. n.a. Investment in infrastructure 1.7 9.7 11.4 3.2 3.0 4.9 16.2 21.1 Support to �rms and transactions Development assistance 1.1 2.0 3.1 n.a. n.a. Other support n.a. n.a. 0.0 13.4 6.8 6.0 18.2 24.2 16.6 9.8 Note: n.a. = not applicable. “Policy-based support� includes quick disbursing operations such as structural adjustment, balance of payment, and general and sectoral programmatic assistance; policies supported under such operations may have been from several sectors. “Technical assistance� includes projects providing technical assistance, training, and other capacity building assistance for legal reform, privatization, research and scienti�c institutions, and employment policy and administration; �nance and banking, trade and tourism and industry, export promotion, mining and construction; and infrastructure policy, administration, and regulation. “Infrastruc- ture� includes physical investments in energy, telecommunications, and transport. “Development assistance to support �rms and transactions� includes �nancial (such as lines of credit) and non�nancial support (such as business development services) provided directly or indirectly to small private �rms. “Other support� includes non- concessional loans and guarantees provided by international development �nance institutions and export credit agencies for periods exceeding one year. Source: Authors’ calculations using OECD CRS data, data gathered by IFC using the methodology de�ned in IFC (2002) and Migliorisi and Galmarini (2004). focus on poverty reduction, as reflected in tion of new international rules and standards. commitments to the Millennium Develop- Many of the lessons of experience in manag- ment Goals. There is greater recognition of ing reform processes—including in relation the key role of government policy in ensur- to priority setting and consensus building— ing that aid is effective, leading to greater and in designing particular interventions are selectivity across countries.9 There are as relevant to donors as they are to develop- greater efforts to ensure country ownership, ing country governments. The key is thus to putting governments in the driver’s seat focus on addressing important constraints, through Country Development Strategies which need to be identi�ed in each case, and and Poverty Reduction Strategies.10 There is to support a process for ongoing improve- a greater focus on results, which has led to ments (chapter 3). The international com- new initiatives that link support to demon- munity can also draw on a growing body of strated performance. These initiatives experience in designing and implementing include efforts to link support to perfor- assistance in each area of support. mance at the country level, such as the out- come-oriented benchmarks for budget sup- Policy-based support. Support to policy port piloted by the EU and the Millennium reforms can take many forms. Policy-based Challenge Account initiative launched by or programmatic support can play an impor- the United States.11 Results-focused efforts tant role, and accounted for an average of $7 also include initiatives at the program or billion per year during 1998–2002—or 33 project level, such as linking the disburse- percent of development assistance for invest- ment of support to the actual delivery of ment climate improvements.13 outputs rather than the �nancing of The focus of this support has changed over inputs.12 Finally, there is a growing empha- time, reflecting the evolution of important sis on development knowledge, including constraints and the emergence of new issues. research on global public good issues, In the 1980s the main focus was on macroeco- knowledge sharing, and more rigorous nomic stability, reducing price and exchange impact evaluation. rate controls, liberalizing �nancial sectors, Development assistance to support invest- and reforming public enterprises. By the ment climate reforms can cover the full 1990s emphasis began to shift to microeco- gamut of issues discussed in this Report, from nomic and institutional reforms to build or improving governance to supporting devel- improve markets. By the end of the 1990s the oping country participation in the negotia- priority areas were improving the business (c) The International Bank for Reconstruction and Development / The World Bank 192 WORLD DEVELOPMENT REPORT 2005 environment, establishing the conditions for alyze policy improvements, bring world- private participation in infrastructure, and class expertise to bear on the design of par- helping to support global integration.14 ticular reforms, and strengthen the capabil- Investment climate improvements are, at ities of policymakers and regulators. While their heart, about improving the quality of some technical assistance may be embodied governance and policymaking, and often in other forms of support, estimates pre- need to confront resistance from those who pared for this Report suggest that technical bene�t from the status quo (chapters 2 and assistance for investment climate improve- 3). In the 1980s and 1990s the international ments averaged $2.7 billion per year from development community used conditional- 1998 to 2002—or just 13 percent of devel- ity attached to international assistance as a opment assistance for investment climate way to encourage policy reforms. While improvements over that period. The sup- often controversial, it proved useful on occa- port ranged from around $200 million a sion and indeed was sometimes sought by year in East Asia and Paci�c to $600 million governments to lock in policy commitments per year in the regions of Sub-Saharan and deal with resistance from local interest Africa, Europe and Central Asia, and Latin groups.15 Too often, however, external actors America and Caribbean. were perceived to be driving country strate- As with any form of assistance, the qual- gies, and when the government was not fully ity, not the volume, determines effectiveness. committed, the promised reforms often In this context donor agencies have been never materialized. New laws were passed to grappling with three main challenges in meet conditionality requirements but not increasing the effectiveness of technical assis- implemented. New regulatory agencies were tance for investment climate improvements. established on paper but not staffed or given political support. When the sustainability of • Supply- vs. demand-driven approaches. reforms was in doubt, they lacked credibility Donors eager to support reforms can easily to �rms, and so elicited a limited investment fall prey to supply-driven approaches. response.16 Indeed, the bookshelves of many ministries Drawing on this experience, new in developing countries are lined with approaches—among them the Comprehen- reports presenting detailed proposals for sive Development Framework—consider the design and implementation of reforms that the way aid is delivered can be as impor- never implemented. To counter this ten- tant as the content of aid in determining its dency, more donors are testing demand for effectiveness. More emphasis is being placed assistance and requiring bene�ciaries to on ensuring country ownership and engag- co�nance or otherwise provide evidence of ing a broader group of actors in society to serious commitment to reform. help build consensus for better policy • Specialist expertise and scale. Technical approaches17—processes especially impor- assistance on many investment climate tant to the effectiveness and sustainability of issues involves the mobilization of exper- investment climate improvements. As part of tise on highly specialized topics—from this effort, the consultative processes that the design of land registries and corpo- have proven effective in supporting invest- rate governance regimes to the regulation ment climate improvements (chapter 3) are of ports. Many technical assistance pro- increasingly being integrated into the process jects in this area are also relatively small of formulating and implementing Poverty in size, averaging $1.1 million each Reduction Strategies and donor Country between 1998 and 2002. Both factors can Assistance Strategies.18 Further progress in increase the design and supervision costs this direction holds great promise. of technical assistance projects relative to other forms of assistance. Technical assistance. Technical assistance • Institutional �t. When recommending the can be one of the most potent ways of help- design of particular policy frameworks or ing governments improve their investment regulatory regimes, too little emphasis can climates. In its many forms it can help cat- be given to questions of institutional �t— (c) The International Bank for Reconstruction and Development / The World Bank How the international community can help 193 ensuring that proposals are well adapted to local conditions. Indeed, advisers from BOX 10.1 Multidonor technical assistance facilities and donor countries very often propose solu- the investment climate tions that bear a striking resemblance to Multidonor facilities for technical assistance the Netherlands, Norway, Sweden, Switzer- those adopted in their home country— leverage resources and expertise and facilitate land, the United Kingdom, United States, regardless of where they are recommend- learning by participating donors, especially United Nations Development Programme ing them.19 It may be understandable that important where the frontiers of knowledge (UNDP), and the World Bank. advisers will be influenced by the are moving quickly, as with the investment By the end of March 2004, PPIAF had climate.They can bring specialist expertise to provided grants of more than $70 million approaches they are most familiar with, bear on the design and implementation of for 310 activities in 88 countries. It but the uncritical transplant of models projects.They can also reduce sensitivities supported the drafting of 32 sets of laws from other countries can lead to poor or associated with technical advice coming from and regulations, the execution of 45 trans- perverse results (chapter 2). donor governments whose �rms may have an actions, the formulation of 14 sector reform interest in any resulting commercial opportu- strategies, the establishment or strengthen- nities.Examples in the investment climate ing of 28 institutions, and the training of One practical response to all three chal- area include the Foreign Investment Advisory more than 1,500 regulators and of�cials. It lenges is to rely more heavily on multidonor Service, the Global Corporate Governance also supported 80 international and technical assistance facilities, which already Forum, and the Public–Private Infrastructure national workshops with over 9,000 partici- Advisory Facility (PPIAF). pants, along with the preparation of numer- play an important role in several areas of the PPIAF, established in 1999, illustrates the ous toolkits and case studies to assist in the investment climate (box 10.1). Concerns approach. PPIAF aims to improve the quality dissemination of emerging lessons of expe- about institutional �t can also be addressed by of infrastructure in developing countries rience.To ensure that assistance is demand- through private sector involvement. Its driven, PPIAF requires recipients of country- expanding the analysis and dissemination of main products include technical advice, speci�c assistance to provide some alternative policy approaches with an empha- capacity building, and the identi�cation and co�nancing or other credible evidence of sis on underlying design principles and trade- dissemination of good practices. Participat- commitment to the project. offs. Ensuring that advisors engage effectively ing donors include the Asian Development Bank, Canada, France, Germany, Italy, Japan, Source: PPIAF (2003) and World Bank staff. with local stakeholders who would be involved in administering and complying with the policy framework can also help. End- ing the tying of aid for technical assistance munity’s role in supporting public sector could also play a role by expanding the pool of investment, particularly with shifting bound- expertise available and reducing concerns that aries between public and private provision of advice might be tainted by the commercial a range of services. interests of �rms from the donor country.20 Experience shows that when governments create an effective policy and regulatory envi- Public investment in infrastructure. Public ronment, many infrastructure services can be investment in infrastructure can improve provided better by private �rms. Indeed, the investment climate, and the interna- engaging private participation in infrastruc- tional development community has long ture provision has been an important part of been an important source of external investment climate improvement strategies �nancing for these investments. Support in in most countries. The international develop- this area accounted for an average of $11.4 ment community thus has to ensure that pro- billion per year from 1998 to 2002, or posed public investments complement rather around 54 percent of development assis- than distract attention from efforts to create a tance for the investment climate. better investment climate for infrastructure To deliver sustainable bene�ts, however, providers. Although the appetite for private investments have to be made in the context of infrastructure investment in developing a sound policy framework—often dif�cult countries has fallen from its peak in the late when the government is both the regulator 1990s, the challenge of striking an appropri- and the service provider. Reflecting this, ate balance remains, particularly for telecom- strategies for improving infrastructure are munications, ports, and power supply. shifting away from an exclusive focus on pub- lic sector providers to creating an effective Supporting �rms and transactions investment climate for commercial providers In addition to helping governments improve of those services (chapter 6). This has impor- their investment climates, bilateral and mul- tant implications for the international com- tilateral agencies provide substantial support (c) The International Bank for Reconstruction and Development / The World Bank 194 WORLD DEVELOPMENT REPORT 2005 directly to �rms and transactions. When tionate burdens in a poor investment cli- that support takes the form of grants or mate, and have more dif�culty obtaining concessional loans, which is often the case access to �nance. with schemes aimed at helping smaller The second debate is whether the support �rms, it is treated as a form of development provided is actually cost-effective. As dis- assistance. Far more substantial support is cussed in chapter 8, schemes aimed at provid- provided to the execution of particular ing special support to small �rms have tended transactions through commercial loans and to have disappointing results. The �rst gener- guarantees that are not treated as forms of ation of schemes for delivering business development assistance. How might both development services used substantial donor types of direct support contribute to invest- funds with little impact. Newer, more market- ment climate improvements? friendly approaches might avoid some of these pitfalls, but have not yet been fully eval- Development assistance to support smaller uated. The provision of subsidized or directed �rms. The main forms of support to �rms credit for small �rms has also had mixed funded from development assistance are results, whether implemented by national �nancial services (lines of credit or micro�- governments or by international donors.21 nance) and business development services, Schemes tend to be dif�cult to sustain, retard both of which tend to be directed to small the development of credit markets, and crowd �rms and microenterprises. Bilateral and out commercial providers. That is why gov- multilateral agencies provided an average of ernments (and donors) are shifting their $3.1 billion a year for this support from emphasis from the provision of �nancial ser- 1998 to 2002—or more than the amount vices to creating a better investment climate allocated to technical assistance to improve for commercial providers of these services the broader investment climate. (chapter 6). This applies to schemes directed There are two main debates in this area. to small �rms in the formal sector, and The �rst is whether small �rms in the for- increasingly even to micro�nance (box 10.2). mal economy merit special attention in this The guidelines suggested in chapter 8 way. As discussed in chapter 3, while many for selective interventions by governments of the bolder claims linking small �rms to are equally applicable to schemes funded economic growth are dif�cult to substanti- by donors and international agencies: ate, small �rms do tend to face dispropor- Have a clear objective and rationale, focus on the sources of the problems rather than the symptoms, match the instrument to BOX 10.2 Knowing when to stop: UNDP’s micro�nance the rationale, impose discipline, be trans- activities in Bangladesh parent, and review regularly. Micro�nance is important for poor house- programs supported by the government, Other support provided to �rms. Developed holds and entrepreneurs. Early ventures, NGOs, and donors. subsidized by governments, donors, and Between 1996 and 2001, UNDP countries and international agencies pro- NGOs, provided important demonstration Bangladesh implemented 14 vide substantial support to �rms and trans- effects.There is now a growing consensus empowerment projects with micro�nance actions on nonconcessional terms that is that for micro�nance to be sustainable and components, with interest rates set without achieve its full potential, it needs to be com- regard to �nancial sustainability.The not regarded as development assistance. mercial.This means that donors and NGOs schemes attracted many clients, at the This includes private sector lending by need to be careful not to unwittingly hinder expense of other micro�nance providers. international �nancial institutions and development of a commercial market. When this and other problems in the pro- loans and guarantees provided by national UNDP’s experience in Bangladesh gram’s administration were revealed epitomizes a donor knowing when to stop. through a review in 2002, UNDP Bangladesh export credit agencies. Support of this kind The micro�nance sector in Bangladesh took swift action to close down all 14 pro- averaged $26.4 billion per year between serves more than 10 million clients in jects. Shutting down programs is not easy 1998 and 2002. While there are dif�culties roughly 70 percent of poor households. Of for donors, but UNDP Bangladesh demon- comparing the value of grants, conces- the many micro�nance institutions in the strated that good donor practice often country, only the two largest are fully self- demands such decisions. sional loans, nonconcessional loans, and suf�cient. Most of the others are small, guarantees, the nominal value of this sup- highly subsidized, and poorly performing Source: Brusky (2003). port was nearly ten times the amount of development assistance provided for tech- (c) The International Bank for Reconstruction and Development / The World Bank How the international community can help 195 nical assistance for investment climate acterizations of “institutions.� The emphasis improvements. is moving to understanding the different Support in this area is often justi�ed by experiences between and within countries de�ciencies in the investment climates in and how various factors influence the perfor- developing countries—and by the bene�ts mance of different types of �rms—a chal- that investment can bring through jobs and lenge, given the lack of comparable data on the transfer of expertise and technology. Such key microeconomic measures. support can also complement broader invest- There is progress. New instruments— ment climate improvements when it helps to including the ones drawn on in this Report— mobilize a supply response and tests and quantify an increasing range of costs, risks, demonstrates investment climate improve- and barriers facing �rms. A wider range of ments. Transactions that support the provi- policy areas and corresponding institutions is sion of better infrastructure, �nancial, or being examined to understand their impact education services can also contribute on incentives. New �rm-level data are pro- directly to investment climate improvements. viding fresh insights into �rm dynamics. The criteria applied when providing this Early results from this work are encouraging, support vary among agencies. In addition to suggesting great promise for ongoing work in commercial criteria, they typically focus on this direction. But a huge knowledge agenda the direct impact of the project on the local lies ahead, and warrants priority attention as economy and, in the case of bilateral agen- an integral part of efforts to accelerate and cies—which account for almost two-thirds of broaden improvements to investment cli- this support—the bene�ts to their national mates in developing countries. �rms.22 One way to strengthen the develop- ment impact of this support is to more fully Better data recognize the potential contribution to the Analysis, understanding, and appropriate broader investment climate. For example, policy responses depend �rst on reliable transactions can be used to create precedents information. Yet policymakers in developing for applying transparent competitive bidding countries are often operating in the dark arrangements, for clarifying policy frame- when it comes to their investment climates. works, and for supporting a more competi- There are opportunities to help in three main tive business environment. International areas: national statistics, cross-country data, �nancial institutions and a growing number and synergies with poverty assessments. of bilateral agencies apply tests of this kind, but there is scope for broadening the National statistics. As noted in chapter 3, approach. substantial efforts are required to improve national statistical systems including on such basic measures as the share of private invest- Tackling the substantial ment in GDP. Work on building the capacity knowledge agenda of statistical agencies in developing countries The last 50 years saw tremendous progress in has increased in recent years, including our understanding of macroeconomics, both through multipartner initiatives.23 These and theoretically and empirically. Consensus is related initiatives need to give due weight to growing on macroeconomic indicators that investment climate issues to help govern- indicate the health of the economy. The chal- ments monitor the performance of their pri- lenge now is to make similar progress on the vate sectors, identify emerging trends and microeconomic determinants of economic problems, and evaluate the impact of alterna- performance—to provide practical guidance tive policy approaches. to policymakers. Understanding the microeconomic deter- Cross-country data. The international com- minants of growth and productivity has gar- munity is well placed to develop more stan- nered much interest in recent years. There is dardized measures of the investment climate growing acknowledgment of the limits of to facilitate comparisons across countries. cross-country regressions and generic char- Recent developments in quantifying many (c) The International Bank for Reconstruction and Development / The World Bank 196 WORLD DEVELOPMENT REPORT 2005 aspects of the investment climate—including to focus policy discussions only on topics or the Bank’s Investment Climate Surveys and measures most easily quanti�ed. There is the the Doing Business Project—are important old joke about the person looking for his lost steps forward in understanding how invest- keys under the streetlight, not because he ment climate policies and behaviors influ- dropped them there, but because that was ence growth and poverty. In addition to where he could see. informing analysis, these measures provide a useful tool for governments to benchmark A large research agenda performance and monitor progress. They can A �eld as broad as the investment climate also act as catalysts for reform. generates a huge research agenda, but four Building up a body of consistent data over larger themes warrant close attention. time can provide insights into the critical links between policy settings and growth The ecology of �rms and growth processes. processes. Many of the measures are new, There is growing understanding of the however, so changes from a baseline cannot processes by which �rms are born and yet be measured. As they build up over time, evolve, including creative destruction, based their explanatory power will increase. Of par- mostly on experience in developed coun- ticular promise is the ability to test more rig- tries. Early research provides insights into orously the impact of different policy how similar processes play out in develop- approaches. Being able to better evaluate the ing countries. But there is a need to deepen impact of policies should encourage more and broaden understanding of these experimentation and competition between dynamics, including the important role of approaches. Evaluations of pilot programs �rms in the informal and rural economies, can identify the ones succeeding—the ones to and the impact of international economic be scaled up. integration. There are also bene�ts in expanding the coverage of these data across several The design of regulatory strategies. Regula- dimensions: tion plays a central role in addressing mar- ket failures, reconciling the interests of • To include the impact on a broader �rms with broader social goals, and shaping range of �rms, including those in the the investment climate. To date most of our informal and rural economies understanding of regulatory policies and • To include the impact on particular sec- strategies is based on experience in devel- tors and on particular supply chains oped countries, and most of that work • To grapple with the measurement of focuses on regulation within relatively nar- critical but hard-to-quantify variables, row �elds, such as infrastructure, �nance, such as policy uncertainty and competi- product safety, or the labor market. Much tive pressure. less attention has been given to how regula- tory strategies might be tailored to different Synergies with poverty assessments. There institutional environments, particularly are opportunities to build synergies those in low-income countries. There are between approaches for assessing invest- also opportunities to explore lessons of ment climates and for assessing poverty. For experience on this question that cut across example, questions on access to infrastruc- �elds of regulation. ture and �nance, and the security of prop- erty rights can be included more systemati- The linkages between the investment climate cally in household surveys. The sampling and migration. The quality of a country’s strategies for household and �rm surveys investment climate not only affects flows of might also be linked. capital—it can influence flows of people, Better data of this kind can advance too. The movement can be from rural areas understanding of many key areas of eco- to urban, from one city to another, or from nomic policy. But care needs to be taken not one country to another. Today the world’s (c) The International Bank for Reconstruction and Development / The World Bank How the international community can help 197 migrants from developing countries total Beyond data and formal research nearly 175 million.24 The $90 billion or Even within the bounds of current data more in remittances they send to their fami- sources, there are opportunities to lies every year is now the second largest advance understanding of many areas of source of private capital (after FDI) for poor the investment climate. Country studies countries.25 Understanding the linkages can help illuminate many important between investment climate conditions and design and implementation details that migration flows will become more impor- remain beyond the reach of cross-country tant as the world deals with major demo- analyses. Country studies can also include graphic shifts over the coming decades. more rigorous evaluations of recent policy experiments to understand their impacts The political economy of investment cli- on �rm performance, productivity, growth, mate improvements. This Report has high- and poverty. There are also opportunities lighted the importance of understanding to expand recent efforts to identify and the political economy considerations that disseminate emerging lessons of experi- influence investment climate policies. ence in the design and implementation of While the subject has attracted signi�cant investment climate improvements. This attention, little is known about the condi- can help policymakers understand the tions under which governments choose to rich menu of options they can choose pursue sound policies in these areas, from in a �eld as broad—and central—as including the implications of alternative the investment climate. political structures and processes. There are Working together on these themes, the also opportunities to deepen understand- international community can do much to ing on strategies for controlling rent-seek- create a better investment climate—for ing and on the dynamics of reform everyone—and so contribute to a more bal- processes more generally. anced, inclusive, and stable world. (c) The International Bank for Reconstruction and Development / The World Bank Bibliographical note This Report draws on a wide range of World Bank documents Patrick Honohan, Robert Hornick, Catherine Hunt, Giuseppe and on numerous outside sources. Background papers, case Iarossi, Gregory Ingram, Naoko Ishii, Roumeen Islam, John studies and notes were prepared by Graham Bannock, Eric Besant Jones, Marc Juhel, William Kalema, Daniel Kaufmann, Bartelsman, Anthony Burns, Charles Byaruhanga, Martha Philip Keefer, Christine Kessides, Mumtaz Hassan Khaleque, Chen, David Christianson, Jacqueline Coolidge, Ramon Anupam Khanna, Homi Kharas, R. Shyam Khemani, Sunita Clarete, Luc De Wulf, Juan Carlos Echeverry Victor Endo, Saul Kikeri, Stephen Knack, Mihaly Kopanyi, Peter Lanjouw, Eliza- Estrin, Simon Evenett, David Finnegan, Peter Fortune, Marco beth Little�eld, Lili Liu, Frannie Leautier, Danny Leipziger, Galmarini, Peter Holmes, Matthew Gamser, Johana Gil Anat Lewin, Syed Mahmood, William Maloney, Ali Mansoor, Hubert, Lars Grava, John Haltiwanger, Ashley Hubka, David Jean-Michel Marchat, Marie-Françoise Marie-Nelly, Keith Irwin, Renana Jhabvala, Mariell Juhlin, Stepan Jurajda, Shamin Maskus, Aaditya Mattoo, Caralee McLiesh, Taye Mengistae, Khan, Sheng Lei, Frances Lund, Raymond Mallon, Katarína Pradeep Mitra, Andrew Morrison, Fergus Murphy, Mamta Mathernová, Gregor Mackinnon, Pradeep Mehta, Klaus Murthi, Mohammad Mustapha, Mustapha Kamel Nabli, John Meyer, Stefano Migliorisi, Winnie Mitullah, Reema Nanavaty, Nasir, Roger Nellist, Richard Newfarmer, Francis Ng, Paul Camilo Navarrete, Anders Olofsgard, Gaelle Pierre, John Pre- Noumba Um, Anders Olofsgard, Jacques Ould-Aoudia, Anto- ston, Amit Ray, Fernando José Salas, Cerstin Sander, Mauricio nio Parra, Guillermo Perry, Axel Peuker, Gaelle Pierre, Miria Santa María, Caroline Skinner, Aleksander Surdej, Sanda Pigato, Tony Polatajko, Sanjay Pradhan, Christine Zhen-Wei Utnina, Dirk Willem te Velde, Brendan Vickers, Elisha Qiang, Brice Quesnel, Firas Raad, Vijaya Ramachandran, Mar- Wasukira, Simon White, Yao Yu, Pu Yufei, and Rita Zaidi. tin Ravallion, Francesca Recanatini, Gerry Rice, Neil Roger, Background papers for the Report are available either on Christian Rogg, Etienne Rolland-Piegue, Jan Rutkowski, Cer- the World Wide Web http://econ.worldbank.org/wdr/ stin Sander, Cecilia Sager, Jamal Saghir, Amartya Sen, Maurice wdr2005/library/ or through the World Development Report Schiff, Robert Schware, Luis Serven, Anne Simpson, Ken of�ce. The views expressed in these papers are not necessarily Sokoloff, Nicholas Stern, Margrete Stevens, Andrew Stone, those of the World Bank or of this Report. Gary Stuggins, Uma Subramanian, Victor Sulla, Gwen Swin- Many people inside and outside the World Bank gave com- burn, Vito Tanzi, Giovanni Tanzillo, Dirk Willem te Velde, ments to the team. Valuable comments and contributions were Francis Teal, Simon Thomas, Nigel Twose, Marilou Uy, Rudolf provided by Alberto Agbonyitor, Daron Acemoglu, Sadiq V. Van Puymbroeck, Dirk Willem te Velde, Milan Vodopivec, Ahmed, Asya Akhlaque, Ian Alexander, Zoubida Allaoua, Shuilin Wang, John Wilson, Peter Woicke, James D. Wolfen- Magdi Amin, Paul Amos, Jim Anderson, Jock Anderson, Doug sohn, Colin Xu, and Tarik Yousef. Andrew, Yaw Ansu, Andreas Antoniou, Robert Bacon, Joseph Other valuable assistance was provided by Jean-Pierre Battat, Simon Bell, Najy Benhassine, Philip Benoit, Lorenzo Djomalieu, Endy Djonokusumo, Jocelyn Dytang, Ines Bertolini, Subhash Bhatnagar, Freddy Bob-Jones, Milan Garcia-Thoumi, Gytis Kanchas, Jimena Luna, Polly Means, Brahmbhatt, David Bridgeman, Harry Broadman, Penelope Nacer Mohamed Megherbi, Christopher Neal, and Jean Gray Brook, Jose Edgardo Campos, Gerry Caprio, Mita Ponchamni. Chakraborty, Vandana Chandra, Shaohua Chen, Pascale The team wishes to thank the many individuals who partic- Marie-Claude Chabrillat, William Cobbett, Louise Cord, ipated in workshops held in Berlin, Dar-es-Salaam, London, Robert Cull, Angus Deaton, Asli Demirguc-Kunt, Jean-Jacques New Delhi, Shanghai, and Washington D.C.; in videoconfer- Dethier, Simeon Djankov, Antonio Estache, Marcel Fafchamps, ences with Brazil, Egypt, Guatemala, Honduras, Japan, Pablo Fajnzylber, Shahrokh Fardoust, Edgardo M. Favaro, Lebanon, Nicaragua, Russia, and Serbia and Montenegro; and Alexander Fleming, Olivier Floris, Francis Fo, Ricardo Fuentes, in on-line discussions of the draft report. The World Bank’s Ahmed Galal, Matthew Gamser, Sushma Ganguly, Alan Gelb, external affairs staff around the world provided valuable sup- Coralie Gevers, Indermit Gill, Sylvie Gregoire, Hannes Holm- port to the consultation process. steinn Gissurarson, Judith Goans, Ian Goldin, Susan Gold- Despite efforts to compile a comprehensive list, some who mark, Carlos Gomez, Charles Grif�n, Pierre Guislain, Naomi contributed may have been inadvertently omitted. The team Halewood, Jonathan Halpern, Kristin Hallberg, Clive Harris, apologizes for any oversights and reiterates its gratitude to all Syed Hashemi, Joel Hellman, Rasmus Heltberg, John Hodges, who contributed to this Report. 198 (c) The International Bank for Reconstruction and Development / The World Bank Endnotes Overview 11. Maddison (1995). 1. For further details on these data sources, see Box 1.2 and 12. De Long (2000). In terms of pure purchasing parity, and Tables A1 and A2 in the Selected Indicators section at the back of looking at how large a bundle of 1900 era goods could be bought the book. with today’s incomes, Thailand’s per capita income is 50 percent 2. See Box 1.5. greater than that of the U.S. in 1900, and Mexico and Uruguay not 3. Johnson, McMillan, and Woodruff (2002b). quite double it. 4. Hall and Jones (1999); Parente and Prescott (2000); Easterly 13. Pritchett (1997). and Levine (2001); and Bosworth and Collins (2003). 14. Pritchett (2002) and Hausmann and Rodrik (2003). 5. Schumpeter (1942). 15. Easterly (2001); Aghion and Durlauf (2004); and Haus- 6. See Figure 1.13. mann, Pritchett, and Rodrik (2004). 7. OECD (2002b); Carlson and Payne (2003). 16. Tanzi and Davoodi (1998). 8. Dollar, Hallward-Driemeier, and Mengistae (2003a). 17. Sala-i-Martin and Vila-Artadi (2002) and Easterly, Devara- 9. Minot and Goletti (2000) and Winters, McCulloch, and jan, and Pack (2001). McKay (2004). 18. Bosworth and Collins (2003). 10. Hoekman, Kee, and Olarreaga (2001). 19. Krugman (1997). 11. See Box 6.12. 20. Barro and Sala-i-Martin (2003); Bosworth and Collins 12. Field (2002). (2003); Easterly and Levine (2001); Hall and Jones (1999); Klenow 13. See Figure 1.18. and Rodríguez-Clare (1997); and Young (1995). 14. World Bank (2004b). 21. Solow (1957); Jones (2002); and Barro and Sala-i-Martin 15. See Figure 1.4. (2003). More recent work still acknowledges the importance of 16. See Table 2.1. technology, but broadens the view of total factor productivity to 17. World Economic Forum (2004). include concepts of institutions and social capital, concepts closely 18. Londoño and Guerrero (2000). related to the investment climate. See Hall and Jones (1999) and 19. Box 5.13. Acemoglu and Johnson (2003). 20. Bartelsman and others (2004). 22. Jovanovic (1995). 21. World Bank (2004d). 23. Parente and Prescott (2000). 22. Migliorisi and Galmarini (2004). 24. Hall and Jones (1999) and Acemoglu, Johnson, and Robin- son (2001). Chapter 1 25. Hicks (1935). 1. Chermak (1992). 26. Baumol (2002). 2. Acemoglu, Johnson, and Robinson (2002); Levine (1997); 27. Schumpeter (1942). Kaufmann, Kraay, and Mastruzzi (2003); La Porta and others 28. A similar result was found in Carlin and others (2001) using (1999); Glaeser and Shleifer (2002); and Glaeser and others (2004). an earlier round of survey data in the same countries. See also Bas- 3. Acemoglu and Johnson (2003); Rodrik, Subramanian, and tos and Nasir (2003). Trebbi (2002); and Knack and Keefer (1995a). 29. Haltiwanger (2000) and Bartelsman, Scarpetta, and 4. Burgess and Venables (2003). Schivardi (2003). 5. Pritchett (2004). 30. Scarpetta and Bartelsman (2003). 6. World Bank (2004b). 31. Potential endogeneity is controlled for using lagged values. 7. Erb, Harvey, and Viskanta (2000). 32. Desai and Mitra (2004) and Caballero, Engel, and Micco 8. Calculated from the World Business Environment Survey data. (2004). 9. World Bank (2004b). 33. World Bank (2002d). 10. Malthus (1798). 34. Kraay (2003). 199 (c) The International Bank for Reconstruction and Development / The World Bank 200 WORLD DEVELOPMENT REPORT 2005 35. While the overall trends are undisputed, the exact levels of 12. Klitgaard (1998). poverty are a matter of some debate due to differences in method- 13. Van Rijckeghem and Weder (2001). ologies in calculating them. For example, household surveys or 14. Laffont and Tirole (1991). national accounts, expenditure or consumption-based measures, 15. Hellman and others (1999). and the challenge of measuring nonmonetary transactions. See 16. Wintrobe (1998). Chen and Ravallion (2004); Ravallion (2003a); and Deaton (2002). 17. Robinson and Verdier (2002); Robinson (1998); Herbst 36. Bourguignon (2004). (2000); and Bates (1981). 37. Dollar and Kraay (2002). 18. On poorly de�ned property rights, see Barzel (2002); on red 38. Bourguignon (2004). tape, see De Soto (2000); on labor markets, see Golden (1997); on 39. World Bank (2003m); Midlarsky (1999); and Fearon and �nance, see Rajan and Zingales (2003); on infrastructure, see World Laitin (2003). Bank (2003p). 40. Fields and Pfeffermann (2003). 19. Patronage and clientelism are often associated with “per- 41. OECD (2002b); Carlson and Payne (2003); and India sonal voting,� whereby promises made by policymakers may only National Sample Survey Organisation. be credible to groups with whom they have personal relationships: 42. World Bank (2004i). Keefer (2002). 43. Dollar, Hallward-Driemeier, and Mengistae (2003a) and 20. See, for example, Morris and Shepsle (1990) and Keefer and Hallward-Driemeier, Iarossi, and Sokoloff (2002). Khemani (2003). 44. Lanjouw and Stern (1998). 21. Khemani (2004) and Desai and Olofsgård (2003). 45. United Nations (2002b). 22. Faccio (2003). 46. ILO (2002b). 23. Fisman (2001). 47. Schneider (2002); Chen, Jhabvala, and Lund (2002); 24. Hellman and Kaufmann (2003). Charmes (2000); Mead and Liedholm (1998); Jhabvala, Sudarshan, 25. For a review of the effects of political �nance on state cap- and Unni (2004); and ILO (2002b). ture, see Kaufmann (2002). For evidence of efforts to reform cam- 48. Minot and Goletti (2000). paign �nance laws in transition economies, see World Bank 49. Winters, McCulloch, and McKay (2004). (2000c). 50. Despite these bene�ts for poor people, many countries 26. Keefer (2002) and Keefer (2003). restrict the sale of second-hand clothing. See Dougherty (2004) 27. Mukherjee (2002) and World Bank (2000a). and Tranberg Hansen (2000). 28. Dixit and Pindyck (1994). The option theory of investment 51. Hoekman, Kee, and Olarreaga (2001). Country-level studies highlights how uncertainty raises the threshold value a project provide additional evidence that greater competitive discipline must meet before �rms will be willing to commit due to the loss of reduced markups in Chile, Colombia, Côte d’Ivoire, Mexico, the option of waiting. However, uncertainty does not necessarily Morocco, and Turkey. See Roberts and Tybout (1996); Harrison always decrease investment. Uncertainty that raises the probability (1994); and Levinsohn (1993). of a bad outcome will lower the expected bene�ts. But if increased 52. ESMAP (2002). uncertainty rises with the marginal revenue product of the invest- 53. Field (2002). ment, then the expected pro�tability can increase. See Serven 54. Palmade (2004). For further discussion of taxes in develop- (1997) and Caballero (1991). ing countries see chapter 5. 29. Pattillo (1998) and Darku (2001). 55. World Bank (2003p). 30. Smith (1997a). 31. Gaston and Wei (2002). Chapter 2 32. Keynes (1936). 1. Appearing before the U.S. Senate Armed Services Committee 33. Thaler (1993); Thaler (2000); Rabin (1998); Kagel and Roth in 1953, Wilson—a former president of GM—actually said, “[F]or (1995); Camerer, Loewenstein, and Rabin (2003); and Kahneman years I have thought that what was good for our country was good and Tversky (2000). for General Motors and vice versa.� 34. Henisz and Delios (2003) review investment patterns by 2. Litvin (2003). Japanese �rms in 49 countries and �nd that market entry strategies 3. Hufbauer and Goodrich (2003a) and Hufbauer and Goodrich were heavily influenced by perceptions of policy uncertainty. (2003b). 35. Hnatkovska and Loayza (2004) and Ramey and Ramey 4. Rajan and Zingales (2003) cite examples in Mexico, Brazil, (1995). and Japan. 36. Svensson (1998); Paunovic (2000); and Rodrik (1991). 5. Stigler (1971) and Peltzman (1976). 37. North (1993). 6. Sutton (1976). 38. North and Weingast (1989). 7. World Bank (1997). 39. See, for example, Henisz (2000); Stasavage (2002); and 8. Murphy, Shleifer, and Vishny (1993). Falaschetti (2003). 9. Klitgaard (2000). 40. Cukierman (1992); Majone (1996); and Levy and Spiller (1994). 10. Reinikka and Svensson (1999) and Transparency Interna- 41. Olofsgård (2004). Also see McCubbins and Lupia (1998). tional Bangladesh (2002). 42. Haber, Razo, and Maurer (2003). 11. Adserà, Boix, and Payne (2003). 43. Perkins (2000). (c) The International Bank for Reconstruction and Development / The World Bank Endnotes 201 44. On the various de�nitions of trust and social capital, see 25. Christianson (2004). Fukuyama (2001); Coleman (1988); and Putnam, Leonardi, and 26. Despite the attention to this issue, credit is not always the Nanetti (1993). binding constraint on small and medium enterprises: McMillan 45. It is also not inevitable that communal, family- or kinship- and Woodruff (2002). based relationships will instill trust and create the basis for richer 27. Baldwin (2003) provides a recent survey of the large litera- civic-associational life. In post-World War II Italy, citizens were ture on growth and openness. unwilling to coordinate in establishing businesses, schools, hospi- 28. Aitken, Hanson, and Harrison (1997). tals, or other voluntary organizations. Organized life tended to 29. The copious literature on this topic is summarized in Tybout depend on the initiative of centralized, distant authority: the (2003) and Keller (2001). church and the state. Ban�eld (1958); Piore and Sabel (1984); and 30. Hoekman, Kee, and Olarreaga (2001); Roberts and Tybout Fukuyama (1995). (1996); Harrison (1994); and Levinsohn (1993). 46. Olivier de Sardan (1999). 31. For a recent illustration see Hufbauer and Goodrich (2003b). 47. Alesina and others (2003b). 32. Bayoumi, Coe, and Helpman (1999) and de Ferranti and 48. Henisz and Zelner (forthcoming) and Kubler (2001). others (2003). 49. Kay (2003). 33. Tybout (2003). 50. For experience in the transition economies, see Center for 34. Fernandes (2003) for Colombia; Muendler (2002) for Brazil; Policy Studies (2003) and The Economist (2003b). and Pavcnik (2003) for Chile. Using industry-level data for India, 51. The Pew Global Attitudes Project (2003). Aghion and others (2003) �nd similar results for India. 52. Muller (2002). 35. Fernandes (2003). 53. Examples include proposals to abolish joint-stock compa- 36. Aghion and others (2003). nies in England after the bankruptcies of 1860s and to nationalize 37. Hu and Jefferson (2002); Aitken and Harrison (1999); and large parts of corporate American by the New Dealers, see Mick- Djankov and Murrell (2002). lethwait and Wooldridge (2003). 38. Blalock and Gertler (2003); Smarzynska (2002); and Kugler 54. Micklethwait and Wooldridge (2003). (2001). 55. Stiglitz (1989). 39. Görg and Strobl (2001) and Blomström and Kokko (1998). 56. OECD (2003g). 40. For countries with low skill levels, the returns from acquir- 57. Stiglitz (1999b). ing foreign technology are higher. In India returns from technology purchase are 44 times higher than the rate of return of domestic Chapter 3 R&D; in scienti�c sectors the rate of return for technology pur- 1. Maloney (2004). chase is 166 percent, but for domestic R&D the rate is 1 percent. See 2. Johnson, McMillan, and Woodruff (2002b). Basant and Fikkert (1996). 3. Reid and Gatrell (2003). 41. Coe and Helpman (1995) and Coe, Helpman, and Hoffmais- 4. Rodrik and Subramanian (2004). ter (1997). 5. Cao, Qian, and Weingast (1999). 42. The level of absorption may depend on human capital and 6. Saavedra (2003). See also Blanchard and Giavazzi (2003); the stock of domestic R&D. See, for example, Crespo, Martin, and Bineswaree and Freund (2004); and Klapper, Laeven, and Rajan Velazquez (2002). (2003). 43. Lederman and Maloney (2003). 7. World Economic Forum (2004). 44. UNCTAD (2003c) and de Ferranti and others (2003). 8. World Bank (2004b). 45. Fernandez and Rodrik (1991). 9. World Bank (2002c) and Commonwealth Secretariat (2003). 46. Krueger (2000); Tommasi (2002); and Hausmann, Pritchett, 10. Blanchard and Giavazzi (2003). and Rodrik (2004). 11. ILO (2002b). 47. Woo-Cumings (2001). 12. Chen, Jhabvala, and Lund (2002). 48. Parente and Prescott (2000). 13. ILO (2002b) and Ellis (2003). See also box 1.12. 49. Berglof and Roland (2000). 14. Galal (2004). 50. Cao, Qian, and Weingast (1999) and Yeung (2003). 15. Hallward-Driemeier and Stone (2004). 51. North and Weingast (1989). 16. Mitullah (2004); Lund and Skinner (2004); and Chen, Jhab- 52. Boix (2003). vala, and Lund (2002). 53. Kingdon (1995). 17. Field (2002) and Bannock and others (2004). 54. Consumers International (www.consumersinternational.org) 18. Stern, Dethier, and Rogers (2004). and Transparency International (www.transparency.org). 19. Lanjouw and Stern (1998). 55. Cabanero-Verzosa and Mitchell (2003). 20. Reardon and others (1998). 56. Georgian Opinion Research Business International 21. Lanjouw and Shariff (1999). (GORBI) (2002) 22. Foster and Rosenzweig (2004). 57. Yufei and others (2004). 23. World Bank (2004k) and Chandra and Rajaratnam (2004). 58. Smith (1997a). 24. Similar results are found in EBRD (1999) and Batra, Kauf- 59. Sidak and Baumol (1995) and Hempling, Rose, and Burns mann, and Stone (2002). (2004). (c) The International Bank for Reconstruction and Development / The World Bank 202 WORLD DEVELOPMENT REPORT 2005 60. Kostrzeva (2003). and others (2004). 61. Wedel (2002). 42. Bigsten and others (2000) and Fafchamps and Minten 62. OECD (1999a). (2001). 63. Argy and Johnson (2003). 43. Djankov and others (2003b). 64. OECD (1999b) and OECD (2002c). 44. Djankov and others (2003b). 65. OECD (2000b) and OECD (2002c). 45. World Bank (2003f). 66. Kirkpatrick and Parker (2003) and Lee (2002). 46. Messick (1999) and Burki and Perry (1998). 67. Stanchev (2003). 47. Chengappa (1999). 68. Regobeth and Ahortor (2003). 48. On Tanzania, see Kahkonen and others (2001). On Bolivia, 69. Stanchev (2003). see Fleisig and De la Peña (2003). 70. Zeruolis (2003) and Vilpisauskas (2003). 49. Inter-American Development Bank (2002). 71. Environmental Resources Management (2004). 50. Ahmadi (1999). 51. UNCTAD (2003a). See also chapter 9. Chapter 4 52. World Bank (2003d). 1. Murdoch and Sandler (2002). 53. Volkov (2002). 2. World Bank (2003m). 54. Polinsky and Shavell (2000). 3. Johnson, McMillan, and Woodruff (2002b). 55. Stone and Ward (2000). 4. Besley (1995) and World Bank (2003f). 56. Schärf (2001). 5. Erb, Harvey, and Viskanta (2000). 57. Greenwood and others (1998) and Waller and Sanfacon 6. Torstensson (1994) and Knack and Keefer (1995b). (2000). 7. Keefer (2004). 58. McDonald (1994). 8. World Bank (2003n). 59. Sherman and others (1998). 9. Jimenez (1984). 60. Villadeces and others (2000) and Mockus (2002). 10. Lanjouw and Levy (2002). 61. Buvinic and Morrison (2000). 11. Friedman, Jimenez, and Mayo (1988). 62. Frye and Zhuravskaya (2000). 12. Feder and others (1988). 63. Rossiter (1961). 13. Do and Iyer (2003). 64. Mattei (2000) and Shavell (2004). 14. Base line data from Bank project cited in Baharoglu (2002). 65. Chifor (2002). 15. Feder and others (1988). 66. Vernon (1971). 16. Field (2002). 67. Wells Jr. and Gleason (1995). 17. World Bank (2003n). 68. West (2001). 18. Macours (2003). 69. For a review of strategies employed by investors in private 19. World Bank (2003n). infrastructure projects, see Smith (1997a). 20. Adlington (2002). 70. Van der Walt (1999). 21. Deininger (2002). 71. Stephenson (2003). 22. International Institute for Environment and Development (2001). Chapter 5 23. Siamwalla (1993) and Stan�eld and others (1990). 1. Coase (1960) and Pigou (1932). 24. United Nations–Habitat (2003) and Baharoglu (2002). 2. Acemoglu, Johnson, and Robinson (2001). 25. Yahya (2002) and Botswana–Ministry of Lands (2002). 3. Pistor and others (2003). 26. Fleisig and De la Peña (2003). 4. Pistor and others (2003). 27. IFC and CIDA (2001). 5. Pistor (2000). 28. World Bank (2002b). 6. Berkowitz, Pistor, and Richard (2003). 29. Greif (1989) and Fafchamps (2004). 7. Alesina and others (2003a) and Nicoletti and Scarpetta (2003). 30. Klein (1992). 8. Numbers based upon estimates presented in Alesina and oth- 31. World Bank (2003a) and Jappelli and Pagano (1999). ers (2003a). 32. Jappelli and Pagano (1999). 9. World Bank (2003a). 33. Fafchamps (2004). 10. Costs are median costs for each group, calculated for �rms 34. Milgrom, North, and Weingast (1990). that report applying for basic activity license in the past three years. 35. Mnookin and Kornhauser (1979). Data is from the Investment Climate Survey for Tanzania. 36. Galanter and Krishnan (2003). 11. World Bank (2003g). 37. Williamson (1996). 12. Schneider (2002). 38. Cristini and Moya (2001) and Castelar-Pinheiro and Cabral 13. Flores and Mikhnew (2004). (2001). 14. OECD (2003a). 39. Laevan and Woodruff (2003). 15. World Bank (2004b). 40. Castelar-Pinheiro (1998); Sereno, de Dios, and Capuano 16. Winston (1993) and OECD (1997b). (2001); and Herrero and Henderson (2001). 17. Guasch and Hahn (1999) and Guasch and Spiller (1999) 41. Johnson, McMillan, and Woodruff (2002a) and Broadman summarize studies for developing countries. (c) The International Bank for Reconstruction and Development / The World Bank Endnotes 203 18. Fernandes (2003) for Colombia; Muendler (2002) for Brazil; ture, and its �nancing mechanisms. For further details, see Pavcnik (2003) for Chile; and Aghion and Burgess (2003) for India. www1.worldbank.org/publicsector/tax/autonomy.html. 19. Bartelsman and others (2004). The countries were Brazil, 47. Bird (2003). Chile, Colombia, Hungary, Latvia, Mexico, Romania, Slovenia, and 48. Taliercio Jr. (2003b). Venezuela. Increases exceeded 10 percent in Brazil, Chile, and Mex- 49. Taliercio Jr. (2003b). ico and exceeded 20 percent in Colombia, Hungary, and Venezuela. 50. Taliercio Jr. (2001). 20. Akiyama and others (2003). 51. Taliercio Jr. (2001); Taliercio Jr. (2003a); and Taliercio Jr. 21. Akiyama and others (2003). (2003b). 22. See, for example, McMillan, Rodrik, and Welch (2002). 