24705 WORLD DEVELOPMENT REPORT 2003 Sustainable Development in a Dynamic World Transforming Institutions, Growth, and Quality of Life A copublication of THE WORLD BANK the World Bank and Oxford University Press (c) The International Bank for Reconstruction and Development / The World Bank About the cover: A montage of two satellite sensor products, the cover image shows the lights of human settlements and (on May 14, 2002) variation in sea surface temperatures. The image illustrates several World Development Report 2003 themes: the link between growth and environment (higher income correlated with greater energy use), the continuing socioeconomic challenge of inequality and poverty reduction (vast disparity in the energy use of industrial countries and that of developing countries), the interconnectedness and impact of human activity (fossil fuel-based energy use raising sea surface temperatures), and the need to gather information (such as that provided by satellite sensors) to anticipate and monitor problems if the world is to shift to a more sustainable development. For more information on the concepts in this report, please visit http://econ.worldbank.org/wdr/wdr2003/ City lights image courtesy of the Defense Meteorological Satellite Program Digital Archive, National Geographic Data Center, U.S. National Oceanic and Aeronautics Administration Sea surface temperatures image courtesy of U.S. National Climatic Data Center Inside art and typesetting by Barton Matheson Willse & Worthington, Baltimore © 2003 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 All rights reserved. 1 2 3 4 04 03 02 A copublication of the World Bank and Oxford University Press. Oxford University Press 198 Madison Avenue New York, NY 10016 The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is copyrighted. No part of this work may be reproduced or trans- mitted in any form or by any means, electronic or mechanical, including photocopying, recording, or inclusion in any information storage and retrieval system, without the prior written permission of the World Bank. The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, fax 202-522-2422, e-mail pubrights@worldbank.org. ISBN 0-8213-5151-6 (clothbound) ISBN 0-8213-5150-8 (paperback) ISSN 0163-5085 (c) The International Bank for Reconstruction and Development / The World Bank Contents Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Acronyms and Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Roadmap to World Development Report 2003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii 1 Achievements and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 The core development challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Act now—for long-term problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2 Managing a Broader Portfolio of Assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Sustainability—an evolving framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Measuring sustainability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 The importance of a range of assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Why the need to manage a broader portfolio of assets? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Tradeoffs and sustainable development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Some assets are overused or underprovided—why? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Correcting the overuse or underprovision of important assets . . . . . . . . . . . . . . . . . . . . . . . . . 31 3 Institutions for Sustainable Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Institutions coordinating human behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Institutions protecting assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Picking up signals, balancing interests, and implementing decisions . . . . . . . . . . . . . . . . . . . . 44 Overcoming barriers to coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Promoting inclusiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Catalysts for change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 4 Improving Livelihoods on Fragile Lands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Inclusion, innovation, and migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Managing fragile land to improve livelihoods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Living on the edge—the arid plains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Living on a precipice—the mountains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Nurturing assets by listening—and by enabling communities to act . . . . . . . . . . . . . . . . . . . . 71 Nurturing women’s human capital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 iii (c) The International Bank for Reconstruction and Development / The World Bank iv WORLD DEVELOPMENT REPORT 2003 Building on traditional social capital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 The use of nonrenewable local resources—balancing interests . . . . . . . . . . . . . . . . . . . . . . . . . 77 Balancing interests among governments, companies, and communities . . . . . . . . . . . . . . . . . . 78 Partnering for change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 5 Transforming Institutions on Agricultural Land . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Land and water constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Eliminating rural poverty and preparing outmigrants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Intensifying the use of land . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Intensifying the use of water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Getting ahead of the frontier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 6 Getting the Best from Cities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 City lights: beacons of hope and warning flares . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Building informed constituencies to address spillovers and anticipate risks . . . . . . . . . . . . . . . 114 Balancing interests to provide urban public goods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Inclusion and access to assets—challenging the institutional roots of urban slums . . . . . . . . . 121 Institutions for sustainable urban development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 7 Strengthening National Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Promoting inclusiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Creating a sound investment climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 Managing the environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 Managing natural resources and using aid effectively. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 Averting violent conflict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 8 Global Problems and Local Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Designing institutions to solve global problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Conserving biodiversity: maintaining current services and future options. . . . . . . . . . . . . . . . 163 Mitigating and adapting to risks of climate change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 9 Pathways to a Sustainable Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Acting today . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Ongoing dialogue: a global vision and accord. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Ongoing dialogue: some open questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 Bibliographical Note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Selected World Development Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 (c) The International Bank for Reconstruction and Development / The World Bank CONTENTS v Boxes 2.1 Not yet able to fully duplicate natural processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.2 Indicators for measuring sustainability—a subset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.3 The Aral Sea—the cost of ignoring the role of an environmental asset . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.4 How keeping the option value of assets can make a serious difference . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.5 Catastrophic ecoshifts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.6 Replacing natural assets with human-made assets can be costly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.7 Perverse subsidies in India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.8 World Development Report 1992: Development and the Environment . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.1 The market as a coordination mechanism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.2 Assets, threats, and protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.3 Natural assets decline when protective institutions are weak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.4 Democracy and environmental policy: picking up signals, shifting the balance . . . . . . . . . . . . . . . . . . . 46 3.5 Local negotiations balance interests and commit parties to clean up Colombia’s rivers . . . . . . . . . . . . . 47 3.6 Policy accountability and accountable ru lemaking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.7 When protective institutions fail: the collapse of Enron and Newfoundland’s cod fisheries . . . . . . . . . . 50 3.8 Fostering inclusiveness: South Africa’s new democracy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 3.9 Mutual reinforcement: environmental movements and democracy. . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.10 Inequality: its long tails in the Americas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 4.1 From degrading soils to degrading water—managing natural assets on the Southern Plains. . . . . . . . . . 64 4.2 Traditional knowledge and voice: sustaining livelihoods on the grasslands of the Sahel . . . . . . . . . . . . . 66 4.3 Balancing public and private goods: biodiversity and coffee production in Chiapas . . . . . . . . . . . . . . . 70 4.4 What worked then (Europe, 1900) is much harder now (developing countries, 2000) . . . . . . . . . . . . . 72 4.5 Addressing risks, changing institutions, and reaching subsistence families in Tunisia . . . . . . . . . . . . . . . 74 4.6 “Cultural translators� as catalysts to upgrade livelihoods in Ait Iktel, Morocco . . . . . . . . . . . . . . . . . . . 76 4.7 Learning to balance interests: two big mines in the Andes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 5.1 More food, greater intensity of land use, fewer farmers per urban resident . . . . . . . . . . . . . . . . . . . . . . 84 5.2 Poverty, equitable growth, and path dependency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 5.3 Land distribution and path dependency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 5.4 Breaking out through zais and tassas—low-input traditional technologies. . . . . . . . . . . . . . . . . . . . . . . 91 5.5 Breaking out through fertilizer: the next green revolution?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.6 Science, technology, and institutions to solve the challenge of nature: obsolete pesticide stockpiles in Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.7 The precautionary principle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 5.8 Institutional commitment and African agriculture: lessons from Asia and South America . . . . . . . . . . . 97 5.9 Weakening the interest of landholders in unproductive land . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 5.10 The race for water—and land—and the displacement of the poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 5.11 Water parliaments in France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 5.12 The Amazon rancher’s decision to deforest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 5.13 Brazil: getting ahead of the frontier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 6.1 The focus of “urban� in this chapter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 6.2 How social networks help the urban poor manage risks and get ahead . . . . . . . . . . . . . . . . . . . . . . . . 110 6.3 Political reform and stakeholder alliances overturning pollution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 6.4 Meeting environmental, social, and economic objectives through urban transport strategy in Bogotá . 120 6.5 Regularizing favelas in Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 6.6 How railway dwellers in Mumbai managed their own resettlement . . . . . . . . . . . . . . . . . . . . . . . . . . 125 6.7 Mexico City’s search for metropolitan management arrangements . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 6.8 Leading the advance on urban settlement growth in Lima . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 7.1 Democracy, leadership, and decentralization in Latin America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 7.2 Brazil: changing the rules of the game for better public services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 (c) The International Bank for Reconstruction and Development / The World Bank vi WORLD DEVELOPMENT REPORT 2003 7.3 Civil society and governance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 7.4 National policy can generate excessive urban concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 7.5 Perverse sugar subsidies in the United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 7.6 Perverse energy subsidies in the Islamic Republic of Iran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 7.7 Aid and compensation to address obstacles to reform in the Russian Federation’s coal sector . . . . . . . . 142 7.8 Cameroon: the path to improved forest governance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 7.9 Partnership for sustainable fisheries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 7.10 Malaysia: ethnic diversity, conflict resolution, and development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 7.11 Improving the process: the Chad-Cameroon Pipeline Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 8.1 An adaptive, learning institution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 8.2 “Coupling institutions� and policy entrepreneurs in Costa Rica and Bolivia . . . . . . . . . . . . . . . . . . . . 161 8.3 Poverty and biodiversity in Madagascar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 8.4 The Nile Basin Initiative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 8.5 Costa Rica’s program of payment for environmental services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 8.6 Municipal incentives for conservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 8.7 Tradable forest obligations efficiently meeting conservation goals. . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 8.8 The Prototype Carbon Fund and the carbon market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 9.1 Think spatially . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 9.2 Problem solving by think-and-do tanks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 9.3 A big push—to address spillovers and seize opportunities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 9.4 Millennium Development Goals (1990–2015) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 9.5 Outcome of the International Conference on Financing for Development, Monterrey, Mexico . . . . . . 193 Figures 1.1 Global population approaching stability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2 Some regions growing fast, others stable. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.3 Dependency ratios on the decline—for a while. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.1 Adjusted net savings rates by per capita GDP level, 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.2 How society’s assets enhance human well-being . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.3 Very different environmental outcomes with the same growth rates . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2.4 Reducing emissions in Mexico City . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.5 Mechanisms to address market and policy failures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.1 Social norms, rules, and organizations for coordinating human behavior . . . . . . . . . . . . . . . . . . . . . . . 38 3.2 Growing participation in civil society organizations, 1981–97 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.3 The relationship between institutional quality and national income. . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.4 Concentration of dust particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 3.5 More mayors in Latin America are elected locally—by citizens or by elected city councils . . . . . . . . . . . 56 4.1 Rural population growth rate relative to share of total population on fragile land . . . . . . . . . . . . . . . . . 61 4.2 Arid lands of the world . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4.3 Rainfall in the Sahel, 1950–2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.4 Mountainous areas of the world. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 5.1 Regional variations in land scarcity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 5.2 Regional variations in water scarcity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 6.1 Many developing countries are undergoing urban transition with relatively high urban population growth rates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 6.2 Poverty in Cali, Colombia: 1999 headcount rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 6.3 High inequality in health outcomes in urban areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 (c) The International Bank for Reconstruction and Development / The World Bank CONTENTS vii 7.1 Lead in gasoline and in blood in the United States, 1975–90 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 7.2 Unsustained growth performance is closely associated with point-source natural resources, and conflict. . 149 7.3 Angola: real GDP per capita, 1960–96. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 8.1 Current land use in closed canopy forest deforested in 1990–2000 . . . . . . . . . . . . . . . . . . . . . . . . . . 166 8.2 Fossil fuel–intensive and climate-friendly scenarios, 1990–2100 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Tables 2.1 Toward adjusted net savings, 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.2 Examples of types of externalities addressed in each spatial arena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.3 The benefits of full-cost energy pricing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.1 Environmental fragility in developing countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.2 Regional distribution of people living on fragile land . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 4.3 Share of population on fragile land, countries in conflict, and rural population growth. . . . . . . . . . . . . 62 5.1 The capacity of institutions to sense problems, balance interests, and implement solutions . . . . . . . . . . 96 6.1 Urban environmental issues and status by level of city development. . . . . . . . . . . . . . . . . . . . . . . . . . 112 6.2 Environmental health, welfare, and living conditions vary by city product . . . . . . . . . . . . . . . . . . . . . 113 7.1 Civil conflict and reported homicides. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 (c) The International Bank for Reconstruction and Development / The World Bank ACKNOWLEDGMENTS This Report has been prepared by a team led by and San Jose (Costa Rica) and a series of video Zmarak Shalizi and comprising Kenneth conferences with East Asia, Africa, and Europe. Chomitz, Christian Eigen-Zucchi, Gunnar The participants in these workshops and video Eskeland, Swati Ghosh, Christine Kessides, conferences included policymakers, academics, Linda Likar, and Robert Schneider. The team and nongovernmental organizations. The team was assisted by Leena Datwani, Claudio E. participated in the United Nations’ WSSD- Montenegro, and B. Bulent Ozbilgin. Valuable related prepcoms in New York. In addition, it contributions were made by Matthew Stilwell organized e-conferences with the help of the and Paul Steinberg. Bruce Ross-Larson was World Bank Institute. The Development Data the principal editor. The work was carried Group contributed to the data appendix and out under the general direction of Nicholas was responsible for the Selected World Devel- H. Stern. opment Indicators. Much of the background The Report received useful advice from a research and external consultations were sup- three-person steering committee consisting of ported by a generous grant from the Norwegian Nicholas Stern, Ian Johnson, and Vinod government. Thomas, and a six-person internal consultative Rebecca Sugui served as executive assistant group consisting of Michele DeNevers, Ian to the team; Leila Search, as program assistant Goldin, Kristalina Georgieva, Steen Jorgensen, and technical support; and Endy Shri Odin Knudsen, and John Shilling. Djonokusomo, Shannon Hendrickson, Joanna Many others inside and outside the World Kata-Blackman, Olivia Kurtz, and Ofelia Bank provided helpful comments, wrote back- Valladolid, as team assistants. Evangeline Santo ground papers, and made other contributions, Domingo served as resource management assis- and participated in consultation meetings. tant. These contributors and participants are listed in the Bibliographical Note. Book design, editing, and production were coor- The team undertook a wide range of consul- dinated by the Production Services Unit of the tations for this Report, from the initial outline World Bank’s Office of the Publisher, under the to the final draft. These consultations included supervision of Susan Graham, Melissa Edeburn, workshops in Berlin, Oslo, Paris, Washington, and Ilma Kramer. viii (c) The International Bank for Reconstruction and Development / The World Bank Foreword with environmental problems, but it underestimated T his year’s Report, the twenty-fifth, is about the growth in income and productivity the capacity of institutions to implement even poli- required in developing countries to eliminate cies that seemed on the surface to be win-win poverty in a way that is environmentally and socially options. The failure to implement them is most often sustainable. The core development challenge is to due to the social and political problems associated ensure productive work and a much better quality of with distributing costs and benefits within and life for the almost 3 billion poor people today earn- between groups and generations. ing less than $2 per day and for the 2–3 billion peo- This Report integrates the findings of the last few ple to be added to the world’s population over the WDRs into a broader and longer term framework to next 30–50 years. To achieve this goal, while taking identify some elements of a process that might do better care of our environmental and social assets, better: will require a global development process that does I The interaction between economic, social, and better than the one followed in the past. environmental problems and opportunities are Even though the world’s population increased by 2 manifested spatially—where people live. For billion people in the last 30 years, there have been this reason, the report takes a spatial perspective significant gains in human welfare in developing on the social transformations and the opportuni- countries as measured by average human develop- ties for growth and poverty reduction—in fragile ment indicators. But the development path has left a lands, in more favored agricultural lands, in urban legacy of accumulated environmental and social areas. Some of the local problems can be handled problems that cannot be repeated. There are many locally—but others must be dealt with nationally drivers of today’s socioeconomic and cultural trans- or globally. formations. Some are ongoing and continuous (such as technological innovation and income growth). I Problems that require lasting solutions often are Others are onetime and transitional, such as the not susceptible to quick fixes. Everyone could be demographic and urban transitions, which should be better off if cooperative solutions were agreed on completed within this century—largely within the and implemented. But often the rules and organi- next 50 years. These historic transitions define the zations to coordinate human behavior do not yet temporal and spatial context for managing sustain- exist, are undeveloped, faulty, or weak—especially ability. How it will be managed is critical. for problems with high transaction costs and Environmental and social assets matter greatly for longer time horizons. To be able to coordinate well well-being and productivity, but they are often neg- requires institutions that: lected. That is why we need to think about managing • Pick up signals about needs and problems, a broader portfolio of assets. The 1992 World especially from the fringes. Development Report identified many policies to deal • Balance competing interests. ix (c) The International Bank for Reconstruction and Development / The World Bank x WORLD DEVELOPMENT REPORT 2003 • Ensure credible commitments and accountabil- I Strengthening the foundations for better insti- ity in executing agreed decisions. tutions requires overcoming the inequitable access to assets and the pervasive barriers to I Institutions need to be improved at many lev- inclusion. The needed institutions (and the solu- els—from the local to the global—to promote tions to tough problems) do not emerge when growth in ways that protect environmental and some interests are dispersed or when some groups social assets. The institutions to manage and pro- in society are poor or in other ways disenfran- tect environmental and social assets are not emerg- chised. This affects the evolution and quality of ing rapidly enough to address the consequences of institutions and their ability to solve problems the growing scale and interconnectedness of over the longer term. The reciprocal relationship human activity. Action is required now—even for between the quality of institutions and the distri- problems that will unfold over a longer period. bution of assets can get countries locked into Societies need to ensure an enabling environment vicious cycles that require a special effort to break for creativity, initiative, and learning. These initia- out. tives can come from the public sector, the private sector, or civil society. Partnerships among these Inclusive societies, within and across countries, various actors are needed within and across coun- ensure that signals of emerging economic, social, or tries. Many innovative institutions are emerging environmental problems are picked up from all which need to be strengthened. The key is to find groups, and that they can cooperate to solve tough ways to scale up these initiatives. problems. Put another way, empowering poor people and the disenfranchised—the people “at the fringes�—and giving them a real stake in society is the key to building the stronger institutions required for longer term sustainable development. James D. Wolfensohn (c) The International Bank for Reconstruction and Development / The World Bank A C R O N Y M S A N D A B B R E V I AT I O N S ANC African National Congress IUCN Global Conservation Union BRAC Bangladesh Rural Advancement IWMI The International Water Management Committee Institute BSE Bovine spongiform encephalopathy MDG Millennium Development Goals (“mad cow disease�) MSC Marine Stewardship Council CCAMLR Convention on the Conservation of NEP New Economic Policy Antarctic Marine Living Resources NEPAD New Partnership for Africa’s Development CCD Convention to Combat Desertification NGOs Nongovernmental organizations CDF Comprehensive Development Framework NIC Newly industrializing country CGIAR Consultative Group on International NOAA U.S. National Oceanic and Atmospheric Agricultural Research Administration CIDA Canadian International Development NSDF National Slum Dwellers Federation (India) Agency ODESYPANO Sylvo-Pastoral Development Authority CIESIN Center for International Earth Science (Tunisia) Information Network OECD Organisation for Economic Co-operation CITES Convention on International Trade in and Development Endangered Species ORNL Oak Ridge National Lab CLRTAP Convention on Long-Range PAN Pesticides Action Network Transboundary Air Pollution PPP Purchasing power parity CSOs Civil society organizations PRSP Poverty Reduction Strategy Paper EDUCO El Salvador’s Community-Managed RSDF Railway Slum Dwellers Federation (India) Schools Program SINAMOS National System for Social Mobilization EPA U.S. Environmental Protection Agency (Peru) EROS Earth Resources Observation System SPARC Society for the Promotion of Area EU European Union Resource Centers (India) EWG Environmental Working Group TI Transparency International FAO Food and Agriculture Organization of the TRIPs Trade-Related Aspects of Intellectual United Nations Property Rights FDA U.S. Food and Drug Administration UCCI Union of Capital Cities of Ibero-America FONAFIFO National forestry fund (Costa Rica) UNDP/GEF United Nations Development FRA Forest Resources Assessment Programme/Global Environment Facility GEF Global Environment Facility UNEP United Nations Environment Programme GHG Greenhouse gas UNFCCC United Nations Framework Convention GIS Geographic information system on Climate Change GMO Genetically modified organism UNSO United Nations Statistical Office HIPC Heavily Indebted Poor Countries USDA U.S. Department of Agriculture ICRG International Country Risk Guide USGS U.S. Geological Survey IIED International Institute for Environment WCD World Commission on Dams and Development WHO World Health Organization IISA International Institute of Applied Systems WRI World Resources Institute Analysis WTO World Trade Organization IPCC Intergovernmental Panel on Climate WWF World Wildlife Fund Change ZIES Special residential zones of social interest IRRI International Rice Research Institute xi (c) The International Bank for Reconstruction and Development / The World Bank Roadmap to World Development Report 2003 Where are the problems? Why are there problems? How can the problems be solved? Chapter 1 Chapter 2 Chapter 3 CHALLENGES: Environmental and social Public goods and externalities create collective • Future growth is needed assets matter for well-being action problems. Coordination requires institutions in developing countries. and productivity but are that perform the following functions: • Past patterns of growth underprovided because of • Picking up signals (information, voice, feedback) worldwide have generated externalities. • Balancing interests (transparency, forums) costs that are not • Execution of agreed-on action (commitment sustainable. capability). • There are new opportunities on the Some problems have horizon. market solutions and some don’t. • Inclusivenenss (voice, access to assets) • Explore the potential for overcomes barriers to coordination. new markets to address • Institutions don’t appear overnight or fully problems amenable to mar- grown; they need catalysts to build momentum These problems and ket fixes. and scale up. opportunities are manifested • Problems that cannot be • Contexts change; therefore, sustainability spatially—in ecosystems solved by markets require requires that institutions adapt to changing and where people live. new approaches. conditions through innovation and learning. Use the institutional approach to address concrete problems where people live Chapters 4–8 Increasing density of human settlements Scale: Local Increasing scale of human activity and hierarchy of Chapter 4 Chapter 5 Chapter 6 Rural—low-productivity Rural—high-productivity Urban—from towns to agriculture, fragile lands agriculture, commercial scale cities to megacities spillovers (subsidiarity) Chapter 7 National (sovereign)—resolution of national problems and coordination of subnational issues Chapter 8 Global and transnational issues Chapter 9 Pathways to a sustainable future (c) The International Bank for Reconstruction and Development / The World Bank Roadmap to World Development Report 2003 How can productive work and a good quality of life Chapter 1 be provided for the 2.5–3 billion people now living Achievements and Challenges on less than $2 a day—and the 3 billion people likely Provision of productive work and a better quality of to be added to developing countries by 2050—in an life for current and future generations in developing environmentally and socially sustainable way? This countries will require substantial growth in income report asks where problems and opportunities are and productivity in these countries. This task will likely to arise, why the problems arise, and how they also require management of the social, economic, and can be solved (in fragile lands, relatively favored agri- environmental problems and opportunities accompa- cultural lands, and urban areas) and at different levels nying the transition to a predominantly urban world (local, national, and global). It argues that many (see figure 1); attention to the needs of the hundreds appropriate policies are known but not implemented of millions of people living on fragile lands; reaping because of distributional issues and institutional of the “demographic dividend� of declining depend- weaknesses. Competent institutions pick up signals, balance interests, and exe- cute agreed-on decisions. Figure 1—Opportunities seized—or lost? Demographic and urban transitions Inclusion of the poor and Population (billions) disenfranchised—giving 9 Global them a stake in society 8 population Megacities 54 stabilizes through voice and access to Cities 7 assets—will enable more Towns 29 6 effective institutions to Other rural Rapid urban Fragile lands 15 emerge. 5 growth 4 3 1 2 0 Still many in fragile areas 1 4 5 5 1 2 0 DTC OECD DTC OECD DTC OECD DTC OECD DTC OECD 1950 1970 2000 2030 2050 Note: DTC refers to developing and transition countries; OECD refers to high-income countries (and not all members of the Organisation for Economic Co-operation and Development). The numbers to the right of the columns show the number of megacities (cities in excess of 10 million people). Towns are classified as hav- ing a population of less than 100,000 and cities, a population of 100,000 to 10 million. Source: Authors; global population projections are based on World Bank estimates; estimates of population shifts in urban and rural areas are based on United Nations data. xiii (c) The International Bank for Reconstruction and Development / The World Bank xiv WORLD DEVELOPMENT REPORT 2003 ency rates and slowing population growth; and avoid- Why are environmental and social assets particu- ance of the social and environmental stresses—local larly threatened and underprovided? Because of and global—that might accompany achievement of externalities: the actions of one person may impose the prospective, mid-century, approximately $140 environmental costs (such as pollution) and social trillion world gross domestic product (GDP). costs (diminished trust in institutions) on other Although these problems and opportunities will play people—costs that the responsible party does not out differently in different places—in the ecosystems bear. Free riders have no incentive to contribute to where people live and the social systems where they the maintenance of public goods from which they interact, at scales that range from a small mountain cannot be excluded. valley to a coastal metropolis to the planetary bio- The solution to these problems is well known: sphere—many of the issues and the mechanisms policies that align individual and social incentives, needed to tackle them are common to all places. either through taxes, subsidies, and regulation, or through the deliberate creation of new market mech- Chapter 2 anisms. Failure to adopt such policies—even when Managing a Broader Portfolio of Assets they appear to be “win-win�—is most often the result of distributional problems and society’s inability to Social and environmental assets are critical— make credible long-term commitments. but underprovided. Policy solutions are under- stood—but not implemented. Chapter 3 Societies need to manage a broad portfolio of Institutions for Sustainable Development assets—not just human and physical capital, but also environmental assets (such as fresh water and fish Problems that require lasting solutions are often not stocks) and social assets (such as trust). These assets susceptible to quick fixes. Such problems require the are not perfectly substitutable. The immediate gains coordination of many actors. Inclusion in the form of of depleting or degrading them can be outweighed by voice and access to assets facilitates coordination: costs in productivity and lost options, as illustrated more inclusive processes lead to more sustainable by forest conversion in Madagascar. outcomes; voice and wider ownership of assets lead to Productivity growth in agriculture is critical to more inclusive processes. poverty reduction in Madagascar, where nearly three- Avoiding inflation and protecting investors, ensur- quarters of the population live in rural areas and ing labor and service delivery, maintaining environ- where three-quarters of that population is poor. But mental assets and systems for using them, preventing conversion of Madagascar’s biodiversity-rich forests, crimes and maintaining peace are all coordination the potential focus of a future ecotourism industry, to problems. Markets work well for addressing some mostly unsustainable, low-yield agriculture has been kinds of coordination problems—matching suppliers costly. Much of the new cropland is degraded, and and demanders of goods, services, and physical hillside erosion clogs downslope waterways. The assets—if supporting institutions such as property country has experienced a decrease in its per capita rights are in place. Mechanisms for other kinds of GDP from $383 (in 1995 dollars) in 1960 to $246 coordination problems, especially those in the social today. Madagascar is not the only country that has and environmental sphere, are often lacking, undevel- depleted or degraded forest assets without realizing oped, faulty, or weak. gains in other assets. On average, forest depletion in Coordination mechanisms typically fail in three low-income countries lowers net savings by 1.5 per- ways. First, they fail to take the long view. Cities grow cent of GDP. without adequate provision for transport right of way. Short-term political fixes evolve into constituencies for perverse subsidies. Second, they fail to represent dis- persed interests. The voices of the many who are affect- ed by pollution may be less audible than the voices of (c) The International Bank for Reconstruction and Development / The World Bank ROADMAP TO WORLD DEVELOPMENT REPORT 2003 xv those who pollute. Third, they fail to commit to allow lems), balancing interests (transparency, voice, assets to thrive. Wasteful destruction of forests, overex- forums for negotiation), and executing agreed-on ploitation of fisheries, plundering of people’s savings decisions (commitment and enforcement mecha- through inflationary monetary policies—all reflect a nisms). Such institutions are often lacking or are lack of social mechanisms for restraint. flawed, when some interests are dispersed or when The collapses of the Newfoundland cod fisheries some groups in society are poor or in other ways dis- and of the U.S. energy and financial and risk man- enfranchised. Groups that lack assets tend also to lack agement services company Enron illustrate these voice, security, and a stake in the larger society, ham- coordination failures—common problems in dis- pering institutions’ ability to perform needed coordi- parate realms. Potentially renewable assets—fish in nation functions. The result is a vicious cycle in one case, trust in the other—were depleted to the which biased institutions implement policies that short-run benefit of some but the long-run loss of lead to an increase in polarization and unequal asset society (see figure 2). distributions (see figure 3). Effective coordination requires institutions (infor- That policies affect institutions and asset distribu- mal and formal rules and organizations) that under- tion is widely understood; less well known is that take the following functions: picking up signals asset distribution affects the quality of institutions (information, feedback, anticipation of future prob- and policies. This cycle can be broken. Certain mechanisms for Figure 2—Failure of institutions to protect assets promoting transparency, feedback, accountability, commitment, and negotiation of interests have been successfully applied in fragile lands, rural areas, and urban areas. These mechanisms do not change insti- tutions overnight but help to build momentum for lasting change. Over the long run, fostering inclusive- ness is essential. South Africa and Malaysia, among other countries, demonstrate that societies can make decisive moves toward inclusiveness when it becomes clear that failure to do so will be unsustainable. Enron, market value, 1985–2002 Figure 3—Policy-institutions-assets loop Policies shape Policies institutions and the distribution Institutions of assets The distribution of assets shapes institutions Newfoundland cod catch, tons, 1850–2000 and policies Distribution of assets Sources: Hannesson (2002); Center for Research in Security Prices, University of Chicago; New York Stock Exchange. (c) The International Bank for Reconstruction and Development / The World Bank xvi WORLD DEVELOPMENT REPORT 2003 A local school and teacher enable girls to attend primary school for the first time in their village’s history. Photo cour- tesy of Robert Clement-Jones Chapter 4 By listening to grassroots organizations and test- Improving Livelihoods on Fragile Lands ing ideas, governments, civil society, and donors can promote creativity, adaptable institutions, relevant Living on fragile lands—in arid zones, on slopes and policies, and workable solutions to address the social, poor soils, or in forest ecosystems—are an estimated environmental, and economic problems affecting 1.3 billion people, (see figures 4 and 5) a number that one-quarter of the people in developing countries. has doubled over the past 50 years. The inhabitants of these fragile lands account for a large share of people in extreme poverty. Living in remote areas and working in the informal economy, these people are invisible to decisionmakers. Remotely located communities in some of the most fragile areas have a modest portfolio of assets that can help bring them out of poverty, but these assets are seldom nurtured by local or national institutions. Deftly combining resources for research and cost-effective services could enable these communities to catch up with more prosperous, less remotely located communities. Indeed, managing land to improve livelihoods underscores the strong link between tra- ditional know-how and outside techni- A Mongolian family moves camp for the winter. Mobility reduces overgrazing cal advice, which results in recognition pressures, promotes sustainable grassland management, and ensures acceptable of the land’s potential and limitations. livelihoods. Photo courtesy of Robin Mearns (c) The International Bank for Reconstruction and Development / The World Bank ROADMAP TO WORLD DEVELOPMENT REPORT 2003 xvii Figure 4—Fragile and nonfragile lands No constraint Aridity and combinations of aridity and soil contraints Soil as the only constraint Slope and combinations of slope and soil or slope and aridity constraints Nonfragile lands cover only a tiny fraction of the earth’s surface, bear most of the world’s population, and receive essentially all of the development attention. That the minority who live on fragile lands are nonetheless numerous, and are especially poor and voiceless, is a quintessential example of failure to balance interests. Figure 5—Rural population on fragile and nonfragile lands Rural population density (people per square kilometer) <1 1–50 50–250 >250 Fragile lands Nonfragile lands (c) The International Bank for Reconstruction and Development / The World Bank xviii WORLD DEVELOPMENT REPORT 2003 Figure 6—Projected water scarcity in 2025 For countries with physical water scarcity, no reallocation of water or construction of water supply structures will satisfy all water needs. These countries will have to transfer water from agriculture to other sectors and import food, or invest in costly desaliniza- tion. Countries with economic water scarcity will have to increase their primary water supply by more than 25 percent through addi- tional storage and conveyance facilities. Source: Prepared by the International Water Management Institute as input for the World Water Vision, The Hague, March 2000. Chapter 5 ing new deforestation control program in Mato Transforming Institutions on Grosso, Brazil, provides hope that these challenges Agricultural Lands can be met. An environmental licensing system uses satellite imagery and ground inspection to regulate Over the next 30 to 50 years, the key development land use by large landowners on five million hectares challenges for rural transformation are to eliminate in the state of Mato Grosso. Through this system, rural poverty and strengthen rural-urban linkages, the product of political will and technological and intensify agricultural production and sustainably institutional innovation, Mato Grosso enhances manage land and water to feed a growing population, enforcement of land use laws and deters wasteful control wasteful land conversion, and create off-farm conversion of Amazonian forest. economic opportunities. In rural areas with potential In the case of both water management and land for commercial agriculture, getting ahead of the management, two principles should apply. First, gov- “scarcity frontier� for both water and land is crucial. ernments must anticipate and attempt to prevent Many areas will experience physical or economic resource management problems before those prob- water scarcity by 2025 (see figure 6). Governments lems lead to severe environmental degradation. must establish institutions to equitably allocate water Second, where possible, allocation of rights to land rights and ensure adequate stream flows to maintain and water should favor poor people. These principles aquatic ecosystems. ensure that non-market environmental values are Similarly, governments must intervene to protect protected as the economic frontier advances and that the environmental values of land before economic economic assets are put in the hands of poor people, scarcity emerges. Mobilizing support for land man- developments that are good for the environment and agement interventions, and implementing them, for the evolution of local, regional, and national insti- presents a host of institutional challenges. A promis- tutions. (c) The International Bank for Reconstruction and Development / The World Bank ROADMAP TO WORLD DEVELOPMENT REPORT 2003 xix A nationwide program to upgrade kampongs (slum settlements) in urban areas of Indonesia has provided basic infrastructural invest- ments (water supply and sanitation, drainage, access roads and footpaths, lighting and other community service facilities) and secu- rity of tenure, dramatically improving living conditions for kampong residents and integrating their neighborhoods into cities. In this kampung in Banjarmasin in Kalimantan province, storm drainage works corrected chronic flooding. Photos courtesy of James Fitz Ford, the World Bank Chapter 6 tenure), stimulating urban investment and job cre- Getting the Best from Cities ation, and building informed constituencies to address environmental and social issues and anticipate risks. Urban areas are expected to grow significantly in the Institutions for urban governance need to link infor- next 30 years. The number of urban residents in mal networks of social capital to formal structures so developing countries and countries in economic that together they can address the increasing scale and transition will almost double through a combination complexity of environmental and social assets in cities, of rural-to-urban migration, natural population while promoting a well-integrated labor market and increases in cities, and reclassification of adjacent improvements in the investment climate. rural areas as urban areas. The growth of urban areas Often the urban poor have been left to fend for will require physical expansion of the urban periph- themselves, leading to the proliferation of large infor- ery as well as redevelopment and densification within mal settlements without services (slums) where resi- cities. dents face serious environmental hazards. This neg- The increase in share of national populations that lect creates high private as well as social costs. These will be living in urban areas (cities and towns) is one costs can be mitigated through corrective measures of the main forces of social and economic transfor- such as upgrading of investments through programs mation. The massive new investment in the capital that involve slum residents in arranging their own stock of cities required for the doubling of urban resettlement when necessary to increase their safety population by 2030 will be critical to environmental and protect environmentally fragile areas. A more outcomes. Urban land use patterns, right of way promising approach to upgrading of investments is to arrangements, and building standards will affect confirm the rights and responsibilities associated with energy and water use. the occupation and use of land, regularizing tenure Some key urban development challenges are antici- pating urban growth and guiding new settlements to prevent future slums, empowering the poor and excluded by providing access to assets (security of (c) The International Bank for Reconstruction and Development / The World Bank xx WORLD DEVELOPMENT REPORT 2003 status and thereby removing a major source of eco- Heavy reliance on natural resources (whether nomic and political insecurity for households and renewable, like forests, or nonrenewable, like miner- communities. Tenure reduces some of the risks that als) for public revenues can, in many cases, retard the discourage residents from investing in their houses emergence of strong institutions (important for both and shops—and gives residents a stronger stake in economic performance and sustainable development) urban society and an incentive to work with local because this reliance weakens government accounta- officials to obtain services. bility. Ensuring that development aid does not have a similar effect is a major focus in current efforts to Chapter 7 improve the effectiveness of such aid. Strengthening National Coordination The tragedy of violent conflict is more likely to visit countries with lootable natural resources and Because many externalities spill outside municipali- extreme poverty than countries without such ties and regions, the nation is often the level at which resources and poverty. Poverty reduction and other interests can be balanced—directly or through facili- forms of conflict prevention are essential, because the tation of negotiation among localities. National risk of conflict increases with poverty, economic stag- actors help to create a framework and solve problems nation, and a history of political turmoil. Providing that cannot be resolved at local levels and are better public goods, reducing negative externalities, and placed than local actors to organize the provision of avoiding conflict will require improved coordination nonlocal public goods and to take advantage of scale at the national level by promoting inclusiveness and economies when beneficiaries are spread among participation (through voice and improved access to many subnational regions. assets) as well as creating the framework to foster National concerns requiring coordination at dif- partnerships among stakeholders from government, ferent levels include the following: promoting inclu- civil society, and the private sector. siveness (by fostering access to assets and voice), gen- erating a sound investment climate (attending to Chapter 8 macroeconomic fundamentals, strengthening gover- Global Problems and Local Concerns nance, providing basic infrastructure), managing the environment (for instance, by regulating pollution Many local environmental and social problems spill and husbanding forests and fisheries), using aid over national borders. How can air pollution, water and natural resources effectively (by avoiding natural pollution, armed conflict, infectious disease, and resource depletion and degradation), and averting other problems be addressed without a global author- conflict. ity? Some institutions are finding ways to align inter- ests within and across borders to address the prob- lems of stratospheric ozone depletion and trans- boundary acid rain. Other institutions are emerging to facilitate international coordination, including broader use of standards and certification and of “coupling institutions� that link policymakers and scientists, nurturing the development of creative new solutions to problems. (c) The International Bank for Reconstruction and Development / The World Bank ROADMAP TO WORLD DEVELOPMENT REPORT 2003 xxi Figure 7—Population density in forests Population per square kilometer No forest <2 2–10 10–50 50+ Source: Authors’ construction based on Columbia University’s Center for International Earth Science Information Network’s Gridded Population of the World dataset (version 2) and Global Land Cover Characterization. Mapped forests include deciduous broadleaf, deciduous needle leaf, evergreen broadleaf, evergreen needle leaf, and mixed forest; not mapped are savannas, shrublands, wooded wetlands, and tundra. Two important global sustainability issues are Chapter 9 deeply connected to local land, water, and energy use Pathways to a Sustainable Future and have proven difficult to resolve: conserving bio- Lack of assets and lack of an effective voice for large diversity and maintaining ecosystems, and mitigating segments of the population block the emergence of and adapting to climate change. Strategies for biodi- competent institutions that can pick up signals early, versity conservation must operate at the level of entire balance interests, and commit to implementation of ecosystems. The ecological and social issues related to decisions. As a result, policies to avoid wasting of sustainable forest management, for instance, will be assets, particularly environmental and social assets, quite different for vast, unpopulated forests than they are not adopted and implemented. The more people will be for highly fragmented and densely populated heard, the fewer the assets that are wasted. These forests (see figure 7). messages of World Development Report 2003 inform a Climate change, if unchecked, could have severe variety of recommendations and suggest some issues consequences. Long lead times, and concerted action requiring further research and dialogue, including a at the required scale, are necessary to effect changes global vision and accord on sustainable development. in both economic systems and the global climate system. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 1 Achievements and Challenges W orld Development Report 2003 is about sus- The core development challenge tainable development. It is about people Most current estimates suggest that 2 billion people and how we deal with each other. It is will be added to the world’s population over the next about our home planet and its fabric of life. And it 30 years and another billion in the following 20 is about our aspirations for prosperity and posterity. years.1 Virtually all of this increase will be in devel- Any serious attempt at reducing poverty requires oping countries, the bulk of it in urban areas. In sustained economic growth in order to increase pro- these same countries, 2.5 billion to 3 billion people ductivity and income in developing countries. But now live on less than $2 a day.2 The core challenge there is more to development than just economic for development is to ensure productive work and a growth—much more. This Report argues that ensur- better quality of life for all these people. This will ing sustainable development requires attention not require substantial growth in productivity and in- just to economic growth but also to environmental comes in developing countries. and social issues. Unless the transformation of soci- The challenge may seem daunting—and it is. But ety and the management of the environment are over the past 30 years world population also rose addressed integrally along with economic growth, by 2 billion.3 And this growth was accompanied growth itself will be jeopardized over the longer term. by considerable progress in improving human well- Environment and social issues, when not ad- being, as measured by human development in- dressed, accumulate over time and have conse- dicators. Average income per capita (population- quences that do not show up in the shorter time weighted in 1995 dollars) in developing countries horizons typical of economic policymaking. That is grew from $989 in 1980 to $1,354 in 2000.4 Infant why this Report adopts a longer time horizon of 20 mortality was cut in half, from 107 per 1,000 live to 50 years. Within this time frame it is possible births to 58, as was adult illiteracy, from 47 to 25 to identify environmental and social problems— percent.5 local, national, and global—that can have very costly Looking back to the 1950s and 1960s, it was or even irreversible consequences if not addressed feared at the time that the developing countries— immediately. For other problems, where the conse- particularly China, India, and Indonesia—would quences are not irreversible, the longer time horizon not be able to feed their rapidly growing popula- provides the lead time to start changing attitudes tions. Thanks to the green revolution in agriculture, and institutions and so make it possible to respond the doomsday scenarios of famine and starvation did before the problems become crises. not materialize in these, the most populous, devel- In short, this Report takes a comprehensive, longer oping countries. In the 1960s and 1970s the Club of term, and dynamic view of sustainability, with a clear Rome and many other groups forecast that the Earth focus on poverty reduction. would rapidly run out of key natural resources. So  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  far, this has not happened, again because changes in The increased scale and reach of human activity technology and in preferences have allowed the sub- have also put great pressure on local and global com- stitution of new resources for existing ones—for ex- mon property resources (water, soil, and fisheries), ample, fiber optics in place of copper. Global action as well as on local and global sinks (the ability of the has also led to major strides in eliminating disease biosphere to absorb waste and regulate climate). scourges (smallpox and river blindness), and in ad- dressing new problems (ozone depletion). Ⅲ Air: polluted. At the local level, hundreds of develop- But accompanying these achievements were some ing-country cities have unhealthy levels of air pollu- negative social and environmental patterns that must tion (see chapter 3, figure 3.4). At the global level, not be repeated in the next 50 years if development the biosphere’s capacity to absorb carbon dioxide is to be sustained. without altering temperatures has been compro- mised because of heavy reliance on fossil fuels for Ⅲ Poverty: declining, but still a challenge. There has energy. Global energy use traditionally has grown at been a significant drop in the percentage of peo- the same rate as gross domestic product (GDP). ple living in extreme poverty (that is, living on less Greenhouse gas (GHG) emissions will continue to than $1 per day). Even the absolute number of grow unless a concerted effort is made to increase very poor people declined between 1980 and energy efficiency and move away from today’s heavy 1998 by at least 200 million, to almost 1.2 billion reliance on fossil fuels.11 In the past 50 years excess nitrogen—mainly from fertilizers, human sewage, in 1998.6 The decrease was primarily due to the and combustion of fossil fuels—has begun to over- decline in the number of very poor people in whelm the global nitrogen cycle, giving rise to a va- China as a result of its strong growth from 1980 riety of ill effects ranging from reduced soil fertility onward.7 Since 1993, there have also been encour- to excess nutrients in lakes, rivers, and coastal wa- aging signs of renewed poverty reduction in India. ters. On current trends, the amount of biologically Sub-Saharan Africa, by contrast, has seen its num- available nitrogen will double in 25 years.12 ber of very poor people increase steadily. Yet in Ⅲ Fresh water: increasingly scarce. Fresh water con- 1998, despite the decline in Asia and the increase sumption is rising quickly, and the availability of in Sub-Saharan Africa, East Asia and South Asia water in some regions is likely to become one of still accounted for two-thirds of the world’s very the most pressing issues of the 21st century. One- poor people, and Sub-Saharan Africa for one- third of the world’s people live in countries that quarter. Development strategies will need to do are already experiencing moderate to high water better in eliminating abject poverty. The estimated shortages. That proportion could (at current pop- 1 billion very poor people is of the same order of ulation forecasts) rise to half or more in the next magnitude as the independently generated figures 30 years unless institutions change to ensure bet- on the number of people who are undernourished ter conservation and allocation of water.13 More and underweight.8 than a billion people in low- and middle-income Ⅲ Inequality: widening. The average income in the countries—and 50 million people in high-income richest 20 countries is now 37 times that in the countries—lacked access to safe water for drink- poorest 20. This ratio has doubled in the past 40 ing, personal hygiene, and domestic use in 1995.14 years, mainly because of lack of growth in the Ⅲ Soil: being degraded. Nearly 2 million hectares of poorest countries.9 Similar increases in inequality land worldwide (23 percent of all cropland, pas- are found within many (but not all) countries. ture, forest, and woodland) have been degraded Ⅲ Conflict: devastating. In the 1990s, 46 countries since the 1950s. About 39 percent of these lands were involved in conflict, primarily civil.10 This are lightly degraded, 46 percent moderately de- included more than half of the poorest countries graded, and 16 percent so severely degraded that (17 out of 33). These conflicts have very high the change is too costly to reverse. Some areas face costs, destroying past development gains and leav- sharp losses in productivity. Grasslands do not fare ing a legacy of damaged assets and mistrust that much better: close to 54 percent show degrada- impedes future gains. tion, with 5 percent being strongly degraded.15 (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������  Ⅲ Forests: being destroyed. Deforestation is proceed- tal concerns in pursuit of the goal of sustained im- ing at a significant rate. One-fifth of all tropical provements in well-being. forests have been cleared since 1960.16 According to the Food and Agriculture Organization of the Windows of opportunity United Nations (FAO), deforestation has been The development process is about change and trans- concentrated in the developing world, which lost formation. Economies evolve. Societies and cultures nearly 200 million hectares between 1980 and evolve. Nature evolves. But they evolve at different 1995. In the Brazilian Amazon annual deforesta- speeds, creating stresses that need to be addressed and tion rates varied between 11,000 and 29,000 square managed.22 Moreover, in an era of globalization, the kilometers a year in the 1990s. Deforestation in growing scale and speed of change in human activity developing countries has several causes, including are in some cases outpacing the rate at which natural the conversion of forests to large-scale ranching processes and life-support systems can adapt.23 Glob- and plantations and the expansion of subsistence alization and faster technological change are also al- farming. At the same time, forest cover in indus- tering the nature of social interaction and affecting trial countries is stable or even increasing slightly, the efficacy of existing institutions. Although global- although the forest ecosystem has been somewhat ization and technological change offer many benefits, altered. According to a 1997 World Resources In- they can have deleterious side effects if institutions at stitute (WRI) assessment, just one-fifth of the local, national, and international levels do not evolve Earth’s original forest remains in large, relatively fast enough to deal with the adverse spillovers. The natural ecosystems.17 consequences of previous patterns of development Ⅲ Biodiversity: disappearing. Through a series of local are also beginning to bind, restricting certain growth extinctions, the ranges of many plants and ani- paths or making them more costly.24 mals have been reduced from those at the be- But these processes, if managed well, can create ginning of the century. In addition, many plants new opportunities. Of the many interrelated drivers and animals are unique to certain areas. One-third of change and transformation, four stand out: scien- of terrestrial biodiversity, accounting for 1.4 per- tific and technological innovation, income growth, cent of the Earth’s surface, is in vulnerable “hot population growth, and urbanization. The first two spots� and is threatened with complete loss in the are likely to continue changing preferences and pro- event of natural disasters or further human en- viding new opportunities to satisfy these preferences. croachment.18 Some statistics suggest that 20 per- The demographic and urban transitions, by contrast, cent of all endangered species are threatened by are one-time changes, and the opportunities they species, introduced by human activity, alien to the offer are perhaps less well recognized. These are dis- locality.19 cussed in the next section. Ⅲ Fisheries: declining. The aquatic environment and its productivity are on the decline. About 58 per- Ⅲ Scientific and technological innovation. The flow of cent of the world’s coral reefs and 34 percent of all information and ideas, boosted greatly by the In- fish species are at risk from human activities.20 ternet, can enable developing countries to learn Seventy percent of the world’s commercial fish- more rapidly from each other and from industrial eries are fully exploited or overexploited and expe- countries. It can also facilitate the emergence of riencing declining yields.21 networks to monitor a wider array of development impacts. Other technological changes can enable None of these social and environmental patterns developing countries to leapfrog stages in the de- is consistent with sustained growth in an interde- velopment process that rely on inefficient uses of pendent world over the long term. Given the social natural resources. Science and technology can help and environmental stresses caused by past develop- address major socioeconomic problems. As noted, ment strategies, the goal of raising human well-being the green revolution was critical in enabling many worldwide must be pursued through a development developing countries to avoid widespread starva- process that “does better�—a poverty-eliminating tion. To benefit from these opportunities, institu- growth path that integrates social and environmen- tions are needed that can stimulate and diffuse (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  technological innovations and avoid or mitigate Figure 1.1 any deleterious consequences. Global population approaching stability Ⅲ Income growth. A projected growth in global in- come of 3 percent a year over the next 50 years im- People (millions) plies a fourfold increase in global GDP. Increasing 12 income growth may place a strain on the environ- Medium-high Total (medium) mental and social fabric if there is too little atten- 10 Medium-low tion to shifting consumption and production pat- terns. But this future economic growth will also 8 require major investments in new human-made capital to expand capacity and to replace existing 6 capacity as it ages. Making these investments (many of which are long lived) more environmen- tally and socially responsible through appropriate 4 investment criteria will go a long way toward put- ting development on a more sustainable path—an 2 opportunity not to be missed. 0 Opportunities in the demographic transition 1950 1970 2000 2030 2050 When today’s industrial countries were themselves Year developing, their population densities and growth Note: Medium-high and medium-low variants based on U.N. projections rates were much lower than those of developing of medium-high and medium-low scenarios scaled to World Bank countries today, and the pressure on their resources aggregates. Source: World Bank estimates. was consequently lower. They also had a more evenly distributed age structure and lower dependency rates, allowing social institutions to adapt gradually to the requirements of a changing population. Ⅲ Greater off-farm opportunities, creating a need for Populations in industrial countries as a group were more education for children fairly stable for most of the second half of the 20th Ⅲ Widespread dissemination of modern contracep- century. As a result, the growth in world population tive technology, making it easier for people to plan in this period has been driven primarily by popula- childbearing. tion growth in developing countries. The stresses and spillovers from this population growth are generally Of the expected population increase, 85 percent observed not, as was originally expected, at the ag- (3 billion) will be born in the next 50 years (figure gregate level (for example, in large-scale famines and 1.1). But the speed of the transition, and the result- food shortages) but, rather, in more insidious ways— ing population size and structure, will vary by region in many smaller interactions between population, (figure 1.2) and by country. If fertility rates do not poverty, and resources.25 The outcomes are felt in fall as rapidly as now projected, aggregate popula- greater pressures on fragile lands, in lower wages, and tions will be larger, putting greater pressures on nat- in persistent unemployment. ural resources and the social fabric. If they drop It is now clear that a global demographic transi- faster, many countries will have to deal sooner than tion is well underway, even if it is not yet complete. expected with another problem—an aging popula- This is a major historic opportunity. World popu- tion. This can have major consequences, especially lation is expected to stabilize by the end of this for rural populations, for whom formal social safety century at 9 billion to 10 billion people, 20 to 30 nets are either nonexistent or not well developed. percent lower than forecast in the 1960s and 1970s. For example, one consequence of China’s one-child Many factors have contributed to this slowdown: policy—which dramatically and successfully low- ered aggregate population—may be that by 2030 Ⅲ More educated, employed women and smaller as much as one-third of the population will be over families age 65.26 (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������  Figure 1.2 members of the working-age population are gain- Some regions growing fast, others stable fully employed and have opportunities to expand their asset base. Eventually, dependency ratios rise People (millions) again as these workers age, and the window of op- 3 portunity starts to close, as it will soon begin doing East Asia and Pacific South Asia in East Asia and Eastern Europe (see figure 1.3). Europe and Central Asia Some regions, notably East Asia, have benefited Latin America and the Caribbean Middle East and North Africa substantially from the drop in the ratio of depen- 2 Sub-Saharan Africa High-income OECD dents to workers.27 Investment in forming a skilled, healthy labor force, combined with policy and insti- tutional settings conducive to using this labor force effectively, helped generate strong economic growth. Two keys to success were maintenance of an open 1 economy and investment in sectors with high growth potential. Since most developing regions will con- tinue to experience relatively low dependency ratios for some decades, careful preparation now can help make the most of their windows of opportunity. 0 Until now, populations have been growing too 1960 1970 2000 2030 2050 rapidly for fiscally constrained governments to ex- Year pand the provision of jobs, infrastructure, and pub- Source: WDI database and SIMA. lic services enough to keep pace with people’s needs. This task will become easier now that the global population is approaching stability. Governments in Influencing the demographic processes in many both urban and rural areas can move from catching countries is the growing incidence of HIV/AIDS, up with the quantitative need for services, to upgrad- malaria, and tuberculosis. For example, current esti- ing their quality. Much of the social tension and mates and projections in Sub-Saharan Africa indi- frustration arising from unemployment and poor cate increasingly large losses of working-age people public services can then be attenuated. to the AIDS epidemic. The economic impact of such Lower rates of population growth will reduce high mortality is especially serious because enormous pressure on natural resources, but this will be offset private and public investments have already been by the increase in per capita consumption. The lat- made in members of this age group. The loss of their ter trend makes it essential to adopt the technologies productive lives leaves large and unpredictable gaps and growth paths for production and consumption in the labor force. Malaria causes high levels of adult that will ensure the sustainable use of natural re- sickness rather than deaths, but this too inflicts sources. To benefit from the opportunities a stabiliz- heavy losses on labor productivity. Changes in the ing population provides, it is critical to anticipate incidence of disease will have profound effects on problems and identify development strategies for health expenditures in these African countries. getting through the transition period (the next 20 to With declining fertility, the age structure of the 50 years) without creating conditions that generate population changes, opening a window of opportu- further conflict or resource degradation. nity in developing countries for a few decades—a window they can use for catching up and raising Opportunities in the urban transition welfare for all. As figure 1.3 shows, the proportion As countries move from poverty to affluence, the re- of the working-age population rises in relation to the quired growth in productivity involves a shift from proportions of children (those under 15) and the el- heavy dependence on agriculture as a primary source derly (over 65), enabling societies to spend less on of employment and income to nonagricultural activ- school construction and on old-age medical expenses ities that do not make intensive use of land. This is and to invest the savings in generating economic generally accompanied by a major shift in popu- growth. But such benefits will materialize only if the lation from rural to urban areas. Indeed, the most (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Figure 1.3 Dependency ratios on the decline—for a while Dependency ratio* 100 90 East Asia and the Pacific 80 Eastern Europe and Central Asia 70 South Asia 60 High-income OECD countries 50 40 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Year Dependency ratio* 100 90 Latin America and the Caribbean 80 Middle East and North Africa 70 Sub-Saharan Africa 60 High-income OECD countries 50 40 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Year * The dependency ratio is the ratio of the non-working-age population (under 15 years old and over 64 years old) to the working-age population (ages 15 to 64). Source: World Bank (2001g) important socioeconomic and cultural transforma- ecosystems, however, imposed limits on risk taking tion over the past 150 years has been the transfor- and innovation. This autonomous development path mation of relatively closed, exclusive, custom-based changes as rural areas become drawn into larger mar- rural societies into relatively open, inclusive, innova- kets and strengthen their links with urban areas, tion-oriented urban societies.28 making trade networks and distance from market Rural communities, especially in less accessible centers more critical features of development oppor- areas, have long adapted to their circumstances, de- tunities and local resource pressures. veloping vibrant, self-sufficient communities. As long Increasing densities in towns and cities, and the as risks could be absorbed locally, these communities greater connectivity between cities, as well as between continued to learn and adapt. Dependence on local urban and rural areas, increases the catchment area of (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������  markets and the returns to economic endeavor. If focusing on GDP and the emergence or obsoles- managed well, this transformation enables the emer- cence of industries, but it is not very helpful for un- gence of new activities and productive job opportuni- derstanding the impact of these changes on society ties. Towns, as market centers for a rural hinterland, and nature. The most fundamental social and eco- start the process of creating economies of scale for nomic transformation—from traditional rural to nonagricultural activities. Urban society also permits modern urban—is manifested spatially. Except in the spreading of risks over larger numbers of people the most populous countries, such as China and and activities. Knowledge flows more readily, through India, rural societies are relatively low in density and increased opportunities for face-to-face contacts among heavily dependent on agriculture as the primary various actors. And the need to accommodate diverse source of employment and output. Modern urban views and meet rapidly changing challenges stimulates societies are generally higher in density and depen- innovation and new applications of technology. As a dent on activities that benefit from proximity and result, larger cities become incubators of new values— do not require a great deal of land, such as manufac- among them, risk taking and innovation. turing and services. These activities and land use pat- Creativity, knowledge flow, the increasing scale of terns generate different types of sociocultural and activities, and larger catchment areas are central to environmental problems. specialization and productivity growth. This is true Most ecosystems, too, are defined spatially. Much not just for the production of goods but also for the flora and fauna is locally unique and adapts gradu- provision of services. A village or neighborhood can ally to changes in local circumstances. Local prob- support a primary school or basic clinic, and the local lems and stresses appear earliest, whether in the form teacher or doctor can be a generalist. But providing of local extinctions, the reduction of the ranges of higher, more sophisticated, and more differentiated many plants and animals, or soil, air, and water pol- education and health care requires more specialized lution. These changes, the result of local develop- skills. Because of the fixed costs of supporting these ment pressures, do not show up at national and specialized skills, a larger catchment area (a town or global levels until they accumulate, but they provide a subsection of a city) is required. The higher popu- early warning of problematic consequences of cur- lation densities, lower transport costs, and lower com- rent development patterns. munications costs in towns and cities make the more The jurisdictions of many institutions that make specialized operations possible. In moving further up or implement rules and laws (legislatures, constitu- the hierarchy of required specialization, the required tions, and government agencies) are also defined spa- catchment area also increases. So, the transition from tially. Often, the spatial jurisdiction of institutions villages to towns, and from cities to metropolitan does not match the spatial nature of the social and en- areas, corresponds to the different functional capa- vironmental problems generated by economic activ- bilities of larger, higher-density conurbations. The ity—one reason for the persistence of these problems. potential benefits of higher densities and greater con- Given our interest in people, where they live, and nectivity can be more easily realized if the investment how they interact with each other and with nature, it climate is improved through better enabling rules is important to look at where people are now and and frameworks and better physical infrastructure. where they are likely to be in the future. The world’s Stimulating and attracting investments—in particu- population increased by more than 3.5 billion people lar, by the small and medium-size enterprises that in the past 50 years, and 85 percent of these added provide most of the jobs for growing urban popula- people were in developing and transition countries tions—is the key to accommodating the expected (see Figure 1 in the Roadmap). The number of peo- growth in urban populations and ensuring their abil- ple living in fragile rural areas in developing countries ity to pay for needed urban services and amenities. doubled, in stark contrast to the declining numbers in this category in high-income countries. The num- Seeing the socioeconomic transformations ber of cities with a population of more than 10 mil- in spatial terms lion people went from 0 to 15 in developing countries Economists and engineers focus on the sectoral but only from 1 to 4 in high-income countries. changes that accompany economic growth and tech- In the next 30 to 50 years the 2 billion to 3 bil- nological innovations. This is understandable when lion increase in the world’s population will be almost (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  exclusively (97 percent) in developing and transition identifies an approach and process that should gen- countries, and virtually all of it will be in urban erate more dialogue and creativity in finding answers. areas. The growth of the urban population is driven The interactions among society, economy, and by natural increase, rural-to-urban migration, and nature vary in the different spatial arenas, although the incorporation of high-density rural areas on the problems across locations are linked. Productivity in- urban fringe. The number of megacities in develop- creases in agriculture help feed the cities. Innovation ing countries is likely to increase to 54, while it will and productivity increases in the cities help raise pro- stabilize at 5 in high-income countries. It is not yet ductivity and the quality of life in rural areas. Geog- clear whether the number of people living in fragile raphy matters because of the characteristics of local areas will continue to increase, but it probably will ecosystems, such as the cost of overcoming local dis- unless migration opportunities change. As many as eases.30 Geography also matters because of geometry 2 billion people will live in two areas that are diffi- in the form of connectivity and distance to central cult to manage: fragile rural areas and megacities.29 nodes and markets; the cost of transport is more im- Dealing with these people’s needs will be a major portant here than that of communication.31 Indeed, challenge, since there is not much experience in in- the strong association between rural poverty in re- dustrial countries that can be adapted to their needs. mote and fragile ecosystems becomes more apparent The following are some of the key questions with when the problem is viewed through a spatial lens. local and global implications that will face the For this reason, the Report is organized by spatial world’s population over the next two to five decades: areas that have different characteristics and require correspondingly different approaches to their devel- Ⅲ Will rural populations—especially those on frag- opment. ile lands, in more commercially active areas, and Fragile lands. The estimated 1.3 billion people liv- on agricultural frontiers—be able to overcome ing on fragile lands have modest assets that can help poverty, improve their livelihoods, and adapt to bring them out of extreme poverty, but these assets new opportunities, including opportunities in towns are seldom nurtured by local or national institutions. and cities? The people have land that is subject to many con- Ⅲ Will the rapidly growing cities of the developing straints, making it vulnerable to degradation, erosion, world live up to their potential as dynamic engines floods, and landslides. They possess human capital, of growth and social modernization, or will they which is handicapped by restrictive traditions, lim- get mired in poverty, pollution, congestion, and ited mobility, lack of voice, and poor access to ser- crime? vices. This is even more true for women, who are thus Ⅲ Will renewable resources—particularly forests, the most marginal group. The mainly poor people on soil, water, biodiversity, and fisheries—be depleted, fragile lands also face circumstances vastly different or will they be managed as indefinitely sustained from their counterparts on Europe’s rural periphery sources of livelihood and well-being? 50 to 100 years ago. Today, international migration Ⅲ Will societies be sufficiently creative, resilient, and is highly restricted, and while rural-to-urban migra- forward-looking as they undergo sweeping trans- tion is important for them, there are limited numbers formations in patterns of growth and migration? of jobs at above-subsistence wages for unskilled work- Will they be able to promote more equitable de- ers, especially in the low-growth economies. As a re- velopment and cope with unexpected shocks? sult, as noted above, instead of declining sharply, the Ⅲ Will poor countries be able to accelerate their number of people living on fragile lands is estimated growth without destabilizing social and environ- to have doubled in the past 50 years—despite some mental stresses? Will the prospective $140 trillion outmigration. world GDP at mid-century generate fewer en- Rural areas with potential for commercial crops. vironmental and social stresses than the much The problem of feeding a growing and more urban smaller global economy today? population calls for better management of the in- teraction with nature, particularly with respect to These are difficult but important questions, which land and water (extensification versus intensifica- this Report cannot answer definitely. However, it tion of agriculture). Whether or not rural families (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������  have land, water, and education is critical to their ternalities that affect wider catchments be addressed, current livelihood, as well as to their ability to move at higher levels—national and global. to cities in the future. More egalitarian access to At the national level. The political, legal, and these assets is also crucial for determining the qual- market domain for coordinating many activities is ity of society’s institutions. A successful rural-urban frequently the nation. Many externalities spill over transition requires the elimination of poverty for beyond local communities and municipalities, and those who stay in the countryside and better prepa- even across regional boundaries. The nation is thus ration of those who move to the cities. It also de- often the level at which interests can be balanced, mands protection of remaining natural ecosystems either directly or by facilitating negotiation among and habitats, given their central role in maintaining localities. National actors may be better placed to life-support systems and biodiversity. This latter re- organize the provision of public goods and to take quirement is one reason to intensify agricultural advantage of scale economies when the beneficiaries production in areas already under commercial crops extend beyond subnational regions. Generating a and pasture. Intensification in such areas not only strong investment climate, including sound macro- minimizes pressure on biodiversity and on marginal economic fundamentals, good governance, and basic agricultural areas but also increases the food avail- infrastructure, requires a framework that is typically able to cities and leads to dynamic rural-urban link- national in scope. Dismantling perverse subsidies, ages. Higher population density in these rural areas husbanding forests and fisheries, and curbing water would also make investments in health and educa- and air pollution in river basins and airsheds are tion more cost-effective and would increase the po- major national challenges. Managing foreign aid and tential for off-farm employment and help farmers avoiding civil conflict are other key national concerns accept risk and innovate. that determine whether development is sustainable. Urban areas. Cities of the developing world face a At the global level. Many economic, environmen- formidable undertaking, given the expected rapid tal, and social processes—knowledge, conflict, dis- rate of growth and sheer numbers of urban residents ease, pollution, migration, and finance—spill over to be employed, housed, and serviced. The charac- national boundaries. A few of these processes gener- teristics of periurban settlements, towns, cities, and ate problems that are purely global: depletion of the megacities—higher density, large scale of settlement, stratospheric ozone layer is an example. But most and greater social diversity—facilitate the creation of global problems and opportunities are experienced productive employment opportunities, efficient pro- at the local level as well. Automobiles that pollute vision of services, and access to ideas and learning. local airsheds also generate greenhouse gases; wet- But having many people at close quarters also creates land destruction that disrupts local water resources the potential for social problems—crime and social also undermines biodiversity of global significance; dislocation—and for environmental spillovers that new ideas that are generated in one place can bene- pose health and safety hazards, especially for those fit people in other places, near and far. The public living in neighborhoods without sanitation or drain- goods nature of many of these issues and the need to age and in potential disaster zones. The long life of address the negative externalities requires coordina- urban physical capital stock can lock in certain de- tion across boundaries. The distinctive challenge for velopment paths, making changes costly. If managed global issues is to balance interests and commit to well, urban areas can be the future engines of growth. solutions in the absence of a global authority. If not, their environmental and social problems will Act now—for long-term problems be concentrated and difficult to fix. The discussion of problems affecting fragile lands, Before proceeding to a discussion of local, national, rural commercial areas, and urban settings, and of and global issues, this Report sets forth a framework possible solutions, is important because many pub- which argues that social and environmental out- lic goods and externalities are local in nature and are, comes have a bearing on human well-being both in principle, amenable to action at the local level. An directly and through their effect on growth. When enabling framework for local action and the princi- social and environmental issues are systematically ple of subsidiarity require that public goods and ex- neglected for long periods, economic growth will be (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  affected. That is why improving the quality of life near term could avert much greater costs and disrup- for those living in poverty today—and for the 2 bil- tion to human development in the longer term. lion to 3 billion people who will be added to the In looking back over past successes and failures in world’s population over the next 50 years—will re- solving development problems, it is clear that there quire a growth path that integrates environmental have been more successes where markets function and social concerns more explicitly. well (for example, in providing food to people with Some problems of sustainability are already ur- effective demand), even where the problems that gent and require immediate action; examples are markets have to solve (such as transport and commu- local ecosystems where population is pressing on nications) are relatively complex. The major prob- deeply degraded soils, and forests and water stocks lems that remain (inclusion, poverty reduction, de- that have been nearly depleted. In such cases produc- forestation, biodiversity, and global warming) are, tivity is already on the decline and opportunities for however, generally not amenable to standard market correction or mitigation may even have been lost; solutions, although markets can help solve subsets of abandonment of existing practices and outmigration these problems. may be necessary. The urgency of some of these One difficulty is that environmental and social as- problems has been overlooked because the people sets suffer from underinvestment and overuse be- most affected are physically remote from centers of cause they have the characteristics of public goods: power, or because their voices are not heard, or both. Some issues call for immediate action because Ⅲ Sometimes, ignorance of the consequences of ac- there are good prospects for reversing the damage to tion leads to overuse or underprovision. The igno- the environment at relatively low cost, as in taking rance is in part due to underinvestment in knowl- measures against air and water pollution. Even then, edge and understanding—itself a public good.32 undoing some of the damage to the affected popu- Ⅲ In other cases there are no mechanisms for facili- lation (such as the respiratory damage caused by tating cooperation among individuals, communi- breathing air laden with particulates) may not be fully ties, or countries even when it is clear to those in- possible. But knowing the health impacts does create volved that the returns to cooperation (especially a moral imperative to protect those affected from fur- in the long run), exceed the returns to unilateral ther exposure, to compensate them to the extent pos- action (especially in the short run). sible, and to prevent others from becoming victims. Ⅲ In still other cases the gains from acting in the Another category of issues unfolds over a longer broader interests of society fail to be realized be- time horizon. The problems may not yet be urgent, cause correcting a spillover has distributional con- but the direction of change is unmistakable. For sequences and the potential losers resist change. these, it is essential to get ahead of the curve and pre- Ⅲ Sometimes underprovision is a response to per- vent a worsening crisis before it is too costly. Biodi- ceived tradeoffs between growth and the costs of versity loss and climate change are in this category: correcting externalities. These tradeoffs may be there is already a need to adapt to the consequences the unfortunate outcome of having been boxed of past and current behavior, but there is also still into a corner through a past failure of foresight. scope for mitigation, though not for complacency. Or there may be genuinely difficult choices in bal- Similarly, the need to anticipate urban growth by fa- ancing legitimate interests and assessing the value cilitating low-income settlements in safe areas and of nonmarket benefits and risk reduction, espe- by setting aside major rights-of-way and spaces for cially if those who would benefit are dispersed public amenities makes it necessary to act now to over current and future generations. avoid greater costs and regrets later. What is clear is that almost all of the challenges of Environmental and social stresses reflect the fail- sustainable development require that action be initi- ure of institutions to manage and provide public ated in the near term, whether to confront immedi- goods, to correct spillovers, and to broker differing ate crises, such as the health risks to children from interests. Because the spatial extent of spillovers varies unsanitary living conditions in existing slums, or to by problem, appropriate institutions are needed at stem the tide of crises where concerted action in the different levels, from local through national to global. (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������  Getting to socially preferred outcomes requires insti- cases, and increasingly, institutions respond too late tutions that can identify who bears the burden of so- or too poorly—or without the capacity to commit cial and environmental neglect and who benefits— to a course of action. In today’s world the lag be- and who can balance these diverse interests within tween the emergence of a problem and the emer- society. This perspective helps in understanding why gence of institutions that can respond to it is too technically sound policy advice (for instance, “elimi- long. We need to see farther down the road. Why? nate perverse incentives� or “impose charges on envi- Because institutions that facilitate and manage na- ronmental damages�) is so seldom taken up. tional economic growth, and even globalization, are The emphasis of this Report is not on identifying still inadequate, yet where such institutions are in a specific set of policies or outcomes considered ad- fact emerging, they are developing faster than com- vantageous but on the processes by which such poli- plementary institutions that might be able to avoid cies and outcomes are selected. Outcomes emerging or cope with the deleterious environmental and so- from strong processes are more robust. In many cial consequences of economic change. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 2 Managing a Broader Portfolio of Assets What we are doing to the forests of the world is but a on a sustained basis requires that society manage a mirror reflection of what we are doing to ourselves and portfolio of assets. Different assets have different to one another. characteristics that limit the extent to which they can substitute for one another in production and in —Mahatma Gandhi human well-being. This chapter discusses the broad concerns that S ustainable development is about enhancing human need to be taken into account when balancing the well-being through time. What constitutes a good objectives of economic growth and attending to en- life is highly subjective, and the relative impor- vironmental considerations and their social under- tance accorded to different aspects of well-being varies pinnings in the short to medium term—recogniz- for individuals, societies, and generations.1 But on some ing that over the longer term prolonged neglect of elements most people could probably agree. Having the environmental and social assets is likely to jeopar- ability and opportunity to shape one’s life—which dize the durability of economic growth. More specifi- increase with better health, education, and material cally, it addresses the following questions: comfort—is certainly one of them. Having a sense of self-worth is another, enhanced by family and social re- Ⅲ What is meant by sustainable development and lationships, inclusiveness, and participation in society. how can progress toward it be measured? Al- So is enjoying physical security and basic civil and po- though the adjusted net savings indicator is a po- litical liberties. And so is appreciating the natural envi- tentially useful headline indicator at the aggregate ronment—breathing fresh air, drinking clean water, liv- level, indicators are most useful when they can be ing among an abundance of plant and animal varieties, disaggregated and used to diagnose and ultimately and not irrevocably undermining the natural processes address specific problems. that produce and renew these features. Indeed, peoples’ Ⅲ Why the need to manage a broader portfolio of as- self-reported happiness and satisfaction with life are sets? What choices can and must be made between closely associated with all of these factors.2 creating, maintaining, and restoring different as- Society’s ability to enhance human well-being sets as part of a long-term, dynamic view of sus- through time depends on choices made by individu- tainability? Although assets are complementary als, firms, communities, and governments on how to and substitutable to a certain degree, they all need use and transform their assets. They might cut down to be managed, since once the quality or level of an forests to build dams and other physical infrastruc- asset falls below a threshold, there can be little fur- ture or to make way for commercial agriculture or ther substitution without jeopardizing the produc- urban expansion. They might clear mangroves to tivity of other assets, as well as overall production. build shrimp farms. Or they may conserve forests and Ⅲ What are alternative development paths to those mangroves to maintain important natural processes followed by developed countries? What tradeoffs or to support tourism. Enhancing human well-being and priorities are justified, and when? By taking  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  advantage of technological innovations and by Recent definitions have focused more explicitly learning from past mistakes of others, countries on the three pillars of sustainability: economic, en- today have the option to manage their portfolio of vironmental, and social. These highlight the need to assets in a different way to ensure they are on a consider not only the environmental, or even the more sustainable development path in the long environmental and economic aspects, but also the term. social aspects of sustainability. The thinking about Ⅲ How to address the almost endemic overuse or social sustainability is not yet as advanced as for the underprovision of environmental and social assets other two pillars. Societies do, and will continue while sustaining growth? Wherever spillovers (ex- to, transform over time. But it seems clear that sig- ternalities) exist, there is a coordination problem nificant social stress—and, at the extreme, social that needs to be dealt with by correcting market conflict—is likely to lead to a breakdown in the ac- and policy failures. This can be done by using a cumulation or preservation of all assets, thereby jeop- variety of mechanisms such as command-and- ardizing intergenerational well-being. control regulations, harnessing market forces, and One concrete approach to thinking about sustain- improving supporting institutions. ability and intergenerational well-being is to ensure that the flow of consumption does not decline over Sustainability—an evolving framework time. But what is needed for this? The academic lit- What is meant by sustainability? erature shows that a country’s ability to sustain a For any given technology, preference structure, and flow of consumption (and utility) depends on the known resource base there are some utilization rates change in its stock of assets or wealth. Intergenera- that cannot be sustained. Drawing attention to these tional well-being will rise only if wealth (measured unsustainable rates is critical to informing decision- in shadow prices and excluding capital gains) in- makers and changing course toward sustainability. creases over time—that is, only if a country’s ad- This will often require altering the pattern of prefer- justed net savings are positive.3,4 (See the section ti- ences, the resource intensity of technologies, or the tled “Measuring sustainability.�) relevant time horizon for different decisions. Since none of these is constant or stable over time, defining Not a steady-state concept sustainability in a broader sense is not easy—but there Does the composition of the asset base matter? In have been many attempts. The most commonly used principle, this depends on the potential for sub- definition is the one provided by the World Commis- stitutability among assets (see the section titled “The sion on Environment and Development (Brundtland importance of a range of assets� ). In the environmen- Commission 1987): “progress that meets the needs of tal economics literature (Pearce and others 1989) a the present without compromising the ability of fu- distinction is made between weak constraints on ture generations to meet their own needs.� growth, known as “weak sustainability� (which pre- While the Brundtland definition highlights the sumes that assets are fully substitutable) and strong need to balance the interests of current and future constraints on growth, known as “strong sustainabil- generations, it does not define the concept of needs ity� (which holds that assets are not fully substitutable or its implications. For instance, does the Brundt- because some natural assets, or more precisely some land definition imply that well-being (utility) should of the functions performed by these assets—such as not fall below some minimum for any subsequent global life support—cannot be replaced by others). generation? Does it imply that each generation should Limits-to-growth type arguments focus on strong sus- enjoy a constant level of well-being? Alternatively, tainability, while arguments in favor of indefinite should well-being be nondeclining for each future growth focus on weak sustainability. So far the former generation? Most later definitions have retained the arguments have not been very convincing because the core ethic of intergenerational equity, emphasizing substitutability among assets has been high for most the current generation’s moral obligation to ensure inputs used in production at a small scale. There is that future generations enjoy at least as good a qual- now, however, a growing recognition that different ity of life as the current generation has now (Pezzey thresholds apply at different scales—local to global. 1989). Technology can be expected to continue to increase (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Box 2.1 actions on complex ecosystems: small changes can Not yet able to fully duplicate natural processes sometimes accumulate and translate into losses of whole ecosystems (see box 2.5). There is also uncer- Biosphere 2—a sealed glass ecosystem that was built in tainty about what technological innovations will be Oracle, Arizona, at a cost of some $200 million in 1991— attempted to create a completely self-contained, human- available and when. Where the costs of human ac- made system to support eight people for two years. It tions today are uncertain, with potential for large could not. and irreversible damage, there is a need for proceed- There is still debate on how to conduct such an exper- ing with greater caution in maintaining environmen- iment. The idea was that there would be no exchange with tal and social assets. the outside world except for the energy supplied to run ap- pliances. The people inside the biosphere would grow all Measuring sustainability their own food. And the system would operate with a fixed volume of air and water, recycled and reused as they There are many important things that are not mea- are on Earth, the original biosphere. surable, but in general, people value what they mea- A year and a half after the sphere was sealed, the oxy- gen content of the atmosphere had fallen from 21 percent sure. One of the biggest challenges is how to measure to 14 percent, a level normally found at 17,500 feet and all our assets and our progress toward sustainable de- barely sufficient to keep people in the biosphere function- velopment. Since the Brundtland Commission, there ing. Carbon dioxide (CO2) and nitrous oxide levels surged. have been many efforts to develop indicators of sus- All pollinators became extinct, so agricultural production could not be sustained. Worse still, the drop in oxygen and tainability. Much of the progress in developing in- rise in CO2 meant that the biosphere’s systems could not dicators for measuring sustainability has been in the replicate the carbon cycle, the most essential cycle for life. economic and environmental sphere (box 2.2). Social indexes, such as transparency, trust, and conflict are Source: Heal (2000). still at early stages of development. The fact that so- cial indicators are less developed reflects the ongoing the potential substitutability among assets over time, debate about the concept of social sustainability: what but for many essential environmental services—espe- it means and what should be measured. cially global life support systems—there are no known alternatives now, and potential technological Green accounting solutions cannot be taken for granted (box 2.1). Early efforts to link economic and environmental ac- The limits to substitutability among assets are counting focused on the measurement of “a green likely to be greater for those assets that enter con- GDP,� motivated by the genuine concern that the sumption untransformed (for example, natural forest traditional measure of gross domestic product (GDP) scenery versus natural desert scenery) rather than as provides only a partial picture of changes in wel- a produced output using the same materials (for ex- fare—capturing mainly, if not exclusively, elements ample, a wooden window shutter or a glass pane). transacted in markets (only a few imputed services, Ensuring that the well-being of future generations such as owner-occupied housing, are included). does not decline requires maintaining sufficient lev- Many environmental assets—especially those that els of some assets for the future, particularly when the function as “sinks� receiving pollution and waste, and drawdown or degradation entails irreversible losses those supporting life—do not operate in markets and and there is a possibility that these assets matter di- are therefore excluded. rectly for the well-being of future generations. Of These early environmental accounting efforts course, the mix of assets that supports improvements tried to modify national accounts to include envi- in human well-being is likely to change over time, as ronmental damages, environmental services, and people’s preferences and technologies change. So the changes in stocks of natural capital. But that proved concept of sustainability will itself evolve over time. problematic mainly because of valuation difficulties and some conceptual issues. For example, should ex- Proceeding with caution penditure for environmental protection be treated as What is more important for sustainability is how to intermediate or final consumption? manage risks by retaining options. There is consid- Later efforts have been directed toward construct- erable uncertainty about the consequences of human ing “satellite accounts� that try to link environmental (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 2.2 Indicators for measuring sustainability—a subset Some of the main approaches to developing indicators of en- Environmental Sustainability Index. World Economic vironmental sustainability are the following: Forum. An aggregate index spanning 22 major factors that contribute to environmental sustainability. • Extended national accounts Green Accounts System of Environmental and Economic • Unequally weighted indexes* Accounts. United Nations. A framework for environmen- Environmental Pressure Indexes. Netherlands, EU. A set tal accounting. of aggregate indexes for specific environmental pressures Adjusted Net Savings. World Bank. Change in total wealth, such as acidification or emissions of greenhouse gases. accounting for resource depletion and environmental Well-being of Nations. Prescott-Allen. A set of indexes damage. that capture elements of human well-being and ecosys- Genuine Progress Indicator, Redefining Progress, and tem well-being and combines them to construct barome- Index of Sustainable Economic Welfare. United Kingdom ters of sustainability. and other countries. An adjusted GDP figure, reflecting • Eco-efficiency welfare losses from environmental and social factors. Resource Flows. World Resources Institute. Total mater- • Biophysical accounts ial flows underpinning economic processes. Ecological Footprint, Redefining Progress. World Wildlife Fund and others. A measure of the productive land and • Indicator sets sea area required to produce food and fiber, and in renew- U.N. Commission for Sustainable Development and many able form, the energy consumed by different lifestyles countries. within and among countries. • Equally weighted indexes* * Equally weighted indexes are those whose components are equally Living Planet Index. World Wildlife Fund. An assessment weighted and then aggregated, while unequally weighted indexes give of the populations of animal species in forests, fresh some components greater weight than others. water, and marine environments. Source: Authors. datasets with (unmodified) national accounts in- over time—that is, only if adjusted net savings is pos- formation. In principle, environmental costs and itive—will intergenerational well-being rise. benefits, natural resource assets, and environmental Ideally, measures of adjusted net savings would protection are all presented in flow accounts and bal- take into account human capital, natural assets, ance sheets. But in practice, given the difficulty in knowledge, and social assets.5 But measurement dif- valuation, the emphasis has often been on using in- ficulties and the lack of available data preclude this. formation on physical quantities from environmen- Estimates of net savings currently account for some tal accounts. The drawback of this approach is the key elements of environmental stocks—energy de- difficulty in making comparisons across accounts in pletion, mineral depletion, net forest depletion, and different units to evaluate priorities or tradeoffs. CO2 emissions.6 They also include education spend- ing, as a proxy for human asset accumulation, but Adjusted net savings they do not yet include changes in the stock of (cod- The focus of more recent efforts to link economic ified) knowledge or social assets (see table 2.1).7 It is and environmental concerns has been on determin- clear that adjusted net savings is an improvement ing changes in wealth (adjusted net savings) as an over traditional savings measures; however, efforts to indicator of sustainability. Change in wealth, appro- refine it further will need to continue. priately defined to include a comprehensive and In practice, also, additional adjustments may need complete set of assets, is a good measure of prospects to be made to deal with specific issues. First, when a for well-being as it indicates a country’s ability to sus- country’s population is growing, it is on a sustain- tain a consumption stream—which is what matters able path on a per capita basis only if the percentage for sustainability—not just the consumption flow at change in wealth (adjusted net savings as a share of a particular time as measured in GDP or green equiv- total wealth) exceeds the population growth rate.8 If alent. In principle, only if wealth (measured in the change in wealth is lower than the population shadow prices and excluding capital gains) increases growth rate, the country is “de-capitalizing� or run- (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Table 2.1 Toward adjusted net savings, 1999 (percentage of GDP) ( )( ) Gross Consumption Carbon Income domestic – of fixed Energy Mineral Net forest dioxide + Education = Adjusted and region savings capital depletion depletion depletion damage expenditure net savings By income Low income 20.3 8.3 3.8 0.3 1.5 1.4 2.9 7.8 Middle income 26.1 9.6 4.2 0.3 0.1 1.1 3.5 14.3 Low and middle income 25.2 9.4 4.1 0.3 0.4 1.2 3.4 13.3 High income 22.7 13.1 0.5 0.0 0.0 0.3 4.8 13.5 By region East Asia and Pacific 36.1 9.0 1.3 0.2 0.4 1.7 1.7 25.2 Europe and Central Asia 24.6 9.1 6.0 0.0 0.0 1.7 4.1 11.9 Latin America and the Caribbean 19.2 10.0 2.8 0.4 0.0 0.4 4.1 9.6 Middle East and North Africa 24.2 9.3 19.7a 0.1 0.0 1.1 4.7 –1.3 South Asia 18.3 8.8 1.0 0.2 1.8 1.3 3.1 8.3 Sub-Saharan Africa 15.3 9.3 4.2 0.6 1.1 0.9 4.7 3.9 Note: Adjusted net savings are equal to net domestic savings (calculated as the difference between gross domestic savings and consumption of fixed capital), plus education expenditure, minus energy depletion, mineral depletion, net forest depletion, and carbon dioxide damage. a Note that the energy depletion figure in the table is stated in terms of GDP. This translates to an annual depletion rate of about 1 percent of proven reserves. Source: World Bank (2001h); for details on the methodology, see Hamilton (2000). ning down its assets on a per capita basis. This would interest rates, and which can be influenced by econ- imply that it is on an unsustainable path to an even- omywide policies—there are no policy-relevant ag- tual decline in welfare per capita. Second, if produc- gregate indexes on the state of the environment. For tion processes are subject to thresholds (nonconstant policy purposes, these indexes need to be disaggre- returns to scale), then again an adjustment to net gated (as in table 2.1) and complemented by such savings needs to be made, if measured net savings are biophysical measures as pressure-response indicators. to correctly indicate sustainability. Not only can the latter be disaggregated to a much The adjusted net savings measure is a useful “head- greater extent, but they also have the added advan- line� indicator for the economy. Like all national ac- tage that they can be used to identify the source of counts or monetary-based indicators, it employs an the problem. integrating framework that permits weighting and ag- While recognizing the need for an aggregate index gregating disparate elements of the economy and the as a headline indicator, it is important to note that environment. In principle an aggregate indicator such indicators are most useful when they address specific as adjusted net savings allows for comparisons across problems. To catalyze change, information and sig- groups of countries—by region or by income. Figure nals have to be picked up by specific groups or insti- 2.1 presents a comparison by GDP per capita, and tutions that can use them to diagnose specific prob- shows that adjusted net savings are negative in some lems, rally support for change, balance interests, and countries—that is, they are de-capitalizing. take action.9 A good example of this process is Silent Spring, A system of indicators the book Rachel Carson wrote in 1962 to alert the As just mentioned, the adjusted net savings indica- public that birds were disappearing or being si- tor is a potentially useful headline indicator at the lenced. She pointed to indicators that no govern- aggregate level. But unlike GDP—which is affected ment agency would have considered important in by economywide prices, such as exchange rates and advance—DDT levels in falcons and the fragility of (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Figure 2.1 tions to accommodate the unusable outputs of Adjusted net savings rates by per capita GDP production and consumption—air, water, and soil level, 1999 receiving pollution and waste generated by human activities.11 More fundamentally, nature performs Adjusted net savings, percentage of GDP 50 critical life-support services on which the well- being of all life depends. So far—despite all the 40 technological advances—no way has been found 30 to fully replace these services through human- 20 made alternatives (box 2.1). 10 Ⅲ Human-made assets—created physical products, particularly those used in production, such as ma- 0 chinery, equipment, buildings, and physical net- –10 works, as well as financial assets. –20 Ⅲ Knowledge assets—“codified knowledge,� which –30 is easily transferable across space and time (unlike “tacit� knowledge, which entails an individual’s 100 1,000 10,000 100,000 experience and learned judgment and thus cannot Log of GDP per capita be easily transferred until codified). Source: World Bank (2001h). Ⅲ Social (or relational) assets—interpersonal trust12 and networks,13 plus the understanding and shared values that these give rise to—which facilitate their eggs. This gave birdwatchers in America a new cooperation within or among groups.14 role and put environmental protection agencies on a track to monitor toxins in nature, industry, and else- The importance of managing human, physical, where that might affect human well-being as well. and financial assets is well known, but how they in- Policy-relevant indicators emerge and are contin- teract with other assets is less well developed. Social, ually validated and refined in an environment where and environmental assets enhance human well-being there is a free flow of information and interaction. directly through their very existence (e.g., the ability To avoid major regrets, there is a need for more cred- to trust another person or enjoy a natural setting).15 ible information and networks that link experts, civil They also enhance human well-being indirectly society, and decisionmakers. through their contribution to production and mate- rial well-being (figure 2.2). A tropical forest provides The importance of a range of assets cut lumber as an input into the production of furni- Action to improve asset management need not await ture and houses. A standing forest’s environmental resolution of debates on how to define and measure services—such as flood control and storm protec- sustainability, but does require a clear understanding tion—can also improve the production of crops. And of what assets matter and why. The capacity of any a forest’s complex ecological functions support life society to meet the “requirements� of individual well- for many species—that are important for the func- being depends on the level and quality of a range of tioning and survival of the forest, which provides hu- assets—and on how society deploys them. Broadly, mans with material and aesthetic pleasures. these assets consist of the following:10 Why the need to manage a broader portfolio of assets? Ⅲ Human assets—the innate skills, talents, compe- tencies, and abilities of individuals, as well as the The complementarity of assets effects of education and health. In improving human well-being, assets generally Ⅲ Natural assets—both renewable and nonrenew- complement each other. For instance, human assets able. These assets have source functions that enter together with social assets can enhance a person’s as inputs into production and utility—forests, “freedom to be and to do.� Assets can also be com- fisheries, mineral ores, and natural forces (such as plements in the production process—that is, the air and water currents). They also have sink func- productivity of one type of asset usually rises with (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Figure 2.2 Ⅲ Social capital can improve the productivity of How society’s assets enhance human well-being physical capital. For example, interfirm social con- tact in the form of interpersonal networks in the ᮢ Well-being clothing industry has a positive impact on learn- ᮢ ing.22 Similarly, interfirm social interactions, as ᮢ ᮢ Amenities and ecological well as customer network ties, have significant ef- functions fects through their impact on knowledge acquisi- Networks, shared values, and trust tion and on new product development in a range of high-tech industries, including pharmaceuti- Consumption of cals, medical devices, and electronic instrumenta- Natural assets material goods Social assets tion.23,24 There is also a clear link between inter- Direct inputs Networks and trust and ecological ᮢ firm trust and firm performance (through conflict functions avoidance and lower negotiation costs).25 ᮢ ᮢ ᮢ Output ᮢ ᮢ This complementarity generally applies to other Buildings, physical networks, Tacit knowledge, and financial assets skills, assets as well. Cleaner air and water, for instance, im- competencies, and prove human health and the productivity of human abilities capital.26 And the synergies from the complementar- Human-made Codified knowledge Human ity of two or more assets raises overall productivity. ᮢ ᮢ assets assets But social and environmental assets are underpro- “Direct� channel “Indirect� channel vided or overused. Source: Authors. Assets and diminishing returns Most assets are also subject to diminishing marginal additional amounts of other assets.16 For instance, returns. The benefits to well-being or to productiv- there is now a growing body of literature that high- ity of an additional unit of an asset declines as the lights the role of social capital (interpersonal net- level or quality of the asset rises (all other assets kept works, shared values, and trust)—an asset that has constant). Why? As J.B. Clark said, “Put one man arguably received little attention in the economics only in a square mile of prairie, and he will get a rich literature so far—in the accumulation, preservation, return. Two laborers on the same ground will get less and productivity of other assets—human-made, en- per man; and if you enlarge the force to ten, the last vironmental, and human:17,18 man will perhaps get wages only.�27 As more people Ⅲ Social capital can improve the management and are added, the returns continue to drop, until some- productivity of environmental assets. For example, one is unable to cover his or her cost. the combined effect of attitudes about participa- Only if there are very strong positive spillovers as- tion—and the actual participation in a collective sociated with an asset is the tendency of diminish- enterprise, along with human capital (literacy)— ing marginal returns offset. That is true for knowl- has significantly improved the management of wa- edge, particularly codified knowledge. Because new tersheds in Rajasthan, India.19 Watershed manage- knowledge complements existing knowledge (there ment has, in turn, been crucial in raising incomes. is no crowding out, as with the laborers), it is more Trust between technicians (agricultural extension valuable the more society already knows. Similarly, agents) and farmers can generate increases in agri- it is true for networks, such as telephones, where the cultural production. And the degree and nature of advantages of owning a telephone increase with each trust between the contact groups and the other new member in the network. members of the community can determine the ef- fectiveness of the groups as catalysts for commu- Limits to substitutability among assets nity development.20 Because assets generally complement each other and Ⅲ Social capital can enhance human capital accumu- because the returns to a particular asset diminish, the lation: higher levels of trust have been associated rate at which one asset can be substituted for another with higher enrollment in secondary education.21 in production (while maintaining a given unit of (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  output) tends to diminish as well. As the level of one The consequences of ignoring the complementarity asset declines relative to another, the rate at which it of environmental assets and breaching thresholds can be further replaced falls. Moreover, when the As an illustration of what can happen when the com- quality or level of an asset falls below a threshold, plementary role of environmental assets is ignored there can be little further substitution without jeop- and certain thresholds are breached, consider the ardizing the productivity of the other assets, as well Yangtze Valley in 1998. Although China has always as overall production. been susceptible to flood and drought, the 1998 When environmental or natural assets are fairly floods were some of the most severe in its history. abundant relative to human-made assets, substitu- Rainfall from June to August that year was 38 per- tion of the former by the latter can be expected to cent above normal, but later analysis found that lead to higher returns. But there are limits to a long- these unusually high levels could only partly explain term strategy that focuses primarily on replacing nat- the floods. The rest was perceived to be due to log- ural assets by human-made assets. Severely degraded ging of the river’s watershed, which eroded the soil. farmland or fisheries will yield little wheat or fish, Deforestation had been so great (forest cover had de- no matter how many plows or boats are used. creased by more than half since the 1950s) that the Development strategy to date has often relied on watershed could no longer stabilize the water flow.33 drawing down environmental resources and replacing The resulting floods had very high costs in human them with human-made assets. This was the strategy lives—tens of thousands dead—and in lost produc- followed by today’s industrial countries.28 Most devel- tion in the area. oping countries’ growth strategies continue to focus Similarly, the degradation of the Aral Sea high- largely on the accumulation of human-made assets lights what can go wrong when there is inadequate (physical capital). Indeed, a review of 60 countries in recognition of the role of environmental assets in the the late 1980s and 1990s shows that the growth of 16 production process—and of the costs for human countries considered to be serious policy reformers was welfare (box 2.3). Expansion of irrigation schemes accompanied primarily by physical capital accumula- in the Aral Sea basin has generated billions of U.S. tion. The increase in per capita GDP growth of this dollars in benefits and millions of jobs. But the over- group of countries—rising from 2.8 percent in the late all costs of these schemes have been high, both in 1980s to 3.5 percent in the 1990s—entailed an in- failing to generate the expected high levels of sus- crease in the rate of physical capital accumulation from tainable production over time and in causing serious 2.1 percent to 3.5 percent. In contrast, spending on health effects in areas immediately surrounding the education—a proxy for human capital accumula- sea. Today, avoiding further declines in the sea level is tion—rose only slightly, from 3.2 percent of GDP in possible only if appropriate operational adjustments the late 1980s to 3.5 percent of GDP in the 1990s. And are made to the existing irrigation systems to improve the rate of deforestation—a proxy for the depletion of their efficiency. natural assets—rose from 0.7 percent to 1.1 percent.29 Some countries’ experience with shrimp farming The limits to focusing on physical capital alone illustrates the costs of ignoring environmental ser- are borne out empirically. An econometric study of vices. Over the past two decades new technologies 70 developing countries found that countries with and production systems have enabled a dramatic in- low physical capital–labor ratios tend to experience a crease in the intensity of shrimp farm operations— rise in their growth rates with increases in the stock the production of farmed shrimp has grown at of physical capital. But after countries reach a certain 20–30 percent a year.34 Compared with traditional capital intensity, the contributions of further physi- systems, however, the more intensive systems require cal capital accumulation to growth—for any given large amounts of feed to support the shrimp and human and natural capital—decline.30 A separate large amounts of water to flush out the wastes.35 Be- study of 20 middle-income countries also found the cause of the high concentration of farm units in areas marginal productivity of physical capital to dimin- of limited water supplies and inadequate flushing, ish.31 So although there may be economies of scale the effluents in many cases exceeded the capacity of and technological spillovers for physical capital,32 the receiving waters (sink), leading to pollution in- these do not seem to be large enough to continually side the ponds as well, which adversely affected pro- offset diminishing marginal productivity. duction since these farms require a lot of water as an (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Box 2.3 The Aral Sea—the cost of ignoring the role of an environmental asset The Aral Sea watershed now spans the national borders of six Shrinking sea, falling yields countries. Over the past 40 years the excessive water diver- sion for irrigation along the Amu Darya and Syr Darya Rivers— Cotton yields in Uzbekistan, 1961–2000 the two main tributaries of the Aral Sea—caused the volume Centners per hectare of the sea to fall by 85 percent, and the sea level by 18 meters, 35 exposing more than 40,000 square kilometers of saline seabed and heavily salinating the remaining water (box figure). Today 30 the Aral is divided into a smaller, less saline sea in the north 25 and a larger, saline sea in the south. 20 ARAL SEA Depth Level 15 0 m (1956) BATIMETRICAL MAP - 5 m (1976) 10 -10 m -15 m (1996) 5 -20 m 1956 - 25 m 0 -30 m 61 65 69 73 77 81 85 89 93 97 00 19 19 19 19 19 19 19 19 19 19 20 -35 m -40 m Year -45 m -50 m Note: 1 centner is equal to 0.1 ton. -55 m 1976 -60 m seabed contaminate lands adjacent to the sea, and increased chemical and pesticide use upstream pollutes drinking water. The people hardest hit live in Karakalpakstan, at the end of the Amu Darya Delta. Reliable data on health costs are hard to ob- KM 1996 0 25 50 75 100 tain. But by some estimates maternal deaths in Karakalpakstan in 1994 were 120 per 100,000 live births (twice the national av- erage) and infant mortality was 60 per 1,000 live births (three Loss of fisheries times the national average). In the past 10–15 years kidney and Although Soviet planners realized that greater irrigation would liver diseases, especially cancers, have increased 30- to 40-fold, lower the sea’s water level, it was thought that the increment arthritic diseases 60-fold, and chronic bronchitis 30-fold. in agricultural output of the whole basin would yield significantly higher benefits than any damage caused. Not recognized, how- Full restoration too costly—how to avoid further decline ever, was that the excessive water withdrawal would make the It may be too late to fully reverse the damage, but it is possible remaining sea water so much saltier that it would become unfit to stabilize agricultural production in the basin and mitigate neg- for higher forms of aquatic life. The once fairly substantial fish- ative downstream effects. Restoring the sea to its former level ing industry has now almost completely disappeared. in the next 50 years would mean suspending all irrigation and other water uses in the basin—impossible today, when even A drop in agricultural output water reductions of 3–5 percent meet with strong local opposi- At the same time the combination of excessive irrigation and tion from people highly dependent on irrigation. Although the poor management of the irrigated land has led to water- rates of return on the incremental irrigation have not been very logging and increased the salinity of the soil in the entire basin. high—ranging from 13 percent in the best case scenario (high Almost one-third of the irrigated land is now degraded. Effec- cotton prices and low irrigation costs) to minus 10 percent in tive management in these areas, with an emphasis on environ- the worst case (low cotton prices and high irrigation costs)*— mental assets, could have helped avert the current problems better returns can be achieved and agricultural production in and the environmental degradation surrounding the sea. the region can be put on a more sustainable path. Estimates It is no longer possible to maintain irrigation and cotton pro- put potential efficiency gains through operational improve- duction at levels experienced during the Soviet period. The land ments and greater participation and collective action in the use degradation, combined with the reduced availability of appro- of irrigation water at 20–30 percent—this, at relatively low fi- priate agricultural inputs for production after the breakup of the nancial cost and without constraining production (World Bank Soviet Union, has greatly reduced cotton production, both total 1998a). With these improvements, the decline in sea level yields and productivity per hectare. The original conversion of could be arrested and some aquatic life could be reintroduced. 7.9 million hectares of desert allowed a rise in Soviet cotton production from 2.2 million tons in 1940 to 9.1 million tons (at * The estimates of the return-to-irrigation investment are based on its peak) in 1980. Cotton production in Uzbekistan, which ac- Uzbekistan, which accounts for more than 70 percent of cotton pro- counted for 70 percent of the total production (4 million tons) duction in the region. The rate of return calculations are very sensi- in 1960, peaked at 5.5 million tons in 1980. By 2000, it was tive to the average raw cotton price and the full cost of irrigation. down to 1960 levels—which may be more optimal and sustain- Cotton prices fluctuated widely over the 1960–2000 period. The av- able—when large-scale irrigation was beginning (box figure). erage cost of cotton is assumed to be $1,200 a ton (2000 prices) for the high-cost case and $850 a ton for the low-cost case. The aver- An increase in health costs age cost of irrigation is assumed to be $500 a hectare for the low The exposed seabed and polluted downstream waters have also case and $300 for the high case. had high human and health costs. Winds carrying salt from the Source: Authors. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  input (source).36 The quality of the water in tradi- capital to GDP ratio is below 15 percent. Even when tional shrimp farms is generally better because of the the ratio is above 15 percent, the probability falls by lower intensity of shrimp, which are thus less prone more than 23 percentage points if the level of social to disease. assets—proxied by an index of (the lack of ) political The collapse of many shrimp farms in China, In- and social tensions—falls below a threshold.44 The donesia, Taiwan (China), and Thailand has meant probability of such durable growth also falls signifi- large losses in physical assets and in labor.37,38 This cantly (from 70 percent to 44 percent) if the educa- was a direct consequence of not recognizing the im- tion Gini—measuring inequality in the distribution portance of ensuring good naturally provided water of education—is greater than 0.30. quality in the production process, especially as the In sum, the long-term neglect of any set of as- volume of shrimp and the capital intensity of farms sets—human, social, or environmental—can at some increased. point sharply reduce the productivity of the other as- Breaching thresholds through the cumulative loss sets, whether for commodities, sectors, regions, or of biodiversity can also lead at a localized level to a nations.45 Therefore, while countries may be able to loss of resilience of an ecosystem—in its capacity to grow for a period based on a strategy of accumulat- absorb disturbances without undergoing fundamental ing physical capital, the prolonged neglect of other changes in functional characteristics. A run-down assets is likely to endanger the durability and sustain- ecosystem, (one degraded by excessive use) can suc- ability of the growth process—for example, allowing cumb to shocks that would not destroy a healthy a country to fall into a state of high social and civil ecosystem. A famous analogy made by Ehrlich and unrest (a drop in social capital) is likely to undermine Ehrlich (1981) relates ecosystem components to rivets sustained economic growth.46,47 Similarly, if envi- in an airplane.39 One by one, biological species may ronmental degradation is irreversible, society can lose disappear and not be missed. Eventually, however, the the option value of an asset that could make a seri- cumulative loss of biodiversity will lead to the crash of ous difference to future productivity (box 2.4). ecosystem functions just as the cumulative loss of re- So far the concern has been with the potential for dundant rivets will lead to the crash of an airplane.40 substituting assets in production. What about the Thresholds are clearest when a renewable asset has potential for substitutions that affect human well- been exploited beyond its capacity to regenerate or being directly? The need to manage all of society’s reproduce. When that threshold is reached, the pro- assets may be even greater. The substitutability of ductivity of other assets decreases—or if the de- assets that enter people’s well-being directly is likely graded asset is the main input, production may cease to be lower than the substitutability technically fea- altogether. The change is often sudden and discon- sible in production. Some minimum bundle of so- tinuous, as in cod fisheries in New England (see cial and environmental assets is likely to be needed chapter 7).41 if one is to achieve a given level of personal well- In some cases there may be no substitute for some being.48 This argument is just as valid for intergen- of the functions of the environmental asset, so breach- erational well-being. ing thresholds can cause irreversible damage. An ex- There will always be much uncertainty about the ample of this is the ozone level: wearing a sunscreen tastes and preferences of future generations—and lotion all day may protect skin from cancer caused by about the technological possibilities open to them. ultraviolet rays, but there is no known substitute for But there is also much uncertainty about the conse- the protection ozone affords to our food chain.42 quences of our current actions. While many ecologi- Thresholds can apply to all assets. Indeed, the ex- cal problems are gradual, some can switch abruptly perience of 80 countries during 1970–99 suggests from one stable state to another (box 2.5). Such that the probability of achieving a relatively high per shifts can cause large losses of ecological and eco- capita growth of 2.5 percent a year for a five-year pe- nomic resources. riod is highly affected by the crossing of certain min- Very often, restoring the desired state would re- imum thresholds of physical assets, human assets, quire drastic and expensive interventions. And some- and social assets.43 That probability drops from 58 times the process of restoration is not even known. percent to 28 percent if the investment of physical Technological solutions to these problems might be (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Box 2.4 Box 2.5 How keeping the option value of assets can Catastrophic ecoshifts make a serious difference Recent studies highlight the possibility of catastrophic In 1970 a new virus—the grassy stunt virus, carried by the shifts in ecosystems. Usually the changes in outside con- brown plant hopper—threatened rice production in Asia. ditions affecting ecosystems—climate, injection of nutri- The virus appeared capable of destroying as much as one- ents or toxic chemicals, groundwater reductions, habitat quarter of the crop in some years, making it critical to de- fragmentations, losses of species diversity—occur very velop a rice strain resistant to the virus. This was done gradually. And sometimes the ecosystems will respond to with the help of the International Rice Research Institute such changes smoothly and continuously. But studies of (IRRI), which researches rice production and maintains a lakes, oceans, coral reefs, forests, and arid lands show huge bank of rice seeds—about 80,000 varieties of rice that these smooth changes can be interrupted by sudden, and near-relatives of rice. In this instance, a single strain drastic switches to another state. The gradual changes of wild rice not used commercially was found to be resis- in external conditions can lead to a loss of resilience and tant to the grassy stunt virus. The appropriate gene was make the ecosystem more vulnerable to catastrophic transferred to commercial rice varieties, yielding commer- shifts. Once a threshold is passed, the shift can occur sud- cial rice crops that were resistant to the virus. denly, with little warning. So under some conditions the Note that this strain was found in only one location, a ecosystem can move from one stable state to another, valley flooded by a hydroelectric dam shortly after the IRRI separated by an unstable state. took the strain into its collection. Without this strain— Coral reef ecosystems can exhibit such dramatic shifts— which apparently had no commercial value—the well-being from having high biodiversity to being overgrown with fleshy of hundreds of millions of people would have been seri- algae. Factors that make them vulnerable to such shifts in- ously affected. clude increased nutrient loading from changed land use and overfishing, and reduction of the number of large, and later Source: Heal (2000). the smaller, herbivorous fish species that control the algae. In the Caribbean, overfishing had already reduced herbivo- rous fish when a pathogen reduced the population of sea available in the future—or they might not be. When urchin Diadema (which also controls the algae). As a result, the potential damage can be very large—where the the reefs became overgrown with fleshy brown macro effects may be irreversible and where substitution algae—the spread is now difficult to reverse because adult algae are less palatable to herbivores and the persistence of possibilities may be limited—a “precautionary prin- the former prevent the settlement of coral larvae. ciple� applies: act more conservatively when you are uncertain about the effects (see chapter 5 and box Source: Scheffer and others (2001). 