.:� ; f�::: - .. д��А;� � . _: ,��рф� - О Р . . �. .���7 ~,я�ь3' ���,1 . . . _ 0Q' �О \ . _. \iI 6 _� _ �?� � 0 о о уа� °� ._�#; �к� � � лл С' �°о�'�Р � О � �� �о �, , � � QJ�ry ��Р �.���д . �° г ��� ' �'��� � � О J � Pi{,Уr,и-.�-� - � е',i �. �� '� ���� 1ii �л .�Г,F . . . . �; -. iitfi � т � • � ' ' и'-,цf�� . .,:г'1�.�т��. . . �р J`' _ ^ • � - � д�и~ r -" -�йн� `УΡ' \\■ • -+*�� ,.�] •- 1� ! ` I'' � - .�лмУ� l/�!� - J.Э�и�•��k1�� .'$J,-�--� -__-.�... � .��� � - . ;.,,�-3у�р„Ч ___,- � ; х� •rv����if. - . . . :�""5:.".�'.�'�^ ' �� ' м-�i�}�, -.--' �.�i �_� 1 �'iг.Ar. _ -_ � . `чг-�г�� - - _.�°^'�"�_ .°Г �а • ''/' . . .. -,. - �гв+г 1� ' '.л,, '- -� _.--_._ � М ` � �. 0 -�������� � - - _ +,��� У ' tr • ,4'' �� .�;а;_;;F���r, i�`�Е � °.;е. �. � -- -- � �--; м ь�-�.2й1'{� � ' _ �м�� . w � , . л ,,,� °�ri<� � - � - , л', i, r � - .. , �, - _- -:: у '; �.м . -- 3� 1' i . �'У Ь�:F _ � J '_ - I � 1,���: � - 1 mi�� 1и = п'1'гр -- �����N _ ��М Jq1 ' 'j�г�� ., й�iн ' i' и�к `.`� :' � �ti�; i ! ^' i• F ��г�и5н��- ' i - ' �`'ь� � .+С � ! �/+ь�М. а �д -- 1996 Global Economic Prospects AND THE Developing Countries The World Bank Washington, D.C. © 1996 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, N. W., Washington, D.C. 20433 U.S.A. All rights reserved Manufactured in the United States of America First printing April 1996 This report has been prepared by the staff of the World Bank. The judgments expressed do not necessarily reflect the views of the Board of Executive Directors or the governments they represent. ISBN 0-8213-3285-6 ISSN 1014-8906 Library of Congress catalog card number: 91-644001 (serial) Editing, layout, and production by American Writing Corporation This report is the result of work by staff drawn from throughout the International Economics Department. The task manager of the report was Ron Johannes, working under the guidance of Uri Dadush. The core team also comprised Milan Brahmbhatt, Alan Winters, and Takamasa Akiyama. General direction was provided by Masood Ahmed. Contents Abbreviations and data notes iv Foreword v Summary 1 1 Long-run prospects for the growth and integration of developing countries 3 Favorable prospects for the coming decade 4 A differentiated outlook for developing countries 17 Conclusion 19 Notes 19 2 Disparities in global integration 20 Explaining disparities in integration 20 Integration and growth 25 Policy reforms and integration 27 Prospects and risks for weak and slow integrators 29 Notes 31 3 Obstacles and opportunities on the path to trade integration 34 Minimizing the adjustment costs of trade reform 35 Obstacles, both real and imagined, to trade integration 36 Regional agreements as a vehicle for integration 42 Conclusion 45 Notes 46 4 Integration and successful commodity producers 47 Defining success 47 Fostering productivity 50 Conclusion 55 Note 55 Appendixes 1. Regional economic prospects 56 2. The speed of integration index: Qualifications and anomalies 66 3. Robustness of results 73 Global economic indicators 75 Technical notes 88 Classification of economies 89 Bibliography 95 iii Abbreviations and data notes APEC Asia-Pacific Economic Cooperation GSP Generalized System of Preferences ASEAN Association of South East Asian Nations IMF International Monetary Fund CFA Communaute financiere africaine LIBOR London interbank offer rate CPI Consumer price index Mercosur Latin America Southern Cone trade bloc EC European Community (Argentina, Brazil, Paraguay, Uruguay) EU European Union (formerly the EC) MFA Multifiber Arrangement FDI Foreign direct investment MUV Manufactures unit value (index) G-3 Germany, Japan, and the United States NAFTA North American Free Trade Agreement G-5 France, Germany, Japan, United NIES Newly industrialized economies Kingdom, and United States OECD Organization for Economic Cooperation G-7 Canada, France, Germany, Italy, Japan, and Development United Kingdom, and United States OPEC Organization of Petroleum Exporting GAT General Agreement on Tariffs and Countries Trade PPP Purchasing power parity GDP Gross domestic product WTO World Trade Organization Data notes The Classification of Economies tables at the end of The following norms are used throughout: this volume classify economies by income, region, * Billion is 1,000 million. export category, and indebtedness. Unless otherwise * All dollar figures are U.S. dollars. indicated, the term "developing countries" as used in * In general, data for periods through 1994 are this report covers all low- and middle-income coun- actual, data for 1995 are estimated, and data for tries, including the transition economies. 1996 onward are projected. iv Foreword Global Economic Prospects and the Developing Countries is an annual report prepared by the staff of the World Bank's International Economics Department. The series, now in its sixth year, provides an annual assessment of global economic prospects as they affect developing countries and analyzes the links between developing countries and the world economy, particularly in the areas of trade, foreign direct investment and other capital flows, and commodity markets. Last year's report focused on the increasing integration of developing countries with the global economy, highlighting the opportunities and challenges that arise from such integration. This year's report starts from the observation that the par- ticipation of developing countries in the accelerated pace of integration over the past decade has been marked by large disparities. While some developing countries have rapidly expanded their engagements in world trade and capital markets, many oth- ers have not. Some have even become less integrated with the global economy. Integration matters because there is an association between growth performance and integration. Fast growth tends to promote a more open economy because of policies common to both, and hence lagging integration is a sign of underlying pol- icy deficiencies. Additionally, integration can be a means to higher growth through the channels of better resource allocation, greater competition, transfer of technol- ogy, and access to foreign savings. Many of the countries that are lagging in global integration are among the poorest. Developing countries that have lagged behind in opening to world trade are often deterred by perceptions of trade barriers. But some of these fears are exag- gerated. For example, tariffs are lower now than at any time in the postwar era. Neither does being a commodity exporter preclude improvements in productivity and diversification: many commodity exporting countries have been among the high-growth, fast integrators. Concerted action on the part of the international community also is needed to remove some of the real obstacles to trade liberalization that are external; for exam- ple, the spread of antidumping practices, agricultural protection, and, for the time being, the Multifiber Arrangement. It is important to ensure continuing efforts to open international markets in these areas. Globalization and its impact on developing countries raise a number of policy and research issues that we are now beginning to understand more clearly. This report represents part of the ongoing work of the International Economics Department to address these issues. Michael Bruno Senior Vice President, Development Economics and Chief Economist The World Bank V Summary The pace of global economic integration-the widen- policies that countries can adopt to become more pro- ing and intensifying of international linkages in trade ductive participants in the world economy are the and finance-has accelerated over the past decade. main themes of this year's report: During 1985-94 the ratio of world trade to GDP rose three times faster than in the preceding ten years and * A review of prospects for the world economy nearly twice as fast as in the 1960s, the celebrated suggests that the next ten years will likely see an "golden age" of rapid world growth and trade expan- acceleration in the pace of international integra- sion. Over the same period foreign direct investment tion-even compared with the rapid pace of the doubled as a share of global GDP, while the share going past ten years. Real interest rates will be moder- to developing countries rose to more than a third. ate, and growth of world trade is likely to exceed The pace of integration among developing coun- 6 percent a year in volume terms, faster than at tries has been very uneven, however. Though in the any time since the 1960s. The international eco- aggregate developing countries kept pace with the nomic environment is not devoid of risks, but it world rate of trade integration, the ratio of trade to appears to have become more stable, and this GDP fell in forty-four of ninety-three developing will help developing countries that have countries in the past ten years. The distribution of for- embarked on outward-oriented reforms. In the eign direct investment was also skewed: eight devel- aggregate, then, the growth of developing coun- oping countries accounted for two-thirds of foreign tries is likely to accelerate. Still, if current poli- direct inflows in 1990-93, while half of all developing cies continue, large differences in performance countries received little or none. among developing countries will persist. As argued in last year's Global Economic Prospects, Countries with bad policies will tend to lag in globalization is underpinned by liberalization of eco- integration and are likely to see only a mild rise nomic policies and by technological advances that in per capita incomes, with the income gap continue to facilitate transport and communications favoring strong integrators continuing to widen. networks. Increased participation in the world econ- * Disparities in the level and speed of integration, omy carries important benefits: improved resource as measured by such indicators as ratios of trade allocation, heightened competition as a spur to and foreign direct investment to GDP, are closely achieving world standards of efficiency, wider associated with differences in growth rates. The options for consumers, the ability to tap international quarter of developing countries that integrated capital markets, and exposure to new ideas, technolo- most quickly over the past decade grew nearly gies, and products. But globalization also demands 3 percentage points faster than the slowest inte- more of governments. As firms in developing coun- grating quarter. Many low-income countries are tries face more intense international competition their among the least integrated, and some became needs for clear rules, a stable environment, access to even more marginalized during this period. But imports, efficient infrastructure, and freedom from other low-income countries-including some of red tape increases. And greater reliance on private the largest-were among the fastest integrators. sources of finance makes it essential to retain the con- The quality of policies is critical to the speed of fidence of international capital markets. both integration and growth, and there also Not all countries have risen to the challenges of appears to be a mutually reinforcing relation- globalization, and this year's report focuses on devel- ship between the two. Policy reforms designed oping countries that are lagging in integration. The to increase an economy's growth and stability causes and implications of laggingintegration and the are likely to influence a country's speed of inte- 1 gration both directly and through their effect on policies that enhance economic stability and growth. Examples include policies that ensure flexibility. macroeconomic stability and realistic exchange * How can commodity-reliant countries enhance rates, and investments in telecommunications productivity and diversify exports? Many slow- and transport infrastructure. growing, lagging integrators rely on primary commodities for exports, but so do (or did) The report also addresses two questions often many countries that are now fast-growing, asked by policymakers in countries that aim to pro- strong integrators. In fact, successful commod- mote growth by opening to the global economy: ity-reliant countries outperformed manufac- tures-exporting developing countries in terms of * Will trade liberalization work? Though some both growth and integration over the past ten important external obstacles remain on the path years. The successful countries followed policies to increased exports from developing coun- that enhanced their ability to invest and compete tries-such as the proliferation of antidumping in international markets in primary commodi- initiations-trade barriers facing developing ties, and not just in manufactures, notably by countries are lower now than at any time in post- strengthening the private sector, encouraging war history. They are much lower now, for foreign direct investment, promoting the appli- example, than during the 1960s, when countries cation of research and technology, and develop- in East Asia began penetrating world markets. ing physical infrastructure. In fact, internal obstacles to export growth, such as excessive transport costs, are sometimes as In summary, the experience of fast-integrating important as external ones. And while there are developing countries provides many powerful, prac- genuine transition costs to trade liberalization, tical lessons in how countries can expand their experience shows that they can be managed. engagement with the world economy in order to Trading arrangements between developing and enhance growth. Countries that are lagging in inte- industrial countries, one avenue of liberaliza- gration will likely face a favorable external environ- tion, can bring important benefits to developing ment in which to undertake reforms, and these countries, securing markets, encouraging reforms have a large role to play in furthering growth. investment, and enhancing the credibility of For the lagging countries, managing the transition reforms by establishing precise timetables. But toward increased integration will be a difficult but such arrangements are not a panacea, and they crucial challenge. require the speedy adoption of complementary 2 Long-run prospects for the growth and integration of developing countries Despite recessions and modest growth in industrial Saharan Africa and the Middle East and North Africa, countries the global economic environment for devel- on the other hand, grew the slowest and advanced oping countries was relatively favorable over the past their integration the least. Experience in Europe and five years. World trade growth averaged 6.1 percent Central Asia, where most countries are in transition a year in 1991-95, more than 4 percentage points toward a market economy, was mixed. Many of these higher than the growth of world output. Private cap- countries accomplished a remarkable reorientation of ital flows to developing countries quadrupled. Infla- exports toward Western (primarily European) mar- tion in the OECD countries (as measured by the kets and attracted significant capital inflows. The consumer price index) remained under 3 percent, real high-income countries, which typically are the most short-term interest rates in the Group of Seven (G-7) integrated with world markets, achieved per capita countries averaged under 2 percent, and real non-oil annual income growth of 1.2 percent, faster than most commodity prices stabilized (and even rose) after a developing countries. decade of decline. Last year's Global Economic Prospects (World Bank Yet marked disparities in growth persisted among 1995b) argued that globalization-the increased inte- and within developing regions. The extent to which gration of world markets for goods, services, and cap- countries benefited from the increased integration of ital-would affect the growth prospects of developing world goods and capital markets was also highly countries by expanding opportunities and intensify- uneven (table 1-1). The fastest-growing regions over ing competition in international markets. Global- the past five years-East and South Asia, Latin ization, consequently, places a high premium on America-also showed the largest advances in inte- sound policies. This year's report focuses on the large gration with the world economy, as measured by the number of developing countries that are lagging size of capital inflows and the growth of exports. Sub- behind. The causes and implications of lagging inte- Bet tcr- ij tegr a ted ecot nOMics tend to achtiev e lighller xrot'tlh Table 1-1 Growth and integration, 1991-95 (percent) Other prizate Real GDP growth Export growth FDI inflows capital flows per capita, per capita, as a share of as a s/hare of Region 1991-95 1991-95 GDP, 1993-95 GDP, 1993-95 East Asia 8.0 14.1 3.1 2.5 South Asia 2.2 8.4 0.3 1.2 High-income countries 1.2 5.0 0.6 0.4 Latin America and the Caribbean 1.1 7.2 1.1 2.0 Middle East and North Africa -0.2 0.4 0.4 0.3 Sub-Saharan Africa -1.5 -1.6 0.9 0.1 Europe and Central Asia -7.7 1.0 1.4 2.1 Source- World Bank data and staff estimates. 3 gration and the policies that countries can adopt to improve the environment for foreign trade and become fuller participants in the world economy are investment. Countries where policies stand still can the main themes of this year's report. fall behind. The rapid pace of world trade growth for a second In reviewing the prospects for growth and integra- successive year in 1995 underscores some important tion of developing countries, this chapter pays special implications of globalization for policies in develop- attention to the external factors that drive the ing countries. Globalization of production processes prospects of regions or countries lagging in integra- increases the need for fast, efficient, and reliable com- tion. The theme of lagging integration is picked up munications and transport infrastructure in develop- again in chapter 2, which examines the disparities in ing countries. Just-in-time management of input global integration, identifies their causes and implica- inventories requires a leaner regulatory framework, tions, and reviews the prospects of lagging integrators including faster customs clearance procedures. A sig- in greater detail. nificant share of world trade is intrafirm and is stim- ulated by foreign direct investment as firms seek to Favorable prospects for the coming decade reduce production costs and tap unexploited domes- tic markets. Developments in the world economy since the start of Thus, to be fully effective, measures that liberalize 1995 have been largely benign, with somewhat lower trade need to be accompanied by complementary poli- than expected G-7 growth accompanied by reduced cies on foreign direct investment. The importance of pressure on already low inflation rates, unusually fast foreign direct investment as a vehicle for integration is growth of world trade, and low nominal and real reflected by the fact that multinational corporations sell interest rates. Reflecting these developments and such more goods through foreign subsidiaries than they policy factors as improved prospects for fiscal consol- export from their home countries. Provision of long- idation in industrial countries, our projections envis- distance commercial services (such as order process- age a favorable international economic environment ing, invoice clearance, and software and database for developing countries in aggregate, although some development) is also growing fast, thanks to computer will benefit much more than others (table 1-2). and telecommunications networks such as the Internet. Compared with the preceding decade, factors favor- Though still in their infancy, long-distance services ing higher growth (besides the recovery of the transi- promise to be a large new segment of world trade. tion economies of Central and Eastern Europe and the In light of these considerations, and the broader former Soviet Union) include the continued liberal- trends discussed in last year's report, this report ization and fast growth of world trade and the lower argues that measures that enhance the integration of level of real interest rates compared with those pre- lagging developing countries are likely to pay high vailing during much of the 1980s. There are, however, dividends and are necessary to avoid the marginaliza- downside risks of slow OECD growth, to which the tion of these countries in the world economy. Growth poorest countries are especially vulnerable. and increased integration are mutually reinforcing, Key features of the long-term projections include: and both depend critically on the quality of policies * The increased probability of a more stretched-out and institutions. Increased participation in interna- industrial country recovery, with growth contin- tional trade improves resource allocation, enhances uing at modest but more steady rates and con- efficiency by increasing competition among firms, and tributing to a more stable external environment induces learning and technology transfer, thus facili- for developing countries. tating growth. And, as the East Asian countries show, * Inflation and world real interest rates over the next faster output growth also tends to project countries decade that are likely to be lower than previously onto regional and world markets. The higher returns expected as a result of firm monetary policies, associated with faster growth increase firms' demand declining inflationary expectations, and con- for imported capital goods. As incomes rise, con- tinued fiscal consolidation in industrial countries. sumers demand more variety, as reflected in higher * Private capital flows to developing countries that imports of consumer products. And faster-growing were unexpectedly resilient in the aftermath of countries tend to attract more foreign investment, the Mexico crisis. Factors such as moderate real The uneven distribution of foreign investment international interest rates, continued liberaliza- across developing countries and the wide differences tion in developing countries, and portfolio diver- in domestic investment rates suggest that, in a more sification in industrial countries are likely to tightly knit world economy, the quality of policies support continued significant growth in private helps determine growth prospects. Other things flows over the coming decade. But fiscal con- being equal, investment flows toward countries that straints and other pressures are unlikely to allow exhibit greater stability and that adopt policies to much room for growth in official aid flows. 4 Dezeloping countries shoild bc'nefit from strong trade growth aM11imod(ratt' intcrest rates over the next i'ecade Table 1-2 Global conditions affecting growth in developing countries, 1974-2005 (average annual percentage change except for LIBOR) Last Current year's forecast forecast Indicator 1974-80 1981-90 7992-93 1994 1995 1996-2005 1995-2004b Real GDP in G-7 countries 3.0 3.1 1.4 2.9 2.3 2.8 2.7 Inflation in G-7 countriesc 10.0 4.3 3.2 2.2 2.1 2.6 2.7 World traded 4.8 4.2 4.1 9.5 8.9 6.3 6.2 Nominal LIBOR (six months, US$) 9.5 10.0 4.6 5.1 6.0 6.1 6.6 Real six-month LIBORC 0.2 5.2 1.0 2.4 3.3 3.3 3.6 Price indices (US$) G-5 export unit value of manufactures (MUV)Y 11.6 3.3 2.1 3.6 4.5 2.5 2.3 Oil prices9 26.7 -5.3 -11.5 -9.0 3.5 -1.4 1.3 Non-oil commodity prices -1.5 -5.4 -4.8 17.8 4.7 -1.6 -1.1 a. Estimated. b. Taken from Global Economic Prospects 1995. c. Consumer price index in local currency, aggregated using 1988-90 GDP weights. d. Average of merchandise export and import volumes. e. Deflated by U.S. consumer price index. f. Data for G-5 countries (France, Germany, Japan, the United Kingdom, and the United States) weighted by their exports of manufactures to developing countries. g. Based on World Bank indices and deflated by the export price of manufactures. Source: World Bank data and staff estimates. * The current boom in world trade, which is in 1981-90 and 5.0 percent in 1991-94 to 5.4 percent expected to be followed by a decade of strong over the next ten years (table 1-3). The transition growth in trade volumes-averaging a little over economies-the countries of Central and Eastern 6 percent a year-including significant growth in Europe and the former Soviet Union-are expected to trade between developing countries. These consolidate their recovery and, indeed, to achieve developments will be supported by implemen- higher growth than all other developing regions tation of the Uruguay Round and other trade lib- except East and South Asia. Though all developing eralization initiatives, continued declines in regions are projected to match or exceed their growth transport and communications costs, and performance of the past decade, the disparities among growth in foreign direct investment, them will remain large, with Sub-Saharan Africa and * The cyclical boom in non-oil commodity prices the Middle East and North Africa registering only faded during the course of 1995. Over the long modest increases in per capita incomes. term commodity prices are expected to continue These are trend projections, and cyclical fluctua- to decline in real terms, though a more gradual tions are bound to occur, though the likelihood of pro- pace of decline than in the past decade is nounced boom-bust cycles is low in the foreseeable expected to contribute to a more stable environ- future. However, there are industrial country sources ment for commodity-reliant developing coun- of longer-term risk, discussed below. These include tries. Real oil prices are projected to decline by problems induced by aging populations in industrial more than 1 percent a year on average over the countries, financial sector weaknesses in Japan, and next decade-rather than rising by the same the persistence of high unemployment in Europe. amount, as anticipated in last year's report- implying heightened pressure for adjustment Industrial countyil outlook: A stretched-out, and reforms in major oil exporters. low-inflation recovery This broadly favorable external environment and the expectation of sustained reform underpin the pro- With few exceptions growth in the major industrial jected acceleration of growth in developing countries countries was lower in 1995 than in 1994 (figure 1-1). (excluding the transition economies) from 3.4 percent An easing in U.S. growth had been widely expected 5 Aggregate developing country, grozwth should be i,narkedlyi higlr than in the past fifteen i1ears Table 1-3 World growth summary, 1966-2005 (annual percentage change in real GDP) Forecasts Region 1966-73 1974-80 1981-90 1991-94 1995a 1996-97 1996-2005 World total 5.1 3.4 3.1 1.5 2.8 3.1 3.5 High-income countries 4.8 3.0 3.2 1.7 2.5 2.6 2.9 OECD countries 4.7 2.9 3.1 1.6 2.4 2.6 2.8 Non-OECD countries 8.8 7.0 5.2 6.2 5.5 5.6 5.5 Developing countries 6.9 5.3 3.0 1.0 3.9 4.8 5.3 East Asia 7.9 7.1 7.9 9.4 9.2 8.2 7.9 South Asia 3.7 4.0 5.7 3.9 5.5 5.5 5.4 Sub-Saharan Africa 4.7 3.5 1.7 0.7 3.8 3.7 3.8 Latin America and the Caribbean 6.4 4.8 1.7 3.6 0.9 2.6 3.8 Europe and Central Asia 6.9 6.1 2.1 -9.0 -0.7 3.0 4.3 Middle East and North Africa 8.6 4.9 0.8 2.4 2.5 3.2 2.9 Mcnorandun itens Eastern Europe and the former Soviet Union 7.0 5.1 1.8 -9.4 -2.5 2.9 4.4 Developing countries excluding Eastern Europe and the former Soviet Union 6.2 5.0 3.4 5.0 4.9 5.1 5.4 Note: GDP measured at market prices and expressed in 1987 prices and exchange rates. Growth rates over historical intervals are computed using least squares regression. a. Estimated. Source: OECD national accounts statistics; World Bank data, staff estimates, and projections. as a result of tightening monetary policy over the course of 1994, but the failure of recovery to take hold in Japan and the slowing in Europe, which had only industrial country growth moderated in 1995 recently emerged from recession, was not anticipated. The outcome, a slippage in overall G-7 growth to Figure 1-1 G-3 and G-7 real GDP growth, 1983-95 2.3 percent, was in keeping with the moderate, low amplitude character of the overall industrial country cycle. Annual G-7 growth never fell below 1 percent 10 United in the 1991-93 recession and reached a recovery peak States of only 2.9 percent in 1994. Such patterns are in sharp 8 contrast to the cycles of the mid-1970s and early 1980s, Japan Germany when growth was negative at recession troughs but 6 approached 5 percent at subsequent recovery peaks. 4 Trends and their causes. In the United States growth is projected to continue within a sustainable 2 2.2-2.7 percent range. This positive outlook is based partly on the strengthening consensus among U.S. G-7 policymakers to move to a balanced budget over the next seven to ten years and partly on the large drop in U.S. long-term interest rates in 1995. In Europe, despite the unexpected slowdown, developments in 1995 such as significant monetary easing should 19 enhance the prospects for a moderate revival in growth. In Japan growth remained stalled for the Source: IMF data. fourth year in succession. A substantial easing in mon- 6 etary and fiscal conditions during the second half of Consistent with moderate growth, average G-7 1995 improved the likelihood of recovery from 1996 inflation (as measured by the consumer price index) on, though financial system debt may contribute to a edged down from 2.2 percent in 1994 to 2.1 percent in more muted and drawn-out pace than would other- 1995 (see table 1-2). Long-term interest rates fell wise be expected (box 1-1). With aggregate G-7 sharply as growth and inflationary expectations growth near most estimates of its long-run potential eased and as the likelihood of cutting industrial coun- rate, inflation under control, and real interest rates at try fiscal deficits improved. Average real yields on moderate levels, the most likely medium-term out- G-7 ten-year bonds, which had risen by more than look is for neither boom nor bust but for an extended 220 basis points in the year to the fourth quarter of period of moderate, low-inflation growth. Smooth 1994, fell back 150 basis points over the subsequent growth will provide a stable environment for contin- four quarters to 3.9 percent, 50 basis points less than ued reforms and international integration by the the 1986-95 average. Short-term rates also fell in developing countries. Germany, Japan, and the United States and, by the Box 1-1 Japan's financial sector problems will affect developing countries Because it accounts for about 20 percent of foreign aid to from which to confront the problems associated with a developing countries and 10 percent of their exports, rapidly aging population. Japan's economic health is important for low- and middle- The impact of Japan's fiscal strain on developing coun- income countries. The collapse of land and other asset tries is likely to be most serious for indebted primary prices in Japan since 1991 contributed to an increase in the commodity exporters. Japanese demand for these prod- bad debts of financial institutions to Y42 trillion ($420 bil- ucts is adversely affected by slow income growth, and lion), or 9 percent of GDP, by September 1995. The associ- these countries benefit least from the demand for ated reduction in the capital base of financial institutions imported manufactures induced by a rising yen. Lower has slowed credit extension and been an important factor growth and higher interest rates in Japan will also con- in the country's long, drawn-out recession. tinue to put pressure on commodity prices. Low-income The concentration of bad debts in smaller financial insti- countries are also likely to be affected by fiscal pressures tutions (regional banks, credit unions, mortgage institu- on Japanese foreign aid. On the other hand many Asian tions) has increased their exposure to bank runs, such as exporters of low-cost manufactures to Japan will remain those affecting several institutions in the Tokyo-Kansai the least affected. Higher long-term interest rates will area in the second half of 1995. The possibility of a sys- raise the debt servicing costs of indebted countries, while temic financial sector crisis is, however, remote. Taking a stronger yen puts pressure on countries with a large account of debt already written off and assuming recov- share of yen-denominated debt. ery of 40 percent of collateral, about V18 trillion of bad debt (4 percent of GDP) remains to be disposed of. Given that bank profitability is supported by low interest rates and Japan's financial markets have been experiencing wide lending spreads, it is estimated that, at the current negative growth pace of writeoffs, it will take another two or three years for financial institutions to eliminate remaining bad debts. The authorities will likely have to restructure severely C a i d s s r f affected institutions, resulting in higher fiscal expendi- 1985-93 tures. Studies of the U.S. savings and loan crisis, for Percent example, suggest that cumulative losses (including mis- 200 allocation of capital and the costs of the bailout) could run as high as 8 percent of GDP between 1982 and 2007 150 Total (Manchester and McKibbin 1994). By the late 1980s U.S. real long-term interest rates are estimated to have been 100 pushed to a peak of about 50 basis points higher than would otherwise have been the case, and the U.S. dollar 50 . Land was 1.5 percent stronger than otherwise. Given the scale of the problem it seems likely that economic costs in 0 Japan will also be large. Growth only averaged a little more than 1 percent a year in 1991-95, with real long- term interest rates averaging a little below U.S. rates and -100 the real value of the yen surging during this time. The 1985 1987 1989 1991 1993 costs of restructuring will add to the structural deficit built up in efforts to combat the recession. These costs Source: Japan Economic Planning Agency data. will add to the problem of a less favorable starting point 7 turn of the year, appeared poised to fall in several 1.5 percent of world GDP, down from 7.0 percent in the other countries as well. early 1980s, even a significant jump in current prices Looking forward, real G-7 growth is expected to would have modest effects relative to the two previous average 2.8 percent a year in 1996-2005, slightly above shocks, which each represented about 3 percent of last year's projections, while inflation in these coun- world GDP. In addition, a sustained increase in oil tries is expected to average 2.6 percent, a tenth of a prices beyond $25 a barrel is considered improbable. point less than in last year's projections. This year's Second, industrial countries have maintained tight lower inflation projections reflect significant output fiscal and monetary policies. Concerns about debt, gaps (the gap between actual output and that which aging populations, and the effect of taxes on incen- could be produced if all resources were fully tives have supported a growing emphasis on fiscal employed) in Japan and some European countries, consolidation. These efforts have helped lower infla- more restrained fiscal and monetary policies in the tionary expectations, and inflation fell through much approach to European monetary union, and increased of the 1980s and into the 1990s. High inflation, often competition arising from international trade. Progress caused by the monetization of fiscal deficits, increases on fiscal consolidation is expected to support lower economic instability, which is sometimes further exac- real interest rates, with real six-month LIBOR expected erbated by the contractionary policies needed to curb to average 3.3 percent in 1996-2005, three-tenths of a inflation. In recent years tight monetary policies have percentage point less than was projected last year. been used to preempt the build-up of inflationary Rigidities in labor and product markets are an pressures early in the recovery process. important element affecting long-run performance in Third, integration through world trade rose industrial countries. Unemployment in Western quickly over the past decade. Higher levels and Europe in particular remains high, at 10.8 percent in speeds of integration across countries are closely asso- 1995, slightly below the 1994 peak of 11.2 percent. ciated with less volatile growth and inflation (see According to the OECD (1995a) high structural unem- chapter 2). The volatility of world trade was also ployment is caused by such labor market rigidities as lower in 1991-93, when continued import growth in restrictions on work time, high nonwage labor costs, developing countries helped mitigate recession in barriers to entry for new firms, and product market industrial countries. A recent analysis suggests that rigidities that limit competition and demand. High the high observed negative correlation between infla- structural unemployment in industrial countries has tion and trade integration may reflect both greater several adverse implications for developing countries competitive pressures and the higher costs of infla- and is an important source of risk in the long-run tionary policies and exchange rate instability in more outlook. It reduces potential output and worsens open economies (Romer 1993). Since 1970 the volatil- inflation performance. It bolsters demands for unsus- ity of output was less than that of domestic demand tainably expansive macroeconomic policies that ulti- in all but ten countries, underscoring the stabilizing mately destabilize global conditions. And it generates effect of trade. pressures for more protectionist trade policies and Fourth, financial integration has increased. more restrictive immigration policies. Countries with sound policies gain access to interna- tional capital markets and benefit from the procyclical An increasingly stable international economic environ- behavior of international interest rates and the possi- ment. Rates of world output growth and inflation have bility of borrowing to adjust to internal and external become more stable in recent years, with important shocks. The international experience following the benefits for developing countries that are increasingly Mexican crisis suggests that the effect of external open to foreign trade and investment (figure 1-2). The financial shocks greatly depends on the soundness of business cycle of the early 1990s was mild compared domestic macroeconomic policies. Countries with with those associated with the oil shocks of the 1970s realistic exchange rates and low budget and current and the debt crisis of the 1980s. G-7 growth, inflation, account deficits were hardly affected by the crisis. and interest rates deviated less from their long-term The outlook for greater stability is also improved by averages in both the up and down phases of the cycle- long-run structural trends whose influence was over- one reason developing country output (excluding the shadowed by the inflationary policies of the late 1960s former Soviet Union and Eastern Europe) grew by and the oil shocks of the 1970s (Zarnowitz 1989). Most more than 4 percent a year even during the recession. countries' share in output of services, a sector whose Four main factors account for the mildness of the growth is not particularly volatile, is rising. For exam- current world business cycle. First, the 1991-93 reces- ple, the share of services in output in the United States sion did not involve supply shocks from oil or other rose from 51 percent to 58 percent (in real terms) commodity markets as large as those in the 1974-75 between 1977 and 1991, while the volatility of services and 1980-82 recessions. Since oil demand is now only growth, at 2 percent,' was less than half that for 8 The global economy is becoming less volatile Figure 1-2 Measures of stability in the global economic environment (Average annual growth) Low- and middle-income countries' GDP G-7 GDP Percent 25-year Percent 5 Recession average 5 Recovery 4 4 Recovery 25-year 3 3 average 3 3 22 0- o Current Average of two previous Current Average of two previous business cycle business cycles business cycle business cycles World exports G-7 inflation Percent Percent 10 Recovery 12 8 10 * 25-year 25-year average 8 average 6 Recession,/ 4 4 Recession 2 2Recovery 0 0- Current Average of two previous Current Average of two previous business cycle business cycles business cycle business cycles US Treasury bills, three-month real annual average yield Real nonfuel commodity prices Percent Percent 2.0, 25-year 20 Recession Recovery / 15 1.5 10 Recovery 1.0 5 5-year 0- 0.5 -5 Recession 0.0 -10-- _ Current Average of two previous Current Average of two previous business cycle business cycles business cycle business cycles Note: Current cycle: recession=1990-93; recovery=1993-94. Previous cycles: recession=1973-75 and 1979-82; recovery=1976-77 and 1983-84. Source: World Bank data and staff estimates. 