52. See Das-Gupta, Engelschalk, and Mayville (1999) and Bird 23. Lawson and Meyenn (2000) describe the program. Data are (2003). from Grameen Telecom’s Web site (www.grameen-info.org/grameen/ 53. Das-Gupta, Engelschalk, and Mayville (1999). gtelecom/) from February 2004. 54. Taliercio Jr. (2003b). 24. La Porta and López-de-Silanes (2001). 55. Fjeldstad (2002) and World Bank (2004e). 25. World Bank (2004d). 56. Gill (2003) and Engelschalk, Melhem, and Weist (2000). 26. Evenett (2004). 57. Bird and Engelschalk (2003). 27.These are based upon the recommendations in UNCTAD (2003d). 58. Bird (2003); Engelschalk, Melhem, and Weist (2000); and 28. See www.internationalcompetitionnetwork.org, “Advocacy Bird and Engelschalk (2003). and Competition Policy� ICN Conference, Naples, Italy, 2002. 59. Gill (2003). 29. Khemani (2002) and Kovacic (1997). 60. Bird (2003). 30. Kee and Hoekman (2003). 61. APEC Committee on Trade and Investment (2003). 31. The Economist (2002c). 62. World Bank (2004d), Table 2.9. 32. The U.S. did, however, levy a temporary income tax in 1862 63. World Bank (2004d), Table 1.9. The gains are estimated to be during the civil war. See www.irs.ustreas.gov/. between $114 billion (in 1997 US$) and $265 billion depending 33. Ebrill and others (2001). upon assumptions about the dynamic effects. 34. Lewis (2004). 64. Dollar, Hallward-Driemeier, and Mengistae (2003a). 35. For a review of ef�ciency arguments in taxation, see Dia- 65. Engelschalk, Melhem, and Weist (2000) discuss computeriz- mond and Mirlees (1971); Stiglitz and Dasgupta (1971); Ebrill and ing customs and tax administration in greater detail. others (2001); Sandmo (1976); and Slemrod (1990). For a discus- 66. De Wulf (2003). sion of the pragmatic considerations driving tax system design in 67. De Wulf and Finateu (2002). developing countries, see Tanzi and Zee (2001). 36. In 2000/01, corporate income taxes accounted for 14 percent Chapter 6 of tax revenues in low-income countries, 12 percent of revenues in 1. Rajan and Zingales (2003). lower middle-income countries, and 9 percent of revenues in upper 2. Harris (2003); World Bank (1994b); World Bank (2004j); and middle-income countries. Direct taxes on goods and services World Bank (2003p). accounted for 41 percent, 42 percent, and 37 percent of revenues 3. King and Levine (1993); Levine, Loayza, and Beck (2000); respectively. Taxes on international trade accounted for 18 percent, Beck, Levine, and Loayza (2000); Bandiera and others (2000); and 14 percent and 8 percent of revenues respectively. Data are averages Demirgüç-Kunt and Maksimovic (1998). for 60 developing countries for which comparable data were avail- 4. Caprio and Honohan (2003). able and calculated based on data from IMF (2003); OECD 5. Li, Squire, and Zou (1998). (2002d); and Dobrinsky (2002). 6. Rajan and Zingales (2003). 37. Mitra and Stern (2003) discuss corporate tax revenues in the 7. Easterly, Islam, and Stiglitz (2000). transition economies. 8. Dehejia and Gatti (2002). 38. Gauthier and Gersovitz (1997) and Gauthier and Reinikka (2001). 9. Stiglitz and Rothschild (1976) and Stiglitz and Weiss (1981). 39. Taxes on goods and services increased on average as a per- 10. Rajan and Zingales (2003). cent of GDP between the mid-1990s and 2000-2001 among all 11. Barth, Caprio Jr., and Levine (2001); Clarke and Cull (2002); income groups, while taxes on international trade fell among all La Porta, López-de-Silanes, and Shleifer (2002); and Sapienza groups. Calculations were based upon data from IMF (2003); (2004). OECD (2002d); and Dobrinsky (2002). 12. Beck, Cull, and Afeikhena (2003); Beck, Crivelli, and Sum- 40. Elstrodt, Lenero, and Urdapilleta (2002). merhill (2003); and Omran (2003). 41. Djankov and others (2002). 13. Berger and others (forthcoming); Demirgüç-Kunt and Mak- 42. Ebrill and others (2001) show that the largest 10 percent of simovic (forthcoming); and Berger, Hasan, and Klapper (2004). �rms account for about 90 percent of turnover in Georgia, Pak- 14. Schreiner and Yaron (2001). istan, Sri Lanka, and Uganda. 15. Townsend and Yaron (2001). 43. Das-Gupta, Engelschalk, and Mayville (1999). 16. Harvey (1991) and World Bank (2001f). 44. Taliercio Jr. (2003b). 17. World Bank (1994a). 45. Bird and Engelschalk (2003) discuss this in greater detail. 18. Vittas and Je Cho (1995). 46. In practice, the autonomy of an agency depends upon many 19. Klapper and Zaidi (2004) and World Bank (1989). factors, including the agency’s legal position, its governance struc- 20. Caprio and Demirgüç-Kunt (1998) and World Bank (1989). (c) The International Bank for Reconstruction and Development / The World Bank 204 WORLD DEVELOPMENT REPORT 2005 21. In Nigeria, for example, 15 percent of guaranteed loans were 55. Chami, Khan, and Sharma (2003). reported to be in arrears; see Njoku and Obasi (1991). 56. Barth, Caprio Jr., and Levine (2004) and Beck, Demirgüç- 22. Management Systems International (1996) and Magno and Kunt, and Levine (2003). Meyer (1988). 57. Stigler and Becker (1977); Stigler (1975); and Rajan and Zin- 23. Graham Bannock and Partners Ltd (1997). gales (2003). 24. Black and Strahan (2002); Cetorelli and Strahan (2002); 58. Stigler (1971). Beck, Demirgüç-Kunt, and Levine (2003); Cetorelli (2003); and 59. Barth, Caprio Jr., and Levine (2001) and Barth, Caprio Jr., Berger, Hasan, and Klapper (2004). and Levine (2004). 25. Demirgüç-Kunt, Laeven, and Levine (2003). 60. Martinez Peria and Schmukler (2001). 26. Berger and others (forthcoming). 61. Calomiris and Powell (2001). 27. Barth, Caprio Jr., and Levine (2004) and Demirgüç-Kunt, 62. Saunders and Wilson (2002). Laeven, and Levine (2003), respectively. 63. Caprio and Honohan (2003). 28. Unite and Sullivan (2003). 64. Stiglitz and Yusuf (2001). 29. Clarke and others (2003); Clarke and others (forthcoming); 65. Caprio and Honohan (2003). and Escude and others (2001). 66. Levy-Yeyati, Martinez Peria, and Schmukler (2004). 30. Faulkender and Petersen (2003). 67. Beck, Demirgüç-Kunt, and Levine (2003). 31. Carmichael and Pomerleano (2002). 68. See, for example, Tanzi and Davoodi (1997); Tanzi and 32. Impavido (2001) and Impavido, Musalem, and Tressel Davoodi (1998); and Devarajan, Swaroop, and Zou (1996). (2003). 69. For discussions of the problem and the history of private 33. Ekmekcioglu (2003). infrastructure, see Gómez-Ibáñez (2003); Gómez-Ibáñez and 34. Shah (1997) and Srinivas, Whitehouse, and Yermo (2000). Meyer (1993); Klein and Roger (1994); Levy and Spiller (1994); 35. Impavido (2001). Levy and Spiller (1996); Smith (1997b); Spiller and Savedoff 36. Deepthi and others (2003). (1999); and Willig (1999). The problems are greatest when 37. Black, Jang, and Kim (2003); Johnson and others (2000); La investors are asked to make large one-off investments and smaller Porta and others (1997); La Porta and others (1998); and Stiglitz when a series of small investments creates a “repeated game� that (1999a). encourages the government not to expropriate the investor. 38. Shleifer and Wolfenzohn (2002). 70. See Wodon, Ajwad, and Siaens (2003). See also Clarke and 39. Demirgüç-Kunt and Maksimovic (1998); Demirgüç-Kunt Wallsten (2003); Estache, Foster, and Wodon (2002); and World and Maksimovic (1999); Giannetti (2003); Claessens and Laeven Bank (1994b). (2003); Allayanis, Brown, and Klapper (2003); and Esty and Meg- 71. For empirical evidence of the effect of various features of the ginson (2003). investment climate on infrastructure, see Bergara, Henisz, and 40. Bae and Goyal (2003). Spiller (1998); Henisz (2002); Henisz and Zelner (2001); Weder and 41. For example, in Mexico domestic credit to the private sector as Schiffer (2000); and Zhang, Parker, and Kirkpatrick (2002). a percentage of GDP was only 12.6 percent in 2002, as compared to 72. See Lamech and Saeed (2003) for selected evidence on prior- 35 percent in Brazil and 141 percent in the U.S. (IMF-IFS statistics). ities of investors in electricity in developing countries. 42. Berkowitz and White (2002). 73. Phillips (1993); Smith (1997a); and Smith (1997b). 43. Pistor, Raiser, and Gelfer (2000). 74. For a collection of published contracts, see http://rru.worldbank. 44. Claessens and Laeven (2003). org/contracts/. 45. Durnev and Kim (2003); Gompers and Metrick (2001); Joh 75. See, for example, Klein and Hadjimichael (2003). (2003); Klapper and Love (forthcoming); and La Porta and others 76. Dollar, Hallward-Driemeier, and Mengistae (2003a). Time (1998). to get a new telephone connection is used as a proxy for the quality 46. McKinsey & Company (2002) and Aggarwal, Klapper, and of telecommunications services generally. Wysocki (2003). 77. Röller and Waverman (2001). 47. Levitt (1998); Frost, Gordon, and Hayes (2002); Hail and 78. Calderón and Servén (2003). Luez (2003); and Lee and Ng (2002). 79. International Telecommunication Union data from SIMA 48. Rajan and Zingales (2003). database for 2001. 49. Glaeser, Johnson, and Shleifer (2001). 80. World Bank Investment Climate Surveys. 50. Miller (2003). 81. Rossotto and others (2003) citing Telegeography, and Inter- 51. Galindo and Miller (2001) and Love and Mylenko (2003). national Telecommunication Union data. 52. World Bank (2003a). 82. Wallsten (2001); Wallsten (2003); Bortolotti and others 53. A recent cross-country study �nds that about 50 percent of (2002); Boylaud and Nicoletti (2001); Galal and others (1994); small �rms report �nancing constraints in countries without a Ramamurti (1996); Ros (1999); Wellenius (1997b); Winston credit bureau as compared to 27 percent in countries with a bureau, (1993); and Fink, Mattoo, and Rathindran (2002). and that 28 percent of �rms are able to obtain a bank loan in coun- 83. Komives, Whittington, and Wu (2003) and Clarke and Wall- tries without a bureau versus 40 percent of �rms in countries with sten (2003). a bureau. Love and Mylenko (2003). 84. World Bank Investment Climate Surveys. 54. Barron and Staten (2003); Bailey, Chun, and Wong (2003); 85. World Bank Investment Climate Surveys and Batra, Kauf- Padilla and Pagano (2000); and Castelar-Pinheiro and Moura (2003). mann, and Stone (2002). (c) The International Bank for Reconstruction and Development / The World Bank Endnotes 205 86. World Energy Council (2001). 14. Tooley (1999). 87. Reinikka and Svensson (2002). 15. IFC (2001). 88. Dollar, Hallward-Driemeier, and Mengistae (2003a). 16. Data are from the U.S. Department of Education. 89. Calderón and Servén (2003). 17. While one might expect a positive relation between the level 90. Hunt and Shuttleworth (1996). of mandated labor protection and income across countries (that is, 91. See, for example, Besant-Jones and Tenenbaum (2001). labor protection is a normal good), the relationship is, in fact, neg- 92. Pollitt (2003); Newbery and Pollitt (1997); and Galal and ative across a large group of countries. others (1994). 18. Bourguignon and Goh (2003); de Ferranti and others 93. World Bank (2004j), citing Spiller in Gilbert and Kahn (1996). (2000); Gill, Maloney, and Sanchez-Paramo (2002); Devarajan, 94. Zhang, Parker, and Kirkpatrick (2002). Dollar, and Holmgren (2001); Rodrik (1997); Freeman (1994); 95. Limão and Venables (2001). Matusz and Tarr (1999); Rama (2003); and World Bank (2002d). 96. Clark, Dollar, and Micco (2002). 19. For a review of the role of unions see Aidt and Tzannatos 97. Limão and Venables (2001). (2002); Brown (2000); and Boeri, Brugiavini, and Calmfors (2001). 98. Radelet and Sachs (1998). 20. Aidt and Tzannatos (2002) and Forteza and Rama (2002). 99. Limão and Venables (2001). Infrastructure includes telecom- 21. Calmfors (1993). munications as well as paved roads, unpaved roads, and railways— 22. Haltiwanger, Scarpetta, and Vodopivec (2003). each having a weight of 25 percent in an index. 23. Eslava and others (2003). 100. Limão and Venables (2001). 24. In Colombia, for every percentage point rise in the mini- 101. World Bank (2004j). mum wage, employment falls by 0.15 percentage points: Maloney 102. Clark, Dollar, and Micco (2002). and Núñez (2004). For Indonesia, see Alatas and Cameron (2003). 103. Inter-American Development Bank (2001). 25. This is possible because of noncompliance in the informal 104. Estache and Carbajo (1996) and Gaviria (1998), for example. economy and exemption for certain workers. 105. Estache and Carbajo (1996); Trujillo and Serebrisky (2003); 26. The relationship between compliance and the level of the and Gaviria (1998). minimum wage is not one-to-one, and depends on the overall 106. World Bank and PPIAF (2003) discusses these options. institutional climate and respect for laws. 107. Galal and others (1994). 27. In Poland the national minimum wage accounts for over 80 108. Gaviria (1998). percent of the going market wage in less developed areas, con- 109. Trujillo and Serebrisky (2003). tributing to high unemployment among low skilled workers, World 110. Calderón and Servén (2003). Bank (2001b). 111. Fernald (1999). 28. For more details on the role of the minimum wage as a price 112. Gómez-Ibáñez and Meyer (1993) and www.worldbank. signal for the informal sector in Latin America, see Maloney and org/html/fpd/transport/roads_ss.htm. Núñez (2004), and World Bank (2004g). 113. Tanzi and Davoodi (1997) and Tanzi and Davoodi (1998). 29. See Feliciano (1998) for Mexico, and Gill, Montenegro, and 114. Liautaud (2001). Dömeland (2002) for the experience of Latin American countries that have introduced apprentice wages. Chapter 7 30. The index of condition of employment reported in �gure 7.5 1. ILO (2004). is the normalized sum of maximum number of hours in the work- 2. Schneider (2002). See also ILO (2002a). week, overtime work, night shifts, holidays, hours of work, vacation 3. Pritchett (2001); Easterly (2001); and Topel (1999). days, and whether paid time off for holidays is mandatory. For more 4. One more year of schooling is estimated to raise wages by 7- details, see Djankov and others (2003a) and World Bank (2003a). 10 percent in many countries. See Psacharopoulos and Patrinos 31. The 39 days in Sierra Leone refer to a worker with 20 years of (2002). service. 5. Recent studies focusing on OECD countries-where differ- 32. Heckman and Pagés (2004) estimate that workers absorb ences in the quality of education are relatively smaller than across between 52 and 90 percent of the cost associated with nonwage developing countries-suggest strong positive effects of enhance- bene�ts in Latin America. Mondino and Montoya (2004) for ment of human capital on GDP per capita growth. See Bassanini Argentina, and MacIsaac and Rama (1997) for Ecuador suggest that and Scarpetta (2002) and De La Fuente and Doménech (2002). compliance with labor regulations implies an increase in labor costs 6. Rosenzweig (1995)shows for India that schooling returns are with possible disemployment effects. high when the returns to learning are also high. 33. In most countries, worker turnover is even larger than job 7. Pritchett (2001) and Pissarides (2000). turnover, because workers not only move directly from one job to 8. Acemoglu and Shimer (1999). For empirical evidence for the another, but also between employment and unemployment and United States, see Abowd and others (2001) and Nestoriak (2004). inactivity. See Alogoskou�s and others (1995). 9. Bresnahan, Brynjolfsson, and Hitt (2002). 34. Heckman and Pagés (2004), using an alternative measure of 10. Van de Walle (2003). job security that takes into account the monetary transfer to be 11. Nicholls (1998). paid to dismissed workers, con�rm that such transfers tend to be 12. World Bank (2003e). larger in Latin America than in industrial countries. 13. IFC (2001); World Bank (2002a); World Bank (2003e); and 35. The synthetic indicator of employment protection for workers El-Khawas, DePietro-Jurand, and Holm Nielsen (1998). with permanent contracts is the normalized sum of: (a) procedural (c) The International Bank for Reconstruction and Development / The World Bank 206 WORLD DEVELOPMENT REPORT 2005 inconveniences; (b) notice and severance payments; (c) standards ica and the Caribbean been able to diversify their idiosyncratic of and penalties for “unfair� dismissals; and (d) procedures for col- aggregate volatility in the 1990s, they would have enjoyed a 7 per- lective dismissals. Indicators of employment protection for tempo- cent higher consumption. rary contracts refer to: (a) the “objective� reasons under which they 56. Bigsten and others (2003) suggests that in African countries could be offered; and (b) the maximum cumulated duration of the with underdeveloped credit and insurance markets, �rms cannot contract. See Djankov and others (2003a) and World Bank (2003a). insure against temporary demand shocks and have to adjust wages 36. World Bank (2003j). and employment. 37. World Bank (2002a). 57. Acemoglu and Shimer (1999) suggest that moderate levels of 38. These results are from econometric analyses based on unemployment bene�ts help improve job matches, with positive industry-level data and controlling for other main drivers of pro- effects on productivity and output growth. ductivity or entry rates. In particular, see Nicoletti and others 58. Tanzanian households with limited liquid assets (livestock) (2001) for the evidence on the relationship between R&D and labor tend to grow proportionally more sweet potatoes, a low return and regulations, and Scarpetta and others (2002) and Scarpetta and low risk crop, than wealthier households, see Dercon (1996). In the Tressel (2004) for evidence of the impact of employment protec- Indian ICRISAT villages, reducing rainfall timing variability tion on productivity and entry rates. (through some mechanism of insurance) is estimated to have a 39. Görg (2002) and Dewit, Gorg, and Montagna (2003) for evi- large effect on farm pro�ts of the poor households, see Rosenzweig dence of the effects of employment protection on FDI. and Binswanger (1993). 40. See Nicoletti and others (2001) on self-employment; Nico- 59. See, for example, Ravallion (2003b) and World Bank (2002d). letti and others (2001) for the evidence on �rm size; and Scarpetta 60. Klasen and Woolard (2001) suggest that the absence of and others (2002) for the evidence on size of entrant �rms and unemployment bene�ts in South Africa forces the unemployed to post-entry expansion. Christianson (2004) suggests strict labor reg- base their location decisions on the availability of economic sup- ulations in South Africa push small and medium size �rms toward port—generally available in rural areas, often in parental house- more capital-intensive technologies. holds—rather than on the availability of job openings. 41. Nonpayment of contractual obligations, or wage arrears, 61. de Ferranti and others (2000) suggest that trade liberaliza- spread to nearly 60 percent of all workers in Russia in 1998 and, tion and greater competition in Latin American countries have despite declining, continued to affect a signi�cant share of the reduced the possibility of de facto pooling unemployment risk via workforce in the following years, see World Bank (2003j). severance pay over a greater population by subsidizing potentially 42. See Pagés and Montenegro (1999) and Montenegro and bankrupt �rms through higher prices. Pagés (2004) for Latin America. Djankov and others (2003a) sug- 62. In Slovenia, unpaid claims amount to more than one-third gest that an increase of 1 point in the employment laws index of total severance pay provisions: Vodopivec (2004). In Peru, poor (ranging from 0.76 to 2.40) is associated with an increase in the workers are less likely to be entitled to severance pay, and also less share of the unof�cial economy in GDP of 6.7 percentage points likely to receive it in case of dismissal: MacIsaac and Rama (2001). and an increase in the share of unof�cial employment of 13.8 per- 63. Gruber (1997) �nds that in the absence of unemployment insur- centage points. ance, average consumption expenditures would fall by 22 percent. 43. Addison and Teixeria (2001) and Nickell and Layard (1999). 64. More than two-thirds of the households with at least one 44. Cavalcanti (2003) and Mondino and Montoya (2004). unemployed worker received such bene�ts in Hungary and Poland 45. Besley and Burgess (2004). in the mid-1990s, see Vodopivec (2004). 46. Kugler and Pica (2003) for evidence in Italy. 65. Martin and Grubb (2001). 47. Pierre and Scarpetta (2004). Evidence from South Africa also 66. Mazza (1999). suggests that more than 90 percent of large �rms are reported to 67. World Bank (1995b). make greater use of temporary workers to increase flexibility of the 68. Tabor (2002). workforce: see Chandra and others (2001). See also Saavedra and 69. Cox, Jimenez, and Jordan (1994) estimate that poverty inci- Torero (2004) for Peru, and World Bank (2003a) for other coun- dence would be 25 percent higher among those receiving transfers tries. had they not received them. 48. Dolado, García-Serrano, and Jimeno (2001); Blanchard and 70. Informal transfers are estimated to offer less than 10 percent Landier (2001); and Hopenhayn (2004). of the size of typical income shocks in bad periods in India; and less 49. Agénor (1996) argues that the effectiveness of structural than 3 percent in the Sahel region, following the 1984 drought, adjustment programs in developing countries is affected by the Morduch (1999a). speci�c characteristics of their labor markets. 71. Ravallion (2003b), Ravallion and Datt (1995); Subbarao 50. Goldberg and Pavcnik (2003) for Colombia, and Aghion and (1997); Teklu and Asefa (1999); Jalan and Ravallion (2003); and others (2003) for India. Chirwa, Zgovu, and Mvula (2002). 51. Winter-Ebmer (2001). 72. Gaiha (2000). 52. Kikeri (1998). 73. Subbarao, Ahmed, and Teklu (1995) for the Philippines and 53. Winter-Ebmer (2001) and Kikeri (1998). World Bank (2002d) for African countries. 54. See Holzmann and Jorgensen (2001) and World Bank 74. Haddad and Adato (2001) �nd little relationship between (2001c). See also World Commission on the Social Dimension of the district level share of public works activity and the relative Globalization (2004). poverty, unemployment, and infrastructure need in a sample of 101 55. de Ferranti and others (2000) suggest that had Latin Amer- public works projects in the Western Cape of South Africa. (c) The International Bank for Reconstruction and Development / The World Bank Endnotes 207 75. Rawlings, Sherburne-Benz, and Van Domelen (2003). 18. Cardenas, Ocampo, and Thorp (2003) mention how notions 76. The percentage of bene�ciaries beneath the national poverty of performance-based and time-bound support were absent from line ranged from 71 percent in Zambia to 55 percent in Nicaragua. the post-World War II import substitution strategies in Latin 77. Rawlings and Rubio (2002). America. 78. Sedlacek, Ilahi, and Gustafsson-Wright (2000) and Bour- 19. Jones and Sakong (1980). guignon, Ferreira, and Leite (2002). 20. Shah (1995a). 21. OECD (2003c). Non-OECD countries include Argentina, Chapter 8 Bolivia, Cambodia, Chile, Kenya, and South Africa. 1. Chang (2002). 22. UNCTAD (2002a) and Noland and Pack (2003). 2. Taylor (1996). 23. OECD (2003e). 3. Recent publications show that the debate continues. Skeptical 24. Chen and others (2004). views on the role of industrial policy in East Asian growth include 25. ILO (2002b). Noland and Pack (2003); Pack (2000); and Smith (2000). More pos- 26. PPIAF and World Bank (2002). itive views are expressed in Lall (2003); Lall (2000); and Amsden 27. World Bank (2003i). and Chu (2003). Wong and Ng (2001) stand somewhere in 28. Charitonenko and Campion (2003) and Yaron, Benjamin, between. See also Hernandez (2004). and Piprek (1997). 4. The potential for positive spillovers from FDI, exporting and 29. Morduch (1999b). R&D provides an economic rationale for selective interventions in 30. Adams (1988). favor of these activities. Proponents of selective interventions also 31. Yaron, Benjamin, and Piprek (1997). point to other possible market failures that might justify special 32. World Bank (2003i). treatment: different learning economies for each technology, see 33. Anderson and Feder (2003). Lall (2000); coordination of competing investments, see Chang 34. Alex, Zijp, and Byerlee (2002) and Anderson and Feder (1999); and helping latecomers achieve economies of scale in (2003). mature industries, see Amsden and Chu (2003). 35. Glaeser (2001). 5. Lipsey and Lancaster (1956). 36. Batra and Mahmood (2003). 6. Reid and Gatrell (2003); Wolman (1988); and Kokko (2002). 37. Klein and Hadjimichael (2003). 7. Hausmann and Rodrik (2003). 38. Nugent and Yhee (2002). 8. References on Bangladesh, Colombia, and India are from 39. Johnson, McMillan, and Woodruff (2002a). Hausmann and Rodrik (2003); on Japan from World Bank (1993); 40. Hallberg (2000) and Batra and Mahmood (2003). and on Kenya from English, Jaffee, and Okello (2004). 41. Hallberg and Konishi (2003). 9. Noland and Pack (2003). 42. Batra and Mahmood (2003). 10. WTO (2003) reviews literature on government efforts to 43. Scott and Storper (2003). restrict competition in East Asia and concludes that cartelization 44. Porter (1998) discusses the cluster literature. Clusters are did not improve economic performance in Japan, and had negative de�ned as “geographically proximate group of interconnected consequences in Korea. companies, suppliers, service providers and associated institutions 11. For example, see World Bank (2004d); Noland and Pack in a particular �eld, linked by externalities of various types.� Porter (2003); Mody (1999); Wong and Ng (2001); and Lall (2000). (2003). 12. Irwin (2004). 45. Sölvell, Lindqvist, and Ketels (2003). 13. Rodríguez-Clare (2001). 46. World Bank (2003b). 14. Greenstone and Moretti (2003). 47. For example, Altenburg and Meyer-Stamer (1999) identify 15. Reid and Gatrell (2003) describe the case of an automotive three types of clusters in Latin America: “survival clusters of SMEs, company that threatened to relocate to a different U.S. city unless it more advanced and differentiated mass production clusters, and received various incentives—which led an incentive package of clusters around transnational corporations.� $322.5 million. While 4,900 jobs were promised in 1997, in 2001 48. Altenburg and Meyer-Stamer (1999). the company announced that it would reduce the number of 49. The Mitchell Group Inc. (2003). employees to 3,600. The authors suggest the threat to relocate was 50. The Mitchell Group Inc. (2003). motivated by opportunism. 51. Includes workers in export processing zones, special eco- 16. Thaler (1993). nomic zones, economic and technological development zones, and 17. The technology policy in India’s ninth �ve-year plan (1997- maquiladoras (�rms in Mexico that process or assemble imported 2002) included the following objectives: (a) optimal utilization of components that are then exported). science and technology to control population growth, improve 52. Madani (1999) mentions at least 18 Sub-Saharan African food security, literacy, and so forth; (b) support best scientists and countries with some type of EPZ initiative. Political unrest halted be at the forefront of selected research �elds; (c) concentrate on the development of EPZ in Togo, and civil war in Liberia and Sierra technological capabilities that can be commercially successful; (d) Leone. Macroeconomic distortions hindered EPZ development in promote environmentally friendly technologies and the like; (e) Kenya. Infrastructure and bureaucratic burden posed obstacles to develop innovative capabilities in education system; (f) increase the development of EPZ in Senegal and Ghana. The success stories resources for R&D in private �rms; (g) promote strategic sectors of Mauritius and Madagascar are anomalies in the region. such as atomic energy and space. Cited by Mani (2001b). 53. Subramanian and Roy (2003). (c) The International Bank for Reconstruction and Development / The World Bank 208 WORLD DEVELOPMENT REPORT 2005 54. Jenkins and Kuo (2000); Panagariya (2000); Radelet (1999); Chapter 9 and Harrold, Jayawickrama, and Bhattasali (1996). 1. Braithwaite and Drahos (2000) and Dollinger (1970). 55. English and De Wulf (2002). 2. For example, in 1928 the Permanent Court of the Interna- 56. UNCTAD (1996); Christiansen, Oman, and Charlton tional Court of Justice ruled that compensation was payable for the (2003); and Easson (2001). expropriation by Poland of private property belonging to a German 57. UNCTAD (2000c). �rm in the Chorzow Factory Case. The court stated that “there can 58. Morisset (2003b). be no doubt that the expropriation . . . is a derogation from the 59. Fletcher (2002). rules generally applied in regard to the treatment of foreigners and 60. UNCTAD (2002b). the principle of respect for vested rights.� 61. Wells and others (2001) and Bergsman (1999). 3. Putnam (1988); Ederington (2001); Staiger and Tabellini 62. MIGA (2002). (1999); and Conconi and Perroni (2003). 63. Wunder (2001b). 4. Dixit and Nalebuff (1991); and Persson and Tabellini (2000). 64. In Thailand a 1984 study from the Ministry of Finance 5. Under the auspices of the joint World Bank-IMF initiatives found that, in aggregate, the share of investment that would have on the Financial Sector Assessment Program and the Reports on taken place even in the absence of incentives was 70 percent. Cited the Observance of Standards and Codes (ROSC), the OECD Princi- by Halvorsen (1995). ples are used as a benchmark in assessing corporate governance 65. Zee, Stotsky, and Ley (2002) and Shah (1995a). institutional frameworks and practices. 66. Wells and others (2001). 6. Shihata (1986) cites the example of the “Jecker� claim, when 67. UNCTAD (2002b). an investment dispute was used by France for armed conflict in 68. Morisset (2003a). Mexico in 1861–62. 69. Morisset and Andrews-Johnson (2003). 7. For example, UNCITRAL Arbitration Rules, the 1958 New 70. Morisset and Andrews-Johnson (2003). The authors mea- York Convention on the Recognition and Enforcement of Foreign sure the investment climate using the Heritage Foundation’s Index Arbitral Awards, and the International Court of Arbitration of the of Economic Freedom, a composite measure that aggregates mea- International Chamber of Commerce. For a review of international sures of macroeconomic stability, openness, taxation, and other commercial arbitration, see Paulsson (1996). factors. 8. For background on ICSID, see Shihata (1986). Information 71. UNCTAD (2003b). on ICSID can be found on its Web site (www.worldbank.org/icsid/) 72. Pursell (2001) and UNCTAD (2003b). and in ICSID Review-Foreign Investment Law Journal, published 73. Moran (2001). by the Johns Hopkins University Press. 74. Moran (2001). 9. ICSID (2003). 75. Battat, Frank, and Shen (1996). 10. Recent cases have been based on interpretations of NAFTA, 76. Zee, Stotsky, and Ley (2002). rather than BITs, although similar issues can arise. The cases are 77. Hall and Van Reenen (1999). discussed in UNCTAD (2003e) and Hallward-Driemeier (2003). 78. Shah and Baffes (1995) and Shah (1995b) found R&D For those considered under ICSID jurisdiction, see also incentives cost effective in Pakistan and Canada, respectively. How- www.worldbank.org/icsid/cases/awards.htm. ever, Hall and Van Reenen (1999) review the literature of tax incen- 11. Weingast (1995). For a skeptical view on the bene�ts of har- tives in the OECD and are more skeptical. Their review of studies of monization relative to competition, see Stephan (1999). For a tax credits in the U.S. concludes that on average it produced review of the tradeoffs in the area of �nancial services, see White roughly a dollar-for-dollar increase in reported R&D spending (1996). since their introduction in 1981. 12. Putnam (1988); see also Maggi and Rodríguez-Clare (1998). 79. OECD (2003f). 13. The International Chamber of Commerce dates from 1919, 80. Kim (1997) and Yusuf (2003). Yet heavy regulation and gov- and has been involved in promoting harmonization of various con- ernment funding may have constrained the development of private tractual terms to facilitate international trade (www.iccwbo.org). venture capitalists. Israel is a counter-example; jump-starting a 14. UNCITRAL is a subsidiary body of the General Assembly of venture capital sector that was successful enough in attracting pri- the United Nations, and was established in 1966 with the general vate capital to render public support unnecessary.Trajtenberg mandate to further the progressive harmonization and uni�cation (2002). of the law of international trade. UNCITRAL has since prepared a 81. David, Hall, and Toole (2000). wide range of conventions, model laws, and other instruments 82. Wallsten (2004) mentions that in the U.S. alone, there were dealing with the substantive law that governs trade transactions or 135 science parks in 1998. He �nds that they are not a major source other aspects of business law having an impact on international of high-tech employment. Literature evaluating science parks is not trade (www.uncitral.org). very developed, and few studies look at their cost-effectiveness. de 15. Hoekman and others (2004); Schiff and Winters (2003); and Ferranti and others (2003). Bhagwati (2002). 83. Feser (2002). 16. Multilateral arrangements on environmental issues date 84. Yusuf (2003). from the 1972 U.N. Stockholm Conference. 85. de Ferranti and others (2003). 17. Siebert (2003). (c) The International Bank for Reconstruction and Development / The World Bank Endnotes 209 18. World Bank (2003b); WTO (2003); and Clarke and Evenett 11. The approach piloted by the EU (for example, in Burkina (2003b). Faso) is based on outcome indicators that are agreed beforehand 19. Hoekman and Mavroidis (2002) and Clarke and Evenett with the recipient government; see Zongo and others (2000). The (2003a). approach adopted under the Millennium Challenge Account 20. The tax harmonization debate in the EU has been ongoing focuses on country selectivity based on three main criteria: ruling for many years. Recent analytical work suggests that the bene�ts justly, encouraging economic freedom, and investing in people; for from tax coordination in the EU may be negligible, see Mendoza more details see www.mca.gov. and Tesar (2003). 12. For a discussion of output-based aid and its applications see 21. For example, although the national government in Brazil Brook and Smith (2001). prohibited the states from exempting �rms from the value-added 13. About one-third of the policy-based support during the tax, the states were able to get around this using various mecha- period was for �nancial sector-related operations, reflecting nisms, including lending enterprises amounts equal to the tax they responses to �nancial crises. The crises also explain the increase in owed on highly subsidized terms, see Tendler (2002). policy-based lending, which had been trending downward. 22. OECD (2003d) and OECD and WTO (2003). 14. An evaluation of World Bank adjustment lending shows that between 1996 and 1999 about 40 percent of investment climate Chapter 10 conditionality related to business environment, a third to privatiza- 1. World Bank (2003b). tion and public enterprise reforms, and one-quarter to supporting 2. The G7 (Group of 7) has also endorsed the importance of private participation in infrastructure. World Bank (2001a). investment climate improvements; see communiqué issued in Sep- 15. Devarajan, Dollar, and Holmgren (2001) and World Bank tember 2003. The importance of investment climate improvements (1998a). in achieving the Millenium Development Goals is also highlighted 16. McMillan, Rodrik, and Welch (2002). in the report of the United Nations Commission on the Private Sec- 17. World Bank (2003k) and Wolfensohn (1998). tor and Development (2004). 18. IDA and IMF (2003) and World Bank (2003l). The experi- 3. United Nations (2002b). ence has not been uniformly adopted across all countries preparing 4. World Bank (2004d); UNCTAD (2000a); and UNCTAD and PRSPs; see World Bank (2004h). WTO (2000). 19. Pistor (2000). 5. IMF estimations from OECD PSE (Producer Support Esti- 20. Technical Assistance is not subject to the 2001 OECD Devel- mate); OECD (2002a); and OECD DAC (2003). opment Assistance Committee recommendation for its members to 6. Estimates based on dynamic gains in real income (relative to untie Of�cial Development Assistance to Least Developed Coun- base year of 1997) using the Global Trade Analysis Project (GTAP) tries; see OECD DAC (2001). However, countries including the database. They are based on a “pro-poor� scenario in which high United Kingdom have already untied the provision of all develop- income countries reduce tariffs and eliminate tariff peaks. Agricul- ment assistance, and similar proposals are being discussed in the ture has a maximum tariff of 10 percent, with an average tariff of 5 EU and by the OECD. See European Commission (2004a) and percent, and manufacturing has a maximum tariff of 5 percent United Kingdom - DFID (2001). with a 1 percent average tariff. Further, export subsidies are elimi- 21. Batra and Mahmood (2003). nated, domestic subsidies decoupled, and speci�c tariffs, tariff rate 22. A recent review of donor support for private sector develop- quotas, and antidumping duties and sanctions eliminated. See ment observed: “One may question the sincerity of those donors World Bank (2004d). who claim to be working towards [private sector development] in 7. Hoekman (2000). developing countries, but whose instruments in the main center on 8. Development assistance data is based on OECD DAC (2004) the promotion of exports and investments by their own private sec- and includes support provided as Of�cial Development Finance tors. This is not to say that the involvement of the donor’s private (ODF). For this Report, Of�cial Assistance provided to the subset sector should be condemned by de�nition, but rather that it is of more advanced developing countries and territories has been facile to claim that investments and exports contribute by de�ni- excluded. Commitment data in the OECD Creditor Reporting Sys- tion to recipient country PSD, let alone to poverty reduction.� See tem (CRS) was mapped to speci�c categories of assistance (policy- Schulpen and Gibbon (2002). based support, capacity building, and infrastructure). For a discus- 23. In particular, the Partnership in Statistics for Development sion of the methodology and associated caveats see Migliorisi and in the 21st Century (PARIS21), (www.paris21.org). Galmarini (2004). 24. United Nations (2002a). 9. World Bank (1998a). 25. World Bank (2004c). 10. World Bank (2004a). (c) The International Bank for Reconstruction and Development / The World Bank References The word processed describes informally reproduced works that Adserà, Alicia, Carles Boix, and Mark Payne. 2003. “Are you Being may not be commonly available through libraries. Served? Political Accountability and Quality of Government.� Abowd, John M., John Haltiwanger, Julia Lane, and Kristin San- Journal of Law, Economics, and Organization 19(2):445–90. dusky. 2001. “Within and Between Changes in Human Capital, Aedo, Cristián, and Sergio Núñez. 2001. The Impact of Training Technology, and Productivity.� U.S. Census Bureau. Washington, Policies in Latin America and the Caribbean: The Case of ‘Pro- D.C. Processed. grama Joven’. Washington, D.C.: Inter-American Development Acemoglu, Daron. 2001. “Directed Technical Change.� Cambridge, Bank. Available on line at http://www.ilo.org/public/english/ Mass.: National Bureau of Economic Research Working Paper 8287. region/ampro/cinterfor/temas/youth/doc/aedo/aedo.pdf. Acemoglu, Daron, and Simon Johnson. 2003. “Unbundling Institu- Agarwal, Manish, Ian Alexander, and Bernard Tenenbaum. 2003. tions.� Cambridge, Mass.: National Bureau of Economic “The Delhi Electricity Discom Privatizations: Some Observa- Research Working Paper Series 9934. tions and Recommendations for Future Privatizations in India Acemoglu, Daron, Simon Johnson, and James A. Robinson. 2001. and Elsewhere.� Washington, D.C.: World Bank, Energy and “The Colonial Origins of Comparative Development: An Empir- Mining Sector Board Discussion Paper 8. ical Investigation.� American Economic Review 91(5):1369–401. Agénor, Pierre. 1996. “The Labor Market and Economic Adjust- ————. 2002. “Reversal of Fortune: Geography and Institutions ment.� IMF Staff Papers 43(2):261–335. in the Making of the Modern World Income Distribution.� Aggarwal, Reena, Leora Klapper, and Peter D. Wysocki. 2003. “Port- Quarterly Journal of Economics 117(4):1231–94. folio Preferences of Foreign Institutional Investors.� Washington, Acemoglu, Daron, and Robert Shimer. 1999. “Ef�cient Unemploy- D.C.: World Bank Policy Research Working Paper Series 3101. ment Insurance.� Journal of Political Economy 107(5):893–28. Aghion, Philipe, Robin Burgess, Stephen Redding, and Fabrizio Acevedo, Germán C., and Patricio A. Eskenazi. 2003. “The Chilean Zilibotti. 2003. “The Unequal Effects of Liberalization: Theory Unemployment Insurance: A New Model of Income Support and Evidence from India.� Centre for Economic Policy Research. Available for Unemployed Workers.� Paper presented at the London. Processed. International Workshop on Severance Pay Reform: Toward Aghion, Philippe, and Robin Burgess. 2003. “Liberalization and Unemployment Savings and Retirement Accounts. Vienna. Industrial Performance: Evidence from India and the UK.� Har- November 7. vard University. Cambridge, Mass. Processed. Acs, Zoltan J., and David B. Audretsch. 1987. “Innovation, Market Aghion, Philippe and Steven Durlauf. Forthcoming. “Handbook of Structure and Firm Size.� Review of Economics and Statistics Economic Growth.� Amsterdam: North Holland. 69(4):567–74. Ahluwalia, Montek. 2002. “Economic Reforms in India Since 1991: Adams, Dale W. 1988. “The Conundrum of Successful Credit Pro- Has Gradualism Worked?� Journal of Economic Perspectives jects in Floundering Rural Financial Markets.� Economic Devel- 16(3):67–88. opment and Cultural Change 36(2):355–67. Ahmadi, Justice A. M. 1999. “ADR Programmes in India.� In Addison, John T., and Paulino Teixeria. 2001. “The Economics of USAID and World Bank, eds., Background Material on ADR. Employment Protection.� Bonn: Institute for the Study of Labor Washington, D.C.: World Bank, Legal Institutions Thematic Discussion Paper 381. Group. Adenikinju, Adeola F. 2003. “Electric Infrastructure Failures in Aidt, Toke, and Za�ris Tzannatos. 2002. Unions and Collective Bar- Nigeria: A Survey-Based Analysis of the Costs and Adjustment gaining—Economic Effects in a Global Environment. Washington, Responses.� Energy Policy 31(14):1519–30. D.C.: World Bank. Adlington, Gavin. 2002. “Land Workshop Material.� Paper pre- Aitken, Brian, Gordon H. Hanson, and Ann E. Harrison. 1997. sented at the Regional Workshop on Land Issues in Europe and “Spillovers, Foreign Investment, and Export Behavior.� Journal of the CIS. Budapest, Hungary. April 3. International Economics 43(1-2):103–32. 210 (c) The International Bank for Reconstruction and Development / The World Bank References 211 Aitken, Brian J., and Ann E. Harrison. 1999. “Do Domestic Firms base.� Washington, D.C.: World Bank Policy Research Working Bene�t from Direct Foreign Investment? Evidence from Paper Series 3127. Venezuela.� American Economic Review 89(3):605–18. Ba, Seydou. 2000. “How Can Effective Strategies Be Developed for Akiyama, Takamasa, John Baffes, Donald Larson, and Panayotis Law and Justice Programs? Are There Models for Legal Reform Varangis. 2003. “Commodity Market Reforms in Africa: Some Programs? An Example of OHADA.� Paper presented at the Recent Experience.� Economic Systems 27(1):83–115. Comprehensive Legal and Juridical Development Conference. Alatas, Vivi, and Lisa Cameron. 2003. “The Impact of Minimum Washington, D.C. May 6. Wages on Employment in a Low Income Country: An Evaluation Bae, Kee-Hong, and Vidhan K. Goyal. 2003. “Property Rights Pro- Using the Difference-in-Differences Approach.� Washington, tection and Bank Loan Pricing.� Korea University and Hong D.C.: World Bank Policy Research Working Paper Series 2985. Kong University of Science and Technology. Seoul. Processed. Alesina, Alberto, Silvia Ardagna, Giuseppe Nicoletti, and Fabio Baharoglu, Deniz. 2002. “World Bank Experience in Land Manage- Schiantarelli. 2003a. “Regulation and Investment.� Cambridge, ment and The Debate on Tenure Security.� Washington, D.C.: Mass.: National Bureau of Economic Research Working Paper World Bank, Urban and Local Government Background Series Series 9560. 16. Alesina, Alberto, Arnaud Devleeschauwer, William Easterly, Sergio Bailey, Andre, Suzi Chun, and Jeffery Wong. 2003. “Wanted: Asian Kurlat, and Rômain Wacziarg. 2003b. “Fractionalization.� Journal Credit Bureaus.� The McKinsey Quarterly, July 9. of Economic Growth 8(2):155–94. Baker, Joanne. 1996. Conveyancing Fees in a Competitive Market. Alex, Gary, Willem Zijp, and Derek Byerlee. 2002. “Rural Extension Sydney: Law Foundation of New South Wales for the Justice and Advisory Services: New Directions.� Washington, D.C.: Research Centre. World Bank, Rural Development Strategy Background Paper 9. Baldwin, Richard E, and Paul Krugman. 2004. “Agglomeration, Allayanis, George, Gregory W. Brown, and Leora F. Klapper. 2003. Integration, and Tax Harmonisation.� European Economic “Capital Structure, Foreign Debt and Financial Risk: Evidence Review 48(1):1–23. from East Asia.� Journal of Finance 58(6):2667–709. Baldwin, Richard E. 2003. “Openess and Growth: What’s the Alogoskou�s, George, Charles R. Bean, Giuseppe Bertola, Daniel Empirical Relationship?� Cambridge, Mass.: National Bureau of Cohen, Juan José Dolado, and Gilles Saint-Paul. 1995. Unemploy- Economic Research Working Paper Series 9578. ment: Choices for Europe. Monitoring European Integration. Lon- Bandiera, Oriana, Gerard Caprio, Patrick Honohan, and Fabio don: Centre for Economic Policy Research. Schiantarelli. 2000. “Does Financial Reform Raise or Reduce Sav- Altenburg, Tilman, and Jorg Meyer-Stamer. 1999. “How to Promote ings?� Review of Economics and Statistics 82(2):239–63. Clusters: Policy Experiences from Latin America.� World Devel- Ban�eld, Edward C. 1958. The Moral Basis of a Backward Society. opment 27(9):1693–713. Glencoe, Ill.: Free Press. Amsden, Alice H., and Wan-wen Chu. 2003. Beyond Late Develop- Banks, Arthur. 2001. Cross-national Time-series Data Archive. Bing- ment: Taiwan’s Upgrading Policies. Cambridge, Mass.: MIT Press. hamton, N.Y.: Center for Comparative Political Research at the Anderson, Jock R., and Gershon Feder. 2003. “Rural Extension Ser- State University of New York. vices.� Washington, D.C.: World Bank Policy Research Working Bannock Consulting. 2001. Best Practices in “One Stop Shops� for Paper Series 2976. SME Development. Bulgaria Reform of Administrative Barriers to APEC Committee on Trade and Investment. 2003. Guide to the SME Development. London, UK: Bannock Consulting. Investment Regimes of the APEC Economies, 5th ed. Singapore: Bannock, Graham, Matthew Gamser, and Mariell Juhlin. 2004. Asian Paci�c Economic Co-operation Secretariat (APEC). “The Importance of the Enabling Environment for Business and Argy, Steven, and Matthew Johnson. 2003. Mechanisms for Improv- Economic Growth: A 10 Country Comparison of Central Europe ing the Quality of Regulations: Australia in an International Con- and Africa.� Case Study Commissioned by the UK Department text. Melbourne: Productivity Commission. for International Development for the 2005 World Development Ariyoshi, Akira, Karl Habermeier, Bernard Laurens, Inci Otker- Report. Robe, Jorge Ivan Canales-Krijenko, and Andrei Kirilenko. 2000. Barro, Robert, and Xavier Sala-i-Martin. 2003. Economic Growth. “Capital Controls: Country Experiences with their Use and Lib- Cambridge, Mass.: MIT Press. eralization.� Washington, D.C.: International Monetary Fund Barro, Robert J., and Jong-Wha Lee. 2001. “International Data on Occasional Paper 190. Educational Attainment: Updates and Implications.� Oxford Eco- Arora, Ashish, Marco Ceccagnoli, and Wesley M. Cohen. 2003. nomic Papers 53(3):541–63. “R&D and the Patent Premium.� Cambridge, Mass.: National Barron, John M., and Michael Staten. 2003. “The Value of Compre- Bureau of Economic Research Working Paper Series 9431. hensive Credit Reports: Lessons from the U.S. Experience.� In Aw, Bee Yan, Sukkyun Chung, and Mark J. Roberts. 2000. “Produc- Margaret Miller, ed., Credit Reporting Systems and the Interna- tivity and Turnover in the Export Market: Micro-level Evidence tional Economy. Cambridge, Mass.: MIT Press. from the Republic of Korea and Taiwan (China).� World Bank Bartelsman, Eric, John Haltiwanger, and Stefano Scarpetta. 2004. Economic Review 14(1):65–90. “Microeconomic Evidence of Creative Destruction in Industrial Ayyagari, Meghana, Thorsten Beck, and Asli Demirgüç-Kunt. 2002. and Developing Countries.� Background paper for the WDR “Small and Medium Enterprises across the Globe: A New Data- 2005. (c) The International Bank for Reconstruction and Development / The World Bank 212 WORLD DEVELOPMENT REPORT 2005 Bartelsman, Eric, Stefano Scarpetta, and Fabiano Schivardi. 2003. sented at the Bank Privatization, World Bank Conference. Wash- “Comparative Analysis of Firm Demographics and Survival: ington, D.C. November 20. Micro-Level Evidence for the OECD Countries.� Paris: Organiza- Beck, Thorsten, Robert Cull, and T. Jerome Afeikhena. 2003. “Bank tion for Economic Co-operation and Development, Economics Privatization and Performance: Empirical Evidence from Nige- Department Working Paper 348. ria.� Paper presented at the Bank Privatization–World Bank Barth, James R., Gerard Caprio Jr., and Ross Levine. 2001. “Banking Conference. Washington, D.C. November 20. Systems Around the Globe: Do Regulation and Ownership Affect Beck, Thorsten, Asli Demirgüç-Kunt, and Ross Levine. 2003. “The Performance and Stability.� In Frederic Mishkin S., ed., Pruden- Impact of SMEs on Growth, Development, and Poverty: Cross- tial Regulation and Supervision: What Works and What Doesn’t. Country Evidence.� Washington, D.C.: World Bank Policy Chicago, I.L.: University of Chicago Press. Research Working Paper Series 3178. ————. 2004. “Bank Regulation and Supervision: What Works Beck, Thorsten, Ross Levine, and Norman Loayza. 2000. “Finance Best?� Journal of Financial Intermediation 13(2):205–48. and the Sources of Growth.� Journal of Financial Economics 58(1- Barzel, Yoram. 2002. A Theory of the State: Economic Rights, Legal 2):261–300. Rights, and the Scope of the State. Cambridge,U.K.: Cambridge Becker, Randy, and J. Vernon Henderson. 2000. “Effects of Air Qual- University Press. ity Regulations on Polluting Industries.� Journal of Political Econ- Basant, Rakesh, and Brian Fikkert. 1996. “The Effects of R&D, For- omy 108(2):379–421. eign Technology Purchase, and Domestic and International Bell, Clive, Shantayanan Devarajan, and Hans Gersbach. 2003. “The Spillovers on Productivity in Indian Firms.� Review of Economics Long-Run Economic Costs of AIDS: Theory and an Application and Statistics 78(2):187–99. to South Africa.� Washington, D.C.: World Bank Policy Research Bassanini, Andrea, and Stefano Scarpetta. 2002. “Does Human Working Paper Series 3152. Capital Matter for Growth in OECD Countries? A Pooled Mean- Bergara, Mario E., Witold J. Henisz, and Pablo T. Spiller. 1998. group Approach.� Economics Letters 74(3):399–405. “Political Institutions and Electric Utility Investment: A Cross- Bastos, Fabiano, and John Nasir. 2003. “Which Dimensions of the Nation Analysis.� California Management Review 40(2):18–35. IC are Most Constraining?� World Bank. Washington D.C. Berger, Allen, Seth Bonime, Lawrence Goldberg, and Lawrence Processed. White. Forthcoming. “The Dynamics of Market Entry: The Basu, Kaushik. 1999. “Child Labor: Cause, Consequence, and Cure, Effects of Mergers and Acquisitions on Entry in the Banking with Remarks on International Labor Standards.� Journal of Eco- Industry.� Journal of Business. nomic Literature 37(3):1083–119. Berger, Allen N., Asli Demirgüç-Kunt, Ross Levine, and Joseph G. Bates, Robert H. 1981. Markets and States in Tropical Africa: The Haubrich. Forthcoming. “Bank Concentration and Competition: Political Basis of Agricultural Policies. Berkeley: University of Cal- An Evolution in the Making.� Journal of Money, Credit and Bank- ifornia Press. ing. ————. 2001. Prosperity and Violence: The Political Economy of Berger, Allen N., Iftekhar Hasan, and Leora F. Klapper. 2004. “Fur- Development. New York: W.W. Norton & Company Ltd. ther Evidence on the Link between Finance and Growth: An Batra, Geeta, Daniel Kaufmann, and Andrew H. W. Stone. 2002. Invest- International Analysis of Community Banking and Economic ment Climate Around the World: Voices of the Firms from the World Performance.