5.6 on the precautionary principle). Tradeoffs and sustainable development from the perspective of people’s well-being than of Balancing objectives and choosing how to act production. So to ensure that the well-being of fu- Improving human well-being over time is a broader ture generations is not compromised, some attention goal than increasing economic growth that focuses has to focus on environmental concerns—in partic- primarily on material comfort. This has some impor- ular the avoidance of irreversibilities that may mat- tant implications. Since social and environmental as- ter for future well-being. sets also affect human well-being directly, a strict The way the economy grows—the pace and pat- policy of “grow now, clean up later� has costs for tern of growth—can matter for the well-being of today’s generation, costs that often fall dispropor- both the current generation and that generation’s tionately on today’s poor.49 children and grandchildren. Developing countries do Moreover, any serious attempt at poverty re- not have to follow the path of development traversed duction requires, at a minimum, durable economic in the last century by the industrial countries. Tech- growth—not economic growth in fits and starts. This nological options have improved and it is now possi- means paying enough attention to social and envi- ble to avoid repeating the mistakes of industrial ronmental concerns to ensure that durable growth is countries in their development (i.e., the use of lead not jeopardized. in gasoline). On the other hand, some options open And while there is potential for substituting assets to industrial countries in their development phase over a range, there are limits to such substitution (see are not open to developing countries now (land- earlier section on this topic), perhaps even more labor ratios, extent of global competition, and so on). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  What do these considerations imply for a coun- tion and low soil fertility. The country has already try’s development strategy—or how does a country lost 80 percent of its original forest cover, more than balance the objectives of addressing environmental half in the past 40 years (see box 8.3 in chapter 8). concerns and pursuing economic growth? Over the In the east of the country, under the tavy agricultural longer term, economic growth is unlikely to be sus- system, rice is grown on steep slopes after slashing tained unless enough attention is paid to environ- and burning of virgin or secondary forests. In the mental assets. But over the short to medium term it central highlands population pressure forces people may be possible to do so, on the grounds that such from the valley bottoms to farm the hillsides, evident short-term growth could generate more resources for in the big increase in rainfed agriculture. The result- addressing environmental concerns later. Indeed, ing erosion causes nutrients to wash off the already having limited resources usually makes it necessary poor soil and to silt irrigation schemes in the valleys. to choose priorities between tradeoffs. But the pri- The annual cost of environmental degradation— orities will not always favor growth over attention to from soil erosion, silting, declining soil fertility, and environmental assets in the short run, or vice versa. lost forest—is high, estimated at over 5 percent of The appropriate ranking of priorities will vary by GDP, and the agricultural resource base has not kept locale (region or nation) and at different times, de- up with population growth. That is why arresting this pending on the issue and on several other factors. cycle—through agricultural intensification to reduce What environmental depletion or degradation has the pressure of cultivating new uplands—is para- already taken place? How important is the asset in mount. Today, little use is made of fertilizers and of either the production process or in utility directly? new higher yielding varieties—for several reasons. Are the poor particularly vulnerable if the issue is left The absence of secure land tenure reduces the incen- unattended? tives for investing in intensification. The lack of credit Three broad cases can be distinguished for differ- and liquidity hampers the use of inputs. And the very ent emphasis and sequencing: poor quality of rural infrastructure constrains the sup- ply of inputs and makes it more expensive. 1. Simultaneously addressing environmental con- Indeed, for countries that rely heavily on renew- cerns along with economic growth, even in the able natural resources and have few alternatives in the short run short to medium term (because they are poor in 2. Placing a higher priority on economic growth, human and human-made assets), it is especially im- while addressing environmental concerns that portant to contain environmental depletion or degra- can be dealt with at relatively low cost in the dation. For these countries, maintaining natural as- short run sets is a critical component of economic growth. For 3. Placing a higher priority on maintaining or re- example, in southern Africa, the Caribbean, and the storing the environment in the short run. Indian and Pacific Oceans, nature-based tourism has become an important source of foreign exchange and Case 1. Win-win: preserve natural assets local income. and keep growing In some cases restoring or maintaining an envi- Addressing both growth objective and preservation ronmental asset may not be critical for economic or restoration of environmental assets can sometimes production (other factors of production could re- be critical to raising production and incomes, even place its functions), but it may be more economi- in the short to medium term. That would be the case cally efficient (box 2.6). in Madagascar, where almost three-quarters of the people, most of them poor, live in rural areas. The Case 2. Tradeoff: place more weight on economic bulk of rural poor people are in agriculture, and pro- growth and only address low-cost environmental ductivity growth in agriculture is critical to poverty concerns reduction. Yet agricultural productivity has been When environmental degradation is reversible and stagnant for the past four decades.50 has limited impact on economic growth in the short One of the deep constraints to increasing agricul- to medium term, placing greater weight on economic tural production in Madagascar is resource degrada- growth entails lower opportunity costs and should be (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Box 2.6 strong inverted-U shape relationship with income for Replacing natural assets with human-made sulfur dioxide and carbon monoxide, and even a one- assets can be costly for-one downward relationship between particulates and per capita income.51 But for water quality the ev- For years the Catskill watershed provided New York City residents with water of such high quality that it needed no idence is mixed. And for per capita emissions of CO2 filtration or chemical treatment. New York could even bot- there is a steady worsening as per capita incomes tle and sell its water to other cities. grow.52 Indeed, recent research concludes that, on the This began to change in the 1990s. The U.S. Environ- whole, there is little evidence of environmental qual- mental Protection Agency warned the city that it would soon have to invest in a filtration plant—for $6 to $8 bil- ity getting worse with initial growth and then getting lion, with annual operating costs of about $300 million. better at higher per capita incomes.53 Given the huge sums, the city began to ask why a water- Second, even for environmental assets that show a shed that performed so well for so long was now begin- positive association with per capita income growth, ning to fail. The main causes were uncontrolled land development in the Catskills and the intensified use of the association is not structural. Instead, the better en- land in and around the watershed. The combination of vironmental outcomes reflect the impact of regula- pollutants from residential communities and farms was tions and other polices put in place in response to pub- overwhelming the soil microbes that naturally filtered and lic action and pressures from society as preferences for cleansed the water as it percolated through. Because there had been little deforestation or soil environmental quality become stronger with higher erosion, and because much of the natural infrastructure of per capita incomes—not to any natural changes in the the watershed was still intact, it was possible to reverse composition of production or consumption.54 the situation. New York City then faced a choice: restore It is important to recognize that, although the re- the watershed, or build and run a filtration plant. Costs of the first option—improving sewerage treatment in the source degradation or depletion may be reversible, watershed and buying lands to prevent development— its impact on human well-being is not (recall the were estimated in the range of $1 to $1.5 billion, one-fifth degradation of Aral Sea described in box 2.3). Fu- the cost of an artificial filtration system. ture remedial action cannot compensate the genera- The choice was clear. As the commissioner of the city’s Department of Environmental Protection commented at tion or generations that live during the transition to the time, “All that human-made filtration does is solve a a better environment. Consider the costs of air and problem. Preventing the problem, through watershed pro- water pollution for human health. Recent estimates tection, is faster, cheaper, and has lots of other benefits.� suggest that about 11 percent of illnesses and prema- Source: Heal (2000). ture deaths in developing countries are due to envi- ronmental health risks from water supply and sani- tation and from urban and indoor air pollution.55 pursued. But as discussed below, this does not justify This is about the same as from malnutrition, which ignoring environmental concerns altogether. accounts for 15 percent of all illnesses and deaths. To justify a strategy of “grow first, clean up later,� The poor are particularly vulnerable since they have policymakers rely on the argument that is only par- fewer alternatives to polluted drinking water and are tially borne out by observation—that environmental more likely to live near heavily traveled roads where degradation gets worse initially and then gets better air pollution is highest. as a country develops—the environmental Kuznets For this reason, there is little justification for not curve. Often they also act as if the relationship is au- addressing at least some of these environmental con- tomatic—so that there is little need to actively ad- cerns along with economic growth. And often a large dress the problem. This could be the case if, say, proportion of the problem can be addressed at rela- shifts in the scale and sectoral composition of output tively low cost (see figure 2.4 on page 31).56 Indeed, and changes in technology within sectors result in a several cost-benefit studies have shown that the costs move away from pollution-intensive production to of addressing a sizable proportion of pollution can less pollution-intensive methods. be relatively low—and that the benefits of doing so But it cannot be assumed that environmental qual- can often be very high. In such cases there would be ity will necessarily improve with economic growth. grounds for stricter pollution control when pursu- First, such a relationship is observed only for some ing a high growth strategy even in very low-income environmental factors. For local air quality, there is a countries.57 (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Although policymakers often worry that pollution icant and long-lasting implications60 and having the control measures hurt the competitiveness of firms, asset may be important to the nation in the future— research does not support their concern.58 What is the environmental concerns need to be addressed observed is that countries can have quite different en- today. vironmental outcomes while achieving the same eco- Forests rich in biodiversity may have little amenity nomic growth rates. There is, for instance, a fairly value to the people in a poor country today. But as wide range of environmental outcomes in countries the country’s per capita income rises, that value is averaging a 3 to 5 percent annual growth (figure 2.3). likely to increase—making it important to have pre- Indeed, environmental outcomes at given incomes vented irreversible losses. Since these assets often are strongly influenced by how parties (citizens, busi- yield significant benefits to poor people in the coun- ness leaders, policymakers, regulators, NGOs, and try today, who rely heavily on it for their livelihood other market actors) react to economic growth and (food, fuel, fodder, and medicinal plants), it may be its side effects.59 This suggests that there can be a de- possible to address the environmental degradation mand in society for better environmental quality and poverty reduction simultaneously through fi- even at fairly low incomes. In a policy formulation nancing or cost-sharing from the larger community setting that allows for participation, voice, and chan- within the country or from abroad. Such schemes nels for feedback, countries are likely to experience need to be appropriately designed to provide, where better environmental outcomes at all levels of income necessary, alternative livelihoods for the local popu- (see chapter 3). lace.61 By avoiding irreversible degradation, these schemes can also keep the option value of the re- Case 3. Tradeoff: place more weight on the source for the nation in the future. Such cost-sharing environment schemes are interim proxies for economic growth in- When current depletion or degradation threatens to sofar as they align the preferences of the current be irreversible—or when the degradation has signif- (poorer) population with those of future (richer) populations. An example of such cost-sharing is Costa Rica’s en- Figure 2.3 vironmental services program. Costa Rica’s forests are Very different environmental outcomes with the attractive to tourists worldwide, given the rich biodi- same growth rates versity there. But the rate of deforestation in the Percentage annual change in environmental index 1970s and 1980s was one of the highest in the world. ᮢ 6 To protect this asset, Costa Rica designed a very in- novative scheme, the Payments for Environmental Worsening environmental outcomes Services Program, in which those who benefit from 4 the environmental services of the forests compensate those who bear the burden of maintaining the forests. 2 Under the scheme, a market has been created for a variety of services, with carbon sequestration among 0 the most successful (see box 8.5 in chapter 8). Some assets are overused or underprovided— –2 why? From the preceeding discussions it should be clear –4 that there is real value in designing development 2.5 3 3.5 4 4.5 5 strategies based on better management of a broader Percentage GDP growth rate per year portfolio of assets. A major problem in pursuit of this Note: The environmental index is constructed by giving equal weights to annual rates of deforestation, water pollution proxied by emissions of goal is that some assets (knowledge, environmental, organic water pollutants in per capita metric tons, and the increase in and social) tend to have characteristics of public CO2 emissions in per capita metric tons between 1987 and 1995. goods or externalities—that is, their use generates Source: World Bank (2001h). spillover benefits or costs to others that are not taken (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  into account. As a result, the stocks of these assets are can remove them. They change the climate in com- generally too small from society’s perspective. This is plex ways. Their global effect is the same regardless a consequence of market or policy failures. of where they are emitted. Again, individuals (and individual countries) do not factor in the spillovers Market failures of their actions on others. It is difficult to exclude people from using many of Underprovision. Knowledge is a public good— the functions of environmental assets—they are since once generated, it is difficult to exclude others nonexcludable. That means there are no well-defined from using it (nonexcludable) and consumption by private (individual or group) property rights, so mar- one individual does not reduce the supply for others kets cannot be used to ration the use of these assets (nonrival). Individuals or groups have less incentive or to expand their provision where that would be to invest in generating information and knowledge justified. Without property rights, it is not possible than is socially desirable. There is a tendency to free- to charge others for the use of a good or service. ride, expecting to benefit from a piece of knowledge Therefore not only does an individual or group have created by someone else. And since an individual’s little incentive to preserve or provide the asset (since use of a piece of extant knowledge does not reduce he or they cannot prevent others from using it), he the knowledge available to others, the generation of or they have every incentive to free-ride on others’ new knowledge can have large positive externalities efforts to preserve or provide those assets. From the or spillovers to society that are not taken into ac- perspective of society, then, the assets will be over- count in decentralized decisions to invest in creating used or underprovided. new knowledge. Thus knowledge also tends to be Overuse. For some renewable assets that are com- underprovided from society’s perspective. mon property goods (nonexcludable but rival) con- The existence of spillovers (externalities) that are sumption by one individual or group will reduce the not taken into account by individuals gives rise to supply for others. Each individual or group can gain the need for a “market for external effects� that can from overexploiting it in the short run, but lose in align the marginal costs and benefits to the individ- the long run as everyone else does the same and the ual with those of society as a whole—so that indi- asset falls below its regenerative capacity. Society viduals take into account their impact on others (in- then ends up worse off. ternalize the externality). When transaction costs are As an example of the common property problem, low, and property rights relatively well defined and consider offshore fisheries, many of which are greatly perfectly and costlessly enforceable, all affected par- depleted by overfishing. The fish biomass of several ties could get together to negotiate an outcome that important fisheries is now a mere tenth of its pre- is efficient from the perspective of society.64 Under exploitation level—90 percent of initial stocks have such circumstances, there is little need for policy been destroyed.62,63 Although all fishermen would intervention. benefit in the long term from a flourishing fishery, But generally transaction costs are significant, and individuals tend to act in their own interests and for many environmental assets private property catch as many fish as they can. This is the “tragedy rights are difficult to define. The costs of transac- of the commons�—or open access, with users over- tions are likely to depend on the number of people exploiting what would otherwise be a renewable re- involved and on whether the parties are concen- source as they race to get their share before others trated or diffused groups. (Clearly, not all problems deplete the resource. The same behavior applies to deserve being addressed—sometimes the transaction ozone depletion and climate change—clear examples costs may be higher than the social benefits.) Usu- of global common property goods (see table 2.2). ally, transaction costs are likely to be higher—and As discussed in chapter 8, the emission of ozone- the problem more difficult to solve—when the ef- depleting substances, or the use of fossil fuels (and, fects of the spillover fall on a large, diffuse group. to a lesser but still important extent deforestation The problem is likely to be even more difficult to and other land use practices that release CO2 and solve when a small concentrated group (that can or- other greenhouse gasses), results in gases accumulat- ganize itself at lower costs) generates the spillover, ing more rapidly in the atmosphere than natural sinks while the effect of the spillover is borne by a diffused (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 2.2 Examples of types of externalities addressed in each spatial arena Nonexcludability leading to market failures Space/scale arena Common property goods (rival)a Public goods (nonrival)a Externality effect of a few (concentrated) on many (dispersed, Usual case: externality effect Externality effect of many on who could also lack voice or be of many on many many (dispersed interests) otherwise excluded) (dispersed interests) Fragile Wells and grazing land (chapter Mines (chapter 4, box 4.7) Schooling for girls (chapter 4, rural 4, box 4.2) box 4.6) Knowledge outreach in Tunisia (chapter 4, box 4.5) Commercial Groundwater (chapter 5, box 5.10) rural Frontier land (chapter 5, box 5.12) Urban Disposal of solid waste in drains Pollution in Cubatão, São Paulo Public works in favelas in Brazil (chapter 6) (chapter 6, box 6.3) (chapter 6, box 6.5) Automotive air pollution Automotive air pollution National Cameroon forestry (chapter 7, Public health services in Ceara, box 7.8) Brazil (chapter 7, box 7.2) Global Ozone depletion (chapter 8) Global warming (chapter 8) Global warming (chapter 8) Note: Nonexcludable means that a user cannot be prevented from consuming that good or service. a Rival means that consumption by one user reduces the supply available to other users. Public goods are nonrival up to a threshold. Once that threshold has been breached they can become rival: for instance, a freeway, as it shifts from being underutilized to being congested; or the atmosphere once the pollutant concentration exceeds the atmosphere’s absorptive capacity. Source: Authors. group that incurs higher costs to organize itself be- competitiveness of certain products, industries, re- cause it lacks the ability or voice to negotiate. Solv- gions, or to support particular social groups—that ing such problems requires policy interventions and have adverse environmental impacts. When the so- supporting institutions (see the section titled “Cor- cial costs outweigh the social benefits, this consti- recting the overuse and underprovision of important tutes a policy failure, requiring offsetting corrections assets� and chapter 3). And as discussed in the rest or even elimination of the policy intervention. of this Report, where such institutions do not exist, So-called perverse subsidies are an example. Many it is necessary to find mechanisms or catalysts that subsidies are introduced initially to stimulate the use may spur their emergence. Table 2.2 shows some ex- of a good or service that is underutilized—fertilizer, amples that are taken up in each of the chapters of electricity, water. But in the absence of sunset clauses this Report, which is organized by space and scale. and with the creation of a constituency based on per- ceived acquired rights, these subsidies can persist be- Policy failures yond their economically useful life and be detrimen- The overuse or underprovision of an environmental tal environmentally. They can be economically costly asset can sometimes be the result of policy interven- if they sustain processes that would otherwise not be tions to correct market failures that in turn have con- viable (for example, producing rice in California). sequences for another set of problems—leading to a They can also be economically harmful if they re- policy failure in the case of the latter. For example, duce the costs of environmental inputs to the point countries may implement policies—to improve the that eventual degradation of this complementary (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Box 2.7 Perverse subsidies in India Power subsidies in India have resulted in overpumping of to power but suffer the opportunity costs of having subsidies aquifers, reducing the availability of drinking water, and encour- go to others. Since State Electricity Boards are not allowed to aging water-intensive crops in areas where water is scarce. charge realistic tariffs, their accumulated deficits are at least In not distinguishing between peak and nonpeak tariffs, the partly serviced by deducting their dues from the Central Plan implicit subsidy has also increased the incentive to overbuild Assistance to the states. This reduced central assistance, capacity. (In fact, the World Bank estimated in 1991 that vari- along with the direct state subsidies to power, means that the ous measures to reduce peak power usage could reduce power poorest do not receive adequate basic services, such as health generation requirements by about 12 percent in 10 years.) care and primary education. In addition to facilitating the excess drawdown of aquifers, the subsidy is costly for poor people, who typically lack access Source: World Bank (2000e). Adapted from box 5.2. asset affects productivity (for instance, power sub- would be real GDP gains associated with their re- sidies in India encouraging the overpumping of moval).72 Social considerations may thus call for ground water—box 2.7) or if in attempting to ben- incentive-compatible transfers and compensation efit one activity, they harm others, so that their net (see chapter 7, box 7.7), as well as other support (vo- impact is negative.65 cational training for other jobs) to enable the transi- Energy subsidies,66 the bulk of which are directed tion out of perverse subsidies. to fossil fuels67 in both industrial and developing countries, entail economic efficiency losses. But they The costs and benefits of correcting underprovision also have highly deleterious effects on the environ- or overuse ment, some reflected in higher economic costs.68 If environmental assets are generally overused or un- Subsidies to fossil fuel and nuclear energy in Orga- derprovided, how can society begin to weigh the re- nisation for Economic Co-operation and Devel- turns to addressing an environmental issue against opment (OECD) countries total $71 billion annu- the costs? It depends on the starting point. ally.69 Studies that simulate the effects of removing Starting from a moderately degraded state, it is coal and other energy subsidies—either for individ- often possible to make significant improvements at ual countries or the world—all find significant envi- fairly low cost. Very simple low-cost measures to abate ronmental benefits in reducing CO2 emissions. And water pollution (for example, installing water filters) most studies that look at the economic effects also can often remove close to half the pollutants. The costs find real GDP gains.70 The problem is not limited of additional reductions in degradation are likely to to industrial countries. While many developing rise more steeply because more sophisticated measures countries significantly reduced their energy subsidies are needed. Consequently initial costs are low and rise in the 1990s, they would still gain considerably by more steeply as the quality of the asset is restored. removing the subsidies altogether (table 2.3). Al- In another example, consider the costs (estimated though it is often argued that these subsidies are in the early 1990s) of reducing air pollution from needed to help poor people, the poor rarely benefit. transport in Mexico City. The cheapest emission re- In general, subsidies encourage the use of the sup- ductions were found initially among the busiest ve- ported inputs, processes, or products—and reduce hicles, especially those that were driven downtown the incentives to find alternatives that may be more during most of the day. Further emission reductions economically efficient. Fuel subsidies to fossil fuels required modifications for a larger part of the vehi- reduce the incentive to develop renewable energy cle fleet—which became more expensive. (Emissions sources.71 reductions for buses and taxis would have cost only Although dismantling perverse subsidies may be $300 a ton because of their higher annual mileage, good for society, some groups would lose. For exam- compared with $1,600 a ton for passenger cars.) The ple, studies looking at the effects of removing energy incremental costs of mandatory inspection and subsidies in industrial countries point to a signifi- maintenance programs for vehicles in use would have cant loss of jobs in the coal sector (although there been higher still (with costs rising as the standards (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 2.3 The benefits of full-cost energy pricing Effects of subsidy removal Average Gain in Reduction in subsidy economic Reduction in carbon (percentage Cost of efficiencyb energy dioxide of reference subsidy (percentage consumption emissions Country price) ($ billion) of GDP) (percent) (percent) Islamic Republic of Iran 80.4 3.6 2.2 48 49 República Bolivariana de Venezuela 57.6 1.1 1.2 25 26 Russian Federation 32.5 6.7a 1.5 18 17 Indonesia 27.5 0.5a 0.2 7 11 Kazakhstan 18.2 0.3 1.0 19 23 India 14.2 1.5 0.3 7 14 China 10.9 3.6 0.4 9 13 South Africa 6.4 0.08 0.1 6 8 Total 21.2 17.2 0.7 13 16 a Based on 1997 (hence, pre–financial crisis) prices and exchange rates. b Transfers in the form of subsidies lead to increases in consumer and producer surplus (defined as the difference between what consumers are willing to pay for a unit of the good and what they actually pay, and as the difference between what producers actually receive when selling a product and the amount that they would be willing to accept for a unit of the good, respectively). These increases however are smaller than the total amount of the transfers (subsidy), which means that the subsidy entails a net loss in social welfare. Hence removing the subsidy would entail net economic efficiency gains. Source: International Energy Agency (1999a); Myers and Kent (2001). were tightened) and the additional reduction in terialize in the long term while costs are incurred in emissions would have been lower assuming the the short term, as with climate change. Sometimes the cheaper alternatives had already been implemented. costs and benefits that occur in the future at different Improvements in the fuel mix are at the high end of points in time can be discounted or converted into an marginal costs, providing even less additional reduc- equivalent set of costs and benefits today, using the tions in emissions (figure 2.4).73 consumption rate of interest—the rate at which con- But if a resource is substantially degraded, the sumption tomorrow can be substituted for consump- costs of restoring it can jump dramatically. For ex- tion today without changing social well-being. But the ample, restoring water to the Florida Everglades is farther the benefits occur into the future, the greater estimated at $7.8 billion. Sometimes, when the de- the bias toward inaction, because discounting auto- pletion or degradation reaches very high propor- matically reduces the valuation of these benefits.74 tions, even if it is technologically feasible to address However, if one recognizes that a longer time- the problem, it may not be economically viable to do horizon also means that there could be significant un- so. And when the problem is technically irreversible, certainty about the interest rate itself, and if the cost- the costs become infinite. For example, once sub- benefit analysis takes this uncertainty into account, stances such as oil, petroleum, and chemical solvents the valuation of benefits over distant horizons in- (which are part of a certain type of contaminants creases. (Since the consumption rate of interest de- known as nonaqueous phase liquids) penetrate an pends in part on the forecast of future consumption, aquifer, they are almost impossible to remove. uncertainty about long-run economic forecasts would imply an uncertainty about the interest rate.)75 As a Complications arising from long time-horizons nonrigorous but illustrative example, an exercise al- and uncertainty lowing for such uncertainty looked at the benefits of One of the difficulties with environmental problems addressing climate change. Using the government is that the costs and benefits of addressing the issue are bond rate (generally taken to be the best proxy for the sometimes highly uncertain. The problem is even consumption rate of interest) of 4 percent as the ini- more complicated when the costs and benefits are re- tial rate, the study assumed that future rates could ei- alized at different points in time—often benefits ma- ther rise to 7 percent or fall to 1 percent, and showed (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Figure 2.4 Reducing emissions in Mexico City Marginal cost of emission reductions (U.S. dollars per ton) 2,500 Fuel improvements Emissions 2,000 standards 1,500 Passengers cars ᮢ Taxis ᮢ ᮢ 1,000 (replacement) ᮢ ᮢ Minibuses Gasoline Strengthened inspection ᮢ trucks ᮢ 500 ᮢ ᮢ ᮢ Inspection of passenger cars ᮢ ᮢ 0 Inspection of high use vehicles ᮢ Retrofitting (natural gas and LPG) 0 200 400 600 800 1,000 1,200 1,400 Cumulative emission reductions (millions of weighted tons) Source: Eskeland and Devarajan (1996). how allowing for this uncertainty could add about 80 Addressing market failures percent to the expected present value of addressing Whenever spillovers (externalities) exist, there is a climate change (carbon mitigation) relative to the val- coordination problem—private marginal costs and uation under a constant interest rate of 4 percent.76 benefits diverge from social marginal costs and bene- Thus, if a dollar’s worth of benefits in the future is fits, and policies that align the two are needed. While worth 25 cents under a constant interest rate of 4 per- the focus is generally on formal policies or mecha- cent, it would be worth 20 cents more (45 cents) al- nisms, informal community institutions, which rely lowing for this uncertainty in interest rates. on informal norms and networks, can also be key Reducing uncertainty generally requires the gen- means for addressing coordination problems. eration of knowledge and information. The possibil- It is usually most efficient to address market fail- ity of hitting thresholds also highlights the impor- ures at the lowest level that can internalize the exter- tance of developing and monitoring key indicators nality—this is known as the principle of subsidiar- that can signal coming problems. Unfortunately, as ity. Note that this can have a bearing on the roles of discussed earlier, such knowledge and information is informal and formal mechanisms.77 Spillovers that also usually underprovided because individuals, in affect people in a single community should be ad- deciding whether to invest in knowledge and infor- dressed at that level. But quite often spillovers ex- mation gathering, do not take into account the pos- tend much beyond a single community and must itive spillovers that this can generate for society. therefore be dealt with in a broader setting. For ex- ample, maintaining a hillside forest is of interest to Correcting the overuse or underprovision groups at many levels. Local communities and those of important assets living near the forest may want to manage it to pro- Developing indicators to determine how assets vide fuel and food. Communities in the larger wa- are being used is a challenge (see earlier section on tershed may have an interest in maintaining the measuring sustainability). Addressing the overuse or same forests to mitigate flooding and siltation down- underprovision of assets is another. This section dis- stream. The nation may want to maintain the eco- cusses the mechanisms to address the two main rea- tourism potential of the forests. The world at large sons for the overuse or underprovision of environ- may be concerned about the forest’s ability to sup- mental assets discussed earlier—market and policy port biodiversity and carbon stocks. This requires failures. corresponding action at all levels. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Appropriate formal mechanisms to address a mar- agency know the effect of a given tax rate. Increas- ket failure can range from using command and con- ingly it is being recognized that a combination of trol regulations and harnessing market forces to cre- command and control and market-based instru- ating markets and engaging the public (figure 2.5). ments is superior to either alone. So if the interest is Usually a mix of mechanisms is required to address in reaching a desired quantity or quality at lowest a problem, although occasionally one is applicable cost, a target can be set for overall emissions, and or sufficient. In general, the choice of mechanisms permits or licenses would allow industry to emit up needs to be guided by the following: to the total but trade amongst themselves to achieve the overall goal at lowest cost to society. Ⅲ The effectiveness of the instrument in meeting the Countries are thus moving to economic instru- objective ments to address environmental concerns. These Ⅲ The efficiency of the instrument—including offer more potential in terms of efficiency now (sta- whether it ensures static efficiency (achieving the tic) and over time (dynamic). They can offer more goal at minimum cost to society) and dynamic ef- flexibility in meeting objectives. And they provide a ficiency (providing incentives for innovation and source of government revenue that can address other the search for alternative, more efficient ways of public concerns. There are difficulties: many envi- meeting the objective)—while minimizing the im- ronmental assets do not have well-defined property plementation costs (monitoring, enforcement) rights, and operating in the market requires that Ⅲ The extent to which the instrument minimizes the property rights be assignable. Even so, some part of costs of meeting other objectives when there are the depletion or degradation of the asset often takes tradeoffs place in the arena of markets—and is thus amenable Ⅲ The effects on distribution and poverty. to correction through economic instruments.78 And technology can sometimes change whether an asset Regulations—command and control can have well-defined property rights and hence op- Regulations, or command and control measures, have erate in the market (meters can foster water markets traditionally been the means of aligning public and that would otherwise not be feasible). private interests. Such measures—which include li- For example, even though private property rights censes, permits, quality standards, emission standards, to clean air—the asset—are not assignable, it is still process standards, product standards, and prohi- possible to deal with aspects of the degradation bitions—have the advantage of targeting a desired within markets. Emissions or fuels can be taxed, or level (quantity) or quality of an asset more easily than vehicle use in the case of vehicles. other mechanisms. For example, air quality can be ad- One proposal for dealing with global common– dressed by process standards or emissions standards. related concerns such as ozone depletion and climate Similarly, management and planning, also a regu- change is to impose user charges or levies at the global latory approach, can sometimes work. For urban pol- level.79 (Of course, curbing air pollution by taxing lution, zoning and land use restrictions can be impor- vehicle and industrial emissions in cities, as men- tant if, for instance, there are economies of scale in tioned above, would also be an important compo- dealing with pollution when firms are in one place. nent of a strategy to deal with climate change.)80 User Though blunt, zoning can be an effective tool in han- charges create incentives to reduce environmental dling environmental damage when the spatial dimen- pressures (the incentive function of user charges). sion matters. Experience suggests, however, that reg- They can also mobilize financial resources that can ulations are sometimes less efficient and effective than be earmarked to fund the conservation and restora- market-based instruments—and costly in the institu- tion of global common goods (the financing function tional capacity they require for implementation. of user charges). Using markets—taxes and subsidies Creating markets: property rights Pollution can also be addressed through such mar- and trading permits ket instruments as a tax, but the impact of tax rates Sometimes it is possible to define and allocate prop- on the levels of emissions cannot be known before erty rights that are supported through regulations and the fact. Only by trial and error would a regulatory institutional arrangements, which then create markets (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  Figure 2.5 Mechanisms to address market and policy failures Government helping to correct market failures ᮢ ᮢ ᮢ ᮢ Command and control Market-based instructions Creating markets Engaging the public regulations that affect that affect prices – property rights – public participation quantities – taxes – tradable permits – information – standards – user fees – international offset disclosure – bans – deposit refund systems systems – permits and quotas – subsidies ᮢ ᮢ Removing bad policies (e.g., perverse subsidies) ᮢ Private sector and civil society helping to correct policy failures Source: Adapted from “Five Years after Rio: Innovations in Environmental Policy,� World Bank (1997a). and allow the advantages of efficiency. Indeed, this plement. Finding the right balance between giving approach (tradable permits for pollution emissions) free play to market forces and monitoring and en- has been a major innovation in the last decade.81 The forcement is a big challenge. use of command and control to regulate the overall allowable pollution levels, together with tradable per- Engaging the public: publicizing mits, creates a market for pollution abatement that and sharing information would not otherwise exist. Making permits tradable Civil society can monitor and ensure compliance gives firms an incentive to look for the most cost- with regulations. A good example is Indonesia’s effective solutions for pollution abatement, because PROPER program, which discloses the noncompli- firms that lower their pollution more effectively or at ance of polluting firms to the public (while reward- a lower cost than do other firms can sell their excess ing compliance), encouraging local communities to credits to those firms. Firms then face an opportunity put pressure on companies that score poorly. The cost of pollution, which creates incentives to find program focused initially on water pollution. It ranked cheaper abatement methods, encourages less pollu- companies by their emissions, and disclosed the re- tion in aggregate, and ensures dynamic efficiency. sults in stages, recognizing good performers first and In OECD countries, tradable permits are seen as giving the bad ones six months to clean up. Within a way of harmonizing environmental protection 18 months, half of the noncomplying firms were ob- with economic efficiency.82 The U.S. sulfur dioxide serving the legally established standards.83 reduction scheme to reduce acid rain is an example, Public participation and monitoring can also relying on tradable rights and credible threats in make voluntary compliance agreements more effec- cases of noncompliance. Similarly, Iceland and New tive. Such agreements with the private sector are Zealand have revived fish stocks by assigning fishing becoming popular in addressing environmental prob- rights at a sustainable level and allowing fishers to lems in many OECD countries, especially when reg- trade their quotas freely. ulatory structures cannot address specific issues. The These arrangements, despite their advantages in agreements can be commitments devised by the gov- providing the right incentives to adopt least-cost ernment (or an environmental agency), with individ- solutions, can still be costly to administer and im- ual firms invited to participate. Or they can be nego- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  tiated commitments for environmental protection energy, and food—especially in industrial countries). developed through bargaining between a public au- Damaging races for property rights abound (individ- thority and industry. They can also be unilateral com- uals or companies pushing to develop the remaining mitments initiated by the private sector. These agree- natural resources ahead of someone else: minerals, ments are not limited to environmental issues. For forests, fisheries). While the world is moving toward instance, tour operators have agreed with the City and greater trade liberalization, trade restrictions (tariffs Borough of Juneau (Alaska) to minimize any adverse and non-tariff barriers) remain on precisely the goods impacts of tourism on the local community. in which developing nations are competitive, includ- Voluntary approaches—designed, implemented, ing agricultural products and textiles. and monitored properly—can work. But they can If the policy recommendations of a decade ago con- also have problems. Control can be weak because in- tinue to be the best route to improving the welfare of dustry does not provide adequate control mecha- millions of people, why have they not been imple- nisms or because of a lack of sanctions. Free-riding mented? In reality, even the win-win policies have been is possible when other firms bear no cost of comply- much harder to implement than initially thought— ing with the agreement while reaping the benefit. vested interests were much more entrenched, and in- Then there is the possibility of regulatory capture— stitutional development was harder to foster. The per- when powerful businesses exert undue influence on sistence of policy failures even when society as a whole the process.84 Encouraging the participation of civil can benefit from their removal often reflects powerful society can help to mitigate these problems. interest groups blocking the necessary reforms. Just as participation by civil society, together with greater in- Addressing policy failures formation disclosure and transparency, can help in Many environmental stresses today are not the result monitoring the implementation of environmental reg- of ignorance about what policies to adopt. Indeed, 10 ulations by individual companies, so too can it be years ago World Development Report 1992 addressed an important means of improving the accountability the complex issues of environment and development of the public sector (see figure 2.5). The blocking of and concluded that several doable, “win-win� policy reforms by powerful groups represents one of the options were available (box 2.8). A decade later these deeper barriers to the emergence of the institutions policy recommendations remain valid, but many of needed to support environmental policies. them have, at best, been adopted or implemented This Report as a whole tries to show that environ- only partially.85 As discussed, the widespread use of mental problems are, at their root, social problems. subsidies remains high across the globe (for water, The distribution of assets, and of the costs and ben- Box 2.8 World Development Report 1992: Development and the Environment World Development Report 1992 identified the challenge of pur- Ⅲ Priorities for action. Removing perverse subsidies, strength- suing development and poverty alleviation in a generation ening property rights over common pool resources, expand- (1990–2030) that would see world population increase by 3.7 bil- ing service provision, increasing voice and participation, care- lion, food production double, and energy use triple. It called for fully evaluating environmental tradeoffs with special regard actions that would mutually reinforce environmental protection for long-term irreversible or large-scale damage, matching the and development: provide clean air, sanitation, and clean water; government’s role to its capability improve management of soils; and protect biodiversity. It saw Ⅲ Policies for sustained development. Where possible, rely- great scope for win-win interventions that would simultaneously ing on incentives rather than regulations; curbing the influ- improve the environment and provide local economic benefits. ence of vested interests That report also called for improved institutions for environ- Ⅲ Partnership for solutions. Partnering with high-income coun- mental regulation, using market-based incentive principles tries to expand market access and to increase development as- where possible, and made a series of policy recommendations: sistance; partnering with high-income countries to finance the costs of global environmental priorities, especially those requir- Ⅲ Win-win policies. Eliminating subsidies for energy inputs, ing the protection of natural habitats in developing countries. pesticides, fertilizer, irrigation water, logging, and ranching (perverse subsidies); taxing urban road emissions Source: Authors; Acharya and Dixon, background paper for the WDR 2003. (c) The International Bank for Reconstruction and Development / The World Bank �������� � ������� ��������� �� ������  efits of different policies, as well as the role of trust, policy instruments and mechanisms to address these are all critical to the ability of societies to develop externalities. As discussed, the nonadoption or non- competent rules and institutions (chapter 3) to ad- implementation of these policies reflect the fact that dress environmental, social, and economic problems. the supporting institutions—with the appropriate This chapter has discussed the importance of characteristics—have not yet emerged. Chapter 3 managing and ensuring a better balance of assets to looks at the characteristics of appropriate institu- enhance human well-being on a sustained basis. It tions, the potential barriers to their emergence, and also covered the externalities and coordination prob- how these may be addressed; it focuses on catalysts lems that generally lead to the overuse or underpro- that may increase the likelihood of the timely emer- vision of some of society’s key assets, detailing the gence of these institutions. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 3 Institutions for Sustainable Development We have to see individual freedom as a social Ⅲ What are institutions? They are the rules and orga- commitment. nizations, including informal norms, that coor- Amartya Sen1 dinate human behavior. They are essential for sus- tainable and equitable development. When they function well, they enable people to work with each T he previous chapter made the case that for peo- other to plan a future for themselves, their families, ple to thrive—especially over a longer time hori- and their larger communities. But when they are zon—a wide range of assets must also thrive. weak or unjust, the result is mistrust and uncer- Managing a broader portfolio of assets can ensure that tainty. This encourages people to “take� rather than the growth process contributes to people’s well-being on “make,� and it undermines joint potential.3 a sustained basis. Policies can be designed to improve Ⅲ What are the key functions of the institutional the management of assets.2 In practice many socially environment in promoting human well-being? It worthwhile policies are not adopted or implemented. must pick up signals about needs and problems— The institutional perspective examines the forces that particularly from the fringes; this involves gen- work to shape and implement policies. erating information, giving citizens a voice, re- If institutions are to protect people and a broad sponding to feedback, and fostering learning. It portfolio of assets, they must respond to and shape must also balance interests—by negotiating change the major changes that will unfold over the next and forging agreements, and by avoiding stale- 50 years: urbanization, technological innovation, mates and conflicts. And it must execute and im- economic growth, shifting social values, changing plement solutions—by credibly following through scarcities for environmental and natural assets, and on agreements. stronger linkages among nations. Institutions thus Ⅲ What are the barriers to the emergence of such an must be stable, but they also must be capable of institutional environment? One is dispersed inter- changing and adapting, and new institutions must ests. Concentrated interests are often given too emerge. much weight, as in the assignment of property Chapter 3 focuses on the coordination of human rights for land and water, and in the operation of behavior that is required for people and assets to government. A second barrier is the difficulty of thrive, particularly institutions that sustain this co- forging credible commitments to protect and nur- ordination—by channeling interests, and by shap- ture persons and assets. And a third is institutions ing the quality and effectiveness of growth. This that are not inclusive. When societies and processes chapter addresses four questions: are unequal and undemocratic, it is more difficult  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  to coordinate dispersed interests and forge credible Because institutions govern behavior, they are so- commitments. cial assets (or liabilities, when bad or weak). So are Ⅲ How can these barriers be overcome? Sometimes the elements of social capital, such as trust and per- social and economic development offers opportu- sonal networks. The distinction between social capi- nities for change. Structural changes—urbaniza- tal and institutions can sometimes be blurred, and tion, the demographic transition, the redistribution there are strong influences between the various so- of wealth (particularly increments of new wealth)— cial assets (see chapter 2, note 14). For example, the unleash dynamic forces and opportunities for insti- exchange of goods and services may be based on per- tutional change. Initiatives to channel information sonal networks and other forms of social capital in can also serve as catalysts for change. Information the village, but on formal institutions in the city. can empower people by giving them more voice in Similarly, general trustworthiness in a society can be public services and allowing greater transparency strong either because of strong personal networks— and accountability in the activities of governments or because of good laws and judicial systems that are and firms. generally accepted. In fact, as societies become more complex, trust in individuals (based on knowledge Institutions coordinating human behavior of character and frequency of interpersonal contacts) Institutions are the rules, organizations, and social is supplemented by trust in institutions (rules and norms that facilitate coordination of human action organizations) when dealing with strangers. (figure 3.1). On the informal end, they go from trust Two very important dimensions of coordination and other forms of social capital (including deeply are others and future. Markets are institutions with rooted norms governing social behavior) to informal coordinating functions (box 3.1). A market coordi- mechanisms and networks for coordination. On the nating the transactions of individuals and firms en- formal end, they include a country’s codified rules and ables them to serve others and invest for the future laws, and the procedures and organizations for mak- (as when a baker builds an oven in response to ing, modifying, interpreting, and enforcing the rules greater demand). But markets need the support of and laws (from the legislature to the central bank). other institutions to ensure confidence, control, and Figure 3.1 Social norms, rules, and organizations for coordinating human behavior INSTITUTIONS Social RULES capital ORGANIZATIONS Informal Formal Easy to Government change Trust Rules ᮢ Regulations agencies Firms Networks Civil society Laws organizations Shared values ᮢ Norms Difficult to Police change Religion Traditions Constitutions Courts Traditional, Modern, ᮢ ᮢ informal formal (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  Box 3.1 and transactions. This Report builds on that foun- The market as a coordination mechanism dation but expands the discussion to aspects of well- being that are not limited to income or easily Adam Smith, 1776, captures well how market forces coor- amenable to markets, including services from envi- dinate human behavior for the common good with his famous “invisible hand� formulation: “It is not from the ronmental, natural, and social assets. The actors in benevolence of the butcher, the brewer or the baker that society partly play under a given set of rules and they we expect our dinner, but from their regard to their own partly shape those rules. Firms, government, and self interest.� Going beyond this, Friedrich von Hayek lu- civil society are positioned to act and to influence cidly depicts the market as a discovery device. Through voluntary exchange, the market generates price signals, the actions of others, playing complementary roles conveying information about scarcity. When markets are in coordination. competitive and decentralized, the “knowledge of the par- ticular circumstances of time and place� will help guide Market players decisionmaking (Hayek 1945, p. 526). Hayek uses the ex- ample of a rise in the scarcity of tin. Users of tin will re- Market participation enables people to specialize and ceive a signal through the higher price that “some of the work together, and to apply their skills and resources tin they used to consume is now more profitably employed in the best way possible, as the price mechanism pro- elsewhere, and that in consequence they must economize vides information to coordinate (see box 3.1, and tin.� Simultaneously producers receive a signal to seek new supplies. The system operates with remarkable econ- chapter 7). But the institutional framework must en- omy of knowledge and gathers dispersed interests. sure that markets function and deliver their expected “The marvel is that in a case like that of scarcity of one benefits. Where formal institutions are weak, activ- raw material, without an order being issued, without more ity will locate in the informal sector—in many de- than perhaps a handful of people knowing its cause, tens of thousands of people whose identity could not be ascer- veloping countries the share of workers in the infor- tained by months of investigation, are made to use the mal sector is more than 60 percent.4 This means that material or its products more sparingly; i.e., they move in a substantial proportion of economic activity is de- the right direction� (Hayek 1945, p. 527). prived of potential productivity-enhancing support, Markets can perform these functions even when there are important public concerns, of course, but they need the and may also escape guidance from institutions that support of other institutions to handle those concerns. For manage environmental and social aspects.5 But mar- instance, an environmental protection agency issuing trad- ket power can also give large domestic firms or able pollution permits or charging for emissions (as with multinational corporations potential coercive pow- sulfur in the United States, or water pollution in Colombia), can use markets to find cheap ways to abate pollution. ers akin to the government’s in terms of serving nar- More typically, emission regulations fail to use markets as row rather then broad interests. Thus, a major chal- much as they could to challenge firms to find cheaper and lenge for government and formal institutions is to more environmentally friendly ways to produce. be more welcoming and supportive of private actors Source: Authors. with appropriate safeguards to ensure the public in- terest is not compromised. the right incentives. Economic agents face transac- Government tion costs, and institutions can coordinate to lower In many areas, government plays a central role in or- those costs. Traders may want to cheat each other— ganizing dispersed interests: meeting national goals or to renege on their promises. But social capital and and balancing competing interests. Unlike social modern institutions—such as the rule of law and the norms and values, government operates a rulemak- enforcement of contracts—reduce this probability (a ing process by which rules can be changed more transaction cost), facilitating mutually beneficial quickly, with vision and design, and still be forceful. transactions. Particular challenges for these other in- But if a government—with its socially sanctioned stitutions are to commit to protect and nurture peo- coercive powers—finds itself unbound by rules (e.g., ple and assets—and to serve dispersed interests. by a constitution or equivalent with the separation World Development Report 2002: Building Institu- of powers it brings), how can it commit itself as a tions for Markets, looked mainly at human-made as- partner? The private sector will be less willing to in- sets, and its focus was narrower than this Report’s. vest and do business if instability and risks of expro- But it looked deeper into what sustains investments priatory consequence have not been curtailed. Un- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  less institutions succeed in separating the powers of commitment and related outcomes are positively government and providing meaningful checks and correlated with democratic practices, though some balances, communities and the private sector will be countries have also done well environmentally using less forward looking, and environmental and natural other channels. assets will be hurt through inappropriate investment and conservation. Civil society Democratic forms of decisionmaking, despite There has been a sharp increase in the active mem- their limitations, are associated with processes that bership of civil society organizations in the past two exhibit desirable institutional features; they pick up decades (figure 3.2). During this period civil society signals, balance interests, and execute chosen ac- organizations have become more capable, sometimes tions. A growing body of literature finds that indi- acting independently, sometimes influencing the ac- cators of voice and accountability are closely associ- tivities of government and the private sector.7 ated with better development outcomes, including Civil society’s role has been most notable in mo- higher national income per capita, lower infant mor- bilizing support for specific issues, supplying infor- tality rates, and lower illiteracy rates (chapter 7, and mation, and providing third-party verification. Civil WDR 2002). Empirical evidence strongly supports society organizations often take initiative and pro- Amartya Sen’s finding that democracy—helped by vide voice for unheard interests—building the trust, free speech—plays a key role in eliminating famine legitimacy, and knowledge needed. In West Bengal, and eliciting effective disaster relief.6 Theoretical and India, the Ramakrishna Mission works with youth empirical evidence also suggests that environmental clubs in 1,500 villages, undertaking needs assess- Figure 3.2 Growing participation in civil society organizations, 1981–97 Religious organizations Environmental organizations Political parties Active members Active members Active members (percent of respondents) (percent of respondents) (percent of respondents) 70 14 14 60 12 12 50 10 10 40 8 8 30 6 6 20 4 4 10 2 2 0 0 0 1981 1990 1995–97 1981 1990 1995–97 1981 1990 1995–97 Year Year Year Mexico South Africa Argentina Republic of Korea Brazil Chile Note: Countries shown are the only developing countries for which longitudinal data are available. In the 1995–97 survey, participation data were collected in 11 additional developing economies: Bangladesh, Colombia, the Dominican Republic, Ghana, India, Nigeria, Peru, the Philippines, Taiwan (China), Uruguay, and República Bolivariana de Venezuela. Active group membership in these additional countries in 1995–97 averaged 25.1 percent for religious organizations, 8.5 percent for political parties, and 6.2 percent for environmental organizations. The survey was not conducted in South Africa in 1990 or in Brazil or Chile in 1981. As a result the true shape of the South Africa trend line may differ from what is shown. Source: The World Values Survey. (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  ments and helping to identify external support for yield dispersed benefits, such institutions do not social welfare and development projects.8 In Thai- emerge easily. land the Appropriate Technology Association has fa- Protective institutions are those that define and cilitated collaborations among farmers, university support control rights in terms of access to and use scientists, and government authorities to reduce of assets central to human well-being; for example, pesticide poisoning through the use of biological who can graze their cows where, who controls a fac- controls.9 tory, who takes home eventual profits—or who is al- Civil society and the media also play an important lowed to discharge emissions—when and where, and monitoring role—for example, improving account- at what price. A special subset of protective institu- ability and reducing incentives for corruption in tions are private property rights. They entail well- government. The commercial exploitation of natural delimited (and not too limited) rights of use and resources is particularly susceptible to corruption;10 decisionmaking for an owner, typically including for example, in Bolivia, a voracious illegal trade in en- rights to sell or lease an asset. These institutions in- dangered species in the mid-1980s was facilitated by clude a commitment from society (extended family, customs officials who, in the absence of independent neighbors, villagers, or governments) to help protect oversight, doctored export permits. An active civil so- these rights. In a modern state, this commitment re- ciety makes it easier for environmental officials and quires an active obligation of enforcement from the others to publicize mismanagement; for example, in government (police or judges) and the assurance that Cameroon civil society plays a role in monitoring a the government itself respects those rights. For this promising new logging concession scheme (chapter reason, the security of property rights is closely asso- 7). When citizens have access to independent sources ciated with the rule of law—so that people can make of information, meaningful channels for political par- assumptions about what will be respected as theirs. ticipation, and legal protection against retribution, But protective institutions also arise to manage as- they can become a strong political force for improved sets that are not amenable to private ownership. A performance in public agencies. pollution control agency defines and protects con- Civil society organizations are particularly impor- trol rights to the air. A central bank protects the in- tant for promoting environmentally sound develop- tegrity of a currency and financial system. And com- ment strategies. Environmental management often munities can solve coordination problems and allow is about “downstream� spillover effects, and NGOs assets to thrive, as the literature on common property can often provide knowledge and links between ge- resource management and social capital demonstrates ographically dispersed locations. The surge in civic (the important interplay between community coop- environmental activity has often been part of a broad eration and modern society’s formal institutions can upward trend in civil society activity generally.11 be beneficial or detrimental).13 Opinion surveys consistently reveal levels of public In the broader institutional environment, firms, environmental concern in developing countries equal government, and civil society organizations together to that in industrial countries.12 give force to rules and norms. For instance, civil so- ciety can help keep the judiciary independent. In- Institutions protecting assets deed, rules and norms are typically backed by sanc- How can there be assets? Is it not easier to take than to tions, and many are effective only when agencies and make? Institutions that restrain the taking of assets— organizations back them up. through norms or under threats of punishment—are Some assets protect other assets—and some are essential for assets to thrive, whether the assets are more vulnerable to predation than others (box 3.2). human-made or natural, whether they are transacted Land is protected by formal institutions (property in markets or not. For example, restraint is necessary rights, courts, and titling agencies) in a modern for the forest and the aquifer to yield sustained ben- urban setting, but it is relatively less protected under efits when population density, changes in technol- periodic, nomadic grazing. Fisheries turn from need- ogy or preferences, and other developments increase ing no protection when stocks are abundant—to the demands on their use. Institutions ensuring such needing costly protective institutions as fishing tech- restraint allow assets to thrive; however, because they nology and demand develop. Savings under mat- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 3.2 Assets, threats, and protection If not properly protected, assets cannot thrive and contribute erty rights in business disputes.� Among the six economies to growth or human well-being. Assets are not all vulnerable where property rights were least secure, income (GDP) fell in to the same threats, but all assets—natural as well as human- the decade of transition—by 40 percent or more. Many coun- made, in the village and the city—depend on protective insti- tries with contractions in GDP also had large increases in death tutions. In the examples here, the protection is weak. rates. With bandits, there is no use in bringing in oxen Without commitments to law and property, In Uganda, animal traction for plowing is now promoted as a well-managed exploitation is unlikely way to raise agricultural productivity. In the subcounty of Nam- Insecure property rights also hurt natural and environmental biti, the administration plans to provide selected farmer groups assets: fish populations are threatened, and trees are cut pre- with trained oxen, in the hope that this will raise productivity maturely. Because people are not sure that they will be able and induce other farmers to put their savings into oxen. Farm- to harvest tomorrow, they take what they can today—initiat- ers there are skeptical, however; the problem of mobile and ing a race for property rights. armed bandits has not been resolved, and farmers expect that For a cross-section of countries, the insecurity of property any oxen would soon be stolen. rights reduces investment in human-made capital and in- creases deforestation. Forests would benefit from reforms With insecure property rights, incomes that strengthen the commitment to law and property. There and well-being collapse are good reasons to believe that benefits will extend to other Massive transfers of state property to private agents have oc- natural and environmental assets as well. But there will be ex- curred in the postcommunist economies in the past decade. In ceptions, as when stocks are naturally protected by exploration almost all of these countries, the private sector went from hav- or extraction costs. For petroleum reserves, exploration and ing a very small share of the economy in 1989 to having the extraction requires investment that is itself vulnerable to own- dominant share in 2000. But in many countries, these transfers ership risk. In these cases when more secure rights increase were not accompanied by the development of institutions that extraction or conversion, the likely effects will be more attrac- would make private property rights well defined and secure. tive, and not limited to reflecting short-term, narrow interests. A staggering 75 percent of firms in the Kyrgyz Republic, Moldova, the Russian Federation, and Ukraine were not “con- Source: Bohn and Deacon (2000); Hellman, Jones, and Kaufmann fident that the legal system will uphold my contract and prop- (2000); Hoff and Stiglitz (2002); EBRD (2002); World Bank staff. tresses need some protection, but those in banks Why would a politician or leader not take steps to need good vaults and such credible institutions as in- strengthen the judiciary and protect property rights? dependent regulatory agencies and central banks. Because a leader who takes steps to build stronger Higher national income can contribute to better institutions would reap benefits from the stronger institutional quality (for example, through more ex- economy and the better environment only in the penditures on courts). But more importantly, better long run, and this requires a stable setting with broad institutional quality can also contribute to higher na- political support.16 tional income, as when good institutions facilitate Valuable assets cannot thrive, and can be in waste- investments or curb overfishing. A large body of ful decline, if there is no social commitment to de- theoretical and empirical studies concludes that there veloping protective institutions (box 3.3). For agri- are strong causal effects from good institutions, mea- cultural land, two facts facilitate the emergence of sured by such variables as rule of law to higher in- protective institutions, as they may have come about comes per capita (figure 3.3). According to one study, historically. First, when land becomes scarce, com- better voice and accountability would raise national peting users cannot avoid meeting face to face, and income per capita by a factor of 2.5.14 Another study can choose to fight or negotiate. Second, when suggests that per capita incomes would grow at least neighbors are settled, they can help each other de- 2 percent per year in all countries if they would only fend their land in a reciprocal fashion. Therefore, protect property rights (public and private) and pur- even without a state or feudal lord, transgressors will sue more competitive market policies.15 face some deterrence.17 So when institutions allow assets to thrive, an To see how protective institutions form under more economy can flourish. But if good institutions are challenging circumstances, consider fisheries.18 Fish- so important, why don’t all countries have them? eries account for 19 percent of total human consump- (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  Figure 3.3 Box 3.3 The relationship between institutional quality Natural assets decline when protective and national income institutions are weak GDP per capita (PPP, 1999) Poor institutions contribute to a poor investment climate 50,000 and to the depletion of natural resources. It is also possi- 45,000 ble that some types of natural resources make institution- building difficult (chapter 7), while others support inclusive 40,000 and development-friendly institutions.* 35,000 Consider a sample of 150 countries for which there are 30,000 data (from the World Bank’s adjusted savings table) on 25,000 capital stocks, including human-made physical, mineral, forest, and agricultural capital. Adjusted savings are nega- 20,000 tive when this stock is declining. Of countries with low 15,000 human-made physical capital stocks (less than 75 percent 10,000 of national income), 24 percent (16 of 67 countries) had 5,000 negative adjusted savings, compared with 10 percent (8 of 83 countries) for countries with higher stocks of human- 0 made physical capital. –2.5 –2 –1.5 –1 –0.5 0 0.5 1 1.5 2 2.5 Thus, countries with high rates of depletion of natural Index of rule of law resources also tend to be countries with low stocks of physical capital. For countries with negative adjusted sav- PPP = Purchasing power parity. ings, this perspective shifts the focus away from recom- Note: As a measure of institutional quality, the rule of law includes considerations such as the security of property rights, or the mending policies to increase savings to recommending confidence with which a group or individual can find tomorrow the improvements in protective institutions. Their low accu- fruits of what is planted or conserved today. mulation of physical assets and the declining natural asset Source: Kaufman, Kraay, and Zoido-Lobaton (1999). base are more the result of an institutional environment in which assets cannot thrive, than of low savings. * The literature on natural resources as a curse makes this con- tion of animal protein, so good institutions for fish- nection, as do recent studies such as Acemoglu, Johnson, and eries are important.19 A typical story of fisheries de- Robinson (2001); Engerman and Sokoloff (1997); and Woolcok, velopment starts with the discovery of new fish stock. Pritchett, and Isham (2001). See also Auty (1997) and Hoff and Stiglitz (2002). As fishermen make good money, more capacity rushes Source: World Bank’s adjusted savings table. in. Though scarcity and wasteful competition ensue, fishermen do not coordinate management. Over time, the catch falls, first per hour of effort, then in total, with individual fishermen barely breaking even.20 The Law of the Sea, with 200-mile exclusive eco- This wasteful race to exploit the resource wastes labor nomic zones, reflects the idea that privatizing a re- and capital, since fish that should have been allowed source can allow for better management—it “priva- to spawn and grow are caught. Government involve- tizes� ownership to coastal states, suitably linking the ment can make matters worse, if subsidies distort sig- resource both to traditional interests and to a natural nals and prolong overfishing (chapter 7). jurisdiction for enforcement. While not all fisher- Contrast this story—of waste and no profit— ies’ management problems are confined to national with alternatives in which the value of the resource waters, nations can build on the law and negotiate. is maximized. A cooperative of fishermen could Many impressive fisheries management schemes manage a resource well if it could control its mem- would not have been possible without this law (chap- bers, if its members could control their organization, ter 7). In many other cases, however, management in- and if the cooperative could control entry into the stitutions have been only partly successful in limiting industry—all big ifs. In theory, a private monopoly entry and stabilizing catches, and as a result there has could also manage a fishing resource efficiently. not been enough reduction in excess capacity and More often in practice, government regulates by harvesting to bring major benefits. The challenge is denying access or enforcing restraint. Restraints that to win the industry’s confidence that it can reap the start with and build on existing resource users are benefits if capacity is reduced and profitability re- more likely to be credible and successful—as seen in stored. If the commitment to rewarding restraint is New Zealand and Iceland (see chapter 7). not credible, overfishing continues. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Picking up signals, balancing interests, A message from chapters 4 through 6 is that urban and implementing decisions as well as rural residents can have difficulty in being A good institutional environment must perform heard and served, and new institutional arrange- many functions. But to credibly and purposefully ments are evolving to overcome this problem (e.g., coordinate actions, three functions stand out—to boxes 6.4 and 6.5, respectively on favela residents in pick up signals, to balance different interests, and to Brazil and on railway station–dwellers in Mumbai). implement decisions. For marginal rural areas, two factors make receptive- ness to signals from the fringes more critical now Picking up signals than before. First, outmigration is less of an option Fisheries being depleted, toxins poisoning children, than it was when high-income countries were indus- corruption weakening emission testing, oppression trializing (chapter 4). Second, other developments, hobbling indigenous groups, violence against women such as mining, happen faster, so that signals about continuing—all these phenomena can be ignored their impact on the community and the environment willfully or accidentally in a setting that is not recep- need to move faster too (see box 4.7). tive to signals. Signals of social and environmental Many countries are rearranging the way they gov- degradation (chapter 2) can be based on scientific ern themselves at the local level. With political de- measurements (as in the case of air quality), or voice centralization, institutions develop to receive more and feedback, but they would be effective only if fine-grained signals—important, since problems and there are constituencies for information and action. priorities differ from place to place. But decentral- The ability to pick up signals is closely associated ization, proceeding in both rich and poor countries, with the ability to balance interests. Creating and re- carries promises as well as risks. It can be hard to get ceiving signals range from the feasibility of detecting a the incentives right, and there may be issues of weak phenomenon in a meaningful way, to the process of institutional capacity and elite capture at lower lev- aggregating signals and getting the attention of deci- els of government as well.22 sion makers. Receptivity to signals thus depends on social and political relations. Are they open and in- Balancing alternatives—and interests clusive, or fragmented and discriminatory? Are they A protective institution such as an air quality protec- pluralistic, meritocratic, and free, or politicized and tion agency operates within an authorizing frame- monolithic? Is there freedom and competition in indi- work that balances interests. It may impose an emis- vidual expression, business, and political organization? sion standard or a tax, or strengthen enforcement. Both citizens and the air-quality protection agency This in effect strengthens the rights of beneficiaries need good information on pollution, and this infor- to air quality—weakening rights of others, and this mation has many uses. Mexico City’s IMECA index, is one of the ways changing social priorities can be published daily in the press, advises people whether implemented. The balancing of interests takes place to keep their children indoors and avoid exercise. at many levels: in national legislatures, in court pro- And it obviously informs citizens on how well the cesses, in marketplaces, in individual norms and vil- agency is doing in improving air quality, strengthen- lage interactions, in the seen or unseen processes in ing their hand in holding politicians and agencies corporate boardrooms and branches. accountable. Evidence shows how information provision (in an But there are also other signals. Voice describes the era of unprecedented quantities of information) can signals from citizens, firms, and civil society to influ- catalyze shifts in political balances and real world de- ence institutions (i.e., through complaints, votes, cisions. Thus, the term “transparency� dominates the court proceedings, and the media). A lesson from re- current campaign for better governance. There is ev- cent research in developing and developed countries idence from rich and poor countries that greater is that influence—including good-natured influence availability of information means better environ- from citizens to the emission reductions by firms— mental performance.23 can travel through many channels, strengthened by Figure 3.4 shows the 1,445 cities in the world— information.21 where, according to World Bank estimates—the pop- (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  Figure 3.4 Concentration of total suspended particles Total suspended particles (TSP) by city <90 91–200 >200 Source: Bolt and others (forthcoming). ulation suffers from exposure to concentration of dust approach to complaints, and national policies to particles, or total suspended particles (TSP) above make local environmental information available to traditional guidelines of 90 micrograms µ/m3.24 In the public (chapter 7). A complaints-driven system less than 2 percent of these cities is air pollution sys- has many qualities, but may be biased toward im- tematically monitored, and in even fewer is informa- mediately noticeable phenomena, such as noise. So tion about that pollution made available to the pub- the combination with objective monitoring data is lic. A good working hypothesis—based on studies valuable. Both mechanisms utilize the strengths of a from industrial and developing countries—is that decentralized system and recognize how national in- monitoring pollution information and making it stitutions (such as assurance of information) can be public would help (in part through political chan- important for local accountability.26 nels) to improve air quality in these cities regardless In Europe information that helped establish a of their level of income.25 common understanding of who suffers from a prob- But balance is not maintained by information lem and who contributes to the problem was essen- provision alone. More open and democratic coun- tial in shifting the balance in favor of reducing trans- tries presumably give weight to dispersed interests, boundary pollution, even when negotiation—not so there are reasons to expect that they would give authority—did the balancing (see box 8.1). For bal- more attention to environmental protection. The ance and for unbiased signals, it is essential to have data are incomplete, but some results support this supportive institutions in place. Corporations rely hypothesis (box 3.4). on laws and auditors for such traditional goals as One might expect a one-party system to have dif- protecting workers’ pensions and shareholders’ as- ficulty being receptive to signals, since unpleasant sets—and they now rely also on civil society for news might be easier to suppress in such a setting. broader corporate social responsibility (chapter 8). But this is not always the case. In China two in- stitutional features play a role in areas such as en- Executing decisions vironmental management, where there has been As an environmental agency executes decisions noticeable progress in the last decade: a systematic through taxes, regulations, and enforcement, citizens (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 3.4 Democracy and environmental policy: picking up signals, shifting the balance There are strong theoretical reasons to think of democracy of powers are best envisaged in a democratic setting; these as conducive to environmental protection and economic effi- institutional features are also the key ingredients in putting ciency in general. new priorities on the table, rebalancing competing interests, Two plausible mechanisms can be observed: and taking action. Ⅲ Democracy helps give weight to dispersed interests. In gen- It is not easy to accurately measure environmental commit- eral, policies will be biased in favor of concentrated interests, ment. Even so, democracies have a greater tendency to do the giving less weight to equally important interests spread following: across a larger number of people. Benefits from environmen- Ⅲ Put their land area under protection tal assets, such as from river water quality, are often consid- Ⅲ Sign and ratify multilateral environmental agreements ered public goods and are thus dispersed across many indi- Ⅲ Belong to environmental intergovernmental organizations viduals, while the cleanup costs may be more concentrated. Ⅲ Meet reporting requirements for the Convention on Inter- Ⅲ Freedom of expression and association helps society pick up national Trade in Endangered Species of Fauna and Flora signals and adjust to change. As population density in- Ⅲ Have a National Council on Sustainable Development creases, knowledge increases, incomes grow, or prefer- Ⅲ Have environmentally relevant information publicly available. ences change, the pressures on the environment change. As the problem of horse manure in London’s streets de- The study found that democracies are more likely to make clined, new problems—such as smog or lead contamination an environmental commitment, regardless of their level of in city air or oil spills in the North Sea—emerged to beg for income. new management institutions and technical solutions. The accountability of politicians to the people and the separation Source: Neumayer, Gleditch, and Gates, background paper for the WDR 2003. and firms change their pollution emissions. The It is not sufficient for society—or a development agency will monitor these emissions and act on them, bank—to make a policy decision. Say society decides either by adjusting its charges for the emissions or by that forests should not be converted if the conserva- assessing penalties for noncompliance with regula- tion benefits are higher than the conversion benefits. tions. Costa Rica’s program to pay for environmental The implementation of this decision can be blunted services program (see box 8.5) is an example where by developers who press ahead and convert a forest, institutions shift the burden of protection, help to asserting that the conservation benefits are minor. balance interests, and ensure better execution. There The developers count on escaping sanctions—even are many other examples, often integrated with de- if the losses turn out to be high—if society is known velopment projects, such as the Global Environmen- to lack the incentives, opportunity, or commitment tal Facility (GEF) and the Prototype Carbon Fund to punish or undo wrongdoing. (chapter 8). Box 3.5 discusses how locally negotiated How then does one ensure that policies are imple- solutions assisted in the implementation of water pol- mented? Good procedures and broad participation lution reductions in Colombia. can help in the execution of high-level decisions. Pro- Implementation is an extension of balancing. cedures requiring ex ante assessments, participation, When the balancing is between suppliers and cus- and public reviews can help. Routine social and en- tomers in a marketplace, the balancing and imple- vironmental impact assessments, enforced with good mentation functions may be one and the same (see quality information and public access to them, can box 3.1). Balanced decisionmaking in board rooms expose consequences before development is irrevers- and national legislatures is not worth much if it is ible. This can make it clear—to the public, to politi- not implemented—or if the steps from policy to im- cal leaders, to courts, and to civil society—that the plementation are too far apart. In many countries, proposed developments do not comply with broader laws and protection through the courts are of little social priorities. The information—and the support- value because they are implemented by ineffective ive institutions—function as a commitment device. or corrupt courts—unless one has connections or In Uzbekistan, as part of the Uzbekistan Water money. In others, budget deliberations are not worth Supply, Sanitation, and Health Project, a social assess- much because the budget is not followed. ment process was undertaken during project prepara- (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  Box 3.5 Local negotiations balance interests and commit parties to clean up Colombia’s rivers In Colombia, as in many countries, most wastewater is re- region is allowed to set goals and timetables to reflect local leased untreated into waterways. With little enforcement, lim- conditions and aspirations. Firms can choose emission re- its on pollution emissions have long been ignored. In 1997, the ductions—and method—in light of per unit charges. environment ministry implemented a new water pollution Ⅲ Devise innovative approaches to program administration. A charge system that is cost-effective and enforceable. Facili- well-respected private bank collects the charges and admin- tated under Colombia’s decentralized structure, the system is isters the funds in return for a percentage of the revenues, implemented by regional environmental agencies. It brings to- reducing the burden of collection but not of auditing by gov- gether municipal authorities, polluting industries, and affected ernment agencies. communities to negotiate local pollution reduction targets and Ⅲ Enhance the community benefits of market-based regula- charges. Polluters are charged per unit of effluent, and the par- tory tools. Local business leaders were initially skeptical, ties agree to timetables for increasing the charges if targets perceiving the program as a new generalized tax burden. are not met. When it was agreed that revenues would fund monitorable All the parties have received extensive capacity building benefits, such as local pollution reduction, this appealed to from the national ministry, and the system holds together im- businesses and communities alike, and helped generate pressively: In the nation’s 135 river basins, biochemical oxy- commitment to implementation. gen demand is already down by 31.5 percent, and suspended solids by 34.2 percent. Nationally the program has generated Future progress will require greater compliance from recal- $9.7 million in revenues, funding pollution reduction projects citrant sectors, such as municipal water companies, who use and regional environmental agencies. various pretexts to avoid paying and investing. If those who do Lessons include the following: not comply are seen to gain, it could threaten the more gen- eral commitment among polluters, a commitment that has Ⅲ Use national commitment to facilitate locally negotiated so- proven to be a strength of negotiated approaches. lutions. Regulated sectors participate because authorities have signaled their intent to enforce the program. But each Sources: World Bank 2000d; Andean Center 2001. tion. The government initially wanted to ensure that sions are not always sharply distinguished. For pri- drinking water would have no more than one gram vate goods traded in markets, the market itself pro- of salinity per liter, although international guidelines duces signals on scarcity—future and present, likely allowed higher salinity. The lower salinity level would and plausible—while balancing needs and executing have been costly and there is no known evidence that trades. But transaction between agents in a market it would be healthier. So with the help of local social can affect others (as with pollution). In this case, sup- scientists, a taste tolerance survey was carried out. It plementary institutions (an air protection agency) found that salinity levels of up to two grams per liter can perform these functions and represent the inter- were socially acceptable. The findings from the ex ests of those affected (box 3.6). ante assessment were accepted by the Uzbek govern- Overcoming barriers to coordination ment, and consequent design changes freed up about $15 million dollars. Parts of the savings expanded the Institutions assist in coordination, but encouraging project’s geographical scope and resulted in additional the emergence of good institutions is itself a coordi- pilot projects.27 nation problem. Three barriers to building and sup- Many countries have a gap between the policy porting institutions are important: decision to teach children with public funds, and the implementation of that decision—to make sure Ⅲ Organizing dispersed interests—it takes more for learning is effective. Studies from Argentina, El Sal- people to come together and build a school than vador, and Nicaragua show that empowering parents it takes for them to build individual homes. (through participation on school boards, for instance) Ⅲ Forging credible commitments—it takes more to can improve the delivery of educational services.28 commit to helping each other in the future and over a longer period than to helping each other Feedback—by and for institutions today. The three functions discussed above of picking up Ⅲ Promoting greater inclusiveness—it takes more to signals, balancing interests, and implementing deci- coordinate for the common good when access (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 3.6 Policy accountability and accountable rulemaking Consider an air-quality protection agency charged with balancing concerns for well-being. It could also give citizens access to the need to pollute with the need for better air quality (see box information and the right to sue polluters (or the agency), or it figure a). It receives signals on pollution levels, damages, emis- could give the agency more powerful tax instruments. All sions, and preferences. It balances the interests of different these examples would shift property rights to the air toward stakeholders. And it implements policies through taxes and reg- the beneficiaries of better air quality. ulations, and through monitoring, enforcement, and publicity. Authorizing the rulemakers Accountable rulemaking The legislature is a permanent rule-making institution, but it is To handle new and evolving problems, the institutional en- also a list of people who are there temporarily (changing with vironment must be adaptive, even though institutions also elections). The legislative process and the individual legislators need to be stable and durable. The needs of coordination will earn their authority and powers from the authorizing environ- change, and so must institutions. When the ozone layer was ment, exemplified by values, religions, and constitutions (box threatened, nations coordinated their actions by creating a new figure b, the outer loop). In a strong authorizing environment— institution, the Montreal Protocol, to protect it (chapter 8). And one giving strength to rules—the rules for rulemaking are fairly most countries now have rules to keep lead out of gasoline stable. since scientific evidence on its deleterious effects has accu- A key element in a strong authorizing environment can mulated and is widely available (chapter 7). be a constitution that is respected, and likely short, which con- If the air is poorly managed, institutional analysis will likely tains core principles only. If this is the case, when the con- find a weakness in the institution charged with adopting or im- stitution is tested against legislation, the executive, or the plementing the policy to protect the air. For example, air qual- judiciary, the constitution typically stands unchallenged and un- ity interests may not have the information or voice they need changed. But more important than a constitution is a set of to counter concentrated polluting interests. In these cases, widely held values, so that support for sudden radical change monitoring and transparency can help shift the effective bal- is unlikely. Even knowing that the world is changing and that ance toward air quality. But maybe there is no air protection tomorrow’s rules have many uncertain elements, some deeper institution—or it is ill equipped or obstructed. values and widely held principles can provide stability and lend This brings the institutional analysis to a deeper level—to strength to the rulemaking. Countries with little commitment accountable rulemaking. At this level, rules will change and in the authorizing environment are more vulnerable to distribu- new institutions will emerge—if the overall institutional envi- tional conflict, as reflected in macroeconomic instability, lack ronment is picking up signals, balancing interests, and imple- of confidence, and credibility. Thus, the great penalty of low- menting policies. For rulemaking in an electoral democracy, quality institutions comes in the longer run, as is confirmed in elections are a powerful channel for signals to the legislature— empirical studies. the rulemaking process (box figure b, the intermediate loop). And when the legislature has done its balancing, it implements Sources: See Williamson (2000) for such a staggered hierarchy in insti- solutions by authorizing a new (or strengthened) air quality pro- tutional analysis. Rodrik (1996) emphasizes the role of institution in tection institution. For example, it could authorize the agency overcoming districutional conflict and negotiating reform. See also Ace- to justify its interventions by health benefits or more general moglu and others (2002). (a) Policy accountability (b) Accountable rulemaking AIR-QUALITY VALUES, Execute PROTECTION CONSTITUTION (authorize legislature) AGENCY ᮢ Balance ᮢ ᮢ interests LEGISLATURE Execute ᮢ Signals (establish rules (new legislature) ᮢ and agency) Signals Execute policies AIR-QUALITY ᮢ (pollution, (taxes, Signals PROTECTION preferences, regulation, (votes) AGENCY damage) enforcement) Signals Execute policies (pollution, (taxes, regulation, preferences, enforcement) damage) ᮢ ᮢ STAKEHOLDERS: FIRMS, CITIZENS STAKEHOLDERS Source: World Bank staff. (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  to assets and voice are exclusive, unequal, and un- Similarly, a government might want to protect an democratic. area from settlement, but it will have difficulty doing so if many settlers can arrive and dig in before the Organizing dispersed interests government gets there. A prior action (assigning and The brokering of political forces is often biased paying for suitable guards to protect the area) can against dispersed interests. A trade regime may dis- serve as a commitment device. For example, the play protection even though many may lose from it, United States posted the U.S. army to protect Yel- since the interests of those benefiting from it are lowstone National Park against incursions before the more concentrated and thus more easily organized. park service was established.31 Or policies may show an urban bias because the rural When assets can be degraded or destroyed without population is less vocal and has more difficulty or- risk of sanction, they are more likely to be depleted, ganizing itself. Or the civil service may be overstaffed often in a wasteful fashion. But the cultivator settling or overpaid because civil servants have a good grip beyond the agricultural frontier traditionally has not on the policy process. asked for permission. At low levels of human impact Institutions face challenges in organizing dis- on the environment, anyone can use something that persed interests even if the counter-interests are not nobody else uses, so this regime makes sense. It allows concentrated. Recall Mexico City: 20 million peo- institutions to advance geographically at a pace with ple benefit from air quality improvements, but there (or slightly behind) settlements, which is both eco- are also millions of polluters (vehicles, households, nomic and realistic.32 and firms). The collective action to generate institu- But there may also be times when it is important tions for air quality improvements may represent a to get ahead of the advancing frontier of settlement, tremendous challenge even when the interests of pol- either because there is scarcely any forest, land, or luters are dispersed as well. water left, or because selective preservation has be- Protecting air quality requires coordination be- come a priority (chapter 5). In many developing yond what the market would accomplish unassisted. countries the area under agricultural cultivation will But how is this coordination accomplished? In many stop expanding or may even reverse in the coming contexts a government intervenes with taxes and reg- 30 to 50 years, so selective preservation now can ulations. But in many others, coordination is accom- yield lasting benefits at modest cost. As incomes rise, plished without state intervention. Communal graz- a time arrives when the forest has value not only to ing grounds and irrigation systems are managed, a potential cultivator (or those who depend on it as sometimes well, by village norms and councils.29 an ecosystem for their livelihood) but also to citizens What is the process that enables such coordination? far away. If these interests can find a suitable chan- Freedom of expression and association, trust, and nel for expressing their interests the first-come, first- political accountability provide some institutional serve regime can be taken off the table selectively. machinery to coordinate dispersed interests, both in It is often difficult, however, to fight the tendency picking up signals and in giving them balance. Dem- to reward early movers. When fish quota systems are ocratic institutions and the popular vote, despite put in place, the race for quotas can become a race many weaknesses, lower the costs of coordinating to the shipyards, or to government offices—races dispersed interests (see boxes 3.4 and 3.8). that can be just as wasteful as the race to fish the de- clining stock (chapter 7). Preventing waste in envi- Forging credible commitments ronmental matters and natural resources requires Society may have an interest in not polluting the confronting those who choose to deplete natural as- water, and it may even prohibit excess discharges in sets with the full social consequences of their actions, the water. But if the water is already polluted, ag- whether through informal mechanisms, taxes, or reg- grieved parties may find it difficult or impossible, for ulations. Similarly, loggers and fishermen must be political or other reasons, to punish polluters.30 Such induced to act as if they faced the consequences of failures to deliver on commitments encourage trans- others who might use the forest or the fish. gressions by those who are well positioned, and are When commitments to protect are not credible, detrimental to accumulating and protecting assets. it is highly profitable to move early to manipulate (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  the (supposedly) protective institutions. Tradable helps, as when participatory procedures ensure that conservation obligations, such as those in Brazil consequences are known beforehand. Box 3.7 draws (chapter 8), reduce the costs of compliance, and are the parallel between commitment by institutions thus one way to make conservation commitments protecting natural assets and by those protecting more credible. As discussed above, information also people’s savings in a modern financial market. Box 3.7 When protective institutions fail: the collapse of Enron and Newfoundland’s cod fisheries Like a forest that has been logged behind a nice facade of someone else’s tomorrow, and a frenzy of harvesting may end trees and billboards, the Enron corporation crumbled and filed in collapse. Collapses are rare, however. It is more common for bankruptcy in 2001. At $60 billion in market value before it that weak institutions result in steady and sustained losses, perished, the darling of financial markets became one of the for both natural as well as produced assets. For instance, many largest bankruptcies in U.S. history. The assets had disap- fisheries steadily yield zero profits, and many managers and peared over time, shielded by misrepresentation. workers in overstaffed firms and public agencies add little Even in a private corporation such as Enron, the dispersed value for their pay. interests of thousands of owners are potentially threatened by A system of protective institutions lies behind the success well-placed individuals. Just as a contracted logger can either of mobilizing savings through financial institutions and stock log sustainably or wastefully, managers and auditors can—by markets, undoubtedly one of the great achievements of the legal or illegal means—serve their own interests rather than United States. Such a system rests on checks and balances those of dispersed owners. Norms and culture, and of course and introduces independent veto players, many of whom must protective institutions (laws about property, accounting, and look the other way for improprieties to occur. But Enron offi- auditing) protect dispersed interests, but the protection will cials commanded impressive contacts and influence. Though never be perfect. The Enron case is still in its fact-finding phase, the potential watchdogs were many, Enron went down with- but the list of techniques used by well-placed officials covers out a peep from uneasy auditors and with enthusiastic “buy� what one would expect in any other public sector or private recommendations from the world’s best paid stock analysts. sector institution that has been corrupted: concealed, misrep- For the United States, the Enron incident is a sobering one resented, and ghost transactions aimed at the enrichment of but not earth-shattering, and it has not eliminated the credibil- a few. ity of the protective institutions. Valuable assets—be they Why, one might ask, if well-placed individuals can abuse an trees and fish, or people’s savings—need protective institu- asset by taking out more than their assigned incomes, would tions to thrive. When those protective institutions are success- they overdo it, killing the golden goose? They may of course ful, trees and fish and air quality and savings will thrive—to the have lost or overtaxed the asset by accident. However, the great benefit of widely dispersed interests. But as assets path of Enron’s stock market value looks suspiciously similar grow, pressures on institutions grow—testing their commit- to that of another failure of protective institutions, the collapse ment. Protective institutions are essential for people to make of Newfoundland’s cod fishery, hinting at how failing protec- their savings available for banks and business people. The real tive institutions lead to wasteful races. cost of weak institutions is not that they result in more fraud Newfoundland’s rich cod catches were growing slowly but and theft, but that, as a result, most people are not forthcom- steadily for about 100 years, until they increased steeply in the ing with their savings. Potential is left unrealized. Ownership 1960s and 1970s, and then collapsed (chapter 7). In fisheries, is concentrated and business is constrained by lack of depth as technology and equipment develop, the stock of fish in the and little competition—serving everybody poorly. sea starts declining. And if protective institutions fail to curtail overfishing, a wasteful race among fishermen follows. Fisher- Source: CRSP, University of Chicago, original data from New York Stock men understand that the fish they do not catch today may be Exchange; The Economist, February 14, 2002; Hannesson, background paper for the WDR 2003. Enron, market value, 1985–2002 Newfoundland cod catch, tons, 1850–2000 (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  Promoting inclusiveness this—in other words, voice and participation are im- The third barrier to coordination mentioned in the portant. A society in which the majority has no voice previous section is fundamental enough to deserve can lose out big in two ways. First, it can lose because expanded discussion. While it is well recognized that the potential creativity and productivity resting in the quality of institutions affects the management of the majority of the people is ignored or valued only assets, it is less well recognized that the distribution in part. Second, because beneficiaries to communal of assets and voice affects the evolution of institu- and natural assets are not heard, the potential of tions in the long term (see figure, roadmap). Poorly these assets may be wasted, too. Institutions such as distributed assets can affect adversely the quality the law necessarily involve coercive powers, and one of institutions and their ability to solve problems. of the potential benefits of broad-based voice—an Because of this reciprocal relationship between insti- inclusive democracy—is that it better commits these tutions and the distribution of assets one can get powers to serving society at large. Box 3.8 relates the locked into vicious or virtuous circles. These circles remarkable and very promising transformation in are not deterministic, but extra effort is needed to South Africa toward a more inclusive society. break out of a vicious circle. This is easier when But even very basic protective measures, such as greater inclusiveness in access to assets is assured shielding families and savings from abuse and theft, from new additions to the asset base, such as with often fail to materialize. Poor people have to accept broad investment in primary education. very costly outlets for their savings, as when they buy gold, are hurt by inflation, or must pay others for The importance of voice and participation safekeeping.34 And the police and courts, responsi- A community that wants to improve air quality—or ble for enforcing the law, often fail to assist or ade- protect trees—may or may not find a channel to ex- quately protect poor or disenfranchised groups. In- press its interests.33 Individuals in the community deed, many institutional development initiatives are could form an association and negotiate with pol- geared to making police and judges more attentive luters. Or they could lean on government to do to the needs of the poor and disenfranchised.35 Box 3.8 Fostering inclusiveness: South Africa’s new democracy One of the more remarkable examples of institutional transfor- of nonviolent protest and resorted to armed struggle. Mass mation toward an inclusive society is South Africa’s transition demonstrations and violence continued throughout the 1980s from a system of white rule to a pluralist democracy founded and early 1990s. The transition to a new path in South Africa on the principles of human rights and reconciliation. When na- began with a series of conversations, initiated by Nelson Man- tional elections were held in 1994, black South Africans, com- dela from his prison cell, involving the ANC and National Party prising three-quarters of country’s population, were able to leaders in the mid-1980s. At the same time, work stoppages vote for the first time. In addition, they were able to exercise and uprisings in overcrowded urban slums were exacting a toll long-denied rights to travel freely and to live and work where on the country and prompting the flight of skilled workers. they please. This transformation has required uprooting the en- Western nations that had long supported the apartheid regime trenched institutional foundations of apartheid and creating a became more vocal in their criticism, and eventually tightened host of new and more inclusive institutions—from the 1996 economic sanctions. Following the collapse of the Soviet Constitution to reformed security agencies, provincial govern- Union, the Cold War logic of support for the staunchly anti- ments, and health and education ministries. How did South communist regime was also abandoned. Africa manage this transition? Leadership and a commitment to reconciliation From violence to negotiations These developments alone could not ensure a successful tran- Under apartheid, legally sanctioned discrimination backed by sition from apartheid to a new set of institutional rules. The violence permeated every aspect of society, as blacks were process depended on the vision and skills of political leaders. denied the most basic liberties and were the victims of wide- Mandela went to great lengths to reassure his supporters that spread human rights abuses. In 1961 Nelson Mandela and the he would not act without the consent of the rest of the ANC. African National Congress (ANC) abandoned their strategy This earned him the trust and respect of his followers, a re- (Box continues on next page). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 3.8 (continued) source he drew on when it came time to reach difficult com- The transition to democratic rule promises with the National Party. Leadership was also demon- These and other measures facilitated the transition to demo- strated by President de Klerk, whose decision in 1990 to free cratic rule while preventing capital flight and preserving valu- Mandela and lift the ban on major black political organizations able social assets such as the skills and expertise of the mostly involved great risk. white civil service. A commitment to human rights and recon- Mandela and other ANC leaders were adept at combin- ciliation provided the winning formula that ensured the suc- ing tough negotiation with a strong public commitment to na- cess of the transition despite efforts to derail the process. tional reconciliation. This allayed the worst fears of National South Africa’s transformation to an inclusive society is an Party leaders and facilitated compromise. The mechanisms example to the world—-but not because it was timely, smooth, of reconciliation included broadly participatory negotiations or bloodless; it was none of those. But South Africa underwent over the new constitution, and a government of national unity the most difficult of institutional transformations in the most that gave former rivals the experience of governing side by trying of circumstances, and did so while fostering a political side. The Truth and Reconciliation Commission, established in culture that emphasizes human rights and reconciliation. Many 1995 under Archbishop Desmond Tutu, provided a high-profile challenges of governance and development remain, and aspi- confidence-building mechanism for addressing past atrocities rations will continue to create tremendous pressures for while restraining fears and pressures for retribution. And the change—a positive force but also a challenge to institutions new constitution devolved significant powers to the provinces, and leaders. which further allayed white concerns over majority rule at the national level. Source: Sparks (1996); personal communication with World Bank staff. Protecting people—and the emergence of protective dians, to look over something of value. An example institutions for assets is when participatory approaches in projects ask peo- When institutions such as the law, and the agencies ple to speak their mind. The presumption behind supporting the law, become more inclusive, more this empowering people’s voice is not only that peo- people are given protection, voice, and support. And ple have a right to speak on their own behalf. It is when institutions are more inclusive—listening to and supporting more people—a broader range of Box 3.9 assets can thrive (boxes 3.4, 3.8, 3.9, and 3.10 illumi- Mutual reinforcement: environmental movements and democracy nate different aspects of this). The reason is that assets need guardians and spokespersons. Assets therefore In many places environmental movements arose in the may fail to be served if the people who benefit from 1980s in the midst of broader social movements for de- these assets are not well served by—or represented mocratization. Democratization and environmentalism have in—institutions. For private assets, more inclusive in- developed together but in diverse ways. In the Republic of stitutions facilitate development and asset accumula- Korea social movements for democratization, labor, and environmental protection joined forces in opposition to tion as more people feel safe in their homes and find authoritarian rule in the 1980s. In Taiwan (China) the envi- promising outlets for their savings. For communal ronmental and prodemocracy movements were the two and natural assets (roads, water, fish, or forests), more strongest social mobilizations. An estimated 582 environ- inclusive institutions deepen the support for their mental protests occurred there between 1983 and 1988— one-fifth of public protests during this period. In Brazil dis- provision, so that their quality and quantity can rise. parate environmental organizations that had kept a low Consider what happened in Cubatão, Brazil, where profile during military rule were animated and united when inclusiveness in the form of democratization and the they helped draft the environmental chapter of the new na- tional constitution during 1985–88. In the former Soviet end of media censorship shifted the balance toward Union, civic environmental organizations flourished in the civil society and a cleaner environment (box 6.8). In early years, were crushed under Stalin, resurfaced in lim- many countries, social movements pressing for de- ited form during the political liberalization of the 1950s, and mocratization and environmental improvements have exploded as a central component of mass movements for democratization in the late 1980s. reinforced each other (box 3.9). How can protective institutions be formed to Source: Mirovitskaya (1998); Anbarasan and Yul (2001); Lee and give dispersed interests effective channels? In a wide others (1999); Hochstetler (1997). range of cases, society relies on guardians, or custo- (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  also that city people, for example, can benefit from with low yields only because she cannot finance mi- hearing from people in more remote areas about gration or a job search (chapter 4).36 Another exam- what goes on in the forest, about the effects of cut- ple is when agricultural potential is wasted because ting trees or damming rivers. of distorted and highly concentrated land-ownership For people to be functional guardians, they must and contracting problems.37 be well-endowed and feel safe. As an illustration, all Inequality in land assets has been found to be societies rely on parents to protect and nurture chil- harmful to growth.38 Good institutions appear to fa- dren. It happens that this protection fails—as when cilitate long-term growth, and more egalitarian soci- children are sold into slavery or prostitution. This is eties appear to have better institutions. Furthermore, not because parents are not their guardians—they good institutions work in part directly, and in part are—but because of the family’s poverty and despair. through schooling and openness.39 This need to have well-endowed guardians places An ambitious quantitative study tested the role of broad-based development and poverty reduction at inclusiveness and institutions using a 500-year per- the heart of concerns for environmental and other spective.40 For colonized countries (not limited to communal assets. More inclusive access to assets the Americas), a major break in power structures and (human capital, a piece of land, or a plot for hous- institutions happened under colonization. Those ing) can change people’s perspectives, allowing them that were richer and more densely populated in 1500 to be more forward looking and engaged in their (before colonization) are poorer now. This reversal communities. When people have assets—and thus a of fortune came about because colonizers in richer, stake in the future and in the community—it is also more densely populated areas could force a large easier to build support for institutions, public goods, supply of labor to work in mines or plantations. and publicly provided goods such as rule of law, wa- Under these extractive institutions, political power tershed management, and schooling. was more concentrated. The lower quality of these institutions for growth reveals itself after 1700, when What does inclusiveness in access to assets have to asset creation became important, rewarding coun- do with sustainable development? tries that had institutions better suited for savings In important ways, high levels of inequality and dep- and investment. rivation can be harmful to efficiency and growth. The Long-lasting harmful effects of institutions that presence or absence of inclusiveness in institutions concentrate ownership are also found in a recent study and in access to assets tends to have long-lasting ef- of India.41 The British colonization in India lasted fects (box 3.10). for 200 years. Where they implemented a landlord- Economic forces that create differences in income based revenue system (by implication, with concen- and wealth serve a positive function by creating in- trated property rights), yields were higher than in centives to allocate resources efficiently. But poverty areas where they implemented cultivator-owned and inequality can be harmful through other impor- rights. In post-independence India, the landlord- tant mechanisms; at the macro level, damage can be based revenue system was abolished, so only the his- done in the political process. Institutions and gov- toric traces of the institutions remain. Yields have ernment policies are essential for assets to thrive, grown significantly faster in the areas where, histori- through the rule of law and macroeconomic stabil- cally, cultivators themselves had property rights. The ity. An equitable distribution can facilitate the emer- differences prove to be particularly important from gence of institutions to negotiate change and thus the 1960s onward, as districts with smallholders ben- help adopt and implement good policies—particu- efited more from the green revolution, with signifi- larly to address externalities and public goods. At the cantly higher application of fertilizer, high-yield vari- micro level, sharp differences in income and wealth eties, and irrigation. Districts that historically had are also costly. Imperfections such as those found in smallholder institutions also had higher investments capital markets may allow individual potential to be in human capital. wasted when individuals are very poor. Examples are The proposition that ownership matters is sup- when a talented child goes uneducated (chapter 7), ported also in other studies. Before 1977 sharecrop- or a worker stays with an employer, or with an asset ping contracts in West Bengal, India, generally in- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 3.10 Inequality: its long tails in the Americas Many of the former European colonies that offered the best sharply from the rest of the hemisphere. Why did the areas pre- economic prospects in early colonial times (based on their na- viously favored fall behind? Development depends not just on tional resource endowments) are today among the poorest in having productive opportunities—it depends on creating a the world. They started to fall behind at the outset of the In- never-ending supply of new opportunities. One key to early in- dustrial Revolution. In 1700, Mexico and the colonies that were dustrialization was the ability of the broad population to invest, to become the United States had very similar per capita in- accumulate human capital, and participate in commercial activ- comes, and the sugar-producing islands of Barbados and Cuba ity. In the Americas, only the United States and Canada pro- were far richer. Indeed, before the 19th century, the North vided the laws, institutions, and government policies to make American mainland was widely considered to offer poorer eco- such investment and participation possible. nomic prospects than the Caribbean and Latin America. All of In the New World colonies of Spanish America, the core Canada, which Voltaire once characterized as “a few acres of natural resources—high-yield ores and agricultural land—were snow,� was considered by the colonial powers to have a value susceptible to large-scale operations. This made possible a comparable to that of the small sugar-producing island of high inequality in income, wealth, and human capital at the be- Guadeloupe. ginning of colonization. This inequality had a great influence on the evolution of institutions. In particular, the institutions that The rapid rise of national income per capita in the emerged in these colonies blocked effective access to oppor- United States and Canada after 1800 tunities for economic and social advancement for a broad cross-section of the population. This persisted long after colo- nization ended and slavery was abolished. These institutions GDP per capita relative to the U.S. (percent) inhibited the accumulation of human capital, the spread of en- Economy 1700 1800 1900 2000 trepreneurship, and the creation of a mass market—factors Argentina — 102 52 36 viewed as important in industrial development. Barbados 150 — — 44 Why should inequality hundreds of years ago matter for de- Brazil — 50 10 22 velopment today? As just noted, societies that had high in- Chile — 46 38 28 equality in the ownership of assets at the outset generated Cuba 167 112 — — institutions that placed restrictions on individuals’ opportuni- Mexico 89 50 35 26 ties for future economic advancement, and this may have tied Peru — 41 20 14 these economies into low growth paths. Canada — — 67 82 Note: Klenow and Rodriguez-Clare (1997) attribute only 3 percent of the United States variation of growth per worker across countries to variations in the (GDP per capita in growth of capital per worker, while variations in technical progress ac- 1985 dollars) 550 807 3,859 34,260 counted for 91 percent. Other studies conclude differently, in part be- — Not available. cause technical progress and capital accumulation move together. But Source: World Bank (2001); Engerman, Haber, and Sokoloff (2000). there are few studies that place major emphasis positively on initial endowments. In chapter 7, the “natural resource curse�—the idea that certain natural resources can be harmful to growth—is discussed in more detail. Once industrialization began in North America in the 19th Sources: Engerman and Sokoloff (1997, 2001); and Hoff, background century, the United States and Canadian economies diverged paper for WDR 2003. volved 50 percent output shares to the tenant for the At early stages of development, owners of land approximately 2 million sharecroppers in the state. may benefit from booms more than others (so ensur- In 1977 a new administration gave high priority to a ing broad-based land ownership and smallholder law giving security of tenure to tenants. The reform agriculture is likely to be more effective in reducing increased most tenants’ share of output from 50 per- poverty—see chapter 5).43 In a similar way, owners cent to 75 percent. In the decade after this reform, may be best positioned to benefit when a commu- West Bengal broke through: Annual growth in the nity performs better, as when schools and roads are production of food grains rose from 0.4 percent in good repair. For this reason, having narrowly based to 5.1 percent, while that for all of India rose only ownership and many citizens without land or secure from 1.9 percent to 3.1 percent. The tenancy reform tenure can be an impediment to forward-looking and program explains about 30 percent of the added constructive collective action, whether for environ- growth.42 Tenancy reform in urban slums in Brazil mental protection or for other purposes.44 seem to also have unleashed growth potential and im- The cited studies show how greater inclusiveness provements in the urban environment (see box 6.6). in access to assets—or lower inequality—can assist in (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  making development more sustainable. One mecha- The studies cited in this chapter represent, but do nism is direct and micro-economic: Ownership mat- not exhaust, a still-young literature on the deeper ters, and access to assets can help a poor family real- institutional preconditions for economic growth. Im- ize its potential. Another is political: A person with portant questions are whether institutions are ev- land or a house is more likely to support institutions erything, whether policies—in part determined by in- protecting assets (rule of law and secure property stitutions—have separate and important effects, and rights, for instance) than a person without a house or finally whether high inequality itself is a major and hope of having one. So insecure property rights— important obstacle to sound institutions. At a practi- with costly policy swings where shifting groups ex- cal level, there are many points of agreement: A key propriate each other’s assets—are less likely if access element in the success of East Asian economies was to assets is broad and inclusive. a focus on shared growth, inclusive schooling, and In figure 3.3, the observation was made that pro- how this served to give political stability and investor tective institutions—such as the rule of law—are confidence (box 7.10, Malaysia). An important ele- typically stronger in high-income countries. It was ment in political discourse in Western Europe and also noted that there are causal effects in both direc- North America in the 20th century has been “to give tions: Not only can protective institutions allow everyone a stake in society,� supporting policies to assets to thrive and incomes to grow, but a society strengthen social safety nets, to subsidize general edu- strengthens its institutional capacity as incomes cation, and to make home ownership more inclusive. grow. In a similar way, there will be causality both Finally, policies that are frequently pursued—wasteful ways between inequality and good institutions. protectionism, unsustainable macropolicies, a bloated Countries with greater inequality have a weaker rule public sector—are best understood as short-term re- of law (and lower incomes). The key point of this distributive games that are costly in the long run. section is that highly unequal access to assets can be These games are played at greatest cost in nations with punishing to asset creation, preservation, and im- poor institutions, giving them low ability to negotiate provements in well-being if institutions are not rock and to commit to mutually beneficial change.45 solid. Groups without assets see themselves as un- Catalysts for change supported by property rights, and are thus less sup- portive of property rights politically. This under- Institutional reform happens when the actors take mines support for the evolution of institutions that advantage of opportunities for change, and use in- enable growth and sustainable development. struments of change at their disposal. The institu- A narrowly based elite is often concerned about tions that mediate social interaction must foster both the risks of more inclusive political empowerment. stability and change. A measure of stability and pre- One concern is that they might be expropriated. Just dictability in the rules governing society is necessary as unequal access to assets can be an obstacle to the for the people to have confidence to work together, emergence of good institutions, improperly designed to challenge each other to improve their commu- and balanced, redistribution of existing assets can also nities, and to invest in their future. A vibrant civil be harmful to the emergence of good institutions. society and such institutions as a democratic legis- If a person without assets is more likely to support lature can provide for dynamism—including that expropriation of assets, then a group that has lost in rulemaking. In a society founded on broader con- through expropriation is also likely to become less sensus and certain ethical principles, these institu- respectful of the law and property. It should be tions are simultaneously given force and anchored to clear that it will be easier to improve inclusiveness give predictability and confidence. through access to new types of assets (as when land A seeming paradox is that democracies—despite for agriculture replaces the importance of natural re- their frequent leadership changes—can better commit sources—minerals, forests, fisheries; or when educa- to the longer run, and do the right things. When char- tion replaces muscle power, etc.) and through the ex- acterized as a stable democracy, countries benefit from pansion of assets that come from the growth process. predictable successions, as if the democratic institu- Redistributive measures must be designed and bal- tions themselves have taken on the role of an owner anced to avoid undermining the emergence of good with a long-range perspective.46 However, studies institutions that enable people and assets to thrive. show that young democratic states have problems (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  similar to those in autocratic states, in terms of pro- in haste. Institutions may also be designed to facili- tecting property rights to allow assets to thrive, at tate change in one direction and not in another. In least until democratic institutions have taken root. many Latin American countries a national park may This is a challenge for fledgling democracies. be created with a simple presidential decree, but dis- Many East Asian countries have done well in the mantling a park requires the approval of both the last 30 years in stability and asset creation, but not president and the legislature. Some countries have an- necessarily through democratic institutions. What chored their commitment to the environment in the these countries had in common, however, was broad constitution by linking the environment to the rights access to land (some had experienced historical of their citizens. Others, such as India and Pakistan, shocks that gave them land reform) or other charac- have supreme court decisions serving the same pur- teristics that made them emphasize shared growth, pose.48 The stability provided by all these institutional such as through broad-based rural development and mechanisms depends on whether they rest in shared broad provision of health and education. values, so that relevant actors abide by them. In gen- Some institutions—such as constitutions—are de- eral, multiple review procedures and systems of checks signed to make change exceptionally difficult.47 Con- and balances work only when political and economic stitutional changes typically require a much higher power are not too concentrated. degree of voter consensus than do lesser legislative re- forms. And a two-chamber legislature requires that Opportunities coalitions be built in alternative ways for changes to Improvements in social conditions and in the insti- pass. Other institutional means of providing stability tutions that shape them often seem unbearably slow. and commitment are so-called checks and balances, But significant and sometimes sweeping institutional that is, independent veto players. High court justices reforms do occur, as with democratization in South appointed for life, and systems with separation of Africa, the successful anticorruption campaign in power among branches of government can ensure Hong Kong (China), and decentralization in Latin that radical departures from the norm are not made America (figure 3.5). Figure 3.5 More mayors in Latin America are elected locally—by citizens or by elected city councils Number of countries 25 Panama, Guyana The Bahamas Paraguay, Chile 20 Rep. B. de Venezuela, Nicaragua 15 Guatemala, Uruguay, Colombia Bolivia El Salvador 10 Ecuador, Argentina Honduras Peru, Belize 5 Trinidad and Tobago, Mexico, Jamaica, Dominican Republic, Costa Rica, Brazil 0 1980 1982 1984 1986 1988 1990 1992 1994 1996 Year Source: IADB (1997). (c) The International Bank for Reconstruction and Development / The World Bank ������������ ��� ����������� �����������  Opportunities for reform often arise from eco- tional reforms. Cultural translators (see chapter 4) nomic or political crises that inspire civil society or po- can bring new ideas and ways of working together litical elites to demand changes in the status quo and that increases the self confidence and voice of groups to search for new solutions to long-standing problems. previously excluded from participating in or autho- In Latin America, perceived crises in the legitimacy of rizing institutional change. In Cubatão, Brazil, the governing institutions have inspired substantial de- state environmental protection agency was able to mocratizing reforms that give greater voice and power reduce pollution by joining forces with a vocal citi- to local communities (chapter 7). On a more routine zens’ movement during Brazil’s transition to democ- basis, opportunities appear as a result of elections, racy (box 6.3). Stakeholder forums can facilitate changes in agency leadership, or discretionary deci- conflict resolution and the sharing of ideas, and the sions by national leaders. And opportunities for re- consensus building associated with international en- form arise with changing public preferences, and with vironmental treaties has allowed steady progress on changes such as education, urbanization, technologi- issues ranging from wetlands conservation to pre- cal change, and income growth. For example, new venting oil pollution at sea (chapter 8). generations of individuals raised in conditions of ma- terial prosperity and stability tend to place greater em- A spatial approach phasis on freedom of expression and quality of life.49 Some changes come about easily, some only with The demands that societies place on their insti- other structural changes—such as changes in tech- tutions also change as a result of observing other nology, endowments, and values. Changes in labor societies. In recent years the transnational social market conditions, combined with public and gov- movement for indigenous rights, the sweep of inde- ernment action, have contributed to phase out slav- pendence movements across Eastern Europe, and the ery and elevate women’s status. And women with multinational campaign for transparency in gover- more voice, clout, and knowledge have delivered nance show that new social demands can spread fertility declines and better-educated children. Eco- rapidly across borders. nomic growth, better transport, and the successful growth of cities give poor people a broader choice of Information and forums as catalysts for change employers, neighborhoods, and service providers. A suite of tools is available for catalyzing changes in Today, perhaps the most significant structural change institutions: tools that provide information to en- is the information revolution, with the potential not hance the voice of neglected stakeholders, and forums only to increase knowledge and the use of knowledge held for collective problem solving. Unaccountable but also to improve accountability. power structures are often vulnerable—and thus po- Dynamic development is sustainable when it is tentially responsive—to the exposure of information, forward looking and responsible. Therefore it must so having more information available can be a catalyst be assessed not only by such indicators as poverty, for change used by reformers, entrepreneurs, and civil natural resources, forest coverage, and ocean temper- society. A broad set of studies shows the power of in- atures but also by the institutional environment that formation disclosure in environmental performance, helps this information emerge, gives it weight, and in both developed and developing countries (chapter ensures that it is acted on. This chapter proposes a 7).50 A growing number of firms and NGOs provide checklist of functions for the institutional environ- investors and markets with information on the social ment: to pick up signals, to balance alternatives and and environmental performance of companies (chap- interests, and to execute decisions. It also highlights ter 8)—information that facilitates decisions by in- some barriers to institutional development: dispersed vestors and customers, and catalyzes changes in the interests and commitment problems. These barriers behavior of firms. In the realm of governing widely are more easily overcome by fostering inclusiveness. held corporations for profits, better financial account- The main messages of this Report are that sus- ing is sought to improve actual behavior. Also, public tained development requires that a broad portfolio of disclosure of weakness drives pressure, both for adher- assets thrives in order for people to thrive, and that ance to rules and for better rules (see box 3.7). managing this broad portfolio well requires better in- Enhancing the voice of stakeholders can dramati- stitutions.51 In this chapter it has been argued that cally shift the balance of forces that favors institu- the quality of institutions themselves influence, and (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  are influenced by, the distribution of assets. As a re- trusting, and see greater value in creating and support- sult, more inclusive access to assets and more inclu- ing good institutions. In such a setting they can bet- sive authorizing institutions allow implementing ter accumulate, manage, and protect a variety of as- institutions to better protect assets and people, and sets, including environmental and social assets. As to facilitate well-being. In short, the more people stakeholders they can become more cooperative, cre- that are heard and the more diverse interests that are ative, and willing to take risks—all necessary for the voiced, the fewer assets that are wasted. transformations described in the rest of this Report. Inclusiveness can be increased in a number of ways. In the next few chapters (chapters 4, 5, and 6), With greater access to education, agricultural land, these ideas are applied in spatial arenas—marginal and security of tenure, people are better-endowed cit- rural, commercial rural, and urban areas—where izens. They have something to lose, and something people live, enjoy life, and engage locally, before the that can grow—they can be more forward looking, ideas are applied at the national and global levels. (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 4 Improving Livelihoods on Fragile Lands The test of our progress is not whether we add more to makers. Because it was assumed that they would the abundance of those who have much; it is whether move out of these areas, few governments took the we provide enough for those who have too little. initiative to gather information about their activities. —Franklin Delano Roosevelt As a result, institutions have not been picking up so- cial and environmental distress signals from the pe- riphery—nor have these institutions been able to Inclusion, innovation, and migration balance interests (particularly dispersed interests) in One-quarter of the people in developing countries— setting their development agendas. For the past 50 1.3 billion in all—survive on fragile lands, areas that years the government and private sector have focused present significant constraints for intensive agriculture the bulk of their attention and agricultural spending and where the people’s links to the land are critical for on the development of lands with commercial po- the sustainability of communities, pastures, forests, and tential—even though much of the rural population other natural resources.1 They account for many of the remains on marginal land. people in extreme poverty, living on less than $1 a day. This focus is beginning to shift. Returns on more The size of this population is a signal that our assump- productive land are diminishing.2 And boosting tions about the extent and speed of outmigration have yields in fragile areas is becoming more pressing— been flawed. The least productive areas should have and feasible. But to address the needs of people on been abandoned first, as people migrate out to better marginal land requires more research on appropriate opportunities. While some people have left, many re- technologies and services and more information on main behind and others are migrating in (the estimated their conditions. Many of the households are headed population on fragile lands has doubled since 1950— by women, constrained by poor education opportu- chapter 1, figure 1.5). Improving their livelihoods is es- nities, little access to information, and no legal land sential for meeting many of the Millennium Develop- tenure. Population pressure, lack of knowledge, and ment Goals for the coming decades. simple fear of change lead to destructive patterns of People living on fragile lands are vulnerable but asset management. Understanding the problems and have a modest portfolio of assets that can help bring finding ways to help these people out of vicious cir- them out of poverty: the land (albeit with con- cles of degrading existing assets, damaging liveli- straints), traditional social capital, human capital, hoods, and blocking paths out of poverty are major and indigenous knowledge and know-how. However, challenges. the potential productivity of even these assets has not People are more likely to break out of vicious cir- been fully developed by either local or national insti- cles when change is introduced gradually but stead- tutions. Living in dispersed settlements and working ily over long periods. And change is more likely in the informal or subsistence economy, people on when the risk factors are addressed openly, in ways the rural periphery are largely invisible to decision- that make the costs less burdensome to those who  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  have most to lose. Long-term advice and grant Table 4.1 money to experiment with innovative institutional Environmental fragility in developing countries solutions should be part of the package—to mitigate risks. Introducing high-tech mining operations in re- Share of Share of population earth’s land mote areas disrupts communities and can harm the Number on fragile surface environment. Setting up community-based schools of people lands affected Characteristics (millions) (percent) (percent) is a major shift from the centralized system and often Aridity 518 40 35 perceived as a threat by ministry officials. But coun- Only 350 tries can benefit from long-term partnering with ex- Arid, slope 36 perienced institutions to help them think through Arid, poor soil 107 Arid, slope, poor the process. Successful strategies combine outmigra- soil, forest 25 tion of a few family members, organization of com- Slope 216 17 7 munity associations, and national programs that up- Only 149 grade the community’s modest portfolio of assets. Slope, poor soil 26 Slope, forest 41 This chapter looks at what governments, commu- nities, and the private sector can do to promote Poor soila 430 33 22 Only 386 growth and improvement in the well-being of peo- Poor soil, forest 44 ple inhabiting fragile lands. The emphasis here is on Forests (only) b 130 10 7 arid areas (because of the many people living there) Total 1,294 100 73 and on mountain slopes (because of the links with Notes: a. FAO data on soils unsuitable for rain fed agriculture. water, forests, and mineral resources). How can pub- b. Total estimated number of people in forests is 237 million, of whom 130 million live in forests that have no other geophysical lic and private (national and local) institutions pro- constraints. These forests are part of fragile ecosystems, mainly mote in situ upgrading and/or outmigration? Some in remote tropical areas (Amazon, Central Africa) and the boreal of the options explored in this chapter include: forests of Asia. Conversion to private commercial use needs to take into account the forest’s private and public good values. Source: Averages of CIESIN and LandScan (see Endnotes, chapter Ⅲ Allowing voice and the inclusion of these groups 4, note 1). The constraints were classified according to dominant constraint, ranking first aridity, followed by slope, poor soil, and in the decisionmaking process. Only in this way forest. This does not include fragility due to weather-related factors. can institutions pick up the signals of what is hap- pening at the periphery so they can design appro- priate solutions. mated 73 percent of the Earth’s land surface, face sig- Ⅲ Nurturing all the assets available to poor commu- nificant problems for agriculture investment and nities—sharing of know-how, upgrading the sta- have limited ability to sustain growing populations. tus of women, applying research on special crops, Sensitive to land use patterns, they are particularly and sharing revenues from mineral and other as- vulnerable to degradation, erosion, floods, and land- sets that have national benefits. slides. Ⅲ Creating environments that motivate entrepre- neurial people to come forward with ideas that ad- Rapid population growth, fragile land, and conflict dress grass-roots realities. East and South Asia have the most people on fragile Ⅲ Establishing long-term public-private-NGO part- land, and Sub-Saharan Africa and the Middle East nerships that promote transparency, accountabil- and North Africa the largest shares, at nearly 40 per- ity, the transfer of knowledge, and solutions that cent each. All regions have several countries where balance everyone’s interests. people living on fragile lands make up half of their total populations. Between 1950 and 2000 several Managing fragile land to improve livelihoods countries with a large share of their population on Half a billion people in developing countries live in fragile land saw their rural populations triple or arid regions with no access to irrigation systems. An- quadruple. And more than three-quarters of the 42 other 400 million are on land with soils unsuitable countries in civil conflict in the 1990s have signifi- for agriculture, 200 million in slope-dominated re- cant populations on fragile lands (tables 4.2 and 4.3). gions, and more than 130 million in fragile forest The size and speed of population growth in devel- ecosystems.3 These areas (table 4.1), covering an esti- oping countries over the past 50 years were unprece- (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  Table 4.2 Regional distribution of people living on fragile land Population on fragile lands by region Population in 2000 Region (millions) Number (millions) Share of total (percent) Latin America and the Caribbean 515.3 68 13.1 Eastern Europe and Central Asia 474.7 58 12.1 Middle East and North Africa 293.0 110 37.6 Sub-Saharan Africa 658.4 258 39.3 South Asia 1,354.5 330 24.4 East Asia and Pacific 1,856.5 469 25.3 OECD groupa 850.4 94 11.1 Other 27.3 2 6.9 Total 6,030.1 1,389 24.7 Total less OECD 5,179.7 1,295 26.9 a. OECD: Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, Luxembourg, Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom, United States (23 original members). Source: Average of CIESIN and LandScan measurement methods (see Endnotes, chapter 4, note 1). dented—faster than the rate experienced in the Figure 4.1 OECD countries at any time in their history. In two Rural population growth rate relative to share of total population on fragile land generations the working-age population increased 3.5 times in North and Sub-Saharan Africa, and in Latin America and the Caribbean, and nearly 3 times in Average annual rural population growth rate (percent) 4 Central and South Asia. Rural population growth 3 rates even now remain higher in countries where 30 percent or more of the population are on fragile land 2 (figure 4.1). Many people are on marginal land be- 1 cause of their higher fertility rates and because of over- 0 crowding on the better land. Refugees and displaced –1 persons have also been forced there, because they have –2 lost their homes—from floods, fires, hurricanes, con- –3 flict, civil war, or high urban unemployment.4 Some –4 of the people in these marginal areas are the estimated –5 250 million indigenous people with distinct lan- –6 guages, cultures, and attachment to the land.5 0 20 40 60 80 100 Estimated population on fragile lands (percent) Living on the edge—the arid plains Source: World Bank estimates for population on fragile land; average Dryland ecosystems are characterized by extreme rain- rural population growth rates from 1995–2000, U.N. secretariat. fall variability, recurrent but unpredictable droughts, high temperatures, low soil fertility, high salinity, grazing pressure, and fires. They reflect and absorb solar radiation, maintain balance in the functioning systems have developed an array of coping mecha- of the atmosphere, and sustain biomass and biodi- nisms that provide resilience and recovery in case of versity. Although the biodiversity of drylands is low fire, drought, and pressure from wildlife. These mech- relative to that of forests or wetlands, the ecosystem anisms are important for climate changes, which are services they provide are considerable. Despite its expected to intensify drought and the variability of fragility the Serengeti Plain of East Africa currently rainfall in Africa.6 supports the largest tonnage of animal wildlife assem- Of the 500 million rural people on arid and dry bled on land, as did the equally fragile Great Plains of semi-arid land,7 most are in Asia and Africa, but there North America in the past. Dryland species and eco- are also large pockets in Mexico and Northeastern (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 4.3 Share of population on fragile land, countries in conflict, and rural population growth Middle East and Latin America Europe and East Asia Sub-Saharan Africa South Asia North Africa and the Caribbean Central Asia and Pacific >70 percent Eritrea a ( 2.77) b Bhutan (2.7) Yemen, Rep. of Niger (3.62) (3.4) Cape Verde (1.19) 70–50 percent Somalia (3.22) Burkina Faso (2.47) Afghanistan Egypt, Arab Papua New Namibia (2.62) (2.21) Rep. of (2.5) Guinea (2.48) Sudan (2.31) Mali (2.47) Swaziland (2.62) Zimbabwe (3.33) Congo, Dem. Rep. of (2.0) 50–30 percent Uganda (3.96) Pakistan (2.70) Morocco (1.99) Guyana (1.54) Kyrgyz Rep. (2.7) Lao PDR (2.48) Sierra Leone (1.54) Nepal (2.44) Syrian Arab Rep. Costa Rica (3.66) Turkmenistan Vanuatu (3.49) Guinea (2.27) (3.04) Guatemala (3.28) (3.93) Solomon Islands Lesotho (2.02) Algeria (1.77) St. Vincent and Tajikistan (4.08) (4.38) Comoros (2.83) Iran, Islamic the Grenadines Uzbekistan (3.62) Chad (2.35) Rep. of (2.19) (0.9) Senegal (2.85) Tunisia (1.34) Haiti (1.83) Equatorial Guinea Grenada (1.07) (1.25) Belize (3.5) Rwanda (3.43) Botswana (1.97) Angola (2.26) Kenya (3.47) South Africa (2.76) Ethiopia (2.95) Mauritania (1.40) Cameroon (1.89) Nigeria (2.38) Tanzania (3.11) Central African Republic (1.98) The Gambia (3.35) Benin (1.86) (30–20 percent) Sri Lanka (2.26) Jordan (4.11) Bolivia (1.85) Kazakhstan (1.72) China (1.78) Togo (2.45) India (2.44) Jamaica (1.10) Azerbaijan (2.21) Vietnam (2.6) Côte d’Ivoire (3.56) Honduras (2.67) Albania (1.89) Indonesia (1.8) Liberia (2.24) Peru (1.42) Armenia (1.47) Cambodia (2.82) Burundi (n/a) Panama (2.26) Bosnia and Malaysia (1.95) Ghana (2.84) Ecuador (1.81) Herzegovina Korea, Rep. of Madagascar (2.88) Dominican Rep. (0.99) (0.53) Guinea-Bissau (2.01) (1.63) Myanmar (2.31) Mozambique (1.81) El Salvador (2.71) Mongolia (1.5) Congo, Rep. of (n/a) Trinidad and Zambia (2.83) Tobago (1.46) Mexico (1.59) Notes: a. Countries in bold italic were reported to be in civil conflict during the 1990s, defined as war that has caused more than a thousand battle deaths, challenged the sovereignty of an internationally recognized state, occurred within the country’s recognized boundary, involved the state as a principal combatant, and subjected the state to an organized military opposition that inflicted significant casualties. Countries in Civil Conflict in the 1990s from Sambanis (2000). b. Numbers in parentheses are ratios of rural population in 2000 to 1950, U.N. Secretariat www.un.org/esa/population/demobase. Source: Country estimates of population on fragile land: Average of CIESIN and LandScan (see Endnotes, chapter 4, note 1). (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  Brazil (figure 4.2). The low volume and extreme vari- indigenous knowledge followed by its recognition, ability of precipitation limit the productive potential the near-extinction of the plains bison and subse- of this land for settled farming and nomadic pastoral- quent efforts to preserve it, the partial understand- ism. Many ways of expanding agricultural production ing of climatic variability followed by technology to in the drylands—shifting cultivation from other areas, neutralize many of the effects of climate, and poverty reducing fallow periods, switching farming practices, followed by massive outmigration and measures to overgrazing pasture areas, cutting trees for fuelwood— expand the resilience of the ecosystem to withstand result in greater environmental degradation. drought and generate wealth (box 4.1). Both state-driven and market-driven agricultural It is also an example of a heavily subsidized, energy- investments neglect dryland agriculture, with its intensive model that is unlikely to be sustainable in the lower returns and higher risks, concentrating instead United States, and is not replicable in other grassland on agriculture in more productive areas. Research regions. Few countries are of continental size, enabling and development (R&D) funding for temperate easier outmigration to better-endowed areas. Few econ- agriculture is 70 percent of total public and private omies are large enough or diversified enough to enable funding for agricultural research. R&D funding for extensive cross-subsidization from other sectors to pay tropical agriculture accounts for 28 percent of the for the technical solutions to the problems of the frag- total (mostly on rice). And R&D that focuses on the ile grasslands. And few have the political and financial problems facing people on fragile lands accounts for commitment to sustain such a high level of support only 7 to 8 percent of total R&D funding.8 With- over such a long period. The solution to preventing and out the capacity to migrate, and without major finan- offsetting Dust Bowl consequences required massive cial and technical support, poor rural inhabitants in transfers from the rest of society. Each affected state arid areas have few prospects for meeting their nu- alone could not have solved the problems with only its tritional needs.9 own resources. The Southern Plains of North America, Africa’s Sahel, and the inner Asian grasslands face similar cli- Rain, floods, or drought? Africa, north and south matic and soil characteristics but different political, of the Sahara financial, and institutional constraints. The case of Throughout much of Africa the plowing and mono- the Southern Plains is an example of the dismissal of cropping on fragile soils of colonial times continued Figure 4.2 Arid lands of the world Aridity as an environmental constraint No constraint Fragile ecosystem Aridity as constraint Source: FAO Global-AEZ Aridity Map; covering hyper-arid, arid, and dry semi-arid. See note 1. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 4.1 From degrading soils to degrading water—managing natural assets on the Southern Plains Many indigenous people in the Southern Plains of North Amer- lands under private ownership), cash payments for leaving ica and around the world recognized and accepted the basic land fallow, and farm loans tied to approved land practices; constraints of drylands that forced a pattern of ecological re- Ⅲ The Civilian Conservation Corps, planting of shelterbelts straint on their behavior. They also designed rules to alter de- with 220 million trees, soil and water conservation tech- structive behavior. Complex and evolving institutions—tradi- niques such as the introduction of contour plowing, small tions, rules, laws, habits, and a conservation ethic—guided dam and pond construction, mixed cropping, replanting of indigenous cultures to conserve scarce natural resources and grasses, and state and federal protection of the remaining to survive in hostile environments by getting the incentives open grasslands under the Bureau of Land Management. right. The colonial settlers on the Southern Plains saw the tra- ditional use of productive land by nomadic groups as inefficient. Beginning in 1940, normal rainfall patterns resumed, and They converted prime grazing land into intensive agricultural outmigration reduced the farm population and increased farm uses (monocropping, usually wheat). This pattern was badly sizes (about 1 million people migrated out of the area between suited to the lighter soils of the Southern Plains. Deep plowing 1930 and 1970). But in the 1950s Dust Bowl II hit, followed in dislodged soils, and monocropping mined soil nutrients. the 1970s by Dust Bowl III. Conservation practices had helped, but to achieve reliable production for the agroprocessing in- Degradation, poverty, and migration dustry, the United States needed to achieve a “climate-free� Large-scale farming in the 1920s pushed the expansion of agriculture on the plains. It needed to get rain by pumping from wheat cultivation further onto native grasslands. By the next deep, underground aquifers. decade overgrazing, overplowing, and monocropping were ex- acerbated by the worst drought in U.S. history. An area of Financial transfers, technology, and “underground� rain about 50 million hectares was affected each year in the “Dust The government responded with an unprecedented and sus- Bowl� of the 1930s. The government mobilized a range of ex- tained political and financial commitment at the national and perts to find solutions—scientists, agronomists, civil engi- local level to address the human and environmental impact of neers, political and social historians, local farmers, business- degradation. The strategy reflected the conviction that ingenu- men, and politicians. The scientists’ solution was to bring back ity and technology must solve the puzzles of nature that our an- indigenous methods of planting a variety of plant species, re- cestors learned to live with as immutable forces. One striking planting grass on the looser soils, and limiting grazing. The feature has been the reliance on fossil fuel–intensive agricul- business view was against giving up the profitability and ease tural production with deep pumping of underground aquifers of monocropping wheat on large farms. While hundreds of (up to 600 feet), and heavy reliance on chemical fertilizers and thousands of destitute people migrated out of the area, the mechanization. The vast aquifer is being pumped faster than New Deal Conservation program spent an estimated $500 mil- replenishment rates, with a net depletion rate of 3.62 million lion on drought relief in the 1930s ($6 billion in 2000 dollars) acre-feet (4.5 billion cubic meters) a year. Government net and introduced a series of measures: spending per head in the Southern Plains is higher than any- where else in the United States, with state farm subsidies Ⅲ Federal Emergency Relief, zoning laws for the most fragile estimated at a cumulative $350 billion from 1960 to 2000.* areas, repurchases of submarginal private land (it was deemed easier to buy problem areas and move the people *The Economist (2001) December 15th. living there to better land than to regulate and rehabilitate Source: Worster (1979). after independence.10 National governments viewed Ⅲ Growing populations convert higher quality pas- common tenure claims as impediments to getting ac- ture land to grow cash crops. Herders lose the bet- cess to more agricultural land for growing popu- ter grazing land, their security against drought. lations. But when traditional common forests and Migratory movements for herders are reduced, lands managed by village elders were broken up, they lower quality land is more intensively grazed, and were not replaced by alternate tenure arrangements overgrazing leads to degradation. and the state could not protect the areas. Neither in- Ⅲ Poor subsistence farmers have to reduce fallow pe- dividuals nor communities owned the land or forests, riods to feed growing families. The reduction in fal- so there were no clearly defined or direct conse- low increases vulnerability to drought and without quences of misuse.11 So the lands were misused. sufficient inputs, depletes soil nutrients. Degrada- Changes in land use can rapidly lower soil qual- tion and soil erosion get worse. ity, and intensive cultivation can deplete soil nutri- ents. Deforestation can cause erosion, washing away More people and animals are concentrated on the layers of soil most suitable for farming. Two pat- semi-arid and arid lands that can sustain cultivation terns are typical in Africa (and the world): or more intensive grazing only when rainfall is higher (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  Figure 4.3 Rainfall in the Sahel, 1950–2000 JJASO—mean normalized Sahel rainfall 2.5 2.0 1.5 1.0 0.5 0 –0.5 –1.0 –1.5 –2.0 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 Year Note: JJASO stands for June, July, August, September, October. Source: National Center for Atmospheric Research, World Monthly Surface Station Climatology. Available on-line at http://tao.atmos.washington.edu/data_set/sahel. than normal. In the Sahel favorable rainfall from the levels in some lakes and major dams, with adverse 1950s to the mid-1960s attracted more people. repercussion for industrial activity and agricultural Rainfall reverted to normal low levels after 1970 (fig- irrigation. Given the diversity of constraints, Africa ure 4.3), and by 1974 an estimated 250,000 peo- faces daunting challenges in adapting to the effects ple had died along with nearly all their cattle, sheep, of climate change (chapter 8). and goats. Some 7 million people had to rely on The poor quality of soils is another constraining emergency food aid. The devastation prompted the environmental factor. Phosphorus deficiency, low United Nations to call a special conference on deser- organic content, and low water infiltration and re- tification in 1977 in Nairobi, Kenya. tention capacity on much of African soil have been The possibility that the Sahel could enter another limiting factors in agriculture. Unlike climate vari- period of favorable rainfall poses the risk of repeat- ability, this problem can be addressed: soil quality ing the same tragedy—as poor people are drawn can be augmented through careful management and back to the land. Scientists do not have enough in- soil nutrient supplementation. More difficult to ad- formation about the effect of climatic disturbances dress are the recurrent droughts (box 4.2). on the resilience and long-term viability of dryland ecosystems; nor do they know the human and nat- The Asian drylands: Managing competing ural stress that these ecosystems can handle.12 One land-use pressures difficulty in distinguishing between human and nat- Population pressure on arable land in Asia is consider- ural causes is the lack of data on the extent of grass- able—and growing. Severe land degradation affects lands before human disturbance and the loss over some 35 percent of productive land. The result has time. been to put more population pressure on the Inner The Intergovernmental Panel on Climate Change Asian drylands. Most affected are Afghanistan, China, (IPCC) reports that Africa is highly vulnerable to cli- India, and Pakistan,14 and Inner Asia’s high steppe, the mate change.13 Although the equatorial region and largest remaining pastureland in the world, which in- coastal areas are humid, the rest of the continent is cludes Mongolia, northwestern China, and parts of dry subhumid to arid. Global warming will reduce Siberia. Over thousands of years, these grasslands have soil moisture in subhumid zones and reduce runoff. been home to nomadic herders of horses, camels, Already, water storage has been reduced to critical goats, sheep, and cattle, practicing elaborate systems (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 4.2 Traditional knowledge and voice: sustaining livelihoods on the grasslands of the Sahel Traditional survival know-how in Nigeria, grass-roots manage- pensation. Water committees, for example, make decisions ment efforts in Burkina Faso, and high-efficiency rangeland that ensure that a maximum number of working boreholes or management in Mali all illustrate important livelihood strate- water ponds are within walking distance of the community dur- gies in the Sahel. ing the dry season, with adequate backups. (See also box 5.5 on zais.) Seasonal migration and hedging techniques in Nigeria There is hope that locally based rural organizations could In Nigeria, as in much of the Sahel, traditional social and insti- make a difference in coping with the climate problems and tutional mechanisms have allowed pastoralists to adapt to fluc- service delivery. Local institutions in Burkina Faso start with tuations in rainfall and other natural changes.* Dryland people equity and solidarity and aim for growth and development. migrate in response to scarcity and environmental change. For They are reducing poverty with little or no outside assistance. some, migration is seasonal, as between the dry and humid areas of Nigeria. After the short rainy season Fulani pastoral- Mali’s high-efficiency traditional pastoral systems ists migrate south to graze livestock and avoid the tsetse. On Earlier research depicted traditional pastoral systems of the arid their return, they bring back root crops grown in the south. tropical areas as inefficient. More recent findings highlight the Other arid land farmers and pastoralists recognize the value of efficiency of those systems in using their resources.‡ A pio- diversity in their hedging strategies against environmental vari- neering study in Mali showed that the mobile pastoral system ability and water scarcity. They plant a variety of crops adapted produced 1.5 to 8 times more protein per hectare in meat and to different stresses and graze a mix of animals. These strate- milk than beef cattle systems under similar climatic conditions gies help people manage risks by understanding the resilience in the United States and Australia, with essentially zero input that biodiversity contributes to dryland ecosystems. of fossil fuel. The more settled, sedentary systems in Mali were less efficient. Later work in Botswana and other coun- Inclusion and grass-roots development in Burkina Faso tries confirmed these indications of higher biological efficiency. The communities inhabiting the Sahel are poor, and erratic The findings shift thinking about rangeland management weather patterns make them just a growing season away from under the highly variable climatic conditions of arid tropical re- destitution. Providing for basic health, education, and food se- gions. Under those “nonequilibrium� systems livestock produc- curity under such vulnerable conditions remains very difficult. ers need to be able to “track� available forage or find new graz- Service-asset management organizations are development ing areas for their animals, which usually requires access to committees formed to manage local infrastructure assets and large areas that encompass a diverse range of landscape niches. indigenous associations that collaboratively manage resources This calls for mobility and flexibility that enable rapid destocking such as land, forests, water, livestock, wildlife, and some vil- in times of drought and restocking when the rains reappear. lage production activities.† They scale up the internal organiza- tion of villages and provinces by implementing a culturally co- Source: * See Niamir-Fuller (1998); † See Donnely-Roark, Ouedraogo, herent strategy that balances equity and enhances productivity, and Ye (2001); ‡ See Breman and Wit (1983); Behnke, Scoones, and using mechanisms for inclusion, equity expansion, and com- Kerven (1993). of seasonal pasture rotation across wide stretches of Mobile pastoralism—Mongolia.16 Mongolia has re- land in response to climate fluctuations. Herd rotation tained many traditional herding customs and custom- has helped sustain the fertility and resilience of grass- ary tenure with land management as a commons. land ecosystems and improve the health of livestock.15 Herders rely on local breeds (which are stronger and Over the past decade population pressures and more resilient) that graze year-round on native grasses. competing uses on these fragile lands have made it These customary practices were effectively supported hard to find the right balance between traditional by the collectives between the 1950s and 1980s. The land management and demand for higher agricul- policy environment allowed people and herds to move tural productivity. Government policies that dis- over large areas and provided the possibility of sustain- couraged a nomadic lifestyle, herd movement, and able grasslands management under controlled-access temporary use of patchy grasses led to dependence conditions. Until 1989 the state helped move families on agricultural livelihoods and sedentary herds, around to different grazing areas and provided subsi- which created greater pressure on local ecosystems, dized schools and clinics. The state also set up several and degraded fragile grasslands. The contrasting ex- public enterprises that offered employment outlets, periences of Mongolia and northwestern China il- reducing the numbers of herders and keeping herd lustrate some of the problems and possible solutions. sizes relatively stable. (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  The economic transition since 1990 has not been mixed farming systems, which put more pressure on conducive to sustainable management. Livestock fragile land than the traditional mobile pastoralism. mobility declined significantly. Many public enter- Economic reforms in the early 1990s granted prises closed. Having few alternatives, people turned households nominal shares in the collective land pool. to herding—often for the first time. The numbers of Shared areas were fenced off, making herd mobility herders more than doubled from 400,000 in 1989 more difficult. Subsidized inputs, income transfers, (17 percent of Mongolia’s population) to 800,000 in and deep pumping of underground aquifers encour- the mid-1990s (35 percent). Poverty also increased aged a rapid increase in farming. From an estimated to 36 percent of the population by 1995 from a very 3 million indigenous pastoralists in the 1950s in the low base in the 1980s. Herds went from the tradi- “Inner Mongolian� part of northwestern China, farm- tional 25 million head to about 30 million. State ers and livestock producers today number 20 million, subsidies for health, education, and relocation ser- and cattle doubled from 17 million head in 1957 to vices were halted, making migration and the acqui- 32 million today. sition of human capital more difficult. Today, an China’s western development plan shares two estimated 10 percent of pastureland is believed to characteristics with the policies followed in the be degraded, causing noticeable increases in the Southern Plains of the Unites States: intensify agri- frequency and intensity of dust storms. cultural production and create “climate-free� agri- The problem is considered manageable in Mon- culture in the grasslands through irrigation from golia because population pressures are not too high. underground aquifers. The objective is to make the Rural population increased by about 50 percent area a bread and meat basket to provide for China’s from 1950 to 2000 (compared with a 700 percent growing demands for improved local diets. But un- increase in neighboring northwestern China). The like the Southern Plains—where about 1 million government is responding to the consequences of farmers left between the 1930s to the 1970s, en- the last 10 years by promoting secure livelihoods in abling reconsolidation of land holdings and con- the pastoral livestock sector through asset diversifi- version of vast grassland areas to protected areas— cation, risk management, microfinance, and assis- population pressures have continued to increase in tance to improve population mobility. The state China’s grasslands. Poverty rates in these degraded is setting up a fund to finance service delivery in and ecologically sensitive areas are well above the remote areas and is trying to foster growth and national average (25 percent in some provinces, new jobs in other parts of the economy, reducing compared with the national average of 6.3 percent). the number of herders. Having fewer more mobile There is little empirical scientific research on what herders should reduce overgrazing pressures, pro- is happening to the land and the aquifers. The fre- mote sustainable grassland management, and ensure quency and intensity of dust storms are increasing. acceptable livelihoods. Estimates of degradation are 50 to 75 percent, com- Mixed farming and intensifying livestock produc- pared with 10 to 15 percent in the grasslands of tion—northwestern China.17 As in Mongolia, the Mongolia. grasslands in China are state-owned. But settled pas- toralism and the conversion of grasslands to arable Combating desertification and a way forward cultivation were more common in northwestern for the drylands China than in Mongolia, beginning in the 1950s The environmental problems of the coming century when state-owned pastureland was allocated to “peo- will almost certainly arise from the worsening of cur- ple’s communes.� The concentration of people in vil- rent problems that are not receiving adequate atten- lages meant declining pasture rotation and expand- tion. Some scientists rank desertification and defor- ing agriculture. Policies encouraged conversion of estation third among environmental issues requiring prime pasturelands into arable crop land, leading to attention, after climate change and water resources.18 salinization and wind erosion in some areas. Com- Many emphasize that the links between climate mon policies were applied to highly diverse circum- change and other environmental problems (water, stances, resulting in perverse outcomes and higher ecosystems) are likely to be important. And as demon- degradation in some places. Subsidies encouraged strated repeatedly, sector policies taken in isolation (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  may solve one problem while aggravating others, par- to South Africa have shown yields that are 34 to 50 ticularly over a long period. We may know more percent higher than currently grown varieties.22 about these links now, but we still do not understand There are opportunities to achieve sustainable exactly how these issues interact or what the most ef- livelihoods in quite a few areas. But developers must fective measures are likely to be.19 More applied re- recognize that the drylands are not homogeneous search and organized dissemination of lessons and and cannot be made to function sustainably as non- techniques are needed. drylands. Since large numbers of people are likely to With the 1992 Rio process and under the auspices remain in the dry grasslands for at least a few more of the United Nations, the Convention to Combat generations, a range of strategies is needed to iden- Desertification (CCD) was negotiated and entered tify the attributes of the land that can be harnessed into force in 1996.20 With 178 signatories and 115 to provide inhabitants with a livelihood:23 countries directly affected by desertification, the con- vention reflects a global commitment to combat the Ⅲ New technologies for drought-resistant crops problem. It is one of the few conventions that incor- Ⅲ Better water harvesting porate socially and environmentally sustainable de- Ⅲ Some intensification, including the use of fertil- velopment objectives. Recognizing the disconnect izers between the wealth of local experience in dryland Ⅲ Advice on better farming and grazing practices management and the cutting-edge science that con- Ⅲ Innovative insurance schemes (such as those estab- nects global environmental changes to societies, the lished in Mongolia, Kenya, and Ethiopia) convention established institutional arrangements Ⅲ Community-based early warning systems (such as that link national goals and global interest in land those in Kenya) and water management. Ⅲ Local knowledge and new initiatives. The convention also recognizes the need to share the risk and management of solutions over a much Some arid areas can take advantage of solar energy larger group of countries (the U.S. Southern Plains potential; others may have scenic value worthy of eco- example illustrates the limited options available to a tourism development. The Mozambique Transfron- small jurisdiction in the absence of wider burden- tier Conservation Area Program and Burkina Faso’s and risk-sharing). It promotes partnering of national wildlife reserve development are two attempts in the and international groups and linking indigenous com- direction of ecotourism that combine local and inter- munities with the scientific community to develop national cooperation. Research and innovations for solutions to desertification by integrating partners, appropriate service delivery—combined with policies financial resources, and land degradation concerns that link human activities (farming, herding, and set- into ongoing programs. tlements) with natural processes (vegetation distribu- Agricultural research in China and India shows di- tion, seasonal growing cycles, and watersheds)—help minishing returns to investments in many high poten- sustain vulnerable ecosystems while enhancing pro- tial areas, but investments in drylands can produce ductivity to support growing populations. large returns in reducing poverty, even if yields are Living on a precipice—the mountains modest.21 Governments, researchers, and donor orga- nizations are beginning to pay some attention to R&D Mountains provide most of the world’s people with on crop breeding varieties for people on marginal fresh water and a substantial portion of their timber lands, but much more needs to be done by the public and minerals.24 They shelter more than half the sector to replace antiquated crop varieties (see notes 7 world’s biodiversity and nurture varied cultures in a and 8). In partnership with South African institutions, wide range of latitudes, from the polar regions to the the CGIAR’s International Maize and Wheat Im- temperate, subtropical, and tropical zones (figure provement Center has developed two maize varieties 4.4). But their slope, altitude, relief, temperature, for small farmers in South Africa’s drought-prone, isolation, and rainfall make them one of the most acidic, nutrient-depleted soils. Both varieties are highly variable and differentiated ecosystems. The drought-resistant, and one matures early, when farm concentration of people dispersed in many small food supplies are at their lowest. Trials from Ethiopia communities in rugged areas has implications for (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  Figure 4.4 Mountainous areas of the world Slope as an environmental constraint No constraint Fragile ecosystem Slope as constraint Note: Slope = 8 percent. See note 1. Source: USGS Slope Map. their subsistence and for the sustainability of moun- Deforestation in mountain areas has contributed tain production systems. Agricultural potential in to lasting changes in land productivity. Large areas mountains is limited by the small size of arable plots, of European mountain forests were cut and have climatic variability, and more difficult growing con- not grown back because of changes in land use and ditions, including shorter growing seasons, which soil loss. Some mountain areas in Africa have been contribute to higher levels of poverty. The people liv- stripped of vegetation by overgrazing and are no ing in these fragile areas are surviving by deftly man- longer capable of supporting sustainable livelihoods. aging the mosaic of land available to them.25 Land conversion (deforestation) and species depletion can often be spread over time spans longer than a nor- Mountain transformations mal human life, so impacts may not be immediately Mountain environments help to even out the rate of perceptible. Institutions need to be improved so that water flow between wet and dry seasons. But human they can pick up these signals before it is too late. activities, such as forest conversion, can disrupt nor- Some mountain attributes change over even longer mal flows and increase siltation, with costs to down- periods, through gradual erosion or uplifting due to stream users. Mining and fertilizer use can contami- tectonic processes. Glacial retreat due to global warm- nate mountain water before it goes downstream. So ing is already occurring and over the next 50 to 100 in some places downstream users have begun to years, nearly all mountain glaciers are likely to have compensate upstream users to mitigate the negative melted, affecting downstream water flows. Some en- impacts (for instance in Costa Rica).26 Many of the vironmental fixes, such as restoring glaciers or refor- commercial activities in mountains contribute to in- esting in higher altitudes where trees grow slowly, may come generation and growth. But mountain people be impossible in any humanly relevant time span. have not always benefited appropriately from the goods and services provided by mountain areas. The Mountain resources: Forests, minerals, biodiversity, objective here is not to stop change in mountains; it and sustainable livelihoods is to manage resources in ways that provide sustain- Logging generates employment and income—as well able livelihoods for mountain dwellers and provide as inputs for production. It can also disrupt local cul- the goods and services valued in lowland areas. tures and production patterns. Unless forest dwellers (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  control their own resources and gain the revenue Box 4.3 from their exploitation, logging may not raise the in- Balancing public and private goods: biodiversity and coffee production in Chiapas come of mountain people—and may even impover- ish them over the medium term (as happened in The El Triunfo Biosphere Reserve has remarkable biodi- India and Indonesia), even in high-income countries versity conservation value, with relatively large tracts of (West Virginia).27 It may provide short-term income still-intact cloud forest and a high diversity of native animal at low wages for loggers, but once an area is logged and plant species, including many which only occur in the out, timber companies move on, leaving the local Sierra Madre of Chiapas and Guatemala.* Inside the El Tri- unfo Reserve’s 120,000 hectares of pristine forest are populace without traditional sources of livelihood in some of the poorest people in Mexico. At 40 percent, the the now-depleted forests. This often leads to destruc- incidence of extreme poverty in Chiapas is more than tive cutting of the remaining wood for other uses, twice the national rate (17 percent) and more than six which is not sustainable. times the incidence in Mexico’s northern states (6 per- Minerals, like forests, are distributed unevenly cent). Some 14,000 people in a buffer zone of private land inside the reserve had been clearing forest to plant moun- among mountain areas and are often extracted by tain-grown coffee, cutting down some 17,000 hectares of enterprises (foreign or national) from outside the re- forest in the last 20 years. Coffee producers were un- gion. The impacts of mining are more localized than aware that tree cover protects the coffee plants and im- for logging, but usually more intense due to the po- proves the quality of the coffee. In July 1999 the Global Environment Facility (GEF) pro- tential for social clashes and possible environmental vided grant funding ($750,000) for a Habitat Enhancement problems. More recently, there has been progress in Project. A local NGO was put in charge of fostering com- addressing sustainable mining. munity cooperatives and local leadership in 20 villages, Biodiversity and amenity benefits are among the helping local leaders prepare natural resource and devel- opment plans. The NGO brought together for the first time most widespread of mountain values and among the local government officials, communities, and NGOs to co- most difficult to assign market prices. Although in- ordinate activities, learn about shade-grown coffee, and dividual species attract attention, most biodiversity improve access to credit and technical assistance. and amenity benefits stem from the integrated func- The El Triunfo farmers were among the first to test an emerging market for environmentally friendly coffee. The tioning of mountain environmental systems. These organic shade-grown coffee and the producer organiza- ecosystems provide important sources of livelihood tion’s skill in marketing the superior quality coffee allow for mountain dwellers. A steady livelihood can be farmers to earn a premium of 40–100 percent over ordi- earned from the sustainable use of mountain forests, nary mountain-grown coffee (and over what they were earning before). Investing in knowledge, local leadership, for example from tourism and recreational uses, or and grass-roots cooperation gave poor farmers an incen- from combining biodiversity preservation and com- tive to protect their natural resource base as one of their mercial crop development (box 4.3). best assets. Integrated mountain systems have aesthetic and Source: Pagiola and Ruthenberg (2002). economic benefits of global value. They reduce risks of landslides and protect biodiversity, which pre- serves genomes for food crops and the development of new medicines. Mountain forest areas can also be dant, use of the public good does not usually pose a important for sequestering carbon dioxide. It is dif- sustainability problem. As pressures increase, overuse ficult to translate these benefits and values into mar- and abuse may arise, usually requiring some type of ket prices and transactions, but work is under way institution to manage the scarcity. Threats can result on a carbon trading system (one example is the se- from degradation due to open access exploitation, questration program in Costa Rica—see box 8.5). from insufficient protection of valuable assets, and Mountains are involved in many ecological pro- from imperfect pricing of the goods provided. Man- cesses: water management, biodiversity, weather in- aging mountain environments often requires more fluences, and cultural, recreational, and amenity val- elaborate consideration of the systematic secondary ues. Human interventions can alter these relationships effects than is the case for lowland areas. in ways that may harm (or benefit) different popula- There are often competing demands on mountain tions. Just as on the arid grasslands, when population resources for increasing resource extraction or pre- and economic pressures are low and resources abun- serving in-place and downstream services. Like dry- (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  lands, mountains are not homogeneous. Each area tries today. The institutions that developing coun- requires a different strategy based on its inherent po- tries inherited were not geared to addressing the tential, the mix of natural resource values, and the problems of large, dispersed groups living in remote, commercial value of some of its renewable and non- fragile areas. Today, in many cases government spend- renewable products. All strategies need to incorpo- ing for social services is highly skewed toward the rate the land’s potential and the voice, capabilities, better-off in urban areas—even when a large share and aspirations of the people living there. of the population inhabits rural areas, marginal lands, and the urban periphery (chapter 6). Many Nurturing assets by listening—and by enabling countries have highly centralized and standardized communities to act education and health delivery systems that simply do In addition to the geophysical constraints, other so- not fit the needs of remote areas—and are costly cioeconomic constraints leave many people in the to administer. Agricultural investments and services rural periphery with little to protect themselves from are concentrated on the more favorable lands, even shocks. Poor health care, limited access to education, when the majority of farmers are on fragile lands. information and technical assistance, and high urban Countries are slowly changing these approaches.28 unemployment reduce the opportunities for outmi- Nurturing women’s human capital gration and lower the remittances sent back to the village communities. Many developing countries have Studies of a wide range of societies find that women been ill prepared to help people on their rural pe- are an important engine of growth and develop- riphery address problems and get connected to the ment.29 Their ability to save and invest in their fam- economic mainstream. ilies is well documented. As the family’s nutritional Lacking access to information, education, and gatekeeper, women fight hunger and malnutrition. training, subsistence-based communities have diffi- Their largely unrecorded role in agriculture explains culty improving their health and diversifying their off- the survival of many traditional subsistence commu- farm activities. The costs of addressing malnutrition nities on marginal lands. Yet in many places, tradi- are manageable, yet micronutrient deficiencies remain tions, limited mobility, and lack of voice or access to serious in an estimated 85 countries, reducing mental information make women the most marginal group. capacity and the ability to learn. Schooling deficien- With the men seeking work elsewhere, women tend cies are poorly measured, since most systems focus the fields and look after the children, the elderly, and more on enrollments rather than on completion rates the farm animals. Traditional communities depend or the relevance of curricula. Poor access to health and on women and girls to fetch fuelwood and water, education services increase the incidence of mental and to produce and prepare food. Are national and handicaps and low productivity, blocking opportuni- local institutions investing in this engine for growth, ties for marginalized communities to advance. or are they handicapping it? This section looks at how communities can nur- Some 80 percent of economically active women in ture their assets and find ways out of poverty through Sub-Saharan Africa and South Asia are in agricultural a combination of public sector or centrally initiated activities—largely subsistence farmers in female- and top-down policies (as found in Tunisia), and lo- headed households or day laborers on larger commer- cally initiated and bottom-up changes that work their cial farms. These economic realities are beginning to way up to power centers (Morocco). Peru’s mining give women more influence. Forward-looking insti- sector, for instance, looks at a recent attempt at shift- tutions are responding with changes in attitude and ing to shared development among communities, service delivery. Bangladesh’s Grameen Bank and companies, and the government. The way marginal Morocco’s Zakoura Foundation offer microcredit for rural groups in some European countries got out of women and schools for girls; women contribute to poverty 100 years ago also reveals important lessons, the design of water, health, and education projects showing how much more difficult it is for develop- in West Africa, Central America, and Baluchistan. ing countries today (box 4.4). Agencies and communities, recognizing the high Industrial country institutions never had to deal returns from raising women’s status, are teaming with many of the problems facing developing coun- with NGOs, local anthropologists, sociologists, and (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 4.4 What worked then (Europe, 1900) is much harder now (developing countries, 2000) At the turn of the last century, many of Europe’s poor peasants Ford’s decision meant that poorly educated workers could inhabited marginal lands. They got out of poverty traps thanks begin to accumulate capital and savings—enabling unskilled to ingenuity, to inclusive and flexible institutions, and to favor- workers to lift themselves and their families out of poverty. He able circumstances that do not exist for the rural periphery today. reduced labor turnover from 300 percent to 23 percent and in- Technical innovations attracted unskilled workers and encour- creased productivity by another 50 percent. What motivated aged the migration of peasants from Europe’s rural periphery to him? He wanted to sell more cars (wages were so low at the factory jobs in North America. A vibrant civil society brought time, that few but the wealthy could afford them). And he about sustained and wider participation in income growth. wanted to block the establishment of a labor union. In the following 50 years, interest groups in the United Migration then . . . but not now States and other OECD countries pushed for shared growth, Institutions never targeted policies to deal with people remain- creating institutions to include more people in a wider prosper- ing on fragile lands, because most of them left. Open migra- ity circle. Top-down measures (such as universal public educa- tion from Europe between 1870 and 1910 reduced pressures tion and health care) and bottom-up measures brought about on Europe’s poor rural areas and boosted productivity in the wider participation in income growth. Labor unions obtained New World. Some 13 percent of Europe’s labor force migrated higher wages through a combination of collective bargaining, to the New World during those 40 years. For Italy and Ireland, increases in productivity, and some tightening of the labor mar- as much as 45 percent of the labor force migrated—for Scan- ket. Women’s rights organizations gained for women the right dinavia, about 25 percent. Some 80 percent of migrants were to vote and later to become active participants in the job mar- peasants or unskilled laborers with no more than primary edu- ket. Social safety nets helped the elderly and unemployed. cation, but they found jobs in factories and mines. The transi- These and other policies all served to bring more people into tion took place with few legal restrictions, and government fa- managing, distributing, and benefiting from the countries’ grow- cilitated the assimilation through public education and health. ing wealth. The policies supported inclusiveness and helped For developing countries today, outmigration from the rural create better institutions. periphery is toward coastal urban centers and the peri-urban shantytowns, not North America, Western Europe, or other de- Technology, wages, and jobs now veloped countries. Cumulative migration to the United States By the end of the 1970s production methods in all countries from 1970 to 2000 accounted for less than 2 percent of the started changing with diminishing returns to unskilled labor labor force in Sub-Saharan Africa and less than 5 percent in and increasing returns to skills. Today, unskilled workers in de- Latin America and the Caribbean (the region with the highest veloping countries face legal migration restrictions and higher migration ratio). Unlike 100 years ago, when peasants made skill requirements. The limited number of jobs with above-sub- up 80 percent of migrants, today professionals, skilled work- sistence wages makes it difficult to improve the incomes of ers, and those with some university training make up more the globally large uneducated, unskilled work force. Since 1990 than half the migrants into the United States. The lowest the high supply of unskilled workers has pointed to a global skilled workers came from Mexico, the highest skilled work- stagnation and convergence of wages at subsistence levels ers from Asia and Africa.* in many developing countries. This makes it difficult for out- migrants from rural areas (both the periphery and the over- Technology, wages, and jobs then crowded commercial rural areas) to find gainful employment The factories of the early 1900s employed unskilled workers in urban and peri-urban areas. with little schooling, at subsistence wages (under well-docu- Even though the informal sector accounts for the largest mented Dickensian working conditions). Henry Ford took the share of employment for the working-age population, it is unprecedented decision to improve working conditions by pur- not visible on the economist’s radar screen. Data on the infor- suing his own interests within the context of the interests of a mal sector are not systematically collected. Wage rates re- wider group. Increasing labor’s access to assets is a distribu- ported in the 1990s for farm labor and unskilled construction tional initiative that has efficiency gains, recognized even by workers (the two most likely jobs for people migrating from hard-nosed businessmen. the rural periphery) remained low and flat in many countries In 1908, after designing a reliable and affordable automo- (box figure). bile, Henry Ford wanted to bring the unit cost down to expand Only the Republic of Korea (at $500 a month) and the Czech sales to a mass market. In 1913 Ford and his engineers intro- Republic, Mauritius, and Tunisia saw unskilled wages approach duced the assembly line, reducing the time to assemble a car $250 a month. Average purchasing power parity (PPP) wages from 12 hours to 2 with the same amount of labor. Productiv- for unskilled work in most of the other countries remained very ity shot up, and the labor required no education and little train- low, at under $100 a month for the past decade. Farm wages ing (half of Ford’s workers were poor immigrants from Italy and show a similar trend. The average PPP equivalent wage in the Eastern Europe’s marginal lands). After a year of record prof- OECD countries for similar work was 16 times higher for farm its, Ford more than doubled unskilled wages and reduced the labor and 22 times higher for unskilled construction work.† work day from 10 hours to 8—even though workers were wait- Such a difference in wages for farm work is partly explained by ing in line for jobs at the lower wage. legal migration restrictions and barriers to agricultural trade in (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  Box 4.4 (continued) Monthly wages of unskilled construction workers and changes, 1989–99 Cambodia Sri Lanka Sierra Leone Initial value Bangladesh Change during the period Chad Nicaragua India China Zambia Myanmar Central African Republic Mexico Mauritus Romania Tunisia Czech Republic Republic of Korea United States Italy United Kingdom Germany –250 0 250 500 750 1,000 1,250 1,500 1,750 2,000 2,250 Monthly wages in U.S. dollars converted to PPP Source: Freeman and Oostendorp (2000). the OECD countries. Wage differences between skilled and un- which now typically accounts for more than one-third of rural skilled activities is even greater, highlighting the importance of income in many countries. education and training, both of which are totally lacking in the rural periphery of most countries. It helps explain why income inequality has become a global issue in the 21st century—and *U.S. Department of Commerce, U.S. Census Brief (2002), figures 4 and 5. why so many people in the rural periphery have remained be- † See Freeman and Oostendorp (2000). hind and in poverty. Many families survive by diversifying in- Source: O’Rourke and Williamson (1995); Williamson (1997); Hatton comes through remittances from family members outside the and Williamson (1998); Raff and Summers (1986); Lacey (1986); World country and from service jobs in the rural nonfarm economy, Bank Rural Development Strategy (2002l). economists to reach women directly—with informa- tus of women. He routinely visited villages, explain- tion, education, and access to credit.30 ing the changes he wanted. The education ministry By tailoring service delivery to local circumstances transported boys and girls in remote areas to school, and empowering remote rural communities, some and the health ministry sent midwives out to villages countries are finding affordable ways to improve ser- to discuss preventive health care and family plan- vices and help people get out of poverty traps. This ning, and to inform women of their rights. starts with a good understanding of a community’s Decades later Germany’s technical assistance values and capabilities. It requires people who can agency, Gesellschaft für Technische Zusammenarbeit marry an appreciation of modernity with an under- (GTZ), recognized that one of Tunisia’s public agen- standing of local traditions (cultural translators). cies would be a good candidate for its participatory Tunisia illustrates the combination of national development approach. The Sylvo-Pastoral Develop- leadership, long-term commitment, and cultural un- ment Authority (ODESYPANO) had been admin- derstanding to achieve broad-based improvements istering a tree-planting program, with mixed success, in the quality of life of all citizens. Soon after inde- along the barren hillsides in northwestern Tunisia to pendence in the 1950s Tunisia’s President Bourguiba reduce the erosion that silted up dams farther down- began introducing legal reforms to improve the sta- stream. GTZ wanted to fund a project that would (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 4.5 Addressing risks, changing institutions, and reaching subsistence families in Tunisia The families in the semi-arid mountainous region of northwest- After their third meeting the men brought along their ern Tunisia are poor, with an annual average per capita income wives and told Leila they trusted her. The barrier had been of $220. In the mid-1990s GTZ approached the Tunisian gov- broken. Leila taught the women animal hygiene, better milk- ernment with a $1 million grant to finance a project that inte- ing methods, and how to make cheese, plant caper bushes, grated female participation at the village level. The director of cultivate saffron flowers, and plant and braid garlic—all prod- ODESYPANO saw several problems and risks in introducing ucts they are beginning to sell in the local market and to re- female agricultural outreach workers. There were few, if any, sort hotels along the coast. Several activities were introduced trained female agricultural specialists. Families would not want as team efforts for the men and their wives, including rabbit young women to go with male agents to villages in remote, husbandry, improved poultry pens, and better water harvest- rugged areas. And the villagers would have difficulty accept- ing techniques. Having the women hear the same messages ing female agents. GTZ argued that this approach had brought that were being given to their husbands reinforced the know- good results in other countries and that the women in Tunisian how and application of new ideas—significantly improving villages had important farm responsibilities. After a year of dis- outcomes. cussions on how to minimize the risks, in 1995 the director A development dynamic is changing traditions, increasing hired seven women for a staff of 40 outreach workers. family incomes (up 7 percent a year from 1996 to 2000), and One of those selected, Leila, 25, was an unemployed uni- promoting social cohesiveness. It is also reinforcing partner- versity graduate in Arabic literature. GTZ put her and the oth- ships between husbands and wives and among families who ers through a six-week training program and teamed her with are starting to pool resources to create larger commercial ac- Ali, an agent with a degree in animal husbandry who had al- tivities. The number of female agricultural workers has nearly ready been working in the villages. On her first visit to a village, doubled, from 7 women in 1997 to 13 women in 2001. Ali requested the men to allow Leila to talk with their wives. They refused. She sat quietly listening to the discussion and continued to accompany Ali to his meetings with the village men once every two weeks, but never spoke, only listened to Source: Bank staff field visits, interviews with ODESYPANO staff, June the men discuss the problems of the village. 1997, and World Bank (2001d). incorporate women. The idea of having female out- villages. Thanks to consistent efforts by Tunisian reach workers accompany male agents had not pre- leaders over a long period, women enjoy equal rights viously been considered. The director of ODESY- in almost every respect under the law except inheri- PANO was supportive of the idea, but saw many tance. When the women in the Tunisian mountain risks that are difficult for a civil servant to assume. village were asked to reflect on what change had the The experience showed that persistence, grant fund- most important impact on their lives in the past ing, and partnership can overcome an agency’s deep- 10–20 years, they responded that it was “the way the seated aversion to risk taking (box 4.5). men’s behavior toward us has changed . . . they are nicer to us—less demanding, more appreciative. Transforming institutions and individuals: Now they call us by name, and we have the right to The role of leadership refuse our husbands.� Tunisia’s political leadership improved the status of It is difficult for economic analysis to capture all women through decades of persistent public pro- the benefits of bringing remote communities into the nouncements, changes in the laws, and concrete ac- mainstream. The costs can be high and easy to cal- tions. These policies have continued and been consis- culate, but the benefits are more difficult to capture. tently applied even after a change in government in It is possible to have some idea of the costs of not un- the late 1980s. Adding female outreach at ODESY- dertaking investments that include minorities and PANO fit in with these efforts, but it entailed risks remote groups in the development process. Drug that needed to be addressed up front. The director cartels in Colombia, the Shining Path movement in emphasized changing local traditions by strengthen- Peru, rebellion in Mindanao in the Philippines and ing women’s position in the context of lifting up the in Chiapas in Mexico—all are partly attributable to entire family. the discontent and poverty of disenfranchised com- Tunisia has made major progress in transforming munities in remote areas. Unless ways are found a master-servant relationship into a partnership be- to meaningfully bring these groups into the main- tween husbands and wives—even in poor, remote stream, they sow the seeds of later conflict. (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  Building on traditional social capital hamed Amahane now works full time in 14 villages Dispersed settlements, far from urban centers, make on community development, but he advises other it costly and physically difficult to provide services. villages only when they initiate the contact. He helps Public servants, especially teachers, are recruited them identify “cultural translators,� and helps them from urban centers and are reluctant to live in vil- come up with projects that are within the village’s lages. Absenteeism is high, and villagers often dis- means and capacity to manage. A national effort to trust outsiders. Agricultural experts sent to marginal support such activities and expand voice in local rural areas sometimes view the local people as too communities is gradually emerging.32 poor or uneducated to develop themselves. The re- sult is either benign neglect or costly (and only par- Support from the top tially successful) interventions. By building on long- Formal government institutions could have blocked standing traditions, one poor village found a way to the community development process, but the late improve its quality of life. King Hassan II allowed some political loosening in the mid-1990s that enabled local advocacy NGOs to Solving collective action problems in the community emerge. Without this opening, Ait Iktel could not By combining traditional assets of trust and sharing have set up an association or sought external grant with modern assets—educated men and women—the funding. Another boost came in 1997 when the min- villagers can move beyond survival to development. ister of basic education introduced a pilot program Local leadership blended a keen understanding of of community-based schools. The program’s budget the village culture with technical, managerial, and is less than 0.01 percent of the ministry’s budget, but fundraising skills acquired through education and ex- it allowed local NGOs to set up schools, benefiting perience outside the village. The project’s technical de- the many girls for whom the public system was not a sign matched the community’s financial capacity and viable alternative. It also allowed communities to engendered a strong sense of community ownership adapt rules to local conditions, identify teachers, and box 4.6). It was important that everyone contribute, promote stronger community involvement in educa- maintain, and benefit from a project. The villagers in tion. The cost of these schools is 25 to 50 percent that Ait Iktel had to be able to afford the project and sub- of public schools, with impressive results. The pro- sequent maintenance costs. As Ali Amahan explained, gram has remained a small pilot. The ministry, cau- “the grant from the Japanese Embassy for the electric- tious about the initiative, is taking time to consider ity generator was vital. We could not have done the the many changes the program introduced. project without it, but it was important the villagers To reach remote populations in cost-effective ways, work hard to get that grant.� If a project is designed, national institutions need to be flexible—open to new built, and entirely paid for by an outside entity, the ideas and to learning by listening.33 Because govern- community will have little sense of ownership. ment administrations can be highly risk averse, chang- Achieving unanimity is difficult, but in this vil- ing behavior is extremely difficult. Prominent leaders lage it was important for sustaining the dynamic and and international agencies can play a catalytic role in guiding traditional social capital in the direction raising awareness and promoting promising initiatives. of development. When consensus is lacking (as for In 1998 a well-known Moroccan writer, Fatema the girls school), it is better to move forward on ac- Mernissi, published a book about the development tivities on which everyone agrees (the access road dynamic in Ait Iktel. Her book was featured at the in- and electricity). Goodwill has time to develop, mak- ternational gathering of the Mediterranean Develop- ing it easier to reach consensus on the next project. ment Forum in Marrakesh. In 1999 a Moroccan busi- By listening to, understanding, and addressing each nessman launched a rural school program, drawing family’s objections to the school, the village reached on the lessons of Ait Iktel. In 2000 King Mohamed a consensus, and the association prepared a highly VI honored the Ait Iktel Association with a national successful project with locally appropriate features merit award and cited Ait Iktel’s development philos- not found in the state education system.31 ophy for the activities of the Mohamed V Foundation Scaling up community-driven development to a for Solidarity (a national grant facility established in large number of villages requires visible commitment 1998). In 2001 the association received an interna- from the communities. It cannot be forced. Mo- tional award from the Aga Khan Foundation. Such (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 4.6 “Cultural translators� as catalysts to upgrade livelihoods in Ait Iktel, Morocco Ait Iktel is in the High Atlas Mountains about 100 kilometers The villagers were not opposed to sending girls to school, from Marrakech.* Per capita incomes are low, 2,500 dirham but they were dissatisfied with the schooling provided by the ($250) a year, much of it from migrant remittances. In the mid- state. The poor quality of instruction did not prepare students 1990s the village had no electricity, and in drought years for jobs in the village or the city, and it cut children off from local potable water was a 3-kilometer walk. Primary enrollment was agricultural and artisanal roots that could provide them some 5 percent for girls and 20 percent for boys, who attended the livelihood. The association selected an unemployed university state public school about 5 kilometers away. graduate from the village to be the teacher. The villagers refur- The village’s most valuable asset was its traditional social bished an abandoned house for the school room using their capital, characterized by village elders managing by consensus- own materials and set school hours to allow time for girls to do building and an equitable, shared distribution of the limited re- their chores. They also wanted year-round classes (with vaca- sources (brush forest, water, and communal grazing land). The tions coinciding with village events, planting, and harvesting). village’s social capital enabled the community and its social, Classes were taught in the native language, and the curri- musical, and religious traditions to survive over the centuries. culum was Arabic, French, math, and on Fridays handicrafts More recently, it has enabled the community to shift toward a taught by the village women. These represented major changes development dynamic unprecedented in this region’s history. from the state system. By the second year enrollment of girls In 1995, when Ait Iktel faced a third consecutive year of ages 6 to 20 went from 5 percent to 90 percent. To accommo- drought, the villagers pooled remittances and two of the vil- date demand the villagers built a second school in 1998. After lagers (Ali Amahan, then director of the National Monuments of three years, many girls had graduated but had no prospect of Fez, and his cousin Mohamed Amahane, a mechanical techni- continuing to the next level. In 2000 a national NGO (Support cian at a phosphate mining company in Casablanca) organized Committee for Rural Girls’ Education)† set up a scholarship pro- the men to construct a well. Assuming the vital role of “cultural gram for girls to continue their education. translators�—people who understand modern management Over a period of three years, each project contributed to a methods and are also steeped in the local traditions—the two development dynamic that expanded the villagers’ modest men noticed how spontaneously and efficiently the women asset base, and that continues to this day. Incomes increased organized the water distribution and maintenance of the well somewhat, but the time budget increased dramatically, so that and decided the community was ready to do more. The water people had more time to devote to advancement rather than to project’s success set off a development dynamic that contin- survival. Electricity allowed children to study at night, women to ues today. continue working on handicrafts, and the villagers to afford an The villagers established an association, Ait Iktel pour electrically operated irrigation pump. Readily available water and Développement, working under the village assembly’s tradi- electricity cut down on girls’ time for fetching water and wood. tional authority. The village assembly, a traditional patriarchal Health advice is now available on video in the community cen- authority structure that brings together all the chefs de familles, ter (65 percent of families have begun using family planning). manages village affairs, resolves disagreements, and makes The irrigation system has doubled summer crops during the dry decisions based on unanimous agreement. The association mo- season and allowed for some crop diversification. Thanks to the bilized the migrant remittances for community development ambulance, there have been no maternal deaths in childbirth. projects and set up a “village work bank.� Each family con- Total project costs of $300,000 ($300 to $400 per person) tributes five labor days a year on projects. were covered by a grant from the Japanese Embassy (60 per- After constructing the well, Mohamed and Ali asked the as- cent), savings from remittances (25 percent), and the villagers’ sembly about building a school for girls, but the village priority labor (15 percent). Maintenance costs are covered by the vil- was to upgrade the access road and purchase an ambulance lagers, and teachers receive their salaries from the association to help reduce maternal deaths. After these two projects were through a transfer from the education ministry. The grant cost completed, Ali and Mohamed again raised the possibility of of scaling up this level of service nationwide would be roughly setting up a girls school. Again the village assembly had an- $1 billion a year over five years. The ministry of agriculture’s other priority: electricity. Mohamed designed a project that fit annual budget is about $2 billion, most of it devoted to invest- the income levels of the villagers: a small generator for all the ments for farmers on more productive land, even though the homes in the central village and solar panels for more remote majority of farm families inhabit marginal lands. locations. It was critical to the building of social capital that everyone contribute to and benefit from the project. In 1997, on the night they all celebrated lighting up the village, the as- Source: * See Amahan (1998); Mernissi (1997); interviews, field visits with sembly agreed to a school for girls. Association Ait Iktel du Développement, 2000; † http://www.cssf.ma/. recognition is important, especially if the authorities level is a long, complex process. Such transformations back it up with concrete actions. Transforming hier- are being prompted by internal and external political archical national government agencies into institu- and economic pressures from local NGOs—and by tions that listen, devolving some decisionmaking to easier access to satellite news and information which communities, and responding effectively at the local make people aware of the possibilities. (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  Scaling up community-driven initiatives national environment combined with grant funding In several countries, government ministries and civil are critical complements. society are working together to strengthen and ex- There is a long history of qualitative studies on pand community-based initiatives. The Bangladesh community development, but careful econometric Rural Advancement Committee (BRAC) is the largest evaluations are more recent. The results of the and one of the most impressive scaled-up examples econometric research on the effectiveness of commu- of community schools. Other promising projects in- nity development initiatives are still sketchy but the clude the Community Support Program for primary findings indicate that community-based projects are education in Baluchistan and El Salvador’s Com- directed to the poor and can improve service deliv- munity-Managed Schools Program (EDUCO). In ery. Much depends on the village context (homoge- Nicaragua, with its diverse and hard-to-reach popu- neous groups have a higher success rate), on whether lations, the ministry of education devolved manage- the design is sensitive to and scaled to local realities, rial and budgetary autonomy to local school coun- whether the government is committed to the proj- cils with reasonable success. Private companies are ects, and whether the approach is gradual, moni- also getting more involved in education promotion tored, and adapted as necessary.35 and in “adopt a school� initiatives. The use of nonrenewable local resources— Health outreach, microsavings, and credit are balancing interests other badly needed services in remote areas.34 Donors and health ministries are teaming with NGOs to get In all countries, marginal rural groups living on or trained midwives and health visitors (rather than ex- near potentially rich natural resources often have the pensive clinics staffed with doctors and nurses) out least voice in matters that concern them, their land, to villages on a routine schedule with medicines, fam- and other resources. But how the environmental and ily planning, and nutrition advice. Other examples economic costs and benefits from resource extrac- include the following: tion are managed and transformed into other types of assets is critical to sustaining the livelihoods of Ⅲ In Orissa, India, the international NGO CARE is poor communities in fragile mountain areas. Are the setting up microenterprises to produce insecticide- revenues from natural resources shared with the local treated mosquito nets to reduce malaria and to community? Are the revenues used to transform the help poor villages generate income. local and national asset base by investing in new Ⅲ A community-based health and antimalaria pro- human, physical, and financial assets? For some de- gram was launched in 1992 in Tigray, Ethiopia, veloping countries resource revenues have been an with 714 volunteers serving more than 1.7 million important opportunity for accelerating development people in some 2,000 villages. (for instance in Botswana, Chile, and Malaysia). For Ⅲ Private banks in Lebanon are sponsoring NGOs others (Algeria, Angola, Liberia, and Peru), mineral to promote microsavings in remote mountainous and oil resources have not generated sustained, broad- areas. Vans go to villages, collecting savings, mak- based economic growth. Institutional rules make the ing small loans, and depositing the savings in the difference (chapter 7, figure 7.3). nearest bank branch. A few combine mobile bank- In the late 1990s attitudes, approaches, and laws ing with health outreach services. concerning the extractive industries began to alter the rules of the game to give more influence to local com- Scaling up community-driven development to a munities. In countries as varied as Australia, Canada, large number of villages is critical to improving Nigeria, Peru, Turkey, and the United States, local livelihoods on fragile lands. Some government min- communities have been making their voices heard, istries are embracing new approaches, but often the organizing themselves to achieve sustainable benefits leadership, will, and know-how of government offi- from large extractive operations. Social and environ- cials are lacking—keeping promising initiatives at a mental considerations are being woven into decision- modest level. Local motivation and capacity for col- making to avoid harming the community or the lective action are the main prerequisites for scaling environment and leaving behind wasted lands and up successful community initiatives, but an enabling dysfunctional communities after an operation closes. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Poor communities in remote areas have high ex- net benefits, local participation, and respect for the pectations that extractive industries will offer them a environment.�37 chance to climb out of poverty through jobs, infra- structure, and tax revenues. When these expectations Governments are not met, and when social and environmental costs Governments have difficult roles in regulation, rev- are incurred, local communities often revolt. In recent enue balancing, and national and local development years community protests have led to the interrup- that are particularly apparent in the extractive indus- tion, or even closure, of extractive industries opera- tries sector. Getting the structure of fiscal arrange- tions—with high direct economic costs to the private ments right to achieve all of these objectives is not sector, the public sector, and everyone else (examples easy, and solutions need to be tailored to local cir- include a copper mine in Papua New Guinea; a gold cumstances.38 Central governments need to put aside and copper mine in Iryan Jaya; a gold mine in Ber- tax receipts legally targeted to the regions, establish gama, Turkey; oilfields in the Niger delta; and a gas transparent procedures on how the revenues will be pipeline in Malaysia and Thailand). used, and assist local officials to organize themselves Done well, extractive activities can help transform to spend these funds efficiently—with accountabil- a society’s asset base, generate growth, and serve the ity. Each of these tasks is difficult, even for countries interests of all stakeholders. Local communities want with well-developed institutions. Governments have to get out of poverty. Central governments want for- several objectives in setting the fiscal terms: eign exchange and tax revenues to meet fiscal obli- gations. And companies want to maximize the returns Ⅲ Protect tax revenues from commodity price fluc- on their investments. But in many cases, govern- tuations ments have difficulty balancing the different in- Ⅲ Ensure some distribution of wealth to affected terests: institutional capacity is weak, officials are communities insufficiently trained, local communities are poorly Ⅲ Support investment decisions that generate the organized, and companies are left to manage poten- highest returns tially chaotic situations that go beyond their tradi- Ⅲ Avoid corruption and prevent misuse of funds tional areas of expertise. International and local Ⅲ Allow some share of tax revenues to be set aside, NGOs are putting pressure on mining companies either for emergencies or for future generations. and governments by helping local groups organize to demand transparent disclosure, environmental The central government can legitimately be asked cleanups, and fair treatment. Companies—and gov- to provide the legal and regulatory framework for ernments—have begun to develop strategies to re- the environmental and social impacts of extractive spond (box 4.7).36 industries and for institutions that monitor and en- force compliance on the ground. Even if mines use Balancing interests among governments, clean, modern processes, they can still create envi- companies, and communities ronmental and health problems that are technically The community’s lack of voice at an early stage of difficult to address. Communication and emergency mine development at Yanacocha meant that signals plans are needed to respond to accidents and em- did not get picked up, and problems and mistrust ployees and local communities need training in the accumulated. Diverging interests became more diffi- steps to be taken. cult to balance, and problems were much harder to The agency responsible for environmental moni- solve. Inclusive institutions, transparency, access to toring and enforcement needs to have autonomy and information, and attention to the decisionmaking professional credibility. In Peru this responsibility process are now recognized as key elements of good rests with a specialized unit within the ministry of practice and social corporate responsibility in the ex- mines. This ensures technical capability, but also sets tractive industries. Shifting from bilateral relation- the agency up for conflicts of interest, since the min- ships to tripartite partnerships among companies, istry’s mandate is to promote mining as well as regu- communities, and government shows promise. In late it. Conflict of interest could be diminished by a the mid-1990s the Canadian government adopted a more autonomous environmental unit with some in- partnership approach based on “a fair distribution of dustry and community representation (similar to the (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  Box 4.7 Learning to balance interests: two big mines in the Andes The Yanacocha and Antamina mines are 4,000 meters (13,120 received is unclear. Transparency of accounts at different in- feet) above sea level in the Andes Mountains of Peru. At this stitutional levels remains an issue. altitude, agriculture is not viable except for small-scale grazing. In June 2000 an accidental mercury spill proved to be a More than 90 percent of the predominantly rural people in these “wake-up call� for the company, as well as the community and two regions live below the poverty line. Malnutrition, infant mor- the government, prodding them to reexamine the project’s im- tality, and illiteracy rates are high, at two to three times national pact on the local community. The spilled mercury did not reach averages. the water system, and the company undertook remedial ac- Exploration of the Yanacocha mine began in 1989 and op- tion. (There has been some dispute about the circumstances erations began in 1992, while explorations in Antamina began and number of contamination cases.) Much concern remained roughly in 1999, and operations began only at the end of 2001. in the community regarding Yanacocha’s commitment to pro- The experiences of these two mines illustrate the degree to tect the environment and the community’s health. which—globally—expectations and industry practices in so- Concerned for some time that the expansion of the mine cially, environmentally, and economically sustainable mining could affect the source of water for Cajamarca, the Cajamarca are beginning to change. And they show how institutions (gov- community organized itself by November 2000. To prevent the ernments, companies, and communities) need to learn to adapt company from mining the large remaining deposits next to the behaviors, anticipate or avoid clashes, and promote broad- existing mine, the municipality of Cajamarca passed an ordi- based development. nance declaring part of the basin a reserve for water recharge (the ordinance is being appealed to the Constitutional Tribunal Yanacocha—turning collision into cooperation over gold by Minera Yanacocha). in Cajamarca? The company now realizes the importance of a wide-reach- By early 1990 gold and other minerals were detected in the ing “social� license for its operation in addition to its “legal� li- Cajamarca region, and Buenaventura, one of Peru’s leading pri- cense. Consultation has shifted to a broad process that includes vate mining companies, teamed with U.S.-owned Newmont, community validation of local development projects, formaliza- one of the world’s largest mining companies, and the Interna- tion of the information and complaint systems, and multistake- tional Finance Corporation (IFC) to form Minera Yanacocha, SRL. holder dialogue. The company is embarking on an urban devel- The deposit is near the city of Cajamarca, a site of symbolic opment program involving investments that will be added to and historical importance. Inhabited by descendants of the the ongoing rural programs. Since the mine has at least another Incan people, Cajamarca is where, in 1532, the Spanish con- 20 years of operation, it is not too late to forge a socially and quistador Francisco Pizarro and his hundred or so men am- environmentally sustainable development compact. bushed and killed thousands of native Incan warriors and cap- tured their emperor Atahuallpa. Pizarro held his captive for eight Antamina—building a development relationship months “while extracting history’s largest ransom in return for with local communities a promise to free him. After the ransom was delivered (enough Antamina is a new mining venture in the central north Andes gold to fill a room 22 feet long, 17 feet wide, and 8 feet high), Mountains, about 300 kilometers south of Yanacocha. Owned Pizarro reneged on his promise and executed Atahuallpa.�* by a consortium of three Canadian companies (90 percent of Gold has once again become a source of tension in the re- the shares) and a Japanese firm (10 percent), Antamina is ex- gion. Community expectations for the mine as an escape from pected to become the world’s third largest zinc producer and poverty were understandably high. But from the outset the seventh largest copper producer. Export earnings were initially company was preoccupied with security, fearing the activities projected at $950 million a year and corporate taxes at $83 mil- of the Shining Path (Sendero Luminoso). The company believed lion (with 20 percent to go back to the region). it had little choice but to keep a low profile to protect its em- The Antamina operation came along some 10 years after ployees and others in the community who supported the oper- Yanacocha, but the two mines share several similarities. Both ation. The company thus refrained from organizing extensive have high economic profiles as modern operations contribut- consultation meetings and stayed away from the more urban ing valuable revenues to the Peruvian economy. Both are in areas, focusing instead on development activities in the rural areas with poor indigenous people who have little or no previ- areas near the mine and limiting consultation to selected repre- ous mining tradition. But the communities around Antamina sentatives of the community. The feeling of distance between have had much less contact with the modern, outside world the company and the town of Cajamarca was difficult to change than those in the Cajamarca region. For both operations the even when the Shining Path was no longer a threat after 1995. central government’s capacity to address social, environmen- The Yanacocha mine was a remarkable financial success. tal, and other institutional development issues has been lim- It is the most profitable, lowest-cost gold mine in the world, ited. But Antamina wants to avoid some of the problems ex- owing largely to the excellent gold reserves. Under full opera- perienced by Yanacocha and is fortunate to begin its activities tion export earnings have reached $500 million a year in the at a peaceful time when more inclusive rules of the game are past few years, and corporate income taxes amount to $45 to becoming internationally recognized. $55 million a year (30 percent of profits). A 1992 law (canon minero) required the central government to return 20 percent of the annual corporate tax collected (about $8 million) to the *See Diamond (1997). local region, but Cajamarca appears to have received only a Source: McMahon and Felix (2001); interviews with World Bank and part of these funds. Whether and in what form the money was IFC staff. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  environmental agency in Chile). In exceptional cases, see their neighbors’ land selling for much more. Of- responsibility may need to be vested in an impartial ten, farmers spend their sale (or resettlement) pro- external agency. ceeds quickly and find themselves destitute shortly thereafter. Companies may need to propose ways to Companies assist local farmers in managing cash incomes. Where Large mining operations are capital intensive, requir- possible, land sales or rental agreements should take ing skilled technicians, who are often from outside. place between the company and organized groups of With few jobs available for locals, communities in- farmers rather than individual farmers. creasingly look for other compensations. One option is for the company to provide intensive training for Communities locals and small enterprises at the earliest stages of Company and NGO efforts to help local communi- development, enabling them to sell goods and ser- ties develop advocacy and operational capabilities are vices to the mining company. Given the opportuni- beginning to bear fruit. Local communities have ties for local outsourcing, such training can have begun to learn how to organize and find their voice, an important impact on the local economy, foster- a major change that has gained momentum in the ing entrepreneurship. The goal would be to transfer past decade. They need to make sure that their views skills—to have a larger share of locals working for are understood and that their goals are geared to pro- the company, directly and indirectly—and to create tecting and developing their communities. Like the more social cohesion between the company and the companies and government, they need motivated local community. Because mines are a finite resource, leadership, open access to knowledge, and a willing- the revenues need to be invested wisely to create al- ness to learn. Around the world a workable approach ternatives for the community when the mine closes. appears to be gradually emerging around “sustain- The arrival of skilled workers from outside and the able mining paradigms� that combine business strat- availability of higher cash incomes often clash with egy and ethics. In a case as complex as Yanacocha, local customs. Companies would benefit from hiring things are moving in the right direction, but much cultural translators, who can link the modern com- remains to be done. mercial world with the local culture, language, and traditions. If cultural translators are part of the deci- Deeper institutional support sionmaking team alongside the engineers and finan- Many governments are struggling to fulfill demanding cial specialists, they can play a vital bridging function and complicated roles in these three-way partnerships. to help the community and the company understand When government cannot meet its obligations—and each other and resolve problems. Antamina’s consor- companies, local officials, and communities cannot tium of local mayors and NGOs and its early engage- agree on their responsibilities—it is difficult for extrac- ment of three international NGOs working full time tive resources to be developed in a way that is sustain- on development issues show promise. Yanacocha able. If mineral extraction is likely to continue anyway, is now strengthening its community relations, train- given the overwhelming interest of most parties in ing, and outsourcing program. It is also setting up a moving forward, then providing short-term technical foundation to promote income-generating activities, assistance for institutional strengthening—while nec- which will help improve communication and trust essary—may not be sufficient. There must be more with the local community. substantial support and genuine learning-by-doing at Large mining and other extractive industries re- an appropriate scale and duration. Managing tripar- quire extensive land areas. How the land is acquired tite arrangements effectively involves a long and costly (and whether the inhabitants have clearly demar- learning process, but it has a potentially high payoff cated titles) affects the negotiations and the trust of for everyone concerned. the local community. The price of the land needed Partnering for change for mining is also difficult to determine. Farmers who sell early at lower prices—even though they re- Several initiatives have emerged to integrate local con- ceived agreed prices—later feel cheated when they sultation in decisionmaking. The Latin American Or- (c) The International Bank for Reconstruction and Development / The World Bank ��������� ����������� �� ������� �����  ganization for Energy for the oil-producing countries country circumstances. By listening at the grass- of the Sub-Andean Basin emphasizes communication roots level and testing ideas against reality, they can among governments, industry, and indigenous com- promote creativity, relevant policies, and workable munities on how to use and distribute oil rents. It also solutions that help governments govern better (chap- strengthens social and environmental standards for ter 9, box 9.2). oil and gas in Latin America. The Chad-Cameroon pipeline project addresses how oil revenues will be Nurturing assets in the community. . . used, with procedures for incorporating community Communities in the rural periphery have assets that views and regulations on environmental and social need nurturing. Combining local know-how with impact (box 7.10). The Nile River Basin initiative stepped-up research and outside technical advice can tries to get all parties to focus on the potential bene- help increase their land’s productivity and sustain fits of cooperating by thinking not only of their coun- critical ecosystems. Such communities have indige- try’s interests but also the interests of neighboring nous knowledge that has guided them to conserve countries that share the resource (box 8.4). scarce natural resources and survive in hostile envi- Donors can do much to help governments set up ronments by getting the incentives right—and na- the right institutions by supporting long-term part- tional institutions need to listen and learn from these nering, but it will require a greater scale of support, insights and combine them with modern technolog- for example, than what is currently practiced for sci- ical approaches. They also have women who are po- entific exchanges or for the CGIAR. Long-term in- tentially strong engines of development but whose stitutional efforts are needed at all levels. Donors can contribution is too often handicapped. And they substantially increase the volume and reliability of have social capital—and some savings from remit- funding for long-term expert advice, timely techni- tances, which, if mobilized, can launch a develop- cal support, technology transfers, staff exchange pro- ment dynamic. Villagers who have experience out- grams for key personnel, and international training side the area can help the community guide its scholarships (chapter 9). traditions toward the design of projects that fit their means, and governments need to find ways to pro- Combining know-how, information, mote and encourage this type of work. and grass-roots understanding Many developing countries copied institutions from the nation . . . the West, but few adapted them to local circum- Governments (and donors) can fund knowledge net- stances, which led to adverse environmental, eco- works, dissemination and village exchanges, enabling nomic, and social consequences. Centralized admin- local community leaders and government counter- istrations had difficulty adapting public services parts to learn firsthand about creative solutions— designed for urban settings to vastly different con- such as Burkina Faso’s service-asset management ditions in remote rural areas. Institutions have groups or Morocco’s Ait Iktel Association. Develop- changed—but slowly, particularly relative to fast- ing countries receive various volunteer programs growing populations and a fast-changing world. from donor countries and NGOs working in mar- They need to adjust services to local conditions, to ginal rural areas. To ensure that the knowledge trans- listen to and understand the people they are try- mitted stays with a country, governments could set ing to help, and to empower communities to help up national volunteer organizations for local gradu- themselves. ates (often fresh out of university and unemployed) Governments need to encourage open interac- to team with the foreign volunteers to work on proj- tions among government, universities, business, and ects together. civil society. Problem-solving institutes (think-and- By working with NGOs, public institutions can ex- do tanks) focusing on concrete challenges facing pand their reach and improve the relevance and cost- communities are needed. At present, many countries effectiveness of their services. Civil servants are often are badly underserved. Think-and-do tanks can help risk averse, perhaps reluctant to cooperate with out- to make sense of imported ideas and adapt them to side associations. Behavioral changes are sometimes (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  easier to introduce when senior officials launch pilot Donors can do much in learning what seems to initiatives that are followed up with clear evaluations work in community development dynamics and in and results-oriented incentives that promote learning, disseminating this information through hands-on changing, working with local groups, and scaling up village exchanges and support for setting up practi- positive experiences. Better monitoring of appropriate cal problem-solving institutes. They can also take indicators will help governments track what is hap- the lead in R&D and technology dissemination for pening in the rural periphery. For example, what share renewable energy, suitable crops for fragile areas, of public expenditures actually reaches the rural mar- land management techniques, and medicines. They ginal areas? Is health, agricultural, and environmental should expand scientific and empirical studies on advice routinely reaching village communities? What what is happening to fragile lands and the climate— share of the nation is employed in unskilled, low-wage proposing economically feasible scientific and tech- jobs, and how have these wages changed? nological remedies as and where appropriate. Be- cause fragile lands are heterogeneous, improving and the world livelihoods is not always easily replicable. But there There is often the potential for local upgrading con- is substantial scope for adapting innovations across tingent on realizing that upgradings requires signifi- countries and regions; such innovations include com- cant change from the bottom up and the top down, munity schools, outreach advice, drought-resistant and both approaches take time. Many people on plant breeding, and other productivity-enhancing fragile lands have begun to organize themselves to technologies developed in one region with good ap- move beyond survival—and onto grass-roots devel- plications in other regions. opment. They are at the cutting edge of social, eco- This chapter explores opportunities for improv- nomic, and environmental advances. But they need ing in situ the well-being of the many people living adaptable and flexible national institutions and global on fragile lands and ecosystems. At present they have commitments for funding, support, and partnerships few options but to remain. International migration for the long term. is highly restricted, compared with 100 years ago Long-term partnering of institutions—experi- (see also chapter 9). Even internal migration is un- enced civil servants helping their counterparts in certain, especially in economies where the numbers other countries implement difficult reforms—could of rural unskilled workers are very high, or where become a much larger part of donor development as- urban-led economic growth has been low. To address sistance. The long duration of these relationships in- population pressures in fragile areas, outmigration creases the relevance of the advice and could include must be encouraged both by better preparing rural donor-sponsored exchanges between agencies, think inhabitants to take on nonrural jobs, and by improv- tanks, academics, and business advisors. More grant ing the ability of commercial rural areas and urban funding (combined with advice) can help national areas to provide these people with more productive governments overcome the aversion to risk taking. opportunities (see chapters 5 and 6). For the people Sharing risks and burdens across a larger number of living on fragile land, as well as for those in the com- better-endowed countries offers the best chance for mercial agricultural or urban areas, developing their addressing some of the most difficult problems trap- human capital is critical for expanding their options ping communities in poverty. for improved livelihoods. (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 5 Transforming Institutions on Agricultural Land T he previous chapter focused on people living Ⅲ Eliminate rural poverty and strengthen rural-urban in remote, low-density settlements on fragile linkages—including preparing outmigrants for a lands, and how, with new institutional improve- productive urban life. ments, they can better manage their portfolio of assets to increase productivity and sustain critical ecosystems. As discussed in chapter 4, rural populations are Chapter 5 is about people living in areas with commer- expected to grow in most low-income countries. In cial agricultural potential, either in frontier areas where much of the world, the combination of subsistence market-driven agriculture is newly emerging, or in areas food production and cash earnings in the hands of closer to larger and increasingly urban markets. These poor people is not enough to yield an adequate diet. areas will help feed the growing and increasingly higher- About 820 million people lack access to enough income world population. This chapter focuses on the food to lead healthy and productive lives, and about management of and interaction of assets such as land and 160 million children are seriously underweight for water and the environment; how to help the poor get bet- their age.1 Some 2 to 2.5 billion rural people will ter access to land and water; and the importance of asset become urban residents between now and 2050. distribution for the development of good institutions (as Whether their families have land, water, or educa- described in chapter 3)—especially in near-market areas tion before they urbanize is critical to their future, where intensification can generate considerable equitable the future of the cities they move to, and the qual- growth. Population in rural areas totals 3 billion people, ity of their societies’ institutions. and more than half of them live in areas with commer- cial agriculture potential (see figure 1.1). Some of these Ⅲ Intensify agricultural production and manage land people will migrate to cities, and many will live in areas and water to feed a growing and increasingly urban that will be reclassified as urban when the areas’ densities population. increase. The total number will remain in the range of approximately 1.5 to 2 billion people over the next three Over the next 30 to 50 years, rural areas will to five decades. Despite widespread concern over the past have to feed an additional 2 to 3 billion people 20 to 30 years about food shortages, the rural develop- globally, and substantially improve the diets of the ing world has exceeded expectations in food production 2.5 to 3 billion people living on less than $2 a day. (box 5.1). Will past trends continue, or is there a real That will require tilting institutional rules to move cause for concern? For the world to make a smooth tran- assets into the hands of smallholders, halting nutri- sition to relative population stability 50 years from now, ent mining, reducing soil erosion, and adopting its rural areas will have to meet a range of challenges. agricultural practices that restore soil fertility. It will Chapter 5 addresses the key development chal- also require sharing rural land and water to serve lenges for rural transformation over the next 30 to the expanding urban population and meet environ- 50 years: mental needs.  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 5.1 This chapter argues that although the rural “sec- More food, greater intensity of land use, fewer tor� has done well in meeting aggregate food needs, farmers per urban resident it has done less well in meeting the broader needs of the rural population and preparing many for an Global food availability has increased. Global food availabil- ity per capita is at an all-time high, with variations among urban future. It also highlights issues surrounding countries and regions. Doubling grain production and the conflicts and complementarities between pro- tripling livestock production since the early 1960s, the moting rural development and protecting the envi- world’s farmers now provide about 2,700 calories per per- ronment. For both, good institutional rules are criti- son a day. India and China, widely considered two decades ago to be Malthusian disasters in the making, satisfy their cal. Because of this focus, the chapter covers issues own demands for cereal. For the developing world rising dealing with property rights in land and water, and incomes enabled increased consumption of meat and intellectual property in agricultural knowledge. It poultry. And despite growing demand for grain, the prices for maize, rice, and wheat came down 50 percent or more does not try to give a complete, or even a balanced, over the past 20 years. Perhaps most important, the pro- treatment of the problems of rural development. The portion of children who suffer from malnutrition fell purpose is to illustrate the importance of thinking sharply—from 45 percent in the 1960s to 31 percent in the late 1990s—though not yet sharply enough. more deeply about the institutional rules that gov- Agriculture has intensified. For most of the world, re- ern behavior and support policies, and how they duced availability of agricultural land has induced a transi- might be improved.3 tion from land-increasing to yield-increasing technology. The main message of the chapter is this: Coun- Africa and South America are the clear exceptions; they both have large remaining areas of unexploited land. How- tries should, where still possible, give a high priority ever, that land may not be very productive. Although the to creating egalitarian endowments of land, water, trend in South America is toward intensification, the ex- and human capital for its people as they make the tensive margin continues to expand into the Amazon for- est. In Africa there are pockets of intensified production, transition from rural to urban human settlement. but the larger story is one of new frontiers of crop produc- The smooth emergence of land and water institu- tion opening areas previously devoted to communal graz- tions is of fundamental importance to a country— ing of livestock—and of shortening the fallow period under shifting or bush-fallow cultivation. In the more marginal because the rules sanctioning property ownership areas these changes have created new problems (as determine the later character of the state and society. noted in chapter 4). Countries that have distributed rural property equi- Many countries have made the transition from rural tably before urbanizing have developed more egali- to urban human settlement, with fewer farmers feeding more city folk. In developing countries in the 1960s, there tarian and democratic societies than those that put were three farmers per urban resident—today, there is assets in the hands of relatively few rural elites. Put one and one-half. another way, countries with rapidly growing popu- Source: Pinstrup-Andersen and others (1999); Rosegrant and oth- lations that have concentrated land in the hands of ers (2001); Crosson and Anderson (2002). the few have urbanized prematurely, educated few, and developed extremely inegalitarian societies. Ex- perience and research show that creating widespread Ⅲ Get ahead of the agricultural frontier to control land ownership is critical to the later development wasteful land conversion. of inclusive institutions. Land and water constraints The expansion of agricultural land has taken a large toll on the world’s repository of biodiversity, Food production increases are slowing. Land is be- with one-fifth of tropical forests cleared since 1960. coming increasingly degraded. Scarcities of land and The remaining biodiversity is concentrated precari- water are more evident. These problems are best ad- ously—more than one-third of it now confined to dressed by thinking of them not as problems of 1.4 percent of the world’s land.2 Some new agricul- global resource scarcities but as problems of poverty tural land is of high quality and yields important local among plenty. benefits in agriculture. But much of the newly con- verted frontier provides little opportunity for the ad- Global food abundance, yet hungry poor vancement of locals, despite imposing large national The prevailing view among agricultural economists and global social costs in GHG emissions and the loss is that the world food problem is one of insufficient of biodiversity and amenity resources. purchasing power in the hands of poor people, not (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  of global constraints on aggregate food produc- through time.9 Empirical studies based on actual tion—even with an expanded global population. time-series data on soil samples (taken throughout The aggregate data support this view, but some poor China and Indonesia over 50 years) find no overall regions have too little food. And it is true, as many loss of agricultural soil depth or quality for China or point out, that annual increases in food production Indonesia.10 Time-series data for 1971–93 from the have been falling. But annual increases in demand India and Pakistan Punjabs, by contrast, suggest that are falling faster. Evidence at the global level—that intensification of land and water use has resulted in the growth of yields (as opposed to production) is resource degradation that is lowering overall produc- slowing—is extremely weak.4 tivity growth. For Pakistan these data indicate that What is incontestable is that a slowdown in food resource degradation has reduced overall productiv- demand relative to production—much of it inappro- ity growth from technical change, education, and in- priately subsidized in OECD countries—has de- frastructure investment by one-third.11 pressed food prices to record lows. With an ever- Studies based on cropping patterns and fertilizer larger portion of the world’s people fed well, rising use in Africa indicate that failure to replace the soil world incomes induce smaller increases in food con- nutrients removed through cropping (nutrient min- sumption. Falling rates of population growth are also ing) is grave, widespread, and poverty induced.12 All slowing the growth in food demand.5 With higher but three countries in Africa show negative balances incomes, food consumption patterns do change. But of nutrients of more than 30 kilograms of nitrogen, simulations of the world food economy suggest that phosphorus, and potassium per hectare annually.13 even a rapid increase in meat consumption in China African farmers have traditionally practiced bush (underway) and India (less likely) would not signif- fallow and shifting cultivation to maintain produc- icantly alter the balance of world food supply and tion, as decreasing soil nutrients begin to affect demand.6 yields. This practice is becoming unsustainable as In short, food will continue to be abundant at a rising population density shortens the fallow period, reasonable price for those people with the income to which lowers fertility. So more land is needed in cul- purchase it. Eliminating hunger tomorrow, however, tivation each year, partly because of higher popula- will require the same solution as eliminating hunger tion and partly to offset the effect of a decline in today—raising the productivity and incomes of poor yields. Poverty-stressed farmers face three choices, people. And here the world must do better. For the with mixed outcomes: more than 70 percent of the world’s poor people who live in the countryside, this means increasing their Ⅲ Expand into forests, permanent pasture, hillsides, ability to produce food to consume and food to sell or wetlands. in markets.7 Ⅲ Continue to intensify labor inputs on existing land. Land degradation—also a poverty problem Ⅲ Complement labor on existing lands through the It is widely reported that erosion, salinization, com- use of inorganic and organic fertilizer and land and paction, and other forms of soil degradation affect 30 water conservation infrastructure, such as grass percent of the world’s irrigated lands, 40 percent of strips, anti-erosion ditches, hedgerows, bunds, and rainfed agricultural lands, and 70 percent of range- terraces. lands. The effect of this degradation on overall pro- ductivity so far is limited, in part because cultivators The first puts farmers on a collision course with bring new lands under cultivation. Cumulative global other land users, such as cattle herders, commons productivity loss due to land degradation over three holders, and biodiversity reserves. It also begins the decades has been estimated at 12 percent of total pro- vicious cycle of land degradation anew, condemning duction from irrigated land, rainfed cropland, and farmers to work on increasingly marginal lands (be- rangeland. This yields an average annual rate of pro- cause of agronomic conditions, disease, and distance ductivity loss of 0.4 percent.8 from markets). The second is unsustainable in the The underlying degradation estimates have weak- absence of new fertility-augmenting soil manage- nesses, however, because most attempt to estimate ment, hastening the downward spiral of falling yields losses through time without data on degradation and shorter fallows.14 The third choice has been pro- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  hibitive in the past because it requires that farmers tles are looming between direct economic use and assume additional risk in the form of purchased in- environmental needs. And in many urbanizing semi- puts.15 As discussed below, new adapted techniques arid regions, conflict between high-value (usually are becoming available that can improve soil fertility urban) use and low-value irrigation will worsen. using resources naturally available in Africa.16 These Land availability. Additional land available in the techniques offer the promise of breaking out of this developing world is of three types (figure 5.1):19 downward spiral. Ⅲ Land in use for annual and permanent cultivation Land and water: Serious regional scarcities Ⅲ Land lost or no longer usable economically for globally abundant cultivation Land and water, now globally abundant, are pro- Ⅲ Land reserve still unused but suitable for sustain- jected to remain adequate throughout the 30- to 50- able agriculture. year time horizon of this Report—even while meet- ing the needs of a growing population and improving “Lost� land has been either consumed by urban nutrition.17 But the aggregate picture masks serious sprawl or degraded beyond the point of economic local and regional water and land shortages in all recovery.20 Globally, agricultural land has been lost continents—as well as a lack of financing and in- through degradation at the rate of about 0.5 percent stitutional capacity to develop and sustain Africa’s wa- a year—and from new infrastructure at 0.1 percent ter resources potential. The World Commission on a year. Severe degradation comes from water erosion Water predicts that water use will increase by 50 per- (particularly in Southeast Asia and Central Amer- cent over the coming 30 years and that 4 billion peo- ica), soil nutrient mining (particularly in Africa), ple—half of the world’s population—will live under and salinization (particularly in some areas with conditions of severe water stress in 2025.18 Conflict large irrigation schemes). Note that almost as much over land and water will worsen, especially in areas land has been lost (303 million hectares) as is now already suffering from water stress in South Asia and in use (307) in Africa and Latin America. Some of the Middle East and North Africa. In addition, bat- this is from shifting “slash and burn� cultivation pat- Figure 5.1 Regional variations in land scarcity Million hectares 1,200 1,000 Land reserve 800 Total land lost 600 Unsustainably cropped land 400 Land in use for cropping 200 0 Sub-Saharan Latin America East Asia South Asia Middle East Africa and the and Pacific and North Africa Caribbean Regions Source: IBSRAM (2001). (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  terns (based on nutrient mining) and may recover already exceed environmental limits in Afghanistan after an extended fallow period. and Pakistan and will exceed limits in India by 2020. The land that is available (and tillable) varies Irrigation already exceeds recharge rates in India’s widely from region to region. Latin America and northwest plains (the major site of its green revolu- Africa stand out as having extremely large reserves of tion). In the Middle East and North Africa only Mo- tillable land: 700 million hectares and 680 million rocco has unexploited water resources. All the rest hectares, respectively. At least 45 percent of the have exceeded environmental limits, and many are 1,400 million hectares of reserve land is in forests. mining groundwater aquifers (figure 5.2). East Asia and the Pacific and the Middle East and Africa is relatively well endowed with water re- North Africa have little additional tillable land. And sources. It has only 1–3 percent of its agriculture South Asia is already farming soils that are not sus- under irrigation, compared with two-thirds in Asia. tainable for cultivation. The potential for expansion in Africa is limited, how- Much of the potentially tillable land in Africa is ever, because more than 60 percent of the irrigation either not accessible for commercial agriculture, is potential is in humid regions, where, because of high subject to human or animal disease, or is already rainfall levels, irrigation would be at most supple- being used for animal grazing. In Latin America, mentary to well-managed rainfed agriculture. In new land is mostly in the Amazon, predominantly many of the regions where irrigation is most needed uneconomic to exploit. But some of it is potentially more than 60 percent of the potential renewable economic and therefore the subject of considerable water resources are already exploited, and most of the dispute between environmentalists and rural devel- highest potential areas are already under irrigation.22 opment interests. Clearly on both continents this re- Figure 6 in the Roadmap shows the projected maining uncultivated land has extreme limitations water scarcity worldwide in 2025 grouped in three relative to land already under production, from the categories: physical water scarcity, economic water agronomic perspective and relative to markets. Much scarcity, and little or no water scarcity.23 of it can be farmed only at high environmental cost. Eliminating rural poverty Water availability. Like land, water resources and preparing outmigrants are also unequally distributed around the world.21 Africa and Latin America again have large unex- The social challenge of the rural sector over the next ploited water resources. According to the Interna- 50 years is enormous. Not only must it feed the tional Water Management Institute (IWMI), in world and prepare some 2–2.5 billion people to be- Africa only South Africa has exploited more than 20 come productive urban citizens but it can also cre- percent of its potentially usable water resources, and ate the preconditions for the evolution of responsive, in Latin America only Cuba (48 percent) and Mex- inclusive local and national institutions (as discussed ico (27 percent). Despite the modest exploitation of in chapter 3). water resources in Africa and Latin America, future For poor developing countries with large agricul- irrigation development is expected to be limited. tural sectors, rural growth has a powerful effect in Most good sites for irrigation—flat, close to water, pulling people out of poverty.24 In rural economies, near good markets—have already been developed. the more equal the incomes and assets, the more Data from IWMI reveal China as the most water- powerful the growth effect in poverty reduction.25 stressed country in East Asia, exploiting 44 percent As inequality increases, the linkage of growth to the of its usable water (in the aggregate) and projected to poor weakens, and in the most unequal of rural exceed 60 percent by 2020. Primary withdrawal of economies, growth tends to bypass poor people com- more than 60 percent is widely considered by water pletely. The quality of rural development is thus a experts to exceed the environmental carrying capac- basic determinant of the quality of the future social ity of a river basin system. Although China’s aggre- development of a country. Countries that let rural in- gate use appears still to be reasonable, it has several comes and assets become concentrated in the hands basins that are severely stressed environmentally, and of a few find it extremely difficult to lift poor people it faces a serious groundwater overdraft in the North out of poverty later (box 5.2). They have painted China Plains. According to IWMI data, withdrawals themselves into a corner. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Figure 5.2 Regional variations in water scarcity Percentage of countries 100 Percentage of usable water 80 Mining* > 100% Environmentally 60 overexploited, 60–100% Comfortable, 20–59% 40 Low exploitation, < 20% 20 *Extraction rate greater than replenishment rate. 0 Sub-Saharan Latin America East Asia South Asia Middle East Africa and the and Pacific (6 countries) and North Africa (29 countries) Caribbean (10 countries) (12 countries) (19 countries) Regions Source: IWMI (2001). Breaking the poverty cycle sume goods of predominantly local origin, produced and preparing outmigrants by people who are also poor. This sets up a chain of The conditions to break the poverty cycle and bring increased demand and incomes that cascades through the rural poor out of poverty are overwhelmingly as- the hands of poor people in the rural and urban sec- sociated with increasing rural-urban interactions and tors. An extra dollar of income in the hands of a poor more intensified use of existing agricultural land. In farmer might lead to 50 cents worth of demand for more dense rural areas with towns, credit markets are products and services from other rural poor, 40 cents more apt to exist, and land is more likely to qualify of demand for products produced by poor urban res- for collateral. Higher farmgate prices associated with idents, and only 10 cents of demand for products better roads and proximity to urban markets, and produced in the formal urban sector or imported more opportunities for spreading risk, encourage from abroad. In the hands of a rich farmer, nearly the higher-input agriculture. This in turn leads to greater whole dollar escapes the economy of poor people. value produced per unit area and generally to more The second pillar of rural-based economic growth off-farm jobs in food processing, transportation, and is the training effect of rural nonfarm employment. in the agricultural service industries. A more diver- Besides allowing the rural family to diversify income sified economic base, dynamically linked to farm sources and reduce risk, rural nonfarm employment towns—and eventually cities—provides more stability is an important stepping stone to urban skills. A of incomes throughout the economy. Education and young rural man or woman who gets a job with the health services can also be provided more efficiently.26 fertilizer or farm implement dealer makes a smooth The first pillar of rural-led economic growth that transition to a rural town economy—developing the reaches poor people is a “virtuous employment mul- urban skills and often the opportunity to later move tiplier� and a transition to urban life through rural to the city. This process of transformation and growth nonfarm jobs.27 Driving the virtuous employment has the added benefit that it leads to greater support multiplier is the tendency of the rural poor to con- for schools. As parents observe children in the com- (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  Box 5.2 the variation in the records of growth of North and Poverty, equitable growth, and path dependency South America.30 While in North America, early set- tlement experience led to a virtuous cycle of inclu- Inequality reduces economic efficiency and traps societies sive institutions, in Latin America inequality and ex- in bad development paths through inequality-perpetuating institutions in three ways: clusive institutions became the rule (box 5.3). Ⅲ Inequality reduces the participation of poor people in po- Increasing the value of smallholders’ assets litical processes, both directly and indirectly. This in turn Urbanization has been rapid over the past several reduces the likelihood that poor people have access to education, health care, and other services that would decades, and in many countries, especially countries contribute to growth. in Latin America, most rural residents have already Ⅲ Inequality can hinder the establishment of independent migrated to the cities without physical or financial and impartial institutions and the enforcement of bind- assets or human capital. At the same time, institu- ing rules, because they might reduce the benefits of the privileged. tions have not evolved to include them in the devel- Ⅲ Inequality makes it easier for the wealthy to hold out in opment process. In Asia and Africa, however, most political bargaining, either directly or through capital people still live in the countryside. And it is from flight. It therefore makes it more difficult for societies to respond quickly and optimally to external shocks. these countries that a substantial part of the 2 to 2.5 billion increase in urban populations expected in de- Each of these effects prevents the emergence of institu- veloping countries by 2050 will occur. In these coun- tions that would distribute incomes, assets, and opportu- tries a strategy to enhance the range of assets to nity more widely. which rural people have access would simultaneously Source: Binswanger and Deininger (1997). strike a blow against rural poverty, stimulate an in- clusive orientation in institutional evolution, and prepare migrants to become good urban citizens.31 Water control is critical to ensure and minimize munity successfully negotiating the rural-urban dy- climatic risk—and to allow farmers to safely invest namic, they place a higher value on better education. in increased production. Unlike Asia, where 30 to 35 percent of agricultural land is irrigated (producing Governance and the distribution of rural assets two-thirds of the agricultural output), Africa irrigates As chapter 3 showed, path dependency arises when only 1 to 5 percent of its agricultural land (depend- institutional rules lock countries into bad develop- ing on the classification of traditional water manage- ment paths that prevent them from meeting future ment).32 Without mechanisms to control risk, on- challenges. For example, the sharpest drops in growth farm investment will be restricted to intensifying after the economic shocks of the 1970s were in coun- labor inputs. Farm and landscape-scale investments tries with divided societies and weak institutions for may be needed to enrich soil nutrients and organic conflict management.28 Similarly, the recent stagna- matter, to protect fields from water and wind ero- tion in Argentina and the República Bolivariana de sion, and to regenerate natural vegetation to provide Venezuela can be traced to an inability to settle dis- local ecosystem services that also benefit agricultural tributive conflicts. Investigations into the determi- production. Where investment can take the form of nants of growth have consistently found a negative improving water control, and the water-retaining relationship between initial inequality (usually un- characteristics of the soil, climatic risk may be low- equal distribution of income or wealth in the rural ered considerably (box 5.4). sector) and subsequent growth.29 This effect can be A broad-based attack to help African farmers traced through the short and medium terms, as dis- break out of the vicious cycle of poverty will require cussed above in the context of the virtuous employ- increased investment on all fronts, including more ment multiplier, as well as through the very long term attention to low-intensity agricultural techniques (relevant for the 30–50 year time horizon here). As and genetically modified crop technology in areas discussed in chapter 3, differences in inequality in remote from markets. And it will take major public wealth, human capital, and political power stem- investments in water control, transport, and tech- ming from colonial experiences account for much of nological advance aimed at staple food crops.33 (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 5.3 Land distribution and path dependency The powerful impact of initial land allocations on subsequent specified number of years. Owner-operated farms dominated agricultural development is well illustrated in Brazil. In most of agricultural production, with rentals and sales merely reallocat- the country in the late 19th century land could be titled only in ing land to more efficient farm families working plots of com- lots of 4 square kilometers (988 acres) or more—an area much parable size. U.S. agriculture became one of the most produc- larger than a family could work. Restrictions on subdivision tive systems in the world—and remains so today. kept landownership highly concentrated. As a consequence, Based on an agrarian structure consisting predominantly of Brazilian agriculture became dependent on wage labor, char- family farms, the major Southeast Asian economies—Indone- acterized by relatively low efficiency and investment. Invest- sia, Malaysia, and Thailand, following the earlier lead of Taiwan ment and productivity rose only after government subsidies (China) and the Republic of Korea—and China reduced agricul- brought about rapid capitalization of the sector. tural taxation in the 1970s and started to support smallholders. Land sales were unable to significantly reduce inequality in These countries, in addition to establishing favorable macro- the size distributions of the holdings. Brazil became a largely economic policies, invested in rural infrastructure and social urban society without ever having developed a significant pres- services. They also provided research and extension services ence of appropriately sized family farms. As a result, much of and supported viable smallholder credit systems. Agricultural the rural population moved as wage labor to the cities, with- output grew rapidly, and the number of rural households living out assets and unprepared for urban life—and over half the re- in poverty fell dramatically. maining rural population is in poverty. Undoubtedly, the dy- In 1978 China abandoned collective agriculture and as- namic of this rural-urban transition is a major contributor to signed most agricultural land to families, giving each a very Brazil’s having one of the world’s highest levels of income in- small holding. It also sharply increased the prices paid for agri- equality. Recently, Brazil has invested heavily to overcome this cultural goods. Over the next 15 years farm output grew more early inequitable start. From 1995 to 2001 Brazil invested more than 6 percent a year. This dramatic increase in agricultural than $10 billion in land reform, settling some 584,000 families productivity precipitated China’s long-running economic boom on nearly 20 million acres. and reduction in poverty. In the United States the Homesteading Acts limited to 160 acres the plots that families could acquire. To retain ownership rights, individuals were required to cultivate the plots for a Source: Based on Binswanger and Deininger (1997). In Africa improved agricultural institutions may in Kenya, Malawi, Mozambique, Tanzania, Uganda, depend on first strengthening asset value through Zambia, and Zimbabwe. This approach uses various water and transport infrastructure and fertilizer (box combinations of fallow, phosphorus, and biomass 5.5). Many countries will follow a two-pronged strat- transfers with consistently good results. A program to egy—encouraging intensification and commercial- scale up these practices from tens of thousands to tens ization through purchased input-intensive agricul- of millions of African farm families would cost $100 ture in productive areas near urban markets and million a year for 10 years.34 transport (and in more distant areas with dense rural populations) and low-external input agriculture in How deep institutional structures impede research more remote areas. With the predominance of con- on the needs of poor smallholders strained soils, high production levels in Africa will Genes are already available that could help food pro- require use of both inorganic fertilizers and inputs to duction in the poorest countries if they were to be maintain soil organic matter and structure. In peri- transferred into poor people’s crops. These include urban farming areas where high levels of fertilizer in- genes that improve tolerances to salt, aluminum, and puts (inorganic or organic) are used, water quality manganese in soils; give plants greater resistance to in- issues may become a concern (as they have become sects, viruses, bacteria, and fungi; enrich beta carotene in intensive production areas of Asia and Europe), to correct vitamin A deficiency; create more nutri- requiring use of filter strips and other waterway tious oils, starches, and amino acids; and improve protections. fatty acid profiles and digestibility for animals.35 A soil fertility replenishment approach, developed Despite this promise for poor people, biotechnol- over the past 10 years by researchers from the In- ogy in general and transgenics research in particular ternational Center for Research in Agroforestry has have barely begun to be put to work to address the been adapted by tens of thousands of farm families problems of poor people. So far large commercial (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  Box 5.4 Breaking out through zais and tassas—low-input traditional technologies Since the early 1980s a technique for reclaiming degraded land The reaction of visiting Ghanaian farmers to what the has spread rapidly in Burkina Faso and Niger and was recently Burkanabé had accomplished bears noting: introduced into Ghana. This technique—called zai in Mossi and “We are very much blessed and yet we are complaining of tassa in Hausa—originated in the Yatenga region of Burkina our poor soils. A large proportion of our soils we have Faso. It involves digging holes that are 20–30 centimeters even discarded as ‘dead’ but, to our surprise, such soils deep and filling them with crop residue, household compost, are being used here [in Burkina Faso] to produce some- and manure. thing even better than we are doing on our best soils.