9 primary products, manufacturing, or utilities and con- tinued fiscal consolidation in industrial countries con- struction. On the other hand inventories, a volatile ele- tributed to these declines. ment of aggregate demand, have halved as a share of Current projections are for real six-month U.S. dol- U.S. GNP over the past ten years. "Automatic" stabi- lar LIBOR to average 3.3 percent and for real ten-year lizers have also become more important in the postwar U.S. bond yields to approach 4 percent over the com- period. Although business cycles have not been abol- ing decade. At these levels real interest rates would be ished, recent experience suggests that the external envi- roughly halfway between the low, recession-affected ronment of developing countries may be more stable levels of 1991-95 and the higher levels in the pre- than it has been in many years. Developing countries recession 1986-90 period. The projections are, how- that integrate most rapidly with the global economy ever, critically dependent on continued progress in while pursuing sound policies will likely score the fiscal consolidation over the coming decade. greatest gains from these favorable developments. G-7 government deficits are projected to fall below 2 percent of GDP by the end of the decade, which Better prospects for lower real interest rates would stabilize gross debt at 75 percent of GDP. Deficits in the G-7 rose from about 1 percent of GDP The dramatic fall in long-term interest rates in many in 1989 to more than 4 percent in 1993 before falling to countries was one of the most important develop- a still-high 3.3 percent in 1995. Deficits range from ments of 1995, suggesting the potential for an 1.6 percent of GDP in the United States to 7.5 percent extended period of low-inflation growth. In the year in Italy. Gross government debt as a share of G-7 GDP to the fourth quarter of 1995, ten-year bond yields has risen 15 percentage points, from 58 percent in 1989 dropped 200 basis points in the United States, 130 to 73 percent in 1995. basis points in Japan, and from 50-100 basis points in Looming over these issues are the longer-term fis- the major European countries (figure 1-3). Broadly cal problems associated with aging populations in the speaking, the declines reversed between half and industrial countries. In most OECD countries these three-quarters of the big jump in rates that had pressures will accelerate over the next thirty years as occurred in 1994. As noted above, reduced inflation- the baby boom generation retires from the labor force. ary expectations and increased confidence about con- The ratio of the elderly, who tend to be low savers or dissavers, to the working population in the G-7 coun- tries (except the United Kingdom) is projected to Lowe'r bond yilds hold ptntial for low-inlfation growth almost double by 2030-40, peaking at 60 percent in France, Germany, and Japan and at 70 percent in Italy Figure 1-3 Real ten-year government bond yields (in contrast to still-low ratios in most developing in G-3 countries, 1981-95 countries). This development will tend to reduce Percent OECD savings rates by an average of I percentage point and significantly exacerbate pressures on gov- 10 Untd S ernment finances. Public pensions account for about 10 percent of national income and 20 percent of gov- ernment spending in most industrial countries. The present value of future net pension liabilities in the Germany G-7 countries, however, is more than 130 percent of 6 GDP (OECD 1995b and Mitchell 1993). This prospect underscores the importance of achieving early progress on fiscal consolidation. For 4 many countries a 1 percent of GDP improvement in the fiscal balance in 2000 will translate into 40-50 percent of GDP lower net debt by 2030. Conversely, fiscal slip- 2J page will have a negative impact on interest rates, especially when the longer-run effects of aging popu- lation on deficits are taken into account. OECD (1995b) 0 n _____ iMr calculations suggest that maintaining the U.S. budget 1981 1983 1985 1987 1989 1991 1993 1995 deficit at current levels would result in a 50 basis point Q1 QI Ql Q1 QI Q1 Q1 Q1 increase in real long-term interest rates. Such a rise in G-7 real rates would eventually reduce G-7 output by Note: Calculated as nominal yields less consumer price index a quarter to one-half percent a year. The effect would inflton ver pevious four quarters. be transmitted to the developing countries through lower exports, commodity prices, and private capital 10 flows, and by higher debt service payments, resulting capital inflows matters: both the balance between for- in 0.15 percent lower annual growth in low- and mid- eign direct investment and portfolio equity flows, and dle-income countries over a ten-year period. The the extent of short-term borrowing; Mexico depended highly indebted and primary commodity-reliant too much on "hot money" portfolio equity and short- countries, such as those in Sub-Saharan Africa, would term borrowing. be most severely affected. The impact of substantially These considerations suggest that occasional set- higher fiscal deficits resulting from aging populations backs to aggregate private flows may-and probably would, of course, be correspondingly larger. will-occur in the future. But a number of factors sug- gest that private flows, while not achieving the dra- Private capital flows resilient, official aid limited matic growth of the early 1990s, are likely to achieve continued modest growth. One is the projection of Private capital flows. The most striking aspect of moderate industrial country real interest rates dis- aggregate private flows to developing countries in the cussed earlier. Growth in foreign direct investment will aftermath of the December 1994 Mexican crisis has likely be underpinned by structural factors such as been their resilience. These flows reached an esti- falling transport and communications costs, the rise of mated $167 billion in 1995, up from $159 billion in regional trading arrangements, and continued eco- 1994 and $154 billion in 1993. Thus after a dramatic, nomic reform in developing countries (including almost fourfold rise between 1990 and 1993, private liberalization of investment regimes and continued flows have stabilized in the $160-170 billion range privatization).2 The ongoing international diversifica- despite such shocks as the rise in U.S. interest rates tion of investment portfolios in industrial countries is during 1994 and the Mexican crisis (World Bank 1996). likely to continue propelling growth in portfolio flows. Foreign direct investment rose 13 percent to a Investment funds dedicated to emerging markets grew record estimated $90 billion in 1995, commercial bank from 232 in 1990, with assets of $14 billion, to more than lending also increased over the preceding year, bond 1,000 by mid-1995, with assets of $123 billion.3 issuance was flat, and only portfolio equity invest- Developing countries' access to private flows has ment fell significantly, from $35 billion to $22 billion. always been highly differentiated. The Mexico crisis Yet equity investment in stock markets is the most increased this differentiation. Today about twenty volatile and fickle type of investment, having multi- developing countries have access to the private mar- plied some tenfold between 1990 and 1993, then kets for bonds, commercial bank loans, and portfolio halved from 1993 to 1995. equity; the rest are shut out by a lack of creditworthi- The Mexican crisis was also not a harbinger of a ness. Among those with market access borrowing generalized crisis comparable, for example, to the terms vary widely. For example, in 1995 new bond 1980s debt crisis. Contagion effects on other emerg- issues by East Asian borrowers averaged a maturity ing markets proved transient, as shown by money of nearly twelve years with a spread of 116 basis market rates, bond yields, and stock market prices in points over comparable government benchmark May 1995, just six months after the onset of the prob- issues. Latin American borrowers could only obtain lem. By then stock prices in Chile and a number of three-year maturities, for which they paid a spread of East Asian markets had already exceeded precrisis 336 basis points (World Bank 1996). For countries that levels. Stock prices in Argentina and Brazil were have been slow to integrate with global markets, the lower but could be explained by weaknesses in bank- message is clear: isolationism carries high costs in ing systems and preexisting bearish market trends, terms of forgone finance, inhibited investment, and rather than pure contagion. expensive and often misallocated domestic credit. The Mexican crisis offered several lessons for coun- tries pursuing greater integration in world capital Foreign direct investment. Foreign direct investment markets. One is that countries trying to establish low falls on the borderline between a financial flow and a inflation credentials cannot adhere to nominal ex- real linkage, like trade. Sales of foreign affiliates of change rate pegs indefinitely without regard to cur- multinational corporations exceed global exports of rent account deterioration. That is, exchange rate goods and services (table 1-4). Their total worldwide management needs to be flexible. Another is that if sales of $5.3 trillion in 1992 (the latest year available) capital inflows fuel a consumption boom dispropor- compare with worldwide exports of goods and non- tionate to investment, then warning signals should factor services of $4.6 trillion. The sales of foreign affil- flash. Consumption in Mexico grew faster than invest- iates located in developing countries totaled $1.3 ment, to the detriment of growth. A third is that capi- trillion, compared with exports to developing coun- tal account liberalization is risky if banking systems tries of $1.0 trillion. are inadequately capitalized, ineffectively regulated, Part of the reason foreign direct investment in and poorly supervised. Finally, the composition of developing countries has grown so fast-and is 11 ing in most industrial countries. Second, the end of the Sales of foreign-owned affiliates of multinational firms Cold War has reduced (though far from eliminated) exceed global exports of goods and services the substantial military and strategic rationale for offi- cial assistance and made it subject to the greater skep- Table 1-4 Sales of foreign affiliates of ticism in recent years about the effectiveness of multinational corporations, 1992 government action in general. Moreover, the breakup (billions of LIS dollars) of the Soviet Union and the collapse of socialism in -__-_-_-_-_-_-_-_ Eastern Europe eliminated a major source of aid for Origin Amount some developing countries while creating a new Worldwide sales of foreign affiliates 5,325 source of demand for it (figure 1-4). Third, in recent Located in developing countries 1,331 years relief aid for humanitarian disasters such as Located in industrial countries 3,994 those in Rwanda and Bosnia may have diverted funds Global exports of goods and services 4,570 from development purposes. Finally, the effectiveness Source: World Bank staff estimates, based on UNCTAD data. of aid as administered over the past thirty years is under considerable scrutiny. The implication of these trends for low-income expected to continue to grow rapidly-is the revolu- countries is that they will become increasingly less tion in developing country investment regimes in the able to rely on development flows and will have to 1990s, embodying the shift in attitudes from sover- develop a more welcoming environment for private eign discretion to the free flow of foreign investment. financing. The effectiveness of official development Half of all investment codes and bilateral treaties on assistance will become an increasingly important cri- investment were drawn up in the 1990s,4 as were terion in its allocation. Thus developing countries will many important multilateral instruments, including have to meet even tougher policy requirements in the Uruguay Round's Trade-Related Investment order to receive aid. For donor countries, improving Measures, Trade-Related Investment Practices, and the coordination of aid delivery at the country and the General Agreement on Trade in Services; World local levels will be essential, so as to minimize the inef- Bank guidelines and Asia-Pacific Economic Cooper- ficiencies that have sometimes plagued assistance ation nonbinding principles; and regional agreements efforts in the past. bearing on foreign investment (North American Free Trade Agreement, Lome IV, Mercosur). Countries that Continued fast growth in world trade have followed open admission policies include many of those in Central and Eastern Europe, the former The rebound in world trade that commenced in 1994 Soviet Union, and Latin America. Nearly all these continued in 1995, with trade volumes growing at laws, treaties, and instruments espouse substantial about 9 percent, just below the 9.5 percent rate of liberalization, especially with respect to scope, admis- 1994. Over the longer term, world trade growth is sion, and the settlement of disputes. projected to average 6.3 percent a year during 1996-2005, about the same pace as achieved during Aid flows. The widely differing prospects for capital the second half of the 1980s. With world output grow- flows by country are brought into focus by the poor ing 3.5 percent a year over the coming decade, global prospects for aid flows. In 1994 official development trade integration (as measured by the ratio of trade to assistance constituted a third of net resource flows to GDP) is expected to rise 2-3 percent a year, continu- low- and middle-income countries (excluding transi- ing the rapid pace experienced in the past decade (fig- tion economies), and nearly two-thirds of net resource ure 1-5 and table 1-5). flows to low-income countries. For many of the poor- Industrial country import growth, which rose est developing countries official development assis- through 1993 to hit 10.5 percent in 1994, decelerated tance represents virtually the only source of external to a still-healthy 7.6 percent in 1995. Slower economic financing and accounts for a significant share of growth in the United States and the European Union income. In Mozambique in the early 1990s, for exam- contributed most to the easing. By contrast, import ple, it equaled two-thirds of GDP. Yet in recent years aid growth remained firm in Japan, where the strong flows have been falling, from 0.35 percent of donors' yen, rising imports of manufactured goods from off- GDP in fiscal 1983 to 0.29 percent in 1994, the lowest shore production sites, and continued market open- level in more than twenty years. Recent cuts have been ing and deregulation outweighed the effects of the almost universal, with just four of twenty-one donors lingering recession. reporting program increases in current dollars. International trade activity in developing countries These trends are likely to continue for several rea- was even more buoyant. Overall import growth sons. First, the need for fiscal consolidation is press- among the low- and middle-income countries rose an 12 Traiisition economies have tiglteized demand for official aid flows Figure 1-4 Developing regions' share in total official development assistance, 1970-93 Percent Percent 40 45 35Middle East and 40 Sub-Saharan Africa 30 North Africa 35 30 25 25 20 South Asia 20 15 Latin America and 15 10 the Caribbean East Asia Europe an10 Central Asia 5 OL0 1970 1973 1976 1979 1982 1985 1988 1991 1970 1973 1976 1979 1982 1985 1988 1991 Source: World Bank data and staff estimates. estimated 10 percent despite a sharp downturn in Latin America. Import demand continued to boom in Steady global output growth should facilitate integration East and South Asia and accelerated in the transition economies of Europe and Central Asia. Developing Figure 1-5 World exports/GDP and GDP, 1971-97 country export growth also jumped, by about 4.5 per- Perceit centage points, to nearly 13 percent. After an ebullient expansion averaging nearly 12 Forecast 16 percent a year in 1991-93, Latin American imports Exports fell an estimated 3 percent in 1995, largely as a result 10E of the downturn in private capital flows to Mexico and Argentina and the macroeconomic adjustments this 8 downturn required. The region's exports boomed, however, rising 10 percent in 1995 as exchange rates 6 were depreciated and lower domestic demand growth freed resources for exports. The main excep- tion to the regional pattern was Brazil, where sub- stantial real exchange rate appreciation and a consumption boom contributed to a surge in imports 2 and a drop in exports. The longer-run prospects for Latin American trade are promising thanks to the 0G extensive liberalization of the past decade: trade vol- umes are projected to average growth of nearly 6 per- -2 cent during 1996-2005,4 percentage points more than 1971 1974 1977 1980 1983 1986 1989 1992 1995 in the 1980s, with export growth outrunning import growth as the region continues to move toward more Source: World Bank data and staff estimates. sustainable external financial positions. Trade activity in East and South Asia continued to boom due to strong domestic demand and increased boom of recent years in South Asia's international capital goods imports to support fast-growing export trade activities continued, with both exports and sectors. East Asian economies were among the princi- imports rising 10-15 percent. Trade volumes also rose pal beneficiaries of the competitiveness boost resulting rapidly in the early reforming economies of Eastern from the yen's appreciation in the first part of 1995. The Europe as economic recovery took hold. Imports of 13 World trade should experience buotlanti growth Table 1-5 World merchandise trade, 1981-2005 (percent) Indicator/region 1981-90 1991-93 1994-95 1996-2005" World trade growthb 4.1 4.0 9.2 6.3 World output growth 3.1 1.4 2.9 3.5 Export growth biy region High-income countries 4.8 3.7 8.7 6.0 OECD countries 4.6 2.9 8.1 5.8 Non-OECD countries 7.8 11.2 13.0 7.5 Low- and middle-income countries 2.4 6.6 10.6 7.5 East Asia 9.3 14.1 17.4 10.2 South Asia 5.6 10.8 10.1 7.2 Sub-Saharan Africa 0.0 0.5 2.6 4.8 Latin America and the Caribbean 4.4 9.3 8.8 6.1 Europe and Central Asia -0.5 -3.8 10.2 5.1 Middle East and North Africa -1.6 4.6 0.6 4.1 lniport growth biy region High-income countries 5.0 3.0 9.4 5.8 OECD countries 4.8 1.8 9.0 5.5 Non-OECD countries 7.3 14.3 12.0 7.4 Low- and middle-income countries 1.4 7.1 9.0 8.2 East Asia 7.0 14.6 16.7 10.7 South Asia 4.2 5.5 14.9 8.1 Sub-Saharan Africa -3.7 2.0 2.5 5.5 Latin America and the Caribbean -0.3 16.3 3.9 5.8 Europe and Central Asia 0.7 -5.2 5.4 5.9 Middle East and North Africa -0.8 3.1 -0.1 5.9 a. Estimated. b. Growth rate of the sum of merchandise export and import volumes. Soiirce: World Bank data and staff projections. capital goods grew quickly as increasing financial Latin America, South Asia, Eastern and Central inflows relaxed foreign exchange constraints. Associa- Europe, and East Asia; changes in Sub-Saharan Africa tion agreements with the European Union contributed and the Middle East and North Africa were relatively to export growth. Import demand in Sub-Saharan negligible. Strong trade growth will also be supported Africa also picked up as the higher level of commod- by falling communications costs and, despite a pause ity prices relative to the early 1990s improved export in early 1995, the long-run trend toward rising private earnings and as recovery continued in South Africa. capital flows to developing countries. Trade between Continued growth in industrial countries and more developing countries has grown fast, showing that realistic exchange rates in the CommunautO financiere rising developing country exports do not necessarily africaine (CFA) zone and other countries contributed encroach on industrial country markets (box 1-2). to a pickup in African export growth. The long-term forecast of buoyant world trade Coinmodity prices: Flattening in real terms growth averaging 6.3 percent a year is underpinned by a number of factors, including the continuing trend The World Bank's index of non-oil commodity prices toward widespread trade liberalization, regional rose by close to 5 percent in real terms in 1995 after an trade arrangements, and unilateral liberalization on 18 percent surge in 1994. In many cases, however, the part of many developing countries. The median prices were slipping over the course of 1995 so that, unweighted average tariff rate in a sample of forty- for the most part, the commodity price boom that eight developing countries, for example, fell by 9 per- started in 1993 appears to be over, having been pri- centage points (or 34 percent) between the mid-1980s marily the result of transitory supply disruptions and and early 1990s, with the largest drops occurring in normal cyclical forces. The projections are for the 14 overall non-oil price index to fall by more than 15 per- 1995. Price increases in 1995 were led by grains, a sig- cent in real terms over 1996-97, returning about two- nificant import item for many low-income countries, thirds of price gains seen in 1994-95, and to remain notably those in Africa. broadly flat in the medium term, in contrast to the Wheat, rice, and corn experienced particularly major declines of the past 15 years (figure 1-6). steep price increases in 1995. The main factors behind Prices are expected to be restrained in the longer grain price increases were rising demand, supported term by rising agricultural productivity in many by the world recovery, poor harvests in the United countries, low income elasticities for most staple food States and the former Soviet Union, and cuts in price items, and the declining material intensity of eco- supports and subsidies in the United States and nomic activity in industrial countries. Thus commod- Europe. Stocks are expected to fall to 13 percent of ity-exporting developing countries cannot look to consumption by the end of the 1995/96 crop year, the significant price increases as a sustainable source of lowest reserve on record. The market is expected to export revenue growth. On the other hand flat prices remain tight, with high and volatile prices through will be an improvement on the protracted and large 1996. The possibility of further supply constraints price declines suffered during the 1980s and early brought about by poor weather, for example, cannot 1990s and will provide a stable environment in which be ruled out, making this potentially the most to pursue long-term economic reforms. volatile situation in the past fifteen years. Thus the In some respects movements in individual com- current situation is not viewed as the beginning of a modity prices in 1995 were less favorable to develop- period of sustained real price increases, but rather as ing countries than in 1994. Prices for beverages, which a short-term response to the current imbalance are important for many small commodity exporters in between supply and demand. If prices were to Africa and Latin America-and which rose by more remain high, production by the major exporters than 70 percent in real terms in 1994-turned down in could increase significantly. And prices may begin to Box 1-2 Developing country regional trade: A bonus of global trade integration The export pessimism that led many developing coun- of the former Soviet Union, the former Yugoslavia, and the tries to enter into the inward-looking and largely unsuc- Czech and Slovak Republics. Trade complementarities for cessful "South-South" regional trading arrangements of the region are not shown for the same reason. The region the 1970s and early 1980s has been widely replaced by is extraordinarily heterogeneous, comprising countries outward-oriented trade policies and a resurgence in from Western Europe (Greece, Turkey), former members global trade integration. Ironically, it is in this liberalized of the Council for Mutual Economic Assistance from trading environment that intra-developing country trade Eastern Europe (Hungary, Poland, the Czech Republic), has flourished, averaging 12 percent annual growth in and the former Soviet republics. Intraregional trade col- nominal dollar terms during 1985-94, even more buoyant lapsed following the end of Council for Mutual Economic than the 10 percent annual growth in developing country Assistance but has since, in Eastern European countries, trade with OECD countries during this period. The wave undergone a sharp reorientation toward the European of trade liberalization in developing countries over the Union (see chapter 3). The complementarity of intrare- past decade-often unilateral-has led to a gradual shift- gional trade-the extent to which a country's exports ing of productive resources toward areas of international match trading partners' import needs-rose markedly comparative advantage. in Latin America, which underwent substantial trade The share of intra-developing country trade in world liberalization over the past decade, and in East Asia, trade increased from 5 percent in 1970 to 13 percent in where trade protection was also significantly reduced 1994 thanks to manufactures, whose share in this trade over this period. rose from 56 percent to 75 percent over the past decade. Other factors affecting the growth of developing coun- Less than 10 percentage points of this shift can be attrib- try intraregional trade include the dynamism of regional uted to relative price changes favoring manufactures. growth and access to external financing. The dramatic The figure shows that trade within the major develop- growth of East Asia's intraregional trade was supported ing country regions in 1985-93 grew most rapidly in East by rapid regional economic growth and large capital Asia and Latin America (average annual rates of 22 per- inflows. In Sub-Saharan Africa low intraregional trade cent and 16 percent). Starting from a small base, it has also complementarities, scarce financing, and relatively stag- grown in South Asia and Sub-Saharan Africa (average nant output growth kept intraregional trade growth low. annual rates of 9 percent and 8 percent). There was no In Latin America the resurgence of capital flows in the growth in intraregional transactions in the Middle East early 1990s allowed higher import levels and thus con- and North Africa during this time. Growth in intrare- tributed to the growth of intraregional trade. gional trade is not shown for Europe and Central Asia because of the lack of reliable trade data for the republics Box continues on next page. 15 Box 1-2 (continued) Trade among developing regions has grown rapidly Developing country intraregional trade and the factors affecting it, 1985-93 Intraregional trade growth Regional real output growth Percent Percent 25 10 8 20 6 15 4 10 2 5 -2 0 -4 Sub- South East Middle East Latin Sub- South East Middle Europe Latin Saharan Asia Asia and North America Saharan Asia Asia East and and America Africa Africa and the Africa North Central and the Caribbean Africa Asia Caribbean Accumulated net private resource inflows Complementarities for intraregional trade US$ billions Index 250 1.0 200 0.8 150 0.6 1993 1985 100 0.4 50 0.2 0 0.0 Sub- South East Middle Europe Latin Sub-Saharan South East Middle East Latin Saharan Asia Asia East and and America Africa Asia Asia and North America Africa North Central and the Africa and the Africa Asia Caribbean Caribbean Source: World Bank data and staff estimates. fall if productivity increases continue to outstrip and minerals prices have stayed on an upward trend, demand. led by copper and aluminum. These prices are likely Prices for agricultural raw materials like cotton and to remain especially firm because of a lack of new rubber rose substantially in 1994 and early 1995 investment in recent years. because of supply scarcities but were relatively flat for Although oil prices recovered somewhat from their the rest of 1995. Rubber prices are likely to remain 1994 lows to an average of $17 a barrel in 1995, real above their long-term trends for the next three years prices remained at their lowest level in more than or so because of a lack of new tree plantings. Metals twenty years. Prices continue to be depressed by sur- 16 plus capacity and by OPEC's inability to increase mar- ket share. In recent years global demand increases Coininodity prices will remain steady have tended to be captured by non-OPEC supplies, particularly from the North Sea and Latin America- Figure 1-6 Real commodity prices, 1960-2005 a situation expected to continue for the next few years. (deflated by the MUV index) The growth in supply has been fostered by improve- In (1990=100) ments in oil exploration and production technologies, increased competition and liberalization in the energy 600 sector, privatization of public oil companies, and a more welcoming stance toward multinational oil 500 companies on the part of such countries as Algeria, Iran, and Venezuela (Streifel 1995). Additional supply may come from resumed Iraqi exports at some point 400 in the future and from an increase in exports by coun- tries of the former Soviet Union. The cumulative 300 Food Nonfuel impact of these bearish supply influences is expected to cause real oil prices to fall significantly in 1996-97, 200 with 1997 prices 12 percent lower than 1995 levels in real terms, and then to remain relatively flat through 2005, implying a 1.4 percent annual average decline. 100 as Oil A differentiated outlook for developing countries 0 1L1 unamir amur _ __ 1mrnI' 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 In some ways the prospective external environment for developing countries summarized in preceding MUV is manufactures unit value. sections may be among the most favorable in many Source: World Bank data and staff projections. decades. Positive factors include moderate low-infla- tion growth in industrial countries, real interest rates substantially lower than during the 1980s, buoyant opportunities offered by an expanding but increas- growth in world trade, and a strong upward trend in ingly competitive global economy. A number of cur- private capital flows to developing countries. In this rent and prospective trends in the external environment developing country GDP (excluding the environment, however, suggest that growth prospects transition economics) is projected to average 5.4 per- for countries that are lagging in integration will be cent annual growth during 1996-2005, up from modest at best (see chapter 2). These trends include a 4.4 percent in 1986-95, or 5.3 percent (including the continuing decline in the share of primary commodi- transition economies), up from 2.1 percent over the ties in world trade, growing and intense competition past decade. In addition to the positive influence of a in labor-intensive manufactures markets from low- better external environment, the projected improve- income countries like China and India, growing con- ment in aggregate developing country growth also straints on foreign aid flows and the complementary reflects the swing to positive growth in the transition need to attract private foreign financing and invest- economies and the rising share of the fast-growing ment (or, indeed, to avert large-scale flight of domes- East Asian region in developing country output. tic capital and skilled labor), and growing constraints There are two important qualifications to this on the growth of worker remittances. favorable baseline scenario. First, it will depend on Much of the estimated acceleration in low- and mid- continued significant policy adjustments in industrial dIe-income countries' growth, from 2.5 percent in 1994 countries. Key assumptions here include substantial to around 4 percent in 1995, derived from improve- reductions in fiscal deficits, sustained anti-inflation- ments in the transition economies of Eutrope and Central ary resolve, structural reforms to address high unem- Asia. Growth in these countries rose to 4.8 percent from ployment, especially in Europe, and continued strong just over 2.0 percent in 1994, led by early reformers momentum on global trade liberalization. such as Poland and the Czech Republic. Many of these Second, the buoyant overall projections for low- countries have moved determinedly toward global and middle-income countries mask a wide range of integration, with booming foreign direct investment individual outcomes. The outlook is most favorable inflows and exports contributing significantly to for those developing countries that continue to imple- revival (see chapter 3). In the republics of the former ment macroeconomic and structural reforms and that Soviet Union economic contraction eased to a little develop institutions and infrastructure to exploit the over 5 percent in 1995 from more than 14 percent in 17 1994, led by postreform lows in inflation and budget to 3.5-4.0 percent in 1995 after averaging 0.7 percent a deficits in Russia. Exports to markets outside the for- year in 1991-94. This recovery was due in part to mer Soviet Union have been buoyant, and positive improved commodity prices but also to the effect of growth is expected to emerge from 1996 on. On the reforms undertaken by a number of countries, includ- downside, economic conditions remain difficult in the ing those of the CFA zone. South Africa's recovery con- Central Asian and Caucasus republics, many of which tinued for a second year, and spillover benefits-in share features with weak and lagging integrators. part through increased trade and investment flows- Overall, though, the 4-5 percent long-run growth fore- should begin to accrue to the Southern African area. cast for the region will contribute half of the 3.2 per- But the large number of countries in the Sub-Saharan centage point increase in developing country growth region and the considerable variation in growth rates over the next decade. suggest the need to look behind the regional averages Much of the rest of the improvement in aggregate (figure 1-7). Many countries persisting with structural developing country growth is expected to come from adjustment (such as Ghana, Tanzania, and Uganda) two regions-East and South Asia-where global inte- have recovered sharply from declines during the gration will remain at or come to the forefront of the 1980s. Others (such as Botswana and Mauritius) con- development agenda. East Asian growth eased mar- tinue their strong performance of earlier years. At the ginally to about 9 percent in 1995 as the Chinese econ- opposite pole, countries suffering from civil strife omy, where growth fell from near 12 percent to just (such as Liberia, Rwanda, Somalia, and Zaire) con- over 10 percent, moved toward a soft landing. But tinue to suffer contraction (accurate data are not avail- growth accelerated or continued at high rates in most able for some of these countries). Expectations for an other countries of the region on the basis of rapid improvement in regional growth over the next decade export growth, led by burgeoning intraregional trade, to 3.5-4.0 percent are conditioned on a number of strong domestic demand, and continued large private assumptions. Among these are a continuation in the capital inflows. In South Asia booming exports and a recent pace of reforms, a slower rate of decline in terms continued rise in foreign direct investment inflows of trade, and a consolidation of civil peace in the sub- supported another year of near 6 percent growth in India. But countries in the region will continue to face tough challenges if they are to sustain the rapid pace S11b-Saharan Africa remains a region of contrasts of growth and integration of recent years. These include dealing with infrastructure bottlenecks, fiscal Figure 1-7 Annual GDP growth in Sub-Saharan deficits, military expenditures, and inefficient public Africa, 1992-94 sectors, as well as lowering high trade and investment (pcrcont) barriers and extending the scope of competition in domestic factor and product markets. Eatnn Latin America and the Caribbean was the only devel- ozn_udan Ur,nd. oping region that experienced a substantial slow- 3i. down in 1995-regional growth dropped from near Ma, Mauritania 5 percent in 1994 to less than 1 percent. The slowdown Gainea-B.,,a. was led by recessions in Mexico and Argentina, where VBe"i a sharp downturn in private capital flows in the wake B,na Mah of the Mexican crisis forced a sudden adjustment in cwizl,nd Tanzania domestic consumption and investment to close large oTndna current account deficits built up after several years of C-0 real currency appreciation, low domestic savings, and enra externally financed growth. Regional growth is Burkin Fauo chaI expected to remain sluggish in 1996-97 as a result of a projected (though moderate) adjustment in Brazil, Coe (or where strong current growth is also associated with n real appreciation, a consumption boom, and bur- geoning external deficits. Many countries in the region that underwent macroeconomic adjustment in Bn 1995 have shown a resolve to persist with and even ME strengthen policy reforms. This factor underpins a __-_-- projection of long-term regional growth (2001-05) -2 0 2 4 6 8 10 reviving to just over 4 percent. Soiircc. World Bank data and staff estimates. Sub-Saharan Africa experienced an upturn in growth is regions where breakthroughs have recently been rience only moderate rates of growth. Industrial coun- achieved. Still, given the low projected rate of per try growth will be modest, but inflation and interest capita income growth, the prospects for poverty rates will remain moderate and world trade will grow reduction remain modest. rapidly. Wide disparities will persist in growth rates The Middle East and North Africa grew a sluggish 2.5 across developing countries, and these disparities will percent in 1995 following barely positive growth in be reflected in (and, to some extent, be the reflection 1994, in part reflecting weak oil prices. Severe drought of) the pace at which countries are becoming inte- also reduced output, particularly in Morocco and grated with the global economy. Tunisia. Real oil price declines averaging more than 6 This outlook suggests a number of issues. What are percent a year in 1981-95 were a major factor in aver- the causes and the implications of lagging integra- age annual regional growth of just over 1 percent in tion? Do countries that want to become more inte- 1986-95. As the section on commodities explained, grated with the world economy face insurmountable current projections for oil prices are notably more obstacles in the form of international trade barriers bearish than in last year's report, with an average and the high cost of liberalization? Can commodity- annual drop in real prices of 1.4 percent expected over dependent countries, which tend to lag in growth and the next decade rather than an average annual rise of integration, break out of their vicious cycle? These 2 percent. Regional growth is projected to average 2.9 questions are addressed in chapters 2, 3, and 4. percent a year over the next decade, the lowest among all developing regions, combined with more differen- Notes tiation in performance within the region. Key issues affecting the region's long-run outlook and relative 1. Calculated as the standard deviation of annual performance include progress on structural reforms, growth rates over the period 1970 to 1994. which up to now has been sporadic at best; greater 2. During 1990-95 developing countries attracted nearly integration with the world economy, especially in $40 billion in foreign direct investment to privatization pro- terms of formal arrangements with the European grams. Union; progress in regional peace; and institutional 3. However, extravagant projections about the impact of development to alleviate sociopolitical tensions. portfolio diversification should be tempered by evidence that investors exhibit "home bias," that is, they tend to Conclusion invest less abroad than diversification theory would suggest (Tesar and Werner 1993). Portfolio diversification also These projections paint an aggregate picture of rapid increases short-term volatility, since emerging markets rep- growth and further advances in the integration of resent a small fraction of total investments. developing countries with the world economy over 4. These numbers refer to all legal instruments known to the next ten years. Some countries will prosper; oth- the International Center for the Settlement of Investment ers, including some of the poorest, are likely to expe- Disputes, an arm of the World Bank Group. 19 Disparities in global integration Developing countries as a group have participated growth. Reforms that promote stable macro- extensively in the acceleration of global integra- economic conditions, realistic exchange rates, tion, although some have done much better than and open trade and investment regimes are others. Over the past decade their overall ratio also important for growth and integration. of trade to GDP-a common measure of integra- * Improvements in the external environment tion-rose 1.2 percentage points a year, while their and modest reforms in many lagging integra- share of global foreign direct investment (FDI) rose tors suggest that their growth rates may show to two-fifths. But there are wide disparities in some improvement in the next decade. But if global economic integration across developing current policies and trends persist, many countries. This chapter reviews developing coun- developing countries can expect to fall further tries' widely varying experience with integration behind OECD countries in per capita GDP. over the past ten years and explores the causes and implications of the large disparities. (Box 2-1 con- Explaining disparities in integration siders the experience of global integration from a historical perspective.) The analysis draws four Global economic integration-participation in the main conclusions: international markets for goods, services, capital, * Changes in integration were highly differenti- and labor-can be measured in several different ated. Many developing countries became less ways. One approach is to try to capture the degree integrated with the world economy over the to which domestic prices and interest rates reflect past decade, and a large divide separates the their international counterparts: if markets were least from the most integrated. It is striking, for perfectly integrated, prices would be the same example, that the ratio of trade to GDP fell in everywhere. A measure of the extent to which a forty-four of ninety-three developing coun- country has absorbed the global stock of technolog- tries over the past ten years, while the ratio of ical and other knowledge would also be useful. But FDI to GDP fell in more than a third. in practice such measures are hard to calculate. This * Countries with the highest levels of integra- chapter examines direct measures of integration, tion tended to exhibit the fastest output such as ratios of trade to GDP and FDI to GDP, as growth, as did countries that made the great- well as indirect measures, such as creditworthiness est advances in integration. Many low-income ratings (a measure of access to international capital countries are among the least integrated, how- markets), tariffs (an indicator of disparities between ever, and some became even more marginal- domestic and international prices), and the share of ized during this period, experiencing both manufactures in exports (an imperfect measure of a falling incomes and reduced integration. country's ability to produce at world standards and But other low-income countries-including absorb technical knowledge). There is an element of some of the largest-were among the fastest arbitrariness in this selection. A number of other integrators. measures could also have been used, including * Sound policies play an important role in deter- exchange controls, quantitative restrictions on mining both growth and the speed of integra- imports, or institutional factors such as member- tion. Policy reforms designed to increase an ship in the World Trade Organization. But whatever economy's growth and stability are likely to the measure, most indicators point to the same con- influence a country's speed of integration, clusion: developing countries show wide dispari- both directly and through their effect on ties in both the level and the rate of integration.' 