� Journal of Financial Services Research 25(2- Business Environment Survey. Washington, D.C.: World Bank. 3):169–202. Batra, Geeta, and Syed Mahmood. 2003. “Direct Support to Private Berglof, Erik, and Gerard Roland. 2000. “From Regatta to Big Bang? Firms: Evidence on Effectiveness.� Washington, D.C.: World The Impact of EU Accession Strategy on Reforms in Central and Bank Policy Research Working Paper Series 3170. Eastern Europe.� International Monetary Fund. Washington, Batra, Geeta, and Hong Tan. 1995. “Enterprise Training in Develop- D.C. Processed. ing Countries: Overview of Incidence, Determinants, and Pro- Bergsman, Joel. 1999. “Advice on Taxation and Tax Incentives for ductivity Outcomes.� Washington, D.C.: World Bank Private Sec- Foreign Direct Investment.� Paper presented at the FIAS Semi- tor Development Occasional Paper 9. nars on FDI Issues: Knowledge Gains. Washington, D.C. May 27. Battat, Joseph, Isaiah Frank, and Xiaofang Shen. 1996. “Suppliers to Berkowitz, Daniel, Katharina Pistor, and Jean-Francois Richard. Multinationals: Linkage Programs to Strengthen Local Compa- 2003. “Economic Development, Legality, and the Transplant nies in Developing Countries.� Washington, D.C.: Foreign Effect.� European Economic Review 47(1):165–95. Investment Advisory Service Occasional Paper 6. Berkowitz, Jeremy, and Michelle J. White. 2002. “Bankruptcy and Baumol, William J. 2002. The Free-Market Innovation Machine: Small Firm’s Acces to Credit.� Cambridge, Mass.: National Analyzing the Growth Miracle of Capitalism. Princeton, N.J.: Bureau of Economic Research Working Paper Series 9010. Princeton University Press. Berman, Jonathan E., and Tobias Webb. 2003. Race to the Top: Bayoumi, Tamim, David T. Coe, and Elhanan Helpman. 1999. Attracting and Enabling Global Sustainable Business. Business Sur- “R&D Spillovers and Global Growth.� Journal of International vey Report. Washington, D.C.: World Bank. Economics 47(2):399–428. Bernard, Andrew B., and J. Bradford Jensen. 1999. “Exceptional Beck, Thorsten, Juan Miguel Crivelli, and William Summerhill. Exporter Performance: Cause, Effect or Both?� Journal of Inter- 2003. “The Transformation of State Banks in Brazil.� Paper pre- national Economics 47(1):1–25. (c) The International Bank for Reconstruction and Development / The World Bank References 213 Bernstein, Peter L. 1996. Against the Gods: The Remarkable Story of Bird, Richard M., and Michael Engelschalk. 2003. “Tax Policy and Risk. New York: John Wiley & Sons. Tax Administration: Lessons from the Last Decade.� World Bank. Bertola, Giuseppe, Tito Boeri, and Sandrine Cazes. 2000. “Employ- Washington, D.C. Processed. ment Protection in Industrialized Countries: The Case for New Black, Bernard, Hansung Jang, and Woochan Kim. 2003. “Does Indicators.� International Labour Review 139(1):57–72. Corporate Governance Affect Firms’ Market Values? Evidence Besant-Jones, John, and Bernard Tenenbaum. 2001. “The Califor- from Korea.� Stanford, C.A.: Stanford Law and Economics nia Power Crisis: Lessons for Developing Countries.� Washing- School, Olin Working Paper 237. ton, D.C.: World Bank, Energy and Mining Sector Board Discus- Black, Sandra E. 1999. “Investigating the Link Between Competi- sion Paper 1. tion and Discrimination.� Monthly Labor Review, December. Besley, Timothy. 1995. “Property Rights and Investment Incentives: Black, Sandra E., and Philip E. Strahan. 2002. “Entrepreneurship Theory and Evidence from Ghana.� Journal of Political Economy and Bank Credit Availability.� Journal of Finance 57(6):2807–33. 103(5):903–37. Blalock, Garrick, and Paul Gertler. 2003. “Technology From Foreign Besley, Timothy, and Robin Burgess. 2004. “Can Labor Regulation Direct Investment and Welfare Gains Through the Supply Hinder Economic Performance? Evidence from India.� Quarterly Chain.� University of California. Berkeley, C.A. Processed. Journal of Economics 119(1):91–134. Blanchard, Olivier. 1986. “Investment, Output, and the Cost of Betcherman, Gordon, Karina Olivas, and Amit Dar. 2003. “Impacts Capital: A Comment.� Brookings Papers on Economic Activity of Active Labor Market Programs: New Evidence from Evalua- 0(1):153–58. tions with Particular Attention to Developing and Transition Blanchard, Olivier, and Francesco Giavazzi. 2003. “Macroeconomic Countries.� Washington, D.C.: World Bank, Social Protection Effects of Regulation and Deregulation in Goods and Labor Network Paper Series 402. Markets.� Quarterly Journal of Economics 118(3):879–907. Bhagwati, Jagdish. 2002. Free Trade Today. Princeton, N.J.: Prince- Blanchard, Olivier J., and Augustin Landier. 2001. “The Perverse ton University Press. Effects of Partial Labor Market Reform: Fixed-term Contracts in Bhatnagar, Subnash, and Rajeb Chawla. 2004. “Bhoomi: Online France.� Economic Journal 112(480):214–44. Delivery of Land Titles in Karnataka, India.� Ahmedabad Indian Blomström, Magnus, and Ari Kokko. 1998. “Multinational Corpora- Institute of Management and Government of Karnataka. tions and Spillovers.� Journal of Economic Surveys 12(3):247–77. Ahmedabad, India and Karnataka, India. Processed. Bloom, David E., Lakshmi. R. Bloom, Davod Steven, and Mark Biers, Dan, and Sadanard Dhume. 2000. “In India, a Bit of Califor- Weston. 2003. Business and HIV/AIDS: Who Me? Geneva: World nia.� Far Eastern Economic Review 163(44):38–40. Economic Forum. Biggs, Tyler. 2003. “Is Small Beautiful and Worthy of Subsidy? Liter- Boeri, Tito, Agar Brugiavini, and Lars Calmfors, eds. 2001. The Role ature Review.� World Bank. Washington, D.C. Processed. of Unions in the Twenty-First Century. New York: Oxford Univer- Biggs, Tyler, Vijaya Ramachandran, and Manju Kedia Shah. 1998. sity Press. “The Determinants of Enterprise Growth in Sub-Saharan Africa: Boix, Carles. 2003. Democracy and Redistribution. New York : Cam- Evidence from the Regional Program on Enterprise Develop- bridge University Press. ment.� Washington, D.C.: World Bank, Regional Program for Bortolotti, Bernardo, Juliet D’Souza, William L. Megginson, and Enterprise Development Discussion Paper 103. Marcella Fantini. 2002. “Privatization and the Sources of Perfor- Biggs, Tyler, Manju Kedia Shah, and Pradeep Srivastava. 1995. mance Improvement in the Global Telecommunications Indus- “Technological Capabilities and Learning in African Enter- try.� Telecommunications Policy 26(5-6):243–68. prises.� Washington, D.C.: World Bank, Africa Department Bosworth, Barry, and Susan M. Collins. 2003. “The Empirics of Series, Technical Paper 288. Growth: An Update.� The Brookings Institution. Washington, Bigsten, Arne, Paul Collier, Stefano Dercon, Marcel Fafchamps, D.C. Processed. Bernard Gauthier, Jan Willem Gunning, Abena Oduro, Remco Oos- Botswana–Ministry of Lands, Housing and Environment. 2002. tendorp, Cathy Patillo, Måns Soderbom, Francis Teal, and Albert Botswana National Land Policy. Gaborone: Natural Resources Zeufack. 2000. “Contract Flexibility and Dispute Resolution in Services Ltd. for the Ministry of Lands, Housing and Environ- African Manufacturing.� Journal of Development Studies 36(4):1–37. ment. Department of Lands. ————. 2003. “Risk Sharing in Labor Markets.� World Bank Eco- Bourguignon, Francois. 2000. “Crime, Violence, and Inequitable nomic Review 17(3):349–66. Development.� In Joseph Stiglitz, ed., Annual World Bank Confer- Bineswaree, Bolasky, and Caroline Freund. 2004. “Trade, Regula- ence on Development Economics. Washington, D.C.: World Bank. tions and Growth.� Washington, D.C.: World Bank Policy ————. 2004. “The Poverty-Growth-Inequality Triangle.� Paper Research Working Paper Series 3255. presented at the Indian Council for Research on International Binswanger, Hans P., Klaus Deininger, and Gershon Feder. 1995. Economic Relations Conference. New Delhi, India. February 4. “Power, Distortions, Revolt, and Reform in Agricultural Land Bourguignon, Francois, Francisco Ferreira, and P. G. Leite. 2002. Relations.� In Jere Behrman and T. N. Srinivasan, eds., Handbook Ex-ante Evaluation of Conditional Cash Transfer Programs: The of Development Economics vol. 3B. Amsterdam: Elsevier B.V. Case of Bolsa Escola. Washington, D.C.: World Bank. Bird, Richard M. 2003. “Administrative Dimensions of Tax Bourguignon, Francois, and C. Goh. 2003. Trade and Labor Mar- Reform.� University of Toronto. Toronto, Canada. Processed. ket Vulnerability in Korea, Indonesia, and Thailand: Evaluating (c) The International Bank for Reconstruction and Development / The World Bank 214 WORLD DEVELOPMENT REPORT 2005 Vulnerability from Repeated Cross-sections. Washington, D.C.: Burgess, Robin, and Tony Venables. 2003. “Towards a Microeconom- World Bank. ics of Growth.� London School of Economics. London. Processed. Boylaud, Olivier, and Giuseppe Nicoletti. 2001. “Regulation, Mar- Burki, Shahid Javed, and Guillermo E. Perry. 1998. An Economic ket Structure, and Performance in Telecommunications.� OECD Theorist’s Book of Tales: Essays that Entertain the Consequences of Economic Studies 32:99–142. New Assumptions in Economic Theory. Washington, D.C.: World Braga, Carlos A. Primo, Carsten Fink, and Claudia Paz Sepúlveda. Bank Latin American and Caribbean Studies. 2000. “Intellectual Property Rights and Economic Develop- Burns, Anthony. 2004. “Thailand’s 20 Year Program to Title Rural ment.� Washington, D.C.: World Bank Discussion Paper 412. Land.� Background paper for the WDR 2005. Braithwaite, John, and Peter Drahos. 2000. Global Business Regula- Burr, Chandler. 2000. “Grameen Village Phone: Its Current Status tion. Cambridge, U.K.: Cambridge University Press. and Future Prospects.� Paper presented at the Business Services Bratton, William J., and William Andrews. 1999. “What We’ve for Small Enterprises in Asia: Developing Markets and Measur- Learned About Policing.� City Journal 9(2):14–27. ing Performance Conference. Hanoi, Vietnam. March 6. ————. 2004. “Driving Out the Crime Wave: The Police Meth- Burra, Neera. 1995. Born to Work: Child Labour in India. Oxford: ods that Worked in New York Can Work in Latin America.� Time, Oxford University Press. July 23. Busse, Matthias. 2003. “Tariffs, Transport Costs and the WTO Doha Brautigam, Deborah, and Michael Woolcock. 2001. “Small States in Round: The Case of Developing Countries.� Estey Centre Journal a Global Economy—The Role of Institutions in Managing Vul- of International Law and Trade Policy 4(1):15–31. nerability and Opportunity in Small Developing Economies.� Buvinic, Mayra, and Andrew Morrison. 2000. “Technical Notes on World Institute for Development Economics/United Nations Violence Prevention.� Washington, D.C.: Inter-American Devel- University. Helsinki. Processed. opment Bank, Social Development Division, Technical Note 5. Brennan, Geoffrey, and James Buchanan. 1980. The Power to Tax: Caballero, Ricardo J. 1991. “On the Sign of the Uncertainty-Invest- Analytical Foundations of a Fiscal Constitution. Cambridge, U.K.: ment Relationship.� American Economic Review 81(1):279–88. Cambridge University Press. ————. 1999. “Aggregate Investment.� In John B. Taylor and Bresnahan, Timothy F., Erik Brynjolfsson, and Lorin M. Hitt. 2002. Michael Woodford, eds., Handbook of Macroeconomics. Amster- “Information Technology, Workplace Organization and the dam: Elsevier Science B. V. Demand for Skilled Labor: Firm-Level Evidence.� Quarterly Jour- Caballero, Ricardo J., Eduardo M. R. A. Engel, and Alejandro Micco. nal of Economics 117(1):339–76. 2004. “Microeconomic Flexibility in Latin America.� Cambridge, Brimble, Atchaka. 2002. “Thailand’s Experience of One-Stop Mass.: National Bureau of Economic Research Working Paper Shops.� Paper presented at the APEC Seminar on Investments’ Series 10398. One Stop Shop. Lima, Peru. February 26. Caballero, Ricardo J., and Mohamad Hammour. 2000. “Creative Broadman, Harry, James Anderson, Constantijn A. Claessens, Destruction and Development: Institutions, Crises and Restruc- Randi Ryterman, Stefka Slavova, María Vagliasindi, and María turing.� Paper presented at the World Bank Conference in Devel- Vincelette. 2004. Building Market Institutions in South Eastern opment Economics. Washington, D.C. July 20. Europe: Comparative Prospects for Investment and Private Sector Cabanero-Verzosa, Cecilia, and Paul Mitchell. 2003. “Communicat- Development. Washington, D.C.: World Bank. ing Economic Reform.� World Bank. Washington, D.C. Processed. Brook, Penelope J., and Suzanne M. Smith, eds. 2001. Contracting Calderón, César, and Luis Servén. 2003. “The Output Cost of Latin for Public Services: Output-Based Aid and Its Applications. Wash- America’s Infrastructure Gap.� In William R. Easterly and Luis ington, D.C.: World Bank. Servén, eds., The Limits of Stabilization: Infrastructure, Public Brown, Drusilla K. 2000. “International Trade and Core Labour De�cits, and Growth in Latin America. Washington, D.C.: World Standards: A Survey of Recent Literature.� Paris: OECD, Direc- Bank. torate for Education, Employment, Labour and Social Affairs, Calderon-Madrid, A., and T. Belem. 2001. The Impact of the Mexi- Labour Market and Social Policy–Occasional Papers 43. can Training Program for Unemployed Workers on Re-employment Brown, Drusilla K., Alan V. Deardorff, and Robert M. Stern. 2003. Dynamics and on Earnings. Washington, D.C.: Inter-American “The Effects of Multinational Production on Wages and Working Development Bank. Conditions in Developing Countries.� Cambridge, Mass.: National Calmfors, Lars. 1993. “Centralization of Wage Bargaining and Bureau of Economic Research Working Paper Series 9669. Macroeconomic Performance: A Survey.� OECD Economic Stud- Brown, J. David, and John S. Earle. 2004. “Economic Reforms and ies 21:161–91. Productivity-Enhancing Reallocation in the Post-Soviet Transi- Calomiris, Charles W., and Andrew Powell. 2001. “Can Emerging tion.� Kalamazoo, M.I.: Upjohn Institute Staff Working Paper 04- Market Regulators Establish Credible Discipline? The Case of 98. Available on line at http://www.upjohninst.org. Argentina. 1992–99.� In Frederic Mishkin, ed., Prudential Super- Brueckner, Jan K. 2000. “Fiscal Decentralization in Developing vision: What Works and What Doesn’t. Chicago, I.L.: University of Countries: The Effects of Local Corruption and Tax Evasion.� Chicago Press. Annals of Economics and Finance 1(1):1–18. Camerer, Colin F., George Loewenstein, and Matthew Rabin, eds. Brusky, Bonnie. 2003. “Knowing When to Stop: The Case of UNDP 2003. Advances in Behavioral Economics. Princeton: Princeton Bangladesh.� Washington, D.C.: CGAP Direct Case Study 4. University Press. (c) The International Bank for Reconstruction and Development / The World Bank References 215 Cannock, Geoffrey. 2001. “Expanding Rural Telephony: Output- Cetorelli, Nicola. 2003. “Life-Cycle Dynamics in Industrial Sectors: Based Contracts for Pay Phones in Peru.� In Penelope J. Brook The Role of Banking Market Structure.� Federal Reserve Bank of and Suzanne M. Smith, eds., Contracting for Public Services: Out- St.Louis Review 85(4):135–47. put-based Aid and its Applications. Washington, D.C.: World Cetorelli, Nicola, and Philip E. Strahan. 2002. “Banking Competi- Bank. tion, Technology and the Structure of Industry.� Federal Reserve Cao, Yuanzheng, Yingyi Qian, and Barry R. Weingast. 1999. “From Bank of Chicago. Chicago, I.L. Processed. Federalism, Chinese Style, to Privatization, Chinese Style.� Eco- CGAP (The Consultive Group to Assist the Poor). 1997. “Scaling nomics of Transition 7(1):103–31. Up in Micro�nance-Evidence from Global Experience.� Wash- Caprio, Gerard, and Asli Demirgüç-Kunt. 1998. “The Role of Long- ington, D.C.: CGAP Focus Note 6. Term Finance: Theory and Evidence.� World Bank Research Chalk, Nigel A. 2001. “Tax Incentives in the Philippines: A Regional Observer 13(2):171–89. Perspective.� Washington, D.C.: International Monetary Fund Caprio, Gerard, and Patrick Honohan. 2003. “Can the Unsophisti- Working Paper WP/01/181. cated Market Provide Discipline?� Paper presented at the Market Chami, Ralph, Mohsin Khan, and Sunil Sharma. 2003. “Emerging Discipline: The Evidence across Countries and Industries Con- Issues in Bank Regulation.� Washington, D.C.: International ference. Chicago, IL. October 30. Monetary Fund Working Paper 101. Cardenas, Enrique, José Antonio Ocampo, and Rosemary Thorp, Chandra, Vandana, Lalita Moorty, Jean-Pascal Nganou, Bala Rajarat- eds. 2003. An Economic History of Twentieth-Century Latin Amer- nam, and Kendall Schaefer. 2001. “Constraints to Growth and ica. Vol. 3. Industrialization and the State in Latin America: The Employment in South Africa: Report No.2 Evidence from the Small, Postwar Years. New York: Palgrave Publishers Ltd. Medium and Micro Enterprise Firm Survey.� Washington, D.C.: Carlin, Wendy, Steven Fries, Mark E. Schaffer, and Paul Seabright. World Bank, Southern Africa Department, Discussion Paper 15. 2001. “Competition and Enterprise Performance in Transition Chandra, Vandana, and Bala Rajaratnam. 2004. “Differences in Economies Evidence from a Cross-country Survey.� London: Firm Types and Obstacles to Growth: Insights from South CEPR Discussion Paper 2840. African Investment Climate Surveys.� World Bank. Washington, Carlson, Ingrid, and Mark J. Payne. 2003. “Estudio Comparativo de D.C. Processed. Estadísticas de Empleo Público en 26 Países de America Latina y Chang, Ha-Joon. 1999. “Industrial Policy and East Asia: The Mira- el Caribe.� In Koldo Echebarria, ed., Red de Gestión y Transparen- cle, the Crisis, and the Future.� Paper presented at the Re-think- cia de la Política Pública. Servicio Civil: Temas para un Diálogo. ing the East Asian Miracle. San Francisco, C.A. Washington, D.C.: Banco Interamericano de Desarrollo. ————. 2002. Kicking Away the Ladder: Development Strategy in Carmichael, Jeffrey, and Michael Pomerleano. 2002. The Develop- Historical Perspective. London: Anthem. ment and Regulation of Non-Bank Financial Institutions. Wash- Charitonenko, Stephanie, and Anita Campion. 2003. “Expanding ington, D.C.: World Bank. Commercial Micro�nance in Rural Areas: Constraints and Castelar Pinheiro, Armando, Indermit S. Gill, Luis Servén, and Opportunities.� Paper presented at the Paving the Way Forward: Mark R. Thomas. 2001. “Brazilian Economic Growth, An International Conference on Best Practice on Rural Finance. 1900–2000: Lessons and Policy Implications.� Paper presented at Washington, D.C. June 2. the GDN Conference. Rio de Janeiro. December 9. Charlton, Andrew. 2003. “Incentive Bidding for Mobile Investment: Castelar-Pinheiro, Armando. 1998. “The Hidden Costs of Judicial Economic Consequences and Potential Responses.� Paris: Orga- Inef�ciency: General Concepts and Calculation for Brazil.� In nization for Economic Co-operation and Development, Devel- Organization of American States, ed., Judicial Reform in Latin opment Centre, Technical Papers 203. America: An Un�nished Task. Washington, D.C.: Organization of Charmes, Jacques. 2000. “The Contribution of Informal Sector to American States. GDP in Developing Countries: Assessment, Estimates, Methods, Castelar-Pinheiro, Armando, and Célia Cabral. 2001. “Credit Mar- Orientations for the Future.� Paper presented at the 4th Meeting kets in Brazil: The Role of Judicial Enforcement.� In Marco of the Delhi Group on Informal sector Statistics. Geneva. Pagano and Oracio Attansio, eds., Defusing Default: Incentives Chaudhry, Kiren Aziz. 1997. The Price of Wealth: Economies and and Institutions. Washington, D.C.: Inter-American Develop- Institutions in the Middle East. Ithaca: Cornell University Press. ment Bank. Chen, Chunlai. 2002. “Foreign Direct Investment: Prospects and Castelar-Pinheiro, Armando, and Alkimar Moura. 2003. “Segmen- Policies.� In OECD, ed., China in the World Economy: The Domes- tation and the Use of Information in Brazilian Credit Markets.� tic Policy Challenges. Paris: Organization for Economic Co-oper- In Margaret S. Miller, ed., Credit Reporting Systems and the Inter- ation and Development. national Economy. Cambridge, Mass.: MIT Press. Chen, Martha Alter, Renana Jhabvala, and Frances Lund. 2002. Sup- Cavalcanti, Carlos B. 2003. “Argentina: Growth, Productivity and porting Workers in the Informal Economy: A Policy Framework. Labor Reallocation.� World Bank. Washington, D.C. Processed. Geneva: International Labor Of�ce. Center for Policy Studies. 2003. Blue Bird Agenda for Civil Society in Chen, Martha Alter, Renana Jhabvala, and Reema Nanavaty. 2004. Southeast Europe. In Search of Responsive Government: State “The Investment Climate for Female Informal Businesses: A Building and Economic Growth in the Balkans. Budapest: Center Case Study from Urban and Rural India.� Background paper for for Policy Studies, Central European University. the WDR 2005. (c) The International Bank for Reconstruction and Development / The World Bank 216 WORLD DEVELOPMENT REPORT 2005 Chen, Shaohua, and Martin Ravallion. 2004. “How Have the Clarke, George R. G., and Scott J. Wallsten. 2003. “Universal Service: World’s Poorest Fared since the Early 1980s?� World Bank Empirical Evidence on the Provision of Infrastructure Services to Research Observer 19(2):141–69. Rural and Poor Urban Consumers.� In Timothy C. Irwin and Chen, Shaohua, and Yan Wang. 2001. “China’s Growth and Poverty Penelope J. Brook, eds., Infrastructure for Poor People: Public Pol- Reduction: Trends between 1990 and 1999.� Washington, D.C.: icy for Private Provision. Washington, D.C.: World Bank. World Bank Policy Research Working Paper Series 2651. Clarke, Julian L., and Simon J. Evenett. 2003a. “A Multilateral Chengappa, A. 1999. “File Management in Secretariat: Present Framework for Competition Policy?� In Simon J. Evenett and Structure.� Institute for Human Resource Development. Hyder- Swiss Secretariat of Economic Affairs, eds., The Singapore Issues abad, India. Processed. and the World Trading System: The Road to Cancun and Beyond. Chermak, Janie M. 1992. “Political Risk Analysis: Past and Present.� Bern: World Trade Institute. Resources Policy 18(3):167–78. ————. 2003b. “The Deterrent Effects of National Anti-Cartel Chifor, George. 2002. “Caveat Emptor: Developing International Laws: Evidence from the International Vitamins Cartel.� Disciplines for Deterring Third Party Investment in Unlawfully Antitrust Bulletin 48(3):689–726. Expropriated Property.� Law and Policy in International Business Clerides, Sofronis K., Saul Lach, and James R. Tybout. 1998. “Is 33(2):179–282. Learning by Exporting Important? Micro-dynamic Evidence Chinoy, Mike. 1998. N. Korea Shows Small Signs of Private Enterprise. from Colombia, Mexico, and Morocco.� Quarterly Journal of Eco- Atlanta, G.A.: CNN News website. Available on line at nomics 113(3):903–947. http://www.cnn.com/WORLD/asiapcf/9809/16/nkorea.economy/. Coase, Ronald H. 1960. “The Problem of Social Cost.� Journal of Chirwa, Ephrain W., Evious K. Zgovu, and Peter M. Mvula. 2002. Law and Economics 3(1):1–44. “Participation and Impact of Poverty-oriented Public Works Coe, David, and Elhanan Helpman. 1995. “International R&D Projects in Rural Malawi.� Development Policy Review Spillovers.� European Economic Review 39(5):859–87. 20(2):159–176. Coe, David T., Elhanan Helpman, and Alexander W. Hoffmaister. Christiansen, Hans, Charles Oman, and Andrew Charlton. 2003. 1997. “North-South R&D Spillovers.� Economic Journal “Incentives-based Competition for Foreign Direct Investment: 107(440):134–49. The Case of Brazil.� Paris: Organization for Economic Co-opera- Cohen, Nevin. 2001. What Works: Grameen Telecom’s Village tion and Development, Development Centre, Working Papers on Phones. Washington, D.C.: World Resources Institute, Digital International Investment 2003/1. Dividend Study. Christianson, David. 2004. “The Investment Climate in South Coleman, James S. 1988. “Social Capital in the Creation of Human Africa Regulatory Issues: Some Insights from the High-Growth, Capital.� American Journal of Sociology 94(Supplement):s95–s120. Export-Oriented SME Sector.� Case Study Commissioned by the Commonwealth Secretariat. 2003. “Promoting Investment into UK Department for International Development for the 2005 ‘Economies with Endowed Handicaps’.� Paper presented at the World Development Report. 2003 Small States Forum. Dubai. September 24. Claessens, Stijn, and Luc Laeven. 2003. “Financial Development, Commonwealth Secretariat and World Bank Joint Task Force on Property Rights, and Growth.� Journal of Finance 58(6):2401–36. Small States. 2004. Small States: Meeting Challenges in the Global Clark, Ximena, David Dollar, and Alejandro Micco. 2002. “Mar- Economy. Washington, D.C. and London: Commonwealth Secre- itime Transport Cost and Port Ef�ciency.� Washington, D.C.: tariat and World Bank. World Bank Policy Research Working Paper Series 2781. Conconi, Paola, and Carlo Perroni. 2003. “Self-enforcing Interna- Clarke, George R. G, and Lixin Colin Xu. 2004. “Privatization, tional Agreements and Domestic Policy Credibility.� Munich: Competition and Corruption: How Characteristics of Bribe Tak- CESIfo Working Paper Series 988. ers and Payers Affect Bribes To Utilities.� Journal of Public Eco- Consejo Nacional de Zonas Francas de Exportación. 2002. Informe nomics 88(9-10):2067–97. Estadístico del Sector Zonas Francas 2002. Santo Domingo, Clarke, George R. G., and Robert Cull. 2002. “Political and Eco- República Dominicana: Consejo Nacional de Zonas Francas de nomic Determinants of the Likelihood of Privatizing Argentine Exportación. Public Banks.� Journal of Law and Economics 45(1):165–77. Coolidge, Jacqueline, Lars Grava, and Sanda Putnina. 2004. Clarke, George R. G., Robert Cull, María Soledad Martinez Peria, “Inspectorate Reform in Latvia 1999–2003.� Background paper and Susana Sánchez. Forthcoming. “Bank Lending to Small Busi- for the WDR 2005. nesses in Latin America: Does Bank Origin Matter?� Journal of Copeland, Brian R., and M. Scott Taylor. 2004. “Trade, Growth and Money, Credit and Banking. the Environment.� Journal of Economic Literature 42(1):7–71. ————. 2003. “Foreign Bank Entry: Experience, Implications for Covin, Jeffrey G., and Dennis P. Slevin. 1989. “Strategic Manage- Developing Countries, and Agenda for Further Research.� World ment of Small Firms in Hostile and Benign Environments.� Bank Research Observer 18(1):25–40. Strategic Management Journal 10(1):75–87. Clarke, George R. G., Robert Cull, and Mary Shirley. 2003. “Synthe- Cox, Donald, Emmanuel Jimenez, and John. Jordan. 1994. “Fam- sis: Empirical Studies of Bank Privatization.� Paper presented at ily Safety Nets and Economic Transition: A Study of Private the World Bank Conference on Privatization. Washington, D.C. Transfers in Kyrgyzstan.� World Bank. Washington, D.C. November 20. Processed. (c) The International Bank for Reconstruction and Development / The World Bank References 217 Crespo, Jorge, Carmela Martin, and Francisco Javier Velazquez. De Wulf, Luc, and Emile Finateu. 2002. “Best Practices in Customs 2002. “International Technology Diffusion Through Imports and Administration Reform — Lessons from Morocco.� Washington, Its Impact on Economic Growth.� Madrid: European Economy D.C.: World Bank PREM notes 67. Group Working Paper 12. Deaton, Angus. 2002. “Counting the World’s Poor: Problems and Cristini, Marcela, and Ramiro A. Moya. 2001. “The Importance of Possible Solutions.� World Bank Research Observer 16(2):125–47. an Effective Legal System for Credit Markets: The Case of Deccan Herald. 2003. “Will China Challenge India?� Deccan Herald, Argentina.� In Marco Pagano and Orazio Attansio, eds., Defusing February 3. Default: Incentives and Institutions. Washington, D.C.: Inter- Deepthi, Fernando, Leora Klapper, Victor Sulla, and Dimitri Vittas. American Development Bank. 2003. “The Global Growth of Mutual Funds.� Washington, D.C.: Cukierman, Alex. 1992. Central Bank Strategy, Credibility, and Inde- World Bank Policy Research Working Paper Series 3055. pendence: Theory and Evidence. Cambridge, Mass.: MIT Press. Dehejia, Rajeev H., and Roberta Gatti. 2002. “Child Labor: The CUTS Center for Competition, Investment and Economic Regula- Role of Income Variability and Access to Credit in a Cross Sec- tion, eds. 2003. Pulling Our Socks Up: A Study of Competition tion of Countries.� Washington, D.C.: World Bank Policy Regimes of Seven Developing Countries of Africa and Asia under Research Working Paper Series 2767. the 7-UP Project. Jaipur, India: Jaipur Printers. Deininger, Klaus. 2002. “Agrarian Reforms in Eastern European Darku, Alexander Bilson. 2001. “Private Investment, Uncertainty Countries: Lessons from International Experience.� Journal of and Irreversibility in Uganda.� African Finance Journal 3(1):1–25. International Development 14(7):987–1003. Das-Gupta, Arindam, Michael Engelschalk, and William Mayville. Deininger, Klaus, Songquing Jin, Berham Andrew, Samuel 1999. “An Anti-corruption Strategy for Revenue Administra- Gebre-Selassie, and Berhanu Nega. 2003. “Tenure Security and tion.� World Bank, Washington, D.C.: PREM Note 33. Land-Related Investment: Evidence from Ethiopia.� Washing- David, Paul A., Bronwyn H. Hall, and Andrew A. Toole. 2000. “Is Pub- ton, D.C.: World Bank Policy Research Working Paper Series lic R&D a Complement or Substitute for Private R&D? A Review of 2991. the Econometric Evidence.� Research Policy 29(4-5):497–529. Demirgüç-Kunt, Asli, Thorsten Beck, and Vojislav Maksimovic. de Carvalho, Antonio Gledson. 2003. “Effects of Migration to Spe- Forthcoming. “Bank Competition, Financing Obstacles and cial Corporate Governance Levels of Bovespa.� University of Sao Access to Credit.� Journal of Money Credit and Banking. Paulo, Economics Department. Sao Paulo. Processed. Demirgüç-Kunt, Asli, Luc Laeven, and Ross Levine. 2003. “The de Ferranti, David, Guillermo E. Perry, Indermit S. Gill, J. Luis Impact of Bank Regulations, Concentration and Institutions on Guasch, William F. Maloney, Carolina Sanchez-Paramo, and Bank Margins.� Washington, D.C.: World Bank Policy Research Norbert Schady. 2003. Closing the Gap in Education and Technol- Working Paper Series 3030. ogy. Washington, D.C.: World Bank. Demirgüç-Kunt, Asli, and Vojislav Maksimovic. 1998. “Law, de Ferranti, David, Guillermo E. Perry, Indermit S. Gill, and Luis Finance and Firm Growth.� Journal of Finance 53(6):2107–37. Serven. 2000. Securing Our Future in a Global Economy. Washing- ————. 1999. “Institutions, Financial Markets and Firm Debt ton, D.C.: World Bank. Maturity.� Journal of Financial Economics 54(3):295–336. De La Fuente, A., and R. Doménech. 2002. “Human Capital in Demombynes, Gabriel, and Berk Özler. 2002. “Crime and Local Growth Regressions, How Much Difference Does Data Quality Inequality in South Africa.� Washington, D.C.: World Bank Pol- Make?� London: CEPR Discussion Paper 3587. icy Research Working Paper Series 2925. De Long, J. Bradford. 2000. “Cornucopia: The Pace of Economic Demsetz, Harold. 1967. “Toward a Theory of Property Rights.� Growth in the Twentieth Century.� Cambridge, Mass.: National American Economic Review 57(May 1967):347–58. Bureau of Economic Research Working Paper Series 7602. Dercon, Stefan. 1996. “Risk, Crop Choice, and Savings: Evidence De Long, J. Bradford. 2003. “India since Independence: An Analytic from Tanzania.� Economic Development and Cultural Change Growth Narrative.� In Dani Rodrik, ed., In Search of Prosperity. 44(3):458–513. Princeton: Princeton University Press. Desai, Padma, and Pritha Mitra. 2004. “Why Do Some Countries De Mooij, Ruud A., and Sjef Ederveen. 2001. “Taxation and Foreign Recover So Much More Easily Than Others?� Paper presented at Direct Investment.� CPB Netherlands Bureau for Economic Pol- the Festschrift in Honor of Guillermo A. Calvo. Washington, icy Analysis: CPB Discussion Paper 003. D.C. ————. 2002. “Taxation and Foreign Direct Investment: A Meta- Desai, Raj M. 2004. “Political Influence and Firm Innovation: Analysis.� CPB Netherlands Bureau for Economic Policy Analy- Micro-Level Evidence from Developing Countries.� Background sis: CPB Report 2002/1. paper for the WDR 2005. De Soto, Hernando. 2000. The Mystery of Capital: Why Capitalism Desai, Raj M., and Anders F. Olofsgård. 2003. “A Political Model of Triumphs in the West and Fails Everywhere Else. New York: Basic the Soft Budget Constraint: Evidence from Transitions Books. Economies.� Georgetown University, School of Foreign Service De Wulf, Luc. 2004. “Tradenet in Ghana: Best Practice in the Use of Working Paper. Washington, D.C. Information Technology.� Background paper for the WDR 2005. Devarajan, Shantayanan, David Dollar, and Torgny Holmgren. ————. 2003. “Uganda: Issues and Lessons in Customs Reform.� 2001. Aid and Reform in Africa Lessons from Ten Case Studies. World Bank. Washington, D.C. Processed. Washington, D.C.: World Bank. (c) The International Bank for Reconstruction and Development / The World Bank 218 WORLD DEVELOPMENT REPORT 2005 Devarajan, Shantayanan, Vinaya Swaroop, and Heng-fu Zou. 1996. Dollar, David, and Aart Kraay. 2002. “Growth is Good for the Poor.� “The Composition of Public Expenditure and Economic Journal of Economic Growth 7(3):195–225. Growth.� Journal of Monetary Economics. 37(2):313–44. Dollar, David, Anqing Shi, Shuilin Wang, and L. Colin. Xu. 2004. Devereux, Michael P., Rachael Grif�th, and Alexander Klemm. Improving City Competitiveness through the Investment Climate: 2002. “Corporate Income Tax Reforms and International Tax Ranking 23 Chinese Cities. Washington, D.C.: World Bank. Competition.� Economic Policy: A European Forum 0(35):449–88. Dollinger, Philippe. 1970. The German Hansa. Stanford, Ca.: Stan- Dewit, G., H. Gorg, and C. Montagna. 2003. “Should I Stay or ford University Press. Should I Go? A Note on Employment Protection, Domestic Dolzer, Rudolf, and Margrete Stevens. 1995. Bilateral Investment Anchorage, and FDI.� Bonn, Germany: Institute for the Study of Treaties. The Hague: Martinus Nijhoff Publishers. Labor, Discussion Paper 845. Domberger, Simon, and Avrom Sherr. 1989. “The Impact of Com- Diamond, Peter A., and James A. Mirlees. 1971. “Optimal Taxation petition on Pricing and Quality of Legal Services.� International and Public Production I: Production Ef�ciency.� American Eco- Review of Law and Economics 9(1):41–56. nomic Review 61(1):8–27. Dougherty, Carter. 2004. “Trade Theory versus Used Clothes in Diop, Chimere. 2003. “Support to Private Sector and Trade Devel- Africa.� The New York Times, June 3. opment: Lessons Learned.� Paper presented at the International Dowell, Glen, Stuart Hart, and Bernard Yeung. 2000. “Do Corpo- Conference on Trade and Investment. Dakhar. April 23. rate Global Environmental Standards Create or Destroy Market Dixit, Avinash, and Barry Nalebuff. 1991. “Making Strategies Cred- Value?� Management Science 46(8):1059–74. ible.� In Richard Zeckhauser, ed., Strategy and Choice. Cam- Dubash, Navroz K., and Sudhir Chella Rajan. 2001. “The Politics of bridge, Mass.: MIT Press. Power Sector Reform in India.� World Resources Institute. Wash- Dixit, Avinash K., and Robert S. Pindyck. 1994. Investment Under ington, D.C. Processed. Uncertainty. Princeton, N.J.: Princeton University Press. Durnev, Art, and E. Han Kim. 2003. “To Steal or Not to Steal: Firm Djankov, Simeon, Rafael La Porta, Florencio López-de-Silanes, Attributes, Legal Environment, and Valuation.� Paper presented Andrei Shleifer, and Juan Botero. 2003a. “The Regulation of at the 14th Annual Conference on Financial Economics and Labor.� Cambridge, Mass.: National Bureau of Economic Accounting (FEA). San Diego. October 31. Research Working Paper Series 9756. Dyer, Geoff. 2001a. “Brazil’s Pitch for Market Credibility.� Financial Djankov, Simeon, Rafael La Porta, Florencio López-de-Silanes, and Times, February 27. Andrei Shliefer. 2003b. “Courts.� Quarterly Journal of Economics ————. 2001b. “UBS Warburg Leads Brasil IPO.� Financial 118(2):453–518. Times, February 22. Djankov, Simeon, Ira Lieberman, Joyita Mukherjee, and Tatiana Easson, Alex. 2001. “Tax Incentives for Foreign Direct Investment, Nenova. 2002. “Going Informal: Bene�ts and Costs.� World Part I: Recent Trends and Countertrends.� Bulletin for Interna- Bank. Washington, D.C. Processed. tional Fiscal Documentation 55(7):266–74. Djankov, Simeon, and Peter Murrell. 2002. “Enterprise Restructur- Easterly, William. 2001. The Elusive Quest for Growth: Economists’ Adven- ing in Transition: A Quantitative Survey.� Journal of Economic tures and Misadventures in the Tropics. Cambridge, Mass.: MIT Press. Literature 40(3):739–792. Easterly, William, Shantayanan Devarajan, and Howard Pack. 2001. Do, Quy-Toan, and Lakshmi Iyer. 2003. “Land Rights and Eco- “Is Investment in Africa Too Low or Too High? Macro and Micro nomic Development: Evidence from Vietnam.� Washington, Evidence.� Journal of African Economies 10(suplement):81–108. D.C.: World Bank Policy Research Working Paper Series 3120. Easterly, William, and Ross Levine. 2001. “It’s Not Factor Accumu- Dobrinsky, Rumen. 2002. “Tax Structures in Transitions lation: Stylized Facts and Growth Models.� World Bank Economic Economies: A Comparative Perspective vis-a-vis EU Member Review 15(2):177–219. States.� Paper presented at the Oesterreichische National Bank ————. 2003. “Tropics, Germs and Crops: How Endowments East-West Conference. Vienna. November 3. Influence Economic Development.� Journal of Monetary Eco- Doig, Alan, and Stephen Riley. 1998. “Corruption and Anticorruption nomics 50(1):3–39. Strategies: Issues and Case Studies from Developing Countries.� In Easterly, William R., Roumeen Islam, and Joseph E. Stiglitz. 2000. UNDP, ed., Corruption and Integrity Improvement Initiatives in Devel- “Shaken and Stirred: Explaining Growth Volatility.� In Boris oping Countries. New York: United Nations Development Program. Pleskovic and Nicholas Stern, eds., Annual World Bank Confer- Dolado, Juan José, Carlos García-Serrano, and Juan F. Jimeno. 2001. ence on Development Economics. Washington, D.C.: World Bank. “Drawing Lessons from the Boom of Temporary Employment in Easterly, William R., and Luis Servén, eds. 2003. The Limits of Stabi- Spain.� Madrid: Fundación de Estudios de Economía Aplicada lization: Infrastructure, Public De�cits, and Growth in Latin Amer- 2001–11. Available on line at http://www.fedea.es/. ica. Washington, D.C.: World Bank. Dollar, David, Mary Hallward-Driemeier, and Taye Mengistae. EBRD (European Bank for Reconstruction and Development). 2003a. “Investment Climate and Firm Performance in Develop- 1999. Transition Report 1999: Ten Years of Transition. London, ing Countries.� World Bank. Washington D.C. Processed. U.K.: European Bank for Reconstruction and Development. ————. 2003b. “Investment Climate, Infrastructure and Trade: A Ebrill, Liam, Michael Keen, Jean-Paul Bodin, and Victoria Sum- Comparison of Latin America and Asia.� World Bank. Washing- mers. 2001. The Modern VAT. Washington, D.C.: International ton, D.C. Processed. Monetary Fund. (c) The International Bank for Reconstruction and Development / The World Bank References 219 Economic and Social Research Foundation (ESRF). 2002. Competi- Estache, Antonio, and José Carbajo. 1996. “Competing Private tion Law and Policy—A Tool for Development in Tanzania. Jaipur, Ports—Lessons from Argentina.� Washington, D.C.: World Bank, India: CUTS Center for International Trade, Economics and Public Policy for the Private Sector Note 100. Environment. Estache, Antonio, Vivien Foster, and Quentin Wodon. 2002. Ederington, Josh. 2001. “International Coordination of Trade and Accounting for Poverty in Infrastructure Reforms: Learning from Domestic Policies.� American Economic Review 91(5):1580–39. Latin America’s Experience. Washington, D.C.: World Bank. Edmonds, Eric V. 2004. “Does Child Labor Decline with Improving Esty, Benjamin C., and William L. Megginson. 2003. “Creditor Economic Status?� Cambridge, Mass.: National Bureau of Eco- Rights, Enforcement, and Debt Ownership Structure: Evidence nomic Research Working Paper Series 10134. from the Global Syndicated Loan Market.� Journal of Financial Edwards, Sebastian. 1999. “How Effective Are Capital Controls?� and Quantitative Analysis 38(1):37–59. Journal of Economic Perspectives 13(4):65–84. Etounga-Manguelle, Daniel D. 2000. “Does Africa Need a Cultural Ekmekcioglu, Rengin. 2003. “Development of Factoring Market in Adjustment Program?� In Lawrence E. Harrison and Samuel Turkey.� Paper presented at the Factoring Industry as a Key Tool Huntington, eds., Culture Matters: How Values Shape Human for SME Development in EU Accession Countries Conference. Progress. New York: Basic Books. Warsaw. October 23. European Commission. 2004a. Proposal for a Regulation of the El-Khawas, Elaine, Robin DePietro-Jurand, and Lauritz Holm European Parliament and of the Council: On the Access to Com- Nielsen. 1998. Quality Assurance in Higher Education: Recent munity External Assistance. Brussels: European Commission. Progress; Challenges Ahead. Report 21199. Washington, D.C.: ————. 2004b. Report on Competition in Professional Services. World Bank, Human Development Network. Brussels: European Commission. Ellis, Amanda. 2003. Engendering Private Sector Development. Evenett, Simon J. 2004. “Competition Law and the Investment Cli- Washington, D.C.: World Bank. mate in Developing Countries.� Case Study Commissioned by Elstrodt, Heinz-Peter, Pablo Ordorica Lenero, and Eduardo the UK Department for International Development for the 2005 Urdapilleta. 2002. “Micro Lessons for Argentina.� The McKinsey World Development Report. Quarterly 2002(2):1–8. Faccio, Mara. 2003. “Politically Connected Firms.� University of Engelschalk, Michael, Samia Melhem, and Dana Weist. 2000. Nashville. Nashville, Tenn. Processed. “Computerizing Tax and Customs Administrations.� Washing- Fafchamps, Marcel. 2004. Market Institutions and Sub-Saharan ton, D.C.: World Bank PREM Note 44. Africa: Theory and Evidence. Cambridge, Mass.: MIT Press. English, Philip, and Luc De Wulf. 2002. “Export Development Poli- Fafchamps, Marcel, and Bart Minten. 2001. “Property Rights in a cies and Institutions.� In Bernard M. Hoekman, Philip English, Flea Market Economy.� Economic Development and Cultural and Aaditya Mattoo, eds., Development, Trade, and the WTO: A Change 49(2):229–68. Handbook. Washington, D.C.: World Bank. Fairbairn, I. J., and Worrell DeLisle. 1996. South Paci�c and English, Philip, Steven Jaffee, and Julius Okello. 2004. “Exporting Caribbean Island Economies: A Comparative Study. Brisbane: Out of Africa: The Kenya Horticulture Success Story.� Paper pre- Foundation for Development Cooperation. sented at the Scaling Up Poverty Reduction: A Global Learning Falaschetti, Dino. 2003. “Credible Commitments and Investment: Process Conference. Shanghai. May 25. Does Opportunistic Ability or Incentive Matter?� Economic Environmental Resources Management. 2004. Contracting Out Inquiry 41(4):660–74. Utility Regulatory Functions. Washington, D.C.: Environmental Fan, Shenggen, Peter Hazell, and Sukhadeo Thorat. 1999. “Linkages Resources Management for the World Bank. between Government Spending, Growth, and Poverty in Rural Erb, Claude B., Campbell R. Harvey, and Tadas E. Viskanta. 2000. India.� Washington, D.C.: IFPRI Research Report 110. Available “The Risk and Expected Returns of African Equity Investment.� on line at www.ifpri.org. In Paul Collier and Catherine Pattillo, eds., Investment and Risk Fan, Shenggen, Linxiu Zhang, and Xiaobo Zhang. 2002. “Growth, in Africa. London: McMillan. Inequality and Poverty in Rural China.� Washington, D.C.: IFPRI Escude, Guillermo, Tamara Burdisso, Marcelo Catena, Laudra Research Report 125. Available on line at www.ifpri.org. D’Amato, George McCandless, and Thomas E. Murphy. 2001. Fan, Shenggen, Xiaobo Zhang, and Neetha Rao. 2004. “Public Expen- “Las MIP y MES y el Mercado de Credito en la Argentina.� diture, Growth, and Poverty Reduction In Rural Uganda.� Washing- Buenos Aires: Banco Central de la República Argentina, Docu- ton, D.C.: IFPRI, Development Strategy and Governance Division mento de Trabajo 15. (DSGD), Discussion Paper 4. Available on line at www.ifpri.org. Eslava, Marcela, John Haltiwanger, Adriana Kugler, and Maurice Faulkender, Michael, and Mitchell A. Petersen. 2003.“Does the Source Kugler. 2003. “Have Market Reforms Changed the Evolution of of Capital Affect Capital Structure?� Cambridge, Mass.: National Productivity and Pro�tability in Colombia?� Paper presented at Bureau of Economic Research Working Paper Series 9930. the IASE–National Bureau of Economic Research Conference on Fearon, James D., and David Laitin. 2003. “Ethnicity, Insurgency “Productivity Dynamics�. Santiago de Chile. and Civil War.� American Political Science Review 91(1):75–90. ESMAP. 2002. Rural Electri�cation and Development in the Philip- Feder, Gershon, Tongroj Onchan, Yongyuth Chalamwong, and pines: Measuring the Social and Economic Bene�ts. Washington, Chira Hongladarom. 1988. Land Policies and Farm Productivity D.C.: Energy Assistance Management Assistance Programme. in Thailand. Baltimore: John Hopkins University Press. (c) The International Bank for Reconstruction and Development / The World Bank 220 WORLD DEVELOPMENT REPORT 2005 Feliciano, Zadia M. 1998. “Does Minimum Wage Affect Employ- Hundred Countries.� Washington, D.C.: World Bank Policy ment in Mexico?� Eastern Economic Journal 24(2):165–80. Research Working Paper Series 2521. Fernald, John G. 1999. “Roads to Prosperity? Assessing the Link Foster, Andrew D., and Mark R. Rosenzweig. 2004. “Agricultural Between Public Capital and Productivity.� American Economic Productivity Growth, Rural Economic Diversity and Economic Review 89(3):619–38. Reforms: India 1970–2000.� Economic Development and Cultural Fernandes, Ana M. 2003. “Trade Policy, Trade Volumes and Plant Change 52(3):509–42. Level Productivity in Colombian Manufacturing Industries.� Frankel, Jeffrey A. 2003. “The Environment and Globalization.� Washington, D.C.: World Bank Policy Research Working Paper Cambridge Mass.: National Bureau of Economic Research Work- Series 3064. ing Paper Series 10090. Fernandez, Raquel, and Dani Rodrik. 1991. “Resistance to Reform: Freeman, Richard. 1994. A Global Labor Market? Differences in Wages Status Quo Bias in the Presence of Individual-Speci�c Uncer- Among Countries in the 1980s. Washington, D.C.: World Bank. tainty.� American Economic Review 81(5):1146–53. Friedman, Joseph, Emmanuel Jimenez, and Stephen K. Mayo. 1988. Ferrarini, Benno. Forthcoming. “A Multilateral Framework for “The Demand for Tenure Security in Developing Countries.� Investment?� In Simon J. Evenett and Swiss Secretariat of Eco- Journal of Development Economics 29(2):185–98. nomic Affairs (eds.) The Singapore Issues and the World Trading Frost, Carol, Elizabeth Gordon, and Andrew Hayes. 2002. “Stock System: The Road to Cancun and Beyond. Bern: World Trade Exchange Disclosure and Market Liquidity: An Analysis of 50 Institute. International Exchanges.� Paper presented at the 13th Annual Feser, Edward. 2002. “The Relevance of Clusters for Innovation Pol- Conference on Financial Economics and Accounting joint with icy in Latin America and the Caribbean.� University of Califor- The 5th Maryland Finance Symposium. College Park, MD. nia, Background paper prepared for the World Bank, LAC group. November 15. Chapel Hill. Processed. Frye, Timothy, and Ekaterina Zhuravskaya. 2000. “Rackets, Regula- Field, Erica. 2002. “Entitled to Work: Urban Property Rights and tion, and the Rule of Law.� Journal of Law, Economics and Organi- Labor Supply in Peru.� Princenton, N.J.: Princeton University, zations 16(2):478–502. Princeton Law and Public Affairs Working Paper 02-1. Fuchs, Victor R, Alan B. Krueger, and James M. Poterba. 1998. Fields, Gary S., and Guy Pfeffermann, eds. 2003. Pathways Out of “Economist’s Views about Parameters, Values and Policies: Sur- Poverty. Washington, D.C.: World Bank. vey Results in Labor and Public Economics.� Journal of Economic Fink, Carsten, Aaditya Mattoo, and Randeep Rathindran. 2002. Literature 36(3):1387–425. “Liberalizing Basic Telecommunications: The Asian Experience.� Fukuyama, Francis. 1995. Trust: The Social Virtues and the Creation HWWA-Institut fïr Wirtschaftsforschung-Hamburg: HWWA of Prosperity. New York: Free Press. Discussion Paper 163. ————. 2001. “Social Capital, Civil Society and Development.� Fisman, Raymond. 2001. “Estimating the Value of Political Connec- Third World Quarterly 22(1):7–20. tions.� American Economic Review 91(4):1095–102. Fundación Paz Ciudadana. 2001. Plan Cuadrante: la Gran Reforma. Fjeldstad, Odd-Helge. 2002. “Fighting Fiscal Corruption: The Case Santiago de Chile: Fundación Paz Ciudadana. Available on line at of the Tanzania Revenue Authority.� Chr. Michelsen Institute. http://www.pazciudadana.cl/haciendo/doc/plan_cuadrante.PDF. Bergen, Norway. Processed. ————. 2002. Plan Comuna Segura, Compromiso 100. Santiago de Fleisig, Heywood, and Nuria De la Peña. 2003. Law, Legal Institu- Chile: Fundación Paz Ciudadana. Available on line at tions, and Development: Lessons of the 1990s for Property Rights, http://www.pazciudadana.cl/haciendo/doc/comuna_segura.PDF. Secured Transactions, Business Registration, and Contract Enforce- Funke, Norbert, and Saleh Nsouli. 2003. “The New Partnership for ment. Washington, D.C.: Center for the Economic Analysis of Africa’s Development (NEPAD): Opportunities and Challenges.� Law. Washington, D.C.: International Monetary Fund Working Paper Fleisig, Heywood W. 1998. How Problems in the Framework for 03/69. Secured Transactions Limit Access to Credit. Washington, D.C.: Gaiha, R. 2000. Rural Public Works and the Poor: A Review of the Center for the Economic Analysis of Law. Employment Guarantee Scheme in Maharashtra. India: University Fletcher, Kevin. 2002. “Tax Incentives in Cambodia, Laos PDR, and of Delhi, Faculty of Management Studies. Vietnam.� Paper presented at the IMF Conference on Foreign Galal, Ahmed. 2004. “The Economics of Formalization: Potential Direct Investment: Opportunities and Challenges for Cambodia, Winners and Losers from Formalization in Egypt.� Cairo: The Laos PDR and Vietnam. Hanoi. August 16. Egyptian Center for Economic Studies Working Paper 95. Flores, Gonzalo, and Andrei Mikhnew. 2004. “Improving the Busi- Galal, Ahmed, Leroy Jones, Pankaj Tandon, and Ingo Vogelsang. ness Environment for SMEs.� World Bank. Washington, D.C. 1994. Welfare Consequences of Selling Public Enterprises: An Processed. Empirical Analysis. New York: Oxford University Press. Fombad, Manga. 1999. “Curbing Corruption in Africa: Some Galanter, Marc, and Jayanth Krishnan. 2003. “Debased Informal- Lessons from Botswana’s Experience.� International Social Sci- ism: Lok Adalats and Legal Rights in Modern India.� In Erik G. ence Journal 51(2):241–54. Jensen and Thomas C. Heller, eds., Beyond Common Knowledge: Forteza, Alvaro, and Martín Rama. 2002. “Labor Market “Rigidity� Empirical Approaches to the Rule of Law. Stanford, C.A.: Stanford and the Success of Economic Reforms across More Than One University Press. (c) The International Bank for Reconstruction and Development / The World Bank References 221 Galindo, Arturo, and Margaret J. Miller. 2001. “Can Credit Reg- Glaeser, Edward L. 2001. “The Economics of Location-Based Tax istries Reduce Credit Constraints? Empirical Evidence on the Incentives.� Cambridge, Mass.: Harvard Institute of Economic Role of Credit Registries in Firm Investment Decisions.� Paper Research, Discussion Paper 1932. presented at the Annual Meetings of the Inter-American Devel- Glaeser, Edward L., and Andrei Shleifer. 2002. “Legal Origins.� opment Bank. Santiago de Chile. March 16. Quarterly Journal of Economics 117(4):1193–229. Gallup, John Luke, Jeffrey D. Sachs, and Andrew D. Mellinger. 1999. Glaeser, Edward L., Rafael La Porta, Florencio López-de-Silanes, “Geography and Economic Development.� International and Andrei Shleifer. 2004. “Do Institutions Cause Growth?� Regional Science Review 22(2):179–232. Cambridge, Mass.: National Bureau of Economic Research Gaston, Gelos R., and Shang-Jin Wei. 2002. “Transparency and Working Paper Series 10568. International Investor Behavior.� Cambridge, Mass.: National Goldberg, Pinelopi K., and Nina Pavcnik. 2003. “The Response of Bureau of Economic Research Working Paper Series 9260. the Informal Sector to Trade Liberalization.� Journal of Develop- Gauthier, Bernard, and Mark Gersovitz. 1997. “Revenue Erosion ment Economics 72(2):463–96. through Exemption and Evasion in Cameroon, 1993.� Journal of Golden, Miriam A. 1997. Heroic Defeats: The Politics of Job Loss. Public Economics 64(3):407–24. Cambridge, U.K.: Cambridge University Press. Gauthier, Bernard, and Ritva Reinikka. 2001. “Shifting Tax Burdens Golub, Stephen, and Ahmadou Aly Mbaye. 