� Many lateritic soils of the area had become impermeable, The zai is a response by farmers to population and climatic sealed by a thin crust, hardened by wind and water. The zais stress. The Yatenga region of Burkina Faso has some of the attract termites, and their underground tunnels increase water highest population densities in the country. Earlier versions of infiltration. Millet is planted in the holes, which protect seed- the zai were used in the Yatenga prior to the 1950s, but on a lings from wind damage. The number of zais per hectare varies much smaller scale. They were abandoned in the 1950s and from 12,000 to 25,000. Digging that hectare takes about 60 the 1960s because rainfall was much above average and the days (averaging five hours a day) in the dry season when work water harvesting properties of the zai were not needed. Re- demands are low. emerging in the drought of the 1960s and 1970s, they began This technique has tripled yields and greatly reduced yield to be rapidly adopted following improvements in the tech- loss in dry years. On the central plateau of the Yatenga some niques around 1980 by Yacouba Sawadogo, a farmer from the 100,000 hectares have been restored, promoted by Projet village of Gourga. The role of donors in dissemination, espe- Agro-Forestier (funded by the Oxford Committee for Famine cially through field visits and farmer trials, has been critical. Relief), and the German-funded Projet Agro-Ecologie. In many cases farmers spontaneously adopt the technique after see- Source: Reij, Scoones, and Toulmin (1996); Millar (1999); Meitzner and ing the results on others’ fields. Price (1996); and IFAD (1999). Box 5.5 little commercial interest in working on crops with Breaking out through fertilizer: the next green limited markets, or funding research for the needs of revolution? poor producers. It is also because the rules that make this technology available worldwide—about the shar- During the 1960s the fundamental cause of declining per ing of proprietary knowledge, products, processes, capita food production in Asia was the lack of short- statured, high-yielding varieties of rice and wheat. Asian and genes—are being defined in U.S. courts, based food security was only effectively addressed with the ad- on U.S. case law. The other high-income countries vent of improved germ plasm. Then other key aspects that are catching up, but the developing countries, where had previously been largely ineffective (enabling govern- needs are greatest, are being left behind.37 ment policies, irrigation, seed production, fertilizer use, pest management, research and extension services) came The use of modern biotechnology (genetic engi- into play in support of the spread of the new varieties. The neering) in agriculture has left the world sharply di- need for soil fertility replenishment in Africa now is analo- vided, though not always along predictable lines.38 gous to the need for the “Green Revolution-type� germ Some applications generate little controversy, such plasm in Asia three decades ago, a belief that is supported by two of the “fathers� of the Green Revolution: Norman as marker-assisted genetic selection. Others, such as Borlaugh and M.S. Swaminathan. transgenic organisms, cause much concern. Comfort with the new technology is determined in large mea- Source: Sanchez and others (2001). sure by a society’s comfort with its scientific and food safety institutions, as well as its feelings about emerg- ing concentrations of economic power in multi- plantings of transgenic crops have been restricted to national life-sciences corporations.39 Poor farmers Argentina, Canada, and the United States, with other and poor consumers have not yet been given signifi- countries planting less than 2 percent of the world cant voice in the decisions on agricultural use of these total.36 This is partly because much of the research new technologies. supporting this technology is locked into patents The current controversies over biotechnology re- held by a small number of multinational, vertically sult from twin revolutions in science and in property integrated life-science organizations, which have had rights: (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Ⅲ A scientific revolution in understanding the struc- have a foundation in related earlier experiences (see ture of genes and regulatory gene sequences (ge- box 5.6 on pesticide stockpiles in Africa). nomics) and in bioinformatics has created an Transgenics and developing countries. The slow unprecedented opportunity to improve the char- progress in applying transgenic research to the prob- acteristics of plants and animals, either through lems of poor farmers is due to the exclusionary as- more rapid and efficient marker-assisted breeding pect of IPRs, as well as issues of biosafety, food safety, or through the insertion of new genes into plants consumer choice, trade, and the publicly funded re- by genetic transformation. search network. Solutions to these complex issues are Ⅲ An institutional revolution in the coverage of intel- all playing out against a backdrop of globalization- lectual property rights (IPRs), especially in the related uncertainty that leaves many people unsettled United States, has resulted in rapid growth of the about their incapacity to control their lives and their private sector in crop genetics. environment. Deepening IPRs have created a revolution in agri- These revolutions have generated associated con- cultural science and a race for (exclusive) property cerns about food and environmental safety, and eth- rights for agricultural and biological knowledge. This ical and equity consequences, all within a context of race has led to greater concentration of the life- society’s lagging institutional capacity to keep up, es- sciences industry, including massive buyouts of seed pecially in the developing world. These concerns companies in all countries. These global giants raise Box 5.6 Science, technology, and institutions to solve the challenges of nature: obsolete pesticide stockpiles in Africa Science and technology are important tools to address the Africa’s shores. The stockpile problem went from being a local forces that damage livelihoods and affect quality of life. To take problem to being a local and global problem, with shared re- advantage of these tools requires effective supporting institu- sponsibility among all parties. African countries do not have the tions. In Africa, weak institutions, unable to regulate and dis- specialized industrial facilities to dispose of the pesticides, tribute pesticides, have left a legacy of dangerous, obsolete hence the stockpiles need to be collected and transported back pesticide stockpiles. to hazardous waste disposal centers in industrial country loca- In agriculture, insects and other pests have a potentially tions, mainly Europe. devastating impact on crop yields throughout the world. One Donor agencies underestimated the institutional prerequi- of the most dramatic examples is locust infestation in Africa. sites necessary for the correct application of pesticides, which To address this and other pest infestation over the past 40 would have allowed African countries to realize the beneficial years, the donor community has provided loans, grants, and effects while minimizing the negative impacts. Donors have other transfers to cover the procurement of pesticides as im- also underestimated the difficulties in mobilizing global action portant components of agricultural projects. However, for this to address the problem. At the initiative of the FAO and a few action to be successful, ministries of agriculture, port authori- other donors, modest clean-up activity began nearly a decade ties, transport and handling networks, storage agencies, shop- ago, but has moved slowly. Less than 5 percent of the esti- keepers, extension agencies and farming communities need mated stockpiles have been disposed of, and new additions of to know how to manage the timely purchase, transport, stor- obsolete pesticides continue to accumulate faster than the dis- age, application, and disposal of these chemicals. If this coor- posal rate. Recognizing that a gradual, piecemeal approach dination is done well, the chemicals contribute to well-being. would not solve the problem, in December 2000 two NGOs If not, they become a social and environmental liability. (WWF and Pesticides Action Network, or PAN U.K.) launched Coordination problems combined with misguided advice or the idea of major clean-up actions under an “Africa Stockpiles ineffective development assistance from donor agencies and Program.� The 15-year program, estimated to cost $250 mil- pesticide manufacturers have resulted in a stockpile of more lion, would be donor funded and managed in cooperation with than 50,000 tons of extremely toxic and now obsolete pesti- the partners and member governments. The program includes cides, some dating back 40 years, and tens of thousands of country level inventories, clean-up and disposal, and extensive tons of severely contaminated soils that have to be shipped technical assistance to avoid recurrence of the problem. The out. An estimated 30 percent of the pesticide waste is believed challenge now is to coordinate the process among the differ- to be made up of highly persistent organic pollutants (POPs), ent parties involved. which have seeped into the water tables and oceans, and which are moving around the globe through ocean currents, Source: GEF, Africa Stockpile Program; interviews with World Bank contaminating the food chain of many marine species far from staff, 2002. (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  fears about biosafety and food safety, heightened by erably strengthened public sector role in doing and tragic institutional failures in the areas of feed, food, directing basic research. and drug safety in Europe. Proprietary agricultural science and the dominance of Who has the greatest need to access these new the private sector. The dominance of the private sec- agricultural technologies? The rural poor in develop- tor in agricultural bioscience is relatively new. As ing countries. The precautionary principle tells us long as farmers could replicate improved plant and that we should err on the side of caution, look at al- animal varieties in the field, there was little scope for ternatives, and ensure a fully transparent and demo- the private sector to recapture the costs of invest- cratic process. Applying this principle suggests doubt ment in improved varieties. So, to promote the pub- about the need for genetically modified organisms lic interest in higher yields and food quality, the (GMOs) in high-income countries, where oversup- public sector has traditionally borne the bulk of agri- plies of food and low food prices are associated with cultural research spending. It was only with the in- subsidies of $360 billion a year. It will certainly lead troduction of hybrid technology in the 1930s that to a different decision in countries in Africa, where the private sector became interested in investing in the rural poor depend critically on one or two crops crop biology. Why? Because hybrid varieties lose grown under precarious conditions. their high-yielding characteristics if seeds harvested Poor people need a stronger voice in international from them are used for planting. Farmers who want discussions of these matters, and science and govern- the benefits from high-yielding hybrids must pur- ments must find mechanisms to improve the quality chase new seeds every year if they want to maintain of the debate. This requires sorting more clearly high yields. This makes it possible for seed com- what is known in current science from what is not, panies holding the parent lines to appropriate the so that the political process can act more effectively benefits from research-induced increased crop yields. on what is not known. If informed, representative Incentives for private-sector agriculture R&D stakeholders decide to move forward with genetic were strengthened in 1980 when the U.S. Supreme engineering, that would also require credible and in- Court decided that although patent protection could dependent monitoring—and systems to identify and not be extended to naturally occurring living things, intervene early, when unforeseen problems arise. The it should apply to living organisms that had been al- worldwide problems with invasive species should tered by human intervention.41 Later interpretations serve as a warning on the need for due diligence in extended this coverage to new processes, which may introducing new organisms to nature. or may not give rise to a new product. The 1995 The twin revolutions in science and property Trade-Related Aspects of Intellectual Property Rights rights have created challenges and opportunities for (TRIPS) agreement has further established an inter- public research institutions, life sciences depart- national institutional framework of minimum stan- ments in universities, national agricultural research dards for international trade involving intellectual institutions, and the international agricultural re- property, including for proprietary agricultural pro- search centers of the CGIAR. The opportunity is cesses and products. that stronger IPRs have unleashed an extremely dy- Three factors explain the private sector’s domi- namic race for exclusive property rights in agricul- nance: tural knowledge. And public research is in the race. The race has generated new agricultural knowl- Ⅲ The deeper protection of intellectual property, which edge at an unprecedented rate. But the challenge is allowed firms to move into more basic research, twofold. First, how can research be channeled to traditionally left to the public sector. benefit poor people in developing countries, who re- Ⅲ The race for property rights among firms and be- ally need it? And second, how can this be done in a tween firms and the public sector, strongly influ- period of declining public financial support for pub- enced by eager capital markets. lic agricultural research? This situation calls for new Ⅲ The weakening of taxpayer support for public sector public-private research—and new institutional mod- agricultural research. Indeed, with annual agricul- els for regional cooperation.40 Ensuring that this tural subsidies in excess of $300 billion, glutted research reaches poor people also calls for a consid- world markets, and record low prices, the case for (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  public support of agricultural innovation was, for tragedies that occurred elsewhere.46 This confidence many taxpayers, weak. The prevailing mood has has enabled the U.S. Food and Drug Administration become “leave it to the private sector.� (FDA) to resist industry pressure to reduce its scru- tiny of new products.47 A divided public. Support for the application of Europe, by contrast, has been buffeted by a series transgenics to the food needs of poor people has of food and drug safety issues, precisely as genetically been nearly unanimous among the major agricul- modified foods were beginning to enter the market. tural development institutions.42 Even so, discussion “Mad cow disease� (bovine spongiform encephalopa- of the use of transgenics to address the development thy, or BSE), suspected of killing at least eight peo- needs of poor people has been heavily influenced by ple and leading to the slaughter of herds in Britain public concern, especially in Europe, and the use of worth $5.5 billion, rocked public confidence in gov- transgenic agricultural technology has been over- ernment and the agri-food industry’s credibility and whelmingly geared to the needs of large, mechanized capacity.48 According to a British parliamentary re- agriculture. port released in February 2000, BSE created a “crises In Europe consumer groups, led by Greenpeace of confidence� in both science and government.49 It and Friends of the Earth, have driven genetically was observed that British citizens were more likely to modified foods off the shelf—and subsequent do- trust science they see as “independent,� with univer- mestic production and imports of genetically modi- sity scientists ranking at the top and government at fied products have largely halted.43 In a European the bottom. The rest of Europe has also suffered re- Commission survey 56 percent of the respondents cent crises of confidence in government’s ability to felt that genetically modified food was dangerous.44 protect them. France suffered from government fail- Similarly, consumer groups and NGOs prevented ge- ure in its scandal over AIDS-tainted blood, which in- netically modified soybeans from entering Brazil, de- fected 3,600 people receiving blood transfusions spite the support of the minister of agriculture and in the mid-1980s. Belgium, too, experienced high- the head of the national agricultural research agency, visibility food and agricultural scares in the 1990s.50 EMBRAPA; local and international NGOs have also European distrust of government food-protection led resistance in China, India, and Kenya.45 In North institutions has led to demands to be directly in- America, by contrast, the public attitude is optimistic formed and directly involved. Given this desire, the about the promise of modern biotechnology, though precautionary principle (box 5.7), widely supported with concern over possible environmental effects. in Europe, has three attractive characteristics. First, Clearly the perception of risk differs strongly it slows the commercialization of new crops. By between scientists and the public in Europe, and putting a greater burden of proof on promoters to between the general public in Europe and North show that new organisms will not create environ- America. Caught in the middle are the rural poor, mental or food hazards, the precautionary principle, especially those on dry or degraded lands who could most benefit from the new technology. Box 5.7 Explaining the differences in perceptions of risk. The The precautionary principle differences in risk perceptions between North Amer- ica and Europe appear to be due in large measure to Ⅲ When an activity raises threats of harm to human health differences in their confidence in their life-science or the environment, precautionary measures should be institutions to accurately pick up risk signals and to taken even if some cause-and-effect relationships are communicate them to the public. These differences not fully established scientifically. Ⅲ The proponent of an activity, rather than the public, have a base in experience. While North America has should bear the burden of proof. had no catastrophic failure in the food and drug pro- Ⅲ The process of applying the precautionary principle must tection system, Europe has experienced numerous be open, informed, and democratic, including potentially failures, especially over the past decade. Rightly or affected parties. It must also involve an examination of the full range of alternatives, including no action. wrongly, the U.S. public’s confidence in its food and drug safety institutions has been constantly re- Source: Adapted from The Wingspread Consensus Statement on inforced. Beginning with Thalidomide at the end the Precautionary Principle (http://www.sehn.org/ wing.html). of the 1950s, it has avoided the food and drug (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  as implemented in some national laws, slows the institutions responsible for biosafety, the institutions approval process and allows more time for the pub- concerned with food safety and consumer choice, the lic to become informed. Second, the principle calls rules for international trade, and the organization for more transparency in environmental, food, and and role of public sector research (table 5.1).53 health-related public sector decisionmaking. Third, For biosafety, food safety, and consumer choice where substantial scientific uncertainty exists, it re- and trade, active sensing mechanisms are in place, quires that decisions be made through an informed largely through NGOs and formal and informal political process. References to precaution have been networks. The Internet has greatly increased the adopted in the Cartagena Protocol on Biosafety, power of networking. Mechanisms to balance in- which regulates the shipping of certain genetically terests are weak across all five dimensions. Much modified organisms (referred to in the protocol as greater attention is required to frame the debate so “living modified organisms�) across national borders. that the public can better understand the benefits and risks by making independent positions, agreed Poverty and the precautionary principle on by scientific leaders, accessible to the public. Im- The precautionary principle requires analysis of the plementation of institutional remedies in each of alternatives to introducing new technology and or- these areas is taking place within a global context, ganisms, not just the risks inherent in their use. In and frameworks are emerging to forge international Africa, in many marginally viable agricultural areas, agreements. Because these frameworks often do not the range of alternatives is currently minimal. For reflect satisfactory consensus at the national level, farmers in these areas, modified crops that can bet- their implementation path can be expected to be ter survive prolonged drought, or that improve diets rocky. through micronutrient enrichment, may be among More importantly, to successfully introduce more the few critical options, to be supplemented with in- science and technology into neglected areas of devel- vestment in farm- and landscape-scale soil and water oping country agriculture, a long-term commitment improvements.51 to the development of agricultural knowledge and Efforts to improve the lives of farmers in arid and supporting institutions must be in place—particu- semi-arid borderline areas through conventional larly in Africa (see box 5.8). breeding programs have had limited success. Crop breeders have had limited success in either raising Intensifying the use of land the yields of robust but low-yielding millet or get- Creating more dynamic, input-intensive agricultural ting improved resistance to moisture stress from communities in near-market areas, where price in- fragile but higher-yielding hybrid maize. Both these centives make intensification attractive, requires objectives have resisted traditional breeding pro- sustained effort on several fronts.54 First, conditions grams but have become credibly feasible with genetic must be established to activate the land market and modification.52 The precautionary principle requires make land affordable to smallholders. Second, small- weighing the alternatives and the costs and benefits holders must have access to credit to make the hold- of helping these farmers. As noted above, the devel- ing a viable economic unit. And third, smallholders opment community supports a broad consensus that must have enough protection from risk to be able to alternatives that can be adopted on a large scale are afford investing in yield-increasing inputs, such as few and that risks from transgenics can be managed. fertilizer and improved seed. The challenge is to focus research and promote Three activities are important to move additional knowledge sharing for improving “orphan crops�— land into the hands of smallholders: clarify and ad- crops important to poor people but that have not judicate property rights, improve the functioning of had the benefit of significant research because of lim- land sales and rental markets, and where necessary, ited market potential. redistribute land through land reform. Disentangling the elements. The ethical, moral, and Clarify and adjudicate property rights as land social dimensions of biotechnology challenge soci- scarcity dictates. Countries making the transition ety’s institutions at all levels, from the field to the from communal to more individual forms of land planet. They challenge notions of property and prop- ownership need to set up a legal framework that per- erty rights institutions, the national and international mits evolution of land rights toward individualized (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 5.1 The capacity of institutions to sense problems, balance interests, and implement solutions Where we are in Where we are Where we are identifying problems in balancing interests in implementing solutions Problems with Innovators (especially in IPR debates not well accompanied IPR implementation varies widely, intellectual property pharmaceuticals) felt investment or understood by public. Minimum as a function of national income.57 rights regime threatened by weak property standards of IPRs established in IPR implementation will strengthen rights. Legal costs of innovative the Uruguay Round. Menu of ac- as countries become potential research now a barrier for public ceptable IPR options large.55 Many exporters (to meet intellectual sector researchers. Concern about developing countries and NGOs be- property (IP) standards of importing effect of strong IPRs on concentra- lieve that IPRs, based on Western countries). tion in life sciences industry. concepts of law, are unethical or impractical in developing countries.56 Biosafety Old concern over invasive, exotic, Developed countries have minimal United States screens GMOs using wild species. Nature documents voice regarding tradeoff of bio- same standards as non-GMOs. Bt-maize pollen threat to Monarch safety risk against food security. U.S. Department of Agriculture butterflies.58 Monsanto sued for Arguments are inaccessible to lay (USDA) responsible for implemen- Bt “creep� into organically grown public. Precautionary principle ac- tation—except for Bt products neighboring crops. Subsequent cepted in Cartagena Protocol on which must also clear EPA. Euro- research shows low threat to Biosafety, but mechanisms to pean legislation tends to require Monarchs under field conditions.59 resolve disagreement weak. screening hypothetical risk. Euro- Genetically modified crops found pean public pressure halts new poorly competitive outside of approvals in 1998. cultivated conditions for which designed. Food safety and No documented cases of human European public pressure pushes U.S. FDA applies the same consumer choice health problems from eating a “strong precautionary principle� standards of food safety as for commercially marketed GMOs. requiring screening for hypothetical non-GMO crops.61 EU adopts Normal testing catches potential risk. U.S. Academy of Science labeling based “on consumers’ allergy problem. Scientific opinion declares no scientific basis for right to know.� GMO food safety in Europe and the United States tougher screening processes for low priority in developing countries agrees: “there is no problem with GMOs than non-GMOs. European compared with clean water and GMOs over and above any other preference for non-GMOs drives uncontaminated meat. food.�60 down price of GMO crops. U.S. corn acreage in GMOs falls. Liability concerns arise for pollen contamination. Trade Developed country consumer and Contradictory positions. World Protocol establishes Biosafety environmental groups question Trade Organization (WTO) require- Clearing-House to exchange trade in GMOs. Concerns related ments, based on Uruguay Round information on living modified to globalization-related “loss of (SPS agreements), require restric- organisms and to assist countries control� and multinational industry tions to be based on scientific to implement. Calls for assistance concentration, as well as to assessment of risks. Cartagena Pro- from developing countries in specific food and environmental tocol on Biosafety, endorses “pre- capacity building. safety issues. Industry concerned cautionary principle�—that “lack of with disguised protectionism and scientific certainty due to insuffi- theft of IP. cient relevant scientific information and knowledge� should not prevent states from banning imports. Public research With food abundant, prices low, Difficulty balancing (a) “pure re- Must develop national and investment and agricultural subsidies high, search,� (b) research on improving international public sector research developed countries’ support for the productivity of crops without programs that (a) are oriented public agricultural research is weak. significant commercial markets, but toward clear national objectives, of importance to poor people, and and (b) complement (through (c) research on yields for major sta- partnerships), and do not duplicate ples where yields are approaching other public sector and private genetic maximums. sector activities.62 (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  Box 5.8 Institutional commitment and African agriculture: lessons from Asia and South America The current pessimism about a continent’s ability to feed itself ful development of agricultural knowledge is measured in has been seen before—in Asia. The pessimists were proved decades, not years, and requires long-term national and donor wrong about Asia. But it took a long-term, coordinated com- support. The complexity of agricultural systems in Africa have mitment to agricultural research, extension, and agricultural foiled attempts to “jumpstart� a Green Revolution in Africa. higher education. These are the lessons to be learned from the Success in Africa will require local adaptation, piloting, and dis- Green Revolution in Asia and South America. semination. Where agricultural institutions are strong, success “Asia’s development experience reveals that a bleak eco- has been achieved. Zimbabwe’s Green Revolution in maize, nomic future for Africa in the 21st century is not foreordained. which led to a reliable maize surplus and generated maize ex- There are scores of cracked crystal balls in economic forecast- ports for 19 of 21 years during 1970–91, was based on 28 years ing. Even Nobel Laureates such as Gunnar Myrdal can widely of indigenous research, which in 1960 produced the high-yield- miss the mark. Myrdal was pessimistic about Asia’s develop- ing maize variety SR-52. Unfortunately the technological lead- ment prospects in the late 1960s because of corruption, ‘soft ership that made this possible has been allowed to erode. states,’ rapid population growth, and the gloomy prospects for The long-term commitment required to develop agricultural agriculture. But Myrdal failed to anticipate Asia’s Green Revo- knowledge institutions does not resonate comfortably with lution, which was taking root at the same time that his book, today’s aid climate. Today’s climate favors rapid “results on the Asian Drama, was rolling off the press in 1968. The rapid ground,� and prefers direct field involvement, often through spread of Green Revolution wheat and rice varieties in Asia NGOs. Donors are right to demand accountability for aid re- in the late 1960s and early 1970s, and China achieving the sources. But experience shows that there is no shortcut to agri- fastest rate of agricultural growth in the world from 1980 cultural knowledge. Sustained agricultural progress in develop- to 1995, highlight the perils of economic forecasting.� Even ing countries will require a long-term commitment from donors Bangladesh, long considered a “basket case,� has recently and partnering with local agricultural research extension and emerged as an agricultural success story. In Latin America, higher education institutions. Brazil and Chile have become aggressive competitors in global food markets. Sources: Carl Eicher, Institutions and the African Farmer, CIMMYT Eco- nomics Program third distinguished economist lecture. On the Web What these countries have in common is a long-term com- at http://www.cimmyt.org/Resources/Publications/cat-log2001/Pub mitment to agricultural research, extension, and higher educa- Cat2001-Economics.htm; Zimbabwe’s Maize-Based Green Revolu- tion—to the development of what Carl Eicher has labeled the tion: Preconditions for Replication, World Development, vol. 23, no. 5, “Agricultural Knowledge Triangle.� The time for the success- pp. 805–808, 1995. tenure as the need emerges with increased commer- connections or power positions.63 Experience in Bo- cialization and land scarcity. Where tenure arrange- livia, El Salvador, Peru, and Thailand demonstrates ments have been severely disrupted by civil strife and that area-based titling can be accomplished by intro- war, collectivist land reform, or land-grabbing by in- ducing titling in combination with a mechanism for fluential individuals (Bolivia, Cuba, Ethiopia, Hon- dispute resolution. duras, Nicaragua, Tanzania, Uganda, and Vietnam), An on-demand approach can be justified under an approach is needed that adjudicates among over- four conditions. Commercial agriculture is just be- lapping claims and establishes clear ownership rights ginning to emerge as a profitable enterprise for the at minimum cost. most innovative and progressive producers. Tradi- Where insecurity of tenure already affects incen- tional community values and norms are still strongly tives, land titling should be initiated. Area-based ti- enforced. Local political power in the formal govern- tling is important where the insecurity results from mental structure is constrained effectively by tradi- attempts by the powerful to wrest land from the less tional political structures and traditional authority. powerful. Under these conditions an “on-demand� And national administrative systems extend the rule program may increase the ease of land grabbing, and of law to the local areas. an area-based system is more appropriate. An exam- Under these conditions the emerging entrepre- ple would be the rehabilitation of an irrigation neurial commercially oriented producers may need scheme in Somalia—where land values will increase more security of ownership than the majority of greatly with improved water access and where the farmers who are still largely subsistence producers. rule of law is tenuous, exposing those less powerful For example, the emerging entrepreneurial group to loss of land to individuals with strong political may need access to formal credit markets, which (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  typically requires greater security of ownership. Un- lump-sum local tax on land has the advantage of der such circumstances an area-wide titling project maintaining incentives to produce, and it provides would be expensive and inefficient because the cost revenue to local governments. In addition, by capi- would exceed the benefits for most parcels. Further, talizing local amenities in land values, a local land tax it may be possible to use the political strength and establishes a direct link between tax levels and bene- local support of traditional authorities to certify fits received by taxpayers.65 boundaries in an equitable manner without area- Depoliticize land reform and stress sustained pro- wide titling. Communities of exactly this type exist ductivity and poverty reduction. Because many land in many areas in central and western Uganda.64 reforms have taken place in response to political Improve the functioning of land sales and rental crises, with little sustained commitment to making markets. Restrictions have often been placed on op- the smallholders’ farms productive, the outcomes erations of land markets to compensate for failures have often been disappointing.66 in credit and risk markets—and for policies that raise Transforming a large farm into a workable set of the price of land above its expected value in agricul- smallholder enterprises requires a new pattern of pro- tural use, such as the use of agriculture as a tax loop- duction, subdivision of the farm, and construction hole, and agricultural credit subsidies directed to of infrastructure.67 So, realizing the productivity ben- large land owners. These restrictions on tenancy and efits of redistribution requires a shift from ad hoc po- sales contracts typically reduce the willingness of litical objectives to productivity and poverty-related landlords to make land available to smallholders objectives. Brazil, Colombia, and South Africa are through sales or rentals and so should be removed. implementing a new model of “negotiated� land Underlying market failures must be addressed di- reform. Although it is too early to draw definitive rectly. Taxing land can reduce the incentive for large conclusions, initial evidence is encouraging. Key landholders to hold unproductive land (box 5.9). A elements of this approach are:68 Ⅲ Emphasize sustainable poverty reduction through Box 5.9 Weakening the interest of landholders elaboration of integrated farm projects by poor in unproductive land people (which are then supported by a land pur- chase grant). In 1995 Brazil transferred collection of its land tax from the Ⅲ Decentralize execution and integrate into develop- land reform agency (INCRA) to the ministry of finance. ment objectives at the local level, with an overarch- With this change, the tax began to be seriously assessed ing emphasis on beneficiary training and human for the first time. In addition, Brazilian Federal Banks began capital formation. foreclosing on bad debt collateralized by land, and land prices fell because of Brazil’s entry to the Southern Com- Ⅲ Involve the private sector in project development, mon Market (MERCOSUR) and the ending of hyperinfla- financing, and implementation. tion (which had led to a flight to real assets). Ⅲ Work to build and maintain a broad-based con- These events began a process of weakening the inter- stituency, including landowners, rural workers’ est of landowners in holding unproductive land. With this weakened interest it became possible to mount a major unions, agribusiness, and agricultural research and campaign against irregular holdings. In the past few years extension. INCRA reclaimed more than 50 million hectares of irregu- lar holdings. Of these holdings 10 million hectares in the Intensifying the use of water Amazon region were turned over to the Brazilian Institute for the Environment to create protected areas. The rest With water scarcity rising, markets tend to emerge— became available for redistribution. According to the min- formal or informal, legal or illegal, peacefully or ister of land reform, there is now no constraint on land through violence. And the tendency will be for water available to be distributed to Brazil’s landless. The con- to go to its highest-value use given the infrastructure straint is budgetary resources required under the Brazilian Constitution to help land reform beneficiaries set up work- in place. Although this provides some comfort, be- able production units. cause the scope for improving efficiency of water use through better pricing is great, the institutions un- Source: World Bank staff. derlying those markets will determine the social cost of the transition to markets and the efficiency and (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  equity of infrastructure put in place to make water Box 5.10 transport and water markets possible.69 The race for water—and land—and the displacement of the poor As for land, emerging legal protection of property usually confers rights on the first users. But gaining In some arid regions of Sub-Saharan Africa, land rights are truly secure property rights is more complicated with less important than rights to use water at specific loca- water than it is for land. Physical distance and dif- tions such as an isolated waterhole. Rights to graze land ferent legal and administrative jurisdiction (states, might be open to the entire community or even several regions, nations) complicate agreements between communities, but rights to water were restricted. Control over the water meant de facto control over the land, just upstream and downstream users. Guaranteed quan- the reverse of the frontier setting of the Americas. So hav- tities become meaningless as quality is degraded. ing a water right was far more important than having a And the complexity of hydrologic systems and un- right to graze the land. predictability of climate make “ownership� a contin- With the advent of a modern technology—the bore- gent concept. For these reasons, property rights in hole well—the constraints to grazing were suddenly re- laxed, and it became possible for wealthy or powerful in- water have emerged only in situations of scarcity, dividuals to gain access to land and grazing opportunities and the nature of the rights varies according to the through implementation of the new borehole technology. water basin’s peculiarities. The tribal lands grazing policy (TGPL) program in Botswana The competition for water has two stages: one for promoted private ownership of grazing land and borehole water points under the theory that individual ownership access to cheap water and one for rights to water. would provide incentives to maximize the returns from The competition for access takes the form of com- grazing, maintain or improve the quality of the range, and petition for property rights for land that is well increase the rate of herd offtake and national income. The endowed with water (box 5.10). This means pur- result was that many of the private ranches came to be owned by wealthy individuals, and grazing, hunting, and chasing good, easy to irrigate land; land above shal- gathering opportunities for poor people declined. low aquifers; or land on which the government can be persuaded to provide low-cost water. In each case Source: Richard Barrows, personal communication. rights to water are less of an issue—in the early stages—than rights to land. Only as water scarcity emerges in a second stage, often from conflict with ing countries with strong central governments such urban or public use, does the irrigator seek to con- as Mexico (where redistributing water rights would, vert traditional uses of water to property rights. This in the words of a prominent reformer, “require a rev- principle of prior appropriation—the squatter’s rights olution�). It is also problematic even in developing to water—tends to be accepted in most settings. countries with a strong redistributional mandate (for Prior appropriation has pluses and minuses. A plus example, South Africa). The benefits of formalizing is that it tends to reward entrepreneurial behavior and the de facto rights of water users, and working with investment in productive resources. A farmer who these users to manage the resource in a sustainable fears losing rights to water is much less likely to invest manner, have in most cases been judged to outweigh in land leveling and irrigation infrastructure than a the drawbacks of reinforcing existing inequities.70 farmer who is sure that water will continue to be avail- If the settlement of rights becomes protracted, ne- able. Communities also prefer to see investment and gotiation strategies of individual claimants will lead economic activity than to see a resource lying idle. to a wasteful drawdown of the resources—and to There are two minuses. First, prior appropriation premature investment. A property rights regime that sharpens inequalities in incomes and assets by re- does not allocate rights expeditiously not only risks warding those with the initial capital to invest. Sec- wasting water, but also leads to uneconomic infra- ond, unless well administered, it leads to a destruc- structure designed to “lock in� water claims before tive race for property rights and loss of environmental other claimants do so—often years if not decades be- services of water. fore it is justified by emerging demand. For exam- Nonvoluntary redistribution of existing water use ple, pressure for a 2,000-kilometer conveyance sys- and associated rights has been proposed but found tem taking water from the San Francisco River to to be impractical—both in industrial countries such Brazil’s northeast—even though most of the urban as Australia and the United States—and in develop- demand justifying the project will not begin to (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  emerge for more than 10 years—comes in part from ronmental flows be established early. If environmen- a concern to guarantee Brazil’s poor and arid north- tal needs emerge only after economic conflict is al- east a claim on the San Francisco’s limited water. ready emerging, the problem of balance becomes Similar pressures exist between states in India.71 much more controversial. The function of a water The environmental use of water will not be pro- agency is to provide quality analysis and technical tected without specific institutional intervention. In information about the hydrological characteristics of the absence of protective institutions the environ- the basin, including analysis of alternative water mental use of water is priced at zero—every other sharing, trading, and pricing scenarios. use will establish a prior claim. If estuaries and fresh- Interests need to be balanced where they con- water ecosystems are to be maintained, institutional flict—the Dublin institutional principle. Where solutions have to be put in place to take into account problems are local, such as managing an irrigation the public goods nature of water. district, and the actors homogeneous, communities What are the major institutional issues for rural typically find mechanisms to resolve the conflict.72 communities in controlling the race for water? The Conflicts among widely differing claimants—for ex- first, “institutional� principle for water resource man- ample, the irrigation district and urban water supply, agement, is that water management should be carried with the hydropower dam an added complication— out at the lowest appropriate level—and be as partic- require basin-level mediation. Water parliaments, or ipatory as possible. The second, “ecological� princi- a similar institutional structure representing all basin ple requires the holistic management of water, in- stakeholders, become essential. Water parliaments cluding management of watersheds and guarantees must have technical backstopping from a strong and of maintenance of environmental values. And finally respected water agency (box 5.11). the third, “instrument� principle requires that water be managed as an economic resource. Widely known Box 5.11 as the Dublin Principles, for the 1992 Dublin con- Water parliaments in France ference where they were first developed and agreed on, these are the three principles that water special- Since 1964 water policy in France has been made at each ists agree that water management must respect. of six major hydrographic levels by a comité de bassin, an authentic river basin water parliament. The number of Picking up signals of environmental decay. Water seats varies from 61 to 114, with the composition fixed by problems are environmental and economic. Signals ministerial order and comprising three groups: of environmental decay are picked up by rural com- munities as a gradual loss of fisheries and the recre- 1. Users, qualified local dignitaries, and representatives of socioprofessional groups (40–45 percent of seats). ational and aesthetic value of water, as well as the 2. Representatives of the regional authorities (régions, dé- quality and supply of water for domestic use, and the partements, communes; 36–38 percent of the seats). presence of water-borne human diseases. These sig- 3. Representatives of the state (19–23 percent of the nals are often missed or interpreted incorrectly. If en- seats). vironmental flows are to be protected, basin-level This organization deliberately limits the influence of the expertise must be mobilized prior to the emergence state and reflects the desire to promote the role and re- of economic conflict for water. The economic signal sponsibility of the different actors—users and elected rep- is a growing disparity among the values of water in resentatives—in each basin and to encourage them to irrigation and in urban and industrial uses. For reach agreement. The basin committee must give its ap- proval on fees and the basis for their calculation. groundwater the early signal of a problem is falling The basin committee’s executing agency is the water levels, often noticed by local communities with shal- board (agences de l’eau), an administrative public body low drinking water wells long before it becomes a under the responsibility of the state. It has a dual role: problem for irrigators. Ⅲ Taking part in the financing of general works in the basin. Balancing the interests of all of water’s claimants. Ⅲ Carrying out water-related research studies. Balancing competing water interests requires a con- sensus on the technical nature of the problem. This Source: Chéret (1993). requires basin-level expertise. It is critical that envi- (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  Executing arrangements to share water. Action re- market reform—delaying or even preventing badly quires either sharing available water better or build- needed improvements in water allocation. If they are ing new infrastructure to bring in additional water, weak, they risk losing their economic base without or increasing the flow and quality of water through clear alternatives, as irrigation water becomes priced rainwater harvesting and landscaping improvements. out of their reach. Water management institutions have historically de- But if the rule of prior allocation were to assign veloped to address allocation of existing flows; much property rights to rural users, the transition to mar- more attention is needed to enhance institutional ac- ket allocation would be greatly smoothed. When tion on water production and use efficiency. Water, local farmers are owners of the water, they are un- even in industrial countries, tends to be very poorly equivocally made better-off by the market-induced allocated.73 So the scope to share water better, with- increased value of the water. They have a range of out resorting to new sources of supplies, is generally choices; they can continue farming as they were, sac- great, but incentives have to be in place. Where the rificing the income they could get by selling water; number of competing users is relatively small, ad hoc or they may upgrade by investing in water-saving solutions can be sought. For example, the water util- irrigation equipment and higher-value crops and sell ity can purchase high-efficiency irrigation devices for the surplus water created; or they may decide to sell the irrigation district (such as the Imperial Valley, all the water and invest in alternative livelihoods. California). Or the utility can pay the farmers not to Similar property rights would have to be assigned to irrigate at all (California water market and others). institutions protecting the environmental services of As the number of players increases, or as all easy so- the basin. lutions become exhausted, water markets become Getting ahead of the frontier most efficient for water allocation and coordination. Water basin commissions, or water agencies, must What drives the expansion of the agricultural fron- be established to open the dialogue for win-win op- tier into different wildernesses? Poverty and oppor- portunities in water sharing. Eventually water mar- tunity. In this section the forest frontier is discussed. kets will emerge through this process of ad hoc nego- But the pattern also applies to the conversion of wet- tiation and comparison of value.74 Water markets lands, grasslands, and other agriculture—or wilder- ensure that water in each basin gets to its highest value ness—boundaries. The settlement in the eastern use. They also ensure that price differentials clearly Amazon of Brazil has been shaped by Northeastern- signal the potential benefits of augmenting flows ers fleeing periodic droughts. Western settlements, through dams, reservoirs, or interbasin transfers. by contrast, tend to be populated by smallholders Moving to market allocations of water requires who sold farms in southern Brazil during the boom- two important acts, however. First, property rights ing land market of the mid-1970s and early 1980s to must be assigned: it must be decided who will have seek opportunities by buying cheaper land in the to pay whom for water. Second, environmental allo- Amazon, often gaining 5–10 hectares of land in the cations must be made—how much water has to be set Amazon for every hectare sold in the south.76 Simi- aside to maintain environmentally critical flows in the lar stories of expansion into the forests of Indonesia, basin? The market will not make these decisions. The Malaysia, the Philippines, and Thailand have ele- first must be determined by existing law and prevail- ments of poverty-induced flight and profit-driven ing notions of fairness, and the second through a opportunity.77 The frontier’s expansion highlights combination of technical and political criteria.75 In the importance of institutions to address a race for the absence of appropriate balancing mechanisms, property rights. many major rivers stop flowing into the ocean or in- Many forests were originally claimed by local land lakes. communities, but these claims were granted to gov- Water markets will eventually price low-value ernments during the colonial or postcolonial period. users out of the market. This can have two bad out- While governments have often faced great difficul- comes. If the community of low-value users is strong, ties in managing these (often huge areas of ) forest, well represented, and politically strategic, it can block lack of local rights and economic benefits from (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  forests have reduced local incentives to protect them. Box 5.12 The past 15 years have seen a major shift in forest The Amazon rancher’s decision to deforest tenure in developing countries back to local com- munities and indigenous peoples, such that 14 per- Pasture in the Amazon often degrades beyond economic use in some 10–15 years. The rancher then has to decide cent are now owned by communities and indige- whether to restore the pasture through plowing, fertiliz- nous groups, and another 4 percent are still publicly ing, and planting new pasture grasses or to plant on newly owned but reserved for exclusive use of communities deforested land. The decision depends largely on the rela- and indigenous groups.78 This shift in forest owner- tive cost of new land (net of sales of commercial value logs) and the cost of fertilizer and limestone to reclaim de- ship, plus the dramatic shift in many forest-scarce graded land. countries to tree growing on farms, means that com- In intensifying areas where the cost of new land is mercial forest product and ecosystem service markets above $300 per hectare, farmers will generally choose to could potentially contribute much more in the fu- reform pasture. But where land values are between $20 and $100 per hectare, farmers find it more profitable to de- ture to rural livelihoods than they have in the past, forest new land and abandon degraded pasture. By a con- particularly with reforms in market policies.79 servative estimate of the value of the carbon storage, the The process of conversion to agriculture usually value of a hectare of Amazon forest in sequestering car- begins with logging. As roads advance and markets bon is over $800. So, although deforestation generates develop, forests become worth seizing. In Asia and value to the rancher of less than $300 and costs society more than $800 in lost carbon storage benefits, no national Latin America there are typically waves of logging, or international arrangement has yet succeeded in devel- successively removing more valuable to less valuable oping institutions to influence the rancher’s decision. trees, followed by the burning of residual noncom- mercial trees, and finally the establishment of large, Source: World Bank estimates. commercially oriented ranches or farms. In Africa the closed forest is more likely to be converted to smallholder farms. These are nonetheless often com- itary purposes can open these regions to encroach- mercially oriented (chapter 8, figure 8.1). ment—and often wasteful exploitation by loggers, Rational occupation of the frontier and conserva- ranchers, and farmers—and to destructive fires. The tion of its biodiversity require better national gover- combination of conversion and sloppy logging pro- nance and policies. Project interventions to improve vides a deadly recipe for forest fires: open canopies community welfare are well meaning and potentially leading to dried-out soil, highly flammable logging useful, but they do not address the scale of the prob- waste on the ground, and escaped land-clearing lem.80 Countries need first to rein in unregulated fires.82 The catastrophic forest fires in Indonesia in logging, which catalyzes conversion and degradation 1999 caused $7.9 billion of damage to the Indone- and appropriates rents to private individuals that be- sian economy and additional health and tourism long to the public or to indigenous people. It is tech- damage to neighboring countries.83 Without inter- nically feasible to do this, but more influential vested vention these areas are likely to experience, over the interests can resist change. Large-scale logging often next 5 to 50 years, the social and environmental benefits government leaders or other powerful inter- problems of earlier frontiers. ests—sometimes for personal gain, but often because The frontier trap. The race for frontier property the forest provides an off-budget source of revenue creates a sharp disparity between what is good for an for projects.81 individual landholder and what is good for society. The people and wildlife of the world’s great trans- First, the rancher or farmer opening new land is un- frontier forests—the western Amazon, parts of the likely to take into account the loss of biodiversity Congo Basin, Siberia, and New Guinea—are pro- and carbon storage (box 5.12). Second, holdings at tected, only partially and for the moment, by their the extensive margin tend to be associated with low inaccessibility. These are the last places where large density and transient communities—raising the costs ecosystem processes represent preindustrial experi- and lowering the quality of government services and ence, and they are home to many indigenous people, creating little opportunity for building human and plants, and animal species. Today they are also sub- social capital. So an extensive and predatory agricul- ject to increasing threats. Roads built for oil and gas tural economy either has a sharply higher cost to the development, for extraction of mahogany, or for mil- state to provide equivalent human services, or has (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  services that are poor, leading to a corresponding loss tensive frontier. Education is rudimentary at best, so in human potential. the poor homesteader, lacking education and urban skills, has few alternatives. To make matters worse, Weak institutions to support communities the predominantly low-input and low-population- and protect biodiversity density pattern of frontier expansion sharply limits Whatever the motive driving settlement onto new off-farm opportunities. So poor settlers lack insur- lands, important institutional and economic condi- ance and risk-management alternatives, except sell- tions are constant. First, the combination of abun- ing the farm and migrating farther out on the fron- dant, inexpensive land and high-cost agricultural tier. This generates a destructive cycle for the poor inputs (owing to high transport costs and poorly de- and for forests. veloped markets) creates economic pressure for land- Although social capital might be substantial in extensive techniques. These include mining the soil frontier settlement areas—especially in settlements nutrients and failing to control soil erosion. Second, where immigrants have moved together from the government and governance are weak, with much same community—the capacity to pool risk through of the frontier population involved in unilaterally collective action is low. Since nearly all poor settlers staking out claims to forested land. Third, the loss are engaged in the same activity, they are all subject of valuable biodiversity and contribution to global to the same risks of pests, drought, sickness, and are warming do not enter the economic calculation of unable to self-insure as a group. farmers opening new land. Under these conditions rapid farm turnover and Getting institutions ahead of the frontier transience for poor people are nearly inevitable. Poor Most remaining large wilderness areas are in remote people lack assets—collateral, access to credit, access areas of low agricultural potential. In these areas to other, nonfarm sources of income, and urban measures to remove land from the land market—by skills—to navigate formal sector input and output establishing parks, indigenous reserves, or biological markets. As a result a speculator or entrepreneur reserves—reinforce complementary efforts by the will almost always, sooner or later, make a purchase national authorities to encourage more intensified offer that the poor homesteader cannot refuse—gen- production on lands already under cultivation. Set- erally under distress, such as crop failure, death of a ting this land aside has important public benefits. It family member, or illness.84 Some poor families es- protects critical ecosystems, and it reinforces eco- cape poverty through repeatedly settling new lands, nomic forces to intensify land use closer to markets. improving them, and selling out and moving on as To the extent that it closes the frontier and raises the frontier advances. Many more stay locked in per- land prices (reflecting decreased land abundance), it petual poverty. discourages nutrient mining, stabilizes communities, The process through which poor people occupy and promotes intensified land use. Park creation, the and gain squatters’ rights to land, and later sell out focus of most conservation projects, has achieved to entrepreneurial agents is repeated in many forms considerable success: 13 percent of the world’s low- and settings. In the pa boei system in the Chon Buri land rainforest is already protected.87 Such parks can hinterland in southeast Thailand, poor small farmers be effective, especially when guards are present and are paid by local entrepreneurs to establish home- local people are involved or compensated.88 steads on federal forest land, with agreements that Financing the maintenance of protected areas is a the land will be turned over to the sponsor after three concern everywhere. Frequent criticism has been to five years.85 Whether formally arranged in advance voiced about “paper parks,� without adequate infra- or resulting from distress sales or different implicit structure or staffing. But recent research shows that discount rates, this pattern of poor people gaining in- the mere designation of parks has an important formal property rights through clearing the forest, impact on future settlement patterns.89 Where pro- and later selling out to the entrepreneurial agent, is tected areas are established well ahead of the fron- common throughout the developing world.86 tier’s advance, a light official presence is enough to This cycle of transience is embedded deeply in stop intruders. Over the next 30 to 50 years, the both the educational and social status of the poor pressure on the frontier is going to increase initially settler and the institutional environment of the ex- and then abate as global population stabilizes, and (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  higher incomes and education create better job op- Box 5.13 portunities. Avoiding irreversible losses during this Brazil: getting ahead of the frontier ebb and flow will likely have a high payoff for future generations. One hundred forty million hectares of the Brazilian Ama- zon, or 28 percent, are in protected areas—national parks, biological reserves, extractive reserves, or indigenous re- Major institutional needs to establish protected serves. Analysis based on satellite imagery and field sur- areas ahead of the frontier veys to detect signs of occupation, forestry potential, and The major institutional needs are (a) pick up signals high biodiversity values shows that without competitive use, 46 million new hectares (9 percent of the Brazilian of biodiversity loss; (b) balance interests of commu- Amazon) could be put into biodiversity protection and 70 nities with biodiversity protection; and (c) execute million hectares (14 percent) into national production for activities to protect biodiversity. sustainable forestry. Pick up signals of biodiversity loss. Signals of ecosys- If this were to come about, more than half the Amazon would be dedicated to either preservation or sustainable tem or species loss are difficult to identify locally be- forest use. Government has pledged to put a representa- cause the loss is often a part of cumulative effects on tive 10 percent of the Amazon forest (41 million hectares) a much larger scale. Warning signals of biodiversity into new protected areas through the WWF–World Bank loss, such as forest conversion, are being increasingly Forest Alliance program. And it is developing a National monitored by government environmental agencies Forest Program to strengthen the forestry system, creat- ing new national forests. The long-run goal is to create a and universities. Specialized monitoring organiza- mosaic of land use to control the advance of the agricul- tions, such as Global Forest Watch, increasingly play tural frontier, support communities through sustainable ac- a vital role. The Global Environment Fund has been tivities, and ensure a strategic buffer for areas of high instrumental in developing biodiversity assessments biodiversity value. and action strategies in many countries. Source: Veríssimo and others (2000). Initiatives to promote transparency can help cat- alyze change. The combination of voice for forest dwellers, better communications technology, and ad- economically sensible, and socially sustainable re- vances in remote sensing means that forest activities gional development approaches. But the agenda has are now more visible than before. NGOs and reform scarcely begun to take shape. groups in government can use this information to call Experience shows that balancing interests in creat- for greater accountability on how logging is con- ing individual parks, protected areas, and forest pro- ducted and how forest revenues are used. Trans- duction reserves is much easier than building con- parency in the award of concessions and in monitor- sensus on more comprehensive zoning. This is largely ing concessionaire performance are important steps a matter of getting far enough ahead of the frontier toward better forest regulation. Recent efforts in that development pressure has not yet emerged. Cameroon show both the challenges and the benefits Once the protected or reserve area is a going concern, in moving toward transparency (chapter 7, box 7.10). little presence is required to keep it intact. Balance the interests of communities and developers. Even so, setting aside land beyond the frontier Ultimately though, these tools can be applied only if will generate resistance from development interests there is popular consensus on regional development and from local traditional communities. By provid- strategies and support for policies that set up the nec- ing both economic and environmental benefits, a essary incentives and disincentives. Provinces and na- land-use pattern based on a “mosaic� of land use—a tions need to debate, for instance, the desirability of mix of production forests, extractive reserves, indig- intensifying and upgrading rural road networks in enous lands, and fully protected areas—can help densely populated rural areas while restricting the build a constituency of environmentalists, foresters, construction of new roads in areas important for bio- and forest dwellers, including indigenous peoples diversity but poorly suited for agriculture. There are (box 5.13). Large protected area initiatives, by con- few good examples of this kind of large-scale land- trast, unaccompanied by job-creating alternatives, use planning, which goes far beyond the discredited face formidable politics. technocratic approach to zoning. It is an area where Execute ecosystem protection activities. Areas be- international resources may be crucial in helping to yond the frontier can be protected through biologi- facilitate domestic agreement on biodiversity-friendly, cal reserves, indigenous reserves, extractive reserves, (c) The International Bank for Reconstruction and Development / The World Bank ������������ ������������ �� ������������ ����  or production forests. International resources are open new land, the environmental costs are high. helping governments sustain land use and protect Third, there is a high probability that marginal fron- biodiversity. For example, with the goal of putting a tier land being opened up today will be abandoned representative 10 percent of all forest ecosystems into as uneconomic in the future. This is becoming more fully protected status and 200 million hectares into evident now than ever, as global food projections in- certified production forest, the World Bank and the dicate little need for additional land to meet antici- WWF are working with governments and local pated growth in population and incomes.92 NGOs to create parks and to create and certify sus- Getting ahead of the frontier with parks, reserves, tainable logging reserves.90 Brazil’s pilot program to and production forests helps end this cycle of tran- conserve the Brazilian rain forest, jointly financed by sience and low-value land conversion. It stabilizes Brazil, the Netherlands, and the G7, has set the stan- the frontier economy. It provides incentives for more dard for NGO and local people’s participation in intensive development nearer to cities. And it re- forest protection activities. duces needless loss of biodiversity. Technical solutions are at hand to ensure the rule of law in areas of difficult access, in part owing to Intensifying agricultural production rapid technological advances. Brazil’s Proarco and Intensifying agricultural production and increasing Amazonia Fique Legal programs have shown the overall agricultural productivity is critical in much of technical feasibility of detecting large-scale illegal de- the developing world, in response to rising pop- forestation through coordination of remote sensing ulations and food demand. It can also be highly de- and ground-based inspection. Using satellite detec- sirable. It can reduce pressure for expansion in wilder- tion the state of Mato Grosso has moved vigorously ness areas and remaining areas of natural habitat to prosecute illegal deforesters. Brazil and Indonesia within settled regions (in conjunction with conserva- have used the Internet to post the location and iden- tion initiatives)—thus reducing pressure on biodiver- tities of lawbreakers.91 sity. It increases the food available to the cities and it leads to dynamic rural-urban linkages. Higher popu- Conclusion lation density and strong rural-urban linkages make This chapter reviewed some key development chal- investments in health and education more effective in lenges for rural areas over the next 30 to 50 years: get rural areas, increase the potential for off-farm employ- ahead of the frontier with biodiversity protection ment, and help farmers accept risk and innovate. and environmentally and socially sustainable activi- These arguments all support a tenure policy promot- ties in frontier areas; intensify agricultural produc- ing relatively small, owner-operated farms. tion; and manage land and water to generate growth, In areas closer to rural towns and cities, nonfarm eliminate rural poverty, and prepare outmigrants to rural employment will be a powerful force for diversi- be productive urban citizens. fying income, allowing greater risk and investment. It can also act as a stepping stone for the rural worker to Getting ahead of the frontier enter productive urban employment. Rental arrange- In many countries the frontier’s advance reflects a ments should thus be encouraged to allow young failure in land tenure policy, and the race for prop- “starter� farmers access to land and often to credit. erty rights leads to excessive farm sizes, underutiliza- Shareholding arrangements, effective for starter farm- tion of land, and lack of opportunity in the more fa- ers to share risk with the landowner, should not be vorable areas nearer cities. It also creates incentives discouraged. to open new land on the frontier. The results are nearly all negative. First, because Eliminating rural poverty and preparing of distance, cost, and transience, the ability of gov- outmigrants ernments to provide for human development on a In poor developing countries with large agricultural frontier is extremely limited (and thus, frontier peo- sectors, growth led by the agricultural sector has a ple are the big losers). Second, the low cost of land powerful effect in pulling people out of poverty, es- at the frontier leads to extremely extensive agricul- pecially when the incomes and assets of the rural sec- ture. As long as biodiversity and carbon values are tor are somewhat equal. Smallholders with assets not taken into account in the farmer’s decision to develop voice and become political players. History (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  has shown that this generates an inclusive develop- This strategy will require reforms in both devel- ment path that helps countries face later challenges. oping and developed countries, however. The devel- But getting assets into the hands of smallholders re- oping-world farmers produce in a world market quires good land and water policies. These policies where world agricultural prices are depressed some also enable poor people to get access to opportuni- 12 percent by tariff barriers and agricultural subsi- ties for building their human and social capital. dies worldwide—but mainly in industrial countries. The value of assets is enhanced through agricul- For many farmers in high-transport-cost developing tural research directed to poor people, and through countries, this may translate into a difference in farm- better agricultural institutions. In Africa improving gate prices of 50 percent or more.93 Similarly, devel- agricultural institutions may depend on strengthen- oping countries’ farmers suffer from lack of agricul- ing the asset value first—with water control and tural knowledge. As noted in chapter 4, only 28 transport infrastructure, and with a concerted pro- percent of public and private agriculture research and gram of fertility enhancement. A reasonable estimate development is applied to tropical agriculture. Sus- for the cost of a program to scale up currently suc- tained agricultural progress in developing countries cessful models is $100 million annually for 10 years. will require a long-term national and donor commit- Many countries will follow a two-pronged strategy ment to the agricultural “knowledge triangle�—agri- that encourages intensification and commercializa- cultural research, extension, and higher education. A tion. This strategy would also promote intensified re- development strategy based on strengthening rural search to adapt staples for high-input agriculture in institutions and a strong smallholder sector will also productive areas near urban markets and transport, facilitate eventual migration to cities. As will be dis- and encourage minimal chemical fertilizer supple- cussed in chapter 6, cities must also be prepared to ments for low-input agriculture in more remote areas. deal with new migrants. (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 6 Getting the Best from Cities C ities of the developing world face a formidable areas—they need stronger institutions to provide undertaking, given the rapid rate of growth and wide access to assets and to balance interests that en- sheer numbers of urban residents to be em- sure the provision of public goods. Such institutions ployed, housed, and serviced. Cities offer proximity, are central to an urban governance that is inclusive which generate externalities both positive and negative. of all residents, responsive to their needs, and con- On the positive side, proximity is a source of produc- ducive to careful management of natural resources tivity; industrial and service activities emerge in cities and wastes. because entrepreneurs and small firms can share mar- This chapter first describes the opportunities and kets, infrastructure, labor, and information. Cost sav- challenges for urban life and then asks the following ings and productivity advantages that accrue to firms questions: when they locate near each other in the same industry or near other economic activities derive directly from Ⅲ How can informed constituencies be built to ad- physical proximity or indirectly from less tangible in- dress spillovers and anticipate risks? Providing in- teraction among economic actors (learning and net- formation, building knowledge, and mobilizing working, leading to innovation). However, large groups dispersed interests are key to creating constituents of people and activities in close proximity also generate that act together to anticipate problems and to negative externalities: poor sanitation, pollution of air prevent and manage disasters. and water, congestion, crime, and so on. This puts a Ⅲ How can competing interests be balanced and premium on the quality of institutions—both formal dispersed interests articulated to provide urban and informal—to ensure the positive externalities and public goods? Foresight, political will, and a gov- to cope with the negative externalities. ernance system that is accountable to a wide array As the previous two chapters have noted, the de- of stakeholders are key ingredients for achieving velopment of urban areas will have to be better co- credible commitments. ordinated with the development of rural areas by Ⅲ How can inclusion and access to assets be encour- providing markets for rural products, by subcon- aged—one key to a city’s sustainability? Security tracting activities to expand nonfarm rural employ- of tenure and guidance of new settlements to ment, and by helping rural migrants adapt more prevent future slums will lessen the inequitable rapidly to city life. Cities and towns facilitate soci- access to assets, thus empowering and enabling ety’s transformations in knowledge, institutions, and poor people to become productive members of economic activity. By bringing together diverse peo- urban society. ple and activities, urban areas offer great opportuni- Ⅲ What institutional mechanisms are necessary for ties for improving the quality of life.1 For cities and good urban governance and sustainable urban de- towns to realize the promises of a better life—espe- velopment? These include an appropriate sharing cially for poor people and for migrants from rural of responsibility and coordination across stake-  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  holder groups; wide participation in strategic plan- Box 6.1 ning; and networking among practitioners and The focus of “urban� in this chapter stakeholders. Two categories of urban issues are relevant to this Report: City lights: beacons of hope and warning flares the spatial system of urban areas in the country and the performance of urban areas. The first topic is discussed in The rising share of people in urban areas and the chapter 7. This chapter takes the second category of urban corresponding economic growth of cities and towns issues—the city and its governance—as its unit of analy- are two defining experiences of economic and social sis. Focusing on the city spotlights the impacts of urban liv- ing (good and bad) and how the relevant institutional frame- development (box 6.1). Urban areas offer possibili- work, both national and local, affects these outcomes. ties for greater welfare because they give individuals Although the analysis of this chapter applies both to the opportunity (through a myriad of functioning cities and to towns, as their smaller counterparts, the dis- cussion mainly refers to cities. Most of the benefits of ag- urban markets) to develop a wider and larger port- glomeration and most of the diseconomies appear as folio of assets—and to achieve higher returns to their towns become recognizable cities. labor. They also exist because of collective concerns The Report examines changing opportunities and chal- to share culture, learning, religious observation, and lenges that appear as population moves along the set- tlement-size continuum, but it does not argue that the mutual protection. prospects for sustainability are a function of scale. Very Yet across cities in all regions of the world there is large settlements (cities of multimillion inhabitants) are evidence that the potential benefits, both individual neither necessarily the best nor the worst cases of sus- and collective, are not being fully realized, and are tainable development. clouded instead by myriad problems. The inadequate Ⅲ In some countries urban air quality tends to be worst provision of jobs, housing, and other goods and ser- in medium-size cities (populations between 100,000– vices stems from imperfections in markets and poli- 500,000).* cies. For many environmental and social concerns, Ⅲ Crime and vulnerability to disaster often become prob- lems in urban agglomerations well below 1 million markets cannot provide the coordination needed to inhabitants, but do not increase proportionally with pop- reveal interests and minimize transaction costs. In ulation size. many countries institutional failures mean that mar- Ⅲ Congestion tends to worsen with city size but also de- pends on such other factors as public transport, traffic kets are less effective than they could be, while alter- management, and road space. native solutions and innovative uses of market in- Ⅲ Many of the economic benefits of urban productivity, struments to address the threats to sustainable such as higher wages and increased human capital, ap- development are inadequately developed. This chap- pear positively correlated with city population, at least to a fairly large threshold. ter does not examine the full range of good urban policies.2 Instead, it focuses on the conditions for What is most clear is that the quality of urban gover- building institutions to protect urban assets—partic- nance and management is critical to gaining the benefits ularly environmental and social assets—because and reducing the negative aspects of cities of any size. As these conditions shed light on the potential for iden- chapter 1 notes, the projected trend of increasing num- bers of people (and possibly shares of urban population) tifying and adopting good policies, and determining living in very large cities over the coming decades in de- society’s ability to respond to future concerns. veloping countries will put a premium on building institu- tions to address the problems of those cities. The role of cities in sustainable development * Observation based on countries with most extensive data by Over the next three decades the urban population of cities. Lvovsky (2001), annex B. developing countries will grow (from natural in- crease, migration, and reclassification of formerly rural areas) by 60 million people a year—equivalent ogy and by creating both the impetus and oppor- to the population of the Arab Republic of Egypt tunity for innovation. or Ethiopia. Urban areas will need to perform a Ⅲ Provide employment and services at a scale suffi- few key functions to support sustainable national cient for current residents and new arrivals. Pro- development: ductive employment is critical so that the fall in demographic dependency ratios projected in most Ⅲ Facilitate social and institutional change, by im- developing countries over the next 20–30 years can proving access to ideas, knowledge, and technol- translate into increased savings and investment. (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  Ⅲ Ensure a healthful and attractive environment for with local buyers, suppliers, competitors, and gov- the urban population while protecting natural re- ernment officials) has a significant and positive ef- sources and reducing deleterious impacts on wider fect on their productivity.8 And proximity to higher regions and later generations. The massive new in- education institutions provides firms with opportu- vestment in the capital stock of cities required for nities to commercialize research ideas, often through the doubling of urban population by 2030 will be university-enterprise partnerships.9 critical to environmental outcomes. Sources of urban productivity. Urban employment and services benefit from the economies of agglom- What enables urban areas to promote change and eration—from cost savings and other advantages that improved quality of life is their scale and density— accrue to firms when they locate near others in the and their social and economic diversity. Proximity same industry, or simply near other economic activi- and heterogeneity make urban areas mechanisms for ties to share markets, services, infrastructure, labor, knowledge and learning, for productivity and mar- and information. The productivity advantage means ket development, and for improved choice and qual- that urban investment has strong multiplier effects ity of services. And they do this even more when the in stimulating other high-value activities. The bene- institutional conditions are right. fits extend to rural areas, which need access to urban The urban stimulus to social transformation and markets to expand and diversify both farm and non- innovation. The shift from rural to urban society, farm production. with greater mingling among diverse people, trans- As a rule, larger urban areas are the most produc- forms social attitudes and behaviors. It reveals the tive since they allow for greater specialization in limits of traditional values and institutions and in- labor use, better matching of skills and jobs, and a tensifies pressures for change in local governance and wider array of consumption choices for workers and intergovernmental relations. Traditional social norms ancillary services for producers.10 As long as this that perpetuate inequalities for women and for cer- greater productivity outweighs higher costs for land, tain minorities tend to be less strictly enforced in the labor, housing, and other necessities, the city can urban environment.3 Urban households are generally thrive. Once the diseconomies become too great, more motivated to limit family size, because of eco- larger cities may lose their edge in creating jobs or nomic alternatives and lifestyles. improving the welfare of residents, unless they can The ferment of urban life generates new forms of shed some activities (those that are more mature and collective action to address the challenges that arise. standardized) to smaller cities to make room for oth- Urban-based constituencies have been the driving ers (more innovative and higher value industry and force behind many of the environmental causes that services) and change land uses. pertain to national and global public goods—creat- For cities to fulfill their potential as engines of na- ing national parks, protecting biodiversity, and man- tional economic growth, they need to ensure that the aging coastal zones. More than 90 percent of China’s labor market is not only deep but well integrated and environmental NGOs are located in cities of the rel- inclusive—with accessible workplaces and residences. atively well-developed eastern coastal region.4 A re- A city can improve its investment climate. However, cent study of Indonesia proposed that the best way cities in general can only improve the national invest- to save the tigers was to teach urban children about ment climate if their overall legal and regulatory them.5 framework complements the national framework to Historically, cities have been centers of learning minimize risks, uncertainties, and transaction costs to and innovation. The growing intensity of knowledge investors. This is especially important for small and exchanges arising from globalization and the infor- informal sector enterprises, which provide most urban mation technology revolution has the greatest im- employment, rely more heavily on publicly provided pact where there is also occasion for interpersonal infrastructure and information, and are particularly communications.6 Informal information or tacit vulnerable to institutional and policy failures. knowledge, important to productivity and to social More affordable and higher quality services. The relationships, thrives on face-to-face contacts.7 Re- greater scope for competition and specialization in search among Mexican firms, for example, shows all goods and services enables urban areas to provide that access to informal networks (business lunches consumer benefits in the form of greater choice and (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  quality.11 But the advantage is especially important more steadily transform and expand their portfolio for services with high fixed costs (increasing returns of assets than is possible in many rural areas. to scale), such as middle- and higher-level education Those who migrate to cities are often better off and health facilities, and network infrastructure.12 than their neighbors back home;16 in cities they re- The cost advantage explains some of the manifestly ceive more education and better skills, and in the better social indicators in urban areas and in coun- longer run they catch up with established urban res- tries that are more urbanized.13 That is, urbanization idents.17 A recent survey in Latin America finds a has a large positive impact on a country’s efficiency potentially large private gain from migration to in achieving health and education outcomes.14 Even urban areas, in part because the returns to human where more expensive services are required for envi- capital tend to be larger there.18 A study of the urban ronmental and health benefits (as for waste disposal, labor market in the Punjab State of India found no sewerage, water treatment, and mass transit), the evidence that migrants remain confined to marginal added cost can often be justified by higher economic jobs or are disproportionately unemployed.19 Mi- returns. grants frequently obtain work, housing, and urban Migrating for choice and change. The offer of new services through the informal sector, and they often opportunity and a better life is often what draws mi- depend on supportive social networks to do so (box grants to towns and cities (the pull factor). Limited 6.2). Nonetheless, imbalances in the demand and employment options in rural areas, whether from supply of jobs, housing, and urban services can con- agricultural dislocations arising from natural and so- tribute to urban poverty initially and for long periods. cial disasters, or from increased agricultural produc- In the short term, the influx of migrants can tivity that reduces the demand for farm labor (chap- sometimes overwhelm some urban areas, particularly ters 4 and 5), also contribute (the push factor).15 when the pace is sharply accelerated by civil conflict Cities and towns allow individuals to substitute their or national disaster. In the medium term, as part of human capital (work effort and skills) for natural, fi- the natural development process, rural-to-urban mi- nancial, or physical assets they may lack—and to gration and the resulting return flow of transfers Box 6.2 How social networks help the urban poor manage risks and get ahead Urban areas are often said to lack social capital. Yet social net- the least security are usually those harboring new migrants and works are important to the survival and mobility strategies of others with relatively limited social networks, who are unable urban poor people and to decisions about migration from rural to negotiate with bureaucrats and politicians. Social networks areas. Risks in the urban context arise mainly from weak prop- among slum residents assist with survival (coping with emer- erty rights (which can result in loss of assets and involuntary gency needs), similar to traditional rural networks, and to up- resettlement), inadequate sanitation, exposure to violent crime, ward mobility of individuals and the community. The transition and unemployment or other effects of macroeconomic shocks. from an “unrecognized� to a “recognized� slum affords access The urban poor, like their rural counterparts, cope by using their to water and sanitation and immunity from demolition. Similarly, social networks and personal assets. networks are used to get ahead occupationally, whether through Social networks in the city are based more on reciprocal formal sector employment or informal self-employment. Slum- links between individuals and friends, than on familial obliga- dwellers also use relationships with local leaders to obtain citi- tions (as in rural areas). Yet, maintaining close links between zenship status and strengthen their legal protection. rural and urban social networks can be crucial to preserving The urban poor, including migrants, move from their initial one’s identity. As in rural areas the ability of urban commu- inherited networks to ones that link them to external benefits nities to engage in collective action is often instrumental in and resources outside their original community. While these obtaining public services. This is particularly the case in large networks are a useful resource, the energies of poor people cities (for instance, Jakarta or Manila) where the urban poor— would be used more productively if basic services and security because of their numbers, and the relative ease of organizing of tenure were provided more consistently and transparently, them—are an important political constituency that can also be thus removing both the need for exhaustive negotiations to ac- manipulated by officials promising services. quire basic entitlements and the opportunity for rent-seeking Recent field research in the slums of Delhi confirms that the by local officials. major source of risk to the home and asset base for residents of squatter settlements is insecure land tenure. The slums with Source: Rao and Woolcock (2002). (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  raise incomes and living standards in both areas. Cir- the issues of environmental degradation in richer cular and temporary migration to and from cities or cities, such as greenhouse gas emissions from hydro- towns help manage risk for both rural and urban carbon fuels, stem from lifestyles entailing high con- households.20 Openness to new ideas and learning sumption and associated waste of natural resources. acquired by urban migrants are transmitted to rural Despite these differences in the incidence of risks communities through social and family links—and and their links to income and consumption, urban through the use of remittances to introduce technol- residents of middle-income countries suffer environ- ogy to rural activities.21 In the longer term, once the mental insults both traditional and modern (such as urban transition has been completed, natural popu- exposure to hazardous wastes and chemical pollu- lation increase in cities, rather than rural-to-urban tants).23 International travel and changing global migration, will account for most urban growth. Mi- weather patterns are bringing environmental profiles gration among cities will continue in response to of cities in industrial and developing countries closer changing economic opportunities. together by spreading some risks (of disease and natu- Making the environment work for urban resi- ral disasters, respectively) to both groups of countries. dents—and saving it for others. Urban living poses en- For the full range of concerns, institutional reform vironmental hazards, which affect the current popu- is required to protect poor people and environmental lation (especially poor people) through immediate, assets in cities both today—and in the future.24 local impacts on health and safety. It also causes en- Allowing the urban potential to transform society vironmental degradation, with longer-term, wider- and to improve welfare, while also protecting the en- area, and intergenerational consequences.22 Varia- vironment, may appear to be harder for the develop- tions in the incidence and relative severity of a range ing world today than it was for industrial countries of environmental problems across cities at different at similar points in their urban transition. Why? Be- levels of development suggest differences in priori- cause of today’s faster urban population growth rates ties for action (table 6.1). (figure 6.1) (approximately one-half from migration In low-income cities fewer than half the house- and the rest from natural growth and reclassification holds are connected to water and sewerage, and per of contiguous areas) and the sheer numbers of urban capita water consumption is half that of cities with residents to be employed, housed, and serviced over lower-middle income ranking (table 6.2). Less than the next few decades. Urban population growth one-third of solid waste in the poorest cities is dis- rates, especially in Sub-Saharan Africa, have been posed of properly; only the richest cities provide unprecedented, though as in other regions they are comprehensive wastewater treatment. Partly reflect- projected to slow. Despite the deceleration, almost ing environmental risks, the average mortality of 20 million new urban residents a year are projected children under five in the poorest cities is more than through 2030 for East Asia. twice that in the next city-income category, and 20 The same characteristics of urban areas—density, times that in the richest cities. scale of settlement, and social diversity—that can Especially in cities at low levels of development bring about the positive potential of more jobs, ser- many residents face environmental risk because of vices, and learning also create the negative potential their living conditions and location. These house- and the need to balance interests. In addition to en- holds are least able to afford protective or mitigating vironmental spillovers, urban areas are associated mechanisms—or to assert claims for improved ser- with other threats to sustainable development and vices. Most vulnerable are children and women, the livability. Problems of land use and accessibility (in- elderly and disabled, and homeworkers, who are cluding congestion) impede the utility, inclusiveness, continuously exposed to hazards in their immediate and enjoyment of urban life, while physical insecu- environment. rity results from poorly managed risks of natural dis- In the poorer cities badly managed urban growth asters, crime, and violence. All of these problems degrades natural resources, especially watersheds, affect poor people even as they reduce the welfare of soils, and coastal environments—because of un- all urban residents. treated sewage discharge, poor solid waste disposal, None of these threats is driven primarily by the and a lack of storm drainage. In contrast, many of rate of urbanization, by the size of city, or by the lack (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 6.1 Urban environmental issues and status by level of city development Sector or problem area Low Lower-middle Upper-middle High Water supply Low coverage, high Low access by poor Generally reliable, but Good supply but high total service bacteria contamination, residents and informal rising demand causing consumption; some con- inadequate quantity for neighborhoods shortages in resource cern with trace pollutants hygiene (high risk of supply food contamination and infectious diseases) Sanitation Very low coverage, open Better coverage of More access to improved Full coverage; most defecation in some neigh- latrines and public toilets, sanitation, but still large wastewater treated borhoods and low ratio but poorly maintained; numbers of residents in public toilets to residents; low sewerage coverage large cities not covered high risk of diarrheal especially in informal diseases settlements; most waste- water discharge untreated Drainage Storm drains very in- Somewhat better than Better drainage; Good drainage; very adequate, poorly main- in low income occasional flooding limited flooding tained; frequent flooding, creating high risk of water-related disease vectors (mosquitos) Water resources Mixed sewerage and Risk of groundwater Private wells drawing High levels of effluent storm water runoff to contamination from down groundwater; controls and treatment water bodies causing bac- poorly maintained latrines severe pollution from to reduce pollution terial pollution and silting and untreated sewage industrial and municipal discharge Solid waste Little organized collection; Moderate coverage of Better organized collec- Increased emphasis on management recycling by informal sec- collection service, little tion; severe problems total waste reduction, tor, open dumping or separation of hazardous but growing capacity for resource recovery, and burning of mixed wastes; waste; mostly uncon- hazardous waste manage- preventing hazardous high exposure to disease trolled landfills ment; semicontrolled waste; controlled landfills vectors (rats, flies) landfills or incineration Air pollution Indoor and ambient air Growing ambient air pollu- Ambient air pollution still Ambient air pollution pollution from low-quality tion from industrial and serious (but greater mainly from vehicles fuels for household uses vehicular emissions (high capacity to control espe- (due to high volume of and power generation per-vehicle, due to ineffi- cially industrial sources) vehicle kilometers) cient fuels and vehicles) Greenhouse gas Very low per capita Low but growing per Rapidly increasing, mainly Very high per capita emissions capita due to motorization Land management Uncontrolled land devel- Ineffective or inappropri- Some environmental Regular use of environ- (environmental opment; intense pressure ate land-use controls, zoning mental zoning; little popu- zoning of fragile from squatter settlements pushing new settlements lation growth, but rising sites and prepara- on open sites toward urban periphery; incomes press for more tion for new continued high population land consumption for settlements) growth existing residents Accident risk In-home and workplace Increased risks of indus- Transport accidents in- Rate of industrial and accidents due to trial workplace and traffic creasing, but some miti- transport accidents re- crowding, fires accidents (pedestrians gation and emergency duced despite increasing and nonmotorizcd treatment response travel (vehicle kilometers) vehicles) Disaster Natural disasters produce Somewhat better than Increasing awareness Good capacity for management massive loss of life and in low-income, although and capacity for disaster mitigation and response property especially in set- with increasing risk of mitigation and emergency tlements in disaster-prone industrial disasters response areas; little capacity for mitigation or emergency response Note: Cities grouped by estimated city product (city average income calculated by national accounts methods). Sample is of cities (including in OECD countries) with available data and is not statistically representative. Low income defined as city product below $750 per capita a year; lower-middle as $751–2,499; upper-middle as $2,500–9,999; high as above $10,000. Source: Adapted from Leitmann (2001) and Hardoy, Mitlin, and Satterthwaite (2001). (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  Table 6.2 Environmental health, welfare, and living conditions vary by city product City product category Lower- Upper- Indicator Low middle middle High Household connections (percent) water 48.0 78.8 92.9 99.9 sewerage 45.6 68.7 84.3 99.9 electricity 72.3 93.6 95.0 100.0 Water consumption liters per person per day in all settlements 88 161 232 247 Wastewater treated percent treated 29.4 56.7 68.2 97.4 Solid waste disposal (percent) sanitary landfill or incinerated 30.7 41.4 37.6 77.7 other (open dump, recycled, burned) 65.9 58.3 62.2 22.3 Under-five mortality per 1,000 104.2 39.7 25.8 5.2 Households below the locally defined poverty line (percent) 31.7 23.2 16.0 6.9 Sample size (cities) 49 36 25 20 Note: Cities grouped by estimated city product (city average income calculated by national accounts methods). Sample is of cities (including in OECD countries) with available data and is not statistically representative. Low income defined as city product below $750 per capita a year; lower-middle as $751–2,499; upper-middle as $2,500–9,999; high as above $10,000. Source: United Nations–Habitat Global Urban Indicators Database 1998. Figure 6.1 Many developing countries are undergoing urban transition with relatively high urban population growth rates Average urban growth rate (percent) 6 1975–2000 2000–25 5 Sub-Saharan Africa East Asia Middle East 4 and North Africa and Pacific Sub-Saharan Africa South Asia All developing countries Latin America 3 South Asia and the Caribbean Developed countries Middle East (1900–25) and North Africa All developing 2 East Asia and Pacific countries Latin America and the Caribbean Europe and Central Asia 1 Europe and Central Asia 0 0 10 20 30 40 50 60 70 80 90 National urban population share (percent) Note: All averages weighted by population. Lines indicate increase in share of urban population between end-point years (25-year increments). Source: Developed country data from Brockerhoff and Brennan (1998); other data and projections from U.N. (1999). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  of fiscal or other resources—though these factors tional governments need to play important leader- (and such others as geography, local culture, and his- ship roles in both cases. torical legacy) can make problems harder to manage. Cities need not suffer crippling diseconomies what- Credible information and incentives—curbing ever their size or income (see box 6.1). Shanghai, one air pollution of the world’s largest urban areas in one of the poor- Air pollution generates large social and economic est countries, devotes more than 3 percent of the city costs. In many developing and transition countries, GDP to environmental protection. And it has man- the damage reaches 4 to 6 percent of urban income, aged to achieve better outcomes (green space, im- and has serious adverse affects on human health. Be- provements in air quality and sewage treatment) than tween 500,000 and 1 million people die prematurely most other developing-country cities.25 every year as a result of air pollution–induced respi- The growth of urban poverty in many countries, ratory problems.28 Vehicle emissions create the great- evidenced especially by the increase in populations est damage to human health because they occur near residing in extremely poor environmental condi- ground level and in dense population centers, while tions, is partly a reflection of the pressures on lim- smokestack sources disperse pollutants more widely ited city resources.26 It also underscores the failures at higher elevations.29 Urban residents in low- and of institutions and their unresponsiveness to certain middle-income countries have greater exposure (well constituencies. As urbanization plays out, growth above WHO guidelines) to some localized air pollu- rates will slow, easing the pressure on cities. But tants, such as suspended particulates, than their coun- catching up over several decades is not satisfactory. terparts in high-income countries, even though the Measures are needed now to accelerate the rate of latter consume more energy per capita.30 improvement in cities and to avoid making it more Countries do not have to suffer worsening air qual- costly to close the gaps later. There are now techno- ity as they industrialize, motorize, and become richer. logical and institutional options that need to be ex- Many technologies and behaviors for curbing urban plored more vigorously. air pollution are cost-effective even at low levels of For cities to contribute to sustainable develop- economic development and limited institutional ca- ment, they need to maximize the positive while min- pacity, as long as there is political commitment and imizing the negative externalities. The favorable eco- public understanding.31 While action by industrial nomic and social impact of cities can exceed their countries to eliminate leaded gasoline, for example, “ecological footprint.�27 Cities can get themselves took a decade to implement, sharing knowledge and into vicious or virtuous circles that become self- demonstrating workable solutions have permitted de- reinforcing, and triggers for change can often be veloping countries to phase out this fuel much more found in institutional innovation. A key institutional rapidly (chapter 7). catalyst—information—can increase urban benefits Curbing stationary sources of urban air pollutants and reduce diseconomies and risks. (concentrated interests) is institutionally easier than curbing mobile sources (dispersed interests) because Building informed constituencies to address there are fewer polluters.32 That the fuel supply was spillovers and anticipate risks the main source of airborne lead made it easier for The spatial concentration of people and economic countries to implement the phaseout administra- activities in urban areas creates spillovers with signif- tively. Pressure from an informed public has been icant impacts on residents—and increasingly on more instrumental in getting governments to rein in wider regional and global populations. Mobilizing other types of pollution and in motivating regula- for action to solve such problems (for example, pol- tory or other action. In China the educated urban lution) requires that the parties affected gain access population has been an important force for such re- to credible information on costs and benefits and forms. But the national government’s willingness to that they perceive a common interest in finding a make information on health costs and risks publicly solution. Building an effective constituency can be available was an essential precondition.33 more difficult where the impacts are uncertain and Curtailing mobile sources of pollution and large infrequent, as in disaster mitigation. Advances in gas guzzling vehicles is most challenging because the technology and knowledge help, and local and na- middle- and upper-income groups are the beneficia- (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  ries of increased motor vehicle travel, and the main work, is being considered for any future extensions source of growing emissions with global and regional of the system.37 impact. These stakeholders are a more influential in- While communities and NGOs can work to- terest group than the general public, and especially gether to provide household and neighborhood san- more so than poor people suffering from the result- itation facilities, the social costs are lower and bene- ing pollution and accident risks. Collective action to fits greater if disposal and treatment of wastewater reduce transport-based GHG, (especially CO2) is are citywide. Few cities in developing countries treat further complicated by the nonlocal and longer-term their wastewater, contributing to the pollution of nature of the damages. downstream water bodies and corruption of fragile Growth in motor vehicle ownership in develop- coastal environments. Because investment costs are ing countries could overwhelm improvements in fuel high and impact areas can extend across many juris- or vehicle efficiency. Effective approaches to reduce dictions, solutions require cooperation among local transport-based pollution therefore involve a range governments and across levels of government. of interventions at different scales (local, national, Water pollution charges have been much less ef- and global), forming part of the integrated transport fective in making municipalities cut their emissions strategies discussed below.34 These measures include than industrial and agrobased polluters. The Wa- improved information on levels and sources of, and tershed Cap and Charge System in Colombia has damages from, pollution, educational campaigns, in- induced industrialists to reduce effluents, yet two- centive systems (including taxation of vehicles and thirds of local governments, which create 70 percent fuels), and technological measures, such as replacing of the water pollution, have been unresponsive.38 high-mileage, heavy-polluting vehicles and installing Rising public demand for improved water quality computerized inspection and maintenance regimes. will be needed to press for such changes, aided by in- It is also necessary to manage supply and demand come growth and good water-system management to across transport modes, through better public trans- reduce waste and sustain net revenues, since treat- port, improved conditions for nonmotorized trans- ment is very costly. port, traffic-management, traffic-calming measures,35 For both air and water pollution, public awareness and road and vehicle user fees. These measures re- and citizen pressure on governments and polluters quire coordination between jurisdictions within ur- are possibly the most important factors motivating ban areas and across levels of government.36 environmental reform in the face of entrenched in- terests or official indifference. Political democratiza- Creating constituencies—for clean water tion and the freedom to associate can be catalytic in and wastewater management making dispersed stakeholder groups see their com- Even though inadequate neighborhood disposal of mon concerns and collective strength (box 6.3). wastewater has unavoidable negative impacts, solu- tions are often limited by weak organizational capac- Mobilizing dispersed interests to anticipate ity for collective action. Technological innovation, problems—preventing and managing disasters 39 spurred by a professional association of progressive- Urban disasters (natural and industrial) can cause minded civil engineers, was behind the introduction large loss of life and have enormous economic and of a low-cost approach of shallow, small-bore sewer- financial costs. They are especially devastating to age networks in Brazil a decade ago. Compared with poor people, who often live and work in precarious conventional systems, this condominial sewerage re- conditions. The drive for mitigation increases as the duces investment costs by half but requires a strong effects of disasters, and the costs of failures to take commitment of households to maintain it collec- action, become more immediate and widely per- tively. Condominial sewerage systems have worked ceived. Institutions are needed that can motivate well where this cooperation was sustained and par- action in advance of crisis and share the costs and ticipating households understood their responsibili- benefits of preventive measures among citizens in a ties. But since it can be difficult to get individual fair manner. Hazard mitigation requires improving households to commit to take part, a more organized knowledge, building constituencies for risk reduc- institutional arrangement, through community as- tion, and strengthening institutions and partnerships sociations that would contract for the maintenance across levels of government and the private sector. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 6.3 Political reform and stakeholder alliances overturning pollution The city of Cubatão, in São Paulo State, Brazil, was castigated changes at the national level in the early 1980s were key: the in the country’s press in the late 1970s as the “valley of death� transition to democracy that allowed the emergence of inde- because of the extreme industrial pollution of its water, air, and pendent social activism, free elections at the state level, and soils that had occurred under decades of military dictatorship. the elimination of media censorship. Poor people lived in the midst of toxic waste dumps in an area In executing the program, CETESB relied on a suitable leg- also prone to natural disasters. In 1983 the state environmen- islative framework and its status as the only agency with both tal protection agency (CETESB) initiated a pollution control pro- the mandate and expertise for pollution control in the state. gram that significantly reduced pollution levels in less than 10 Still, the environmental clean-up of Cubatão was possible only years. The agency managed to challenge the privileged posi- after changes in the political rules of the game meant that pro- tion of entrenched industrial interests and make industrialists gressive bureaucrats could ally with informed citizens to chal- bear most of the costs of pollution control. lenge the powerful economic elite that had stymied previous How was this achieved? A citizens’ association of “victims reform efforts. The experience of APVM forged a collective of pollution and bad living conditions� (APVM) focused the identity among victimized residents that moved them to act public debate on the human toll of pollution and attracted wide- and set new terms for a collaborative relationship between the spread support for reform. While the program was advanced citizens and economic powers of the city. by high-level support for environmental improvement in the state government and aided by international opinion, three Source: De Mello Lemos (1998). Climate change and natural disasters are closely hazard information, create new knowledge, build related. Global warming, projected to raise sea levels the capacity of institutions, and train and raise the as much as 0.8 meter this century,40 is particularly awareness of decisionmakers and the communities threatening to coastal cities—where most of the at risk. megacities in developing countries will be located by Estimating losses can make the financial case for 202541—as well as to small island states.42 Climate- preparedness. Memphis, Tennessee, calculated a $0.5 related events directly affecting urban areas include million cost for retrofitting water pumping stations floods, mudslides, heat inversions, wind storms, and to be disaster resistant—compared with $17 million storm surges. Much of the economic and human toll to replace each pump and $1.4 million for each day will strike at the advancing frontier between the built the system is out of service.44 But developing coun- environment and nature in the cities of developing tries rarely have well-documented, location-specific, countries, which also serve as reception areas for en- and hazard-specific costing of hard and soft mitiga- vironmental refugees. tion measures. Even more rare is costing used sys- Although earthquakes claim fewer lives than tematically for public education. Without an edu- weather-related events on an annual basis, between cated constituency collective decisions on disaster 40 and 50 of the fastest-growing cities in developing policies are usually dominated by better-off mem- countries are in earthquake zones.43 Exposure to bers of the community. Their priorities can differ such hazards and the reduced ability to provide basic greatly from those of poor people, who risk a larger services after a crisis can jeopardize a city’s attractive- share of their assets in a disaster.45 ness as a business location. Creating incentives and constituencies for risk reduc- Building knowledge. Knowledge about the hazards tion. Indispensable for mitigation strategies are may be scant or absent, even among residents most strong disaster prevention proponents and the polit- at risk, yet community awareness of physical hazards ical will to lead regulatory changes and financial ap- is fundamental for mitigation efforts. Comprehen- propriations. With limited resources developing sive vulnerability assessments using remote sensing, countries must rely on partnerships of all actors. In satellite imagery, risk and loss estimation modeling the United States disaster prevention started with can help document and reduce physical, social, and coalitions of scientists, emergency relief organiza- economic vulnerability. Changing physical infra- tions, professional associations, and other civic structure and innovative techniques for retrofitting groups who lobbied governments to fund research buildings can improve disaster prevention. So can and hazard mitigation strategies. This movement re- “soft,� nonstructural methods—those that increase ceived impetus when the Federal Emergency Man- (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  agement Agency (FEMA), armed with a federal companies, and, indirectly, on the urban economy. mandate and incentives, took the lead and promoted Low-income residents living in harm’s way will need local and state initiatives (such as the regional Earth- particular assistance. quake Preparedness Projects in California), but still Balancing interests to provide worked through civic and professional partners. urban public goods The public needs to decide on acceptable levels of risk, comparing the immediate benefits of expendi- Urban areas can enhance and enrich social integra- tures on other social priorities with the delayed ben- tion through the provision of public goods and cul- efits of reduced loss of life and asset replacement cost tural and environmental amenities. Achieving these following a potential disaster. These tradeoffs can be benefits requires institutions to channel dispersed in- eased when well-designed incentives change private terests of a pluralistic public—to give expression to behavior to help prevent hazards. Examples include the social value of equitable access to publicly pro- reducing insurance premiums on residential prop- vided assets, and to identify future needs in land de- erty when basic hazard-resistant steps are taken, of- velopment and redevelopment—and sometimes to fering disaster insurance with strict enforcement overcome powerful vested interests. To provide other of building code provisions, or providing tax holi- urban public goods, similar mechanisms are needed days or grants for mitigation.46 Poor residents, for to balance competing interests (for a well-integrated whom insurance or fiscally based incentives may not transport system, and sanitary solid waste disposal),48 be practical, would benefit from urban planning for and to express dispersed interests (for drainage). slum prevention, enforceable environmental zon- Foresight, political will, and a governance system ac- ing in cities, and resettlement combined with com- countable to a wide array of stakeholders are key in- munity-based upgrading and tenure regularization gredients for achieving credible commitments. schemes (discussed below). Recent disasters can motivate countries to under- Balancing private and public interests in land use take some of these measures and instill longer-term and committing to priorities in the public interest thinking. Gujarat State in India is trying to establish The challenge in anticipating urban population effective disaster management institutions following growth is to focus on the most socially, environmen- the January 2001 earthquake that killed 15,000 peo- tally, and economically important aspects of future ple. The state has a new disaster management author- land uses, and commit to credibly executing these ity to coordinate all aspects of the response, working public choices. New settlements in or near existing with NGOs, the private sector, universities, local com- urban areas require the following actions: munities, and external donors. The program includes predisaster preparedness and postdisaster response, re- Ⅲ Setting aside rights of way for primary transport construction, and disaster prevention. Incentives are arteries being introduced to build constituencies for disaster Ⅲ Proscribing settlement or other development of prevention by capitalizing on the population’s height- areas that are unsuitable because of environmental ened awareness and willingness to change. fragility or vulnerability to disasters (steep hill- Adapting to climate change. Adapting to climate sides, flood zones)—and protecting fragile envi- change may be more difficult since the risks mount ronmental resources (urban watersheds, wetlands) gradually and less visibly—but no less urgently.47 Ⅲ Reserving areas for amenities, especially parks, and Coastal cities and other population centers (espe- developing other public spaces with social and cul- cially small island states) will need to invest in pro- tural value. tective barriers and possibly to relocate residences and essential public facilities through managed re- This effort has to take into account emerging sup- treat. Priorities for such adaptation should be given ply and demand and avoid overdetermining the city’s to built areas and infrastructure that require urgent future. A frequent problem is that city master plans attention in any case, such as vulnerable informal may exclude large high-value sites, especially at the settlements and outgrown sanitation and drainage periphery, from urban development, while failing to systems. Adaptive expenditures will place a signif- fence off environmentally vulnerable or risky sites in icant burden on the public sector, private utility an enforceable way. Although urban expansion into (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  agricultural areas may pose real social welfare trade- city’s economic decline has been reversed and it now offs, much so-called agricultural zoning around cities enjoys a new distinction for tourism as the only city is outdated. It neglects the greater economic and fis- in China preserved within old city walls.50 cal benefits of urban land uses and fosters opportu- The use and reuse of public spaces provides an nities for corruption and speculation. Urban munici- opportunity to cultivate a city’s natural resources— palities often have much less say over land conversion by preserving or creating parks, architecturally dis- at the periphery than do national governments or tinctive streetscapes and squares, and waterfronts. powerful elites. Such urban amenities are part of a city’s portfolio of To identify minimal, high-priority, and enforce- assets and broad access to them enhances well-being able limits on land use, local institutions must first and strengthens social capital, since poor and wealthy identify the socially desired outcomes for urban de- alike can enjoy them. In the fast-growing cities of the velopment (such as which environmentally sensitive developing world reserving open spaces requires areas to protect—recall the Catskill example in chap- strong commitment (championship) and forward ter 2, and how much green space to set aside) and thinking to speak for these dispersed interests. A for- then commit credibly to achieving them. In Conakry, mer mayor of Bogotá saw in the hundreds of kilo- Guinea, the municipality developed a basic structure meters of drainage canals crisscrossing the city a way plan for the city in the early 1980s. Thirteen years of connecting all parts of the city by converting them after the plan was adopted to create the primary roads to walkways and bikeways. Despite resistance from and infrastructure networks essential for urban mo- some quarters, the city realized this vision, creating bility and productivity most of the networks are in one of the world’s largest pedestrian paths lined with place. Now the city’s focus is on upgrading densely trees, lighting, open sculptures, and benches—and populated neighborhoods by providing internal and linking some of the poorest neighborhoods of the secondary roads linking to this network, and basic city with golf courses and parks.51 municipal services (drainage and solid waste collec- Remediation of despoiled sites and abandoned tion), as part of an integrated citywide program. structures (brownfields) can also present many ben- Urban redevelopment for public goods. The eco- efits to a city. Overcoming the institutional and nomic and social vitality of cities is enhanced by financial barriers to brownfield redevelopment is promoting and protecting their cultural characters, important to correct the environmental and social developing public spaces and other amenities as eq- blight in surrounding areas and to prevent the flight uitable social assets, and converting abandoned and of new (greenfield) investments to the urban periph- degraded land and waterfronts to new uses. Civic ery. Extensive experience in industrial countries un- groups and private entrepreneurs can motivate such derscores the social and economic benefits when collective action, but often formal partnership and brownfields are redeveloped as an integral part of political leadership from local and national govern- neighborhood renewal processes.52 Public–private ment are needed. investment partnerships require clear legal frame- A city’s historical heritage and social culture em- works for property rights, risk sharing, and assign- bedded in its neighborhoods and structures are valu- ing liability for pollution clean-up. Efforts in Bu- able assets.49 Vision and voice for dispersed and fu- dapest to identify a redevelopment program for the ture interests are necessary to give appropriate weight Csepel steel works on an island in the Danube River to such intangible values, to counteract pressures in the middle of the city have been delayed by diffi- to rebuild and modernize for commercial or high- culties in negotiating with more than 200 owners of income uses, and to prevent the gradual degradation the site, which was hastily privatized during the post- of the built environment because of poor house- socialist transition. holds’ need for affordable places to live. In the early Leadership and foresight are also key to urban re- 1980s the city government of a historic Chinese city, newal. In the mid-1980s the mayor of Istanbul led Ping-yao, almost demolished the old city wall to an initiative to clean up an inlet of the Bosphorus, build a wide road. Protests by scholars and residents, known as the Golden Horn, which had become pu- with help from officials and news media elsewhere trid from the dumping of sewage and solid waste. in China and abroad, persuaded the government to The program—part of a large metropolitan sewerage locate new development outside the old city. The investment to extend sanitation to some 300,000 (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  low-income residents and treat municipal water— vocal (and often competing) lobbies, the interests of also relocated the polluting industries and ware- poor people are less well expressed politically.57 Yet houses. The water quality of the Golden Horn has poor people lose when the dominance of private vehi- been restored. Recreational and tourist activity has cle traffic undermines support for public transport reemerged. And public and private investment has re- and space for nonmotorized options. Poor people juvenated the area’s historic and cultural assets.53 become more restricted in their mobility—and as pedestrians, they suffer most from road accidents. Balancing competing interests for accessibility 54 High-speed roads often carve up low-income neigh- Urban transportation is closely tied to urban land borhoods, increasing noise and ground-level pollu- development and can create both positive and nega- tion. During the construction of the U.S. interstate tive externalities as cities grow. Urban transport is highway system in the 1950s, for example, the plan- best addressed as part of integrated urban strategies ning criterion of selecting “least cost� sites caused in- that can attend to the interests of all user groups trusion not only into environmentally sensitive areas (including poor people, women, and the mobility- (wetlands and so on) because their market land value impaired) and anticipate long-term needs that have was low, but also into the poorest urban neighbor- no vocal constituency. Most cities in developing and hoods, severing them from the rest of the city and has- transition economies are sufficiently densely popu- tening their deterioration. The massive public spend- lated to support extensive public transport,55 and ing on the highway system, unbalanced by sustained often include (depending on physical and climatic support for other transport modes, accelerated the conditions) walking and cycling as major modes of flight to the suburbs of wealthier city residents and the travel. It is important that institutions are developed economic decline of older inner-city areas, contribut- that balance and give weight to these interests. And ing to persistent social problems in U.S. cities. since motorization is still at an early stage, it is likely Urban transport strategies that focus on the that an urban transport strategy will emerge in such mobility of all residents, not just a few, to make an institutional setting that focuses on balancing transportation more sustainable environmentally, so- roads and private cars with other alternatives within cially, and economically contain a balanced array of a broader urban perspective. measures: Traffic congestion is one of the major negative ur- ban externalities. As cities grow and become richer, Ⅲ Managing the existing road infrastructure to im- vehicle ownership and use increase more rapidly than prove the traffic flow and calm speeds around available road space. Expanding road space tends to densely populated areas stimulate more car use, so the imbalance continues. Ⅲ Giving weight to the effects of induced traffic and Experience shows that more road building is not the impacts on nonmotorized transport and the envi- route to a congestion-free future. Most important is ronment when evaluating new road projects how the space devoted to roads is used and man- Ⅲ Internalizing the social costs of road use by charg- aged—for example, it should be organized hierarchi- ing land developers impact fees to finance new cally to separate traffic flows for different purposes. roads, introducing road congestion pricing (or Managing demand, through taxes and impact fees on some proxy), and charging the full social costs of road construction and use, is politically much more parking difficult once car dependency becomes entrenched. Ⅲ Improving the viability of public transport by giv- Cities need a minimal amount of space for circula- ing priority to buses on restricted lanes, ensuring tion relative to their size to operate efficiently, and adequate financing, and improving operational ef- early reservation of rights-of-way for major transport ficiency through regulated competition routes is essential to good urban planning.56 Trans- Ⅲ Protecting pedestrians and nonmotorized trans- port infrastructure costs rise sharply as cities become port users by providing safe walkways and bicycle more densely developed, and the investments are paths large and “lumpy,� so requirements need to be con- Ⅲ Providing rail-based mass transit in very large sidered well in advance of actual demand. cities with high transport demand, where it can Although the environmental and efficiency costs of also serve low-income users, as in some Latin motorization and related traffic congestion attract American cities. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 6.4 posal is a coordination problem. Waste collection is Meeting environmental, social, and economic usually the responsibility of municipal government, objectives through urban transport strategy but in many cities the formal service covers at best in Bogotá half the waste generated. Informal private operators do much of the pick-up, sorting, and recycling of Since 1998 Bogotá has implemented a comprehensive urban mobility strategy that includes promotion of nonmo- garbage, and communities sometimes provide the torized transport (bicycle paths), restriction of automobile service for themselves.59 But these informal solutions use during certain hours and days (approved by public ref- are rarely well integrated into a safe disposal system. erendum), and a bus rapid transit system (Transmilenio). Improper disposal of solid waste damages drainage Using exclusive busways on central lanes of major roads and a network of feeder buses and stations, the system systems, contaminates groundwater, and releases provides express and local services and carries 45,000 pas- methane, a potent greenhouse gas. sengers an hour per direction. Vehicle operations, passen- A key constraint to collective action and coordi- ger access, and ticketing services are carried out by private nation is the NIMBY syndrome (“not in my back companies through competitively tendered concessions. The new bus system is attracting ridership by former car yard�)—no community wants the waste disposal users and restoring respect for public transport.* By mid- site. Disposal is then neglected, or a facility is located 2001 the system had achieved high productivity (630,000 without public discussion near the population with trips per weekday) at a fare that fully covers operating the least political clout. Successes in public decision- costs, with no traffic fatalities. Some air pollutants have been reduced by 40 percent, and user travel time is down making on solid waste disposal facilities (sanitary by 32 percent. landfills or incinerators) suggest several lessons:60 Bogotá’s transport strategy benefited from strong mayoral leadership in articulating a long-term vision and Ⅲ First, public discussion should be early and open, representing the interests of noncar users despite the re- with site selection based on transparent criteria sistance of motor vehicle lobbies. The program also re- quired partnership between the private concessionaires agreed on in advance by candidate communities.61 and the municipal government, which financed and imple- Ⅲ Second, communities adversely affected should be mented the physical infrastructure and provided the dedi- compensated, through financial transfers or access cated road space.† to other desired investments. In Canada and the * Presentation by Peñalosa (2001), updated April 2002. United States the selected community typically re- † World Bank (2002a), box 8.2 ceives “host fees.� Ⅲ Third, the project sponsor needs to be credible in meeting commitments to minimize environmen- Planning and managing transport requires bal- tal impacts, through proper operation and man- ancing conflicting interests in an environment of un- agement. The facility should be monitored by the certainty and risks. Many decisions have long-term local community and local authority—and de- impacts and high costs, so coordination across the signed to retain functions for informal collectors, various modes of transport is needed. Cities that so that they can acquire less risky livelihoods. have managed to execute a comprehensive transport strategy, such as Curitiba, Bogotá, and Singapore,58 The problem of safe disposal is particularly acute have combined political will and leadership with in large cities and metropolitan areas where several technical and professional competence (box 6.4). municipal governments need to reach agreement Similar integrated approaches are necessary to ensure on siting and cooperate to share costs and reach traffic safety. economies of scale. Many large cities have been stymied by the absence of an appropriate governance Reaching a consensus and compensating losers— arrangement. In Monterrey, Mexico, a special-pur- sanitary solid waste disposal pose metropolitan authority runs a public company The production of solid waste (including hazardous to operate a sanitary landfill serving eight munici- waste) increases as economic activities shift from palities. In Santiago, Chile, 14 communes created a agriculture to industry, incomes rise, and lifestyles jointly owned corporation to build and operate a change. Its improper disposal can have environmen- landfill and gas recovery system through voluntary tal consequences. Managing waste removal and dis- cooperation rather than a formal metropolitan au- (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  thority. These examples show that collective action litical representation. They also reflect the failures of by governments, as by other social entities, requires government to guide and facilitate the growth of goodwill, trust, and a conviction that interests are low-income housing and basic services for incoming fairly balanced (in this case, reflected in burden shar- migrants through appropriate policy and planning. ing of costs and environmental spillovers from the These communities grow through the enormous en- disposal facility).62 trepreneurial energy of residents who build the city and provide its labor. With the right institutional en- Finding voice for dispersed interests—drainage vironment they can evolve more quickly into safe, Many cities also lack effective storm drainage sys- healthful, and hospitable urban neighborhoods. tems, and ill-planned construction closes off natural water courses. In Algiers, where massive flooding in Geographical and environmental manifestations November 2001 caused 800 deaths (700 in densely of exclusion populated neighborhoods), a natural water runoff Poverty and its many manifestations in cities can be channel in the city had been converted to a paved appreciated only by looking at disaggregated (espe- road. Overflowing of clogged storm drains and sew- cially spatially detailed) data. In Cali, for example, ers during high rainfall is projected to become a the incidence of income poverty is highest in periur- greater source of disasters in major cities than river ban neighborhoods with precarious environmental flooding.63 and infrastructural conditions. The eastern area (for- The key institutional issue is that drainage has no merly a lagoon) and the western zone (steeply sloped) clear constituency until major problems occur. Local are settled mainly by poor migrants and minorities governments may become motivated to act on living in very crowded housing (figure 6.2).64 drainage issues when flooding affects the business district, as in Cabanatuan in the Philippines, where Figure 6.2 the local business community put pressure on the Poverty in Cali, Colombia: 1999 headcount rates mayor to invest in drainage infrastructure. In Kam- pala the local authorities had neglected for years to protect past investments in the Nakivubo chan- 6 N nel from settlement encroachment and obstruction 2 with solid waste. There, and in communities in 5 Ethiopia, recent reforms expanding local democracy 4 7 have raised the profile of drainage as a priority for 1 3 8 public expenditure. 