20 Box 2-1 Global integration at the start of the twentieth century From a longer-run perspective the recent increases in exports to GDP in 1913 was 12.9 percent, not much global integration can be viewed as the quickening of a below their 1993 level of 14.5 percent. Tariff levels var- trend that has been evident in the postwar era and that, ied widely across countries, with many countries engag- to some extent, constitutes a recovery from the interwar ing in virtually free trade-that is, low tariffs. The stock slump in trade, foreign investment, and capital flows. But of foreign capital invested in developing countries is these developments are not without historical precedent. estimated to have risen by 3.7 percent a year in real Until 1914 the global economy experienced a rapid terms. Further, international migration was scarcely expansion of private investment, largely in the form of restricted and migration flows were substantial. Stable foreign direct investment and public sector bonds. exchange rates were the rule, with widespread adher- Recipient countries included Argentina, Australia, Brazil, ence to the gold standard. Canada, India, Mexico, Russia, and the United States; Thus the start of the twentieth century was a period of major creditor countries included France, Germany, and considerable global economic integration supported by the United Kingdom. Between 1885 and 1895 the United relatively liberal economic policies. Still, it differed from Kingdom's current account surplus averaged 4.9 percent the 1990s in a number of ways. First, the composition of of GDP. In contrast with the present day, most investment trade is different today, with a lower share of commodity was concentrated on the primary sector (petroleum, min- trade in the total and higher shares of intraindustry trade ing, smelting, agriculture), not manufacturing. and trade in services. The falling share of commodities Export volumes rose by 4.3 percent a year in real terms partly reflects the decline in commodity prices relative to during 1900-13 for thirty-two countries representing manufactures over the course of the century. Second, some four-fifths of world output, population, and gross (as distinct from net) capital flows are very high exports (Maddison 1989). For seventeen industrial today, and come from a wider variety of sources. Third, countries for which data are available, the ratio of migration is more restricted today. Measures of integration 'Trade expanded in mtany regions over the past decade Trade ratios advanced strongly in some regions in the past ten years, though they fell in Sub-Saharan Africa, Figure 2-1 Changes in real trade/GDP, 1960-94 were flat in the Middle East and North Africa, and (percent) barely edged forward in South Asia (figure 2-1).2 -0.6 -0.3 0.0 0.3 0.6 0.9 1.2 1.5 Overall, trade ratios fell in forty-four of ninety-three developing countries, representing more than one bil- World lion people, or 26 percent of the sample population.3 A further seventeen countries experienced only mod- High-income est rises. And although the aggregate ratio of devel- countries oping country trade to GDP rose strongly over the Low- and middle- decade, three-quarters of the increase was accounted income countries for by just ten countries. Changes in ratios of trade to GDP matter because competition in global markets exposes exporters to new technologies, designs, prod- Latin America and ucts, and management techniques and provides the Caribbean access to imports that were previously unavailable or that embody new technologies that can contribute to euroe and productivity gains. Ratios of trade to GDP may also provide indirect measures of competition and price South Asia allocation effects. For example, Coe, Helpman, and Hoffmaister (1995) find that the size of spillovers from Middle East and research and development in industrial countries on North Africa productivity in developing countries rises with devel- Sub-Saharan oping country imports from OECD countries. Such Africa spillovers may account for most of the rise in devel- oping countries' total factor productivity in 1971-90.' Levels of and changes in tariffs also vary widely across developing regions (figure 2-2). Tariff levels Source: World Bank data and staff estimates. matter because they, together with other trade restric- 21 Tariffs 1l17e follen1 inl r-Ceent /eIrS The regioiial distributioni of foreigni direct inz'estitieWt is hkihti skeed Figure 2-2 Average unweighted tariff rates for selected developing regions, 1985-93 Figure 2-3 Foreign direct investment flows/GDP, Percent 1981-93 70 Percent 2.01 60 50 M 1989-89 Li1990-93 1.5 1981-83 1991-93 30 20 10 05 1. South Middl Sb Lain East Central Asia East and Saaa mrica Asia" and0. -oF9 (5) North Africa and the (5) Eastern0. Africa (1l) Caribbean E11rope High- East Europe Latin Middle Sub- South (5) (1)(4) incomne Asia and America East Saharan Asia countries Central and the and Africa Asia Caribbean North Note: Number Of COUntries iii parentheses. Africa a. EXClUdes China. Source: World Bank data and staff estimates. Source: World Bank data and staff estimates. tions, are a main source of distortion between domes- FDI inflows were less than 0.25 percent of GDP in tic and international prices, resulting in resource mis- 1991-93. Regions with particularly low ratios of FDI to allocations and a reduced competitive spur from GDP included South Asia, Sub-Saharan Africa, and the imports.5 High levels of protection in particular can Middle East and North Africa. Over the past decade seriously impair growth. Despite the trend toward ratios of FDI to GDP fell in thirty-seven of the ninety- trade liberalization in developing countries over the three countries studied. Of these, twenty were in Sub- past decade, there are still large differences in the Saharan Africa, nine were in Latin America and the extent of liberalization undertaken and the levels of Caribbean, and seven were in the Middle East and protection remaining. Tariffs in South Asia, averaging North Africa. In many cases these falls reflected a loss around 45 percent in the early 1990s, remain far higher of country creditworthiness due to macroeconomic than in other regions, while those in the Middle East instability, adverse terms of trade shocks, political and North Africa and Sub-Saharan Africa, both in the uncertainties, and civil strife. The large increase in FDI 25-30 percent range, have shown little change since in high-income countries, together with a substantial the second half of the 1980s. However, several transi- rise in their trade ratios, underscores the fact that these tion economies in Central and Eastern Europe have countries, already the most integrated, are continuing achieved average tariffs of about 10 percent, while to deepen their interaction with the world economy. rates in Latin America and East Asia (excluding As with trade, FDI is a significant indicator of integra- China) are down to around 15 percent. Tariffs in tion in part because of its potential for diffusing tech- industrial countries on imports from developing nology and skills (box 2-2)!6 countries are now down to about 4 percent. Two other indicators that shed light on disparities The distribution of FDL across developing countries in integration are country credit ratings and the share is also highly skewed (figure 2-3). Eight countries that of manufactures in exports. Credit ratings are a mea- account for 30) percent of developing country GDP gar- sure of both access to private capital markets and the nered two-thirds of overall FDI flows in 1990-93. For terms of that access; the share of manufactures in half of the ninety-three developing countries reviewed exports measures exposure to international technolo- 22 Box 2-2 The success story of India's Maruti Udyog The 1982 establishment of Maruti Udyog Limited (MUL), laboration with vendors have "changed the market's per- a joint venture between the government of India and ception of design and quality and revolutionized the com- Suzuki Motor Company of Japan, was a watershed in the ponents industry through its philosophy of vendor development of the Indian automobile industry. Since upgradation" (Maxton 1994). M-UL actively nurtured some then, car sales have soared 16 percent a year, with MUL critical component industries by establishing eleven joint selling more units in its first five years of operation than venture companies to help push quality and productivity all other domestic manufacturers combined in the previ- concerns upstream. As a result about 75 percent of com- ous forty years. Productivity levels have reached nearly ponents are now sourced from domestic suppliers, with fifty-three cars per employee per year, compared with the rest manufactured in-house or imported. The local five for the next big Indian producer and twenty-three content of MUL's most popular model is now 95 percent. cars or trucks for General Motors, the biggest producer in The improvements spawned by MUL have supported North America, and are still rising. The company pro- 22 percent annual growth in the components industry duces nearly three-quarters of the passenger cars in India since 1986 and led to the industry achieving $2 billion in and has entered foreign markets, with about 10 percent of output and $300 million in exports in 1995. Though most production exported. exports are replacement parts for older models in indus- MUL's presence also has generated significant spillover trial countries, supplies to original equipment manufac- benefits for Indian industry, particularly the auto compo- turers are on the rise. For instance, Sundram Fasteners, a nents sector. A recent study of Asian auto markets found company in Madras, supplies 85 percent of General that the company's strict quality standards and close col- Motors' need for radiator caps in the U.S. market. gies, although it also reflects the stage of development cally borrow at rates that are 50 basis points or less and factor endowments. above benchmark U.S. rates. Borrowing costs rise as Country credit ratings generated by banks or rat- ratings fall, with many countries with C ratings pay- ing agencies show wide disparities across developing ing 500 or more basis points over the benchmark, countries, with many countries completely shut out while most in the D category have no access to pri- of medium- and long-term private markets.7 vate lending. Highly rated countries can also borrow Institutional Investor's recent ratings for 126 countries more relative to their economic size, and for longer are summarized in table 2-1. For the purposes of this maturities. table the ratings are divided into four equal ranges More than 40 percent of the countries in table 2-1 labeled, from best to worst, A through D. Such ratings were in the lowest rating category. These include half are an important influence on the cost of funds in or more of the countries in Sub-Saharan Africa, Latin international markets. Countries with A ratings typi- America (chiefly in Central America and the Caribbean), and Europe and Central Asia (almost all states of the former Soviet Union). They also include Few developing countries have good access nearly half the countries in the Middle East and North to private capital markets Africa, as well as Bangladesh, the Democratic People's Republic of Korea, Myanmar, Nepal, and Vietnam in Table 2-1 Country credit ratings, March 1995 Asia. The countries in the C category broadly comprise - economies undertaking reforms whose payoffs still lie Credit rating mainly in the future, including Argentina, Brazil, Region A B C D Total Colombia, and Mexico in Latin America, Morocco and Tunisia in the Middle East, India and Pakistan in South High-income countries 19 7 1 27 Asia, Hungary, Poland, and Turkey in Europe, and East Asia 5 2 3 10) Ai,Hnay oad n ukyi uoe n South Asia 3 2 5 Botswana, Ghana, and Mauritius in Africa. While the Latin America highest-rated segment comprises exclusively high- and the Caribbean 1 10 13 24 income countries, the B category contains the most suc- Middle East and cessful developing country exporters (Chile, China, North Africa 4 4 b 14 Republic of Korea, Malaysia, and Thailand), as well as Sub-Saharan Africa 6 19 25 a number of Gulf countries with strong oil collateral. Europe and The share of manufactures in exports may provide Central Asia 3 7 11 21 some information on countries' access to learning and Total 19 20 33 54 126 technology transfer gains and ability to produce at Source: histitutiital Izestor, March 1995. world standards. This ratio varies a great deal: two- thirds of the ninety-three developing countries 23 reviewed had a share of manufactures in exports of a On the basis of this index, developing countries are third or less in 1983-92, while half stood at less than grouped in four categories ranging from "fast inte- 20 percent. Sub-Saharan Africa's share of manufac- grators" (those with the highest index values) to tures in exports was less than 10 percent, while the "slow integrators" (those with the lowest; table 2-2). Middle East and North Africa and Latin America and This classification is not intended to derive a precise the Caribbean had average shares of 20-25 percent categorization of individual countries but rather to (figure 2-4). These regions also experienced the lowest develop evidence about the factors that might account rate of growth in shares of manufactured exports over for large differences in the speed of integration among the past ten years; shares fell in fourteen of thirty-six groups of countries, and the consequences of this for Sub-Saharan African countries. performance. The complexity and variety of factors that affect the pace of integration of individual coun- A summary view of disparities tries require detailed treatment for each country, something beyond the scope of this report. The foregoing discussion suggests that integration There are striking disparities in developing coun- has a number of different dimensions and that a vari- tries' speeds of integration. Most of the fast-growing ety of indicators can be used to evaluate overall East Asian exporters were among the fast integra- trends. To summarize integration trends, the analysis tors as a result of exceptionally large increases in that follows uses a speed of integration index derived trade, manufactures exports, and FDI ratios. This from changes between the early 1980s and early 1990s group also contained reformers such as Argentina, in four of the indicators discussed above: the ratio of Chile, and Mexico in Latin America, Morocco in the real trade to GDP, the ratio of FDI to GDP, Institutional Middle East, Ghana and Mauritius in Sub-Saharan Investor credit ratings, and the share of manufactures Africa, and the Czech Republic, Hungary, Poland, in exports.' The speed of integration index is the sim- and Turkey in Europe. The concentration of transi- ple average of changes in the four indicators over the tion economies in this and the next category of mod- period expressed as standardized scores.10 erate integrators results from the marked rise in these countries' trade shares, FDI inflows, and credit ratings after the fall of communism.]] Mtviiifachitres cxports tire aii iMpOrtafit co)ip1eWt Among the anomalies that an index of this nature f hinevitably generates is that China falls into the group of weak integrators. The main reason is a reduction in its Institutional Investor credit rating, though the Figure 2-4 Average share of manufactures in level of its credit rating remains among the highest merchandise exports, 1973-92 given to developing countries. Since the credit rating Percent component of the index differs from the other three 801 components in being a summary of capital market 1973-82 1983-92 judgments at a point in time (rather than a direct mea- 70 sure of flows), it is of interest to know how the index would appear without the credit rating component. 60 On that basis, the broad analysis and conclusions remain unchanged, but many individual countries 50 fall into a different quartile, notably China, which moves up to near the top of the group of moderate 40 integrators, and India, which rises to the group of fast integrators. Another instance is Saudi Arabia, which was one of the strongest integrators at the start of the 20 1980s. Over the past ten years its speed of integration j index was negative, reflecting falls in FDI consequent 10 on reduced infrastructure needs and a drop in its very 0 ~ ~ high rating in Institutional Investor. Yet Saudi Arabia remained one of the stronger integrators in the 1990s. High- Europe South East Middle Latin Sub- In any event, the purpose of the analysis is not to rank income and Asia Asia East America Saharan countries Central and and the Africa countries but to draw broad conclusions about inte- Asia North Caribbean gration, growth, and policies (see appendix 2). Africa South Asian countries were concentrated in the fast Somrce: World Bank data and staff estimates. or moderate integrator categories but retain consider- able potential for furthering integration. Bangladesh, 24 India, and Pakistan had average tariffs in the early the weak and slow groups.12 There was also a strong 1990s of 50 percent or more. Among other Asian coun- association between growth in 1984-93 and the level tries, China had average tariffs around 40 percent. of integration prevailing at the beginning of the Nontariff barriers in some of these countries were also period (figure 2-5b).'3 high. Similarly, Tunisia (a moderate integrator) had Per capita growth rates based on each of the four average tariffs in the 20-30 percent range. For all coun- constituent integration variables are shown in table tries, lowering tariffs facilitates integration. 2-3. Broadly speaking, the association between The weak and slow integrators in table 2-2 include growth and the composite speed of integration index not only most of the low-income countries in Sub- extends to each of these individual constituents, Saharan Africa but also many middle-income coun- although the correlations are far from perfect. tries in Latin America and the Middle East and North For the most part, only fast-integrating developing Africa. Median trade ratios and credit ratings in Sub- countries saw per capita incomes converge toward Saharan Africa and the Middle East and North Africa industrial country levels over the past decade. In sev- fell, while median FDI ratios either fell or were stag- enty-five of ninety-three developing countries-most nant. The countries in these categories also experi- of them in the slower integration categories-per enced much greater macroeconomic instability than capita GDP growth in 1984-93 was less than the 2.0 other countries (see below). Many suffered substan- percent median of high-income countries, and in fifty tial terms of trade declines over the past decade countries per capita income levels fell over the decade. because of falling primary commodity prices. Finally, Finally, countries with better integration performance many are or have been subject to severe internal or enjoyed not only higher but also more stable growth: external political conflicts and weak institutional the median standard deviation of per capita GDP development. growth among fast integrators was a little over half as high as among the weak and slow integrators.14 Integration and growth A number of interesting contrasts between the speed of integration and growth reflect the influence Using the integration indicators discussed in the pre- of factors other than integration on growth. Many vious section, figure 2-5a documents the empirical transition economies in Central and Eastern Europe association between faster speeds of integration and (Poland, the Czech Republic) integrated rapidly after higher growth. The high-income countries and fast the fall of communism even though output fell integrators among the developing countries achieved sharply during the adjustment to a market system. median per capita GDP growth of about 2 percent a Mexico achieved a high integration speed over the year over the past decade. Moreover, the experience past decade, but a combination of inadequate domes- of the fast integrators was not merely a reflection of tic savings, real exchange rate appreciation, and rising high-growth East Asian countries, although they are external deficits contributed to slowing growth in the important. Excluding East Asian countries, fast inte- early 1990s. grators still achieved median per capita growth of 1.5 It is also important to note that there is a significant percent a year, well above the other classes of integra- positive association between income and integration tors. Median incomes fell among the other groups of levels-that is, poor countries tend to have the lowest integrators, with the largest declines occurring among trade ratios (after adjusting for size), less FDI, lower Dcvcloping cotntrics shozo wvide disparities in their speed of integration Table 2-2 Speed of integration of developing countries, early 1980s to early 1990s (niumber of countries) Latin Middle East Sub- Europe and America and and North Saharan Central Ranking East Asia South Asia the Caribbean Africa Africa Asia Fast integrators 6 3 5 2 2 5 Moderate integrators 2 5 4 10 2 Weak integrators 3 9 2 10 Slow integrators 2 5 14 2 Total 9 5 21 13 36 9 Source: World Bank staff estimates. 25 The faster and higher the integration, the greater the growth Figure 2-5a Median real per capita GDP growth Figure 2-5b Median real per capita GDP growth, and speed of integration, 1984-93 1984-93, and initial level of integration, 1981-83 Percent Percent 4 4 Quartile Q I Third 0 Median Third - First 0 - - -2 -1 First -3 - -4- -3- High- Fast Moderate Weak Slow High- High Moderate Weak Low income integrators integrators integrators integrator m:come integrators integrators integrators integrators countries countries Source: World Bank data and staff estimates. credit ratings, and lower shares of manufactures and learning, improved resource allocation, greater exports. But even though they start out with low levels competition, and access to foreign capital help countries of integration, poor countries are not necessarily slow grow. But this is only one part of the interaction between integrators, just as they are not necessarily doomed to integration and growth. Growth itself tends to promote slow growth. In fact, the association between the speed integration. Imports rise faster than incomes as con- of integration and the level of initial per capita income sumers satisfy their desire for diversity. The rising in 1981-83 is weak. Fast-integrating developing coun- returns to capital associated with faster growth raise tries had a substantially higher speed of integration developing country capital goods imports. Fast-grow- than high-income countries even though their per ing countries attract more FDI and obtain better credit capita income levels were less than a third as high. Some terms. As noted in chapter 1, East Asia most clearly illus- large low-income countries such as India and Indonesia trates how rapid growth tends to project countries onto were among the fast and moderate integrators. And, regional and world markets while lack of growth, as in conversely, high- and middle-income countries did not Sub-Saharan Africa, leads to marginalization in world necessarily achieve a fast rate of integration. goods and capital markets.15 The positive effect of freer trade and foreign invest- The close association between growth and the ment on growth is undoubtedly one of the most critical speed of integration also suggests that both are likely factors explaining the relation between integration and to be affected by a number of common factors, includ- growth. Simply put, factors such as technology transfer ing changes in the external environment,16 the evolu- InidiVidlial indicators of fist rates of integration are associated with high growth Table 2-3 Real per capita GDP growth sorted by individual integration variable, 1984-93 (percent) Developing countries High- Fast Moderate Weak Slow Variable income coutries integrators integrators integrators integrators Composite index 2.03 2.09 -0.40 -1.04 -0.92 Trade/GDP 2.03 -0.36 0.01 -0.12 -0.92 FDI/GDP 2.03 1.15 -0.67 -0.57 -0.84 Country credit rating 2.03 1.52 -0.67 -0.67 -0.36 Manufactures/exports 2.03 1.45 0.63 -0.67 -0.93 Soirce: World Bank staff estimates. 26 tion of the institutional setting,'7 and the policies pur- economy. Macroeconomic volatility deriving in large sued by governments. Policies that are good for part from poor macroeconomic policies is estimated growth are also apt to be good for integration, though to have reduced Latin America's investment rate in some aspects of policy will have particular relevance 1960-85 by 5 percentage points of GDP relative to for integration. The next section focuses on the critical what it would have been had the region experienced role of such policies, the variable that governments the same volatility as industrial countries (Hausman have the most control over, and Gavin 1995). Macroeconomic volatility is esti- mated to have reduced developing country growth in Policy reforms and integration 1960-90 by as much as 0.9 percentage point a year. (Schmidt-Hebbel 1995).8 Volatility is especially Policy reforms designed to increase growth and stabil- likely to discourage foreign investors, who know less ity are likely to influence a country's speed of integra- about the country than domestic investors, have tion both directly and through their effect on growth, greater choice in pursuing alternate opportunities which, as noted above, also helps promote integration. outside the country, and are likely to attach a higher Three types of policies affect the speed of integration risk premium to a more unstable economy.' A relatively quickly: those relating to macroeconomic volatile real exchange rate, one manifestation of policy, trade and FDI regimes, and telecommunications macroeconomic instability, has been found to be a and transport infrastructure. deterrent to investment in the export sector. The Macroeconomic polcy affects integration either inverse empirical relationships between the volatility directly, or indirectly through its effect on growth. of exchange rates (adjusted for inflation) and ratios of Poor macroeconomic policies may affect integration trade to GDP and FDI to GDP (adjusted for popula- directly, for example, through the impact of macro- tion) are shown in figures 2-6a and 2-6b.20 economic instability on FDI, other foreign capital It is empirically the case that countries that suffer inflows, and investment in the export sector of the from higher macroeconomic instability tend to be A volatile exchange rate discoirages trade and foreign investient Figure 2-6a Real exchange rate volatility and Figure 2-6b Real exchange rate volatility and population-adjusted trade/GDP foreign direct investment /GDP, 1991-93 Percent Percent 140 * 6 120 100 80 4 60.mU U 40 0 U 20 M .* U -20 . I ** *s * . .* * mU * -60 -- -- --_- - -1 L 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Real exchange rate volatilitil, 1984-93 Real exchange rate volatility, 1984-93 Source: World Bank data and staff estimates. 27 slower integrators (table 2-4). Median inflation rates quality and process efficiency. Because many foreign were 7-11 percentage points lower in fast than in weak investors operate complex international supply net- and slow integrators in 1984-93. Inflation volatility works, protection and its associated red tape reduces was also much lower among fast integrators. Median a country's attractiveness in these networks. In some real exchange rate volatility among fast integrators developing countries (for example, China, Malaysia, was a third the level in slow integrators. Large fiscal and Hungary in recent years) foreign direct investors deficits are the most important source of macroeco- account for a large share of exports. As tariffs rise, nomic instability.21 Slower integrators tend to have countries tend to have a lower ratio of trade to GDP, higher and more volatile deficits and to shrink their even adjusting for the tendency of big economies to deficits at a slower rate (indeed, the deficits of the have small ratios (figure 2-7).` Using estimates by slowest integrators actually expanded during the Sachs and Warner (1995a) of when countries achieved 1980s). Large fiscal deficits can also affect integration open policy regimes, we find a statistically significant in other ways, particularly when they are externally increase in countries' speed of integration in the five financed, by underpinning real exchange rate appre- years after opening compared with the five years ciation, which impedes integration by favoring non- before it. The increase in speed derives from more tradables over tradables, deterring exporters and rapid gains in ratios of trade to GDP and better credit foreign investors, and encouraging capital flight.22 ratings. Trade liberalization is discussed at greater Investors, both foreign and domestic, demand macro- length in chapter 3. economic stability. A final reason for the association The third policy area affecting the pace of integra- between macroeconomic stability and integration is tion is the availability and proper maintenance of that countries undertaking stabilization after severe adequate economic infrastructure, in particular macroeconomic crises often use the opportunity to telecommunications and transport facilities. High- open their economies because it is easier to gain polit- quality communications are essential for countries ical support to do so. that aim to participate in the globalized production Restrictive trade and FDI policies limit integration by structures established by multinational corporations, restricting imports and inhibiting capital inflows. to respond promptly to rapidly changing market con- Protection also reduces the profitability of exporting ditions in industrial countries, or to participate in new relative to serving the domestic market and blunts export markets for long-distance services such as data incentives to adopt international standards of product processing, software programming, back-office ser- Macrocconiic instability uidermines fast integration Table 2-4 Integration policy and performance, 1984-93 (percent) Developing countries High- income Fast Moderate Weak Slow Performance indicator countries integrators integrators integrators integrators CPI inflation 3.63 13.40 16.86 23.86 19.89 Change in CPI inflation -5.22 -2.81 -0.22 8.21 19.77 CPI inflation volatility 1.65 7.24 7.63 14.21 13.27 Black market premium 0.00 0.12 0.56 0.41 0.48 Real exchange rate volatilitya 0.06 0.13 0.20 0.27 0.40 GDP volatility' 1.89 2.61 3.09 4.39 3.60 Budget balance/GDP -2.46 -2.37 -6.66 -3.70 -5.92 Change in budget deficit 1.16 1.88 0.79 0.38 -2.54 Budget balance volatility' 2.27 2.31 2.82 2.79 4.53 CPI is consumer price index. Note: Contains data for eight-eight countries for which there were adequate tariff data. The balanced data set for this sample allows compar- ison across policy areas such as macroeconomic and trade policy. Values for integration classes are medians. Changes in inflation and bud- get deficit are between averages for 1981-83 and 1991-93. See appendix 3 for details on robustness of results. a. Standard deviation. Source: World Bank data and staff estimates. 28 The forty-seven countries in the group of weak and Higher tariffs mean lower trade ratios slow integrators (hereafter called the slower integra- tors) carry a significant economic weight among devel- Figure 2-7 Average tariff rates and population- oping countries, accounting for about a quarter of GDP adjusted trade/GDP ratios and a third of developing country population.25 Of the forty-seven, twenty-seven were low-income countries, Trade ratio (percent) twenty-one of them in Sub-Saharan Africa. The twenty 80 . middle-income slower integrators were concentrated in Latin America and the Middle East and North Africa. 6* One reason to expect modestly better growth and 60 *integration performance are the partial efforts at eco- nomic reform undertaken by many countries and the 40 likelihood that these efforts will continue. During the 1980s many countries in Sub-Saharan Africa made 20 E progress on reducing trade barriers (especially non- . tariff barriers), foreign exchange controls, exchange * rate overvaluation, and heavy taxation of exporters 0 - (especially farmers), while a smaller number also U. . M " * * made progress on macroeconomic policies (World * N. * " Bank 1994). Many slower integrators in Latin -20 America, the Middle East and North Africa, and the U " transition economies made similar efforts at policy -40 reform. A second reason is that, after an unwinding of * their recent strength in the next year or two, real pri- mary commodity prices are projected to be relatively -60 - - - - - - - - 0 20 40 60 80 flat compared with the 1980s and early 1990s, when sharp declines reduced slower integrators' terms of trade by 2 percent a year and impeded both growth and progress in integration. Still, these factors are insufficient to prevent a vices, and customer support (box 2-3; also see World continued decline in these countries' incomes and inte- Bank 1995b, chapter 3). As chapter 3 argues, high gration levels relative to high-income countries and transport costs deriving in part from domestic policies fast-integrating developing countries. Despite some are a serious impediment to African exports (Amjadi progress, macroeconomic, trade, infrastructure, public and Yeats 1995). sector, and financial sector policies remain far from international standards. Even Africa's best performers Prospects and risks for weak and slow integrators have worse macroeconomic policies than Asia's newly industrialized economies (World Bank 1994). Despite Baseline projections for the next decade indicate a mod- trade reforms most African countries retain high and est improvement in growth relative to the poor out- dispersed tariffs, and little progress has been made in comes of the past decade (figure 2-8).24 Median real per reforming and reducing the size of swollen public sec- capita growth among weak and slow integrators is pro- tors or in strengthening weak financial sectors. And jected to rise to just over 1 percent a year in 1996-2005, while flat real commodity prices might not act as a compared with a -0.5 percent annual contraction in major drag on growth and integration, they are also 1981-95. Median ratios of trade to GDP, which fell over unlikely to provide a boost to performance in the com- the past decade, are expected to rise 0.5 percentage ing decade. In addition, world import demand for pri- point a year in 1996-2005. But in the absence of accel- mary commodities is projected to grow by only 3-4 erated reforms, this performance will be considerably percent a year compared with more than 6 percent for less impressive than that of the fast and moderate inte- merchandise volumes as a whole. grators and high-income countries. One implication of Competition in commodity markets (from efficient these projections is that the absolute gap in per capita producers like Chile and Malaysia) and manufactures incomes between high-income countries and some markets (from low-income exporters like India and three-quarters of developing countries will continue to China) will remain fierce. Given their overwhelming rise. Further, the projected improvement in growth abundance of labor relative to other factors of pro- among the weak and slow integrators is subject to sig- duction and assuming continued reforms, China and nificant uncertainties. India will likely retain their comparative advantage in 29 Box 2-3 Telecommunications growth in developing economies Telecommunications forms an indispensable part of the rate in main lines, which was offset by higher population infrastructure of a modern economy, providing the means growth. The number of main lines also grew fast in Latin to both transmit and process information. As information America (about 8 percent a year) and Asia, notably in the continues to grow in importance as a factor of produc- Republic of Korea, Singapore, and Taiwan (China). tion-one estimate suggests that by 2000 70 percent of One indicator of demand is the waiting period for OECD employment will be information-related-the installation of a telephone connection. In developing demand for telecommunications services is rising fast. countries the average is 1.5 years; again, the figure varies Developing countries have increasingly recognized that a widely across developing countries and in some high-quality telecommunications infrastructure is essen- instances exceeds 10 years (see box figure). With the tial to compete successfully in the global marketplace. advent of cellular radio, however, main lines do not tell Telephone density, defined as the number of main lines the whole story. Globally, cellular lines grew by an aver- per 100 people, varies widely across countries. In high- age of 45 percent in 1990-93, making it one of the highest- income countries telephone density ranges from 40 to 60 growth industries. In many developing countries cellular percent, whereas in developing countries it is less than 20 lines are fast becoming a basic mode of provision, since percent, ranging from 16 percent in Russia to less than 0.2 they can be supplied rapidly to both urban and rural percent in Uganda (box figure). Telephone density is sig- areas. nificantly correlated with per capita GDP (World Bank The level of international telephone tariffs is a primary 1995c). Some countries, however, have achieved much determinant of the balance of telephone traffic between higher telephone density than their per capita output countries. Thus they can significantly influence the might suggest (for example, Egypt and Poland), and degree to which a country can participate in international some much lower (C6te d'Ivoire). business. Some developing countries follow distor- At the end of 1993 the global number of telephones tionary policies, maintaining high collection charges to totaled some 607 million, up more than 150 percent from discourage outgoing calls and hence earn a foreign ten years earlier. In the OECD countries, which represent exchange surplus. For example, the charge for a one- about 15 percent of the world's population, telephone minute telephone call from Tanzania to the United States density rose from 39 percent to 49 percent. Growth in the is about $7.50 equivalent, whereas the charge for the same rest of the world went from 2.1 to 3.5 percent. Among the call in the reverse direction, from the United States to regions, Sub-Saharan Africa had the third highest growth Tanzania, averages about $2. Telephone coverage is still spotty in many developing countries Telephone density and waiting period for selected countries, 1993 Main lines per 100 residents Waiting period Number Years 20 10 >1o >10 15 8 6 10 4 II 2 0 0 labor-intensive manufactures for some time, espe- competition as they try to establish footholds in these cially given these countries' large income disparities markets and to some extent will be forced to focus on between regions and between urban and rural popu- niche manufactures markets that are based on unique lations.26 Slower integrators are likely to face intense or specialized factors or skills. 30 past decade, world trade is expected to grow briskly, Fast integrators will continue to score the greatest gains and private financial flows to developing countries are expected to be significant, albeit selective. Each of these Figure 2-8 Median real per capita GDP growth by assumptions is subject to significant risk, as discussed integration category, 1971-2005 earlier in chapter 1. Percent Still, the baseline projections should not be a cause for despair. The experience of fast-integrating devel- 4, * 1971-80 1981-95 . 1996-2005 oping countries provides powerful, practical lessons for countries that want to expand their engagement with the world economy in order to enhance their 3 growth performance. After all, some of today's fast integrators were yesterday's weak performers. These issues are taken up in chapters 3 and 4. 2 Notes 1 1. Migration of labor, one significant dimension of inte- gration, is not addressed in this report. For a recent exten- sive discussion see World Bank (1995d). 2. The trade ratio is defined as exports plus imports of goods and nonfactor services in real terms as a percentage of real GDP. High-income Fast Moderate Weak Slow 3. The analysis excludes countries with inadequate countries integrators integrators integrators integrators data, including many small (often island) economies and certain current or formerly socialist countries, including Source: World Bank data and staff projections. Cambodia, the Democratic People's Republic of Korea, Vietnam, and the countries emerging from the former Soviet Union. Fast growth in China and India will, of course, also 4. Other recent studies of the role of trade as a channel generate markets for capital- and technology-intensive of technology transfer include Pack and Page (1994). products on the one hand and for land- and natural- 5. Easterly (1993) finds that a one standard deviation resource intensive products on the other. Products increase in distortions of input prices relative to world based on land and natural resources may provide new prices was associated with 1.2 percentage point lower per export markets for the slower integrators if they are capita GDP growth in 1970-85. Levinsohn (1993) and nimble and efficient enough to take advantage of them. Harrison (1994) discuss evidence for the "imports as market Exploiting these opportunities will require upgrading discipline" hypothesis. from raw material exports to higher value added, nat- 6. Borensztein, de Gregorio, and Lee (1995) find that a ural resource-based manufactures and facing up to one percentage point increase in the ratio of FDI to GDP in competition from fast integrators with flourishing developing countries in 1971-89 was associated with 0.4-0.7 primary commodity sectors like Chile, Indonesia, percentage point higher per capita GDP growth. The impact Malaysia, and Thailand. varied positively with educational attainment, an indicator The pressures on slower integrators' export markets of a country's ability to absorb technology. may also be aggravated by constraints on other sources 7. Properties of these ratings and their relationship with of foreign exchange. As chapter 1 noted, the growth of underlying economic performance are discussed in Haque foreign aid, a major source of foreign transfers in many and others (1995). Formal measures of financial integration slower integrators, is likely to slow. Worker remittances include interest rate differentials, savings-investment corre- inflows may also be sluggish as a result of flat demand lations, and consumption tests; see Montiel (1993). in labor-importing oil producers, a shift in demand 8. Sachs and Warner (1995b) find that a 10 percentage toward more highly skilled migrant workers, and point higher share of manufactures in exports is associated tighter immigration controls in many industrial coun- with 0.3 percentage point higher per capita GDP growth tries suffering from high unemployment. (after accounting for other influences on growth such as It is also worth stressing the fragility of even the mod- investment and education). est baseline assessment of the slower integrators' 9. The analysis focuses on changes in rather than on prospects, which assumes a relatively favorable inter- level of integration for two reasons. First, the acceleration of national economic environment. Primary commodity globalization over the past decade makes the question of prices are expected to decline less steeply than in the changes and their trends of inherent interest. Second, mea- 31 sures of the level of integration such as the trade ratio are market premiums, export marketing monopolies, and the heavily influenced by the size of the economy, the abun- existence of a socialist economic system). dance of its natural resources, and other structural factors, 15. In this vein Rodrik (1995) argues that rapid trade which is not the case for the speed of integration. As it turns integration in the Republic of Korea and Taiwan (China) ini- out, the distinction between levels and rates of change may tially resulted from government policies in the 1960s that be less important than it first appears because, as shown in raised returns to capital and generated a domestic invest- figure 2-5a and 2-5b and discussed in appendix 3, the main ment boom, which raised demand for capital goods that nei- conclusions of the analysis hold whether speeds or levels of ther country had a comparative advantage in producing. integration are used. The indicators also refer to integration The resulting faster import growth necessitated faster outcomes rather than to policies affecting integration. export growth, which was facilitated by appropriate macro- Keeping policies and outcomes separate allows for later economic and exchange rate policies. evaluation of the association between them. 16. Slow integrators suffered median terms of trade 10. The standardized score is the variable, less its mean, declines that were more than 2 percentage points a year larger divided by its standard deviation. The resulting standard than the fast integrators in 1984-93. Note, however, that these variable has a mean of zero and a standard deviation of one. differences could account for income losses of perhaps 0.5 This procedure prevents the composite index from being percentage point a year (assuming import to GDP ratios of 20 dominated by constituents with the highest volatility. percent), compared with a gap in GDP growth between the Appendix tables A2.1 and A2.2 present underlying data for two groups of more than 3 percentage points a year. these countries and for twenty-three high-income countries. 