2002. “Obstacles and Through Exemptions and Evasion: An Empirical Investigation of Opportunities for Senegal’s International Competitiveness: Case Uganda.� Washington, D.C.: World Bank Policy Research Work- Studies of the Peanut Oil, Fishing and Textile Industries.� Wash- ing Paper Series 2735. ington, D.C.: World Bank Africa Region Working Paper Series Gaviria, Juan. 1998. “Port Privatization and Competition in 37. Colombia.� Washington, D.C.: World Bank, Public Policy for the Golub, Stephen S. 2003. “Measures of Restrictions on Inward For- Private Sector, Note 167. eign Direct Investment for OECD Countries.� Paris: Organiza- Ge, Wei. 1999. “Special Economic Zones and the Opening of the tion for Economic Co-operation and Development, Economics Chinese Economy: Some Lessons for Economic Liberalization.� Department Working Papers 357. World Development 27(7):1267–85. Gómez-Ibáñez, José A. 2003. Regulating Infrastructure: Monopoly, Georgian Opinion Research Business International (GORBI). 2002. Contracts, and Discretion. Cambridge, Mass.: Harvard University Legal Reform Monitoring Study. Tbilisi, Georgia: Georgian Opin- Press. ion Research Business International. Available on line at Gómez-Ibáñez, José A, and John R. Meyer. 1993. Going Private, The www.gorgi.com. International Experience with Transport Privatization. Washing- Ghatak, Maitreesh, and Timothy W. Guinnane. 1999. “The Eco- ton, D.C.: Brookings Institution. nomics of Lending with Joint Liability: Theory and Practice.� Gompers, Paul A., and Andrew Metrick. 2001. “Institutional Journal of Development Economics 60(1):195–228. Investors and Equity Prices.� Quarterly Journal of Economics Giannetti, Mariassunta. 2003. “Do better Institutions Mitigate 116(1):229–59. Agency Problems? Evidence from Corporate Finance Choices.� Gordon, Roger, and James R. Hines. 2002. “International Taxation.� Journal of Financial and Quantitative Analysis 38(1):185–212. In Alan J. Auerbach and Martin Feldstein, eds., Handbook of Pub- Gilbert, Richard J., and Edward P. Kahn, eds. 1996. International lic Economics, Vol. 4. Amsterdam: Elsevier B. V. Comparisons of Electricity Regulation. Cambridge, U.K.: Cam- Görg, H. 2002. “Fancy a Stay at the ‘Hotel California’? Foreign bridge University Press. Direct Investment, Taxation and Firing Costs.� Bonn: Institute Gill, Indermit S., William F. Maloney, and Carolina Sanchez- for the Study of Labor, Discussion Working Paper 665. Paramo. 2002. Trade Liberalization and Labor Reform in Latin Görg, Holger, and Eric Strobl. 2001. “Multinational Companies and America and the Caribbean in the 1990s. Washington, D.C.: Productivity Spillovers: A Meta-analysis.� Economic Journal World Bank. 111(475):F723–F739. Gill, Indermit S., Claudio E. Montenegro, and Dörte Dömeland. Graham Bannock and Partners Ltd. 1997. Credit Guarantee Schemes 2002. Crafting Labor Policy: Techniques and Lessons from Latin for Small Business Lending: A Global Perspective. London: Gra- America. New York: Oxford University Press. ham Bannock and Partners Ltd. Gill, Jit B. S. 2003. The Nuts and Bolts of Revenue Administration Graham, Carol, and Stefano Pettinato. 2001. “Frustrated Achievers: Reform. Washington, D.C.: World Bank. Winners, Losers, and Subjective Well Being in New Market Gillespie, Kate, Liesl Riddle, Edward Sayre, and David Sturges. Economies.� Washington, D.C.: Brookings Institution, Center on 1999. “Diaspora Interest in Homeland Investment.� Journal of Social and Economic Dynamics Working Paper Series 21. International Business Studies 30(3):623–34. Graham, Edward M. 2000. Fighting the Wrong Enemy: Anti-Global Gissurarson, Hannes H. 2000. “Over�shing: The Icelandic Solu- Activists and Multinational Enterprises. Washington, D.C.: Insti- tion.� Sussex: The Institute of Economic Affairs, Studies on the tute for International Economics. Environment 17. Greenstone, Michael. 2002. “The Impact of Environmental Regula- Glaeser, Edward L., Simon Johnson, and Andrei Shleifer. 2001. tions on Industrial Activity: Evidence from the 1970 and 1977 “Coase Versus the Coasians.� Quarterly Journal of Economics Clean Air Act Amendments and the Census of Manufactures.� 116(3):853–99. Journal of Political Economy 110(6):1175–1219. (c) The International Bank for Reconstruction and Development / The World Bank 222 WORLD DEVELOPMENT REPORT 2005 Greenstone, Michael, and Enrico Moretti. 2003. “Bidding for Hallward-Driemeier, Mary, and David Stewart. 2004. “How Do Industrial Plants: Does Winning a ‘Million Dollar Plant’ Increase Investment Climate Conditions Vary Across Countries and Welfare?� Cambridge, Mass.: National Bureau of Economic Types of Firms?� Background paper for the WDR 2005. Research Working Paper Series 9844. Hallward-Driemeier, Mary, and Andrew H. W. Stone. 2004. “The Greenwood, Peter W., Karyn E. Model, C. Peter Rydell, and James R. Investment Climate for Informal Firms.� Background paper for Chiesa. 1998. Diverting Children from a Life of Crime: Measuring the WDR 2005. Cots and Bene�ts. Santa Monica, CA: Rand Corporation. Hallward-Driemeier, Mary, L. Colin. Xu, and Scott Wallsten. 2003. Greif, Avner. 1989. “Reputation and Coalitions in Medieval Trade: “The Investment Climate and the Firm: Firm-Level Evidence Evidence on the Maghribi Traders.� Journal of Economic History from China.� Washington, D.C.: World Bank Policy Research 49(4):857–82. Working Paper Series 3003. Gruber, Jonathan. 1997. “Consumption Smoothing Effects of Unem- Haltiwanger, John. 2000. “Aggregate Growth: What Have we ployment Insurance.� American Economic Review 87(1):192–205. Learned from Microeconomic Evidence.� Paris: Organization for Guasch, J. Luis, and Robert W. Hahn. 1999. “The Costs and Bene�ts Economic Co-operation and Development Economics Depart- of Regulation: Implications for Developing Countries.� World ment Working Paper 267. Bank Research Observer 14(1):137–58. Haltiwanger, John, Stefano Scarpetta, and Milan Vodopivec. 2003. Guasch, J. Luis, and Pablo Spiller. 1999. Managing the Regulatory “How Institutions Affect Labor Market Outcomes: Evidence Process: Design, Concepts, Issues, and the Latin American and from Transition Economies.� Paper presented at the 2003 World Caribbean story. Washington, D.C.: World Bank. Bank Economists’ Forum. Washington, D.C. April 10. Gupta, S. P. 1999. Interim Report of the Study Group on Development Halvorsen, Robert. 1995. “Fiscal Incentives for Investment in Thai- of Small Enterprises. New Delhi: Planning Commission Village land.� In Anwar Shah, ed., Fiscal Incentives for Investment and and Small Industries Division, Government of India. Innovation. New York: Oxford University Press. Haber, Stephen, Armando Razo, and Noel Maurer. 2003. The Poli- Harris, Clive. 2002. “Private Rural Power: Network Expansion Using tics of Property Rights: Political Instability, Credible Commitments, an Output-Based Scheme in Guatemala.� Washington, D.C.: and Economic Growth in Mexico, 1876–1929. Cambridge, U.K.: World Bank, Private Sector and Infrastructure Network, Note 245. Cambridge University Press. ————. 2003. “Private Participation in the Infrastructure of Haddad, L., and M. Adato. 2001. “How Ef�cient do Public Works Developing Countries: Trends, Impacts, and Policy Lessons.� Programs Transfer Bene�ts to the Poor? Evidence from South Washington, D.C.: World Bank Working Paper 5. Africa.� Washington, D.C.: International Food Policy Research Harrison, Ann E. 1994. “Productivity, Imperfect Competition and Institute Discussion Paper 108. Trade Reform: Theory and Evidence.� Journal of International Hail, Luzi, and Christian Luez. 2003. “International Differences in Economics 36(1-2):53–73. Cost of Capital: Do Legal Institutions and Securities Regulation Harrison, Ann E., and Edward Leamer. 1997. “Labor Markets in Matter?� Brussels: European Corporate Governance Institute Developing Countries: An Agenda for Research.� Journal of Labor (ECGI)–Law Working Paper 15. Economics 15(3):S1–S19. Hall, Bronwyn H., and John Van Reenen. 1999. “How Effective are Harrold, Peter, Malathi Jayawickrama, and Deepak Bhattasali. 1996. Fiscal Incentives for R&D? A Review of the Evidence.� Research “Practical Lessons for Africa from East Asia in Industrial and Policy 29(4-5):449–69. Trade Policies.� Washington, D.C.: World Bank Africa Technical Hall, Robert E., and Charles I. Jones. 1999. “Why Do Some Coun- Department Series 310. tries Produce so much more Output per Worker than Others?� Harsch, Ernest. 2002. “Making Regional Integration a Reality: New Quarterly Journal of Economics 114(1):83–116. Strategies Aim for Public Involvement, Practical Results.� Africa Hallberg, Kristin, and Yasuo Konishi. 2003. “Bringing SMEs into Recovery 16(2-3):10–10. Global Markets.� In Gary S. Fields and Guy Pfeffermann, eds., Harsh, Vivek. 2003. Small Scale Industries and Reservation Policy. Pathways out of Poverty: Private Firms and Economic Mobility in Anand-Gujarat, India: Institute of Rural Management. Available Developing Countries. Boston, Mass.: Kluwer Adademic Publish- on line at http://www.irma.ac.in/pub/network/9-3doc2.html. ers for the International Finance Corporation. Hart, David M. 2003. “Entrepreneurship Policy: What It Is and Where Hallberg, Kristin. 2000. “A Market-Oriented Strategy For Small and It Came From.� In David M.Hart, ed., Emergence of Entrepreneur- Medium-Scale Enterprises.� Washington, D.C.: International ship Policy: Governance, Start-Ups, and Growth in the U.S. Knowl- Finance Corporation Discussion Paper 40. edge Economy. Cambridge, U.K.: Cambridge University Press. Hallward-Driemeier, Mary. 2003. “Do Bilateral Treaties Attract For- Harvey, Charles. 1991. “On the Perverse Effects of Financial Sector eign Direct Investment? Only a Bit...and They Could Bite.� World Reform in Anglophone Africa.� South African Journal of Econom- Bank. Washington, DC. Processed. ics 59(3):258–86. Hallward-Driemeier, Mary, Giuseppe Iarossi, and Kenneth L. Haufler, Andreas. 2001. Taxation in a Global Economy. Cambridge, Sokoloff. 2002. “Exports and Manufacturing Productivity in East U.K.: Cambridge University Press. Asia: A Comparative Analysis with Firm-Level Data.� Cam- Hausmann, Ricardo, Lant Pritchett, and Dani Rodrik. 2004. bridge, Mass.: National Bureau of Economic Research Working “Growth Accelerations.� Cambridge, Mass.: National Bureau Paper Series 8894. Economic Research Working Paper Series 10566. (c) The International Bank for Reconstruction and Development / The World Bank References 223 Hausmann, Ricardo, and Dani Rodrik. 2003. “Economic Develop- Hicks, John R. 1935. “Annual Survey of Economic Theory: The ment As Self-Discovery.� Journal of Development Economics Theory of Monopoly.� Econometrica 3(1):1–20. 72(2):603–33. Hines, James R. Jr. 1999. “Lessons from Behavioral Responses to Head, Keith, and John Reis. 1998. “Immigration and Trade Cre- International Taxation.� National Tax Journal 52(2):305–23. ation: Econometric Evidence from Canada.� Canadian Journal of Hnatkovska, Viktoria, and Norman A. Loayza. 2004. “Volatility and Economics 31(1):47–62. Growth.� Washington, D.C.: World Bank Policy Research Work- Heckman, James, and Carmen Pagés. 2004. Law and Employment: ing Paper Series 3184. Lessons from the Latin America and the Caribbean. Cambridge, Hoekman, Bernard. 2000. “The Next Round of Services Trade Mass.: National Bureau of Economic Research. Negotiations: Identifying Priorities and Options.� Federal Reserve Hellman, Joel S., Geraint Jones, Daniel Kaufmann, and Mark Bank of St.Louis Review 82(4):31–48. Schankerman. 1999. “Measuring Governance and State Capture: Hoekman, Bernard, Hiau Looi Kee, and Marcelo Olarreaga. 2001. The Role of Bureaucrats and Firms in Shaping the Business Envi- “Markups, Entry Regulations, and Trade: Does Country Size ronment.� London, U.K.: European Bank for Reconstruction and Matter?� Washington, D.C.: World Bank Policy Research Work- Development Working Paper 51. ing Paper Series 2662. Hellman, Joel S., and Daniel Kaufmann. 2003. “The Inequality of Hoekman, Bernard, Aadita Mattoo, and Philip English, eds. 2002. Influence.� World Bank. Washington, D.C. Processed. Development, Trade, and the WTO: A Handbook. Washington, Hempling, Scott, Kenneth Rose, and Robert E. Burns. 2004. The D.C.: World Bank. Regulatory Treatment of Embedded Costs Exceeding Market Prices: Hoekman, Bernard, and Petros C. Mavroidis. 2002. “Economic Transition to a Competitive Electric Generation Market. Colum- Development, Competition Policy, and the WTO.� Washington, bus, O.H.: National Regulatory Research Institute. D.C.: World Bank Policy Research Report Working Paper Series Henderson, David. 2000. The MAI Affair: A Story and its Lessons. 2917. London: Royal Institute of International Affairs. Hoekman, Bernard, Constantine Michalopoulos, and Alan L. Win- Henisz, Witold J. 2000. “The Institutional Environment for Eco- ters. 2003. “More Favorable and Differential Treatment of Devel- nomic Growth.� Economics and Politics 12(1):1–31. oping Countries: Toward a New Approach in the World Trade ————. 2002. “The Institutional Environment for Infrastructure Organization.� Washington, D.C.: World Bank Policy Research Investment.� Industrial and Corporate Change 11(2):355–89. Working Paper Series 3107. Henisz, Witold J., and A. Delios. 2003. “Policy Uncertainty and the Hoekman, Bernard, and others. 2004. Leveraging Trade for Develop- Sequence of Entry by Japanese Firms, 1980–1998.� Journal of ment: The World Bank Research Agenda. Washington, D.C.: International Business Studies 34(3):227–41. World Bank. Henisz, Witold J., and Bennet A. Zelner. Forthcoming. “Legitimacy, Hoff, Karla. 2003. “Paths of Institutional Development: A View Interest Group Pressures and Change in Emergent Institutions: from Economic History.� World Bank Research Observer The Case of Foreign Investors and Host Country Governments.� 18(2):205–26. Academy of Management Review. Hoff, Karla, and Joseph E. Stiglitz. 2004. “After the Big Bang? Obsta- Henisz, Witold J., and Bennet A. Zelner. 2001. “The Institutional cles to the Emergence of the Rule of Law in Post-Communist Environment for Telecommunications Investment.� Journal of Societies.� American Economic Review 94(3):753–63. Economics & Management Strategy 10(1):123–47. Hofstede, Geert. 1984. Culture’s Consequences: International Differ- Herbst, Jeffrey I. 2000. States and Power in Africa: Comparative ences in Work Related Values. Beverly Hills, C.A.: Sage Publications. Lessons in Authority and Control. Princeton, N.J.: Princeton Uni- Holmgren, Torgny, Louis Kasekende, Michael Atingi-Ego, and versity Press. Daniel Ddamulira. 2001. “Uganda.� In Shantayanan Devarajan, Hernandez, Zenaida. 2004. “The Debate on Industrial Policy in East David Dollar, and Torgny Holmgren, eds., Aid and Reform in Asia: In Search for Lessons.� Background paper for the WDR 2005. Africa: Lessons from ten Case Studies. Washington, D.C.: World Herrero, Alvaro, and Keith Henderson. 2001. “The Cost of Resolv- Bank. ing Small Business Conflicts: The Case of Peru.� Inter-American Holzmann, Robert, and Steer Jorgensen. 2001. “Social Risk Man- Development Bank, Sustainable Development Department, Best agement: A New Conceptual Framework for Social Protection Practices Series. Washington, D.C. Processed. and Beyond.� International Tax and Public Finance 8(4):529–56. Herzberg, Benjamin. 2004. “Investment Climate Reform: Going the Hopenhayn, Hugo. 2004. “Labor Market Policies and Employment Last Mile: The Bulldozer Initiative in Bosnia and Herzegovina.� Duration: The Effects of Labor Market Reform in Argentina.� In World Bank. Washington, D.C. Processed. James Heckman and Carmen Pagés, eds., Law and Employment: Hess, Ulrich, and Leora Klapper. 2003. “The Use of New Products, Lessons from the Latin America and the Caribbean. Cambridge, Processes and Technology for the Delivery of Rural and Micro�- Mass.: National Bureau of Economic Research. nance Loans in India.� World Bank. Washington, D.C. Processed. Hu, Albert G. Z, and Gary H. Jefferson. 2002. “FDI Impact and Hettige, Hemamala, Mainul Huq, Sheoli Pargal, and David Spillover: Evidence from China’s Electronic and Textile Indus- Wheeler. 1995. “Determinants of Pollution Abatement in Devel- tries.� World Economy 25(8):1063–76. oping Countries: Evidence from South East Asia.� World Devel- Hubka, Ashley, and Rita Zaidi. 2004. “Innovations in Micro�- opment 24(12):1891–904. nance.� Background paper for the WDR 2005. (c) The International Bank for Reconstruction and Development / The World Bank 224 WORLD DEVELOPMENT REPORT 2005 Hufbauer, Gary Clyde, and Ben Goodrich. 2003a. “Next Move in India–Ministry of Small Scale Industries. 2003. Annual Report Steel: Revocation or Retaliation.� Washington, D.C.: Institute for 2002–2003. New Delhi: India–Ministry of Small Scale Industries. International Economics, International Economics Policy Briefs Inglehart, Ronald and others. 2000. World Values Surveys and Euro- PB03-10. pean Values Surveys, 1981–1984, 1990–1993, and 1995–1997. Ann ————. 2003b. “Steel Policy: The Good, the Bad and the Ugly.� Arbor, MI: Inter University Consortium for Political and Social Washington, D.C.: Institute for International Economics, Inter- Research [distributor]. national Economics Policy Briefs PB03-1. Inter-American Development Bank. 2001. Economic and Social Hunt, Sally, and Graham Shuttleworth. 1996. Competition and Progress in Latin America 2001 Report. Washington, D.C.: Inter- Choice in Electricity. New York: John Wiley & Sons. American Development Bank. Hussain, Abid. 1997. Report of the Expert Committee on Small Enter- ————. 2002. Evaluation of MIF Projects: Alternative Commercial prises. New Delhi: Government of India, Ministry of Industry. Dispute Resolution Methods. Washington, D.C.: Inter-American ICSID. 2003. “Disputes before the Centre.� News from ICSID Development Bank, Of�ce of Evaluation and Oversight. 20(1):2–2. International Institute for Environment and Development. 2001. IDA, and IMF. 2003. Poverty Reduction Strategy Papers—Detailed From Abundance to Scarcity—The Closing of the Frontier: Analysis of Progress in Implementation. Washington, D.C.: Inter- Addressing Land Issues in West & Central Africa. London: Inter- national Development Association and International Monetary national Institute for Environment and Development, Drylands Fund. Program. IFC. 2001. Investing in Private Education: IFC’s Strategic Directions. Ireland–IDA. 2002. IDA Annual Report 2002: Enriching Ireland. Washington, D.C.: International Finance Corporation, Health Dublin: Industrial Development Agency. and Education Group. Irwin, Timothy C. 2004. “Accounting for Public-Private Partner- ————. 2002. The Private Sector Financing Activities of Interna- ships: How Should Governments Report Guarantees and Long- tional Financial Institutions (2001Update). Washington, D.C.: term Purchase Contracts.� Background paper for the WDR 2005. International Finance Corporation. Iyigun, Murat F., and Dani Rodrik. 2003. “On the Ef�cacy of IFC, and CIDA. 2001. Financial Leasing in Russia: Market Survey Reforms: Policy Tinkering, Institutional Change, and Entrepre- 2000–2001. Washington, D.C.: International Finance Corpora- neurship.� Cambridge, Mass.: National Bureau of Economic tion and Canadian International Development Agency. Research Working Paper Series 10455. ILO (International Labour Organisation). 2002a. “Compendium of Jadresic, Alejandro. 2000. “Promoting Private Investment in Rural Of�cial Statistics on Employment in the Informal Sector.� Geneva: Electri�cation: The Case of Chile.� Washington, D.C.: World International Labor Organization, STAT Working Paper 1. Bank Viewpoint Note 214. ————. 2002b. Women and Men in the Informal Economy: A Sta- Jaffe, Adam B., Steven R. Peterson, Paul R. Portney, and Robert N. tistical Picture. Geneva: International Labor Organization. Stavins. 1995. “Environmental Regulation and the Competitive- ————. 2003a. Employment and Social Policy in Respect of Export ness of U.S. Manufacturing: What Does the Evidence Tell Us?� Processing Zones (EPZs). Geneva: International Labor Organiza- Journal of Economic Literature 33(1):132–63. tion Governing Body. Jalan, Jyotsna, and Martin Ravallion. 2003. “Estimating the Bene�t ————. 2003b. Investing in Every Child—An Economic Study of Incidence of an Anti-Poverty Program by Propensity-Score the Costs and Bene�ts of Eliminating Child Labor. Geneva: Inter- Matching.� Journal of Business and Economic Statistics 21(1):19–30. national Labor Organization. Jappelli, Tullio, and Marco Pagano. 1999. “Information Sharing, ————. 2004. Global Employment Trends. Geneva: International Lending and Defaults: Cross-Country Evidence.� Salerno, Italy: Labor Organization. Centro Studi in Economia e Finanza, Working Paper 22. IMF (International Monetary Fund). 2003. Government Finance Jenkins, Glenn P., and Chun-Yan Kuo. 2000. “Promoting Export- Statistics. Washington, D.C.: International Monetary Fund. Oriented Foreign Direct Investment in Developing Countries: Tax and Customs.� Cambridge, Mass.: Harvard Institute for ————. 2004. World Economic Outlook. Washington, D.C.: Inter- International Development Consulting Assistance on Economic national Monetary Fund. Reform II 65. IMF, and IDA. 2003. Joint Staff Assessment of the Poverty Reduction Jhabvala, Renana, Ratna M. Sudarshan, and Jeemol Unni, eds. 2004. Strategy Paper Annual Progress Report. Washington, D.C.: Interna- Informal Economy Centre Stage: New Structures of Employment. tional Monetary Fund and International Development Association. New Delhi: Sage Publications India. Impavido, Gregorio. 2001. Assessment of Implementation of the IAIS Jimenez, Emmanuel. 1984. “Tenure Security and Urban Squatting.� Insurance Supervisory Principles. Washington, D.C.: World Bank. Review of Economic and Statistics 66(4):556–67. Impavido, Gregorio, Alberto R. Musalem, and Thierry Tressel. Joh, Sung Wook. 2003. “Corporate Governance and Firm Prof- 2003. “The Impact of Contractual Savings Institutions on Secu- itability: Evidence from Korea before the Economic Crisis.� Jour- rities Markets.� Washington, D.C.: World Bank Policy Research nal of Financial Economics 68(2):287–322. Working Papers Series 2948. Johnson, Simon, Peter Boone, Alasdir Breach, and Eric Friedman. India–Ministry of Finance. 2003. Economic Survey 2002–2003. New 2000. “Corporate Governance in the Asian Financial Crisis.� Delhi: India–Ministry of Finance. Journal of Financial Economics 58(1-2):141–86. (c) The International Bank for Reconstruction and Development / The World Bank References 225 Johnson, Simon, John McMillan, and Christopher Woodruff. ————. 2003. “The Political Economy of Public Spending Deci- 2002a. “Courts and Relational Contracts.� Journal of Law, Eco- sions in the Dominican Republic.� World Bank. Washington, nomics and Organization 18(1):221–77. D.C. Processed. ————. 2002b. “Property Rights and Finance.� American Eco- ————. 2004. “A Review of the Political Economy of Gover- nomic Review 92(5):1335–56. nance: From Property Rights to Voice.� Washington, D.C.: World Jones, Charles I. 2002. Introduction to Economic Growth. New York: Bank Policy Research Working Paper Series 3315. W. W. Norton and Company. Keefer, Philip, and Stuti Khemani. 2003. “Democracy, Public Jones, Leroy, and Il Sakong. 1980. Government, Business and Entre- Expenditures, and the Poor.� Washington, D.C.: World Bank Pol- preneurship in Economic Development: The Korean Case. Cam- icy Research Working Paper Series 3154. bridge, Mass.: Harvard University Press. Keller, Wolfgang. 2001. “International Technology Diffusion.� Cam- Jorgensen, Helle Bank, Michael Peder Pruzan-Jorgensen, Margaret bridge, Mass.: National Bureau of Economic Research Working Jungk, and Aaron Cramer. 2003. Strengthening Implementation of Paper Series 8573. Corporate Social Responsibility in Global Supply Chains. Washing- Keller, Wolfgang, and Arik Levinson. 2002. “Pollution Abatement ton, D.C.: World Bank and International Finance Corporation. Costs and Foreign Direct Investment Inflows to U.S. States.� Jovanovic, Boyan. 1995. “Learning and Growth.� Cambridge, Mass.: Review of Economics and Statistics 84(4):691–703. National Bureau of Economic Research Working Paper Series Keynes, John M. 1936. General Theory of Employment, Interest and 5383. Money. London: MacMillan and Co. Ltd. Kabeer, N. 2003. Gender Mainstreaming in Poverty Eradication and Khemani, R. S. 2002. “Competition Policy, Economic Adjustment the Millennium Development Goals: A Handbook for Policy-mak- and Competitiveness.� In Ijaz Nabi and Manjula Luthria, eds., ers and other Stakeholders. London: Commonwealth Secretariat. Building Competitive Firms: Incentives and Capabilities. Washing- Kagel, John H., and Alvin E. Roth, eds. 1995. The Handbook of ton, D.C.: World Bank. Experimental Economics. Princeton, N.J.: Princeton University Khemani, Stuti. 2004. “Political Cycles in a Developing Economy: Press. Effects of Elections in the Indian States.� Journal of Development Kahkonen, Satu, Young Lee, Patrick Meagher, and Haji Semboja. Economics 73(1):125–54. 2001. “Contracting Practices in an African Economy: Industrial Kikeri, Sunita. 1998. “Privatization and Labor: What Happens to Firms and Suppliers in Tanzania.� College Park, MD: University Workers when Governments Divest?� Washington, D.C.: World of Maryland, IRIS Center Working Paper 242. Bank Technical Paper 396. Kahneman, Daniel, and Amos Tversky, eds. 2000. Choices, Values Kim, Linsu. 1997. Imitation to Innovation: The Dynamics of Korea’s and Frames. Cambridge, UK: Press Syndicate of the University of Technological Learning. Cambridge, Mass.: Harvard Business Cambridge. School Press. Kaminsky, Graciela L., Carmen M. Reinhart, and Carlos A. Végh. King, Robert G., and Ross Levine. 1993. “Finance and Growth: 2003. “The Unholy Trinity of Financial Contagion.� Journal of Schumpeter Might be Right.� Quarterly Journal of Economics Economic Perspectives 17(4):51–74. 108(3):717–37. Kapur, Devesh. 2001. “Diasporas and Technology Transfers.� Jour- Kingdon, John W. 1995. Agendas, Alternatives, and Public Policies. nal of Human Development 2(2):265–86. New York: Longman. Katrak, Homi. 1999. “Small-Scale Enterprise Policy in Developing Kirkpatrick, Colin, and David Parker. 2003. “Regulatory Impact Countries: An Analysis of India’s Reservation Policy.� Journal of Assessment: Developing its Potential for Use in Developing International Development 11(5):701–15. Countries.� Manchester, UK: Centre on Regulation and Compe- Kaufmann, Daniel. 2002. “Rethinking Governance: Empirical tition, Working Paper Series 56. Lessons Challenge Orthodoxy.� World Bank. Washington, D.C. Klapper, Leora, and Elke Kraus. 2002. “Credit Information Infra- Processed. structure and Political Economy Issues.� World Bank. Washing- ————. 2003. “Governance Redux: The Empirical Challenge.� ton, D.C. Processed. World Bank. Washington, D.C. Processed. Klapper, Leora, Luc Laeven, and Raghuram G. Rajan. 2003. “Busi- Kaufmann, Daniel, Aart Kraay, and Massimo Mastruzzi. 2003. ness Environment and Firm Entry: Evidence from International “Governance Matters III: Governance Indicators for 1996–2002.� Data.� Cambridge, Mass.: National Bureau of Economic Washington, D.C.: World Bank Policy Research Report Series Research Working Paper Series 10380. 3106. Klapper, Leora F., and Inessa Love. Forthcoming. “Corporate Gov- Kay, John. 2003. The Truth About Markets: Their Genius, Their Lim- ernance, Investor Protection, and Performance in Emerging its, and Their Follies. London: Allen Lane. Markets.� Journal of Corporate Finance. Kee, Hiau Looi, and Bernard Hoekman. 2003. “Imports, Entry, and Klapper, Leora F., and Rida Zaidi. 2004. “A Survey of Government Competition Law as Market Disciplines.� Washington, D.C.: Regulation and Intervention in Financial Markets.� Background World Bank Policy Research Working Paper Series 3031. paper for the WDR 2005. Keefer, Philip. 2002. “The Political Economy of Corruption in Klasen, Stephan. 1999. “Does Gender Inequality Reduce Growth Indonesia.� World Bank. Washington, D.C. Processed. and Development? Evidence from Cross-country Regressions.� (c) The International Bank for Reconstruction and Development / The World Bank 226 WORLD DEVELOPMENT REPORT 2005 Washington, D.C.: World Bank Policy Research Report on Gen- Krueger, Anne. 1996. “Observations on International Labor Stan- der and Development Working Paper Series 7. dards and Trade.� Cambridge, Mass.: National Bureau of Eco- Klasen, Stephan, and Francesca Lamanna. 2003. “The Impact of nomic Research Working Paper Series 5632. Gender Inequality in Education and Employment on Economic ————, ed. 2000. Economic Policy Reform: The Second Stage. Growth in the Middle East and North Africa.� University of Chicago, IL: University of Chicago Press. Munich. Munich. Processed. Krugman, Paul. 1997. The Age of Diminished Expectations: U.S. Eco- Klasen, Stephan, and Ingrid Woolard. 2001. “Surviving Unemploy- nomic Policy in the 1990s. Cambridge, Mass.: MIT Press. ment without State Support: Unemployment and Household Krumm, Kathie, and Homi Kharas. 2004. “Overview.� In Kathie Formation in South Africa.� Bonn, Germany: Institute for the Krumm and Homi Kharas, eds., East Asia Integrates: A Trade Pol- Study of Labor, Discussion Working Paper 237. icy Agenda for Shared Growth. Washington, D.C.: World Bank Klein, Daniel B. 1992. “Promise Keeping in the Great Society: A and Oxford University Press. Model of Credit Information Sharing.� Economics and Politics Kubler, Dorothea. 2001. “On the Regulation of Social Norms.� Jour- 4(2):117–36. nal of Law, Economics, and Organization 17(2):449–76. Klein, Michael U., and Bita Hadjimichael. 2003. The Private Sector Kugler, Adriana. 2002. “From Severance Pay to Self-Insurance: in Development: Entrepreneurship, Regulation, and Competitive Effects of Severance Payments Savings Accounts in Colombia.� Disciplines. Washington, D.C.: World Bank. Bonn, Germany: Institute for the Study of Labor Discussion Klein, Michael U., and Neil Roger. 1994. “Back to the Future: The Paper 434. Potential in Infrastructure and Privatization.� In Richard Kugler, Adriana, and Giovanni Pica. 2003. “Effects of Employment O’Brian, ed., Finance and the International Economy Vol. 8: The Protection and Product Market Regulations on the Italian Labor AMEX Bank Review Prize Essays: In Memory of Robert Marjolin. Market.� Bonn, Germany: Institute for the Study of Labor, Dis- New York: Oxford University Press. cussion Paper 948. Klenow, Peter, and Andrés Rodríguez-Clare. 1997.“Economic Growth: Kugler, Maurice. 2001. “Externalities from Foreign Direct Invest- A Review Essay.� Journal of Monetary Economics 40(3):597–617. ment: The Sectoral Pattern of Spillovers and Linkages.� Univer- Klitgaard, Robert. 1998. “International Cooperation Against Cor- sity of Southampton. Southampton. Processed. ruption.� Finance and Development 35(1):3–6. La Porta, Rafael, Florencio López-de Silanes, Andrei Shleifer, and ————. 2000. “Subverting Corruption.� Finance and Develop- Robert Vishny. 1997. “Legal Determinants of External Finance.� ment 37(2):2–5. Journal of Finance 52(3):1131–50. Knack, Stephen, and Philip Keefer. 1995a. “Institutions and Eco- La Porta, Rafael, and Florencio López-de-Silanes. 2001. “Creditor nomic Performance: Cross-Country Tests Using Alternative Protection and Bankruptcy Law Reform.� In Stijn Claessens, Institutional Measures.� Economics and Politics 7(3):207–227. Simeon Djankov, and Ashoka Mody, eds., Resolution of Financial ————. 1995b. “Institutions and Economic Performance: Cross- Distress: An International Perspective on the Design of Bankruptcy Country Tests Using Alternative Institutional Measures.� Eco- Laws. Washington DC: World Bank. nomics and Politics 7(3):207–27. La Porta, Rafael, Florencio López-de-Silanes, and Andrei Shleifer. Kokko, Ari. 2002. “Globalization and FDI Incentives.� Paper pre- 2002. “Government Ownership of Banks.� Journal of Finance sented at the World Bank ABCDE–Europe Conference. Oslo, 57(1):265–301. Norway. June 22. La Porta, Rafael, Florencio López-de-Silanes, Andrei Shleifer, and Komives, Kristin, Dale Whittington, and Xun Wu. 2003. “Infra- Robert Vishny. 1998. “Law and Finance.� Journal of Political Econ- structure Coverage and the Poor: A Global Perspective.� In Tim- omy 106(6):1113–55. othy C. Irwin and Penelope J. Brook, eds., Infrastructure for Poor ————. 1999. “The Quality of Government.� Journal of Law, Eco- People: Public Policy for Private Provision. Washington, D.C.: nomics, and Organization 15(1):222–79. World Bank. Laevan, Luc, and Christopher Woodruff. 2003. “The Quality of the Kostrzeva, Karina. 2003. “New Quality of Law-drafting Process and Legal System, Firm Ownership, and Firm Size.� Washington, Public Administration in Poland Concerning the Harmonization D.C.: World Bank Policy Research Working Paper Series 3246. with EU Legislation.� Paper presented at the Regulatory Impact Laffont, Jean-Jacques, and Jean Tirole. 1991. “The Politics of Gov- Assessment Conference. American University of Bulgaria. Febru- ernment Decision Making: A Theory of Regulatory Capture.� ary 27. Quarterly Journal of Economics 106(4):1089–127. Kovacic, William E. 1997. “Getting Started: Creating New Competi- Lal, Sumir. 2004. “Can Good Economics Ever Be Good Politics? tion Policy Institutions in Transition Economies.� Brooklyn Jour- Case Study of the Power Sector in India.� World Bank. Wasing- nal of International Law 23(2):1197–225. ton, D.C. Processed. Kraay, Aart. 1999. “Exports and Economic Performance: Evidence Lall, Sanjaya. 2000. “Selective Industrial and Trade Policies in from a Panel of Chinese Enterprises.� Revue d’Economie du Developing Countries: Theoretical and Empirical Issues.� Developpement 0(1-2):183–207. Oxford, U.K.: Queen Elizabeth House Working Paper Series 48. ————. 2003. “What Can Cross-Country Regressions Tell Us ————. 2003. “Reinventing Industrial Strategy: The Role of About the Determinants of Pro-Poor Growth?� World Bank. Government Policy in Building Industrial Competitiveness.� Washington, D.C. Processed. Oxford, U.K.: Queen Elizabeth House Working Paper Series 111. (c) The International Bank for Reconstruction and Development / The World Bank References 227 Lamech, Ranjit, and Kazim Saeed. 2003. “What International Levy, Brian, and Pablo T. Spiller. 1994. “The Institutional Founda- Investors Look for When Investing in Developing Countries: tions of Regulatory Commitment: A Comparative Analysis of Results from a Survey of International Investors in the Power Telecommunications Regulation.� Journal of Law, Economics, and Sector.� Washington, D.C.: World Bank Energy and Mining Sec- Organization 10(2):201–46. tor Board Paper Series 6. ————, eds. 1996. Regulations, Institutions, and Commitment: Lanjouw, Jean O., and Philip I. Levy. 2002. “Untitled: A Study of Comparative Studies of Telecommunications. Cambridge, U.K.: Formal and Informal Property Rights in Urban Ecuador.� Eco- Cambridge University Press. nomic Journal 112(482):986–1019. Levy-Yeyati, Eduardo, María Soledad Martinez Peria, and Sergio L. Lanjouw, Peter, and A. Shariff. 1999. “Rural Poverty and Non-Farm Schmukler. 2004. “Market Discipline in Emerging Economies: Employment in India: Evidence from Survey Data.� Charles Uni- Beyond Bank Fundamentals.� In William C. Hunter, George G. versity. Prague. Processed. Kaufman, Claudio Borio, and Kostas Tsatsaronis, eds., Market Lanjouw, Peter, and Nicholas Stern. 1998. Economic Development in Discipline across Countries and Industries. Cambridge, Mass.: Palanpur over Five Decades. Oxford and New York: Oxford Uni- MIT Press. versity Press. Lewis, William W. 2004. The Power of Productivity: Wealth, Poverty Lawson, Cina, and Natalie Meyenn. 2000. “Bringing Cellular Phone and the Threat to Global Stability. Chicago: University of Chicago Service to Rural Areas: Grameen Telecom and Village Pay phones Press. in Bangladesh.� Washington, D.C.: World Bank Public Policy for Li, Hongyi, Lyn Squire, and Heng-fu Zou. 1998. “Explaining Inter- the Private Sector Note 205. national and Intertemporal Variations in Income Inequality.� Lax, David A., and James K. Sebenis. 1986. The Manager as Negotia- Economic Journal 108(406):26–43. tor: Bargaining for Co-operation and Competitive Gain. New Li, Shaomin, Shuhe Li, and Weiying Zhang. 1999. “Cross-regional York: Free Press. Competition and Privatization in China.� MOCT-MOST 9(1):75–8. Lederman, Daniel, and William F. Maloney. 2003. “R&D and Devel- Liautaud, Gerard. 2001. “Maintaining Roads: Experience with Out- opment.� Washington, D.C.: World Bank Policy Research Work- put-Based Contracts in Argentina.� In Penelope J. Brook and ing Paper Series 3024. Suzanne M. Smith, eds., Contracting for Public Services: Output- Lederman, Daniel, and Laura Saenz. 2003. “Innovation and Devel- based Aid and Its Applications. Washington, D.C.: World Bank. opment Around the World 1960–2000.� World Bank. Washing- Libecap, Gary D. 1994. Contracting for Property Rights. Cambridge, ton, D.C. Processed. U.K..: Cambridge University Press. Lee, Charles, and David Ng. 2002. “Corruption and International Lifsher, Marc. 2001. “If He Can Fight Crime There, He’ll Fight It Valuation: Does Virtue Pay?� Cornell University. Cornell. Anywhere.� Wall Street Journal, March 8. Page: A18. Processed. Limão, Nuno, and Anthony J. Venables. 2001. “Infrastructure, Geo- Lee, Norman. 2002. “Developing and Applying Regulatory Impact graphical Diadvantage, Tansport Costs, and Trade.� World Bank Assessment Methodologies in Low and Middle Income Coun- Economic Review 15(3):451–79. tries.� Manchester, U.K.: Centre on Regulation and Competition, Lin, Justin Yifu. 1992. “Rural Reforms and Agricultural Growth in Working Paper Series 30. China.� American Economic Review 82(1):34–51. Lee, Sang M., and Suzzane J. Peterson. 2000. “Culture, Entrepre- Lipsey, Richard, and Kevin Lancaster. 1956. “The General Theory of neurial Orientation, and Global Competitiveness.� Journal of Second Best.� Review of Economic Studies 25(1):11–32. World Business 35(4):401–16. List, John A., Daniel L. Millimet, Per G. Fredricksson, and W. War- Leite, Carlos, and Jens Weidmann. 1999. “Does Mother Nature Cor- ren McHone. 2003. “Effects of Environmental Regulations on rupt? Natural Resources, Corruption, and Economic Growth.� Manufacturing Plant Births: Evidence from a Propensity Score Washington, D.C.: International Monetary Fund Working Paper Matching Estimator.� Review of Economics and Statistics WP/99/85. Available on line at 85(4):944–52. http://www.imf.org/external/pubs/cat/wp1_sp.cfm?s_year=1997 Litvin, Daniel. 2003. Empires of Pro�t: Commerce, Conquest and &e_year=2001&brtype=default. Corporate Responsibility. New York: Texere. Levi, Margaret. 1988. Of Rule and Revenue. Berkeley: University of Liu, Jin-Tan, Meng-Wen Tsou, and James K. Hammitt. 1999. California Press. “Export Activity and Productivity: Evidence from the Taiwan Levine, Ross. 1997. “Financial Development and Economic Electronics Industry.� Weltwirtschaftliches Archiv 135(4):675–91. Growth: Views and Agenda.� Journal of Economic Literature Lobo, Albert, and Suresh Balakrishnan. 2002. Report Card on Ser- 35(2):688–726. vice of Bhoomi Kiosks: An Assessment of Bene�ts by Users of the Levine, Ross, Norman Loayza, and Thorsten Beck. 2000. “Financial Computerized Land Records System in Karnataka. Bangalore, Intermediation and Growth: Causality and Causes.� Journal of India: Public Affairs Centre. Monetary Economics 46(1):31–77. Londoño, Juan Luis, and Rodrigo Guerrero. 2000. “Violencia en Levinsohn, James. 1993. “Testing the Imports-as-Market-Discipline America Latina: Epidemiología y Costos.� In Rodrigo Guerrero, Hypothesis.� Journal of International Economics 35(1-2):1–22. Alejandro Gaviria, and Juan Luis Londoño, eds., Asalto al Desar- Levitt, Arthur. 1998. “The Importance of High Quality Accounting rollo: Violencia en América Latina. Washington, D.C.: Inter- Standards.� Accounting Horizons 12:79–82. American Development Bank. (c) The International Bank for Reconstruction and Development / The World Bank 228 WORLD DEVELOPMENT REPORT 2005 López, Humberto. 2003. Pro Growth, Pro Poor: Is there a Trade off? from Latin America and the Caribbean. Cambridge, Mass.: Washington, D.C.: World Bank. National Bureau of Economic Research. Love, Inessa, and Nataliya Mylenko. 2003. “Credit Reporting and Maloney, William F., and Eduardo Pontual Ribeiro. 2001. “Employ- Financing Constraints.� Washington, D.C.: World Bank Policy ment and Wage Effects of Mexican Unions—A Case of Extreme Research Working Paper Series 3142. Ef�cient Bargaining.� World Bank. Washington, D.C. Processed. Lovei, Laszlo, and Alastair McKechnie. 2000. “The Costs of Corrup- Malthus, Thomas. 1798. Essays on the Principle of Population. Lon- tion for the Poor.� Washington, D.C.: World Bank Public Policy don: J. Johnson. for the Private Sector Note 207. Management Systems International. 1996. Case Study of Impact on Lumpkin, G. T., and Gregory G. Dess. 1996. “Clarifying the Entre- Sub-Borrower Enterprises in Tow PRE Projects, Philippines. Wash- preneurial Orientation Construct and Linking It to Perfor- ington, D.C.: Of�ce of Program Review, Bureau of Private Enter- mance.� Academy of Management Review 21(1):135–72. prise, USAID. Lund, Frances, and Caroline Skinner. 2004. “The Investment Cli- Mani, Sunil. 2001a. “Globalization, Markets for Technology and the mate for the Informal Economy: A Case of Durban, South Relevance of Innovation Policies in Developing Countries.� Africa.� Background paper for the WDR 2005. Nairobi: African Technology Policy Studies Network Special Luthria, Manjula, and Keith E. Maskus. 2004. “Protecting Industrial Paper 2. Inventions, Authors’ Rights, and Traditional Knowledge: Rele- ————. 2001b. “Role of Government in Promoting Innovation vance, Lessons, and Unresolved Issues.� In Keith E. Krumm and in the Enterprise Sector: An Analysis of the Indian Experience.� Homi Kharas, eds., East Asia Integrates: A Trade Policy for Shared Maastricht, The Netherlands: United Nations University Insti- Growth. Washington, D.C.: World Bank and Oxford University tute for New Technologies Discussion Paper 2001-3. Press. Martin, John P., and David Grubb. 2001. “What Works and for MacIsaac, Donra, and Martin Rama. 1997. “Determinants of Whom: A Review of OECD Countries’ Experiences with Active Hourly Earnings in Ecuador: The Role of Labor Market Regula- Labour Market Policies.� Swedish Economic Policy Review tions.� Journal of Labor Economics 15(3):136–65. 18(2):9–56. MacIsaac, Donra, and Martin Rama. 2001. “Mandatory Severance Martin, Will, and Keith E. Maskus. 2001. “Core Labor Standards Pay: Its Coverage and Effects in Peru.� Washington, D.C.: World and Competitiveness: Implications for Global Trade Policy.� Bank Policy Research Working Paper Series 2626. Review of International Economics 9(2):317–28. Macours, Karen. 2003. “Insecurity of Property Rights and Matching Martinez Peria, María Soledad, and Sergio Schmukler. 2001. “Do in Land Rental Markets in Latin America.� PhD thesis. University Depositors Punish Banks for ‘Bad’ Behaviour?: Market Disci- of California at Berkeley. pline, Deposit Insurance, and Banking Crisis?� Journal of Finance Madani, Dorsati. 1999. “A Review of the Role and Impact of Export 56(3):1029–51. Processing Zones.� Washington, D.C.: World Bank Policy Maskus, Keith E. 1997. “Should Core Labor Standards be Imposed Research Working Paper Series 2238. Through International Trade Policy?� Washington, D.C.: World Maddison, Angus. 1995. Monitoring the World Economy. Paris: Bank Policy Research Working Paper Series 1817. Organization for Economic Co-operation and Development. ————. 2002. Bene�ting from Intellectual Property Protection. ————. 2003. The World Economy: Historical Statistics. Paris: Washington, D.C.: World Bank. Organization for Economic Co-operation and Development. Mattei, Ugo. 2000. Basic Principles of Property Law: A Comparative Maggi, Giovanni, and Andres Rodríguez-Clare. 1998. “The Value of Legal and Economic Introduction. Westport, Conn: Greenwood Trade Agreements in the Presence of Political Pressures.� Journal Press. of Political Economy 106(3):574–601. Matusz, Steven J., and David Tarr. 1999. “Adjusting to Trade Policy Magno, Marife T., and Richard L. Meyer. 1988. “Guarantee Reform.� Washington, D.C.: World Bank Policy Research Work- Schemes: An Alternative to the Supervised Credit Program.� ing Paper Series 2142. Paper presented at the Financial Intermediation in the Rural Sec- Max�eld, Sylvia, and Ben Ross Schneider. 1997. Business and the tor: Research Results and Policy Issues Conference. Manila, State in Developing Countries. Ithaca, N.Y.: Cornell University Philippines. September 26. Press. Majone, Giandomenico. 1996. “Temporal Consistency and Policy Mazza, Jacqueline. 1999. “Unemployment Insurance: Case Studies Credibility: Why Democracies Need Non-Majoritarian Institu- and Lessons for Latin America and the Caribbean.� Washington, tions.� Florence: European University Institute Robert Schu- D.C.: Inter-American Development Bank, Technical Study mann Center Working Paper 57. RE2/SO2. Mallon, Raymond. 2004. “Managing Investment Climate Reforms: McArthur, John W., and Jeffrey D. Sachs. 2001. “Institutions and Vietnam Case Study.� Background paper for the WDR 2005. Geography: Comment on Acemoglu, Johnson and Robinson Maloney, William F. 2004. “Informality Revisited.� World Develop- (2000).� Cambridge, Mass.: National Bureau of Economic ment 32(7):1159–78. Research Working Paper Series 8114. Maloney, William F., and Jairo Núñez. 2004. “Measuring the Impact McCubbins, Marthew, and Arthur Lupia. 1998. “Political Credibil- of Minimum Wages: Evidence from Latin America.� In James ity and Economic Reform.� World Bank. Washington, D.C. Heckman and Carmen Pagés, eds., Law and Employment: Lessons Processed. (c) The International Bank for Reconstruction and Development / The World Bank References 229 McDonald, G. J. 1994. “Testimony Given to the U.S. Senate.� In Milgrom, Paul R., Douglas C. North, and Barry R. Weingast. 1990. Ronald G. Slaby, Renée Wilson-Brewer, and Kimberly Dash, eds., “The Role of Institutions in the Revival of Trade: The Law Mer- Aggressors, Victims and Bystanders: Thinking and Acting to Pre- chant, Private Judges, and the Champagne Fairs.� Economics and vent Violence. Newton, Mass.: Education Development Center. Politics 2(1):1–23. McGrath, Rita G., Ian C. MacMillan, and Sari Scheinberg. 1992. Miller, Danny. 1983. “The Correlates of Entrepreneurship in Three “Elitists, Risk-Takers, and Rugged Individualists? An Exploratory Types of Firms.� Management Science 29(7):770–91. Analysis of Cultural Differences Between Entrepreneurs and Miller, Danny, and Peter H. Friesen. 1982. “Innovation in Conserv- Non-Entrepreneurs.� Journal of Business Venturing 7(2):115–135. ative and Entrepreneurial Firms: Two Models of Strategic McKinsey & Company. 2002. McKinsey Global Investor Opinion Survey Momentum.� Strategic Management Journal 3(1):1–25. on Corporate Governance. Washington, D.C.: McKinsey & Company. Miller, Margaret. 2003. “Credit Reporting Systems around the McKinsey Global Institute. 2003. New Horizons: Multinational Globe: The State of the Art in Public Credit Registries and Pri- Company Investment in Developing Economies. San Francisco, vate Credit Reporting Firms.� In Margaret Miller, ed., Credit C.A.: McKinsey & Company. Reporting Systems and the International Economy. Cambridge, McMillan, John. 2002. Reinventing the Bazaar: A Natural History of Mass.: MIT Press. Markets. New York: W.W. Norton and Company. Minot, Nicholas, and Francesco Goletti. 2000. Rice Market Liberal- McMillan, John, and Christopher Woodruff. 2002. “The Central ization and Poverty in Vietnam. Washington, D.C.: International Role of Entrepreneurs in Transition Economies.� Journal of Eco- Food Policy Research Institute, Research Report 114. nomic Perspectives 16(3):153–70. Miralles, Dennis R. 2002. “Government Administrative Practices as McMillan, Margaret, Dani Rodrik, and Karen Horn Welch. 2002. Obstacles in Investment.� Paper presented at the APEC Seminar “When Economic Reform Goes Wrong: Cashews in Mozam- on Investments’ One Stop Shop. Lima, Peru. bique.� Cambridge, Mass.: National Bureau of Economic Mitra, Pradeep, and Nicholas Stern. 2003. “Tax Systems in Transi- Research Working Paper Series 9117. tion.� Washington, D.C.: World Bank Policy Research Working Mead, Donald C., and Carl Liedholm. 1998. “The Dynamics of Paper Series 2947. Micro and Small Enterprises in Developing Countries.� World Mitullah, Winnie V. 2004. “Street Vending in African Cities: A Syn- Development 26(1):61–74. thesis of Empirical Findings From Kenya, Cote D’ivoire, Ghana, Megginson, William L., and Jeffry M. Netter. 2001. “From State to Zimbabwe, Uganda and South Africa.� Background paper for the Market: A Survey of Empirical Studies on Privatization.� Journal WDR 2005. of Economic Literature 39(2):321–89. Mnookin, Robert, and William Kornhauser. 1979. “Bargaining in Mendoza, Enrique G., and Linda L. Tesar. 2003.“Winners and Losers of the Shadow of the Law: The Case of Divorce.� Yale Law Journal Tax Competition in the European Union.� Cambridge, Mass.: 88:950–97. National Bureau of Economic Research Working Paper Series 10051. Mockus, Antanas. 2002. “Cultura Ciudadana: Programa Contra la Menezes Filho, N., H. Zylberstajn, J. P. Chahad, and E. Pazello. 2002. Violencia en Santa Fe de Bogotá, Colombia, 1995–1997.� Wash- “Unions and the Economic Performance of Brazilian Establish- ington, D.C.: Inter-American Development Bank, Sustainable ments.� Washington, D.C.: Inter-American Development Bank, Development Department, Technical Paper 127. Research Network Working Paper R464. Mody, Ashoka. 1999.“Industrial Policy After the East Asian Crisis: From Messick, Richard. 1999. “Reducing Court Delays: Five Lessons from Outward Orientation to New Internal Capabilities?� Washington, the United States.� Washington, D.C.: World Bank, Poverty Reduc- D.C.: World Bank Policy Research Working Paper Series 2112. tion and Economic Management Network (PREM) Note 34. Mondino, Guillermo, and Silvia Montoya. 2004. “The Effects of Micklethwait, John, and Adrian Wooldridge. 2003. The Company: A Labor Market Regulations on Employment Decisions by Firms: Short History of a Revolutionary Idea. New York: Modern Library. Empirical Evidence for Argentina.� In James Heckman and Car- Middleton, John, Adrian Ziderman, and Arvil van Adams. 1993. men Pagés, eds., Law and Employment: Lessons from the Latin Skills for Productivity: Vocational Education and Training in America and the Caribbean. Cambridge, Mass. and Chicago, I.L.: Developing Countries. Washington, D.C.: World Bank. National Bureau of Economic Research and University of Chicago. Midlarsky, Manus I. 1999. The Evolution of Inequality: War, State Montenegro, Claudio E., and Carmen Pagés. 2004. “Who Bene�ts Survival, and Democracy in Comparative Perspective. Stanford, from Labor Market Regulations? Chile 1960–1998.� In James C.A.: Stanford University Press. Heckman and Carmen Pagés, eds., Law and Employment: Lessons from the Latin America and the Caribbean. Cambridge, Mass.: MIGA (Multilateral Investment Guarantee Agency). 2002. Foreign National Bureau of Economic Research. Direct Investment Survey. Washignton, D.C.: Multilateral Invest- ment Guarantee Agency. Montiel, Peter, and Carmen M. Reinhart. 1999. “Do Capital Con- trols and Macroeconomic Policies Influence the Volume and Migliorisi, Stefano, and Marco Galmarini. 2004. “Donor Assistance Composition of Capital Flows? Evidence from the 1990s.� Jour- to Investment Climate Reforms.� Background paper for the nal of International Money and Finance 18(4):519–35. WDR 2005. Moore, Mick. 1998. “Death Without Taxes: Democracy, State Miles, Rebecca. 2002. “Employment and Unemployment in Jordan: Capacity, and Aid Dependence in The Fourth World.� In Mark the Importance of the Gender System.� World Development Robinson and Gordon White, eds., The Democratic Developmen- 30(3):413–27. (c) The International Bank for Reconstruction and Development / The World Bank 230 WORLD DEVELOPMENT REPORT 2005 tal State: Political and Institutional Design. New York: Oxford Mwangi, Anthony. 2003. “Final Report on the Survey on Customs University Press. Reform and Modernization in Mozambique.� World Bank. Moran, Theodore H. 1998. Foreign Direct Investment and Develop- Washington, D.C. Processed. ment. Washington, D.C.: Institute for International Economics. Mwangi, Ester, Paul Ongugo, and Jane Njuguna. 2000. “Decentral- ————. 