9 12 Inclusion and access to assets: Challenging the 20 11 13 14 19 institutional roots of urban slums 10 16 As noted in chapter 3, the evolution of good institu- 15 tions to solve coordination problems is itself deter- 18 mined by the extent of inclusion and access to assets by a wider public. Access to urban land—the city’s scarce natural resource and most durable asset—is Low: < 20% key to a city’s economic, social, and environmental Middle: 21–40% sustainability. Institutions need to allow people to 17 settle securely, so that they can envision a future for High: 41–50% their families and their city, while allowing for flexi- Very High: 51–70% bility in land use. Informal, illegal, or quasilegal Ext. High: > 70% neighborhoods with seriously substandard living conditions, often generically called slums, are an ob- Source: Departamento Administrativo de Planeación Municipal de Cali, vious manifestation of inequitable access to physical Colombia, 1999. and financial assets, to secure land tenure, and to po- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Reducing the disparities in welfare among resi- cities are combining census data and household sur- dents within cities is one of the starkest challenges veys with a geographic-referenced information sys- to a sustainable urban future.65 These disparities are tem to map service access and health outcomes by often masked by official data on access to water and neighborhoods. In Porto Alegre, Brazil, for example, sanitation, which do not accurately reflect problems the local authorities use a detailed environmental with the quantity, quality, and reliability of ser- atlas for planning and management and for educa- vices—or the numbers of people sharing facilities in tion in schools.70 dense settlements. In Accra, Ghana, for example, The multiple environmental health and safety only 12 percent of the richest quintile of the popu- risks in urban areas are related largely to the condi- lation, but more than 66 percent of the poorest tions and location of settlement. Hundreds of mil- quintile, share one toilet or latrine with more than lions of urban dwellers have few affordable options 10 other households.66 Some 44 percent of house- other than to live on sites (usually public lands) holds in Mysore, India, have water connections, but where development has not been approved and only 8 percent of those in informal settlements do.67 where residents are therefore not officially entitled to Such inequities help account for infant mortality urban services or protections. Such informal neigh- rates three or more times higher among the low- borhoods remain in squalid condition for decades. income households of many cities than among high- Since the home is also a major source of income income households (figure 6.3).68 Poor people in (both from rental and home-based industry) and the Accra and São Paulo have higher death rates not only household’s main private asset, the social and eco- from communicable diseases of childhood but also nomic burden of such physical conditions is pro- from respiratory and circulatory diseases and injuries found. Many slums are also disaster-prone sites—on from traffic accidents and homicides. The threats af- hillsides or floodplains, or near factories. Monsoon fect all age groups, creating a web of insecurity.69 flooding in Mumbai claims hundreds of victims Local groups and agencies concerned with envi- among the illegal occupants of hazardous areas—in- ronmental health problems and epidemiology in cluding the canals meant to drain the excess water. The population of urban slums is estimated by Figure 6.3 U.N.-Habitat at 837 million in 2001. Based on High inequality in health outcomes 1993 regional breakdowns, more than half are in in urban areas Asia accounting for one-third of the region’s urban population. Slums house more than one-half of all Urban infant mortality rate, second poorest income quintile urban residents in Africa and about one-quarter in 160 Latin America and the Caribbean.71 140 Slum neighborhoods typically have dispropor- MLI 120 tionately high concentrations of low-income people 100 EGY IND (though not necessarily the extreme poor, such as the 80 TUR homeless). They may also house middle-income res- IDN 60 DOM BOL idents in cities where formal provision of infrastruc- NIC PHL PER BRA ture and housing markets are very weak. Residents 40 COL of inner-city slums, typically settled for many years, 20 generally have better availability of infrastructure 0 (though it is often poor in quality and unreliable). 0 20 40 60 80 100 120 140 160 They also have more established communities and Urban infant mortality rate, richest income quintile less physical isolation than residents of newer settle- ments, usually on the outskirts. Both groups suffer Note: Data shown for all countries with data base for which adequate from the stigma of their neighborhood that impedes size sample is available (lowest quintile inadequately represented to be shown). Infant mortality rates measured as deaths before one year of their access to employment and to wider networks age per 1,000 live births. of social capital.72 Source: Demographic and Health Surveys (see www.measuredhs.com). Factors associated with crime and violence are also common in zones of deprivation within cities. The (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  highest homicide rates in Cali are found in its poor- paign for Secure Tenure and the Cities without est neighborhoods. Surveys of urban residents in Slums program of the multidonor Cities Alliance. Guatemala and Colombia identified tensions over Lack of secure tenure in urban areas has not been access to water as a cause of violence.73 Analysis of systematically measured. Even the designation of res- administrative regions or municipalities in São Paulo idents as homeowners or tenants does not convey State categorized their territorial exclusion based on protection from summary eviction when land regis- physical hazard, provision of urban services, and se- tration and the rule of law are poorly enforced.77 Se- curity of tenure. Municipalities with the most pre- cure tenure is part of a country’s hierarchy of rights, carious living conditions had the highest homicide ranging from legal titles and contracts to customary rates, and those with the least territorial exclusion recognition of use rights.78 Providing secure tenure were the least violent. The regions with the worst therefore does not pit the rights of squatters or ten- outcomes also had very high income inequality.74 ants against those of private property owners and Such exclusion contributes to frustrated expecta- landlords, who should be protected under contract tions, defeated hopes, and mistrust in society’s future. law. But countries permitting arbitrary eviction often Even in some transition economies many of the also fail to enforce private real estate contracts and urban poor live in severely substandard, quasilegal otherwise obstruct the private rental market, further settlements (for example, 15–25 percent of urban res- disadvantaging low-income citizens. idents in the Former Yugoslav Republic of Macedo- By confirming the rights and responsibilities asso- nia, many of them ethnic minorities).75 But not all in- ciated with the occupation and use of land, regulariz- formal settlements feature low-quality housing. Some ing tenure status removes a major source of economic illegal or irregular housing is produced by commer- and political insecurity for households and for com- cial developers or politically influential parties who munities. It reduces some of the risks that discourage speculate that property investments will be regular- residents from investing in their houses and shops— ized later (akin to the race for land rights at the agri- and gives them a stronger stake in urban society and cultural frontier described in chapter 5). Such land an incentive to work with local officials to obtain ser- speculation is encouraged when countries lack clear vices. A study in Indonesia found that stronger tenure policies on tenure security, and authorities are unable security increased the probability of demanding to balance interests and articulate public choices re- garbage collection.79 And surveys of slum dwellers in garding land use, or commit to enforcing them. Bangalore reveal that better tenure status has a signif- icant and positive impact on willingness to engage in Empowerment through access to assets: security collective action to obtain urban services, even in cul- of tenure turally heterogeneous communities.80 Although slums reflect institutional failures in hous- A growing commitment by city, state, and na- ing policy, housing finance, urban planning, public tional governments of Brazil to regularize slums or utilities, and local governance, one of the most fun- favelas has put in train a process of transformation damental failures is the absence of tenure security. (box 6.5). A key turning point was the 1988 federal Security of tenure means “protection from involun- constitution, which strengthened the role of local tary removal from land or residence except through government and encouraged municipal policies to le- due legal process.�76 The emphasis is thus on pre- galize and improve tenure conditions in these infor- venting forcible and arbitrary eviction, whether of mal settlements. A groundbreaking new city statute individual households or entire settlements. The sig- enacted at the federal level in July 2001 provides a nificance for urban poverty underlies its inclusion legal underpinning for municipalities to regularize under the Millennium Development Goals (MDGs) favelas as part of concerted plans to combat spatial as indicator 31 (“Proportion of people with access to segregation and social inequity—and to create more secure tenure�), related to Target 11 (“By 2020, to inclusive and democratic urban governance.81 have achieved a significant improvement in the lives Security of tenure is both a collective good and a of at least 100 million slum dwellers�) and Goal 7 private good in the urban context.82 Whole commu- (“Ensure Environmental Sustainability�). The issue nities are threatened when shanty towns are bull- is also a focus of the U.N.–Habitat Global Cam- dozed, while residents individually gain security (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 6.5 Regularizing favelas in Brazil In many cities in Brazil, large shares of the population—25 per- favela or informal settlement upgrading program and about cent of the residents of Rio de Janiero and 40 percent in met- 506 of these include some form of land tenure regularization. ropolitan Recife—live in informal or illegal settlements, often Where tenure regularization policies have intended to on public lands. These favelas are home to an essential work- transfer full individual freehold titles to the occupiers of public force—a workforce subject to terrible health conditions, fre- or private land, as in Belo Horizonte and Rio, this aspect of the quent natural disasters such as mudslides and floods, and program has been problematic to implement and less success- crime. Official policy toward favelas in the past was that of ne- ful than the physical upgrading and service provision. Other glect (with occasional introduction of services when politically municipalities, such as Porto Alegre and Recife, have used an expedient or necessitated by emergency) and threats of evic- innovative alternative legal instrument to promote individual tion. Not until the 1970s did most municipalities begin to even and community security of tenure. This formulation, the “con- include such settlements on planning maps as provisional, de- cession of the real right to use� (CRRU), is a leasehold that spite their existence for decades in many cases. Transforma- confers private property rights to publicly owned land for a pe- tion in these settlements has started to occur in recent years riod of up to 50 years, either for an individual or a community.† where local governments, supported by their state and the na- Combined with the designation of settlements as “zones of tional government, have made commitments to sociopolitical social interest,� the CRRU protects residents from eviction as well as physical inclusion of the favelas into the city. and gives them broad property rights. This instrument permits Beginning in the early 1980s a number of cities, most no- the state to protect access for the low-income communities tably Belo Horizonte and Recife, initiated efforts to regularize to land they occupy in order to promote socioeconomic inte- or integrate the favelas into the urban fabric and give them gration of the city; it also serves to preserve scarce public land legal recognition. New planning instruments were introduced for current and future social uses. Settlements granted such at the national level to permit designating certain settlements use rights have gained physical improvements from private as “special residential zones of social interest� (ZEIS), which and public investment in housing and infrastructure, and in- permitted planning and zoning regulations to be adapted to the creasingly take on the appearance of working class neighbor- land use requirements of these communities. In Recife, a fur- hoods physically integrated with the adjacent areas. ther mechanism (PREZEIS) was established in 1987 to institu- Though still largely untested, the new national City Statute tionalize, for the first time, the process of integrating irregular gives municipalities the tools to go even further in regularizing settlements into the formal planning apparatus, with commu- informal settlements. The Statute includes, for example, provi- nity participation, and allow for the provision of services and sions for facilitating the transfer of privately owned land to the infrastructure to reduce disparities. Under this law Recife cre- existing occupiers in cases where occupation has gone uncon- ated a land tenure legalization commission charged to identify tested for at least five years. To complete the transformation in and address specific problems in each ZEIS through participa- the quality of life and social inclusion of the urban poor, these tion of multiple stakeholders—a device credited with enforc- political and legal commitments and investments need to be ing the government’s commitment to follow through with its supplemented by a broader set of policies that also promote regularization program despite resistance from conservative economic opportunity and counter other dimensions of poverty. sectors. The state of Pernambuco has joined Recife’s efforts by bringing investment resources to help cover the settle- * The first phase upgraded about 60 favelas and 20 irregular subdi- ments designated for regularization across the metropolitan visions, affecting some 250,000 people, supported by an Inter- area. American Development Bank loan in 1995. A second phase and loan Programs with similar objectives have been adopted in in 2000 applies to 56 favelas and 8 irregular subdivisions, for about other Brazilian cities, including Porto Alegre, Rio de Janiero, the same number of residents. The total costs of upgrading under and São Paulo. The Rio program (where the state government these projects average $4,000 per household, which compares has also reversed its past resistance to favelas, by providing favorably to most housing programs and to many sanitation and finance for building materials for residents without requiring social services projects. Brakarz (2002). † In Brazil, the CRRU can be registered, allows transfer of right to collateral) is notable for its scale.* In Belo Horizonte and Porto legal heirs, selling, renting and use of land as collateral. It can be used Alegre the programs entail a strong emphasis on participatory as both individual and collective right (a form of condominium), and budgeting and planning for investments in the settlements. A has a gender dimension, as women are given priority treatment in 1998 study by the Brazilian Institute for Applied Economics in- granting of use rights (Payne and Fernandes 2001; Fernandes 2002). dicates that at least 794 municipalities have some kind of Source: Cira, background note for the WDR 2003. when their settlement is accepted as an integral part for associated rights as urban citizens.83 The Railway of the urban fabric. Often there is enough stability Slum Dwellers Federation in Mumbai (RSDF) suc- of inhabitants that they confirm each others’ rights cessfully managed the resettlement of member house- to residence. Community organizations in informal holds to permit a transport project with citywide settlements use their strongest asset (social networks) benefits (box 6.6). to protect their area from encroachment by new en- How important is a legal title? Experience in many trants, to resist involuntary resettlement, and to push developing countries confirms that households and (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  Box 6.6 vide their own housing.86 In the transition econo- How railway dwellers in Mumbai managed their mies of Europe and the Former Soviet Union, some own resettlement of the incentive effects of housing privatization have been limited by the informational and financial im- In Mumbai, the commercial capital of India and home to 12 million people, some 24,000 families have lived for al- pediments in the real estate markets.87 most two decades along heavily traveled suburban rail Titling can even worsen poor people’s overall ac- lines, with some huts hardly a meter from the tracks. Be- cess to affordable land and housing. That can hap- sides risking death and injury, these residents suffer from pen when titles are extended only to certain settle- a near-total absence of basic services. A project to improve the city’s traffic and transportation ments, when tenants are forced out by higher rents system required the resettlement of these slumdwellers. after titling, or when slumdwellers are resettled into To represent civil society in the resettlement plan, the Ma- new neighborhoods with deeds of ownership but no harashtra Government task force sought the participation of an alliance of The Society for the Promotion of Area Re- assistance in acquiring infrastructure services.88 Ef- source Centers (SPARC; a registered NGO), the National fective demand for legal titles is therefore often less Slum Dwellers Federation (NSDF), and a savings coopera- than would be expected by observed increases in tive of women slum and pavement dwellers. A constituent land values.89 unit of the NSDF is the RSDF, made up of the Mumbai families who would have to move for the railway project. Legal titling can also have significant administra- By June 2001 the alliance had resettled 10,000 families, tive costs, especially since many cities lack good in just over a year, without force, to accommodations with cadastres, leading to protracted legal battles over assurance of secure tenure and basic amenities of water, ownership status. The COFOPRI program in Peru sanitation, and electricity. How was this done? The Mum- bai Metropolitan Regional Development Authority in charge achieved a much higher volume of legal titling (one of the railway project was willing to give up some of the million deeds issued in four years) than did similar powers normally held by government agencies in resettle- efforts in other countries. But this has been possible ment and rehabilitation—determining eligibility, obtaining because of the large tracts of government-owned baseline information on the community, allocating housing. Such functions, which provide opportunities for rent-seek- land on the urban periphery. Titling in older urban ing and corruption, were ceded to the NGO alliance. areas of the country has progressed more slowly be- Long before the railway project was initiated, the RSDF cause of ownership disputes.90 An alternative to for- had collected information on the railway dwellers as a means of community mobilization and had the trust of its mal titling and cadastres that can be quite effective own members as a resource for the resettlement process. and easy to implement to aid service delivery and The households agreed on the criteria for allocating perma- acknowledge occupancy is street addressing—the nent and temporary accommodations. In the new settle- mapping and naming or numbering of streets and ments the families have formed lending cooperatives to compensate for income forgone as a result of the move. homes in informal settlements. This system, used in The experience shows that a mobilized and self-gov- 15 West African countries, aids utilities in billing for erning community of poor people can act collectively for services and permits the simple taxation of plots.91 its own good and for the good of the larger urban society When residents in informal neighborhoods do not when there is mutual trust and flexibility on the part of the community and government agencies. fear arbitrary eviction, they can devote their social cap- ital to negotiate claims for services with local govern- Source: Burra (2001a). ment or utility companies and take collective action to improve their settlements.92 In Pune, India, a resi- dents’ organization used similar techniques of com- communities realize significant benefits in moving munity mobilization and self-assessment as in Mum- from highly insecure tenure to de facto tenure (more bai to respond to their own demands for sanitation, secure though not fully legal).84 Granting individual after these had been long neglected by the municipal legal titles (freehold) is sometimes expected to con- authorities. The Pune slumdweller alliance (mainly fer additional advantages, such as greater access to the women) assessed the needs of residents, managed housing loans and a more active housing market.85 the construction with innovative designs to meet the But this requires supportive banking and real estate needs of different user groups, and set up effective institutions, which have often not materialized be- payment and maintenance arrangements. Their ef- cause of other limitations, including low incomes forts resulted in a record outturn of latrines in just a and the reluctance of lenders to finance incremental few years, serving about half of the city’s 1 million home improvements, the main way poor people pro- slumdwellers.93 In Santo Domingo, the Dominican (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Republic, residents’ associations of three low-income Empowering the appropriate level of actors settlements took similar steps. These settlements, lo- and ensuring coordination cated on the sides of ravines are vulnerable to frequent The environmental and social assets needed for sus- landsides. Association members, mainly the women, tainable urban development get more complex with designed and managed their own disaster mitigation increasing scale of settlement. Moving from neigh- program by building retaining walls and making other borhood to city, to region, and to nation implies infrastructure improvements.94 more extensive environmental and social linkages and impacts, increasing the divergence of interests Institutions for sustainable urban development and potential for conflict, and greater technical and Good urban governance requires institutions to re- institutional requirements for coordinating those di- veal and balance divergent interests and to commit vergent interests. to solutions for collective welfare. Some institutional Taking responsibility for urban services and spill- arrangements are particularly important to ensure overs at the lowest practical level—the principle performance of these functions across the range of of subsidiarity—is a basic condition for mobilizing urban issues: collective action. Subsidiarity empowers those with most at stake and strengthens the legitimacy of higher Ⅲ A structure of responsibility sharing and coordi- government through power sharing. Decentralizing nation that links the community, local govern- urban services to local government is desirable for ment, and provincial and national levels of gov- enhancing the voice of urban citizens and the access ernment and empowers the appropriate actors to to credible information, but it requires the respec- address problems at each level tive authorities to be accountable and have the Ⅲ A forum for wide participation in strategic think- means to address problems at their level—and that ing, to enable common understanding and con- depends on the framework of intergovernmental fi- sensus, motivate actions, and assess progress nancial relationships.96 Ⅲ Networks for communication and capacity build- Political tensions across levels of government are ing among practitioners and stakeholders. common. Central governments frequently impose unfunded mandates on local governments, and local As noted, some informal arrangements that work governments may innovate without getting adequate in rural areas can be paralleled in urban communi- support from central governments. Local govern- ties, especially in neighborhoods. But such informal ments can also be less progressive than central gov- institutions are often asked to do too much. They ernments and obstruct needed reforms. In China, can stimulate private enterprise but not enable firms some local governments resisted disseminating in- to grow. And they can provide support to house- formation on city environmental conditions well holds but not confer all the services, economic secu- after the State Environmental Protection Agency had rity, or political legitimacy that the population de- authorized it, while others (in Jiangsu province) serves at the scale the city requires. The greater scale moved on disclosure policies—experimentally—in and complexity of urban life thus require effective advance of the central government’s commitment.97 formal institutions that operate with greater pre- Collective action in neighborhoods. Local environ- dictability, transparency, and adaptability. mental problems, such as removing solid waste from What prompts such institutions to emerge and the neighborhood, can often be addressed by coordi- grow? Often major changes seem to come about nation at the community level. The commonality of through sudden crises—a disease outbreak or nat- interests makes collective action possible either to ural disaster—or the rise of a charismatic leader. In solve problems internally or to obtain what is needed recent years democratization and fiscal decentraliza- from other parties (government or utilities). Grass- tion have given new legitimacy and authority to local roots and community organizations that move be- governments (chapter 7).95 And globalization has yond confrontation to engagement with city govern- been creating new opportunities, bringing in new ment have greater prospects for obtaining sustainable knowledge, and raising new expectations for address- benefits for the urban poor.98 Experience in Pakistan, ing urban challenges. the Philippines, and Thailand with programs for (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  basic infrastructure and housing improvements in Ⅲ Increased democratization (electoral processes) low-income neighborhoods demonstrates what com- Ⅲ Good practices and incentives for sound financial munity associations can achieve as key players in management partnership with government and the private sector. Ⅲ Public participation and access to information in Those partnerships, however, require a long-term budgeting and investment planning commitment to sustain them.99 Ⅲ Upgrading of skills and professionalism of govern- The main weakness in many such programs has ment staff been the lack of continuity in financial and political Ⅲ Monitoring and evaluation based on benchmark- support from formal institutions. In particular, city ing and client feedback. agencies and utilities need to assume responsibility for scaling up and maintaining infrastructure net- Many cities demonstrate innovative relationships works and such services as drainage, lighting, and between civil society and local government to in- parks beyond the neighborhood. Links with off-site crease pressures for performance in the execution of infrastructure (roads or solid waste disposal sites) are basic functions. Reform initiatives often gain ground often inadequate because of weak coordination with through collaboration on concrete activities of the responsible government agencies, reducing the envi- local government—such as the public review of mu- ronmental and other benefits. nicipal budgeting and procurement in Obninsk, Many urban neighborhoods have also used their Russia,101 and the citywide referendum to affirm res- internal social networks to create protective institu- idents’ willingness to pay for infrastructure improve- tions to ensure local public safety, such as through ments after flooding in Tijuana, Mexico.102 Such ini- community “crime watch� activities—but this also tiatives can then launch wider, deeper processes of requires the close collaboration and support of for- reform in other areas. Obninsk is now influencing mal institutions, such as the police. In the Warwick reforms at the provincial level, and Tijuana trans- Junction district of Durban, South Africa, 50 traders formed its disaster recovery plan into a series of in- from the community voluntarily patrol around the novations linking taxation and public work improve- clock. The group was trained by metropolitan police ments. Participatory budgeting, which was initiated on citizen’s arrests, constitutional rights of individu- in Porto Alegre and has now spread to over 80 cities als, and court procedures to ensure successful prose- in Brazil, has dramatically transformed relationships cutions. Their efforts have contributed to a reduc- between civil society and local government.103 tion in crime, and new trust between the community As part of urban capacity building and decentral- and police has improved the rate of successful police ization in Senegal and Guinea some local govern- investigations and prosecutions. But relations are ments have institutionalized public consultation in still strained by the citizens’ perceptions of inade- the production of their public investment and main- quate formal policing in the city.100 tenance programs and embedded these programs in Strong local (municipal) government. Most issues a contractual agreement between the city and the of sustainable development in urban areas extend central government. This municipal contract com- beyond an individual neighborhood and require a mits the authorities at both levels to enforcing the permanent formal mechanism of collective action, agreed financial implications and reform measures through effective local government that works with promised to constituents.104 communities but shoulders broader responsibilities. Collaboration across jurisdictions—metropolitan This is where clashes of interests among larger and management.105 The wide impact areas of many en- more powerful stakeholder groups—and tradeoffs vironmental spillovers, and the interdependence of about priorities and who gets served—begin to occur, major economic activities sharing spatially contigu- underscoring the need for representative governance. ous infrastructure networks and other services, re- Reforms to make municipal governments and quire collaboration among local governments, as well their agencies more accountable and transparent to as regional and national governments. This is espe- all constituents are thus at the core of improving sus- cially true for systems implying economies of scale— tainable development in cities. The reform agenda citywide transportation, water resources manage- comprises: ment, pollution control, landfills, and wastewater (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  treatment. But this next level in the hierarchy of col- Box 6.7 lective action presents an even greater potential di- Mexico City’s search for metropolitan management arrangements vergence of interests, especially over sharing costs and benefits. And fragmentation and excessive competi- Mexico City has explored several organizational arrange- tion among municipalities are more the norm. In the ments in recent decades. The urban agglomeration in the more than 70 countries undergoing decentralization, Valley of Mexico starts with the Federal District, the seat with the notable exception of India, municipal laws of the central government, an area with 10 million people and constitutional reforms have had the largely un- divided into 16 boroughs or delegations. Surrounding the Federal District, but seamlessly connected in functional intended effect of weakening the prospects for met- terms, are 12 contiguous municipalities (another 7 million ropolitan solutions to large city problems. people) in the State of Mexico. An even more encompass- A variety of organizational arrangements—formal ing definition (the Mexico City Region) covers nearly 100 and informal, in developing and industrial coun- municipalities in five states. tries—have evolved over the years to meet the chal- Various metropolitan commissions have been created to cover key areas of need. A water commission was lenges of managing large cities. In one common pat- formed more than 30 years ago to plan and implement a tern (as in Dhaka and São Paulo), jurisdiction for mammoth system of interbasin transfers to supply water specific functions is assigned by geographic area, cre- to Mexico City. Similarly, an air quality commission has ating many general purpose local governments that steadily grown to manage mobile sources of pollution. The commission has successfully eliminated lead from fuel may cooperate for specific purposes, such as gather- and is working to reduce the volume of traffic and to im- ing regional data or sharing the costs of expensive prove vehicle operating efficiency. equipment and facilities. Sometimes this collabora- A persistent problem, despite half a dozen planning tion results in a specialized metropolitan area or dis- and special purpose commissions, is coordinating the city’s growth with water and transport infrastructure—for trict service, in which limited powers—usually for example, having mass transit connect residential areas to planning or preservation, as of watersheds or re- key concentrations of employment. Recent political re- gional parks—are ceded to special authorities. forms making the office of the mayor an elected position Formally constituted metropolitan authorities, have sparked an active political debate about creating an entirely new metropolitan authority for Mexico City. created or authorized by national law, are less fre- quent. In a pattern called functional fragmentation, Source: Campbell, background note for the WDR 2003. lower-tier governments are limited to basic functions such as water distribution and street lights, while a second tier handles areawide functions, such as free- all followed an iterative path, moving first (in the late ways and water trunk supply lines, as in Mexico City 1980s) from formal metropolitan bodies with execu- (box 6.7). These second-tier governments can be au- tive authority to a system of fragmented, independent tonomous bodies, sometimes with executive powers municipalities. By the end of the 1990s these cities to carry out projects such as the development author- had shifted back to various centralized systems, but ity in Kolkata. In other cases, as was common in with more democratic input from elected or appointed Latin America in the 1970s and 1980s, the metro- citizen groups. politan agencies have only a consultative role. The As these and other arrangements evolve, interna- most formal arrangement—that of centralized met- tional sharing of experience will be important. Lead- ropolitan organizations as in Bangkok, Kuala Lum- ership from national or provincial governments is pur, and Seoul—is fairly rare and usually imposed by often necessary to provide functional assignments central government to manage a capital city. and funding for metropolitan arrangements, since Some cities, such as Johannesburg, have blended local governments do not cede powers easily, espe- these models—and the original prototypes are also cially when it involves redistributing tax revenues changing as demands shift. Globalization exposes across jurisdictions.106 cities to forces requiring a wider basis for manage- ment and planning—and more strategic direction. Building consensus and strategies for sustainable Metropolitan arrangements also seem to be respond- urban development ing to the need to connect voter-taxpayers with mech- An essential condition for coordination is having a anisms of public choice. In the past few decades, Lon- shared understanding of the problem, knowing the don, Montreal, New York, Ottawa, and Toronto have costs and benefits of alternative solutions, and ac- (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  commodating the concerns of different stakehold- vironmental conditions and prospects, and spell out ers. Two relevant sources of experience in revealing priorities and action plans for both policy and invest- interests and reaching a consensus on what to do are ment. Some examples: local environmental action plans and city develop- ment strategies. Ⅲ A city development strategy in Cali, Colombia, Since the Earth Summit in 1992 some 6,400 local helped explore stakeholder views on major public authorities in 113 countries have either formally projects and led to changes in the city’s investment committed or actively undertaken to produce local priorities. The first phase, which also prompted environmental action plans (called “Local Agenda discussion of violence as a key issue for the city, 21s�).107 These initiatives integrate environmental pointed out the need for better understanding of objectives into development plans, emphasizing par- the local economy and employment constraints. ticipation and accountability. They articulate local Ⅲ In San Fernando, in the Philippines, a city devel- concerns and motivate local stakeholders around opment strategy helped reorder investment prior- shared priorities for the area’s future. And they pro- ities for sanitation, among other outcomes. vide a basis for coordinating the work of different Ⅲ Santo Andre in São Paulo State focused its strate- levels of government and sectoral agencies. gic planning on social inequities and exclusion— The city of Manizales, Colombia, with about and undertook social mapping to target and mon- 360,000 inhabitants, has formulated a local environ- itor actions to reduce disparities. mental action plan with wide consultation and inte- grated it into the municipal development plan and A good practice for institutionalizing city strate- budget. The plan included measures to revitalize the gies is to incorporate their key elements into the reg- city’s architectural heritage, improve public transport, ular systems of city planning. While sustaining and strengthen watershed management of the Chinchina implementing strategic planning remains a chal- River which provides water to the city aqueduct, re- lenge, such efforts can reveal the priorities of various duce risks of landslides, create ecoparks, and define stakeholders and contribute to public pressure for community environmental action plans. The plan change.110 Two items that should be on the agenda also devised an innovative indicators program of of most city development strategies are: Getting ahead “environmental traffic lights� to signal progress.108 of the expanding urban frontier; and making urban Not all Agenda 21 exercises have been as partici- density affordable and livable. patory as planned, and the momentum for longer- Getting ahead of the expanding urban frontier— term implementation has often waned—the dif- guiding new settlements to prevent future slums. Cities ficulty of forging long-term commitments being one and towns in developing countries will need to ac- of the barriers to successful coordination noted in commodate the projected doubling of urban pop- chapter 3. Success has been greater with sustained ulation over the next generation. Even with insti- support by successive mayoral administrations, strong tutional reforms to upgrade and integrate existing participation by a local NGO or university, and ef- slums, new ones may form. Local governments have forts to mobilize funding for local economic develop- often shied away from acknowledging the need to ment. Most Local Agenda 21s have been undertaken anticipate and facilitate the growth of low-income in smaller cities, perhaps indicating the difficulty of settlements, instead letting them fend for them- consensus-building at a metropolitan scale. National selves. Providing infrastructure networks after the government leadership has also been instrumental in fact is much more costly, however, especially for very replicating Local Agenda 21s within countries. dense settlements with irregular layouts or requiring City development strategies are similar efforts in resettlement. In Bogotá the urban development participatory strategic planning, but with a poten- agency estimates that installing drainage networks is tially broader focus of integrating environmental and about three times more expensive in informal settle- productivity concerns within a propoor perspec- ments than in planned neighborhoods.111 tive.109 They have been used to build consensus on Political will is essential to create an institutional a vision for the city and on the steps to achieve the environment that senses and anticipates the de- vision. These strategies typically include a participa- mands from new entrants and permits forward tory assessment of the city’s economic, social, and en- thinking and partnership among government, pri- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 6.8 Leading the advance on urban settlement growth in Lima Lima, Peru, 1977—As we stood on a hill overlooking a new set- of the dynamics of settlements that John F.C. Turner had de- tlement on the edge of a vast desert plain about 10 kilometers veloped in Arequipa and Lima, SINAMOS set out to meet the northeast of downtown Lima, the boy, about 7 or 8 years old, challenge of rapidly growing squatter settlements on the out- said he was an engineer. An engineer? “Well,� he said, “I am skirts of Lima.* They called the low-income communities pue- helping the people draw the lines for the lots and build our bar- blos jóvenes, which gave the squatter settlements a positive rio.� The settlement, which consisted of little more than chalk image. lines and shanties of woven reeds and plastic sheets, was SINAMOS developed a two-pronged approach. It upgraded growing day by day, as new arrivals were trucked in by the Na- existing pueblos jóvenes, relying on community participation tional System for Social Mobilization (SINAMOS). The settlers and relatively little public investment. Then, as squatter settle- helped the surveyors lay out the plots and clear areas for play- ments continued to appear, it began a massive slum preven- grounds and community facilities. tion program, providing surveyed plots to meet anticipated de- The core group of settlers had initially organized itself to mand for low-income housing. The first minimal structure plan squat on public land under high-tension power lines. They had was gradually detailed down to neighborhood layouts and en- come mostly from slums in the city center, where they rented gineering studies for trunk infrastructure. rooms or lived with family. On the agreed-on night, walking in Twenty years later the pueblos jóvenes had become well- small groups and with plastic sheets wrapped around their consolidated, low-income and low-middle-income neighbor- bodies, they converged on the selected site and built their hoods with most urban services, schools, clinics, markets, and tents and shacks overnight. Having established themselves as other amenities. The key to the program’s success was pick- a squatter settlement, they knew SINAMOS would remove ing up signals of demand, balancing interests, and committing them to a permanent, if unserviced, settlement, one of the to implementation—matching social coordination with the as- new pueblos jóvenes, or young towns. pirations of poor people. SINAMOS, staffed mostly by young and deeply committed engineers, architects, and social workers, was established in * Turner and Fichter (1972). the mid-1970s. Armed with enthusiasm and an understanding Source: Chavez, background note for the WDR 2003. vate investors, and households. Valuable experience replication of the scheme. But the authorities could in planning low-income settlements has come from not commit to fully implementing this part of the sites and services programs, usually initiated by local plan when no private developers joined in, and re- government or its agents, to provide basic plot lay- sulting delays led to land disputes as squatters moved out and minimal infrastructure (such as core sani- into the area. In El Salvador, however, a private com- tary facilities) in advance of spontaneous devel- mercial company (ARGOZ) has carried out a prof- opment. Such a program in Lima aimed to prevent itable land development scheme for low-income the growth of squatter areas by anticipating demand households for more than 25 years, with the help of (box 6.8). a conducive legal framework for urban land conver- Forging long-term commitments is key to suc- sion and a determination to keep design standards cessfully getting ahead of the frontier. A strong sup- affordable.112 ply response is important in making such schemes Making urban density affordable and livable. Ac- sustainable. Many well-meaning programs have been commodating the growth of population in cities will stymied by the lack of affordable land and hous- involve both physical expansion at the periphery ing—even for the middle-income groups. This has and, in many cases, increased density of settlement reduced political support for minimal design/mini- in the city. Average urban densities are already vastly mal cost approaches, and poor people have been greater in major cities in developing countries, espe- pushed to the end of the service queue. In Conakry cially in Asia, than in North America and Europe— the municipality’s basic structure plan aimed in part for example, Mumbai has almost 400 persons per to set aside periurban areas for new settlements and hectare and Shanghai 500, compared with about to test public-private partnerships for the production 170 for the Barcelona metropolitan area, and 40 for of serviced plots. The government planned to facili- New York.113 The key issue for mature cities of de- tate connection of the sites to infrastructure net- veloping countries is to provide the infrastructure works while the private investors, through payment and services (especially sanitation, public transport, of an “equipment contribution,� were to ensure and green spaces such as parks and playgrounds) (c) The International Bank for Reconstruction and Development / The World Bank ������� ��� ���� ���� ������  needed to make already high densities livable and ef- security, and they already pay dearly for whatever in- ficient, with ease of access to homes, work, and other frastructure they are able to acquire from informal places. The challenge for cities that have not yet markets. So, regularizing informal settlements and fa- filled in is to avoid making such spatial development cilitating low-cost land and housing development unaffordable to their growing population. should be the highest priorities to ensure more equi- Governments often try to control city size and in- table access to urban assets and more healthful, attrac- fluence the spatial form of city growth by regulating tive living conditions in developing-country cities. land use—through rules on minimum plot sizes and These measures, coupled with appropriate allocation road widths, for example. It is common for cities to of urban land for public purposes—right of way, en- present inconsistent signals to investors: an official vironmental easements, and so on—can transform stance extolling compact urban form, countered by the institutional basis for the evolution of urban form. regulations and financial practices promoting low density land uses that favor middle- and higher- Promoting institutional learning and leadership income groups. At the start of the economic transi- through networks tion in Cracow, Poland, municipal officials advo- Institutions for sustainable urban development need cated higher density development of the inner city— to embody incentives and processes for learning to yet rigid zoning persisted, inconsistent with both the better solve existing problems, and anticipate and planners’ intentions and the market incentives.114 prepare to deal with new problems. Increasingly this When land-use regulations aiming to limit densi- stimulus occurs through networking by local govern- ties are strictly enforced, as in Brasilia, they drive up ment and nongovernmental groups, through both the cost of inner-city housing and force poor people associations and Web-based communications. Net- to the periphery, where infrastructure services and works foster communication among peers, dissemi- transport are unavailable or expensive.115 Even cities nate innovations, and enhance reputational pressures that try to encourage development around public for change. They can also instill professionalism and transport zones (such as Curitiba, Brazil) or to curb high standards of performance, cultivating leader- periurban expansion by imposing a green belt (such ship. Increasing forums for public feedback also helps as Portland, Oregon) have seen denser land uses shift identify mistakes and make mid-course corrections. to the outskirts. In Mexico, the lack of financing for A prime illustration of networking is the asso- rehabilitation of existing housing or for most multi- ciation of local, national, and international NGOs family units, as well as vestiges of rent control, deter promoting the empowerment of communities and improvement of inner-city neighborhoods and pro- women in India and other countries. Through the mote development on the urban periphery.116 alliance of SPARC, Mahila Milan (a women’s savings Urban growth controls aimed at tightly regulating cooperation), NSDF of India, and Slum Dwellers densities and building codes can make access to urban International (see box 6.6), practitioners and the assets of land and housing more inequitable.117 Zon- urban poor share experiences on housing, urban ser- ing that permits mixed land uses—consistent with vices, and security of tenure. And by distributing in- how low-income neighborhoods develop naturally— formation gathering methods (such as a self-census is more advisable to keep jobs, services, and housing of slumdwellers), negotiating skills, and encourage- accessible. Fiscal and other policies that charge devel- ment across the city and country and communicat- opers the full costs of providing the incremental in- ing with similar groups elsewhere, the alliance is in- frastructure required for new settlements are also nec- creasing the scale and sustainability of its efforts. essary to internalize the social cost of expanding urban Local governments are also networking interna- development. These charges, which should be intro- tionally to learn from each other. National, regional, duced before spatial expansion patterns are locked in, and international associations of local governments can be a combination of development impact fees and diffuse technical assistance, training, and ideas to general taxation linked to property values. member cities on a wide range of planning, opera- But such charges will not reduce the demand for tional, and fiscal issues.118 The Union of Capital low-income settlements at the periphery in develop- Cities of Ibero-America (UCCI) runs workshops ing-country cities. Residents in these areas are not and a Web site to help member cities learn about subject to formal taxation as long as they lack tenure municipal modernization, solid waste management, (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  urban transport, cultural heritage protection, and ment can help, but city stakeholders should take other issues.119 The first cohort of Philippine cities much of the initiative. that carried out city development strategies is help- More fundamental changes would have wider and ing others do the same, as part of a growing urban more lasting impact—and would be more instru- knowledge network involving the Philippine League mental in building new institutions. These deeper of Cities, the China Association of Mayors, and reforms include granting secure tenure, which can other national groups in East Asia.120 The Clean Air transform the balance of power between urban poor Initiative, a consortium of donor and private fund- people and the rest of the urban society. Increasing ing, is helping build capacity among cities in several the openness and accountability of local govern- regions—for example, to extend to African cities the ment, through democratic processes and partici- Asian and Latin American experiences with remov- patory procedures, would also increase the respon- ing lead-based fuel (chapters 3 and 7). siveness to the interests and problems of poorer Networks also help to create incentives for sus- constituencies and the legitimacy of government tained collective action by building reputational pres- actions. Although these reforms are long in impact, sures within peer groups. Professional associations of experience shows that they can be initiated fairly local governments provide advice and standards on quickly—provided there is political will. More com- performance indicators that can be compared or plex measures—such as devising metropolitan man- benchmarked among member cities. Numerous ex- agement arrangements, and helping cities mitigate ternal rankings of cities on quality of life or attrac- disaster risks and adapt to threats from climate tiveness to investors have been widely publicized— change—may require more creativity, leadership, and have sometimes provoked corrective action.121 and resources. They also require greater and sus- tained support from national institutions. Conclusion Much of the future physical development that Achieving sustainable urban development requires cities require can come at a lower cost when problems forward-looking institutions that sense emerging are recognized sooner rather than later; for example, problems, balance interests (especially by heeding setting aside rights-of-way for primary transport routes the disadvantaged in society), commit to effective and parks and green spaces, facilitating new low-cost execution of agreed solutions, learn, and adapt. To settlement, and guiding land development away from make such institutions emerge and function well, it precarious or environmentally fragile areas. Invest- is necessary to confront basic inequities in access to ments to protect environmental health locally can also assets, to empower dispersed interests and balance be made effectively and cheaply with the participation them against vested interests, and to build consti- of residents to identify and carry out appropriate so- tuencies that can represent and commit to longer- lutions. Significant advances in the quality of life of term concerns. the less advantaged urban residents are possible when Priority actions to reveal problems and divergent there is a shared commitment to integrating them interests include developing disaggregated datasets, fully into the life of the city; when there is flexibility such as mapping environmental hazards within a on the part of government and private service pro- city. Wide dissemination of such information, along viders; and when there is an openness by formal insti- with the costs and benefits of alternative solutions, tutions to creative solutions developed by a diverse is essential to building constituencies for action. Bal- array of residents, actors, and networks in a city. Many ancing interests and forging consensus can be facili- activities in one location or community have conse- tated through participatory strategic planning, aided quences that affect other locations or communities. by networking among practitioners and local gov- The principle of subsidiarity requires that these spill- ernments for sharing local and global knowledge, in- overs be addressed at higher levels—a principle based novations, and reputational pressures that stimulate on matching the span of the spillover with the span leadership. These measures to strengthen the work- of the jurisdiction best able to internalize the prob- ings of institutions are fairly low in cost and can be lem. The principle of inclusion ensures that people’s implemented in the short term with existing capaci- well-being is a priority to be addressed at national and ties and resources. Support from the central govern- global levels, as discussed in the next chapters. (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 7 Strengthening National Coordination The biggest gains to society cannot be picked up Nelson Mandela’s contribution to the formation of a through uncoordinated individual actions. modern South Africa (chapter 3). National actors also —Adapted from Mancur Olson engage in and frame the pursuit of international goals, as elaborated in chapter 8. This chapter focuses on the reasons that better M any welfare enhancing opportunities can be outcomes emerge when the institutional basis for realized only through coordinated activity, coordination is strong, encouraging the make, ex- guided by institutions and policy.1 Because change, and preserve, which leads to growth and sus- of differences in laws, norms, and systems of govern- tainable development, and discouraging the take, ment, the political and legal domain for coordinating which leads to waste and conflict. The emphasis is activity is frequently the nation. Much private sector ac- on coordination at the national level—to manage a tivity is national in scope and currencies, trade policies, broader portfolio of assets (chapter 2). The national safety regulations, and the like substantially circum- concerns analyzed here are critical for sustainable scribe the markets. Furthermore since many externali- development and highlight the institutional levers ties spill outside municipalities and regions, the nation that can strengthen coordination: is also the level at which interests can be balanced— either directly or by facilitating negotiation among lo- Ⅲ Promoting inclusiveness and the participation of calities. In fact, the national government plays a special poor people (fostering access to assets and voice) role in providing a legal framework and in creating Ⅲ Generating a sound investment climate (attending an enabling environment for partnerships in which the to macroeconomic fundamentals, strengthening private sector, civil society, and all levels of government governance, and providing basic infrastructure) can contribute. Partnerships between government and Ⅲ Managing the environment (dismantling perverse think-and-do-tanks are one example (chapter 9); those subsidies, husbanding forests and fisheries, and between loggers, NGOs, and local and national govern- curbing air pollution) ment, to create mosaics conserving the productive land Ⅲ Using resources effectively (avoiding the natural uses ahead of the agricultural frontier are another (chap- resource curse and ensuring that external aid does ter 5). The smooth evolution of property rights from not undermine government accountability) communal to private (chapter 5) or the enhancement of Ⅲ Averting violent conflict (using natural resources tenure security in urban slums (chapter 6) depends crit- well and confronting extreme poverty, thereby ically on a facilitating national framework of norms, limiting their inciting effects). laws, and organizations. National leadership, including parliamentarians, can be fundamental in directly bring- Reducing poverty and providing a sound invest- ing about change, as with Tunisian President Bour- ment climate strengthens the incentive to work and guiba’s support for the rights of women (chapter 4) and build communities, and fosters the better use and  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  protection of the environment. Strengthening the South Asia, North Africa, and western and central management of natural resources, especially certain Africa, and poor people have limited access to educa- concentrated resources, helps confront widespread tion in almost all countries studied. destitution and reduces the likelihood of conflict. The study also shows that educated adults, espe- The treatment of each of these issues will attempt cially women, tend to have fewer, but healthier and to answer three questions: better-educated children. Beyond the generalized productivity gains associated with schooling, the sur- Ⅲ Does the institutional environment perform the veys suggest that major opportunities are missed in key functions of picking up signals, balancing in- poor countries. The expected returns to a nation of terests, and executing decisions? expanding education, particularly for girls, are now Ⅲ What institutional levers might improve the pro- well documented, and education is a key part of the cess of organizing dispersed interests and forging MDGs. credible commitments? The wider extension of access to assets and ser- Ⅲ Do today’s processes permit the emergence and vices will help realize people’s potential and improve evolution of more inclusive institutions over time? dynamic efficiency. The institutional environment is central in redistribution through growth. Partner- Whether the development process is strong, sus- ships with donors are a promising avenue for financ- tainable, and responsive to new challenges depends ing greater inclusiveness, including spending on largely on the answers to these questions. education and market-based land reform projects where the landless poor receive help in buying land Promoting inclusiveness and establishing farms (see chapter 5). This shifts the As noted in the previous chapters, the growth of as- potential disincentive among taxpayers to higher- sets, the production of assets, and the sustainability income countries, where the effect is easier to absorb. of social transportation depends on the quality of in- Regardless of source, national or international, the stitutions whose evolution depends on voice and in- support for assistance from rich to poor will be more clusivity of all members of society. There are not feasible politically when resources to provide public many simple adjustments that can be implemented services are used more efficiently. In the case of in- in a 3- to 5-year period, but over a long period, sus- ternational development assistance, for example, this tained broadening is possible if it starts now and is means mitigating the negative incentive effects that steadily reinforced at the national level. aid can potentially generate, such as weaker govern- Promoting inclusiveness can help realize the po- ment accountability (as discussed in greater detail tential of a nation more broadly. Poor or excluded below). Levers for addressing such incentive issues people often cannot seize opportunities because of and improving services include involving the private high barriers to participation. Without a stake in the sector and civil society in delivery; promoting decen- social, economic, and political system, they do not tralization (fiscal, administrative, and political); and have the incentive to partake, contribute, and be for- facilitating parental and community influence in ward looking. Promoting wider participation, with schools (voice and other means to raise transpar- greater access to assets and voice, is a pillar of steady ency). Such measures also encourage the emergence growth and sustainable development, and the insti- of better institutions over time (chapter 4). tutional environment is central to how national ac- tors address this pressing issue. Strengthening voice Mechanisms that strengthen voice—such as a free Fostering access to assets and services press, elections, and participation in civil society orga- Much human potential is wasted because of the un- nizations—are keys to including the disenfranchised, derprovision of services. A recent study of primary building mandates, and generating consensus. Those and secondary education based on surveys from 41 excluded, without voice, will not take ownership of a countries shows major gaps in educational attain- process, action, or outcome. This effect also applies at ments associated with gender and family income.2 different scales—local, national, global. Voice can do There is a large female disadvantage in education in much to channel dispersed interests (chapter 3). (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  Box 7.1 Democracy, leadership, and decentralization in Latin America Between 1989 and 1994 the República Bolivariana de Venezuela Leadership played an important role in the decision to de- and Bolivia implemented sweeping institutional reforms that de- centralize power and in the choice of specific institutional ap- centralized political power and enhanced opportunities for partici- proaches. Political leaders exercised substantial discretion in pation. In the República Bolivariana de Venezuela, citizens were seeking to balance interests through decentralization, overcom- given the right to vote for mayors and governors for the first time, ing opposition from vested interests. In Bolivia, for example, reversing a long trend toward centralization. In Bolivia, the Law for powerful unions and departmental civic committees opposed Popular Participation created hundreds of municipalities and the idea of municipal empowerment. granted them 20 percent of government revenues. In addition, political leaders had at their disposal a stock of In each case, institutional change resulted from a combination specific, ready-made proposals that they could offer as solu- of three factors: widely felt social problems, political leadership, tions for addressing general social dissatisfaction with the per- and ready-made proposals crafted by small groups of analysts. In formance of public institutions. The proposals were devised by the República Bolivariana de Venezuela, riots in Caracas forcibly technical analysts in think tanks and presidential commissions demonstrated to political elites the depth of public dissatisfaction that were intimately familiar with national conditions—in many with the country’s governing institutions. President Pérez re- cases through direct contact with grass-roots organizations. sponded by proposing a package of democratic reforms designed Often the proposals were developed many years in advance years earlier by a commission established under his predecessor. and were kept alive in discussions among experts and social The creation of municipalities in Bolivia was an idea developed in leaders who made the proposals available for adoption when a think tank associated with the party of President Sánchez de the political opportunity arose. Lozada. The president became convinced that such an initiative held promise for addressing the concerns of increasingly vocal in- digenous groups and local communities. Source: Grindle (2000). Opportunities for reform often arise from eco- The study shows that information, political account- nomic or political crises that inspire civil society or ability, and an educated and informed public are political elites to demand changes in the status quo critical for picking up dispersed signals of distress. and to search for new solutions to long-standing Ensuring a free press and democracy also represents problems. In Latin America, perceived crises in the a commitment device to acting. Unfortunately, this legitimacy of governing institutions have inspired mechanism has limitations. As noted by Sen, “the substantial reforms that give greater voice and power deprivation has to be dramatic to be ‘newsworthy’ to local communities (box 7.1). and politically exploitable. The Indian political sys- Organizing dispersed interests for improved coor- tem may prevent famines but . . . it seems unable to dination depends largely on voice. In the form of deal effectively with endemic malnutrition.�5 votes, voice increases accountability and responsive- Education is another example of the importance ness, as policymakers in competitive democratic con- of voice. And it is a good example of the challenge texts are concerned about reelection. A good example hinted at by Sen: how can institutions be geared to is famine—perhaps the worst sort of coordination fail- ensure commitment to activities that are longer ure. In observing that India has not suffered a famine term, less visible, but as important? One resounding since independence, Nobel Laureate Amartya Sen message from research on education systems is that notes that “with the present political system in India, the participation of families and communities can it is almost impossible for a famine to take place. The increase inputs and discipline. Decentralization and pressure of newspapers and diverse political parties the promotion of private sector and NGO involve- make it imperative . . . to organize swift relief.�3 ment can help society meet educational goals. In El The proposition that newspapers in a competitive Salvador, decentralization encouraged greater family democracy elicit swifter disaster relief has been and community involvement in schooling, expand- investigated empirically in 16 Indian states, showing ing educational opportunities for the poor in rural that newspaper circulation strongly increases public areas.6 In Colombia, a government voucher program food distribution during shortages.4 The effects are used the private sector to expand secondary school large, robust, and significant, even after controlling enrollment and choice for poor children in a cost- for income, urbanization, and population density. effective way.7 (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 7.2 a broader portfolio of assets to thrive. Whether Brazil: changing the rules of the game planting rice saplings in a field, making furniture on for better public services the side of a road, setting up a new enterprise, or buying securities on a stock exchange, investors like The state of Ceara—in Brazil’s poorest region, the north- east—was legendary for clientelism, patronage, and poor return and dislike risk. Aspects of the broader invest- public administration. In 1987 a newly elected reformist ment climate, such as the rate of inflation, the extent government, led by Tasso Jereissati, took the reins. Only a of corruption, or the cost of shipping, help deter- few years after the new government launched a public mine the incentive to invest.9 Institutions will be im- health program, vaccination coverage for measles and po- portant in addressing these, affecting not only the lio had tripled to 90 percent of the child population, and in- fant deaths had fallen from 102 per 1,000 infants to 65. drive for new investment, but also the productivity How could a long tradition of clientelism and political oppo- and value of the existing stock of assets. sition be overcome so rapidly? How did reputedly mediocre The investment climate encompasses: state agencies deliver and sustain beter performance span- ning several years and changes in administration? The governor and the state administration had to com- Ⅲ the macroeconomic fundamentals (including sound pel reluctant mayors to join the program. Pressure on the money, strong fiscal accounts, and stable exchange mayors came from neighboring municipios that had joined rates), the program, and from an unusual and unending flurry of Ⅲ the operation of government (emphasizing effec- publicity through radio and other means. By creating an in- formed and demanding community, the state had initiated tive corruption control and a sound regulatory a dynamic in which the mayors saw political rewards for framework that promotes private sector competi- supporting the program. In so doing, the government con- tion while addressing negative externalities), and tributed to replacing the old patronage dynamic with a Ⅲ basic infrastructure (underpinning commerce and more service-oriented one, exploiting an opportunity as a strong third party to improve municipal accountability. the development of human capital). Source: Tendler (1997). Set in a broader context that facilitates initiatives like self-regulation through the adoption of the triple bottom line to evaluate a company’s perfor- Studies on secondary schools in Argentina show mance (gauging profits, social impacts, and environ- that institutional features, such as school autonomy mental effects), these components are critical deter- and parental participation, make a substantial differ- minants of transactions costs that shape what is ence in educational outcomes.8 Parental participa- produced and how.10 Improved criteria are especially tion (such as through a school board) empowers important in guiding investment that is long lived families to influence how the school is run—and an- (lasting 20–50 years), or where impacts are irre- chors the school’s commitment to serving their chil- versible. The investment climate will also play an im- dren’s needs. Autonomy and participation are found portant role in the emergence of better institutions to have a positive and significant effect on student over time. learning, as measured by test scores. Interestingly, the effect is at least as strong for children from poor Attending to macroeconomic fundamentals households as it is for the general population. Sound management of macroeconomic variables Voice combined with transparency and access to is necessary but not sufficient to foster sustainable credible information is a powerful lever for improved development. Macroeconomic crises—especially fi- coordination. This makes new technologies that pro- nancial and budget crises—can undo development mote transparency and bring relevant information and undermine sustainability. The macroeconomic closer to all stakeholders all the more exciting. Voice landscape is broad, and only a few elements are also plays a major role in creating a setting that is con- touched on here: inflation, fiscal balances, and the ducive to the emergence of better institutions over exchange rate. time (box 7.2). Inflation—particularly unexpected inflation— weakens the incentive to produce. There is an exten- Creating a sound investment climate sive literature on inflation, with commitment prob- Institutions are crucial to fostering a sound invest- lems as a major feature: can the government or ment climate that favors make over take and enables central bank commit to not printing money? Some (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  institutional remedies involve granting the central exchange rates, the gold standard, currency boards, bank independence and adopting transparent targets and even using the U.S. dollar. and rules in managing the money supply. Other commitment devices include tying the exchange rate Strengthening the operation of government to an appropriate nominal peg.11 The incentive to invest is strongly influenced by The fiscal accounts are also an important part of maintaining property rights and the rule of law, en- the investment climate. Public finance research has forcing contracts, regulating activity, and engaging addressed many important aspects of fiscal manage- in other interventions that support make over take. ment, such as spending countercyclically, building a But regulation, for example, can be captured by con- safety net, redistributing income, providing public centrated interests to the detriment of the many. goods, and the like. One important aspect that illus- Similarly, bribe seeking by government officials typ- trates well the approach taken in this Report is fiscal ically motivates arbitrary harassment, also harming decentralization, which has attracted increased at- broader interests and the incentive to invest. And tention based on the promise of improved account- once someone has accumulated wealth through ability and the more responsive provision of services judicious investment, it may be subject to seizure, and public goods financed closer to the level of the either by private individuals (crime), or by govern- ultimate beneficiaries. ment (expropriation). Can government commit to The potential gains from decentralization, how- protecting assets and avoiding predation? If not, ever, face several pitfalls, especially the problem of much activity will be discouraged, relocate to where “soft budget� constraints.12 If localities feel confi- it is protected, or move to the informal sector. dent that a higher-level government will bail them As noted in earlier chapters, substantial growth in out in the event of a financial crisis, spending may productivity and employment is required over the be excessive, creating uncertainty and gyrations that next 50 years in order to improve the quality of life undermine the incentive to invest. In the extreme, for growing populations in developing countries. regional fiscal profligacy can lead to national fiscal Small- and medium-scale enterprises are often the problems that precipitate inflation and exchange rate most dynamic segments of developing economies, crises. So institutions are important in generating a and are central to the expansion of employment op- “hard budget constraint� that protects the dispersed portunities, especially for poor people.14 Yet these interests of taxpayers in other municipalities, and fa- enterprises face many hurdles in difficult business en- cilitates commitment by the higher levels of govern- vironments. For example, seminal research on entry ment.13 For example, in 1996 Hungary adopted a regulation was done by Hernando de Soto and his law establishing formal bankruptcy proceedings for team by opening a garment shop with one employee municipalities. Such arrangements are central to en- in Lima, Peru. They found that it took 289 days abling the sound fiscal management that contributes working 6 hours a day to legally register the business, to a strong investment climate. and the cost was more than 30 times the monthly The exchange rate is another macroeconomic minimum wage.15 A similar study subsequently looked fundamental, important for international transac- at the obstacles and costs of formally establishing a tions on both the current account (trade in goods business in 85 countries, to see whether the quality and services) and the capital account (borrowing and of public and private goods—ostensibly the justifi- foreign direct investment). Large gyrations in the ex- cation for regulation—is better in countries with change rate can undermine the incentive to invest. more regulations.16 If the quality is better, govern- In addition, poor people perceive adverse effects dis- ment works in the public interest—if it is lower, gov- proportionately, because they have little access to the ernment has a grabbing hand. The authors find that sophisticated financial instruments needed to pro- the quality of goods is not better in high-regulation tect or hedge against exchange rate movements. countries; instead, the additional regulation leads Much depends on confidence. Although a lively de- to corruption and forces activities to locate in the in- bate continues on how best to manage the exchange formal sector, hampering more productive arrange- rate, maintaining strong fiscal and monetary policy ments. The authors conclude that where government is essential. And many governments have sought in- institutions are weak it is better to have a smaller gov- stitutional commitment mechanisms including fixed ernment that intervenes and regulates less.17 This is (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  consistent with one of the main themes of the World Corruption is a major risk entrepreneurs face, be- Development Report 1997, arguing for “a good fit cause government officials seeking bribes can under- between the states’ institutional capabilities and its mine the viability of an entire investment. Paying action.�18 bribes to continue business operations also lowers re- Another important avenue for broad-based invest- turns, reducing the incentive to invest. A series of ment is housing and land. Formally acquiring land World Bank–sponsored surveys of businesses in de- for housing often involves many steps and takes veloping countries show that corruption is a major years.19 The result again is discouraged investment, concern, frequently imposing additional costs on the inefficient location of activity in the informal smaller enterprises disproportionately. Ongoing re- sector, and corruption. Institutional responses in- search continues to illuminate the deleterious effects clude better tenure arrangements, streamlining for- of corruption, and suggests potential remedies that are mal registration procedures, and introducing more mainly institutional, including partnering with civil transparency and accountability into the process. society (box 7.3), providing access to information, Box 7.3 Civil society and governance Civil society has emerged as a major force in efforts to improve pending on the project, the basic idea is a commitment by the governance. High levels of corruption are associated with authorities not to demand or accept, and by firms not to offer, poverty, inequality, reduced domestic and foreign direct invest- any bribe or gift in exchange for an advantage in the project ment, and weak overall economic performance.* Corruption award process. These pacts are currently forging coalitions control that ensures a well-functioning government is substan- among bidders, government agencies, and CSOs in about 50 tially a public good and an intermediate input into the provision competitive situations around the world, and are underpinned of other public goods. This collective action problem is aggra- by transparency and disclosure procedures, independent third- vated by difficult incentives: not only would beneficiaries pre- party oversight, and specific sanctions for noncompliance. fer to free-ride and not contribute to governance if others are TI has also promoted international collective action to curb contributing, but the incentive to defect and engage in corrupt corruption, most significantly the OECD Convention. With the practices is stronger when other actors curb their own illicit support of business leaders, governments, and CSOs, the activities. OECD Convention on Combating Bribery of Foreign Public Of- International aspects of corruption further complicate the ficials in International Business Transactions was signed in issue. When the United States enacted the Foreign Corrupt 1997, went into force in 1999, and, at the end of 2001, had Practices Act of 1977, prohibiting the payment of bribes by been ratified by 34 of 35 signatory states, including several non- U.S. citizens or companies and their subsidiaries, no other na- OECD members.*** tion followed suit with similar legislation. In many countries, The Convention is an attempt to make a coordinated es- foreign bribery was so commonplace that bribes were treated cape from the corruption trap, such that firms are not disad- as tax-deductible business expenses. Government subsidies vantaged by refraining from corrupt practices. It has no formal through export financing and insurance accommodated foreign mechanism of punishing signatories who cheat and do not bribery as an unpleasant but necessary ingredient of doing contribute to controlling corruption by enforcing antibribery business in emerging markets. laws, but publicity and national and international condemna- Civil society organizations (CSOs), such as Transparency In- tion are expected to ensure good faith efforts. It does stipu- ternational (TI), helped change attitudes towards corruption in late an elaborate monitoring system, however, aided in part by the 1990s. This organization’s basic approach is a nonconfronta- CSOs, to further reassure signatories that other participants tional effort to build “national, regional, and global coalitions are complying with the agreed requirements. The experience that embrace the state, civil society, and the private sector to with international efforts to combat corruption is an important fight domestic and international corruption.�† The objective is example of how CSOs can contribute to development. pursued by building public awareness, lobbying governments, and facilitating agreements to reduce corruption. One of TI’s most important and enduring contributions has been to help * Lambsdorff (1999). † Galtung (2000, p. 26). place corruption firmly on the agenda in both developing and ‡ For a discussion of how corruption impacts the poor disproportion- developed countries. This was accomplished mainly by empha- sizing the negative effects of corruption (falling disproportion- ately, see the World Bank (2001c, p. 102). ** TI (2000). Source Book 2000: Confronting Corruption—The Elements ately on the poor), and generating publicity through the annual of a National Integrity System. Berlin. [http://www.transparency.org/ publication of the Corruption Perceptions Index.‡ sourcebook]. The cornerstone of TI’s efforts to foster agreements curb- *** See the OECD Web site for the text: http://www.oecd.org/pdf/ ing corruption is the integrity pact between governments and M00017000/M00017037.pdf. companies bidding for projects.** While the details vary de- Source: Eigen and Eigen-Zucchi (2002). (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  creating anticorruption agencies, ensuring media and Box 7.4 judicial independence, and decentralization.20 National policy can generate excessive urban concentration Providing basic infrastructure Urban concentration is positively associated with eco- Infrastructure, the third main part of a sound invest- nomic growth in the early stages of development. Focus- ment climate, is analytically distinct from other ing development in a few urban centers generates scale investment in that it tends to exhibit a substantial economies in production and consumption and conserves public goods component and economies of scale in scarce infrastructure such as roads, skilled workers, and managerial resources. But excessive concentration is production. It also requires the coordination of net- costly, as these gains are increasingly offset by conges- worked facilities and a sound regulatory regime. tion and diminishing returns to additional investments. Fur- Hence, it is often cheaper to provide infrastructure ther, disproportionately high concentration of economic services—sewerage, piped water, power, telecommu- activity in a few urban areas is likely to have an adverse ef- fect on the interregional distribution of opportunities and nications, ports, and roads—in ways that capture welfare. Empirical research shows that urban concentra- scale economies. Greater population density raises tion rises and then falls with economic development. For the returns to infrastructure investment.21 Infra- example, recent work provides evidence that countries’ urban concentration increases with growth at low levels structure also has profound long-term effects on de- of income, peaks at low-middle levels of income (around velopment, with significant path dependence in en- 1987 PPP income per capita of $2,500), and then declines ergy efficiency, mode of transport, and the spatial at higher levels of income. distribution of activity. Policy and politics also influence urban concentration. Moving from the least to the most democratic regimes re- Proximity matters.22 Not all areas have the same duces urban concentration by 10 percent from its mean. potential. As a result, the spatial distribution of eco- Urban concentration is also lower in countries with higher nomic activity in general, and of urban centers in par- levels of fiscal decentralization (box figure). This relation- ticular, is important to sustainable development (box ship holds even when controlling for the scale of the urban population. Infrastructure investments in waterways and 6.1). The spatial view of urban settlement as a system roads, usually driven by national policies, also help reduce of cities and towns (linking urban and rural activities) primacy, as these investments open the hinterland to ex- deals with the extent to which the urban population ternal trade, domestic and international. is concentrated in the largest city (measured as “pri- Urban concentration declines with fiscal macy�), and how smaller urban units and secondary decentralization cities are dispersed across regions. Excessive primacy can have real economic efficiency costs to countries. Index of urban concentration Poor accessibility of the rural hinterland to urban 2.5 markets—caused by a weakly developed system of 2.0 towns and cities reflecting a thin rural to urban and intercity transport network—also limits options for 1.5 growth and diversification of the rural economy (chapter 5), and weakens the incentive to invest. 1.0 Disproportionate urban concentration often re- sults from an imbalance in national institutions for 0.5 signaling and balancing interests, especially across 0 regions and levels of government. Democratization, 0 0.5 1.0 1.5 2.0 2.5 3.0 fiscal decentralization, and investments in intercity Index of fiscal decentralization roads, waterways, and communication can signifi- cantly reduce such concentration, permitting wider Note: The graph shows that urban concentration (as measured access to urban assets across the country (box 7.4).23 by the size distribution of cities) falls with greater fiscal decen- tralization (as measured by the share of sub-national revenues in By providing access to urban opportunities and as- total revenues). This note draws on findings reported in Ades sets, and linking spatially dispersed activity across the and Glaeser (1995); Henderson (2000); Henderson and Davis (2002); Lall (2002); and Wheaton and Shishido (1982). country, basic infrastructure helps determine the re- Data Sources: World Bank, WDI, and Government Financial turn on other investments. Farmers need roads to Statistics. bring produce to the market. Exporters require tele- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  communications and ports to make and execute trans- incentives for the supply of public goods or provid- actions. Small industrial enterprises demand electric- ing the public goods directly, financed out of tax rev- ity to run equipment. In addition, the reliability of enues. But these remedies are typically difficult to these infrastructure services is a major determinant implement—complicated by disputes about burden of risk. Farmers, exporters, and entrepreneurs are ex- and benefit sharing and the nature, importance, and posed to large losses in the event of infrastructure fail- value of the externalities and public goods. ure. Recent research emphasizes the importance of in- Many environmental issues were not recognized in frastructure for economic performance, especially in the early stages of the activities that precipitated the the long run—and its positive spillovers on private in- problems. Furthermore, powerful constituencies have vestment.24 Institutional remedies for improving the emerged to obstruct change and relegate environmen- delivery of infrastructure services include commercial tal concerns to lower levels of priority. Not only does management, competition, stakeholder involvement, policy often fail to discourage environmentally dam- and public-private partnerships in financing infra- aging activities, but, in the form of perverse subsidies, structure projects.25 These will be central to building it frequently encourages such activities. The gover- the basic infrastructure underpinning a strong invest- nance and institutional challenges of effectively man- ment climate for sustainable development. aging the environment are examined here by elabo- rating on perverse subsidies before turning to three Managing the environment environmental concerns at the national level that il- Many developing countries are responding to envi- lustrate well the importance of nationally supported ronmental concerns sooner than industrial countries institutions for coordination—managing forests, fish- did at similar levels of development. Indeed, as eries, and airsheds. noted in chapter 2, the development approach of “grow first, clean up later,� and the notion of an en- Dismantling perverse subsidies vironmental Kuznet’s curve is increasingly being Perverse subsidies encourage activities that are harm- challenged in recent research.26 For some pollutants ful to the environment and the economy.28 Subsidiz- the relationship holds, but for others, like carbon ing coal, for instance, can hurt the economy by allo- dioxide, emissions continue to grow as per capita in- cating scarce tax revenue to an activity that is not comes rise.27 New technologies and the spreading competitive at international prices, and damage the knowledge about the impacts of environmental environment by encouraging the use of one of the degradation also mean that developing countries can most polluting sources of energy. Subsidies have make better environmental choices, as in the case of been instituted for a wide variety of reasons, includ- switching to unleaded gasoline (see below). Still, the ing many laudable ones such as alleviating poverty environment remains a major challenge for national or addressing market failures. However, they often actors and institutions. Managing forests, water re- fail to achieve these aims or achieve them only at ex- sources, coastal zones, fisheries, and airsheds pose orbitant cost. The removal of perverse subsidies several well-known coordination problems involving would simultaneously benefit the economy and the externalities and public goods. In many cases, when environment, releasing public funds for other pur- fisheries, forests, bodies of water, and air are treated poses. World Development Report 1992 described such as common pool resources, efforts to preserve the en- policy actions as win-win.29 vironment are underprovided; if coordination were Unfortunately, policy reforms are often very diffi- implemented, the gains could be large. cult to implement because in reality the situation is Public finance approaches to these problems are to win-win-lose. Once in place, constituencies develop tax the activities generating negative externalities, sub- to keep subsidies in place past their publicly useful sidize the activity yielding positive externalities, or role. Some politically important stakeholders stand regulate the activity directly (chapter 2). The tendency to lose. Another factor obstructing the removal of to underprovide public goods can be remedied by subsidies—mainly in industrial countries, is the tran- using the political process to determine the type and sitional gains trap, whereby the expected subsidy the level at which public goods are to be provided— stream becomes capitalized in the associated asset.30 and then levying taxes and subsidies to strengthen the The removal of an agricultural subsidy, for example, (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  would reduce land prices and hurt current farmers, Box 7.5 who may perceive no special benefit from the sub- Perverse sugar subsidies in the United States sidy program once the higher price they paid initially for the land is included in the calculation. The Sugar in the United States is an extreme case of perverse subsidies, and the classic example used by Anne Krueger biggest beneficiaries of the subsidy stream were the to help spawn the literature on rent seeking.* Because of farmers who owned land at the time the subsidy was a program of price supports and import quotas, American first introduced. So, current farmers can be expected consumers pay at least twice the world price for sugar. to resist the removal of subsidies strenuously (com- More than $3 is lost to consumers and the economy for each $1 transferred to the sugar growers, and the subsidy pensation in exchange for greater dynamic efficiency to each grower is more than twice the average family in- is very difficult to address). come in the United States. Furthermore, most sugar pro- Agriculture is subsidized worldwide for a variety duction is concentrated in southern Florida, drawing water from the Everglades and returning it with fertilizer, leading of reasons—among them ensuring the supply of to eutrophication. food, reducing the variance of farm prices, maintain- The U.S. Army Corps of Engineers and the South Flor- ing the farming sector, and supporting rural com- ida Water Management District have developed a plan to munities. The bulk of global subsidies to agriculture reverse some of the environmental degradation in the Florida Everglades, at an estimated cost of $8 billion and are dispensed in OECD countries (most developing to be implemented over 20 years.† Generating information countries impose net taxes on their agricultural sec- about the environmental and economic cost will be central tor), where they are estimated to have exceeded $325 to organizing dispersed interests to overcome the vested billion in 2000 (compared with official net resource interests surrounding current U.S. sugar policies. Some in- stitutional progress is being made. The Environmental flows to developing countries of about $37 billion Working Group (EWG), an NGO, is publishing the amount in 2001), and much of this is perverse.31 Agricultural of subsidies received by agricultural producers and others, production spurred by subsidies degrades the envi- which may eventually help catalyze a change in policy.‡ In- ronment by encouraging the following practices: ternational pressure is also growing for liberalizing trade to promote sustainable development, because the U.S. pol- icy, and similar policies in the EU that protect sugarbeet Ⅲ Overly rapid conversion of forested land or wet- farmers, undermines the development prospects of more lands for agricultural use competitive sugar growers in developing countries. Ⅲ Overgrazing on range land * Krueger (1974, 1996). Ⅲ Excessive application of pesticides and fertilizers † See http://www.discovery.com/news/features/everglades/ Ⅲ High water use and aquifer depletion. everglades.html for details. ‡ See EWG’s Web site at www.ewg.org. Source: Myers and Kent (2001), p. 47. Many agricultural subsidies were intended to be temporary—to provide assistance in a farm crisis, for example (see chapter 5). However, their removal has the underpricing of water in the Aral Sea basin (box proven difficult, even in cases in which the initial 2.3 in chapter 2). legislation contained a sunset clause. Agricultural Perverse subsidies also abound in energy (box 7.6). subsidies benefit a concentrated group and hurt dis- Much of the production and distribution of energy is persed interests, including consumers, taxpayers, and controlled by governments. When the fiscal and regu- those who bear the environmental costs, especially latory framework fails to price energy in a way that in- in the OECD countries, where farming is a small ternalizes the full social costs of its use, consumption part of the economy (box 7.5). is excessive, with high costs of pollution. Developing Perverse subsidies to agriculture also seriously un- and transition countries provide subsidies of more dermine development. Those in the North distort than $120 billion to electricity production, and in domestic economies and depress world prices, block- eight large developing countries, total energy subsidies ing trading opportunities in agricultural commodi- exceed $83 billion.33 In the OECD, as noted in chap- ties that generate more than half the export earnings ter 2, total energy subsidies in this group are about for 40 developing countries.32 Perverse agricultural $71 billion, and coal subsidies are about $8 billion a subsidies in the South primarily involve specific in- year.34 The dispersed interests of those who bear the puts like water, fertilizer, and pesticides, but can have environmental and other costs are not captured ade- similarly adverse environmental effects, as shown by quately in the decisionmaking process. While energy (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 7.6 Box 7.7 Perverse energy subsidies in the Islamic Aid and compensation to address obstacles to Republic of Iran reform in the Russian Federation’s coal sector The Iranian economy is heavily distorted, such that re- The Russian Government’s coal sector restructuring pro- forms would yield welfare gains estimated at about 19 per- gram is aimed at reducing subsidies and redirecting re- cent of the GDP. A large part of this stems from petroleum sources from loss-making production (among other reform prices that are only about 10 percent of world prices, with goals). It illustrates how aid combined with increased an implicit subsidy to petroleum products that amounts to transparency and effective institutional arrangements can more than 18 percent of GDP. Such subsidies encourage help overcome obstacles to reform. Total subsidies to the excessive and wasteful energy consumption, with sub- coal sector in Russia were $2.76 billion in 1994 (1.02 per- stantial foreign exchange earnings forgone. cent of the GDP), and production was inefficient. Initial ef- Poor people usually do not benefit from these sub- forts to restructure the coal sector by redirecting govern- sidies and their removal would release vast resources ment support ran into difficulties, mainly due to a lack that could be redirected toward environmental, social, and of transparency in the transfer system, and implementa- other expenditures underpinning sustainable develop- tion problems in the early phases of transition. During the ment. The prospects for reform hinge on the opening of second stage of the restructuring program, supported by the political process, allowing greater voice and participa- the World Bank, new institutional mechanisms, including tion. The first step in addressing fuel subsidies is to dis- checks and balances, were put in place with the explicit play them explicitly in the budget. This would highlight aim of bringing transparency into the allocation and trans- their magnitude in relation to other priority areas, and fa- fer of support. This included funds for the social protec- cilitate a process of gradually lowering the subsidy to allow tion of laid-off workers, physical works at closing mines, fuel prices in Iran to rise to world levels over a 3-year pe- job-creation programs, and safety measures. The World riod once reforms are initiated. Bank also provided a separate technical assistance loan to the Russian government in advance, in order to build ca- Source: World Bank (2001d). pacity and help address implementation problems more effectively. This loan was instrumental to the success of the larger sectoral restructuring program, which led to a fall in total coal subsidies to $0.28 billion in 2000 (0.12 per- use is broad based, the subsidies generally benefit cent of the GDP). After 10 years of lowering subsidy- induced overproduction, the output of coal began to grow high-income groups disproportionately. The impact again in 1998, and the industry’s production is now in- of such subsidies is also felt for a long time because the creasingly market-driven without distortive subsidies but stock of capital with low energy efficiency (transport, not yet subject to externality-correcting taxes. buildings, and production processes) takes many years Source: Kudat, Ozbilgin, and Borisov (1997) and World Bank staff. (and some- times decades) to change. Perverse subsidies also exist in other sectors, such as water management, roads, fisheries, and forestry. And debate continues on how some of these subsi- of forests for maintaining biodiversity and sequester- dies serve such policy goals as reducing poverty and ing carbon needs to be more fully recognized as an in- inequality. Still, the staggering sums suggest that pol- tegral part of sustainable development strategies. For- icy reform could substantially improve environmen- est resources underpin the livelihoods of 90 percent tal conditions, strengthen economic performance, of the 1.2 billion people living in extreme poverty— and release vast resources for more equitable devel- and are important to the agriculture and food security opment. There has been some progress in the last of almost half the people in low-income countries.36 decade and international development assistance can Illegal logging and mismanagement lead to substan- facilitate the process (box 7.7). Indeed, reforming tial losses and forgone government revenues that are policy to stop encouraging environmental degrada- larger than total World Bank lending to these coun- tion is the first step to implementing broader mea- tries.37 In addition, at present, much of the carbon sures to ensure that environmental assets such as emitted by developing countries, between 10 and 30 forests, fish, and airsheds are not squandered. percent of global emissions, is from burning forests.38 When traditional methods of managing forest re- Getting the most from forests—governance, markets, sources are supplanted by mechanisms that are for- and partnerships mal but weak, the resource is squandered, often be- Natural forests are being converted at an unprece- cause those who are in and near the forest perceive dented 12.6 million hectares net a year.35 The value less secure future control. Races for property rights (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  can be very damaging to forests, accelerating the rate World Bank’s Prototype Carbon Fund, discussed in of conversion into agricultural land (chapter 5). And chapter 8). Partnerships with CSOs and private sec- perverse subsidies to industrial forestry—through tor actors can also encourage sustainable forest man- tax breaks, underpricing for commercial loggers, in- agement practices by rewarding responsible behav- frastructure spending, and lack of enforcement of ior, driving rent seekers and corrupt actors out of regulations—are substantial. business, and creating the basis for credible, inde- Institutional levers to correct this include the pendent certification by internationally recognized following: bodies. Many of these features are illustrated in the promising steps taken toward reforming forest man- Ⅲ Improving the policy and legal frameworks to es- agement practices in Cameroon (box 7.8). tablish and maintain the property rights of local communities Getting the most from fisheries—overcoming the Ⅲ Promoting partnerships with local groups and tragedy of the commons and improving information CSOs to integrate forestry, small-scale enterprise, As discussed in chapter 3, fishery resources present and conservation activities into rural development major opportunities and challenges. Managing this and watershed management renewable resource to maximize well-being in a sus- Ⅲ Seeking reforms of governance and timber conces- tainable way is critical for development. About 30 sion policies million people living in developing countries are en- Ⅲ Building international markets that channel funds gaged in fisheries, and “coastal fishing communities to countries providing the global public goods of are frequently among the poorest of the poor.�39 For preserving biodiversity and sequestering carbon about 1 billion people living in poor countries, fish Ⅲ Fostering partnerships with private actors and are also a vital source of protein, and an important CSOs to move to sustainable harvesting and for- component of food security.40 Yet, as noted in chap- est management with credible performance-based ter 1, 70 percent of the world’s fisheries are being certification by independent third parties. overexploited or are at capacity. Subsidies in fishing are between $14 billion and $20 billion a year glob- Each of these levers is important for strong forest ally, of which half are given in OECD countries, management that recognizes the different benefits to leading to substantial overcapacity.41 By some esti- different stakeholders, and performs the functions mates, the maximum sustainable yield worldwide of picking up signals, balancing interests, and exe- could be caught with half the existing fishing fleet.42 cuting decisions. When property rights are well spec- The two main reasons for the difficulty in managing ified, it is more likely that actors will be receptive to fisheries are the open access or common pool nature signals about the state of their asset, and will weigh of the resource, and the lack of information. This tradeoffs between present and deferred uses. When leads to spectacular coordination failures in the form transparency and governance are strong (i.e., the of waste and sometimes even in the collapse of fish public is an active partner of government and bu- stocks. The institutional framework is key to pre- reaucrats are not accepting bribes), government will venting such failures and in underpinning strong have a greater capacity to balance broader societal in- fishery management. terests with the narrower interests of loggers. The rel- The common pool nature of fisheries gives rise to evant government agency is also better able to im- dispersed interests and commitment problems. In plement informed concession policies that mobilize the absence of clear property rights, individual fish- fiscal revenues. Forest management requires nested ermen are unable to manage fish resources effectively institutions that can ensure that property rights, and commit to limiting the catch to a sustainable assigned to local communities for food, fuel, and level. The gains from reduced fishing effort by an in- shelter, do not overlook the regional or provincial dividual fisherman accrues to the whole group, and watershed benefits of forest conservation, nor global cooperative arrangements are complicated by strong interests in a forest’s carbon and biodiversity services. incentives to cheat. The inability to forge credible The latter set of dispersed interests can be channeled commitments undermines cooperation, and the com- with funds from around the globe through newly mon pool is overfished. When fish stocks and fishing created markets (including initiatives such as the incomes fall, the problem is compounded by politi- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 7.8 Cameroon: The path to improved forest governance As Africa’s largest wood exporter, Cameroon generates sub- important was the widespread formal use of independent ob- stantial income from the exploitation of forest resources, and servers. Respected Cameroonians are hired to observe the the rich biodiversity and other environmental services provided concession allocation process. Cameroonian and international by these forests make them important globally. As in some NGOs are contracted to assist in verifying concessionaire com- other forest-rich nations, logging has been poorly regulated. pliance with logging regulations, both through on-the-ground Until the mid-1990s, logging permits were awarded through inspection and through the use of satellite imagery. This helps an opaque administrative process linked to deep-seated pa- to ensure that the bidding system does not encourage over- tronage. Permit holders were not required to adhere to forest exploitation of the forests. management plans. There was little enforcement of basic logging regulation. Furthermore, the five-year license provided Progress so far no incentive for long-term care of forest resources. Without Ⅲ Improved transparency. Global Forest Watch Cameroon enforcement, loggers built roads deep into the forest. This concluded, “The June 2000 round of concession allocations opened previously inaccessible areas to agricultural conversion was far more transparent than those of 1997 . . . the new and facilitated the bushmeat trade, imperiling rare wildlife and allocations appear to be in compliance with government the forest’s long-term ecological integrity. The system of allo- guidelines, which have recently been clarified. This change cating forest permits was prone to corruption and was gener- attests to Cameroon’s commitment to develop a complex ally insecure for long-term investors. Moreover, the forests market-based auction system aimed at increased trans- yielded very low, unsteady revenues for the state, local admin- parency and rents captured from logging.� The annual area istrations, and local communities. fee per hectare increased from $0.14 in 1996 to $6.00 in 2002, and the total value of multiyear contracts awarded A new law to auction concessions since June 2000 exceeds $600 million. After years of policy dialogue with donors, Cameroon’s execu- Ⅲ Introduction of the first legal framework for community tive branch brought to parliament a new forestry law in 1994, forestry in central Africa, including priority access to forest providing for the auction of forest concessions on the basis of land secured to local communities. per-hectare bids by prequalified bidders. This was intended to Ⅲ Clarification and simplification of forest management regu- increase transparency, discourage unnecessary expansion of lations, resuming field inspections and prosecuting illegal the logging road networks, and encourage high value-added in- logging, with fines up to $15 million being levied on a single dustrial activity. The law also required management plans and company. allocated half the revenues to local governments and commu- Ⅲ Gradual exit of short-term speculators and increase in long- nities. But the law’s implementation decrees ran into public term investors with a positive impact on high value-added opposition, both from those who perceived it as benefiting for- industry and local employment. eign companies and from those who preferred the status quo. Ⅲ Enhanced revenue capture in the interest of the nation and In the end the proposed reforms were adopted to comply with rural communities, as annual forest revenues increased structural adjustment loan conditions but without strong initial from less than $3 million in 1995 to more than $30 million local ownership. in 2001 (excluding timber export taxes and duties). Fiscal revenues accruing to local communities increased from neg- Support builds ligible levels in 1998 to $8 million in 2002. After a slow start, the reforms developed local support and Ⅲ Stronger commitment for biodiversity conservation, includ- began to yield results. The first round of concession alloca- ing the creation of new protected areas and ongoing ne- tions, in 1997, was flawed and drew criticism from the World gotiations on conservation concessions with international Bank and from local and international NGOs. Soon afterwards, NGOs. a third World Bank–supported structural adjustment operation provided a new, more flexible framework for debating and sup- Time will tell if these reforms hold up. An encouraging sign porting the implementation of the forestry law. Progress was is that the forest sector’s lessons in negotiating reforms, deal- due in part to the emergence of constituencies for reform, in- ing with vested interests, and increasing transparency are now cluding a strong team of reformers in government, and the being applied to other sectors including the national procure- communities that began to benefit from increased revenue ment system. This may prompt emulation elsewhere in the sharing. Congo Basin, and suggests that the existing process is con- Consultations among stakeholders have intensified. And ducive to institutional strengthening over time. transparency has increased: newspapers publish details de- scribing which companies are authorized to operate in which Sources: Bank staff; Essama-Nssah and Gockowski (2000); Brunner and locations, helping local residents identify illegal operators. Also Ekoko (2000); Collomb and Bikié (2001). cal economy dynamics that frequently enable the Poor information is also a major problem. Not concentrated interests of fishermen to elicit govern- only is it difficult to monitor fishing and landed ment subsidies, harming dispersed interests (other catch, but it is also hard to monitor stocks. The con- taxpayers) and prolonging overfishing (see box 3.4). cern about reliable fishery statistics is illustrated by (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  the ongoing debates about whether the total world A major difficulty in restricting catch for im- catch has peaked (in part depending on the accuracy proved fishery management is the international race of Chinese fishery data), and the extensive efforts for property rights over marine resources. The Third made by governments and international organizations U.N. Conference on the Law of the Sea (passed in to ensure dependable statistics.43 Such data form the 1982, ratified in 1994) brought an important insti- basis for establishing maximum sustainable yields and tutional change in this regard, giving coastal nations other key aspects of fishery management. Aggregate a 200-mile exclusive economic zone.48 This facilitates data on caught fish by weight mask the composition more productive use of fishery resources since most of the fish, making it hard to account for changes in fishing takes place within this zone and responsibil- the quality of fish. There is also little information on ity for good management falls to a smaller group of the relationship between fishing and marine ecosys- stakeholders. Although many types of fish migrate tems, an area that is generating growing interest.44 between the territorial waters of different countries, The challenge of sound fishery management is and swim into international waters outside the ex- magnified in developing countries, where fishing clusive zone, cooperation is easier when it involves often represents employment of last resort, “attract- fewer nations, and fishery issues can often be re- ing increasing numbers of poorly remunerated fish- solved on a bilateral basis. workers who have to resort to damaging fishing While establishing property rights over fish stocks practices to make a meager living.�45 When fishing is rare, efforts to address the common property prob- is on a small scale and done by poor communities in lem have focused on some form of exclusive use the informal sector, authoritative intervention is rights, such as individual transferable quotas, licenses more challenging administratively and politically, to own and operate fishing vessels, and territorial use and the capacity for monitoring and enforcement is rights.49 These mechanisms exhibit varying strengths likely to be weak. A broader development process and weaknesses, depending upon the context, but that creates alternative employment opportunities is they generally aim to improve the private incentives important not only for reducing poverty, but also for faced by fishermen to ensure sustainable manage- improving fisheries management. ment, sometimes by creating tradable fishing rights. Institutions are central to improving coordination These practices have generated some success stories in the management of fish resources. Mechanisms (Iceland and New Zealand), but were unsuccessful in need to be found to restrict fishing to levels that max- other instances, such as the collapse of cod stocks off imize the value of the resource. Such improvements Newfoundland, Canada, in the early 1990s (box 3.4). would significantly raise the total value of landed fish. Although fishery priorities vary—with developed This will depend on stronger protective institutions, countries seeking to reduce capacity and allow the re- including improved monitoring and enforcement of generation of fully or overexploited stocks, and devel- fishing rules. For example, the Thai government has oping countries aiming to develop new resources and made extensive efforts to improve fishery manage- improve technology—all nations are trying to maxi- ment during the 1990s, more than doubling related mize the benefits from this great renewable re- expenditures to reach 1.6 percent of the total gross source.50 Actors must be able to pick up signals about revenues from marine fish landings in 1999.46 Still, the status and trends of fisheries, develop confidence a ban on push-net fishing from 1972 remains un- in each other, and commit to restraint. The FAO is enforced. The same fishery management–spending involved in improving reporting and strengthening ratio in Iceland is 3 percent; in Norway it is about 8 capacity in developing countries.51 A Code of Con- percent, and in Newfoundland, Canada, it is esti- duct for Responsible Fisheries, promoting stronger mated to be 20 percent of the catch value.47 Improv- fishery management in national legislation, underpins ing monitoring, enforcement, and research capacity the efforts of governments, civil society, and private is costly, but a growing trend in fisheries management sector actors.52 There is also a growing recognition is to use part of the increased wealth generated by that balancing the interests of fishermen, present and strengthened management of the fishery to finance future, requires consensus and partnerships among all this capacity. This is a marked change from financ- stakeholders, and that CSOs working through the ing those activities from general taxation. market can play an important role (box 7.9). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 7.9 Partnership for sustainable fisheries As one of the world’s largest buyers of fish from around the ardship MSC illustrates how partnerships ew globe, Unilever shares the growing concern about the sustain- st between private sector actors, e in ability of fishing practices in many areas, especially since a re- NGOs, and government can gener- ar portedly stable gross tonnage of landed fish masks a fall in ate an institutional response that m © recent years of white fish, the type used most in Unilever prod- il improves coordination for sustain- c n ucts. In response, Unilever joined with the WWF to found the cou able development. A large private sec- Marine Stewardship Council (MSC) in 1997. The MSC’s mis- tor firm recognizes the threat to its core sion is “to safeguard the world’s seafood supply by promoting businesses from overfishing, and partners with an NGO to the best environmental choice.�* Through extensive negotia- generate a credible, independent certification agency. By estab- tions with key fishery stakeholders around the world, the MSC lishing environmental standards and promoting an ecolabel, developed a set of strong environmental standards. The MSC the new agency connects the consumer to the management encourages the certification of fisheries to this standard, pro- of fisheries, helping to underpin government efforts to estab- vides an ecolabel for products from certified fisheries (pictured lish, maintain, and enforce sound rules for exploiting fish re- at right), and promotes sourcing from MSC certified fisheries. sources. While some of the certification decisions have been The MSC became fully independent in 1998, and continues to controversial, the partnership process offers some hope that refine the MSC standard, accredit independent third party cer- the management of fish resources will improve over time. The tifiers, and build public awareness. Unilever supports these ini- example of the MSC may also spark ideas for how partner- tiatives and seeks to promote sustainable fishing practices in ships can help in other areas as well. part through its sourcing decisions. At the end of 2001, 5 per- cent of Unilever’s fish inputs were purchased from certified sustainable fisheries (through the certification of the Hoki fish- * See the pages on vision, mission, and values within the MSC Web site. ery in New Zealand). By the end of 2002, Unilever hopes to in- Source: MSC Web site at: http://www.msc.org/, and Frozen Fish Inter- crease this share to 25 percent (as certifications expand to national GmbH presentation “Verpflichtung des Marktes zur bestand- cover larger fisheries), and by 2005, all fish are to be purchased serhaltenden Fischerei—Fortschritte und Herausforderungen,� at the exclusively from sustainability certified fisheries. Berlin Forum—Committed to Sustainability, Berlin, May 20–23, 2002. Curbing air pollution at the national level: the role ready been developed. Second, institutional mecha- of information nisms and policy approaches, developed elsewhere, As noted in chapter 6, air pollution causes serious can help countries and communities avoid mistakes damage at the local level in many developing and and escape some regrets.55 Third, and perhaps most transition country cities.53 Air pollution is mainly an important, in a realm where political economy con- issue about externalities, where the dispersed inter- siderations are paramount, the increasing global flow ests of people negatively affected are not accounted of information about people’s experiences elsewhere for by polluters. Although many actions and initia- strengthens the position of those adversely affected tives are taken at the local level, control strategies, in any given country. such as standards for fuels and new cars, generally Lead in gasoline: latecomers escape some regrets .56 have—and need—support in a broader national Leaded gasoline is a good example. By the end of the framework. National level actors are also engaged in 1990s, most developing countries outside Africa had addressing air pollution problems, such as ground- made substantial progress switching to unleaded level ozone and sulfur, that affect larger regions and gasoline, and about 16 had completely phased out neighboring countries (see chapter 8 for a discussion leaded gasoline.57 Tetra-ethyl lead is an inexpensive of sulfur in Europe). In the absence of broader in- additive used to raise the power of gasoline. It was stitutional support, traditional public finance ap- introduced in the United States in the 1920s, and proaches are often not adopted or implemented. The was not removed until more than half a century later. role of information is critical.54 Initially, there was scant information about the ef- As noted in chapter 2, developing countries do fects of lead, and over time, industry interests be- not need to go through a prolonged period of haz- came entrenched and blocked efforts to remove lead ardous air quality on their path to development. from gasoline. As the United States switched to un- They can benefit as latecomers at many levels. First, leaded gasoline, however, it quickly became clear cost-effective, low-emission technologies have al- that removing lead from gasoline is associated with (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  Figure 7.1 Lead in gasoline and in blood in the United States, 1975–90 Vehicular lead emissions (1,000 tons) Blood levels (�g/dl) 120 18 Vehicular lead emissions 100 15 Median blood lead levels United States 80 12 60 9 40 6 20 3 0 0 1975 1978 1981 1984 1987 1990 Year Source: Lovei (1999). falling lead levels in human blood (figure 7.1), low- rapid switch would be technically feasible and cost- ering hypertension and heart attacks, and avoiding effective, and Vietnam switched virtually overnight harm to children’s intelligence (removing lead from to unleaded gasoline on July 1, 2001. Leaded gaso- paint also played a role). line was banned in September 2001. Vietnam was The phasing out of leaded gasoline in Vietnam is well placed to switch, since competitively priced un- a good example of how an institutional context that leaded gasoline was already available on international encourages actors to pick up signals, balance inter- markets, and the absence of substantial refining in- ests, and execute decisions can help address a difficult terests within Vietnam also meant that there was challenge, even at relatively low levels of economic little opposition from this sector. Still, information development.58 Initial attempts to phase out lead in dissemination and stakeholder involvement to help gasoline began in 1995, but were resisted due to con- overcome misperceptions about leaded gasoline were cerns about the cost of switching and the impact on central to the process and may be replicated to ad- older vehicles. A workshop was held in December dress other similar challenges. 1999, which included representatives from govern- Information helps curb air pollution in China. Infor- ment, the military, the police, industry, academia, mation was important to the process of switching to and the media, as well as experts from neighboring unleaded gasoline, described above, and can catalyze countries (such as Thailand) with fresh experience significant improvements in air quality more gener- and experts from other countries with earlier experi- ally, as evidenced in China. Air pollution in China is ence. The “South-South� consultations were very im- a serious problem. Thousands of city-dwellers die pre- portant to allaying continuing concerns. Following maturely from it every year. Millions more are ill with the meeting, the government of Vietnam started re- pollution-related diseases. Despite problems in Chi- evaluating more carefully the costs of switching, and nese policy for air pollution, there have been suc- launched a program to raise public awareness. Con- cesses, and new tools are being developed to improve sultations with various stakeholders indicated that a the process. These tools are based on experience (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  gained with information disclosure programs about official policy instruments, but clear complements firms’ environmental performance, which have “been which lower the eventual cost of adopting and imple- characterized as the ‘third wave’ of environmental reg- menting policy measures. Information empowers dis- ulation, after command-and-control and market- persed interests through multiple channels: product based approaches.�59 markets—labeling and certification; legislatures—in- Chinese policymakers and institutions clearly rec- creasing voter pressure for reform; capital markets— ognize the value of environmental monitoring and shareholders; courts—citizen suits, tort actions, and how it can shift balances and strengthen accountabil- complaint processes. None of these channels works ity. “One of the strongest elements of the State Envi- without adequate information. ronmental Protection Agency’s environmental strat- Managing natural resources egy has been its work on public participation, public and using aid effectively dissemination of environmental information, and en- vironmental information at all levels.�60 The central The exploitation of natural resources is often a major government explicitly emphasizes openness, and Pre- challenge for national actors and institutions. The mier Zhu Rongji stated in the spring of 2001 that all exploitation of natural resources creates revenue for environmental information should be public.61 government that in a weak institutional context The province of Jiangsu is a good example of how makes it less accountable and responsive to citizen access to information can improve environmental taxpayers, encourages rent seeking and other unpro- outcomes. A pilot industrial performance rating and ductive competition, and can lead to macroeco- disclosure program was established in the city of nomic imbalances, weaker economic performance, Zhenjiang in 2000.62 After a series of workshops, the and in extreme cases, violent conflict. It thus has the provincial government scaled up the initiative to potential to impair the emergence of good institu- cover the entire province of Jiangsu. By mid-2002, tions, which in the longer term are critical for strong, about 2,500 industrial companies, responsible for in sustained growth. International aid can have similar excess of 80 percent of the total industrial pollution effects in impeding the emergence of inclusive and in the province, were included in the program.63 responsive institutions. The challenge lies in avoid- In these programs information is generated in ing these negative potentials from being realized. order to promote the involvement of the wider com- munity in the regulatory process, to help put pres- Natural resources—blessing or curse? sure on polluters with “informal regulation.�64 Where There is much discussion about whether and under formal regulatory capacity is weak, as in many devel- what circumstances natural resources are “treasure or oping countries, information catalyzes the participa- trouble.�67 Will the exploitation of vast new-found tion of civil society, political leaders, and the private oil in central Asia be a source of great wealth and eco- sector. Concerns about environmental liabilities may nomic dynamism, or will it bring misfortune? The also induce private creditors to avoid making loans importance of institutions has emerged as a key aspect to heavy polluters, and consumers to avoid their of the debate. In the context of a viable social contract, products. In addition, consumers and CSOs in in- based on widely agreed formal and informal rules for dustrial countries pay close attention to the activities the allocation of resources and the settlement of griev- of multinationals. Firms that face social, political, ances, institutional arrangements can be sufficient to and economic sanctions are more likely to negotiate restrain opportunistic behavior and the violent ex- with local actors.65 pression of grievance. An extensive World Bank study Environmental outcomes are sensitive to institu- on natural resources in Latin America “found . . . that tional arrangements. This highlights the need to the key to success is to complement natural resource avoid focusing exclusively on emission standards wealth with good institutions, human capital, and and pollution taxes and instead to complement the knowledge.�68 Where institutions are weak, however, latter with the voice of the environment’s beneficia- exploiting natural resources can have negative conse- ries and other mechanisms to improve environmen- quences, with some authors even pointing to a “re- tal choices.66 Such initiatives are not substitutes source curse.�69 One central mechanism is that the for efforts to develop more formal institutions and availability of revenue from such resources may free (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  the government from developing institutions that are in Peru). In a context where government is responsive accountable to the populace at large. For example, the to its citizens, these adverse effects are muted. question of whether oil and minerals hinder the emer- Data on real per capita GDP show that develop- gence of democratic institutions has been tested ing countries with few natural resources grew 2–3 empirically and found to hold for a panel of 113 times faster between 1960 and 1990 than natural countries between 1971 and 1997.70 resource–abundant countries, an observation that ap- The adverse effects are magnified when the natural pears to be robust with regard to alternative methods resource endowment—say, petrochemicals and min- for classifying countries as natural resource–poor or erals—is “point-sourced,� meaning that its produc- natural resource–abundant.73 Indeed, figure 7.2 pre- tion and revenue patterns are concentrated.71 Coffee sents a list of 45 countries whose growth was not sus- and cocoa take on point-source characteristics when tained, in that they achieved their 1999 level of GDP shipped, and are similar to petrochemicals and min- per capita in an earlier decade—many as far back as erals.72 When governments control the revenues from the 1960s.74 Of these 45 countries, all but six are resource extraction, their activities can be financed point-source economies.75 The majority of the coun- to a greater extent without the consent of taxpayers, tries with point-source natural resources also suffered reducing accountability. Where government is less violent conflict in the 1990s (indicated in bold in open, easily appropriable rents weaken governance figure 7.2). and institutions, both of which are critical for long- Angola is a striking example of this tragedy. Civil term economic performance (see box 4.7 on mining strife has persisted in Angola since the mid-1970s, Figure 7.2 Unsustained growth performance is closely associated with point-source natural resources, and conflict* COUNTRIES WITH UNSUSTAINED GROWTH 1999 GDP per capita (1995 US$) was reached: In 1960 During the During the During the or before 1960s 1970s 1980s Senegal Honduras Somalia Mali Philippines Zimbabwe Angola Bolivia Algeria Ecuador Benin Oman Cen. African Rep. Côte d’Ivoire Burundi Ethiopia Botswana Papua New Guinea Chad Jamaica Cameroon Iraq Brazil South Africa SOURCE NATURAL RESOURCES Dem. Rep. of Congo Mauritania El Salvador Jordan Burkina Faso Sudan COUNTRIES WITH POINT Ghana Togo Gabon Kenya Chile Syrian Arab Republic Haiti Guatemala Paraguay Colombia Tunisia Liberia Guyana Rep. of Congo Costa Rica Uganda Madagascar Iran, Islamic Rep. of Tanzania Dominican Rep. Nicaragua Malawi Trinidad and Tobago Egypt, Arab Rep. of Niger Peru Fiji Nigeria Guinea Rwanda Indonesia Sierra Leone Mauritius Venezuela, Rep. B. de Mexico Zambia Morocco * Countries in italic were reported to be in civil conflict in the 1990s; civil conflict is defined as war that has caused more than 1,000 battle deaths, challenged the sovereignty of an internationally recognized state, occurred within its recognized boundary, involved the state as a principal combatant, and subjected the state to an organized military opposition that inflicted significant casualties (Sambanis 2000). Note: Countries for which data are available show unsustained growth if their 1999 per capita GDP (1999 U.S. dollars) was reached in some prior period (Murshed 2002). Countries are categorized as point-source economies if their two most important exports are of fuels, minerals or plantation crops (with some judgments made where there was some ambiguity) (Isham and others 2002). Source: Authors. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Figure 7.3 Angola: real GDP per capita, 1960–96 Real GDP per capita (U.S. dollars, 1985 international prices) 1,400 1,200 Civil war restarts after 1992 Independence general election. 1,000 ᮢ 800 600 ᮢ 400 200 0 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Year Source: Araujo and Costa (1999). and “has been associated with weak (sometimes other claimants are uncertain about future control non-existent) institutions—political instability and over the resource, they are likely to extract the re- violence, little rule of law, and an underpaid and sources as fast as possible—with little regard for mar- corrupt bureaucracy—which have presided over an ket conditions, the impact on local communities, the average annual change in GDP per capita since 1973 environment, or sustainable yields. In the midst of vi- of –4.3 percent� (figure 7.3).76 This, in a country olent conflict, resource management is even less likely with abundant diamond and oil resources. to be optimal, because future control is very uncer- The institutional context is crucial. Where norms tain. If it is inherently difficult to establish property and rules are weak, greater endowments of natural rights and the resource remains in a common pool, resources lead to worse economic performance in management is also likely to be wasteful. the long run, compared with countries that have While formal participants enjoying property rights smaller resource endowments. Existing institutions typically are able to receive market signals, exploiting are eroded, and the emergence of new institutions is natural resources involves substantial externalities hampered.77 There are two key issues in dealing with with wide social and environmental impacts. Imple- natural resources: how are resources managed, and menting standard remedies, such as regulation and how are the resulting proceeds deployed ? taxes, is difficult. Where accountability is weak and Effective natural resource management hinges on government leaders have short time horizons, the re- property rights, whether ownership is held individu- source is again likely to be extracted as fast as possi- ally, as a group, or by government. While there are ex- ble—with little regard for the broader ramifications.78 ceptions involving some multinational corporations In these instances, the institutional context is cen- promoting sustainable practices due to concerns about tral to the ability of national actors to improve coor- reputation, in general when property rights have not dination. Secure property rights rely on an enforce- been established or are poorly defended, time horizons able commitment by all stakeholders, especially the are short, and damaging races for control ensue. As government and other nexuses of power that employ chapter 5 noted for forests, if private sector actors and force, to refrain from looting. Only then will the re- (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  sources be husbanded in the process of extraction. slows urbanization and the demographic transition The same goal will be served by finding ways of for- to slower population growth.80 And by inhibiting malizing the rights of those who have de facto con- competitively honed industrialization and urbaniza- trol, such as landless farmers in the Amazon forest tion, it also impedes the emergence of favorable in- (chapter 5). stitutions that ultimately are more important for Managing natural resources also involves major long-term economic performance and the creation problems in channeling dispersed interests. As a re- of a continuous stream of new opportunities. sult of resource extraction, communities may need to In the presence of large accumulated funds, can move or the local environment may suffer substantial budget constraints still be binding? How can govern- degradation. There is an important role for govern- ment leaders commit to honest use in the interest of ment in managing these impacts—setting a frame- broader sustainable development? Overcoming short- work for responsible private sector participation and term rent seeking is a major challenge, requiring a facilitating the efforts of CSOs. Social cohesion and strong “no bailout commitment� for struggling sec- institutional development will be served by using re- tors of the economy or local governments. Trans- source wealth to promote the participation of poor parency, independent audits, and open dialogues with people and to share the benefits broadly (even though all stakeholders are central to effective processes— it is more difficult to compensate affected communi- especially when creating investment funds (sometimes ties that operate outside the cash economy). The offshore) for future generations as part of a long-term Cameroonian forests project described in box 7.8 and strategy to ward off Dutch disease–type problems.81 the Chad-Cameroon oil pipeline described in box Once the share of total revenues to be placed in 7.11 are promising examples of how broad partner- these investment funds is determined, their effective ships can improve the process. management would benefit from an independent Another set of problems arises with deploying the board whose sole objective is to maximize returns. proceeds from natural resource exploitation. The Civil society can perform a key monitoring function, availability of associated funds in government coffers as in Cameroon (box 7.8) and Chad (box 7.11). And complicates the relationship between government reaching broad agreement on corporate codes of leaders and the citizenry. When government must fi- conduct can be important, as with the attempts by nance its activities through general taxation, it must some international oil companies to improve ac- interact and negotiate with taxpayers, giving citizens countability by publicizing the beneficiaries and greater opportunities for holding their leaders ac- accounts of their royalty payments in Angola. countable. But having funds available from natural That natural resources can undermine institutions resources, especially when production involves a con- and hamper their evolution is not destiny. Malaysia centrated few, enables government leaders and oth- (box 7.10) and Botswana have both leveraged their ers with de facto control to pursue their own agenda. natural resources to foster development.82 In addi- The funds confer power, facilitate patronage, and tion to sound macroeconomic policies, the “good� provide a basis for co-option. And the country is examples of resource-rich countries point to the likely to experience more corruption, rent seeking, importance of shared growth—or inclusive access to and other unproductive activities in the costly com- assets—by which resource rents are invested in edu- petition for the resources, including civil conflict.79 cation and agriculture improvements targeting the Such a country is also likely to end up with poor. Taking advantage of resource rents to promote macroeconomic imbalances stemming from price inclusiveness in access to assets also encourages social volatility and “Dutch disease,� whereby the real ef- cohesion and institutional development, where such fective exchange rate appreciates, increasing the rela- initiatives financed from taxing the affluent might tive price of nontradables and impairing the produc- have been more difficult. Reducing polarization and tion of tradables. Domestic manufacturing loses fostering a capacity for negotiations and coalition competitiveness and is undermined, weakening the building as well as an ability to commit to and sus- opportunities for human capital development and tain reforms, can also be decisive in responding to ex- generating unemployment. The phenomenon also ternal shocks.83 When managed and deployed well, (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 7.10 Malaysia: ethnic diversity, conflict resolution, and development Despite the rich natural resource base (tin, rubber, forestry prod- relationship after independence between the federal govern- ucts, oil and gas, and large tracts of cultivable land), Malaysia’s ment and the states and blunted potential competitive claims successful economic transition was not obvious in the years fol- to natural resource rents. lowing independence from British colonial rule in 1957. A com- munist insurgency raged in rural areas, Singapore separated in Good distributional solutions 1965 in acrimony, and urban centers simmered with ethnic ten- Malaysia’s delicate ethnic balance was also a strong incentive sion.