17. Survey data indicates that FDI in Sub-Saharan 11. Rising trade shares in these economies were due in African countries, most of which were poor performers, was part to the fall in their output in the early 1990s. But it especially affected by the possibility of extreme events (civil remains true that a larger share of their (smaller) output is war, expropriation), high uncertainty about policies, and the internationally traded and that they are more closely inte- lack of adequate restraint mechanisms on the arbitrary exer- grated with the world economy than before. cise of power. (See Collier 1995a on these and other institu- 12. In a regression for 115 countries of per capita growth tional factors in the marginalization of Africa). Though on the speed of integration index, which is measured in difficult to measure, the quality of institutional develop- standard deviations, the regression coefficient is 2.6 with a t ment is likely to be important for both growth and integra- statistic of more than 6. The correlation coefficient between tion. Institutions promote domestic and foreign trade and the two variables is about 0.5. investment by reducing the transaction costs of economic 13. This association between the level of initial integra- activity, such as the costs of contracting, protecting property tion and growth is consistent with studies finding growth rights, obtaining information, and dealing with uncertainty to be positively associated with open policy regimes (Sachs (North 1991). Foreign investors and traders are likely to be and Warner 1995a; Ben-David 1993). The index for the level especially sensitive to these considerations. of integration is constructed as the simple average of the 18. The effects of uncertainty on investment when four constituent variables (trade-GDP ratio, FDI-GDP ratio, investments are irreversible are analyzed by Dixit and country credit rating, share of manufactures in exports) Pindyck (1994). Leahy and Whited (1995), Findyck and expressed as standardized scores. Developing countries are Solimano (1994), and Fischer (1993) provide empirical evi- then classified as high, moderate, weak, or low integrators. dence for the adverse impact of volatility on investment, Trade ratios are adjusted for the systematic influence of productivity, and output growth. country size by using the residuals from a regression of 19. Thus lower domestic investment is associated with trade ratios on population: bigger countries tend to trade lower FDI (Edwards 1990). Credit ratings, a measure of less proportionately. China, which had a low score on the access to other types of foreign capital, are also adversely speed of integration index, has among the highest scores on affected by elements of macroeconomic instability such as the level of integration index. Appendix table A2.2 presents inflation, external indebtedness, and current account underlying data for the initial level of integration. deficits (Haque and others 1995). 14. Recent evidence supports the contention that coun- 20. Caballero and Corbo (1989) find that higher real tries with more open trade and investment policies also exchange rate volatility is associated with less export grow faster. Sachs and Warner (1995a) find that, after con- growth. Roberts, Sullivan, and Tybout (1995) find a similar trolling for other influences on growth such as education result for a panel of firms from Columbia, Mexico, and and investment, countries with open policy regimes in Morocco and argue that volatility discourages the sunk-cost 1971-89 had average per capita GDP growth that was 2.5 investments needed to export. The strength of the relation- percentage points a year higher than countries with closed ships between tariffs and real exchange rate volatility on the regimes. They were also less prone to macroeconomic crises, one hand and the ratios in figures 2.6a and 2.6b might be defined as rescheduling of foreign debt, arrears on external overstated if there was a strong association between the first payments, or inflation greater than 100 percent. (Open pol- two factors. That is unlikely because the correlation between icy regimes were defined in relation to tariffs, quotas, black the two in this sample is only 0.15. 32 21. Schmidt-Hebbel (1995) finds a strong association population. Lee (1992) also documents the strong inverse between fiscal deficits and a measure of central bank inde- relation of trade restrictions to trade integration. pendence on the one hand and an index of macroeconomic 24. These projections are a subset of the standard pro- instability based on inflation, the volatility of the real jections summarized in chapter 1. exchange rate, the ratio of external debt to GDP, and the 25. To emphasize the importance of the gmup of slower black market premium on the other. integrators this calculation excludes China, which would dom- 22. Macroeconomic stability variables retain a signifi- inate GDP and population totals on its own. See earlier com- cant association with the separate constituent speed of inte- ments and appendix I for more details on the position of China. gration variables in a significant number of cases. 26. China and India's combined labor force of 1.2 billion 23. Population-adjusted trade ratios, which attempt to people is three times that of the slower integrators (exclud- purge the systematic influence of country size on trade ing China), while their ratios of labor to arable land are ratios, are the residuals from a regression of trade ratios on higher than those in 90 percent of the slower integrators. 33 Obstacles and opportunities on the path to trade integration International trade is the main engine driving global face three main issues. First, what are the costs of trade economic integration. In the aftermath of the Uruguay reform? Second, do world markets in fact offer a level Round the volume of trade continues to grow at a playing field and, hence, what is the probability of brisk pace, more than twice that of world output. reform being successful? Third, do regional trading Tariffs are at new lows, averaging just under 4 percent arrangements provide a feasible alternative to multi- in industrial countries. Services, which account for lateral or unilateral liberalization? roughly one-third of world trade (half traded services This chapter addresses these issues, particularly in and half investment and factor incomes), are now sub- light of the experience of successful reformers accu- ject to multilateral disciplines, albeit weak ones. A mulated over the past ten years. The central message start has been made on establishing global intellectual is that the adjustment costs of trade reform can be property rights. And countries in transition, repre- quite low if appropriate complementary policies are senting about a quarter of the world's population adopted, and that while genuine external impedi- (including China), are progressively becoming part of ments to successful liberalization do exist-and are the world economy. Over the past decade developing sizable for some countries-some major obstacles are countries have reduced tariffs, notably in Latin avoidable. The main conclusions are: America (from an average of 27 percent to 15 percent) * Though we understand less than we should and South Asia (from more than 60 percent to about about the transition costs of liberalization, the 45 percent). Besides changing attitudes toward trade case for opening domestic markets to imports policy, an important impetus to growth has come from (without favoring certain trading partners) has the continued fall in the costs of trade, especially com- been strengthened by the record of the past ten munications costs. Thus it is no exaggeration to say years. While there are genuine transition costs to that world trade at the end of the twentieth century is trade liberalization, these costs tend to be over- as healthy as at any time in the postwar era. estimated and can be minimized by appropriate Many developing countries have lagged behind, complementary macroeconomic and regulatory however, in opening to world trade, as discussed in reform. chapter 2. Policymakers in developing countries are * Important and real external obstacles to growth often deterred by perceptions of trade barriers and the in trade remain, including the spread of uncertain benefits and apparently certain costs of lib- antidumping practices, agricultural protection, eralization. This hesitation often feeds back to gradu- and, for the time being, the Multifiber Arrange- alism and timidity in pursuing reform. As a result the ment. These are significant problems for some characteristic features of global integration are also countries, so it is important for OECD countries found in trade: big disparities across developing to continue efforts to open markets in these countries, an associated divergence in growth and liv- areas. In addition, excessive transport costs ing standards, and a set of underlying policy defi- remain a major handicap for some countries, ciencies. This chapter looks behind these features to especially in Sub-Saharan Africa. the issues with which the slow integrators in particu- * Regionalism is a sufficiently diverse and new lar are grappling. phenomenon as to defy simple conclusions. But The gradual reformers are becoming more aware of recent experience-notably, the growth of East lost opportunities and are contemplating measures to Asian trade-suggests that it is neither a neces- liberalize their trade regimes. The debate revolves sary nor a sufficient condition for successful around the desirable breadth and depth of trade integration with either the world or a regional reform and its pace and sequencing. These countries economy. Nevertheless, preferential trading 34 arrangements between developing and indus- region's population; absence of macroeconomic stabil- trial countries can carry important benefits for ity, a further 40 percent; and absence of adequate developing countries by securing markets, resource allocation (a composite measure of trade and encouraging investment, and enhancing the dis- exchange rate regimes, financial sector depth, and cipline and credibility of reform. The benefits of public sector intrusiveness), a further 12 percent. An trade arrangements between developing coun- additional 8 percent of the population in middle- tries tend to be much smaller. income countries is excluded, leaving 29 percent of Sub-Saharan countries with a minimum adequate Minimizing the adjustment costs of trade reform environment necessary for growth. The countries in this category are in fact performing quite well, with Trade experts know less than they would like to about annual growth of about 6 percent, contrasted with neg- the conditions required to secure the benefits of trade ative growth in the other low-income countries. reform and about how to minimize the transition costs. A second concern is that trade reform can prove costly Still, some important lessons have been learned. This in terms of declining industrial sectors and higher unem- section analyzes the main concerns and reviews how ploymient. Though it has proved difficult to isolate and they have been addressed by successful reformers. measure this transition cost-because reforms often Afirst concern is that the supply response to trade reform are accompanied by contractionary stabilization poli- may vary from country to country and in some cases may be cies-the evidence suggests that there is no systematic weak. The trouble with assessing supply response-for connection between reform and unemployment or instance, the volume of production and the alignment short-run growth. Indeed, unemployment in manu- of relative prices to world prices-is that the evidence facturing increases about as often as it decreases often gives rise to conflicting interpretations. In most (Harrison and Revenga 1995). An earlier study of cases, though, it is less an issue of whether trade reform eighteen trade reform episodes similarly concluded works than an issue of whether it has been tried. The that no association can be established in individual pace of reform is often slower than is sometimes countries between trade liberalization and short-run claimed, for instance in Sub-Saharan Africa, where unemployment in the manufacturing sector much agricultural reform only started around 1991, (Papageorgiou, Michaely, and Choksi 1990). several years after other reforms. And when the supply How to explain these surprising findings? In part, response is poor, there is almost always a negative pro- by the capacity of new export-oriented firms to tection or regulatory barrier to explain it, such as cash quickly enter the market and of existing firms to crops that are heavily controlled by the buying power quickly switch production strategies toward exports (monopsony) of state marketing boards. (provided reforms are credible). This capacity has A related issue concerns what complementary poli- been underestimated because of the lack of data at a cies are necessary, for example, macroeconomic stabi- sufficiently disaggregated (that is, firm) level. lization. Large fiscal deficits are probably the single Industry-level data miss the shifts that go on among most important reason for the collapse of trade firms within industries, which is where the vast reforms, followed by inflationary financing and over- majority of micro-level reallocation occurs. valued real exchange rates. Since tariffs are often an A new study of micro-patterns of turnover, pro- important source of government revenue for devel- ductivity, and market structure suggests that indus- oping countries, tax mobilization is an indispensable trial evolution in developing countries is much more accompaniment to tariff reduction. Supply response dynamic than previously thought (Roberts and also depends on the credibility of reforms. Ways to Tybout forthcoming). Among the study's findings is signal credibility include commitment to the World that substantial job creation and destruction take Trade Organization or to adjustment programs sup- place at all phases of the business cycle, implying that ported by multilateral institutions. economies must continually adapt to changing cir- Assessments of supply response, then, first require cumstances, whether or not they open to trade. a sorting out of facts about regulation and movements These findings are in accord with the experience of in incentives and relative prices. One study of Sub- Ghana following its trade reform of the mid-1980s Saharan Africa illustrates the importance of comple- (World Bank 1994). After the reforms were intro- mentary policies (Collier 1995b). Collier proposes a set duced, export-oriented firms expanded rapidly while of conditions that constitute a "minimum adequate import-substituting textile firms contracted. In some environment" for prospective reformers; countries cases old export markets were revived, for example, that do not meet any of the criteria are deemed not to in furniture, aluminum utensils, African prints, and have established conditions satisfactory for growth. milk. On balance the manufacturing sector's employ- The conditions are absence of civil war, which ment increased by some 10 percent a year between excludes countries accounting for 11 percent of the 1984 and 1991. 35 Nevertheless, the workings of factor markets may be exchange rate (C6te d'Ivoire was the exception) and slow, especially in countries that are the least integrated experienced a faster growth rate of exports. In this with the global economy. Leads and lags and the real- context of highly respectable output and trade per- location of resources from nontradables to tradables formance, most of the countries were able to improve mean that governments have a role to play in remov- their fiscal and current account balances through com- ing impediments in these markets, in addressing social plementary fiscal policies. costs through targeted social safety net programs, and Since there are likely to be transition costs associated in managing the political economy of trade reform. with the reallocation of labor and capital among indus- A third concern is that trade liberalization can have an trial sectors following trade reform, and since these adverse impact on fiscaland current account balances. Trade costs will affect categories of workers, regions, and sec- liberalization is motivated by concerns of efficiency tors unevenly, managing this transition is one of the and growth rather than macroeconomic balance. policy challenges posed by reform. Successful reform- Nonetheless, although it reduces the cost of imports, it ers have complemented trade liberalization with need not have an adverse impact on the budget balance macroeconomic stability, including the control of fiscal or the balance of payments when accompanied by deficits and inflation; avoidance of overvalued appropriate exchange rate and fiscal policies. To the exchange rates, which often includes real exchange rate contrary, a study of nine developing countries that devaluation; adequate compensatory tax mechanisms received thirty-five trade-related adjustment loans where necessary; and efficient domestic resource allo- during the 1980s suggests that output and exports can cation mechanisms, such as financial sector reform and increase as a result of trade reform, improving fiscal a reduced role for parastatals. Poland's recent experi- and current account balances, provided that strong ence provides a good example of such efforts (box 3.1). complementary fiscal measures are taken (table 3-1). Adopting appropriate complementary policies helps In six of the nine countries real GDP growth rates minimize the adjustment costs associated with trade were higher after reform: intensive adjusters such as liberalization and reduces the chances of a reversal. Ghana, Indonesia, Morocco, and Turkey were able to achieve high rates of growth. All the countries Obstacles, both real and imagined, to trade reduced quantitative restrictions on imports, and integration those that reduced tariffs experienced substantial increases in productivity. Most of the countries also Another important concern of policymakers is that engineered a substantial depreciation of the real world markets in trade do not offer a level playing field, Trade reform can be coinpleennted by inproved fiscal and current account policies Table 3-1 Impact of trade reform for World Bank trade adjustment loan recipients, 1978-89 (percent) CCte Period" Colombia d1oirc Ghana Indonesia Jamaica Mexico Morocco Pakistan Turket Auerage Real GDP growth rate Before 4.2 4.4 -0.2 5.8 -3.1 6.8 4.6 7.3 2.1 3.6 During 2.8 1.2 3.0 3.9 3.0 0.7 3.5 6.4 4.9 3.3 After 4.6 -1.7 5.4 7.7 3.2 0.5 4.9 6.3 5.1 4.0 Fiscal balance as a share of GDP Before -3.6 -13.9 -9.4 -2.3 -13.5 -10.4 -11.4 -6.9 -7.0 -8.7 During -5.8 -3.8 -2.2 -2.0 -14.4 -8.9 -9.8 -6.9 -6.1 -6.7 After -1.3 -6.4 0.3 -3.2 -10.1 -12.6 -4.0 -7.8 -6.3 -5.7 Current account balance as a share of GDP Before -5.9 -9.6 -2.1 -0.6 -8.0 -4.1 -10.4 -4.5 -3.1 -5.4 During -7.8 -4.9 -2.7 -4.1 -13.3 2.2 -6.9 -3.5 -2.8 -4.9 After 0.3 -2.9 -2.9 -3.4 -1.4 -0.6 0.4 -3.5 -0.6 -1.6 a. Before refers to 1978-82, during to 1983-85, and after to 1986-89. Source: World Bank data and staff estimates. 36 Box 3-1 How Poland sustained its trade turnaround, despite the costs Poland's experience shows that trade reform in conjunc- How did Poland manage to stay the macroeconomic tion with macroeconomic stabilization does carry transi- course? Faced with sluggish exports in early 1991, the tion costs, but that with perseverance the benefits can be authorities moved from a pegged exchange rate linked much greater than the costs. It also shows the importance to the U.S. dollar to a crawling exchange rate linked to a of complementary reforms, such as privatization of basket of currencies of Poland's major trading partners. state-owned enterprises. Rigid monetary and interest rate policies discouraged Poland's "big bang" reform featured simultaneous sta- the continuation of soft budget constraints, which in turn bilization, liberalization, and privatization, as well as the created incentives to shift production toward import establishment of market-supporting institutions. On substitutes and exportables. Privatization of initially January 1, 1990, the prices of all exportables were liberal- successful state-owned enterprises locked in improved ized, exchange rates were unified, the domestic currency performance by insulating them from potential political became convertible (for current account transactions), a pressures. The supply response was driven not by easily positive real interest rate was introduced, and wages in depletable energy products or raw materials, but by the public sector were frozen. More gradual complemen- manufactures. The share of manufactures in total exports tary reforms in public finance, the banking sector, and to OECD countries increased from 47 percent in 1988 to state-owned enterprises, including privatization, were almost 70 percent in 1994. Opening of the economy to also announced. external competition produced adjustment in relative Poland's output fell by more than 20 percent in factor intensities more in line with factor endowments. 1990-91 but rebounded sharply with 2.5 percent growth External factors were also important, especially in 1992, an average 4.6 percent in 1993-94, and an esti- improved access to EU markets. Generalized System of mated 7 percent in 1995. Initially, unemployment-an Preferences status and later the trade provisions (the unknown phenomenon under the command econ- Interim Trade Agreement) of an EU association agree- omy-started to emerge. The contraction in output was ment eliminated tariffs on some industrial products and temporarily exacerbated by state-owned enterprises gradually reduced tariffs on others. In 1992, the first year shedding input inventories and other assets. Polish of the Interim Trade Agreement, about 60 percent of trade underwent a rapid reorientation from members of industrial exports obtained duty-free access to EU (then the Council for Mutual Economic Assistance (CMEA) to EC) markets. Thus, while the value of exports to the for- members of the OECD. In 1990 alone the value of mer Soviet Union fell by 78 percent over 1988-94, exports exports to OECD markets increased by 46 percent. The to the European Union increased by 355 percent. Foreign share of CMEA countries in total exports fell from 41 direct investment helped as well. Sales of the Fiat joint percent in 1989 to 13 percent in 1994, and by 1992 venture, Fiat Auto Poland, Inc., made it the sixth largest Germany had replaced the Soviet Union as Poland's Polish exporter in 1993. In 1992 Fiat was not even among major trading partner. the 500 largest enterprises. making for a low probability of success in opening to by developing countries. World agriculture remains trade. Developing countries' fear that global markets highly protected and subsidized, with little change for goods and services are stacked against them helps likely in the foreseeable future. The Multifiber explain why some (particularly in the Middle East, Arrangement continues to be a major distortion in Africa, and Central America) have adopted a gradual- global trade; its phaseout is heavily weighted toward ist and overly cautious approach to trade reform. But the end of the transition process in 2005. We take up many of these fears are exaggerated. For example, the these topics in turn, beginning with the real obstacles. fear that industrial country tariff and nontariff barriers preclude export success is at odds with the facts: tariff Antidinping barriers are much lower now than when the newly industrializing countries of East Asia started their Because they are rarely motivated by sensible eco- industrialization drive (box 3-2). Also, the fear that tar- nomic criteria, antidumping actions remain a menace iff escalation-rates that rise with the degree of pro- to free trade and can have serious repercussions, as cessing-inhibits the processing of commodities is exemplified by the recent EU action against central almost certainly exaggerated, given the influence of European steel. Disturbingly, developing countries other constraints (as discussed later in this section). are increasingly turning to antidumping actions. Distorted transport costs, a self-imposed impediment, The number of antidumping actions reported to the are likely to be much more important. General Agreement on Tariffs and Trade (GATT) Still, there are some serious external obstacles. Secretariat has fluctuated widely over the past decade, Antidumping actions remain a favored device of pro- but since 1990 there has been a steady rise in activity tectionists, and one that is increasingly being adopted among antidumpers (figure 3-1). The number of 37 Box 3-2 Now and then: Industrial country tariffs on developing country trade The importance of current OECD tariff barriers to devel- average of the mid-1960s. Tariffs are even lower for oping country exports can be put into perspective by com- some countries if trade preferences are taken into paring them with those that faced the newly industrialized account. Thus industrial country tariff barriers to devel- countries when they began their successful export drive in oping country trade are far lower today than those over- the 1960s. During the late 1960s countries like Hong Kong, come by the newly industrialized countries. the Republic of Korea, Singapore, and Taiwan (China) faced average tariffs of 17 percent in major OECD markets Average tariff on imports of (see table). At that time industrial countries' tariffs dis- manufactures from criminated considerably against developing countries. developing countries, Furthermore, Generalized System of Preferences schemes Import market mid-1960s had not yet been adopted, so the newly industrialized European Community 14.3 countries had to compete with other suppliers on an equal, Japan 18.0 most favored nation basis. The degree of escalation in Sweden 9.8 OECD tariffs was far greater than it is today (UNCTAD United Kingdom 19.5 1968; Amjadi, Reincke, and Yeats 1996). United States 17.9 Today developing countries as a whole face OECD All OECD countries 17.1 tariffs of less than 4 percent, well below the 17 percent antidumping initiations reported in the year preceding study has found that by competition policy standards, June 1994 was 2.5 times greater than the number exporters posed no threat to the competitiveness of reported just five years earlier. These cases did not the importing country or world industry in more than always result in an antidumping duty. Often the threat 90 percent of U.S. and EU antidumping duties of a formal antidumping duty or the loss of customers imposed in the 1980s (de Jonquires 1995). Another resulting from uncertainty created by the secretariat's piece of evidence that antidumping cases do not investigation induced an exporter to agree to a volun- involve predatory pricing is the fact that the 1916 U.S. tary export restraint.' antidumping law has never been used successfully. No direct estimates of the impact of these antidump- This law offers the reward of triple damages for ing cases have been made, but two telling comparisons proven dumping cases, but it also imposes the burden are available. Messerlin (1990) has pointed out that EU antidumping cases during the 1980s covered a sizable amount of EU imports that was, for example, roughly Developing countries are increasingly turning to equal to the value of EU imports of agricultural goods. antidumping actions Another comparison builds on the fact that the eco- nomics of antidumping and of OPEC price increases are Figure 3-1 Antidumping initiations reported to the the same. In both, the objective and the effect is to have GATT Secretariat, 1986-94 importing users pay more for imports. In the United States the cost imposed on the U.S. economy by its own Number antidumping actions during the 1980s comes to about 250 Industrial half the cost that the 1974 OPEC price increase imposed countries on the U.S. economy (Finger 1991). 200 Competition policy laws distinguish between dis- count pricing and predatory pricing. Predation refers to a situation in which a seller holds prices below cost 150 sufficiently long to drive competitors out of business, then exploits the resulting monopoly situation to impose high prices. But there is little if any evidence to suggest that antidumping cases aim at predatory Developing/ pricing. One legal scholar with considerable direct 50 countries experience with antidumping cases has argued that predatory pricing was not even remotely present in 0 any of the 767 affirmative antidumping determina- tions reached by Australia, Canada, the European 1986 1987 1988 1989 1990 1991 1992 1993 1994 Community, and the United States between 1980 and Sotirce: GATT Secretariat. 1986 (Palmeter 1988). More conservatively, an OECD 38 of having to meet the legal standards of competition Agriculture after the Uruguay Roinid policy. Predatory pricing is part of the rhetoric of Progress toward trade liberalization in agriculture has antidumping, not part of the substance. In economics been limited and uneven in recent years. While the dumping is a precisely defined form of pricing. In pol- Uruguay Round successfully brought agriculture itics it simply means trouble from imports. In law under multilateral discipline, the extent of liberaliza- dumping follows its legal meaning, not its economic tion achieved fell short of expectations and is subject one. Antidumping is ordinary protection with a good to the vagaries of implementation. A ceiling has been public relations program. Accordingly, there is a need placed on export subsidies and aggregate domestic to strengthen the World Trade Organization's (WTO) support. Tariffication has abolished nontariff barriers rules on the application of such measures. and replaced them with transparent, bound tariffs. The developing world's involvement in anti- Still, restrictions on market access remain substantial. dumping actions is a recent development. During Even though the reductions in export subsidies are 1980-85 developing countries initiated only 34 of important, trade in many products will still be sub- the 1,019 antidumping cases reported to GATT. In jected to significant distortions at the end of the imple- recent years, however, six developing countries rentation period. With trade at current levels, accounted for more than 25 percent of the subsidized exports can account for a third or more of antidumping actions initiated in the five years prior trade in beef, veal, pork, wheat, and vegetable oils. to June 1994.2 Mexico alone was responsible for More than a fifth of trade in poultry and coarse grains nearly 25 percent of the developing country total, can still be subsidized. surpassing Canada in the number of cases reported. The average tariff equivalents of post-Uruguay In addition, thirty-four of the forty-seven countries Round border measures in agriculture often are that have brought antidumping legislation to the higher than the pre-Round applied rates. Many of the WTO's Committee on Antidumping Practices since tariffs resulting from the tariffication process are pro- January 1995 are developing countries. While most hibitive and, while they will be reduced during the of these antidumping systems have yet to be used, implementation period, will remain high in many they represent significant potential for the spread of countries. Primary commodity exports of developing antidumping activity. countries such as sugar, rice, and certain fruits will A subsidiary issue is the treatment of nonmarket remain under high tariffs in many OECD countries. In economies-the socialist or formerly socialist ones. many developing countries the high tariff bindings Contingent protection, such as antidumping initia- and the continuation of state trading enterprises with tions, is sometimes alleged to hit the exports of non- monopoly control of imports and exports will limit market economies harder than others. It is true that the benefits of the agreement. these economies, including China, account for a far higher share of antidumping cases than their share Multfiber Arrangement of U.S. or EU imports and that the law treats them differently. The main difference in the treatment is Multifiber Arrangement (MFA) quotas constitute a that the "fair value" of a nonmarket economy's huge compulsory export tax on developing country exports is determined by valuing the inputs they use exports of textiles and clothing. Their eventual at the prices found in "similar" market economies. removal will have highly beneficial welfare effects in Despite the hurdles, however, a careful examination industrial countries, with varying effects in develop- of the law and practice of contingent protection sug- ing countries. But because the phasing out will likely gests that nonmarket economies are not discrimi- be delayed until the last moment-in 2005-the MEA, nated against; the frequency of their cases is which covers virtually all fabrics and all significant probably just due to their strong competitiveness in developing country exporters, continues to represent certain sectors. a major distortion to world trade. How can the nonmarket economies avoid these The export quotas imposed under the MFA are essen- problems? Joining the WTO helps a little, but the WTO tially set by importing countries but are administered allows the nonmarket economies to be treated differ- by the governments of exporting countries. The alloca- ently. Proving that they are now market economies tion mechanisms used for quotas typically generate could also help, but market economy members of the substantial costs by creating incentives for rent-seeking WTO still face antidumping cases. Nonmarket behavior or by stimulating production and exports that economies must therefore try to fight their cases more are unprofitable for the economy as a whole (Trela and effectively, for example, by using expert lawyers and Whalley 1995). For exporters the quota scheme is effec- providing documents that substantiate their market tively the same as (though less transparent than) a set of status. export taxes that vary by product and destination. 39 relative to their underlying competitiveness will lose MFA quotas represent a significant tax on from the abolition of the MFA, with Central America developing country exports and the Caribbean (for example, the Dominican Republic) being vulnerable unless they can enhance Table 3-2 Export tax equivalents of MFA quotas on productivity Alternatively, they must begin to shift to apparel, 1992 other areas of comparative advantage, a course already (percent) taken by some East Asian countries (Hong Kong, the United States European Republic of Korea, Singapore). The message of the Supplier and Canada Union MFA phaseout is that genuinely competitive produc- Lai Ation whose development has been inhibited by quotas South Asia 40 36 wvill have an opportunity to flourish. Hong Kong 18 16 Indonesia 47 48 Tariff escalation Korea, Rep. of 23 19 Malaysia 37 32 Tariffs and other trade barriers in major markets, and Philippines 34 28 their tendency to increase or escalate with the level of Taiwan, China 19 22 a product's processing, are often cited as working Thailand 35 36 against the efforts of commodity-dependent countries Source: Chyc and others 1995. to increase domestic processing. Given the special relations and conditions of market access many devel- oping countries have in OECD markets, the issues relating to tariff escalation are not as clear-cut as was Broad estimates suggest that the export tax equiv- once thought. Account needs to be taken of the alents are enormous, ranging from 16-48 percent and Generalized System of Preferences (GSP) and the affecting exporting countries differently (table 3-2). Lome IV Convention preferences some developing The protection provided by the MFA quotas is all the countries receive, including the least developed coun- more remarkable given that it is imposed on top of try category under the GSP. some of the highest tariffs on manufactured goods- To analyze the structure of EU tariffs, Safadi and about 13 percent on textiles and clothing imports from Yeats (1994) developed a processing chain scheme and developing countries. applied it to primary commodities exported from Product groups accounting for just over half of the Sub-Saharan Africa. This analysis shows that the pos- total imports falling under MFA categories will be sibility of African countries receiving GSP, Lome IV, or integrated with the GATT in three tranches during the least developed country preferences eliminates the first seven years of the agreement, with the rest to be effect of any nominal tariff escalation. For example, eliminated at the end of the ten-year transition period. African preferential tariffs are zero at every stage in The importing countries are free to choose the prod- the four processing chains: hides and skins, leather, ucts that will be integrated and, not surprisingly, have and leather manufactures; raw cotton, cotton yarns, usually chosen to integrate products that are not sub- and cotton fabrics; zinc ores, unwrought zinc, and ject to quotas or on which quota restrictions are not a zinc bars and sheets; and palm nuts and kernels and binding constraint. As a result most of the liberaliza- palm kernel oil. Thus, thanks to the influence of pref- tion provided for under the MFA will likely be erences, tariffs do not appear to have been a major deferred until the tenth year (2005), and even then will constraint to the further processing and export of be subject to safeguard clauses. African commodities. The liberalization proposed under the Agreement If the importance attached to tariff escalation has on Textiles and Clothing will generate substantial wel- been exaggerated, then what explains the bias in the fare gains. The largest gains will likely accrue to the structure of many developing countries' exports away industrial countries that imposed the MFA's peculiar from processed products? For example, compared and perverse form of protection. Among exporters, the with imports from all suppliers, the structure of greatest gains will likely go to highly efficient produc- OECD imports of primary and processed commodi- ers whose current exports to industrial country mar- ties from low-income Sub-Saharan Africa is concen- kets are tightly constrained by quotas. Thus China, trated in unprocessed commodities (table 3-3). Indonesia, Thailand, and South Asia should gain from First, some commodity processing operations, such the abolition of the MFA. Also gaining will be countries as petroleum refining, ferrous metals, and tobacco that have demonstrated the capacity to develop cloth- manufacturing, involve highly capital-intensive and ing exports, only to be faced with a newly imposed mechanized operations in which middle- and high- quota, for instance Kenya. Exporters with large quotas income countries hold a comparative advantage. 40 Sub-Saharan Africa's exports are concentrated on primary commodities Table 3-3 Share of 1993 OECD imports of primary and processed commodities (percent) Commodity chain Primary stage Semiprocessed products Final processed stage Agricultural materials Low-income Sub-Saharan Africa 37 32 31 All countries 6 47 47 Ores and metals Low-income Sub-Saharan Africa 31 34 35 All countries 8 46 46 Tobacco Low-income Sub-Saharan Africa 100 0 All countries 42 a 58 a. No semiprocessed stages are defined for tobacco. Source: Calculated using UN COMTRADE data based on Safadi and Yeats 1994. Second, a number of comprehensive commodity- foster the development of national fleets. In fiscal 1991 specific studies have shown that various barriers to Sub-Saharan Africa's net freight and insurance pay- market entry (aside from escalating trade barriers) ments were $3.9 billion, or 15.0 percent of total work against further processing in some developing exports, compared with 5.8 percent for all developing countries. In some cases international transport costs countries. Net transport and insurance payments (discussed below) have been found to escalate with average more than 25 percent of total exports for one- processing. And the final consumer market for some third of Sub-Saharan Africa countries. And the trend commodities like coffee, cocoa, and tea have high of costs is rising. entry barriers that are associated with the market Summary statistics on 1993 transport costs show power of a few large oligopolistic firms (Scherer 1984). that Africa is at an important cost disadvantage rela- In summary, some of the externally imposed barriers tive to its competitors. Figure 3-2 shows median costs, to trade are significant for some developing countries. represented by a horizontal bar, and the range of costs The remedy lies in OECD country efforts to further around the median, represented by a vertical bar open their domestic markets, whose endpoints mark the first and third quartiles of export volume. The shaded bars show the margin of Transport costs African costs over those of other countries. For exam- ple, the median vessel nominal freight rate for middle- The influence of transport costs on developing country income West Africa-10 percent-is about 2 trade has been a neglected subject, in part because of percentage points higher than that paid by other the lack of easily accessible data on their incidence and developing countries. This compares with the average in part because of presumptions that freight costs are 2.4 percentage point reduction in industrial country less important than tariffs and are beyond the control tariffs (to 3.9 percent) achieved in the Uruguay Round. of exporters. Recent analysis suggests that both pre- Some African exports encounter even higher transport sumptions are erroneous, and that in some instances costs. For example, 25 percent (the third quartile) of transport costs can outweigh the impact of tariffs. Sub-Saharan Africa's air exports encounter freight The freight costs for African exports to the United rates exceeding 26 percent, and about a quarter of low- States are usually considerably higher than on similar income West Africa's vessel shipments have nominal goods originating in other countries. These charges freight rates of more than 19 percent. conceal very high rates of effective protection for The disincentive effects of these high transport costs processed goods-which may be thought of as the are captured by the effective rates of protection. For ani- cost margin below domestic producers that exporters mal and vegetable oils the transport effective rate of must achieve on value-added production in order to protection of 56 percent is more than five times higher be able to compete-a point that significantly reduces than the corresponding nominal rate of protection.3 incentives for new investment in export-oriented pro- Thus while Africa is an important producer of several duction activities. In the past many African countries types of oilseeds, like groundnuts and copra, further adopted anticompetitive cargo reservation policies to processing for export is inhibited by transport costs. 41 Many developing countries' governments have African exporters payv high transport costs adopted anticompetitive cargo reservation schemes in the hopes of achieving objectives such as the develop- Figure 3-2 African transport costs as a share of ment of national fleets. These schemes have been export value largely unsuccessful in promoting national lines, but their anticompetitive effects have resulted in markedly Percent higher freight rates. World Bank studies have shown 35 Sub-Saharan Oil- that deregulation and the adoption of measures to pro- Africa West Africa exporting mote competition for international transport services 30 countries can significantly lower freight costs for developing Air Low Middle countries (Bennathan and others 1989). 