2001. Parental Supervision: The New Paradigm for For- izing Institutions for Forest Conservation in Kenya: Comparative eign Direct Investment and Development. Washington, D.C.: Insti- Analysis of Resource Conservation Outcomes under National tute for International Economics. Park and Forest Reserve Regimes in Mt. Elgon Forest Ecosystem.� ————. 2002. Beyond Sweatshops: Foreign Direct Investment and Paper presented at the 8th Conference of the International Asso- Globalization in Developing Countries. Washington, D.C.: Brook- ciation for the Study of Common Property. Bloomington, Indi- ings Institution. ana. May 31. Morduch, Jonathan. 1997. “The Micro�nance Revolution.� Harvard Narayan, Deepa, Robert Chambers, Meera Kaul Shah, and Patti University. Cambridge, Mass. Processed. Petesch. 2000. Voices of the Poor: Crying Out for Change. Wash- ington, D.C.: World Bank. ————. 1999a. “Between the State and the Market: Can Informal Insurance Patch the Safety Net?� World Bank Research Observer Nathan Associates Inc. 2003. Intellectual Property and Developing 14(2):187–207. Countries: An Overview. Arlington, VA: Nathan Associates Inc. ————. 1999b. “The Micro�nance Promise.� Journal of Economic Navarrete, Camilo. 2004. “Managing Investment Climate Reforms: Literature 37(4):1569–614. Colombian Ports Sector Reform Case Study.� Background paper for the WDR 2005. Morduch, Jonathan, Elizabeth Little�eld, and Syed Hashemi. 2003. “Is Micro�nance an Effective Strategy to Reach the Millennium Nestoriak, Nicole. 2004. “Endogenous Technology and Local Labor Development Goals?� Washington, D.C.: CGAP Focus Note 24. Market Skill.� University of Maryland. College Park, M.D. Processed. Morisset, Jacques. 2003a. “Does a Country Need a Promotion Agency to Attract Foreign Direct Investment? A Small Analytical Newbery, David, and Michael Pollitt. 1997. “The Restructuring and Model Applied to 58 Countries.� Washington, D.C.: World Bank Privatization of Britain’s Central Electricity Generating Board Policy Research Working Paper Series 3028. (CEGB)—Was it Worth it?� Journal of Industrial Economics 45(3):269–303. ————. 2003b. “Tax Incentives: Using Tax Incentives to Attract Foreign Direct Investment.� Washington, D.C.: World Bank Pri- Newell, Richard G., James N. Sanchirico, and Suzi Kerr. 2002. “An vate Sector and Infrastructure Network Viewpoint Note 253. Empirical Analysis of New Zealand’s ITQ Markets.� Paper pre- sented at the Fisheries on the Global Economy (IIFET) Confer- Morisset, Jacques, and Kelly Andrews-Johnson. 2003. “The Effec- ence. Wellington. August 19. tiveness of Promotion Agencies at Attracting Foreign Direct Investment.� Washington, D.C.: Foreign Investment Advisory Nicholls, J. 1998. “Student Financing in the Developing World: Service Occasional Paper 16. Applying Income-contingent Approaches to Cost Recovery.� University of Melbourne, Department of Science and Maths Morris, Fiorina, and Kenneth Shepsle. 1990. “A Positive Theory of Education. Melbourne, Australia. Processed. Negative Voting.� In John A. Ferejohn and James J. Kuklinski, eds., Information and Democratic Processes. Urbana,I.L.: Univer- Nicholson, Michael W. 2003. “Quantifying Antitrust Regimes.� Fed- sity of Illinois Press. eral Trade Commission. Washington, D.C. Processed. Morris, Sebastian, Rakesh Basant, Keshab Das, K. Ramachandran, Nickell, Stephen, and Richard Layard. 1999. “Labor Market Institu- and Abraham Koshy. 2001. The Growth and Transformation of tions and Economic Performance.� In Orley Ashenfelter and David Small Firms in India. New Delhi: Oxford University Press. Card, eds., Handbook of Labor Economics. Amsterdam: Elsevier B. V. Muendler, Marc-Andreas. 2002. “Trade, Technology, and Produc- Nicoletti, Giuseppe, Robert Haffner, Stephen Nickell, Stefano Scar- tivity: A Study of Brazilian Manufacturers.� University of Cali- petta, and Gyl� Zoega. 2001. “European Integration, Liberaliza- fornia, San Diego. San Diego, CA. Processed. tion and Labor-Market Performance.� In Giuseppe Bertola, Tito Boeri, and Giuseppe Nicoletti, eds., Welfare and Employment in a Mukherjee, Amitabba. 2002. “Lessons from Armenia’s Institutional United Europe. Cambridge, Mass.: MIT Press. and Governance Review.� Washington, D.C.: World Bank, Public Sector, PREM Note 76. Nicoletti, Giuseppe, and Stefano Scarpetta. 2003. “Regulation, Pro- ductivity and Growth: OECD Evidence.� Economic Policy Muller, Jerry Z. 2002. The Mind and the Market: Capitalism in Mod- 18(36):9–51. ern European Thought. New York: Anchor Books. Njoku, J. E., and P. C. Obasi. 1991. “Loan Repayment and its Deter- Mulligan, Casey. 2002. “Capital Tax Incidence: First Impressions minants Under the Agricultural Credit Guarantee Scheme in from the Time Series.� Cambridge, Mass.: National Bureau of Imo State, Nigeria.� Savings and Development 2:167–80. Economic Research Working Paper Series 9374. Noland, Marcus, and Howard Pack. 2003. Industrial Policy in an Era Murdoch, James C., and Todd Sandler. 2002. “Economic Growth, of Globalization: Lessons from Asia. Washington, D.C.: Institute Civil Wars, and Spatial Spillovers.� Journal of Conflict Resolution for International Economics. 46(1):91–110. North, Douglass C. 1993. “Institutions and Credible Commit- Murphy, Kevin M., Andrei Shleifer, and Robert W. Vishny. 1993. ment.� Journal of Institutional and Theoretical Economics “Why is Rent-Seeking so Costly to Growth?� American Economic 149(1):11–23. Review 83(2):409–14. (c) The International Bank for Reconstruction and Development / The World Bank References 231 North, Douglass C., and Barry R. Weingast. 1989. “Constitutions ————. 2003e. Public Sector Transparency and International and Commitment: The Evolution of Institutions Governing Investment Policy. Paris: Organization for Economic Co-opera- Public Choice in Seventeenth-century England.� Journal of Eco- tion and Development. nomic History 49(4):803–32. ————. 2003f. Tax Incentives for Research and Development: Norton, Seth W. 2002. “Population Growth, Economic Freedom, Trends and Issues. Paris: Organization for Economic Co-opera- and the Rule of Law.� Montana: The Property and Environment tion and Development. Research Center (PERC) Policy Series 24. Available on line at ————. 2003g. The E-Government Imperative. Paris: Organiza- http://www.perc.org/pdf/ps24.pdf. tion for Economic Cooperation and Development. Nugent, Jeffrey B., and Seung-Jae Yhee. 2002. “Small and Medium OECD, and WTO. 2003. Second Joint WTO/OECD Report on Trade- Enterprises in Korea: Achievements, Constraints and Policy related Technical Assistance and Capacity Building. Paris and Issues.� Small Business Economics 18(1-3):85–119. Geneva: World Trade Organization and Organization for Eco- O’Malley, Chris. 2004. “New Hope for a United Base.� The Indi- nomic Co-operation and Development. anapolis Star, April 30. OECD DAC. 2001. DAC Recommendation on Untying Of�cial Oates, Wallace E. 2001. “Fiscal Competition and Harmonization? Development Assistance to the Least Developed Countries. Paris: Some Reflections.� National Tax Journal 54(3):507–12. Organization for Economic Co-operation and Development. OECD. 1997a. Employment Outlook—1997. Paris: Organization for Development Assistance Committee. Economic Co-operation and Development. ————. 2003. Development Cooperation Report 2002. Paris: ————. 1997b. Regulatory Reform. Paris: Organization for Eco- Organization for Economic Co-operation and Development. nomic Co-operation and Development. Development Assistance Committee. ————. 1999a. Regulatory Reform in Japan. Paris: Organization ————. 2004. Development Cooperation Report 2003. Paris: for Economic Co-operation and Development. Organization for Economic Co-operation and Development. ————. 1999b. Regulatory Reform in Mexico. Paris: Organization Development Assistance Committee. for Economic Co-operation and Development. Olivier de Sardan, J. P. 1999. “A Moral Economy of Corruption in ————. 2000a. International Trade and Core Labour Standards. Africa.� Journal of Modern African Studies 37(1):25–52. Paris: Organization for Economic Co-operation and Develop- Olofsgård, Anders F. 2004. “The Political Economy of Reform: ment. Institutional Change as a Tool for Political Credibility.� Back- ————. 2000b. Regulatory Reform in Korea. Paris: Organization ground paper for the WDR 2005. for Economic Co-operation and Development. Omran, Mohammed. 2003. “Privatization, State-Ownership, and ————. 2001. Codes of Corporate Conduct: Expanded Review of Bank Performance in Egypt.� Paper presented at the Bank Priva- their Contents. Paris: Organization for Economic Co-operation tization, World Bank Conference. Washington, D.C. November and Development. 20. ————. 2002a. Agricultural Policies in OECD Countries: Monitor- Open Society Institute. 2002. Corruption and Anti-Corruption Pol- ing and Evaluation 2002. Paris: Organization for Economic Co- icy in Lithuania. Monitoring the EU Accession Process: Corruption operation and Development. and Anti-Corruption Policy. New York: Open Society Institute (OSI). ————. 2002b. Highlights of Public Sector Pay and Employment: 2002 Update. Paris: Organization for Economic Co-operation Ostrom, Elinor. 2000. “Collective Action and the Evolution of and Development. Social Norms.� Journal of Economic Perspectives 14(3):137–58. ————. 2002c. Regulatory Policies in OECD Countries: From Pack, Howard. 2000. “Industrial Policy: Growth Elixir or Poison.� Interventionism to Regulatory Governance. Paris: Organization World Bank Research Observer 15(1):47–67. for Economic Co-operation and Development. Padilla, A. Jorge, and Marco Pagano. 2000. “Sharing Default Infor- ————. 2002d. Revenue Statistics 1965–2001:2002 Edition. Paris: mation as a Borrower Discipline Device.� European Economic Organization for Economic Co-operation and Development. Review 44(10):1951–80. ————. 2003a. OECD Economic Surveys: Chile. Paris: Organiza- Pagés, Carmen, and Claudio E. Montenegro. 1999. “Job Security tion for Economic Co-operation and Development. and the Age Composition of Employment: Evidence from Chile.� Washington, D.C.: Inter-American Development Bank, Of�ce of ————. 2003b. OECD Investment Policy Reviews—China: the Chief Economist Working Paper Series 398. Progress and Reform Challenges. Paris: Organization for Eco- nomic Co-operation and Development. Palmade, Vincent. 2004. “The Importance of Sector Level Perspec- tive: Findings and Methodology of the McKinsey Global Insti- ————. 2003c. “OECD/World Bank Budget Practices and Proce- tute.� World Bank. Washington, D.C. Processed. dures Database�. Paris, Organization for Economic Co-operation and Development. Available on line at http://ocde.dyndns.org/. Panagariya, Arvind. 2000. “Evaluating the Case for Export Subsi- dies.� Washington, D.C.: World Bank Policy Research Working ————. 2003d. Overview of Donor and Agency Policies in Paper Series 2276. Trade-related Technical Assistance and Capacity Building. Paris: Organization for Economic Co-operation and Develop- ————. 2003. “India in the 1980s and 1990s: A Triumph of ment. Reforms.� Paper presented at the Tale of Two Giants: India’s and (c) The International Bank for Reconstruction and Development / The World Bank 232 WORLD DEVELOPMENT REPORT 2005 China’s Experience with Reform and Growth Conference. New Pissarides, Christopher. 2000. “Human Capital and Growth: A Syn- Delhi. November 14. thesis Report.� Paris: OECD Development Centre, Technical Parente, Stephen L., and Edward C. Prescott. 2000. Barriers to Paper 168. Riches. Cambridge, Mass.: MIT Press. Pistor, Katharina, Yoram Keinan, Jan Kleinheisterkamp, and Mark Parra, Antonio. 2000. “Applicable Substantive Law in ICSID Arbi- D. West. 2003. “Evolution of Corporate Law and the Transplant trations Initiated under Investment Treaties.� Paper presented at Effect.� World Bank Research Observer 18(1):89–112. the New Trends in Governing Law Conference. Washington, D.C. Pistor, Katherina. 2000. “Patterns of Legal Change: Shareholder and November 2. Creditor Rights in Transition Economies.� European Business Pattillo, Catherine. 1998. “Investment, Uncertainty, and Irre- Organization Law Review 1(1):59–108. versibility in Ghana.� IMF Staff Papers 45(3):522–53. Pistor, Katherina, Martin Raiser, and Stanislav Gelfer. 2000. “Law Paulsson, Jan. 1996. “Dispute Resolution.� In Robert Pritchard, ed., and Finance in Transition Economies: Lessons from Six Coun- Economic Development, Foreign Investment and Law. London, tries.� Economics of Transition 8(2):325–68. The Hague, and Boston: International Bar Association. Polinsky, Mitchell A., and Steven Shavell. 2000. “The Economic Paunovic, Igor. 2000. “Growth and Reforms in Latin America and Theory of Public Enforcement of Law.� Journal of Economic Lit- the Caribbean in the 1990s.� Santiago de Chile: United Nations erature 38(1):45–76. Economic Commission for Latin America and the Caribbean Pollitt, Michael G. 2003. “Electricity Reform in Chile and (ECLAC), Economic Development Division, Serie Reformas Argentina: Lessons for Developing Countries.� Paper presented Económicas 70. at the Cambridge-MIT Institute Electricy Power Autumn Pavcnik, Nina. 2003. “Trade Liberalization, Exit, and Productivity Research Seminar. Cambridge, Mass. November 7. Improvements: Evidence from Chilean Plants.� Review of Eco- Porter, Michael E. 1998. “Clusters and Competition: New Agendas nomic Studies 69(238):245–76. for Companies, Governments, and Institutions.� In Michael E. Peltzman, Sam. 1976. “Toward a more General Theory of Regula- Porter, ed., On Competition. Cambridge, Mass.: Harvard Busi- tion.� Journal of Law and Economics 19(2):211–40. ness Review. Pender, John, Berhanu Gebremedhin, Samuel Benin, and Simeon ————. 2000. “Attitudes, Values, Beliefs, and the Microeconom- Ehui. 2001. “Strategies for Sustainable Agricultural Development ics of Prosperity.� In Lawrence E. Harrison and Samuel P. Hunt- in the Ethiopian Highlands.� American Journal of Agricultural ington, eds., Culture Matters: How Values Shape Human Progress. Economics 83(5):1231–40. New York, N.Y.: Basic Books. Perkins, Dwight. 2000. “Law, Family Ties, and the East Asian Way of ————. 2003. “The Economic Performance of Regions.� Business.� In Lawrence Harrison and Samuel Huntington, eds., Regional Studies 37(6-7):549–78. Culture Matters: How Values Shape Human Progress. New York: Pound, Roscoe. 1959. Jurisprudence vol. III. Saint Paul, Minnesota: Basic Books. West Publishing Company. Persson, Torsten, and Guido Tabellini. 2000. “Double-edged Incen- PPIAF. 2003. FY2003 Annual Report. Washington, D.C.: Public-Pri- tives: Institutions and Policy Coordination.� In Gene M. Gross- vate Infrastructure Advisory Facility, World Bank. man and Kenneth S. Rogoff, eds., Handbook of International Eco- PPIAF, and World Bank. 2002. Private Solutions for Infrastructure in nomics Vol. 3. Amsterdam; New York and Oxford: Elsevier, Cambodia. Washington, D.C.: World Bank. North-Holland. Pritchett, Lant. 1997. “Divergence, Big Time.� Journal of Economic Peters, Alan, and Peter Fisher. 2004. “The Failures of Economic Perspectives 11(3):3–17. Development Incentives.� Journal of the American Planning Asso- ————. 2001. “Where Has All the Education Gone?� World Bank ciation 70(1):27–37. Economic Review 15(3):367–91. Philipsen, Niels J. 2003. “Overview of the Commission’s Stocktak- ————. 2002. “Understanding Patterns of Economic Growth: ing Exercise.� Paper presented at the Liberal Professions Confer- Searching for Hills among Plateaus, Mountains, and Plains.� ence. Brussels. October 28. World Bank Economic Review 14(2):231–50. Phillips, Charles F. Jr. 1993. The Regulation of Public Utilities: The- ————. 2003. “Who is Not Poor? Proposing a Higher Interna- ory and Practice (Third Edition). Arlington, V.A.: Public Utilities tional Standard for Poverty.� Washington, D.C.: Center for Reports. Global Development Working Paper 33. Pierre, Gaëlle, and Stefano Scarpetta. 2004. “Do Employers’ Percep- ————. 2004. “Reform is Like a Box of Chocolates: Understand- tions Square with Actual Labor Regulations?� Background paper ing the Growth Disappointments and Surprises.� Kennedy for the WDR 2005. School of Government, Harvard University. Cambridge, Mass. ————. 2004. “How Labor Market Policy Can Combine Work- Processed. ers’ Protection and Job Creation.� Background paper for the Psacharopoulos, George, and Harry Anthony Patrinos. 2002.“Returns WDR 2005. to Investment in Education: A Further Update.� Washington, D.C.: Pigou, Arthur C. 1932. The Economics of Welfare, 4th Edition. Lon- World Bank Policy Research Working Paper Series 2881. don: MacMillan and Co. Ltd. Pursell, Gary. 2001. “Australian Experience with Local Content Pro- Piore, Michael J., and Charles F. Sabel. 1984. The Second Industrial grams in the Auto Industry: Some Lessons for India and Other Divide: Possibilities for Prosperity. New York: Basic Books. Developing Countries.� Journal of World Trade 35(2):379–94. (c) The International Bank for Reconstruction and Development / The World Bank References 233 Putnam, Robert D. 1988. “Diplomacy and Domestic Politics: The Ravallion, Martin, and Gaurav Datt. 1995. “Is Targeting through a Logic of Two-Level Games.� International Organization Work Requirement Ef�cient?: Some Evidence from Rural India.� 42(3):427–60. In D. van de Walle and K. Nead, eds., Public Spending and the Putnam, Robert D., Robert Leonardi, and Raffaella Y. Nanetti. 1993. Poor: Theory and Evidence. Baltimore, M.D.: Johns Hopkins Uni- Making Democracy Work: Civic Traditions in Modern Italy. versity Press for the World Bank. Princeton, N.J.: Princeton University Press. Rawlings, Laura B., Lynne. Sherburne-Benz, and Julie Van Dome- Qian, Yingyi. 2003. “How Reform Worked in China.� In Dani len. 2003. Evaluating Social Funds: A Cross-country Analysis of Rodrik, ed., In Search of Prosperity: Analytic Narratives on Eco- Community Investments. Washington, D.C.: World Bank. nomic Growth. Princeton, N.J.: Princeton University Press. Rawlings, Laura B., and Gloria M. Rubio. 2002. Evaluating the Qiang, Christine Zhen-Wei, Alexander Pitt, and Seth Ayers. 2004. Impact of Conditional Cash Transfer Programs: Lessons from Latin “Contribution of Information and Communication Technolo- America. Washington, D.C.: World Bank. gies to Growth.� Washington, D.C.: World Bank Working Paper Raworth, Kate. 2004. Trading Away our Rights: Women Working in 24. Global Supply Chains. Oxford: Oxfam International. Rabin, Matthew. 1998. “Psychology and Economics.� Journal of Eco- Ray, Amit S. 2004. “Managing Port Reforms in India: Case Study of nomic Literature 36(1):11–46. Jawaharlal Nehru Port Trust (JNPT) Mumbai.� Background Radelet, Steven. 1999. “Manufactured Exports, Export Platforms, paper for the WDR 2005. and Economic Growth.� Cambridge, Mass.: Harvard Institute for Reardon, Thomas, Kostas Stamoulis, María Elena Cruz, Arsenio International Development Consulting Assistance on Economic Balisacan, Julio Berdegue, and Kimseyinga Savadogo. 1998. Reform II 43. “Diversi�cation of Household Incomes into Non-Farm Sources: Radelet, Steven, and Jeffery Sachs. 1998. “Shipping Costs, Manufac- Patterns, Determinants and Effects.� Paper presented at the tured Exports, and Economic Growth.� Paper presented at the IFPRI Conference: Strategies for Stimulating Growth of the American Economic Association Annual Meeting. Chicago, I.L. Rural Non-Farm Economy in Developing Countries. Arlie January 8. House, V.A. May 17. Rajan, Raghuram G., and Luigi Zingales. 2003. Saving Capitalism Recanatini, Francesca, and Randi Ryterman. 2001. “Disorganiza- from the Capitalists: Unleashing the Power of Financial Markets to tion of Self-organization? The Emergence of Business Associa- Create Wealth and Spread Opportunity. New York: Crown Busi- tions in a Transition Economy.� World Bank. Washington, D.C. ness. Processed. Rama, Martin. 2002. “The Gender Implications of Public Sector Regobeth, Christian, and Ko� Ahortor. 2003. “Regulatory Impact Downsizing: The Reform Program of Vietnam.� World Bank on Ghana.� Paper presented at the Regulatory Impact Assess- Research Observer 17(2):167–89. ment: Strengthening Regulation Policy and Practice Conference. ————. 2003. “Globalization and Workers in Developing Coun- Centre on Regulation and Competition, Manchester, U.K. tries.� Washington, D.C.: World Bank Policy Research Working November 23. Paper Series 2958. Reid, Neil, and Jay D. Gatrell. 2003. “Uncertainty, Incentives, and Rama, Martin, and Raquel Artecona. 2002. “A Database of Labor the Preservation of an Industrial Icon: The Case of Toledo Jeep.� Market Indicators Across Countries.� World Bank. Washington, In Nicholas Phelps and Philip Raines, eds., The New Competition D.C. Processed. for Inward Investment: Companies, Institutions and Territorial Development. Cheltenham, U.K.: Edward Elgar. Ramamurti, Ravi. 1996. “The New Frontier of Privatization.� In Ravi Ramamurti, ed., Privatizing Monopolies: Lessons from the Reinikka, Ritva, and Jakob Svensson. 1999. “Confronting Competi- Telecommunications and Transport Sectors in Latin America. Bal- tion: Investment Response and Constraints in Uganda.� Wash- timore, M.D.: Johns Hopkins University Press. ington, D.C.: World Bank Policy Research Working Paper Series 2242. Ramey, Garey, and Valerie A. Ramey. 1995. “Cross-Country Evi- dence on the Link Between Volatility and Growth.� American ————. 2002. “Coping with Poor Public Capital.� Journal of Economic Review 85(5):1138–51. Development Economics 69(1):51–69. Ratnam, V. 1999. Trade Unions in the Informal Sector: Finding their Reynolds, Paul D., William D. Bygrave, Erkko Autio, and others. Bearings. Nine Country Papers. Geneva: International Labor 2004. Global Entrepreneurship Monitor Global 2003 Executive Organization. Report. http://www.gemconsortium.org/: Global Entrepreneur- ship Monitor. Rauch, James E., and Vitor Trindade. 2002. “Ethnic Chinese Net- works in International Trade.� The Review of Economics and Sta- Rhee, Yung Whee. 1990. “The Catalyst Model of Development: tistics 84(1):116–30. Lessons from Bangladesh’s Success with Garment Exports.� World Development 18(2):333–46. Ravallion, Martin. 2003a. “Measuring Aggregate Welfare in Devel- oping Countries: How Well do National Accounts and Surveys Roberts, Mark J., and James R. Tybout, eds. 1996. Industrial Evolu- Agree?� Review of Economics and Statistics 85(3):645–52. tion in Developing Countries. Washington, D.C.: Oxford Univer- sity Press for the World Bank. ————. 2003b. “Targeted Transfers in Poor Countries: Revisiting the Trade-offs and Policy Options.� Washington, D.C.: World Robinson, James A. 1998. “Theories of ‘Bad Policy’.� Journal of Pol- Bank Policy Research Working Paper Series 3048. icy Reform 2(1):1–46. (c) The International Bank for Reconstruction and Development / The World Bank 234 WORLD DEVELOPMENT REPORT 2005 Robinson, James A., and Thierry Verdier. 2002. “The Political Econ- Consensus: Restarting Growth and Reform in Latin America. omy of Clientelism.� Paris: Centre for Economic Policy Research, Washington, D.C.: Institute for International Economics. Discussion Paper Series 3205. Available on line at Sachs, Jeffrey, and Andrew M. Warner. 2001. “The Curse of Natural http://www.cepr.org/pubs/new-dps/dp_papers.htm. Resources.� European Economic Review 45(4-6):827–38. Rodríguez-Clare, Andrés. 2001. “Costa Rica’s Development Strat- Sachs, Jeffrey D. 2003. “Institutions Don’t Rule: Direct Effects of egy Based on Human Capital and Technology: How it Got There, Geography On Per Capita Income.� Cambridge, Mass.: National the Impact of Intel, and Lessons for Other Countries.� Journal of Bureau of Economic Research Working Paper Series 9490. Human Development 2(2):311–24. Sader, Frank. 2003. “Do ‘One-Stop’ Shops Work?� World Bank. Rodrik, Dani. 1991. “Policy Uncertainty and Private Investment in Washington, D.C. Processed. Developing Countries.� Journal of Development Economics Sala-i-Martin, Xavier, and Elsa Vila-Artadi. 2002. “Economic 36(2):229–42. Growth and Investment in the Arab World.� New York: Colum- ————. 1997. Has Globalization Gone Too Far? Washington, bia University, Department of Economics Discussion Paper D.C.: Institute for International Economics. Series 0203-08. ————. 1999. The New Global Economy and Developing Coun- Samuel, Benin, and John Pender. 2002. “Impacts of Land Redistrib- tries: Making Openness Work. Washington, D.C.: Overseas Devel- ution on Land Management and Productivity in the Ethiopian oping Council. Highlands.� Addis Ababa: International Livestock Research Insti- Rodrik, Dani, and Arvind Subramanian. 2004. “From ‘Hindu tute, Working Paper 43. Growth’ to Productivity Surge: The Mystery of the Indian Sander, Cerstin. 2004. “Less is More: Better Compliance and Growth Transition.� Harvard University. Cambridge, Mass. Increased Revenues by Streamlining Business Registration in Processed. Uganda.� Case Study Commissioned by the UK Department for Rodrik, Dani, Arvind Subramanian, and Francesco Trebbi. 2002. International Development for the 2005 World Development Institutions Rule: The Primacy of Institutions over Geography and Report. Integration in Economic Development. Cambridge, Mass: Harvard Sandmo, Agnar. 1976. “Optimal Taxation: An Introduction to the University. Literature.� Journal of Public Economics 6(1-2):37–54. Röller, Lars-Hendrik, and Leonard Waverman. 2001. “Telecommu- Sapienza, Paola. 2004. “The Effect of Government Ownership on nications Infrastructure and Economic Development: A Simul- Bank Lending.� Journal of Financial Economics 72(2):357–84. taneous Approach.� American Economic Review 91(4):909–23. Saunders, Anthony, and Berry Wilson. 2002. “An Analysis of Bank Ros, Augustin J. 1999. “Does Ownership or Competition Matter? Charter Value and Its Risk-Constraining Incentives.� Journal of The Effects of Telecommunications Reform on Network Expan- Financial Services Research 19(2-3):185–95. sion and Ef�ciency.� Journal of Regulatory Economics Scarpetta, Stefano, and Eric Bartelsman. 2003. “Firm Dynamics in 15(1):65–92. Developing Countries.� World Bank. Washington, D.C. Processed. Rosen, Harvey. 1995. Public Finance, 4th Edition. Chicago, I.L.: Scarpetta, Stefano, Philip Hemmings, Thierry Tressel, and Jaejoon Irwin Publishers. Woo. 2002. “The Role of Policy and Institutions for Productivity Rosenzweig, Mark R. 1995. “Why Are There Returns to Schooling?� and Firm Dynamics: Evidence from Micro and Industry Data.� American Economic Review 85(2):153–58. Paris: OECD Economics Department Working Paper 329. Avail- Rosenzweig, Mark R., and Hans P. Binswanger. 1993. “Wealth, able on line at http://www.olis.oecd.org/olis/2002doc.nsf/ Weather Risk and the Composition and Pro�tability of Agricul- linkto/eco-wkp(2002)15. tural Investments.� Economic Journal 103(416):56–78. Scarpetta, Stefano, and Thierry Tressel. 2004. “Boosting Productiv- Ross, Michael L. 2001. “Does Oil Hinder Democracy?� World Poli- ity via Innovation and Adoption of New Technologies: Any Role tics 53(3):325–61. for Labor Market Institutions?� Washington, D.C.: World Bank Rossiter, Clinton, ed. 1961. The Federalist Papers. New York City: Working Research Paper Series 3273. New American Library. Schärf, Wilfried. 2001. “Community Justice and Community Polic- Rossotto, Carlo Maria, Anat Lewin, Carlos Gomez, and Bjõrn ing in Post-Apartheid South Africa.� IDS Bulletin 32(1):74–82. Wellenius. 2003. “Competition in International Voice Communi- Schiff, Maurice, and L. Alan Winters. 2003. Regional Integration and cations.� World Bank, Policy Paper of the Global Information Development. Washington, D.C. and Oxford: World Bank and and Communications Technology Department, Policy Division. Oxford University Press. Washington, D.C. Processed. Schmukler, Sergio. 2003. “Financial Globalization: Gain and Pain Saavedra, Jaime, and Máximo Torero. 2004. “Labor Market Reforms for Developing Countries.� World Bank. Washington DC. and Their Impact on Formal Labor Demand and Job Market Processed. Turnover: The Case of Peru.� In James Heckman and Carmen Schneider, Friedrich. 2002. “Size and Measurement of the Informal Pagés, eds., Law and Employment: Lessons from Latin America Economy in 110 Countries Around the World.� Paper presented and the Caribbean. Cambridge, M.A. and Chicago, I.L.: National at the Workshop of Australian National Tax Centre, ANU. Can- Bureau of Economic Research and University of Chicago. berra, Australia. July 17. Saavedra, Jaime. 2003. “Labor Markets during the 1990’s.� In Pedro- Schor, Adriana. Forthcoming. “Productivity, Embodied Technology Pablo Kuczynski and John Williamson, eds., After the Washington and Heterogeneous Responses to Tariff Reduction: Evidence (c) The International Bank for Reconstruction and Development / The World Bank References 235 from Brazilian Manufacturing Firms.� Journal of Development Segmented Markets.� In Avishay Braverman, Karla Hoff, and Economics. Joseph E. Stiglitz, eds., The Economics of Rural Organization: The- Schrank, Andrew. 2001. “Export Processing Zones: Free Market ory, Practice, and Policy. New York: Oxford University Press for Islands or Bridges to Structural Transformation.� Development the World Bank. Policy Review 19(2):223–42. Sidak, Gregory J., and William J. Baumol. 1995. Transmission Pric- Schreiner, Mark, and Jacob Yaron. 2001. Development Finance Insti- ing and Stranded Costs in the Electric Power Industry. Washington, tutions: Measuring their Subsidy. Washington, D.C.: World Bank. D.C.: AEI Press. Schulpen, Lau, and Peter Gibbon. 2002. “Private Sector Develop- Siebert, Horst, ed. 2003. Global Governance: An Architecture for the ment: Policies, Practices and Problems.� World Development World Economy. Berlin, Heidelberg, New York: Springer. 30(1):1–15. Slemrod, Joel. 1990. “Optimal Taxation and Optimal Tax Systems.� Schumpeter, Joseph. 1942. Capitalism, Socialism and Democracy. Journal of Economic Perspectives 4(1):157–78. New York: Harper & Row. Smarzynska, Beata. 2002. “Does Foreign Direct Investment Increase Scott, Allen J., and Michael Storper. 2003. “Regions, Globalization, the Productivity of Domestic Firms? In Search of Spillovers Development.� Regional Studies 37(6-7):579–93. through Backward Linkages.� Washington, D.C.: World Bank Sedlacek, Guilherme, Nadeem Ilahi, and Emily Gustafsson-Wright. Policy Research Working Paper Series 2923. 2000. “Targeted Conditional Transfer Programs in Latin Amer- Smith, Gare, and Dan Feldman. 2003. Company Codes of Conduct ica: An Early Survey.� World Bank Report Prepared for the and International Standards: An Analytical Comparison, Part I Regional Study on “Securing our Future�. Washington, D.C. and II. Washington, D.C.: World Bank and International Finance Processed. Corporation. Sereno, Lourdes Ma, Emmanuel de Dios, and Joseph J. Capuano. Smith, Heather. 2000. Industry Policy in Taiwan and Korea in the 2001. “Justice and the Social Cost of Doing Business: The Case of 1980s: Winning with the Market. Northampton, Mass.: Edward Philippines.� World Bank. Manila, Philippines. Processed. Elgar Publishing. Serven, Luis. 1997. “Uncertainty, Instability, and Irreversible Invest- Smith, Warrick. 1997a. “Covering Political and Regulatory Risks: ment : Theory, Evidence, and Lessons for Africa.� Washington, Issues and Options for Private Infrastructure Arrangements.� In D.C.: World Bank Policy Research Working Paper Series 1722. Timothy Irwin, Michael Klein, Guillermo E. Perry, and Mateen Shah, Anwar, ed. 1995a. Fiscal Incentives for Investment and Innova- Thobani, eds., Dealing with Public Risk in Private Infrastructure. tion. New York: Oxford University Press. Washington, D.C.: World Bank. ————. 1995b. “Research and Development Investment, Indus- ————. 1997b. “Utility Regulators: The Independence Debate.� trial Structure, Economic Performance, and Tax Policies.� In Washington, D.C.: World Bank, Public Policy for the Private Sec- Anwar Shah, ed., Fiscal Incentives for Investment and Innovation. tor Note 127. New York: Oxford University Press. Solow, Robert. 1957. “Technical Change and the Aggregate Produc- Shah, Anwar, and John Baffes. 1995. “Do Tax Policies Stimulate tion Function.� Review of Economics and Statistics 39(3):312–30. Investment in Physical and R&D Capital?� In Anwar Shah, ed., Sölvell, Örjan, Göran Lindqvist, and Christian Ketels. 2003. The Fiscal Incentives for Investment and Innovation. New York: Oxford Cluster Initiative Greenbook. Stockholm: Ivory Tower AB. University Press. Spar, Deborah. 1998. “Attracting High Technology Investment: Shah, Hemant. 1997. “Toward Better Regulation of Private Pension Intel’s Costa Rica Plant.� Washington, D.C.: World Bank, Foreign Funds.� Washington, D.C.: World Bank Policy Research Working Investment Advisory Service, Occasional Paper 11. Paper Series 1791. Spiller, Pablo, and Cezley I. Sampson. 1996.“Regulating Telecommuni- Shavell, Steven. 2004. Foundations of Economic Analysis of Law. cations in Jamaica.� In Brian Levy and Pablo T. Spiller, eds., Regula- Cambridge, Mass.: Harvard University Press. tions, Institutions, and Commitment: Comparative Studies of Sherman, Lawrence W., Denise C. Gottfredson, Doris L. MacKen- Telecommunications. Cambridge, U.K.: Cambridge University Press. zie, John Eck, Peter Reuter, and Shawn D. Bushway. 1998. Pre- Spiller, Pablo T., and William Savedoff. 1999. “Commitment and venting Crime: What Works, What Doesn’t, What’s Promising: A Governance in Infrastructure.� In Federico Basañes, Evamaría Report to the U.S. Congress. Washington, D.C.: National Institute Uribe, and Robert Willig, eds., Can Privatization Deliver? Infra- of Justice. structure for Latin America. Washington, D.C.: Inter-American Shihata, Ibrahim. 1986. “Towards a Greater Depolitization of Development Bank. Investment Disputes: The Roles of ICSID and MIGA.� ICSID Srinivas, P. S., Edward Whitehouse, and Juan Yermo. 2000. “Regu- Review–Foreign Investment Law Journal 1(1):1–25. lating Private Pension Funds’ Structure, Performance and Invest- Shirley, Mary, and Patrick Walsh. 2000. “Public versus Private Own- ments: Cross-country Evidence.� Washington, D.C.: World Bank ership: The Current State of the Debate.� Washington, D.C.: Social Protection Discussion Paper 113. World Bank Policy Research Working Paper Series 2420. Staiger, Robert W., and Guido Tabellini. 1999. “Do GATT Rules Shleifer, Andrei, and Daniel Wolfenzohn. 2002. “Investor Protection Help Governments Make Domestic Commitments?� Economics and Equity Markets.� Journal of Financial Economics 66(1):3–27. and Politics 11(2):109–44. Siamwalla, Ammar. 1993. “The Thai Rural Credit System and Ele- Stanchev, Krassen. 2003. “Applicable RIA Instruments in an Imma- ments of a Theory: Public Subsidies, Private Information, and ture Administrative Environment: The Bulgarian Experience.� (c) The International Bank for Reconstruction and Development / The World Bank 236 WORLD DEVELOPMENT REPORT 2005 Paper presented at the Regulatory Impact Assessment: Best Prac- Subbarao, Kalanidhi. 1997. “Public Works as an Anti-poverty Pro- tices and Lesson-drawing in Europe Conference. So�a, Bulgaria. gram: An Overview of Cross-country Experience.� American February 27. Journal of Agricultural Economics 79(2):678–83. Stan�eld, David, Edgar Nesman, Mitchell Seligson, and Alexander Subbarao, Kalanidhi, Ahhtes Ahmed, and Tesfaye Teklu. 1995. Coles. 1990. “The Honduran Land Titling and Registration “Philippines: Social Safety Net Programs: Targeting, Cost-effec- Experience.� University of Wisconsin, Land Tenure Center. tiveness and Options for Reform.� Washington, D.C.: World Madison, Wisconsin. Processed. Bank Poverty and Social Policy Discussion Paper 77. Stasavage, David. 2002. “Private Investment and Political Institu- Subramanian, Arvind, and Devesh Roy. 2003. “Who Can Explain tions.� Economics and Politics 14(1):41–63. the Mauritian Miracle?: Meade, Romer, Sachs, or Rodrik?� In Stephan, Paul B. 1999. “The Futility of Uni�cation and Harmoniza- Dani Rodrik, ed., In Search of Prosperity: Analytic Narratives on tion of International Commercial Law.� Virginia Journal of Inter- Economic Growth. Princeton, N.J.: Princeton University Press. national Law 39(3):743–97. Sutton, John. 2002. “Rich Trades, Scarce Capabilities: Industrial Stephenson, Matthew. 2003. “When the Devil Turns: The Political Development Reconsidered.� Economic and Social Review Foundations of Independent Judicial Review.� Journal of Legal 33(1):1–22. Studies 32(1):59–90. Sutton, Willie. 1976. Where the Money Was: The Memoirs of a Bank Stern, Nicholas, Jean-Jacques Dethier, and Halsey Rogers. 2004. Robber. New York: Viking Press. “Growth and Empowerment: Making Development Happen.� Svensson, Jakob. 1998. “Investment, Property Rights and Political World Bank. Washington, D.C. Processed. Instability: Theory and Evidence.� European Economic Review Stern, Nicholas H. 2002. A Strategy for Development. Washington, 42(7):1317–41. D.C.: World Bank. Tabor, Steven R. 2002. “Assisting the Poor with Cash: Design and Stigler, George. 1975. The Citizen and the State: Essays on Regula- Implementation of Social Transfer Programs.� Washington, D.C.: tion. Chicago, I.L.: University of Chicago Press. World Bank Social Protection Discussion Paper Series 0223. Stigler, George, and Gary S. Becker. 1977. “De Gustibus Non Est Taliercio Jr., Robert. 2001. “Raising Revenues and Raising Hackles: Disputandum.� American Economic Review 67(2):76–90. The Problem of Semi-Autonomous Tax Agencies in Developing Stigler, George J. 1971. “The Theory of Economic Regulation.� Bell Countries.� World Bank. Washington, D.C. Processed. Journal of Economics and Management Science 2(1):3–21. ————. 2003a. “Administrative Reform as Credible Commitment: Stiglitz, Joseph E. 1989. “Markets, Market Failures, and Develop- The Impact of Autonomy of Revenue Authority Performance in ment.� American Economic Review 79(2):197–203. Latin America.� World Bank. Washington, D.C. Processed. Stiglitz, Joseph E. 1999a. “Bankruptcy Laws: Some Elementary Eco- ————. 2003b. “Designing Performance: The Semi- nomic Principles.� Paper presented at the Annual World Bank Autonomous Revenue Authority Model in Africa and Latin Conference on Development Economics. Paris. June 21. America.� World Bank. Washington, D.C. Processed. ————. 1999b. “On Liberty, the Right to Know, and Public Dis- Tan, Eng Pheng. 2004. “The Singapore E-Government Experience.� course: The Role of Transparency in Public Life.� Paper pre- Paper presented at the I Foro Internacional: Hacia una Sociedad sented at the Oxford Amnesty Lecture. Oxford, U.K. January 27. Digital. Panamá City, Panamá. April 6. Stiglitz, Joseph E., and Partha Dasgupta. 1971. “Differential Taxa- Tanzi, Vito. 1995. “Fiscal Federalism and Decentralization: A tion, Public Goods and Economic Ef�ciency.� Review of Eco- Review of Some Ef�ciency and Macroeconomic Aspects.� Paper nomic Studies 38(114):151–74. presented at the World Bank Conference on Development Eco- nomics. Washington, D.C. Stiglitz, Joseph E., and Michael Rothschild. 1976.“Equilibrium in Com- petitive Insurance Markets: The Economics of Markets with Imper- Tanzi, Vito, and Hamid Davoodi. 1997. “Corruption, Public Invest- fect Information.� Quarterly Journal of Economics 90(4):629–49. ment, and Growth.� Washington, D.C.: International Monetary Fund Working Paper WP/97/139. Stiglitz, Joseph E., and Andrew Weiss. 1981. “Credit Rationing in Markets with Imperfect Information.� American Economic ————. 1998. “Roads to Nowhere: How Corruption in Public Review 71(3):393–410. Investment Hurts Growth.� Washington, D.C.: International Monetary Fund Economic Issues 12. Stiglitz, Joseph E., and Shahid Yusuf, eds. 2001. Rethinking the East Asian Miracle. Washington, D.C.: World Bank and Oxford Uni- Tanzi, Vito, and Howell H. Zee. 2001. “Tax Policy for Developing versity Press. Countries.� Washington, D.C.: International Monetary Fund, Economic Issues 27. Stijns, Jean-Philippe C. 2000. “Natural Resource Abundance and Economic Growth Revisited.� University of California. Berkeley, Taylor, Alan M. 1996. “On the Costs of Inward Looking Develop- Calif. Processed. ment: Historical Perspectives on Price Distortions, Growth, and Divergence in Latin America from the 1930s to the 1980s.� Cam- Stoica, Christina I., and Valeriu Stoica. 2002. “Romania’s Legal bridge, Mass.: National Bureau of Economic Research Working Regime for Security Interests in Personal Property.� Law in Tran- Paper Series 5432. sition Spring: 62–6. Teklu, Tesfaye, and Sisay Asefa. 1999. “Who Participates in Labor- Stone, Christopher E., and Heather H. Ward. 2000. “Democratic Intensive Public Works in Sub-saharan Africa? Evidence from Policing: A Framework for Action.� Policing and Society Rural Botswana and Kenya.� World Development 27(2):431–38. 10(1):11–12. (c) The International Bank for Reconstruction and Development / The World Bank References 237 Tendler, Judith. 2002. “The Economic Wars Between the States.� Torstensson, Johan. 1994. “Property Rights and Economic Growth: Paper presented at the OECD/State Government of Ceará Meet- An Empricial Study.� Kyklos 47(2):231–47. ing on Foreign Direct Investment and Regional Development. Townsend, Robert M., and Jacob Yaron. 2001. “The Credit Risk- Fortaleza, Brasil. December 12. Contingency System of an Asian Development Bank.� Federal Thaler, Richard H. 1993. The Winner’s Curse: Paradoxes and Anom- Reserve Bank of Chicago Economic Perspectives 25(3):31–48. alies of Economic Life. Princeton, N.J.: Princeton University Press. Trajtenberg, Manuel. 2002. “Government Support for Commercial ————. 2000. “From Homo Economicus to Homo Sapiens.� R&D: Lessons from the Israeli Experience.� Innovation Policy and Journal of Economic Perspectives 14(1):133–41. the Economy 2(1):79–134. The Economist. 1999. “Sweatshop Wars.� The Economist, February Tranberg Hansen, Karen. 2000. Salaula: The World of Secondhand 25. Clothing in Zambia. Chicago: University of Chicago Press. ————. 2001. “How Countries go High-Tech.� The Economist, Transition. 2003. “Russia’s Unpopular Billionaires on Forbes’ List.� December 28. Transition 14:14–15. ————. 2002a. “Digging Deep: Anglo American Offers AIDS Transparency International. 2004. Global Corruption Report. Berlin: Drugs to All Its Employees.� The Economist, August 8. Transparency International. ————. 2002b. “Enlisting Gamblers.� The Economist, September Transparency International Bangladesh. 2002. Corruption in 26. Bangladesh: A Household Survey. Dakka: Transparency Interna- ————. 2002c. Globalisation—Making Sense of an Integrating tional Bangladesh. Available on line at http://www.ti- World. London, UK: Economist Books. bangladesh.org/. ————. 2003a. “Into Africa: Investing in Angola.� The Economist, Treisman, Daniel. 2000. “The Causes of Corruption: A Cross- June 12. National Study.� Journal of Public Economics 76(3):399–457. ————. 2003b. “Never Had it so Good.� The Economist, Septem- Tropin, Mitchell. 2003. Transfer Pricing: Special Report. Washing- ber 11. ton, D.C.: Tax Management, Bureau of National Affairs. ————. 2003c. “The Diaspora is ‘Discovered’.� The Economist, Trujillo, Lourdes, and Tomas Serebrisky. 2003. “Market Power: February 23. Ports: A Case Study of Postprivatization Mergers.� Washington, The Mitchell Group Inc. 2003. Promoting Competitiveness in Prac- D.C.: World Bank Public Policy for the Private Sector Note 260. tice: An Assessment of Cluster-based Approaches. Washington, Tullock, Gordon. 1983. Economics of Income Distribution. Boston: D.C.: The Mitchell Group Inc., prepared for the U.S. Agency for Kluwer-Nijhoff. International Development. Tybout, James R. 2003. “Plant- and Firm-Level Evidence on ‘New’ The Pew Global Attitudes Project. 2003. Views of a Changing World. Trade Theories.� In E. Kwan Choi and James Harrigan, eds., Hand- Washington, D.C.: The Pew Research Center for the People & the book of International Trade. Malden, Mass.: Blackwell Publishers. Press. Available on line at http://people- Uchitelle, Louis. 2003. “States Pay for Jobs, but it Doesn’t Always press.org/reports/pdf/185.pdf. Pay Off.� The New York Times, November 10. Tiebout, Charles M. 1956. “A Pure Theory of Local Expenditures.� UNAIDS. 2003. AIDS Epidemic Update 2003. New York: Joint Journal of Political Economy 64(5):416–24. United Nations Programme on HIV/AIDS. Tomkins, Ray. 2001. “Extending Rural Electri�cation: A Survey of UNCTAD. 1996. Incentives and Foreign Direct Investment. Geneva, Innovative Schemes.� In Penelope J. Brook and Suzanne M. New York: United Nations Conference on Trade and Development. Smith, eds., Contracting for Public Services: Output-based Aid and ————. 1998. Bilateral Investment Treaties in the Mid-1990s. Its Applications. Washington, D.C.: World Bank. New York and Geneva: United Nations Conference on Trade and Tommasi, Mariano. 2002. “Crisis, Political Institutions, and Policy Development. Reform: It is not the Policy, it is the Polity, Stupid.� Paper pre- ————. 2000a. A Positive Agenda for Developing Countries: Issues sented at the Annual World Bank Conference on Development for Future Trade Negotiations. New York and Geneva: United Economics. Oslo. June 22. Nations Conference on Trade and Development. Tooley, James. 1999. The Global Education Industry: Lessons from ————. 2000b. Bilateral Investment Treaties 1959–1999. New Private Education in Developing Countries. London, U.K.: IEA York and Geneva: United Nations Conference on Trade and Education and Training Unit in association with the Interna- Development. Available on line at http://www.unctad.org/en/ tional Finance Corporation. docs/poiteiiad2.en.pdf. Topalova, Petia. 2003. “Trade Liberalization and Firm Productivity: ————. 2000c. “Tax Incentives and Foreign Direct Investment: A The Case of India.� Massachusetts Institute of Technology. Cam- Global Survey.� Geneva: United Nations Conference on Trade bridge, Mass. Processed. and Development ASIT Advisory Studies 16. Available on line at Topel, Robert. 1999. “Labor Markets and Economic Growth.� In http://www.unctad.org/en/docs//iteipcmisc3_en.pdf. Orley Ashenfelter and David Card, eds., Handbook of Labor Eco- ————. 2001a. Social Responsibility. New York and Geneva: nomics (Volume 3C). Amsterdam: Elsevier Science B.V. United Nations Conference on Trade and Development. Avail- Tornell, Aaron, and Philip R. Lane. 1999. “The Voracity Effect.� able on line at http://www.unctad.org/en/docs//psiteiitd22. American Economic Review 89(1):22–46. en.pdf. (c) The International Bank for Reconstruction and Development / The World Bank 238 WORLD DEVELOPMENT REPORT 2005 ————. 2001b. World Investment Report 2001: Promoting Link- United Nations–Habitat. 2003. The Challenge of Slums: Global ages. Geneva: United Nations Conference on Trade and Develop- Report on Human Settlements 2003. London: Earthscan Publica- ment. Available on line at http://www.unctad.org/Templates/ tions Limited. WebFlyer.asp?intItemID=2434&lang=1. United Nations Commission on the Private Sector and Develop- ————. 2002a. Investment Policy Review. Botswana. Geneva: ment. 2004. Report to the Secretary-General of the United Nations. United Nations Conference on Trade and Development. Avail- Unleashing Entrepreneurship: Making Business Work for the Poor. able on line at http://www.unctad.org/Templates/WebFlyer. New York: United Nations Development Program. asp?intItemID=2811&lang=1. United Nations Economic Commission for Africa. 2000. ————. 2002b. World Investment Report 2002: Transnational Cor- “HIV/AIDS and Economic Development in Sub-Saharan porations and Export Competitiveness. Geneva: United Nations Africa.� Paper presented at the African Development Forum Conference on Trade and Development. Available on line at 2000, AIDS: The Greatest Leadership Challenge. Addis Ababa, http://www.unctad.org/Templates/WebFlyer.asp?intItemID= Ethiopia. December 3. 2399&lang=1. Urízar H., Carmen, and Sigfrido Lee. 2003. “The Effects of Unions ————. 2003a. Dispute Settlement: Investor-state. Geneva: United on Productivity: Evidence from Large Coffee Producers in Nations Conference on Trade and Development. Guatemala.� Washington, D.C.: Inter-American Development ————. 2003b. Foreign Direct Investment and Performance Bank, Research Department Working Paper R-473. Requirements: New Evidence from Selected Countries. New York Van de Walle, Dominique. 2003. “Are Returns to Investment Lower and Geneva: United Nations Conference on Trade and Develop- for the Poor? Human and Physical Capital Interactions in Rural ment. Available on line at http://www.unctad.org/Templates/ Vietnam.� Review of Development Economics 7(4):636–53. webflyer.asp?docid=4378&intItemID=1397&lang=1. Van der Walt, A. J. 1999. “Reducing Regulatory Risk in Infrastruc- ————. 2003c. Investment and Technology Policies for Competi- ture by Requiring Compensation for Regulatory Takings: A tiveness: Review of Successful Country Experiences. New York and Comparative Perspective.� Paper presented at the Private Infra- Geneva: United Nations Conference on Trade and Development structure for Development, International Conference. Rome, Technology for Development Series. Available on line at Italy. September 10. http://www.unctad.org/en/docs//iteipc20032_en.pdf. Van Rijckeghem, Caroline, and Beatrice Weder. 2001. “Bureaucratic ————. 2003d. “Model Law on Competition.� Geneva, Switzer- Corruption and the Rate of Temptation: Do Wages in the Civil land: United Nations Conference on Trade and Development Service Affect Corruption, and by How Much?� Journal of Devel- Series on Issues in Competition Law and Policy opment Economics 65(2):307–31. TD/B/RBP/CONF.5/7/Rev.1. Varshney, Ashutoth. 1998. “Mass Politics or Elite Politics? India’s ————. 2003e. World Investment Report 2003: FDI Policies for Economic Reforms in Comparative Perspective.� Journal of Policy Development National and International Perspectives. Geneva: Reform 2(4):301–35. United Nations Conference on Trade and Development. Avail- Velde, Dirk Willem te. 2001. Policies Towards Foreign Direct Invest- able on line at http://www.unctad.org/Templates/WebFlyer. ment in Developing Countries: Emerging Best-practices and Out- asp?intItemID=2979&lang=1. standing Issues. London: Overseas Development Institute. UNCTAD, and WTO. 2000. The Post-Uruguay Round Tariff Envi- Vernon, Raymond. 1971. Sovereignty at Bay: The Multinational ronment for Developing Country Exports: Tariff Peaks and Tariff Spread of U.S. Enterprises. New York: Basic Books. Escalation. Geneva: United Nations Conference on Trade and Villadeces, Andrés, Peter Cummings, Victoria E. Espitia, Thomas D. Development and World Trade Organization. Koepsell, Barbara McKnight, and Arthur L. Kellermann. 2000. Unite, Angelo A., and Michael J. Sullivan. 2003. “The Effect of “Effect of a Ban on Carrying Firearms on Homicide Rates in Two Foreign Entry and Ownership Structure on the Philippine Colombian Cities.� Journal of the American Medical Association Domestic Banking Sector.� Journal of Banking and Finance 283:1205–9. 27(12):2323–45. Vilpisauskas, Ramunas. 2003. “The Introduction of Regulatory United Kingdom–DFID. 2001. Untying Aid. London: Department Impact Assessment in Lithuania: From Contributing to EU for International Development. Accession to Improving the Performance of Public Administra- ————. 2004. What is Pro-Poor Growth and Why Do We Need to tion.� Paper presented at the Politics of Regulatory Impact Know? London: Department for International Development. Assessment: Best Practices and Lesson-drawing in Europe Con- United Nations. 2000. The World’s Women 2000: Trends and Statis- ference. So�a, Bulgaria. February 27. tics. New York: United Nations. Available on line at Vittas, Dimitri, and Yoon Je Cho. 1995. “Credit Policies: Lessons http://unstats.un.org/unsd/demographic/ww2000/. from East Asia.� Washington, D.C.: World Bank Policy Research ————. 2002a. International Migration Report 2002. New York: Working Paper Series 1458. United Nations Department of Economic and Social Affairs Pop- Vodopivec, Milan. 2004. Income Support for the Unemployed: Issues ulation Division. and Options. Washington, D.C.: World Bank. ————. 2002b. World Population Prospects: The 2002 Revision Volkov, Vadim. 2002. Violent Entrepreneurs: The Use of Force in the Population Database. New York: United Nations Department of Making of Russian Capitalism. New York: Cornell University Economic and Social Affairs Population Division. Press. (c) The International Bank for Reconstruction and Development / The World Bank References 239 Waelde, Thomas W., and George Ndi. 1996. “Stabilizing International Wheeler, David. 2001. “Racing to the Bottom? Foreign Investment Investment Commitments: International Law versus Contract and Air Pollution in Developing Countries.� Journal of Environ- Interpretation.� Texas International Law Journal 31(2):216–67. ment and Development 10(3):225–45. Waller, Irwin, and Daniel Sanfacon. 