* There was a strong perception in the majority Malay eth- for the Malay political elite to seek out distributional solutions nic group that they had lost out to the Chinese in securing well- that delivered resources to low-income Malays. Even a small paid urban jobs and in benefiting from high-income growth.† In shift of the Malay vote could undermine the parliamentary ma- the federal elections of 1969 political parties promoting non- jority and control of government, and tilt the balance toward Malay interests did well, and Malay frustration ended in riots non-Malay elites. Strengthening the civil service and other po- with hundreds of deaths. Parliament was suspended, and a state litical, economic, and social institutions for cost-effective deliv- of emergency was declared. Malaysia’s future looked bleak. ery of resources to poor Malays was thus a political necessity. The pragmatic political leadership was committed to an Response to crisis outward orientation of Malaysia in trade and investment, re- In response to the crisis, the government announced the New quiring a long-term perspective on development and distribu- Economic Policy (NEP) in 1970, a 20-year program aimed at tional outcomes, and a consensual approach to decisionmak- sharply reducing poverty, improving living standards through the ing. This was acceptable to the electorate because the Malay expansion of education and health services, and eliminating the political leadership enjoyed credibility stemming from success identification of race with economic function. Employment and in steering the country to independence from colonial rule.# output growth were to be sustained by large investments in land A critical but less well-documented factor in Malaysia’s development programs, and smallholder agriculture, an expan- success lay in fostering ethnic harmony through a tacit under- sion of the public sector, and a welcoming foreign investment standing with Chinese citizens on distributional mechanisms. regime. The NEP strategy was to redistribute growth in favor of ethnic By 1990 many objectives were met. The incidence of Malays, while avoiding expropriation of existing assets for poverty fell countrywide from 44 percent in the early 1970s redistribution.†† Furthermore, while natural resource rents to 15 percent in the late 1980s, with Malays registering the helped fund public programs aimed primarily at low-income largest decline. Primary education was made universal, and life Malays, taxation of income, trade, and commerce continued expectancy at birth increased from 58 to 71 years, with all to be light, primarily helping Chinese groups whose presence ethnic groups registering impressive gains. Unemployment in in activities that generated such tax revenues was stronger.‡‡ peninsular Malaysia was reduced to less than 4 percent, and While several important challenges persist, Malaysia’s sub- Malays and Chinese had an equal opportunity of securing a stantial development success lies in fostering inclusiveness by well-paid urban job. Annual GDP growth during the NEP period using natural resource rents to create an ethnically diverse was sustained at an average of 6.5 percent and the structure middle class with a stake in conflict resolution and economic of the economy was modernized with manufactured goods re- progress. Contributing factors include the absence of a feudal placing natural resources for the bulk of exports. class, a delicate ethnic balance, and a credible and forward- What were the factors that largely enabled Malaysia to looking political leadership. avoid the resource curse, defuse complex ethnic conflict, and successfully transform the economy? * With about 55 percent Malays, 35 percent Chinese, and 10 percent Indians, Malaysia is an ethnically diverse country. An egalitarian ethos † In the 1960s, average Malay per capita income was half that of the av- One explanation is the absence in Malaysia of a feudal elite erage Chinese, 80 percent of Malays worked in rural areas in agricul- that might have siphoned away rents and prevented institu- ture, and the incidence of poverty among rural Malays was high (Ross- tional development. As a result of legislation enacted during Larson 1980). ‡ The legislation was the Malay Reservation Enactment of 1913. the British colonial rule, 60 percent of the agricultural land on peninsular Malaysia was owned by smallholders and 30 per- ** Steinberg (1987). Independent Malaysia’s Constitution continues to protect the privileges of the Sultans. The hereditary rulers and gover- cent was under estates owned mainly by non-Malaysians at nors constitute the Conference of Rulers. The hereditary rulers elect the time of independence.‡ While reservations and smallhold- one of their number to serve a term of five years as the Supreme Head ings discouraged Malays from playing a role in urban life, they of State or Yang di-Pertuan Agong. promoted an egalitarian Malay ethos that contributed to the # Parliamentary government resumed in Malaysia after launching the success of NEP. NEP and elections were held every four years starting in 1974. Barisan Furthermore, Malaysia developed a sultanate system that Nasional (National Front), a pragmatic multiethnic alliance constituted cooperated with the federal government in consensus building along the lines of the alliance that launched the independence struggle in 1951, was voted into power throughout the NEP period. and conflict resolution. The British had a protectorate agree- †† Chinese per capita income was 64 percent higher than Malay per ment with the Malay sultans whereby the resident British offi- capita income in 1973; in 1987 it was still higher but by a more moder- cer did not interfere with the sultan in matters of Malay religion ate 44 percent. and customs even though the sultan’s status was that of a ti- ‡‡ World Bank (1992a). tled courtier.** This set a good precedent for the cooperative Source: World Bank staff. (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  the rents from natural resources present an impor- costs. In a weak institutional context, donors need tant opportunity for accelerating more sustainable to ensure that the limited accountability of govern- development. ment officials to their constituents is enhanced (or temporarily offset) by other accountability mecha- Ensuring that aid does not make government nisms, such as detailed project involvement or con- less accountable ditionality, until the development of more perma- Development assistance can help governments lever- nent domestic measures based on transparency and age domestic resources to bring about change. How- participation. Donors and other creditors need to ever, aid and natural resource rents share a key at- be receptive to signals from the dispersed interest of tribute: the potential for weakening the accountability the populations shouldering the debt service, for of government to its citizens. Aid provides a source example, when providing funds to concentrated in- of funds to government leaders, again without the terests—those who control the proceeds from new need for interaction and negotiation with members international obligations. of the polity. This effect is clearly identified in re- Donors have become more selective, directing aid search on odious debts, where government leaders to countries with good policies and institutions (“in- raise international loans—and proceed to misallo- stitutional conditionality�).88 International financial cate funds to poorly conceived projects or, in the ex- institutions increasingly seek to promote the emer- treme, to steal and squander the funds without the gence of better institutions and participatory ap- consent of those in whose name the obligations were proaches, aiming to strengthen coordination by di- contracted.84 That the loans are still extended and rectly addressing the problem of finding channels debt service is still demanded from subsequent gov- for the dispersed interests harmed by ill-conceived ernments, despite a weak relationship between the policies. Debt relief seeks to encourage reform and initial contracting government and its citizens, shows channel funds to development. Recipient govern- that in a weak institutional context, official lending ments as well as donors are permitting greater trans- can have damaging effects. The 1998 World Bank parency in the process, especially in providing access aid effectiveness study and other research offer sober- to information and facilitating the participation of ing observations:85 civil society (box 7.11). Support is growing for the preparation of Poverty Reduction Strategy Papers, Ⅲ In the past, countries with poor policies received as which aim: “To strengthen country ownership of much aid as those making positive reform efforts. poverty reduction strategies; to broaden the repre- Ⅲ Aid has sometimes prolonged bad policies. sentation of civil society—particularly the poor Ⅲ Aid is fungible: Funds for schools may facilitate themselves—in the design of such strategies; to im- the acquisition of planes.86 prove coordination among development partners; and to focus the analytical, advisory, and financial The study found that aid amounting to 1 percent resources of the international community on achiev- of national income contributed 0.5 percent to ing results in reducing poverty.�89 growth in countries with good policies, zero percent in countries with mediocre policies, and –0.3 per- Averting violent conflict cent in countries with bad policies. Other studies Violent conflict breaks down the institutional frame- find evidence that less corrupt countries were not work that enables people to get the most out of life preferentially funded by donors, and that foreign aid and to work together toward sustainable develop- eroded the quality of governance, as measured by in- ment. When death and destruction are widespread, dexes of corruption, the rule of law, and bureaucratic prevention, reconciliation, and reconstruction must quality.87 be the first order of business. Since 1990, more than Similar to the natural resource curse, develop- half of all low-income countries have experienced sig- ment aid gives rise to commitment and dispersed in- nificant conflict, generating substantial transnational terest problems. But as with natural resources, it is spillovers that demand a response from the global possible to find innovative solutions to increase aid community—for example, in helping refugees, pro- effectiveness without incurring undue transaction viding third-party mediation, underpinning commit- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 7.11 Improving the process: the Chad-Cameroon Pipeline Project The development experience in Chad suggests that aid can be Ⅲ The World Bank is also committed to monitoring the use of made more effective in reducing poverty. The exploitation of oil oil revenues, and violations of agreements could trigger ac- is a major new opportunity to accelerate development in one celerated repayment of loans under the program, and affect of the world’s poorest countries. To facilitate extraction, con- the level of development assistance more generally from tribute to poverty reduction, and avoid adverse environmental the World Bank and other donors. and social impacts, the World Bank is supporting the construc- Ⅲ The project will comply with World Bank safeguard policies tion of the Chad-Cameroon oil pipeline through three projects. for environmental assessments, natural habitats, indigenous But the institutional environment in Chad is weak. Extract- peoples, cultural property, resettlement, and forests. ing point-source natural resources and managing the associ- Ⅲ A small area of tropical forest (10–15 square kilometers) will ated revenues often add to the institutional challenges. More- be lost as a result of the pipeline construction, but to com- over, part of the pipeline passes through forest areas and is pensate for this loss two large new national parks (5,000 close to indigenous communities in Cameroon. square kilometers) have been created in Cameroon, to be International aid to facilitate the exploitation of oil in Chad managed for better biodiversity conservation in those areas. aims to address these concerns with the following measures to strengthen the management of oil revenues and to mitigate These measures, which emphasize capacity building, trans- any negative environmental and social impacts: parency, and accountability, combined with continuous civic engagement in spending decisions, should help ensure that Ⅲ By law, direct net incomes (dividends and royalties) are to revenues are targeted to the poor. be deposited in an offshore escrow account to ensure an- The program still faces challenges. Some of the first pay- nually published audits according to international account- ments received by the government of Chad in mid-2000 were ing standards. used for military purposes. Since these funds were “prepay- Ⅲ By law, 10 percent of revenues will be deposited in a Future ments�—and therefore not included in the framework of the Generations Fund. The remainder will be allocated as fol- project—they did not violate the letter of the pipeline agree- lows: 80 percent for priority sectors, including public health ments but they highlight the potential difficulties in improving and social affairs, education, infrastructure, rural develop- aid effectiveness. The military spending was made public, pro- ment, environment, and water resources; 5 percent for de- voking debate, and the government later took steps to make velopment in the producing area (Doba); and 15 percent for spending plans for the remainder of the signing bonus more government operating and investment expenses. transparent. The hope is that the process will enable further Ⅲ An Oil Revenues Control and Monitoring Board, established institutional strengthening over time. to authorize and monitor disbursements, comprises nine members, four of whom are representatives from civil soci- ety and one who is a parliamentarian from the opposition. Source: World Bank staff. ment guarantees, and supporting reconstruction.90 gued in chapter 2, there are interdependencies in the In Africa major conflict has visited virtually every portfolio of assets, and their destruction can spawn country or an immediate neighbor over the last vicious cycles, as evidenced by the greater likelihood decade.91 Although some aspects of security may be of conflict returning to areas that have recently suf- resolved at the local level, the formation of the state fered conflict.92 and challenges to the state are national in scope, mak- Civil conflict and homicides (table 7.1) stem ing conflict a major national issue. from similar underlying factors, including the po- Chapter 3 argued that people and assets—human- tential gains from violence, such as the available made and natural—need to be protected to thrive. loot—and from the breakdown or weakness of de- Threats to people and assets are held in check by in- fensive structures, such as traditional norms, sanc- formal norms, values, and sanctions, but also by such tions, and government-enforced rule of law.93 Table institutions as laws and a state that upholds them. 7.1 shows that: This protection may fail on the small scale, leading to extortion and crime—or on the large scale, result- Ⅲ For the world as a whole, the average number of ing in challenges to the state and civil war. It is worth annual deaths from civil wars was four to five noting that in addition to destroying lives and physi- times greater than from homicides between 1944 cal assets, violent conflict also assaults social capital, and 1996. undermining trust and social networks, and devas- Ⅲ In Sub-Saharan Africa, both civil conflict deaths tates the environment, wasting natural assets. As ar- and homicides are higher than in other regions. (c) The International Bank for Reconstruction and Development / The World Bank ������������� �������� ������������  Table 7.1 tinct from the broader impact of point-source, nat- Civil conflict and reported homicides ural resources elaborated earlier.98 There is also a link (per 100,000 people) between natural resource extraction and civil war based on grievances related to “land expropriation, Total civil conflict Total deaths per year homicides environmental damage, and labor migration.�99 The (1944–96) per year potential access to resource rents makes it easier for All regions private armies or warlords to acquire the arms that (population-weighted average) 10.3 2.3 contribute to the incidence of civil conflict. Similar mechanisms may affect the duration and Sub-Saharan Africa 30.3 7.4 intensity of conflict.100 Resource wealth makes it East Asia and Pacific 10.1 0.6 easier for weak groups to continue rather than be Middle East and North Africa 9.4 1.2 crushed, and their leaders might derive greater re- South Asia 7.4 1.9 source rents during conflict than they would obtain Europe and Central Asia 3.4 1.8 in peace.101 The presence of natural resources may Latin America and the Caribbean 3.2 4.5 also make it harder to reach peace accords, because Sources: Sambanis (2000); Jacobs, Aeron-Thomas, and Astrop leaders have difficulty controlling looting by subordi- (2000); Fajnzylber, Lederman, and Loayza (2000). nates. For 13 cases of civil conflict—in countries such as Angola, Sierra Leone, and the Democratic Repub- lic of Congo—these mechanisms were evident.102 In addition, the government and the international Ⅲ East Asia and Pacific, and the Middle East and community may be unable to convince violent rebel North Africa, by contrast, have relatively high civil groups that they will not benefit from the resources conflict deaths and low homicides. This suggests extracted. To the contrary, rebel groups successfully that state institutions have less success in mediat- pay for weapons and other support with “booty ing tension and containing civil violence than futures,� trading diamond concessions for merce- norms and sanctions do in limiting homicide. nary services, for example, before the conflict has Ⅲ In Latin America, where incomes and inequality are even begun.103 Rebels agree to peace in exchange for somewhat higher, civil conflict deaths are relatively regional autonomy only if the government can cred- low, but homicides are high. This suggests that the ibly commit to its promises, but the presence of nat- state has been well “defended,� partly through pe- ural resources increases the incentive for the govern- riods of military rule, but that factors such as high ment to renege on peace agreements.104 levels of inequality continue to fuel homicides.94 Institutional remedies are emerging, but they need to be strengthened. Civil society organizations “Atypically severe grievances� and religious or eth- continue to focus international attention on the nic divisions in society have been cited as important practices of failing governments and violent rebel causes of civil conflict.95 Studies by Easterly (forth- groups. The publicity is also encouraging private coming) and others suggest that the institutional actors to adopt codes of conduct for resource extrac- framework is central to constraining the divisive ef- tion, illustrated by efforts to label diamonds and fects of such social characteristics, and that other fac- exclude “conflict diamonds� originating in Sierra tors may be more important.96 Empirical findings Leone and Angola from the market. International suggest that two other factors are important to the pressure is growing to curb the looting by neighbors, incidence and duration of civil conflict: the presence now fueling conflict in the Democratic Republic of of natural resources, and the incidence of extreme Congo, and to promote democracy and participation poverty and unsustained growth. to help prevent the accumulation of grievances that boil over into violence. Indeed, as noted above, re- Natural resources and civil war cent research on the “resource curse� suggests another There is evidence that dependence on natural re- less direct mechanism by which resources raise the sources increases the likelihood of conflict, especially probability of violent conflict—by undermining the if they are location specific and lootable, giving op- institutional framework that is critical to economic position groups funding opportunities.97 This is dis- growth and poverty reduction. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Extreme poverty, inclusiveness, and civil war and commitment to reconstruction is thus facilitated The high level of conflict in Africa can be largely ex- not only by moving from the local level to engaging plained by low incomes and poor growth.105 Still, institutions at the national level, but further by invit- the likelihood of conflict rises as poverty is concen- ing global participation. trated in a group—distinct by ethnicity, religion, or Conclusion region. Systematic discrimination, denied access to public services, “extreme poverty and poor social Fostering sustainable development requires the ef- conditions (including refugee camps) also facili- fective management of broad portfolios of assets— tate conflict by providing more readily available human-made, human, natural, intellectual, and so- combatants.�106 cial. This requires improving coordination across The concentration of poverty in a particular group communities and localities at the national level, par- suggests that decisionmakers are not receptive to sig- ticularly for assets with beneficiaries that are widely nals, especially from the fringes, and national poten- dispersed or poor. The institutional environment is tial is being wasted. Government leaders may also be central to this process, enabling government as well politically unable to take better care of aggrieved as private and civil actors to be receptive to signals, groups to prevent violence. Adapting institutions, by balance divergent interests, and execute collective expanding democracy, for example, can help. Rulers decisions. Good processes also promote the emer- of England strengthened the hand of the nobility gence of better institutions over time—discouraging by creating a parliament.107 Indeed, democracy (or the propensity to take, and providing incentives to a dramatic extension of suffrage to new groups) can make, exchange, and preserve. be a commitment to redistribution. In many states Motivated by the take, and engendered largely by democracy has been extended in response to social extreme poverty and unproductive competition for tensions—bringing about successive reductions in in- natural resources, conflict breaks institutions down. equality.108 After the cessation of hostilities, the de- But strengthening voice, broadening participation, centralization of power is sometimes sought to reduce and ensuring more inclusive access to assets can limit tensions by allowing greater autonomy. But it does the tendencies to take. It can contribute to better in- not guarantee peace.109 It can also allow regions to stitutions—with support and reach that are broader drift farther apart, and the center often reasserts its and deeper. Inclusive institutions can liberate creative powers over time. For this reason, offers of greater au- potential by extending protection to groups and as- tonomy often suffer the same commitment problems sets previously unprotected. Discouraging the take as other elements of peace proposals. hinges on rendering institutions more inclusive and Governments need to move to nondiscriminatory receptive to dispersed beneficiaries, which is crucial policies and help those in extreme poverty. Recall for better management of the environment. Encour- Malaysia, which used natural resource rents (rather aging the make hinges on fostering a sound invest- than aggressive asset redistribution) to finance an im- ment climate that provides the conditions for growth provement in poor and excluded people’s access to and for assets to thrive, which is crucial for reducing assets, helping overcome ethnic and socioeconomic poverty and generating innovations to reduce the de- conflict (see box 7.10). International aid agencies mands on the environment. may be able to contribute in a similar way, by pro- Partnerships among government, civil society, and viding aid to leverage a new base of assets for poor private sector actors offer substantial promise for im- and excluded people that enables greater participa- proving development outcomes. These nexuses of tion. In addition, by helping to guarantee agree- interests, especially national governments, also play ments, the international community can help over- key roles in addressing transnational issues, and im- come the serious commitment problems associated proving global coordination, as will be elaborated in with laying down arms. The resolution of conflicts the next chapter. (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 8 Global Problems and Local Concerns S ocial and environmental problems often spill over political roots, and lessons for nonenvironmental national boundaries. Many of the issues described global problems. in earlier chapters—risk management in the frag- ile lands (chapter 4), races for property rights in water Designing institutions to solve global problems and land (chapter 5), urban pollution (chapter 6), and Who would have thought that leaky refrigerators, conflict (chapter 7)—have international ramifications. fire extinguishers, and aerosol spray cans could seri- Dealing with them requires the same kind of insti- ously damage the entire biosphere? The story of how tutional apparatus described in chapter 3: problems stratospheric ozone depletion was diagnosed as a must be detected and diagnosed, and interests must be problem, and how the global community organized balanced within and across borders. However, there is to address it, illustrates how adaptive, learning insti- one big difference: at the global level, there is no cen- tutions can successfully address global issues. tral authority to enforce agreements. Nations have to Refrigerators began using chlorofluorocarbons devise ways to keep themselves on agreed paths. (CFCs) around 1930.1 By 1970 the world used This chapter cannot treat in detail the long, var- about 1 million tons of these substances each year as ied, and growing list of challenges that require inter- coolants, as propellants in aerosol cans, and for man- national cooperation: transboundary river basin ufacturing. In that year, James Lovelock used re- management; international fisheries management; cently invented techniques to detect trace amounts control of infectious diseases; mitigation of acid rain; of CFC in the atmosphere over London. His request and prevention of armed conflict and terrorism, to for a grant to measure CFC concentrations over the name a few. Instead, it draws general lessons from Atlantic was denied: “One reviewer commented that the experience with some environmental problems even if the measurement succeeded, he could not regarding the design and development of institutions imagine a more useless bit of knowledge.� that can handle more difficult transnational issues. Lovelock persisted, though, and showed that Chapter 8 features progress on two transnational CFCs were detectable far from land. Four years environmental problems: protecting the stratospheric later, chemists F. Sherwood Rowland and Mario ozone layer, and mitigating acid rain in Europe. It ap- Molina realized that even tiny concentrations of plies these lessons to two fundamental but unresolved CFCs could, theoretically, erode the stratospheric sustainability issues that are the subjects of contro- ozone layer that shields life from ultraviolet radia- versy and emerging global environmental conven- tion, an insight that won them the 1995 Nobel Prize tions: mitigating and adapting to climate change, and in chemistry. It was known, too, that CFCs had a conserving biodiversity. (A third issue, desertification, long lifetime in the atmosphere and that increased is addressed in the context of chapter 4.) Though usu- exposure to ultraviolet radiation would increase the ally characterized as environmental issues, these prob- risk of skin cancer. Although a definitive cause-and- lems have causes and solutions with deep social and effect relationship had not yet been demonstrated,  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  circumstantial evidence was strong enough in the produced most ozone-depleting substances also pro- early 1980s to support a precautionary approach to duced most substitutes. the threat of ozone depletion. The Vienna Conven- The record of success in tackling this problem pro- tion (1985) committed the nations of the world to vides both hope and inspiration for other global ini- addressing the problem, but imposed no obligations. tiatives. It also shows the key components in global Meanwhile, scientists had been monitoring strato- problem-solving: spheric ozone since the 1920s in a widening global network that extended to Antarctica in 1957. A sci- Ⅲ Pick up signals of the problem and agree on its entist at the British Antarctic Station, noticing de- nature. clining ozone readings in the late 1970s, published Ⅲ Build local capacity and international networks to definitive data by 1984. Shortly thereafter, dramatic support adaptive learning. satellite images of the Antarctic ozone “hole� cap- Ⅲ Reconcile domestic and international interests. tured public attention. This deepening evidence prompted the Montreal Protocol of 1987, an out- These components are explored in detail below, growth of the Vienna Convention, to impose obli- together with an emerging fourth: gations on developed countries to reduce the use of ozone-depleting substances. The Montreal Protocol Ⅲ Harness decentralized mechanisms to establish in- also set up panels to assess the impacts of ozone de- centives for socially responsible actions. pletion and the technology and economics of miti- gating ozone-depleting substances. Pick up signals of the problem By 1990 there was firmer evidence of a causal im- and agree on its nature pact of chlorine and bromine compounds on ozone. Solving problems requires some consensus on the In that year the London Protocol to the Vienna facts and on the costs and impacts of action (or inac- Convention took effect. Under this protocol, devel- tion). The first step is to detect the problem and put oping countries agreed to take on obligations, with it on the public agenda. Initial detection of environ- a grace period, and developed countries underwrote mental problems is often by scientists, sometimes a trust fund to assist them. drawing serendipitously on information gathered for The process remains dynamic. Two more amend- entirely different purposes. Acid rain, for instance, was ments to the Vienna Convention have been adopted. taken seriously in Europe only after a Swedish scien- Technical panels, involving multistakeholder cooper- tist, Svante Odén, in 1967, used data from a long- ation, have helped identify technological approaches standing network of precipitation monitors to link to phasing out ozone-depleting substances. More foreign emissions to acidic rain in Sweden, and to link than $1.3 billion have been committed to help devel- the rain to deteriorating surface water quality.2 But oping countries. The result: a foreseeable reduction in detection is not enough. Especially where dispersed atmospheric concentrations of ozone-depleting sub- interests need to be mobilized, activists (sometimes stances and an eventual recovery of the ozone layer. including scientists) can put a problem on the public The problem of protecting the global ozone layer agenda. NGOs such as TI, Global Witness, and was, for a variety of reasons, easier to tackle than Global Forest Watch gather and publicize evidence on other global problems. The production and use of corruption and human rights abuse, especially in re- ozone-depleting substances is not central to any lation to management of forests and natural resources. economy—unlike greenhouse gases, whose produc- In the future, the new Aarhus Convention on Access tion is deeply embedded in the energy and transport to Information, Public Participation in Decision- sectors. It has been easy to find less harmful substi- making and Access to Justice in Environmental Mat- tutes for most substances, at modest cost. The polit- ters may facilitate detection and discussion of envi- ical economy of reaching agreement has also been fa- ronmental and social problems. vorable. At the national level, the wealthy industrial The next step is achieving some consensus on the nations responsible for most production were also problem’s gravity, threats, and potential solutions. At those at the greatest risk from skin cancer, in part be- the outset, activists use data to demand action, and cause ozone depletion is far more severe at temper- defenders of the status quo attack the data and in- ate than tropical latitudes. And the corporations that terpretation as inaccurate, incomplete, and biased. (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  Progress in resolving the issue requires better infor- work for assessing options and decisionmaking mation and some consensus on the diagnosis. This processes for water resource development, and to is not always easy. To understand such problems as produce guidelines related to all aspects of dam de- acid rain and global warming, we need to under- velopment. Convened by the World Bank and the stand how thousands of factories and millions of IUCN, the commission’s members represented a households behave—and how chemicals mix and broad range of stakeholders. It succeeded in produc- react across the entire atmosphere. These processes ing a consensus report whose core values and strate- can be understood only through sophisticated simu- gic priorities have been widely endorsed. But the in- lation models, and the models can be validated only formal authorizing environment has resulted in weak against rich and accurate observations of physical, engagement of national governments in the result, biological, and social systems. There is scope for according to an independent evaluation.4 And there honest disagreement on interpreting data and mod- is less consensus on the WCD’s specific recommen- els. And naturally, each stakeholder group will pro- dations for implementation. It remains to be seen mote interpretations favorable to its own interests. whether the report will be a one-off outcome—or What is needed is a credible, legitimate forum for fos- will have initiated a sustained process of learning and tering consensus on diagnosis and action.3 engagement. Combining credibility and legitimacy in a policy institution is a fine balancing act, especially for Learning and adapting global issues. Credibility requires scientific and tech- The diagnostic process is most effective when it feeds nical input, insulated as much as possible from po- into an adaptive process of balancing interests, set- litical pressures. Legitimacy, by contrast, is properly ting goals, taking actions, and learning from results. political. Parties to an international agreement need The Convention on Long-Range Transboundary Air to legitimate and accept the scientists’ interpreta- Pollution (CLRTAP) illustrates adaptive learning tions. So do the citizenries who will be asked to com- (box 8.1). This Convention has forged increasingly ply with the agreement. Mediating institutions need ambitious agreements among European nations (in- somehow to broker problem analyses that are politi- cluding economies in transition) on reducing emis- cally palatable and yet have scientific integrity. How sions that cause acid rain, eutrophication, ground- can this be done? level ozone, and other environmental problems. It The IPCC is one example. The IPCC was char- has done so in part by encouraging the collection, tered by the World Meteorological Organization harmonization, and analysis of data on emissions and the United Nations Environment Programme and environmental conditions. This process has (UNEP) to assess the risk of human-induced climate fostered communication among policymakers and change. It has produced three large assessments, car- scientists, facilitated agreement on an operational ried out by an international team of volunteer ex- definition of goals, and promoted a rational, cost- perts, who evaluate and synthesize the vast and some- effective approach to achieving those goals. times contradictory scientific literature through an The CLRTAP and the Montreal Protocol illus- elaborate set of working groups, subgroups, and re- trate the appeal of adaptive learning in forging inter- views. Because the reports are thick, densely techni- national agreements. Countries are averse to taking cal documents, attention focuses on distilling sum- on binding commitments when there is great uncer- maries for policymakers. Each summary is approved, tainty about the costs or impacts, about their ability line by line, by representatives of all IPCC member to induce citizens to comply, and about the compli- governments in a forum where scientists can defend ance of other parties. Adaptive learning allows coun- their conclusions. The process results in political buy- tries—and groups whose behavior is targeted for in to scientific findings. Over the past 10 years, the change—to understand the problem and to acquire IPCC’s work has contributed greatly to promoting confidence in their own ability and others’ to deal consensus on the nature and causes of climate change. with it. The World Commission on Dams (WCD) is an- Two routes are available: other pioneering assessment effort, emphasizing so- cial issues. The commission’s goals were to review Ⅲ One route is through “soft law�: nonbinding state- the effectiveness of large dams, to provide a frame- ments of principles and sometimes targets. By grad- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 8.1 An adaptive, learning institution The CLRTAP has concentrated mostly on mitigating European EMEP worked to ensure consistency in data collection and re- pollution, though it includes North American parties. Its first porting methods among its diverse members. substantive agreement, the Helsinki Protocol (1985), required By 1990 the data were deemed good enough to support parties to reduce sulfur emissions by 30 percent relative to a credible simulation model, RAINS, to assess the costs and those in 1980. Many observers consider this to have been a impacts of alternative emissions reductions scenarios. This modest goal. But it established a track record of cooperation model, developed at IIASA, was used by negotiators in setting that has so far resulted in six subsequent (and increasingly commitment levels for the Second Protocol on Sulfur Reduc- more ambitious) protocols on emissions reductions. tion. It and subsequent analyses have shown that the near- In setting, refining, and implementing reduction targets, term cost of fully meeting environmental goals was unafford- CLRTAP has been aided by the Cooperative Programme for the able, facilitating agreement on achievable interim measures. Monitoring and Evaluation of the Long-Range Transmission The process of data-gathering, model-building, and model application facilitated communication among scientists and of Air Pollutants in Europe (EMEP) and the acid-rain model- policymakers, fostering a virtuous cycle of continuous refine- ing group at the International Institute of Applied Systems ment of data and models. This has helped the Convention Analysis (IIASA). EMEP was established in 1977 with a U.N. tackle additional pollutants and provides a basis for all stake- mandate, but was “adopted� and given permanent funding by holders to monitor nations’ compliance with the protocols, CLRTAP in a 1984 protocol. EMEP has worked to compile data increasing mutual confidence in the Convention. Integrated on emissions and air quality—and to model atmospheric trans- assessment modeling has now been formally incorporated port of pollutants. Several reviews by political scientists have into EMEP, though it remains based at IIASA. pointed to EMEP as catalytic in promoting better understand- ing of the pollution problems and facilitating agreements on Sources: Jäger and others (2001); Jäger, van Eijndhoven, and Clark (2001); more stringent emissions limits. Over more than a decade, Di Primio (1998); Chayes and Chayes (1995). ually establishing norms, soft law lays the foun- dards affect the workers, owners, and customers of dation for negotiation on binding arrangements. low-wage assembly plants. The Montreal Protocol Nonbinding but ambitious targets can also encour- touches multinational and local chemical companies, age experimentation that would be too risky under people who risk developing skin cancer, and poor a binding regime.5 families that dream of affording a small used refriger- Ⅲ The other route is to start with a binding agree- ator. Negotiations on climate change affect coal min- ment that is easy to achieve, but that sets up a ers, oil companies, Sahelian herders, atoll dwellers, car process that allows parties to learn more about costs owners, and wind turbine entrepreneurs. and benefits and to build confidence in their part- To work, these agreements must reconcile interests ners’ behavior and in newly created institutions. within and between countries. This requires mobiliz- ing concern, and demands for action, among the many For both routes, the seemingly mundane require- who would gain some benefit from the agreement, but ment of reporting can be key.6 Reporting—for who are less vocal than the few who perceive their greenhouse gas emissions under the Kyoto Protocol, main interests to be at risk. It thus requires creative for consumption of ozone-depleting substances un- ways of framing problems and solutions to increase the der the Montreal Protocol, or for compliance with perceived congruence of interests, within and across labor standards under the International Labour Or- countries. And it often depends on strengthening the ganisation—deepens domestic understanding of the capabilities of people and organizations in the devel- problem and strengthens external confidence in the oping world to assess options, to negotiate provisions, country’s commitment to compliance. and to finance and undertake actions. Bolivia and Costa Rica have countless pressing Build local capacity for assessment, domestic concerns, yet both have taken the lead in negotiation, and action pursuing biodiversity conservation goals with global How can a hundred or more governments, represent- implications. Their experience illustrates the critical ing billions of people, forge sustainable agreements role of networks of experts and policy entrepreneurs that touch those people’s lives? These agreements in mobilizing domestic concern and finding creative need to balance the diverse interests of groups that cut ways to link civil society, domestic policymakers, across national boundaries. International labor stan- and global interests. In both countries, research or- (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  Box 8.2 “Coupling institutions� and policy entrepreneurs in Costa Rica and Bolivia In Costa Rica and Bolivia strong communities of policy entrepre- vative explorations of environmental finance, including its en- neurs have grown around a unique brand of environmental re- vironmental services payments system, forest-based carbon search organization that serves as a site for collaboration and in- offsets, and biodiversity prospecting agreements. tellectual exchange between national and foreign environmental Three characteristics of these research institutions are experts. The Tropical Science Center, The Tropical Agriculture noteworthy: Research and Higher Learning Center, the Organization for Trop- ical Studies, and the Ecology Institute provide training in tropical Ⅲ They are physically located in the countries of interest. This ecology. They also facilitate networking among environmental is essential for networking and community building among scientists who wish to apply their knowledge toward creating national scientists, and produces a cadre of experts who institutions such as environmental laws, agencies, and pro- often assume leadership roles in environmental agencies tected areas. These same experts have taken the lead in build- and NGOs. ing national support for sustainable environmental management, Ⅲ They ensure extensive participation by both domestic and creating environmental education programs in schools and help- foreign scientists, which encourages international coopera- ing to “mainstream� environmental concerns in their societies. tion in support of national goals. The institutional accomplishments in Bolivia include the Ⅲ They are nonpartisan, which facilitates constructive work- world’s first debt-for-nature swap, the world’s largest forest- ing relationships among experts and reformers associated based climate mitigation project, and some of the world’s with diverse political parties—a key ingredient for ensuring most innovative approaches to park management, involving that policy reform efforts continue across administrations. indigenous peoples, NGOs, and local stakeholders. Among Costa Rica’s successes are its national park system and inno- Source: Steinberg (2001). ganizations linking national and international scien- Reconcile domestic and international interests— tists nurtured a group of policy entrepreneurs who with commitments and cash could blend scientific knowledge and international International agreements are possible because of the financial resources with the domestic political skills overlap between domestic and global interests—and and experience needed to usher through and imple- because participating nations agree that the benefits ment major policy reforms (box 8.2). Attuned to they gain outweigh the costs that they accept. But ideas from abroad but deeply immersed in domestic environmental and social agreements usually involve social movements and policy debates, these countries balancing opposing domestic interests, often sup- have been at the forefront of an impressive record of porting a broad constituency of dispersed interests environmental policy innovations. And they have against one that is more narrowly focused but in- helped to stimulate national dialogues on environ- fluential. And national compliance is not usually mental quality and sustainable development. achieved with the simple stroke of an executive pen, Capacity building of this kind is important for requiring instead the cooperation of a multitude developing countries to assess, negotiate, and imple- of citizens, government officials, corporate leaders, ment international agreements. Lacking experts and and others. Think, for instance, of the issues sur- money, poorer countries are often at a disadvantage rounding worker rights, pollution, and protection of in international negotiations. For instance, lower- privately owned wetlands or forests. A nation that income countries fielded substantially smaller dele- agrees to international commitments on these issues gations at the sixth Conference of Parties of the has to deploy domestic carrots and sticks to coax its Kyoto Protocol, handicapping their ability to partic- citizens into compliance. However, international ipate in the wide range of simultaneous, technically agreements themselves can help provide some of specialized sessions.7 And without a pool of experts, those carrots and sticks. it is difficult for these countries to design policies Sometimes, international agreements can be a and implement projects. For these reasons, it is im- welcome tool to reinforce domestic legislation and portant to develop expert networks and organiza- regulation. The Ramsar Convention on wetlands re- tions within developing countries (and in some cases quires that each participant commits to the conser- shared between developing countries)—and sustain vation and sustainable use of at least one wetland them over the long term. It is not enough to assem- site of “international importance.� (Almost 1 million ble teams for temporary assignments.8 square kilometers of wetland are now listed, in both (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  developed and developing countries.) Listing may erating information about performance, using that restrict the ability of farmers or developers to drain information to set up incentives for socially respon- and convert wetlands—or that of factories and waste- sible behavior. treatment plants to pollute them. But these restric- Intentional oil pollution at sea was curbed through tions may also confer domestic benefits such as clever use of standards and performance reporting. groundwater recharge and flood prevention, while The problem had long been intractable: empty also providing global benefits such as maintenance tankers filled their still-oily tanks with water for bal- of migratory wildlife populations. last, then discharged the polluted mix. The 1958 In- Although the Ramsar Convention has little en- ternational Convention for the Prevention of Pollu- forcement power, preliminary analysis shows that tion of the Sea by Oil prohibited this practice, but protection of listed wetlands has improved. This sug- enforcement was impossible on the wide dark seas. gests that listing with Ramsar helps strengthen do- A new convention, MARPOL (1978) tackled the mestic commitments to wetlands protection. Simi- problem afresh, requiring that all new ships have a larly, accession to human rights conventions can ballast tank separate from the oil tank. Independent strengthen implementation of domestic laws on verification bodies inspect ships and issue certificates human rights.9 The Aarhus Convention, for exam- of compliance. Ships find it hard to get insurance ple, appears to strengthen domestic commitments to without a certificate. The problem was partially freedom of information on environmental issues. solved10—though the lack of port facilities for oil Financial transfers are often designed to align disposal remains a problem.11 local actions with global interests. Many interna- Private firms have great leeway in their choice of tional agreements recognize that developing coun- production processes. These choices have environ- tries may be unable to finance their commitments to mental and social consequences, both local and improve the global environment, even when those global. They affect the quantity of industrial and commitments provide some domestic benefits. The agrochemical pollutants dumped into waterways, GEF has approved about $2.7 billion in grants to the care with which fish and timber are harvested, reduce ozone-depleting substances, mitigate cli- the treatment of low-wage workers, the release of mate change, protect biodiversity, and protect in- greenhouse gases. But these choices are generally not ternational waters. Depending on how the Kyoto easily observable by outsiders. Protocol is implemented, developing countries and Systems for environmental and social perfor- economies in transition could get billions of dollars mance reporting (or certification) might help shift annually in market payments that would promote firms toward more socially responsible production clean energy technologies. processes, for a variety of reasons. Consumers may preferentially patronize more responsible firms—for Standards, certification, and performance instance, those that produce sustainable timber or reporting—inducing socially responsible behavior fish products. Communities may apply pressure to How can society reward people, firms, organizations, firms that flout legal or social norms.12 and governments that behave well? Locally, a com- Perhaps most importantly, financial markets may munity might patronize merchants who are friendly, reward companies with good performance indica- civic-minded, and environmentally responsible— tors. Why? A growing literature suggests that better and do so happily even if their prices are a bit higher environmental and social performance is no burden than those of less respectable competitors. Outside and at best is associated with higher profits.13 One the community, the scope for doing this diminishes, econometric study found that multinational firms as information about reputation thins. Citizens may that apply self-imposed higher-than-U.S. standards appeal to the government to regulate or tax bad be- throughout their global operations had higher mar- havior, and sometimes that works. But it does not ket value than otherwise comparable firms.14 An- always work—it fails at the global scale, or when other study, of 614 U.S. firms, found that a 10 per- government is unresponsive. An emerging set of in- cent reduction in waste generation was associated stitutions and networks tries to fill this gap by gen- with a 0.3 percentage point increase in the return on (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  assets.15 These associations may be causal: good information enables more discerning investment; practices reduce waste of valuable materials, improve greater interest in ethical investment elicits better in- worker morale and productivity, smooth community formation. Similarly, as certification starts to become relations, and reduce liability. And it may be that the norm in an industry, noncertified products find managers who deal well with complex environmen- it harder to compete. tal and social issues are also good at other aspects of Who sets the standards and defines the indica- running a business. Either way, if environmental and tors—and how? This is crucial to the future of such social performance is a proxy for profitability, then “bottom-up� approaches to regulation. Already there financial markets will welcome and act on improved are disputes about how strictly to set standards for information on such performance. certification. Overly lax standards could defeat the Various initiatives are beginning to publicize purpose of certification. But so too could overly information about environmental and social per- strict standards, if they are too expensive for firms to formance—and there is some evidence of firms re- adopt and for outsiders to monitor. In logging regu- sponding. Indonesia’s government-led PROPER pro- lation, overly strict standards can impose high costs gram, which instituted audited self-reporting of firms’ on loggers without yielding environmental bene- pollution levels, has now been emulated in China, fits.18 This tradeoff is of crucial interest in trade ne- India, the Philippines, and Vietnam (see World Bank gotiations, especially where developing countries fear 2000d and Wang and others forthcoming). Non- that onerous standards would freeze them out of ex- governmental evaluation and certification systems are port markets. It is worth considering whether global developing quickly. The International Organization environmental assessment institutions could serve a for Standardization has formalized certification for role in evaluating potential standards. environmental management processes—systems that Standards and indicators are also being applied to give firms the kind of internal feedback mechanisms governments. TI assesses corruption in national gov- that figure prominently throughout this Report. ernments, with ratings that catalyze domestic politi- Some NGOs have developed certification systems for cal pressure for reform and affect private sector in- timber, labor standards in shoe and apparel assembly, vestment decisions. It has been credited with helping organic food production, and other products and to spur international efforts to reduce corruption (see processes.16 For instance, the NGO-initiated Forest chapter 7, box 7.3).19 The International Monetary Stewardship Council has set up criteria for sustain- Fund (IMF) has recently been promoting standards able forest management and now accredits private for reporting basic economic data, such as GDP, certifiers. By 2001, 25 million hectares of forest inflation, employment, and balance of payments. It (mostly plantation) were certified. Several private in- has also prepared Codes of Good Practice on Fiscal, vestment firms have developed “triple bottom line� Monetary, and Financial Transparency, with the ex- rating systems to assess firms’ social, environmental, plicit aim of promoting good governance. Countries and financial performances. And the Global Report- naturally have different capacities to comply with the ing Initiative, a UNEP-sponsored organization, is try- standards, and the IMF gives them assistance to up- ing to develop auditable standards for environmental grade their capabilities. Ultimately, though, markets and social reporting, analogous to those for financial and the global community may look at progress to- reporting. ward compliance as one indicator of a country’s com- There has been rapid growth in mutual funds and mitment to good governance. other investment vehicles that screen investments on Conserving biodiversity: Maintaining current social and environmental performance. In 1984, $40 services and future options billion in professionally managed assets were socially screened; in 2001, $2 trillion, of $19 trillion in pro- In a remote corner of Ethiopia a farmer clears wood- fessionally managed assets.17 The growing demand land for planting. In the process he eliminates one for socially responsible investment and the growing of the few remaining stands of the wild coffee from supply of environmental and social performance which all commercial coffee is descended—and indicators can interact in a virtuous circle. Better which contains genes that protect against leaf rust, a (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  peril to worldwide coffee production. In the Atlantic thought to be threatened over the next 30 years, Forest of Brazil a prosperous cocoa grower chops and 11 percent have already been lost.22 down the forest trees that shaded his now-diseased Ⅲ Three-quarters of all fish stocks are being ex- cacao plants, but which also provided habitat for the ploited at or above their sustainable limits. Total golden-headed lion tamarin, an endangered species harvests from capture fisheries have leveled off or that could be the prime attraction in a future eco- declined. Some fisheries, such as the Northwest tourism industry. In the lowlands of Sumatra large Atlantic cod, have completely collapsed.23 In oth- companies convert forests of rich biodiversity to oil ers, depletion of prized predatory fish have led to palm plantations. shifts in ecosystem structure. Almost 15 million In all these cases, actors pursuing private profit square kilometers of ocean bottom have been not only threaten biodiversity of global interest— scraped by ocean trawlers, possibly causing long- they also damage resources valuable to their neigh- lasting damage to bottom-dwelling species. bors and country. The damage may be immediate and palpable, but sometimes it is hard to measure in What drives ecosystem degradation? financial terms and its full impact may be deferred, People deliberately degrade ecosystems for profit. To since doomed ecosystems can take decades to un- reach any kind of social consensus on policies to re- ravel. That makes it hard to pick up signals of biodi- duce ecosystem degradation, it is essential to under- versity damage, difficult to balance diffuse nonmon- stand the actors involved, and the incentives that drive etary interests against focused profit-driven ones, them. Forest loss, for instance, results largely from and challenging to implement policies that shift in- conversion to agriculture by small, medium, and large centives from degradation toward sustainable use. farmers, though logging often plays a crucial catalytic The complexity of the problem, along with the pos- role in providing access and financing for conversion. sibility of irreversible losses, motivates the attention Until recently, impoverished shifting cultivators were to biodiversity in this chapter. thought to cause much tropical deforestation. While The message here is that maintaining biodiversity there are localized, poignant examples of poverty- and ecosystem functions is not an agenda solely of driven deforestation of this type—for instance, in wealthy countries, as some hold. To the contrary: Madagascar (box 8.3)—shifting cultivation appears biodiversity has a local constituency that values it for to account for only a small proportion of the degra- economic and noneconomic reasons. But where bio- dation of closed-canopy tropical forests (figure 8.1).24 diversity’s services do not yield revenues, it can be Other small and medium farmers account for much difficult for those constituencies to protect their en- of African deforestation, and a small proportion but vironmental assets against liquidation. Poor societies significant quantity of deforestation in closed forests may be unable, by themselves, to finance the option elsewhere. This is a diverse group, including some values of ecosystem conservation. The challenge subsistence farmers but many commercially oriented then is to find ways to ally domestic and global in- and prosperous operators. And large-scale agriculture, terests that support conservation and sustainable use. including ranches and plantations, accounts for most deforestation in Latin America and Asia. Poverty, The scale of the problem therefore, is not the immediate driver of most tropi- Ecosystems are being disrupted on a large scale: cal deforestation, but tropical deforestation can exac- erbate the poverty of communities dependent on the Ⅲ A global satellite survey estimated a pantropical forest for their livelihood. gross deforestation rate of 0.52 percent annually Returns to forest conversion by smallholders are over 1990–2000, or 9.2 million hectares a year, an variable but often modest. Often the returns are lower area the size of Portugal.20 than the option value of the forests (see section titled Ⅲ Coral reefs are being lost to bleaching,21 pollution, “Act now to reduce today’s emissions�) for carbon se- and destructive fishing. A worldwide bleaching questration alone. Farmers’ conversion of Ecuadorian event in 1998, associated with El Niño, harmed forest has been estimated to yield a present value of 16 percent of the world’s coral reefs, with possibly $376 to $1,721 a hectare,25 depending on the prox- half damaged irreversibly. Another 32 percent are imity to roads and access to credit.26 In Sumatra, con- (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  Box 8.3 Poverty and biodiversity in Madagascar Madagascar’s biodiversity is among the richest and most un- a ton per hectare after a year or two. Subsequently, the land is usual in the world, an asset difficult to value in monetary terms used for even lower productivity uses, such as cattle, or it is but with great potential to support ecotourism and perhaps bio- abandoned. In drier parts of the country, grazing and fuelwood prospecting industries. Of its 12,000 plant species, 85 percent extraction spur forest degradation. So, while 115,000 square are found only on Madagascar. Its 32 endemic lemur species kilometers of forest have been lost since 1960, the area under are an attraction for ecotourists. Alkaloids extracted from its cultivation for staple crops has expanded by only 15,000 square rosy periwinkle plant form the basis for some of the most ef- kilometers. fective cancer treatment drugs, achieving a 90 percent remis- Forest destruction has not only failed to yield new produc- sion rate against childhood leukemia. Yet over the past 40 tive land; it has degraded the productivity of existing farmlands years, Madagascar has liquidated about half its forests, which and infrastructure. Denuded hillsides are easily eroded: 130,000 contain the overwhelming majority of its biodiversity assets, hectares of irrigated land have sustained damage or are threat- without realizing offsetting gains in other assets. The country ened by sediment. Sediment also clogs hydropower facilities has fallen deeper into poverty, with its GDP per capita falling and threatens freshwater and marine ecosystems. from $383 (in 1995 dollars) in 1960 to $246 today. In 1997, 16 Madagascar hopes to alleviate poverty and reduce pressure percent of children died before age five. on its biodiversity by boosting agricultural productivity. Improv- What happened? Agricultural productivity stagnated while ing roads in agriculturally productive areas may increase farmer population tripled. Madagascar’s people depend heavily on rice revenues, reduce fertilizer prices, promote intensification, and and a few other staple crops. In 1960 average rice productivity absorb labor—reducing incentives for farmers to migrate to the was 1.8 tons a hectare—about the same as Indonesia, and forest frontier. In addition, transferring property and manage- much more than the 1 ton a hectare average in Mali. By 2000 ment rights of natural resources to local communities is gen- productivity had doubled in Mali and more than doubled in In- erating incentives for more sustainable use and conservation donesia, but it was almost unchanged in Madagascar. Static of these resources. The country also aims to scale up promis- productivity—despite a substantial increase in irrigated rice ing sustainable agricultural technologies, such as conservation area—reflects in part the implosion of the nation’s road net- tillage, that better protect natural resources and that have the work, which fell from 55,000 kilometers in 1960 to 33,000 in capacity to improve profitability. Expansion of the tiny indus- 2000. It reflects also a low and declining rate of fertilizer use: trial sector may also relieve pressure on the land. only 4 kilograms per hectare, against a Sub-Saharan average In the medium to long run, Madagascar’s unique natural as- of 12 and a developing country average of 96.* Meanwhile, sets may provide the basis for a lucrative tourist industry based population grew from 5.4 million to 15.5 million. on ecotourism and resorts. The country may also be able The combination of an expanding population and stagnant to benefit from global markets for biodiversity and carbon productivity generated pressures for agricultural expansion sequestration services—if these markets develop on a large through forest conversion. Small farmers expanded slash-and- scale. burn cultivation of rice into forest lands officially belonging to the state. The practice is attractive to farmers because of its * WRI 2000. low labor and input requirements and relatively attractive yields Source: World Bank staff. Forest area, cultivated area, and yields from in the first two years. But yields rapidly decline to less than half FAOSTAT database; child mortality from Gwatkin (2000). versions of forest to cassava, upland rice, or rubber ing the $876 that might be realized from sale of tim- agroforest yield negligible returns to land (that is, ber.30 But large-scale conversion sometimes yields after the cost of labor is deducted).27 In Cameroon only modest private and social returns. In the Brazil- long and short fallow cultivation of food crops yield ian Amazon, almost 90 percent of cleared farm prop- present values of, respectively, $288 and $644 per erty is in extensive pasture or abandoned. Although hectare. Intensive cultivation of cocoa offers returns some ranching may be sustainable, average stocking of $785 to $1,236 per hectare, depending on assump- rates are very low: 40 percent of active pasture has tions about cocoa prices; interplanting with fruit less than 0.5 cattle per hectare. And more than half boosts returns further.28 In the Atlantic Forest of the converted land is in the 1 percent of properties Bahia, a highly fragmented long-occupied ecosystem, larger than 2,000 hectares.31 mean land values are only about $275 a hectare—and Similar variation in actors and profits is found remaining forested land is worth (per hectare) 30 per- in marine ecosystem degradation. Poor fishers in cent of equivalent land under agriculture.29 Southeast Asia practice cyanide fishing to gain a Large-scale conversion also yields varying returns. mere $50 a month, threatening reef ecosystems in The returns to large-scale monoculture oil palm in the process.32 But highly capitalized and industrial Sumatra are estimated at $617 a hectare, not includ- vessels, often subsidized, deplete large fisheries. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Figure 8.1 local damage affects lives and livelihoods directly and Current land use in closed canopy forest immediately: deforested in 1990–2000 Ⅲ Run-downs of renewable stocks of fish, timber, or Thousand square kilometers wildlife 450 Ⅲ Decreased flood buffering and nutrient filtering Large-scale agriculture 400 Smallholders due to the loss of wetlands 350 Shifting agriculture Ⅲ Increased flooding and sedimentation in small, 300 steep watersheds due to upland land-use change Ⅲ Loss of water yield from cloud forests 250 Ⅲ Degraded drinking water quality 200 Ⅲ Health and other impacts of air pollution from 150 forest and land fires. 100 These damages can be large. The Indonesian for- 50 est fires of 1997–98 caused an estimated $7.9 billion 0 in domestic damages.36 Africa Asia Latin America Other keenly felt local ecosystem values are dif- Source: Authors’ calculations based on FAO (2000) Table 46-3. ficult to assign a dollar value. Their constituencies may therefore find it hard to counterbalance the more focused interests that derive benefits from ecosystem Sometimes ecosystem degradation is an unin- degradation. For instance, natural habitats may be lo- tended consequence of other activities. Irrigation cally valued for recreational, spiritual, and aesthetic and flood control, for instance, have altered many reasons. In a world where incomes are rising and ecosystems. And there are growing threats to coastal transport costs are dropping, rare ecosystems may ecosystems as, worldwide, coastal cities grow, stimu- have an option value as the basis for a future eco- lated by booming transocean trade. Already, 20 per- tourism industry. And more speculatively, the genetic, cent of the world’s people live within 25 kilometers biophysical, and ecological information embodied in of the coast, 39 percent within 100 kilometers.33 As biodiversity may be valuable to future agricultural, urban populations grow along the coasts and major pharmaceutical, chemical, materials, and information rivers, waste streams grow too. The combination of industries.37 For instance, gene bank collections cur- human waste, animal waste, fertilizer runoff, and ni- rently hold 15 percent or less of the genetic diver- trous oxide emissions generates massive flows of ni- sity of wild relatives of important crop species, includ- trogen into coastal waters. Nitrogen contributes to ing maize, rice, sorghum, millets, and peas.38 Loss of eutrophication, a major problem in coastal waters, some of the remaining 85 percent might constrain de- and to the related phenomenon of hypoxia: oxygen- velopment of improved varieties of these crops. starved “dead zones.�34 It may also be associated with algal blooms, some of which are harmful to Biodiversity as a global public good people. Concentrated human populations also load The purely global interest in biodiversity focuses on coastal waters with sediments, pathogens, and toxic two aspects: diversity itself and the maintenance of chemicals. And coastal population growth leads to global processes. The term biodiversity is often used destruction of mangroves and other habitats that loosely to refer to biological resources. But those who nurture biological resources, including the more see biodiversity as a truly global public good see a than 90 percent of the world’s marine fish harvest problem akin to Noah’s: making sure that a represen- that comes from coastal waters.35 tative selection of the diverse range of genes, organ- isms, and ecosystems survives the current onslaught Who has an interest in maintaining ecosystems? of habitat loss, invasions of alien species, overex- While some people gain from ecosystem damage, ploitation, pollution, and climate change. The Noah’s others suffer, both locally and globally. Some of the Ark strategy reflects people’s desires, grounded in (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  ethics and aesthetics as well as economics, to ensure level—or the political level that governs access to that future generations can benefit from biodiversity. land and water. Second, actions in one part of an This is not just a concern of the wealthiest nations. ecosystem can affect a distant part, as when water A 1992 survey found that world species loss was con- pollution harms a distant reef. Third, to reduce po- sidered a “very serious� problem by a larger propor- tential conflict, efficiency is necessary—through in- tion of people in Brazil, Chile, Mexico, and Poland centives that keep agriculture on land with high eco- than in Germany, Norway, Switzerland, or the United nomic value and low ecological value. Kingdom.39 Ecosystem management institutions will take Maintaining global biodiversity requires global co- quite different forms, depending on the biodiversity operation. Think about Noah’s problem: how do we involved and the prevailing systems of tenure and maintain a representative set of the world’s biodiver- governance. Consider a stylized typology of situa- sity? Conservationists have attempted to identify sets tions (actual situations will often have aspects of of ecosystems, which taken together contain much of more than one type): the world’s biological variety. One such exercise iden- tified a priority set of 233 terrestrial, freshwater, and Ⅲ Aquatic ecosystems, marine and freshwater, are far marine ecoregions based on distinctiveness of species ranging, involve many types of actors, and often and ecological processes.40 More than half cross na- spill over national boundaries. tional boundaries, and so would require some kind Ⅲ Frontier forests are sparsely settled sites of conflict of coordination for conservation and sustainable use. and exploitation as both corporate and popular in- And as the Convention on International Trade in En- terests rush to seize rents and claim property. Bio- dangered Species demonstrates, international coop- diversity conservation here is an outgrowth of the eration in trade can help to reshape local incentives more fundamental need to establish governance driving ecosystem degradation. and rationalize land use. These important issues In addition to considerations of pure diversity, are discussed at length in chapter 5 and so are not biodiversity is of global interest because the loss of treated here. key species or ecosystems could have transborder or Ⅲ Commons in transition are areas, often with fairly global impacts. This is particularly the case for ma- high population densities, where management of rine ecosystems, where the loss of one species can fray forests, rangelands, or fisheries has broken down, the food web half an ocean away, and for migratory caused often by government appropriation and bird species. Large-scale changes in land cover can mismanagement of commons, in some cases exac- contribute to regional climate change. There is evi- erbated by population growth. Sustainable use of dence that a loss of vegetation in West Africa and biodiversity depends on resolving disputes among in the Eastern Amazon can start a self-reinforcing communities, and clarifying community and gov- cycle of reduced rainfall and further vegetation die- ernment rights and responsibilities. offs.41 Deforestation is a major contributor to global Ⅲ Fragmented habitats with less-disputed tenure pose climate change. And there is reason to apply the difficult policy questions. They tend to be mosaics precautionary principle: the global consequences of of agriculture and natural habitat, where both the massive biodiversity loss are unknown. private opportunity cost and social benefits of sus- tainable use are high. They include some of the Landscape approaches to biodiversity conservation: “hotspot� areas where the risk of losing an entire Ecosystems meet social systems ecosystem is highest. Balancing interests in biodiversity for the public good is going to require a new breed of ecosystem To give some flavor of how these stylized types management institutions. For the most part, prob- differ, consider the global map of population density lems of biodiversity loss cannot be solved at the in forests (see figure 7 in the roadmap). The great, farmer’s plot or fisherman’s territory. Solutions need relatively unbroken, sparsely populated forests of to consider entire ecosystems and social systems for Amazonia, the Congo Basin, and Siberia exemplify several reasons. First, the incentives driving biodi- frontier forests. The densely populated strands of versity loss must often be addressed at the market forest in India and Nepal include commons areas (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  under transition from government to community Many coastal and marine ecosystems cross na- administration. And the populated forests of Central tional boundaries and demand coordinated transna- America, coastal Brazil, and Madagascar are exam- tional action, particularly for enclosed seas and in- ples of biodiversity-rich fragmented habitats. ternational lakes. The GEF, operating under various Described here are some of the institutional chal- mandates to support 45 international waters projects lenges in addressing the maintenance of these eco- by 2000, has pioneered transboundary diagnostic systems. The point emphasized is that to a large ex- analysis to identify problems and balance interests tent these are challenges for local management. The across stakeholders. The science-based analysis pro- global interest is in supporting these local institu- vides a way of objectively assessing the nature of the tions in maintaining assets of global significance and problem and engaging stakeholders. It then serves as in coordinating action where management issues the basis for agreeing on a Strategic Action Plan. A cross national boundaries. GEF study found that the analysis and planning, when completed, substantially improved priority set- Aquatic ecosystem management ting and consensus forging.43 The need for an ecosystem-wide approach to fish- Integrated coastal management is an approach eries has long been obvious, underlined by the recent that systematically engages stakeholders in the di- disastrous crash of some fisheries. The advent of the agnosis and solution of coastal problems. A recent 200-mile exclusive economic zone places most (but count found 621 national and subnational examples not all) fish stocks under predominantly national of integrated coastal management worldwide, with control—and that puts nations in a position to reg- 284 in 99 developing and transition economies.44 ulate these resources for sustainability. (Chapter 7 But many of these efforts exist only on paper. Ex- discusses some of the factors that determine nations’ cluding the 110 integrated coastal management ef- success in doing so.) forts in the United States (where the track record is But some fisheries require international manage- generally longer), only 45 percent are in implemen- ment. The Convention on the Conservation of Ant- tation, and data are lacking on their effectiveness. arctic Marine Living Resources (CCAMLR) repre- While integrated coastal management exemplifies sents an international effort at sustainable ecosystem the institutional approach to collective action prob- management—in this case, the 35 million square kilo- lems championed by this Report, it has not yet fully meters of the circumpolar Southern Ocean. The Con- demonstrated its potential. vention’s goal is to manage this area with attention not Sixty percent of the earth’s freshwater resources are just to economically exploitable species, such as krill, found in international river basins, within the bor- but to the ecosystem as a whole, encompassing other ders of more than one state.45 Forty percent of the species of concern, such as penguins and seals. world’s people live in those shared basins, all with Similar to the CLRTAP, CCAMLR aims to be expectations of using the rivers’ resources. Histori- an adaptive, learning system. Two working groups, cally, competing demands for shared waters have led under the supervision of a scientific committee, mon- to tensions and conflict. As populations grow and itor ecosystem and fishery data. The data help to cal- economies develop, more pressure will be brought ibrate ecosystem models and guide decisions by to bear on these shared resources. To promote peace, CCAMLR on conservation measures, operationaliz- to sustain river basin ecosystems, and to meet the ing the precautionary principle to ensure that fish development needs of all those who depend upon stocks do not crash. The Convention faces particular shared water resources, it will become imperative that challenges in deterring illegal, unreported, and un- countries cooperatively sustain, manage, and develop regulated capture of Patagonian toothfish (Chilean international river basins. The Nile Basin Initiative sea bass), a valuable but very slow-reproducing spe- responds to this challenge (box 8.4). cies. But innovations in monitoring and reporting— including requirements for vessel monitoring systems Commons in transition that permit satellite tracking and implementation In South Asia, much of Africa, and parts of Southeast of a catch documentation scheme for landings and Asia there are regions where people have used forests transshipments of fish—are changing incentives and and woodlands for generations. Historically, some of improving information for management.42 these common property resources were well managed (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  Box 8.4 The Nile Basin Initiative An extraordinary example of cooperation in the management of united in common pursuit of the sustainable development international river basins is evolving in the Nile River Basin. The and management of Nile waters. Its Strategic Action Program Nile, at almost 7,000 kilometers, is the world’s longest river. The is guided by a shared vision “to achieve sustainable socio- basin covers 3 million square kilometers and is shared by 10 economic development through the equitable utilization of, and countries: Burundi, the Democratic Republic of Congo, the Arab benefit from, the common Nile Basin water resources.� The Republic of Egypt, Eritrea, Ethiopia, Kenya, Rwanda, Sudan, Tan- program includes basinwide projects to lay the foundation for zania, and Uganda. Tensions, some ancient, arise because all ri- joint action, and two subbasin programs of cooperative invest- parians rely to a greater or lesser extent on the waters of the ments that will promote poverty alleviation, growth, and better Nile for their basic needs and economic growth. For some, the environmental management. The initiative enjoys the strong waters of the Nile are perceived as central to their survival. support of many donor partners through an International Con- The countries of the basin are characterized by extreme sortium for Cooperation on the Nile, chaired by the World Bank. poverty, widespread conflict, and increasing water scarcity The Nile waters embody both potential for conflict and po- in the face of growing water demands. This instability com- tential for mutual gain. Unilateral water development strate- pounds the challenges of economic growth in the region, as gies in the basin could lead to serious degradation of the river does a growing scarcity of water relative to the basin’s bur- system and greatly increase tensions among riparians. But co- geoning population. About 150 million people live in the basin operative development and management of Nile waters in sus- today, with growing water demand per capita. More than 300 tainable ways could increase total river flows and economic million people are projected to be living in the basin in 25 years. benefits, generating opportunities for “win-win� gains that can The pressures on scarce water resources will be very great. be shared among the riparians. The Initiative provides an insti- The countries of the Nile have made a conscious decision tutional framework to promote this cooperation, built on strong to use the river as a force to unify and integrate—rather than riparian ownership and shared purpose and supported by the divide and fragment—the region, committing themselves to co- international community. Cooperative water resources man- operation. Together they have launched the Nile Basin Initia- agement might also serve as a catalyst for greater regional in- tive, led by a Council of Ministers of Water Affairs of the Nile tegration beyond the river, with benefits far exceeding those Basin, with the support of a Technical Advisory Committee, and from the river itself. a Secretariat in Entebbe, Uganda. The initiative is a regional partnership within which the countries of the Nile Basin have Source: World Bank staff. by community institutions. Elsewhere the resources and land tenure—and changing the incentives and were so abundant that there was no need for elaborate organizational culture of the national forestry or management. In both cases, many of these woodlands land management authority. It also requires nego- were appropriated by colonial governments, often tiating rights among traditional users of common eager for timber revenues. The problem was that these property resources—and building social capital and governments and their independent successors often management capacity in local communities. These lacked the ability to manage and protect these re- are formidable challenges, but a decade of effort has sources—and the interest in involving the communi- yielded some encouraging results—as well as cau- ties that used them. As population and economic tionary lessons. Projects in India and Nepal show pressures increased, these woodlands have become de- that communities can realize greater income and en- graded through conversion and overexploitation. vironmental gains through management and recu- Since 1985 many countries have begun to trans- peration of highly degraded forest areas. But there fer control of woodlands back to local communities. has sometimes been less willingness of government Bolivia, Colombia, and Peru transferred almost 50 to relinquish areas that still contain valuable timber million hectares to community ownership; Bolivia, resources.47 Brazil, India, and Peru set up community manage- ment over 111.1 million hectares. Indonesia, Nepal, Fragmented habitats with less-disputed tenure Sudan, Tanzania, and a number of other countries The tradeoff between biodiversity goals and private have undertaken similar programs.46 profits is most problematic in more extensively mod- Projects in these countries seek to foster commu- ified areas where most of the original habitat has nity institutions for forest management, as well as been lost. These lands, attractive to settlement, re- formally transfer authority. Doing so often requires tain less-disturbed natural habitat patches within changing the national policies and laws for forests mosaics of agricultural land. One study identified a (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  set of such areas, the hotspots that have lost more sufficient to create a sustainable source of value in than 70 percent of their original area and now hold biodiversity, one that the landholder would then be about one-third of the world’s terrestrial biodiversity motivated to maintain rather than mine. An elegant on just 1.4 percent of the Earth’s surface.48 pilot project in Peru illustrates the principle. There, Fragmentation raises the risk of extinction. Smaller villagers will “ranch� valuable poison dart frogs in fragments support fewer species. Species caught in the forest, using a technique that hatches and har- shrinking fragments may vanish locally; if they are un- vests in a sustainable manner more juveniles than lucky enough to be restricted to just a few fragments, would normally grow. Only juveniles will be ex- they risk extinction. It takes time, though, for species ported; since they are impossible to catch in the to vanish in a newly isolated fragment, as their popu- wild, the scheme will not induce poaching if prop- lations dwindle slowly. In a 10 square kilometers frag- erly enforced. The frogs fetch high prices, so there is ment, half the threatened species—those unsupport- a strong incentive to keep the forest in place. able by the smaller fragment—are lost in 50 years;49 But there is a growing consensus that this kind of in a 1 square kilometer fragment the half life is just 10 fully self-reinforcing approach, while locally impor- years.50 So over the coming decades there is the risk tant and worth pursuing where possible, may have of an avalanche of extinctions—and the consequent limited scope. Few wild biological resources are ex- loss of entire ecosystems—if habitat loss and fragmen- tremely profitable, resistant to domestication, and tation continue. But there is also the possibility of re- more attractive for a landowner to maintain than to versing the decline if action is swift enough now. liquidate. Large trees, for instance, grow more slowly Because these areas have been settled longer, parts than money in the bank, so they are always tempting of them may exhibit reasonably well-defined land to liquidate in the absence of regulation or strongly tenure for individuals or groups—though rarely felt nonmarket values. Ecotourism today rarely con- without some degree of dispute. And tenure gener- fers substantial per hectare returns, though there are ally carries with it some measure of legal or tradi- some examples of success (often partially subsidized tional rights to modify land cover. So the problem by donors) in community wildlife management in of establishing governance is less pressing than in Africa.51 Integrated conservation-development proj- frontier forests (though rarely absent), and attention ects, premised on the idea that improved local liveli- focuses on reconciling the interests of landholders hoods would reduce pressure on habitats, have also with those of the wider community. been disappointing. In some cases local agents were The proximity of people and habitats increases not responsible for habitat degradation; in others un- the value of environmental services, such as flood conditional provision of additional income did noth- prevention and recreation. But favorable agrocli- ing to diminish the attractiveness of overexploiting matic conditions and dense populations motivate natural resources. landholders to drain wetlands, to appropriate stream A more promising approach to self-reinforcing in- flows, to “mine� forests, and to expand their towns, centives seeks to shift farmers to more biodiversity- croplands, and pastures. These areas thus have high friendly forms of land management.52 This includes conservation values for the local and global commu- promoting agroforestry systems that mimic and com- nity—and often high exploitation values for the plement the biodiversity and hydrological functions landholder. How can these values be reconciled? of the original ecosystem while providing more in- The general approach is to use markets, regula- come and employment than annual crops. In Suma- tions, or inducements to change the landholder’s in- tra, smallholder rubber agroforests improved plant- centives. It helps to distinguish between incentives ing stock may be able to maintain half the species that are self-enforcing and those that require exter- richness and carbon levels of primary forest, while of- nal monitoring and enforcement. fering profits and employment generation superior Much project-oriented work in promoting biodi- to that of biodiversity-poor oil palm plantations.53 versity (apart from establishing protected areas) has Such systems may help to restore biodiversity in de- been aimed at setting up self-enforcing incentives graded ecosystems dominated by agricultural pro- through new technologies or through new markets. duction, to reduce habitat damage downstream from The dream is that a one-time intervention would be intensive agricultural areas, and to enhance the con- (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  servation effectiveness in protected areas by enhanc- Funds are then used to purchase five-year renewable ing the habitat quality of surrounding land uses. conservation easements on forested property. But those introducing ecofriendly farming ap- Brazilian states have recently introduced two ex- proaches walk a knife-edge. Not profitable enough, tremely innovative financing mechanisms. One, the and the approach will be shunned. Too profitable, ICMS Ecológico (box 8.6), modifies state revenue- and it could displace the habitats it is supposed to sharing rules to reward municipios (districts) that save. So agroforests and similar approaches can com- create public or private protected areas, or protect plement but not substitute for the maintenance of watersheds. The other (box 8.7) introduces tradabil- some areas of natural habitat, and may not always be ity of a long-standing obligation of landholders to self-enforcing. maintain a set proportion of each property as a for- est reserve. With tradability, farms that are out of Equity and efficiency in blending development compliance can potentially pay others to maintain and conservation and expand high-quality forest of biodiversity value, Disillusionment with the self-enforcing approach has rather than uproot profitable, employment-generat- prompted interest in an alternative that compensates ing crops in a vain and expensive attempt to recreate landholders for agreeing to externally verified restric- a vanished forest. This reduces compliance costs by tions on land use.54 Payments may be ongoing, or creating a market for conservation services. Paraná where legal institutions are strong, landholders may state has recently used tradability as a means of se- agree to permanent conservation easements on their curing stakeholder support for a new law that seeks property in return for a one-time payment. Payments to secure universal enforcement of the forest reserve may be directly financed by the state on behalf of the obligation. beneficiaries of environmental services. Or the state These examples point the way toward ecosystem may create a market for these services by imposing reg- management institutions with three important fea- ulatory requirements on environmental service users. tures aimed at balancing interests and forging long- A well-known example is the U.S. Conservation term commitments. First, they would foster a partic- Reserve Program, which spends about $1.5 billion ipatory formulation of a vision and specific goals for a year in incentives for landholders to remove envi- regional development and landscape management. ronmentally sensitive land from production and es- Environmental goals might well include mainte- tablish vegetation that prevents erosion. Funding is nance of representative ecosystems over areas large based on a scoring system that considers a range of enough to ensure their long-term viability. Second, environmental benefits as well as the farmer’s asking they would allow for flexibility in achieving those price, resulting in a cost-effective award system. Eu- goals, reducing the scope for conflict among stake- rope spends a comparable amount for conservation set-asides.55 holders and reducing social and private costs of out- In the developing world, Latin American coun- comes that are valued but hard to measure in finan- tries are leading the way. Costa Rica’s Payment for cial terms. Third, they would set up incentives for Environmental Services Program (box 8.5) aggre- landholders to realize the regional vision. gates financing for forest conservation from a vari- International contributions of funds to such do- ety of dispersed beneficiaries: mestic landscape management institutions might be one way to meet both international and domestic Ⅲ Urban water users (who pay for sediment reduc- goals, while keeping land ownership and manage- tion). ment firmly in domestic hands. The domestic in- Ⅲ Run-of-river hydropower facilities (which care stitution would assess local goals and priorities, set about regulation of water flow). up transparent rules for providing and distributing Ⅲ Domestic taxpayers (concerned with biodiversity incentives, establish compliance and enforcement and scenic beauty, for their own enjoyment and as mechanisms, and receive domestic and international a source of tourism and bioprospecting revenue). financing, both public and private. It might well be Ⅲ Foreigners (seeking carbon sequestration credits integrated with regional development authorities to comply with voluntary or regulatory limits on and use funding to address poverty alleviation needs net CO2 emissions). that are only indirectly tied to land use but are per- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 8.5 Costa Rica’s program of payment for environmental services Costa Rica has pioneered a program that allows those who reforestation. In all cases, participants must present a forest benefit from the environmental services of forests to compen- management plan, certified by a licensed forester, that de- sate those who bear the burden of maintaining those forests. scribes the biophysical condition of the land, sets up a moni- The Payments for Environmental Services Program is an out- toring schedule, and specifies actions for the prevention of for- growth of a landmark 1996 Forestry Law, which recognizes est fires, illegal hunting, and illegal harvesting. The landholders four environmental services provided by forest ecosystems: cede rights to environmental services (such as sequestered mitigation of GHG emissions; hydrological services, including carbon) to FONAFIFO. Payments differ by contract type. For- water for human consumption, irrigation, and energy produc- est conservation contracts, which constitute 85 percent of the tion; biodiversity conservation; and scenic beauty for recreation contracted area, pay $42 per hectare a year for five years and ecotourism. Under the program, users of these services against the completion of specified tasks. By the end of 2001, finance a national forestry fund (FONAFIFO), which in turn con- 4,461 contracts covered 283,384 hectares, with 14 percent of tracts with private landholders for forest conservation and ap- the area belonging to indigenous groups. plication of sustainable management practices. The fund finances the program in part through the sale of The program arose from a growing awareness of forests’ these services. Hydropower producers, including both small importance against a backdrop of rapid deforestation. In 1950 private facilities and the state-owned Compañía Nacional de forests covered approximately half of Costa Rica. But in the Fuerza y Luz, are interested in purchasing environmental ser- 1970s and 1980s, the country’s deforestation rate was among vices such as stream-flow regulation, sediment retention, and the highest in the world. Appreciation of the importance of erosion control. These private and public sector companies Costa Rica’s biodiversity—both as an element of the national have signed multiyear contracts totaling more than $5.5 mil- patrimony and as a source of revenue through ecotourism— lion. International sales of carbon offsets (carbon sequestration prompted the creation of an extensive national park system. Even so, much of the nation’s forest remained in private hands. services) have netted $2 million. The GEF, through the World And from a landholder’s viewpoint, extraction of all salable Bank, recently provided $5 million to support forestry conser- timber and conversion to pasture was more profitable than vation contracts in priority areas of the Mesoamerican Biologi- sustainable forestry, and certainly more profitable than strict cal Corridor as well as an additional $3 million to strengthen forest conservation. By 1995 forest cover had fallen to just one- program implementation. This is supplemented by a $32 mil- quarter of Costa Rica’s territory. But from the early 1990s there lion World Bank loan to support the program while long-term had been increasing attention by NGOs and government agen- financing mechanisms are developed and institutionalized. So cies to the environmental services of forests, catalyzed in part far, the bulk of the $57 million expended has come from a na- by a World Bank study that tried for the first time to place eco- tionwide fuel tax. nomic values on forest environmental services. These discus- As a pioneering effort, the program faces a variety of chal- sions culminated in the new forestry law. lenges—among them, reducing the costs of monitoring and The national forestry fund contracts with individuals (for up enforcing thousands of small contracts, optimizing the Pro- to 300 hectares of primary and mature secondary forest), with gram’s impact on environmental quality, and securing long- indigenous reserves, and with NGO groups representing small- term sustainable sources of finance. holders. There are three types of contracts: for conservation of existing forests, for sustainable forest management, and for Source: Ortiz Malavasi and Kellenberg, background note for WDR 2003. ceived as being part of a comprehensive vision of sus- lating more rapidly in the atmosphere than can be tainable local development. Having such an insti- removed by natural sinks, these gases trap heat, tution in place as a precondition for international changing climate in complex ways, with widespread conservation finance would allay international fears impacts. This is quintessentially a global problem be- that the promise of funding would perversely induce cause GHGs mix rapidly in the atmosphere and have greater habitat destruction. It would also allay do- the same impact on climate change regardless of mestic fears of foreign control of land and threats to where they are emitted. And it is a long-term prob- sovereignty. lem because the great inertia in social, economic, and physical systems means that it would take decades to Mitigating and adapting to risks moderate the rate of change substantially. of climate change Because of its characteristics, climate change has People are changing the planet’s climate. Burning been a particularly difficult problem to solve. It has fossil fuels—and to a lesser but important extent, de- been difficult for society to pick up signals—to un- forestation and other land use practices—releases derstand the causes, magnitude, and consequences of CO2 and other greenhouse gases (GHGs). Accumu- climate change. Atmospheric CO2 has been increas- (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  Box 8.6 Municipal incentives for conservation A major source of state finance in Brazil is a value added tax, While the ICMS Ecológico represents only a small propor- the ICMS. One-quarter of the tax is rebated by states to mu- tion of total ICMS disbursements, it constitutes a relatively nicipalities. Of this payment, three-quarters must be pro- large incentive by the standards of conservation programs. portional to the municipality’s contribution; the rest may be Annual budgets have been about R$50 million in Paraná and distributed according to criteria set by the state. Four states— R$15 million in Minas Gerais. (Until 1999, the Brazilian real and Paraná, São Paulo, Minas Gerais, and Rondônia—now use the U.S. dollar were roughly equivalent.) area under protection as one criterion for redistribution. Since the programs were adopted, about 1 million hectares The ICMS Ecológico is a unique Brazilian mechanism that have been placed under environmental zoning restrictions in uses state-to-municipality (including rural districts) transfers to Paraná, and about 800,000 in Minas Gerais. Field interviews reward the creation and maintenance of protected areas for suggest that municipal authorities deploy local incentives to in- biodiversity conservation and watershed protection. The intent duce landholders to undertake these restrictions, in order to is to counteract local perceptions that maintenance of pro- attract state funding. The ICMS Ecológico is thus an interest- tected areas reduces municipal revenue. This provides an in- ing mechanism because it affects landholder incentives with- centive for local authorities and communities to support the out incurring the large transactions costs associated with pay- establishment of protected areas rather than permit, say, the ments directly to landholders. Its effectiveness, however, expansion of extensive cattle ranching. But the revenue trans- depends on the ability of the state to monitor and enforce land- fers are untied and need not be devoted to park management. holders’ compliance with conservation commitments. The proportion devoted to protected area incentives varies from 0.5 percent in Minas Gerais to 5 percent in Paraná and Rondônia. In Minas Gerais much of the redistributive portion of the ICMS is used to support social objectives other than Source: May and others (forthcoming). See also Bernardes (1999); environment. Grieg-Gran (2000). Box 8.7 Tradable forest obligations efficiently meeting conservation goals The Brazilian state of Paraná has created a market for conser- est reserve requirements are enforced property-by-property, vation by allowing trade in landholder obligations to maintain landholders with less than 20 percent forest cover achieve forests. A long-standing Brazilian law has required that prop- compliance by abandoning their land to spontaneous regrowth. erty owners maintain 20 percent of each property under native Because this land is heavily worked and has sparse seed vegetation (50 percent to 80 percent in the Amazon region). sources, this regrowth is likely to be of low quality, with little But noncompliance was common. real environmental benefit. The private costs of compliance are Paraná’s new law allows landowners to satisfy their forest estimated at about R$1.5 billion. In the trading scenarios, land- reserve requirements off site, on areas of greater ecological holders may achieve compliance in part by purchasing for- significance but lower opportunity cost. “Trading� of forest re- est protection or regeneration from others who have more serve is allowed only within biome–river basin combinations in than 20 percent forest cover. Because of the proximity of for- order to ensure full representation of the state’s biodiversity. est remnants, regeneration from this source is likely to be As an incentive for compliance, landholders must prove that more vigorous and of substantially greater ecological value. they are registered with SISLEG to carry out any legal transac- When landholders are free to trade within the same biome, tion related to their land, such as sales. compliance costs drop by almost three-quarters, while the A preliminary analysis of a hypothetical similar program for proportion of higher-ecological-quality forest reserve increases the nearby state of Minas Gerais illustrates how efficiency- to 72 percent. enhancing programs such as this might increase biodiversity conservation and economic output. In a scenario in which for- Source: Chomitz, Thomas, and Salazar Brandão (forthcoming). ing since 1750. Svante Arrhenius surmised in 1896 to climate change are poor, live in remote regions— that this might affect global climate, but emerging or have not even been born. Even the vulnerable consensus on aspects of the problem has been wealthy—owners of oceanfront property, for in- achieved only a century later with the IPCC (de- stance—may not yet rank climate change among their scribed earlier). Dispersion of interests in mitigating greatest current concerns. The voice of these numer- climate change has been a barrier to achieving agree- ous but diffuse interests is weaker than that of indus- ment on actions. Many of the people most vulnerable tries and consumers, especially wealthy ones, that are (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  heavily reliant on fossil fuels and would bear the bur- longer lasting. Saltwater intrusions, and more severe den of control costs. Finally, climate change is an dry seasons, will reduce fresh water availability in extreme example of the commitment problem de- coastal areas. As coastal populations swell world- scribed in chapter 3. Mitigation of climate change will wide, a 40-centimeter rise in the sea level would in- require a concerted, decades-long effort. crease the number of coastal dwellers at risk of an- With these barriers in mind this section starts by nual flooding by 75 to 206 million—90 percent of reviewing the consequences and sources of climate them in Africa and Asia.58 The starkest local impacts change. Using this information, it assesses institu- are faced by the low islands of the Pacific, some of tional aspects of undertaking the long-run miti- which could lose their freshwater and be largely in- gation of climate change. Then, it examines issues undated during storm surges if sea levels rise. related to adapting to the climate change that past Climate change could damage developing-country actions have already made inevitable—and that lack agriculture. Even taking into account crop substitu- of progress in mitigation will exacerbate. tion possibilities, one study finds that a 2° C temper- ature increase decreases the value of Indian agricul- Consequences and causes of climate change tural land by 36 percent.59 Arid and semi-arid areas Climate change is already here.56 Over the past cen- in Africa and Asia will probably face higher tempera- tury, mean global surface temperature has increased tures. Feedback between vegetation loss and reduced by 0.4° to 0.8° Celsius (C). According to the IPCC, rainfall could result in faster desertification.60 GHGs released by human action are likely to have Impacts on industrial countries are thought to be been responsible for most of the warming of the past mixed, but may be generally negative.61 Agricultural 50 years. Other observed changes are consistent with productivity will likely improve, in the medium term, this warming. Sea levels rose 10 to 20 centimeters in some northern areas. But southern Europe will over the past century. Over the past 50 years, the likely suffer drier summers; much of Europe could summer extent of arctic sea-ice has shrunk by 10 per- experience river flooding. The Atlantic coast of the cent or more, and its thickness by 40 percent. Out- United States will be vulnerable to rising sea levels, side the polar regions, glaciers are retreating, affect- and Australia will likely be more subject to drought. ing mountain ecosystems and water flows. Droughts Current understanding also depicts the global cli- have become more frequent and intense in Asia and mate as a finely balanced mechanism that goes awry Africa. Many of the world’s coral reefs have been when stressed, with prehistoric instances of 10° C damaged by bleaching (see note 21), associated with global temperature changes occurring within the higher sea temperatures. Animals and plants have span of a decade.62 There is a risk of catastrophic shifted their geographic ranges and behavior. Ex- consequences of climate change that could be irre- treme weather events may have increased. versibly set in motion during this century. There Unchecked, these impacts are predicted to inten- could, for instance, be an abrupt failure of the great sify, posing risks of varying kinds for different coun- ocean “conveyor belt� currents that warm the North tries. Impacts will fall heavily on many developing Atlantic and mix deep with surface waters. Biodiver- countries, including those that have not contributed sity losses could be massive as habitat fragmentation to climate change. They are physically vulnerable. makes it impossible for plants and animals to mi- Climate-sensitive agriculture bulks large in their grate in response to rapidly changing temperatures. economies. And they have less institutional capacity The risks are difficult to evaluate, but they affect in- to adapt to change. dustrial as well as developing countries and are cred- Low-lying islands and coastal areas everywhere ible enough to demand attention. At the very least will be exposed to flooding and storm damage. they put a premium, or option value, on maintain- Bangladesh, for instance, may be severely hit. A re- ing lower levels of atmospheric GHGs while the cent study predicts that by 2030 an additional 14 world more carefully examines the consequences and percent of the country would become extremely vul- develops options for mitigation. nerable to floods caused by increased rainfall.57 A What drives climate change? GHGs have built up 10-centimeter increase in sea level would perma- in the atmosphere as a consequence of 250 years of nently inundate 2 percent of the country, with the emissions from burning fossil fuel, deforestation, additional effect of making floods more severe and and other sources. Currently, about 40 percent of the (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  human-induced heating effect63 is from increased at- income, across countries, though there is consider- mospheric concentrations of methane (from land- able variation in emissions per dollar of GDP and fills, rice paddies, and cows), nitrous oxide (from in- emissions per capita among the wealthier countries. dustry and agriculture), and halocarbons such as The remaining 5.1 billion tons come from tropical CFCs. The remaining 60 percent is CO2. Of the deforestation. approximately 28.2 billion tons of annual CO2 A look at two scenarios64 for future CO2 emissions emissions, 23.1 billion are from energy and other in- will help provide background for understanding the dustrial sources. This component is closely linked to challenge of mitigating climate change (figure 8.2). Figure 8.2 Fossil fuel–intensive and climate-friendly scenarios 1990–2100 Fossil fuel–intensive scenario CO2 emissions GtC (billions of tons carbon-equivalent) 30.00 OECD members as of 1990 25.00 Economies in transition Developing countries in Asia 20.00 Africa, Latin America, and the Middle East 15.00 10.00 5.00 0.00 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 Year Time to develop Lifetime of power plants new technologies and buildings Climate-friendly scenario CO2 emissions GtC (billions of tons carbon-equivalent) 30.00 OECD members as of 1990 25.00 Economies in transition Developing countries in Asia 20.00 Africa, Latin America, and the Middle East 15.00 10.00 5.00 0.00 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 Year Fossil scenario: +3° to 6.9° C temperature increase by 2100. Climate-friendly scenario: +1.2° to 3.3° C temperature increase by 2100. Source: Emissions scenarios A1FI and B1 from Nakicenovic and Swart (2000); temperature predictions from Stott and Kettleborough (2002). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Both scenarios start in 1990, with emissions per capita of human-made capital—incorporating greater en- in OECD countries six times the level in Asia (exclud- ergy efficiency and increased use of renewables—has ing Japan), and with total emissions about equally di- to begin now. vided between the developing and developed world. In sum, those whose actions cause climate change, Both scenarios posit rapid economic growth—and and those who bear its risks, form two diverse and substantial convergence of per capita GDP between only partially overlapping sets of actors. This diver- developed and developing countries. The top panel sity raises issues of equity and efficiency in seeking scenario is not a static extrapolation of current tech- options for climate change mitigation—and financ- nologies. It already incorporates rapid technological ing for both mitigation and adaptation. progress, with a 75 percent reduction in energy use per dollar of GDP, and increased use of renewables Mitigating climate change (up to 17 percent from 5 percent in 1999). Concerned about climatic risks, most of the world’s Nonetheless, emissions increase radically over the nations agreed in 1992 to the UNFCCC. The con- century, and industrial country emissions in 2100 are vention’s objective is defined as the “stabilization of far above world emissions in 1990. By 2100 the im- greenhouse gas concentrations in the atmosphere at plied mean increase in global temperature is 3.0° to a level that would prevent dangerous anthropogenic 6.9° C.65 The bottom panel scenario posits more vig- interference with the climate system.� But the Con- orous technological change, with a much less energy- vention itself did not quantify this level or specify intensive economy and a 52 percent share of renew- how to achieve it. able energy. This holds the temperature increase to As a first step the Kyoto Protocol to the UNFCCC the range 1.2° to 3.3° C.66 In both scenarios, OECD was negotiated in 1997. This agreement would re- emissions per capita are still twice the level of the de- quire industrial nations and economies in transi- veloping countries at the end of the century. tion—the Annex B countries—to accept specified These scenarios are illustrative rather than predic- limits on emissions of GHGs for 2008–12. The Pro- tive. But they convey three points that are essential tocol would decrease compliance costs by allowing to understanding the problems of balancing inter- Annex B countries to trade their emissions allow- ests and executing agreements. First, emissions per ances. It would also allow these countries to purchase capita in industrial countries are much higher than emissions reductions from developing countries, the in developing countries and are likely to remain reductions being reckoned against assumed “business higher for some time. In response to this imbal- as usual� levels, since the developing countries’ emis- ance—richer countries imposing higher per capita sions were not capped. The subsequent Marrakech externalities—the U.N. Framework Commission on Accords of 2001 allowed for developing countries to Climate Change (UNFCCC) established different- generate emissions reductions from forestry projects iated responsibilities for developed countries, requir- in only a limited way. At this writing, the Kyoto Pro- ing them to take the lead in addressing climate tocol has not entered into force. change and providing needed technology to the de- It is important to recognize that the Protocol’s veloping world. Second, developing countries will commitments for 2008–12, even if observed by all nonetheless emit substantially more than developed major emitters, would be only a first step toward countries in the future and therefore must be in- the UNFCCC goal. Keeping this in mind, this chap- volved in implementation. Third, pursuing the more ter outlines some strategic considerations in pur- climate-friendly scenario requires starting now. Much suing that long-run goal, a cornerstone of global capital stock—such as power plants and buildings— sustainability. has a working life of 50 years or more. And many If the world is to stabilize atmospheric concentra- of the renewable and low-carbon energy technolo- tions and provide good living standards to all its gies required for the favorable scenario will require citizens, it must switch in the long run to energy 10–20 years of research and development to bring technologies (such as wind, solar power, and hydro- to market. To have high-efficiency, low-carbon capi- gen, among others) that emit near-zero net amounts tal in place in the latter half of this century, the of CO2. Simple arithmetic shows why. The world’s process of research, development, and deployment population is now expected to stabilize at about 9 bil- (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  lion around mid-century. Suppose that people then erators, and buildings—is energy efficient, to aspire to the current lifestyle of a prosperous country. encourage agricultural intensification and main- Among the prosperous countries, Norway has one of tenance of carbon stocks in forests, and to shift the lowest ratios of CO2 emissions per capita from urban structures toward lower energy use. energy, owing in part to ample use of hydropower. Ⅲ Start now on research and development to ensure Yet if the global population of 2050 emitted CO2 on that zero-emission energy technologies can be de- average at this rate, the total would be about 2.5 veloped and widely deployed by mid-century. times current global emissions,67 which would greatly Ⅲ Building on current efforts, create adaptive inter- exceed the planetary absorptive capacity. national institutions for fostering cooperation and Between now and the time the world switches en- burden sharing. tirely to near-zero-emissions technologies, GHGs will accumulate in the atmosphere. The amount of Act now to reduce today’s emissions damage, and the risk of catastrophic changes, will be Although non-OECD countries use only about 20 related to the cumulative amount. To reduce the percent as much energy per capita as OECD coun- damage, the world needs to accelerate the shift to tries, they use 3.8 times as much energy per dollar of lower-emissions energy technologies, increase the ef- GDP.68 This disparity suggests looking for ways that ficiency of energy use, and reduce the emissions of developing and transition countries can increase ef- GHGs. ficiency and reduce fuel costs—with reduced GHG Although these actions provide some immediate emissions as a welcome side-benefit. Why are these side benefits in addition to their cumulative effect apparent “win-win� opportunities so elusive? Two on reducing climate damages, they involve costs. Be- types of institutional failures get in the way. First, cause emissions reductions represent a global public distortions in energy policy may disadvantage soci- good, burden sharing is inevitably contentious. To ety at large, but benefit special interests. Second, facilitate global coordination in this effort, a strat- firms and households neglect profitable ways of sav- egy has to reduce the overall cost of mitigating emis- ing energy because it is simply too much trouble to sions and seek to align local and global interests as pursue them. Fortunately, there are institutional so- far as possible. It also has to avoid free-rider prob- lutions to both of these problems—though neither lems. This requires further institutional innovation is easy to solve. at both national and global levels. Many energy-rich countries subsidize energy con- An adaptive strategy for mitigating climate change sumers or producers, resulting in inefficient fuel use, provides incentives for taking action now to reduce an inappropriate fuel mix, and needless CO2 emis- GHG emissions over three time horizons: near term sions. Box 7.6 discussed Iran, which spends 18 per- (5–10 years), medium term (10–20 years), and long cent of its GDP on petroleum product subsidies. term (20–50 years). The global climate change strat- Coal subsidies in OECD countries were $8 billion egy has to be adaptive because climate change miti- in 1997.69 gation will take most of this century to accomplish. Dismantling subsidies to energy—or to ineffi- Economic, environmental, and political conditions— cient energy-using industries—is no easy task, for and our understanding of climate change—will cer- reasons that this report has discussed at length. But tainly change markedly over this period. Some actions it is possible. China reduced CO2 emissions by 7.3 need to be undertaken now—the impact of those ac- percent over 1996–2000, largely through industrial tions will play out over these three time horizons: restructuring and fuel improvements, while increas- ing its GDP by 36 percent.70 These reductions were Ⅲ Vigorously pursue current options to cheaply abate accompanied by a 32 percent reduction in particu- GHG emissions, thus reducing the possibility of lates, which have severe health effects and contribute triggering catastrophic climate changes and buy- to global warming.71 ing time for longer-term, more fundamental ac- In both industrial and developing countries house- tions to take hold. holds and firms pass up energy-saving investments Ⅲ Set up incentives to ensure that the next genera- with extraordinarily high financial rates of return— tion of long-lived capital stock—transport, gen- on paper. Investments such as efficient electric mo- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  tors, compact fluorescent lights, improved boil- side management program in 1993.72 The program ers, and insulation can often pay for themselves in a first targeted lighting, which accounts for 20 percent year or two, in the process yielding reductions of of Thai electricity consumption. The program per- both GHGs and local air pollutants. But it takes ef- suaded Thai manufacturers of fluorescent lights to fort and attention to discover these opportunities, switch to a new design that consumed 10 percent which may appear burdensome and risky to pursue. less energy. The program eased consumer acceptance Consumers may legitimately wonder if an expensive through a combination of advertising and imposi- light bulb is really going to last long enough to pro- tion of standards for light quality and durability. duce the advertised savings, or if the spectrum of the Within a year the new lighting commanded 100 per- illumination will be unpleasant. They may not know cent of the market. Estimated benefit-cost ratios or much care that some appliances draw a couple of were 54.6 for consumers and 13.8 for society as a watts of stand-by power, though on a national scale whole, taking account of the program costs. those watts add up to entire generating stations. Cor- There appear to be many opportunities for devel- porate executives or government facility managers oping countries to reduce GHG emissions at a cost may not have the information or incentive to find just high enough to be a local deterrent, but quite opportunities for reducing heating bills. low for the world. The capture of methane from New sets of institutions are making it easier for landfills is an example with global applicability (box consumers, business, and governments to take ad- 8.8). Examples such as these motivate the “carbon vantage of energy efficiency opportunities. These market,� which mobilizes funds from the industrial include government initiatives to set standards and world to tip the balance toward clean energy in the disseminate information about efficiency. These ini- developing and transition economies. tiatives, pioneered in industrial countries, are now Agricultural intensification, combined with pro- being extended to developing and transition econo- tection of forests from wasteful destruction, has the mies. Thailand introduced a $189 million demand- potential to dramatically reduce CO2 emissions while Box 8.8 The Prototype Carbon Fund and the carbon market The Prototype Carbon Fund is a pilot effort to “show how proj- The Prototype Carbon Fund is a learning-by-doing enter- ect-based greenhouse gas emissions reduction transactions prise to help answer these questions. With $145 million con- can promote and contribute to sustainable development and tributed by six national governments and 17 private firms, it lower the cost of compliance with the Kyoto Protocol.