25 income income Regional agreements as a vehicle for integration 20 Vessel Vse Regional integration agreements-regionalism for 15 Vessel short-are a hot item on the international trade agenda. Regionalism covers an enormous diversity of arrange- 10- ments, from fully fledged common markets to customs unions (characterized by common external tariffs) to ordinary free trade areas. This diversity is one reason it 51 is hard to make a conclusive assessment of such arrange- 0 Quariments. Another is that many agreements-the North American Free Trade Agreement (NAFTA), the EU asso- art of exports ciation agreements, the nascent Asia-Pacific Economic Third Margin of African costs over Cooperation arrangement-are of recent vintage. Median other developing countries Contrary to popular belief, the regional concentra- First tion of trade has not grown much in the second half of Source: Nominal transport costs calculated from U.S. Department the twentieth century except within the European of Census trade tapes. Pre-Uruguay Round tariff averages are Union, which is the principal example of deep integra- from Erzan and Svedburg (1989). tion (involving harmonization of standards, policies, and institutions). In fact, the fastest-growing trade area has been East Asia, where regionalism is largely absent What type of corrective policy measures could be except for the Free Trade Agreement of the Association used to address these distortions? Any approach of Southeast Asian Nations. This experience suggests should recognize that two types of factors affect trans- that joining a trading bloc is not necessary for success- port costs-those that are not subject to policy control ful trade performance, but can in some instances (par- and those where considerable scope for corrective ticularly in arrangements between developing and action exists. Distance is one factor that is not subject to industrial countries) generate better and more secure policy control-freight rates vary directly with dis- market access, increase policy credibility, attract invest- tance, and countries that are geographically removed ment, and improve access to technology. The benefits from their major markets obviously cannot change are far from automatic, however. their location. Similarly, countries that export a high share of bulky low-value products face higher nominal Arrangements between developing and industrial freight costs than countries that ship high-value prod- countries ucts with low stowage factors. However, governments and traders generally have far more policy options for The expansion of NAFTA to include Mexico (in addi- reducing transport costs than is generally recognized. tion to Canada and the United States) and the network These options include such measures as cargo bulking of special agreements tying various developing coun- to achieve economies of scale in transport, rationaliza- tries to the European Union illustrate the wide array tion of shipping services and improved scheduling for of new forms of integration between developing and liners, adoption of procedures to speed vessel turn- industrial countries. There are different types of pref- around, utilization of lower-cost tramp services where erential arrangements: unilateral preferences (the feasible, development or improvement of coastal European Union's Lom& Convention), free trade areas feeder services, adoption of new transport technolo- (NAFTA with a possible extension to Chile), customs gies, promotion of shippers associations, and port and unions (the European Union and Turkey), association storage improvements. or preferential agreements (the agreements between 42 the European Union and Central and Eastern The second part of the answer is that regional inte- European countries and the Baltic states, and between gration is not sufficient for economic success. For the European Union and certain Mediterranean coun- instance, Central and Eastern European countries tries), and enlargement of a regional integration inherited huge distortions in their trade patterns that agreement through accession (Greece, Portugal, and were a major factor behind the frequently big changes Spain joining the European Union). in their exports to the European Union during the first While integration between developing and indus- stages of their transitions. Indeed, the fact that not trial countries started in the 1980s, it has assumed far every country with an EU association agreement expe- greater significance in the 1990s. Between 1991 and rienced an export boom (Kaminski, Wang, and Winters 1995 the European Union signed association agree- 1995) and that many of the export gains occurred in ments with nine European transition economies, con- products not subject to extensive liberalization mea- cluded similar agreements with Israel, Morocco, and sures (Kawecka-Wyrzykowska 1995) suggests that the Tunisia, and is currently negotiating with several other EU agreements were not sufficient for export success. Middle Eastern and North African countries. The main One frequently cited-but overstated-benefit of parts of these agreements deal with establishing free these agreements to developing countries is access to trade areas for industrial products over a period of ten industrial markets. The European Union and the to twelve years along with, at least in Europe, some lib- United States have, with a few notable exceptions, rel- eralization in services. The European Union recently atively liberal import regimes for their developing created a deeper but narrower arrangement with country trade partners (through low most favored Turkey establishing a customs union in industrial nation tariffs and widespread application of General- goods and perhaps eventually in agriculture. The cre- ized System of Preferences rates). Thus the benefits of ation of NAFTA in 1993 is thought by some as laying improved market access are rather limited for most the groundwork for a major North-South free trade developing country partners unless they gain access area in the Western hemisphere. Discussions are under to industrial countries' protected "sensitive" markets, way to establish a free trade area of the Americas and such as agriculture, textiles and clothing, and steel. another in the Asia-Pacific region. It is too early to However, these are the very sectors that are most often know whether these efforts will succeed, but they excluded from or subject to constraint under regional illustrate the enthusiasm for such arrangements. integration agreements between developing and Thanks to their exclusiveness in most cases-it is too industrial countries. early to know, for instance, whether liberalization The benefits of integration (global or regional) are under the Asia-Pacific Economic Cooperation agree- not static. The depth of a regional agreement-as ment will be done on a most favored nation basis-and measured by the scope of precommitments to coop- despite the inherent dangers involved in regionalism, eration, harmonization, and liberalization-also preferential arrangements have become an alluring influences the dynamic gains to integration. Major model. Many developing countries view preferential gains stem not only from changes in the level of arrangements as a means of securing access to technol- imports but also from higher investment, increased ogy, capital, and markets, and sometimes of fostering competition, better use of economic resources, the increased (or at least nondecreasing) financial trans- economies of scale that develop from access to larger fers. In addition, the arrangements are seen as a mech- markets, and the higher potential for export-led anism for achieving greater international credibility of growth. Gains also arise to the extent that a develop- policy reforms. In short, some developing countries see ing country's national institutions become like those integration as both a potential engine of economic of its industrial country partners. In this regard the growth and as a way to develop domestic institutional enlargement of the European Union to include rela- settings conducive to sound economic policies. tively poor southern European countries represents But is regionalism actually advantageous to devel- the deepest integration between developing and oping countries? The first part of the answer is that industrial countries (Winters 1993). Other agree- they are not necessary for success. East Asia's spec- ments are less inclusive but nevertheless address tacular economic success has evolved largely inde- issues of nonborder policies such as competition poli- pendently of special arrangements with industrial cies, subsidies, technical standards, environmental countries or institutions of regional integration. Chile regulations, and intellectual property rights. and Mauritius provide further examples. Their dra- A very important benefit of the institutional reform matic economic growth can be attributed to sound associated with integration between developing and domestic policies-observing market-oriented funda- industrial countries is the enhancement of the devel- mentals, vigorous accumulation of physical and oping country's credibility in its commitment to eco- human capital, and the willingness to exploit the nomic reform and its attractiveness as a location for opportunities offered by international markets. investment. Once the process of accession to the 43 European Union had begun, Spain and Portugal expe- Arrangements between developing cotitries rienced large inflows of foreign direct investment. Anticipating NAFTA, foreigners poured about $30 Trade arrangements between developing countries billion of foreign direct investment (FDI) into Mexico are a different proposition from developing-industrial during 1990-93. While some portion of this flow can ones. Because of their small size and limited scope for be attributed to the liberalizing measures that had economies of scale, such arrangements may do little been introduced earlier, at least some of it is likely to to enhance participants' policy credibility in the way have been directly attributable to NAFTA. that, for example, commitment to an EU arrangement In a similar vein, the combination of domestic liber- would. And although the empirical evidence on the alization and market access has lured foreign investors most recent arrangements is still too preliminary to to Eastern Europe. In 1994 the three most advanced support firm judgments, past experience suggests reformers-the Czech Republic, Hungary, and that unless such arrangements are pursued in the con- Poland-received FDI inflows of $19 billion. There is text of liberal policies relative to the rest of the world, already evidence that multinational corporations they will be harmful. investing in these countries have contributed to indus- Earlier regional integration agreements between trial restructuring, the diffusion of managerial and mar- developing countries were carried out in the context keting skills, and the growth of the export sector. Polish of inward-looking, import-substituting industrializa- Fiat became Poland's largest exporter to the European tion strategies and a large public sector. With high pro- Union in 1993, and multinational corporations have tection, these agreements generated more trade contributed immensely to the expansion of Hungary's diversion than trade creation. For instance, trade intraindustry trade with the European Union. among members of the Central American Common Regional arrangements are neither necessary nor Market went from 2 percent of their total trade before sufficient to stimulate FDI inflows, however. Chile, for the formation of the common market to about 20 per- example, has attracted substantial volumes of FDI, cent after, with most of the increase in light manufac- thanks to private sector reform and a liberal invest- tures that had previously been imported from ment regime. By contrast, Bulgaria and Romania, industrial countries. Labor can also be diverted. For although partners to EU association agreements, have instance, protection of labor-intensive rice production failed to attract foreign investors-mainly because of in C6te d'lvoire led to an inflow of labor from neigh- the less-advanced state of their market-oriented boring countries and to misallocation of resources. reforms. The more recent, outward-looking agreements are Regionalism can enhance growth by improving among countries with more liberal trade regimes that efficiency and stimulating investment. But the degree have privatized and deregulated large parts of their of success depends on a country's commitment to economies. Thus they are less likely to fail. In Latin macroeconomic discipline. The experiences of Greece, America the main agreement is Mercosur. Bilateral free Portugal, and Spain offer important insights. Greece trade agreements have also mushroomed, including has experienced slow growth, low investment, and Mexico-Venezuela, Chile-Mexico, Chile-Colombia, persistently high inflation since its accession to the Chile-Venezuela, and more. Chile is also negotiating a European Union, while Portugal and Spain experi- preferential trade agreement with Mercosur. In Asia the enced an impressive turnaround in growth and falling Association of Southeast Asian Nations (ASEAN) has inflation. The difference is that generous transfers initiated a process of integration, as have South Asian from the European Union (amounting to about 5 per- countries.The ASEAN Free Trade Agreement has been cent of GDP) have weakened pressures to necessary in place for three years-almost as long as NAFTA- adjustment in Greece, while neither Portugal nor and during that time intra-ASEAN trade has risen sub- Spain has allowed these transfers to affect their com- stantially, trailing only the group's trade with Japan and mitment to rigid fiscal policy (box 3-3). the United States in value. In Sub-Saharan Africa new The growing trend toward regionalization changes integration efforts include the Cross-Border Initiative. the economic possibilities for the countries that are The possibility of trade diversion rather than trade excluded from regional blocs, who face losses from creation suggests that regional agreements between EDI diversion and more difficult market access to developing countries need to integrate further with the regional blocs than bloc members. For instance, the world market by continuing to lower external tariffs preferences given to EU producers and the invest- (common or not) and by integrating with the major eco- ments by multinational corporations in the automo- nomic entities (NAFTA, the European Union, Japan, and bile industries of European associates have been Asia). For instance, Chile is negotiating an agreement largely diversionary. Japanese and Korean producers with Mercosuras well as with NAFTA, and it has started and Eastern European consumers lose from such talks with the European Union. It is important that such developments. efforts do not exclude Japan and the rest of Asia, or they 44 Box 3.3 Liberalization is not enough: Greece, Portugal, and Spain following EU accession Joining the European Union (then the European Within a few years of joining the Union inflation rates Community) led to very different outcomes for Greece fell by half in Portugal and Spain, partly as a result of the than for Portugal, and Spain. Output growth accelerated price discipline imposed by real exchange rate apprecia- in Spain and Portugal but fell in Greece, where unem- tion following accession. This effect was strengthened ployment soared. These different outcomes were largely when Portugal and Spain joined the exchange rate mech- the result of different domestic policies. anism. Greece, on the other hand, experienced a jump in Before joining the Union-Greece in 1981 and Portugal inflation to nearly 20 percent after joining, and did not and Spain in 1986-the three countries embarked on sim- get the inflation benefit of currency appreciation or ilar trade liberalization policies. They also received sig- exchange rate mechanism membership. However, nificant EU structural transfers intended to offset the imports of cheaper manufactures from the Union after adjustment costs associated with accession. By 1993 net liberalization did help dampen inflationary pressures in official transfers-comprised mainly of EU transfers- the tradable sectors of all three countries. had reach 5.6 percent of GDP in Greece and 3.3 percent in Portugal. Net transfers also rose in Spain, though they only amounted to 0.6 percent of GDP. Despite higher aid flows, Greece's government budget deficit rose from an average 4.5 percent of GDP in the five EU membership brought considerable benefits for years before joining the Union to 10.1 percent in 1983-87. Portugal and Spain By contrast, Portugal and Spain consolidated their deficits as they introduced new value-added taxes. Portugal also embarked on a major privatization pro- Inflation and budget deficits (five-year averages) gram. Monetary policy was looser in Greece than in the Percent other two countries, in part because of its fiscal deficits. 0.25 P Money growth in Spain and Portugal fell after EU entry, Pre-accession Post-accession while in Greece it continued at nearly 25 percent a year, 0.20 as it had in the five years before entry. As a result Portugal 0.15 and Spain were eventually able to join the European exchange rate mechanism; Greece was not. 0.10 Improved macroeconomic stability in Portugal and Spain bolstered the confidence of foreign investors, who 0.05 Deficit/ Deficit/ Deficit/ .GDP GDP G'DP were also attracted by low labor costs. Both countries o.oo .: attracted large amounts of private capital after their entry Inflation Inflation nflation into the Union-foreign direct and portfolio investment -0.05 after accession rose by some 3 percent of each country's -0.10 GDP. Greece's credit ratings also rose in the period before EU entry, but they were later downgraded, there was no -0.15- inflow of portfolio capital after accession, and foreign Portugal Spain Greece direct investment actually fell a little. Source: Alogoskoufis 1995. will run the risk of generating significant amounts of * Regional trade arrangements are neither neces- trade diversion (Leipziger and Winters forthcoming). sary nor sufficient to successful trade reform, though they may help cement it. Conclusion Countries that lag others in integrating with the global economy are confronted with the choice of how much We can draw several conclusions about the costs of to liberalize, how fast, and which path to follow: multi- trade reform and the obstacles: lateralism, regionalism, unilateralism, or some combi- * The costs of trade liberalization and the associated nation. While the benefits and costs of liberalization are problems of political economy must be managed not fully understood, the benefits tend to be higher than by reforming governments, but the costs must be theory would suggest (granted appropriate comple- weighed against the opportunity costs (not least, mentary policies) and the costs more moderate. low credibility) of a gradualist approach. The barriers to developing country exports have * While some of the obstacles to trade integration never been lower-putting aside self-imposed barri- are external, others are domestic-and under ers-and the later a country turns to trade reform, the the control of reforming governments. fiercer the competition will be. Regionalism may help 45 countries fight the battle through enhanced access and 2. Brazil, Colombia, India, the Republic of Korea, policy credibility, but there is no substitute for a gen- Mexico, and Turkey notified GATT of 215 of the 815 uinely nondiscriminatory and liberal import regime, antidumping initiations and 164 of the 521 provisional which benefits producers and consumers alike. duties reported between July 1989 and June 1994 (GATT Secretariat). Notes 3. Since the effective rate of protection measures the difference between value added in the presence of protec- 1. During the 1980s almost half of U.S. antidumping tion (and without trade barriers), relatively low nominal cases were superseded by voluntary export restraints freight costs can produce quite high effective rates of pro- (Finger 1993). tection for low value-added products. 46 Integration and successful commodity producers Most of the countries that have been the slowest to This chapter draws on practical examples from suc- integrate with the global economy are primary cessful commodity exporters that illustrate their poli- commodity-dependent countries-primary corn- cies. These examples suggest that the persistent modities account for more than half of their exports. dependence on a few traditional commodities that Although many commodity exporters-notably in often accompanies slow growth and failure to inte- Sub-Saharan Africa, the Middle East and North grate with the global economy is a symptom of policy Africa, and Central America-suffer from low pro- and institutional weaknesses. Thus there is no ductivity in the primary sector and an inability to inevitable causal link running from commodity diversify their production and export structures, oth- dependence to the failure to improve productivity ers are among the fast growing. During the early and to diversify. 1980s sixty-nine of eighty-four developing countries were commodity-dependent (in the sample for which Defining success comparable data were available). Of these, fifteen were among the fast-growing fast integrators iden- What distinguishes successful from unsuccessful tified in chapter 2 and sixteen were among the commodity producers are high levels of productivity slow-growing slow integrators. Furthermore, the in commodity production, as evidenced by invest- commodity-dependent fast-growing countries, which ment and growth in the commodity sector; an ability included Chile, Indonesia, Malaysia, Mauritius, to diversify into nontraditional commodities, as evi- Mexico, Morocco, the Philippines, Turkey, and denced by a high share of new export markets; and Thailand, have actually grown faster in per capita macroeconomic stability. Underlying all three char- terms than the fifteen developing countries that are acteristics is the ability to attract domestic and for- primarily manufactures exporters. eign investment. This section examines these What characteristics distinguished the perfor- characteristics, while the subsequent section exam- mance of the commodity-exporting strong integra- ines some of the practices adopted by successful com- tors, and what underlying policies accounted for modity producers. these characteristics? The successful exporters had highly productive commodity sectors, possessed the High t7rodnctivity iii existinig connnodities capacity to diversify to nontraditional (typically, high value-added) commodity exports, and maintained Successful and unsuccessful commodity-dependent macroeconomic stability. Improved productivity and exporters exhibited a pronounced difference in per diversification were the consequence of policies con- capita growth in capital stock and income during ducive to high investment, of which macroeconomic 1980-90 (figure 4-1). Countries that invested more stability was the essential ingredient. And successful tended to grow much faster-nearly 3 percentage exporters typically were able to implement their poli- points a year-and have been fast integrators. The cies with the support of strong institutional and GDP of moderate, weak, and slow integrators, on the human capacity and sound and sustainable public other hand, fell during this period. Though strong investment programs. performers diversified faster into manufacturing, this Investment was also encouraged by policies that: was not at the expense of slower growth in the pri- " Strengthened the private sector mary sector. In fact, the disparity in investment per- " Attracted foreign direct investment form ance is especially striking in agricultural " Developed research and infrastructure investment, where the capital stock actually fell for " Promoted financial soundness and flexibility. weaker performers. 47 work, effective institutions, and adequate transport Suiccessful connodity e'xporters invested more infrastructure. Figure 4-1 Annual per capita growth in output, Diversification into nontraditional conmodities capital, and investment for commodity-dependent exporters, 1980-90 Commodity-dependent countries have varied widely Percent in their capacity to increase diversification and pro- ductivity, as evidenced by the composition of their 3 Fast integrators export growth in real terms over 1972-92 (figure 4-2). Moderate, weak, The four successful producers chosen-Chile, 2 and slow imtegrators Indonesia, Malaysia, and Thailand-experienced spectacular growth in nontraditional exports such as fruits and vegetables and shrimp, and steady growth in other agricultural categories. By contrast, slow inte- grators (excluding oil exporters) made little headway in nontraditional exports (shrimp is a partial excep- tion) and actually regressed in traditional products such as sugar and fats and oilseeds. Opportunities for dynamic growth in nontradi- -3 tional commodities are particularly strong in horti- _ -- _ -_ -_ --cultural products. Between 1985 and 1994 world Total Total Total Agri- Agri- Agri- imports of fresh and simply preserved vegetables GDP invest- capital cultural cultural cultural increased from $8.2 billion to $18.8 billion; fresh and ment stock GDP invest- capital dried fruits and nuts grew from $12.3 billion to $24.2 ment stock billion. In 1994 the total value of these imports was Sowrce: World Bank data and staff estimates, about twice as much as the export value of coffee from developing countries (the most important agricul- tural commodity of developing countries in terms of export value). Major exporters of vegetables include Most commodity-exporting slow integrators have the United States ($1.78 billion in 1994), China ($1.42 retained the traditional production, marketing, and billion), Mexico ($1.26 billion), and Thailand ($738 export systems established during the colonial period. million). Production is undertaken by small producers with The horticultural market is competitive because, limited resources and education, while pricing, mar- unlike tropical products, developing country horti- keting, and export systems are government-controlled cultural exporters must compete with industrial and often noncompetitive. Such systems provide lim- country exporters. Horticultural trade also requires ited opportunities for producers and traders to exert sophisticated marketing skills. Delivering perishable initiative, introduce innovations, or learn new goods requires not only well-organized transport, approaches. Many of these less productive commod- communications, and other infrastructure but also ity exporters are subsistence economies that are unable to generate significant domestic savings, and their unstable macroeconomic environment deters foreign InveCstmcnt in mineral exploration varies widely investors. For example, Latin America and Asia have managed to attract large inflows of private capital into Table 4-1 Average annual world mineral mining projects; Sub-Saharan Africa has not. A recent exploration expenditure, 1980-89 study shows that countries with major mining sectors (millions of 1989 U.S. dollars) usually invest 5-10 percent of production value in Australia 560 exploration. Many Sub-Saharan African countries, on Canada 600 the other hand, attracted investment of only about 1 United States 360 percent of production value (table 4-1). A number of South Africa 180 Sub-Saharan African countries are endowed with Sub-Saharan Africa 100 high-quality mineral resources and low labor costs Other 700 and face low or insignificant tariff barriers to exports Total 2,500 of primary products. Thus the figures in table 4-1 Note: Excludes Soviet Union and Eastern Europe. underscore the importance for sustainable growth in Source: World Bank data and staff estimates. mineral production of a sound macroeconomic frame- 48 Cut flowers are another booming commodity. Successful comnnoditi exporters find opportunities in World cut flower exports went from $1.25 billion in nontraditional products 1985 to almost $4 billion in 1993. In that year the world's largest exporter was the Netherlands, with Figure 4-2 Real annual growth in commodity exports of $2.2 billion, followed by Colombia with exports, 1972-92 $382.5 million. Other major developing country Porcent flower exporters are Mexico, Costa Rica, Thailand, Kenya, and Ecuador. Colombia has been the most suc- 12 Slow integrators cessful cut flower exporter among developing coun- 10 Ctries because of the cost advantages that come from its 8 h 1 e Madasia, abundant land, favorable climate, and low-wage Thailand labor force. These factors, as well as the development of air transportation and major changes in govern- ment policies in the late 1960s (such as reduction of import restrictions and devaluation of the domestic 0 -currency), enabled Colombia to compete with local -2 producers in the U.S. market, where demand was 4 increasing rapidly. -6 Macroeconomlic stabilityi -10- The influence of macroeconomic policy deficiencies is Total Fats Total Fruits Sugar Shrimp evidenced by the correlation between the size and agriculture and fishery and oilseeds vegetables volatility of budget deficits and the degree of integra- tion (table 4-2). Among commodity exporters at the Source: World Bank data and staff estimates. start of the period (defined as those with at least half their exports in primary commodities) the fast inte- grators ran much lower deficits-a median of 4.3 per- efficient organization and management of these cent of GDP compared with 7.1 percent for the slow systems. Information and skills allowing penetration integrators. And the commodity-dependent countries of markets protected by complicated tariff and non- that invested more and grew faster were also the tariff barriers are also required. As discussed below, fastest integrators. Indeed, the growth of the fast inte- foreign direct investment is one way to acquire this grators even outpaced that of non-commodity depen- information and these skills. dent countries. Fast integrators also suffered smaller Fas t-it'a'iltil, ounodity-dependent exporters achi t d bettter ou tcomS Table 4-2 Integration, policy, and performance in commodity-dependent developing countries, 1984-93 (percent) Connnodityi-dependent countries Non-coninodity Fast Slow Indicator dependent countries integrators in tegrators Number of countries 15 15 16 Budget deficit as a share of GDP -4.98 -4.29 -7.09 Budget deficit volatility 2.60 2.65 5.82 Inflationa 14.00 13.94 7.25 Inflation volatilitya 0.53 0.57 1.82 Real per capita GDP growth 1.24 2.09 -1.80 Per capita investment growth -0.45 3.35 -7.20 Note: Commodity reliance is defined as countries with at least half their exports in primary commodities. See chapter 2 for definitions of fast and slow integrators. a. Measured by the consumer price index. Source: World Bank data and staff estimates. 49 terms of trade losses than slow integrators, thanks in importance needing to be preserved by a public sector part to much faster growth in the share of manufac- monopoly. But that strategy has frequently proved tures in exports over the period analyzed. counterproductive. Cross-country experience shows Perhaps surprisingly, there was no major difference that removing of a host of public sector interven- between fast and slow integrators within the com- tions-for example, in marketing, production, and modity-dependent group with respect to real price control-can produce large efficiency gains. exchange rate volatility, and the slow integrators actu- Examples of successful liberalization in commod- ally experienced lower (median) inflation and black ity sectors come not only from the dynamic economies market premiums. Part of the explanation lies in the of East Asia (which were heavily commodity reliant a CFA franc zone countries, which feature prominently few decades ago) but also from fast-growing countries among the slow integrators. Throughout the period from Chile to Uganda. A number of other countries in the CFA franc was stable in value against the French Africa, South Asia, and Latin America have also franc, even though it became progressively overval- started to liberalize. ued in real terms, and the CFA countries exhibited Market liberalization is often accompanied by elimi- slow growth and growing dependency on foreign aid. nation or privatization of monopolistic state-owned In addition to low investment, other causes of enterprises. For example, one study of cocoa marketing dependency include mishandling of commodity price costs shows that in 1989 costs (excluding explicit taxes) booms, inefficient public sector processing, and, for a were two to three times higher in countries where mar- few large producers, an inability to expand produc- keting is controlled by state-owned enterprises, such as tion without inducing falls in prices (for instance, in C6te d'lvoire and Ghana, than in countries with liberal- bananas, cocoa, coffee, cotton, tea, and tobacco).' ized markets, such as Indonesia and Malaysia. At that time marketing costs in the first group accounted for as Fostering productivity much as half the selling price (figure 4-3). The privati- zation of marketing boards in several African countries Successful commodity producers, as discussed above, has been an important element in the market liberaliza- have established climates that are conducive to for- tion that revived many commodity subsectors. eign and domestic investment. This section outlines In the case of C6te d'lvoire a further element has some of the practices implemented by successful com- been the attainment of macroeconomic stability with modity producers using a series of case studies. These the devaluation of the CFA franc. Before devaluation practices include: C6te d'Ivoire was able to maintain a high market share * Strengthening the private sector and achieving large gains in efficiency by abolishing public sec- tor monopolies in production and marketing. S * Attracting foreign capital and technology by fos- tering market incentives in the private sector and by removing regulatory barriers, especially Figure 4-3 Cocoa marketing costs as a share of the to foreign firms. selling price, 1989 * Enhancing research and development and infra- Perccit structure capacities to maintain competitiveness. 60 * Promoting financial flexibility to strengthen export and inventory financing of commodities. 50 Strengthening the private sector 40 Private and public sectors have important comple- mentary roles to play in commodity-dependent coun- 30 tries. The private sector provides the engine of investment and productivity growth, while the public 20' sector provides a liberalized regulatory framework, physical infrastructure, an educated workforce, and 10k perhaps sponsorship of research and technology developments. Too often the public sector in com- 0 modity-dependent countries has cosseted production Cameroon Cte Ghana Malaysia Nigeria Brazil Indonesia I. d'Ivoire and protected it from effective competition. Large pri- mary commodity sectors have sometimes been viewed Source: World Bank staff estimates based on Ruf 1993. as the "family silver"-a strategic sector of national 50 State-owned enterprises are less efficient and profitable than their private counterparts Table 4-3 Tea productivity and profitability in selected producing countries, 1991 Sri Lanka India, private Indicator Public Private North South Kenya, private Yield (kilograms per hectare) 1,268 2,442 2,127 2,300 2,237 Intake per tea plucker (kilograms per day) 13.52 24.59 26.22 25.24 48.0 Labor per hectare 3.2 2.7 2.7 2.5 2.2 Cost of production (dollars per kilogram) 1.87 1.54 1.52 1.39 0.94 Revenue (dollars per hectare) 2,574 4,957 4,318 4,669 4,338 Gross profit (dollars per hectare) 203 1,196 1,085 1,472 2,438 Sourcc: World Bank data and staff estimates. despite overvaluation, largely because it adopted mar- The experiences of Ghana and Peru show thatdeclin- ket-based reforms in the cocoa sector. With devalua- ing performance in the gold mining sector can be turned tion, the benefits of these past reforms have had a around through privatization, legal stability, and a busi- chance to come through. ness climate that welcomes both domestic and foreign In Tanzania the market liberalization process that investors. In the early 1960s Ghana's annual gold pro- began a few years ago abolished the state-controlled duction was about 31 tons; by 1983 production had marketing boards and significantly changed the sys- dropped to 8.6 tons because of insufficient investment tem of export marketing and pricing for cashews, cof- in exploration and extraction technologies (figure 4-4). fee, and cotton. As a result a large number of private Low investment was mainly the result of a poor busi- traders, many of them foreign or foreign-affiliated, ness environment. In 1983 the government embarked became active in purchasing and exporting, particu- on a privatization program that included measures larly in cashews. Farmgate prices for cashews aimed at encouraging foreign direct investment, espe- increased sharply-more than doubling in real terms cially in the mining sector. Three state-owned mines between 1988/89 and 1991/92. Cashew production were divested and a public offering of government which had stood at 20,000 tons in the late 1980s, stock of the largest gold producer, Ashanti, was held. In increased to 47,600 tons by 1993/94, and export volume increased sixfold between 1990 and 1993. Farmgate Gbaia's gold productioniiumped m7cc iines werc prices also increased sharply for coffee, by about 40 per- ti cent (in real terms) between 1988/89 and 1992/93, at a time when world prices were declining. The low productivity of state-owned tea estates in Figure 4-4 Gold production in Ghana, 1974-94 Sri Lanka illustrates the desirability of effective priva- Metric toiis tization (table 4-3). In 1991 per hectare yields and gross profit per hectare on the state-owned estates were sig- nificantly lower than those of private estates and con- siderably lower than in other countries. In 1992 the Sri 40 Lankan government consolidated 422 loss-making, state-owned tea, rubber, and coconut estates into 22 joint stock companies and contracted management out 30 to private firms in an attempt to make them more effi- cient. Thus far the plan's objectives have been frus- 20 trated and the state-owned estates continue to record large deficits. The main reason for this failure seems to have been the absence of an institutional framework 10 for transferring effective control to the private compa- nies. The managing companies were not allowed to 0 ImW acquire ownership in the ventures, and because the 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 estates continued to be state-owned, wage costs eaesed c d to b Source: World Bureau of Metal Statistics, World Metal Statistics increased substantially. Moreover, the financial incen- Yearbook. tives for investment were not sufficiently attractive. 