2000. “Investing Wisely in White, Lawrence J. 1996. “Competition versus Harmonization: An Crime Prevention: International Experiences.� Washington, Overview of International Regulation of Financial Services.� In D.C.: Department of Justice, Of�ce of Justice Programs, Bureau Claude E. Bar�eld, ed., International Financial Markets: Harmo- of Justice Assistance, Monograph 1. nization Versus Competition. Washington, D.C.: AEI Press. Wallsten, Scott. 2001. “An Econometric Analysis of Telecom Com- Wild, Volker. 1997. Pro�t Not For Pro�t’s Sake: History & Business petition, Privatization, and Regulation in Africa and Latin Amer- Culture of African Entrepreneurs in Zimbabwe. Harare, Zim- ica.� Journal of Industrial Economics 49(1):1–19. babwe: Baobab Books. ————. 2003. “Privatizing Monopolies in Developing Countries: Williamson, Oliver E. 1996. “The Institutions and Governance of The Real Effects of Exclusivity Periods.� Washington, D.C.: AEI- Economic Development and Reform.� In Oliver E. Williamson, Brookings Joint Center for Regulatory Studies 03-17. ed., Mechanisms of Governance. New York: Oxford University ————. 2004. “The Role of Government in Regional Technology Press. Development: The Effects of Science Parks and Public Venture Willig, Robert. 1999.“Economic Principles to Guide Post-privatization Capital.� In Timothy Bresnahan and Alfonso Gambardella, eds., Governance.� In Federico Basañes, Evamaría Uribe, and Robert Building High-Tech Clusters: Sillicon Valley and Beyond. Cam- Willig, eds., Can Privatization Deliver? Infrastructure for Latin Amer- bridge, U.K.: Cambridge University Press. ica. Washington, D.C.: Inter-American Development Bank. Warner, Mark A. 2000. “The Potential for Restraint Through an Wilson, James Q. 1991. On Character. Washington, D.C.: American International Charter for FDI.� In Paul Collier and Catherine Enterprise Institute. Patillo, eds., Investment and Risk in Africa. Basingtone: McMillan. Wilson, John Douglas. 1999. “Theories of Tax Competition.� Webb-Vidal, Andy. 2001. “New York Supercop Takes a Crack at National Tax Journal 52(2):269–304. Caracas.� Financial Times, March 22. Winston, Clifford. 1993. “Economic Deregulation: Days of Reckon- Wedel, Paul. 2002. Southeast Asia Competitiveness Programs: Thai- ing for Microeconomists.� Journal of Economic Literature land. Bangkok: Kenan Institute Asia. 31(3):1263–89. Weder, Beatrice, and Mirjam Schiffer. 2000. “Catastrophic Political Winter-Ebmer, Rudolph. 2001. “Long-term Consequences of an Risk versus Creeping Expropriation: A Cross-Country Analysis Innovative Redundancy-retraining Project: The Austrian Steel of Risks in Private Infrastructure Financing in Emerging Mar- Foundation.� Washington, D.C.: World Bank Social Protection kets.� University of Basel. Basel, Switzerland. Processed. Discussion Paper 0103. Weingast, Barry R. 1995. “The Economic Role of Political Institu- Winters, Alan, Neil McCulloch, and Andrew McKay. 2004. “Trade tions: Market-Preserving Federalism and Economic Develop- Liberalization and Poverty: The Evidence so Far.� Journal of Eco- ment.� Journal of Law, Economics, and Organization 11(1):1–31. nomic Literature 42(1):72–115. Weiss, Ricardo. 2002. “Advances in Corporate Governance in Wintrobe, Ronald. 1998. The Political Economy of Dictatorship. Brazil.� Institutional Investor 36(9): B16–B16. Cambridge: Cambridge University Press. Wellenius, Bjõrn. 1997a. “Extending Telecommunications Service Wodon, Quentin, Mohamed Ihsan Ajwad, and Corinne Siaens. to Rural Areas—The Chilean Experience: Awarding Subsidies 2003. “Lifeline or Means-testing? Electric Utility Subsidies in Through Competitive Bidding.� Washington, D.C.: World Bank Honduras.� In Penelope J. Brook and Timothy C. Irwin, eds., Public Policy for the Private Sector Note 105. Infrastructure for Poor People: Public Policy for Private Provision. ————. 1997b. “Telecommunications Reform-How to Succeed.� Washington, D.C.: World Bank. Washington, D.C.: World Bank Public Policy for the Private Sec- Wolfensohn, James D. 1998. The Other Crisis: Address to the Board tor Note 130. of Governors. Washington, D.C.: World Bank. Wells Jr., Louis T., and Eric S. Gleason. 1995. “Is Foreign Infrastruc- Wolman, Harold. 1988. “Local Economic Development Policy: ture Investment Still Risky?� Harvard Business Review What Explains the Divergence Between Policy Analysis and Sept.–Oct.:44–54. Political Behavior.� Journal of Urban Affairs 10(1):19–28. Wells, Louis T. Jr., Nancy J. Allen, Jacques Morisset, and Neda Pir- Wong, Poh-Kam, and Chee-Yueng Ng, eds. 2001. Industrial Policy, nia. 2001. “Using Tax Incentives to Compete for Foreign Invest- Innovation and Economic Growth: The Experience of Japan and ment: Are They Worth the Cost?� Washington, D.C.: Foreign the Asian NIEs. Singapore: Singapore University Press. Investment Advisory Service Occasional Paper 15. Woo-Cumings, Meredith. 2001. “Miracle as Prologue: The State West, Gerard T. 2001. “Comment on Securitizing Political Risk and the Reform of the Corporate Sector in Korea.� In Joseph E. Insurance.� In Theodore H. Moran, ed., International Political Stiglitz and Shahid Yusuf, eds., Rethinking East Asia. Washington, Risk Management: Exploring New Frontiers. Washington, D.C.: D.C.: Oxford University Press for the World Bank. World Bank–MIGA. World Bank. 1989. World Development Report 1989. Financial Sys- Westphal, Larry E. 2002. “Technology Strategies for Economic tems and Development. New York: Oxford University Press. Development in a Fast Changing Global Economy.� Economics of ————. 1993. The East Asian Miracle: Economic Growth and Innovation and New Technology 11(4-5):275–320. Public Policy. New York: Oxford University Press. (c) The International Bank for Reconstruction and Development / The World Bank 240 WORLD DEVELOPMENT REPORT 2005 ————. 1994a. Mexico: Agricultural Sector Memorandum. Wash- ————. 2002c. World Bank Group Work in Low Income Countries ington, D.C.: World Bank, Natural Resources and Poverty Divi- Under Stress: A Task Force Report. Washington, D.C.: World Bank. sion, Latin America and Caribbean Regional Of�ce, Country ————. 2002d. World Bank Policy Research Report 2002. Global- Department II. ization, Growth, and Poverty: Building an Inclusive World Econ- ————. 1994b. World Development Report 1994: Infrastructure omy. New York: Oxford University Press. for Development. New York: Oxford University Press. ————. 2003a. Doing Business in 2004—Understanding Regula- ————. 1995a. World Bank Policy Research Report. Bureaucrats in tion. Washington, D.C.: World Bank. Business: The Economics and Politics of Government Ownership. ————. 2003b. Global Economic Prospects and the Developing New York: Oxford University Press. Countries: Investing to Unlock Global Opportunities. Washington, ————. 1995b. World Development Report 1995: Workers in an D.C.: World Bank. Integrating World. New York: Oxford University Press. ————. 2003c. India: Sustaining Reform, Reducing Poverty. ————. 1996a. Morocco-Socioeconomic Influence of Rural Roads: Report 25797-IN. Washington, D.C.: World Bank. Fourth Highway Project. Washington, D.C.: World Bank, Opera- ————. 2003d. Jamaica: The Road to Sustained Growth. Report tions Evaluation Department. 26088-JM. Washington, D.C.: World Bank. ————. 1996b. World Development Report 1996: From Plan to ————. 2003e. Lifelong Learning in the Global Knowledge Econ- Market. New York: Oxford University Press. omy: Challenges for Developing Countries. A World Bank Report. ————. 1997. World Development Report 1997: The State in a Washington, D.C.: World Bank. Changing World. New York: Oxford University Press. ————. 2003f. Nicaragua Land Policy and Administration: ————. 1998a. Assessing Aid: What Works, What Doesn’t and Toward a More Secure Property Rights Regime. Report 26683-NI. Why. New York: Oxford University Press. Washington, D.C.: World Bank. ————. 1998b. Project Appraisal Document: Ghana Trade and ————. 2003g. Peru. Microeconomic Constraints to Growth: The Investment Gateway Project. Washington, D.C.: World Bank. Evidence from the Manufacturing Sector. Washington, D.C.: ————. 2000a. Bolivia: From Patronage to a Profesional State. World Bank. Washington, D.C.: World Bank. ————. 2003h. “Productivity and Investment Climate Survey�. ————. 2000b. Chad/Cameroon Petroleum Development and Washington, D.C., World Bank. Pipeline Project (Annex 11). Washington, D.C.: World Bank. ————. 2003i. Rural Financial Services: Implementing the Bank’s Available on line at http://www.worldbank.org/afr/ccproj/pro- Strategy to Reach the Rural Poor. Report 26030. Washington, D.C.: ject/pro_document.htm. World Bank, Agriculture and Rural Development Department. ————. 2000c. Reforming Public Institutions and Strengthening ————. 2003j. The Russian Labor Market: Moving from Crisis to Governance. Washington, D.C.: World Bank, Public Sector Recovery. Washington, D.C.: World Bank. Group, Poverty Reduction and Economic Management (PREM) ————. 2003k. Towards Country-Led Development: A Multi- Network. partner Evaluation of the Comprehensive Development Frame- ————. 2001a. Adjustment Lending Retrospective. Washington, work. Synthesis Report. Washington, D.C.: World Bank. D.C.: World Bank, Operations Policy and Country Services. ————. 2003l. World Bank Group Private Sector Development ————. 2001b. Poland Labor Market Study: The Challenge of Job Strategy Implementation Progress Report. Washington, D.C.: Creation. Washington, D.C.: World Bank. World Bank. ————. 2001c. Social Protection Sector Strategy: From Safety Net ————. 2003m. World Bank Policy Research Report 2003. Break- to Springboard. Washington, D.C.: World Bank. ing the Conflict Trap: Civil War and Development Policy. New ————. 2001d. Uganda. Country Assistance Evaluation: Policy, York: Oxford University Press. Participation, People. Washington, D.C.: World Bank, Operations ————. 2003n. World Bank Policy Research Report 2003. Land Evaluation Department. Policies for Growth and Poverty Reduction. New York: Oxford ————. 2001e. “World Bank Group Activities in the Extractive University Press. Industries.� World Bank. Washington, D.C. Processed. ————. 2003o. World Development Report 2003: Sustainable ————. 2001f. World Bank Policy Research Report 2001. Finance Development in a Dynamic World. New York: Oxford University for Growth: Policy Choices in a Volatile World. New York: Oxford Press. University Press. ————. 2003p. World Development Report 2004: Making Services ————. 2001g. World Bank Policy Research Report 2001: Engen- Work for Poor People. New York: Oxford University Press. dering Development Through Gender Equality In Rights, Resources ————. 2004a. 2003 Annual Review of Development Effectiveness: And Voice. New York: Oxford University Press. The Effectiveness of Bank Support for Policy Reform. Report 28290. ————. 2002a. Brazil Jobs Report 24480-BR. Washington, D.C: Washington, D.C.: World Bank Operations Evaluation Department. World Bank. ————. 2004b. Doing Business in 2005: Removing Obstacles to ————. 2002b. “Review of Leasing Across Selected Middle East Growth. Washington, D.C.: World Bank. and North Africa Countries.� World Bank. Washington, D.C. ————. 2004c. Global Development Finance 2004: Harnessing Processed. Cyclical Gains for Development. Washington, D.C.: World Bank. (c) The International Bank for Reconstruction and Development / The World Bank References 241 ————. 2004d. Global Economic Prospects 2004: Realizing the Development Studies and Monographs Series 14. Development Promise of the Doha Agenda. Washington, D.C.: Yergin, Daniel, and Joseph Stanislaw. 2002. The Commanding World Bank. Heights : The Battle for the World Economy. New York: Simon & ————. 2004e. Investment Climate Assessment: Enterprise Perfor- Schuster. mance and Growth In Tanzania. Washington, D.C: World Bank. Yeung, Godfrey. 2003. “Scramble for FDI: The Experience of Guan- ————. 2004f. MENA Development Report: Gender and Develop- dong Province in Southern China.� In Nicholas A. Phelps and ment in the Middle East and North Africa Region: Women in the Philip Raines, eds., The New Competition for Inward Investment: Public Sphere. Washington, D.C.: World Bank. Companies, Institutions and Territorial Development. Cheltenham, ————. 2004g. “Minimum Wages in Latin America and the U.K. and Northhampton, Mass.: Edward Elgar Publishing. Caribbean: The Impact on Employment, Inequality and Young, Alwyn. 2000. “Gold into Base Metals: Productivity Growth Poverty.� World Bank Of�ce of the Chief Economist for LAC. in the People’s Republic of China during the Reform Period.� Washington, D.C. Processed. Journal of Political Economy 111(6):1220–61. ————. 2004h. The Poverty Reduction Strategy Initiative: An ————. 2003. “The Razor’s Edge: Distortions and Incremental Independent Evaluation of the World Bank’s Support Through Reform in the People’s Republic of China.� Cambridge, Mass.: 2003. Washington, D.C: World Bank. National Bureau of Economic Research Working Paper Series ————. 2004i. Towards a Private Sector Development Strategy for 7828. Cambodia. Washington, D.C.: World Bank. Young, Alwyn. 1995. “The Tyranny of Numbers: Confronting the ————. 2004j. World Bank Policy Research Report 2004. Reform- Statistical Realities of the East Asian Growth Experience.� Quar- ing Infrastructure: Privatization, Regulation, and Competition. terly Journal of Economics 110(3):641–80. New York: Oxford University Press. Yufei, Pu, Sheng Lei, and Yao Yu. 2004. “Policy Reform on Invest- ————. 2004k. World Development Indicators. Washington, D.C.: ment Climate: Case Analysis of Hangzhou.� Background paper World Bank. for the WDR 2005. World Bank, and IFC. 2003. Race to the Top: Attracting and Enabling Yusuf, Shahid. 2003. Innovative East Asia: The Future of Growth. Global Sustainable Business. Washington, D.C.: World Bank and Washington, D.C.: World Bank. International Finance Corporation. Zee, Howell H., Janet G. Stotsky, and Eduardo Ley. 2002. “Tax World Bank, and PPIAF. 2003. Port Reform Toolkit: Effective Deci- Incentives for Business Investment: A Primer for Policy Makers sion Support for Policymakers. Washington, D.C.: World Bank. in Developing Countries.� World Development 30(9):1497–516. World Commission on the Social Dimension of Globalization. Zeruolis, Darius. 2003. “The Politics of Impact Assessment across 2004. A Fair Globalization—Creating Opportunities for All. the Government and in Euro-institutions. Lessons from Lithua- Geneva: International Labor Organization. nia.� Paper presented at the Regulatory Impact Assessment: Best World Economic Forum. 2002. Global Competitiveness Report Practices and Lesson-drawing in Europe Conference. So�a, Bul- 2001/02. Geneva, Switzerland: World Economic Forum. garia. February 27. ————. 2004. The Global Competitiveness Report 2003–2004. Zhang, Yin-Fang, David Parker, and Colin Kirkpatrick. 2002. “Elec- Geneva: World Economic Forum. tricity Sector Reform in Developing Countries: An Econometric Assessment of the Effects of Privatization, Competition, and World Energy Council. 2001. Pricing Energy in Developing Coun- Regulation.� Manchester: University of Manchester, Centre on tries. London, U.K.: World Energy Council. Available on line at Regulation and Competition, Institute for Development Policy http://www.worldenergy.org. and Management Working Paper 13. Wrong, Michela. 2001. In the Footsteps of Mr. Kurtz: Living on the Zongo, Tersius, Siaka Coulibaly, Gilles Hervio, Javier Nino Perez, Brink of Disaster in Mobutu’s Congo. New York: Harper Collins and Stefan Emblad. 2000. Conditionality Reform: The Burkina Publishers. Faso Pilot Case. Brussels: SPA Task Team on Contractual Rela- WTO. 2003. Working Group on the Interaction between Trade and tionships and Selectivity. Competition Policy. Study on Issues Relating to a Possible Multilat- eral Framework on Competition Policy: Note by the Secretariat. Geneva: World Trade Organization. Background Papers for the WDR 2005 Wunder, Haroldene F. 2001a. “Tanzi (1987): A Retrospective.� Bartelsman, Eric, John Haltiwanger, and Stefano Scarpetta. “Micro- National Tax Journal 54(4):763–70. economic Evidence of Creative Destruction in Industrial and Developing Countries.� ————. 2001b. “The Effect of International Tax Policy on Busi- ness Location Decisions.� Tax Notes International 24:1331–55. Burns, Anthony. “Thailand’s 20 Year Program to Title Rural Land.� Yahya, Saad S. 2002. “The Certi�cate of Rights Story in Botswana.� Byaruhanga, Charles. “Managing Investment Climate Reforms: In Geoffrey Payne, ed., Land, Rights & Innovation: Improving Case Study of Uganda Telecommunications.� Tenure Security for the Urban Poor. London: ITDG Publishing. Chen, Martha Alter, Renana Jhabvala, and Reema Nanavaty. “The Yaron, Jacob, McDonald P. Benjamin, and Gerda L. Piprek. 1997. Investment Climate for Female Informal Businesses: A Case “Rural Finance: Issues, Design, and Best Practices.� Washington, Study from Urban and Rural India.� D.C.: World Bank Environmentally and Socially Sustainable Clarete, Ramon L. “Customs Valuation Reform in the Philippines.� (c) The International Bank for Reconstruction and Development / The World Bank 242 WORLD DEVELOPMENT REPORT 2005 Coolidge, Jacqueline, Lars Grava, and Sanda Putnina, “Inspectorate Ray, Amit S. “Managing Port Reforms in India: Case Study of Jawa- Reform in Latvia, 1999–2003.� harlal Nehru Port Trust (JNPT) Mumbai.� De Wulf, Luc. “Tradenet in Ghana: Best Practice of the Use of Infor- Salas, Fernando José. “Mexican Deregulation: Smart Tape on Red mation Technology.� Tape.� Desai, Raj M. “ Political Influence and Firm Innovation: Micro- Surdej, Aleksander. “Managing Labor Market Reforms: Case Study Level Evidence from Developing Countries.� of Poland.� Echeverry, Juan Carlos, and Mauricio Santa María. “The Political Yufei, Pu, Sheng Lei, and Yao Yu. “Policy Reform on Investment Economy of Labor Reform in Colombia.� Climate—Case Analysis of Hangzhou.� Endo, Victor. “Managing Investment Climate Reforms: The Peru Urban Land Reform Case Study Draft No.2.� Case studies commissioned by the U.K. Finnegan, David. “Tanzania’s Commercial Court.� Department for International Development Gil Hubert, Johana. “The Mexican Credit Reporting Industry for the 2005 World Development Report Reform: A Case Study. “ Bannock, Graham, Matthew Gamser and Mariell Juhlin. “The Hallward-Driemeier, Mary, and Andrew H. W. Stone. “The Invest- Importance of the Enabling Environment for Business and Eco- ment Climate for Informal Firms.� nomic Growth: A 10 Country Comparison of Central Europe Hallward-Driemeier, Mary, and David Stewart. “How Do Invest- and Africa.� ment Climate Conditions Vary Across Countries and Types of Christianson, David. “The Investment Climate in South Africa— Firms?� Regulatory Issues: Some Insights from the High-Growth, Hernandez, Zenaida. “The Debate on Industrial Policy in East Asia: Export-Oriented SME Sector.� In Search for Lessons.� Estrin, Saul and Klaus Meyer. “Lessons for Development from Lon- Hubka, Ashley, and Rita Zaidi. “Innovations in Micro�nance.� don Business School’s Centre for New and Emerging Markets Irwin, Timothy. “Accounting for Public-Private Partnerships: How (CNEM) Research Project ‘Investment in Emerging Marketing’.� Should Governments Report Guarantees and Long-term Pur- Evenett, Simon J. “Competition Law and the Investment Climate in chase Contracts.� Developing Countries.� Jurajda, Stepan, and Katarína Mathernová. “How to Overhaul the Fortune, Peter. “Investment Climate Reform in Ukraine.� Labor Market: Political Economy of Recent Czech and Slovak Gamser, Matthew. “Improving the Business Trade Licensing Reforms.� Reform Environment.� Khan, Shamin Ahmad. “Business Registration Reforms in Pak- Holmes, Peter. “Some Lessons from the CUTS 7-Up: Comparative istan.� Competitive Policy Project.� Klapper, Leora F., and Rida Zaidi. “A Survey of Government Regu- Irwin, David. “Creating FDI Friendly Environments in South Asia.� lation and Intervention in Financial Markets.� Mackinnon, Gregor. “Lessons from CBC Business Environment Lund, Frances, and Caroline Skinner. “The Investment Climate for Surveys 1999, 2001, 2003.� the Informal Economy: A Case of Durban, South Africa.� Mehta, Pradeep. “Investment for Development Project (IFD)— Mallon, Raymond. “Managing Investment Climate Reforms: Viet- Civil Society Perceptions.� nam Case Study.� Preston, John. “Investment Climate Reform: Competition Policy Migliorisi, Stefano, and Marco Galmarini. “Donor Assistance to and Economic Development: Some Country Experiences.� Investment Climate Reforms.� Sander, Cerstin. “Migrant Remittances and the Investment Climate: Mitullah, Winnie V. “Street Vending in African Cities: A Synthesis of Exploring the Nexus.� Empirical Findings From Kenya, Cote D’Ivoire, Ghana, Zim- Sander, Cerstin. “Less is More: Better Compliance and Increased babwe, Uganda and South Africa.� Revenues by Streamlining Business Registration in Uganda.� Navarrete, Camilo. “Managing Investment Climate Reforms: Velde, Dirk Willem te . “OECD (UK & EU) Home Country Mea- Colombian Ports Sector Reform Case Study.� sures and FDI in Developing Countries; A Preliminary Analysis.� Olofsgård, Anders F. “The Political Economy of Reform: Institu- Vickers, Brendan. “Investment Climate Reform in South Africa.� tional Change as a Tool for Political Credibility.� White, Simon. “Donor Experiences in Supporting Reforms to the Pierre, Gaëlle, and Stefano Scarpetta. “Do Employers’ Perceptions Investment Climate for Small Enterprise Development.� Square with Actual Labor Regulations?.� Pierre, Gaëlle, and Stefano Scarpetta. “How Labor Market Policy Can Combine Workers’ Protection and Job Creation.� (c) The International Bank for Reconstruction and Development / The World Bank Selected Indicators Measuring the investment climate 244 Introduction Table A1. Investment climate indicators: World Bank Investment Climate Surveys Table A2. Investment climate indicators: expert polls and other surveys Technical notes Selected world development indicators 253 Introduction Classi�cation of economies by region and income Table 1. Key indicators of development Table 2. Poverty and income distribution Table 3. Economic activity Table 4. Trade, aid, and �nance Table 5. Key indicators for other economies Technical notes (c) The International Bank for Reconstruction and Development / The World Bank Measuring the investment climate Firms evaluating alternative investment options, governments approaches to regulation and access to modern telecom- interested in improving their investment climates, and econo- munications services. The many factors can also interact mists seeking to understand the role of different factors in in various ways. The lack of secure property rights can explaining economic performance—all have grappled with lead to dif�culties in getting �nance on reasonable terms. de�ning and measuring the investment climate. The number And the level of taxes affects the ability of governments to of organizations working in this area has expanded the vari- provide public services, including those that bene�t ables available. For example, The PRS Group’s International �rms. Similarly, the level of corruption is not only a direct Country Risk Guide and Business Environment Risk Intelli- cost to �rms but can also lead to deep distortions across gence give measures on various sources of country risk based the policymaking apparatus of government. Reducing on evaluations of international experts (additional examples such details into a single measure misses the insights from and their websites are in a table at the end of the technical a more disaggregated analysis, and hides the degree of notes). The World Economic Forum looks at a wider range of variation within a country. factors thought to affect competitiveness based on relatively • Some dimensions are inherently dif�cult to measure. Cer- small samples of mostly multinational �rms. The Worldwide tain investment climate constraints are relatively easy to Governance Research Indicators Dataset draw on sources from identify and measure, such as the reliability of the power 18 different organizations to create six governance indicators, supply or the time it takes to register a business. Others including “rule of law,� “government effectiveness� and “con- are more sensitive, such as issues dealing with corruption, trol of corruption.� While these and related variables have con- and can lead to underreporting. Other dimensions are tributed to cross-country analysis, such broad assessments do harder to quantify such as competitive pressures and pol- not translate easily into diagnoses of speci�c problems or cap- icy related risks. However, omitting important dimen- ture the nuances of different institutional settings. sions because the measurements have not been perfected To complement and extend these efforts, the World Bank, would give a distorted assessment. Alternatively, the col- working with client governments and others, recently pio- lection of the wider set of information can be pursued, neered new measures of the investment climate. The Invest- with the evaluation of the responses taking into account ment Climate Surveys measure speci�c constraints facing the nature of the subject matter being reported. �rms, and relate them to measures of �rm performance, • Differences in perspective across �rms and activities. Even a growth, and investment. The Doing Business Project collects single dimension of the investment climate can affect country level data on the details of a set of regulations. The �rms or activities in different ways. For example, de�- Report draws on both sets of data and presents selections from ciencies in port and customs infrastructure can be a these databases in the following tables. major impediment to �rms engaged in exporting and have only more limited and indirect effects on other Challenges in measuring �rms. Similarly, some �rms may bene�t from govern- the investment climate ment-mandated monopolies, while other �rms lose by being denied the opportunity to compete or by paying All efforts to develop more speci�c insights and related data higher prices for products from the protected industry. have to contend with �ve main challenges: Burdens that represent �xed costs also result in a dispro- • Multidimensional nature of the concept being measured. portionate burden on smaller �rms. In addition, some Stability and corruption are important, but so are variables that may impose a burden on �rms may provide 244 (c) The International Bank for Reconstruction and Development / The World Bank Measuring the investment climate 245 other social bene�ts. Examples include levying taxes to New investment climate measures from the World Bank improve public services or meet other social goals, or reg- Investment Climate Doing Surveys Business Project ulations to safeguard the environment or consumers. Simply relying on opinions from �rms could lead to Country coverage Launched in 2001, this Report Initially covering 130 questionable policy advice. But even objective responses draws on over 26,000 �rms countries in 2003, in 53 countries. Each year additional countries can vary by type of respondent. Ideally measures would an additional 15–20 surveys are being added. capture the range of perspectives and evaluations of con- are �elded. straints. Investment climate The standard questionnaire Beginning with 5 areas dimensions covered of 82 questions covers of regulation (business • Differences between locations within countries. Investment regulations, governance, registration, insolvency, climate conditions are not uniform in each country, with access to �nance, and contract enforcement, infrastructure services. It hiring and �ring workers signi�cant differences often evident across locations. This also collects data on �rm and accessing credit), is most obviously the case in large countries with federal productivity, investment, and additional topics are structures, where sub-national governments may differ in employment decisions. being added. their policies and behaviors. But it also true with more Types of variables Covers both objective and Objective measures of perception data. The the number of centralized governments, where there are often important objective data includes the procedures, the time to differences within the country in matters like infrastruc- time to complete processes compete them, and the and monetary costs of fees and costs ture provision and even the enforcement of national laws various disruptions and associated with and regulations. regulations. In addition, compliance. respondents give • The experience on the ground does not always reflect formal perceptions of potential policies. The policies as they exist on the books are not constraints and assessments always implemented. In some countries, the gap between of risks and competition. the formal policy and its implementation is substantial. Whose perspective Surveys cover a diverse Use a single, de�ned, range of sizes and activities, hypothetical �rm and Variations in the degree of discretion of�cials have, the with random samples of transaction. Judgments resources put into implementation and the political will several hundred �rms. Data based on assessment is gathered through face-to- of up to 5 local experts to enforce existing regulations can have a big impact. The face interviews conducted (lawyers, accountants). distinction can be important in determining the priori- with senior managers and ties and expected bene�ts of reform initiatives. accountants. Differences within Samples cover multiple A single indicator is In grappling with these issues, objective and perception based a country locations within each given for the largest data each can make a contribution. Objective measures have country. city in the country. For some large countries, advantages of allowing more precise and consistent bench- additional cities are marking of conditions. But for some factors, subjective indica- available. tors may be the only effective way to reflect differences across Basis of assessment Indicators are based on the Indicators measure experience reported by �rms, formal regulatory locations or types of �rms. As investment decisions ultimately providing ranges of how requirements. depend on subjective judgments, measures that reflect �rm policies are implemented perceptions add additional insights. in practice. The World Bank’s new measures The table illustrates how Investment Climate Surveys and the Guatemala, India, Indonesia, Kenya, Pakistan, Senegal, Tanza- Doing Business Project address these challenges, providing nia, and Uganda. complementary sources of indicators. Together, they provide Additional information and access to these datasets can be new insights in the investment climates of a growing number obtained at: of countries. econ.worldbank.org/wdr/wdr2005 The WDR team also adapted the Investment Climate Sur- vey methodology to surveys of micro and informal �rms in 11 iresearch.worldbank.org/ics countries. These comprise Bangladesh, Brazil, Cambodia, rru.worldbank.org/DoingBusiness. (c) The International Bank for Reconstruction and Development / The World Bank 246 WORLD DEVELOPMENT REPORT 2005 Table A1. Investment climate indicators: World Bank Investment Climate Surveys Policy Uncertainty Corruption Courts Crime Unpredictable Report Lack con�dence Report Major interpretation Major bribes Av. Major courts uphold Major losses Av. loss constraint of regulations constraint are paid Bribe constraint property rights constraint from crime from crime Survey Sample % of % of year size % % % % sales % % % % sales Albania 2002 170 48.5 54.5 47.5 84.5 4.6 32.9 50.6 21.2 11.8 1.4 Algeria 2003 557 .. 44.8 35.2 75.0 8.6 .. 27.3 .. 11.0 12.2 Armenia 2002 171 32.0 51.6 13.5 35.7 4.8 8.2 44.1 3.6 9.4 14.1 Azerbaijan 2002 170 6.7 48.3 19.5 63.5 6.0 4.4 31.0 2.6 6.5 12.9 Bangladesh 2002 1,001 45.4 21.4 57.9 97.8 2.8 .. 83.0 39.4 23.5 2.3 Belarus 2002 250 59.0 77.6 17.9 62.0 3.4 11.2 48.1 12.3 21.6 3.8 Bhutana 2002 96 .. .. .. .. .. .. .. 2.3 .. .. Boliviaa 2001 671 .. .. .. 40.5 .. .. .. .. .. .. Bosnia & Herzegovina 2002 182 40.5 47.0 34.8 62.6 3.0 22.6 38.0 18.7 13.7 1.7 Brazil 2003 1,642 75.9 66.0 67.2 51.0 .. 32.8 39.6 52.2 22.7 2.8 Bulgaria 2002 250 59.5 62.3 25.4 75.9 4.2 17.9 50.6 18.8 34.4 2.7 Cambodia 2003 503 40.1 44.4 55.9 82.3 6.0 31.4 61.0 41.7 20.1 7.0 China 2002/3 3,948 32.9 33.7 27.3 55.0 2.6 .. 17.5 20.0 10.4 2.6 Croatia 2002 187 35.9 51.4 22.5 48.7 2.6 27.6 33.3 8.5 13.4 2.1 Czech Rep. 2002 268 20.2 56.0 12.5 55.5 2.9 11.1 47.1 14.3 33.6 3.1 Ecuador 2003 453 60.7 68.0 49.2 58.9 5.4 34.1 70.8 27.8 36.4 3.5 Eritreaa 2002 78 31.5 .. 2.7 64.1 3.8 .. .. 1.3 .. .. Estonia 2002 170 12.0 45.1 5.4 48.8 1.1 4.8 28.6 6.5 35.9 0.5 Ethiopiaa 2002 427 39.3 .. 39.0 .. .. .. .. 9.5 11.5 7.1 Georgia 2002 174 44.3 73.4 35.1 81.5 4.4 11.2 59.0 19.0 27.6 7.0 Guatemala 2003 455 66.4 89.5 80.9 57.6 7.4 36.7 71.3 80.4 42.2 4.8 Honduras 2003 450 47.0 65.9 62.8 50.0 6.0 21.8 56.1 60.9 3.3 3.1 Hungary 2002 250 21.1 42.7 8.8 60.4 2.4 4.5 40.3 4.9 33.6 1.1 Indiab 2003 1,827 20.9 64.1 37.4 .. .. .. 29.4 15.6 .. .. Indonesia 2004 713 48.2 56.0 41.5 50.9 4.6 24.7 40.8 22.0 15.6 3.1 Kazakhstan 2002 250 18.5 52.7 14.2 69.2 3.8 4.0 48.5 8.4 29.2 3.5 Kenya 2003 284 51.5 45.5 73.8 75.5 5.5 .. 51.3 69.8 31.0 4.1 Kyrgyzstan 2002/3 275 34.7 67.0 31.4 82.4 4.6 15.7 66.3 18.5 27.3 8.2 Latvia 2002 176 27.4 71.4 11.7 62.6 2.3 3.2 49.1 6.4 33.0 2.7 Lithuania 2002 200 33.5 61.9 15.6 52.0 1.9 12.0 59.5 16.2 38.0 2.8 Macedonia, FYR 2002 170 37.3 42.3 31.2 68.7 1.5 27.1 50.6 20.4 14.1 6.7 Malaysia 2003 902 22.4 .. 14.5 .. .. .. 19.1 11.4 19.1 3.0 Moldova 2002/3 277 57.0 79.0 40.2 77.6 3.0 19.8 72.1 26.5 17.3 3.9 Moroccoa 2001 859 .. .. .. .. .. .. .. .. .. .. Nicaragua 2003 452 58.2 66.4 65.7 45.5 7.0 33.3 60.4 39.2 2.7 7.0 Nigeriaa 2001 232 .. 55.1 .. .. .. .. .. 36.3 .. .. Pakistan 2002 965 40.1 64.8 40.4 59.0 3.6 .. 62.6 21.5 8.8 2.5 Perua 2002 583 71.1 78.7 59.6 .. .. .. 34.7 51.6 21.8 10.2 Philippines 2003 719 29.5 49.1 35.2 50.6 4.0 .. 33.8 26.5 27.1 4.2 Poland 2002/3 608 59.1 68.0 27.6 52.4 3.1 27.0 46.2 24.9 31.6 2.8 Romania 2002 255 43.3 54.5 34.9 73.3 4.7 20.9 45.8 19.8 24.7 3.8 Russia 2002 506 31.5 75.1 13.7 78.0 2.3 9.5 65.3 12.4 36.4 2.9 Senegal 2004 262 31.3 42.5 39.9 45.2 1.8 13.3 40.5 15.4 47.0 2.1 Serbia & Montenegro 2002 250 47.8 42.9 16.3 61.6 4.0 13.8 28.6 8.9 22.4 4.6 Slovakia 2002 170 44.6 55.1 27.5 68.1 2.6 25.3 53.9 15.4 42.9 1.8 Slovenia 2002 188 11.8 47.8 6.1 36.2 5.4 8.0 45.6 3.3 19.7 2.8 Tajikistan 2002/3 283 24.4 56.3 21.0 76.7 3.7 9.1 48.2 3.0 20.1 4.2 Tanzania 2003 276 31.5 58.6 51.1 42.9 2.9 20.0 55.1 25.5 25.7 3.2 Turkey 2002 514 53.8 40.6 23.7 71.8 0.6 11.9 33.1 12.9 5.8 2.7 Uganda 2003 300 27.6 40.0 38.2 39.0 4.9 .. 30.1 26.8 .. .. Ukraine 2002 463 46.9 67.5 27.8 70.2 4.4 15.3 49.0 19.6 27.9 4.7 Uzbekistan 2002/3 360 27.2 42.3 8.7 57.7 2.6 7.6 25.4 7.0 6.7 10.4 Zambia 2003 207 57.0 70.1 46.4 49.5 3.8 38.6 36.0 48.8 79.7 4.4 (c) The International Bank for Reconstruction and Development / The World Bank Measuring the investment climate 247 Table A1. Investment climate indicators: World Bank Investment Climate Surveys—continued Regulation and tax administration Finance Electricity Labor Small Labor Tax rates Tax admin. Licensing Mgt. time Avg. days �rms Losses Skills as regul. as major as major as major dealing with to clear Major with Major Firms reporting from major major constraint constraint constraint of�cials customs constraint a loan constraint outages outages constraint constraint % % % % mgt time Days % % % % % of sales % % Albania 37.1 25.0 22.9 13.6 2.4 20.1 7.8 57.1 .. .. 13.2 7.3 Algeria 44.8 36.2 27.4 .. 21.6 51.3 27.1 11.5 58.9 8.9 25.5 12.9 Armenia 35.5 37.7 9.0 7.4 3.7 25.9 11.1 15.8 .. .. 6.0 1.8 Azerbaijan 18.8 17.5 10.1 7.3 2.6 12.3 4.9 20.2 .. .. 4.5 1.3 Bangladesh 35.8 50.7 22.5 4.6 11.5 45.7 48.8 73.2 58.5 5.2 19.8 10.8 Belarus 47.0 44.2 25.8 11.0 2.4 30.1 8.3 2.8 .. .. 8.4 9.3 Bhutan .. .. .. .. 3.1 .. 50 5.6 .. .. .. .. Bolivia .. .. .. .. 9.3 .. .. .. .. .. .. .. Bosnia & Herzegovina 26.9 26.0 11.9 11.7 3.6 27.9 23.2 5.6 .. .. 5.7 9.1 Brazil 84.5 66.1 29.8 9.4 13.8 71.7 51.6 20.3 40.1 3.8 39.6 56.9 Bulgaria 33.1 13.0 15.1 8.5 4.2 40.3 9.0 8.0 .. .. 10.2 7.8 Cambodia 18.6 20.7 11.7 14.6 .. 9.9 7.9 12.7 38.6 5.2 6.6 5.9 China 36.8 26.7 21.3 19.0 7.9 22.3 52.0 29.7 38.0 5.0 30.7 20.7 Croatia 27.8 7.7 9.2 9.0 3.8 21.6 33.3 1.1 .. .. 8.7 5.4 Czech Rep. 25.6 19.8 10.2 5.5 4.4 23.1 32.2 5.3 .. .. 9.1 3.5 Ecuador 38.1 28.5 13.0 17.7 16.4 42.2 54.6 28.3 46.4 5.7 22.3 14.1 Eritrea 31.1 16.2 2.7 5.9 9.1 53.7 26.3 38.2 41.0 12.8 41.0 5.2 Estonia 16.7 4.5 11.2 6.2 1.6 8.4 46.0 10.1 .. .. 23.8 4.2 Ethiopia 73.6 60.3 8.3 5.7 13.5 40.2 26.3 42.5 65.6 7.7 17.9 4.6 Georgia 30.5 47.1 9.9 14.7 3.2 14.2 19.6 22.4 .. .. 8.6 4.0 Guatemala 56.5 34.8 15.6 17.4 9.4 38.7 43.5 26.6 60.7 3.7 31.4 16.7 Honduras 35.6 23.2 21.1 14.2 5.1 55.4 46.9 36.4 58.0 5.2 26.4 14.2 Hungary 30.2 13.7 3.3 8.7 4.3 20.2 18.5 1.2 .. .. 12.5 7.3 India 27.9 26.4 13.4 15.3 6.7 19.2 51.1 28.9 69.2 11.6 12.5 16.7 Indonesia 29.5 23.0 20.5 14.6 5.8 23.0 16.7 22.3 33.0 6.1 18.9 25.9 Kazakhstan 13.8 14.3 9.0 14.6 5.3 14.0 13.3 3.6 .. .. 6.3 0.8 Kenya 68.2 50.9 15.2 13.8 8.9 58.3 59.3 48.1 58.5 14.9 27.6 22.5 Kyrgyzstan 32.5 35.1 11.6 13.2 3.3 27.7 9.3 4.7 46.1 3.2 7.7 4.5 Latvia 27.3 27.6 9.2 10.7 1.2 7.6 23.2 4.0 .. .. 15.5 4.1 Lithuania 36.5 19.8 8.1 10.0 2.4 7.0 21.1 4.5 .. .. 7.5 8.5 Macedonia, FYR 21.0 15.1 17.4 13.5 5.0 16.6 11.1 5.4 .. .. 3.7 4.6 Malaysia 21.7 13.3 10.9 10.2 3.6 17.8 57.3 14.8 40.6 5.2 25.0 14.5 Moldova 54.9 47.6 24.6 7.1 2.1 39.6 26.4 5.4 15.5 0.8 11.0 5.2 Morocco .. .. .. .. 2.7 .. 34.2 .. .. .. .. .. Nicaragua 34.7 18.1 10.6 17.3 5.8 57.6 42.0 34.7 59.5 7.1 17.0 6.9 Nigeria .. .. .. .. 17.8 .. 11.1 97.4 .. .. .. .. Pakistan 45.6 46.1 14.5 10.6 17.2 40.1 11.2 39.2 81.3 6.7 12.8 15.0 Peru .. .. .. .. 7.9 55.8 43.6 11.1 30.5 6.3 12.5 .. Philippines 30.4 25.1 13.5 11.0 2.8 18.2 16.8 33.4 41.6 9.6 11.9 24.7 Poland 64.7 41.0 13.5 12.3 3.1 42.6 31.5 5.8 18.5 0.7 12.2 25.2 Romania 51.6 33.2 23.2 10.7 1.4 32.3 25.5 9.5 .. .. 10.8 8.1 Russia 24.6 31.8 14.6 14.1 6.9 17.0 8.8 4.6 .. .. 9.9 3.3 Senegal 50.8 48.2 7.5 13.8 6.5 60.0 23.2 30.7 49.4 9.6 18.5 16.3 Serbia & Montenegro 35.3 29.3 7.8 15.1 5.5 28.3 11.3 6.2 .. .. 11.9 6.9 Slovakia 31.7 19.8 17.9 9.5 2.2 30.1 41.2 3.0 .. .. 9.7 7.4 Slovenia 11.2 5.9 3.2 7.7 3.1 11.2 23.8 0.5 .. .. 4.3 2.7 Tajikistan 26.2 21.8 14.2 8.3 9.6 20.1 2.0 17.1 63.6 5.7 2.4 2.3 Tanzania 73.4 55.7 27.4 16.2 17.5 53.0 13.3 58.9 .. .. 25.0 12.1 Turkey 38.1 33.1 5.8 8.0 3.7 23.2 11.3 17.3 .. .. 12.8 8.7 Uganda 48.3 36.1 10.1 5.0 .. 52.8 14.1 44.5 41.7 13.1 30.8 10.8 Ukraine 39.6 34.9 18.2 15.4 5.8 29.1 6.5 5.9 .. .. 13.0 5.8 Uzbekistan 19.9 22.7 7.7 12.1 6.0 20.6 2.3 4.8 19.0 5.6 4.9 1.7 Zambia 57.5 27.5 10.1 14.1 4.8 67.7 29.6 39.6 63.8 6.6 35.7 16.9 Data are based on enterprise surveys conducted by the World Bank and its partners in the year indicated. While averages are reported, there are signi�cant variations across �rms. The data are not intended for the ranking of countries. The WDR Survey of Micro and Informal Firms was also conducted in 11 countries: Bangladesh, Brazil, Cambodia, Guatemala, India, Indonesia, Kenya, Pakistan, Senegal, Tanzania, and Uganda. The �ndings of these surveys are not reflected in this table. For more information, see Hallward-Driemeier and Stone (2004). “..� indicates data is not available. a. In 2002 the survey was expanded, so the earliest surveys include the �rm performance measures, but not the full set of investment climate variables. b. India’s �rst round survey of 895 �rms was conducted in 2000. (c) The International Bank for Reconstruction and Development / The World Bank Table A2. Investment climate indicators: expert polls and other surveys World Bank’s Doing Business Project Intensity Transparency Regional Starting a Enforcing a Registering Resolving Investment of local of gov’t disparities of business contract property insolvency Pro�le competition policymaking bus. environ. Days Procedures Days Procedures Days Procedures Years ICRG WEF index WEF index WEF index Jan-04 Jan-04 Jan-04 Jan-04 Jan-04 Jan-04 Jan-04 2003 2003/4 2003/4 2003/4 Albania 47 11 390 39 47 7 4 8 .. .. .. Algeria 26 14 407 49 52 16 3.5 8 3.5 3.6 2.7 Angola 146 14 1011 47 335 8 4.7 8.5 2.4 2.5 2.8 Argentina 32 15 520 33 44 5 2.8 5 4.4 2 2.8 Armenia 25 10 195 24 18 4 1.9 8 .. .. .. Australia 2 2 157 11 7 5 1 10 5.4 5.6 5.1 Austria 29 9 374 20 32 3 1 12 5.1 4 5.1 Azerbaijan 123 14 267 25 61 7 2.7 9 .. .. .. Bangladesh 35 8 365 29 .. .. 4 5.25 4.8 3 2.9 Belarus 79 16 250 28 231 7 5.8 5.5 .. .. .. Belgium 34 4 112 27 132 2 0.9 11.5 5.6 3.9 3.8 Benin 32 8 570 49 50 3 3.1 .. .. .. .. Bhutan 62 11 275 20 44 4 .. .. .. .. Bolivia 59 15 591 47 92 7 1.8 9.5 3.8 3 3 Bosnia & Herzegovina 54 12 330 36 331 7 3.3 .. .. .. .. Botswana 108 11 154 26 69 4 2.2 11.5 4.1 5.1 3.8 Brazil 152 17 566 25 42 14 10 7.5 5.2 3.6 2.1 Bulgaria 32 11 440 34 19 9 3.3 11.5 4.6 2.7 3 Burkina Faso 135 13 458 41 107 8 4 9 .. .. .. Burundi 43 11 512 51 94 5 4 .. .. .. .. Cambodia 94 11 401 31 56 7 .. .. .. .. .. Cameroon 37 12 585 58 93 5 3.2 6.5 4.1 4.4 2.8 Canada 3 2 346 17 20 6 0.8 12 5.5 4.5 4.1 Central African Rep. 14 10 660 45 69 3 4.8 .. .. .. .. Chad 75 19 526 52 44 6 10 .. 3.6 2.5 2.3 Chile 28 10 305 28 31 6 5.6 11 5.6 4.5 3.3 China 41 12 241 25 32 3 2.4 7.5 5.3 4.2 3.3 Hong Kong, China 11 5 211 16 56 3 1.1 11.5 5.6 5.4 5.2 Colombia 43 14 363 37 23 7 3 9.25 4.6 4 2.8 Congo, Dem. Rep. 155 13 909 51 106 8 5.2 6 .. .. .. Congo, Rep. 67 8 560 47 103 6 3 8.5 .. .. .. Costa Rica 77 11 550 34 21 6 3.5 8.5 4.7 3.9 3.7 Côte d’Ivoire 58 11 525 25 340 7 2.2 6 .. .. .. Croatia 49 12 415 22 956 5 3.1 9 4.6 3.1 2.8 Czech Rep. 40 10 300 22 122 4 9.2 12 5.1 3.5 3.2 Denmark 4 4 83 15 42 6 3.4 11.5 5.5 5.2 5 Dominican Rep. 78 10 580 29 107 7 3.5 8.5 4.5 3.4 3.3 Ecuador 92 14 388 41 21 12 4.3 6 3.5 2.5 2.9 Egypt, Arab Rep. 43 13 410 55 193 7 4.2 6.5 4.4 3.4 3.6 El Salvador 115 12 275 41 52 5 4 6 5 4 3.3 Eritrea .. .. .. .. .. .. .. .. .. .. Estonia 72 6 150 25 65 4 3 10 5.3 4.2 2.7 Ethiopia 32 7 420 30 56 15 2.4 7 3.6 3 2.2 Finland 14 3 240 27 14 3 0.9 12 5.4 5.5 4.3 France 8 7 75 21 193 10 1.9 12 5.4 4.4 4.2 Gambia, The .. .. .. .. .. .. .. 8.5 4.2 4.7 3.4 Georgia 25 9 375 18 39 8 3.2 .. .. .. .. Germany 45 9 184 26 41 4 1.2 12 5.5 4.5 4.8 Ghana 85 12 200 23 382 7 1.9 7 4.3 4.3 3 Greece 38 15 151 14 23 12 2 11 5.1 3.6 3 Guatemala 39 15 1459 37 55 5 4 11 4.1 2 2.7 Guinea 49 13 306 44 104 6 3.8 6.5 .. .. .. Haiti 203 12 368 35 195 5 5.7 5 4 2.7 1.5 Honduras 62 13 545 36 36 7 3.7 8 3.4 2.9 3.5 Hungary 52 6 365 21 79 4 2 12 4.9 3.9 2.3 Iceland .. .. .. .. .. .. .. 11 5.3 5.3 4.3 India 89 11 425 40 67 6 10 8 5.6 4.1 2.5 Indonesia 151 12 570 34 33 6 6 4.5 4 3.6 3.6 Iran, Islamic Rep. 48 9 545 23 36 9 4.5 6 .. .. .. Ireland 24 4 217 16 38 5 0.4 12 5.2 4.2 3.8 Israel 34 5 585 27 144 7 4 9 5.6 4.2 5 Italy 13 9 1390 18 27 8 1.2 12 5.3 3.9 2.6 Jamaica 31 7 202 18 54 5 1.1 9.5 4.9 3.5 4 Japan 31 11 60 16 14 6 0.5 12 5.5 3.9 4.5 Jordan 36 11 342 43 22 8 4.3 9.5 5.2 4.4 3.4 Kazakhstan 25 9 400 41 52 8 3.3 7.5 .. .. .. Kenya 47 12 360 25 39 7 4.5 9 5.2 3.6 2.8 Korea, Rep. 22 12 75 29 11 7 1.5 9.5 5.3 4.4 3.8 Kuwait 35 13 390 52 75 8 4.2 11 .. .. .. Kyrgyz Rep. 21 8 492 46 15 7 3.5 .. .. .. .. Lao PDR 198 9 443 53 135 9 5 .. .. .. .. Latvia 18 7 189 23 62 10 1.1 11 5 4.1 3.6 Lebanon 46 6 721 39 25 8 4 9 .. .. .. Lesotho 92 9 285 49 101 6 2.6 .. .. .. .. Lithuania 26 8 154 17 3 3 1.2 11 5.1 3.8 3 Luxembourg .. .. .. .. .. .. .. 12 4.4 5.3 5.1 Macedonia, FYR 48 13 509 27 74 6 3.7 .. 4.3 3.8 3.7 Madagascar 44 13 280 29 .. .. .. 8 4.2 3.5 1.9 Malawi 35 10 277 16 118 6 2.6 8 4.2 4 2.9 Malaysia 30 9 300 31 143 4 2.3 8.5 5.3 5 3.9 248 (c) The International Bank for Reconstruction and Development / The World Bank Table A2. Investment climate indicators: expert polls and other surveys—continued World Bank’s Doing Business Project Intensity Transparency Regional Starting a Enforcing a Registering Resolving Investment of local of gov’t disparities of business contract property insolvency Pro�le competition policymaking bus. environ. Days Procedures Days Procedures Days Procedures Years ICRG WEF index WEF index WEF index Jan-04 Jan-04 Jan-04 Jan-04 Jan-04 Jan-04 Jan-04 2003 2003/4 2003/4 2003/4 Mali 42 13 340 28 44 5 3.6 7.5 3.8 3.5 2.5 Malta .. .. .. .. .. .. .. 11.5 5 4.8 5.5 Mauritania 82 11 410 28 49 4 8 .. .. .. .. Mauritius .. .. .. .. .. .. .. .. 4.9 4.5 4.4 Mexico 58 8 421 37 74 5 1.8 11.5 4.9 3.7 2.5 Moldova 30 10 280 37 81 5 2.8 6.5 .. .. .. Mongolia 20 8 314 26 10 4 4 8 .. .. .. Morocco 11 5 240 17 82 3 1.8 9 4.4 4.2 2.5 Mozambique 153 14 580 38 33 7 5 8.5 3.2 3.4 2.1 Myanmar .. .. .. .. .. .. 4 .. .. .. Namibia 85 10 270 31 28 9 1.0 10 4.4 4.2 3 Nepal 21 7 350 28 .. .. 5 .. .. .. .. Netherlands 11 7 48 22 5 4 1.7 12 5.6 4.8 5.1 New Zealand 12 2 50 19 2 2 2 11.5 5.7 5.2 4.9 Nicaragua 45 9 155 18 65 7 2.2 6 3.2 2.9 2.9 Niger 27 11 330 33 49 5 5 7.5 .. .. .. Nigeria 44 10 730 23 274 21 1.5 3.5 4.7 3.5 2.9 Norway 23 4 87 14 1 1 0.9 11.5 5.1 3.8 3.9 Oman 34 9 455 41 16 4 7 11.5 .. .. .. Pakistan 24 11 395 46 49 5 2.8 4.5 5 3.5 2.8 Panama 19 7 355 45 44 7 2 9.5 4.5 2.8 3.4 Papua New Guinea 56 8 295 22 72 4 2.8 8 .. .. .. Paraguay 74 17 285 46 48 7 3.9 8.5 4.1 2.2 3.3 Peru 98 10 441 35 31 5 3.1 7.5 4.6 2.9 2.2 Philippines 50 11 380 25 33 8 5.6 10 5 3.7 2.5 Poland 31 10 1000 41 204 7 1.4 11 4.8 2.9 2.8 Portugal 78 11 320 24 83 5 2.5 12 5 3.7 2.8 Puerto Rico 7 7 270 43 .. .. 3.8 .. .. .. .. Romania 28 5 335 43 170 8 4.6 8.5 3.6 2.6 2.8 Russian Federation 36 9 330 29 37 6 1.5 9 4 2.5 2.3 Rwanda 21 9 395 29 354 5 .. .. .. .. .. Saudi Arabia 64 12 360 44 4 4 2.8 11 .. .. .. Senegal 57 9 485 36 114 6 3 8 4.3 3.9 2.6 Serbia & Montenegro 51 11 1028 36 186 6 2.6 8 4.1 4.1 2.8 Sierra Leone 26 9 305 58 58 8 2.5 6.5 .. .. .. Singapore 8 7 69 23 9 3 0.8 12 5.4 6.2 5.8 Slovak Rep. 52 9 565 27 22 5 4.7 12 4.7 3.4 2.2 Slovenia 61 10 1003 25 391 6 3.6 10 4.9 4.2 3.4 South Africa 38 9 277 26 20 6 2 10.5 5.3 4.3 2.9 Spain 108 7 169 23 20 4 1 12 5.5 4.2 3.9 Sri Lanka 50 8 440 17 63 8 2.2 8.5 4.7 3.7 3.4 Sweden 16 3 208 23 2 1 2 12 5.5 5.2 4.1 Switzerland 20 6 170 22 16 4 4.6 11.5 5.1 5.3 4.7 Syrian Arab Rep. 47 12 672 48 23 4 4.1 6.5 .. .. .. Tajikistan .. .. .. .. .. .. .. .. .. .. .. Tanzania 35 13 242 21 61 12 3 7.5 4.7 4.1 2.6 Thailand 33 8 390 26 2 2 2.6 8.5 5.3 4.3 4.1 Togo 53 13 535 37 212 6 3 7.5 .. .. .. Trinidad & Tobago .. .. .. .. .. .. .. 11.5 4.8 3.9 4.3 Tunisia 14 9 27 14 57 5 1.3 8 4.5 5.1 3.4 Turkey 9 8 330 22 9 8 2.9 7.5 4.7 3.4 2.2 Turkmenistan .. .. .. .. .. .. .. .. .. .. .. Uganda 36 17 209 15 48 8 2.1 8.5 4.4 3.9 2.7 Ukraine 34 15 269 28 93 9 2.6 6 4.1 2.2 2.7 United Arab Emirates 54 12 614 53 9 3 5.1 11.5 .. .. .. United Kingdom 18 6 288 14 21 2 1 12 6 5 4.3 United States 5 5 250 17 12 4 3 12 5.9 4.9 5.2 Uruguay 45 11 620 39 66 8 2.1 10.5 4.3 3.3 3.9 Uzbekistan 35 9 368 35 97 12 4 .. .. .. .. Venezuela, RB 116 13 445 41 34 8 4 5.5 3.8 2.1 3.3 Vietnam 56 11 404 37 78 5 5.5 7.5 4.9 4.3 2.8 Yemen, Rep. 63 12 360 37 21 6 3 8 .. .. .. Zambia 35 6 274 16 70 6 2.7 6 4.1 4.5 2.8 Zimbabwe 96 10 350 33 30 4 2.2 2.5 3.6 2.6 3.5 World 50.8 9.9 388.3 31.2 81.4 6.2 3.2 8.8 4.7 3.9 3.4 Low income 65.8 10.8 416.0 34.5 99.6 6.8 3.9 6.8 4.2 3.6 2.7 Middle income 50.0 10.6 422.1 32.6 80.4 6.5 3.4 8.7 4.6 3.5 3.1 Lower middle income 50.0 11.3 424.9 33.1 66.4 7.0 3.4 7.8 4.5 3.4 3.0 Upper middle income 49.9 9.5 417.2 31.8 104.2 5.6 3.3 10.0 4.8 3.7 3.3 Low & middle income 57.5 10.7 419.2 33.5 89.3 6.6 3.6 7.9 4.4 3.6 3.0 East Asia & Paci�c 72.9 9.9 373.8 31.0 59.4 5.2 4.2 7.2 5.0 4.2 3.4 Europe & Central Asia 41.7 9.9 389.0 30.2 120.3 6.7 3.3 9.2 4.6 3.3 2.8 Latin America & Carib. 73.5 12.0 471.7 35.1 56.8 6.9 3.6 8.1 4.4 3.1 3.1 Middle East & N. Africa 39.3 10.2 412.6 37.3 48.3 6.7 3.7 8.1 4.4 4.1 3.1 South Asia 46.8 9.3 375.0 30.0 55.8 5.8 4.8 6.6 5.0 3.6 2.9 Sub-Saharan Africa 63.2 11.2 434.2 35.2 114.2 6.9 3.6 7.2 4.2 3.8 2.9 High income 27.2 7.0 280.2 23.2 49.9 4.7 2.0 11.4 5.4 4.7 4.4 The aggregates are unweighted averages. See p. 255 for country groupings. “..� indicates data is not available. 249 (c) The International Bank for Reconstruction and Development / The World Bank 250 WORLD DEVELOPMENT REPORT 2005 Technical notes Crime constraint measures the share of senior managers that ranked “crime, theft and disorder� as a major or very Table A1. Investment climate indicators: World severe constraint. Report losses from crime is the share of Bank’s Investment Climate Survey of Firms �rms reporting a loss to the establishment due to theft, vandal- ism or arson in the previous year. The average loss from crime Investment Climate Surveys have been implemented in over 53 is the loss as a share of sales for those reporting a crime. countries since 2001. A standardized questionnaire is used to Tax rate constraint measures the share of senior managers ensure comparability of responses. It was re�ned based on that ranked “tax rates� as a major or very severe constraint. Tax extensive �eld testing and reviews by academics and of�cials administration constraint measures the share of senior man- from census departments. The World Bank works with partner agers that ranked “tax administration� as a major or very severe agencies in each country to implement the survey and to con- constraint. Licensing constraint measures the share of senior duct the interviews. In most countries, national statistical managers that ranked “business licenses and permits� as a major of�ces assist with the sampling. The sampling focuses on man- or very severe constraint. Management time dealing with of�- ufacturing establishments according to their contribution to cials with regard to requirements imposed by government regu- GDP. The samples are strati�ed by size to ensure suf�cient cov- lations [e.g. taxes, customs, labor regulations, licensing and reg- erage of larger �rms. The 27 countries in Eastern Europe and istration etc.] in a given week. Average days to clear customs is Central Asia were conducted jointly with the European Bank the time to clear an imported good through customs. for Reconstruction and Development under the name of Busi- Finance constraint is the average of the shares of senior ness Environment and Enterprise Performance Surveys II managers that ranked “access to �nance� or “cost of �nance� as (BEEPS II). In �ve countries in this region the World Bank a major or very severe constraint. Small �rms with a loan is extended the samples to gather additional information on �rm the share of �rms with less than 20 employees that have a loan performance. The Asia Development Bank is a partner in a from a formal �nancial intermediary. number of countries in Asia. Electricity constraint measures the share of senior man- For each of the 8 sets of variables, the �rst column reports agers that ranked “electricity� as a major or severe constraint. on the perception of senior managers of whether the issue rep- Firms reporting outages is the share of �rms that report los- resents a problem for the operation and growth of their busi- ing sales due to power interruptions and outages during the ness. They were given a �ve-point scale, ‘no obstacle,’ ‘minor previous year. Losses from outages is the average value of sales obstacle,’ ‘moderate obstacle,’ ‘major obstacle’ and ‘very severe lost due to power interruptions and outages is expressed as a obstacle.’ This is followed by more speci�c information on the share of sales for those reporting outages. issue, including objective measures in terms of monetary and Skills constraint measures the share of senior managers time costs. that ranked “skills of available workers�as a major or severe Policy uncertainty constraint measures the share of senior constraint. Labor regulations constraint measures the share managers that ranked “economic and regulatory policy uncer- of senior managers that ranked “labor regulations� as a major tainty� as a major or very severe constraint. Unpredictable or severe constraint. interpretation of regulations reports the share of senior man- agers that disagreed with the statement that the interpretation of regulations by of�cials was predictable. Table A2. Investment climate indicators: Corruption constraint measures the share of senior man- expert polls and other surveys agers that ranked “corruption� as a major or very severe con- The World Bank’s Doing Business Project straint. Report bribes are paid is the share of senior managers The Doing Business Project collects information on the num- that report that establishments like theirs can sometimes be ber of calendar days, the number of procedures and the costs it required to make gifts or informal payments to public of�cials takes to complete various business transactions. The �rst two to “get things done� or are paid to pass inspections, get licenses are reported here. It uses a de�ned hypothetical case to stan- or permits, get a public utility connection or to get a govern- dardize comparisons and report the time if all procedures ment contract. The average bribe paid is the average size of the mandated by law are followed and are completed within the bribe as a percentage of sales for those �rms that identify that of�cially designated time for each step. bribes are paid to “get things done�. Days to start up a business refers to the number of calen- Courts constraint measures the share of senior managers dar days needed to complete all the required procedures for that ranked “courts and dispute resolution systems� as a legally operating a business. The number of procedures is also major or very severe constraint. Lack con�dence in courts reported. If a procedure can be speeded up at additional cost, to uphold property rights is the share of managers that dis- the fastest procedure, independent of cost, is chosen. Time agreed with the statement: “I am con�dent that the judicial needed to gather information about the registration proce- system will enforce my contractual and property rights in dures is not included. The hypothetical �rm is a domestic lim- business disputes.� ited liability company of 50 employees. (c) The International Bank for Reconstruction and Development / The World Bank Measuring the investment climate 251 Days to enforce a contract are the number of calendar days Global Competitiveness Report from the moment a plaintiff �les the lawsuit in court until the The World Economic Forum’s Global Competitiveness moment of �nal determination and, in appropriate cases, pay- Report ranks 102 countries using their Executive Opinion ment. The number of procedures is also reported. The stan- Survey, with samples averaging 76 respondents per country. dardized hypothetical case is one involving an unpaid check Answers are scored on a seven point scale. Transparency of worth 50 percent of per capita GDP and is assessed by local government policymaking is based on “Firms in your coun- lawyers based on of�cial times each procedure should take. try are usually informed clearly and transparently by the The time and number of procedures to register property government on changes in policies and regulations affecting looks at the requirements to of�cially register property in a your industry (1 = never informed, 7 = always fully and peri-urban area. clearly informed). Intensity of local competition is “compe- Resolving insolvency measures the number of calendar tition in the local market is (1 = limited in most industries days from the moment of �ling for insolvency in court until and price-cutting is rare, 7 = intense in most industries as the moment of actual resolution of distressed assets. The hypo- market leadership changes over time). Regional disparities thetical case is a hotel whose only asset is real estate. in quality of business environment is “differences among regions within your country in the quality of the business International Country Risk Guide environment (human resources, infrastructure and other The PRS Group’s International Country Risk Guide (ICRG) col- factors) are (1 = large and persistent, 7 = modest). lects information on various components of risk, grouping Other institutions provide additional measures of the them into a number of indices. Lower numbers indicate higher investment climate. The following table provides examples, risk on a scale of 1 to 12. Reported here is the investment pro�le focusing on measures of risk and competition. that combines assessments of contract viability/expropriation, The WDR thanks the PRS Group and the World Economic the ability to repatriate pro�ts and payment delays. Forum for making their data available. Other sources of investment climate–related indicators—selected examples Index Publisher Sample Assessment Business Risk Service Business Environment Risk Intelligence Country risk in 50 countries based on Assessments by in-house experts www.beri.com evaluation of 3 sub-categories. Updated trimestrally. Country Credit Ratings Euromoney Institutional Investor Credit ratings of 151 countries based Surveys of outside �nancial and www.euromoneyplc.com on nine areas of country risk. Updated investment analysts. semi-annually. Country Risk Indicators World Markets Research Center Country risk in 186 countries based Assessments by in-house experts www.wmrc.com on evaluation of 6 risk factors. Updated daily. Country Risk Service Economist Intelligence Unit Country risk in 100 emerging Assessments by in-house experts www.eiu.com economies and 6 regions based on evaluation of 13 risk attributes. Updated monthly. Economic Freedom of the World Fraser Institute Freedom from government regulation Assessments by in-house experts www.freetheworld.com in 123 countries covering 8 areas. and existing surveys, including GCR Updated annually. and ICRG. FDI Con�dence Index A. T. Kearney Attractiveness of 62 countries to FDI. Surveys of 1,000 multinational www.atkearney.com Updated annually. company CEOs Global Competitiveness Report World Economic Forum Competitiveness of 102 countries. Surveys of executives of local and www.weforum.org Updated annually. global companies Global Risk Service Global Insight Country risk in 117 countries based Assessments by in-house experts www.globalinsight.com on an evaluation of 51 risk attributes. Updated quarterly. Index of Economic Freedom Heritage Foundation Freedom from government regulation Assessments by in-house experts www.heritage.org in 142 countries, based on evaluation of 10 factors. Updated annually. International Country Risk Guide Political Risk Services International Country risk in 140 countries based Assessments by in-house experts www.prsgroup.com on evaluation of 22 variables in 3 sub-categories. Updated monthly. World Competitiveness Yearbook International Institute for Management Competitiveness of 51 countries, 9 Compiled from international and Development sub-national regions. Updated regional organizations and private www.imd.ch annually. institutes, executive opinion surveys Worldwide Governance Indicators World Bank Governance indicators for 199 Aggregation of existing surveys and www.worldbank.org/ countries covering six dimensions of indicators. wbi/governance/data governance. Updated biennially. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank Selected world development indicators In this year’s edition, development data are presented in four the capabilities and resources devoted to basic data collec- tables presenting comparative socioeconomic data for more tion and compilation. For some topics, competing sources of than 130 economies for the most recent year for which data data require review by World Bank staff to ensure that the are available and, for some indicators, for an earlier year. An most reliable data available are presented. In some instances, additional table presents basic indicators for 75 economies where available data are deemed too weak to provide reliable with sparse data or with populations of less than 1.5 million. measures of levels and trends or do not adequately adhere to The indicators presented here are a selection from more international standards, the data are not shown. than 800 included in World Development Indicators 2004. The data presented are generally consistent with those in Published annually, World Development Indicators reflects a World Development Indicators 2004. However, data have been comprehensive view of the development process. Its opening revised and updated wherever new information has become chapter reports on the Millennium Development Goals available. Differences may also reflect revisions to historical which grew out of agreements and resolutions of world con- series and changes in methodology. Thus data of different ferences organized by the United Nations (UN) in the past vintages may be published in different editions of World decade, and reaf�rmed at the Millennium Summit in Sep- Bank publications. Readers are advised not to compile data tember 2000 by member countries of the UN. The other �ve series from different publications or different editions of the main sections recognize the contribution of a wide range of same publication. Consistent time-series data are available factors: human capital development, environmental sustain- on World Development Indicators 2004 CD-ROM and ability, macroeconomic performance, private sector devel- through WDI Online. opment and the investment climate, and the global links that All dollar �gures are in current U.S. dollars unless other- influence the external environment for development. World wise stated. The various methods used to convert from Development Indicators is complemented by a separately national currency �gures are described in the Technical published database that gives access to over 1,000 data tables notes. and 800 time-series indicators for 225 economies and Because the World Bank’s primary business is providing regions. This database is available through an electronic sub- lending and policy advice to its low- and middle-income scription (WDI Online) or as a CD-ROM. members, the issues covered in these tables focus mainly on these economies. Where available, information on the high- Data sources and methodology income economies is also provided for comparison. Readers Socioeconomic and environmental data presented here are may wish to refer to national statistical publications and drawn from several sources: primary data collected by the publications of the Organisation for Economic Co-opera- World Bank, member country statistical publications, tion and Development (OECD) and the European Union for research institutes, and international organizations such as more information on the high-income economies. the United Nations and its specialized agencies, the Interna- tional Monetary Fund (IMF), and the OECD (see the Data Changes in the System of National Accounts Sources following the Technical notes for a complete listing). This edition of the Selected World Development Indicators, Although international standards of coverage, de�nition, as in last year’s edition, uses terminology in line with the and classi�cation apply to most statistics reported by coun- 1993 System of National Accounts (SNA). For example, in tries and international agencies, there are inevitably differ- the 1993 SNA gross national income replaces gross national ences in timeliness and reliability arising from differences in product. See the technical notes for tables 1 and 3. 253 (c) The International Bank for Reconstruction and Development / The World Bank 254 WORLD DEVELOPMENT REPORT 2005 Most countries continue to compile their national From time to time an economy’s classi�cation is revised accounts according to the 1968 SNA, but more and more are because of changes in the above cutoff values or in the econ- adopting the 1993 SNA. A few low-income countries still use omy’s measured level of GNI per capita. When such changes concepts from older SNA guidelines, including valuations occur, aggregates based on those classi�cations are recalcu- such as factor cost, in describing major economic aggregates. lated for the past period so that a consistent time series is maintained. Classi�cation of economies and summary measures Terminology and country coverage The summary measures at the bottom of each table include The term country does not imply political independence but economies classi�ed by income per capita and by region. GNI may refer to any territory for which authorities report sepa- per capita is used to determine the following income classi�- rate social or economic statistics. Data are shown for cations: low-income, $765 or less in 2003; middle-income, economies as they were constituted in 2003, and historical $766 to $9,385; and high-income, $9,386 and above. A fur- data are revised to reflect current political arrangements. ther division at GNI per capita $3,035 is made between Throughout the tables, exceptions are noted. lower-middle-income and upper-middle-income economies. See the table on classi�cation of economies at the end of this Technical notes volume for a list of economies in each group (including Because data quality and intercountry comparisons are often those with populations of less than 1.5 million). problematic, readers are encouraged to consult the Technical Summary measures are either totals (indicated by t if the notes, the table on Classi�cation of Economies by Income aggregates include estimates for missing data and nonre- and Region, and the footnotes to the tables. For more exten- porting countries, or by an s for simple sums of the data sive documentation see World Development Indicators 2004. available), weighted averages (w), or median values (m) cal- Readers may �nd more information on the WDI 2004, culated for groups of economies. Data for the countries and orders can be made online, by phone, or fax as follows: excluded from the main tables (those presented in Table 1a) have been included in the summary measures, where data For more information and to order online: are available, or by assuming that they follow the trend of http://www.worldbank.org/data/wdi2002/index.htm. reporting countries. This gives a more consistent aggregated measure by standardizing country coverage for each period To order by phone or fax: 1-800-645-7247 or 703-661-1580; shown. Where missing information accounts for a third or Fax 703-661-1501 more of the overall estimate, however, the group measure is reported as not available. The section on Statistical methods To order by mail: The World Bank, P.O. Box 960, Herndon, in the Technical notes provides further information on aggre- VA 20172-0960, U.S.A. gation methods. Weights used to construct the aggregates are listed in the technical notes for each table. (c) The International Bank for Reconstruction and Development / The World Bank Classi�cation of economies by region and income, FY2005 East Asia and the Paci�c Latin America and the Caribbean South Asia High income OECD American Samoa UMC Antigua and Barbuda UMC Afghanistan LIC Australia Cambodia LIC Argentina UMC Bangladesh LIC Austria China LMC Barbados UMC Bhutan LIC Belgium Fiji LMC Belize UMC India LIC Canada Indonesia LMC Bolivia LMC Maldives LMC Denmark Kiribati LMC Brazil LMC Nepal LIC Finland Korea, Dem. Rep. LIC Chile UMC Pakistan LIC France Lao PDR LIC Colombia LMC Sri Lanka LMC Germany Malaysia UMC Costa Rica UMC Greece Marshall Islands LMC Cuba LMC Sub-Saharan Africa Iceland Micronesia, Fed. Sts. LMC Dominica UMC Angola LIC Ireland Mongolia LIC Dominican Republic LMC Benin LIC Italy Myanmar LIC Ecuador LMC Botswana UMC Japan Northern Mariana Islands UMC El Salvador LMC Burkina Faso LIC Korea, Rep. Palau UMC Grenada UMC Burundi LIC Luxembourg Papua New Guinea LIC Guatemala LMC Cameroon LIC Netherlands Philippines LMC Guyana LMC Cape Verde LMC New Zealand Samoa LMC Haiti LIC Central African Republic LIC Norway Solomon Islands LIC Honduras LMC Chad LIC Portugal Thailand LMC Jamaica LMC Comoros LIC Spain Timor-Leste LIC Mexico UMC Congo, Dem. Rep. LIC Sweden Tonga LMC Nicaragua LIC Congo, Rep. LIC Switzerland Vanuatu LMC Panama UMC Côte d’Ivoire LIC United Kingdom Vietnam LIC Paraguay LMC Equatorial Guinea LIC United States Peru LMC Eritrea LIC Europe and Central Asia St. Kitts and Nevis UMC Ethiopia LIC Other high income Albania LMC St. Lucia UMC Gabon UMC Andorra Armenia LMC St. Vincent and the Gambia, The LIC Aruba Azerbaijan LMC Grenadines UMC Ghana LIC Bahamas, The Belarus LMC Suriname LMC Guinea LIC Bahrain Bosnia and Herzegovina LMC Trinidad and Tobago UMC Guinea-Bissau LIC Bermuda Bulgaria LMC Uruguay UMC Kenya LIC Brunei Croatia UMC Venezuela, RB UMC Lesotho LIC Cayman Islands Czech Republic UMC Liberia LIC Channel Islands Estonia UMC Middle East and North Africa Madagascar LIC Cyprus Georgia LMC Algeria LMC Malawi LIC Faeroe Islands Hungary UMC Djibouti LMC Mali LIC French Polynesia Kazakhstan LMC Egypt, Arab Rep. LMC Mauritania LIC Greenland Kyrgyz Republic LIC Iran, Islamic Rep. LMC Mauritius UMC Guam Latvia UMC Iraq LMC Mayotte UMC Hong Kong, China Lithuania UMC Jordan LMC Mozambique LIC Isle of Man Macedonia, FYR LMC Lebanon UMC Namibia LMC Israel Moldova LIC Libya UMC Niger LIC Kuwait Poland UMC Morocco LMC Nigeria LIC Liechtenstein Romania LMC Oman UMC Rwanda LIC Macao, China Russian Federation LMC Saudi Arabia UMC São Tomé and Principe LIC Malta Serbia and Montenegro LMC Syrian Arab Republic LMC Senegal LIC Monaco Slovak Republic UMC Tunisia LMC Seychelles UMC Netherlands Antilles Tajikistan LIC West Bank and Gaza LMC Sierra Leone LIC New Caledonia Turkey LMC Yemen, Rep. LIC Somalia LIC Puerto Rico Turkmenistan LMC South Africa LMC Qatar Ukraine LMC Sudan LIC San Marino Uzbekistan LIC Swaziland LMC Singapore Tanzania LIC Slovenia Togo LIC Taiwan, China Uganda LIC United Arab Emirates Zambia LIC Virgin Islands (U.S.) Zimbabwe LIC This table classi�es all World Bank member economies, and all other economies with populations of more than 30,000. Economies are divided among income groups according to 2003 GNI per capita, calculated using the World Bank Atlas method. The groups are: low income (LIC), $765 or less; lower middle income (LMC), $766–3,035; upper middle income (UMC), $3,036–9,385; and high income, $9,386 or more. Source: World Bank data. 255 (c) The International Bank for Reconstruction and Development / The World Bank Table 1. Key indicators of development Population Gross national PPP income (GNI) a gross national income (GNI) b Millions Avg. Density Billions Per capita Billions Per capita Gross Life Under-5 Adult Carbon annual people per of dollars dollars of dollars dollars domestic expectancy mortality Literacy dioxide % growth sq. km product at birth rate rate emissions per capita Per % of people Millions % growth Years 1,000 15 and above of tons 2003 1990–2003 2003 2003 2003 2003 2003 2002–2003 2002 2002 2002 2000 Albania 3.2 –0.3 116 6 1,740 15 4,700 6.9 74 24 99 c 2.9 Algeria 31.8 1.9 13 60 1,890 189 d 5,940 d 5.2 71 49 69 89.4 Angola 13.5 2.8 11 10 740 26 d 1,890 d 1.4 47 260 .. 6.4 Argentina 38.4 1.3 14 140 3,650 419 10,920 3.3 74 19 97 138.2 Armenia 3.1 –1.1 108 3 950 12 3,770 11.9 75 35 99 c 3.5 Australia 19.9 1.2 3 431 21,650 563 28,290 1.2 79 6 .. 344.8 Austria 8.1 0.3 97 215 26,720 239 29,610 0.6 79 5 .. 60.8 Azerbaijan 8.2 1.1 95 7 810 28 3,380 10.5 65 96 .. 29.0 Bangladesh 138.1 1.7 1,061 55 400 258 1,870 3.5 62 73 41 29.3 Belarus 9.9 –0.2 48 16 1,590 59 6,010 6.1 68 20 100 59.2 Belgium 10.3 0.3 342 267 25,820 299 28,930 1.0 79 6 .. 102.2 Benin 6.7 2.7 61 3 440 7 1,110 2.9 53 151 40 1.6 Bolivia 9.0 2.4 8 8 890 22 2,450 –0.8 64 71 87 c 11.1 Bosnia & Herzegovina 4.1 –0.6 82 6 1,540 26 6,320 3.0 74 18 95 19.3 Botswana 1.7 2.3 3 6 3,430 14 7,960 4.0 38 110 79 3.9 Brazil 176.6 1.4 21 479 2,710 1,322 7,480 –1.4 69 37 86 c 307.5 Bulgaria 7.8 –0.8 71 17 2,130 60 7,610 4.9 72 16 99 42.3 Burkina Faso 12.1 2.4 44 4 300 14 d 1,180 d 4.1 43 207 .. 1.0 Burundi 7.2 2.1 281 1 100 4d 620 d –2.9 42 208 50 0.2 Cambodia 13.4 2.9 76 4 310 28 d 2,060 d 5.8 54 138 69 0.5 Cameroon 16.1 2.5 35 10 640 32 1,980 0.5 48 166 68 e 6.5 Canada 31.6 1.0 3 757 23,930 941 29,740 0.9 79 7 .. 435.9 Central African Rep. 3.9 2.1 6 1 260 4d 1,080 d –8.8 42 180 49 e 0.3 Chad 8.6 3.0 7 2 250 9 1,100 4.3 48 200 46 0.1 Chile 15.8 1.4 21 69 4,390 155 9,810 2.0 76 12 96 c 59.5 China 1,288.4 1.0 138 1,417 1,100 6,435 f 4,990 f 8.4 71 38 91 c 2,790.5 Hong Kong, China 6.8 1.4 .. 173 25,430 196 28,810 2.9 80 .. .. 33.1 Colombia 44.4 1.8 43 80 1,810 290 d 6,520 d 2.0 72 23 92 58.5 Congo, Dem. Rep. 53.2 2.7 23 5 100 34 d 640 d 1.9 45 205 .. 2.7 Congo, Rep. 3.8 3.2 11 2 640 3 710 –1.7 52 108 83 1.8 Costa Rica 4.0 2.1 78 17 4,280 36 d 9,040 d 3.9 78 11 96 5.4 Côte d’Ivoire 16.8 2.7 53 11 660 23 1,390 –5.6 45 191 .. 10.5 Croatia 4.5 –0.5 80 2 5,350 48 10,710 4.0 74 8 98 c 19.6 Czech Rep. 10.2 –0.1 132 69 6,740 160 15,650 2.9 75 5 .. 118.8 Denmark 5.4 0.4 127 182 33,750 168 31,213 0.2 77 4 .. 44.6 Dominican Rep. 8.7 1.6 181 18 2,070 54 d 6,210 d –2.2 67 38 84 25.1 Ecuador 13.0 1.8 47 23 1,790 45 3,440 0.9 70 29 91 c 25.5 Egypt, Arab Rep. 67.6 1.9 68 94 1,390 266 3,940 1.4 69 39 .. 142.2 El Salvador 6.5 1.9 315 14 2,200 32 d 4,890 d 1.8 70 39 80 6.7 Eritrea 4.4 2.6 43 1 190 5d 1,110 d 2.8 51 80 .. 0.6 Estonia 1.4 –1.2 32 7 4,960 17 12,480 5.3 71 12 100 c 16.0 Ethiopia 68.6 2.3 69 6 90 49 d 710 d –5.7 42 171 42 5.6 Finland 5.2 0.3 17 141 27,020 141 27,100 1.7 78 5 .. 53.4 France 59.7 0.4 109 1,523 g 24,770 g 1,640 27,460 –0.3 79 6 .. 362.4 Georgia 5.1 –0.5 74 4 830 13 d 2,540 d 9.4 73 29 .. 6.2 Germany 82.6 0.3 237 2,085 25,250 2,267 27,460 –0.1 78 5 .. 785.5 Ghana 20.4 2.2 90 7 320 45 d 2,190 d 2.5 55 97 74 5.9 Greece 10.7 0.4 83 147 13,720 213 19,920 4.2 78 5 97 89.6 Guatemala 12.3 2.6 114 23 1,910 50 d 4,060 d –0.5 65 49 70 9.9 Guinea 7.9 2.4 32 3 430 17 2,100 0.0 46 165 .. 1.3 Haiti 8.4 2.0 306 3 380 14 d 1,630 d –1.8 52 123 52 1.4 Honduras 7.0 2.8 62 7 970 18 d 2,580 d –0.5 66 42 80 c 4.8 Hungary 10.1 –0.2 110 64 6,330 139 13,780 0.7 72 9 99 54.2 India 1,064.4 1.7 358 568 530 3,068 d 2,880 d 6.4 63 90 61 c 1,070.9 Indonesia 214.5 1.4 118 173 810 689 3,210 2.8 67 43 88 269.6 Iran, Islamic Rep. 66.4 1.5 41 133 2,000 477 7,190 4.4 69 41 77 e 310.3 Ireland 3.9 0.9 57 106 26,960 120 30,450 1.1 77 6 .. 42.2 Israel 6.7 2.8 324 105 16,020 128 19,200 –0.8 79 6 95 63.1 Italy 57.6 0.1 196 1,243 21,560 1,543 26,760 0.4 78 6 99 428.2 Jamaica 2.6 0.8 244 7 2,760 10 3,790 1.1 76 20 88 10.8 Japan 127.2 0.2 349 4,390 34,510 3,641 28,620 2.7 82 5 .. 1,184.5 Jordan 5.3 4.0 60 10 1,850 23 4,290 0.5 72 33 91 15.6 Kazakhstan 14.9 –0.7 6 27 1,780 92 6,170 8.7 62 99 99 121.3 Kenya 31.9 2.4 56 13 390 33 1,020 –0.7 46 122 84 9.4 Korea, Rep. 47.9 0.9 485 576 12,020 859 17,930 2.4 74 5 .. 427.0 Kuwait 2.4 0.9 134 38 16,340 42 d 17,870 d –3.3 77 10 83 47.9 Kyrgyz Rep. 5.1 1.0 26 2 330 8 1,660 3.9 65 61 .. 4.6 Lao PDR 5.7 2.4 25 2 320 10 1,730 2.6 55 100 66 0.4 Latvia 2.3 –1.1 37 9 4,070 24 10,130 8.1 70 21 100 c 6.0 Lebanon 4.5 1.6 440 18 4,040 22 4,840 1.4 71 32 .. 15.2 Lesotho 1.8 1.0 59 1 590 6d 3,120 d 20.9 38 132 81 e .. Lithuania 3.5 –0.5 53 16 4,490 38 11,090 7.0 73 9 100 c 11.9 Macedonia, FYR 2.0 0.6 81 4 1,980 14 6,720 2.5 73 26 .. 11.2 Madagascar 16.9 2.9 29 5 290 13 800 6.5 55 135 .. 2.3 Malawi 11.0 2.0 117 2 170 7 600 3.8 38 182 62 0.8 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those speci�ed. 256 (c) The International Bank for Reconstruction and Development / The World Bank Table 1. Key indicators of development—continued Population Gross national PPP income (GNI) a gross national income (GNI) b Millions Avg. density Billions per capita Billions per capita Gross Life Under-5 Adult Carbon annual people per of dollars dollars of dollars dollars domestic expectancy mortality Literacy dioxide % growth sq. km product at birth rate rate emissions per capita Per % of people Millions % growth Years 1,000 15 and above of tons 2003 1990–2003 2003 2003 2003 2003 2003 2002–2003 2002 2002 2002 2000 Malaysia 24.8 2.4 75 94 3,780 222 8,940 3.2 73 8 89 c 144.4 Mali 11.7 2.5 10 3 290 11 960 3.5 41 222 19 c 0.6 Mauritania 2.7 2.2 3 1 430 5d 2,010 d 2.9 51 183 41 3.1 Mexico 102.3 1.6 54 637 6,230 915 8,950 –0.1 74 29 91 c 424.0 Moldova 4.2 –0.2 129 2 590 7 1,750 6.5 67 32 99 6.6 Mongolia 2.5 1.3 2 1 480 4 1,800 3.4 65 71 98 c 7.5 Morocco 30.1 1.7 67 40 1,320 119 d 3,950 d 3.8 68 43 51 36.5 Mozambique 18.8 2.2 24 4 210 20 d 1,070 d 5.0 41 205 46 1.2 Myanmar 49.4 1.5 75 .. .. h .. .. .. 57 108 85 9.1 Namibia 2.0 2.8 2 4 1,870 13 6,620 –6.7 42 67 83 1.8 Nepal 24.7 2.4 172 6 240 35 1,420 0.7 60 83 44 3.4 Netherlands 16.2 0.6 479 427 26,310 464 28,600 –0.9 78 5 .. 138.9 New Zealand 4.0 1.2 15 64 15,870 85 21,120 0.9 78 6 .. 32.1 Nicaragua 5.5 2.8 45 4 730 13 d 2,400 d –0.2 69 41 77 e 3.7 Niger 11.8 3.3 9 2 200 10 d 820 d 1.0 46 264 17 1.2 Nigeria 135.6 2.6 149 43 320 122 900 8.3 45 201 67 36.1 Norway 4.6 0.6 15 198 43,350 170 37,300 –0.2 79 4 .. 49.9 Pakistan 148.4 2.4 193 69 470 306 2,060 3.3 64 101 .. 104.8 Panama 3.0 1.7 40 13 4,250 19 d 6,310 d 2.3 75 25 92 6.3 Papua New Guinea 5.5 2.5 12 3 510 12 d 2,240 d 0.2 57 94 .. 2.4 Paraguay 5.6 2.4 14 6 1,100 27 d 4,740 d –0.3 71 30 92 e 3.7 Peru 27.1 1.8 21 58 2,150 138 5,090 2.4 70 39 85 e 29.5 Philippines 81.5 2.2 273 88 1,080 379 4,640 2.5 70 37 93 c 77.5 Poland 38.2 0.0 125 201 5,270 437 11,450 4.9 74 9 .. 301.3 Portugal 10.2 0.2 111 124 12,130 183 17,980 –0.9 76 6 93 59.8 Romania 22.2 –0.3 96 51 2,310 159 7,140 5.6 70 21 97 c 86.3 Russian Federation 143.4 –0.3 8 375 2,610 1,279 8,920 7.8 66 21 100 1,435.1 Rwanda 8.3 1.3 334 2 220 11 d 1,290 d 2.1 40 203 69 0.6 Saudi Arabia 22.5 2.7 10 187 8,530 281 d 12,850 d –1.8 73 28 78 374.3 Senegal 10.0 2.4 52 6 550 17 d 1,660 d 6.0 52 138 39 4.2 Serbia & Montenegro 8.1 .. 79 16 i 1,910 i .. .. 5.5 73 19 .. 39.5 Sierra Leone 5.3 2.2 75 1 150 3 530 4.5 37 284 .. 0.6 Singapore 4.3 2.6 6,967 90 21,230 103 24,180 –1.0 78 4 93 c 59.0 Slovak Republic 5.4 0.1 110 26 4,920 72 13,420 4.8 73 9 100 c 35.4 Slovenia 2.0 –0.1 98 23 11,830 38 19,240 3.5 76 5 100 14.6 South Africa 45.3 1.9 37 126 2,780 465 d 10,270 d –2.0 46 65 86 327.3 Spain 41.1 0.4 82 698 16,990 905 22,020 1.9 78 6 98 282.9 Sri Lanka 19.2 1.3 297 18 930 72 3,730 4.3 74 19 92 10.2 Sweden 9.0 0.3 22 258 28,840 238 26,620 1.2 80 3 .. 46.9 Switzerland 7.3 0.7 186 293 39,880 235 32,030 –1.2 80 6 .. 39.1 Syrian Arab Rep. 17.4 2.8 95 20 1,160 60 3,430 0.0 70 28 83 54.2 Tajikistan 6.3 1.3 45 1 190 7 1,040 7.8 67 116 99 c 4.0 Tanzania 35.9 2.6 41 10 j 290 j 22 610 3.5 43 165 77 4.3 Thailand 62.0 0.8 121 136 2,190 462 7,450 6.1 69 28 93 c 198.6 Togo 4.9 2.6 89 1 310 7d 1,500 d 0.9 50 140 60 1.8 Tunisia 9.9 1.5 64 22 2,240 68 6,840 4.4 73 26 73 18.4 Turkey 70.7 1.8 92 197 2,790 473 6,690 4.2 70 41 87 c 221.6 Turkmenistan 4.9 2.2 10 5 1,120 28 5,840 15.3 65 86 .. 34.6 Uganda 25.3 2.9 128 6 240 36 d 1,440 d 0.8 43 141 69 1.5 Ukraine 48.4 –0.5 83 47 970 262 5,410 10.2 68 20 100 342.8 United Kingdom 59.3 0.2 246 1,680 28,350 1,639 27,650 2.1 77 7 .. 567.8 United States 291.0 1.2 32 10,946 37,610 10,914 37,500 2.0 77 8 .. 5,601.5 Uruguay 3.4 0.7 19 13 3,790 27 7,980 1.9 75 15 98 5.4 Uzbekistan 25.6 1.7 62 11 420 44 1,720 3.0 67 65 99 118.6 Venezuela, RB 25.5 2.1 29 89 3,490 121 4,740 –10.9 74 22 93 157.7 Vietnam 81.3 1.6 250 39 480 202 2,490 6.1 70 26 .. 57.5 Yemen, Rep. 19.2 3.7 36 10 520 16 820 0.7 57 114 49 8.4 Zambia 10.4 2.2 14 4 380 9 850 3.5 37 182 80 1.8 Zimbabwe 13.1 1.9 34 6 480 28 2,180 –6.7 39 123 90 14.8 World 6,271.7 s 1.4 w 48 w 34,491 t 5,500 w 51,314 t 8,180 t 1.4 w 67 w 81 w 79 w 22,994.5 t Low income 2,310.3 2.0 76 1,038 450 5,052 2,190 4.9 58 126 61 2,066.7 Middle income 2,990.1 1.1 43 5,732 1,920 17,933 6,000 3.9 70 38 90 9,129.1 Lower middle income 2,655.2 1.1 47 3,934 1,480 14,617 5,510 4.5 69 40 90 7,116.3 Upper middle income 334.9 1.3 26 1,788 5,340 3,317 9,900 1.7 73 22 91 2,012.0 Low & middle income 5,300.3 1.5 53 6,762 1,280 22,894 4,320 3.8 65 88 78 11,196.2 East Asia & Paci�c 1,854.5 1.2 117 2,011 1,080 8,675 4,680 6.8 69 42 90 3,752.3 Europe & Cen. Asia 472.7 0.1 20 1,217 2,570 3,579 7,570 6.0 69 37 97 3,162.6 Latin Am. & Carib. 534.2 1.6 27 1,741 3,260 3,780 7,080 –0.1 71 34 89 1,357.4 Mid. East & N. Africa 311.6 2.1 28 689 2,250 1,743 5,700 1.2 69 54 69 1,227.2 South Asia 1,424.7 1.8 298 726 510 3,795 2,660 5.7 63 95 59 1,220.3 Sub-Saharan Africa 702.6 2.5 30 347 490 1,243 1,770 1.3 46 174 65 478.8 High income 971.4 0.7 31 27,732 28,550 28,603 29,450 1.4 78 7 .. 11,804.3 a. Preliminary World Bank estimates calculated using the World Bank Atlas method. b. Purchasing power parity; see the Technical Notes. c. National estimates based on census data. d.The estimate is based on regression; others are extrapolated from the latest International Comparison Programme benchmark estimates. e. National estimates based on survey data. f. Estimates based on bilateral comparison between China and the United States (Ruoen and Kai, 1995). g.GNI and GNI per capita estimates include the French Overseas departments of French Guiana, Guadeloupe, Martinique, and Réunion. h. Estimated to be low income ($765 or less). i. Data for Kosovo is excluded. j. Data refer to mainland Tanzania only. 257 (c) The International Bank for Reconstruction and Development / The World Bank Table 2. Poverty and income distribution National poverty lines International poverty line Population below the poverty line (%) Percentage share of Population Poverty Population Poverty income or below gap at below gap at consumption Survey Survey $1 a day $1 a day $2 a day $2 a day Survey Gini Lowest Highest Economy year Rural Urban National year % % % % year index 20% 20% Albania 2002 29.6 .. 25.4 2002 a <2.0 <0.5 11.8 2.0 2002 c,d 28.2 9.1 37.4 Algeria 1998 16.6 7.3 12.2 1995 a <2.0 <0.5 15.1 3.8 1995 c,d 35.3 7.0 42.6 Angola .. .. .. .. .. .. .. .. .. .. Argentina 1998 .. 29.9 .. 2001 b 3.3 0.5 14.3 4.7 2001 e,f 52.2 3.1 56.4 Armenia 1998–99 44.8 60.4 53.7 1998 a 12.8 3.3 49.0 17.3 1998 c,d 37.9 6.7 45.1 Australia .. .. .. .. .. .. .. 1994 e,f 35.2 5.9 41.3 Austria .. .. .. .. .. .. .. 1997 e,f 30.0 8.1 38.5 Azerbaijan 2001 .. .. 49.6 2001 a 3.7 <1.0 9.1 3.5 2001 c,d 36.5 7.4 44.5 Bangladesh 2000 53.0 36.6 49.8 2000 a 36.0 8.1 82.8 36.3 2000 c,d 31.8 9.0 41.3 Belarus 2000 .. .. 41.9 2000 a <2.0 <0.5 <2.0 0.1 2000 c,d 30.4 8.4 39.1 Belgium .. .. .. .. .. .. .. 1996 e,f 25.0 8.3 37.3 Benin 1995 .. .. 33.0 .. .. .. .. .. .. .. Bolivia 1999 81.7 .. 62.7 1999 a 14.4 5.4 34.3 14.9 1999 c,d 44.7 4.0 49.1 Bosnia & Herzegovina 2001–02 19.9 13.8 19.5 .. .. .. .. 2001 c,d 26.2 9.5 35.8 a Botswana .. .. .. 1993 23.5 7.7 50.1 22.8 1993 c,d 63.0 2.2 70.3 Brazil 1990 32.6 13.1 17.4 2001 b 8.2 2.1 22.4 8.8 1998 e,f 59.1 2.0 64.4 Bulgaria 2001 .. .. 12.8 2001 a 4.7 1.4 16.2 5.7 2001 e,f 31.9 6.7 38.9 Burkina Faso 1998 51.0 16.5 45.3 1998 a 44.9 14.4 81.0 40.6 1998 c,d 48.2 4.5 60.7 Burundi 1990 36.0 43.0 .. 1998 a 58.4 24.9 89.2 51.3 1998 c,d 33.3 5.1 48.0 Cambodia 1997 40.1 21.1 36.1 1997 a 34.1 9.7 77.7 34.5 1997 c,d 40.4 6.9 47.6 Cameroon 2001 49.9 22.1 40.2 2001 a 17.1 4.1 50.6 19.3 2001 c,d 44.6 5.6 50.9 Canada .. .. .. .. .. .. .. 1998 e,f 33.1 7.0 40.4 Central African Rep. .. .. .. 1993 a 66.6 38.1 84.0 58.4 1993 c,d 61.3 2.0 65.0 Chad 1995–96 67.0 63.0 64.0 .. .. .. .. .. .. .. Chile 1998 .. .. 17.0 2000 b <2.0 <0.5 9.6 2.5 2000 e,f 57.1 3.3 62.2 China 1998 4.6 <2.0 4.6 2001 a 16.6 3.9 46.7 18.4 2001 c,d 44.7 4.7 50.0 Hong Kong, China .. .. .. .. .. .. .. 1996 e,f 43.4 5.3 50.7 Colombia 1999 79.0 55.0 64.0 1999 b 8.2 2.2 22.6 8.8 1999 e,f 57.6 2.7 61.8 Congo, Dem. Rep. .. .. .. .. .. .. .. .. .. .. Congo, Rep. .. .. .. .. .. .. .. .. .. .. Costa Rica 1992 25.5 19.2 22.0 2000 b 2.0 0.7 9.5 3.0 2000 e,f 46.5 4.2 51.5 Côte d’Ivoire .. .. .. 1998 a 15.5 3.8 50.4 18.9 1998 c,d 45.2 5.5 51.1 Croatia .. .. .. 2000 a <2.0 <0.5 <2.0 <0.5 2001 c,d 29.0 8.3 39.6 Czech Rep. .. .. .. 1996 b <2.0 <0.5 <2.0 <0.5 1996 e,f 25.4 10.3 35.9 Denmark .. .. .. .. .. .. .. 1997 e,f 24.7 8.3 35.8 Dominican Rep. 1998 42.1 20.5 28.6 1998 b <2.0 <0.5 <2.0 <0.5 1998 e,f 47.4 5.1 53.3 Ecuador 1994 47.0 25.0 35.0 1998 b 17.7 7.1 40.8 17.7 1998 c,d 43.7 3.3 58.0 Egypt, Arab Rep. 1999–00 23.3 22.5 16.7 2000 a 3.1 <0.5 43.9 11.3 1999 c,d 34.4 8.6 43.6 El Salvador 1992 55.7 43.1 48.3 2000 b 31.1 14.1 58.0 29.7 2000 e,f 53.2 2.9 57.1 Eritrea 1993–94 .. .. 53.0 .. .. .. .. .. .. .. Estonia 1995 14.7 6.8 8.9 1998 a <2.0 <0.5 5.2 0.8 2000 e,f 37.2 6.1 44.0 Ethiopia 1999–00 45.0 37.0 44.2 1999–00 a 26.3 5.7 80.7 31.8 2000 c,d 30.0 9.1 39.4 Finland .. .. .. .. .. .. .. 2000 e,f 26.9 9.6 36.7 France .. .. .. .. .. .. .. 1995 e,f 32.7 7.2 40.2 Georgia 1997 9.9 12.1 11.1 2001 a 2.7 0.9 15.7 4.6 2001 c,d 36.9 6.4 43.6 Germany .. .. .. .. .. .. .. 2000 e,f 28.3 8.5 36.9 Ghana 1998 49.9 18.6 39.5 1999 a 44.8 17.3 78.5 40.8 1999 c,d 30.0 5.6 46.6 Greece .. .. .. .. .. .. .. 1998 e,f 35.4 7.1 43.6 Guatemala 2000 74.5 27.1 56.2 2000 b 16.0 4.6 37.4 16.0 2000 e,f 48.3 2.6 64.1 Guinea 1994 .. .. 40.0 .. .. .. .. 1994 c,d 40.3 6.4 47.2 Haiti 1995 66.0 .. 65.0 .. .. .. .. .. .. .. Honduras 1993 51.0 57.0 53.0 1998 b 23.8 11.6 44.4 23.1 1999 e,f 55.0 2.7 58.9 Hungary 1997 .. .. 17.3 1998 b <2.0 <0.5 7.3 1.7 1999 c,d 24.4 7.7 37.5 India 1999–00 30.2 24.7 28.6 1999–00 a 34.7 8.2 79.9 35.3 1999–00 c,d 32.5 8.9 41.6 Indonesia 1999 .. .. 27.1 2002 a 7.5 0.9 52.4 15.7 2002 c,d 34.3 8.4 43.3 Iran, Islamic Rep. .. .. .. 1998 a <2.0 <0.5 7.3 1.5 1998 c,d 43.0 5.1 49.9 Ireland .. .. .. .. .. .. .. 1996 e,f 35.9 7.1 43.3 Israel .. .. .. .. .. .. .. 1997 e,f 35.5 6.9 44.3 Italy .. .. .. .. .. .. .. 2000 e,f 36.0 6.5 42.0 Jamaica 2000 25.1 .. 18.7 2000 a <2.0 <0.5 13.3 2.7 2000 c,d 37.9 6.7 46.0 Japan .. .. .. .. .. .. .. 1993 e,f 24.9 10.6 35.7 Jordan 1997 .. .. 11.7 1997 a <2.0 <0.5 7.4 1.4 1997 c,d 36.4 7.6 44.4 Kazakhstan 1996 39.0 30.0 34.6 2001 a <2.0 <0.5 8.5 1.4 2001 c,d 31.3 8.2 39.6 Kenya 1997 53.0 49.0 52.0 1997 a 23.0 6.0 58.6 24.1 1997 c,d 44.5 5.6 51.2 Korea, Rep. .. .. .. 1998 b <2.0 <0.5 <2.0 <0.5 1998 e,f 31.6 7.9 37.5 Kuwait .. .. .. .. .. .. .. .. .. .. Kyrgyz Rep. 1999 69.7 49.0 64.1 2001 a <2.0 <0.5 27.2 5.9 2001 c,d 29.0 9.1 38.3 Lao PDR 1997–98 41.0 26.9 38.6 1997–98 a 26.3 6.3 73.2 29.6 1997 c,d 37.0 7.6 45.0 Latvia .. .. .. 1998 a <2.0 <0.5 8.3 2.0 1998 e,f 32.4 7.6 40.3 Lebanon .. .. .. .. .. .. .. .. .. .. Lesotho .. .. .. 1995 a 36.4 19.0 56.1 33.1 1995 c,d 63.2 1.5 66.5 Lithuania .. .. .. 2000 a <2.0 <0.5 13.7 4.2 2000 c,d 31.9 7.9 40.0 Macedonia, FYR .. .. .. 1998 a <2.0 <0.5 4.0 0.6 1998 c,d 28.2 8.4 36.7 Madagascar 1999 76.7 52.1 71.3 1999 a 49.1 18.3 83.3 44.0 2001 c,d 47.5 4.9 53.5 Malawi 1997–98 66.5 54.9 65.3 1997–98 a 41.7 14.8 76.1 38.3 1997 c,d 50.3 4.9 56.1 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those speci�ed. 258 (c) The International Bank for Reconstruction and Development / The World Bank Table 2. Poverty and income distribution—continued National poverty lines International poverty line Population below the poverty line (%) Percentage share of Population Poverty Population Poverty income or below gap at below gap at consumption Survey Survey $1 a day $1 a day $2 a day $2 a day Survey Gini Lowest Highest Economy year Rural Urban National year % % % % year index 20% 20% Malaysia 1989 .. .. 15.5 1997 b <2.0 <0.5 9.3 2.0 1997 e,f 49.2 4.4 54.3 Mali 1998 75.9 30.1 63.8 1994 a 72.8 37.4 90.6 60.5 1994 c,d 50.5 4.6 56.2 Mauritania 2000 61.2 25.4 46.3 2000 a 25.9 7.6 63.1 26.8 2000 c,d 39.0 6.2 45.7 Mexico 1988 .. .. 10.1 2000 b 9.9 3.7 26.3 10.9 2000 e,f 54.6 3.1 59.1 Moldova 1997 26.7 .. 23.3 2001 a 22.0 5.8 63.7 25.1 2001 c,d 36.2 7.1 43.7 Mongolia 1995 33.1 38.5 36.3 1995 a 13.9 3.1 50.0 17.5 1998 c,d 44.0 5.6 51.2 Morocco 1998–99 27.2 12.0 19.0 1999 a <2.0 <0.5 14.3 3.1 1998–99 c,d 39.5 6.5 46.6 Mozambique 1996–97 71.3 62.0 69.4 1996 a 37.9 12.0 78.4 36.8 1996–97 c,d 39.6 6.5 46.5 Myanmar .. .. .. .. .. .. .. .. .. .. Namibia .. .. .. 1993 b 34.9 14.0 55.8 30.4 1993 e,f 70.7 1.4 78.7 Nepal 1995–96 44.0 23.0 42.0 1995 a 37.7 9.7 82.5 37.5 1995–96 c,d 36.7 7.6 44.8 Netherlands .. .. .. .. .. .. .. 1994 e,f 32.6 7.3 40.1 New Zealand .. .. .. .. .. .. .. 1997 e,f 36.2 6.4 43.8 Nicaragua 1998 68.5 30.5 47.9 2001 a 45.1 16.7 79.9 41.2 2001 e,f 55.1 3.6 59.7 Niger 1989–93 66.0 52.0 63.0 1995 a 61.4 33.9 85.3 54.8 1995 c,d 50.5 2.6 53.3 Nigeria 1992–93 36.4 30.4 34.1 1997 a 70.2 34.9 90.8 59.0 1996–97 c,d 50.6 4.4 55.7 Norway .. ... .. .. .. .. .. 2000 e,f 25.8 9.6 37.2 Pakistan 1998–99 35.9 24.2 32.6 1998 a 13.4 2.4 65.6 22.0 1998–99 c,d 33.0 8.8 42.3 Panama 1997 64.9 15.3 37.3 2000 b 7.2 2.3 17.6 7.4 2000 e,f 56.4 2.4 60.3 Papua New Guinea 1996 41.3 16.1 37.5 .. .. .. .. 1996 c,d 50.9 4.5 56.5 Paraguay 1991 28.5 19.7 21.8 1999 b 14.9 6.8 30.3 14.7 1999 e,f 56.8 2.2 60.2 Peru 1997 64.7 40.4 49.0 2000 b 18.1 9.1 37.7 18.5 2000 e,f 49.8 2.9 53.2 Philippines 1997 50.7 21.5 36.8 2000 a 14.6 2.7 46.4 17.2 2000 c,d 46.1 5.4 52.3 Poland 1993 .. .. 23.8 1999 b <2.0 <0.5 <2.0 <0.5 1999 c,d 31.6 7.3 42.5 Portugal .. .. .. 1994 b <2.0 <0.5 <0.5 <0.5 1997 e,f 38.5 5.8 45.9 Romania 1994 27.9 20.4 21.5 2000 a 2.1 0.6 20.5 5.2 2000 c,d 30.3 8.2 38.4 Russian Federation 1994 .. .. 30.9 2000 a 6.1 1.2 23.8 8.0 2000 c,d 45.6 4.9 51.3 Rwanda 1993 .. .. 51.2 1983–85 a 35.7 7.7 84.6 36.7 1983–85 c,d 28.9 9.7 39.1 Saudi Arabia .. .. .. .. .. .. .. .. .. .. Senegal 1992 40.4 .. 33.4 1995 a 26.3 7.0 67.8 28.2 1995 c,d 41.3 6.4 48.2 Serbia & Montenegro .. .. .. .. .. .. .. .. .. .. Sierra Leone 1989 76.0 53.0 68.0 1989 a 57.0 39.5 74.5 51.8 1989 c,d 62.9 1.1 63.4 Singapore .. .. .. .. .. .. .. 1998 e,f 42.5 5.0 49.0 Slovak Republic .. .. .. 1996 b <2.0 <0.5 2.4 0.7 1996 e,f 25.8 8.8 34.8 Slovenia .. .. .. 1998 a <2.0 <0.5 <2.0 <0.5 1998–99 e,f 28.4 9.1 35.7 South Africa .. .. .. 1995 a 7.1 1.1 23.8 8.6 1995 c,d 59.3 2.0 66.5 Spain .. .. .. .. .. .. .. 1990 e,f 32.5 7.5 40.3 Sri Lanka 1995–96 27.0 15.0 25.0 1995–96 a 6.6 1.0 45.4 13.5 1995 c,d 34.4 8.0 42.8 Sweden .. .. .. .. .. .. .. 2000 e,f 25.0 9.1 36.6 Switzerland .. .. .. .. .. .. .. 1992 e,f 33.1 6.9 40.3 Syrian Arab Rep. .. .. .. .. .. .. .. .. .. .. Tajikistan .. .. .. 1998 a 10.3 2.6 50.8 16.3 1998 c,d 34.7 8.0 40.0 Tanzania 2000–01 38.7 .. 35.7 1993 a 19.9 4.8 59.7 23.0 1993 c,d 38.2 6.8 45.5 Thailand 1992 15.5 10.2 13.1 2000 a <2.0 <0.5 32.5 9.0 2000 c,d 43.2 6.1 50.0 Togo 1987–89 .. .. 32.3 .. .. .. .. .. .. .. Tunisia 1995 13.9 3.6 7.6 2000 a <2.0 <0.5 6.6 1.3 2000 c,d 39.8 6.0 47.3 Turkey .. .. .. 2000 a <2.0 <0.5 10.3 2.5 2000 c,d 40.0 6.1 46.7 Turkmenistan .. .. .. 1998 a 12.1 2.6 44.0 15.4 1998 c,d 40.8 6.1 47.5 Uganda 1997 .. .. 44.0 .. .. .. .. 1999 c,d 43.0 5.9 49.7 Ukraine 1995 .. .. 31.7 1999 b 2.9 0.6 45.7 16.3 1999 c,d 29.0 8.8 37.8 United Kingdom .. .. .. .. .. .. .. 1999 e,f 36.0 6.1 44.0 United States .. .. .. .. .. .. .. 2000 e,f 40.8 5.4 45.8 Uruguay .. .. .. 2000 b <2.0 <0.5 3.9 0.8 2000 e,f 44.6 4.8 50.1 Uzbekistan 2000 30.5 22.5 27.5 2000 a 21.8 5.4 77.5 28.9 2000 c,d 26.8 9.2 36.3 Venezuela, RB 1989 .. .. 31.3 1998 b 15.0 6.9 32.0 15.2 1998 e,f 49.1 3.0 53.4 Vietnam 1993 57.2 25.9 50.9 1998 a 17.7 3.3 63.7 22.9 1998 c,d 36.1 8.0 44.5 Yemen, Rep. 1998 45.0 30.8 41.8 1998 a 15.7 4.5 45.2 15.0 1998 c,d 33.4 7.4 41.2 Zambia 1998 83.1 56.0 72.9 1998 a 63.7 32.7 87.4 55.4 1998 c,d 52.6 3.3 56.6 Zimbabwe 1995–96 48.0 7.9 34.9 1990–91 a 36.0 9.6 64.2 29.4 1995 c,d 56.8 4.6 55.7 a. Based on expenditure. b. Based on income. c. Refers to expenditure shares by percentiles of population. d. Ranked by per capita expenditure. e. Refers to income shares by percentiles of population. f. Ranked by per capita income. 259 (c) The International Bank for Reconstruction and Development / The World Bank Table 3. Economic activity Value added as % of GDP Agricultural productivity Agr. Value added per Gross domestic agricultural worker product 1995 dollars Agricultural Industry Services External GDP Household General gov’t. Gross balance of implicit Avg. �nal cons. �nal cons. capital goods and deflator Millions annual expenditure expenditure formation services Avg. annual of dollars % growth % of GDP % of GDP % of GDP % of GDP % growth 2003 1990–2003 1988–90 2000–2002 2003 2003 2003 2003 2003 2003 2003 1990–2003 Albania 6,124 4.6 1,137 1,868 25 19 56 93 8 23 –24 26.9 Algeria 65,993 2.4 1,781 1,919 11 65 24 45 8 32 14 14.7 Angola 13,189 3.2 218 137 9 65 27 63 .. a 32 5 518.4 Argentina 129,735 2.3 7,282 10,317 11 35 54 63 11 15 11 4.9 Armenia 2,797 1.5 .. 2,827 24 38 38 85 10 20 –15 119.9 Australia 518,382 3.8 24,500 36,327 4 26 71 60 18 24 –3 1.9 Austria 251,456 2.1 15,593 33,828 2 32 66 58 19 22 1 1.7 Azerbaijan 7,124 2.4 .. 1,029 16 54 29 60 10 52 –23 65.6 Bangladesh 51,897 4.9 244 318 22 27 52 77 5 23 –6 3.8 Belarus 17,493 0.6 .. 3,038 10 37 53 60 21 22 –3 252.3 Belgium 302,217 2.1 30,479 57,462 1 27 72 55 21 19 4 1.8 Benin 3,499 5.0 397 621 36 14 50 80 13 19 –12 7.0 Bolivia 8,024 3.5 681 754 15 33 52 77 15 11 –3 7.1 Bosnia & Herzegovina 6,963 17.8 .. 7,634 17 35 49 88 25 19 –32 3.5 Botswana 7,388 4.7 777 575 2 48 50 28 32 25 14 9.0 Brazil 492,338 2.6 2,982 4,899 6 21 73 58 20 20 2 118.9 Bulgaria 19,859 –0.2 3,409 8,282 12 27 61 69 17 21 –8 75.1 Burkina Faso 4,182 4.2 148 185 31 19 50 83 13 19 –15 4.7 Burundi 669 –1.5 176 151 49 19 32 93 8 10 –10 12.8 Cambodia 4,299 6.6 .. 422 36 28 36 80 6 22 –8 3.4 Cameroon 12,449 2.7 837 1,213 45 19 37 71 12 17 –1 4.4 Canada 834,390 3.2 29,425 43,064 .. .. .. 56 19 20 5 1.5 Central African Rep. 1,198 1.8 383 502 61 25 14 75 13 18 –6 3.9 Chad 2,648 3.0 171 211 38 17 46 81 7 45 –33 6.7 Chile 72,416 5.6 4,854 6,226 9 34 57 63 11 22 3 7.0 China 1,409,852 9.5 227 338 15 53 32 44 13 42 1 4.9 Hong Kong, China 158,596 3.7 .. .. 0 12 88 57 11 23 9 1.8 Colombia 77,559 2.3 3,889 3,619 14 31 55 71 14 16 –2 17.8 Congo, Dem. Rep. 5,600 –3.9 250 212 58 19 23 92 4 7 –3 617.0 Congo, Rep. 3,510 1.8 486 469 6 61 33 35 18 23 24 7.9 Costa Rica 17,482 4.8 3,721 5,270 8 29 63 69 15 18 –2 14.9 Côte d’Ivoire 13,734 2.4 779 1,046 28 21 52 63 12 10 16 7.3 Croatia 28,322 1.7 .. 9,741 8 29 62 61 21 27 –9 53.0 Czech Republic 85,438 1.4 .. 6,382 4 40 57 53 21 28 –2 9.2 Denmark 212,404 2.4 29,551 63,131 3 27 71 48 26 20 6 2.0 Dominican Rep. 15,915 5.7 2,061 3,281 11 32 57 80 7 22 –9 9.1 Ecuador 26,913 1.9 4,726 3,310 9 29 62 70 12 22 –4 3.9 Egypt, Arab Rep. 82,427 4.5 1,000 1,316 16 34 50 72 13 17 –2 7.0 El Salvador 14,396 4.0 1,619 1,678 9 32 59 88 11 17 –16 5.7 Eritrea 734 4.0 .. 68 15 24 61 104 34 22 –60 10.3 Estonia 8,383 1.5 .. 3,650 5 30 65 62 18 32 –12 35.5 Ethiopia 6,638 4.3 .. 154 42 11 47 79 19 21 –19 5.4 Finland 161,549 2.9 23,140 42,306 3 33 64 51 22 20 8 2.0 France 1,747,973 1.9 30,635 59,243 3 25 72 55 24 19 2 1.5 Georgia 3,937 –3.2 .. .. 21 23 56 81 10 21 –12 185.8 Germany 2,400,655 1.5 16,783 33,686 1 30 69 59 19 18 4 1.6 Ghana 7,659 4.3 542 571 35 25 40 83 11 19 –14 26.4 Greece 173,045 2.7 10,578 13,860 7 22 70 67 16 23 –6 7.5 Guatemala 24,730 3.8 1,932 2,115 22 19 58 90 5 17 –12 9.3 Guinea 3,626 4.2 228 286 25 36 39 83 6 14 –4 5.2 Haiti 2,745 –0.8 .. .. 27 16 57 103 .. a 21 –24 19.4 Honduras 6,978 3.0 856 1,037 13 31 56 74 14 29 –17 16.2 Hungary 82,805 2.4 5,133 5,625 4 31 65 67 11 24 –2 16.4 India 598,966 5.8 342 401 23 26 52 65 13 24 –2 6.8 Indonesia 208,311 3.5 674 748 17 44 40 69 9 16 6 15.3 Iran, Islamic Rep. 136,833 4.0 2,613 3,737 11 37 53 64 10 30 –3 24.6 Ireland 148,553 7.6 .. .. 3 42 54 47 15 24 15 3.8 Israel 103,689 4.3 .. .. .. .. .. 60 31 16 –7 8.2 Italy 1,465,895 1.6 13,990 27,064 3 29 69 60 19 20 1 3.4 Jamaica 7,817 0.7 1,232 1,487 5 29 66 74 18 27 –19 18.6 Japan 4,326,444 1.3 25,293 33,077 1 31 68 56 17 26 1 –0.5 Jordan 9,860 4.6 1,810 1,145 2 26 72 80 23 23 –26 2.5 Kazakhstan 29,749 –0.6 .. 1,753 8 39 53 59 13 26 2 120.2 Kenya 13,842 1.8 265 213 17 19 64 70 19 16 –5 12.2 Korea, Rep. 605,331 5.5 .. 13,747 3 35 62 55 13 29 3 4.8 Kuwait 35,369 2.9 .. .. .. .. .. 56 26 9 9 2.6 Kyrgyz Rep. 1,737 –1.5 .. 1,861 39 23 38 68 19 18 –4 72.2 Lao PDR 2,036 6.3 462 621 51 23 26 .. .. 22 .. 28.6 Latvia 9,671 –0.1 .. 2,773 5 24 71 62 18 31 –10 31.5 Lebanon 19,000 4.6 .. 29,874 12 20 68 96 13 17 –26 12.2 Lesotho 1,135 3.4 591 575 16 42 42 85 33 34 –52 9.5 Lithuania 18,213 0.0 .. 3,431 7 34 59 64 20 21 –6 45.8 Macedonia, FYR 4,705 0.1 .. 4,243 12 30 57 85 12 22 –18 48.8 Madagascar 5,459 2.1 160 155 29 15 55 82 10 16 –8 16.0 Malawi 1,731 3.1 77 124 38 15 48 85 20 8 –13 30.9 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those speci�ed. 260 (c) The International Bank for Reconstruction and Development / The World Bank Table 3. Economic activity—continued Value added as % of GDP Agricultural productivity Agr. Value added per Gross domestic agricultural worker product 1995 dollars Agricultural Industry Services External GDP Household General gov’t. Gross balance of implicit Avg. �nal cons. �nal cons. capital goods and deflator Millions annual expenditure expenditure formation services Avg. annual of dollars % growth % of GDP % of GDP % of GDP % of GDP % growth 2003 1990–2003 1988–90 2000–2002 2003 2003 2003 2003 2003 2003 2003 1990–2003 Malaysia 103,161 5.9 5,678 6,912 9 49 42 46 14 22 18 3.4 Mali 4,326 4.9 251 274 36 27 37 79 10 22 –11 6.0 Mauritania 1,128 4.4 382 447 19 30 51 82 18 41 –41 5.6 Mexico 626,080 3.0 1,579 1,913 4 26 70 69 13 20 –2 16.5 Moldova 1,964 –5.9 .. 971 23 25 53 95 18 22 –34 78.9 Mongolia 1,188 1.7 1,124 1,444 28 15 57 63 19 31 –13 40.5 Morocco 44,491 2.7 1,823 1,513 18 30 52 64 20 23 –6 2.3 Mozambique 4,320 7.0 126 136 23 34 43 59 11 45 –15 24.8 Myanmar .. .. .. .. .. .. .. .. .. 15 .. 24.6 Namibia 4,658 3.7 1,055 1,545 10 31 59 58 28 24 –10 10.3 Nepal 5,835 4.6 188 203 40 21 39 79 10 26 –14 6.9 Netherlands 511,556 2.7 34,647 59,476 3 26 71 50 24 20 5 2.4 New Zealand 76,256 3.2 20,966 28,740 .. .. .. 60 19 20 2 1.6 Nicaragua 4,100 4.3 1,255 1,618 18 25 57 78 16 31 –25 28.3 Niger 2,730 2.7 211 197 40 17 43 82 12 16 –10 5.2 Nigeria 50,202 2.7 509 729 37 29 34 57 26 22 –5 23.1 Norway 221,579 3.4 21,358 37,073 2 38 60 43 20 20 17 3.2 Pakistan 68,815 3.6 544 719 23 23 53 73 12 15 0 8.6 Panama 12,916 4.1 2,192 2,967 6 14 81 70 7 26 –3 3.0 Papua New Guinea 3,395 2.8 695 823 26 39 35 .. .. .. .. 7.6 Paraguay 5,814 1.7 3,261 3,318 21 27 52 81 8 25 –15 11.2 Peru 61,011 3.9 1,399 1,863 8 29 64 72 10 19 –1 18.1 Philippines 80,574 3.5 1,354 1,458 14 32 53 72 11 19 –2 7.7 Poland 209,563 4.7 .. 1,879 3 31 66 70 16 19 –5 17.7 Portugal 149,454 2.6 5,391 7,567 4 30 66 61 21 28 –10 4.8 Romania 60,358 0.2 2,340 3,588 12 36 52 76 9 21 –5 78.1 Russian Federation 433,491 –1.8 .. 3,826 5 34 61 53 16 20 11 106.4 Rwanda 1,637 2.3 220 254 42 22 36 85 14 20 –19 10.6 Saudi Arabia 188,479 2.1 7,348 15,796 5 51 44 37 26 20 18 1.7 Senegal 6,496 4.0 352 354 17 21 62 75 14 20 –9 3.8 Serbia & Montenegro 19,176 0.5 .. .. .. .. .. 86 19 18 –23 52.9 Sierra Leone 793 –3.1 766 359 52 31 17 92 20 18 –31 24.6 Singapore 91,342 6.3 27,156 42,920 0 35 65 41 12 13 33 0.6 Slovak Rep. 31,868 2.5 .. .. 4 30 67 55 21 25 –1 9.3 Slovenia 26,284 4.0 .. 37,671 3 36 61 53 22 25 0 9.6 South Africa 159,886 2.3 3,428 4,072 4 31 65 67 14 15 4 9.0 Spain 836,100 2.8 12,860 22,412 3 30 66 58 18 26 –2 3.8 Sri Lanka 18,514 4.7 677 725 20 26 54 76 9 23 –7 9.0 Sweden 300,795 2.3 30,186 40,368 2 28 70 49 28 17 6 1.8 Switzerland 309,465 1.0 .. .. .. .. .. 61 14 21 4 1.1 Syrian Arab Rep. 21,517 4.3 2,056 2,636 23 29 48 66 11 24 0 6.6 Tajikistan 1,303 –3.2 .. 617 23 20 56 91 9 19 –19 147.0 Tanzania b 9,872 3.7 174 187 43 17 40 77 15 18 –10 17.4 Thailand 143,163 3.7 768 863 9 41 50 62 9 23 6 3.4 Togo 1,759 2.1 458 503 41 22 37 83 9 22 –14 5.9 Tunisia 24,282 4.6 2,228 3,115 13 30 58 64 15 25 –4 3.9 Turkey 237,972 3.1 1,848 1,848 13 22 65 67 14 23 –3 68.7 Turkmenistan 6,010 0.8 .. 690 25 44 30 55 13 33 0 226.6 Uganda 6,198 6.8 285 346 33 22 45 76 15 23 –14 8.8 Ukraine 49,537 –5.3 .. 1,576 14 40 46 60 16 19 5 155.0 United Kingdom 1,794,858 2.6 29,138 32,918 1 26 73 66 20 16 –2 2.8 United States 10,881,609 3.2 27,975 53,907 2 23 75 70 16 18 –4 2.0 Uruguay 11,182 1.5 6,832 8,177 9 27 64 73 12 11 3 23.9 Uzbekistan 9,949 1.2 .. 1,449 35 22 43 57 19 17 7 162.4 Venezuela, RB 84,793 0.5 4,449 5,399 3 43 54 70 6 12 12 39.5 Vietnam 39,157 7.5 192 256 23 39 38 66 6 32 –4 11.6 Yemen, Rep. 10,831 5.8 329 412 15 40 45 74 14 17 –5 18.6 Zambia 4,299 1.4 188 194 19 30 51 84 11 16 –11 41.8 Zimbabwe 8,304 1.1 292 355 17 24 59 72 17 8 2 32.3 World 36,356,240 t 2.6 w .. w 1,051 w 4w 28 w 68 w 62 w 17 w 20 w 1w Low income 1,101,435 4.7 329 383 25 25 50 68 13 22 –3 Middle income 5,995,502 3.3 .. 818 11 38 51 60 13 25 2 Lower middle income 4,146,612 3.4 522 716 12 40 48 58 13 27 2 Upper middle income 1,830,894 3.0 .. 4,027 7 32 61 65 13 18 4 Low & middle income 7,086,806 3.4 492 627 13 36 51 61 13 24 2 East Asia & Paci�c 2,050,713 7.2 .. .. 14 49 38 52 12 33 3 Europe & Cen. Asia 1,394,511 0.2 .. 2,376 9 31 60 61 16 21 2 Latin Am. & Carib. 1,733,889 2.7 2,770 3,591 7 25 68 62 16 19 3 Mid. East & N. Africa 676,986 3.2 1,917 2,340 11 41 48 54 18 23 5 South Asia 755,772 5.5 343 412 23 25 52 68 12 23 –2 Sub-Saharan Africa 417,336 2.7 382 360 14 29 57 68 16 18 –1 High income 29,270,317 2.5 .. .. 2 27 71 63 18 19 0 a. Data on general government �nal consumption expenditure are not available separately; they are included in household �nal consumption expenditure. b. Data cover mainland Tanzania only. 261 (c) The International Bank for Reconstruction and Development / The World Bank Table 4. Trade, aid, and �nance Merchandise trade exports imports External debt High Manufactured technology Current Net private Foreign Of�cial exports exports account capital direct development Domestic credit % of total % of balance flows investment assistance a Total Present provided by Millions Millions merchandise manufactured Millions Millions Millions Dollars Millions value banking sector of dollars of dollars exports exports of dollars of dollars of dollars per capita of dollars % of GNI % of GDP 2003 2003 2002 2002 2003 2002 2002 2002 2002 2002 2002 Albania 450 1,879 86 1 –408 136 135 101 1,312 20 43.6 Algeria 25,300 12,850 2 4 .. 1,023 1,065 12 22,800 42 29.1 Angola 9,075 4,175 .. .. –1,431 1,420 1,312 32 10,134 120 5.5 Argentina 29,349 13,813 31 7 9,559 681 785 0 132,314 66 62.4 Armenia 678 1,269 61 2 –186 108 111 96 1,149 34 7.3 Australia 70,358 88,618 29 16 –30,675 .. 16,364 .. .. .. 93.9 Austria 96,187 97,678 82 15 –2,392 .. 886 .. .. .. 124.3 Azerbaijan 2,592 2,626 6 8 –2,021 1,313 1,392 43 1,398 21 8.5 Bangladesh 6,820 9,660 92 0 739 132 47 7 17,037 22 40.2 Belarus 9,964 11,505 64 4 –505 227 247 4 908 7 17.5 Belgium 267,179 b 250,399 b 79 b 11 9,392 .. 73,635 b .. .. .. 115.4 Benin 425 765 6 0 –153 41 41 34 1,843 36 c 5.8 Bolivia 1,560 1,575 17 7 –347 601 677 77 4,867 23 c 62.3 Bosnia & Herzegovina 1,440 4,645 .. .. –2,096 299 293 143 2,515 34 35.8 Botswana 2,480 2,085 91 0 .. 35 37 22 480 8 –29.5 Brazil 73,084 50,665 54 19 –7,696 9,861 16,566 2 227,932 48 63.6 Bulgaria 7,439 10,742 61 –1,648 808 600 48 10,462 79 23.7 Burkina Faso 340 710 19 7 –449 8 8 40 1,580 16 c 12.4 Burundi 38 155 1 2 –39 –2 0 24 1,204 110 32.1 Cambodia 1,623 1,724 .. .. –64 54 54 37 2,907 68 6.0 Cameroon 1,885 1,970 7 1 .. 38 86 40 8,502 57 c 15.7 Canada 272,054 245,618 63 14 18,630 .. 20,501 .. .. .. 92.6 Central African Rep. 130 97 .. .. .. 4 4 16 1,066 78 13.2 Chad 230 852 .. .. .. 900 901 28 1,281 37 c 10.9 Chile 20,875 19,320 18 3 –594 2,781 1,713 –1 41,945 62 73.9 China 438,370 412,840 90 23 35,422 47,107 49,308 1 168,255 14 166.4 Hong Kong, China 224,040 d 207,168 95 d 17 17,414 .. 9,682 1 .. .. 144.5 Colombia 13,010 13,744 38 7 –1,417 947 2,023 10 33,853 46 36.7 Congo, Dem. Rep. 1,260 1,489 .. .. .. 32 32 16 8,726 171 0.2 Congo, Rep. 2,645 1,110 .. .. –62 331 331 115 5,152 228 11.4 Costa Rica 6,112 7,621 63 37 –946 602 662 1 4,834 33 36.9 Côte d’Ivoire 6,059 3,750 21 3 767 117 230 65 11,816 91 20.7 Croatia 6,164 14,199 73 12 –2,039 3,604 980 37 15,347 76 62.9 Czech Republic 48,723 51,306 89 14 –4,485 10,382 9,323 38 26,419 46 45.8 Denmark 67,887 58,749 66 22 4,991 .. 6,410 .. .. .. 156.6 Dominican Rep. 5,547 7,970 34 1 –875 1,351 961 18 6,256 30 45.1 Ecuador 5,988 6,534 10 7 –1,222 2,103 1,275 17 16,452 95 28.0 Egypt, Arab Rep. 5,750 13,280 35 1 622 437 647 19 30,750 28 109.9 El Salvador 3,136 5,763 58 6 –384 1,419 208 36 5,828 46 .. Eritrea 56 600 .. .. –223 21 21 54 528 40 148.9 Estonia 5,618 7,967 72 12 –1,150 1,586 285 51 4,741 86 49.6 Ethiopia 535 2,015 14 .. –70 71 75 19 6,523 63 c 61.9 Finland 52,834 41,312 85 24 9,295 .. 8,156 .. .. .. 64.7 France 384,662 388,373 81 21 25,744 .. 52,020 .. .. .. 105.0 Georgia 444 1,058 35 38 –392 149 165 60 1,838 42 19.6 Germany 748,375 493,712 86 17 53,513 .. 35,547 .. .. .. 144.7 Ghana 1,945 3,225 16 3 –106 27 50 33 7,338 73 c 31.9 Greece 13,040 45,379 52 10 –10,405 .. 53 .. .. .. 109.5 Guatemala 2,395 6,150 35 7 –1,193 61 110 21 4,676 21 15.7 Guinea 824 764 28 0 –41 0 0 32 3,401 47 12.5 Haiti 330 1,200 .. .. .. 6 6 19 1,248 23 37.3 Honduras 1,332 3,276 26 2 –266 100 143 64 5,395 50 34.1 Hungary 42,697 47,747 86 25 –2,644 221 54 46 34,958 64 53.8 India 54,740 69,743 75 5 4,656 4,944 3,030 1 104,429 17 58.5 Indonesia 60,650 32,390 54 16 6,085 –6,966 –1,513 6 132,208 89 59.4 Iran, Islamic Rep. 33,360 27,580 9 3 .. 816 37 2 9,154 7 45.3 Ireland 92,695 52,789 88 41 –2,990 .. 24,697 .. .. .. 110.6 Israel 31,577 36,430 93 20 –174 .. 1,649 115 .. .. 93.6 Italy 290,231 289,017 88 9 –21,942 .. 14,699 .. .. .. 99.6 Jamaica 1,215 3,815 64 0 –1,119 540 481 9 5,477 82 27.6 Japan 471,934 382,959 93 24 136,215 .. 