� The seeks to purchase emission reductions from 25–30 projects. Protocol, if it comes into force, sets up opportunities for Its first project finances methane capture and electricity gen- developing countries and economies in transition to adopt eration at a municipal landfill in Latvia. Without this financing cleaner technologies and sell the resulting reductions in GHG the city of Liepaja would not have found it attractive to capture emissions to industrial countries that have committed to limits the methane that landfills emit. The capital costs are high rela- on their own net emissions. (Indeed such a market may come tive to the value of electricity; the economic rate of return into being even if the Kyoto Protocol fails, arising from national- would have been only 2.6 percent. A combination of carbon level policies and voluntary markets for emission reductions.) and grant financing for the initial investment boosts the city’s The carbon market offers tremendous potential. It could re- return to 22 percent. It will also result in an estimated reduc- duce the cost to industrial countries of achieving any agreed tion of 681,000 tons (CO2 equivalent) of GHG emission, be- goal for emissions reductions. It could stimulate the develop- cause methane is a powerful heat-trapping gas; using it for ment of renewable energy technologies. And it could provide electricity not only directly reduces emissions, but also re- technology, environmental benefits, and export revenues to duces combustion of fossil fuels. The project also provides the the developing and transition world. city with a landfill built to higher environmental standards. Achieving this potential, however, requires resolving a host In undertaking this transaction, the Prototype Carbon Fund of technical and institutional problems. Can emission reduc- pioneered the development of institutional tools for contracting, tions be produced at reasonable cost? How do you credibly monitoring, and verifying emission reductions. This information measure them? How do you contract for them? Do they really has been widely disseminated as a global public good, reducing contribute to sustainable development? Answering these transactions costs for future methane-capture projects. questions is important not only for the implementation of car- bon markets, but for fostering consensus on their feasibility. Source: World Bank (2002h). (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  reducing rural poverty, protecting biodiversity, and problems.75 Most important, a global decision to in- providing local environmental services. As mentioned vest in a portfolio of forest and agriculture carbon earlier, land-use change contributes 5.1 billion tons sinks diversifies this risk. Running these investments per year of CO2 to the atmosphere, plus or minus 50 through locally controlled landscape management percent—that is, 10 percent to 30 percent of total institutions ensures that the arrangements are ac- human emissions.73 Most of the land-use emissions ceptable. And it also helps shape the long-term in- result from the conversion of tropical forests. A sub- centives for agricultural intensification that averts stantial portion of this conversion yields pasture or the long-term pressure for deforestation. croplands with modest returns. The agricultural in- tensification strategy described in chapter 5 keeps Act now to reduce emissions over the medium these forests in place, for future sustainable use, and and long terms promotes labor-intensive cropping in more suitable Actions now to affect the evolution of the capital lands. Improved soils and denser crops also serve to stock—vehicles, buildings, and generators—can yield absorb CO2, which increases the land’s productivity huge and long-lasting reductions in GHG emissions and resilience. Timber plantations, agroforestry, and and improvements in economic efficiency. Producing biomass plantations could add substantially to seques- this equipment generates vast amounts of emissions. tration while improving rural livelihoods. And once in place the equipment drives emissions for Incentives for forest conservation and soil carbon decades. Turnover times are about 10 years for vehi- present implementation problems but offer a vast cles, 30–50 years for power plants, and 80 years for payoff. Throughout the tropical world farmers may residential buildings. This means great opportunities burn a hectare of rainforest to get a one-off gain of a for reducing long-term emissions and fuel costs by few hundred dollars—while releasing hundreds of using energy efficient technologies to expand the tons of CO2 and destroying priceless biodiversity. capital stock, or to replace equipment that is being Each year, according to FAO data, deforestation retired. claims 3.8 million hectares of tropical forest with The opportunities are particularly great for devel- biomass greater than 200 tons per hectare, equiva- oping countries, which will be investing massively in lent to about 370 tons per hectare of CO2 emissions long-lived infrastructure as a keystone of develop- if fully cleared. This implies an abatement cost of ment. Between 1997 and 2020 developing countries only a dollar or two per ton.74 are expected to expand their electricity-generating Meanwhile, energy users in industrial countries capacity by a factor of 2.5, investing $1.7 trillion in who desire to abate the same amount of CO2 at new plants and perhaps more in transmission and home—for voluntary reasons, or to meet a regula- distribution.76 Of China’s building stock in 2015, tory requirement—may end up spending consider- half is expected to be built between now and then.77 ably more. In today’s nascent carbon market, buyers Once erected, those buildings are likely to be in are paying $4.40 to $8 per ton to comply with na- place for half a century or more. But current build- tional regulations, and some scenarios for a global ing practices use antiquated technologies that leak carbon market predict substantially higher prices. heat, do not allow users to adjust heat levels, and The potential gains from trade appear to be large. By consume 50–100 percent more energy than build- splitting the difference, energy users in the devel- ings in similar climes elsewhere. The coal to heat oped world could, in principle, save money in meet- these buildings already generates 350 million tons of ing their CO2 reduction obligation, help maintain CO2 a year and much of northern China’s unhealth- the many services and values of the tropical forest, ful levels of sulfur dioxide (SO2) and particulates. and invest in a superior livelihood for the tropical Clearly, a vigorous shift in building practices could farmer. As part of that livelihood improved soils and have tremendous long-term benefits both for China plantings would sequester even more CO2. and for the world’s climate. According to a World There are many practical problems in realizing Bank study, such a shift will take substantial reforms this vision, not least the danger that any particular in energy policies so that consumers have an incen- plot of forest may eventually get burned or cut. But tive to conserve energy, but in a way to protect the there are practical approaches to addressing these poorest. It will simultaneously require research, de- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  velopment, and dissemination of improved building research has declined in Europe and the United designs appropriate to local conditions. States.80 Only 21.8 percent of the energy research Actions now can determine whether development budgets of countries belonging to the International paths “lock in� to high- or low-energy regimes, with Energy Association is devoted to renewable energy self-reinforcing patterns of policy, infrastructure, and conservation.81 capital, and lifestyle. Land-rich countries includ- Increased funding for research could substantially ing Australia, Canada, and the United States have advance the time at which low-emissions energy tech- evolved energy-intensive lifestyles featuring low fuel nologies are deployed and thereby reduce the burden prices and heavy reliance on automobiles.78 Social of GHG emissions controls. This in turn could facil- norms, infrastructure placement, and relative prices itate international agreements on such controls. New discourage individuals from opting for lifestyles that technologies could also provide a wide range of envi- consume fewer resources. And because individuals ronmental benefits. Most important, new technolo- are locked in to high energy consumption, there is gies—especially those related to energy use and effi- likely to be little political support for increasing en- ciency—may be able to reduce the energy bill of the ergy prices to levels that reflect environmental im- developing world. This provides a powerful rationale pacts. Once this lock-in occurs, it may take a gener- for collaborative global research on energy, involving ation or more to change. Lock-in is prevalent in the scientists and engineers from both the developing energy supply sector as well. Coal dependence, for and developed world. It also suggests efforts to en- instance, creates infrastructure, communities, and sure that technologies derived from this research are powerful political constituencies, making it difficult available on favorable terms to all. to shift to less carbon-intensive fuel sources. Over the longer term the atmosphere’s level of International cooperation to reduce emissions greenhouse gases can be stabilized only by switching Short horizon or long, these agendas require com- the world to zero-emission energy sources: wind plementary actions now. Taxes and carbon markets power, solar power, renewable biomass, fusion, and have a number of advantages. They can provide price fossil fuel (with equivalent physical sequestration of signals that spur cost-efficient energy conservation CO2). A few, such as wind power in favorable loca- and forest conservation. These signals may provide tions, have good short-term prospects. But most of the demand stimulus that drives renewable tech- these technologies are thought to be decades from nologies such as wind power and solar power down large-scale commercial realization—and then only the learning curve, making them competitive with if basic and applied research are more vigorously fossil fuels in some areas. This mechanism can there- pursued. Historically, new energy technologies have fore support the development and transfer of tech- taken half a century or more to displace earlier ones. nology adapted to developing countries. Properly set Accelerated development and deployment of new up, carbon markets (such as those envisioned under technologies are therefore essential for substantial re- the Kyoto Protocol) can result in the decentralized ductions of emissions in this century. transfer of resources and technology to sustainable There is an urgent need to boost basic research in development projects in developing and transition energy technologies. The lag times between basic re- economies. search and large-scale commercial deployment are Initiatives to encourage the adoption of low-emis- sobering. Private industry is not willing to undertake sion capital equipment, and development of low- the necessary basic research in areas such as fusion, emissions or efficient technologies, can complement geological carbon sequestration, high-efficiency coal carbon markets and carbon taxes. Imposing energy combustion, or high-efficiency building technolo- standards (for instance, on cars or buildings) could gies for tropical climates. Moreover, there is at least be economically inefficient, but such regulations anecdotal evidence of high returns to government might have advantages. They might fight market funding even in relatively applied research. For in- failures for which price remedies are not apt, such as stance, a $3 million public investment in technolo- a tendency for building developers to shift recurrent gies for efficient windows is projected to yield $15 energy costs to ill-informed renters or buyers. Or billion in energy savings through 2015 in the United they might prove to be more politically acceptable, States alone.79 Yet public funding for basic energy and more amenable, to a long-term commitment (c) The International Bank for Reconstruction and Development / The World Bank ������ �������� ��� ����� ��������  than taxes. And as more people switch to efficient then, that adaptation efforts are necessary—but the equipment, it becomes easier to support tighter lim- adaptation agenda has only begun to be addressed. its on emissions associated with carbon markets. Some impacts of climate change are relatively pre- Similarly, accelerating research on new technologies dictable and will play out inexorably over coming can nicely complement price policies and other poli- decades. Dealing with them will require foresight, cies that encourage rapid development, dissemina- commitment, and resources. For instance, an obvi- tion, and uptake of those technologies. ous way to reduce vulnerability to a rise in the sea How can emissions reductions—beyond those level is to avoid the emergence of large settlements that pay for themselves—be financed? This remains in low-lying areas. However, it is generally difficult the most contentious issue in climate change mitiga- to exclude urban settlers from areas attractive to tion. In carbon markets, for instance, the allocation them. Adaptation considerations may therefore re- of emission allowances determines who pays for re- quire larger current investments in developing set- ductions. In the view of many, equal per capita tlement alternatives, as a complement to the pro- allocation of allowances across the world—perhaps tection of areas that are at increasing risk. Other entailing transfers from rich emitters to poor long-horizon issues include advance planning to re- countries—would constitute an equitable allocation. place threatened water supplies, developing drought- But such an allocation rule, if imposed abruptly, resistant crop varieties, and maintaining biodiversity might disrupt the rich emitters’ economies and thus corridors so that wildlife can migrate in response to would not secure their participation in the scheme. changing temperature. On the other hand, a strong link between past emis- An immediate and enduring effect of climate sions and current allowances, applied globally, would change is to increase climate-related risks, such as hurt the development prospects of poor nations and droughts, floods, and storms. This occurs both be- thus be unacceptable. Hybrid allocation schemes that cause the climate itself becomes more volatile and blend per capita and “grandfathered� allocations and because the past becomes an ever less reliable guide shift toward the former over time have been proposed to the future, especially for infrequent catastrophic as a compromise. Other alternatives include coordi- events. A recent study found that large floods are be- nated national carbon taxes, whereby each country coming more frequent, as climate change models retains the tax revenue and combinations of al- would predict.82 This suggests that the cost of build- lowances and taxes, and the taxes serve as a “safety ing (or insuring) infrastructure to a given risk stan- valve,� limiting compliance costs if allowance prices dard (say, to withstand a once-in-100-years flood) is rise too high. Agreements on burden sharing are rising even now. stymied in part by uncertainty about the actual eco- There is growing appreciation that developing nomic burden that any of these systems would entail. countries, especially, are not dealing optimally with The experience of the CLRTAP suggests that it current weather-related risks, let alone future ones. may not be necessary to work out long-run burden- So efforts to reduce current vulnerabilities will not sharing formulas in great detail in advance. A prac- only have immediate payoffs—they will increase the tical alternative is to engage all parties by starting countries’ capacity to deal with increasing vulnera- with confidence-building steps, while maintaining bilities to climate change. momentum to tackle progressively more ambitious One emerging set of innovative coping mecha- goals, more difficult decisions, and longer-term com- nisms involves the use of long-term weather forecast- mitments as options are better understood. It is ur- ing and insurance markets to mitigate the risks of gent, however, to develop a framework that does not extreme weather events. These events can be partic- penalize nations or other actors that voluntarily re- ularly devastating to poor rural dwellers, whose duce their emissions in advance of commitments. entire network of mutual support can be disabled by droughts, floods, and storms. An interesting by- Adapting to climate change product of global climate research has been the in- The climate system has considerable inertia. Even creasing ability to forecast seasonal climate patterns if GHG emissions were magically halted today, the ef- months in advance. For instance, sea temperatures fect of past emissions would continue to raise temper- in the eastern Pacific can be used to predict season- atures and sea levels for centuries to come. It follows, ahead climate in Zimbabwe and thus potentially (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  to help poor farmers optimize their planting deci- The most general and effective way to help vul- sions.83 These predictions could also help marketing nerable poor countries adapt to climate change is to agents prepare for droughts, significantly reducing promote rapid and sustainable development. Over the impacts on household welfare.84 the coming decades more vigorous growth rates and There is also more interest in using insurance mar- accelerated investments in human capital will shift kets to help poor farmers cope with weather risks—a these countries out of climate-sensitive sectors and role that traditional crop insurance has never been improve their capacity to cope with climate-related able to play well because of the costs of enrolling risks. small farmers, measuring damages and processing Conclusion claims, and avoiding moral hazard and adverse se- lection.85 Weather insurance, in contrast, depends The distinctive feature of global problems is the lack on easily measurable temperature and precipitation of a central authority for coordination and enforce- data—and facilitates reinsurance. A current Interna- ment. Despite this obstacle, there are encouraging ex- tional Finance Corporation pilot project is exploring amples of successful transnational institution build- the potential for this kind of insurance in the devel- ing to tackle transborder environmental problems. oping world. These initiatives underline the value of Success has been greatest in cases such as stratospheric weather data as both local and global public goods. ozone and acid rain, where the problem can be made Management of large-scale climate risks will be- operational in precise technical terms; where interna- come more important at the subnational and na- tional action can therefore focus on tightly defined tional scale. Indeed, the financial damages from interventions; and where the perceived benefits of weather-related catastrophes are increasing rapidly, collective action have been high, for key actors, rela- though it is difficult to separate greater exposure tive to the cost. It will be more difficult for other from the higher frequency of extreme events. There environmental and social problems—where the re- is a strong role for individual nations, and the world lationship between action and impact is less well un- at large, to insure poor vulnerable regions against derstood, and where the costs and benefits of action these catastrophes, a role already filled (on an ad hoc do not coincide. Yet an adaptive strategy of the type and sometimes inadequate basis) through a patch- described in this chapter has much to recommend it- work of disaster relief responses. A key commitment self because the frequency and urgency of such prob- problem in designing a more comprehensive system lems is bound to increase as globalization progresses is providing adequate insurance without encourag- along many dimensions. The next chapter illustrates ing risk-seeking or environmentally damaging be- some ways of approaching the linkages among social, havior, such as settlements in areas that are at risk of economic, and environmental issues within and be- landslides, or agriculture in fragile areas.86 tween countries in a shrinking world. (c) The International Bank for Reconstruction and Development / The World Bank CHAPTER 9 Pathways to a Sustainable Future A ccelerated growth in productivity and in- that the current development path, though possible come can eliminate poverty and enhance for a while, is not sustainable. Science and technol- prosperity in developing countries. This ogy are providing some answers, but they will not be growth needs to be achieved at the same time criti- sufficient without complementary changes in insti- cal ecosystem services are improved and the social tutions. The world community, in confronting some fabric that underpins development is strengthened. of the challenges, is grappling with new strategies A close look at what is happening on the ground and goals: (chapters 4–8) reveals both cause for concern and cause for hope. Ⅲ Development agencies are shifting to more partici- Concern stems from evidence that getting the patory and holistic approaches with a medium- world on a sustainable path is problematic: term perspective, through the Comprehensive Development Framework (CDF) and focused Ⅲ In many developing countries, productivity is low, poverty reduction strategies, backed by actions on growth is stagnant, and unemployment is high. the ground through partnerships, broader inclu- Ⅲ The number of people living on less than $1 a day sion in the preparation of assistance strategies, and (1.2 billion) is dropping but it is still a challenge, some shifts in lending and grant aid. and more people are living on fragile lands. Ⅲ The private sector is more committed to sustain- Ⅲ Income inequality is rising. Average income in the able development, with greater use of triple-bot- wealthiest 20 countries is 37 times that in the poor- tom-line accounting by firms and greater use of est 20 countries—twice the ratio in 1970. environmental and social criteria by investors. Ⅲ Many of the poorest countries are wracked by civil Ⅲ Governments and civil society are supporting the conflict, with animosities deep and prolonged. Convention on Biodiversity, the Convention to Ⅲ Stress on the environment is increasing. Fisher- Combat Desertification, the Convention on Cli- ies are being overexploited, soils degraded, coral mate Change, and the Millennium Development reefs destroyed, tropical forests lost, air and water Goals. And civil society is demanding more pub- polluted. lic and private accountability. Ⅲ The financial transfers to address these issues are far from adequate, even though the resources are Although encouraging, these advances are small available. relative to the many challenges of sustainable devel- opment. Sustainability requires thinking long term, Hope springs from the genuine progress made al- but acting now—it also requires coordination. This ready in boosting average per-capita incomes in de- Report recognizes the importance of economic in- veloping countries and reducing infant mortality and centives and policies in changing behavior, but it illiteracy rates—and from the greater awareness of the does not focus on specific policies or organizational problems that remain. Disparate groups now agree designs. (Nor does it evaluate projections based on  (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  different policy or organizational scenarios.) Instead attention of society’s many institutions than those of it argues that well known and appropriate policies strong and vested interest groups. That is why greater have not been adopted or implemented because equality in access to assets and voice makes a differ- of distributional problems and institutional weak- ence in the kinds of environmental and social con- nesses, and that sustainable development with faster cerns that society addresses. growth and higher productivity, capable of eliminat- This Report argues that, for countries and local ing poverty and achieving a more just and sustainable communities, extreme inequalities in assets, power, development path, requires much stronger institu- and voice are corrosive, linked, and self-perpetuating. tions. To support improvements in well-being, these When the poor lack voice and a stake in society, so- institutions would need to manage a broader portfo- cial assets (such as trust) and environmental assets lio of assets and adapt to new problems and opportu- (on which the poor depend) are eroded, stability is nities. Technologies and preferences will change, and undermined, and the ability to solve economic, so- resource allocation and distribution issues cannot be cial, and environmental problems (that require col- neatly separated. lective action) dissipates. Managing risk is important. Not all risks are in- As the world comes to resemble a single commu- surable. Not all irreversible changes are bad, but nity, these lessons may apply even at the global level. some are. For the assets most at risk—the natural At the very least, inequality and the lack of hope for and social—markets cannot provide the basic coor- the poorest countries will inhibit attempts at solving dinating functions of sensing problems, balancing global problems—not just the current preoccupa- interests, and executing policies and solutions. These tions with cross-border spillovers of conflict and ter- types of assets have impacts, good and bad, that ex- rorism, but also the currently unimagined problems tend beyond individual transactions and thus require that will require global cooperation 20 or 50 years coordination to promote good spillovers and mini- hence. That is why ending global poverty is much mize the bad. This coordination reveals the need for more than a moral imperative—it is the cornerstone institutions that are capable of organizing dispersed of a sustainable world.1 interests, confronting vested interests, and ensuring The next 20 to 50 years are a demographic win- credible commitments in execution. dow of opportunity, created by the deceleration of This Report shows that rising income can facili- population growth rates and the decline in depen- tate but not guarantee better environmental and so- dency ratios. This period will also witness comple- cial outcomes by permitting countries to simply tion of the urban transition in most countries. The “grow out of � pollution or civil conflict. It also shows demographic transition will permit greater savings if that low income does not condemn people to a de- the working age population has jobs and investment teriorating environment or social climate. What opportunities. The urban transition will facilitate in- makes the difference? Public action, through compe- come generation, but over the next 50 years it will tent institutions. also require massive investments. Many of these Mobilizing institutional responses is more diffi- investments are likely to be long-lived—so getting cult for some problems than others. Problems with them right by incorporating environmental and so- impacts and risks that are diffuse and long term (such cial concerns in their design now is critical. as climate change and biodiversity loss) are less read- Even the next 15 years (2003–2018) could bring ily perceived and appreciated than those immediately a record period of economic growth in developing felt and measured (i.e., some forms of local air and countries. Driven by growth in China and India, in- water pollution and deforestation). Assets that are come in the low- and middle-income countries will public goods or common property goods—such as almost double—accounting for more than a third of clean air, forests, fisheries, and water—are a challenge the 60 percent increase in world output.2 This pe- to manage sustainably. The reason is that private riod offers the opportunity to lay the foundation for property rights are difficult to assign or enforce, or if inclusive growth—which will require confronting improperly designed they fragment and undermine barriers to change. Institutions that can manage the the underlying joint functions of an asset—say, an social and economic transitions, by partially com- interconnected ecosystem. In addition, the concerns pensating losing interests, are much easier to create of the poor and powerless are less likely to attract the in rapidly growing economies than in economies (c) The International Bank for Reconstruction and Development / The World Bank �������� �� � ����������� ������  where inclusion requires a battle over stagnant riorate after costly initial investments for lack of rou- shares.3 For many countries whether the opportu- tine funding and maintenance. This speaks to the nities generated by new growth are inclusive, or challenge of organizing beneficiaries for sustained whether they lock in vested interests and exclusive commitments. institutions, will depend on decisions taken in the This Report shows that even with imperfect insti- next few years. tutions it is possible—indeed imperative—to build now on the many institutional innovations already Acting today out there that show the way forward. This Report outlines actions that can be taken now Many, if not all, of the institutional innovations to improve the ability of institutions to identify, cited in this Report already show signs of being adopt, and implement policies that facilitate growth replicable and capable of being scaled up to meet the while addressing critical environmental and social challenges ahead. For example, the pilot experience issues. with security of tenure in favelas in Brazil is being extended to hundreds of thousands of households. Institutional and sectoral approaches The multistakeholder pilot pollution disclosure pro- are complementary grams in China have moved from 2 pilot municipal- Sustainable development requires action across ities to 13; countrywide implementation is currently many sectors and disciplines, including water, en- under discussion. The village initiative in Morocco ergy, health, agriculture, biodiversity, and others. In is being replicated in a dozen other villages. The for- a companion document, the World Bank sets forth est concession program in Cameroon has created a some recommendations and action plans to advance constituency for expanding it to other sectors. There sectoral goals.4 is much creativity under way with initiatives emerg- This Report takes a different approach—but one ing from the public sector, the private sector, and intended to complement and support the sectoral civil society. An enabling environment is needed to perspectives. Its message is that proposing and en- encourage such initiatives, to facilitate partnerships, dorsing a set of action plans are important first step, and to help mobilize the resources needed to scale but realizing them requires an institutional apparatus up promising activities within countries, and across that cuts across sectors. Achieving all the broad sec- countries. toral goals will involve problem diagnosis, decisions with distributional consequences, and coordinated Picking up signals early and sustained commitments to action. These func- Being sensitive to early signs of problems, especially from tional capabilities require general improvement in the fringes, is important if society is to avoid costly crises coordinating institutions within and across countries. later. HIV/AIDS is a case in point. In the early stages Making progress on the sectoral issues, for exam- of the epidemic, HIV/AIDS received little attention ple, requires first a better understanding of local con- since no one knew how contagious it was or the tra- ditions and a better ability to diagnose local prob- jectory of future costs. By now, with the evidence lems. Domestic and international institutions will be available, all countries should have programs to required to fund and implement R&D where local identify the problem early and to stay ahead of the capacities are deficient. Second, some difficult distri- epidemic. The same is true for environmental is- butional issues must be resolved: How to divide the sues—the status of biodiversity, air pollution, lead water among claimants? How to allocate health re- in gasoline, or the drawdown of acquifers in arid sources among preventive, primary, and secondary regions. care? What sort of land uses to permit, for whom, if Creating information for constituencies and con- environmental processes are to be maintained? With- stituencies for information. There is a need for signif- out institutions that represent fairly the interests of icant investment in information and indicators at dispersed and (usually) voiceless interests, institutions global and local levels, where this information would affecting these sectors are prone to capture by vested find users and audiences. The initiative must go well interests, and they are unlikely to implement efficient beyond current attempts to devise indicators of sus- or equitable solutions. Third, commitment problems tainable development, including summary indexes loom large. Health and water services often dete- that try to capture sustainable development in a sin- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 9.1 ing world. They also impede the formulation and Think spatially execution of strategies to combat desertification, other forms of ecosystem degradation, and biodiver- The usefulness of the spatial approach adopted in this Re- sity loss. And they muddy the discussion of the na- port is evident from the map below. The problem, in this case, is the incidence of poverty in Ecuador at varying scales. ture and impacts of global inequality. Fortunately, Pockets of poverty at the canton level are obscured by ag- rapid changes in communication and information gregation, potentially undermining national level responses. technology make it more feasible to gather this infor- mation through a combination of surveys, reports Spatial disaggregation of poverty from ground observers, and information from satel- lites and other sensors. As information costs continue to plummet, the scope for expanding the collection of this kind of information is immense. Innovative information systems can track and deter industrial polluters (as in Indonesia’s PROPER system) and il- legal deforestation and forest fires (as in Mato Grosso’s environmental control system). Although information provision can sometimes be a catalyst for change, the most effective systems are those that create constituencies for information. The con- stituencies may be public agencies that use the infor- Source: Elbers, C., Lanjouw, J. and Lanjouw, P. Forthcoming. ‘Micro-Level Estimation of Poverty and Inequality’, Economet- mation to plan and assess strategies for pollution rica, and Demombynes, G., C. Elbers, J.O. Lanjouw, P. Lan- control, crime prevention, or public health delivery. jouw, J. Mistiaen and B. Ozler (2002), Producing an improved geographic profile of poverty. Methodology and evidence from Or they may be civil society organizations that use three developing countries, WIDER Discussion paper No. information to mobilize dispersed interests. It is de- 2002/39, United Nations University, Helsinki. mand by users that stimulates and maintains the production of relevant, reliable information. This Report proposes an intensive global effort to develop and fund a program to fill data gaps, which gle headline number. If society is to sustain develop- would include: ment, it needs good indicators, but it also needs in- dividuals and groups who demand and use detailed, Ⅲ More local poverty, health, and education data and quality information to solve problems, and oth- linked to national totals ers who produce information to meet that demand. Ⅲ More spatially disaggregated data on economic Summary indicators should rest on a solid founda- and environmental health conditions for the tion of supporting data. analysis of local problems and impacts Environmental and social problems are easy to Ⅲ More coupling of satellite-based remote sensing misdiagnose when there is a lack of reliable, current, data with local “ground-truthing� information, to and geographically disaggregated information. For measure the extent and quality of land under dif- many important aspects of the environment, global ferent types of habitat or land cover conditions or trends are worrisome, but society lacks Ⅲ More monitoring of soil degradation, its causes the detailed data to monitor, diagnose, and manage and impacts the problems at local, national, and global levels. Ⅲ More effort at measuring hydrological conditions. Data for the social sphere are similarly limited. For Ⅲ More investments in geospatial information, such most countries, reliable, up-to-date, spatially disag- as the availability of infrastructure services, and ac- gregated information is lacking on poverty and many curate current maps of road networks other social concerns (health, education, crime). Ⅲ More coordination and augmentation of national These data gaps inhibit understanding of—and efforts to monitor the Millennium Development consensus on—the impacts of policy reforms, na- Goals, to increase comparability, and to provide tional and international, on poverty in the develop- desirable levels of frequency and geographic detail. (c) The International Bank for Reconstruction and Development / The World Bank �������� �� � ����������� ������  Balancing interests age their own compliance processes and help build To equitably and efficiently balance interests within a confidence among parties to forge agreements. society, two elements are necessary: first, getting Ⅲ Independent audits of public programs—especially everyone fairly represented at the bargaining table; programs related to the management of public as- second, facilitating negotiation once everyone is there. sets, such as land, water, forests, minerals, and fish- For many problems at the level of the community or eries and the regulation of pollution—can deter nation—especially those relating to environmental corruption and promote better management of en- and social issues—the responsibility lies with the gov- vironmental resources. Regular audits of national ernment. But governments vary in their capacity to environmental assets can be useful in detecting undertake these tasks. An expanding set of institu- and diagnosing problems. Public ratings of gov- tional tools can assist or complement governments in ernments’ transparency and corruption can affect balancing interests, and they can assist citizens in en- investment and provide a check on government suring that their governments are fair and responsive capture by vested interests. Assessment of the dis- in doing so. These tools become even more important tributional impact of government expenditures at the international level, where they must substitute provides an essential basis for renegotiating them. for government. It is fundamentally difficult to bal- Ⅲ Performance reporting for cities and local govern- ance interests in heterogeneous societies. But taken ments provides a tool for citizens to ensure that together, in a context of increasingly democratic in- governments are responsive to public needs and stitutions and decreasing costs of information and are equitably and efficiently implementing agreed- communication, these tools give some hope. on programs in health, sanitation, water supply, Transparency, performance reporting, and account- public safety, and other areas of public concern. ability. Often the biggest barrier to balancing in- Ⅲ Performance reporting by private firms—financial, terests is an imbalance in the power or influence of environmental, and social—helps society to iden- parties. Devices for accountability—including trans- tify actors with disproportionately large impacts, parency and performance reporting—are useful tools both good and bad, and to understand trade-offs for countering the tendency of entrenched interests and complementarities between economic perfor- to capture institutions or to be unresponsive to dis- mance, and economic, environmental, and social persed or less powerful interests. Providing this kind performance. of information helps level the playing field for nego- tiation, since less powerful interests will typically Forums and networks for negotiation. Governments have poorer access to information. Performance re- and civil society can seek ways to facilitate negotia- porting can also help governments, companies, and tion between affected parties. For instance, Colom- other organizations understand the effectiveness and bia’s regional environmental authorities have set up impacts of their own actions, for instance structured negotiations between water polluters and water users to help determine tolerable levels of pol- Ⅲ National reporting—transparency in fiscal affairs, lution. Participatory budgeting has transformed the reliability of legal institutions, and adequacy of en- budgetary process in more than 80 Brazilian cities. vironmental impact reviews—can all provide the At the international level, the Convention on Long- information and incentives necessary to improve Range Transboundary Air Pollution forums on air governance and the balancing of interests. Exam- pollution and the Global Environment Facility- ples include the recent International Monetary sponsored transboundary diagnostic assessments for Fund Codes of Good Practice on transparency in international waters provide structured means of as- fiscal, monetary, and financial policies, and WTO sessing options. These forums become more effective requirements on subsidy reporting. Another exam- when they build up social capital through dense net- ple is the Aarhus Convention—a voluntary com- works of trust and information linking technical ex- mitment to environmental transparency. National perts, government officials and legislators, civil soci- reporting requirements under the WTO, multina- ety, and special interests. tional environmental agreements, and other inter- Compensation and incentives. Even win-win out- national treaties can help nations track and man- comes usually have a loser—a party whose losses are (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  outbalanced by social gains. Basic principles for bal- Second, careful attention is needed to monitor the ancing interests therefore include minimizing the possibility that complying with certification proce- losses, compensating losers, or providing incentives dures may be too burdensome for small firms. Third, that reconcile private and social objectives. and probably most important, there is a need to en- One way to do this in the environmental sphere sure the integrity of the evolving certification sys- is through market-based permits. Where actors have tems. For these systems to work on a large scale, they different costs of complying with social objectives, will probably require a massive expansion of “mar- trading rights or obligations can substantially reduce kets for honesty.� These are the networks of private compliance costs for individuals and for society as a auditors and certifiers, and their accreditation agen- whole (see box 8.7). Assignment of valuable permit cies, that provide ISO certification, shipping certifi- rights (pollution permits, fishery quotas) is also a cation, and audits of corporate earnings. means of compensating potential losers and ensur- ing participation in a reformed system. When the Implementing potential losers are wealthy interests, however, there Implementing and executing policies and programs may be an unavoidable trade-off between equity and requires appropriate institutional capacity. Mention cooperation. of capacity-building evokes respect and approval— Certification systems are another means of align- but not always excitement or enthusiasm. It has usu- ing private and social incentives. Certification of ally been seen as an add-on to projects and programs firms helps investors identify companies with better with other, more important purposes. It is often environmental and social performance, thus pro- identified with failed technical assistance projects moting incentives for more socially responsible be- that relied heavily on the ephemeral input of foreign havior. Certification of financial institutions (includ- consultants. The same is true of monitoring and ing private banks, bilateral export-import banks, and evaluation, another project add-on that often fails to multilateral development banks) can promote stan- be executed with rigor or to provide much insight. dards for assessing the environmental and social im- Yet institutional competence, especially the ability pacts of investment projects. Certification of prod- to learn, is crucial to efficient use of development in- ucts (wood, coffee, fish, beach resorts, garments) can vestments and to solving the coordination problems allow consumers and investors to reward firms that described in this Report. employ environmentally and socially sustainable Promoting capacity-building and problem-solving production processes. Certification of diamonds and in the developing world. The need to emphasize long- other lootable commodities can be part of a strategy term capacity-building has been recognized for more to avert civil conflict over point-source wealth. Even than a decade.5 Yet, despite technical cooperation governments may find it advantageous to participate grants of $201.3 billion over 1990–2001,6 progress in voluntary certification schemes. For instance, the in institution-building in developing countries has Alaska state government applied for, and received, been disappointing. Most of those funds, to be sure, certification by the Marine Stewardship Council for have been devoted to project preparation. Such in- management of its salmon fishery (see box 7.9). vestments may have been effective in achieving proj- The success of these efforts depends on how they ect goals through reliance on expatriate inputs, but influence producer, consumer, and investor behav- at the cost of forgone opportunities to bolster local ior. And that depends on the costs of certification, capacity. There are failures, too, in the demand for the proportion of the industry that is sensitive to cer- institution-building. Often, lack of ownership, lack tification, and the legitimacy, integrity, and reliabil- of government interest in bolstering capacity, and ity of the certification criteria and process. pressures for rent-seeking have kept competent in- To become more widespread and effective, certi- stitutions from taking root.7 fication systems require coordinated international Several new approaches could help place learning attention on several grounds. First, there are impor- and institutional development at the core of sustain- tant links to the world trade system and a need for able development efforts—that is, projects and pro- coordination with WTO activities—to ensure that grams designed around institutional development, standards are not used to disguise protectionism. rather than vice versa. These approaches recognize (c) The International Bank for Reconstruction and Development / The World Bank �������� �� � ����������� ������  that capacity-building is not accomplished in a few ing countries and in the developed world, to gov- years and thus is not well suited to lending for indi- ernment line agencies in their own countries, and vidual projects or adjustment programs. These ap- to local governments and community organiza- proaches also recognize that capacity has a strong so- tions working on the ground (box 9.2). cial capital component, involving not just people but Ⅲ Scientific research organizations, based in the devel- enduring networks that link policymakers, experts, oping world, that combine local and global exper- civil society, and the private sector. And these ap- tise and help to nurture “policy entrepreneurs� at proaches generate virtuous circles when information home in both scientific and policy worlds. and indicators are made available to groups that de- Ⅲ Learning networks for sharing knowledge, such mand and act on that information. as the Union of Capital Cities of Ibero-America— The new approaches use a variety of new types of real-time, demand-driven learning networks run learning organizations to stimulate both the demand by national and international associations of may- for and supply of institutional skills; they include: ors, local government officials, and city practition- ers share experiences among their members and Ⅲ Think-and-do tanks, policy analytic and action- other cities through the Internet and workshops. oriented organizations that build links (twinning) Ⅲ CSOs for monitoring and evaluating government to universities and think tanks in other develop- and corporate performance. These groups that in- Box 9.2 Problem solving by think-and-do tanks Think-and-do tanks fill the knowledge management need for Ⅲ Brazil’s IMAZON, Amazon Institute of People and the Envi- policy innovation by promoting open interaction between aca- ronment, founded in 1990, a nonprofit research institution, demia, government, business and civil society.* They combine in the Eastern Amazonian city of Belem, is based on the idea rigorous analytical research with practical policy-oriented analy- that the power of scientists to influence patterns of land use sis. Although new research institutes have spread in develop- lies in the appropriateness and quality of the information ing countries, few have the quality, funding, and prestige to in- produced and in their ability to make that information acces- fluence development priorities and policy choices. The best sible to those working on sustainable resource use at re- ones use their capabilities to innovate and solve problems gional, municipal, and community levels. unique to their country. Funding and support Ⅲ Singapore’s Institute for South East Asian Studies, estab- The prerequisites for establishing quality institutions are lished in 1968, describes its philosophy as, “linking people, recognition of need, a strong champion in government, avail- ideas and capital for the purpose of bringing about wide- ability of core funding, and an independent legal status. Gov- spread prosperity.� ernments dedicated to dynamic development and facilitating Ⅲ Malaysia’s Institute of Strategic and International Studies, the transformation of society recognize the need and rely established in 1986, sees its role in a development context, heavily on problem-solving institutes. Funding of sufficient “a country bent on reform and achieving high economic scale and duration determines the ability of an institute to con- growth needs ideas. Some of the best ideas come from the solidate analytical strength, maintain autonomy, and achieve a economically successful countries, and we need to organize reputation for quality advice. Funding includes public and pri- ourselves to learn from these experiences.� vate seed money, often in the form of endowments supple- Ⅲ Korean Development Institute, an autonomous institute set mented by additional grants from individuals, corporations, up in 1971 with government funding through a $15 million philanthropic organizations, donor agencies, and government grant from USAID, reports directly to the president and pro- for particular studies. Institutes that receive all their funding vides independent advice on long and short-term domestic from government must be assured of an independent legal economic policies, and more recently social development. status, reporting to the head of government. Independence is Ⅲ Demos, the United Kingdom’s independent, nonprofit insti- reinforced through the quality of the work, which should be open to scrutiny by the public. The Global Development Net- tute established in 1993 to “solve people’s problems,� has work, founded in 1999, provides funding and other support for an advisory board of successful and innovative business policy research institutions in seven regional networks span- leaders, university professors, notable members of civil so- ning the developing world.† It helps build capacity and re- ciety, and a range of former elected officials. Its strong tech- search expertise by linking these networks of researchers nical reputation gives it good contacts with senior levels of to their counterparts in Europe, North America, and Japan. The government. (Box continues on next page). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 9.2 (continued) Euro-Mediterranean Network also promotes peer reviewed Ⅲ Open access to senior policymakers is critical, but maintain professionalism. Set up in 2000 with EU funding,‡ it groups 97 independence. With autonomy comes responsibility. Make economic institutes from 27 countries around the Mediter- policy recommendations constructive. Garner high-level na- ranean. These networks promote professionalism through tional support based on the quality and creativity of the funding and peer review of research. work. Seek peer review by publishing articles in profes- sional journals as a means of reinforcing quality. Other ingredients for success Ⅲ Stimulate healthy debate of complex issues by being infor- A variety of features are important in ensuring that an institute mative, bringing out the substance behind political pro- produces good analysis, valued by decisionmakers. Some ad- grams, evaluating the costs, benefits, options and choices. vice from the institute directors: Ⅲ Create a team-based approach—no hierarchy—where ex- perts coach, coordinate the work of others, and the director Ⅲ Combine a strong analytical base with a good understand- ensures ultimate quality control. ing of on-the-ground realities. Learn by listening to the peo- Ⅲ Oversee the implementation of your institute’s recommen- ple the authorities are not listening to (the fringes of soci- dations. This provides valuable lessons and continuously im- ety) and by remaining close to the realities of your country. proves your policymaking capabilities. This promotes creativity and relevant policy solutions. Ⅲ Attract the best and brightest nationals back to your country to * Unpublished background note, “Management knowledge and Innova- take advantage of the knowledge they gained abroad and their tion & the Role of ‘Idea Institutes,’� January 2001, World Bank and Al understanding of local culture and traditions to function as cul- Akhawayn University, Ifrane Morocco; Grindle (1997). tural translators. Recruit a good mix of people to bring the nec- † http://www.gdnet.org/ essary breadth of knowledge to solve complex problems. ‡ http://www.femise.org/Presentation/presentation.htm dependently monitor and report on government design projects with learning as a central output, and corporate expenditures and activities have providing hands-on monitoring, evaluation, and im- proliferated recently.8 They can help improve plementation experience to learning organizations. government and corporate accountability, create Expanding the scope of global assessment institutions incentives for monitoring and evaluation within to address emerging issues. At the transnational and government and corporations, and mobilize pub- global level, assessment institutions such as the lic demand for institutional improvements. IPCC have shown their value in forging consensus Ⅲ Mainstreaming monitoring and evaluation functions on the problems and the options for addressing inside government agencies. Responding to both in- them.10 More institutions like the IPCC are needed ternal and public demand for increased efficiency to address the new global problems that continue to and accountability, some governments are improv- emerge all the time. For instance, questions at the ing internal capabilities for monitoring and evalu- intersection of trade and environment are sure to ating programs and projects.9 proliferate, as the discussion of product certification illustrates. Trade policy is an area where more sys- Donors and multilateral development banks can tematic analysis of options might help in forging support the evolution of this intellectual ecosystem agreements. Deepening scientific knowledge exposes of organizations that learn—and apply that learning overlooked transnational environmental processes. to improving policies and projects. Donors can sup- For instance, there is increasing attention to the port these organizations, through direct funding and global nitrogen cycle and its effect on marine ecosys- twinning arrangements on a large scale. These are tems. Technical change, too, brings new problems long-term efforts that bear fruit over a decade or two and risks as well as opportunities. Balancing the en- as the institutions train people, enhance the prestige vironmental risks and benefits of genetically modi- of necessary but neglected professions, such as policy fied organisms is a clear example. And social changes analysis, and build dense networks of trust and require ongoing and forward-looking attention. For knowledge. Funding must be committed over pe- instance, intensifying pressures for international mi- riods much longer than traditional projects, and gration have far-reaching ramifications. In all of funders must accept that the impacts of these in- these areas, there is a strong argument for concerted vestments, though potentially enormous, will be international attention—and for achieving some deferred and difficult to quantify. Donors can also consensus on the relevant issues. (c) The International Bank for Reconstruction and Development / The World Bank �������� �� � ����������� ������  Ensuring greater inclusion But negotiating this great transformation in the Increased voice and major increases in substantive de- next 20 to 50 years requires a renewed commitment mocratization. Inclusiveness can be expanded through by all countries—developed and developing—to this significant changes in governance that increase repre- overarching common vision. The vision is ambi- sentation and accountability, such as empowering tious, but achievable. Many small, poor countries— local government through well-designed decentral- Denmark, Ireland, Japan, Malaysia, Norway, and the ization reforms; electing rather than appointing may- Republic of Korea—have made the leap at different ors (Mexico City); replacing military with elected times from illiteracy and mass poverty to literacy and regimes (Cubatão), or empowering groups excluded affluence within two generations. They were late from decisionmaking—women, indigenous people, industrializers in a global economy already domi- and other disadvantaged groups, who may be in the nated by giants. The European Union shows how majority (see boxes 3.8 and 7.10). the prospect of mutually beneficial integration can Better distribution of access to assets. Dynamic induce poorer countries to adopt higher standards growth and development processes create more assets of environmental and economic management while and new types of assets. It will be much easier to in- the richer ones provide resources and help to boost crease inclusiveness by ensuring that the poor and dis- capacity. The experience of the Dust Bowl in the enfranchised, as well as the middle class, have greater United States shows how small, individual states access to these newly created assets. How? By increas- could not solve their problems without the migra- ing access to education, which build human capital; tion opportunities offered by other states, or the by expanding market-based rural land reforms to in- channeling of knowledge and financial resources crease smallholders’ access to agricultural land and from other states that helped to restore economic complementary assets (water, roads, and knowhow); health over many decades. by expanding the provision of secure tenure (protec- The European Recovery Program (the Marshall tion from arbitrary eviction) in urban slums or other Plan) after World War II showed how mobilizing informal urban settlements; by increasing access to resources on a grand scale can build economies and knowledge (the new asset frontier). Any remaining transform enmity into partnership. The architects of need to improve access to assets by redistributing ex- the Marshall Plan accepted the challenge of tackling isting assets must be based on carefully designed mea- “hunger, poverty, desperation, and chaos� by rebuild- sures that balance interests so that good institutions ing a continent in the interest of political stability, that enable people and assets to thrive can emerge. social development, and a healthy world economy. They had learned the hard lessons of history: the Ongoing dialogue: a global vision and accord Treaty of Versailles ending the First World War in To overcome the barriers to solving collective prob- 1919 had imposed unilateral conditions and enforced lems more rapidly and systematically requires mu- severe reparations on the vanquished, paving the way tual commitments by developing and developed for political extremism. The designers of the Marshall countries to a bold global vision and accord. This vi- Plan avoided these mistakes and paved the way for sion requires a massive and steady effort to eliminate peace. The Treaty of Versailles courted conflict. The poverty and to protect and manage a broader port- Marshall Plan broke a vicious cycle of poverty and re- folio of assets that will ensure the well-being of fu- gret; it supported economic reconstruction and so- ture generations. cial order; and it injected money and ideas to rebuild Europe and herald more than 50 years of unprece- A global vision dented peace, prosperity, and partnership. Today the lessons of history are clearer than ever, for Balancing interests and forging credible commit- instance: ments for the long haul is difficult at the national level but even more so at the global level. Yet it is in- Ⅲ Prosperity and well-being, like peace, are indivisi- creasingly necessary because national action is insuf- ble and must be shared if they are to be maintained. ficient to deal with the scale of spillovers (box 9.3) Ⅲ Two generations—fifty years—are enough to elim- generated by a more interconnected world and global inate poverty and to move to a more sustainable de- economy. So increased global coordination is neces- velopment path. sary to expand the capacity and opportunities of the (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Box 9.3 A big push—to address spillovers and seize opportunities The many global challenges are deeply linked—to each other sions and toxins. Coordination at the international level and and to local concerns. So are their solutions. Managing global actions at the national level are important to pursue syner- spillovers, both environmental and social, and taking advantage gies among goals and avoid unintended consequences. So- of a window of opportunity over the next 20–50 years, will re- cial spillovers, from institutional and environmental weak- quire a big push by global institutions—and by national and ness, also tightly bind the interests of rich and poor nations local institutions. and motivate common interests in sustainable development. There are important biophysical links among the spillovers. Ⅲ Infectious diseases are a global concern. Urbanization and faster travel speed the transmission of disease, increasing Ⅲ Deforestation, an important cause of biodiversity loss, con- the chances of epidemics. The misuse of antibiotics—over- tributes to climate change. Climate change, in turn, puts use in wealthy countries and underuse in poor ones—stim- stress on ecosystem resources, including grazing lands, ulates faster evolution of microbes, against which there is a water resources, and coral reef fisheries that nourish some diminishing supply of fall-back drugs.* And human popula- of the world’s poorest people. It transforms grazing lands tions—particularly the poor, displaced to fragile lands—are to desert. And it threatens cities and coastal populations in closer contact with disease reservoirs in forests and wet- with more storms and flooding. lands and among domestic animals. Ⅲ Poverty alleviation and global growth are linked to biodiver- Ⅲ Poverty contributes to civil conflict and the potential col- sity conservation. In the poorest countries, good governance lapse of the state, with transnational impacts. Poverty and is necessary to protect renewable resources and the people the failure of governance are strong risk factors for civil con- who depend on them—and it is a prerequisite for the ecosys- flict. Nations that experience conflict are at great risk of re- tem management organizations advocated here. Further- lapse. Domestic conflicts often have international repercus- more, vigorous local and global development may pull farm sions, as refugees and violence spill across borders. And populations away from forest lands that are marginal for agri- terrorism and crime take root in a state vacuum. culture but valuable for environmental services. So faster de- Ⅲ Income disparities create strong pressures for migration, velopment and the creation of better institutions may avert and when these pressures meet closed borders, tensions the sacrifice of valuable ecosystems for ephemeral gain. arise. One study estimates that each 10 percent increase in Ⅲ Atmospheric greenhouse gas concentrations cannot be sta- the foreign-to-local wage difference increases emigration by bilized if the poor countries follow the same emissions path 1 per thousand population in African countries.† Historically, as the developed countries—even if the developed coun- such labor movements have been powerful avenues of tries were to cut their emissions to zero. Global sustainabil- poverty alleviation (chapter 4). But strong migration pres- ity requires that all countries work together to pursue long- sures, combined with a growing population and blocked out- term paths to low emissions. But equity, and international lets, lead to tensions and fuel the illegal market in smuggling cooperation, require that greenhouse gases be stabilized people, estimated at $7 billion a year.‡ without jeopardizing the development aspirations of poor countries. Ⅲ Trade and other internationally negotiated policies can affect * WHO (2001). poverty, biodiversity loss, greenhouse gas emissions, and † Hatton and Williamson (2001). other global environmental spillovers such as nitrogen emis- ‡ NIC (2000). weaker segments of the global community. A self- gender equity, and ensuring environmental sustain- enforcing global accord may be required to get the ability (box 9.4). One calculation puts the cost of commitment to finance such a scaled-up effort, to meeting just the nonenvironmental targets at $40 to build capacity to use the funds wisely, and to take on $70 billion a year above the current $50 billion in difficult reforms in developing as well as industrial development assistance. Another estimate puts the countries. cost of reaching the environmental goals over a longer period at $25 billion per year.11 A global accord The recent International Conference on Financ- There is growing recognition of the need for mutual ing for Development in Monterrey also confirmed commitments and for accelerated improvements in the need for more aid, trade, and debt forgiveness key development indicators. Support for the Millen- by industrial countries, in tandem with domestic re- nium Development Goals, which propose to cut the forms in developing countries to increase domestic proportion of people in extreme poverty by half by resource mobilization and facilitate foreign direct in- 2015 is now widespread. The goals set ambitious vestment (box 9.5). quantitative targets for reducing hunger, increasing The recent proposal for the New Partnership for primary schooling, improving health, promoting Africa’s Development (NEPAD) is also based on an (c) The International Bank for Reconstruction and Development / The World Bank �������� �� � ����������� ������  Box 9.4 Box 9.5 Millennium Development Goals (1990–2015) Outcome of the International Conference on Financing for Development, Monterrey, Mexico 1. Eradicate extreme poverty and hunger Ⅲ Halve the proportion of people with less than $1 a Ⅲ Mobilize domestic financial resources for develop- day. ment—by improving governance, macroeconomic poli- Ⅲ Halve the proportion of people who suffer from cies, and social safety nets. hunger. Ⅲ Mobilize foreign direct investment and other private 2. Achieve universal primary education flows—by improving the climate for business. Ⅲ Ensure that boys and girls alike complete primary Ⅲ Make international trade an engine for growth and de- schooling. velopment—by engaging in a true development round. 3. Promote gender equality and empower women Ⅲ Increase international financial cooperation for develop- Ⅲ Eliminate gender disparity at all levels of education. ment—by doubling official development assistance and focusing it effectively in the most needy. 4. Reduce child mortality Ⅲ Provide sustainable debt financing and external debt Ⅲ Reduce by two-thirds the under-five mortality rate. relief—by matching financing needs and repayment 5. Improve maternal health capacities. Ⅲ Reduce by three-quarters the maternal mortality ratio. Ⅲ Address systematic issues—by enhancing the coher- 6. Combat HIV/AIDS, malaria and other diseases ence and consistency of the international monetary, fi- Ⅲ Reverse the spread of HIV/AIDS nancial, and trading systems. 7. Ensure environmental sustainability Ⅲ Integrate sustainable development into country poli- Source: International Conference on Financing for Development, cies and reverse loss of environmental resources. Monterrey, Mexico, (March 2002). Ⅲ Halve the proportion of people without access to potable water. Ⅲ Significantly improve the lives of at least 100 million unsustainable to sustainable paths over the next 50 slum dwellers. 8. Develop a global partnership for development years. Many of these long-horizon initiatives will Ⅲ Raise official development assistance. yield benefits in the medium term in support of the Ⅲ Expand market access. Millennium Development Goals. But because many Ⅲ Encourage debt sustainability. future problems cannot be foreseen, it is important that institutional foundations be strong and that a Source: www.developmentgoals.org process and a framework be developed that are ro- bust in picking up new signals, balancing a broader range of interests, and maintaining commitments to arrangement in which developing countries take re- the global vision. sponsibility for improved governance, and industrial As noted in chapter 1, at a modest 3 percent an- countries help through more aid, debt forgiveness, nual rate of growth, the global economy in 50 years and market access. The Council of the European will be four times the size it is now. Will that larger Union’s proposed Global Deal has many of the same economy generate less environmental and social elements.12 stress than the much smaller economy does today? In the spirit of these initiatives and to maintain Most of the physical capital required for the econ- the momentum of the Millennium Development omy 50 years hence has not yet been created. This Goals beyond 2015, this Report calls for extending provides an opportunity to incorporate inclusiveness the following goals: and sustainability criteria in new investments now. The potential is there to shift development paths, Ⅲ To fully eliminate global poverty, and provided institutions that adapt and implement bet- Ⅲ To put the global economy on a more sustainable ter policies can be put in place. development path by the middle of this century. For development strategies and development as- sistance, this means placing a greater emphasis on: The two features added to existing initiatives are a deeper target over a longer time horizon, and a Ⅲ Identifying vicious circles that keep the pace of greater focus on institutional development. It will growth low and the distribution of assets un- require 10 or 20 years—starting now—to build up equal—and developing strategic interventions to the institutions that can help shift trajectories from break these vicious circles. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Ⅲ Investing in projects, programs, and initiatives resources that can to be freed up say, by eliminat- that bring about better, more inclusive institutions ing perverse subsidies in industrialized and devel- and ensure systematic learning. oping countries and redirecting them to support institutional building and the investment require- Greater inclusion, better information flows, more ments of a shift to a more sustainable path. How- transparency, and wider forums for balancing inter- ever, a serious data and analytic effort is needed to ests will help to improve the functioning of global confirm this information at greater levels of detail, and local institutions to fight poverty and promote country by country. sustainability. Investments in global poverty reduc- Ⅲ Undertaking difficult reforms for both the devel- tion and in greater inclusiveness will help ensure the oped and the developing world. representation of all interests in the design of the new and improved institutions. Main responsibilities of developing countries There is no understating the difficulty of these A development strategy that emphasizes inclusive- challenges. Nor is there an easy solution. Social in- ness, shared growth, and better governance places ertia is great, and institutional change can take de- large demands on leaders in developing countries. cades. Overcoming the inertia to tackle these difficult They must commit to better economic, social, and problems—the fears and risks of unilateral action— environmental management—and thus to better requires coordination. The Report suggests mutually governance. To manage their resources, and what reinforcing ways to catalyze institutional change, to they receive from outside, they need to: mobilize dispersed constituencies, and to support ca- pacity development. The core components of a glo- Ⅲ Strengthen institutions. The rule of law and good bal accord include: governance allow families and firms to have con- fidence—in other words, to save and invest. Ⅲ Building capacity to use resources wisely. It is now well Ⅲ Broaden inclusiveness in the access to assets. School- established that the effectiveness of development ing, health care, and provision of environmental assistance—indeed, of all investment—depends assets that protect health, market-based rural land greatly on the quality of economic policies and the reform, and regularization of urban tenure (pro- reliability and capability of market and nonmarket viding protection from eviction without due legal institutions.13 But there is a lot more to building process) all promote asset generation for the poor. capacity than technical assistance, as discussed in Ⅲ Increase transparency. An open and verifiable flow this chapter’s earlier section titled “Implementing.� of information is important to tighten accountabil- Ⅲ Providing the necessary funding. Capital markets ity in government and the private sector through (foreign direct investment and private financing) such steps as opening procedures for bidding, can cover much of the funding required to shift to strengthening meritocracy in the civil service, and a more sustainable path if appropriate policies are making sure that public and corporate budgeting in place. For example, they can cover investment and resource management are governed by law, in new and replacement capital (buildings and open to the public, and under proper oversight equipment) to improve energy efficiency and meet institutions. the demands of an urban population that will double. But expanded domestic resources and de- The success of the reforms would be long term; velopment assistance will be needed to cover the they would secure opportunities and voice for fami- part of these costs that involve the provision of lies and their children so they can save, invest, and local, national, and global public goods. Institu- engage in their communities. But good policies, to tion building is one of these public goods. The es- be sustained, will require committed support and the timated funding requirements for the Millennium legacy of reversals and stalemates is stunning. Re- Development Goals, would not be enough to sup- search on aid and policy shows that aid is not worth port a broader and deeper agenda of institution- much without good policies and institutions; in fact, building. There are some rough estimates14 of the it can even be harmful. Research also shows that good (c) The International Bank for Reconstruction and Development / The World Bank �������� �� � ����������� ������  policies and institutions are essential to growth and vaccines, and adaptation strategies that would im- development. This Report has argued that often but prove the livelihood for poor people. not always good policies presuppose good institu- tions, and these take time to evolve. Funds and assis- Joint responsibilities of developing tance will not be available without the conviction and developed countries that there is capacity to use them effectively. But Together the developed and developing countries capacity-building requires patient investment to re- must address the most urgent problems facing hu- move critical barriers because its payoffs are large but manity. Their joint responsibility is to establish a take time to be realized. Fundamental reforms require global partnership to set the rules for making rules and better institutions, which evolve slowly. These re- the modalities of burden sharing. As the world be- quirements are interlinked. So developing country comes more interconnected—environmentally, eco- leaders need to know that they will have long-term, nomically, socially—new institutions and rules must reliable support from the larger development com- be agreed on and implemented. These will include munity. Without such support—and the quid pro rules for international trade; rules to avert conflict; quo on reform to support it—many developing rules on migration; rules governing the use of the countries will remain in cycles of promise and disap- biosphere; and rules affecting property rights in pointment. With support, these cycles can be broken. ideas, technological processes, and genetic informa- tion. The consequences will be enduring. If these in- Main responsibilities of developed countries stitutions are to be effective, the rules for making Enabling developing countries to develop more rap- rules have to be fair—in process and in outcome. idly through increased aid, trade, migration, and ac- There is a role for all actors in the global system: cess to knowledge and technology will place big de- Governments of developing, transition, and developed mands on leaders and voters in developed countries. nations; provinces, cities, and local communities; civil The actions required of them include: society organizations; private firms; individuals. Ⅲ Increase aid and make it more effective. Developed Ⅲ All governments can improve the accountability countries should strengthen the ability of develop- of public agencies and the provision of informa- ing countries to pursue sustainable development tion about social and environmental conditions— by providing development assistance that supports to improve the ability of the general public and public goods and attracts private investment.15 civic groups to identify problems, balance inter- Ⅲ Reduce debt. This has started under the Heavily ests fairly, and come up with solutions. Indebted Poor Countries Debt (HIPC) Initiative, Ⅲ Civil society organizations can help to aggregate and it is essential to go farther for all developing the voices of dispersed interests and provide inde- countries by agreeing on poverty reduction strate- pendent verification of public, private, and non- gies and improving accountability. governmental performance. Academia needs to be Ⅲ Open agricultural, industrial, and labor markets. recognized as a key actor in learning, monitoring, Developed country trade barriers impede exports and evaluating. from developing countries and undermine the Ⅲ The private sector can advance economic, social livelihoods of the poor. Unrestricted access to de- and environmental objectives by helping to con- veloped country markets in textiles and clothing struct a framework that provides appropriate could yield $9 billion a year, and access to agricul- incentives for firms to be accountable in all three tural markets $11.6 billion a year.16 dimensions. Ⅲ Improve developing country access to technology and knowledge. Implement incentives to promote the If the global community sees merit in such an ac- transfer and dissemination of technologies to de- cord, the accord’s elements will need more careful veloping countries—including those for climate work over the next few years to develop an imple- mitigation, disease prevention, and agricultural mentable program that can adjust to contingencies development. Support more research on crops, without undermining the promise of the accord. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Ongoing dialogue: some open questions shift resources from competitive individual con- To make more headway on the accord and to define sumption to consumption of public goods. But these a process and framework that is “fair� will require externalities need to be much better understood be- dealing with some global issues of sustainable devel- fore there can be any agreement on the actions to ad- opment that remain the subject of heated debate. dress them. Mentioned here are four important and controver- sial topics whose resolution has important policy and What is the future of agriculture and genetically institutional implications, requiring credible global modified organisms? assessments. Despite great promise for improving the agriculture of the poor, biotechnology in general and transgenics When is consumption overconsumption? research in particular have barely begun to address Concern is often expressed about “overconsump- the problems of the poor. Some applications gener- tion� in wealthy countries and about the threats ate little controversy, such as marker-assisted genetic to sustainability of increasing levels of global con- selection. Others, such as the creation of transgenic sumption. But what kind of consumption qualifies organisms, have generated much concern about food as overconsumption, why is it harmful, and what safety and potential environmental impacts. should be done about it? Does overconsumption Comfort with the new technology is determined imply that there should be a limit on total global con- in large measure by societies’ comfort with their sci- sumption (and that as a result, the already high levels entific and food safety institutions and their feelings of consumption in developed countries need to be about emerging concentrations of economic power reduced to enable increased consumption in poor in multinational “life-sciences� corporations. Solu- countries)? On these questions there is little clarity. tions to these complex issues are all playing out One interpretation of overconsumption is that it against a backdrop of globalization-related uncer- refers to the environmental externalities associated tainty, which has left many people unsettled over with consumption at higher levels of per capita in- their capacity to control their lives and their envi- come. For example, carbon dioxide emissions, and ronment. It is the rural poor in developing countries their contribution to climate change, are highly cor- who most need access to these new agricultural tech- related with consumption of electricity, home heat- nologies. The precautionary principle tells us that ing, transport services, and energy-intensive manu- we should err on the side of caution, look at alterna- factured goods—all of which tend to increase strongly tives, and ensure a fully transparent and democratic with income. In these cases, the over prefix is justi- process. This requires more clearly sorting what is fied, since the externalities are by definition ineffi- known from current science from what is not, so cient (there is no balancing of costs against benefits) that the political process can act more effectively. and usually inequitable (wealthier people impose the damages upon poorer people). But the overall level How to balance interests and avoid the race for of consumption is not the source of the problem. It property rights at the intellectual frontier? is the combination of the specific consumption mix Intellectual property rights (IPRs) is the next genera- and the production processes that generates the ex- tion of assets that can increase or decrease inclusive- ternality. And for these there are well-established pol- ness with consequences for the evolution of quality in- icy prescriptions from public finance. stitutions. IPRs represent a compromise between the Another interpretation of overconsumption, much interests of users, owners, and creators. It often costs more difficult to document, has to do with social ex- very little to duplicate a seed, a computer program, a ternalities. People judge the adequacy of their con- song, a drug, a blueprint for a printed circuit or a me- sumption—clothing, automobiles, housing—in part teorological database. Once these products have been against norms set by others. If this is true, consump- created, their widespread dissemination would bring tion takes on some of the aspects of an arms race. great consumer benefits. So why not make them avail- What are the policy implications? Mutual restraint able for only the cost of reproduction? Because there is needed (a coordination problem par excellence) to would then be no incentive for private actors to create (c) The International Bank for Reconstruction and Development / The World Bank �������� �� � ����������� ������  the information and innovation behind these prod- resistant to automation, such as care of the elderly. ucts. Intellectual property rights such as patents and At the same time, costs of migration will decline— copyrights, balance these static and dynamic aspects especially information costs, but also transport. In of efficiency. sum, there’s likely to be a drastically greater supply The Trade-Related Aspects of Intellectual Prop- of, and demand for, international migrants over the erty Rights (TRIPS) agreement under the WTO next half century. represents a global strengthening of the rights of pro- Dealing with this pressure is a global challenge. ducers in terms of users. Its immediate effect will be There are collective decisions to be made, and every to increase royalty payments to intellectual property option has costs and benefits. There are many reasons rights holders, who are overwhelmingly in the devel- to support both more long-term as well as circular mi- oped world. One estimate found that full applica- gration (the latter can help speed up learning in lag- tion of TRIPS would increase annual net patent ging regions), but migration remains a politically sen- rents to the United States, Germany, and Japan alone sitive issue in receiving countries. There are issues by $31 billion.17 related to assimilation of newcomers—as well as ex- Will the emerging intellectual property rights aggerated fears and misconceptions. The decision regime be detrimental to the long-run interests of of one potential receiving country to restrict immi- developing countries? The issue is hotly debated. In gration has implications for other receiving countries, principle, TRIPS provides a wide latitude for a de- and for the sending countries. Stresses associated with veloping country to fine-tune an intellectual prop- immigration may be related more to the rate of change erty rights system appropriate to its needs.18 In prac- than to the level of change. Assimilation processes for tice, developing countries’ ability to maneuver may immigrants are best measured in decades rather than be more limited, and the potential for unequal out- years. Advance preparation in both sending and re- comes is worrisome. While the outlines of the global ceiving societies over the next generation could yield intellectual property rights regime are clear, many a much preferred world in 2050 compared with one crucial details may not yet be established. Global dis- in which no foresight is exercised. cussions might address ways for developing coun- tries to strengthen (a) collaborative efforts at patent *** examination, (b) standards for the breadth and nov- This Report argues that the lack of assets, oppor- elty of patent claims, (c) protection of rights to ge- tunity, and effective voice for large segments of the netic resources and traditional knowledge, (d) global population blocks the emergence of general welfare- competition policy, and (e) the rationale for public enhancing policies, impedes growth, and under- funding and dissemination of census data, environ- mines the potential for positive change. At the na- mental and meteorological data, and genetic data. tional level, it robs us of the talents of those left out Strengthening the capacity of developing countries in society. And at the international level, it deprives to participate in these discussions might also lead to us of the contribution poor countries can make to a more equitable outcomes. more just and sustainable future. A more sustainable development path is more socially inclusive; it en- What are the prospects for global migration? ables societies to transform and solve collective ac- Global inequality, combined with global demo- tion problems. The challenge, now and in the future, graphic trends, will create more pressure for migra- is to develop the courage and commitment to man- tion. An extensive literature shows that wage differ- age the processes that underpin human life and well- entials drive migration. Fertility rates are highest in being and to bring about a transformation that im- places that have the least capacity for absorbing proves the quality of the environment, strengthens labor. Meanwhile, aging in the developed world will our social fabric, and enhances the quality of peo- drastically reduce the size of the labor force and in- ple’s lives. The more people heard, the less assets crease the demand for low-skilled labor for tasks wasted. (c) The International Bank for Reconstruction and Development / The World Bank Bibliographic note Jorgensen, David Kaimowitz, Hirochi Kawashima, Phil This Report draws on a wide range of World Bank docu- Keefer, Charles Kenny, Homi Kharas, John Kellenberg, ments and on numerous outside sources. Background pa- Elizabeth King, Kenneth King, Nalin Kishor, Agi Kiss, pers and notes were prepared by Sherburne Abbott, G. Stephen Knack, Somik Lall, Manuel Lantin, Frannie Acharya, Alain Bertaud, José Brakarz, Kjell Arne Brekke, Leautier, Franck Michel Lecocq, Johannes Linn, Josef Tim Campbell, Roberto Chavez, Monica das Gupta, John Lloyd Leitmann, Andres Liebenthal, Stephen Malpezzi, Dixon, Scott Gates, Nils Petter Gleditsch, Rognvaldur Robin Mearns, Gerhard Menkhoff, Fatema Mernissi, Alan Hannesson, Karla Hoff, Pernille Holtedahl, Eckard Janeba, Miller, Pradeep Mitra, Augusta Molnar, Caroline Moser, John Kellenberg, Stein Kuhnle, Huck-ju Kwon, Desmond Desmond McNeill, Mohan Munasinghe, Mustapha McNeill, Edgar Ortiz Mtialavasi, S. Mansoob Murshed, Kamel Nabli, Aksel Naerstad, Andrew Nelson, Martien Eric Neumayer, Jelena Pantelic, Sanjeev Prakash, Jane Pratt, van Nieuwkoop, Letitia Obeng, Alexandra Ortiz, Edgar Per Selle, Guttorm Schjeldrup, Haakon Vennemo, Nicolas Ortiz, Amy Nolan Osborn, Elinor Ostrom, Mead Over, Sambanis, Paul Steinberg, Arne Tesli, and Ahmed Zainabi. Stefan P. Pagiola, Guillermo Perry, Guy Pfefferman, Background papers for the report are available either Robert Picciotto, Robert Prescott-Allen, Lant Pritchett, on the World Wide Web http://econ.worldbank.org/ Felicity Proctor, C. Sanjivi Rajasingham, Vijayendra Rao, wdr/wdr2003/ or through the World Development Re- John Redwood, Francisco Reifschneider, Ritva Reinikka, port office. The views expressed in these papers are not Felix Remy, Jozef Ritzen, F. Halsey Rogers, David Rosen- necessarily those of the World Bank or of this Report. blatt, Michael L. Ross, Ina-Marlene Ruthenberg, Maria Many people, both inside and outside the World Bank, Sarraf, David Satterthwaite, Sara J. Scherr, Richard Scur- gave comments to the team. Valuable comments and con- field, Louis Scura, Luis Serven, Cosma Shalizi, Priya tributions were provided by Herbert Acquay, Sadiq Shyamsundar, David Simpson, Anil Sood, Lyn Squire, Ahmed, Eleodoro O. Mayorga Alba, Mir A. Altaf, Ali Andrew Steer, Vivek Suri, Lee Summer Travers, Timothy Amahan, Ivar Andersen, Jock Anderson, Shlomo Angel, S. Thomas, Tom Tietenberg, Jane Toll, Thomas Tomich, William Ascher, Robert Bacon, Deniz Baharoglu, Tulio John Underwood, Keshav Varma, Haakon Vennemo, Barbosa, Scott Barrett, Carl Bartone, Richard Barrows, David G. Victor, Jeffrey Vincent, Tara Vishwanath, Tamsyn Barton, Esra Bennathan, Alain Bertaud, Derek Joachim von Amsberg, Michael Walton, Hua Wang, Beyerlee, Anthony Bigio, Hans Binswanger, Slyvain Bis- Robert Watson, Jaime Webbe, Monika Weber-Fahr, arre, James Bond, Maria Borda, Milan Brahmbhatt, José Brakarz, Marjory-Anne Bromhead, Dan Bromley, Lester Anna Wellenstein, David Wheeler, Anthony J. Whitten, Brown, Piet Buys, Tim Campbell, Franklin Cardy, David Jeff Williamson, Roland White, Julie Viloria-Williams, S. Cassells, Christophe Chamley, Nadereh Chamlou, Michael Woolcock, Sven Wunder, and Hania Zlotnik. Roberto Chavez, Ajay Chhibber, Tanzib Chowdhury, Other valuable assistance was provided by Trinidad Hoon Mok Chung, Dean Cira, Kevin Cleaver, Robert Angeles, Carey Ann Cadman, Meta de Coquereaumont, Clement-Jones, William Cobbett, Paul Collier, Maureen Jean-Pierre S. Djomalieu, Kristyn Ebro, Ines Garcia- Cropper, Csabi Csabi, Partha Dasgupta, Laura De Brular, Thoumi, John Garrison, Anita Gordon, Rita Hilton, Ser- Klaus Deininger, Shanta Devarajan, Simeon Djankov, gio Jellinek, Lawrence MacDonald, Nacer Mohamed David Dollar, Ahmed Eiweida, Enos Esikuri, David Eller- Megherbi, Joyce Msuya, Jean Gray Ponchamni, William man, Antonio Estache, Ke Fang, Marianne Fay, Peter Reuben, Carolyn Reynolds, Roula Yazigi. Fallon, Shahrokh Fardoust, Catherine Farvacque, John The team wishes to thank individuals who partici- Flora, David Freestone, Maria Emilia Freire, Alan Gelb, pated in the following events during the planning and Alan Gilbert, Gershon Feder, Robert Frank, Scott Gates, drafting stages of this Report. The participants in these Hafez Ghanem, Indermit Gill, Sumila Gulyani, Kenneth workshops and video conferences are listed in the Web Gwilliam, Agi Kiss, Kirk Hamilton, Jeff Hammer, David Page. Berlin, Brussels, China, Costa Rica, London, Hanrahan, Rognvaldur Hamnesson, Jarle Harstad, Marea New York, Nigeria, Oslo, Paris, South Africa, Viet- Eleni Hatziolos, John Henderson, Jesko Hentschel, Ver- nam, and Washington. non Henderson, Mark Hildebrand, Rafik Fatehali Hirji, Despite efforts to compile a comprehensive list, some Daniel Hoornweg, J. C. Hourcade, Sarwat Hussain, Gre- who contributed may have been inadvertently omitted. gory Ingram, William Jaeger, Vijay Jagannathan, Em- The team apologizes for any oversights and reiterates its manuel Jimenez, Todd M. Johnson, Olga Jonas, Steen Lau gratitude to all who contributed to this Report.  (c) The International Bank for Reconstruction and Development / The World Bank ��������  Endnotes 24. Until recently, the carbon emissions generated by energy-intensive activities (that rely on fossil fuels, such as Chapter 1 coal) did not affect global temperatures because they had not 1. World Bank 2001h CD Rom (SIMA 349). exceeded the bioshere’s absorptive capacity. Now more ex- 2. Chen and Ravallion (2000). pensive alternatives are needed to avoid further damage. 3. World Bank 2001h CD Rom (SIMA 349). 25. Dasgupta (2002). 4. World Bank 2001h CD Rom (SIMA 349). 26. Yi (2002). 27. Bloom and Williamson (1997). 5. World Bank 2001h CD Rom (SIMA 349). 28. Much like the dynamics by which teams become 6. The quality and coverage of the household survey data more creative, populations moving to cities go through stages used to measure poverty have improved dramatically in the of forming, storming, norming, and performing. Forming past 10 to 15 years, and the World Bank has played an im- occurs when individuals with different backgrounds come portant role in facilitating this improvement. Since 1990, the together; storming, when their different perspectives clash; Bank’s $1 per day poverty estimates have drawn fully on norming, when more inclusive norms evolve; and performing, these new data. However, the paucity of adequate survey data when constructive behavior replaces destructive behavior. for the past naturally makes estimation over longer periods The result is that cities, in the best cases, become centers more hazardous. In Globalization, Growth and Poverty (World where different cultural values come together and jointly de- Bank 2002g), it was estimated that the number of people liv- velop more inclusive values to accommodate different per- ing below $1 per day had fallen by 200 million between spectives and provide space for different subgroups to spe- 1980 and 1998. As noted in the Report, that estimate had cialize and innovate. to draw on two different sources that used different meth- 29. The complete series for developing and for high- ods. Further checks using more consistent methods corrobo- income countries for 1950–2050 were created using vari- rate the earlier estimate. These estimates also suggest that if ous interpolations and extrapolations of existing data while China were excluded, there would have been little or no net maintaining consistency with available World Bank and decline in the total number of poor people. U.N. control totals. Estimates for size classes of 100,000 7. In 1978 China abandoned its reliance on collective population and more were made using the following sources: agriculture, sharply increased the prices paid for agricultural U.N. and World Bank control totals for urban population goods, and dramatically increased the role of market signals in developing and high-income countries; U.N., World Ur- and foreign investment. banization Prospects, 1999 Revision, digital files from the 8. Brown and others (2001). World Watch Institute’s es- U.N. Population Division; and the database of cities above timates based on national-level surveys of body weight by the 100,000 population compiled for the U.N.-HABITAT Suc- United Nations (U.N.) and the World Health Organization cessful Cities project. (World Bank projections for urban (WHO). population are lower than those of the U.N. and closer 9. World Bank 2001i. to those of the International Institute for Applied Systems 10. World Bank 2001h; Sambanis (2000, p. 13). Analysis, as they assume a slower growth rate for most coun- 11. UNDP, UNEP, and others (1999). tries.) The populations of smaller towns (those with popula- tions of less than 100,000) were calculated as the residual of 12. UNDP, UNEP, and others (1999). total urban population as indicated by U.N. and World 13. UNDP, UNEP, and others (1999). Bank sources minus the total estimated population of cities 14. Available at World Bank Group, “Access to Safe Wa- larger than 100,000. ter,� (2000). 31. Henderson, Shalizi, and Venables (2001). 15. UNEP (1997b); Scherr (1999); Scherr and Yadav 32. Meaningfully evaluating the consequences or proba- (1996); White, Murray and others (2002); Cosgrove and Ri- bilities of outcomes, tradeoffs, and priorities becomes diffi- jsberman (2000). cult, if not impossible, without the appropriate data and in- 16. World Bank (2001c). formation. For environmental and social variables, there may 17. UNDP, UNEP, and others (1999). be some time-series data at the local level, but there are rarely 18. Myers, Mittermeier, and others (2000). equivalent disaggregated data for GDP variables. At the na- 19. UNDP, UNEP, and others (1999). tional level, the situation is often reversed. This obstructs any 20. UNDP, UNEP, and others (1999). attempt to quantitatively model or assess changes over time 21. World Bank (2001c). or their determinants. This Report relies heavily on case 22. Social change and cultural evolution have also been studies that are thought to be representative. speeding up, but not uniformly within or across societies. Some cultures are less able to adapt to speed of change even Chapter 2 if they wanted to, while others may not even want to. 1. Utility or well-being has always been an inherently un- 23. Inconsistencies between human and natural processes measurable concept. Despite the limitations, measures of self- manifest themselves spatially (location-specific sources and reported happiness or overall satisfaction with life have some- sinks) and at different scales. times been used as a proxy. See Oswald (1997). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  2. The importance of social participation to human well- the nature of the analysis, to draw finer distinctions within being is reflected, for instance, in the negative correlation be- the categories mentioned here—for instance, between physi- tween people’s self-reported happiness and unemployment, cal and financial assets within human-made assets. controlling for income (Blanchflower and Oswald 2000). 11. The waste from other species is usually more easily Consistent with the happiness data, suicidal behavior is also biodegradable in natural processes. more prevalent among the unemployed (Oswald 1997). The 12. Trust is usually accumulated through repeated inter- importance of the environment is reflected, for example, in actions. the results of a survey of over 35,000 people in the G-20 13. Interpersonal networks can be thought of as assets, countries, in which about one in four citizens spontaneously but networks that fail to give rise to mutual beliefs that sus- mentions environmental issues as a major concern facing his tain good outcomes are not socially productive. or her country. Although the focus of the environmental 14. Although there is still a great deal of debate concern- concern varies, in part by the country’s level of development ing social capital (centering on what constitutes social capi- (in Asia, for example, people are more concerned about the tal, whether the word capital should be used in this context effect of pollution on human health, whereas elsewhere peo- at all, and on how to analyze and measure the effects of so- ple in the G-20 seem equally concerned about the effect of cial capital), there is a growing body of evidence that shared pollution on human health and the loss of natural resources), values, informal ties, and interpersonal networks can have an the percentage of people citing water pollution, air pollution, important impact on outcomes—notwithstanding some of and loss of natural areas and species was high—ranging from the difficulties in measuring social capital. See Grootaert and 63 percent (for species loss) to 71 percent (for water pollu- van Bastelaer (2001) for different definitions of what consti- tion). See: Environics International. (see International Envi- tutes social capital; Solow (2000) and Bowles and Gintis ronmental Monitor Survey Oct–Dec 2001 on line at www. 1999) for examples of objections to the use of the term “cap- environicsinternational.com). ital�; and Stone (2001) on measurement problems in empir- 3. Dasgupta (2001a) has shown that along any arbitrary ical work. In the social capital literature, institutions and or- consumption path, the aggregate present value (discounted ganizations are included under the concept of structural integral) of utility increases during a short interval of time if social capital, while another set of elements (trust, shared val- and only if wealth (estimated at shadow prices defined along ues, norms) are included as cognitive social capital. This Re- the consumption path) increases during the interval, at con- port reserves the term social capital for interpersonal net- stant shadow prices. See also Hamilton (2000). works and the trust and shared values they generate, but 4. Adjusted net savings are also referred to as genuine sav- excludes norms, which function as informal rules. Instead, ings in the literature. as in the institutional economics literature, we separate the 5. Whether social assets should be included in a measure term “institutions� from social capital and use institutions for of net wealth and savings is still an issue of debate. See note the mechanisms or rules of the game (both formal and infor- 14. 6. The adjustment for CO2 emissions reflects the damage mal) that determine how individuals and groups interact, co- to global assets associated with economic activity in a given ordinate, and allocate resources. (Although as chapter 3 country. If we assume certain property rights, in particular shows, social capital narrowly defined as cognitive social cap- that each country has the right not to be damaged by CO2 ital and institutions broadly defined to include embedded or- emissions from its neighbors, then the estimate of global ganizations share critical asset-like characteristics that under- damage represents the sum of a) the damage a country’s CO2 pin a society’s ability to put other assets to good use.) This emissions does to its own assets over time and b) the notional report also includes embedded organizations as institutions damage payments owing to all other countries affected by (but not organizations as agencies). these emissions. 15. Natural assets can also have “intrinsic� values. Ecosys- 7. The effects of improved life expectancy, which have a tems preceded the evolution of humans and can function direct bearing on human assets, are also not included. without humans, but humans cannot survive without ecosys- 8. After accounting for population growth Dasgupta tems. However, in that humans are increasingly acquiring (2001b) finds that changes in wealth (adjusted net savings the knowledge and technology to preserve or destroy ecosys- per capita) are negative in the Indian subcontinent and tems, the report focuses on instrumental values. Africa but positive in China. 16. In the more narrow economic definition, two inputs 9. Some NGOs have begun experimenting with Internet- in the production function are gross complements if their based information provision as a way of stimulating debate cross-price elasticity of demand (the extent to which the de- about public policy; increasing the transparency of the ac- mand for one input rises when the price of the other input tions of private corporations, public agencies, and legislators; increases) is positive. Where markets do not exist, the price and facilitating feedback from civil society to elected officials. elasticity of demand cannot be determined, even though tech- 10. Assets can be classified in many different ways de- nically the assets are complementary. We use this broader pending on the analytic purposes at hand, and the list pre- concept throughout. sented in this chapter is by no means an exhaustive one. For 17. The examples given in the text are at a micro level. A instance, cultural assets can also be important in affecting more macro-level example is one by Knack and Keefer human well-being. It may also be appropriate, depending on (1997), which considers the role of social capital using mea- (c) The International Bank for Reconstruction and Development / The World Bank ��������  sures of trust and civic norms and shows a positive associa- ized by increased stocking rates and requiring corresponding tion with growth. feed and water management inputs. 18. Both social capital and informal and formal institu- 36. Primavera (1994). tions can lower transaction costs and increase efficiency. In 37. Note that the economic losses associated with the facilitating innovation, they can also increase productivity mangrove destruction that often accompanies shrimp farm- growth. Hence good social capital and institutions are cru- ing is not considered here. cial to economic growth. 38. When the industry is highly developed, specialization 19. Krishna and Uphoff (1999). includes producers of farm equipment, algal feeds, formulated 20. Reid and Salmen (2000). feeds, spanners, and services. Hence, it can involve many jobs 21. Galor and Zeira (1993). and large amounts of capital equipment. For example, the 22. Uzzi (1997). Ecuadorian shrimp industry in 1990 had a total capitaliza- 23. Yli-Renko (1999). tion of $1.66 million and employed around 100,000 people. 24. See Suvanto (2000) for a discussion of how social cap- Similarly, in Thailand around 114,000 people were employed ital enhances innovation and accelerates product develop- in 19,000 shrimp farms in 1991, and in India shrimp pro- ment in firms. cessing plans employed some 500,000 people (Primavera 25. Zaheer, McEvily, and Perrone (1998). 1994). 26. A measure that captures both the shortening of life 39. See Dasgupta (2000). and the amount of time lived in varying degrees of dysfunc- 40. Because ecosystems are integrated systems, parceling tion because of illness is the disability adjusted life year an ecosystem into different parts that are then privatized can (DALY). Estimates (Murray and Lopez 1996) suggest that create problems since each individual owner may allocate his environmental “bads� account for a significant proportion— land to a different use without regard to the needs of the ranging from 10–20 percent of total DALYs—of the total ecosystem as a whole, resulting in a loss of biodiversity and burden of disease across developing country regions. ultimately the resilience of the ecosystem itself. 27. Clark (1898). 41. See Dorsey (1998). 28. Even in the most advanced industrial countries, how- 42. Reduction in the ozone layer results in an increase in ever, physical assets account for only a small proportion of the ultraviolet radiation reaching the earth’s surface. Expo- total assets; human, social, and environmental assets account sure to more solar ultraviolet radiation can lead to a very for the bulk. large increase in skin cancer rates and deaths. 29. World Bank (2000f). 43. The exercise does not look at the effects of environ- 30. World Bank (2000f). mental assets. Physical capital is proxied by investment as a 31. World Bank (2000f). share of GDP. Social assets are measured by the lack of racial 32. Meaning that there is evidence of embodied techni- and nationality tensions. Political terrorism is measured cal progress. using International Country Risk Guide (ICRG) indexes. 33. In addition to collecting water, watersheds perform The distribution of human capital is proxied by average years two other big roles: cleaning water and stabilizing its flow. of schooling. The distribution of education, measured by Stabilization is valuable because rainfall is generally very un- the Gini coefficient, is taken from Thomas, Wang, and Fan even. The watershed automatically compensates for the mis- (2001). Why does the distribution of education matter? In match between the supply of rainfall and the flow down- countries with a very skewed distribution of education, edu- stream because the soil in the watershed is absorbent and cation may not match the level of dispersion of ability. If the releases water gradually. Trees play a central role in this sys- dispersion of education is lower than the dispersion of abil- tem by holding the soil in place (important for water stabi- ity in society, then widening the dispersion of education can lization and cleansing, because the soil acts as a filter), and increase per capita income. by interacting with the fungi and micro-organisms in the soil 44. The minimum threshold level of the (principal com- to break down pollutants and purify water. While there may ponent) of the International Country Risk Guide (ICRG) be substitutes for the cleansing function of watersheds, there index of racial and nationality tensions and political terror- are few substitutes for their flow control function, even in ism was –0.23. industrial communities (Heal 2000). 45. This focus on assets is not to detract from the impor- 34. In 1993 more than 600,000 metric tons of shrimp tance of total factor productivity (TFP) growth in sustaining were harvested from 960,000 hectares of ponds worldwide. growth. In cross-country regressions much of the differences About 80 percent of the total production came from Asia, in growth performance across countries are accounted for by and the rest from Latin America. The availability of shrimp differences in TFP growth. It should be noted, however, that larvae—as commercial hatcheries were established through some of the observed differences could actually reflect mis- the Asia Pacific region during the 1970s and 1980s—the measurement and omitted variables rather than true differ- marketing of formulated feeds, and active support of the gov- ences in TFP growth, and part of TFP growth itself could ernment, set the stage for the industry’s takeoff in the 1980s ultimately be due to asset accumulation. Indeed a main ve- (Primavera 1994). hicle for TFP growth is new technology introduced through 35. Farming systems fall into four broad categories—tra- imported capital and new intermediate goods (embodied ditional, extensive, semi-intensive, and intensive—character- technical progress). It should also be noted that, in looking (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  at the determinants of growth over time within a country Kuznets curve relationship: if structural changes inherent in (rather than at differences in growth performances across the development process lead to cleaner industries (Syrquin countries), asset accumulation is found to account for the 1989); if abatement technologies exhibit increasing returns bulk of GDP growth. to scale (Andreoni and Levinson 1998); if development is ac- 46. For example, Collier (1999) finds that during civil companied by demand for a better environment to which wars GDP per capita declines at 2 percent a year owing partly policies and institutions respond (Grossman 1995); and if to a direct reduction in production and partly to a gradual the stock of environmental assets declines over time while loss of capital stock through destruction, dissaving, and the demand rises. portfolio substitution of capital abroad. 55. Based on Sebastian, Lvovsky, and de Koning (1999), 47. Once such low levels of social assets are reached, there Murray and Lopez (1996), Smith (1998), and World Bank can be a downward spiral in which violence promotes the estimates, 9 percent of DALYs in developing countries are ac- emergence of perverse social assets—those based on crime counted for by water supply and sanitation and urban air pol- and violence, which benefit a few members of the group but lution. Indoor air pollution accounts for an additional 2 per- harm the community at large (Moser and McIlwaine 2001). cent of DALYs. While not an externality, indoor air pollution 48. As mentioned earlier (see note 2), surveys indicate justifies public funding from a poverty reduction perspective. that people value the environment. Research also shows that 56. And the costs of delay in addressing pollution prob- people’s health is positively affected by exposure to, and in- teraction with, the natural world. For instance, in an article lems can sometimes be very high. The experience of indus- in the American Journal of Health Promotion, Dr. Frumkin trial countries with pollution remediation to reduce harmful (2001) examined a series of studies looking at human health health effects is illustrative. For the cases of Itai-itai disease and animals, plants, landscapes, and wilderness. Within each (from cadmium poisoning), the Yokkaichi asthma (from sul- of these domains he found a wide body of evidence linking fur emissions), and the Minamata disease (mercury poison- the domain with human health. For example, one study ing) in Japan, the costs of cleanup and compensation to vic- found that pet owners had significantly lower blood pressure tims are estimated at 1.4 to 102 times the costs of prevention. and cholesterol levels and fewer minor health problems than More important, the costs of prevention were affordable at people who did not own pets. Honeyman (1990) has found the time, given Japan’s per capita income and fiscal resources. that people who are shown urban scenes with vegetation re- The problems were lack of knowledge of the consequences of cover from stress more quickly than people who are shown neglect, and different priorities. Moreover, there can be costs urban scenes without vegetation. In fact, there is a theory— associated with “lock-in�: delays in implementing changes in expanded by Wilson and Kellert (1994)—which asserts that incentives to address pollution problems can lead to invest- human evolutionary history has made a human connection ments and technological lock-ins that cumulatively increase with nature a necessity, not a luxury. the costs of reversing the environment-unfriendly policies 49. The poor can be affected by adverse environmental later on. outcomes in three broad ways: natural resource degradation 57. For instance, a study on China—notable for its analy- can affect their livelihoods; environmental degradation can sis of both the costs and the benefits of addressing air pollu- affect their health; and ecological fragility and the likelihood tion based on firm-level data—shows that a “statistical life� of disasters can affect the poor more than others because of can be saved by removing 100 tons of sulfur dioxide annually their greater vulnerability (effects of climate change on poor from Beijing’s atmosphere. Estimates of abatement costs for nations). See Department for International Development and large plants were $3 a ton, when 10 percent of the emissions others (2002). See also Cavendish (1999) who documents are controlled. So abating 100 tons—at a cost of $300— empirically how environmental resources make a significant would save one life. (The abatement costs for small plants contribution to average rural incomes in Zimbabwe. were considerably higher, but large plants are a much larger 50. The agricultural sector in Madagascar suffered from source of air pollution in Beijing.) (World Bank 2000d). discriminatory policies in the past (with negative rates of pro- 58. However, costs of abatement can be disproportion- tection for rice, for example, of up to 43 percent, and with ately greater for small- and medium-size firms. It may still only irrigated wheat and sugar enjoying protection). While the reforms of the mid-1990s (devaluation of the exchange be the case, however, that policy and regulatory levels are not rate, reduction in import barriers, liberalization of markets, high enough to affect economic growth, because industry privatization of most state enterprises) contributed to a more within individual countries may be resisting greater regula- balanced incentive structure, productivity continues to be tion for fear of becoming uncompetitive—with the result hampered by a lack of fertilizer, inadequate road infrastruc- that pollution levels are also suboptimal. ture, and segregated markets (Paternostro, Razafindravonona, 59. Dasgupta and others (2002). and Stifel 2001). See also box 8.3. 60. The U.S. Superfund, for example, shows how high 51. Cole, Rayner, and Bates (1997). the costs of cleaning up severely polluted areas after the fact 52. Easterly (1999). can be: the program has allocated more than $100 billion. 53. See Borghesi (1999) and Shalizi and Kraus (2001). Part of the problem is that the consequences of the pollution 54. See Dasgupta and others (2002). The theoretical lit- were not known at the time—many of the sites were polluted erature has identified several factors that could give rise to a long ago. (c) The International Bank for Reconstruction and Development / The World Bank ��������  61. See Pagiola and Rothenberg (2002) which provides a 76. Newell and Pizer (2001). Example: Suppose the cur- good compilation of case studies analyzing different market- rent consumption rate of interest is 4 percent, and you know based approaches to forest conservation. that over the next 100 years the rate could either rise to 7 62. Heal (2000). percent or fall to 1 percent. And suppose your project were 63. Classic examples of fish population collapse where to yield a benefit of $100 a hundred years from now. The overfishing may have played a role include the sardine stocks lower rate path would value the $100 dollars at $20.28 of California and Japan in the late 1940s and the anchovy today; the higher at only $0.20 today. If you recognize the stocks of Peru and Chile in 1972. More recent examples of uncertainty in future interest rates, and place equal weight overfishing include the collapse of the Canadian cod fishery on these two outcomes, the expected value of $100 in 2102 and several New England groundfish stocks. Groundfish are would be $10.24. Now suppose we evaluated the expected marine fish that live and feed on or near the bottom of the value one year into the future, in 2103. Based on the lower ocean. Their now-reduced numbers include edible species rate of 1 percent in 2102, the same $20.28 is worth $20.08 that New Englanders have relied on for generations. The ($20.28/1.01 = $20.08), and the $0.20 is worth $0.19. Av- most important are haddock, cod, and yellowtail flounder eraging these, the expected value of $100 delivered in the (Botsford, Castilla, and Peterson 1997). year 2103 would be $10.13. This is very close to the value 64. Coase (1960). of the lower rate of 1 percent $20.28 multiplied by the prob- 65. Myers and Kent (2001). ability of that outcome, 50 percent ($10.14). In this way, the 66. The term energy subsidies can refer to transfers to con- change in value between periods comes to depend solely on sumers through underpricing or transfers to producers the lower rate. Why? Intuitively, discounting benefits 100 through overpricing. years hence depends only on the lower rate because the 67. Some 85 percent of total primary energy supply is higher rate discounts future benefits to such an extent that it adds very little to the expected value. See also Weitzman from fossil fuels and 7 percent from nuclear energy. (1998). 68. Fossil fuels cause many environmental problems apart 77. As discussed in the following chapter, both informal from the better-known ones of oil spills and mining tailings. and formal mechanisms shape the incentive structure facing They cause pollution (sulfur dioxide, nitrogen oxide partic- individuals and hence affect environmental outcomes. ulates, and carbon dioxide). They harm health and affect 78. The broad range of economic instruments includes production both directly and through acid rain (which in taxes and charges, tradable quotas, tradable emission permits, turn damages forests and water bodies). They are also the environmental subsidies, deposit-refund systems, perfor- largest contributor to global warming. mance bonds, noncompliance fees, resource pricing, and re- 69. Myers and Kent (2001). source royalties (OECD 1988; OECD 2001c). 70. OECD (2001c). 79. German Advisory Council on Global Change (2002a). 71. R&D for renewables is also subsidized, but the net 80. Fossil fuel combustion is the largest source of human- subsidy (in terms of the relative price effect) is biased toward caused greenhouse gas emissions—so there are both present fossil fuels, and the total budget drain is higher than if fossil and future costs to society. fuels were not subsidized at all. 81. World Bank (2000d), and www.worldbank.org/nipr. 72. Data Resources Inc. (1997) estimates that there would 82. OECD (2001c). be a loss of 104,000 mining jobs in Europe and Japan 83. World Bank (2000c). (OECD 2001c). 84. OECD (1999, 2001c). 73. Eskeland and Devarajan (1996). 85. Acharya and Dixon, background paper for WDR 74. There are two reasons why a small additional amount 2003. of consumption to be made available at some future year could be socially less valuable than that same additional Chapter 3 amount made available today (why the consumption rate of 1. Sen (1999). interest could be positive). One is impatience, and the possi- 2. See WDR 1992 for policies specifically addressing en- bility of no tomorrow. Another is the expectation that con- vironmental assets. sumption will be greater in the future than it is today, which 3. Analysts such as Ronald Coase, Avner Greiff, Douglass means that the benefit from additional future consumption North, Mancur Olson, and Robert Fogel have greatly con- will be less. Thus the consumption rate of interest is equal tributed to the development of ideas presented in this chap- to the pure rate of time preference (reflecting the first con- ter, even though they are not directly cited (and bear no re- sideration) plus the product of the percentage increase in sponsibility for the content here). marginal well-being consequent on a percentage increase in 4. Ihrig and Moe (2000). consumption (called the elasticity of marginal well-being) 5. de Soto (2000). and the percentage rate of change of consumption (reflect- 6. Besley and Burgess (2001). ing the second consideration) (Dasgupta (2001a)). 7. Steinberg (2001). 75. Since the consumption rate of interest is made up of 8. Farrington and Bebbington (1993, p. 106). two components (see note 74), there is no reason to believe 9. Farrington and Bebbington (1993, p. 73). that rates should remain constant over time. 10. Rose-Ackerman (1999). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  11. Steinberg (2001). 36. Aghion, Caroli, and Garcia-Peñalosa (1999) review 12. Dunlap, Gallup, and Gallup (1993); Dunlap and models of the microlevel links; Rodrik (1996) and Nelson Mertig (1995); Brechin and Kempton (1994); Kidd and Lee and Morrisey (1998) emphasize the links through political (1997); Steinberg (2001, pp. 27–45). support, and negotiating change. 13. Baland and Platteau (1996); Ostrom and Gardner 37. See de Janvry and others (2001) for a review. (1993). 38. Deininger and Squire (1998). 14. Kaufmann, Kraay, and Zoido-Lobatón (1999). 39. Easterly (2002). 15. Sachs and Warner (1995). 40. Acemoglu, Johnson, and Robinson (2001). 16. Svensson (1998). 41. Banerjee and Iyer (2002) 17. The threat of invasions, in turn, provides an interpre- 42. Banerjee, Gertler, and Ghatak (2001). tation of why a system of feudal lords or states emerge. See 43. Findlay and Lundahl (1994). Grossman and Kim (1995); Skaperdas (1992); and Hirsh- 44. Hoff and Sen (2001). leifer (1996) for an analysis of emerging institutions. 45. “Our perspective suggests that, as in Bates’ (1981) 18. This section draws heavily on Hannesson (back- analysis of the political economy of Africa, bad economic ground paper for WDR 2003). policies should be understood as part of a package of often 19. Botsford, Castilla, and Peterson (1997). inefficient redistributive tools� (Acemoglu and others 2002). 20. In some instances the stock collapses and is gone for “Institutions that provide dependable property rights, man- many years. See box 3.4. age conflict, maintain law and order, and align economic 21. World Bank (2000d); Dasgupta, Laplante, and incentives with social costs and benefits are the foundation Mamingi (2001); Dasgupta and Wheeler (1997). of long term growth . . . China, Botswana, Mauritius and 22. See Ter-Minassian (1997); Eskeland, Litvack, and Australia—four cases of success in our sample—all owe their performance to the presence (or creation) of institutions that Rodden (2002); Bardhan and Mookherjee (2000). have generated market-oriented incentives, protected prop- 23. World Bank (2000d). erty rights . . . and enabled social and political stability� (Ro- 24. Bolt and others (forthcoming). The model uses mon- drik 2002); Acemoglu and others (forthcoming). itored concentrations of TSP and PM10 (small dust), city 46. McGuire and Olson (1996); Clague and others and county information to estimate determinants of dust and (1999). small dust particles, and then uses this to project pollution 47. Buchanan (2001). levels for a larger number of cities. The results yield good es- 48. Boyce (2002). timates, but are not accurate at the city level. 49. Abramson and Inglehart (1995). 25. World Bank (2000d). 50. World Bank (2000d). 26. Holtedahl and Vennemo (Background paper for the 51. Yes, markets can help in this. Indeed, they are cen- WDR 2003); Dasgupta and Wheeler (1997). tral. But when markets fail, policies are needed to correct 27. World Bank (1998a). those failures. However, policies also fail. So institutions are 28. King and Özler (1998); Jimenez and Sawada (1999); needed that learn and adapt to support better policies. Eskeland and Filmer (2002). 29. Ostrom and Gardner (1993); Baland and Platteau Chapter 4 (1996). 1. Drylands are classified as arid and dry semi-arid land 30. This is one of the reasons regulations focus on instal- without access to irrigation. Aridity is defined according to lations and procedures, not only on results. an aridity index that is the ratio of precipitation and poten- 31. Conroy (2001). tial evapo-transpiration (UNEP World Atlas of Desertifica- 32. Alston, Libecap, and Schneider (1996); Anderson and tion, 1992 and 1997). Terrain constraints are purely based Hill (1990). on steepness of slope (8 percent or more and would not in- 33. The logic in this section is inspired by the works of clude people on mountain plateaus or valley floors). Poor Grossman and Kim (1995); Skaperdas (1992); Sugden soils are identified by the United Nations’ Food and Agri- (1986); Posner (1981); who all deserve credit. culture Organization (FAO) as unsuitable for rainfed agri- 34. Events of hyperinflation and arrests in savings are of culture. (For details on soil constraints, see http://www.fao. course spectacularly brutal—but poor people and the mid- org/ag/AGL/agll/gaez, Plate 27.) Forests are defined accord- dle class often are hurt through their savings under inflation ing to LandScan data and include deciduous broadleaf, de- and fiscal repression (Easterly and Fischer 2001). Formal in- ciduous needle leaf, evergreen broadleaf, evergreen needle stitutions often fail to protect and support the savings of the forest, and mixed forest of the U.S. Geological Survey Global poor. Rutherford (2000) documents how the poor pay dearly Land Cover Characterization classification. to find adequate outlets for their savings. de Soto (2000) The definition of fragile lands does not include weather documents how the poor are harmed when formal institu- related fragility factors (areas prone to floods, storms and tions fail to welcome and support their assets and activities. cold temperatures) which would significantly increase the es- 35. See Moser and Grant (2000); Fajnzylber, Lederman, timated population on fragile land. Fragile and marginal and Loayza and Fruhling (forthcoming); Tulchin, and Gold- agricultural lands are used interchangeably, as distinct from ing (forthcoming). people or groups marginalized in society, although often the (c) The International Bank for Reconstruction and Development / The World Bank ��������  people living on fragile land are among the most marginal- and World Bank population data estimated that about 2 bil- ized groups in society. lion rural people are on marginal lands with little or no ac- Population estimates combine digital maps on popula- cess to technology, and are remote from services (Pretty tion distribution with maps of the geophysical character- 1995). istics (aridity, slope, soils, and forests). Estimates of rural The lower estimates in this report are due to a more nar- population distribution are based on two global population row definition of arid lands, terrain constraints and forests datasets. The Center for International Earth Science In- (see note 1). The difference may also be due to the higher formation Network (CIESIN) Gridded Population of the resolution population grids selected and the fact that geo- World (GPW v2) is based on total population estimates for graphic information system (GIS) datasets are available now 120,000 administrative reporting units (districts or coun- that were not available in 1995 and 1997. The World Bank’s ties)—see http://sedac.ciesin.org/plue/gpw/index.html. Pop- estimate of 1.3 billion people is intended to focus on the ulation distribution within each unit is assumed to be con- poorest rural groups and indicates a large population for stant. The Oakridge National Lab (ORNL) LandScan whom appropriate services and creative solutions have been dataset uses larger administrative units and adjusts popula- lacking. tion distribution based on proximity to roads and settle- 4. See Cardy (2002). There were an estimated 25 million ments, steepness of slope, nighttime lights from satellite data, environmental refugees in 1995 (excluding temporary refu- and land cover (see http://www.ornl.gov/gist/projects/Land- gees from flooding, who return to their land). The United Scan/SIMPLE/smaps.htm). For each population dataset States Committee on Refugees available on line at www. urban areas are masked out using a global map of nighttime refugees.com estimates that 14.1 million refugees officially city lights from the U.S. National Oceanic and Atmospheric crossed a border in 1999 and another 21 million are inter- Administration (NOAA). GPW (CIESIN) tends to yield nally displaced people. roughly 10 percent higher estimates for fragile lands than 5. See Davis (1993) and World Bank Operational Direc- LandScan, which already reduced population numbers in tive 4.2 on “Indigenous Peoples.� The 250 million estimate areas of steep slopes and unsuitable land cover categories. For of indigenous people is a conservative estimate. Precision is some countries, the proportion was even higher. This Report difficult, since such data are not systematically collected in uses an average of the two results as a best estimate of the true many countries. This chapter looks at a broader population population distribution. Population was uniformly adjusted in the rural periphery, which would include some indigenous for each country to match the World Bank’s rural popula- groups. It does not focus on the problems of indigenous tion estimates for the corresponding year. groups, but recognizes their disproportionate level of poverty Forest data source: Global Land Cover Characterization, and their stores of local knowledge, traditions, and links with U.S. Geological Survey (USGS) Earth Resources Observa- natural environments. tion System (EROS) Data Center, University of Nebraska- 6. See Bonkoungou (2001); UNDP/UNSO, Office to Lincoln (UNL), and Joint Research Centre of the European Commission, 30 arc seconds resolution (approx. 1km), Combat Desertification and Drought. http://edcdaac.usgs.gov/glcc/glcc.html. 7. See UNEP (1992, 1997). The authors estimate that 2. Pratt and Shilling (background paper for WDR 2003). slightly more than 1 billion people are dependent on agricul- 3. Our analysis suggests that 1.4 billion people worldwide ture for their livelihoods and live on drylands. UNEP’s esti- inhabit fragile lands, of whom 1.3 billion are in developing mate includes subhumid and arid or semi-arid irrigated areas countries. This includes an estimated 130 million people liv- (which we have omitted). ing in forests with no other geophysical constraints. Many 8. See http://www.ifpri.cgiar.org/pubs/fps/fps36.htm and of these forests are in fragile ecosystems in remote tropical Bank staff estimates of R&D funding for fragile lands. areas of the Amazon and Central Africa or the boreal forests 9. See Doble (2001); UNDP/UNSO, Office to Combat of Asia. Forest conversion to agriculture may be possible, but Desertification and Drought. with short-lived benefits and unsustainable yields in many 10. See Swearingen and Bencherifa (1996); McNeill places. Moreover, conversion to agricultural or other com- (2000); Leach and Mearns (1996). mercial uses may ignore important public goods benefits 11. Okoth-Ogendo, H.W.O. (2001); Gibson, McKean, (such as the livelihoods of local people who depend in part Ostrom (2000). on forest products, watershed and ecosystem management, 12. See Williams (2001). See http://ag.arizona.edu/ soil maintenance, biodiversity, or aesthetic values—see also OALS/ALN/aln49/Williams.html. chapters 5 and 8 for further discussion of forest conversions). 13. See International Panel on Climate Change (2002), The estimates of people living on fragile land may be on pages 44–46 of the Technical Summary. the conservative side. A 1997 CGIAR study on the priorities 14. See FAO, UNEP, and UNDP (1994); ESCAP (1993); for marginal lands estimated that 1.7 billion people inhabit (UNEP/ISRIC 1990). marginal lands (CGIAR 1997). According to this work, 15. See Ojima (2001). Available from START Secre- roughly 70–75 percent of the rural populations in Sub- tariat, Washington, D.C. Saharan Africa and North Africa and the Middle East and 16. See Ojima (2001); UNDP and others (2000); World 65 percent in Asia and Central and South America live on Resources Institute (2000); Mearns (2001) and (2002). fragile lands. A 1995 study using FAO land classifications 17. Mearns (2001); Mearns (2002). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  18. See Munn, Whyte, and Timmerman (2000). Deser- commodity prices these yield averages have been influenced tification is defined as land degradation in arid, semi-arid, by relatively poor land going out of production (increases av- and dry subhumid areas resulting from various factors, in- erage yields) and lower fertilizer use (lowers average yields). cluding climate variations and human activities. 5. Alexandratos (1995). 19. National Research Council (1999). 6. Rosegrant and others (2001). 20. UNDP (1997). 7. Of the 1 billion poor people identified in 58 poverty 21. Hazell (1998); Hazell and Fan (2000); Fan, Hazell, profiles completed by the World Bank, 75 percent live in and Thorat (2000); Wood and others (1999), variously cited rural areas. in recent overviews by several international organizations, 8. Crosson (1995). such as Dixon, Gulliver, and Gibbon (2001); World Bank 9. Crosson (1995); Lindert (2000). (2002k); and IFAD (2001). 10. Lindert (2000). 22. http://www.cimmyt.cgiar.org/Research/Maize/map/ 11. Murgai, Mubarik, and Byerlee (2001). developing_world/nmaize/new_maize.htm. 12. Sanchez (forthcoming). 23. See Pagiola (1999) and Hassan and Dregne (1997). 13. Pinstrup-Andersen, Panya-Lorch, and Rosegrant 24. Pratt and Shilling, Background paper for the WDR (1999). 2003. 14. Low external input systems have a role to play, espe- 25. Scherr, Sara J., A. White, and D. Kaimowitz. 2002. cially in remote, poor areas. Phosphorous inputs, in addition 26. Pratt and Shilling, Background paper for the WDR to that made available by green manuring, is inevitably nec- 2003. essary on phosphorous-poor soils, however, because green 27. Pratt and Shilling, Background paper for the WDR manure from phosphorus-poor soils is also poor in phospho- 2003. rous (Alexandratos 1995). Green manuring competes with 28. Reinikka and Svensson (2001 and 2002). land for food and often has high labor cost during periods of 29. See World Bank (2001b). peak labor demand. See Hazell (2001); Reardon and others 30. Hemmati and Gardiner (2002); and Lubbock and 1999; Ruttan (1990). Ruttan estimates that low external Bourqia (1998). input systems have the potential to increase food output by 31. See Abraham and Platteau (forthcoming). only about 1 percent a year in Africa, roughly the same rate 32. Zainabi, World Development Report 2003 background observed over the past 20 years, and well short of the expected paper. 3–3.5 percent annual growth in Africa’s food demand. 33. See Donnely-Roark, Ouedraogo, and Ye (2001); Kim, 15. African farmers pay three to five times the world mar- Alderman, and Orazem (1998); de Umanzor and others ket price for fertilizer and receive only 30 to 60 percent of (1997); Fuller and Rivarola (1998). the market value for their products (Hazell 2001). This is 34. See Harper (2000). due in part to high African transport costs (owing to many 35. See Rao (2002). landlocked countries and poor quality of infrastructure) 36. See de Ferranti and others (2002); McMahon and (Limao and Venables 2001) and in part to well-documented Felix (2001); World Bank and International Financing Cor- policy distortions, related to pro-urban bias and implicit tax- poration (2002); Sachs and Warner (1995). ation of agriculture (Lipton 1977; Berg 1981). Under such 37. Chapter 6, McMahon and Felix (2001). conditions it usually does not pay for an African farmer to 38. World Bank and International Financing Corpora- apply fertilizer. tion (2002); McMahon and Felix (2001); Davis and others 16. Sanchez (forthcoming). (2001); Heilburnn (2002). 17. Rosegrant and others (2001). 18. World Water Council (2000). Chapter 5 19. This section is based on IBSRAM (2001). 1. Pinstrup-Andersen, Panya-Lorch, and Rosegrant (1999). 20. Note that these estimates of land lost through degra- 2. Meyer and others (2000). dation suffer from the estimation problem discussed in the 3. Readers seeking a comprehensive treatment of issues re- previous section. Lindert’s (2000) results suggest strongly lated to rural development, water resources, and agricultural that they are upwardly biased. research are invited to read the World Bank’s recently com- 21. This section is based on International Water Manage- pleted publications: World Bank (2001a, 2002a, and 2002b). ment Institute (IWMI 2001). We would also like to recommend IFAD (2001), Eicher and 22. FAO (1997). Statz (1998), and Alexandratos (1995) for comprehensive, 23. IWMI (2001). authoritative, and balanced treatments of rural development 24. Ravallion and Datt (1996). issues in a forward-looking context. 25. Timmer (1997); Deininger and Squire (1998). 4. Crosson and Anderson (2002) show that in develop- 26. Ravallion and Datt (1996); Lanjouw and Lanjouw ing countries yield trends for rice, wheat, and coarse grains (2001). were linear over the 39 years from 1960 through 1998. For 27. Mellor (2000). the decade 1988 to 1998 rice yields were above the trend for 28. See Rodrik (1999). 9 of the 11 years, below for 8 of 11 years for wheat, and 8 of 29. See for example, Alesina and Rodrik (1994); Persson 11 years for coarse grains. In an environment of record low and Tabellini (1994); and Deininger and Squire (1998). (c) The International Bank for Reconstruction and Development / The World Bank ��������  30. Engerman and Sokoloff (1997). genetically modified food equivalent. These include compar- 31. See Tomich, Kilby, and Johnson (1995) for a thor- ing toxin levels, nutrient composition, potential introduced ough treatment of these issues. allergens, new composition, marker genes that could trans- 32. IFAD (2000). fer antibiotic resistance to clinically significant organisms, 33. IFAD (2000). plants not originally developed as food products, and nutri- 34. Sanchez (forthcoming). ents or toxins making the product unacceptable for animal 35. Salamini (1999). feed (ESCOP 2000, cited in Paarlberg 2001). 36. Paarlberg (2001). 62. See Byerlee and Fischer (2000) for an excellent review 37. See Gaskell and others (1999); Wambugu (1999). of institutional models for public-private cooperation. 38. See Paarlberg (2001) for a discussion of the political 63. Personal communication with Richard Barrows. economics of countries’ positions on biotechnology. 64. Personal communication with Richard Barrows. 39. Paarlberg (2001). 65. This tax would be an annual tax calculated on either 40. See Byerlee and Fischer (2000) for an excellent review a market ad valorem or per hectare basis, but invariant with of these institutional models. the use of the land. 41. Siamond vs. Chakrabarty, 447 U.S. 303, 206 U.S.P.Q. 66. De Janvry and others eds. (2001). (2002). 193 (1980). 67. Of course, it is also necessary to have prices that make 42. See, for example, IFAD (2001), chapter 4; Alexan- the venture profitable, transportation and marketing systems dratos (1995), chapter 12; IFPRI (1999); Lele, Lesser, and that allow the product to move into urban markets, and Horstkotte-Wesseler (1999); Foundation News (1999). credit that encourages commercial activity, among other 43. See Paarlberg (2001); also Moore (2001). conditions. 44. Available on-line at http://europa.eu.int/comm/ 68. Deininger and Binswanger (2001). research/press/2001/pr2312en.html. 69. For example, whether a city can divert water from a 45. Paarlberg (2001). small farmer irrigator without compensation will depend on 46. That Thalidomide was not marketed in the United how property rights are allocated. States had nothing to do with suspected teratogenic effects. 70. Personal communication with John Briscoe. The delay in the approval process stemmed from other con- 71. Personal communication with John Briscoe. cerns about the drug. Thalidomide was still in the approval 72. Ostrom and Gardner (1993). process in the U.S. FDA when it was discovered in Europe 73. OECD (1998). that it caused birth defects. 74. The evolution of informal markets has been docu- 47. See, for example, http://www.house.gov/waxman/ mented in the state of Gujarat, India. See Shah (1993). FDA/FDAMA/fdama.html. 75. The process is partially political because the political 48. Seventy-four percent of European citizens hold the process will determine the standards of environmental qual- agri-food industry responsible for the BSE problem. Sixty- ity that society will demand. In several basins with function- nine percent hold politicians responsible as well. Respon- ing water markets in the United States environmental groups dents showed high regard for European scientists but wished and even the U.S. EPA purchased water to augment flow for that scientists would keep them better informed (http:// environmental purposes. europa.eu.int/comm/research/press/2001/pr2312en.html). 76. Schneider (1995). 49. House of Lords (2000). 77. For the Philippines see Coxhead, Rola, and Kim 50. See Echikson (1999). (2001); for Philippines and Thailand see Uhilg (1988). 51. Wambugu (1999). 78. White and Martin (2002). 52. IFAD (2001, p. 138). 79. Scherr, White, and Kaimowitz (2002). 53. See Paarlberg (2001). 80. Wells and others (1999). 54. This section borrows heavily from Deininger and 81. See an extensive discussion with case studies in Ascher Feder (1998). (1999). 55. For a good review of these issues, see Lele, Lesser, and 82. Cochrane and others (1999). Horstkotte-Wesseler (1999). 83. World Bank (2001e). 56. Lele, Lesser, and Horstkotte-Wesseler (1999). 84. Carter and Zimmerman (2000). 57. See World Bank (2001b). 85. Scholz (1985). 58. Losey, Rayor, and Carter, (1999). 86. For example, for Amazon see Alston, Libecap, and 59. See Nature Science Update, “Monarchs safe from BT.� Schneider (1995) and Schneider (1995); for Thailand see Available on-line at: http://www.nature.com/nsu/010913/ Uhilg (1988). 010913-12.html. 87. Iremonger, Ravilious, and Quinton (1997). 60. Statement by European Commissioner for Health 88. Bruner and others (2001). and Consumer Protection David Byrne (Birchard 2000, 89. Bruner and others (2001); Mahar and Ducrot (1998). p. 321, cited in Paarlberg 2001). 90. Available on line at: http://www/worldwildlife.org/ 61. Applying these standards requires that new generally forests/forest.cfm?sectionid=181&newspaperid=17. modified food products be assessed for unexpected genetic 91. “Carrasco da Mata—o massacre de Motoserra.� Veja, effects that might mislead consumers relative to the non- April 17, 1999. English language version available on line at (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  http://www.iucn.org/reuters/1999/articles/winningbrasil. 21. A study in Kenya found that remittances from rural- htm urban migrants, measured by the urban-based nonfarm in- 92. Rosegrant and others (2001), p. 6. come component of smallholder household income, were the 93. Henderson, Shalizi, and Venables (2001). most important determinant of innovation by smallholders and of rates of smallholder poverty. Probability of migration, Chapter 6 in turn, was determined by access to primary education in 1. For ease of reference the terms “urban� and “city� are rural areas (Collier and Lal 1980). used interchangeably here. A “town� can be considered the 22. McGranahan and Satterhwaite (2000); Pugh (1996); smallest class of urban area, having a less diverse economy Bartone and others (1994). than the next city-size category and minimal administrative 23. With increasing economic development and urban or fiscal autonomy. This chapter also uses the terms “local growth, industrial activities become a major source of nonor- government� or “local authority� mainly in reference to a ganic pollutants. However, not all of the accompanying “municipality,� understood here as the lowest organized unit structural changes are unfavorable to the environment— within the administrative apparatus of the state. Many large shifting production from raw materials processing to assem- urban areas, often called “metropolises,� span more than one bly and services generates less water pollution (World Bank municipality. 2000a). 2. This trend appeared earliest and has been documented 24. Yusuf (2001). most extensively for Latin America and the Caribbean: 25. Aggregate, comparable cross-country data are lacking Campbell (1997, background note for WDR 2003); Reilly on the urban incidence of income poverty. The U.N.-Habi- (1995); for a global review, see World Bank (2000b). tat Urban Indicators database for over 200 cities (nonran- 3. At the same time, certain conservative cultural and re- dom sample) indicates income poverty rates averaging about ligious norms may be more strictly enforced by people who 15 percent in the Asian and Middle Eastern cities to over 40 feel these values to be challenged by contact with urban percent in Sub-Saharan Africa; however, the population liv- society. ing in precarious conditions, subject to environmental haz- 4. Holtedahl and Vennemo, background note for WDR ards and vulnerable to both economic and physical risks (as 2003. in informal slum settlements) is estimated to be higher than 5. Wells and others 1999. income poverty rates, ranging from 30 percent to more than 6. Sassen (2001); Wheeler, Aoyama, and Wolf (2000). 60 percent of residents in many cities (U.N.-Habitat 1996). 7. Glaeser (1998). 26. McGranahan and others (2001); World Bank 8. Lall and Ghosh (2002). (2000a). 9. Glaeser and others (1992). 27. “Ecological footprint� refers to the estimated land and 10. Quigley (1998); Mills (2000); Ciccone and Hall forest area-equivalent required to meet the needs of cities for (1996); Prud’homme (1994). energy and waste disposal. Rees (1997) argues that the con- 11. Urban areas can provide more options to use compe- centration of urban populations permits influencing this tition for service provision, as in Colombia’s national school footprint through changes in consumption choices. voucher program which enabled municipalities to help low- 28. Kojima and Lovei (2001); Lvovsky (2001). income households obtain places in private schools (Angrist 29. For the poorest cities and their poorest residents, in- and others forthcoming). door pollution from burning biomass and low-quality coal 12. Hardoy, Mitlin, and Satterthwaite (2001), box 1.4. for cooking and heating remains a major health issue as well. 13. Urbanization, beyond its positive impact on income, 30. Lvovsky and others (2000). is an important determinant of nonmonetary indicators of 31. World Bank (2000a). well-being at the national level, including education (liter- 32. Depending on the country, stationary sources (such acy, school enrollment), health (infant and child mortality, as coal combustion) may be mainly responsible for local pol- life expectancy, malnutrition), and access to basic infrastruc- lutants such as nitrogen oxides, sulfur dioxide, and sus- ture (water, sanitation, electricity, telecommunications). Ur- pended particulate matter, as in China, but in other coun- banization can have a larger impact than economic growth tries motor vehicles account for most nitrogen oxide and on these social indicators (Ryan and Wodon 2001). particulates. Generally, motor vehicles also cause most of the 14. The effect of urbanization on education outcomes is emissions of carbon monoxide, ground-level ozone, and car- also greater than the effects of measures of bureaucratic effi- bon dioxide. ciency and of corruption (Jayasuriya and Wodon, back- 33. Holtedahl and Vennemo, background paper for the ground paper for the WDR 2003). WDR 2003. 15. Williamson (1988). 34. World Bank (2002a), chapter 4. 16. Mazumdar (1987); Wodon and Konig (2001). 35. Improving traffic flow reduces fuel consumption per 17. Lucas (1998). kilometer. 18. Wodon and others (2001). 36. Kojima and Lovei (2001). 19. Oberai and Manmohan Singh (1984). Williamson 37. Gambrill, Foster, and Katakura (2001). (1988) cites further evidence in concurrence. 38. Black-Arveláez (2001). 20. De Haan (1999, 2000). 39. Pantelic (background note for the WDR 2003). (c) The International Bank for Reconstruction and Development / The World Bank ��������  40. Intergovernmental Panel on Climate Change (IPCC) 67. U.N.-Habitat, Urban Indicators Program database forecasts. (1998 data). 41. United Nations (1999), 1999 revisions. 68. Such disparities are also evident between lower- and 42. In most small island states the bulk of population and higher-income neighborhoods (Hardoy, Mitlin, and Sat- infrastructure is located in coastal plains (IPCC 2001). terthwaite 2001, box 4.1). 43. As quoted in the International Federation of Red 69. Stephens and others (1997). Cross and Red Crescent Societies (1999). 70. Rualdo Menegat, Atlas Ambiental de Porto Alegre, 44. Federal Emergency Management Agency (FEMA 1998. Cited with other examples in Hardoy, Mitlin, and Sat- 1998). terthwaite (2001). 45. Anderson (2000). 71. Slum population measured by a composite index of 46. Badly designed insurance schemes create perverse in- housing condition, legal compliance of structure, and access centives—for example, in Florida the creation of a public un- to basic services (water and sewerage). UN-Habitat, “Millen- derwriting agency by the state government to cover private ium Development Goal 7, Target 11, Indicator 31: Propor- insurers for hurricane losses encouraged property develop- tion of Population with Secure Tenure,� April 2002 estimates. ment in hazardous zones, because risks were shared with tax- 72. Perceived stigma and/or job discrimination on the payers and property owners were given a false sense of secu- basis of residence is reported by slumdwellers in many stud- rity (Dunn and Flavin 1999). ies; for Perlman (2002) see Rio de Janeiro; for Baker (2001) 47. Bigio (forthcoming). see Montevideo, Uruguay; and for Jamaica, see Moser and 48. Strictly speaking, a solid waste disposal facility and a Holland (1995). mass transport system (bus or metro) are not fully public 73. McIlwaine and Moser (2001). goods because they can become congested (reducing avail- 74. Rolnik (1999). able use) and access can be restricted; such goods would tech- 75. FYR Macedonia, “Poverty Reduction Strategy Paper: nically be considered club goods. However, the main inter- Urban Poverty Chapter,� January 2002 draft. est here is in activities that affect urban public goods—in this 76. U.N.-Habitat Global Campaign for Secure Tenure case, an environment free of solid waste and a highly accessi- (Nairobi, 1999). See www.urbanobservatory.org. 77. Estimates of “squatters and others� (a residual exclud- ble city. ing formal owner-occupants and tenants) average around 15 49. Serageldin, Shluger, and Martin-Brown (2001). percent of residents of low- and middle-income cities (300 50. Fang (2000). million people), according to the U.N.-Habitat’s Urban In- 51. Peñalosa (2001) (updated April 2002). dicators Database (1998 data). Documented cases of actual 52. OECD, Territorial Development, Urban Brownfields or threatened forced eviction affected almost 8 million peo- (http://www.oecd.org//tds/bis/brownfields-chap3.htm). ple over 1998–2000 worldwide. See Centre on Housing 53. Interview (January 29, 2002) with Dr. Ahmet Sam- Rights and Evictions (COHRE 2001). sunlu, Technical University of Istanbul, and former minister 78. Analysts have identified a continuum of at least 10 cat- of housing, city planning, resettlement and the environment; egories of tenure in cities of developing countries. De facto Hacaoglu (2001); Yildizcan (2002). tenure security may include customary rights and occupancy 54. This section draws heavily on World Bank (2002a). rights (important particularly in African cities), which are 55. Typically, they have average population densities of based on duration of use and recognition by the community. well over 30 residents per hectare, considered the minimum Governments often regularize tenure status by acknowledg- needed to sustain a public transport system. ing these informal rights, thereby granting de facto tenure se- 56. Koster and de Languen (1998). curity (Payne 2001; Payne and Fernandes 2001); Durand- 57. See chapter 4, “Social Sustainability,� in World Bank Lasserve, and Royston (2002). (1996a). 79. The probability of households’ demanding garbage 58. Willoughby (2000). collection increased by 32 percent in going from squatter to 59. International Council for Local Environmental Ini- moderate security status and by 44 percent when the squat- tiatives (ICLEI 1991); Kamel (2000). ters were compared with the high-security (legal certifica- 60. Bartone (2002); see also Wilson, Whiteman, and tion) case (Hoy and Jimenez 1998). Tormin (2001). 80. Lall, Deichmann, and Lundberg (2002). 61. Linz (2002). 81. Fernandes (forthcoming). 62. Bartone (2002). 82. Jimenez (1985). 63. Rosenweig and Solecki (2000). 83. Appadurai (2001). 64. Hentschel (2001). 84. Payne (2001); Gilbert (2002). 65. Numerous country poverty assessments (based on na- 85. de Soto (2000). tional survey data) reveal relatively high intraurban inequal- 86. Gilbert (2002); Calderon Cockburn (2002). ity in income and access to services—often greater than in- 87. Struyk (1997); Grover, Munro-Faure, and Solovier equality within rural areas. See Hentschel and Bump (1999); (chapter 3 in Payne 2002). Eastwood and Lipton (2000), table II.2. 88. Ward (1998); Payne 2002 (chapter 1 and conclu- 66. McGranahan and others (2001), table 4.3. sion). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  89. Mutual protection within the community can be suf- coalition of cities and donors—the Cities Alliance—created ficient to fend off property challenges from other individu- in 1999 to fund poverty-focused city development strategies als. But female-headed households are least able to assert and scaled-up slum upgrading in cities of developing and their rights and may benefit more from the greater security transition economies. (www.citiesalliance.org). of formal titles (Lanjouw and Levy 1998). 110. UN-Habitat and Urban Management Program 90. Payne (2001). (2001). 91. Addressing is not a substitute for formal land registry 111. Andres Escobar, General Manager, Metro Vivienda, but is fully complementary and upgradeable. (See http:// Bogotá, Colombia, December 2001 communication. web.mit.edu/urbanupgrading/upgrading/issues-tools/tools/ 112. Sevilla (2000). street-addressing.html.) 113. Figures for years in the 1990s. Bertaud background 92. World Bank (2000b). paper for the WDR 2003. 93. Burra (2001b), and local press reports in Pune. 114. Bertaud (1999). 94. World Bank (2002c). 115. Bertaud (2000). 95. Readers interested in urban policy more broadly may 116. World Bank (2002b). consult the urban development Web site of the World Bank 117. Land-use regulations and growth controls that re- (www.worldbank.org/html/fpd/urban), or U.N.-Habitat strict the supply of developed land are a major contributor (www.unchs.org). See also chapters 6 and 7 of World Bank to higher prices of serviced land and housing (Shlomo 2000). (2000b) and annex E of World Bank (2001f). 118. The largest are the International Union of Local 96. See also World Bank (2000b), which discusses decen- Authorities, Metropolis, World Federation of United Cities, tralization. and World Associations of Cities and Local Authorities 97 Holtedahl and Vennemo, background paper for WDR Coordination. 2003. 119. See UCCI Web site at www.ayudaurbana.com. 98. This shift is characterized as moving from “expose- 120. City networks in East Asia, such as that of the Chi- oppose� to “expose-oppose-propose,� in Devas and others nese mayors association (http://en.townsfuture.com) and (2001). Philippine city league (ww.cdsea.org) are linking in a re- 99. This section refers to the Orangi Pilot Project’s work gional Web site (www.infocity.org). on sanitation in Karachi and the many successful com- 121. For example, the city of Detroit, Michigan, after re- munity organization–municipal partnerships in Thailand, ceiving a poor ranking in a national rating of U.S. cities, cre- supported by what is currently part of the Community Or- ated its own “Comeback Index,� monitored by an indepen- ganizations Development Institute (Hardoy, Mitlin, and dent institution, to track its improved performance. Satterthwaite 2001). The Philippines example is from Vilo- ria-Williams (2002) background note for the WDR 2003. Chapter 7 100. Durban Metro, Safer City Project, May 2000; Janu- 1. Olson (1996, p. 22). ary 2002 interview with Richard Dobson, Programme 2. Filmer (2000). Leader, Inner Thekwini Renewal and Urban Management 3. Sen (1983). Program, Durban. 4. Besley and Burgess (2000). 101. See the city’s Web site (www.obninsk.ru) for details 5. Sen (1983). (Gonzales de Asis and Acuña-Alfaro 2002). 