51 addition, a stronger legal and regulatory framework keting systems to host countries to produce and was established and government mining institutions export products that match the particular tastes of were reformed. As a result of these reforms gold pro- Japanese consumers. In most cases these products are duction (excluding that of small miners) soared from 8.8 exported exclusively to the Japanese market through tons in 1984 to nearly 45 tons in 1994. The introduction marketing channels that are often arranged up to the of the Small-Scale Gold Mining Law of 1989 resulted in retail level by Japanese investors. Shrimp in China, an increase in government purchases of gold from Indonesia, and Thailand, vegetables in China and small-scale miners from 0.3 ton in 1989 to 1.0 ton in 1993, Vietnam, and processed (tinned) fruits in Indonesia, or from $3.7 million to $12.6 million in dollar terms. Malaysia, and Thailand are the most sizable examples In Peru, as in Ghana, the mining sector was stag- of this type of investment. nant for a long time. In 1991 the government initiated The risks of shutting out foreign technologies are a radical stabilization and structural reform program. shown by the use of seed lists, which often constitute a The Free Market Marketing Decree promotes privati- counterproductive barrier to the import of agricultural zation and provides a favorable legal and financial technology. Governments in many countries, particu- environment for foreign investors, including allowing larly developing countries, maintain lists of plant vari- foreign acquisition of domestic mining concerns. In eties for which commercial trade in seeds and other 1994 foreign acquisitions of mining enterprises in planting materials is allowed; selling seeds of unlisted Peru totaled $714 million and investment commit- varieties is illegal. These lists limit farmer access to new ments totaled $2.2 billion. varieties from domestic as well as foreign research. Seed varieties are tested and listed for the avowed pur- Applying foreign technology pose of protecting farmers from economic loss with underperforming varieties. In other words, govern- Commodity-dependent fast integrators have generally ments select technologies. For example, in 1993 the succeeded in attracting foreign capital and technology government of Malawi allowed just one private maize to develop new export commodities. The lesson to be hybrid (which is about twenty years old) along with drawn from the following cases is that foreign technol- five maize hybrids from government research, whereas ogy is best utilized when prices have been liberalized the OECD listed more than 400 new maize varieties. and excessive regulatory barriers removed. Variety limits on seed imports and domestic trade The chicken industry in Thailand provides a good can have an impact on domestic production and example of technology and knowledge transfer from income that dwarfs any impact on seed imports. In foreign investor to recipient. During the early 1970s Turkey seed trade liberalization and related reforms several Thai feed milling enterprises merged and allowed private companies to introduce hybrid maize expanded into poultry production. They developed varieties in the mid-1980s. Within four years private contractual ties with existing producers; built slaugh- hybrids had spread to roughly a third of maize- terhouses and modern processing facilities to produce planted areas and national average maize yields had value-added chicken cuts that met Japanese con- doubled. Research suggests that seed reforms pro- sumers' taste; imported technology from Japanese vided Turkish farmers with $100 million a year in addi- investors in the areas of genetics, nutrition, and dis- tional income from private maize hybrids alone, with ease control; and obtained technical assistance and gains in other crops (Gisselquist and Pray 1995). The capital from those investors. The government offered reforms generated a $1 million increase in average net investment incentives and research and quality con- imports of hybrid maize seed during 1983-85, but once trol services. Thailand also adopted the streamlined, the Turkish market for maize hybrids was established, vertically integrated production, processing, and domestically produced seed took over the domestic overseas distribution systems developed by Japanese market and maize seed exports surged. A similar tran- companies. As a result the Thai chicken industry sition occurred roughly a decade earlier in Chile, with became one of the largest, most cost-efficient, and hybrid maize seed imports giving way to net exports. technically advanced meat industries in the world, Effective seed trade liberalization-allowing foreign with exports of $351 million in 1993. seed to enter and compete within domestic markets- The transfer of Japanese technology and know- cannot be achieved by simply doing away with import how to East Asian developing countries is of particu- controls. For one thing, phytosanitary controls on for- lar interest. Japanese foreign direct investment often eign pests and diseases will remain even after compre- takes a distinctive form, especially if undertaken in hensive reforms eliminate seed import barriers. Also, Asia, where it is called kailiatsu-yunuu. Its main char- government variety lists are often written as limits on acteristic is that Japanese investors, in many cases domestic seed trade, so import liberalization alone does major trading companies, bring most of the input not remove restrictions that block imports and sales of materials as well as production, processing, and mar- seed varieties that are new to the country. Aside from 52 quarantine restrictions, many governments require and Indonesia and the potential growth of coffee in that imported seed be certified, whereas domestic seed Uganda (see below) are examples that underscore the is allowed to be sold without certification. importance of research in commodities. The main rea- Governments that want to achieve effective liber- son Southeast Asian countries have been competitive alization of seed trade must recognize and communi- in the export of commodities despite rising labor costs cate with national and international associations of is the sharp increase in productivity these countries private seed companies, do away with compulsory have achieved through research, extension, and variety registration, review and revise phytosanitary importation of new technologies. restrictions to focus more effectively on realistic pest The importance of agricultural research in main- and disease threats, and eliminate other seed import taining competitiveness is illustrated by Malaysia's barriers that have nothing to do with disease control. performance in the palm oil industry. Over the past The private sector and governments have worked thirty years the world market for palm oil has grown together in some prominent cases of rapid commodity from 1.3 million tons to 12.2 million tons. In 1961-63 production and export expansion (figure 4-5). Success Malaysia and Indonesia together accounted for just 17 has been due largely to transfers of foreign capital and percent of world palm oil production. Africa, on the technology (at least at the initial stage), ranging from other hand, produced about 75 percent, with Nigeria material inputs in production (seeds, tractors, chemi- and Zaire accounting for the majority of African pro- cals) to transfer of capital into commodity production duction. By 1992-93 Malaysia's share of world palm oil or processing. Government contributions to these suc- production had risen to 51 percent while Indonesia's cess stories include funding for research and develop- share rose to 24 percent and Africa's share fell to 14 per- ment and physical infrastructure, complementing the cent. (figure 4-6). The stagnation of African production role of the private sector. is due in large part to the destruction and abandon- ment of palm grove processing mills during the Developing research and infrastructure Nigerian civil war. The surge of palm oil production in Malaysia and The public sector has a role to play in research, exten- Indonesia, on the other hand, is due to high rates of sion, and importation of new technologies. Rapid investment in the sector. Malaysia's palm oil indus- growth in the cocoa and palm oil exports of Malaysia try, for example, is profitable largely because of rapid and sustained gains in yields (even after allowance for some unsustainable elements in pro- duction growth). Research efforts-public and pri- Governments and foreign firms play important roles in vate-resulted in the development of planting developing country exports materials and techniques that were well-suited to the climates of both countries. As an example, pro- Figure 4-5 Contribution of governments and duction was boosted in 1982 by releasing weevils to foreign firms to the development of nontraditional improve pollination; as a result oil extraction rates commodity exports from the palm oil kernel rose by 33 percent. Malaysian growers routinely generate twice as Government Foreign tirms Bth much palm oil per hectare (adjusting for the life cycle of trees) as African producers. Research and development Cotton production in francophone Africa experi- Pya enced sharp growth as a result of the activities of the n cture semipublic enterprise, Conpagnie frangaise pour le Techno og developpement des fibres textiles. In addition to equity participation, the company played a significant role in aroduction the provision of research, technical assistance, and con- sultant services aimed at enhancing productivity As a Produ on result average cotton vields per hectare and the rate of ginning production (the amount of fiber per kilogram Proeg of seed cotton) in the region-which includes Benin, and mareig financing Burkina Faso, C6te d'Ivoire, and Mali-are generally BaZ Chile China Cooa Mexco Thnd Thaan higher than those in anglophone Africa. soybeans grapes fresh cut tomatoes poultry frozen The importance of the policy environment and the sh,imp Ct... ,, hrinip avoidance of unnecessary barriers to technology is evi- Source: Jaffee 1993. dent in the case of coffee in Uganda, which developed hybrid, high-yielding coffee varieties several decades 53 Malaisia and Indonesia have gained considerable ground in paln oil production Figure 4-6 Palm oil production, selected years, 1961 to 1993 1961/62 1981/82 1992/93 Others 8% Indonesia 10% Others 9% Indonesia 14% Others 11% Malaysia 7% . Indonesia 24% Africa 23%. Afrca14 Africa 75% Malaysia 54% Malaysia 51% Source: United Nations Food and Agriculture Organization data. ago. These new varieties were not made available to usage fees declined sharply. Finally, the lack of ade- farmers because of the social and political turmoil that quate transport systems to handle bulk goods is one of the country was undergoing and because excessive the main reasons investment in the mining sectors of government intervention in the subsector essentially slow integrators has been concentrated in metals and shut out private initiatives. With the recent liberaliza- minerals with high value-weight ratios, such as dia- tion of export markets, the private sector has started to monds and gold. play a major role in building nurseries and selling seedlings of high-yielding varieties to farmers. Promoting financial flexibility Another example of strategic development of non- traditional commodities comes from Chile, which has One of the main obstacles to the development of com- developed a number of new export products includ- modity sectors in slow integrators is their weak finan- ing salmon, flounder, forest products, and black- cial and risk management systems. The poor quality berries. These enterprises are driven by the research, of links to international financial markets is a major development, production, and marketing efforts of signal of this weakness. There is considerable scope Fundaci6n Chile, an organization funded equally by for increased use of market-based risk management the Chilean government and the private sector specif- instruments such as futures, options, and swaps, in ically to develop nontraditional exports. Once a new commodity-dependent developing countries. Such industry is developed, Fundaci6n sells it to the private instruments reduce short-term price uncertainty and sector. Because of its success other countries, including increase the predictability of revenues from the com- Bolivia and Colombia, have imitated this approach. modity exports of governments and private firms. Finally, the role of the public sector in developing This reduction in uncertainty and increase in pre- transport and communications infrastructure is evident dictability is especially important for the long-term in a number of areas, as illustrated by the association viability of private traders in these countries. between slow integrators and poor infrastructure (fig- In some of these countries many banks and other ure 4-7). Good transport and communications systems financial institutions are insolvent, inefficient, or both. have become critical to competitiveness not only in ser- As a result real interest rates tend to be extremely vice and manufacturing industries but also in com- high. In these circumstances the warehouse receipt modity markets. The rapid growth of horticulture system, which is widely used in industrial countries, exports, for example, in Chile, Colombia, and Kenya could significantly strengthen export and inventory would not have been possible without such systems. A financing for commodities. This system has already recent examination of the cocoa subsector in Indonesia been implemented in some developing countries, shows that the availability of adequate transport facili- including Brazil, Indonesia, and Mexico, and is under ties was one of the main reasons for the rapid expansion consideration in Poland and Uganda. of cocoa on the island of Sulaweis, where smallholder Warehouse receipts are used in industrial countries production of cocoa increased from 18,000 tons in 1984 as secure collateral to obtain financing for commodi- to 201,000 tons in 1993 (Akiyama and Nishio 1996). ties. They are often tradable in a secondary market. Similarly, Chile's commodity exports benefited greatly The use of warehouse receipts is limited in develop- when the port labor unions were dissolved and port ing countries, however, because of the lack of appro- 54 Fast integration is underpinned byt investments in infrastructure Figure 4-7 Infrastructure development in four groups of countries, 1992 Electric power production Telephone main lines per 1,000 people Paved road density per million people Kilowatt hours per person Numnber Kilometers 1,600 120 2,500 1,400 10 ,0 1,200102,0 1,000 80 1,500 800 6 600 401,000 400 20500 0 0 Fast Moderate Weak Slow Fast Moderate Weak Slow Fast Moderate Weak Slow integrator integrators integrators integrators integrators integrators integrators integrators integrators integrators integrators Integrators Note: Does not include countries of the former Soviet Union. Source: World Bank data and staff estimates. priate legal and institutional frameworks and, in system include the existence of verification and physi- some countries, government policies that crowd out cal controls, a system of warehouse certification, a sys- the private sector's incentive to store commodities. tem for grading, the existence of property and casualty For example, government-decreed fixed prices insurance, and provision of performance guarantees by throughout the crop year create a disincentive for warehouse owners. farmers to store and sell later. Warehouse receipts offer the alternative of market-based financing of agri- Conclusion cultural exports and inventories, especially where prefinancing may be limited by poor creditworthi- The country examples cited in this chapter demon- ness. strate that commodity-dependent countries are not With warehouse receipt financing, a local buyer condemned to low productivity and an inability to with initial working capital makes crop purchases and diversify production and exports. Although the deposits the crop in a bonded warehouse. The ware- volatility of commodity prices and their decline in real house registers the delivery and issues a receipt that terms pose serious problems, these problems can be is usually formally registered. The receipt can then be overcome by policies that strengthen the private sec- used to obtain credit from local banks. In practice the tor (for example, abolition of state monopolies), warehouse itself frequently operates as a bank, enter- encourage foreign direct investment (joint ventures in ing into repurchase agreements with the local buyers. production and export marketing), promote research Warehouse receipt systems offer a number of ben- and its application (minimizing restrictions on new efits to developing countries. First, they provide for a agricultural technology), and develop infrastructure secure collateral to lending institutions. Second, they (telecommunications reform). For these policies to be can form the basis for a developing domestic spot successful they must be complemented by efforts to market for countries in the process of market liberal- develop institutional and human resources. While ization. Third, warehouse receipts provide an alterna- there is no quick fix for the inadequate investment of tive to government involvement in physical markets the past, the potential gains from liberalization of and thus reduce the direct role of the government in commodity sectors adds urgency to the task. crop commercialization. Certain conditions are required to establish a ware- Note house receipt system. For warehouse receipts to be secure collateral, an appropriate legal and institutional 1. Because of low price and income elasticities and the environment must guarantee the sanctity of contracts high concentration of supply among a few producers, and give priority to the claim in case of lender default. expansion of output in these commodities induces stagnant Other preconditions for creating a warehouse receipt or even falling net revenues for the producers as a whole 55 Appendix 1 Regional economic prospects East Asia money supply. Annual growth averaging more than 8 percent in the 1990s has contributed to escalating Remarkable export performance and strong invest- wage pressuresand widening current account deficits ment continued to underpin East Asia's rapid output in Malaysia. Nevertheless, consumer price inflation growth in 1995. For the second year in a row the edged downward in these economies in 1995 and is region's GDP growth exceeded 9 percent, and almost expected to come down further. all countries continued to outstrip their respective Continued high domestic savings and even higher ten-year trend growth rates. Current estimates sug- rates of investment are the central assumptions gest that the region's export volumes surged 21 per- underlying the region's high-growth forecast. Partly cent over 1994 while imports rose 18 percent. This because of mounting needs for investments in infra- growth exerted a significant pull on the economies of structure, the regional share of investment in GDP is trading partners both within and outside the region. projected to increase from 36 percent during 1991-94 Rapid growth of world import demand and the com- (already much higher than the developing country petitive boost provided by the yen's appreciation in average of 25 percent) to more than 40 percent in the the first part of 1995 contributed to the region's export next ten years. External financing is expected to success and helped indirectly stimulate investment in amount to about 1.5 percent of GDP ($50 billion a export and domestic sectors. year) in the coming decade, up from the 1980s aver- The Chinese economy appears to be achieving a soft age of just over 1 percent. Foreign direct investment landing. Output growth slowed from the 13 percent (FDI), which topped $42 billion in 1994 and has pace sustained in the three years to 1994 but remained accounted for about half the region's foreign financ- over 10 percent in 1995, and a further mild slowdown ing in the 1990s, is expected to continue to provide the is projected in 1996. Consumer price inflation, which bulk of these requirements. China has received the peaked at 24 percent in 1994, is estimated to have come lion's share of the region's FDI but Indonesia, down to 17 percent in 1995. Tight monetary policy has, Malaysia, Thailand, and Vietnam also attracted sub- however, complicated the process of improving the stantial inflows. In Indonesia FDI approvals increased financial performance of indebted state-owned enter- fivefold in the two years to 1995 to about $40 billion, prises, and this remains an overriding challenge in fur- partly in response to new foreign investment liberal- ther reforming the economy. The official forecast for ization measures. The excellent credit ratings of East annual GDP growth is 8-9 percent for 1996-2000, Asian countries will allow them to tap international while inflation is targeted at less than 10 percent. bond markets regularly, and the region is also The Philippines' recovery continued to progress expected to account for the bulk of commercial bank last year, with output growth estimated to have loans to developing countries. topped 5 percent for the first time since 1989. Despite Growth in the region is forecast to slow slightly weaknesses in agricultural production, investment over the next ten years, though only in line with slow- growth contributed strongly to the advance in output. ing population growth; per capita incomes are Nevertheless, the country's domestic savings rate, expected to increase by an annual average of 6.9 per- near 14 percent of GDP, remains less than half that of cent, the same as in 1986-95 (table Al-1). The income its Asian neighbors, where savings rates average more advance will be accompanied by further strong than 35 percent of GDP. progress in countries' integration with global and To offset the effect of continuous fast growth on regional economies (figure Al-1). Various regional inflation the authorities in Indonesia, the Republic of and unilateral trade liberalization measures (such as Korea, and Thailand continue to keep a tight grip on China's recent announcement of planned tariff reduc- 56 made it more resilient (though not impervious) to East Asia's growth will remain impressive cyclical shocks emanating from outside the region. Table Al-1 East Asia forecast summary South Asia (percentage annual growth) Indicator 1986-95 Baseline 1996-2005 Despite political strains countries in South Asia posted respectable growth in 1995. India, which GDP 8.5 7.9 accounts for 80 percent of the region's GDP, continued Consumption per capita 5.6 6.8 to draw substantial benefits from the trade and invest- GDP per capita 6.9 6.9 Expr oluma 13.4 10.1 ment liberalization undertaken after the 1991 balance Median inflationb 1.8 .8 of payments crisis. Helped by favorable monsoons, Current account/GDP -0.7 -1.5 output growth in the fiscal year to March 1995 reached se-es - - --6.3 percent and is expected to remain in a 5.5-6.0 a. Goods and nonfactor services. percent range in the year to March 1996. Exports in Source: World Bank baseline forecast, February 1996. dollar terms grew around 20 percent while manufac- turing output growth averaged more than 10 percent in the past two years. In Pakistan growth improved to an estimated 4 percent in the year to June 1995. tions) and growing investment links within the region Political difficulties reduced investor confidence in are expected to continue to fuel intraregional trade, Bangladesh and held back growth to 4-5 percent. In which has advanced at rates in excess of 20 percent a Sri Lanka continued military operations against Tamil year in current U.S. dollar terms in the 1990s. separatists dampened private sector dynamism and There are downside risks to this favorable outlook held growth to an estimated 5 percent. Inflation edged but they are, on balance, modest. In some countries higher throughout the region except in India, largely infrastructure constraints may not be addressed in a reflection of missed fiscal targets. Pakistan and Sri time, and risks to macroeconomic stability emanating Lanka did not meet the International Monetary from large budget deficits or overheating might Fund's (IMF) Extended Structural Adjustment undermine investor confidence. The region's strong Facility criteria to draw on final loan installments and competitive position in export markets and its grow- also had the highest inflation in the region, above 10 ing reliance on intraregional demand and trade have percent.' Fast-integrating East Asia will post rapid gains in per capita incomes Figure Al-1 Integration levels and per capita incomes in East Asia, 1970-2005 Real trade/GDP ratio Real per capita income/OECD per capita income Index (1970=100) Index (1970=100) Forecast Forecast 300 500 2500 400, 200 China 300 150 Asian NIEs 200 IAsian NIEs 10- - - - - - - I ASEAN 4 100 SEAN 4 50 Z7 I 0 I I I0 1970 1975 1980 1985 1990 1995 2000 2005 1970 1975 1980 1985 1990 1995 2000 2005 Source: World Bank data and staff projections. 57 Regional export performance remained vigorous, pages in meeting fiscal deficit targets will be cor- reflecting trade reforms, exchange rate adjustments, rected, something that would raise overall savings and buoyant world demand. Though liberalization of and investment rates by 4-5 percentage points for the intraregional trade moved a modest step forward region as a whole. A competitive and stable real with a preferential regional trade agreement covering exchange rate policy is also assumed to be in place, 222 items, the short-run impact on exports will be supporting 7.3 percent real export growth and imply- small since only 5 percent of trade is conducted within ing a gradually rising share in world trade. the region. Portfolio inflows to the region slowed to The external environment for South Asia in the $2.6 billion in 1995 from a perhaps unsustainable $3.9 coming decade is expected to be favorable. World billion in 1994. However, the overall level of private trade is expected to grow by nearly 7 percent in the capital inflows in 1995 remained at about the same near term and at a little over 6 percent in the longer level as in 1994 thanks to a sizable jump in FDI inflows term. Several sectors of South Asian comparative and approvals in India. FDI doubled in each of the advantage-clothing, textiles, processed foods, and past two years to reach an estimated $2.6 billion in fis- beverages-are set to expand because of the liberal- cal 1995. Aided by buoyant capital inflows, most ization under the Uruguay Round. Geographic prox- South Asian countries have replenished their foreign imity to the dynamic economies in East Asia will also exchange reserves, with India's amounting to several help trade. Oil prices are expected to fall significantly months worth of imports. in 1996-97 and to remain flat in real terms thereafter, Progress on privatization was mixed. The sale of a reassuring to a region that imports more than half its partial stake in Pakistan Telecommunications con- oil. Private capital flows to developing countries are tributed to portfolio inflows of $1.1 billion in fiscal currently around $160 billion and poised to expand at 1995, while in India revenues from telecom licensing perhaps 10 percent a year. With South Asia's share of fees are expected to be substantial. But privatization this total currently under 3 percent, there is significant was delayed in the power sector, and foreign partici- untapped potential for help in financing South Asian pation in oil exploration was slow to pick up. growth. The relatively low cost of English-speaking Privatization of Indian public enterprises is proceed- skilled labor throughout the region and the large size ing more slowly than anticipated, and less than 10 per- of domestic markets in India and Pakistan are some of cent of the targeted sums have been realized. This slow the main attractions for FDI. pace is one reason for the failure to meet fiscal targets. An important downside risk in the projections is The long-term outlook for South Asia projects that the trend toward lower fiscal deficits that began in growth averaging 5.4 percent in the next decade, 1991 seems to be faltering (figure A]-2). In India slow reaching close to 6 percent in the later years (table progress on privatization will contribute to keeping A1-2). The implied per capita GDP growth of 3.7 per- cent would represent a substantial improvement both on the past ten years and in terms of poverty reduc- Once-falling fiscal deficits are flattening out tion. This outlook assumes that the reforms initiated in the early 1990s will be pursued throughout the Figure A1-2 Budget deficits as a share of GDP period. In particular, it is assumed that recent slip- in selected South Asian countries, 1991-95 Percent A favorable external enironment will spur groweth 12 in South Asia Sri Lanka 10 Table A1-2 South Asia forecast summary (percentage aoiag )** . .** India Indicator 1986-95 Baseline 1996-2005 ** 6 ** GDP 5.1 5.4 Consumption per capita 2.4 3.1 4 Pakistan GDP per capita 2.9 3.7 Export volumea 9.2 7.3 Median inflation 9.1 6.4 2 Current account/GDP -2.1 -2.3 - O a. Goods and nonfactor services. 1991 1992 1993 1994 1995 b. GDP deflator Source: World Bank baseline forecast, February 1996. Source: World Bank data and staff estimates. 58 the general government fiscal deficit above 8 percent Brazil, Chile, Colombia, Mexico, and Venezuela- of GDP in fiscal 1995. In Pakistan poor tax collection, and often masks wide disparities in individual coun- cotton crop failures, and a weak economy kept the try growth rates. Nevertheless, the countries in the budget deficit ratio at 5.6 percent of GDP in 1995. In Sri region share common characteristics, namely rela- Lanka military spending kept the fiscal deficit ratio tively low ratios of domestic savings to GDP (aver- close to 10 percent. Coming after a period of fiscal lax- aging less than 20 percent) and a high concentration ity in the 1980s, the room for maneuver in these coun- of primary commodities in merchandise exports. In tries is limited. As much as a third of government recent years tight fiscal and monetary policies cou- revenue currently goes for interest on debt, under- pled with a nominal exchange rate anchor have led to scoring the formidable size of structural fiscal deficits. successful stabilization in a number of countries and A second area of risk is policy toward the private resulted in a decline in average inflation for the sector, in particular its role in easing infrastructure region. In this environment of declining domestic bottlenecks in such areas as power generation and inflation and improved trade prospects (NAFTA, ports. Given illustrative estimates of infrastructure Mercosur), expectations were raised for a higher level investment that can reach up to $500 billion over the of future income. In Argentina and Mexico general next ten years and savings rates around 21 percent government balances moved from deficits of around (compared with 35 percent in East Asia), a positive 2.5 percent of GDP in 1990 to zero by 1994, but over- and credible policy toward FDJ is critical. A third area all public sector consumption also rose in both coun- of risk is good governance to enable a business envi- tries. With the fall in private savings these countries' ronment conducive to long-term development. Good current account balances deteriorated by 4.5-6.0 per- governance goes beyond maintaining law and order centage points of GDP between 1990 and 1994, while to include transparent rules of the game and public real effective exchange rates rose by 68 percent in accountability, so that the fruits of development are Argentina and 23 percent in Mexico. widely shared. A fourth area of risk or constraint over Latin America experienced accelerating inflation the long term is the low stock of human capital in from 1970 to the early 1990s, and most countries in South Asia, with average schooling of about three the region only began addressing this problem in years compared with higher levels in East Asia. 1990 (figure AI-3). Median inflation for the region was more than 20 percent a year in 1985-94 with a Latin America and the Caribbean standard deviation of 12.5 percent compared with rates for East Asia of 6.8 percent a year and a 2 per- Mexico's balance of payments crisis in December 1994 cent standard deviation. A high-inflation, high- produced repercussions throughout the region in volatility environment reflects macroeconomic 1995. Argentina was immediately affected by dwin- instabilitv and is not conducive to raising private sav- dling private capital inflows in the first half of 1995, as ings and investment. Now that progress has been were most countries in the region except Chile and made on reducing inflation, macroeconomic stability Colombia. In the second half of the year most coun- has improved but remains fragile. Efforts to continue tries were able to regain access to private capital mar- this decelerating trend in inflation will require con- kets through the issuance of bonds but at much higher tinued tight fiscal and monetary policies for the rest spreads-an average of 336 basis points over LIBOR of the 1990s. compared with 285 basis points in 1991-94. In 1996-97 the region is expected to grow by an The fallout from the financial crisis led to steep reces- average of 2.6 percent a year compared with 3 percent sions in Argentina and Mexico. Output is estimated to a year in 1991-95. Recovery from the recession in have fallen by more than 3 percent in Argentina (where Mexico is expected to be sluggish in 1996 as domestic unemployment reached almost 20 percent) and by interest rates remain high to help stabilize the peso nearly 7 percent in Mexico. In Brazil output growth and restrain inflation, as well as to reflect the credit slowed to just over 4 percent due to extremely high real constraints caused by the banking crisis. Tight mone- interest rates. GDP growth for the region fell from 4.9 tary policy aimed at preserving the convertibility plan percent in 1994 to 0.9 percent in 1995 because of the in Argentina is likely to slow recovery in that country. steep recessions in Argentina and Mexico. Growth With Brazil exhibiting a strong commitment to lower- slowed in many other countries as well, as indicated by ing inflation, fiscal slippage in 1995 is likely to be the decline in median growth from 4.1 percent in 1994 addressed in 1996, while tight monetary policy will to 3.4 percent in 1995. Chile was the only country in the continue to slow domestic demand and overall region to experience rapid GDP growth, accelerating growth. Real coffee prices, especially important for from 4.1 percent in 1994 to around 8 percent in 1995. Central American economies, are projected to fall. The growth rate for the region as a whole tends to Terms of trade for the region as a whole are likely to be dominated by a few large countries-Argentina, decline 1.5-2.0 percent a year in the short term. 59 savings rates. Ratios of domestic savings to GDP are Latin America's growth should remain steadY; relatively low for most countries in the region when inflation will fall compared with the average for developing countries. In the wake of the Mexican peso crisis more countries Figure A1-3 Ratios of growth and inflation to GDP are turning their attention to this problem. Many have in Latin America and the Caribbean, 1970-2005 started to analyze the benefits of the Chilean model of Percent privatizing pension schemes. But implementing these new policies for raising private savings will take time, 6 and the fruits are unlikely to materialize in the short term. At the same time projections for foreign savings 5 (as measured by the current account deficit) indicate that the annual average will be about $5 billion less oo 20than the 1991-95 average. Investment rates are there- fore unlikely to rise rapidly over the forecast horizon, and growth will increasingly depend on productivity 3 Projected gains. 1990 Export volumes are projected to grow 6.1 percent a 2 year over the forecast period compared with 6.4 per- cent in the past ten years. The high concentration of primary commodities in the regional export basket together with a slowing of effective import demand in principal markets (from 7.3 percent a year in 1986-95 0 L 120 to 5.5 percent a year in the next decade) are the main 0 5 TO 15 20 25 30 reasons for the slowdown. I11flation (CPI) Europe and Central Asia Note: Figures are five-year moving averages. Source: IMF and World Bank data and World Bank staff projections. The fall in aggregate output in Europe and Central Asia eased to less than 1 percent in 1995 from 8.5 per- cent in 1994. The experience of different country GDP growth is projected to average 3.8 percent a groups in the region was, however, quite diverse.2 year in 1996-2005, or 2.3 percent on a per capita basis Turkey rebounded from its financing difficulties and (table A1-3). In the longer term (2001-2005) growth is was buoyed by the recovery in the European Union. expected to pick up to 4 percent. This would be a Among transition economies, growth in Central and marked improvement over the region's performance Eastern Europe and the Baltic states accelerated by 1.5 of the past fifteen years. Two factors impede even bet- points to 5 percent. Expansion in this group, ranging ter performance: inflation needs to be reduced further from some 7 percent in Poland to sluggish gains in and policies need to be implemented that raise low Bulgaria and Latvia, continued to be led by the strong performance of exports (which rose some 30 percent in dollar terms), a 20 percent increase in real private The Mcxican crisis will not impedc long-tcri, recoUTr investment, and further growth in emerging services sectors. But the diversity of country performance was Table Al-3 Latin America and the Caribbean highlighted by developments in Hungary and the erecat snuar Czech Republic: sharp austerity measures were t arequired in Hungary to stem a potential balance of Indicator 1986-95 Baseline 1996-2005 payments crisis, while the Czech Republic became the GDP 2.4 3.8 first Eastern European country to join the OECD Consumption per capita 0.8 1.5 group of industrial economies. GDP per capita 0.4 2.3 Despite pre-election political uncertainties, eco- Export volumea 6.4 6.1 nomic developments in Russia were largely encour- Median inflationb 20.4 8.7 aging during 1995. The output decline slowed to 4 Current account/GDP -1.7 -1.5 percent after surpassing 12 percent in 1994. Progress a. Goods and nonfactor services. in stabilization was apparent as inflation (3.2 percent b. CDP deflator. a month in December) and budget deficits (4.8 per- Source: World Bank baseline forecast, February 1996. cent of GDP) reached postreform lows, while the ruble traded within a narrow band during the second 60 half of the year. Buoyant exports to markets beyond privatization has slowed, banking and financial sectors the republics of the former Soviet Union led to a $18 remain fragile, and issues such as the restructuring of billion trade surplus and, together with rapid large enterprises, pension system reform, and the pro- advances in construction and in private and informal vision of safety nets require urgent attention. The sector activity (the latter not fully recorded in official buildup of arrears in taxes, wages, and interenterprise statistics), contributed to the likely bottoming out of debts and accumulation of bad loans in the banking GDP. Conditions in Ukraine showed signs of system are symptoms that could undermine the improvement early in the year, but stabilization mea- progress achieved thus far and threaten prospects for sures were hampered by difficult relations between sustainable growth. For Central and Eastern Europe the president and parliament, and inflation reaccel- and the Baltics the goal of EU membership may serve erated and GDP fell 15 percent in the year. In as an important driving force to undertake the second- Armenia, Georgia, and Moldova the initial results of generation measures needed to move toward Western stronger reform efforts and rising economic activity European norms. It is hoped that as the fruits of stabi- are becoming evident. Finally, the Central Asian lization become apparent to today's lagging reformers, states have continued to exhibit diversity in policy a growing political constituency in favor of reform will and performance, ranging from stronger results in support more active policy. stabilization for Kazakstan and the Kyrgyz Republic Strong exports and private investment are likely to to slippage in Uzbekistan and little progress in support 4-5 percent growth for the Central and Tajikstan and Turkmenistan. Eastern European countries during 1996-a milestone Recent developments underscore the virtuous cir- for Poland, since it means that real incomes will return cle linking stabilization, integration, and growth for to pretransition levels (table Al-4). And as the benefits the leading reformers among the Central and Eastern European countries. By contrast, conditions akin to a vicious cycle of continued rapid inflation and output losses have afflicted countries that have been slow or lagging reformers (figure A1-4). Among advanced reformers, sound policies implemented over the past Figure A1-4 Developments and projections for few years have fostered an environment that is sup- real GDP, Europe and Central Asia, 1990-2005 portive of private sector activity-especially ser- Index (1989=100) vices-and domestic and foreign investment. Since 1990 the private sector's share of GDP in this group 150 has risen from 20 to 60 percent and private investment from 3.5 to 5.0 percent, while FDI has accumulated to $22 billion (with flows averaging a strong 2 percent of 120 Advant, reformem GDP). The improvement in macroeconomic perfor- mance has boosted creditworthiness (investment 90 grade ratings were recently granted to the Czech Republic and Poland) and has attracted private capi- tal in increasing amounts. Finally, stabilization has strengthened countries' abilities to liberalize trade and regulatory regimes under EU association agree- -Iow reformer ments and underpinned prospects for eventual mem- 30 . . . . bership in the Union. Indeed, integration through 3e&nal ension trade with the West has progressed rapidly: 60 percent of the region's exports and imports are now traded 0 with EU countries (twice the share in 1990), and links 0 through FDI have strengthened considerably-for example, foreign ventures now account for about 40 Note: Groupings are based on a system developed in World Bank percent of Hungary's exports to Western Europe. (forthcoming). Leading reformers include the Czech and Slovak Although the Central and Eastern European coun- Republics, Hungary, Poland, and Slovenia. Advanced reformers fries and several states of the former Soviet Union- include Albania, Bulgaria, Romania, and the Baltic states. Intermediate reformers include Kazakstan, the Kyrgyz Republic, including Georgia, the Kyrgyz Republic, Moldova, and Moldova, and Russia. Slow reformers include Belarus, Russia-have made solid progress in stabilization, and Turkmenistan, Ukraine, and Uzbekistan. Countries afflicted by though recovery (or bottoming out) is now more regional tensions include the Transcaucasus and several republics of former Yugoslavia. broadly established, the agenda for reform in most Source: World Bank data and staff projections. transition economies remains large. In some countries 61 of stabilization in Bulgaria and the Baltics materialize enues will be required to meet debt service obliga- more fully, as Hungary emerges from near-term tions. Under a favorable policy scenario for Russia adjustment, and as the war-torn republics of the for- and other republics of the former Soviet Union the mer Yugoslavia are given respite, regional growth is current bottoming out of output declines could give likely to be maintained near 4 percent in the medium way to growth of 3-4 percent by 1997, and to longer- term. Though export growth is strong, import growth term advances near 5 percent. with recovery is likely to be stronger still (having risen about 40 percent in dollar terms in 1995) and trade Middle East and North Africa deficits will widen. But these trends should prove sus- tainable, since private capital continues to be attracted Preliminary estimates indicate a limited upturn to 2.5 to the region. For a number of countries integration percent growth in the Middle East and North Africa with the European Union could provide support for region in 1995. In part the improvement can be attrib- growth in the longer term-assuming that a founda- uted to somewhat higher growth in the major oil tion of sound structural policies continues to be built exporters-Algeria, Iran, and Saudi Arabia-tied to a over the next few years. This scenario suggests that the modest increases in oil prices, which averaged 3.5 per- leading reformers will have achieved pretransition cent above 1994 levels. But prices weakened signifi- income levels by 1998, and that by 2005 all the Central cantly in the latter part of 1995, and current and Eastern European countries will have done so. projections suggest that oil exporters are likely to face Despite political and other uncertainties and a longer-term trend of gradual erosion in real prices. assuming sustained stabilization efforts, Russia Real oil prices fell by nearly 8 percent a year in should achieve positive growth in 1996 and Ukraine, 1981-95. The possibility of a further decline is a sig- in 1997. As in Eastern Europe, a rise in consumer nificant risk to the long-term forecast, which expects spending may provide the initial impetus for growth. growth in the region to pick up to about 3 percent from Exports of primary products, supported by the 2 percent in 1991-95 (table A1-5). Other key issues removal of impediments to oil and gas trade, should affecting the long-run outlook include progress on also contribute to medium-term recovery in Russia. structural reforms, which up to now has been spo- But policy challenges are abundant. For growth to be radic at best; greater integration with the world econ- sustainable in the longer run, domestic and foreign omy, especially in terms of formal arrangements with investment will need to be supported by successful the European Union; progress in the regional peace stabilization, fiscal reform, and a credible legal and process; and institutional development to alleviate institutional environment. And gains in export rev- high internal sociopolitical tensions. Regional oil output has been nearly flat in the past few years due mainly to a steady increase in non- Reforms and stabilit' will generate OPEC supply. This trend shows little sign of abating sustainable regional growth in the near future and, in conjunction with high levels of oil stocks and projections for fairly moderate Table Al-4 Europe and Central Asia forecast increases in demand, indicates that oil prices may be summary (percentage anntual growth) A recovery in growth rests on faster progress Indicator 1986-95 Baseline 1996-2005 o1 economic oan1d structural reforms GDP -3.2 4.3 Consumption per capita -2.5 3.6 Table Al-5 Middle East and North Africa forecast GDP per capita -3.8 3.7 summary Export volumea 0.7 5.2 (percentage annual growth) Median inflationb 25.9 18.1 Current account/CDP -0.5 -0.5 Indicator 1986-95 Baseline 1996-2005 Memnorandun items GDP 2.0 2.9 Middle-income Consumption per capita -1.0 0.5 Western Europe GDP 3.4 3.6 GDP per capita -0.9 0.4 Central and Eastern Export volume" 6.0 3.8 Europe GDP -1.3 4.1 Median inflationb 8.1 6.3 Former Soviet Union GDP -5.1 4.6 Current account/GDP -4.2 -1.6 a. Goods and nonfactor services. a. Goods and nonfactor services. b. GDP deflator. b. GDP deflator. Source: World Bank baseline forecast, February 1996. Source: World Bank baseline forecast, February 1996. 