9,087 .. .. .. 312.5 Jordan 3,000 5,579 68 3 –619 –31 56 103 8,094 83 89.6 Kazakhstan 12,900 8,327 19 10 –69 4,431 2,583 13 17,538 80 13.0 Kenya 2,395 3,735 24 10 –530 39 50 13 6,031 40 43.2 Korea, Rep. 194,325 178,784 92 32 6,092 .. 1,972 –2 .. .. 101.9 Kuwait 21,550 11,165 .. .. 4,192 .. 7 2 .. .. 105.8 Kyrgyz Rep. 582 717 33 6 –32 –54 5 37 1,797 93 11.4 Lao PDR 371 508 .. .. –82 25 25 50 2,664 85 12.3 Latvia 2,896 5,248 59 4 –956 496 382 37 6,690 85 39.6 Lebanon 1,458 7,035 69 3 –3,587 4,803 257 103 17,077 102 185.7 Lesotho 427 914 .. .. –119 73 81 43 637 45 10.7 Lithuania 7,252 9,870 58 5 –1,214 760 712 42 6,199 49 18.0 Macedonia, FYR 1,336 2,206 70 1 –177 113 77 136 1,619 37 15.9 Madagascar 626 843 .. .. –270 8 8 23 4,518 33 c 18.4 Malawi 460 720 0 3 –174 6 6 35 2,912 51 c 21.6 Taiwan, China* 150,646 127,258 94 42 25,678 .. .. 0 .. .. .. Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those speci�ed. 262 (c) The International Bank for Reconstruction and Development / The World Bank Table 4. Trade, aid, and �nance—continued Merchandise trade exports imports External debt High Manufactured technology Current Net private Foreign Of�cial exports exports account capital direct development Domestic credit % of total % of balance flows investment assistance a Total Present provided by Millions Millions merchandise manufactured Millions Millions Millions Dollars Millions value banking sector of dollars of dollars exports exports of dollars of dollars of dollars per capita of dollars % of GNI % of GDP 2003 2003 2002 2002 2003 2002 2002 2002 2002 2002 2002 Malaysia 100,726 81,067 79 58 7,190 4,807 3,203 4 48,557 57 154.2 Mali 985 1,010 .. .. –310 102 102 42 2,803 47 c 16.5 Mauritania 369 471 .. .. .. 16 12 135 2,309 56 c –8.2 Mexico 165,334 178,990 84 21 –9,150 10,261 14,622 1 141,264 26 38.0 Moldova 791 1,403 31 4 –92 77 111 33 1,349 78 29.1 Mongolia 516 787 36 0 –105 78 78 85 1,037 69 17.1 Morocco 8,701 14,158 66 11 413 15 428 21 18,601 51 c 84.5 Mozambique 730 1,305 8 3 –657 381 406 112 4,609 27 c 13.4 Myanmar 2,802 2,515 .. .. –309 69 129 2 6,556 .. 35.1 Namibia 1,155 1,590 52 1 130 .. .. 68 .. .. 49.0 Nepal 650 1,730 67 0 –165 9 10 15 2,953 31 43.2 Netherlands 293,437 261,135 74 28 16,467 .. 28,534 .. .. .. 160.4 New Zealand 16,505 18,559 28 10 –3,530 .. 823 .. .. .. 118.2 Nicaragua 590 1,865 19 5 –888 206 174 97 6,485 77 93.0 Niger 350 510 3 8 .. 0 8 26 1,797 26 c 8.5 Nigeria 20,255 10,890 0 0 .. 639 1,281 2 30,476 82 26.5 Norway 68,130 39,895 22 22 28,643 .. 502 .. .. .. 54.0 Pakistan 11,901 13,034 85 1 3,597 379 823 15 33,672 45 43.5 Panama 905 2,980 12 1 –408 180 57 12 8,298 84 90.7 Papua New Guinea 2,146 1,193 2 19 286 –46 50 38 2,485 82 25.9 Paraguay 1,289 2,079 15 3 376 34 –22 10 2,967 42 28.8 Peru 8,864 8,494 21 2 –1,116 3,131 2,391 18 28,167 56 23.9 Philippines 37,065 39,301 50 65 2,060 3,549 1,111 7 59,342 77 60.5 Poland 52,285 66,887 82 3 6,178 5,075 4,131 30 69,521 37 35.8 Portugal 31,172 44,821 86 7 –7,549 .. 1,790 .. .. .. 149.9 Romania 17,618 24,003 81 3 –1,525 3,173 1,144 31 14,683 37 13.2 Russian Federation 135,162 74,496 22 13 35,905 8,011 3,009 9 147,541 50 26.7 Rwanda 60 240 3 1 –192 3 3 44 1,435 40 c 11.3 Saudi Arabia 88,500 34,089 10 0 11,889 .. .. 1 .. .. 70.1 Senegal 1,330 2,270 51 4 .. 94 93 45 3,918 53 c 22.6 Serbia & Montenegro 2,522 7,140 .. .. –1,750 507 475 237 e 12,688 f 102 .. Sierra Leone 91 320 .. .. .. 5 5 68 1,448 103 c 48.4 Singapore 144,134 d 127,898 85 d 60 18,704 .. 6,097 2 .. .. 83.5 Slovak Rep. 22,035 22,318 85 3 .. 5,460 4,012 35 13,013 61 51.7 Slovenia 12,738 13,812 90 5 15 .. 1,865 87 .. .. 46.0 South Africa 36,452 g 38,141 g 63 g 5 –1,456 783 739 14 25,041 22 147.5 Spain 151,876 200,088 78 7 –23,676 .. 36,727 .. .. .. 129.6 Sri Lanka 5,060 6,455 74 1 –264 206 242 18 9,611 48 43.6 Sweden 100,939 82,317 81 16 10,624 .. 11,828 .. .. .. 75.2 Switzerland 100,550 96,345 93 21 26,011 .. 3,599 .. .. .. 174.4 Syrian Arab Rep. 5,980 4,835 7 1 1,440 224 225 5 21,504 117 27.9 Tajikistan 798 881 13 42 –41 –10 9 27 1,153 89 21.3 Tanzania 990 2,120 17 2 –964 214 240 35 7,244 19 c,h 10.0 Thailand 80,253 75,679 74 31 7,965 –1,992 900 5 59,212 49 116.0 Togo 425 558 43 1 –169 75 75 11 1,581 92 17.0 Tunisia 8,027 10,909 82 4 –844 1,625 795 49 12,625 65 74.4 Turkey 46,573 67,734 84 2 –1,521 7,582 1,037 9 131,556 77 59.1 Turkmenistan 3,403 2,516 7 5 –74 .. 100 8 .. .. 19.1 Uganda 525 1,240 8 12 –353 149 150 26 4,100 22 c 15.4 Ukraine 17,954 23,021 67 5 2,891 –576 693 10 13,555 35 27.5 United Kingdom 303,890 388,282 79 31 –26,713 .. 29,179 .. .. .. 145.3 United States 724,006 1,305,648 81 32 –541,834 .. 39,633 .. .. .. 246.6 Uruguay 2,169 2,190 37 3 354 107 177 4 10,736 65 93.3 Uzbekistan 2,936 2,576 .. .. 659 –11 65 7 4,568 38 .. Venezuela, RB 23,650 9,306 13 3 7,423 –1,639 690 2 32,563 33 15.0 Vietnam 19,660 24,020 .. .. –604 759 1,400 16 13,349 35 44.8 Yemen, Rep. 4,355 2,892 .. .. 340 114 114 31 5,290 40 –0.5 Zambia 940 1,503 14 2 .. 186 197 63 5,969 127 46.7 Zimbabwe 1,225 2,835 38 3 .. –3 26 15 4,066 .. 58.7 World 7,479,592 t 7,624,797 t 78 w 21 w .. s 630,827 s 11 w .. s 179.5 w Low income 176,218 198,033 47 4 7,151 i 12,941 i 12 523,464 i 46.9 Middle income 1,813,068 1,675,174 60 18 146,679 i 134,145 i 9 1,815,384 i,j 82.9 Lower middle income 1,147,024 1,066,326 60 17 98,852 i 91,104 i 8 1,147,339 i 97.9 Upper middle income 666,731 608,848 60 21 47,828 i 43,041 i 12 668,045 i,j 53.0 Low & middle income 1,989,214 1,873,207 60 17 153,831 147,086 10 2,338,848 j 77.7 East Asia & Paci�c 746,144 676,038 79 32 47,524 54,834 4 497,354 143.8 Europe & Central Asia 458,205 k 474,286 k 57 10 53,739 32,931 27 545,842 36.8 Latin America & Carib. 374,300 359,950 48 16 34,544 44,682 10 727,944 46.8 Middle East & N. Africa 222,781 155,327 19 2 5,359 2,653 21 189,010 72.1 South Asia 79,505 102,282 77 4 5,697 4,164 5 168,349 55.3 Sub-Saharan Africa 109,680 105,324 35 4 6,968 7,822 28 210,350 65.0 High income 5,491,151 5,741,481 82 23 .. 483,741 204.1 a. Regional aggregates include data for economies that are not speci�ed elsewhere. World and income group totals include aid not allocated by country or region. b. Includes Luxembourg. c. Data are from debt sustainability analysis undertaken as part of the Heavily Indebted Poor Countries (HIPC) initiative. d. Includes re-exports. e. Aid to the states of the former Socialist Federal Republic of Yugoslavia that is not otherwise speci�ed is included in regional and income group aggregates. f. Data are estimates and reflect borrowing by the former Socialist Federal Republic of Yugoslavia that are not yet allocated to the successor republics. g. Data on total exports and imports refer to South Africa only. Data on export commodity shares refer to the South African Customs Union (Botswana, Lesotho, Namibia, South Africa, and Swaziland). h. GNI refers to mainaland Tanzania only. i. The aggregates reflect country groupings from Global Development Finance 2004. j. Includes data for Gibraltar not included in oher tables. k. Data include the intratrade of the Baltic states and the Commonwealth of Independent States. 263 (c) The International Bank for Reconstruction and Development / The World Bank Table 5. Key indicators for other economies PPP Gross national gross national Population income (GNI) a income (GNI) b Gross Under-5 Adult Carbon density domestic Life mortality Literacy dioxide Avg. people Millions per Millions per product expectancy rate rate emissions annual per of capita of capita per capita at birth Per % of people Thousands Thousands % growth sq. km dollars dollars dollars dollars % growth Years 1,000 15 and above of tons 2003 1990–2003 2003 2003 2003 2003 2003 2002–2003 2002 2002 2002 2000 Afghanistan 28,766 c 3.7 44 .. .. d .. .. .. 43 257 .. 905 American Samoa 70 .. 353 .. .. e .. .. .. .. .. .. 286 Andorra 69 1.8 136 .. .. f .. .. .. .. 7 .. .. Antigua & Barbuda 79 1.6 179 719 9,160 753 9,590 0.4 75 14 .. 352 Aruba 97 .. 511 .. .. f .. .. .. .. .. .. 1,924 Bahamas, The 317 1.6 32 4,684 15,110 5,067 16,140 –0.6 70 16 .. 1,795 Bahrain 712 2.7 1,003 7,569 11,260 11,288 16,170 1.8 73 16 88 19,500 Barbados 271 0.4 630 2,512 9,270 4,080 15,060 0.8 75 14 100 1,176 Belize 259 2.4 11 807 3,190 1,476 5,840 1.8 74 40 77 g 780 Bermuda 64 0.4 1,280 .. .. f .. .. .. .. .. .. 462 Bhutan 874 2.9 19 578 660 .. .. 4.0 63 94 .. 396 Brunei 356 2.5 68 .. .. f .. .. .. 77 6 .. 4,668 Cape Verde 470 2.5 117 701 1,490 2,558 h 5,440 h 2.4 69 38 76 139 Cayman Islands 39 .. 150 .. .. f .. .. .. .. .. .. 286 Channel Islands 149 0.3 745 .. .. f .. .. .. 79 .. .. .. Comoros 600 2.5 269 269 450 1,056 h 1,760 h 0.1 61 79 56 81 Cuba 11,299 0.5 103 .. .. i .. .. .. 77 9 97 30,913 Cyprus 770 0.9 83 9,373 12,320 15,042 h 19,530 h 3.3 78 6 97 g 6,423 Djibouti 705 2.8 30 643 910 1,550 h 2,200 h 1.8 44 143 .. 385 Dominica 71 –0.1 95 239 3,360 362 5,090 –0.7 77 15 .. 103 Equatorial Guinea 494 2.6 18 437 930 .. .. 12.8 52 152 .. 205 Faeroe Islands 46 –0.2 33 .. .. f .. .. .. .. .. .. 649 Fiji 835 1.0 46 1,969 2,360 4,517 h 5,410 h 3.5 70 21 .. 725 French Polynesia 243 1.6 66 .. .. f .. .. .. 74 .. .. 542 Gabon 1,344 2.6 5 4,813 3,580 7,656 5,700 1.2 53 85 .. 3,499 Gambia, The 1,421 3.3 142 442 310 2,591 h 1,820 h 6.3 53 126 .. 271 Greenland 56 0.0 0 .. .. f .. .. .. 69 .. .. 557 Grenada 105 0.8 308 396 3,790 702 6,710 1.4 73 25 .. 213 Guam 162 1.5 295 .. .. f .. .. .. 78 .. .. 4,071 Guinea-Bissau 1,489 2.9 53 202 140 983 660 –16.9 45 211 .. 264 Guyana 769 0.4 4 689 900 3,035 h 3,950 h –1.0 62 72 .. 1,598 Iceland 286 0.9 3 8,813 30,810 8,619 30,140 1.2 80 4 .. 2,158 Iraq 24,700 2.4 56 .. .. i .. .. .. 63 125 .. 76,336 Isle of Man 74 0.7 125 .. .. f .. .. .. .. .. .. .. Kiribati 96 2.2 132 84 880 .. .. 0.4 63 69 .. 26 Korea, Dem. Rep. 22,612 1.0 188 .. .. d .. .. .. 62 55 .. 188,857 Liberia 3,374 2.5 35 445 130 .. .. –2.3 47 235 56 399 Libya 5,559 2.0 3 .. .. e .. .. .. 72 19 82 57,125 Liechtenstein 33 1.3 207 .. .. f .. .. .. .. 11 .. .. Luxembourg 448 1.2 171 19,683 43,940 24,385 54,430 0.3 78 5 .. 8,482 Macao, China 444 1.4 .. 6,335 14,600 j 9,624 h 21,920 h .. 79 .. 91 g 1,634 Maldives 293 2.5 977 674 I 2,300 .. .. 6.1 69 77 97 498 Malta 399 0.8 1,247 3,678 9,260 7,096 17,870 .. 78 5 93 2,814 Marshall Islands 53 1.1 265 143 2,710 .. .. 2.0 65 66 .. .. Mauritius 1,225 1.1 603 5,012 4,090 13,789 11,260 2.1 73 19 84 2,895 Mayotte 166 .. 400 .. .. e .. .. .. 60 .. .. .. Micronesia, Fed. Sts. 125 2.0 174 261 2,090 .. .. –0.1 69 24 .. .. Monaco 32 1.1 16,842 .. .. f .. .. .. .. 5 .. .. Netherlands Antilles 220 1.1 275 .. .. f .. .. .. 76 .. 97 9,929 New Caledonia 225 2.2 12 .. .. f .. .. .. 74 .. .. 1,667 Northern Mariana Islands 80 .. 159 .. .. e .. .. .. .. .. .. .. Oman 2,599 3.6 8 19,877 7,830 32,985 13,000 .. 74 13 74 19,775 Palau 20 2.2 43 150 7,500 .. .. 1.5 70 29 .. 242 Puerto Rico 3,898 0.7 439 42,057 10,950 62,674 16,320 .. 77 .. 94 8,735 Qatar 624 1.9 57 .. .. f .. .. .. 75 16 .. 40,685 Samoa 178 0.8 63 284 1,600 1,015 h 5,700 h 1.9 69 25 99 139 San Marino 28 1.5 277 .. .. f .. .. .. .. 6 .. .. São Tomé & Principe 157 2.4 164 50 320 .. .. 2.5 66 118 .. 88 Seychelles 84 1.4 186 626 7,480 1,336 15,960 –6.5 73 16 92 g 227 Solomon Islands 457 2.8 16 273 600 746 h 1,630 h 0.7 69 24 .. 165 Somalia 9,626 2.3 15 .. .. d .. .. .. 47 225 .. .. St. Kitts & Nevis 47 0.8 130 321 6,880 516 11,040 2.4 71 24 .. 103 St. Lucia 161 1.4 263 650 4,050 839 5,220 0.8 74 19 .. 322 St. Vincent & the Grenadines 109 0.2 280 361 3,300 719 6,590 2.8 73 25 .. 161 Sudan 33,546 2.3 14 15,372 460 63,145 h 1,880 h 3.6 58 94 60 5,221 Suriname 438 0.7 3 841 1,990 .. .. .. 70 40 .. 2,118 Swaziland 1,106 2.8 64 1,492 1,350 5,359 4,850 0.6 44 149 81 381 Timor-Leste 810 0.7 54 351 430 .. .. .. .. 126 .. .. Tonga 102 0.5 142 152 1,490 703 h 6,890 h 1.7 71 20 .. 121 Trinidad & Tobago 1,313 0.6 256 9,538 7,260 12,405 9,450 3.1 72 20 98 26,362 United Arab Emirates 4,041 6.3 48 .. .. f 78,977 h 21,040 h –5.0 75 9 77 58,913 Vanuatu 210 2.7 17 248 1,180 605 2,880 –0.2 69 42 .. 81 Virgin Islands (U.S.) 112 0.6 329 .. .. f .. .. .. 78 .. .. 13,106 West Bank & Gaza 3,367 4.1 .. 3,734 1,110 .. .. –5.2 73 .. .. .. Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those speci�ed. a. Preliminary World Bank estimates calculated using the World Bank Atlas method. b. Purchasing power parity; see the Technical Notes. c. Estimate does not account for recent refugee flows. d. Estimated to be low income ($765 or less). e. Estimated to be upper middle income ($3,036 to $9,385). f. Estimated to be high income ($9,386 or more). g. National estimates based on census data. h. The estimate is based on regression; others are extrapolated from the latest Internaional Comparison Programme bencmark estimates. i. Estimated to be lower middle income ($766 to $3,035). j. Refers to GDP and GDP per capita. 264 (c) The International Bank for Reconstruction and Development / The World Bank Selected world development indicators 265 Technical notes cise quantitative measures of those differences. Discrepancies These technical notes discuss the sources and methods used to in data presented in different editions reflect updates by coun- compile the indicators included in this edition of Selected tries as well as revisions to historical series and changes in World Development Indicators. The notes follow the order in methodology. Thus readers are advised not to compare data which the indicators appear in the tables. Note that the series between editions or between different editions of World Selected World Development Indicators uses terminology in Bank publications. Consistent time series are available from line with the 1993 System of National Accounts (SNA). For the World Development Indicators 2004 CD-ROM. example, in the 1993 SNA gross national income replaces gross national product. See the technical notes for tables 1 and 3 for Ratios and growth rates other examples. For ease of reference, the tables usually show ratios and rates of growth rather than the simple underlying values. Values in Sources their original form are available from the World Development The data published in the Selected World Development Indi- Indicators 2004 CD-ROM. Unless otherwise noted, growth cators are taken from World Development Indicators 2004. rates are computed using the least-squares regression method Where possible, however, revisions reported since the closing (see Statistical methods below). Because this method takes into date of that edition have been incorporated. In addition, newly account all available observations during a period, the result- released estimates of population and gross national income ing growth rates reflect general trends that are not unduly (GNI) per capita for 2003 are included in table 1. influenced by exceptional values. To exclude the effects of infla- The World Bank draws on a variety of sources for the statis- tion, constant price economic indicators are used in calculat- tics published in the World Development Indicators. Data on ing growth rates. Data in italics are for a year or period other external debt are reported directly to the World Bank by devel- than that speci�ed in the column heading—up to two years oping member countries through the Debtor Reporting Sys- before or after for economic indicators and up to three years tem. Other data are drawn mainly from the United Nations for social indicators, because the latter tend to be collected less and its specialized agencies, from the International Monetary regularly and change less dramatically over short periods. Fund (IMF), and from country reports to the World Bank. Bank staff estimates are also used to improve currentness or Constant price series consistency. For most countries, national accounts estimates An economy’s growth is measured by the increase in value are obtained from member governments through World Bank added produced by the individuals and enterprises operating in economic missions. In some instances these are adjusted by that economy. Thus, measuring real growth requires estimates staff to ensure conformity with international de�nitions and of GDP and its components valued in constant prices. The concepts. Most social data from national sources are drawn World Bank collects constant price national accounts series in from regular administrative �les, special surveys, or periodic national currencies and recorded in the country’s original base censuses. year. To obtain comparable series of constant price data, it For more detailed notes about the data, please refer to the rescales GDP and value added by industrial origin to a common World Bank’s World Development Indicators 2004. reference year, currently 1995. This process gives rise to a dis- crepancy between the rescaled GDP and the sum of the rescaled Data consistency and reliability components. Because allocating the discrepancy would give rise Considerable effort has been made to standardize the data, but to distortions in the growth rate, it is left unallocated. full comparability cannot be assured, and care must be taken in interpreting the indicators. Many factors affect data availabil- Summary measures ity, comparability, and reliability: statistical systems in many The summary measures for regions and income groups, pre- developing economies are still weak; statistical methods, cover- sented at the end of most tables, are calculated by simple addi- age, practices, and de�nitions differ widely; and cross-country tion when they are expressed in levels. Aggregate growth rates and intertemporal comparisons involve complex technical and and ratios are usually computed as weighted averages. The conceptual problems that cannot be unequivocally resolved. summary measures for social indicators are weighted by popu- Data coverage may not be complete because of special circum- lation or subgroups of population, except for infant mortality, stances or for economies experiencing problems (such as those which is weighted by the number of births. See the notes on stemming from conflicts) affecting the collection and report- speci�c indicators for more information. ing of data. For these reasons, although the data are drawn For summary measures that cover many years, calculations from the sources thought to be most authoritative, they should are based on a uniform group of economies so that the com- be construed only as indicating trends and characterizing position of the aggregate does not change over time. Group major differences among economies rather than offering pre- measures are compiled only if the data available for a given (c) The International Bank for Reconstruction and Development / The World Bank 266 WORLD DEVELOPMENT REPORT 2005 year account for at least two-thirds of the full group, as de�ned recent round of surveys, completed in 1999; the rest are either for the 1995 benchmark year. As long as this criterion is met, from the 1996 survey, or data from the 1993 or earlier round economies for which data are missing are assumed to behave and extrapolated to the 1996 benchmark. Estimates for coun- like those that provide estimates. Readers should keep in mind tries not included in the surveys are derived from statistical that the summary measures are estimates of representative models using available data. aggregates for each topic and that nothing meaningful can be PPP GNI per capita is PPP GNI divided by midyear deduced about behavior at the country level by working back population. from group indicators. In addition, the estimation process may Gross domestic product (GDP) per capita growth is based result in discrepancies between subgroup and overall totals. on GDP measured in constant prices. Growth in GDP is con- sidered a broad measure of the growth of an economy. GDP in Table 1. Key indicators of development constant prices can be estimated by measuring the total quan- Population is based on the de facto de�nition, which counts all tity of goods and services produced in a period, valuing them residents, regardless of legal status or citizenship, except for at an agreed set of base year prices, and subtracting the cost of refugees not permanently settled in the country of asylum, intermediate inputs, also in constant prices. See the section on who are generally considered part of the population of the statistical methods for details of the least-squares growth rate. country of origin. Life expectancy at birth is the number of years a newborn Average annual population growth rate is the exponential infant would live if patterns of mortality prevailing at its birth rate of change for the period (see the section on statistical were to stay the same throughout its life. methods below). Under-5 mortality rate is the probability that a newborn Population density is midyear population divided by land child will die before reaching age 5, if the child is subject to cur- area. Land area is a country’s total area excluding areas under rent age speci�c mortality rates. The probability is expressed as inland bodies of water and coastal waterways. Density is calcu- a rate per 1,000. lated using the most recently available data on land area. Adult literacy rate is the percentage of persons aged 15 and Gross national income (GNI—formerly gross national above who can, with understanding, both read and write a product or GNP), the broadest measure of national income, short, simple statement about their everyday life. In practice, measures total value added from domestic and foreign sources literacy is dif�cult to measure. To estimate literacy using such a claimed by residents. GNI comprises gross domestic product de�nition requires census or survey measurements under con- (GDP) plus net receipts of primary income from foreign trolled conditions. Many countries estimate the number of lit- sources. Data are converted from national currency to current erate people from self-reported data. Some use educational U.S. dollars using the World Bank Atlas method. This involves attainment data as a proxy but apply different lengths of school using a three-year average of exchange rates to smooth the attendance or level completion. As de�nition and methodolo- effects of transitory exchange rate fluctuations. (See the section gies of data collection differ across country—and even over on statistical methods below for further discussion of the Atlas time within countries—data need to be used with caution method.) Carbon dioxide emissions (CO2) measures those emis- GNI per capita is GNI divided by midyear population. It is sions stemming from the burning of fossil fuels and the manu- converted into current U.S. dollars by the Atlas method. The facture of cement. These include carbon dioxide produced World Bank uses GNI per capita in U.S dollars to classify during consumption of solid, liquid, and gas fuels and from economies for analytical purposes and to determine borrow- gas flaring. ing eligibility. The Carbon Dioxide Information Analysis Center PPP Gross national income, which is GNI converted into (CDIAC), sponsored by the U.S. Department of Energy, calcu- international dollars using purchasing power parity (PPP) lates annual anthropogenic emissions of CO2. These calcula- conversion factors, is included because nominal exchange rates tions are derived from data on fossil fuel consumption, based do not always reflect international differences in relative prices. on the World Energy Data Set maintained by the UNSD, and At the PPP rate, one international dollar has the same purchas- from data on world cement manufacturing, based on the ing power over domestic GNI that the U.S. dollar has over U.S. Cement Manufacturing Data Set maintained by the U.S. GNI. PPP rates allow a standard comparison of real price levels Bureau of Mines. Each year the CDIAC recalculates the entire between countries, just as conventional price indexes allow time series from 1950 to the present, incorporating its most comparison of real values over time. The PPP conversion fac- recent �ndings and the latest corrections to its database. Esti- tors used here are derived from price surveys covering 118 mates exclude fuels supplied to ships and aircraft engaged in countries conducted by the International Comparison Pro- international transportation because of the dif�culty of appor- gram. For Organisation for Economic Co-operation and tioning these fuels among the countries bene�ting from that Development (OECD) countries data come from the most transport. (c) The International Bank for Reconstruction and Development / The World Bank Selected world development indicators 267 Table 2. Poverty and income distribution Problems also exist in comparing poverty measures within Survey year is the year in which the underlying data were countries. For example, the cost of living is typically higher in collected. urban than in rural areas. So the urban monetary poverty line Rural poverty rate is the percentage of the rural population should be higher than the rural poverty line. But it is not living below the rural poverty line. Urban poverty rate is the always clear that the difference between urban and rural percentage of the urban population living below the urban poverty lines found in practice properly reflects the difference poverty line. National poverty rate is the percentage of the in the cost of living. In some countries the urban poverty line total population living below the national poverty line. in common use has a higher real value than does the rural National estimates are based on population weighted sub- poverty line. Sometimes the difference has been so large as to group estimates from household surveys. imply that the incidence of poverty is greater in urban than in Population below $1 PPP a day and $2 PPP a day are the rural areas, even though the reverse is found when adjustments percentages of the population living on less than $1.08 a day and are made only for differences in the cost of living. As with $2.15 a day at 1993 international prices.As a result of revisions in international comparisons, when the real value of the poverty PPP exchange rates, they cannot be compared with poverty rates line varies, it is not clear how meaningful such urban-rural reported in previous editions for individual countries. comparisons are. Poverty gap at $1 PPP a day and Poverty gap at $2 PPP a The problems of making poverty comparisons do not end day is the mean shortfall below the poverty line (counting the there. More issues arise in measuring household living stan- non-poor as having zero shortfall), expressed as a percentage dards. The choice between income and consumption as a wel- of the poverty line. This measure reflects the depth of poverty fare indicator is one issue. Income is generally more dif�cult to as well as its incidence. measure accurately, and consumption accords better with the International comparisons of poverty data entail both con- idea of the standard of living than does income, which can vary ceptual and practical problems. Different countries have differ- over time even if the standard of living does not. But consump- ent de�nitions of poverty, and consistent comparisons tion data are not always available, and when they are not there between countries can be dif�cult. Local poverty lines tend to is little choice but to use income. There are still other problems. have higher purchasing power in rich countries, where more Household survey questionnaires can differ widely, for exam- generous standards are used than in poor countries. Is it rea- ple, in the number of distinct categories of consumer goods sonable to treat two people with the same standard of living— they identify. Survey quality varies, and even similar surveys in terms of their command over commodities—differently may not be strictly comparable. because one happens to live in a better-off country? Can we Comparisons across countries at different levels of develop- hold the real value of the poverty line constant across coun- ment also pose a potential problem, because of differences in tries, just as we do when making comparisons over time? the relative importance of consumption of nonmarket goods. Poverty measures based on an international poverty line The local market value of all consumption in kind (including attempt to do this. The commonly used $1 a day standard, consumption from own production, particularly important in measured in 1985 international prices and adjusted to local underdeveloped rural economies) should be included in the currency using purchasing power parities (PPPs), was chosen measure of total consumption expenditure. Similarly, the for the World Bank’s World Development Report 1990: Poverty imputed pro�t from production of nonmarket goods should be because it is typical of the poverty lines in low-income coun- included in income. This is not always done, though such omis- tries. PPP exchange rates, such as those from the Penn World sions were a far bigger problem in surveys before the 1980s. Tables or the World Bank, are used because they take into Most survey data now include valuations for consumption or account the local prices of goods and services not traded inter- income from own production. Nonetheless, valuation methods nationally. But PPP rates were designed not for making inter- vary. For example, some surveys use the price in the nearest national poverty comparisons but for comparing aggregates market, while others use the average farm gate selling price. from national accounts. Thus there is no certainty that an Wherever possible, consumption has been used as the wel- international poverty line measures the same degree of need or fare indicator for deciding who is poor. Where consumption deprivation across countries. data are unavailable, income data are used, though there is a This year’s edition (like those of the last four years) uses change in this year’s edition in how income surveys are used. In 1993 consumption PPP estimates produced by the World the past, average income was adjusted to accord with con- Bank. The international poverty line, set at $1 a day in 1985 sumption and income data from national accounts. This PPP terms, has been recalculated in 1993 PPP terms at about approach was tested using data for more than 20 countries for $1.08 a day. Any revisions in the PPP of a country to incorpo- which the surveys provided both income and consumption rate better price indexes can produce dramatically different expenditure data. Income gave a higher mean than consump- poverty lines in local currency. tion but also greater income inequality. These two effects (c) The International Bank for Reconstruction and Development / The World Bank 268 WORLD DEVELOPMENT REPORT 2005 roughly canceled each other out when poverty measures based Because the underlying household surveys differ in on consumption were compared with those based on income method and in the type of data collected, the distribution from the same survey; statistically, there was no signi�cant dif- data are not strictly comparable across countries. These ference. So this year’s edition uses income data to estimate problems are diminishing as survey methods improve and poverty directly and no longer adjusts the income mean. become more standardized, but achieving strict comparabil- In all cases the measures of poverty have been calculated ity is still impossible. from primary data sources (tabulations or household data) Two sources of noncomparability should be noted. First, rather than existing estimates. Estimation from tabulations the surveys can differ in many respects, including whether they requires an interpolation method; the method chosen was use income or consumption expenditure as the living standard Lorenz curves with flexible functional forms, which have indicator. The distribution of income is typically more unequal proved reliable in past work. Empirical Lorenz curves were than the distribution of consumption. In addition, the de�ni- weighted by household size, so they are based on percentiles of tions of income used usually differ among surveys. Consump- population, not households. tion is usually a much better welfare indicator, particularly in Gini index measures the extent to which the distribution of developing countries. Second, households differ in size (num- income (or, in some cases, consumption expenditure) among ber of members) and in the extent of income sharing among individuals or households within an economy deviates from a members. And individuals differ in age and consumption perfectly equal distribution. A Lorenz curve plots the cumula- needs. Differences among countries in these respects may bias tive percentages of total income received against the cumula- comparisons of distribution. tive number of recipients, starting with the poorest individual World Bank staff have made an effort to ensure that the or household. The Gini index measures the area between the data are as comparable as possible. Wherever possible, con- Lorenz curve and a hypothetical line of absolute equality, sumption has been used rather than income. Income distribu- expressed as a percentage of the maximum area under the line. tion and Gini indexes for high-income countries are calculated Thus a Gini index of zero represents perfect equality, while an directly from the Luxembourg Income Study database, using index of 100 implies perfect inequality. an estimation method consistent with that applied for devel- Percentage share of income or consumption is the share oping countries. that accrues to subgroups of population indicated by quintiles. Inequality in the distribution of income is reflected in the Table 3. Economic activity percentage shares of income or consumption accruing to Gross domestic product is gross value added, at purchasers’ segments of the population ranked by income or consump- prices, by all resident producers in the economy plus any taxes tion levels. The segments ranked lowest by personal income and minus any subsidies not included in the value of the prod- receive the smallest shares of total income. The Gini index ucts. It is calculated without deducting for depreciation of fab- provides a convenient summary measure of the degree of ricated assets or for depletion or degradation of natural inequality. resources. Value added is the net output of an industry after Data on personal or household income or consumption adding up all outputs and subtracting intermediate inputs. come from nationally representative household surveys. The The industrial origin of value added is determined by the data in the table refer to different years between 1989 and 2002. International Standard Industrial Classi�cation (ISIC) revi- Footnotes to the survey year indicate whether the rankings are sion 3. The World Bank conventionally uses the U.S. dollar and based on per capita income or consumption. Each distribution applies the average of�cial exchange rate reported by the Inter- is based on percentiles of population—rather than of house- national Monetary Fund for the year shown. An alternative holds—with households ranked by income or expenditure per conversion factor is applied if the of�cial exchange rate is person. judged to diverge by an exceptionally large margin from the Where the original data from the household survey were rate effectively applied to transactions in foreign currencies available, they have been used to directly calculate the income and traded products. (or consumption) shares by quintile. Otherwise shares have Gross domestic product average annual growth rate is cal- been estimated from the best available grouped data. culated from constant price GDP data in local currency. The distribution data have been adjusted for household Agricultural productivity refers to the ratio of agricultural size, providing a more consistent measure of per capita income value added, measured in constant 1995 U.S. dollars, to the or consumption. No adjustment has been made for spatial dif- number of workers in agriculture. ferences in cost of living within countries, because the data Value added is the net output of an industry after adding needed for such calculations are generally unavailable. For fur- up all out-puts and subtracting intermediate inputs. The ther details on the estimation method for low and middle- industrial origin of value added is determined by the Interna- income economies, see Ravallion and Chen (1996). tional Standard Industrial Classi�cation (ISIC) revision 3. (c) The International Bank for Reconstruction and Development / The World Bank Selected world development indicators 269 Agriculture value added corresponds to ISIC divisions 1–5 banks by visiting and resident World Bank missions. Data for and includes forestry and �shing. high-income economies come from the Organization for Eco- Industry value added comprises mining, manufacturing, nomic Co-operation and Development data �les. construction, electricity, water, and gas (ISIC divisions 10–45). Services value added correspond to ISIC divisions 50–99. Table 4. Trade, aid, and �nance Household �nal consumption expenditure (private con- Merchandise exports show the f.o.b. value of goods provided sumption in previous editions) is the market value of all to the rest of the world valued in U.S. dollars. goods and services, including durable products (such as cars, Merchandise imports show the c.i.f. value of goods (the washing machines, and home computers), purchased by cost of the goods including insurance and freight) purchased households. It excludes purchases of dwellings but includes from the rest of the world valued in U.S. dollars. Data on mer- imputed rent for owner-occupied dwellings. It also includes chandise trade come from the World Trade Organization payments and fees to governments to obtain permits and (WTO) in its annual report. licenses. Here, household consumption expenditure includes Manufactured exports comprise the commodities in Stan- the expenditures of nonpro�t institutions serving house- dard Industrial Trade Classi�cation (SITC) sections 5 (chemi- holds, even when reported separately by the country. In prac- cals), 6 (basic manufactures), 7 (machinery and transport tice, household consumption expenditure may include any equipment), and 8 (miscellaneous manufactured goods), statistical discrepancy in the use of resources relative to the excluding division 68. supply of resources. High technology exports are products with high R&D General government �nal consumption expenditure intensity. They include high-technology products such as in (general government consumption in previous editions) aerospace, computers, pharmaceuticals, scienti�c instruments, includes all government current expenditures for purchases of and electrical machinery. goods and services (including compensation of employees). It Current account balance is the sum of net exports of goods also includes most expenditures on national defense and secu- and services, net income, and net current transfers. rity, but excludes government military expenditures that are Net private capital flows consist of private debt and non- part of government capital formation. debt flows. Private debt flows include commercial bank lend- Gross capital formation (gross domestic investment in ing, bonds, and other private credits; nondebt private flows are previous editions) consists of outlays on additions to the �xed foreign direct investment and portfolio equity investment. assets of the economy plus net changes in the level of invento- Foreign direct investment is net inflows of investment to ries and valuables. Fixed assets include land improvements acquire a lasting management interest (10 percent or more of (fences, ditches, drains, and so on); plant, machinery, and voting stock) in an enterprise operating in an economy other equipment purchases; and the construction of buildings, than that of the investor. It is the sum of equity capital, re- roads, railways, and the like, including commercial and indus- investment of earnings, other long-term capital, and short- trial buildings, of�ces, schools, hospitals, and private dwellings. term capital, as shown in the balance of payments. Data on the Inventories are stocks of goods held by �rms to meet tempo- current account balance, private capital flows, and foreign rary or unexpected fluctuations in production or sales, and direct investment are drawn from the IMF’s Balance of Pay- “work in progress�. According to the 1993 SNA net acquisitions ments Statistics Yearbook and International Financial Statistics. of valuables are also considered capital formation. Of�cial development assistance or of�cial aid from the External balance of goods and services is exports of goods high-income members of the Organisation for Economic Co- and services less imports of goods and services. Trade in goods operation and Development (OECD) are the main source of and services comprise all transactions between residents of a of�cial external �nance for developing countries, but of�cial country and the rest of the world involving a change in owner- development assistance (ODA) is also disbursed by some ship of general merchandise, goods sent for processing and important donor countries that are not members of OECD’s repairs, non-monetary gold, and services. Development Assistance Committee (DAC). DAC has three The GDP implicit deflator reflects changes in prices for all criteria for ODA: it is undertaken by the of�cial sector; it pro- �nal demand categories, such as government consumption, motes economic development or welfare as a main objective; capital formation, and international trade, as well as the main and it is provided on concessional terms, with a grant element component, private �nal consumption. It is derived as the ratio of at least 25 percent on loans. of current to constant price GDP. The GDP deflator may also ODA comprises grants and loans, net of repayments, that be calculated explicitly as a Paasche price index in which the meet the DAC de�nition of ODA and are made to countries weights are the current period quantities of output. and territories in part I of the DAC list of aid recipients. Of�- National accounts indicators for most developing countries cial aid comprises grants and ODA-like loans, net of repay- are collected from national statistical organizations and central ments, to countries and territories in part II of the DAC list of (c) The International Bank for Reconstruction and Development / The World Bank 270 WORLD DEVELOPMENT REPORT 2005 aid recipients. Bilateral grants are transfers in money or in kind variable in the relevant period. The regression equation takes for which no repayment is required. Bilateral loans are loans the form extended by governments or of�cial agencies that have a grant ln Xt = a + bt, element of at least 25 percent and for which repayment is required in convertible currencies or in kind. which is equivalent to the logarithmic transformation of the Total external debt is debt owed to nonresidents repayable compound growth equation, in foreign currency, goods, or services. It is the sum of public, publicly guaranteed, and private non-guaranteed long-term Xt = Xo(1 + r)t. debt, use of IMF credit, and short-term debt. Short-term debt In this equation, X is the variable, t is time, and a = log Xo and includes all debt having an original maturity of one year or less b = ln (1 + r) are the parameters to be estimated. If b* is the and interest in arrears on long-term debt. least-squares estimate of b, the average annual growth rate, r, is Present value of debt is the sum of short-term external obtained as [exp(b*)–1] and is multiplied by 100 to express it debt plus the discounted sum of total debt service payments as a percentage. due on public, publicly guaranteed, and private nonguaranteed The calculated growth rate is an average rate that is repre- long-term external debt over the life of existing loans. sentative of the available observations over the entire period. It The main sources of external debt information are reports does not necessarily match the actual growth rate between any to the World Bank through its Debtor Reporting System from two periods. member countries that have received World Bank loans. Addi- tional information has been drawn from the �les of the World Exponential growth rate Bank and the IMF. Summary tables of the external debt of The growth rate between two points in time for certain demo- developing countries are published annually in the World graphic data, notably labor force and population, is calculated Bank’s Global Development Finance. from the equation Domestic credit provided by banking sector includes all credit to various sectors on a gross basis, with the exception of r = ln (pn/p1)/n, credit to the central government, which is net. The banking sector includes monetary authorities, deposit money banks, where pn and p1 are the last and �rst observations in the period, and other banking institutions for which data are available n is the number of years in the period, and ln is the natural log- (including institutions that do not accept transferable deposits arithm operator. This growth rate is based on a model of con- but do incur such liabilities as time and savings deposits). tinuous, exponential growth between two points in time. It Examples of other banking institutions include savings and does not take into account the intermediate values of the series. mortgage loan institutions and building and loan associations. Note also that the exponential growth rate does not corre- Data are from the IMF’s International Finance Statistics. spond to the annual rate of change measured at a one-year interval which is given by Statistical methods (pn – pn–1)/pn–1. This section describes the calculation of the least-squares growth rate, the exponential (endpoint) growth rate, and the The Gini index World Bank’s Atlas methodology for calculating the conver- The Gini index measures the extent to which the distribution sion factor used to estimate GNI and GNI per capita in U.S. of income (or, in some cases, consumption expenditure) dollars. among individuals or households within an economy deviates from a perfectly equal distribution. A Lorenz curve plots the Least-squares growth rate cumulative percentages of total income received against the Least-squares growth rates are used wherever there is a suf�- cumulative number of recipients, starting with the poorest ciently long time series to permit a reliable calculation. No individual or household. The Gini index measures the area growth rate is calculated if more than half the observations in a between the Lorenz curve and a hypothetical line of absolute period are missing. equality, expressed as a percentage of the maximum area under The least-squares growth rate, r, is estimated by �tting a lin- the line. Thus a Gini index of zero represents perfect equality, ear regression trendline to the logarithmic annual values of the and an index of 100 percent implies perfect inequality. (c) The International Bank for Reconstruction and Development / The World Bank Selected world development indicators 271 World Bank Atlas method The following formulas describe the calculation of the Atlas In calculating GNI and GNI per capita in U.S. dollars for cer- conversion factor for year t : tain operational purposes, the World Bank uses the Atlas con- version factor. The purpose of the Atlas conversion factor is to 1  p ptS $   p ptS $   et* = et − 2  t S$  + et −1  t S$  + et  reduce the impact of exchange rate fluctuations in the cross- 3   pt − 2 pt − 2   pt −1 pt −1    country comparison of national incomes. The Atlas conversion factor for any year is the average of a country’s exchange rate and the calculation of GNI per capita in U.S. dollars for year t: (or alternative conversion factor) for that year and its exchange rates for the two preceding years, adjusted for the difference Y$ * t = (Yt /Nt)/e t between the rate of inflation in the country and that in Japan, where et* is the Atlas conversion factor (national currency to the United Kingdom, the United States, and the Euro Zone. A the U.S. dollar) for year t, et is the average annual exchange rate country’s inflation rate is measured by the change in its GDP (national currency to the U.S. dollar) for year t, pt is the GDP deflator. The inflation rate for Japan, the United Kingdom, the deflator for year t, pt S$ is the SDR deflator in U.S. dollar terms United States, and the Euro Zone, representing international for year t, Yt $ is the Atlas GNI per capita in U.S. dollars in year inflation, is measured by the change in the SDR deflator. (Spe- t, Yt is current GNI (local currency) for year t, and Nt is the cial drawing rights, or SDRs, are the IMF’s unit of account.) midyear population for year t. The SDR deflator is calculated as a weighted average of the these countries’ GDP deflators in SDR terms, the weights being Alternative conversion factors the amount of each country’s currency in one SDR unit. The World Bank systematically assesses the appropriateness of Weights vary over time because both the composition of the of�cial exchange rates as conversion factors. An alternative SDR and the relative exchange rates for each currency change. conversion factor is used when the of�cial exchange rate is The SDR deflator is calculated in SDR terms �rst and then judged to diverge by an exceptionally large margin from the converted to U.S. dollars using the SDR to dollar Atlas conver- rate effectively applied to domestic transactions of foreign cur- sion factor. The Atlas conversion factor is then applied to a rencies and traded products. This applies to only a small num- country’s GNI. The resulting GNI in U.S. dollars is divided by ber of countries, as shown in Primary data documentation the midyear population to derive GNI per capita. table in World Development Indicators 2004. Alternative con- When of�cial exchange rates are deemed to be unreliable or version factors are used in the Atlas methodology and else- unrepresentative of the effective exchange rate during a period, where in the Selected World Development Indicators as single- an alternative estimate of the exchange rate is used in the Atlas year conversion factors. formula (see below). (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank world development report A Better Investment Climate for Everyone Firms and entrepreneurs of all types—from microenterprises to multinationals—play a central role in growth and poverty reduction. Their investment decisions drive job creation, the availability and affordability of goods and services for consumers, and the tax revenues governments can draw on to fund health, education, and other services. The contribution they make to society depends largely on the way governments shape the investment climate in each location—through the protection of property rights, regulation and taxation, strategies for providing infrastructure, and interventions in finance and labor markets. New sources of data from the World Bank highlight how investment climates vary dramatically across, as well as within, countries and underline the potential for improvement. The World Development Report 2005: A Better Investment Climate for Everyone argues that improving the investment climates of their societies should be a top priority for govern- ments. Drawing on surveys of nearly 30,000 firms in 53 developing countries, country case studies, and other new research, the Report explores questions such as: • What are the key features of a good investment climate, and how do they influence growth and poverty? • Why is progress in improving the investment climate often slow and difficult? • What practical lessons can be drawn from country experiences on how to tackle such a broad agenda? • What has been learned about good practice in each of the main areas of the investment climate? • What role might selective interventions and international arrangements play in improv- ing the investment climate? • What can the international community do to help developing countries improve the investment climates of their societies? In addition to detailed chapters examining these and related questions, the Report contains selected data from the World Bank’s new program of Investment Climate Surveys, the Bank’s Doing Business Project, and the World Development Indicators 2004, which is an appendix of economic and social data for over 200 countries. Now in its 27th edition, the World Development Report offers practical insights for policymakers, executives, scholars, and all those with an interest in economic development. ™xHSKIMBy356821zv":;:<:):# ISBN 0-8213-5682-8 (c) The International Bank for Reconstruction and Development / The World Bank