62 subject to significant pressure during 1996-2005. In 1995 major steps were taken toward the normal- Current projections anticipate declines in real oil ization of relations between Israel and individual prices of 7 percent in 1996 and 5 percent in 1997 before Arab countries, as well as the Palestinians, but diffi- stabilizing in the later years of the forecast period, cult obstacles remain. There was significant progress implying a 1.4 percent annual decline in real prices in Israeli-Palestinian negotiations, but negotiations on over the next decade (see chapter 1). These projections the most difficult issues-including water rights and represent a considerable worsening of the prospects the final status of Jerusalem-are likely to be pro- for oil-producing countries than those in last year's tracted. And domestic opposition to a lasting peace Global Economic Prospects, which expected real oil persists on both sides. A notable development is a sig- prices to increase by 1.3 a year over the next decade. nificant upturn in tourism in some countries. This year's projections will further tighten the finan- Reflecting improved perceptions of security, the first cial constraints on the region's oil-producing coun- five months of 1995 saw the number of tourists to tries and are likely to make necessary difficult fiscal Egypt increase by 24 percent and to Jordan, by 36 per- and other adjustments. This increasingly less favor- cent over the same period in 1994. Syria, which earned able environment implies an urgent need for struc- $700 million from two million tourists in 1994, expects tural adjustments and reforms. to have increased its tourism revenues by 14 percent Economic outcomes in 1995 were also affected by in 1995. erratic weather-related factors, adding significantly to Current projections (which assume the return of the volatility of growth. Among the region's more Iraq to the world oil market) are for regional GDP to diversified economies (Egypt, Jordan, Morocco, and register annual growth of 2.9 percent during the next Tunisia) severe drought contributed to a 5 percent decade, compared with 2 percent during the past ten decline in Moroccan GDP while reducing growth in years (figure AI-5). As oil export volumes rise by Tunisia. Improved rainfall during the planting season, 2.5-3.5 percent over the period, modest regional however, is expected to support a rebound of 7-10 spillover effects from worker remittances and other percent in 1996 in both countries. Among the oil income flows should follow. In addition, the emer- exporters better weather conditions and a surge in gence of new natural gas pipelines to Europe will sig- agricultural output contributed to 3-4 percent growth nificantly boost Algeria's export revenues and help in Algeria after two years of output declines, a result support GDP growth in the range of 4 percent or more, aided by increased hydrocarbon exports to Europe and some abatement of sociopolitical tensions. Still, longer-term growth performance will require further Oil prices and GDP grow1th will remain steady4 progress in macroeconomic stabilization and struc- tural reform as well as closer integration with the Figure Al-S Real oil price and GDP growth in the world economy. Middle East and North Africa, 1970-2005 Morocco and Tunisia have signed important eco- nomic agreements with the European Union in the Oil price index - - GDP growth rate context of the Union's Mediterranean Initiative, (1975=100) (percent) which aims at closer economic and political links with 250 15 the southern Mediterranean countries (see chapter 3). 12 Several other countries in the region are expected to sign similar agreements in the near future. In 200 9 November 1995 representatives from the European Union and the southern Mediterranean countries 6 signed a declaration calling for regular political dia- 150 logue, enhanced EU aid, cooperation to control immi- 3 gration to the Union, and a schedule for the elimination of trade barriers on industrial goods by 100 0 2010. The Union also pledged $6 billion in grant aid -3 and another $6 billion in loans for education and infrastructure projects over the next five years. These 50 agreements are expected to lead to an across-the- board trade liberalization in the long run that should 09 lead to export expansion, increased investment, and ovrl efiiec gans 1970 1975 1980 1985 1990 1995 2000 2005 overall efficiency gamns. Political issues-especially the politics of the Arab- Source: World Bank data and staff projections. Israeli peace process-will continue to be important. 63 despite political uncertainties. Morocco and Tunisia result of drought, which has had a sporadic but seri- are best positioned to benefit from the continued ous effect on output over the past decade. strong recovery of EU import demand, both for non- A second main contribution to higher African fuel merchandise and for services such as tourism. growth came from the CFA countries in West and Although it will take a number of years for the direct Central Africa, which conducted a major realignment effects of the EU's Mediterranean Initiative to become of their currencies two years ago. Exports have gener- evident, it is hoped that participants' commitments to ally responded well to the boost in competitiveness, reform will set the stage for improved competitive- and there are clear signs of a rise in intra-CFA trade, ness and higher FDI inflows in the medium term. primarily through increased imports of coastal coun- tries from the inland. Inflation has been well controlled Sub-Saharan Africa since the initial pass-through, with eleven of thirteen countries enjoying annual inflation rates of less than 5 Preliminary estimates suggest that regional growth in percent. The success of the devaluation has varied Sub-Saharan Africa picked up to 3.5-4.0 percent in with the extent to which it was complemented by sup- 1995. If confirmed, this would represent the highest porting reforms and adjustments, with the West growth so far in the 1990s (growth in 1991-94 aver- African group of CFA countries, centered on C6te aged 0.7 percent) and the first year of positive per d'Ivoire, appearing set to benefit most from export-led capita GDP growth since 1989. Growth also appears growth. Countries in the CFA zone also benefited from to have been more widespread, exceeding 3 percent in favorable movements in their commodity export nearly thirty countries. This moderately heartening prices over the past two years. Average coffee prices in outcome in a region badly in need of good news was 1994-95 were more than twice their level in the previ- the result of a combination of factors. Some, such as ous two years, cocoa prices were more than 50 percent the high commodity prices of the past two years, are higher, and cotton prices were more than 30 percent likely to be short term in nature. Others, such as bet- higher. The small upturn in oil prices early in 1995 was ter economic policies, political transition in South also helpful for several oil exporters. Africa, and greater civil peace in some areas, ought to Prospects for Nigeria, the second largest economy be more sustainable. in the region, remain uncertain. The domestic political In South Africa, the region's largest economy, the situation is still volatile, and the country faces the pos- recovery accompanying the transition from apartheid sibility of international sanctions. But oil production continued for a second year, with growth rising to rebounded from the strike-induced lows of 1994 and nearly 3.5 percent. Capital inflows have increased agricultural production also showed some increase. from low apartheid-era levels, but a large portion While the country's ability to sustain a recovery is still appears to be of a short-term and portfolio nature. doubtful, policies pursued in 1995 and the recently This probably reflects concerns about relatively high announced 1996 budget may provide a foundation for and downwardly rigid unit labor costs and social ten- more stable growth and controlled inflation. sions (as evidenced by rising crime rates). The coun- High commodity prices also boosted growth in try is committed to trade liberalization under GATT East African nations such as Kenya, Tanzania, and and the WTO and is considering a free trade agree- Uganda. Several of these countries, Uganda and ment with the European Union. If it can come to terms Tanzania in particular, are in a position to make sig- with difficult policy issues aimed at improving labor nificant progress in exchange rate and trade policy market flexibility, reducing unemployment, and deal- reform and the transition to addressing second-gen- ing with social tensions without endangering macro- eration macroeconomic and fiscal policy issues. economic stability, South Africa has the potential to Uganda is estimated to have grown by nearly 10 per- grow at rates much higher than the current projections cent in 1995, and if the pace of reforms is maintained of 3-4 percent a year. is expected to achieve long-term growth rates of more But even this modest growth, if it endures, should than 5 percent a year. carry spillover benefits-in part through increased On balance the largely encouraging developments trade and investment flows-for the Southern Africa of 1995 could help set the stage for better medium- and subregion, the prospects for which have also bright- long-term growth prospects. But such projections are ened with movement toward civil peace. dependent on a number of assumptions, important Mozambique appears to have settled down to normal among which are expectations for commodity prices economic activity, and the current d6tente between and the region's terms of trade. Sharp declines in terms government and opposition forces in Angola signifies of trade during 1980-94-averaging almost 4 percent the first time in a generation that the subregion has a year-represented an annual loss of income for the been without a war. But not all countries in the area subcontinent of 0.8 percent (assuming a ratio of enjoyed higher growth: GDP in Zimbabwe fell as a imports to GDP of 20 percent). Although the longer- 64 term trend decline in primary commodity prices is the early 1980s, and the successful CFA parity realign- likely to continue, the pace of decline should moder- ment of January 1994 resulted in a real depreciation of ate, helping to reduce the degree of drag on growth about 30 percent after the initial inflation pass- (figure Al-6). Regional terms of trade declines should through. Progress has been slower in other key areas, also slow, to about 1 percent a year. Nevertheless, since including fiscal policy, public and financial sector the bulk of projected oil and non-oil commodity price reform, and human resource and infrastructure devel- declines are projected to occur in 1996, the region's opment. The projections here assume that reforms will major commodity exporters may face a significant continue, though at no faster pace than over the past adjustment challenge during this period. Among more decade. Against this background output growth for favorable external factors are the steady growth paths the region over the next ten years is anticipated to aver- projected for world trade and for output in industrial age 3.8 percent (4.2 percent excluding South Africa and countries, including Europe, the region's main trading Nigeria), compared with 1.7 percent over 198645, and partner. The forecast also assumes a consolidation of should be underpinned by modest improvements in civil peace in subregions where breakthroughs have export performance and domestic investment (table recently been achieved, such as the Horn of Africa and AI-6). Southern Africa. Elsewhere, however, the situation remains precarious, as in strife-torn countries such as Notes Burundi, Liberia, Rwanda, and Sierra Leone. Improvements in policy implementation will be the 1. However, in late 1995 Pakistan agreed to a standby key determining factor in the longer-term outlook. arrangement with the IMF. Over the past decade many countries in the region 2. Europe and Central Asia spans a broad range of undertook significant liberalization of exchange rate economies across middle-income Western Europe (Greece, regimes and trade policy. Non-CFA countries have Malta, and Turkey), Eastern and Central Europe, and the generally been able to achieve competitiveness states of the former Soviet Union. See the Classifications of through large devaluations in real exchange rates since Economies tables at the end of this volume. Investment and terms of trade should improve nmarginally A promising outlook for a struggling region Figure A1-6 Sub-Saharan Africa investment/GDP Table A1-6 Sub-Saharan Africa forecast summary and terms of trade, 1975-2005 (percentage annual growth) Terms of trade Indicator 1986-95 Baseline 1996-2005 Percent index (1987=100) GDP 1.7 3.8 25 Investment/GDP 160 Consumption per capita -1.5 1.2 Forecast GDP per capita -1.2 0.9 Export volumea 2.6 4.6 Median inflationb 9.6 8.0 140 Current account/GDP -1.1 -2.2 20 a. Goods and nonfactor services. 120 b. GDP deflator. Source: World Bank baseline forecast, February 1996. 100 15 80 10 60 1975 1980 1985 1990 1995 2000 2005 Source: World Bank data and staff projections. 65 Appendix 2 The speed of integration index: Qualifications and anomalies The speed of integration index tabulated in table A2- Conversely, speed of integration rankings for I is derived purely from changes in various integra- countries such as Iran and Zimbabwe were boosted tion variables between the early 1980s and the early because they had large improvements in credit rat- 1990s rather than from the levels of those variables. ings from levels that were exceptionally low at the The interpretation of a particular change, however, start of the 1980s (mainly for political reasons), even often depends on the initial level of integration. For though ratings remained relatively low in the early example, most industrial countries already had high 1990s even with this increase. Nepal scored a high levels of the four integration variables (the ratio overall ranking on the basis of a large increase in its of real trade to GDP, the ratio of FDI to GDP, share of manufactures in exports, though from a Institutional Investor credit ratings, and the share of modest initial level. Zaire attracted little FDI in either manufactures in exports) in the early 1980s, and a the early 1980s or 1990s but did better on this score decline in one still left the country at a high level in than countries like Tunisia or Uruguay, where the the early 1990s. Thus despite a small decline in ratio of FDI to GDP fell over the period even though Switzerland's Institutional Investor credit rating over it was still significant in the early 1990s. Zaire pro- the period it had the highest rating in the world, vides an example of a sitiation where the ratio of meriting a different interpretation than a similarly trade to GDP rises in a declining economy because sized decline in, say, Congo. exports from a well-protected mineral enclave sector This issue has been partly addressed by treating the that provides government revenue fall less dramati- industrial countries as a separate category and defin- cally than overall GDP. Finally, problems in evaluat- ing the four quartiles (fast, moderate, weak, and slow) ing trade data for the former members of the Council only for developing countries. Nevertheless, it is a for Mutual Economic Assistance during the recent major factor influencing some apparent anomalies or period of transition (and assessing comparability counterintuitive placements in the ranking of coun- with earlier periods) can introduce bias in trade- tries in table A2-1. For example, China is pushed into related measures. The sharp decline in Bulgaria's the group of weak integrators largely because of a trade ratio is a case in point. decline in its Institutional Investor credit rating at the Table A2-2 presents underlying data for the level of end of the 1980s.' But even so, China's rating integration index referred to in the text. remained higher than that of all but a handful of developing countries. Another factor restraining Note China's rating was that, having had a high share of manufactures in exports to start with, the increase in 1. Without this decline in credit rating China's speed of this share was less than for many countries that integration ranking would have been twenty-six places started with a lower base of manufactures in exports. higher, near the top of the moderate integrator group. 66 Table A2-1 Speed of integration index and underlying indicators Change in Change in FDI Change in Speed of Change in real trade Institutional as a share of GDP manufacturing integration as a share of GDP Investor rating 1980-82 to export share Region/country index 1980-83 to 1990-93 1983-85 to 1993-95 1990-92 1981-83 to 1991-93 High income 0.31 1.33 -0.21 0.017 0.55 Australia -0.66 1.02 -1.62 -0.026 1.24 Austria 0.34 1.95 0.35 0.014 0.48 Belgium-Luxembourg 2.24 3.00 0.64 0.384 0.71 Canada 0.54 1.81 -0.42 0.094 1.26 Denmark 0.90 1.13 0.49 0.129 0.88 Finland 0.03 0.43 -0.22 0.018 0.98 France 0.93 0.91 0.79 0.147 0.42 Germany -0.07 1.30 -0.21 0.015 0.35 Hong Kong 2.29 19.10 -0.22 0.000 -0.12 Ireland 0.59 3.30 0.71 4.074 1.21 Italy 0.28 1.05 0.46 0.013 0.54 Japan -0.39 0.24 -0.29 0.003 0.12 Kuwait -0.47 3.51 -1.54 0.000 0.32 Netherlands 1.14 2.07 0.22 0.146 1.36 New Zealand -0.61 1.43 -0.67 -0.083 0.56 Norway -0.53 1.37 -0.77 -0.016 0.12 Singapore 3.52 11.11 0.15 0.367 1.93 Spain 1.07 1.54 1.22 0.099 0.52 Sweden 0.70 1.07 -0.04 0.172 0.54 Switzerland 0.09 0.94 -0.23 0.083 0.17 United Kingdom 0.28 1.02 -0.32 0.055 1.17 United States -0.28 0.78 -0.69 -0.019 1.04 Low and middle income -0.18 0.06 0.04 0.002 0.27 East Asia 0.77 1.37 -0.05 0.03 0.64 China -0.29 0.58 -0.86 0.041 0.61 Indonesia 0.81 -1.40 -0.05 0.021 3.81 Korea, Rep. of 0.63 1.37 1.30 0.011 0.26 Malaysia 1.80 5.74 -0.41 0.054 3.80 Myanmar -0.42 -0.21 -0.08 0.000 -0.06 Papua New Guinea -0.40 -1.11 -0.73 0.056 0.64 Philippines 0.99 1.84 0.13 0.021 2.69 Taiwan, China 0.77 2.72 0.97 0.028 0.36 Thailand 2.12 3.25 1.03 0.046 3.83 Europe and Central Asia 0.46 1.06 0.57 0.03 0.29 Bulgaria -1.73 -6.02 -1.64 0.014 0.64 Czechoslovakia, former 0.46 -0.07 0.57 0.097 0.38 Greece H.11 3.14 -0.48 0.031 0.29 Hungary 0.95 1.06 -0.18 0.289 0.05 Poland 0.58 2.10 1.57 0.048 -1.05 Portugal 1.89 3.61 1.58 0.150 0.94 Romania 0.25 0.63 0.76 0.011 0.24 Soviet Union, former -2.23 -0.13 -2.53 0.000 -1.94 Turkey 1.87 1.65 1.54 0.029 3.27 This table continues on the following page. 67 Table A2-1 Speed of integration index and underlying indicators (continued) Chang'e in Change in FDI Change in Speed of Change in real trade Institutional as a share of GDP manufacturing integration as a share of GDP Investor rating 1980-82 to export share Region/country index 1980-83 to 1990-93 1983-85 to 1993-95 1990-92 1987-83 to 1991-93 Latin America and the Caribbean -0.23 0.45 0.21 0.01 0.32 Argentina 0.59 0.37 0.21 0.141 0.56 Bolivia 0.33 0.69 0.69 0.007 0.58 Brazil -0.28 0.50 -0.80 -0.042 1.61 Chile 0.65 0.97 1.41 -0.011 0.59 Colombia -0.54 0.77 -0.98 -0.010 0.59 Costa Rica 0.73 2.82 1.07 0.034 -0.01 Dominican Republic -0.04 -0.34 0.39 0.037 0.00 Ecuador -0.52 0.28 -0.69 0.028 -0.02 El Salvador 0.08 -0.26 0.73 0.007 0.22 Guatemala -0.17 -0.32 0.51 -0.015 0.08 Haiti -0.37 0.42 -0.21 -0.008 0.10 Honduras -0.12 -0.52 0.39 0.022 0.04 Jamaica 1.19 3.31 0.54 0.184 0.06 Mexico 1.44 1.62 0.44 0.007 3.85 Nicaragua -0.28 -0.05 0.22 0.008 -0.22 Panama -0.74 0.45 -1.48 -0.029 1.11 Paraguay -0.25 2.36 -0.86 0.028 0.14 Peru -0.95 -1 41 -1.14 0.01 0.318 Trinidad and Tobago -0.23 3.12 -2.05 -0.01 2.051 Uruguay -0.34 1.06 0.19 -0.12 0.674 Venezuela -0.24 0.26 -0.78 0.04 0.820 Middle East and North Africa -0.19 -0.27 -0.39 0.00 0.180 Algeria -1.51 -0.97 -2.27 0.00 0.180 Egypt -0.19 -1.88 -0.70 -0.04 2.696 Iran 0.20 0.04 1.09 0.00 -0.001 Iraq -1.68 -4.85 -1.33 0.00 0.000 Israel 0.66 0.73 1.01 0.04 0.674 Jordan -0.39 2.39 -1.32 -0.02 0.880 Morocco 0.97 0.39 0.57 0.03 2.516 Oman -1.00 -1.61 022 -0.14 0.063 Saudi Arabia -3.40 1.31 -1.39 -0.62 0.010 Syria 0.42 -1.05 0.23 0.00 2.356 Tunisia 0.16 -0.27 -0 51 -0.08 3.339 United Arab Emirates -0.18 1.97 -0.27 0.00 0.000 Yemen -0.79 -1.63 na -0.01 -0.001 South Asia 0.87 0.05 -0.08 0.00 2.559 Bangladesh 0.32 0.06 1161 0.00 1.040 India 0.01 0.09 -0.45 0.00 1.613 Nepal 1.35 0.00 -0.08 0.00 5.112 Pakistan 0.87 0.05 0.63 0.01 2.559 Sri Lanka 0.95 -0.33 -(.25 0.00 4.271 68 Table A2-1 Speed of integration index and underlying indicators (continued) Change in Change in FDI Change in Speed of Change in real trade Institutional as a share of GDP manufacturing integration as a share of GDP Investor rating 1980-82 to export share Region/country index 1980-83 to 1990-93 1983-85 to 1993-95 1990-92 1981-83 to 1991-93 Sub-Saharan Africa -0.46 -0.36 -0.03 0.00 0.079 Angola -0.70 0.02 -0.19 -0.05 -0.209 Benin -0.83 -4.53 na 0.00 0.701 Botswana -0.87 -2.69 0.72 -0.17 0.601 Burkina Faso -0.17 -0.26 na -0.01 0.875 Burundi -0.47 -0.37 na 0.00 0.000 Cameroon -0.72 1.81 -1.31 -0.10 1.161 Central African Republic -0.46 -2.38 na -0.03 1.157 Chad -0.25 0.83 na 0.01 0.000 Congo -1.52 -2.83 -0.27 -0.12 -0.383 C6te d'Ivoire -1.08 -0.75 -1.44 0.01 0.116 Ethiopia -0.19 -0.11 0.04 0.01 0.317 Gabon -1.13 -0.20 -0.87 -0.08 -0.063 Ghana 0.56 4.49 0.49 0.00 0.042 Guinea -0.23 0.19 na 0.02 0.193 Kenya 0.00 -0.27 -0.18 -0.01 1.460 Lesotho -0.71 -3.28 na 0.00 0.451 Liberia -1.18 -1.77 -0.42 -0.08 -0.113 Madagascar -0.12 -1.43 na 0.03 1.107 Malawi -0.39 -0.35 0.08 0.01 -0.293 Mali -0.15 1.53 na -0.01 0.242 Mauritania -0.68 -1.44 na -0.03 0.118 Mauritius 2.35 3.82 1.90 0.01 3.478 Niger -0.73 -1.70 na -0.02 -0.045 Nigeria -1.87 -5.59 -1.49 0.00 0.033 Rwanda -0.07 0.45 na -0.04 1.340 Senegal -0.04 -0.46 0.31 -0.01 0.790 Sierra Leone -1.79 -3.53 -0.10 0.03 -3.057 Somalia -0.83 -3.37 na 0.02 -0.008 South Africa -0.79 0.16 -1.84 -0.01 1.350 Sudan --0.56 -1.01 -0.18 0.00 -0.005 Tanzania -0.16 -0.65 0.34 0.01 0.225 Togo -0.43 1.28 na -0.04 -0.153 Uganda -0.13 0.24 0.32 0.00 -0.002 Zaire 0.05 1.82 0.23 0.00 0.035 Zambia -0.56 -).29 0.13 0.01 0.000 Zimbabwe 0.03 0.31 0.79 0.01 -0.289 na is not available. Note: Regional values are the median for all countries in the region. Speed of integration index is an average of the standardized scores for the four underlying variables with a mean of 0 and a standard deviation of 1. Changes are in annual averages and for trade, FDI, and manu- factures exports are in percentage point terms. Source: Institutional Investor and World Bank data and staff estimates. 69 Table A2-2 Initial level of integration index and underlying indicators Initial level of Population-adjusted Institutional FDI as a share Manufacturing integration index trade ratio Investor rating of PPP GDPa export share Region/country 1981-83 1981-83 1981-83 1981-83 1981-83 High income 1.25 25.06 81.80 0.32 70.20 Australia 0.87 -0.17 89.30 1.70 22.00 Austria 1.26 18.51 83.60 0.32 84.47 Belgium-Luxembourg 1.87 77.80 78.07 1.25 74.11 Canada 0.81 19.67 90.53 -0.34 53.68 Denmark 0.79 25.98 71.17 0.18 56.79 Finland 0.96 16.34 75.80 0.13 74.35 France 1.16 18.83 85.10 0.33 73.59 Germany 1.46 25.90 94.73 0.19 86.79 Hong Kong 1.73 79.59 75.13 0.00 96.67 Ireland 1.19 39.27 68.40 0.95 64.47 Italy 1.00 11.13 72.57 0.18 84.07 Japan 1.41 11.82 95.33 0.03 96.71 Kuwait 0.51 52.75 69.20 0.00 19.05 Netherlands 1.60 71.18 87.83 1.07 49.36 New Zealand 0.34 -7.01 75.50 0.83 20.62 Norway 1.35 64.34 87.87 1.16 30.67 Singapore 5.88 295.66 78.23 8.80 56.51 Spain 0.79 -1.60 66.40 0.65 70.62 Sweden 1.24 30.89 80.00 0.29 78.87 Switzerland 1.58 27.97 96.10 0.23 92.51 United Kingdom 1.41 24.21 89.00 1.06 67.39 United States 1.26 10.52 97.23 0.51 69.79 Low and middle income -0.45 -13.69 39.20 0.14 12.96 East Asia 0.40 11.70 56.02 0.12 28.89 China 0.74 14.52 65.60 0.06 70.40 Indonesia -0.17 11.70 55.73 0.12 4.94 Korea, Rep. of 0.93 23.40 56.30 0.07 90.95 Malaysia 1.10 21.13 72.10 2.58 22.73 Myanmar -1.02 -17.65 na 0.00 4.52 Papua New Guinea 0.40 4.19 44.17 2.57 8.56 Philippines -0.04 -6.96 40.10 0.10 48.21 Taiwan, China 1.16 33.20 67.67 0.12 89.14 Thailand -0.06 -6.72 51.30 0.28 28.89 Europe and Central Asia 0.32 -3.72 50.73 0.00 69.49 Bulgaria 0.37 8.55 42.80 0.00 71.82 Czechoslovakia, former 0.51 -9.65 50.73 0.00 88.10 Greece 0.47 -8.14 58.77 0.89 51.06 Hungary 0.32 -0.62 51.83 0.00 63.41 Poland -0.09 -3.72 18.03 0.02 69.49 Portugal 0.49 -9.58 54.93 0.31 73.78 Romania 0.23 6.76 32.60 0.00 72.79 Soviet Union, former 0.28 5.24 63.73 0.00 41.64 Turkey -0.42 -6.71 17.63 0.07 42.28 70 Table A2-2 Initial level of integration index and underlying indicators (continued) Initial level of Population-adjusted Institutional FDI as a share Manufacturing integration index trade ratio Investor rating of PPP GDPa export share Region/country 1981-83 1981-83 1981-83 1981-83 1981-83 Latin America and the Caribbean -0.46 -20.05 40.07 0.32 14.13 Argentina -0.33 -21.49 48.03 0.29 20.09 Bolivia -1.07 -37.98 16.87 0.42 2.45 Brazil 0.15 -4.14 49.37 0.55 40.28 Chile -0.23 -13.18 50.17 0.76 7.46 Colombia -0.10 -20.05 56.67 0.47 24.03 Costa Rica -0.44 -24.15 23.90 0.75 29.01 Dominican Republic -0.71 -27.17 23.43 0.32 19.65 Ecuador -0.60 -20.31 44.23 0.22 2.45 El Salvador -0.70 -19.92 10.00 0.11 36.79 Guatemala -0.67 -28.53 17.87 0.47 26.80 Haiti -0.77 -35.72 na 0.14 41 78 Honduras -0.87 -17.07 16.40 0.21 9.74 Jamaica -0.28 2.26 16.43 -0.34 60.54 Mexico -0.02 -6.58 57.00 0.72 14.13 Nicaragua -1.23 -36.40 8.23 0.00 8.59 Panama -0.46 -13.69 40.53 0.34 10.68 Paraguay -0.61 -31.80 43.53 0.31 9.15 Peru -0.54 -18.69 39.60 0.14 14.07 Trinidad and Tobago 0.56 7.63 54.57 2.45 10.95 Uruguay -0.21 29.90 39.33 0.93 30.78 Venezuela -0.17 0.77 63.50 0.20 2.02 Middle East and North Africa -0.06 -5.11 39.25 0.26 9.52 Algeria -0.36 -0.74 55.93 -0.02 0.67 Egypt -0.42 -6.37 35.03 0.57 9.52 Iran -0.95 -5.53 12.93 0.00 3.56 Iraq -0.06 32.42 39.20 0.00 20.37 Israel 0.52 13.7b 3o.50 0.26 80.70 Jordan -0.05 -5.11 39.30 0.28 42.37 Morocco -0.33 -13.34 34.50 0.16 33.75 Oman 1.01 76.24 46.77 2.51 4.87 Saudi Arabia 2.36 66.04 73.03 6.24 0.81 Syria -0.81 -20.14 26.13 0.00 11.15 Tunisia 0.38 -7.39 47.07 1.51 39.49 United Arab Emirates 0.29 61.75 60.27 0.00 3.67 Yemen -0.57 -14.74 na 0.33 0.05 South Asia -0.30 -12.78 25.92 0.00 53.31 Bangladesh -0.30 -12.78 14.25 0.00 63.83 India 0.24 10.22 47.60 0.00 53.31 Nepal -0.41 -28.50 na 0.00 49.49 Pakistan -0.20 - 8.22 21.53 0.07 58.56 Sri Lanka -0.53 -21.73 30.30 0.23 25.58 This table continues on the following page. 71 Table A2-2 Initial level of integration index and underlying indicators (continued) Initial level of Population-adjusted Institutional FDI as a share Manufacturing integration index trade ratio Investor rating of PPP GDP' export share Region/country 1981-83 1981-83 1981-83 2981-83 1981-83 Sub-Saharan Africa -0.83 -20.55 18.50 0.11 7.57 Angola -0.45 -17.85 17.20 1.45 12.52 Benin -0.72 -32.31 na 0.02 30.59 Botswana 0.07 3.28 na 2.37 10.49 Burkina Faso -0.87 -27.27 na 0.06 12.96 Burundi -1.05 -32.46 na 0.04 2.29 Cameroon -0.50 -19.81 34.55 0.83 5.87 Central African Republic -0.62 -36.35 na 0.32 34.84 Chad -1.04 -33.70 na 0.00 5.27 Congo -0.29 16.67 15.73 1.23 7.66 C6te d'Ivoire -0.38 -2.20 40.20 0.28 10.83 Ethiopia -1.11 -16.52 9.97 0.01 0.95 Gabon -0.50 -27.86 35.57 1.08 4.79 Ghana -1.10 -37.74 na 0.08 0.78 Guinea -0.89 -29.01 na 0.00 14.27 Kenya -0.58 -13.69 35.83 0.11 11.68 Lesotho -0.87 -30.48 na 0.25 10.38 Liberia -0.77 -20.39 15.23 1.00 1.72 Madagascar -0.90 -20.72 na 0.00 7.48 Malawi -0.99 -26.73 20.93 0.04 5.82 Mali -0.90 -22.27 na 0.11 5.39 Mauritania -0.66 -8.49 na 0.54 4.57 Mauritius -0.84 -43.25 19.80 0.03 32.59 Niger -0.96 -25.56 na 0.16 1.73 Nigeria -0.01 24.80 50.23 0.63 1.03 Rwanda -0.75 -34.57 na 0.49 17.85 Senegal -0.82 -18.30 20.87 0.13 11.64 Sierra Leone -0.52 -27.82 12.13 0.21 52.69 Somalia -0.92 -8.42 na -0.16 0.49 South Africa 0.40 -8.61 59.97 0.09 64.76 Sudan -1.16 -21.54 10.30 0.00 0.78 Tanzania -0.91 -14.19 12.83 0.00 13.52 Togo -0.76 -21.52 na 0.37 9.67 Uganda -1.22 -19.94 5.23 0.00 0.29 Zaire -1.13 -18.62 6.13 0.00 5.89 Zambia -0.92 -4.42 13.20 0.15 0.75 Zimbabwe -0.50 -14.37 25.43 0.00 34.37 a. PPP is purchasing power parity. Note: Regional values are the median for all countries in the region. Level of integration index is an average of the standardized scores for the four underlying variables with a mean of 0 and a standard deviation of 1. Population-adjusted trade ratios are residuals from a regression of the ratio of trade to PPP GDP on population. Trade ratios, FDI, and manufactures are in percent terms. Credit ratings are index values. Source: Institutional Investor and World Bank data and staff estimates. 72 Appendix 3 Robustness of results The results in chapter 2 on the relation of pace of inte- relation of real exchange rate volatility to the speed of gration to growth and various policy measures were integration is stronger for the smaller data set of tested for robustness with respect to changes in several eighty-eight countries for which results are shown in conditions of the analysis. First, the analysis was table 2.4. In addition, real exchange volatility is sig- repeated using average levels of the integration vari- nificantly related to elements of the level of integra- ables over the past decade rather than changes in them. tion, as shown in figures 2.6a and 2.6b.) Inflation rates The composition of the integration index was also and inflation volatility do not, however, emerge as tested to see if the results were being dominated by one significantly different in either case because of a small of its constituent variables by dropping these con- number of extreme observations that inflate variance stituents from the index one at a time. Representative within groups. values for the various integration classes were also derived using averages rather than medians. In these Table A3-1 Summary of analysis of variance various tests, comparing fast integrators with weak and slow integrators taken together, the fast integrators Null hypothesis: equal means continued to have fiscal deficits that were 1.5 to 5.0 per- between (a) fast integrators and centage points of GDP smaller and about one standard (b) weak and slow integrators deviation less volatile than the weak and slow integra- F statistic tors, real exchange rate volatility around 40-50 percent (critical value at less, annual per-capita GDP growth that was 1-3 per- Variable 95% level=3.98) P value centage points higher, and so on. Budget deficit The statistical significance of differences between as a share of GDP 5.18 0.0259 the average values for various integration classes of Budget deficit volatility 11.27 0.0013 certain key variables was also tested using analysis of Consumer price inflation 0.68 0.4127 variance (table A3-1). It is possible to reject with 95 Consumer price volatility 1.05 0.3098 percent confidence the hypotheses that fast integra- Real exchange rate volatility 3.12 0.0819 tors on the one hand and weak and slow integrators Average tariff 3.98 0.0519 taken together on the other have the same average Terms of trade change 6.82 0.0111 values for budget deficits, budget deficit volatility, Per capita real GDP growth 18.53 0.0000 per-capita GDP growth and volatility, and terms of Volatility in per capita trade changes. The hypotheses suggesting equal GDP growth 11.96 0.0009 means for real exchange rate volatility and tariffs can Source: World Bank data and staff estimates. be rejected at the 90 percent confidence level. (The 73 Global Economic Indicators Table 1 Growth of real GDP, 1966-2005 (current 1994 dollars and 1987 prices and exchange rates-average annual percentage growth) 1994 GDP 1995 1996-2005 (US$ billions) 1966-73 1974-90 1997-94 1994 estimate forecast World 25,677 5.1 3.0 1.5 2.9 2.8 3.5 High-income 20,397 4.8 2.8 1.7 3.0 2.5 2.9 Industrial countries 19,677 4.7 2.7 1.6 2.9 2.4 2.8 G-7 countries 17,174 4.7 2.8 1.6 2.9 2.3 2.8 United States 6,738 3.0 2.6 2.5 4.1 3.2 Japan 4,590 10.0 4.0 1.2 05 0.4 G-4 Europe 5,296 4.4 2.3 0.9 2.9 2.5 Germany6 1,835 43 2.0 1.3 2.9 2.1 Other industrial 2,500 4.9 2.3 1.2 3 1 32 2 7 Other high-income 720 8.8 56 62 59 55 5.5 Low- and middle-income 5,280 6.9 3.6 1.0 2.5 3.9 5.3 Excluding Eastern Europe and former Soviet Union 4,490 6.2 3.8 5.0 5.5 4 9 5.4 Asia 1,910 5.9 6.8 77 8.5 8.2 7.2 East Asia and Pacific 1,520 7.9 7.9 9.4 9.7 92 7.9 China 520 8.5 8.7 12.9 12.2 10.2 Korea, Rep. of 380 11.2 8.5 6.6 8.0 9.3 Indonesia 175 64 6.7 7.6 7.3 7 5 South Asia 395 3.7 49 3 9 5.7 5.5 5.4 India 295 3.7 4.8 3.8 6.1 5.8 Latin America and the Caribbean 1,627 6.4 2.6 3.6 4.9 0.9 3.8 Brazil 555 9.8 3.6 2 2 5 8 4.2 Mexico 380 6.8 3.4 2.5 3.6 -6.9 Argentina 280 4.3 0.5 7.6 7 4 -3.2 Europe and Central Asia 1,030 6.9 3.3 -9.0 -8.4 -0.7 4.3 Russian Federationb 380 6.9 3.6 -10.6 -12.6 -4.0 Turkey 130 6.1 4.5 3.2 -5.3 5.3 Poland 90 73 0.2 1 8 5.2 7.0 Middle East and North Africa 510 8 6 1.4 2.4 2.1 2.5 2.9 Iran, Islamic Rep. 65 10.2 -0.3 5.2 24 2.7 Algeria 42 6.3 4.4 -0.6 -1.0 3.5 Egypt 43 3.8 7.3 1.1 2.0 2.5 Sub-Saharan Africa 278 4.7 2.2 0.7 1.7 3.8 3.8 South Africa 120 4.7 2.1 -0 1 2.4 3.4 Nigeria 35 6.5 1.1 2.8 25 0.5 Note. Growth rates over intervals are computed using least squares method. a. Data prior to 1991 cover Federal Republic of Germany b Data prior to 1992 cover former Soviet Union Source World Bank data and staff estimates Real GDP growth, 1996-2005 Percent 8 6 4 World 2 0 High-income East South Latin Europe Middle Sub- countries Asia Asia America and East and Saharan and the Central North Africa Caribbean Asia Africa 76 Table 2 Growth of real GDP per capita, 1966-2005 (current 1994 dollars and 1987 prices and exchange rates-average annual percentage growth) 7994 GDP 1995 1996-2005 per capita (US$) 1966-73 1974-90 1991-94 1994 estimate forecast World 4,560 3.0 1.2 -0.1 1.3 1.4 2.0 High-income 23,980 3.8 2.1 1.0 2.3 1.8 2.4 Industrial countries 24,645 3.8 2.1 0.9 2.3 1.8 2.4 G-7 countries 25,650 3.8 2.2 1 0 2.2 1.6 2 4 United States 25,800 1.9 1.6 1.4 3.0 2.1 Japan 36,730 8 7 3.2 1.0 0.2 0.1 G-4 Europe 20,770 3.8 2 1 0.5 2 5 2.2 Germanya 22,445 3.9 2.0 0.7 2.3 1.5 Other industrial 19,440 4.0 1.7 0.6 2.5 2.7 2.4 Other high-income 14,900 6.0 3.5 5.1 4.9 4 5 4.4 Low- and middle-income 1,110 4.3 1.6 -0.7 0.8 2.3 3.7 Excluding Eastern Europe and former Soviet Union 1,040 3.5 1 7 2.9 3.6 32 3.7 Asia 650 3.3 4.9 6.0 6.9 6.7 5.8 East Asia and Pacific 880 5 1 6.2 8 0 8.4 8.0 6 8 China 440 5.7 7.1 11.6 11.0 9.2 Korea, Rep. of 8,560 8 8 7 0 5.6 6.9 8.3 Indonesia 920 3.9 4.7 5.8 5 7 6.0 South Asia 320 1.3 2 7 1.8 3.7 36 3.7 India 320 1.4 2.7 1.7 4.3 4.0 Latin America and the Caribbean 3,460 3.7 0.4 1 6 2.9 -0.7 2.2 Brazil 3,490 7.1 1.4 0.5 4.0 24 Mexico 4,240 3.5 0.9 0.4 1.5 -8.7 Argentina 8,290 2.7 -0.9 6.1 6.1 -43 Europe and Central Asia 2,110 58 23 -9.3 -8.7 -0.8 3.7 Russian Federationb 2,540 5.9 2.8 -10.6 -12.0 -3.3 Turkey 2,150 3.5 2 1 1.2 -7.2 3.3 Poland 2,370 6 5 -0.6 1.6 5.1 6.9 Middle East and North Africa 1,600 5.7 -1.8 -03 -0.3 0.1 0.4 Iran, Islamic Rep. 1,010 7.0 -4 0 2.4 -0 1 04 Algeria 1,550 3.3 1 3 -2.9 -3.2 1.2 Egypt 750 1.7 4.7 -1.0 00 0.5 Sub-Saharan Africa 490 2.0 -0.7 -2.2 -1.4 1.1 0.9 South Africa 2.970 2.1 -0.5 -2.4 0.1 1.2 Nigeria 330 3.8 -1.7 -0.1 -0.3 -1.8 Note: Growth rates over intervals are computed using least squares method. a Data prior to 1991 cover Federal Republic of Germany b. Data pror to 1992 cover former Soviet Union Source. World Bank data and staff estimates Real per capita GDP growth, 1996-2005 Percent 8 6 4 2 World High-income East South Latin Europe Middle Sub- countries Asia and Asia America and East and Saharan Pacific and the Central North Africa Caribbean Asia Africa 77 Table 3 Inflation: GDP deflators, 1966-2005 (1987=100; percentage change) 1995 7996-2005 7966-73 1974-90 1991-94 1994 estimate forecast World 5.0 7.6 4.0 3.0 3.2 3.5 High-incomea 5.4 6.6 2.7 1.8 2.3 2.8 Industrial countries 5.4 6.4 2.7 1 7 2 2 2.7 G-7 countries 5 3 6.1 2.6 1.6 2 1 2.7 United States 4.9 6.0 2.4 2.1 2.0 Japan 5 8 3.3 1.1 0.1 0.6 G-4 Europe 5.4 8.1 4.0 2.3 3 1 Germanyb 4.9 3.5 4.5 2.3 2.9 Other industrial 5.8 7.8 3.0 2 4 2.7 2.6 Other high-income 5.1 18.3 5 4 44 4.7 5 0 Low- and middle-income' 4.5 10.7 11.5 15.8 10.1 8.5 Excluding Eastern Europe and former Soviet Union 4 5 10.9 10.9 15.2 9.3 7.5 Asia 5 6 8.4 9.5 6.6 8 9 5.5 East Asia and Pacific 5.4 7.4 6 8 6.1 8 6 5.8 China -1 7 3 8 10.8 16.3 127 Korea, Rep of 143 12.1 6.3 5.5 47 Indonesia 65.6 133 7.4 6 6 8.9 South Asia 5.8 8.5 10.1 10.4 9.0 6.4 India 6.2 8.1 10 1 10.5 9.0 Latin America and the Caribbean 5.1 18.0 20.0 159 11.1 8.7 Brazil 228 145.0 1232.0 2284.0 90 1 Mexico 5.9 47.5 13.1 7 3 55 7 Argentina 24 0 255.0 27.6 1.8 1 3 Europe and Central Asia 2 4 5.1 54 0 41.1 38.6 18.1 Russian Federationc 2 7 137 512.0 320.0 1900 Turkey 10.8 462 71.7 1067 82.3 Poland 1.2 31 4 36.4 26 8 28.4 Middle East and North Africa 4.2 9 6 5.1 4 6 5.9 6 3 Iran, Islamic Rep. 5 6 16.9 30.3 35.4 46.1 Algeria 4 5 10.5 27.1 27.9 24.1 Egypt 2.2 11.3 149 8.2 7.3 Sub-Saharan Africa 4 1 10.1 94 23.4 9.7 8.0 South Africa 6.4 14.5 11 9 10.3 102 Nigeria 107 14.5 36.8 37.0 47.1 Note: Deflators are in local currency units. Growth rates over intervals are computed using least squares method a High-income group inflation rates are GDP-weighted averages of local currency inflation, low- and middle-income group rates are medians. b Data prior to 1991 cover West Germany. c Data prior to 1992 cover former Soviet Union Source: World Bank data and staff estimates GDP inflation, 1996-2005 Percent 20 15 10 World 0 High-income East South Latin Europe Middle Sub- countries Asia and Asia America and East and Saharan Pacific and the Central North Africa Caribbean Asia Africa 78 Table 4 Current account balance, 1970-2005 (percentage of GDP) 1994 current account balance 1995 1996-2005 (US$ billions) 1970-80 1987-90 1991-94 7994 estimate forecast World -82.0 0.0 -0.6 -0.4 -0.3 -0.3 -0.1 High-income 8.5 0.1 -0.3 0.0 0.0 0.0 0.5 Industrial countries -12.0 -0.1 -0.5 -0.1 -0.1 -0.1 0.4 G-7 countries -40.5 0.1 -0 5 -0.2 -0.2 -0.3 0.2 United States -151.0 0.0 -2.1 -1.3 -2.3 -2.4 Japan 1290 0.6 2.3 2.8 2.8 2 2 G-4 Europe -1 4 0.2 0.3 -0.7 0.0 0.1 Gerrnanya -21.7 0.6 2.4 -1.0 -1.1 -0.8 Other industrial 28.5 -1 0 -0.7 0.4 1.1 1.4 1 6 Other high income 20 5 9.0 9.6 2.6 3.5 2 5 1.4 Low- and middle-income -90.4 -0.4 -1.4 -2.1 -1.9 -1.5 -1.4 Excluding Eastern Europe and former Soviet Union -84.6 -0.5 -2.0 -2 3 -1.9 -1 6 -1.6 Asia -22.0 -1.2 -1 6 -1.4 -1.2 -1.4 -1 6 East Asia and Pacific -188 -1 3 -1 2 -1.4 -1.3 -1.3 -1.5 China 7.2 -0.4 0.1 0.9 1.4 2.6 Korea, Rep. of -5.3 -5.3 0 6 -1 4 -1.4 -1 9 Indonesia -3.8 -1 2 -3 1 -2.2 -2.2 -3.5 South Asia -3 2 -0.9 -2.4 -1.2 -0.8 -1 9 -2.3 India -1 3 -0.1 -2.2 -0.7 -0.4 -1.7 Latin America and the Caribbean -46 6 -2 5 -1.7 -2.6 -3.0 -1.8 -1.6 Brazil -1 9 -4.0 -1.5 0.2 -0.3 -2.4 Mexico -28 9 -2.4 -1 0 -6 6 -7.7 -0.2 Argentina -10.0 -0.3 -2.1 -2.4 -3.6 -1.3 Europe and Central Asia -41 -0 5 -0.1 -1.1 -0.6 -0.6 -0.5 Russian Federationt 0.1 0.1 0 2 0 2 0 0 0 4 Turkey 3.1 -2.2 -1 6 -0.4 2.4 1.5 Poland -0.5 -3 9 -1.4 -1.8 -0.6 -03 Middle East and North Africa -15.6 7.2 -2.7 -5.7 -3.1 -1.7 -1.6 Iran, Islamic, Rep. 2.6 5 2 -0 4 -3.4 1.3 1.5 Algeria -1.8 -4.7 -0.5 0.9 -4 4 -7.5 Egypt 0 2 -4.9 -4.6 4 8 0.5 -3.0 Sub-Saharan Africa -2 1 -1.6 -2.4 -1.4 -0.8 -2.7 -2.2 South Africa -0.4 -1.5 0.6 1.1 -0.3 -1.7 Nigeria 1 6 0.6 0 0 4 7 4.5 -1.0 Note: Shares over intervals are period averages a. Data prior to 1991 cover Federal Republic of Germany b. Data prior to 1992 cover former Soviet Union Source World Bank data and staff estimates Ratio of current account balance to GDP, 1996-2005 Percent 0.5 -0.5 - -1.5 -2.0 -2.5 High-income East South Latin Europe Middle Sub- countries Asia and Asia America and East and Saharan Pacific and the Central North Africa Caribbean Asia Africa 79 Table 5 Exports of goods, 1994 (percent) Merchandise Average Effective Merchandise Average Effective Merchandise Average Effective exports annual marAe? exports annual marAet exports annual market Region! (us$ growth growth Region/ rUS$ growth growth Region! (us$ growth growth country milions) 7987-93 7987-93 countiv mdihons) 7987-93 1987-93 country mitions) 1981-93 1987-93 World 4,237,696 5.4 4.9 Europe and Central Asia (continued) Sub-Saharan Africa (continued) Bulgaria 4,165 -11 7 -1 5 Nigeria 9,378 2.3 5 7 All developing 946,558 5.3 5.0 Czech Repubnc 14.252 Senegal 914 9 3 7 Asia 427,305 10.8 6.5 Eston a I 299 South Afrca 25.000 7 4 4 5 East Asia andGeorgia 90 Sudan 465 -3 8 5 9 Pa386,383 11.2 6.7 Greece 9,384 6.4 3 6 Zambia 1,071 -1 1 4 9 nea 4 18 Hngary 10733 0 8 0 6 Zimbabwe 1,260 0 7 3 8 Indonesia 40,054 8.6 6 2 Kazakrstan 1.100Hihncm 329,7 5. 48 Korea. Rep of 96,000 11.2 6 1 Kyrgyz Republic 16 High-income 3,291,137 5.4 4.8 Malaysia 58.756 132 7 1 Latvia 972 Industrial 2,891,953 4.9 4.8 Myanmar 771 0 5 LIhuania 2,040 aG-7 2a133,929 4 6 5 1 Papua New Guinea 2.640 6 9 5 3 Moldova 144 Canada 166,000 5 2 6 1 Phlilppines 13,304 4.3 6 2 Poland 17,000 4 6 0 Fance 235,905 4 4 4 2 Thailand 45.262 166 6 0 Romania 6.151 -11 2 1 3 Italy 789,805 43 40 Vetnam 3,770 Russian Federation 64,912 Japan 397,000 40 6.6 South Asia 38,922 7.5 4.7 Slovak Repubic 6.587 Bangladesh 2,661 8.8 4 9 Slovenia 6.828 United Kingdom 205.000 4 3 43 India 25,000 7 1 4 3 Taikistan 320 Uited States 513,000 59 5 7 Nepal 363 9 7 5 1 Turkmenistan 382 Other industrial 758,024 5 5 4 0 Pakistan 7,370 10 1 5 4 Turkey 18.106 6 6 a 8 Australia 47.538 6 8 6.2 Sri Lanka 3,210 8 2 6.1 Ukrane 4,570 . Austria 45,200 7 0 3 3 Latin America and Uzbekistan 944 Belgium and the Caribbean 182,201 5.0 5.4 Middle East and DLu,e 394 4 6 3 7 Boia 15.839 3 0 47 North Africa 96,741 1.2 5.4 Finland 29,700 2 2 3.0 Bolivia 1,032 2 9 5.4 Algeria 8,594 2 9 4 6 ca760 21 3 Brazil 43,600 4.7 5 2 oAa 83594 3 9 4 6 Ireand 1,620 2 1 3 8 Chile 11,539 7 0 5 1 Bahrain 3,454 3 9 6 1 Ireland 34,370 9 4 4 2 Colombia 8,399 12 2 5 0 Egypt 3463 0 9 3 6 Netherands 155,554 49 3 8 CsaRcIran Islamic Rep 13,900 8 7 4 6 Nev, Zealand 12,200 3 9 5.7 Costa Rica 2,215 6 2 5 2 Iraq 500 -203 46 Nowav 34700 75 36 Dominican Republic 633 -5 1 6 2 Jordan 1 424 4 8 12 2 Ecuador 3,820 3.8 56 Morocco 4.013 43 43 Portugal 17,540 106 3 8 El Salvador 844 -1 5 4 7 Oman 5,418 1 7 7.0 Spain 73.300 7 1 4 1 Guatemala 1.522 0 5 5 5 Sa A,6 8 5 9 Sweden 61,292 3 3 3 8 JmiaSaudi Arabia 38,600 -0.8 59 Switzeiarrd 66,200 6.2 4.5 Jamaica 1,192 2 6 4.9 Syrian Arab Rep 3,547 8 3 24 Mexico 61,964 10.6 6 3 Tunisia 4,660 8 0 4 2 Other high- Panama 584 4 9 4 1 Yemen. Rep 667 7 5 '03 income 399,184 10.1 4.9 Prguay 4 3 5 3 Sub-Saharan Africa 59,065 2.4 4.6 Brues 2, 4 -08 4 1 Angola 3.550 11 5 47 Cpu Trinidad and Tobago 1,867 -1 3 5 0 Botswana 1,845 6 5 3 1 Hong Kong 751,395 16.3 6 7 Uruguay 1,913 1.9 4 6 Cdte d lvoire 2652 1 4 38 sre 16.881 59 51 VenezuJela 15,480 2 8 5 9 Cameroor 2.062 8 7 4 4 Kuwait 11,614 -4 9 5 8 Europe and Ethiopia 372 -3 5 3 4 Qatar 2,942 1 5 Central Asia 181,246 2.1 1.2 Gabon 2.252 2 6 4 5 Singapore 96.800 13.3 6.8 Armenia 43 Ghana 1,298 6 0 3 8 Taiwan, China 92,847 9 8 Azerbailan 366 Kenya 1,609 4 9 1 8 United Arab Belaras 968 . . Madagascar 277 0 7 49 Emirates 19,700 7 3 5 9 Note Trade data for countries of the former Soviet Union (FSU) refer to trade with non-FSU countries only Growth rates over intervals are computed using least squares method Growth of industrial couniry exports covers 1981-94 Source. See technical notes. Merchandise exports as share of GDP, 1994 Average annual growth of exports, 1981-93 Percent Percent 30 15 20 World 10 10 5 World 105 0- O High- East South Latin Europe Middle Sub- High- East South Latin Europe Middle Sub- income Asia Asia America and East and Saharan income Asia Asia America and East and Saharan countries and and the Central North Africa countries and and the Central North Africa Pacific Caribbean Asia Africa Pacific Caribbean Asia Africa 80 Table 6 Imports of goods, 1994 (percentage growth) Merchandise Average Merchandise Average Merchandise Average imports annual Merchandise impoits annual Merchandise imports annual Merchandise Regionl (US$ growth imports/ Region/ (US$ growth imports/ Region! (US$ growth imports/ country milions) 7981-93 GDP country millions) 1981-93 GDP country milions) 1981-93 GDP World 4,317,548 5.6 17.1 Europe and Central Asia (continued) Sub-Saharan Africa (continued) All developing 1.010,283 3.4 19.1 Bulgara 4,160 -10.8 40.8 Madagascar 434 -2 0 22 6 Czech Republic 15,636 43.4 Nigeria 6,511 -10.1 185 Asia 451,874 8.7 23.6 Estonia 1,660 36.3 Senegal 1,411 1.6 364 East Asia and Georgia 200 9.3 South Africa 23,400 0 2 19 2 Pacific 404,292 9.8 26.6 Greece 21,466 79 27.6 Sudan 1,269 -45 22 0 China 115,681 9.8 22 2 Hungary 14,438 0.8 34.9 Zambia 501 -2 5 144 Indonesia 31,985 4.0 183 Kazakstan 514 2.8 Zimbabwe 1,585 -03 29 2 a sep of 1 2 Kyrgyz Repubic 92 3.4 High-income 3,307,266 6.2 16.6 Malaysia 59,581 10.2 844 Latvia 1,170 20.1 Industrial 2,860,366 5.8 14.8 Myanmar 886 23 1 2 Lithuania 2.353 45.0 G-7 2,126,850 60 12 7 Papua New Guinea 1,521 -0.1 28 1 Moldova 174 4.7 Canada 155,072 60 286 Philippines 22,546 6.4 35.1 Poland 217400 54 23 1 France 230,203 52 173 Thaland 54,459 15.4 380 Romania 7,109 -1 6 23.6 Germany 381,890 6 2 187 Vietnam 4,440 285 Russian Federation 28,200 . 7.5 Italy 167,685 54 16.4 South Asia 47,582 3.4 12.0 Slovak Republic 6,823 . 552 Japan 275,000 68 60 Bangladesh 4,701 1.9 18.0 Slovenia 7,304 52.0 United Kingdom 227,000 5.4 22 3 India 26,846 4.0 9.1 Tapkistan 318 . 15.4 United States 690.000 6.2 104 Nepal 1,176 4.4 29.1 Turkmenistan 304 . . Other industrial 733.516 5.3 27 8 Pakistan 8,890 3.3 17.1 Turkey 23,270 10 7 17 8 Australia 53,400 4.3 16 1 Sri Lanka 4,780 4.7 40.8 Ukraine 2,610 32 Austria 55,300 65 28 1 Latin America and Uzbekistan 1.150 5 3 Belgium and the Caribbean 212,638 3.4 13.1 Middle East Luxembourg 125,762 4.4 553 Argentina 21,527 3.1 7.6 and North Africa 107,306 -4.1 25.2 Denmark 34,800 3.6 23 8 Bolivia 1,209 1 0 22.0 Algeria 8,000 -5 8 19 Finland 23,200 2 3 23 7 Brazil 36,000 1 8 6.5 Bahrain 3,737 0 5 82 2 Iceland 1,470 1.1 237 Chile 11,800 66 22.7 Egypt 10,185 -3 9 23 7 Ireland 25,508 5.3 490 Colombie 11,883 0.7 17.7 Iran, Islamic Rep 20,000 1 2 31 4 Netherlands 139,795 4.9 42.4 Costa Rica 3,025 64 36.5 Iraq 2,000 -27 1 .. New Zealand 11,900 4.3 234 Dominican Republic 2,630 3 1 253 Jordan 3,382 -2 5 55 4 Norway 27,300 3.2 249 Ecuador 3.690 -11 22.3 Morocco 7,188 4 2 23 3 Portugal 26,680 10.5 306 EG Salvador 2,250 3.7 27 7 Oman 3.915 0 4 33 7 Spain 92,500 11.5 19.2 Guatemala 2,604 33 202 Saudi Arabia 22,796 -5 6 194 Sweden 51,800 3.8 264 Jamaica 2,164 2.5 51 0 Syrian Arab Rep. 5.369 -5 6 . Switzerland 64,100 3.0 246 Mexico 80,100 11 9 21 2 Tunisia 6.580 4.0 41 7 Panama 2,404 0 3 34.5 Yemen. Rep 2.835 -4 9 50 6 Other high- Paraguay 2,370 7 8 33income 446,900 10.3 71.3 Perua 6,7 12 8 303 Sub-Saharan Brunei 1,712 3.4 44.0 Peru 6,794 1.2 136 Africa 63,330 -1.5 22.9 Cyprus 3,020 7 0 42.0 Trnidad and Tobago 1,131 -7.8 236 Angola 1,300 1 1 189 Hong Kong 162,000 124 122.8 Uruguay 2,770 4.6 178 Botswana 1,638 8 4 408 Israel 25,237 5 7 32.4 Venezuela 7,710 -2 8 13 2 Cote dIvoire 2,000 -3.4 298 Kuwait 21,716 -52 89.4 Europe and Cameroon 1,100 -2 9 14 7 Oatar 2,051 -1 1 26.8 Central Asia 175,134 5.0 17.0 abon 737 -2.6 187 Singapore 103,000 100 149.4 Armenia 120 4 6 Taiwan. China 85,507 14 1 35.5 Azerbaian 296 . 8.1 Ghana 1,729 5.2 31.9 United Arab Belarus 534 . 2 6 Kenya 2,156 0.3 31 4 Emirates 21,100 3 9 57.1 Note. Trade data for countries of the former Soviet Union (FSUI refer to trade with non-FSU countries only Growth rates over intervals are computed using least squares method. Growth of industrial country imports covers 1981-1994 Source See technical notes. Merchandise imports as share of GDP, 1994 Average annual growth of imports, 1981-93 Percent Percent 30 10 20 World World 100 High- East South Latin Europe Middle Sub- High- East South Latin Europe Middle Sub- income Asia Asia America and East and Saharan income Asia Asia America and East and Saharan countries and and the Central North Africa countries and and the Central North Africa Pacific Caribbean Asia Africa Pacific Caribbean Asia Africa 81 Table 7 Direction of merchandise trade, 1994 (percentage of world trade) Industrial importers Developing importers Other All Western Source United European indus- indus- Developing Middle hemi- All of exports States Union Japan trial trial Africa Asia Europe East sphere developing World Industrial countries 9.2 29.7 2.6 6.3 47.8 1.3 9.4 2.6 2.1 3.8 19.3 67.7 United States .. 2.6 1.3 32 7.0 0.2 2.2 0 2 0.4 2 2 5.2 12.2 European Union (15) 2 9 23.6 0.8 24 297 1 0 2.6 23 1.3 1 0 8.1 38.3 Japan 2.8 1 5 0.5 48 0.1 3.8 0 1 0.3 04 4.7 9.4 Other industrial 3.5 2.1 0.6 0.3 6.4 0 1 0.9 01 0.1 02 1.3 7 8 Developing countries 6.6 7.1 3.2 1.2 18.1 0.6 8.5 1.8 1.0 1.5 13.3 32.3 Africa 03 09 0 1 0.1 1.3 02 0.1 0.0 0.0 00 0.4 2.0 Asia 3.9 2 7 2 3 07 9.5 02 7.1 0.3 0.5 0.4 8.5 18.2 Developing Europe 0.2 1.9 0.1 0.1 2.3 0.0 03 1.4 0.1 0.0 1.9 4.2 Middle East 04 0.8 06 0.1 1.9 0 1 0.8 0.1 0.2 0.1 1 3 3 5 Western Hemisphere 1 9 0.8 0.2 0.2 3.1 0.0 0.3 0.0 0.1 0.9 1.3 4.4 World 15.9 36.8 5.8 7.5 65.9 1.9 17.9 4.5 3.1 5.3 32.6 100.0 Note. Table based on IMF income and geographic classifications Percentages may not sum to totals, given unallocated and miscellaneous trade flows Source. IMF Direction of Trade Statistics. Share of merchandise imports from developing Direction of merchandise exports, 1994 countries, 1994 Developing Percent to industrial 60 (190) Industrial Idsra 50 to industrial (48%) Industrial 40 to developing 30 20 Developing 10 to developing 0 1-m II 1(14%)o United European Japan Industrial Developing States Union Table 8 Growth of merchandise trade in nominal dollars, 1983-94 (average annual percentage growth) Industrial importers Developing importers Other All Western Source United European indus- indus- Developing Middle hemi- All of exports States Union Japan trial trial Africa Asia Europe East sphere developing World Industrial countries 8.8 8.6 9.5 8.7 8.7 2.2 12.9 8.5 -0.2 11.1 8.4 8.6 United States . 7.1 8.5 100 8.5 1.1 11 3 6.7 1.1 124 9 5 8.9 European Union (15) 8.5 8.9 14.4 7.8 8.9 2 3 14.4 10.1 0.2 100 7 3 8.5 Japan 9 6 10.1 .. 5.9 9.3 3.3 13.3 -2.4 -40 106 9.7 9.4 Other industrial 8 5 7.2 7 1 9.7 8.0 1 4 11.5 0 6 0.8 5 5 7.1 7.5 Developing countries 9.3 6.1 6.2 8.4 7.3 5.3 14.9 3.5 1.4 8.2 10.4 8.1 Africa 1.5 1.7 0.8 3.2 1 7 8.4 10.7 -1.3 6.0 1 9 6.8 2.5 Asia 12.7 15.2 10.5 12.2 12 7 8.9 18.2 8.4 3.8 16.4 15.6 14.0 Developing Europe 14.5 7 0 4.8 9.6 7.5 -3.8 9.2 4.4 -3.0 6.1 7.4 5.4 Middle East 3.9 -1.3 -1.1 1.6 -0 3 2 5 3.9 -3 0 1.4 -6 2 1.5 0.9 Western Hemisphere 7.0 3 9 4.8 4 6 5.8 -0.6 11.1 -6 5 -3.0 9.6 6.8 5.9 World 9.0 8.1 7.5 8.6 8.3 3.0 13.7 5.9 0.2 10.2 9.1 8.4 Note Table based on IMF income and geographic class ificaiions, growth rates are compound averages. Source IMF Direction of Trade Statistics 82 Table 9 Structure of long-term public and publicly guaranteed (PPG) debt, 1994 (percentage of long-term PPG debt) Nonconcessional Nonconcessional Nonconcessional Con- Con- Con- cessional Variable Fixed cessional Variable Fixed cessional Variable Fixed All developing 29.2 33.9 36.9 Europe and Sub-Saharan Africa 47.1 13.2 39.6 Asia 46.5 20.5 32.9 Central Asia 11.9 50.2 37.9 Angola 164 3.7 799 East Asia and Armenia 249 729 2.2 Botswana 450 15.7 393 Pacific 37.8 23.5 38.6 Azerbaijan 00 1000 0.0 Cote dlvoire 29.9 48 2 21 9 China 189 26.7 543 Belarus 41.9 50.7 7.4 Cameroon 446 140 41.4 Indonesia 452 304 24.4 Bulgaria 0.1 77.1 22 7 Ethiopia 83 1 2.0 15.0 Korea, Rep. of 183 21 5 602 Czech Republic 1.4 48 2 50 4 Gabon 15 7 158 68.5 Malaysia 243 263 49.4 Estonia 44.6 446 108 Ghana 82 7 0.6 16.6 Myanmar 92.8 0.0 7 2 Georgia 0.1 87.3 126 Kenya 632 6.1 30.7 Papua New Guinea 52 1 25.7 22 1 Greece . . Madagascar 62 1 8 1 29.8 Philippines 38.0 28 8 33 2 Hungary 1.5 33.8 64.7 Nigeria 4.8 18 5 76.8 Thailand 41 2 15.2 436 Kazakstan 0.1 99.0 0 9 Senegal 74.6 7 2 18.2 Vietnam 92 3 3 1 4.6 Kyrgyz Republic 30 1 66.3 3 6 South Africa South Asia 64.2 14.5 21.4 Latvia 57 7 35.2 7 1 Sudan 50.7 15 5 33.8 Bangladesh 98.3 0 2 1.5 Lithuania 365 24.8 38.7 Zambia 66.2 6 1 277 India 542 17 2 28.6 Moldova 25.2 73.7 1 1 Zimbabwe 44 1 176 38.3 Nepal 97 1 0.0 2.9 PolandSource. World Bank data Pakistan 675 18.5 13.9 Romania 8 2 57.0 348 Sn Lanka 91.5 3 2 5.3 Russian Federation 8 2 55 5 36 2 Latin America and Slovak Republic 2 6 35.5 61 8 Ten highest ratios of non- the Caribbean 9.6 46.8 43.6 Slovenia 2 7 69.6 27 7 concessional debt to GDP, 1994 Argentina 4.4 453 50.3 Talikistan 4 1 95 0 0 9 (percent) Bolivia 59.8 12.0 28.2 Turkmenistan 6 0 83.1 109 0 50 100 150 200 250 300 Brazi 2.8 59.9 373 Turkey 16 7 25.4 579 Chile 3.5 65.7 30.8 Ukraine 1 3 966 2 2 Nicaragua Colombia 7.4 47.8 448 Uzbekistan 16 5 81 9 1 6 Costa Rica 29.4 21.7 48 9 Middle East and Congo Dominican Republic 52.7 27.0 20.3 North Africa 38.6 34.5 26.9 Mozambique Ecuador 174 57.9 247 Algeria 7 2 50 9 42.0 El Salvador 552 17.6 27 2 Bahrain . C6te divoire Guatemala 50.2 16.3 33 5 Egypt 64.7 4 2 31.2 Jamaica 38.5 26.9 34 7 Iran, Islamic Rep 1.0 95 2 3.8 Angola Mexico 2 1 39.4 58 5 Iraq Panama 11 3 67.1 21.6 Jordan 41 3 286 30 1 Guyana Paraguay 566 7 5 35 9 Morocco 27.1 50 3 22.6 Bulgaria Peru 292 37.9 32 9 Oman 20.0 52.2 27.8 Trinidad and Tobago 3.5 56.3 40 2 Saudi Arabia Sudan Uruguay 6 8 52.0 41.2 Syrian Arab Rep 88.3 0 0 11.7 Venezuela 1 9 530 45 1 Tunisia 37.6 24.4 379 Nigeria Yemen, Rep. 63.5 1.5 350 Gabon Structure of long-term PPG debt, by group, 1994 Structure of long-term PPG debt, by region, 1994 Percent Percent 100 100 80 80- 60 60 40 40 20 20 0 - -0 E Severely Moderately Severely Moderately Other East Asia South Latin Europe Middle Sub- indebted indebted indebted indebted and Asia America and East Saharan low-income low-income middle- middle- Pacific and the Central and North Africa income income Caribbean Asia Africa U Fixed rate Variable rate iE Concessional U Fixed rate U Variable rate Concessional 83 Table 10 Long-term net resource flows to developing countries, 1994 (US$ millions) Official Private Official Percentage Net debt Portfolio development Total of GDP Total flows FDI equity flows Total assistance Other All developing countries 207,402 3.93 158,789 43,775 80,120 34,894 48,614 47,416 1,198 Asia 99,067 5.17 84,704 21,590 44,279 18,835 14,362 12,668 1,694 East Asia and Pacific 85,267 5.61 77,289 21,639 43,037 12,613 7,978 6,794 1,184 China 50,004 9.58 46,555 8,853 33,787 3,915 3,449 1,918 1,530 Indonesia 9,093 5.21 7,408 1.627 2,109 3,672 1,685 1,570 115 Korea, Rep. of 7,445 1 98 8,132 4,798 809 2,525 -687 -240 -447 Malaysia 6,882 974 6,661 993 4,348 1,320 221 208 13 Myanmar 148 0.20 34 1 4 29 114 115 -1 Papua New Guinea -23 -042 -231 -235 4 0 208 224 -16 Philippines 4,552 7.09 4,107 1,700 1,000 1,407 445 614 -168 Thailand 4,612 3.22 4,138 4,036 640 -538 474 357 117 Vietnam 1,014 651 272 -111 100 283 741 687 55 South Asia 13,800 3.49 7,415 -49 1,242 6,222 6,384 5,874 510 Bangladesh 1,536 5.87 47 -11 11 47 1,489 1,498 -9 India 7,085 2.41 5,497 149 620 4,729 1,588 1,234 354 Nepal 329 8.12 -3 -10 7 0 332 335 -3 Pakistan 3,341 6.42 1,657 -108 430 1,335 1,684 1,505 180 SriLanka 661 565 213 -66 166 112 449 460 -11 Latin America and the Caribbean 51,127 3.15 49,669 15,698 20,811 13,160 1,458 4,133 -2,675 Argentina 8,975 3.18 8,214 5,809 1,200 1,205 761 246 516 Bolivia 480 8 73 -5 -25 20 0 486 424 62 Brazil 9,824 1.77 11,871 3,717 3,072 5,082 -2,047 -57 -1,990 Chile 4,118 7.93 4,300 1,638 1,795 867 -182 61 -243 Colombia 1,518 2.26 1,860 590 950 320 -342 260 -602 Costa Rica 28 0.33 29 -62 87 4 -1 14 -15 Dominican Republic 100 0.96 113 -107 220 0 -13 5 -18 Ecuador 822 4.97 705 171 531 4 117 135 -18 El Salvador 158 1.94 -40 -60 20 0 198 134 64 Guatemala 240 1.86 84 46 38 0 156 113 43 Jamaica 75 1.77 123 6 117 0 -48 71 -119 Mexico 16,817 4.46 17,394 4,895 7,978 4,521 -576 162 -739 Panama 560 8.02 633 -31 549 115 -73 8 -81 Paraguay 181 2.31 135 -45 180 0 46 36 10 Peru 3,770 7.53 3,214 -88 2,326 977 556 382 173 Trinidad and Tobago 391 8.17 343 -173 516 0 48 12 36 Uruguay 508 3.27 378 183 170 25 130 87 43 Venezuela 482 083 70 -735 764 42 411 152 259 Europe and Central Asia 26,777 2.60 15,581 5,285 8,362 1,934 11,196 9,249 1,947 Armenia 182 6.99 0 0 0 0 182 151 31 Azerbaijan 134 3.67 0 0 0 0 134 66 68 Belarus 334 1 64 105 90 15 0 229 107 121 Bulgaria 7 0.07 -376 -487 105 6 383 112 271 Czech Republic 2,795 7.76 2,642 1,650 878 114 152 132 20 Estonia 253 5.53 211 -13 214 10 42 24 19 Georgia 188 8.76 10 10 0 0 178 102 77 Hungary 2,912 7.04 2,717 1,233 1,144 340 195 254 -58 Kazakstan 792 432 394 209 185 0 398 25 373 Kyrgyz Republic 160 5.88 10 0 10 0 150 126 24 Latvia 302 5.19 222 8 214 0 80 49 31 Lithuania 95 1.82 13 -18 31 0 82 68 14 Moldova 189 5.07 23 0 23 0 166 87 79 Poland 3,473 3 75 1,244 -636 1,875 5 2,229 1,631 598 Romania 1,278 4.25 787 371 340 76 491 126 365 84 Official Private Official Percentage Net debt Portfolio development Total of GDP Total flows FD/ equity flows Total assistance Other Russian Federation 2,819 0.75 658 -613 1,000 271 2,162 1,767 395 Slovak Republic 810 6.55 577 374 203 0 233 119 114 Slovenia 336 2.39 368 284 84 0 -32 9 -41 Talikistan 237 11.46 10 0 10 0 227 45 182 Turkmenistan 57 .. 13 13 0 0 45 16 29 Turkey 1,234 0.94 1,530 -137 608 1,059 -295 235 -530 Ukraine 736 089 424 174 250 0 313 236 77 Uzbekistan 37 0.17 52 2 50 0 -15 12 -27 Middle East and North Africa 10,296 2.42 4,110 323 3,681 106 6,186 5,882 304 Algeria 1,385 330 424 401 18 5 961 503 458 Egypt, Arab Rep. 2,623 6.11 1,006 -260 1,256 10 1,617 1,613 5 Iran, Islamic Rep. -1.277 -2.00 -1,579 -1,569 -10 0 302 164 138 Jordan 255 4.18 -159 -162 3 0 415 417 -2 Morocco 1,154 3.75 877 213 601 63 277 585 -308 Oman 451 3.88 395 238 130 26 57 73 -17 Syrian Arab Rep. 331 2.12 50 -26 76 0 281 342 -61 Tunisia 422 268 80 -114 194 0 342 83 258 Yemen, Rep. 131 2.33 12 -5 17 0 119 101 17 Sub-Saharan Africa 20,136 5.10 4,725 878 2,987 860 15,411 15,482 -71 Angola 736 10.72 409 59 350 0 328 337 -9 Botswana -18 -0.45 -50 -2 -48 0 32 57 -25 Cote d'lvoire 1,525 22.70 30 6 17 7 1,495 1,383 112 Cameroon 648 8.68 59 -46 105 0 589 632 -43 Ethiopia 841 1794 -13 -20 7 0 853 849 4 Gabon 80 204 -128 -25 -103 0 209 192 16 Ghana 1,286 23.73 838 48 233 557 448 473 -25 Kenya 106 1.54 -272 -276 4 0 378 508 -131 Madagascar 224 11 66 2 -4 6 0 222 225 -4 Nigeria 1,871 5.32 1,885 -91 1,959 17 -14 -113 99 Senegal 512 13.19 -9 -9 0 0 521 531 -10 Sudan 370 6.42 0 0 0 0 370 370 0 Zambia 475 13.64 -4 -64 60 0 479 544 -65 Zimbabwe 340 6.26 -70 -155 35 50 410 391 19 Source. World Bank data. Distribution of long-term net resource flows, 1994 Change in share of private long-term flows, 1990-94 Percent X Official U Private Percentage points 100 . 50 80 - 40 60 30 40 20 20 10 0 - - - - - 0 East South Latin Europe Middle Sub- East South Latin Europe Middle Sub- Asia and Asia America and East and Saharan Asia and Asia America and East and Saharan Pacific and the Central North Africa Pacific and the Central North Africa Caribbean Asia Africa Caribbean Asia Africa 85 Table 11 Manufactures unit value, LIBOR, and commodity prices, selected years, 1965-95 7965 1970 1972 1973 1974 1975 1976 7977 1978 7979 1980 G-5 unit value index of manufactures' 22 25 29 33 41 45 46 50 58 66 72 LIBORb 5.0 8.9 6.0 94 10.8 7.7 6.1 6.4 9.2 12.2 14.0 Commodity price Weights indexesc (percent) Petroleum 6 5 8 12 48 46 51 55 57 135 161 Nonfuel commodities 41 44 45 68 92 75 88 109 102 117 126 Agriculture 69.1 42 46 48 76 100 81 99 128 117 130 139 Food 29.4 43 47 52 92 137 101 86 90 99 113 139 Beverages 16.9 48 59 57 75 88 83 158 271 203 211 185 Raw materials 22.8 37 36 36 57 63 54 71 71 76 93 104 Metals and minerals 28.1 37 41 36 50 61 53 61 66 68 85 95 Fertilizers 2.7 39 30 43 60 195 158 76 75 73 100 129 Commodity prices, Units Agriculture Cocoa cents/kg 37 67 64 113 156 125 204 379 340 329 260 Coffee cents/kg 100 115 110 137 145 144 315 517 359 382 347 Tea cents/kg 129 110 105 106 140 139 154 269 219 216 223 Sugar cents/kg 5 8 16 21 66 45 26 18 17 21 63 Banana $/mt 159 165 161 165 184 247 257 275 287 326 379 Wheat $/mt 59 55 70 140 180 149 133 103 128 160 173 Rice $/mt 119 126 129 293 517 341 235 252 346 313 411 Maize $/mt 55 58 56 98 132 120 112 95 101 116 125 Coconut oil $/mt 348 397 234 513 998 394 418 578 683 985 674 Palm oil $/mt 273 260 217 378 669 434 407 530 600 654 584 Soybeans $/mt 117 117 140 290 277 220 231 280 268 298 296 Soybean oil $/mt 270 286 241 436 832 563 438 580 607 662 598 Cotton cents/kg 63 63 79 136 142 116 169 155 157 169 205 Rubber cents/kg 50 41 33 68 75 56 77 81 99 126 142 Other Logs $/cm 35 43 41 68 82 68 92 93 97 170 196 Sawnwood $/cm 157 175 171 224 247 223 264 265 272 366 396 Urea $/mt 48 59 95 316 198 112 127 145 173 222 Metals and minerals Copper $/mt 1,290 1,413 1,071 1,786 2,059 1,237 1,401 1,310 1,367 1,985 2,182 Aluminum $/mt 474 556 511 589 674 797 896 1,050 1,088 1,230 1,456 Nickel $/mt 1,735 2,846 3,080 3,373 3,825 4.570 4,974 5,203 4,610 5,986 6,519 Gold ($/toz) 35 36 58 97 159 161 125 148 193 307 608 Phosphate rock $/mt 13 11 12 14 55 67 36 31 29 33 47 Steel products index (1990 =100) 25 31 30 46 66 52 54 53 68 76 79 Energy Crude petroleum $/bbl 1.4 1.2 1.8 2 8 11.0 10.4 11.6 12.6 12.9 31.0 36.9 Coal $/mt . .. . .. . .. .. 33.4 39.6 35.4 43.1 a. Unit value index in U S dollar terms (1990=100) of manufactures exported from the G-5 countries (France, Germany, Japan, United Kingdom, and United States), weighted by the country's exports to developing countres b. London interbank offered rate on six-month U.S dolar deposits. Price indexes relative to manufactures unit value index, 1985-95 Index (1990=100) 250 Beverages 200 Food 150 L. Metals Sand minerals . 100 Raw materials 50 0. 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 86 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 72 71 70 68 69 81 89 95 95 100 102 107 106 110 115 167 13.6 9.9 11.3 8.6 6.8 7.3 8.1 9.3 84 6.1 3.9 3.4 5.1 6.1 155 143 130 125 119 63 79 64 78 100 85 83 74 69 75 109 96 103 105 92 93 93 111 108 100 95 92 92 112 122 118 104 112 118 101 104 99 110 106 100 98 94 99 124 132 123 97 105 107 86 77 84 107 108 100 99 100 99 107 117 149 150 157 180 165 196 136 141 115 100 94 79 85 150 152 90 80 88 87 71 70 90 91 97 100 99 98 110 126 135 83 75 82 74 70 65 78 114 111 100 89 86 74 85 102 122 105 98 98 89 89 94 109 106 100 102 96 84 93 104 208 174 212 240 225 207 199 158 124 127 120 110 112 140 143 287 309 291 319 323 429 251 303 239 197 187 141 156 331 333 202 193 233 346 198 193 171 179 202 203 184 200 186 183 164 37 19 19 11 9 13 15 22 28 28 20 20 22 27 29 401 374 429 370 380 382 393 478 547 541 560 473 443 439 445 175 160 157 152 136 115 113 145 169 136 129 151 140 150 177 459 272 257 232 197 186 215 277 299 271 293 268 235 268 321 131 109 136 136 112 88 76 107 112 109 107 104 102 108 123 570 464 730 1,155 590 297 442 565 517 337 433 578 450 608 670 571 445 501 729 501 257 343 437 350 290 339 394 378 528 628 288 245 282 282 224 208 216 304 275 247 240 236 255 252 259 507 447 527 724 572 342 334 463 432 447 454 429 480 616 625 185 160 185 179 132 106 165 140 167 182 168 128 128 176 213 112 86 106 96 76 81 98 118 97 86 83 86 83 113 158 155 146 138 157 122 139 202 201 191 177 191 210 390 308 256 349 339 328 352 307 329 401 402 485 533 553 607 758 821 740 216 159 135 171 136 107 117 155 132 157 172 140 107 148 212 1,742 1,480 1,592 1,377 1,417 1,374 1,783 2,602 2,848 2,662 2,339 2,281 1,913 2,307 2,936 1,263 992 1,439 1,251 1,041 1,150 1,565 2,551 1,951 1,639 1,302 1,254 1,139 1,477 1,806 5,953 4,838 4,673 4,752 4,899 3,881 4,872 13,778 13,308 8.864 8,156 7,001 5,293 6,340 8,228 460 376 423 360 318 368 446 437 381 383 362 344 360 384 384 50 42 37 38 34 34 31 36 41 41 43 42 33 33 35 82 71 67 70 61 62 72 94 106 100 99 88 91 93 107 355 32.7 29.7 28.6 27.2 14.4 18.2 14.7 17.8 22.9 194 19.0 16.8 15.9 17.2 56.5 52.2 44.5 48.6 46.6 43.9 36.2 37.1 40.5 41.7 41.5 40.6 38.0 36.5 39.2 c Indexes are n current U S. dollar terms (1990=100) d. World Bank commodity prces were revised in April 1995 Price seres for wheat, rice, rubber, sawnwood and crude petroleum are not comparable to those n previous editions Source World Bank data. Price indexes relative to manufactures unit value index, 1985-95 Index (1990= 1001 300 Coffee 250 200 150 Wheat_. 50 Copper Petroleum 0 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 87 Technical notes The principal sources for the data contained in this Tables 7 and 8. Growth rates are compound averages statistical annex are the World Bank's central data- and are computed for current dollar measures of trade. bases. Table 9. Long-term debt covers public and publicly Regional aggregates are based on the classifica- guaranteed external debt but excludes IMF credits. tion of economies by income group and region, fol- Concessional debt is debt with an original grant ele- lowing the Bank's standard definitions (see country ment of 25 percent or more. Nonconcessional variable classification tables that follow). Debt and finance interest rate debt includes all public and publicly data refer to the 136 countries that report to the guaranteed long-term debt with an original grant ele- Bank's Debtor Reporting System (see World Debt ment of less than 25 percent whose terms depend on Tables 1996). Small economies have generally been movements of a key market rate. This item conveys omitted from the tables but are included in the information about the borrower's exposure to regional totals. changes in international interest rates. For complete Current price data are reported in U.S. dollars. definitions, see World Debt Tables 1996. Table 10. Long-term net resource flows are the sum Notes on tables of net resource flows on long-term debt (excluding IMF) plus non-debt creating flows. Foreign direct Tables I through 4. Projections are consistent with investment refers to the net inflows of investment those highlighted in chapter I and appendix 1. from abroad. Portfolio equity flows are the sum of Tables 5 and 6. Merchandise exports and imports country funds, depository receipts, and direct pur- exclude trade in services. Imports are reported on a chases of shares by foreign investors. For complete c.i.f. basis. Growth rates are based on constant price definitions, see World Debt Tables 1996. data, which are derived from current values deflated Table 11. Commodity price data are collected by the by relevant price indexes. Effective market growth is International Economics Department of the World the export-weighted import growth rate of the coun- Bank. World Bank commodity price series for wheat, try's trading partners. The UNCTAD trade database rice, rubber, sawnwood, and crude petroleum were is the principal source for data through 1993; data for revised in April 1995. As a result, commodity price 1994 are from IMF and UN COMTRADE databases or indexes are not strictly comparable to previous edi- World Bank staff estimates. tions of Global Economic Prospects. 88 Classification of Economies Table 1 Classification of economies by income and region, 1996 Sub-Saharan A]rica Asia Europe and Central Asia Middle East and North Africa East and Eastern Income Southern East Asia and Europe and Rest of Middle North group Subgroup Africa West Africa Pacific South Asia Central Asia Europe East Africa Americas Bumundi Benin Cambodia Afghanistan Albania Yemen, Rep. Egypt, Arab Guyana Comoros Burkina Faso China Bangladesh Armenia Rep. Haiti Eritrea Cameroon Lao PDR Bhutan Azerbaijan Honduras Ethiopia Central African Mongolia India Bosnia and Nicaragua Kenya Republic Myanmar Nepal Herzegovina Lesotho Chad Vietnam Pakistan Georgia Madagascar Congo Sri Lanka Kyrgyz Republic Malawi C6te divoire Tajikistan Mozambique Equatorial Guinea Rwanda Gambia, The Low- Somalia Ghana income Sudan Guinea Tanzania Guinea-Bissau Uganda Liberia Zaire Mali Zambia Mauritania Zimbabwe Niger Nigeria S5o Tom6 and Principe Senegal Sierra Leone Togo Angola Cape Verde Fiji Maldives Belarus Turkey Iran, Islamic Rep. Algeria Belize Botswana Indonesia Bulgaria Iraq Morocco Bolivia Djibouti Kiribati Croatia Jordan Tunisia Colombia Namnibia Korea, Dem. Rep. Estonia Lebanon Costa Rica Swaziland Marshall Islands Kazakstan Syrian Arab Cuba Micronesia, Latvia Republic Dominica Fed Sts Lithuania West Bank Dominican N Mariana Macedonia. and Gaza Republic Islands FYRa Ecuador Lower Papua New Moldova El Salvador Guinea Poland Grenada Philippines Romania Guatemala Solomon Islands Russian Jamaica Thailand Federation Panama Tonga Slovak Republic Paraguay Vanuatu Turkmenistan Peru Western Samoa Ukraine St. Vincent and Uzbekistan the Grenadines Middle- Yugoslavia. Suriname income Fed. Rep. Venezuela Mauritius Gabon American Samoa Czech Republic Greece Bahrain Libya Antigua and Mayotte Guam Hungary Isle of Man Oman Barbuda Reunion Korea, Rep. Slovenia Malta Saudi Arabia Argentina Seychelles Malaysia Barbados South Africa New Caledonta Brazil Chile French Guiana Guadeloupe Upper Martinique Mexico Puerto Rico St. Kitts and Nevis St. Lucia Trinidad and Tobago Uruguay Subtotal: 165 27 23 25 8 27 4 10 5 36 90 Table 1 (continued) Sub-Saharan Africa Asia Europe and Central Asia Middle East and North Africa East and Eastern Income Southern East Asia and Europe and Rest of Middle North group Subgroup Africa West Africa Pacific South Asia Central Asia Europe East Africa Americas Australia Austria Canada Japan Belgium United States New Zealand Denmark Finland France Germany Iceland Ireland OECD Italy Countries Luxembourg Netherlands Norway High- Portugal income Spain Sweden Switzerland United Kingdom Brunei Andorra Israel Aruba French Polynesia Channel Islands Kuwait Bahamas, The Hong Kong Cyprus Qatar Bermuda NonOECD Macao Faeroe Islands United Arab Cayman Islands Countries Singapore Greenland lmirates Netherlands OAE Liechtenstein Antilles Monaco Virgin Islands (US) Total: 210 27 23 34 8 27 28 14 5 44 a. Former Yugoslav Republic of Macedonia. b. Other Asian economies-Taiwan, China. For operational and analytical purposes, the World Bank's main Definitions of groups criterion for classifying economies is gross national product (GNP) per capita. Every economy is classified as low-income, middle- These tables classify all World Bank member countries, and all other income (subdivided into lower-middle and upper-middle), or high- economies with populations of more than 30.000. income. Other analytical groups, based on geographic regions, exports, and levels of external debt, are also used. Income group: Economies are divided according to 1994 GNP per capita, calculated using the World Bank Atlas method. The groups are: Low-income and middle-income economies are sometimes referred low-income, $725 or less; lower-middle-income, $726-$2,895; upper- to as developing economies. The use of the term is convenient; it is middle-income, $2,896-$8,955; and high-income. $8,956 or more. not intended to imply that all economies in the group are experiencing similar development or that other economies have The estimates for the republics of the former Soviet Union are reached a preferred or final stage of development. Classification by preliminary and their classification will be kept under review. income does not necessarily reflect development status. 91 Table 2 Classification of economies by major export category and indebtedness, 1996 Low- and middle-income Low-income Middle-income High-income Severely Moderately Less Severely Moderately Less Not classified Group indebted indebted indebted indebted indebted indebted by indebtedness OECD nonOECD India Armenia Bulgaria Russian Belarus Canada Hong Kong Pakistan China Federation Czech Republic Finland Israel Georgia Estonia Germany Macao Kyrgyz Korea, Dem. Ireland Singapore Republic Rep. Italy OAE2 Korea, Rep Japan Exporters of Latvia Sweden manufactures Lebanon Switzerland Lithuania Malaysia Moldova Romania Thailand Ukraine Uzbekistan Burundi Albania Mongolia Bolivia Chile Botswana American Samoa Iceland Faeroe Islands C6te d'Ivoire Chad Cuba Namibia French Guiana New Zealand Greenland Equatorial Guinea Malawi Peru Solomon Islands Guadeloupe Ghana Zimbabwe Suriname Reunion Guinea Swaziland Guinea-Bissau Guyana Honduras Liberia Madagascar Mali Exporters Mauritania of nonfuel Myanmar primary Nicaragua products Niger Rwanda Sio Tom6 and Principe Somalta Sudan Tanzania Togo Uganda Vietnam Zaire Zambia Congo Algeria Venezuela Bahrain Brunei Nigeria Angola Iran, Islamic Qatar Gabon Republic United Arab Exporters Iraq Libya Emirates offuels Oman (mainly oil) Saudi Arabia Trinidad and Tobago Turkmenistan Cambodia Benin Bhutan Jamaica Cape Verde Antigua and Martinique United Aruba Ethiopia Comoros Burkina Faso Jordan Dominican Barbuda Kingdom Bahamas, The Mozambique Egypt, Arab Lesotho Panama Republic Barbados Bermuda Yemen, Rep. Rep. Greece Belize Cayman Islands Gambia, The Morocco Djibouti Cyprus Haiti Western Samoa El Salvador French Polynesia Nepal Fiji Kuwait Exporters Grenada Monaco of services Kiribati Maldives Paraguay Seychelles St. Kitts and Nevis St. Lucia Tonga Vanuatu 92 Table 2 (continued) Low- and middle-income Low-income Middle-income High-income Severely Moderately Less Severely Moderately Less Not classifted Group indebted indebted indebted indebted indebted indebted by indebtedness OECD nonOECD Afghanistan Bangladesh Azerbaijan Argentina Colombia Costa Rica Australia Netherlands Cameroon Lao PDR Sri Lanka Brazil Hungary Dominica Austria Antilles Central African Senegal Tajikistan Ecuador Indonesia Guatemala Belgium Republic Mexico Papua New Kazakstan Denmark Kenya Poland Guinea Malta France Diversified Sierra Leone Syrian Arab Philippines Mauritius Luxembourg exporters' Republic Tunisia South Africa Netherlands Turkey St. Vincent and Norway Uruguay the Grenadines Portugal Yugoslavia, Spain Fed. Rep. United States Croatia Bosnia and Andorra Macedonia, Herzegovina Channel Islands FYR' Eritrea Liechtenstein New Caledonia Guam Virgin Slovak Republic Isle of Man Islands (US) Not classified Slovenia Marshall Islands by export Mayotte category Micronesia. Fed. Sts. N. Mariana Islands Puerto Rico West Bank and Gaza Number of economies 210 36 15 11 17 16 55 15 22 23 a. Other Asian economies-Taiwan, China. b Economies in which no single export category accounts for 50 percent or more of total exports. c. Former Yugoslav Republic of Macedonia. Definitions of groups These tables classify all world Bank member economies, plus all either of the two key ratios is above critical levels: present value of other economies with populations of more than 30,000. debt service to GNP (80 percent) and present value of debt service to exports (220 percent). Moderately indebted means either of the Major export category: Major exports are those that account for 50 two key ratios exceeds 60 percent of, but does not reach, the critical percent or more of total exports of goods and services from one levels. For economies that do not report detailed debt statistics to category, in the period 1990-93. The categories are: nonfuel the World Bank Debtor Reporting System (DRS), present-value primary (SITC 0,1,2.4, plus 68): fuels (SITC 3); manufactures calculation is not possible. Instead. the following methodology is (SITC 5 to 9, less 68); and services (factor and nonfactor service used to classify the non-DRS economies. Severely indebted means receipts plus workers' remittances). If no single category accounts three of four key ratios (averaged over 1992-94) are above critical for 50 percent or more of total exports, the economy is classified as levels: debt to GNP (50 percent); debt to exports (275 percent); debt diversified service to exports (30 percent); and interest to exports (20 percent). Moderately indebted means three of the four key ratios exceed 60 Indebtedness: Standard World Bank definitions of severe and percent of, but do not reach, the critical levels. All other classified moderate indebtedness, averaged over three years (1992-94) are low- and middle-income economies are listed as less-indebted. used to classify economies in this table. Severely indebted means 93 Bibliography Aitken, Brian. 1992. "Measuring Trade Policy Intervention: Chyc, K., M. Gehlhar, D. Gray, T. Hertel, E. lanchovichina, B. A Cross-Country Index of Relative Price Dispersion." McDonald, R. McDougall, and M. Tsigas. 1995. "Overview Policy Research Working Paper 838. World Bank, of the GTAP Database." In Global Trade Analysis: Modeling Country Economics Department, Macroeconomic and Applications. Cambridge: Cambridge University Press. Adjustment and Growth Division, Washington, D.C. Coe, David, Elhanan Helpman, and Alexander Hoffmaister. Akiyama, Takamasa, and Don Larson. 1994. "The Adding- 1995. "North-South R&D Spillovers." CEPR Discussion Up Problem: Strategies for Primary Commodity Exports Paper 1133. Centre for Economic Policy Research, U.K. in Sub-Saharan Africa." Policy Research Working Paper Collier, Paul. 1995a. "The Marginalization of Africa." 1245. World Bank, International Economics Department, International Labor Organization Review. 134(4-5): 541-57. International Trade Division, Washington, D.C. - . 1995b. "Recent African Performance and Its Akiyama, Takamasa, and Akihiko Nishio. 1996. "Indonesia's Implications." Cocoa Boom: Hands-Off Policy Encourages Smallholder de Jonquires, Guy. 1995. "Report Counts Cost of Dynamism." Policy Research Working Paper 1580. World Antidumping." Financial Times, 21 October 1995, p. 5. Bank, International Economics Department, Commodity Dixit, Avinash, and Robert Pindyck. 1994. Investment Under Policy and Analysis Unit, Washington, D.C. Uncertainty. Princeton, N.J.: Princeton University Press. Alogoskoufis, George. 1995. "The Two Faces of Janus: Easterly, William. 1993. "How Much Do Distortions Affect Institutions, Policy Regimes and Macroeconomic Growth?" Journal of Monetary Economics 32:187-212. Peformance in Greece." Economic Policy (April):149-91. Easterly, William, Carlos Rodriguez, and Klaus Schmidt- Amjadi, Azita, and Alexander Yeats. 1995. "Nontariff Hebbel. 1994. Public Sector Deficits and Macroeconomic Barriers Africa Faces: What Did the Uruguay Round Performance. New York: Oxford University Press. Accomplish, and What Remains to be Done?" Policy Edwards, Sebastian. 1990. "Capital Flows, Foreign Direct Research Working Paper 1439. World Bank, International Investment, and Debt-Equity Swaps in Developing Economics Department, International Trade Division, Countries." NBER Working Paper 3497. National Bureau Washington, D.C. of Economic Research, Cambridge, Mass. Amjadi, Azita, Ulrich Reincke, and Alexander Yeats. 1996. Erzan, Refik, and Peter Svedberg. 1989. "Protection Facing "Tariffs, Nontariff Measures and Transport Barriers Facing Exports from Sub-Saharan Africa in the EC, Japan, and Sub-Saharan Africa's Exports: An Assessment of the the United States." Policy Research Working Paper 320. Potential Impact of Positive OECD Trade Policies." World World Bank, International Economics Department, Bank, International Trade Division, Washington, D.C. International Trade Division, Washington, D.C. Bailey, Martin, and Hans Gersbach. 1995. "Efficiency in Finger, J. Michael. 1991. "Trade Policy in the United States." Manufacturing and the Need for Global Competition." In D. A. Salvatore, ed., A Handbook of Trade Policies. Brookings Papers on Economic Activity. Washington, D.C.: Westport, Conn.: Greenwood Publishing Group. The Brookings Institution. - . 1993. Antidumping: How It Works and Who Gets Hurt. Ben-David, Dan. 1993. "Equalizing Exchange: Trade Ann Arbor, Mich.: University of Michigan Press. Liberalization and Income Convergence." Quarterly Fischer, Stanley. 1993. "The Role of Macroeconomic Factors Journal of Economics 108(August):653-79. in Growth." Journal of Monetary Economics 32:485-512. Bennathan, Ezra, Luis Escobar, and George Panagakos. Haque, Nadeem, Manmohan Kumar, Nelson Mark, and 1989. Deregulation of Shipping: What Is to Be Learned from Donald Mathieson. 1995. "Determinants of Developing Chile. World Bank Discussion Paper 67. Washington, D.C. Country Creditworthiness." International Monetary Borensztein, Eduardo, Jos6 de Gregorio, and Jong-Wha Lee. Fund, Research Department, Washington D.C.. 1995. "How Does Foreign Direct Investment Affect Harrison, Ann. 1991. "Openness and Growth: A Time Series, Growth?" NBER Working Paper 5057. National Bureau Cross-Country Analysis for Developing Countries." of Economic Research, Cambridge, Mass. Background paper prepared for World Development Caballero, Ricardo, and Vittorio Corbo. 1989 "The Effect of Report 1991. World Bank, Washington, D.C. Real Exchange Rate Uncertainty on Exports: Empirical - . 1994. "Productivity, Imperfect Competition and Evidence." World Bank Economic Review 3(2):263-78. Trade Reform." Journal of International Economics 36:53-73. 95 Harrison, Ann, and Ana Revenga. 1995. "The Effects of OECD (Organization for Economic Cooperation and Trade Policy Reform: What Do We Really Know?" NBER Development). 1994. Economic Outlook. Paris. Working Paper 5225. National Bureau of Economic . 1995a. Economic Outlook. Paris. Research, Cambridge, Mass. . 1995b. The OECD Jobs Study. Paris. Hausman, Ricardo, and Michael Gavin. 1995. Owens, T., and A. Wood. 1995. "Export-Oriented "Macroeconomic Volatility in Latin America: Causes, Industrialization Through Primary Processing." Consequences and Policies to Assure Stability." Inter- University of Sussex, Sussex, U.K. American Development Bank, Office of the Chief Pack, Howard, and John Page. 1994. "Accumulation, Exports, Economist, Washington, D.C. and Growth in the High-Performing Asian Economies." Jaffee, Steven. 1993. Exporting High-Value Food Commodities: Carnegie-Rochester Conference Series on Public Policy A Review of Major Developing Country Success Stories. 40:199-250. World Bank Discussion Paper 198. Washington, D.C. Palmeter, N. David. 1988. "The Anitdumping Emperor." Kaminski, Bartlomiej, Z. X. Wang, and L. Alan Winters. Journal of World Trade 22 (2):73-80 1995. " Foreign Trade in the Transition: The International Papageorgiou, Demetrious, Michael Michaely, and Environment and Domestic Policy." Studies of Economies Armeane Choksi. 1990. "Liberalizing Foreign Trade in in Transformation 20. World Bank, Washington, D.C. Developing Countries." In Lessons Of Experience in the Kawecka-Wyrzykowska, Elzbieta. 1995. "From Association Developing World, vol. 7. Oxford: Basil Blackwell. to Accession: The Impact of the Association Agreements Petersen, Christian, and T. G. Srinivason. 1995. "Effects of a on Central Europe's Trade and Integration with the Rise in G-7 Real Interest Rates in Developing Countries." European Union." In David Vines and David Currie, eds., North-South Leahy, John, and Toni Whited. 1995. "Effect of Uncertainty Linkages and International Macroeconomic Policy. New on Investment: Some Stylized Facts." NBER Working York: Cambridge University Press. Paper 4986. National Bureau of Economic Research, Pindyck, Robert, and Andr6s Solimano. 1994. "Economic Cambridge, Mass. Instability and Aggregate Investment." World Bank, Lee, Jong-Wha. 1992. "International Trade, Distortions and Washington, D.C. Long-Run Economic Growth." IMF Working Paper. Roberts, Mark, and James Tybout. Forthcoming. Industrial International Monetary Fund, Washington, D.C. Evolution in Developing Countries: Micro Patterns of Leiderman, Leonardo, and Mario Blejer. 1988. "Modeling Turnover, Productivity, and Market Structure. Washington, and Testing Ricardian Equivalence: A Survey." IMF D.C.: World Bank. Working Paper 87/35. International Monetary Fund, Roberts, M., T. Sullivan, and J. Tybout. 1995. "What Makes Fiscal Affairs Department, Washington, D.C. Exports Boom? Evidence from Plant-Level Panel Data." Leipziger, Danny, and L. Alan Winters. Forthcoming. "Chile World Bank, International Trade Division, Washington, and NAFTA: Lessons and Future Directions." In Maurice D.C. Schiff and Claudio Sapelli, eds., Chile in NAFTA: Pros and Rodrik, Dani. 1995. "Getting Interventions Right: How Cons. San Francisco, Calif. and Santiago, Chile: South Korea and Taiwan Grew Rich." Economic Policy International Center for Economic Growth. 20:55-107. Levinsohn, James. 1993. "Testing the Imports-as-Market- Romer, David. 1993. "Openness and Inflation: Theory and Discipline Hypothesis." Journal of International Economics Evidence." Quarterly Journal of Economics 108:869-903. 35(August):1-22. Ruf, F 1993. "Comparisons of Cocoa Production Costs in Maddison, Angus. 1989. "The World Economy in the 20th Seven Producing Countries." The Planter 69(807). Century." OECD Development Centre Studies, Paris. Sachs, Jeffrey, and Andrew Warner. 1995a. "Economic Manchester, Joyce, and Warwick McKibbin. 1994. Reform and the Process of Global Integration." Brookings "Macroeconomic Consequences of the Savings and Loan Papers on Economic Activity 1. Washington, D.C.: The Debacle." Review of Economics and Statistics 76:579-84. Brookings Institution. Maxton, Graeme. 1994. "The Automotive Sector of the - . 1995b. "Natural Resource Abundance and Economic Pacific Rim and China: Moving Into the Fast Lane." Growth." NBER Working Paper 5398. National Bureau of Economist Intelligence Unit, London. Economic Research, Cambridge, Mass. Messerlin, Patrick. 1990. "Antidumping Regulation of Safadi, Raed, and Alexander Yeats. 1994. "Escalation of Procartel Law? The E.C. Chemical Cases." World Economy Asian Trade Barriers." Asian Economic Journal 13:465-92. 8(2):147-79. Mitchell, Olivia. 1993. "Trends in Retirement Systems and Scherer, Frederic. 1984. Industrial Market Structure and Lessons for Reform." Policy Research Working Paper Economic Performance. Chicago, Ill.: Rand-McNally. 1118. World Bank, Education and Social Policy - . 1994. "Competition Policies for an Integrated World Department, Washington, D.C. Economy." The Brookings Institution, Washington, D.C. Montiel, Peter. 1993. "Capital Mobility in Developing Schmidt-Hebbel, Klaus. 1995. "Fiscal Adjustment and Countries: Some Measurement Issues and Empirical Growth: In and Out of Africa." World Bank, Policy Estimates." Policy Research Working Paper 1103. World Research Department, Macroeconomics and Growth Bank, International Economics Department, Debt and Division, Washington, D.C. International Finance Division, Washington, D.C. Streifel, Shane. 1995. Review and Outlook for the World Oil North, Douglass. 1991. "Institutions." Journal of Economic Market. World Bank Discussion Paper 301. Washington, Perspectives 5 (Winter):97-112. D.C. 96 Tesar, Linda, and lingrid Werner. 1993. U.S. Equity - . 1995a. Claiming the Future: Choosing Prosperity in Investment in Emerging Stock Markets. World Bank the Middle East and North Africa. Washington, D.C. Discussion Paper 228. Washington, D.C. - . 1995b. Global Economic Prospects and the Developing Trela, Irene, and John Whalley. 1995. "Global Effects of Countries. Washington, D.C. Developed Country Trade Restrictions on Textiles and - 1995c. "Performance Indicators for the Tele- Apparel." Economic Journal 100. communications Sector." Industry and Energy UNCTAD-DTCI. Forthcoming. World Investment Report Department, Telecommunications and Informatics 1995: Transnational Corporations and Competitiveness. New Division, Washington, D.C. York: United Nations. - . 1995d. World Development Report 1995: Workers in Winters, L. Alan. 1993. "Expanding EC Membership and an Integrating World. New York: Oxford University Association Accords: Recent Experience and Future Press. Prospects." In Kym Anderson and Richard Blackhurst, - . 1996. World Debt Tables. Baltimore, Md.: Johns eds., Regional Integration and the Global Trading System. Hopkins University Press. New York: Harvest Wheatsheaf. . Forthcoming. World Development Report 1996: World Bank. 1993. The East Asian Miracle: Economic Growth From Plan to Market. New York: Oxford University and Public Policy. A World Bank Policy Research Report. Press. Washington, D.C. Zarnowitz, Victor. 1989. "Facts and Factors in the Recent - . 1994. Adjustment in Africa: Reforms, Results, and the Evolution of Business Cycles in the United States." Road Ahead. A World Bank Policy Research Report. NBER Working Paper 2865. National Bureau of Washington, D.C. Economic Research, Cambridge, Mass. 97 Distributors of World CHINA FRANCE ISRAEL NEW ZEALAND RUSSIAN FEDERATION Wenergren-Williams AB BcChina Financial & Economic World Bank Publications Yozmoit Lierature Ltd. EBSCO NZ Ltd. Isdatelstvo