6. See “Examples of Good Practice in Bank Projects Fo- 102. Katz and Campbell (1996). cused in Primary Education: El Salvador EDUCO Basic Edu- 103. Souza (2001); Municipality of Porto Alegre, “Porto cation Modernization Project� available at: www.worldbank. Alegre Participatory Budget,� September 2000. U.N.— org/oed/oeddoclib. Habitat and The Together Foundation, Best Practices for 7. King, Orazem, and Wohlgemuth (1999); Angrist and Human Settlements Database (www.bestpractices. org). others (forthcoming). 104. World Bank–supported Urban Development and 8. Eskeland and Filmer (2002). Decentralization Program Project in Senegal (ID P002365, 9. Eigen-Zucchi (2001). approved in 1997); Guinea Third Urban Development Project 10. Eigen-Zucchi (2001). (ID P001974, approved in 1999). 11. Calvo (2000). 105. Campbell background paper for the WDR 2003. 12. Rodden, Eskeland and Litvack, forthcoming. 106. Stephens and Wikstrom (2000). 13. World Bank (2002n, p. 113, box 5.7). 107. ICLEI Initiatives (2002). 14. See the homepage of the World Bank’s Small & 108. Velasquez (1998). Medium Enterprise Department, at: http://www.ifc.org/sme/ 109. Other terms may be used for city strategic planning index.html. efforts, which have many antecedents and sources of support 15. de Soto (2000, p. 20). in both industrial and developing countries, such as the U.N. 16. Djankov and others (2002). Urban Management Program’s Sustainable Cities Program 17. Djankov and others (2002); Glaeser and Schleifer (www.unhabitat.org/ump/cityconsultation.htm). The term (2001). city development strategy is used here since it was adopted 18. World Development Report 1997, p. 6. as one of the main activities supported by an international 19. de Soto (2000, p. 21). (c) The International Bank for Reconstruction and Development / The World Bank ��������  20. See, for example, Huther and Shah (2000); Transpar- release detailed environmental data in all the provinces. Air ency International (2000). quality information in 41 large cities is the exception: 21. Ciccone and Hall (1996). progress in other areas is still pending. 22. Henderson and others (2001). 62. Wang and others (forthcoming, p. 2). 23. Henderson (2000); Lall, Shalizi, and Deichmann 63. Communication with Hua Wang. See the World (forthcoming). See also World Bank (2000b, chapter 6). Bank’s website on New Ideas in Pollution Regulation at: 24. Canning (1999); Canning and Bennathan (2000). http://www.worldbank.org/nipr/index.htm for additional 25. World Bank (1994), and a World Bank Policy Re- information on the approach of combining traditional en- search Report that is forthcoming on regulatory reforms in forcement measures with information and disclosure to im- infrastructure. prove incentives and reduce air poluution. 26. Borghesi (1999), Shalizi and Kraus (2001), and World 64. Wang and others (forthcoming, p. 6). Bank (2000a). 65. World Bank (2000d, p. 57). 27. Easterly (1999). 66. World Bank (2000d, p. 63). 28. Myers and Kent (2001), p. 22. 67. See Weber-Fahr (2002) for a discussion of both per- 29. World Bank (1992b). spectives with regard to mining. 30. Tullock (1975). 68. de Ferranti, Perry, Lederman and Maloney (2002, 31. OECD (2001a), and World Bank (2002d). p. 4). 32. Moore (2002). 69. See Murshed (2002a); Auty (1997); Auty and Gelb 33. World Bank (2001d). (2001); Isham and others (2002); Sachs and Warner (1995); 34. OECD (2001a, p. 153). Auty (2002). 35. FAO (2000). 70. Ross (2001a). 36. World Bank (2001e). 71. Auty (2001); Isham and others (2002). 37. World Bank (2001e), p. 2. 72. Isham and others (2002). Revenue flows from other 38. World Bank (2001e), p. 2. types of resources such as agriculture are more diffused 39. World Bank (2000c). throughout the economy (diffuse-source economies) and do 40. Myers and Kent (2001, p. 149). not present as many challenges to the emergence of inclusive 41. Myers and Kent (2001, p. 153). and accountable institutions. 42. Myers and Kent (2001, p. 152). 73. Auty and Kiiski (2001, p. 3). Although causality may 43. See the FAO Fishery Statistics Web page at: http:// be unclear if natural resource sector is the only activity able www.fao.org/fi/statist/nature_china/30jan02.asp. to continue in weak institutional environments, empirical 44. FAO, 2002b, p. 11. studies using lagged variables for the natural resource depen- 45. Willmann, Boonchuwong, and Piumsombun (2002, dency criterion suggest that causality runs from resource fac- p. 187). tors to institutions to economic performance. See for exam- 46. Willmann, Boonchuwong, and Piumsombun (2002, ple, Isham and others (2002). p. 191). 74. Murshed (2002b, p. 1). 47. Willmann, Boonchuwong, and Piumsombun, (2002, 75. Former socialist countries in Europe, Asia, and Africa p. 191). are excluded owing to incomplete data. 48. Hannesson (background paper for the WDR 2003, 76. Isham and others (2002 p. 5). p. 13). 77. See, for example, Auty (1997); Auty and Gelb (2001); 49. Hannesson (background paper for the WDR 2003, Isham and others (2002); Murshed and Perälä (2001); p. 6 and 7). Rodriguez and Sachs (1999); Sachs and Warner (1999a, 50. FAO (2002b, p. 19). 1999b). 51. FAO (2002b, p. 20). 78. McGuire and Olson (1996). 52. FAO (2002b, p. 22). 79. Addison, Le Billon, and Murshed (2000). 53. Holtedahl and Vennemo, background paper for 80. Auty and Kiiski (2001, p. 25). WDR 2003. See also World Bank (2001a). 81. Heilbrunn (2002). 54. Kojima and Lovei (2001, p. 1). 82. Malaysia’s natural resources may be considered dif- 55. Dasgupta and others (2002). fuse, but had many point-source features at the time of in- 56. This section draws on Lovei (1999); Rosner and dependence. For a study on Botswana, see Sarraf and Jiwanji Markowitz (1985); and Holtedahl Vennemo, background (2001). paper for WDR 2003. 83. Rodrik (1999). 57. Lovei, 1999, p. 11. 84. See Kremer and Jayachandran (2002). 58. This section draws heavily on World Bank (2002c). 85. World Bank (1998b). 59. Wang and others (2002, p. 3). 86. The fungibility effect, explored by Devarajan and 60. World Bank (2001a, p. xx). Swaroop (1998), does not depend on fraud. An inflow— 61. Holtedahl and Vennemo, background paper for even in kind, such as for a school project—has a substitution WDR 2003 report that this statement is well known among effect and an income effect. The income effect boosts expen- environmental authorities, but it is not by itself sufficient to diture on all normal goods, including schools and planes, (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  while the substitution effect of a school project—which can 7. Data for COP-6B of the Kyoto Protocol. boost school expenditures beyond the income effect—exists 8. GEF (1999, p. 7). only if the donor succeeds in reducing the marginal cost of 9. Chayes and Chayes (1995). schools to the recipient government. The illustrative use, 10. Hunter, Salzman, and Zaelke (2001). here, of planes, is not accidental: See Devarajan and Haque 11. Mitchell (1995); a background paper for WDR 2003. (2002) for an example. 12. See an extensive discussion in Greening Industry. 87. Alesina and Weder (forthcoming); Knack (2001). 13. Sustain Ability Ltd. and UNEP (2001). 88. Alesina (1998); Burnside and Dollar (1998); World 14. Dowell, Hart, and Yeung (2000) Bank (2002g). 15.King and Lennox (forthcoming). 89. World Bank (2000d, p. 3). 16. Gereffi, Garcia-Johnson, and Sasser (2001). 90. World Bank (1998c); Sambanis (2000). 17. Social Investment Forum (2001). 91. World Bank (1998c, p. v). 18. Boscolo and Vincent (2000). 92. Sambanis, background paper for WDR 2003. 19. Richardson (2001). 93. See Fajnzylber, Lederman, and Loayza (2000); Col- 20. FAO (2000, p. 312). This Forest Resources Assess- lier and Hoeffler (2000); Sambanis, Nicolas, background ment (FRA) remote-sensing survey is based on random sam- paper for WDR 2003. pling of 10 percent of the pantropical forest area. It is distinct 94. Fajnzylber, Lederman, and Loayza (1998); Demom- from the better-known FRA country-based deforestation es- bynes and Özler (2002). timates, which rely on national reports of varying consis- 95. Collier and Hoeffler (2001), abstract. tency, periodicity, and accuracy. 96. See for example Collier and Hoeffler (2000, 2001). 21. Bleaching events—loss of coral’s symbiotic algae, 97. Collier and Hoeffler (2000); Ross (2001b). often leading to death of the coral—are associated with tem- 98. See Isham and others (2002). porary spikes in sea temperature comparable to the perma- 99. Ross (2001a, p. 9). nent rises expected from global warming, and are probably 100. Ross (2001a). exacerbated by other stresses. 101. Addison and Murshed (2002) argue that the failure 22. Wilkinson, ed. (2000). to credibly commit to peace is partly a consequence of the 23. Rannesson, background paper for WDR 2003. impatience to consume resource rents. 24. These categorizations are based on a global sample of 102. Ross (2001a) chose the 13 conflicts from a larger set satellite images that recognizes 9 classes of land cover. Fol- of 21 recent conflicts where scholars and journalists indicated lowing FAO (2000, pp. 313–14), deforestation of closed that natural resources played an important role, including canopy forest has been categorized as shifting agriculture if Afghanistan, Angola, Angola-Cabinda, Cambodia, the Dem- it resulted in open forest or long fallow; smallholder agricul- ocratic Republic of Congo (the 1996–97 and 1998–2001 ture if it resulted in fragmented forest, shrubs, or short fal- conflicts), Indonesia-Aceh, Indonesia–West Papua, Liberia, low; and large-scale agriculture if it resulted in plantations, Papua New Guinea, the Republic of Congo, Sierra Leone, other land cover, or water bodies. and Sudan. 25. These figures represent the total net present value of 103. Ross (2001b). returns over time. 104. Fearon (2001). 26. Wunder (2000, p. 210). 105. Collier and Hoeffler (2000). 27. Tomich and others (1998). 106. Murshed (2002a, p. 3). 28. Kotto-Same and others (2000, p. 35). 107. North and Weingast (1989). 29. Chomitz and others, forthcoming. 108. Acemoglu and Robinson (2000a, 2000b). Stewart, 30. Tomich and others (1998). 2000, also emphasizes the importance of addressing inter- 31. Chomitz and Thomas (2001). group or horizontal inequality for conflict prevention. 32. Burke, Selig, and Spalding (2002, p. 30). 109. Sambanis, Nicolas, background paper for WDR 33. Burke and others (2001, p. 23). 2003. 34. Tilman and others (2001). 35. Burke and others (2001, p. 51). Chapter 8 36. World Bank (2001e). 1. This account draws on National Academy of Sciences 37. As Simpson, Sedjo, and Reid (1996) elegantly (NAS 1996); UNEP (1999); and Jager and others (2001). demonstrate, bioprospecting as currently practiced is un- The quote is from NAS (1996). likely to yield significant per hectare rents, because genetic 2. Munton and others (1999); Jager and others (2001). information in one hectare is likely to be duplicated in an- 3. Clark and Dickson (1999); the framework in this other part of the same habitat. But the loss of an entire, dis- chapter has benefited greatly from this study of global prob- tinctive ecosystem might carry with it significant aggregate lem solving institutions, however, it should not be consid- losses of biological information. ered representative of the Social Learning Group’s views. 38. Wood, Sebastian, and Scherr (2000) p. 71. 4. Dubash and others (2001). 39. Steinberg (2001, p. 39). 5. Victor, Raustiala, and Skolnikoff (1998). 40. Olson and Dinerstein (1998). 6. Victor, Raustiala, and Skolnikoff (1998). 41. Zheng and Eltahir (1997). (c) The International Bank for Reconstruction and Development / The World Bank ��������  42. Lant and Sant (2001). 70. Streets and others (2001). 43. Bewers and Uitto (2001). 71. However, a concomitant 21 percent reduction in SO2 44. Sorensen (2002, p. 9). emissions yielded health benefits but contributed to global 45. Postel and Wolf (2001). warming because SO2 aerosols have a cooling effect. 46. White and Martin (2002); Wily and Dewees (2001). 72. The description of the Thai program is based on 47. OED (2002); Ross (2001b). Singh and Mulholland (2000). 48. Myers and others (2000). 73. Houghton and others (2001, p. 39). 49. Brooks, Pimm, and Oyugi (1999). 74. Author’s calculation based on data on biomass den- 50. Ferraz and others (2002). sity and deforestation from FAO 2000. Carbon-to-biomass 51. Roe and others (2000). ratio assumed to be 0.5. 52. For a review, see McNeely and Scherr (2001). 75. Chomitz (2002). 53. Tomich and others (1998). 76. International Energy Agency (1999b, pp. 99, 105). 54. Ferraro (2002). 77. This example is based on World Bank (2002b). 55. OECD (1997). 77. Schipper, Murtishaw, and Unander (2001). 56. IPCC, 2001. 79. President’s Committee of Advisors on Science and 57. World Bank (2000a). Technology (1997). 58. Nicholls, Hoozemans, and Marchand (1999). Cited 80. President’s Committee of Advisors on Science and in McCarthy and others (2001, p. 396). Technology (1997). 59. Such predictions should be taken only as indicative, 81. OECD (2001b, p. 153). given the uncertainty about many dimensions of climate 82. Milley and others (2002). change. Higher mean precipitation would ameliorate the im- 83. Dilley (2000). pacts, according to the study. Increased summer monsoon 84. Arndt and Bacou (2002). variability, not accounted for in the model, would have the 85. Skees and others (2002). opposite effect. 86. Skees and others (2002). 60. McCarthy and others (2001, p. 517). 61. Watson and others (2000); IPPC (2002). Chapter 9 62. A recent authoritative review is National Research 1. World Bank (2002b). Council (2001). 2. World Bank (2000a), p. 29 63. To be precise, 40 percent of the radiative forcing from 3. Rodrik (2002). well-mixed greenhouse gases in 2000 as compared with 1750 4. World Bank (2002g, 2001f, forthcoming) and DFID (Houghton and others (2001, p. 351). (2002). 64. These are IPCC scenarios A1FI and B1 from Naki- 5. World Bank (1996b). cenovic and Swart (2000). Underlying data are available at 6. World Bank (2002a), page 93. sres.ciesin.org. Emissions include gases other than CO2, and 7. World Bank (2002c). emissions from land use. The ratio in 2100 of per capita 8. Mackay (2000), pp. 43–56; Sustainability Ltd, and GDP in the non-OECD (membership as of 1990) to OECD UNEP (2001); Dowell, Hart, and Yeung (2000). countries is 57 percent in A1FI and 51 percent in B1. In 9. Guerrero O. (1999). A1FI, the 2100 share in primary energy of renewables is 17 10. These institutions have sometimes become fully in- percent; for B1, the share, including ‘nonfossil electric’ is 53 tegrated with operational international efforts to negoti- percent. The actual 1999 share was about 5 percent (IEA ate and implement agreements, as in the case of CLRTAP 2001, p. 312). and with transboundary assessments of international waters 65. Stott and Kettleborough (2002). management. 66. Stott and Kettleborough (2002). 11. Devarajan, Miller, and Swanson (2002). 67. Basis of calculation: 1999 Norway emissions, 34.3 12. Council of the European Union (2002). million tons CO2. Population: 4 million. Global emissions: 13. World Bank (2000b); World Bank (2002d), p. 98. 6.3 billion tons per year; carbon equivalent from fossil fuels 14. Myers and Kent (2001); Fischer and Toman (1998); and cement, 1.7 billion tons per year; carbon equivalent Ascher (1999). from land-use change; 3.66 tons CO2 per ton. 15. See World Bank (2002a), chapter 4. 68. IEA (2001). 16. World Bank (2002b), p. 58–9. 69. OECD (2001b, p. 153). 17. World Bank (2002ee), p. 133. 18. World Bank (2002ee), pp. 141–142. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  References Anbarasan, Ethirajan, and Choi Yul. 2001. “The Greening of Abraham, Anita, and Platteau, Jean-Philippe. Forthcoming. Korea.� UNESCO Courier 54(2):47–51. “Participatory Development: Where Culture Creeps In.� In Andean Center for Economics in the Environment. 2001. “Eco- Vijayendra Rao and Michael Walton, eds. Culture and Pub- nomic Instruments and Environment: Can Economic In- lic Action. Stanford, Calif.: Stanford University Press. struments Work in Developing Countries?� Available on line Abramson, Paul R., and Ronald Inglehart. 1995. Value Change at www.andeancenter.com. in Global Perspective. Ann Arbor: University of Michigan Anderson, Mary B. 2000. “The Impacts of Natural Disasters on Press. the Poor: A Background Note.� World Bank, Washington, Acemoglu, Daron, Simon Johnson, and James A. Robinson. D.C. Processed. 2001. “Reversal of Fortune: Geography and Institutions in Anderson, Terry. 1996. “Conservation—Native American the Making of the Modern World Income Distribution.� Style: A Summary.� Political Economy Research Center Pol- National Bureau of Economic Research Working Paper icy Series PS–6. Bozeman, Montana. Available online at 8460. Cambridge, Mass. http://www.perc.org/publications/policyseries/conserva- Acemoglu, Daron, Simon Johnson, James Robinson, and Yuny- tion_full.html. ong Thaicharoen. Forthcoming. “Institutional Causes, Macro- Anderson, Terry L., and Peter J. Hill. 1990. “The Race for Prop- economic Symptoms: Volatility, Crisis and Growth.� Journal erty Rights.� Journal of Law and Economics 33: 177–97. of Monetary Economics. Andreoni, J., and A. Levinson. 1998. “The Simple Analytics of Acemoglu, Daron, and James A. Robinson. 2000a. “Inequality, the Environmental Kuznets Curve.� National Bureau of Growth and Development: Democratization or Repression?� Economic Research Working Paper 6739. Cambridge, Mass. European Economic Review 44:683–93. Angrist, Joshua D., Eric Bettinger, Erik Bloom, Elizabeth King, ———. 2000b. “Why Did the West Extend the Franchise? and Michael Kremer. Forthcoming. “Vouchers for Private Democracy, Inequality and Growth in Historical Perspec- Schooling in Colombia: Evidence from a Randomized Nat- tive.� Quarterly Journal of Economics 115(4): 1167–99. ural Experiment.� American Economic Review. Adams, James D. 2001. “Comparative Localization of Academic Appadurai, Arjun. 2001. “Deep Democracy: Urban Govern- and Industrial Spillovers.� National Bureau of Economic Re- mentality and the Horizon of Politics.� Environment and search Working Paper 8292. Cambridge, Mass. Urbanization 13(2):23–43. Addison, Tony, Philippe Le Billon, and S. Mansoob Murshed. Araujo, J., and M. Neto Costa. 1999. “The Impact of Civil 2000. “On the Economic Motivation for Conflict in Africa.� Conflict on Long-Term Performance in the Height of the Paper presented at World Bank ABCDE–Europe Confer- Angolan Experience.� World Bank Institute, World Bank, ence, Paris, June 2000. Washington, D.C. Processed. Addison, Tony and Mansoob Murshed. 2002. “Credibility and Arndt, C., and M. Bacou. 2002. “Economy–Wide Effects Reputation in Peacemaking.� Forthcoming in the Journal of of Climate Variability and Climate Prediction in Mozambi- Peace Research July. que.� American Journal of Agricultural Economics 82(3): Aghion, Philippe, Eve Caroli, and Cecilia Garcia–Peñalosa. 750–754. 1999. “Inequality and Economic Growth: The Perspective Ascher, William. 1999. Why Governments Waste Natural Re- of the New Growth Theories.� Journal of Economic Litera- sources: Policy Failures in Developing Countries. Baltimore, ture 37(4):1615–60. Md.: Johns Hopkins University Press. Alesina, Alberto. 1998. “The Political Economy of Macroeco- Auty, Richard M., ed. 2001. Resource Abundance and Economic nomic Stabilizations and Income Inequality: Myths and Re- Development. Oxford: Oxford University Press. ality.� In V. Tanzi and Ke–young Chu, eds., Income Distri- ———. 2002. “Best Practice for Economic Diversification in bution and High Quality Growth. Cambridge, Mass.: MIT Mineral Economies.� Paper presented at the Managing Press. Volatility Workshop. World Bank, Washington, D.C. Alesina, Alberto, and Dani Rodrik. 1994. “Distributive Policies Auty, Richard M., and Alan H. Gelb. 2001. “Political Economy and Economic Growth.� Quarterly Journal of Economics of Resource–Abundant States.� In R. M. Auty, ed., Resource 109(2):465–90. Abundance and Economic Development. New York: Oxford Alesina, Alberto, and Beatrice Weder. Forthcoming. “Do Cor- University Press. rupt Governments Receive Less Foreign Aid?� American Eco- Auty, R. M., and Sampsa Kiiski. 2001. “Natural Resources, nomic Review. Capital Accumulation, Structural Change, and Welfare.� In Alexandratos, Nikos, ed. 1995. World Agriculture: Toward R. M. Auty, ed., Resource Abundance and Economic Develop- 2010, A FAO Study. Rome: Food and Agriculture Organiza- ment. New York: Oxford University Press. tion of the United Nations; and John Wiley and Sons, Ltd. Baker, Judy L. 2001. “Social Exclusion in Urban Uruguay.� Alston, Lee, Gary D. Libecap, and Robert Schneider. 1995. Latin America and the Caribbean Region, World Bank, “Property Rights and the Preconditions for Markets: The Washington, D.C. Processed. Case of the Amazon Frontier.� Journal of Institutional and Baland, Jean-Marie, and Jean-Philippe Platteau. 1996. Halting Theoretical Economics 151(1):89–107. Degradation of Natural Resources. Food and Agriculture Or- ———. 1996. “The Determinants and Impact of Property ganization of the United Nations, Oxford: Clarendon Press. Rights: Land Titles on the Brazilian Frontier.� The Journal Available online at www.fao.org. of Law, Economics and Organization 12(1):25–61. Banerjee, Abhiji, Paul J. Gertler, and Maitreesh Ghatak. 2001. Amahan, Ali. 1998. Mutations sociales dans le Haut Atlas: Les “Empowerment and Efficiency: Tenancy Reform in West Ghoujdama. Paris, Rabat: Maison des Sciences de l’Homme, Bengal.� Department of Economics, University of Chicago. Paris, and Editions La Porte, Rabat. Processed. (c) The International Bank for Reconstruction and Development / The World Bank ����������  Banerjee, Abhijit, and Lakshmi Iyer. 2002. “History, Institu- Bloom, D. E., and J. G. Williamson. 1997. “Demographic tions and Economic Performance: The Legacy of Colonial Transition and Economic Miracles in Emerging Asia.� Na- Land Tenure Systems in India.� Cambridge, Mass.: MIT. tional Bureau of Economic Research Working Paper 6268. Banerjee, Abhiji, Dilip Mookherjee, Kaivan Munshi, and De- Cambridge, Mass. braj Ray. 1997. “Inequality, Control Rights and Rent Seek- Bolt, K., U. Deichmann, K. Hamilton, B. Ostro, K. Pandy, and ing—A Theoretical and Empirical Analysis of Sugar Coop- D. Wheeler. Forthcoming. “The Human Cost of Air Pollu- eratives in Maharashtra.� Boston University, Institute for tion: New Estimates for Developing Countries.� World Economic Development 80. Bank Policy Research and Environment Department Work- Bardham, Pranab K., and Dilip Mookherjee. 2000. “Capture ing Paper, Washington, D.C. and Governance at Local and National Levels.� American Bohn, Henning, and Robert T. Deacon. 2000. “Ownership Economic Review 90(2):135–39. Risk, Investment, and the Use of Natural Resources.� Amer- Bartone, Carl. 2002. “Institutional Arrangements for Solid ican Economic Review 90(3):526–49. Waste Management in Metropolitan Areas.� Presentation in Bonkoungou, Eduoard. 2001. “Biodiversity in Drylands: Chal- Metropolitan Governance Seminar Series, World Bank In- lenges and Opportunities for Conservation and Sustainable stitute, March 6, Washington, D.C. Processed. Land Use.� United Nations Development Programme, Bartone, Carl, Janis Bernstein, Josef Leitmann, and Jochen Nairobi. Eigen. 1994. “Toward Environmental Strategies for Cities.� Boonchuwong, P., W. Dechboon, and M. Ahmed. 2002. “Bioe- United Nations Development Programme/ United Nations conomic modeling of the fisheries in the Gulf of Thailand.� Commission on Human Settlements/ World Bank Urban PRIAP Working Paper 8. Policy Research and Impact As- Management Program 18. Washington, D.C. sessment Program. International Center for Living Aquatic Behnke, R. H., Jr., I. Scoones, and C. Kerven, eds. 1993. Range Resources Management: The World Fish Center, Penang, Ecology at Disequilibrium: New Models of Natural Variability Malaysia. and Pastoral Adaptation in African Savannas. London: Over- Borghesi, Simone. 1999. “The Environmental Kuznets Curve: seas Development Institute. A Survey of the Literature.� European University Institute, Berg, Elliot. 1981. “Accelerated Development in Sub- Florence. Available online at http://www.feem. it/web/activ/ Saharan Africa.� World Bank, Washington, D.C. wp/abs99/85–99.pdf. Processed. Bernardes, Aline Tristao. 1999. “Some Mechanisms for Biodi- Boscolo, Marco, and Jeffrey R. Vincent. 2000. “Promoting Bet- versity Protection in Brazil, with Emphasis on Their Appli- ter Logging Practices in Tropical Forests: A Simulation cation in the State of Minas Gerais.� World Bank: Washing- Analysis of Alternative Regulations.� Land Economics 76(1): ton, D.C. Processed. 1–14. Bertaud, Alain. 1999. “Cracow in the Twenty-First Century: Botsford, L. D., J. C.Castilla, and C. H. Peterson. 1997. “The Princes or Merchants? A city’s structure under the conflict- Management of Fisheries and Marine Ecosystems.� Science ing influences of land markets, zoning regulations, and a so- 277: 509–15. cialist past.� Europe and Central Asia Region, Infrastructure Bowles, Samuel, and Herbert Gintis. 1999. “ ‘Social Capital’ and Department Working Paper 8. World Bank, Washington, Community Governance.� University of Massachusetts, D.C. Amherst. Available online at http://www-unix.oit. umass. ———. 2000. “The Costs of Utopia: Brasilia, Johannesburg edu/~bowles/papers/Socap.PDF. Processed. and Moscow.� City of Helsinki Web site, Gavle. Available Boyce, James K. 2002. The Political Economy of the Environment. online at http://www.hel.fi/english/index.html. Processed. Cheltenham, U.K.: Edward Elgar. Besley, Timothy, and Robin Burgess. 2000. “Land Reform, Brakarz, José, with Margarita Greene and Eduardo Rojas. 2002. Poverty Reduction and Growth: Evidence from India.� Ciudades para todos: La experiencia reciente en programa de Quarterly Journal of Economics 115(2):389–430. majoramiento de barrios. Washington, D.C.: Inter-American ———. 2001. “Political Agency, Government Responsiveness Development Bank. and the Role of the Media.� European Economic Review Brechin, Steven R., and Willett Kempton. 1994. “Global Envi- 45(4–6):629–40. ronmentalism: A Challenge to the Postmaterialism Thesis?� Bewers, J. Michael, and Juha I. Uitto. 2001. “International Wa- Social Science Quarterly 75(2):245–69. ters Program Study.� Global Environment Facility, Wash- Breman, H., and C. T. Wit. 1983. “Rangeland Productivity and ington, D.C. Exploitation in the Sahel.� Science 221(4618): 1341–46. Bigio, Anthony G. Forthcoming. “Cities and Climate Change.� Brockerhoff, Martin, and Ellen Brennan. 1998. “The Poverty In Building Resilience in Metropolitan Areas. Washington, of Cities in Developing Regions.� Population and Develop- D.C.: World Bank. ment Review 24(1):1–40. Binswanger, Hans P., and Klaus Deininger. 1997. “Explaining Brooks, Thomas M., Stuart L. Pimm, and Joseph O. Oyugi. Agricultural and Agrarian Policies in Developing Countries.� 1999. “Time Lag between Deforestation and Bird Extinc- Journal of Economic Literature 35(4): 1958– 2005. tion in Tropical Forest Fragments.� Conservation Biology Black-Arveláez. 2001. “Economic Instruments and the Environ- 13(5):1140–50. ment.� Andean Center for Economics and the Environment Brown, Lester R., Flavin Christopher, and Hilary French. 2001. Bulletin 4(1):1. State of the World 2001. New York: W. M. Norton. Blanchflower, David G., and Andrew J. Oswald. 2000. “Well- Brundtland Commission. 1987. “Our Common Future.� The Being over Time in Britain and the USA.� National Bureau World Commission on Environment and Development, of Economic Research Working Paper 7487. Cambridge, Geneva. Available online at http://geneva-international.org/ Mass. GVA/WelcomeKit/Environnement/chap_5.E.html. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Bruner, Aaron G., Raymond E. Gullison, Richard E. Rice, and Carter, Michael R., and Frederick J. Zimmerman. 2000. “The Gustavo A. B. de Fonseca. 2001. “Effectiveness of Parks in Dynamic Cost and Persistence of Asset Inequality in an Protecting Tropical Biodiversity.� Science 291: 125–8. Agrarian Economy.� Journal of Development Economics 63(2): Brunner, Jake, and Francois Ekoko. 2000. “Cameroon.� In 265–302. Frances J. Seymour and Navroz K. Dubash, eds. The Right Cavendish, William. 1999. “Poverty, Inequality and Environ- Conditions: The World Bank, Structural Adjustment, and Forest mental Resources: Quantitative Analysis of Rural House- Policy Reform. Washington, D.C.: World Resources Institute. holds.� T. H. Huxley School, Imperial College of Science, Buchanan, James M. 2001. “The Constitutional Way of Think- Technology and Medicine Working Paper Series 99–9. Lon- ing.� George Mason University. Processed. don. Burke, Lauretta, Yumiko Kura, Ken Kassem, Carmen Revenga, CGIAR (Consultative Group on International Agricultural Re- Mark Spalding, and Don McAllister. 2001. Pilot Analysis of search). 1997. “Report of the Study on CGIAR Research Global Ecosystems: Coastal Ecosystems. Washington, D.C.: Priorities for Marginal Lands.� Technical Advisory Commit- World Resources Institute. Available online at http://www. tee Working Document. Food and Agriculture Organization wri.org/wr2000/coast_page.html. of the United Nations, Rome. Burke, Lauretta, Elizabeth Selig, and Mark Spalding. 2002. Chayes, Abraham, and Antonia Handler Chayes. 1995. The Reefs at Risk in Southeast Asia. Washington, D.C.: World Re- New Sovereignty: Compliance with International Regulatory sources Institute. Agreements. Cambridge, Mass.: Harvard University Press. Burnside, Craig, and David Dollar. 1998. “Aid, the Incentive Chéret, Ivan. 1993. “Managing Water: The French Model.� In Regime, and Poverty Reduction.� World Bank Policy Re- Ismael Serageldin and Andrew Steer, eds., Valuing the Envi- search Department Working Paper No. 1937. June. Wash- ronment. Washington, D.C.: World Bank. ington, D.C. Chen, Shoahua, and Martin Ravallion. 2000. “How Did the Burra, Sundar. 2001a. “Resettlement and Rehabilitation of the World’s Poorest Fare in the 1990s?� World Bank Policy Re- Urban Poor: The Mumbai Urban Transport.� Society for search Working Paper 2409. World Bank, Policy Research Promotion of Area Resource Centers, Mumbai. Processed. Department, Washington, D.C. ———. 2001b. “Slum Sanitation in Pune: A Case Study.� So- Chomitz, Kenneth. 2002. “Baseline, Leakage and Measurement ciety for Promotion of Area Resource Centres, Mumbai. Issues: How Do Forestry and Energy Projects Compare?� Available online at www.sparcindia.org. Processed. Climate Policy 2(1):35–49. Byerlee, Derek, and Ken Fischer. 2000. “Accessing Modern Sci- Chomitz, Kenneth M., Keith Alger, Timothy S. Thomas, ence: Policy and Institutional Options for Agricultural Bio- Heloisa Orlando, and Paulo Vila Nova. Forthcoming. “Op- thechnology in Developing Countries.� World Bank, Wash- portunity Costs of Conservation in a Biodiversity Hotspot: ington, D.C. Processed. The Case of Bahia.� World Bank, Washington, D.C. Calvo, Guillermo A. 2000. “The Case for Hard Pegs in the Chomitz, Kenneth, and T. Thomas. 2001. “Geographic Pat- Brave New World of Global Finance.� Paper presented at terns of Land Use and Land Intensity in the Brazilian Ama- ABCDE Europe, Paris, June 26. zon.� World Bank Policy Research Working Paper 2687. Calderon Cockburn, Julio. 2002. “The Mystery of Credit.� Lin- Washington, D.C. coln Institute of Land Policy, Cambridge. Processed. Ciccone, Antonio, and Robert E. Hall. 1996. “Productivity and Campbell, Tim. 1997. “Innovations and Risk Taking: The En- the Density of Economic Activity.� American Economic Re- gine of Reform in Local Governments in Latin America and view 86(1):54–70. the Caribbean.� World Bank Discussion Paper 357. Wash- Clague, Christopher, Philip Keefer, Stephen Knack, and Man- ington, D.C. cur Olson. 1999. “Contract-Intensive Money: Contract En- ———. Forthcoming. The Quiet Revolution: The Rise of Politi- forcement, Property Rights, and Economic Performance.� cal Participation and Local Government in Latin America and Journal of Economic Growth 4:185–209. the Caribbean. Washington, D.C: World Bank. Clark, J. B. 1898. “The Future of Economic Theory.� Quarterly Canning, David. 1999. “Infrastructure’s Contribution to Ag- Journal of Economics 13(2):1–14. gregate Output.� World Bank Policy Research Working Clark, William, and Nancy Dickson. 1999. “The Global Envi- Paper 2246. Washington, D.C. ronmental Assessment Project: Learning from Efforts to Canning, David, and Esra Bennathan. 2000. “The Social Rate Link Science and Policy in an Interdependent World.� Ac- of Return on Infrastructure Investment.� World Bank Pol- climations 8:6–7. icy Research Working Paper 2390. Washington, D.C. Coase, Ronald. 1960. “The Problem of Social Cost.� Journal of Cardy, W. Franklin. 2002. “Environment and Forced Migra- Law and Economics 3:1–44. tion: A Review.� In Michael B. K. Darkosh and Apollo Cochrane, Mark, Ane Alencar, Mark D. Schulze, Carlos M. Rwonmire, eds., Human Impact on Environmental and Sus- Souza, Daniel C. Nepstad, Paul Lefebvre, and Eric A. David- tainable Development in Africa. Oxford: Ashgate Publishing. son. 1999. “Positive Feedbacks in the Fire Dynamic in Carlin, Elaine. 2002. “Oil Pollution from Ships at Sea: The Closed Canopy Tropical Forests.� Science 284: 1832–35. Ability of Nations to Protect a Blue Planet.� In Edward COHRE (Centre on Housing Rights and Evictions). 2001. Miles, Arild Underdal, Steinar Andresen, Jorgen Wettestad, “Forced Evictions: Violations of Human Rights.� Global Jon Birger Skjaerseth, and Elaine M. Carlin, eds., Environ- Survey on Forced Evictions 8. Geneva. mental Regime Effectiveness: Confronting Theory with Evi- Cole, M. A., A. J. Rayner, and J. M. Bates. 1997. “The Envi- dence. Cambridge, Mass.: MIT Press. ronmental Kuznets Curve: An Empirical Analysis.� Environ- Carson, R. 1962. Silent Spring. Boston: Houghton Mifflin. ment and Development Economics 2(4):401–16. (c) The International Bank for Reconstruction and Development / The World Bank ����������  Collier, Paul. 1999. “On the Economic Causes of Civil War.� tions.� IMF Occasional Paper 205. International Monetary Oxford Economic Papers 51(1):168–83. Fund, Washington, D.C. Collier, Paul, and Anke Hoeffler. 2000. “On the Incidence of Davis, Shelton. 1993. “Indigenous Views of Land and the En- Civil War in Africa.� World Bank, Washington, D.C. Avail- vironment.� World Bank Discussion Paper 188. Wash- able online at http://www.worldbank.org/research/conflict/ ington, D.C. papers.htm. Processed. de Ferranti, David, Guillermo E. Perry, Daniel Lederman, and ———. 2001. “Greed and Grievance in Civil War.� World William F. Maloney. 2002. “From Natural Resources to the Bank, Washington, D.C. Processed. Knowledge Economy: Trade and Job Quality.� World Bank, Collier, Paul, and Deepak Lal. 1980. “Poverty and Growth in Washington, D.C. Available online at http://..1nweb18. Kenya.� World Bank Staff Working Paper 389. Washing- worldbank.org/External/laconst. ton, D.C. De Haan, Arjan. 1999. “Livelihoods and Poverty: The Role of Collomb, Jean-Gael, and Heriette Bikié. 2001. “1999–2000 Al- Migration—A Critical Review of the Migration Literature.� location of Logging Permits in Cameroon: Fine-Tuning Cen- Journal of Development Studies 36(2):1–47. tral Africa’s First Auction System.� Global Forest Watch, ———. 2000. “Migrants, Livelihoods, and Rights: The Rele- Cameroon. Available online at www.globalforestwatch.org. vance of Migration in Development Policies.� World Bank Conroy, Mike. 2001. “Yellowstone Honors Army for Sav- Social Development Working Paper 4. Washington, D.C. ing Park.� U.S. Army Environment Center. Available de Janvry, Alain, Gustavo Gordillo , Jean-Philippe Platteau, and online at http://www.dtic.mil/armylink/news/Oct2000/ Elisabeth Sadoulet, eds. 2001. Access to Land, Rural Poverty a20001004 yellowstone2.html. and Public Action. New York: Oxford University Press. Cosgrove, W. J., and F. R. Rijsberman. 2000. World Water Vi- de Mello Lemos, Maria Carmen. 1998. “The Politics of Pollu- sion: Making Water Everybody’s Business. London: Earthscan tion Control in Brazil: State Actors and Social Movements for the World Water Council. Cleaning up Cubatão.� World Development 26(1): 75–87. Council of the European Union. 2002. “Towards a Global Part- de Soto, Hernando. 2000. The Mystery of Capital: Why Capital- nership for Sustainable Development.� Communication ism Triumphs in the West and Fails Everywhere Else. New from the Commission to the European Parliament, the York: Perseus Book Group. Council, the Economic and Social Committee, and the Com- de Umanzor, Sandra, Isis Soriano, Marta Rosa Vega, Emmanuel mittee of the Regions, February 25. Brussels. Jimenez, Laura Rawlings, and Diana Steele. 1997. “El Sal- Coxhead, Ian, Agnes Rola, and Kwansoo Kim. 2001. “How Do vador’s EDUCO Program: A First Report on Parent’s Par- National Markets and Price Policies Affect Land Use at the ticipation in School-Based Management.� World Bank Forest Margin? Evidence from the Philippines.� Land Eco- Working Paper Series on Impact Evaluation of Education nomics 77(2):250–67. Reforms 4. Washington, D.C. Crosson, Pierre. 1995. “Soil Erosion and Its On-Farm Produc- Di Primio, Juan Carlos. 1998. “Data Quality and Compliance tivity Consequences: What Do We Know?� RFF Discussion Control in the European Air Quality Regime.� In David G. Paper 95–29. Resources for the Future, Washington, D.C. Victor, Kal Raustiala, and Eugene B. Skolnikoff, eds., The Crosson, Pierre, and Jock R. Anderson. 2002. “Technologies for Implementation and Effectiveness of International Environ- Meeting Future Global Demands for Food.� RFF Discus- mental Commitments: Theory and Practice. Cambridge, sion Paper 02–02. Resources for the Future, Washington, Mass.: MIT Press. D.C. Deininger, Klaus, and Hans Binswanger. 2001. “The Evolution Dasgupta, Partha. 2000. “Valuing Biodiversity.� In Simon of the World Bank’s Land Policy.� In Alain de Janvry, Gus- Levin, ed., Encyclopedia of Biodiversity. New York: Academic tavo Gordillo, Jean-Philippe Platteau, and Elisabeth Sadou- Press. Forthcoming. let, eds., Access to Land, Rural Poverty, and Public Action. Dasgupta, Partha. 2001a. Human Well-Being and the Natural New York: Oxford University Press. Environment. Oxford: Oxford University Press. Deininger, Klaus, and Gershon Feder. 1998. “Land Institutions ———. 2001b. “Valuing Objects and Evaluating Policies in and Land Markets.� World Bank Policy Research Working Imperfect Economies.� Economic Journal 111(471): C1–29. Paper 2014. Washington, D.C. ———. 2002. “Population and Resources: An Exploration of Deininger, K., and Lyn Squire. 1998. “New Ways of Looking Reproductive and Environmental Externalities.� Population at Old Issues: Inequality and Growth.� Journal of Develop- and Development Review 26(4):643–89. ment Economics 57(2):259–87. Dasgupta, S., B. Laplante, and N. Mamingi. 2001. “Pollution Demombynes, Gabriel, and Berk Özler. 2002. “Inequality, and Capital Markets in Developing Countries.� Journal of Property Crime, and Violent Conflict in South Africa.� Environmental Economics and Management 42(3). World Bank, Washington, D.C. Processed. Dasgupta, Susmita, Benoit Laplante, Hua Wang, and David Devarajan, Shantayanan, and Trina Haque. 2002. “Human De- Wheeler. 2002. “Confronting the Environmental Kuznets velopment Policy Note.� World Bank, Washington, D.C. Curve.� Journal of Economic Perspectives 16(1):147–68. Processed. Dasgupta, Susmita, and David Wheeler. 1997. “Citizen Com- Devarajan, Shantayanan, Margaret J. Miller, and Eric V. Swan- plaints as Environmental Indicators: Evidence from China.� son. 2002. “Goals for Development: History, Prospects and World Bank Policy Research Working Paper 1704. Wash- Costs.� World Bank, Washington, D.C. Processed. ington, D.C. Devarajan, Shantayanan, and Vinaya Swaroop. 1998. “The Im- Davis, Jeffrey, Rolando Ossowski, James Daniel, and Steven plications of Foreign Aid Fungibility for Development As- Barnett. 2001. “Stabilization and Savings Fund for Nonre- sistance.� World Bank Policy Research Working Paper newable Resources: Experience and Fiscal Policy Implica- 2022. Washington, D.C. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Devas, Nick, Philip Amis, Richard Batley, Ursula Grant, Fiona Durand-Lasserve, Alain, and Lauren Royston, eds. 2002. Hold- Nunan, and Elyzabeth Vidler. 2001. “Urban Governance, ing Their Ground: Secure Land Tenure for the Urban Poor in Partnership and Poverty: ESCOR–Funded Research in Ten Developing Countries. London: Earthscan. Cities: 1998–2001.� University of Birmingham, U.K. Avail- Durban Metro. 2000. Safer City Project. Available online at able online at http://www.bham.ac. uk/IDD/activities/ SaferCities@durban.gov.za. urban/urbgov.htm. Easterly, William. 1999. “Life during Growth.� Journal of Eco- DfID (Department for International Development, United nomic Growth 4(3):239–75. Kingdom), European Commission Directorate General for ———.2002. “Inequality Does Cause Underdevelopment: Development, UNDP (United Nations Development Pro- New Evidence.� Center for Global Development Working gram), and World Bank. 2002. “Linking Poverty Reduction Paper 1. Washington, D.C. and Economic Management: Policy Challenges and Oppor- ———. Forthcoming. “Can Institutions Resolve Ethnic Con- tunities.� Washington, D.C. Processed. flict?� Economic Development and Cultural Change. Diamond, Jared. 1997. Guns, Germs and Steel: The Fate of Easterly, William, and Stanley Fischer. 2001. “Inflation and the Human Societies. New York: W. W. Norton. Poor.� Journal of Money, Credit and Banking 33(2): 160–78. Diener, E. 1984. “Subjective Well-Being.� Psychological Bulletin Eastwood, Robert, and Michael Lipton. 2000. “Rural-Urban Di- 95:542–75. mensions of Inequality Change.� World Institute for Devel- Dilley, Maxx. 2000. “Reducing Vulnerability to Climate Vari- opment Economics Research Working Paper 2003. Helsinki. ability in Southern Africa: The Growing Role of Climate In- EBRD (European Bank for Reconstruction and Development). formation.� Climatic Change 45: 63–73. 2002. Transition Report 2000. London. Dixon, John, Aidan Gulliver, and David Gibbon. 2001. “Glo- Echikson, William. 1999. “Food Scares Whet Europe Appetite bal Farming Systems Study: Challenges and Priorities to for U.S.-Type Oversight.� Christian Science Monitor 28 July. 2030.� Rome. Food and Agriculture Organization of the Available online at http://www.csmonitor. com/durable/ United Nations. 1999/07/28/fp7s1-csm.shtml. Djankov, Simeon, Rafael La Porta, Rafael Lopez-de-Silanes, and Economist. 2001. “The Future of Farming in the Great Ameri- Andrei Schleifer. 2002. “The Regulation of Entry.� The can Desert.� The Economist (December 15): 26–7. Quarterly Journal of Economics 117(1): 1–37. Ehrlich P., and Ehrlich A. 1981. Extinction: The Causes and Doble, Philip. 2001. “Poverty and the Drylands.� United Na- Consequences of the Disappearance of Species. New York: Ran- tions, Nairobi. dom House. Donnely-Roark, Paula, K. Ouedraogo, and X. Ye. 2001. “Can Eicher, Cark K., and John M. Staatz. 1998. International Agri- Local Institutions Reduce Poverty? Rural Decentralization cultural Development. Baltimore, Md.: Johns Hopkins Uni- in Burkina Faso.� World Bank Policy Research Working versity Press. Paper 2677. Washington, D.C. Eigen, Peter, and Christian Eigen-Zucchi. 2002. “Corruption Dorsey, Eleanor. 1998. The Road to Groundfish Collapse and and Global Public Goods.� In Kaul, Inge, eds., Providing Turning the Corner to Recovery: A Brief History of the New Global Public Goods: Managing Globalization. New York: England Fisheries Crisis. Boston, Mass.; Rockland, Me.; Con- Oxford University Press. cord, N.H.; Providence, R.I.; Montpelier, Vt.: Conservation Eigen-Zucchi, Christian. 2001. The Measurement of Transac- Law Foundation. tions Costs, Fairfax, Virginia, George Mason University. Dowell, Glen, Stuart Hart, and Bernard Yeung. 2000. “Do Cor- Ph.D. Dissertation. Available online at http://eigen1.tripod. porate Global Environmental Standards Create or Destroy com/tpi. pdf. Market Value?� Management Science 46(8): 1059–74. Engerman, Stanley L., Stephen H. Haber, and Kenneth L. DRI (Data Resources Inc.) 1997. “Effects of Phasing Out Coal Sokoloff. 2000. “Inequality, Institutions, and Differential Subsidies in OECD Countries.� In OECD, eds., Environ- Paths of Growth among New World Economies.� In Claude mental Taxes and Green Reforms. Paris: DRI/McGraw-Hill. Menard, ed., Institutions, Contracts, and Organizations: Per- Dubash, Navroz, Mairi Dupar, Smitu Kothari, and Tundu spectives from New Institutional Economics. London: Edward Lissu. 2001. “A Watershed in Global Governance? An Inde- Elgar. pendent Assessment of the World Commission on Dams.� Engerman, Stanley L., and Kenneth L. Sokoloff. 1997. “Factor World Resources Institute, Washington, D.C. Available on- Endowments: Institutions, and Differential Paths of Growth line at http://www.wri.org/wri/pdf/wcd_ full.pdf. Among New World Economies: A View from Economic Dunlap, Riley E., and Angela G. Mertig. 1995. “Global Con- Historians of the United States.� In Stephen H. Haber, ed., cern for the Environment: Is Affluence a Prerequisite?� Jour- How Latin America Fell Behind: Essays in the Economic His- nal of Social Issues 51(4):121–37. tories of Brazil and Mexico, 1800–1914. Stanford, Calif.: Dunlap, Riley E., George H. Gallup, Jr., and Alec M. Gallup. Stanford University Press. 1993. Health of the Planet: Results of a 1992 International ———. 2001. “The Evolution of Suffrage Institutions in the Environmental Opinion Survey of Citizens in 24 Nations. New World.� National Bureau of Economic Research Princeton, N.J.: The George H. Gallup International Working Paper 8512. Cambridge, Mass. Institute. ESCAP. 1993. “State of Urbanization in Asia and the Pacific.� Dunn, Seth, and Christopher Flavin. 1999. “Destructive Storms United Nations, New York. Drive Insurance Losses Up—Will Taxpayers Have to Bail Eskeland, Gunnar S., and Shantayanan Devarajan. 1996. Tax- Out Insurance Industry?� Worldwatch Institute, Worldwatch ing Bads by Taxing Goods: Pollution Control with Presump- News Brief 99–3: Washington, D.C. tive Charges. Washington, D.C.: World Bank. (c) The International Bank for Reconstruction and Development / The World Bank ����������  Eskeland, Gunnar S., and Deon Filmer. 2002. “Autonomy, Par- Ferraz, Goncalo, Gareth J. Russell, Philip C. Stouffer, Richard ticipation, and Learning in Argentine Schools.� World Bank O. Bierregaard, Stuart L. Pimm, and Thomas E. Lovejoy. Policy Research Working Paper 2766. Washington, D.C. 2002. “Rate of Species Loss from Amazonian Forest Frag- Eskeland, Gunnar S., Jennie Litvack, and Jonathan Rodden. ments.� Wooster College, Wooster. Processed. 2002. Decentralization and the Challenge of Hard Budget Filmer, Deon. 2000. “The Structure of Social Disparities in Ed- Constraints. Boston: MIT Press. ucation: Gender and Wealth.� World Bank Policy Research Essama-Nssah, B., and James Gockowski. 2000. “Cameroon: Working Paper 2268. Washington, D.C. Forest Sector Development in a Difficult Political Econ- Findlay, Ronald, and Mats Lundahl. 1994. “Natural Resources, omy.� Evaluation Country Case Study Series. World Bank, ‘Vent-for-Surplus’ and the Staples Theory.� In Gerald M. Washington, D.C. Meir, ed. From Classical Economics to Development Econom- Fajnzylber, Pablo, Daniel Lederman, and Norman Loayza. ics. New York: St. Martin’s Press. 1998. “Determinants of Crime Rates in Latin America and Fischer, Carolyn, and Michael Toman. 1998. “Environmentally the World.� World Bank, Latin America and Caribbean and Economically Damaging Subsidies: Concepts and Illus- Studies Viewpoints. Washington, D.C. trations.� Resources for the Future: Climate Issue Brief 14. ———. 2000. “Crime and Victimization: An Economic Per- Washington, D.C. spective.� Economía 1(1):219–302. Foundation News. 1999. “The Rockefeller Foundation and Plant ———. Forthcoming. “Inequality and Violent Crime.� Jour- Biotechnology.� June 24. nal of Law and Economics. Frank, Robert H. 1985. Choosing the Right Pond: Human Be- Fan, Shenggen, Peter Hazell, and Sukhadeo Thorat. 2000. havior and the Quest for Status. New York: Oxford Univer- “Government Spending, Agricultural Growth and Poverty sity Press. in Rural India.� American Journal of Agricultural Economics Frank, Robert H., and Philip J. Cook. 1995. The Winner-Take- 82(4):1038–51. All Society : Why the Few at the Top Get So Much More than Fang, Ke. 2000. Redevelopment in the Inner City of Contempo- the Rest of Us. New York: The Free Press. rary Beijing: Survey, Analysis and Investigation. Beijing: China Freeman, Richard B., and Remco H. Oostendorp. 2000. Architectural Industry Press. “Wages around the World: Pay across Occupations and FAO (Food and Agriculture Organization of the United Na- Countries.� National Bureau of Economic Research Work- ing Paper 8058. Cambridge, Mass. tions). 1997. “Irrigation Potential in Africa: A Basin Ap- Fruhling, Hugo, Joseph S. Tulchin, and Heather A. Golding. proach.� Rome. Forthcoming. Crime and Violence in Latin America: Citizen ———. 2000. “Global Forest Resources Assessment 2000.� Security, Democracy, and the State. Washington, D.C.: Rome. Available on line at http://www.fao.org/forestry/ Woodrow Wilson Center Press. fo/fra/main/index.jsp. Frumkin, Howard. 2001. “Beyond Toxicity: Human Health ——- 2002a. “FAO Fishery Statistics.� Available online at and the Natural Environment.� American Journal of Preven- http://www.fao.org/WAICENT/FAOINFO/FISHERY/sta- tive Medicine 20(3):234–40. tist/statist.asp. Fuller, Bruce, and Magdalena Rivarola. 1998. “Nicaragua’s Ex- ——— 2002b. “The State of World Fisheries and Aquacul- periment to Decentralize Schools: Views of Parents, Teach- ture.� Available online at http://www.fao.org/sof/sofia/ ers and Directors.� World Bank Working Paper Series on Im- index_en.htm. pact Evaluation of Education Reforms 5. Washington, D.C. FAO, UNEP, and UNDP (Food and Agriculture Organization Gallup, John, and Jeffrey Sachs. 1998. “Geography and Eco- of the United Nations, United Nations Environment Pro- nomic Growth.� Paper presented at World Bank Annual Con- gramme, and United Nations Development Programme). ference on Economic Development. Washington, D.C., April 1994. “Land Degradation in South Asia: Its Severity, Causes 20–21. Available online at http://www. worldbank.org/ and Effects upon the People.� World Soil Resources Report html/rad/abcde/html/sachs.htm. Rome. Available online at http://www.fao.org/docrep/ Galor, Oded, and Joseph Zeira. 1993. “Income Distribution V4360E/V4360E00. htm. and Macroeconomics.� Review of Economic Studies 60(1): Farrington, John, and Anthony Bebbington, eds. 1993. Reluc- 35–52. tant Partners? Non-governmental Organizations, the State and Galtung, Fredrik. 2000. “A Global Network to Curb Corrup- Sustainable Agricultural Development. New York: Routledge. tion: The Experience of Transparency International.� In Fearon, James. 2001. “Why Do Some Civil Wars Last So Much Ann M. Florini, ed., The Third Force: The Rise of Transna- Longer than Others?� World Bank–University of California, tional Civil Society. Washington, D.C.: Carnegie Endow- Irvine, Conference. Processed. ment for International Peace. FEMA (Federal Emergency Management Agency). 1998. “Re- Gambrill, Martin, Vivien Foster, and Yoko Katakura. 2001. port on Costs and Benefits of Natural Hazard Mitigation.� “Lessons of Experience with Condominial Water and Sew- Washington, D.C. erage Programs in Brazil and Bolivia.� World Bank, Wash- Fernandes, Edesio. Forthcoming. “Combining Tenure Policies, ington, D.C. Processed. Urban Planning and City Management in Brazil.� In Payne, Garreau, Joel. 2001. “Flocking Together to the Web.� The Geoffrey, ed., Land, Rights and Innovations. London: Inter- Washington Post May 9:C1. mediate Technology Development Group. Gaskell, George, Martin W. Bauer, John Durant, and Nicholas Ferraro, Paul J. 2002. “Global Habitat Protection: Limitations of C. Allum. 1999. “Worlds Apart? The Reception of Geneti- Development Approaches and a Role for Conservation Per- cally Modified Foods in Europe and the U.S.� Science, formance Payments.� Conservation Biology 15(4):990–1000. 285–384. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  GEF (Global Environment Facility). 1999. “An Interim Assess- Hacaoglu, Selcan. 2001. “Turkey’s Bay of the Ottoman Sultans ment of Biodiversity Enabling Activities: National Biodiver- Recovers from Industrial Filth.� Associated Press November sity Strategies and Action Plans.� Washington, D.C. 27. Gereffi, Gary, Ronie Garcia-Johnson, and Erika Sasser. 2001. Hall, Robert E., and Charles I. Jones. 1999. “Why Do Some “The NGO-Industrial Complex.� Foreign Policy July/Au- Countries Produce So Much More Output per Worker than gust:56–65. Others?� Quarterly Journal of Economics 114(1): 83–116. German Advisory Council on Global Change. 2002a. “Charg- Hamilton, Kirk. 2000. “Genuine Saving as a Sustainability In- ing the Use of Global Commons.� German Advisory Coun- dicator.� World Bank Environmental Economics Series 77. cil on Global Change Policy Paper 2. Berlin. Washington, D.C. ———. 2002b. “Charging the Use of Global Commons, Spe- Hardoy, Jorge E., Diana Mitlin, and David Satterthwaite. 2001. cial Report.� German Advisory Council on Global Change Environmental Problems in an Urbanizing World. London: Secretariat. Berlin. Earthscan. Gibson, Clark C., Margaret A. McKean, and Elinor Ostrom, Harper, Malcolm. 2000. Public Services through Private Enter- prises: Micro Privatization for Improved Delivery. New Delhi: eds. 2000. People and Forests: Communities, Institutions and Chapman Enterprises. Governance. Cambridge, Mass.: MIT Press. Hassan, Hassan, and H. E. Dregne. 1997. “Natural Habitats Gilbert, Alan. 2002. “On the Mystery of Capital and the Myths and Ecosystems Management in Drylands.� World Bank of Hernando de Soto: What Difference Does Legal Title Environment Department Paper 51. Washington, D.C. Make?� International Development Planning Review 24:1–20. Hatton, Timothy J., and Jeffrey Williamson. 1998. The Age of Glaeser, Edward L. 1998. “Are Cities Dying?� Journal of Eco- Mass Migration: Causes and Economic Impact. New York and nomic Perspectives 12(2):139–60. Oxford: Oxford University Press. Glaeser, Edward L., and Andrei Schleifer. 2001. “The Rise of ———. 2001. “Demographic and Economic Pressures on Em- the Regulatory State.� National Bureau of Economic Re- igration out of Africa.� National Bureau of Economic Re- search Working Paper 8650. Cambridge, Mass. search Working Paper 8124. Cambridge, Mass. Glaeser, Edward L., Hedi D. Kallal, José A. Scheinkman, and Hayek, Friedrich A. 1945. “The Use of Knowledge in Society.� Andrei Shleifer. 1992. “Growth in Cities.� Journal of Politi- American Economic Review 35(4):519–30. cal Economy 100(6):1126–52. Hazell, Peter. 1998. “Why Invest More in the Sustainable De- Gonzales de Asis, Maria, and Jairo Acuña-Alfaro. 2002. “Civic velopment of Less-Favored Lands?� International Food Pol- Participation in Local Governance.� Governance Training icy Research Institute Report No. 20. Washington, D.C. Module, World Bank Institute, Washington, D.C. Available Available online at http://www. ifpri.cgiar.org/reports/ online at http://www.worldbank.org/ wbi/governance/pdf/ 0798rpt.htm. fy02brief/civic_local_annex.pdf. Processed. ———. 2001. “Shaping Globalization for Poverty Alleviation Grieg-Gran, Maryanne. 2000. “Fiscal Incentives for Biodiver- and Food Security: Technology Change.� 2020 Vision for sity Conservation: The ICMS Ecológico in Brazil.� Interna- Food, Agriculture, and the Environment No. Focus 8, Pol- tional Institute for Environment and Development Discus- icy Brief 8 of 13. International Food Policy Research Insti- sion Paper 00–01. London. tute, Washington, D.C. Available online at http://www. Grindle, Merilee S., ed. 1997. Getting Good Government: Ca- ifpri.cgiar.org/ 2020/focus/focus08.htm. pacity Building in the Public Sectors of Developing Countries. Hazell, Peter, and Fan Shenggen. 2000. “Should Developing Cambridge, Mass.: Harvard University Press. Countries Invest More in Less-Favored Areas?: An Empiri- ———. 2000. Audacious Reforms: Institutional Invention and cal Analysis of Rural India.� Economic and Political Weekly Democracy in Latin America. Baltimore: Johns Hopkins Uni- 35(17): versity Press. Heal, G. 2000. Nature and the Market Place: Capturing the Grootaert, Christian, and Thierry van Bastelaer. 2001. “Under- Value of Ecosystem Services. Washington, D.C.: Island Press. Heilburnn, John R. 2002. “Governance and Oil Funds.� Col- standing and Measuring Social Capital: A Synthesis of Find- orado School of Mines, Golden, Colo. Processed. ings and Recommendations from the Social Capital Initia- Hellman, Joel S., Geraint Jones, and Daniel Kaufmann. 2000. tive.� World Bank Social Capital Initiative Working Paper “ ‘Seize the State, Seize the Day’: State Capture, Corruption, 24. Washington, D.C. and Influence in Transition.� World Bank Policy Research Grossman, Gene. 1995. “Pollution and Growth: What Do We Working Paper 2444. Washington, D.C. Know?� In I. Goldin and L. Alan Winters, eds., Economics Hemmati, Minu, and Rosalie Gardiner. 2002. “Gender Equity of Sustainable Development. Cambridge: Cambridge Univer- and Sustainable Development: Towards Earth Summit sity Press. 2002.� Social Briefing 2. Available online at http:// Grossman, Herschel I., and Minseong Kim. 1995. “Swords or www.earthsummit2002.org/es/issues/ gender/gender.PDF. Plowshares? A Theory of the Security of Claims to Property.� Henderson, Vernon. 2000. “How Urban Concentration Affects Journal of Political Economy 103(6):1275–88. Economic Growth.� World Bank Policy Research Working Guerrero, O., and R. Pablo. 1999. “Comparative Insights from Paper 2326. Washington, D.C. Colombia, China, and Indonesia.� World Bank Operations Henderson, J. Vernon, Zmarak Shalizi, and Anthony J. Ven- Evaluation Department 5. Washington, D.C. ables. 2001. “Geography and Development.� Journal of Eco- Gwatkin, Davidson. 2000. “Socio-Economic Differences in nomic Geography 1:81–105. Health, Education, and Nutrition in Madagascar.� World Hentschel, Jesko. 2001. “Poverty in Cali, Colombia.� World Bank, Washington, D.C. Processed. Bank, Washington, D.C. Processed. (c) The International Bank for Reconstruction and Development / The World Bank ����������  Hentschel, Jesko, and Jesse Bump. 1999. “Urban Poverty Di- tainable Development, Second Preparatory Session (28 mensions: Some Cross-Country Comparisons.� World Jan–8 Feb, 2002). New York: United Nations, Department Bank, Washington, D.C. Processed. of Economic and Social Affairs. Also available online at Hirsch, Fred. 1978. Social Limits to Growth. London, New www.iclei.org. York: Routledge. IFAD (International Fund for Agricultural Development). Hirshleifer, Jack. 1996. “Anarchy and its Breakdown.� In 1999. “Improving Tassa Planting Pits—Using Indigenous Michelle R. Garfinkel and Stergios Skaperdas, eds., The Po- Soil and Water Conservation Techniques to Rehabilitate litical Economy of Conflict and Appropriation. Cambridge: Degraded Plateaus in the Tahoua Region of Niger.� Cambridge University Press. MOST/CIRAN Best Practices on Indigenous Knowledge. Hochstetler, Kathryn. 1997. “The Evolution of the Brazilian Available online at http://www.unesco.org/ most/bpik10. Environmental Movement and Its Political Roles.� In Dou- htm. glas A. Chambers and others, eds. The New Politics of In- ———. 2000. Rural Poverty Report 2000.Oxford University equality in Latin America: Rethinking Participation and Rep- Press. resentation. New York: Oxford University Press. ———. 2001. “Rural Poverty Report 2001: The Challenge of Hoff, Karla, and Arijit Sen. 2001. “Empowerment and Home- Ending Rural Poverty.� Available online at http://www. Ownership.� World Bank, Washington, D.C. Processed. ifad.org/poverty/. Hoff, Karla, and Joseph Stiglitz. 2002. “After the Big Bang? Ob- IFPRI (International Food Policy Research Institute). 1999. stacles to the Emergence of the Rule of Law in Post-Commu- “Biotechnology for Developing-Country Agriculture: Prob- nist Societies.� World Bank, Washington, D.C. Processed. lems and Opportunities.� 2020 Focus 2. Washington, D.C. Honeyman, M. 1990. “Vegetation and Stress: A Comparison Ihrig, Jane, and Karine Moe. 2000. “The Dynamics of Informal Study of Varying Amounts of Vegetation in Countryside and Employment.� Federal Reserve Board, International Finance Urban Scenes.� In D. Relf, ed., The Role of Horticulture in Discussion Paper 664. Washington, D.C. Available online Human Well-Being and Social Development: A National Sym- at http://www.federalreserve.gov/pubs/ ifdp/2000/664/de- posium. 1992 Timber Press, 143–145. fault.htm. Hoogeveen, Hans. 2001. “Assessing Uganda’s Decentraliza- International Council for Local Environmental Issues. 2002. tion.� World Bank, Washington, D.C. Processed. “Second Local Agenda 21 Survey.� Paper presented at the Houghton, John, Ding Yihui, David J. Griggs, Maria Noguer, Commission on Sustainable Development, Second Prepara- Paul J. van der Linden, and Xiaosu Dai, eds. 2001. Climate tory Session, available on line at www.iclei.org. Change 2001: The Scientific Basis. Cambridge, Mass.: Cam- IEA (International Energy Agency). 1999a. World Energy Out- bridge University Press. look—1999 Insights. Looking at Energy Subsidies: Getting the House of Lords, Select Committee on Science and Technology. Price Right . London: IEA Publications. 2000. “Science and Society.� London. Available online ———. 1999b. World Energy Outlook 2000. London: IEA Pub- at http://www.parliament.the-stationery-office.co. uk/pa/ lications. ld199900/ldselect/ldsctech/38/3801.htm. International Federation of Red Cross and Red Crescent Soci- Hoy, Michael, and Emmanuel Jimenez. 1998. “The Impact on eties. 1999. World Disasters Report 1999. Geneva. the Urban Environment of Incomplete Property Rights.� IPCC (Intergovernmetal Panel on Climate Change). 2001. World Bank, Washington, D.C. Processed. “Human Settlements, Energy and Industry.� In IPCC, ed., Hunter, David, James Salzman, and Durwood Zaelke, eds. Third Assessment Report, 2001. Cambridge: Cambridge Uni- 2001. International Environmental Law and Policy. New versity Press. York: Foundation Press. Available online at http://www. ———. 2002. “Climate Change 2001: Impacts, Adaptation, wcl.american.edu/environment/iel/. and Vulnerability.� In James J. McCarthy, Osvaldo F. Huther, Jeff, and Anwar Shah. 2000. “Anti-Corruption Policies Canziani, Neil A. Leary, David J. Dokken, and Kasey S. and Programs: A Framework for Evaluation.� World Bank White, eds., Cambridge: Cambridge University Press. Policy Research Working Paper Series 2501. Washington, Iremonger, S., C. Ravilious, and T. Quinton. 1997. “A Statisti- D.C. cal Analysis of Global Forest Conservation.� In S. Iremon- IADB (Inter-American Development Bank). 1997. Economic ger, C. Ravilious, and T. Quinton, eds., A Global Overview and Social Progress in Latin America: Latin America after a of Forest Conservation. Cambridge: Center for International Decade of Reforms. Baltimore and London: Johns Hopkins Forestry Research and WCMC. University Press. Available online at http:// www.iadb.org/ Isham, Jonathan, Michael Woolcock, Lant Pritchett, and Gwen oce/ipes/. Busby. 2002. “The Varieties of Rentier Experience: How IBSRAM (International Board for Soil Research and Manage- Natural Resource Endowments Affect the Political Economy ment). 2001. “Background Study on Land Degradation in of Economic Growth.� Middlebury College, Middlebury, Selected Regions and Some Consequences for Rural Devel- Vt., World Bank, Washington, D.C., Harvard University, opment.� World Bank, Washington, D.C. Available online Cambidge, Mass., and Yale University, New Haven, Conn. at http://wbln0018.worldbank.org/essd/rdv/vta.nsf/Gweb/ Processed. Studies. Processed. Iskandar Kamel, Laila. 2000. “Urban Governance: The Informal ICLEI (International Council for Local Environmental Initia- Sector and Municipal Solid Waste in Cairo.� Available on- tives). 1991. “Solid Waste Management: Bandung, Indone- line at http://www.archis.org/english/ archis_ art_e_2000/ sia.� ICLEI Case Study 3. Toronto. archis_art_0012b_e.html. ———. 2002. “Second Local Agenda 21 Survey.� Background IWMI (International Water Management Institute). 2001. Paper No. 15. Paper presented at the Commission on Sus- “Water for Rural Development.� Background paper pre- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  pared for the World Bank Agricultural Strategy Paper. Paper Series on Impact Evaluation of Education Reforms 9. Colombo, Sri Lanka. Processed. Washington, D.C. Jacobs, Goff, Amy Aeron-Thomas, and A. Astrop. 2000. “Esti- Klenow, Peter J., and Andrés Rodriguez-Clare. 1997. “The mating Global Road Fatalities.� Transport Research Labora- Neoclassical Revival in Growth Economics: Has It Gone tory Report 445. Berkshire, U.K. Available online at Too Far?� In Ben S. Bernanke and Julio S. Rotemberg, eds., http://www.grsproadsafety.org/activities/campaigns/ 5/10. NBER Macroeconomics Annual. Cambridge and London: pdf. MIT Press. Jäger, Jill, Nancy M. Dickson, Adam Fenech, Peter M. Haas, Knack, Stephen. 2001. “Aid Dependence and the Quality of Edward A. Parson, Vassily Sokolov, Ferenc L. Tóth, Jeroen Governance: Cross-Country Empirical Tests.� Southern Eco- van der Sluis, and Claire Waterton. 2001. “Monitoring in nomic Journal 68(2):310–29. the Management of Global Environmental Risk.� In Knack, Stephen, and Philip Keefer. 1997. “Does Social Capital William C. Clark, Jill Jäger, Josee van Eijndhoven, and Have an Economic Payoff? A Cross-Country Investigation.� Nancy M. Dickson, eds., Learning to Manage Global Envi- Quarterly Journal of Economics 112(4):1251–88. ronmental Risks Vol. II. Cambridge, Mass. and London: MIT Kojima, Masami, and Magda Lovei. 2001. “Urban Air Quality Press. Management: Coordinating Transport, Environment and Jäger, Jill, Josee van Eijndhoven, and William C. Clark. 2001. Energy Policies in Developing Countries.� World Bank “Knowledge and Action: An Analysis of Linkages among Technical Paper, Pollution Management Series 508. Wash- Management Functions for Global Environmental Risks.� ington, D.C. In William C. Clark, Jill Jäger, Josee van Eijndhoven, and Koster, J. H., and M. de Languen. 1998. “Preventive Transport Nancy M. Sdickson, eds., Learning to Manage Global Envi- Strategies for Secondary Cities�. In P. Freman and C. Jamet, ronmental Risks Vol. II. Cambridge, Mass., and London: eds., Urban Transport Policy: A Sustainable Development MIT Press. Tool. Rotterdam: Balkema. Jensen, Jesper, and David Tarr. 2002. “Trade, Foreign Ex- Kotto-Same, J., A. Moukam, R. Njomgang, T. Tiki-Manga, J. change, and Energy Policies in the Islamic Republic of Iran: Tonye, C. Diaw, J. Gockowsky, S. Hauser, S. Weise, D. Reform Agenda, Economic Implications, and Impact on the Nwaga, L. Zapfack, C. Palm, P. Woomer, A. Gillison, D. Poor.� World Bank Policy Research Working Paper 2768. Bignell, and J. Tondoh. 2000. “Alternatives to Slash-and- Washington, D.C. Burn: Summary Report and Synthesis of Phase II in Jimenez, Emmanuel. 1985. “Urban Squatting and Community Cameroon.� Consultative Group on International Agricul- Organization in Developing Countries.� Journal of Public tural Research, Nairobi. Economics 27(1):69–92. Kremer, Michael. 2000. “Creating Markets for New Vaccines.� Jimenez, Emmanuel, and Yasuyuki Sawada. 1999. “Do Com- National Bureau of Economic Research Working Paper munity-Managed Schools Work? An Evaluation of El Sal- 7716. Cambridge, Mass. vador’s EDUCO Program.� World Bank Economic Review Kremer, Michael, and Seema Jayachandran. 2002. “Odious 13(3):415–41. Debt.� Paper presented at the Conference on Macroeco- Kamel, Laila Iskandar. 2000. “Urban Governance: The Infor- nomic Policies and Poverty Reduction. Washington, D.C. mal Sector and Municipal Solid Waste in Cairo.� ARCHIS Available online at http://www.imf.org/external/ NP/Res/ December. Available online at www.archis.org. seminars/2002/poverty/. Katz, Travis, and Tim Campbell. 1996. “Manos a la obra.� Krishna, Anirudh, and Norman Uphoff. 1999. “Mapping and World Bank, Latin America and the Caribbean Technical Measuring Social Capital: A Conceptual and Empirical Department, Washington, D.C. Processed. Study of Collective Action for Conserving and Developing Kaufmann, Daniel, Aart Kraay, and Pablo Zoido-Lobatón. Watersheds in Rajasthan, India.� World Bank Social Capi- 1999. “Governance Matters.� World Bank Policy Research tal Initiative Working Paper 13. Washington, D.C. Working Paper 2196. Washington, D.C. Krueger, Anne. 1974. “The Political Economy of the Rent-Seek- Kidd, Quentin, and Aie-Rie Lee. 1997. “Postmaterial Values ing Society.� American Economic Review 64(3): 291–303. and the Environment: A Critique and Reappraisal.� Social ———. 1996. “The Political Economy of Controls: American Science Quarterly 78(1):1–15. Sugar.� In Lee J. Alston, Thrainn Eggertsson, and North Kim, Jooseop, Harold Alderman, and Peter Orazem. 1998. Douglass, eds., Empirical Issues in Institutional Change. Cam- “Can Cultural Barriers Be Overcome in Girls’ Schooling? bridge, New York, and Melbourne: Cambridge University The Community Support Program in Rural Balochistan.� Press. World Bank Working Paper Series on Impact Evaluation of Krugman, Paul. 1998. “The Role of Geography in Develop- Education Reforms 10. Washington, D.C. ment.� Paper presented at the Annual World Bank Confer- King, Andrew, and Michael Lennox. Forthcoming. “Exploring ence in Development Economics, Washington, D.C. the Locus of Profitable Pollution Reduction.� Management Kudat, A., and B. Ozbilgin. 1996. “Uzbekistan Water Supply, Science. Sanitation and Health Project: Salinity Taste Tolerance As- King, Elyzabeth, Peter Orazem, and Darin Wohlgemuth. 1999. sessment.� World Bank, Washington, D.C. “Central Mandates and Local Incentives: The Colombia Ed- Kudat, A., B. Ozbilgin, and V. Borisov. 1997. “Reconstructing ucation Voucher Program.� World Bank Economic Review Russia’s Coal Sector.� In Michael M. Cernea and Ayse 13(3):467–91. Kudat, eds., Social Assessments for Better Development: Case King, Elyzabeth, and Berk Özler. 1998. “What’s Decentraliza- Studies in Russia and Central Asia. Washington, D.C.: World tion Got to Do With Learning?� World Bank Working Bank. (c) The International Bank for Reconstruction and Development / The World Bank ����������  Lacey, Robert. 1986. Ford: the Man and the Machine. Boston: Lubbock, A., and Rahman Bourqia. 1998. “Gender and House- Little, Brown. hold Food Security, Survival, Change and Decision- Making Lall, Somik, Uwe Deichmann, and Mattias Lundberg. 2002. in Rural Households: Three Village Case Studies from East- “Tenure, Diversity and Commitment: Community Partici- ern Morocco.� International Fund for Agricultural Research, pation for Urban Service Provision.� World Bank, Washing- Rome. Available on line at http://www.ifad.org/gender/the- ton, D.C. Processed. matic/morocco/mo_ toc.htm. Lall, Somik, and Sudeshna Ghosh. 2002. “Learning by Dining: Lucas, Robert E. B. 1998. “Internal Migration and Urbaniza- Informal Networks and Productivity in Mexican Industry.� tion: Recent Contributions and New Evidence.� World World Bank Policy Research Paper 2789. Washington, D.C. Bank, Washington, D.C. Processed. Lall, Somik, Zmarak Shalizi, and Uwe Deichmann. Forthcom- Lvovsky, Kseniya. 2001. “Health and Environment.� World ing. “Agglomeration Economies and Productivity in Indian Bank Environment Strategy Papers 1. Washington, D.C. Industry.� Journal of Economic Literature. Lvovsky, K., G. Hughes, D. Maddison, B. Ostro, and D. Lambsdorff, Johann Graf. 1999. “Corruption in Empirical Re- Pearce. 2000. “Environmental Costs for Fossil Fuels: A search—A Review.� Transparency International Working Rapid Assessment Method with Application to Six Cities.� Paper. Berlin. Available online at www.transparency. org. World Bank Environment Department Paper 78. Washing- Lanjouw, Jean A., and Peter Lanjouw. 2001. “The Rural Non- ton, D.C. Farm Sector: Issues and Evidence from Developing Coun- Mackay, Keith. 2000. “New Evaluation Trends in Public Pol- tries.� Agricultural Economics 26:1–23. icy Reform and Governance.� Paper presented at a seminar Lanjouw, Jean A., and Philip I. Levy. 1998. “Untitled: A Study and workshop organized by the Development Bank of South of Formal and Informal Property Rights in Urban Ecuador.� Africa, the African Development Bank, and the World Bank, Economic Growth Center, Center Discussion Paper 788. Johannesburg. Yale University, New Haven, Conn. Available online at Mahar, Dennis J., and Cecile E. H. Ducrot. 1998. “Land-Use http://www.library.yale.edu/socsci/egcdp788. pdf. Zoning on Tropical Frontiers: Emerging Lessons from the Lant, M., and G. Sant. 2001. “Patagonian Toothfish: Are Con- Brazilian Amazon.� World Bank Institute EDI Case Study servation and Trade Measures Working?� Traffic Bulletin 19674. Washington, D.C. 19(1):1–18. Mason, Edward S., Mahn Je Kim, Dwight H. Perkins, Kwang Leach, Melissa, and Robin Mearns. 1996. The Lie of the Land, Suk Kim, and David C. Cole. 1980. The Economic and So- Challenging Received Wisdom in African Environmental cial Modernization of the Republic of Korea. Cambridge, Change and Policy. Oxford: James Currey Publishers Ltd. Mass., and London: Harvard University Press. and Heinemann. May, Peter H., Fernando Veiga Neto, Valdir Denardin, and Lee, Su-Hoon, Michael Hsin-Huang Hsiao, Hwa-Jen Liu, On- Wilson Loureiro. Forthcoming. “The ‘Ecological’ Value- Kwok Lai, Francisco Magno, and Alvin Y. So. 1999. “The Added Tax: Municipal Responses in Paraná and Minas Impact of Democratization on Environmental Movements.� Gerais, Brazil.� In Stephano Pagiola, Joshua Bishop, and In Su-Hoon Lee and Alvin Y. So, eds., Asia’s Environmental Natasha Landell Mill, eds., Selling Forest Environmental Ser- Movements: Comparative Perspectives. M.E. Sharpe. vices: Market-Based Mechanisms for Conservation. Leitmann, Joseph. 2001. Sustaining Cities: Environmental Plan- Mazumdar, Dipak. 1987. “Rural-Urban Migration in Develop- ning and Management in Urban Design. New York: Mc- ing Countries.� In Peter Nijkamp, ed., Handbook of Regional Graw-Hill. and Urban Economics Vol. II. Amsterdam: Elsevier Science Lele, Uma, William Lesser, and Gesa Horstkotte-Wesseler, eds. Publishers. 1999. Intellectual Property Rights in Agriculture: The World McCarthy, James J., Osvaldo F. Canziani, Neil A. Leary, David Bank’s Role in Assisting Borrower and Member Countries. J. Dokken, and Kasey S. White. 2001. “Climate Change Washington, D.C.: World Bank. 2001: Impacts, Adaptation, and Vulnerability.� Cambridge: Limão, Nuno, and Anthony J. Venables. 2001. “Infrastructure, Cambridge University Press. Geographical Disadvantage, Transport Costs, and Trade.� McGranahan, Gordon, Pedro Jacobi, Jacob Songsore, Charles The World Bank Economic Review 15(3):451–79. Surjadi, and Marianne Kjellén. 2001. The Citizens at Risk: Lindert, Peter H. 2000. Shifting Ground: The Changing Agri- From Urban Sanitation to Sustainable Cities. London: Earth- cultural Soils of China and Indonesia. Cambridge, Mass.: scan. MIT Press. McGranahan, G., and D. Satterhwaite. 2000. “Environmental Linz, Austria. 2002. “Citizens Participation in the Siting of Health or Ecological Sustainability? Reconciling the Brown Waste Facilities.� Case Study 15. and Green Agendas in Urban Development.� In Cedric Lipton, Michael. 1977. Why Poor People Stay Poor: A Study of Pugh, ed., Sustainable Cities in Developing Countries. Lon- Urban Bias in World Development. Canberra: Australian Na- don: Earthscan. tional University Press. McGuire, M., and Mancur Olson. 1996. “The Economics of Losey J.E., L.S. Rayor, and M.E. Carter. 1999. “Transgenic Autocracy and Majority Rule.� Journal of Economic Litera- Pollen Harms Monarch Larvae.� Nature 399:214. ture 34(1):72–96. Loureiro, Wilson. 1998. “Uma Experiencia Brasileira exitosa no McIlwaine, Cathy, and Caroline O. N. Moser. 2001. “Violence incentivo economico para a conservacao da biodiversidade.� and Social Capital in Urban Poor Communities: Perspec- Cadernos de Biodiversidade 1(2):25–47. tives from Colombia and Guatemala.� Journal of Interna- Lovei, Magda. 1999. “Eliminating a Silent Threat: World Bank tional Development 13(7):965–84. Support for the Global Phaseout of Lead from Gasoline.� McMahon, Gary, and Remy Felix, eds. 2001. Large Mines and World Bank, Washington, D.C. the Community: Socioeconomic and Environmental Effects in (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Latin America, Canada, and Spain. Washington, D.C.: World Moser, Caroline, and Cathy McIlwaine. 2001. Violence in a Bank and International Development Research Center. Post-Conflict Context: Urban Poor Perceptions from Guate- McNeely, Jeffrey R., and Sara J. Scherr. 2001. “Common mala. Washington, D.C.: World Bank. Ground: How Ecoagriculture Can Help Feed the World and Moser, Caroline O. N., and Jeremy Holland. 1995. “A Partici- Save Wild Biodiversity.� International Union for the Con- patory Study of Urban Poverty and Violence in Jamaica.� servation of Nature and Natural Resources—The World World Bank Transportation, Water and Urban Develop- Conservation Union Report 5/01. Available online at ment Department, Washington, D.C. http://www.futureharvest.org/earth/common_ ground_bio. Munn, R. E., A. Whyte, and P. Timmerman, eds. 2000. Emerg- shtml. ing Issues for the 21st Century: A Study for GEO–2000. McNeill, J. R. 2000. Something New under the Sun: An Envi- Nairobi: United Nations Environment Programme, Divi- ronmental History of the Twentieth-Century World. New sion of Environmental Information, Assessment and Early York: W. W. Norton. Warning. Mearns, Robin. 2001. “Contextual Factors in the Management Munton, Don, Marvin Soroos, Elena Nikitina, and Mark A. of Common Grazing Lands: Lessons from Mongolia and Levy. 1999. “Acid Rain in Europe and North America.� In Northwestern China.� Proceedings of the XIX International Oran Young, ed., The Effectiveness of International Enviro- Grassland Congress. mental Regimes: Causal Connections and Behavioral Mecha- ———. 2002. “Taking Stock: Policy, Practice, and Profession- nisms. Cambridge, Mass.: MIT Press. alism in Rangeland Development.� Symposium on Range- Murgai, Rinku, Ali Mubarik, and Derek Byerlee. 2001. “Pro- lands Professionals and Policy. Kansas City, Missouri, Feb- ductivity Growth and Sustainability in Post-Green Revolu- ruary 2002. tion Agriculture: The Case of the Indian and Pakistan Pun- Meitzner, Laura S., and Martin L. Price. 1996. Ideas for Grow- jabs.� World Bank Research Observer 16(2): 199–218. ing Food Under Difficult Conditions. North Fort Myers, Fla.: Murray, C. J., and A. D. Lopez. 1996. The Global Burden of ECHO. Disease. Cambridge, Mass.: Harvard University Press. Mellor, John. 2000. “Faster More Equitable Growth: The Rela- Murshed, S. Mansoob. 2002. “Conflict, Civil War and Under- tion between Growth in Agriculture and Poverty Reduc- development.� Institute of Social Studies. Available online at tion.� Harvard University Consulting Assistance on Eco- http://www.wider.unu.edu/whoswho/murshedpapers.htm. nomic Reform II Discussion Paper 70. Cambridge, Mass. Processed. Available online at http://www. cid.harvard.edu/ caer2. Murshed, S. Mansoob, and Maiju Perälä. 2001. “Does the Type Mernissi, Fatema. 1997. Les Ait Debrouille Haut-Atlas. Rabat: of Natural Resource Endowment Influence Growth?� Editions Le Fennec. UNU/WIDER, Helsinki. Available online at http://www. Millar, David. 1999. “Farmer’s Path of Experimentation: The wider.unu.edu/whoswho/murshedpapers.htm. Processed. PTD Process in Northern Ghana.� ILEIA Newsletter Sep- Myers, Norman, and Jennifer Kent. 2001. Perverse Subsidies: tember: 43–6. How Misused Tax Dollars Harm the Environment and the Milley, P. C. D., R. T. Wetherald, K. A. Dunne, and T. L. Del- Economy. Washington, D.C., Covelo, London: Island Press worth. 2002. “Increasing Risk of Great Floods in a Chang- for International Institute for Sustainable Development. ing Climate.� Nature 415:514–17. Myers, Norman, Russell A. Mittermeier, Cristina G. Mitter- Mills, Edwin S. 2000. “The Importance of Large Urban Areas meier, Gustavo A. B. Fonseca, and Jennifer Kent. 2000. and Governments’ Roles in Fostering Them.� In Shahid “Biodiversity Hotspots for Conservation Priorities.� Nature Yusuf, Weiping Wu, and Simon Evenett, eds., Local Dynam- 403:853–58. ics in an Era of Globalization. Washington, D.C.: World Nakicenovic, N., and R. Swart, eds. 2000. Special Report of the Bank. Intergovernmental Panel on Climate Change on Emission Sce- Mirovitskaya, Natalia. 1998. “The Environmental Movement narios. Cambridge: Cambridge University Press. in the Former Soviet Union.� In Andrew Tickle and Ian NAS (National Academy of Sciences). 1996. “The Ozone De- Welsh, eds., Environment and Society in Eastern Europe. New pletion Phenomenon.� Washington, D.C. Available online at York: Addison Wesley Longman. http://www. beyonddiscovery.org/. Mitchell, Ronald B. 1995. “Compliance with International National Intelligence Council. 2000. “Global Trends 2015: A Treaties: Lessons from Intentional Oil Pollution.� Environ- Dialogue about the Future with Nongovernment Experts. ment 37:10–12, 36–41. NIC 2000–02. Washington, D.C. Moore, Julia A. 2001. “Frankenfood or Doubly Green Revolu- National Research Council. 1999. Our Common Journey: A tion: Europe vs. America on the GMO Debate.� In Albert Transition toward Sustainability. Washington, D.C.: Na- H. Teich, Stephen D. Nelson, Ceilia McEnaney, and tional Academy Press. Stephen J. Lita, eds., AAAS Science and Technology Policy ———. 2001. Abrupt Climate Change: Inevitable Surprises. Yearbook 2001. Washington, D.C.: American Association Washington, D.C.: National Academy Press. for the Advancement of Science. Nelson, Douglas R., and Oliver Morrisey. 1998. “East Asian Moore, Mike. 2002. “Trade Gains Are What the Poor Need.� Economics Performance: Miracle or Just a Pleasant Sur- International Herald Tribune March 14. prise?� World Economy 21(7):855–79. Moser, Caroline, and Emma Grant. 2000. “Violence and Secu- Newell, Richard, and William Pizer. 2001. “Discounting the rity in Urban Areas: Their Implications for Governance, Benefits of Climate Change Mitigation.� Pew Center on Health and Labor Markets.� National Academy of Sciences Global Climate Change, Arlington, Va. Available online at Panel on Urban Population Dynamics. Processed. http://www.pewclimate.org/projects/econ_discounting.cfm. (c) The International Bank for Reconstruction and Development / The World Bank ����������  Newsletter. 1998. “Stories from the Cities 1997–98: Durban- Pagiola, Stefano. 1999. “Global Environmental Benefits of Land Diakonia Council of Churches.� Newsletter July/August Ar- Degradation Control on Agricultural Land.� World Bank ticle 3. Environment Paper 16. World Bank, Washington, D.C. Niamir-Fuller, Maryam. 1998. “The Resilience of Pastoral Pagiola, Stefano, and Ina-Marlene Ruthenberg. 2002. “Selling Herding in Sahelian Africa.� In Fikret Berkes and Carl Biodiversity in a Coffee Cup: Shade-grown coffee and Con- Folke, eds., Linking Social and Ecological Systems. Cam- servation in Mesoamerica.� In Stefano Pagiola, Joshua Bishop, bridge: Cambridge University Press. and Natasha Landell-Mills, eds., Selling Forest Environmen- Nicholls, J. R., F. M. J. Hoozemans, and M. Marchand. 1999. tal Services: Market-Based Mechanisms for Conservation and “Increasing Flood Risk and Wetland Losses Due to Global Development. London: Earthscan. Sea-Level Rise: Regional and Global Analyses.� Global En- Pardey, Philip G., and Nienke M. Beintema. 2001. “Slow vironmental Change 9:S69–87. Magic: Agricultural R&D a Century After Mendel.� Inter- North, Douglas, and Barry Weingast. 1989. “Constitutions and national Food Policy Research Institute, Washington, D.C. Commitment: The Evolution of Institutions Governing Paternostro, Stefano, Jean Razafindravonona, and David Stifel. Public Choice in the Seventeenth-Century England.� Jour- 2001. “Madagascar. Poverty and Socio Economic Develop- nal of Economic History 49(4):803–32. ments: 1993–1999.� World Bank, Africa Region Working Oberai, A. S., and H. K. Manmohan Singh. 1984. “Migration, Paper Series 19. Washington, D.C. Employment and the Urban Labor Market: A Study of the Pavot, W. 1991. “Further Validation with the Satisfaction of Indian Punjab.� International Labor Review 123(4): 507–23. OECD (Organisation for Economic Co-operation and Devel- Life Scale: Evidence for the Cross-Method Convergence of opment). 1988. “Economic Instruments for Pollution Con- Well-Being Measures.� Journal of Personality 57(1):149–61. trol and Natural Resources Management.� OECD Countries: Payne, Geoffrey. 2001. “Urban Land Tenure Policy Options: A Survey. Document ENV/EPOC/GEEI (98)35/REV1/ Titles or Rights?� Habitat International 25(3):415–29. FINAL. Payne, Geoffrey, ed. 2002. Land, Rights and Innovations: Im- ———. 1997. “The Environmental Effects of Agricultural proving Tenure Security for the Urban Poor. London: ITDG Land Diversion Schemes.� Paris. Publishing. ———. 1998. “Water Management: Performance and Chal- Payne, Geoffrey, and Edesio Fernandes. 2001. “Legality and Le- lenges in OECD Countries.� gitimacy in Urban Tenure Issues.� Lincoln Institute of Land ———. 1999. Voluntary Approaches for Environmental Policy— Policy Working Paper WP01GP1. Assessment. Paris. Peñalosa, Enrique. 2001. “The Livable City: Experiences in Bo- ———. 2001a. Agricultural Policies in OECD Countries: Moni- gota, Colombia.� Paper presented at a seminar on Empower- toring and Evaluation. Paris. ment, Culture and Civic Engagement, July 24–25. World ———. 2001b. “OECD Environmental Outlook.� Paris. Bank, Washington, D.C. Unpublished proceedings. ———. 2001c. Sustainable Development: Critical Issues. Paris. Pender, John, and Peter Hazell. 2000. “Promoting Sustainable OED (Operations Evaluation Department). “Community Development in Less-Favored Lands.� Focus 4: (brief 9 Forestry in Nepal.� Précis. World Bank, Washington, D.C. of 9). Ojima, Dennis. 2001. “Critical Drivers of Global Environmen- Perlman, Janice. 2002. “The Metamorphosis of Marginality: tal and Land Use Changes in Temperate Asia.� Paper pre- Rio’s Favelas 1969–2002.� The Mega-Cities Project and sented at the Open Symposium on Change and Sustainabil- Trinity College. Presentation at the World Bank, May 7. ity of Pastoral Land Use Systems in Temperate and Central Washington, D.C. Asia. Ulaanbaatar, Mongolia. Persson, Torstein, and Guido Tabellini. 1994. “Representative Olson, David M., and Eric Dinerstein. 1998. “The Global 200: Democracy and Capital Taxation.� Journal of Public Eco- A Representation Approach to the Conserving the Earth’s nomics 55(1):53–70. Distinctive Ecoregions.� Conservation Science Program, Pezzey, J. 1989. “Economic Analysis of Sustainable Growth and World Wildlife Fund–U.S., Washington, D.C. Processed. Sustainable Development.� World Bank Environment De- Olson, Mancur. 1996. “Distinguished Lecture on Economics partment Working Paper 15. Washington, D.C. in Government. Big Bills left on the Sidewalk: Why Some Pinstrup-Andersen, Per, Rajul Panya-Lorch, and Mark W. Nations are Rich and Others Poor.� Journal of Economic Per- Rosegrant. 1999. “World Food Prospects: Critical Issues for spectives 10(2):3–24. O’Rourke, Kevin, and Jeffrey Williamson. 1995. “Around the the Early Twenty-First Century.� International Food Policy European Periphery: 1870 to 1910.� National Bureau of Research Institute, Washington, D.C. Economic Research Working Paper 5392. Cambridge, Mass. Pollard, Sidney. 1997. Marginal Europe: The Contribution of the Oshima, Harry T. 1987. Economic Growth in Monsoon Asia: A Marginal Lands since the Middle Ages. Oxford: Clarendon Comparative Survey. Tokyo: University of Tokyo Press. Press. Ostrom, Elinor, and Roy Gardner. 1993. “Coping with Asym- Posner, Richard A. 1981. The Economics of Justice. Cambridge, metries in the Commons: Self-Governing Irrigation Systems Mass., and London: Harvard University Press. Can Work.� Journal of Economic Perspectives 7(4):93–112. Postel, Sandra L., and Aaron T. Wolf. 2001. “Dehydrating Oswald, A. J. 1997. “Happiness and Economic Performance.� Conflict.� Foreign Policy (September/October):2–9. Economic Journal 107(445):1815–31. President’s Committee of Advisors on Science and Technology. Paarlberg, Robert L. 2001. The Politics of Precaution: Genetically 1997. “Report to the President on Federal Energy Research Modified Crops in Developing Countries. Baltimore: Johns and Development for the Challenges of the Twenty-First Hopkins University Press. Century.� Washington, D.C. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Pretty, Jules N. 1995. Regenerating Agriculture: Policies and Rodden, Jonathan, Gunnar S. Eskeland and Jennie Litvack. Practice for Sustainability and Self-Reliance. Washington, Forthcoming. Fiscal Decentralization and the Challenge of D.C.: Joseph Henry Press. Hard Budget Constraints. Cambridge, Mass.: MIT Press. Primavera, J. Honculada. 1994. “Shrimp Farming in the Asia- Rodriguez, Francisco, and Jeffrey D. Sachs. 1999. “Why Do Re- Pacific: Environmental and Trade Issues and Regional source Abundant Economies Grow More Slowly? A New Ex- Cooperation.� Nautilus Institute Workshop on Trade and planation and an Application to Venezuela.� Journal of Eco- Environment in Asia-Pacific: Prospects for Regional Coop- nomic Growth 4(3):277–303. eration, September 23–5. Berkeley, Calif. Available online Rodrik, Dani. 1996. “Understanding Economic Policy Re- at http://www.nautilus.org/ papers/enviro/trade/shrimp.html. form.� Journal of Economic Literature 34(1):9–41. Processed. ———. 1999. “Where Did All the Growth Go? External Prud’homme, Remy. 1994. “On the Economic Role of Cities.� Shocks, Social Conflict, and Growth Collapses.� Journal of Paper presented at Cities and the New Global Economy, in Economic Growth 4(4):385–412. Melbourne Nov. 20–23. The Centre for Developing Cities, ———. 2002. “Institutions, Integration, and Geography: In Search of Deep Determinants of Economic Growth.� John F. University of Canberra and Organisation for Economic Co- Kennedy School of Government, Harvard University, Boston, operation and Development (endnote 4); available online at Mass. Available online at http://ksghome.harvard.edu/~.dro- http://cities/publications/ OECDpaper/endnotes.htm drik.academic.ksg/papers.html. Processed. Pugh, Cedric, ed. 1996. Sustainability, the Environment, and ———, ed. Forthcoming. In Search of Prosperity: Analytic Nar- Urbanization. London: Earthscan. ratives on Economic Growth. Princeton, N.J.: Princeton Uni- Quigley, John M. 1998. “Urban Diversity and Economic versity Press. Growth.� Journal of Economic Perspectives 12(2):127–138. Roe, Dilys, James Mayers, Maryanne Grieg-Gran, Ashish Raff, Daniel M. G., and Lawrence Summers. 1986. “Did Henry Kothari, Christo Fabricius, and Ross Hughes. 2000. Evalu- Ford Pay Efficiency Wages?� Working Paper Series 2101. ating Eden: Exploring the Myths and Realities of Community National Bureau of Economic Research, Cambridge, Mass. Based Wildlife Management. London: International Institute Rao, Vijayendra. 2002. “Community Driven Development: A for Environment and Development. Brief Review of the Research.� World Bank, Washington, Rolnik, Raquel. 1999. “Territorial Exclusion and Violence: The D.C. Processed. Case of São Paulo, Brazil.� Comparative Urban Studies Oc- Rao, Vijayendra, and Michael Woolcock. 2001. “Social Net- casional Papers 26. Woodrow Wilson International Center works and Risk Management Strategies in Poor Urban Com- for Scholars. munities: What Do We Know?� World Bank, Washington, Rose-Ackerman, Susan. 1999. Corruption and Government: D.C. Processed. Causes, Consequences, and Reform. New York: Cambridge ———. 2002. “Networks, Mobility and Survival in Delhi University Press. Slums: A Mixed-Method Analysis.� World Bank, Washing- Rosegrant, Mark W., Michael S. Paisner, Meijer Siet, and Julie ton, D.C. Processed. Witcover. 2001. “2020 Global Food Outlook: Trends, Alter- Ravallion, Martin, and Gaurav Datt. 1996. “How Important to natives, and Choices.� International Food Policy Research In- India’s Poor Is the Sectoral Composition of Economic stitute, Washington, D.C. Growth?� The World Bank Economic Review 10(1):1–26. Rosenweig, C., and W. D. Solecki, eds. 2000. Climate Change Reardon, Thomas, Christopher Barrett, Valerie Kelly, and Kim- and Global City: Two Metropolitan East Coast Regional Asses- seyinga Savadogo. “Policy Reforms and Sustainable, Agricul- ments. New York: Columbia Earth Institute. tural Intensification in Africa.� Development Policy Review Rosner, D., and G. Markowitz. 1985. “The Public Health Con- (U.K.) 17(4):375–95. troversy over Leaded Gasoline during the 1920s.� American Rees, Williams. 1997. “Is Sustainable City an Oxymoron?� Journal of Public Health 75:344–52. Local Environment 2(3):303–10. Ross, Michael. 2001a. “How Does Natural Resource Wealth In- fluence Civil War?� University of California at Los Angeles Reid, Catherine, and Lawrence Salmen. 2000. “Understanding Political Science Department, Los Angeles. Available online Social Capital. Agricultural Extension in Mali: Trust and So- at: http://www.eireview.org/. Processed. cial Cohesion.� World Bank Social Capital Initiative Work- Ross, Michael. 2001b. Timber Booms and Institutional Breakdown ing Paper 22. Washington, D.C. in Southeast Asia. Cambridge: Cambridge University Press. Reij, Chris P., I. Scoones, and C. Toulmin. 1996. Sustaining Ross-Larson, Bruce. 1980. “Social and Political Setting.� In the Soil; Indigenous Soil and Water Conservation in Africa. Kevin Young, Willen C. M. Bussink, and Parvez Hasan, London: Earthscan. eds., Malaysia: Growth and Equity in a Multiracial Society. Reilly, Charles A., eds. 1995. New Paths to Democratic Develop- Washington, D.C.: International Bank for Reconstruction ment in Latin America: The Rise of NGO-Municipal Collabo- and Development. ration. Boulder, Colo.: Lynne Rienner Publishers. Rutherford, Stuart. 2000. The Poor and Their Money. New Reinikka, Ritva, and Jakob Svensson. 2002. “Explaining Leak- Delhi: Oxford University Press. age of Public Funds.� Centre for Economic Policy Research Ruttan, V. 1990. “Models of Agricultural Development.� In C. Discussion Paper Series 3227:1–45. United Kingdom. and J. Staatz Eicher, eds., Agricultural Development in the Richardson, Peter. 2001. “Corruption.� In P. J. Simmons and Third World. Baltimore: Johns Hopkins University Press. Chantal De Jonge Oudraat, eds., Managing Global Issues: Ryan, B., and Quentin Wodon. 2001. “Assessing the Realism Lessons Learned. Washington, D.C.: Carnegie Endowment of International Development Goals.� World Bank, Wash- for International Peace. ington, D.C. Processed. (c) The International Bank for Reconstruction and Development / The World Bank ����������  Sachs, Jeffrey D., and Andrew M. Warner. 1995. “Economic Sen, Amartya. 1983. “Development: Which Way Now?� Eco- Convergence and Economic Policies.� National Bureau of nomic Journal 93(372):742–62. Economic Research Working Paper 5039. Cambridge, Mass. ———. 1999. Development as Freedom. New York: Anchor ———. 1999a. “Natural Resource Intensity and Economic Books. Growth.� In Jörg Mayer, Brian Chambers, and Ayisha Serageldin, Ismael, Ephim Shluger, and Joan Martin-Brown. Farooq, eds., Development Policies in Natural Resources Econ- 2001. Historic Cities and Sacred Sites: Cultural Roots for omies. Cheltenham, U.K.: Edward Elgar. Urban Futures. Washington, D.C.: World Bank. ———. 1999b. “The Big Push, Natural Resource Booms and Sevilla, Manuel. 2000. “Participación privada en el mejo- Growth.� Journal of Development Economics 59(1): 43–76. ramiento urbano: Sondeo de cuatro casos en El Salvador.� Salamini, Francesco. 1999. “North-South Innovation Transfer.� SACDEL, San Salvador. Nature Biotechnology 17 (Supplement) 11–12. Shah, Tushaar. 1993. Groundwater Markets and Irrigation De- Sambanis, Nicolas. 2000. “Partition as a Solution to Ethnic velopment. Mumbai: Oxford University Press. War: An Empirical Critique of the Theoretical Literature.� Shalizi, Zmarak, and Christine Kraus. 2001. “Globalization, World Politics 52:437–483. Openness and the Environment.� World Bank, Washing- Sanchez, Pedro. 2002. “Soil Fertility and Hunger in Africa.� ton, D.C. Processed. Science 295(5562): 2019–20. Shlomo, Angel. 2000. Housing Policy Matters. Oxford: Oxford Sanchez, P. A., Bashir Jama, Amadou I. Niang, and Cheryl A. University Press. Palm. 2001. “Soil Fertility, Small-farm Intensification and Simpson, R. David, Roger A. Sedjo, and John W. Reid. 1996. the Environment in Africa.� In D. R. Lee and C. B. Barret, “Valuing Biodiversity for Use in Pharmaceutical Research.� eds., Tradeoffs or Synergies? Agricultural Intensification, Eco- Journal of Political Economy 104(1):163–185. nomic Development and the Environment. New York: CABI Singh, Jas, and Carol Mulholland. 2000. “DSM in Thailand: a Publishing. Case Study.� World Bank Energy Sector Management Assis- Sarraf, Maria, and Moortaza Jiwanji. 2001. “Beating the Re- tance Programme Technical Series 008. Washington, D.C. source Curse: The Case of Botswana.� World Bank Environ- Skaperdas, Stergios. 1992. “Cooperation, Conflict, and Power ment Department Papers 83. Washington, D.C. in the Absence of Property Rights.� American Economic Re- Sassen, Saskia. 2001. The Global City: New York, London, Tokyo. view 82(4):720–739. Princeton, N.J.: Princeton University Press. Skees, Jarry, Panos Varangis, Donald Larson, and Paul Siegel. Scheffer, Marten, Steve Carpenter, Jonathan A. Foley, Carl 2002. “Can Financial Markets Be Tapped to Help Poor Peo- Folke, and Brian Walker. 2001. “Catastrophic Shifts in Eco- ple Cope with Weather Risks?� World Bank Policy Research systems.� Nature 413: 591–596. Working Paper 2812. Washington, D.C. Scherr, Sara. 1999. “Soil Degradation—A Threat to Develop- Smith, Adam. 1776/1981. An Inquiry into the Nature and ing Country Food Security by 2020?� Food, Agriculture, Causes of the Wealth of Nations. R. H. Campbell and A. S. and the Environment Discussion 27. International Food Skinner, eds. Indianapolis: Liberty Classics. Policy Research Institute, Washington, D.C. Smith, Kirk R. 1998. “Indoor Air Pollution in India: National Scherr, Sara J., A. White, and D. Kaimowitz. 2002. “Strategies to Improve Rural Livelihoods through Markets for Forest Health Impacts and the Cost-Effectiveness of Intervention.� Products and Services.� Forest Trends and Center for Inter- Goregaon, Mumbai. national Forestry Research, Washington, D.C. Social Investment Forum. 2001. “2001 Report on Socially Re- Scherr, Sara, and S. Yadav. 1996. “Land Degradation in the De- sponsible Investing Trends in the United States.� Social In- veloping World: Implications for Food, Agriculture, and the vestment Forum. Washington, D.C. Available online at Environment to 2020.� International Food Policy Research www.socialinvest.org. Institute Discussion Paper 14. Washington, D.C. Sokoloff, Kenneth L., and Stanley L. Engerman. 2000. “History Schipper, Lee, Scott Murtishaw, and Fridtjof Unander. 2001. Lessons: Institutions, Factor Endowments, and Paths of De- “International Comparisons of Sectoral Carbon Dioxide velopment in the New World.� Journal of Economic Perspec- Emissions Using a Cross-Country Decomposition Tech- tives 14(3):217–232. nique.� Energy Journal 22(2):35–75. Solow, Robert. 2000. “Notes on Social Capital and Economic Schneider, Robert R. 1995. “Government and Economy on the Performance.� In Partha Dasgupta and Ismail Serageldin, Amazon Frontier.� World Bank Environment Working eds., Social Capital: A Multifaceted Perspective. Washington, Paper 11. Washington, D.C. D.C.: World Bank. Scholz, Ulrich. 1985. “Types of Spontaneous Settlement in Sorensen, Jens. 2002. “Baseline 2000 Background Report: The Thailand.� In Walther Manshard and William B. Morgan, Status of Integrated Coastal Management as an International eds., Agricultural Expansion and Pioneer Settlements in the Practice.� Harbor and Coastal Center, University of Massa- Humid Tropics. Tokyo: United Nations University Press. chusetts, Boston. Available online at http://www. uhi.umb. Scoones, Ian, eds. 1994. Living with Uncertainty: New Direc- edu/b2k/baseline 2000.pdf. tions in Pastoral Development in Africa. London: Intermedi- Souza, Celina. 2001. “Participatory Budgeting in Brazilian Cities: ate Technology Development Group Publishing. Limits and Possibilities in Building Democratic Institutions.� Sebastian, Iona, Kseniya Lvovsky, and de Henk Koning. 1999. Environment and Urbanization 13(1): 159–184. “Decision Support System for Integrated Pollution Control: Sparks, Allistar. 1996. Tomorrow Is Another Country: The Inside Software for Education and Analysis of Pollution Manage- Story of South Africa’s Road to Change. Chicago: University ment. User Guide.� World Bank, Washington, D.C. of Chicago Press. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  State Environmental Protection Agency (China). 2001. “New and D. Swackhamer. 2001. “Forecasting Agriculturally Dri- Countermeasures for Air Pollution Control in China, Final ven Global Environmental Change.� Science 292:281–284. Report.� Beijing. Timmer, C. Peter. 1997. “How Well Do the Poor Connect to the Steinberg, David Joel, ed. 1987. In Search of Southeast Asia: A Growth Process?� Harvard University Consulting Assistance Modern History. Honolulu: University of Hawaii Press. on Economic Reform II Discussion Paper 17. Cambridge, Steinberg, Paul F. 2001. Environmental Leadership in Develop- Mass. Available online at http://www.cid.harvard.edu/caer2. ing Countries: Transnational Relations and Biodiversity Policy Tomich, Thomas P., Peter Kilby, and Bruce F. Johnson. 1995. in Costa Rica and Bolivia. Cambridge, Mass.: MIT Press. Transforming Agrarian Economies: Opportunities Seized, Op- Stephens, C., M. Akerman, S. Avle, P. B. Maia, P. Campanario, portunities Missed. Ithaca, N.Y. and London: Cornell Uni- B. Doe, and D. Tetteh. 1997. “Urban Equity and Urban versity Press. Health: Using Existing Data to Understand Inequalities in Tomich, Thomas P., M. van Noordwijk, S. Budidarsono, A. Health and Environment in Accra, Ghana and São Paulo, Gillion, T. Kusumanto, M. Murdiyarso, F. Stolle, and A. M. Brazil.� Environment and Urbanization 9(1):181–202. Fagi. 1998. “Alternatives to Slash-and-Burn in Indonesia: Stephens, G. Ross, and Nelson Wikstrom. 2000. Metropolitan Summary Report and Sysnthesis of Phase II.� Alternatives Government and Governance: Theoretical Perspectives, Empir- to Slash-and-Burn Programme Indonesia Report 8. CGIAR, ical Analysis, and the Future. New York: Oxford University Bogor. Available online at http://www. asb.cgiar.org/CR_ Press. Indonesia.shtm. Stewart, Frances. 2000. “Crisis Prevention: Tackling Horizontal TI (Transparency International). 2000. Confronting Corruption: Inequalities.� Oxford Development Studies 28(3): 245–62. The Elements of a National Integrity System. Berlin. Stone, Wendy. 2001. “Measuring Social Capital: Towards a Tullock, Gordon. 1975. “The Transitional Gains Trap.� Bell Theoretically Informed Measurement Framework for Re- Journal of Economics 6(2):671–78. searching Social Capital in Family and Community Life.� Turner, John F. C., and Robert Fichter, eds. 1972. Freedom to Australian Institute of Family Studies Research Paper 24. Build. New York: Macmillan. Melbourne, Australia. Uhilg, Harald. 1988. “Spontaneous and Planned Settlement in Stott, Peter, and J. A. Kettleborough, 2002. “Origins and Esti- South East Asia.� In Walther Manshard and William B. mates of Uncertainty in Predictions of Twenty-First Cen- Morgan, eds., Agricultural Expansion and Pioneer Settlements in the Humid Tropics. Tokyo: United Nations University tury Temperature Rise.� Science 416:723–725. Press. Streets, David G., Kejun Jiang, Xiulian Hu, Jonathan E. Sinton, United Nations. 1999. “World Urbanization Prospects: The Xiao-Quan Zhang, Deying Xu, Mark Z. Jacobson, and James 1999 Revision. Part 2: Urban Agglomerations.� New York. E. Hansen. 2001. “Recent Reductions in China’s Greenhouse United Nations–Habitat, eds. 1996. An Urbanizing World: Gas Emissions.� Science 294:1835–1837. Global Report on Human Settlements. New York: United Na- Struyk, Raymond J. 1997. Restructuring Russia’s Housing Sector: tions Center for Human Settlements (Habitat), and Oxford 1991–1997. Washington, D.C.: The Urban Institute. University Press. Sugden, Robert. 1986. The Economics of Rights, Cooperation, ———. 1998. “Global Urban Indicators Database.� Nairobi. and Welfare. Oxford and New York: Basil Blackwell. ———. 1999. “Global Campaign for Secure Tenure.� Nairobi. Summers, Robert, and Alan Heston. 1991. “Penn World Ta- Pamplet. bles (Mark 5): An Expanded Set of International Compar- United Nations–Habitat and Urban Management Programme. isons, 1950–1988.� Quarterly Journal of Economics 106(2): 2001. “City Development Strategy: Consolidation of 327–368. Lessons from UMP/UNCHS Experience.� Paper presented SustainAbility Ltd., and United Nations Environment Pro- at Cities Alliance Public Policy Forum, Kolkuta. December gramme. 2001. Buried Treasure: Uncovering the Business Case 9–12. for Corporate Sustainability. London: SustainAbility. UNDP (United Nations Development Programme). 1997. Suvanto, Tina. 2000. “Social Capital and Value Creation: A “Aridity Zones and Dryland Populations: An Assessment of Theoretical Approach.� 47891N. Helsinki. Population Levels in the World’s Drylands.� UNDP and Svensson, Jakob. 1998. “Investment, Property Rights and Polit- United Nations Office to Combat Desertification and ical Instability: Theory and Evidence.� European Economic Drought, Nairobi. Review 42(7):1317–1341. UNDP (United Nations Development Programme), UNEP Swearingen, Will D., and Abdellatif Bencherifa, eds. 1996. (United Nations Environment Programme), World Bank, North African Environment at Risk: State, Culture and Soci- and WRI (World Resources Institute). 1999. “World Re- ety in Arab North Africa. Boulder, Colo.: Westview Press. sources 1998–99: A Guide to the Global Environment.� Syrquin, Moshe. 1989. “Patterns of Structural Change.� In H. ———. 2000. World Resources 2000–2001: People and Ecosys- Chenery and T. N. Srinivasan, eds., Handbook of Develop- tems, the Fraying Way of Life. Oxford: Elsevier Science Ltd. ment Economics Vol. I. Amsterdam: North Holland. UNEP (United Nations Environment Programme). 1992. Tendler, Judith. 1997. Good Governments in the Tropics. Balti- World Atlas of Desertification. New York: Oxford University more: Johns Hopkins University Press. Press. Ter-Minassian, Teresa, ed. 1997. Fiscal Federalism in Theory ———. 1997. World Atlas of Desertification. New York: Oxford and Practice. Washington, D.C: International Monetary University Press. Fund. ———. 1999. “Synthesis of the Reports of the Scientific, Envi- Tilman, D., J. Fargione, B. Wolff, C. D’Antonio, A. Dobson, ronmental Effects, and Technology and Economic Assess- R. Howarth, D. Schindler, W. H. Shlesinger, D. Simberloff, ment Panels of the Montreal Protocol: A Decade of Assess- (c) The International Bank for Reconstruction and Development / The World Bank ����������  ment for Decision Makers Regarding the Protection of the White, Andy, and Alejandra Martin. 2002. “Who Owns the Ozone Layer: 1998–1999.� Nairobi. World’s Forests? Forest Tenure and Public Forests in Tran- UNEP (United Nations Environment Programme) and GRID sition.� Forest Trends, Washington, D.C. Processed. (Global Resource Information Database)–Arendal. 1997. As- White, Robin, Siobhan Murray, and Mark Rohweder. 2002. sessment of Human Induced Soil Degradation (GLASOD) Ex- “Pilot Analysis of Global Ecosystems (PAGE): Grasslands tract of West Africa Dataset. GRID-Arendal online GIS and Ecosystems.� World Resources Institute, Washington, D.C. Maps database. Available online at www.grida.no. Available online at http://www.wri.org/wr2000/grasslands_ USDA (U.S. Department of Agriculture). 2001. Food and Agri- page.html. cultural Policy: Taking Stock for the New Century. Washing- Wilkinson, Clive R., ed. 2000. Status of Coral Reefs of the World. ton, D.C. Available online at http://www.usda. gov. Queensland, Australia: Coral Reef Monitoring Center. U.S. Embassy in China. 1998. “The Fading of Chinese Envi- Williams, Martin A. J. 2001. “Interactions of Desertification ronmental Secrecy.� U.S. Embassy in China, Beijing. and Climate: Present Understanding and Future Research Uzzi, B. 1997. “Social Structure and Competition in Interfirm Imperatives.� Arid Lands Newsletter 49 (May– June). Networks: The Paradox of Embeddedness.� Administrative Williamson, Jeffrey. 1988. “Migration and Urbanization.� In Science Quarterly 29(4):598–621. Hollis Chenery and T. N. Srinivasan, eds., Handbook of De- Velásquez, Luz Stella. 1998. “Agenda 21: A Form of Joint En- velopment Economics Vol. II. Amsterdam: Elsevier Science vironmental Management in Manizales, Colombia.� Envi- Publishers. ronment and Urbanization 10(2):9–36. ———. 1997. “Growth, Distribution and Demography: Some Veríssimo, C. Souza Jr., R. Salomão, and R. Barreto. 2000. Lessons From History.� National Bureau of Economic Re- “Identificação de areas com potencial para a criação de flo- search Working Paper 6244. Cambridge, Mass. restas públicas de produção na Amazônia legal.� Food and Williamson, Oliver E. 2000. “The New Institutional Econom- Agriculture Organization of the United Nations. Processed. ics: Taking Stock, Looking Ahead.� Journal of Economic Lit- Victor, David G., Kal Raustiala, and Eugene B. Skolnikoff. erature 38(3):595–613. 1998. The Implementation and Effectiveness of International Willmann, Rolf, Pongpat Boonchuwong, and Somying Environmental Commitments: Theory and Practice. Cam- Piumsombun. 2002. “Fisheries Management Costs in Thai Marine Fisheries.� Food and Agriculture Organization of the bridge, Mass.: MIT Press. United Nations and Kasetsart University Campus, Bangkok. Wambugu, Florence. 1999. “Why Africa Needs Agricultural Processed. Biotech.� Nature 400:15–16. Willoughby, Christopher. 2000. “Singapore’s Experience in Wang, Hua, Jinnan Wang, Genfa Lu, David Wheeler, and Jun Managing Motorization, and Its Relevance to Other Coun- Bi. Forthcoming.“Public Ratings of Industry’s Environmen- tries.� World Bank, Transportation, Water, and Urban De- tal Performance: China’s Greenwatch Program.� World velopment Discussion Paper 43. Washington, D.C. Bank Policy Research Working Paper series. Washington, Wilson, David, Andrew Whiteman, and Angela Tormin. 2001. D.C. “Strategic Planning Guide for Municipal Solid Waste Man- Ward, Peter M. 1998. “International Forum on Regularization agement.� World Bank, Washington D.C. and Land Markets.� Land Lines July. Available online at Wilson, Edward, and Stephen Kellert, eds. 1994. The Biophilia www.lincolninst.com. Hypothesis. Washington, D.C.: Island Press. Watson, D., and L. A. Clark. 1991. “Self Versus Peer Ratings Wily, Liz, and Peter A. Dewees. 2001. “From Users to Custo- of Specific Emotional Traits: Evidence of Convergent and dians: Changing Relations between People and the State in Discriminant Validity.� Journal of Personality and Social Psy- Forest Management in Tunisia.� World Bank Policy Re- chology 60(6):927–940. search Working Paper 2569. Washington, D.C. Watson, Robert T., Ian R. Noble, Bert Bolin, N. H. Ravin- Wodon, Quentin, Rodrigo Castro, Kihoon Lee, Gladys Lopez- dranath, David J. Verardo, and David J. Dokken, eds. 2000. Acevedo, Corinne Siaens, Carlos Sobrado, and Jean-Philippe Land Use, Land-Use Change, and Forestry: A Special Report Tre. 2001. “Poverty in Latin America: Trends (1986–1998) of the IPCC. Cambridge: Cambridge University Press. Avail- and Determinants.� Cuadernos de Economía 114:127–153. able online at http://www.grida.no/climate/ipcc/land_use/. Wodon, Quentin, and Gonzalez G. Konig. 2001. “The Impact Weber-Fahr, Monika. 2002. “Treasure or Trouble? Mining in of Remittances on Income Distribution.� World Bank, Latin Developing Countries.� World Bank and the International America and Caribbean Region, Washington, D.C. Processed. Finance Corporation. Available online at http:// www.eire- Wood, Stanley, Freddy Nachtergaele, Daniel Nielsen, and view.org/eir/eirhome.nsf Aiguo Dai. 1999. “Spatial Aspects of the Design and Target- Weitzman, Martin L. 1998. “Why the Far-Distant Future ing of Agricultural Development Strategies.� International Should Be Discounted at Its Lowest Possible Rate.� Journal Food Policy Research Institute Discussion Paper 44. Wash- of Environmental Economics and Management 36(3): 201–8. ington, D.C. Available online at http://www. ifpri.org/. Wells, Michael, Scott Guggenheim, Asmeen Khan, Wahjudi Wood, Stanley, Kate Sebastian, and Sara J. Scherr. 2000. “Pilot Wardojo, and Paul Jepson. 1999. “Investing in Biodiversity: Analysis of Global Ecosystems: Agroecosystems.� Washing- A Review of Indonesia’s Integrated Conservation and Devel- ton D.C.: International Food Policy Research Institute and opment Projects.� World Bank, Washington, D.C. World Resources Institute. Wheeler, James O., Yuko Aoyama, and Barney Wolf, eds. 2000. Woolcock, Michael, Lant Pritchett, and Jonathan Isham. 2001. Cities in the Telecommunications Age: The Fracturing of Ge- “The Social Foundations of Poor Economic Growth in Re- ography. New York: Routledge. source-Rich Countries.� In R. M. Auty, eds., Resource Abun- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  dance and Economic Development. New York: Oxford Univer- ———. 2001d. “Implementation Completion Report No. sity Press. 22390–TUN: North West Mountainous Areas Development World Bank. 1991. “Growth, Poverty Alleviation, and Im- Project.� Washington, D.C. Processed. proved Income Distribution in Malaysia: Changing Focus ———. 2001e. “Indonesia: Environment and Natural Resource of Government Policy Intervention.� Report 8667–MA. Management in a Time of Transition.� Environment and Washington, D.C. Social Development Unit Report. Washington, D.C. ———. 1992a. “Malaysia: Fiscal Reform for Stable Growth.� ———. 2001f. Making Sustainable Commitments: An Environ- Report 10120–MA. Washington, DC. ment Strategy for the World Bank. Washington, D.C: World ———. 1992b. World Development Report 1992: Development Bank. and the Environment. New York: Oxford University Press. ———. 2001g. “Trade and Foreign Exchange Policies in Iran: ______. 1993. “Indigenous Peoples.� Operational Directive Reform Agenda, Economic Implications and Impact on the 4.2. Poor.� Processed. ——— 1994. World Development Report 1994: Infrastructure ———. 2001h. World Development Indicators. Washington, for Development. New York: Oxford University Press. D.C., and London: World Bank. ———. 1996a. “Sustainable Transport: Priorities for Policy Re- ———. 2001i. World Development Report 2000/2001: Attack- form.� Development in Practice Series. Washington, D.C. ing Poverty. New York: Oxford University Press. ———. 1996b. “Technical Assistance.� Operations and Evalu- ———. 2002a. “A Revised Forest Strategy for the World Bank ation Department Lessons and Practices No. 7. Washington, Group.� Washington, D.C. D.C. ——— 2002b. A Case for Aid: Building Consensus for Develop- ———. 1997a. “Five Years after Rio: Innovations in Environ- ment Assistance. Washington, D.C. mental Policy.� Environmentally Sustainable Development ———. 2002c. “China: Opportunities to Improve Energy Ef- Studies and Monographs Series 18. Washington, D.C. ficiency in Buildings.� World Bank Discussion Paper, ———. 1997b. World Development Report 1997: The State in a Washington, D.C. Changing World. New York: Oxford University Press. ———. 2002d. “Cities on the Move: A World Bank Urban ———. 1998a. “Aral Sea Basin Program (Kazakhstan, Kyrgyz Transport Strategy Review.� Washington, D.C. Processed. Republic, Turkmenistan, and Uzbekistan): Water and Envi- ———. 2002e. “Global Development Finance: Financing the ronmental Management Project Document. Washington, Poorest Countries 2002.� Washington, D.C. D.C. World Bank. 2002ee. Global Economic Prospects. Washington, ———. 1998b. Assessing Aid: What Works, What Doesn’t and D.C. Why. New York: Oxford University Press. ———. 2002f. “Global Poverty Monitoring Website.� Wash- ———. 1998c. “Post-Conflict Reconstruction.� Washington, ington, D.C., World Bank. Available online at http://www. D.C. worldbank.org/research/povmonitor/index.htm. ———. 1999. World Development Report 1998/1999: Knowl- ———. 2002g. Globalization, Growth, and Poverty: Building edge for Development. New York: Oxford University Press. an Inclusive World Economy. New York: Oxford University ———. 2000a. “Bangladesh: Climate Change and Sustainable Press. Development.� Report 21104–BD.� Washington, D.C. ———. 2002h. “Mexico Low-Income Housing: Issues and Op- ———. 2000b. “Global Economic Prospects 2000.� Washing- tions.� Report 22534–ME. Washington, D.C. Processed. ton, D.C. ———. 2002i. “Prototype Carbon Fund, First Annual Report ———. 2000c. Global Development Finance. Washington, D.C. and Project Concept Note, Liepaja Region Solid Waste Man- ———. 2000d. Greening Industry: New Roles for Communities, agement Project.� Washington, D.C. Processed. Markets, and Governments. New York: Oxford University ———. 2002j. “Public Expenditure and Institutional Review Press. of the Dominican Republic.� Washington, D.C. Processed. ———. 2000e. India: Reducing Poverty, Accelerating Develop- ———. 2002k. “Reaching the Rural Poor: An Updated Strategy ment. New Delhi: Oxford University Press and World Bank. for Rural Development.� Washington, D.C. Processed. ———. 2000f. “Partners in Transforming Development: New ———. 2002l. Rural Development Strategy. Washington, D.C. Approaches to Developing Country-Owned Poverty Reduc- ———. 2002m. “The World Bank’s Experience with Institu- tion Strategies.� Available online at http://www.poverty/ tional Development.� Operations and Evaluation Depart- strategies/prspbroc.pdf. ment, Lessons and Practices 14. Washington, D.C. ———. 2000g. The Quality of Growth. New York: Oxford Uni- ———. 2002n. “Water Resources Sector Strategy: Strategic Di- versity Press. rections for World Bank Engagement.� Washington, D.C. ———. 2000h. World Development Report 1999/2000: Enter- Processed. ing the 21st Century. New York: Oxford University Press. ———. 2002o. World Development Report 2002: Building In- ———. 2001a. “China: Air, Land and Water—Environmental stitutions for Markets. New York: Oxford University Press. Priorities for a New Millennium.� Washington, D.C. ———. Forthcoming. “Johannesburg and Beyond: An Agenda ———. 2001b. Engendering Development through Gender for Action.� Paper to be presented by Environmentally and Equality in Rights, Resources, and Voice. New York: Oxford Socially Sustainable Development in run-up to World Sum- University Press. mit on Sustainable Development, August 2002. ———2001c. “Environment Strategy Paper: Making Sustain- World Bank and International Finance Corporation. 2002. able Commitments (Revised Draft).� Washington, D.C. “Treasure or Trouble? Mining in Developing Countries.� Processed. Washington, D.C. (c) The International Bank for Reconstruction and Development / The World Bank ����������  WHO (World Health Organization), Department of Commu- Campbell, Tim. “The Evolution of Governance in Metropoli- nicable Disease Surveillance and Response. 2001. “WHO tan Areas.� Global Strategy for Containment of Antimicrobial Resis- Chavez, Roberto. “Supported Peru Sites and Services Develop- tance.� Washington, D.C. ment Projects.� World Values Surveys and European Values 1981–4, 1990–3, Das Gupta, Monica. “Population and Sustainable Develop- 1995–7. ICPSR 2790. Available from Inter-university Con- ment.� sortium for Political and Social Research at http://www. Gates, Scott, Nils Petter Gleditsch, and Eric Neumayer. “Envi- icpsr.umich.edu/. ronmental Commitment, Democracy and Inequality.� World Water Council. 2000. “A Water Secure World: Vision Hannesson, Rognvaldur. “The Development of Economic In- for Water, Life, and the Environment.� Commission Report. stitutions in World Fisheries.� Marseille. ———. “Trends in World Fish Catches. Do We Face a Crisis?� Worster, Donald. 1979. Dust Bowl: The Southern Plains in the Hoff, Karla. “Paths of Development and Institutional Barriers 1930s. New York: Oxford University Press. to Economic Opportunities.� WRI (World Resources Institute). 2000. World Resources Holtedahl, Pernille, and Haakon Vennemo. “Environmental 2000–01 People and Ecosystems: The Fraying Web of Life. Challenges in China: Success, and Failures, Determinants, Washington, D.C. Driving Forces.� Wunder, Sven. 2000. The Economics of Deforestation: The Ex- Janeba, Eckard, and Guttorm Schjeldrup. “The Future of Glob- ample of Ecuador. New York: St. Martin’s Press. alization: Tax Competition and Trade Liberalization.� Yi, Zeng. 2002. “Old-Age Insurance and Sustainable Develop- Jayasuriya, Ruwan, and Quentin Wodon. “Measuring and Ex- ment in Rural China.� In Kochendörfer-Lucius and Ples- plaining Country Efficiency in Improving Health and Edu- kovic, eds., Villa Borsig Workshop Series 2001. Berlin. cation Indicators: The Role of Urbanization.� Yildizcan, Guzin. 2002. “Drive to Resuscitate the Golden Kuhnle, Stein, Sanjeev Prakash, Huck-ju Kwon, and Per Selle. Horn.� World Bank, Operations and Evaluation Depart- “Political Institutions, Democracy and Welfare: A Compar- ment Précis 34. Washington, D.C. ative Study of Norway and Korea.� Yli-Renko, H. 1999. “Dependence, Social Capital, and Learn- Murshed, Mansoob S. “On Natural Resource Abundance and ing in Key Customer Relationships: Effects of the Perfor- Underdevelopment.� mance of Technology-Based New Firms.� Master’s thesis. Pratt, Jane, and John D. Shilling. “High Time for Mountains: Helsinki University of Technology. A Program for Sustaining Mountain Resources and Liveli- Yusuf, Shahid. 2001. “East Asia’s Urban Regions: A Strategy for hoods.� the Coming Decade.� World Bank, Washington, D.C. Sambanis, Nicolas. “Preventing Violent Civil Conflict: The Processed. Scope and Limits of Government Action.� Zaheer, A., B. McEvily, and V. Perrone. 1998. “Does Trust Steinberg, Paul. “Civic Environmentalism in Developing Coun- Matter? Exploring the Effects of Interorganizational and In- tries: Opportunities for Innovation in State-Society Rela- terpersonal Trust on Performance.� Organisation Science tions.� 9(2):141–159. Tesli, Arne. “The Use of EIA and SEA Relative to the Objective Zheng, X., and E. A. B. Eltahir. 1997. “The Response to De- of Sustainable Development.� forestation and Desertification in a Model of West African Zainabi, Ahmed. “Pérennité des actions d’autopromotion com- Monsoons.� Geophysical Research Letters 24(2): 155–158. munautaire: Cas de la Vallée du Dra Moyen (Province de Zagora).� Background papers for WDR 2003 Acharya, G., and John Dixon. “No one Said it was Going to be Background notes for WDR 2003 Easy: An Analysis of the Recommendations made by the Chavez, Roberto. “Revisiting the Peru Sites and Services Devel- 1992 World Development Report and the Experience in the opment Projects.� Last Decade.� Cira, Dean. “Regularizing Favelas in Brazil.� Bertaud, Alain. “The Spatial Organization of Cities: Deliberate Ortiz Malavasi, Edgar, and John Kellenberg. “Program of Pay- Outcome or Unforeseen Consequence?� ments for Environmental Services in Costa Rica.� ———. Metropolitan Structure, Densities and Livability.� Pantelic, Jelena. “Hazard Mitigation Through Collective Ac- Brekke, Kjell Arne, and Desmond McNeill. “Identity Signaling tion.� in Consumption: A Case for Provision of More Public Viloria-Williams, J. “Notes on Urban Upgrading (UU) Experi- Goods.� ences in Two Countries.� (c) The International Bank for Reconstruction and Development / The World Bank Introduction to Selected World Development Indicators I n this year’s edition, development data are pre- collected by the World Bank, member country sta- sented in four tables presenting comparative so- tistical publications, research institutes, and interna- cioeconomic data for more than 130 economies tional organizations such as the United Nations and for the most recent year for which data are available its specialized agencies, the International Monetary and, for some indicators, for an earlier year. An ad- Fund (IMF), and the OECD (see the Data Sources ditional table presents basic indicators for 75 econ- following the Technical notes for a complete listing). omies with sparse data or with populations of less Although international standards of coverage, defi- than 1.5 million. nition, and classification apply to most statistics re- The indicators presented here are a selection from ported by countries and international agencies, there more than 800 included in World Development In- are inevitably differences in timeliness and reliability dicators 2002. Published annually, World Develop- arising from differences in the capabilities and re- ment Indicators reflects a comprehensive view of the sources devoted to basic data collection and compi- development process. Its opening chapter reports on lation. For some topics, competing sources of data the Millennium Development Goals which grew out require review by World Bank staff to ensure that the of agreements and resolutions of world conferences most reliable data available are presented. In some organized by the United Nations (UN) in the past instances, where available data are deemed too weak decade, and reaffirmed at the Millennium Summit to provide reliable measures of levels and trends or in September 2000 by member countries of the UN. do not adequately adhere to international standards, The other five main sections recognize the contribu- the data are not shown. tion of a wide range of factors: human capital devel- The data presented are generally consistent with opment, environmental sustainability, macroeco- those in World Development Indicators 2002. How- nomic performance, private sector development, and ever, data have been revised and updated wherever the global links that influence the external environ- new information has become available. Differences ment for development. World Development Indica- may also reflect revisions to historical series and tors is complemented by a separately published data- changes in methodology. Thus data of different vin- base that gives access to over 1,000 data tables and tages may be published in different editions of World 500 time-series indicators for 225 economies and Bank publications. Readers are advised not to com- regions. This database is available through an elec- pile data series from different publications or differ- tronic subscription (WDI Online) or as a CD-ROM. ent editions of the same publication. Consistent time-series data are available on World Development Data sources and methodology Indicators 2002 CD-ROM and through WDI Online. Socioeconomic and environmental data presented All dollar figures are in current U.S. dollars un- here are drawn from several sources: primary data less otherwise stated. The various methods used to  (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  convert from national currency figures are described mies. Data for the countries excluded from the main in the Technical notes. tables (those presented in Table 1a) have been in- Because the World Bank’s primary business is cluded in the summary measures, where data are providing lending and policy advice to its low- and available, or by assuming that they follow the trend middle-income members, the issues covered in these of reporting countries. This gives a more consistent tables focus mainly on these economies. Where aggregated measure by standardizing country cover- available, information on the high-income econo- age for each period shown. Where missing informa- mies is also provided for comparison. Readers may tion accounts for a third or more of the overall esti- wish to refer to national statistical publications and mate, however, the group measure is reported as not publications of the Organisation for Economic Co- available. The section on Statistical methods in the operation and Development (OECD) and the Eu- Technical notes provides further information on ropean Union for more information on the high- aggregation methods. Weights used to construct the income economies. aggregates are listed in the technical notes for each table. Changes in the System of National Accounts From time to time an economy’s classification is This edition of the Selected World Development In- revised because of changes in the above cutoff values dicators, as in last year’s edition, uses terminology in or in the economy’s measured level of GNI per capita. line with the 1993 System of National Accounts When such changes occur, aggregates based on those (SNA). For example, in the 1993 SNA gross national classifications are recalculated for the past period so income replaces gross national product. See the tech- that a consistent time series is maintained. nical notes for tables 1 and 3. Terminology and country coverage Most countries continue to compile their national accounts according to the 1968 SNA, but more and The term country does not imply political indepen- more are adopting the 1993 SNA. A few low-income dence but may refer to any territory for which au- countries still use concepts from older SNA guide- thorities report separate social or economic statistics. lines, including valuations such as factor cost, in de- Data are shown for economies as they were consti- scribing major economic aggregates. tuted in 2000, and historical data are revised to re- flect current political arrangements. Throughout the Classification of economies and summary tables, exceptions are noted. measures Technical notes The summary measures at the bottom of each table include economies classified by income per capita Because data quality and intercountry comparisons and by region. GNI per capita is used to determine are often problematic, readers are encouraged to the following income classifications: low-income, consult the Technical notes, the table on Classifica- $745 or less in 2001; middle-income, $746 to tion of Economies by Income and Region, and the $9,205; and high-income, $9,206 and above. A footnotes to the tables. For more extensive docu- further division at GNI per capita $2,975 is made mentation see World Development Indicators 2002. between lower-middle-income and upper-middle- Readers may find more information on the WDI income economies. See the table on classification of 2002, and orders can be made online, by phone, or economies at the end of this volume for a list of fax as follows: economies in each group (including those with pop- ulations of less than 1.5 million). For more information and to order online: Summary measures are either totals (indicated by http://www.worldbank.org/data/wdi2002/index.htm. t if the aggregates include estimates for missing data To order by phone or fax: 1-800-645-7247 or and nonreporting countries, or by an s for simple 703-661-1580; Fax 703-661-1501 sums of the data available), weighted averages (w), To order by mail: The World Bank, P.O. Box 960, or median values (m) calculated for groups of econo- Herndon, VA 20172-0960, U.S.A. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 1. Key indicators of development PPP Population Gross national gross national Gross Under-5 Adult Carbon income (GNI)a income (GNI)b domestic Life mortality illiteracy dioxide Avg. density product expectancy rate rate emissions annual people per Billions per capita Billions per capita per capita at birth Per % of people Millions Millions % growth sq. km of dollars dollars of dollars dollars % growth Years 1,000 15 and above of tons 2001 1990–2001 2001 2001 2001 2001 2001 2000–2001 2000 2000 2000 1998 Albania 3.4 0.4 126 4.2 1,230 13 3,880 5.6 74 .. 15 1.6 Algeria 30.9 1.9 13 50.4 1,630 159c 5,150c 0.3 71 39 33 106.6 Angola 13.5 3.1 11 6.7 500 21 c 1,550 c 0.3 47 208 .. 5.9 Argentina 37.5 1.3 14 261.0 6,960 438 11,690 –4.8 74 22 3 136.9 Armenia 3.8 0.7 135 2.1 560 11 2,880 9.4 74 17 2 3.4 Australia 19.4 1.2 3 383.3 19,770 500 25,780 1.3 79 7 .. 331.5 Austria 8.1 0.5 98 194.5 23,940 220 27,080 0.9 78 6 .. 63.9 Azerbaijan 8.1 1.1 94 5.3 650 25 3,020 8.2 72 21 .. 38.8 Bangladesh 133.4 1.8 1,025 49.9 370 224 1,680 3.3 61 83 59 23.4 Belarus 10.0 –0.2 48 11.9 1,190 80 8,030 4.4 68 14 0d 60.5 Belgium 10.3 0.3 313 239.8 23,340 290 28,210 0.8 78 7 .. 101.3 Benin 6.4 2.8 58 2.3 360 7 1,030 3.1 53 143 63 0.7 Bolivia 8.5 2.4 8 8.0 940 20 2,380 –1.2 63 79 14 12.1 Botswana 1.6 2.1 3 5.9 3,630 14 8,810 4.8 39 99 23 3.8 Brazil 172.6 1.4 20 528.5 3,060 1,286 7,450 0.2 68 39 15 299.6 Bulgaria 8.1 –0.6 73 12.6 1,560 48 5,950 5.1 72 16 2 47.4 Burkina Faso 11.6 2.4 42 2.4 210 12c 1,020c 3.2 44 206 76 1.0 Burundi 6.9 2.2 270 0.7 100 4 c 590 c 1.3 42 176 52 0.2 Cambodia 12.3 2.7 69 3.3 270 19 1,520 3.2 54 120 32 0.7 Cameroon 15.2 2.4 33 8.7 570 25 1,670 3.1 50 155 24 1.8 Canada 31.0 1.0 3 661.9 21,340 864 c 27,870 c 0.6 79 7 .. 467.2 Central African Republic 3.8 2.2 6 1.0 270 4c 1,180c 0.0 43 152 53 0.2 Chad 7.9 2.9 6 1.6 200 7 930 5.8 48 188 57 0.1 Chile 15.4 1.5 21 66.9 4,350 145 9,420 1.7 76 12 4 60.2 China 1,271.9 1.0 136 1,131.0 890 5,415 4,260 6.5 70 39 16 3,108.0 Hong Kong, China 6.9 1.7 .. 176.2 25,920 179 26,050 –0.1 80 .. 6 35.8 Colombia 43.0 1.9 41 82.0 1,910 258 5,980 –0.2 72 23 8 67.8 Congo, Dem. Rep. 52.4 3.2 23 .. . .e .. .. .. 46 163 39 2.4 Congo, Rep. 3.1 3.0 9 2.2 700 2 580 0.1 51 106 19 1.8 Costa Rica 3.9 2.2 76 15.3 3,950 31 8,080 –1.0 77 13 4 5.1 Côte d’Ivoire 16.4 3.0 52 10.3 630 24 1,470 –3.3 46 180 53 13.2 Croatia 4.4 –0.8 78 19.9 4,550 37 8,440 4.1 73 9 2 19.8 Czech Republic 10.3 –0.1 133 54.1 5,270 149 14,550 3.6 75 7 .. 118.3 Denmark 5.4 0.4 126 166.3 31,090 150 27,950 0.7 76 6 .. 53.4 Dominican Republic 8.5 1.7 176 19.0 2,230 50 5,870 1.1 67 47 16 20.3 Ecuador 12.9 2.1 47 16.0 1,240 40 3,070 3.3 70 34 8 26.3 Egypt, Arab Rep. 65.2 2.0 65 99.4 1,530 247 3,790 1.4 67 52 45 105.8 El Salvador 6.4 2.0 309 13.1 2,050 29 4,500 0.0 70 35 21 6.1 Eritrea 4.2 2.7 42 0.8 190 4 970 2.5 52 103 44 .. Estonia 1.4 –1.3 32 5.2 3,810 14 10,020 5.3 71 11 0 17.0 Ethiopia 65.8 2.3 66 6.8 100 47 710 5.4 42 179 61 2.0 Finland 5.2 0.4 17 124.2 23,940 131 25,180 0.5 77 5 .. 53.3 France 59.2 0.4 108 1,377.4f 22,690f 1,495 25,280 1.6 79 6 .. 369.9 Georgia 5.0 –0.8 72 3.1 620 14 2,860 4.6 73 21 .. 5.2 Germany 82.2 0.3 230 1,948.0 23,700 2,098 25,530 0.5 77 6 .. 825.2 Ghana 19.7 2.4 87 5.7 290 39c 1,980c 1.9 57 112 28 4.4 Greece 10.6 0.4 82 124.6 11,780 189 17,860 3.9 78 8 3 85.2 Guatemala 11.7 2.6 108 19.6 1,670 45 3,850 –0.6 65 49 31 9.7 Guinea 7.6 2.5 31 3.0 400 15 1,980 0.7 46 161 .. 1.2 Haiti 8.1 2.1 294 3.9 480 12c 1,450c –3.5 53 111 50 1.3 Honduras 6.6 2.7 59 5.9 900 16 2,450 0.1 66 44 25 5.1 Hungary 10.2 –0.2 110 48.9 4,800 128 12,570 4.0 71 11 1 58.7 India 1,033.4 1.8 348 474.3 460 2,530 2,450 2.7 63 88 43 1,061.0 Indonesia 213.6 1.6 118 144.7 680 628 2,940 1.8 66 51 13 233.6 Iran, Islamic Rep. 64.7 1.6 40 112.9 1,750 403 6,230 3.0 69 41 24 289.9 Ireland 3.8 0.8 56 88.4 23,060 105 27,460 5.6 76 7 .. 38.3 Israel 6.4 2.8 309 104.1 16,710 121 19,330 .. 78 7 5 60.3 Italy 57.7 0.2 196 1,123.5 19,470 1,404 24,340 1.8 79 7 2 414.9 Jamaica 2.7 0.9 246 7.3 2,720 10 3,650 0.4 75 24 13 11.0 Japan 127.1 0.3 349 4,574.2 35,990 3,487 27,430 –0.6 81 5 .. 1,133.5 Jordan 5.0 4.2 57 8.8 1,750 21 4,080 1.2 72 30 10 13.9 Kazakhstan 14.8 –0.8 5 20.1 1,360 94 6,370 13.5 65 28 .. 122.9 Kenya 30.7 2.5 54 10.3 340 31 1,020 –1.0 47 120 18 9.1 Korea, Rep. 47.6 1.0 483 447.7 9,400 863 18,110 2.3 73 10 2 363.7 Kuwait 2.0 –0.4 115 35.8 18,030 37 18,690 .. 77 13 18 49.1 Kyrgyz Republic 5.0 1.1 26 1.4 280 13 2,710 4.2 67 35 .. 6.4 Lao PDR 5.4 2.4 23 1.6 310 9 c 1,610 c 2.9 54 .. 51 0.4 Latvia 2.3 –1.2 38 7.6 3,260 18 7,870 9.0 70 17 0d 7.9 Lebanon 4.4 1.7 429 17.6 4,010 20 4,640 0.0 70 30 14 16.3 Lesotho 2.1 1.9 68 1.1 550 6c 2,670c 1.7 44 143 17 .. Lithuania 3.5 –0.5 54 11.4 3,270 27 7,610 4.3 73 11 0d 15.6 Macedonia, FYR 2.0 0.6 80 3.4 1,690 10 4,860 –4.7 73 17 .. 12.4 Madagascar 16.0 2.9 27 4.2 260 14 870 3.7 55 144 34 1.3 Malawi 10.5 1.9 112 1.8 170 7 620 0.7 39 193 40 0.7 Malaysia 23.8 2.4 72 86.5 3,640 198 8,340 –1.8 73 11 13 120.5 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  PPP Population Gross national gross national Gross Under-5 Adult Carbon income (GNI)a income (GNI)b domestic Life mortality illiteracy dioxide Avg. density product expectancy rate rate emissions annual people per Billions per capita Billions per capita per capita at birth Per % of people Millions Millions % growth sq. km of dollars dollars of dollars dollars % growth Years 1,000 15 and above of tons 2001 1990–2001 2001 2001 2001 2001 2001 2000–2001 2000 2000 2000 1998 Mali 11.1 2.5 9 2.3 210 9 810 –0.9 42 218 59 0.5 Mauritania 2.8 2.9 3 1.0 350 5 1,680 1.4 52 164 60 2.9 Mexico 99.4 1.6 52 550.5 5,540 872 8,770 –1.8 73 36 9 374.0 Moldova 4.3 –0.2 130 1.4 380 10 2,420 6.3 68 22 1 9.7 Mongolia 2.4 1.3 2 1.0 400 4 1,800 0.4 67 71 1 7.7 Morocco 29.2 1.8 65 34.6 1,180 108 3,690 4.8 67 60 51 32.0 Mozambique 18.1 2.2 23 3.7 210 18c 1,000c 6.7 42 200 56 1.3 Myanmar 48.3 1.6 73 .. . .e .. .. .. 56 126 15 8.2 Namibia 1.8 2.4 2 3.5 1,960 12 c 6,700 c 2.6 47 112 18 0.0 Nepal 23.6 2.4 165 5.9 250 34 1,450 3.4 59 105 58 3.0 Netherlands 16.0 0.6 473 385.4 24,040 424 26,440 0.4 78 7 .. 163.8 New Zealand 3.8 1.0 14 47.6 12,380 74 19,130 1.3 78 7 .. 30.0 Nicaragua 5.2 2.8 43 .. . .e .. .. .. 69 41 33 3.4 Niger 11.2 3.4 9 2.0 170 9c 770c 1.7 46 248 84 1.1 Nigeria 129.9 2.7 143 37.1 290 108 830 1.6 47 153 36 78.5 Norway 4.5 0.6 15 160.6 35,530 138 30,440 0.8 79 5 .. 33.6 Pakistan 141.5 2.5 183 59.6 420 271 1,920 0.9 63 110 57 97.1 Panama 2.9 1.7 39 9.5 3,290 17 c 5,720 c –1.3 75 24 8 5.8 Papua New Guinea 5.3 2.5 12 3.0 580 11c 2,150c –5.8 59 75 36 2.3 Paraguay 5.6 2.6 14 7.3 1,300 25 c 4,400 c –3.0 70 28 7 4.6 Peru 26.1 1.7 20 52.1 2,000 122 4,680 –1.4 69 41 10 27.9 Philippines 77.0 2.1 258 80.8 1,050 336 4,360 1.5 69 39 5 76.0 Poland 38.7 0.1 127 163.9 4,240 359 9,280 1.2 73 11 0d 321.7 Portugal 10.2 0.3 112 109.2 10,670 177 17,270 –0.3 76 8 8 54.6 Romania 22.4 –0.3 97 38.4 1,710 156 6,980 5.5 70 23 2 92.4 Russian Federation 144.8 –0.2 9 253.4 1,750 1,255 8,660 5.5 65 19 0d 1,434.6 Rwanda 8.7 2.0 353 1.9 220 9 1,000 4.3 40 203 33 0.5 Saudi Arabia 21.4 2.8 10 149.9 7,230 236 11,390 .. 73 23 24 283.0 Senegal 9.8 2.6 51 4.7 480 15 1,560 3.2 52 129 63 3.3 Sierra Leone 5.1 2.3 72 0.7 140 2 480 3.1 39 267 .. 0.5 Singapore 4.1 2.7 6,726 99.4 24,740 100 24,910 .. 78 6 8 82.3 Slovak Republic 5.4 0.2 112 20.0 3,700 63 11,610 3.2 73 10 .. 38.1 Slovenia 2.0 0.0 99 19.4 9,780 36 18,160 2.9 75 7 0d 14.6 South Africa 43.2 1.9 35 125.5 2,900 411c 9,510c 1.2 48 79 15 343.7 Spain 39.5 0.2 79 586.9 14,860 796 20,150 2.7 78 6 2 247.2 Sri Lanka 19.6 1.3 304 16.3 830 70 3,560 1.0 73 18 8 8.1 Sweden 8.9 0.3 22 225.9 25,400 219 24,670 1.0 80 4 .. 48.6 Switzerland 7.2 0.6 182 266.5 36,970 226 31,320 0.9 80 6 .. 41.8 Syrian Arab Republic 16.6 2.9 90 16.6 1,000 57 3,440 1.0 70 29 26 50.6 Tajikistan 6.2 1.5 44 1.1 170 7 1,150 4.1 69 30 1 5.1 Tanzania 34.5 2.7 39 9.2 g 270 g 19 g 540 g 2.3 44 149 25 2.2 Thailand 61.2 0.9 120 120.9 1,970 401 6,550 0.9 69 33 5 192.4 Togo 4.7 2.7 86 1.3 270 7 1,420 –0.1 49 142 43 0.9 Tunisia 9.7 1.6 62 20.1 2,070 62 6,450 4.2 72 30 29 22.4 Turkey 66.2 1.5 86 168.3 2,540 440 6,640 –7.8 70 43 15 202.0 Turkmenistan 5.3 3.3 11 5.0 950 24 4,580 18.4 66 43 .. 27.9 Uganda 22.8 3.0 116 6.3 280 29 c 1,250 c 2.0 42 161 33 1.3 Ukraine 49.1 –0.5 85 35.2 720 204 4,150 10.0 68 16 0 d 353.6 United Kingdom 59.9 0.4 249 1,451.4 24,230 1,466 24,460 1.9 77 7 .. 542.3 United States 284.0 1.2 31 9,900.7 34,870 9,902 34,870 0.3 77 9 .. 5,447.6 Uruguay 3.4 0.7 19 19.0 5,670 29 8,710 –3.7 74 17 2 5.8 Uzbekistan 25.1 1.8 61 13.8 550 62 2,470 2.6 70 27 1 109.2 Venezuela, RB 24.6 2.1 28 117.2 4,760 145 5,890 0.7 73 24 7 155.4 Vietnam 79.5 1.7 244 32.6 410 169 2,130 4.7 69 34 7 43.9 Yemen, Rep. 18.0 3.8 34 8.3 460 14 770 –1.0 56 95 54 14.2 Yugoslavia, Fed. Rep. 10.6 0.1 108 .. . .h .. .. 4.9 72 15 .. .. Zambia 10.3 2.5 14 3.3 320 8 790 3.2 38 186 22 1.6 Zimbabwe 12.8 2.0 33 6.2 480 30 2,340 –9.8 40 116 11 14.1 World 6,132.8 s 1.4 w 47 w 31,500.0 t 5,140 w 46,403 t 7,570 w 0.2 w 66 w 78 w . . w 22,825.0 s Low income 2,510.6 2.0 76 1,069.1 430 5,134 2,040 2.4 59 115 37 2,418.7 Middle income 2,667.2 1.2 40 4,922.0 1,850 15,235 5,710 1.7 69 39 14 8,830.1 Lower middle income 2,163.5 1.1 48 2,676.5 1,240 10,867 5,020 3.3 69 42 15 6,660.4 Upper middle income 503.7 1.3 24 2,247.7 4,460 4,397 8,730 –0.5 71 30 10 2,169.6 Low & middle income 5,177.8 1.5 52 5,990.3 1,160 20,338 3,930 1.5 64 85 25 11,248.8 East Asia & Pacific 1,825.2 1.2 115 1,649.4 900 7,383 4,040 4.5 69 45 15 4,021.6 Europe & Central Asia 474.6 0.2 20 930.5 1,960 3,319 6,990 2.4 69 25 3 3,134.8 Latin America & Carib. 523.7 1.6 26 1,861.8 3,560 3,704 7,070 –1.1 70 37 12 1,309.8 Middle East & N. Africa 300.7 2.1 27 601.3 2,000 1,544 5,230 .. 68 54 35 1,076.0 South Asia 1,379.8 1.9 289 615.6 450 3,176 2,300 2.5 62 96 45 1,194.4 Sub-Saharan Africa 673.9 2.6 29 317.0 470 1,094 1,620 0.7 47 162 39 512.2 High income 955.0 0.7 31 25,506.4 26,710 26,431 27,680 0.6 78 7 .. 11,576.2 a. Preliminary World Bank estimates calculated using the World Bank Atlas method. b. Purchasing power parity; see the Technical Notes. c. The estimate is based on regression; others are extrapolated from the latest International Comparison Programme benchmark estimates. d. Less than 0.5. e. Estimated to be low income ($745 or less). f. GNI and GNI per capita estimates include the French Overseas departments of French Guiana, Guadeloupe, Martinique, and Réunion. g. Data refer to mainland Tanzania only. h. Estimated to be lower middle income ($745 to $2,975). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 2. Poverty and income distribution Percentage National poverty lines International poverty line share of income or Population below the Population Poverty Population Poverty consumption poverty line (%) below gap at below gap at Survey Survey $1 a day $1 a day $2 a day $2 a day Gini Lowest Highest Economy year Rural Urban National year % % % % Survey year index 10% 10% Albania 1996 .. 15 .. .. .. .. .. .. .. .. Algeria 1995 30.3 14.7 22.6 1995 <2 <0.5 15.1 3.6 1995 a,b 35.3 2.8 26.8 Angola .. .. .. .. .. .. .. .. .. .. Argentina 1993 .. .. 17.6 .. .. .. .. .. .. .. Armenia .. .. .. 1996 7.8 1.7 34.0 11.3 1996 a,b 44.4 2.3 35.2 Australia .. .. .. .. .. .. .. 1994 c,d 35.2 2.0 25.4 Austria .. .. .. .. .. .. .. 1995 c,d 31.0 2.5 22.5 Azerbaijan 1995 .. .. 68.1 1995 <2 <0.5 9.6 2.3 1995 c,d 36.0 2.8 27.8 Bangladesh 1995–96 39.8 14.3 35.6 1996 29.1 5.9 77.8 31.8 1995–96 a,b 33.6 3.9 28.6 Belarus 2000 .. .. 41.9 1998 <2 <0.5 <2 <0.5 1998 a,b 21.7 5.1 20.0 Belgium .. .. .. .. .. .. .. 1996 c,d 28.7 3.2 23.0 Benin 1995 .. .. 33.0 .. .. .. .. .. .. .. Bolivia 1995 79.1 .. .. 1999 14.4 5.4 34.3 14.9 1999 a,b 44.7 1.3 32.0 Botswana .. .. . . 1985–86 33.3 12.5 61.4 30.7 .. .. .. Brazil 1990 32.6 13.1 17.4 1998 11.6 3.9 26.5 11.6 1998 c,d 60.7 0.7 48.0 Bulgaria .. .. .. 1997 <2 <0.5 21.9 4.2 1997 c,d 26.4 4.5 22.8 Burkina Faso .. .. .. 1994 61.2 25.5 85.8 50.9 1998 a,b 55.1 2.0 46.8 Burundi 1990 .. .. 36.2 .. .. .. .. 1998 a,b 42.5 1.8 32.9 Cambodia 1997 40.1 21.1 36.1 .. .. .. .. 1997 a,b 40.4 2.9 33.8 Cameroon 1984 32.4 44.4 40.0 1996 33.4 11.8 64.4 31.2 1996 a,b 47.7 1.9 36.6 Canada .. .. .. .. .. .. .. 1994 c,d 31.5 2.8 23.8 Central African Republic .. .. .. 1993 66.6 38.1 84.0 58.4 1993 a,b 61.3 0.7 47.7 Chad 1995–96 67.0 63.0 64.0 .. .. .. .. .. .. .. Chile 1998 .. .. 21.2 1998 <2 <0.5 8.7 2.3 1998 c,d 56.7 1.3 45.6 China 1998 4.6 <2 4.6 1999 18.8 4.4 52.6 20.9 1998 c,d 40.3 2.4 30.4 Hong Kong, China .. .. .. .. .. .. .. 1996 c,d 52.2 1.8 43.5 Colombia 1992 31.2 8.0 17.7 1998 19.7 10.8 36.0 19.4 1996 c,d 57.1 1.1 46.1 Congo, Dem. Rep. .. .. .. .. .. .. .. .. .. .. Congo, Rep. .. .. .. .. .. .. .. .. .. .. Costa Rica 1992 25.5 19.2 22.0 1998 12.6 6.2 26.0 12.8 1997 c,d 45.9 1.7 34.6 Côte d’Ivoire 1995 .. .. 36.8 1995 12.3 2.4 49.4 16.8 1995 a,b 36.7 3.1 28.8 Croatia .. .. .. 1998 <2 <0.5 <2 <0.5 1998 c,d 29.0 3.7 23.3 Czech Republic .. .. .. 1996 <2 <0.5 <2 <0.5 1996 c,d 25.4 4.3 22.4 Denmark .. .. .. .. .. .. .. 1992 c,d 24.7 3.6 20.5 Dominican Republic 1992 29.8 10.9 20.6 1996 3.2 0.7 16.0 5.0 1998 c,d 47.4 2.1 37.9 Ecuador 1994 47.0 25.0 35.0 1995 20.2 5.8 52.3 21.2 1995 a,b 43.7 2.2 33.8 Egypt, Arab Rep. 1995–96 23.3 22.5 22.9 1995 3.1 <0.5 52.7 13.9 1995 a,b 28.9 4.4 25.0 El Salvador 1992 55.7 43.1 48.3 1998 21.0 7.8 44.5 20.6 1998 c,d 52.2 1.2 39.5 Eritrea 1993–94 .. .. 53.0 .. .. .. .. .. .. .. Estonia 1995 14.7 6.8 8.9 1998 <2 <0.5 5.2 0.8 1998 c,d 37.6 3.0 29.8 Ethiopia .. .. .. 1995 31.3 8.0 76.4 32.9 1995 a,b 40.0 3.0 33.7 Finland .. .. .. .. .. .. .. 1991 c,d 25.6 4.2 21.6 France .. .. .. .. .. .. .. 1995 c,d 32.7 2.8 25.1 Georgia 1997 9.9 12.1 11.1 1996 <2 <0.5 <2 <0.5 1996 c,d 37.1 2.3 27.9 Germany .. .. .. .. .. .. .. 1994 c,d 30.0 3.3 23.7 Ghana 1992 34.3 26.7 31.4 1999 44.8 17.3 78.5 40.8 1999 a,b 40.7 2.2 30.1 Greece .. .. .. .. .. .. .. 1993 c,d 32.7 3.0 25.3 Guatemala 1989 71.9 33.7 57.9 1998 10.0 2.2 33.8 11.8 1998 c,d 55.8 1.6 46.0 Guinea 1994 .. .. 40.0 .. .. .. .. 1994 a,b 40.3 2.6 32.0 Haiti 1995 66.0 .. .. .. .. .. .. .. .. .. Honduras 1993 51.0 57.0 53.0 1998 24.3 11.9 45.1 23.5 1998 c,d 56.3 0.6 42.7 Hungary 1993 .. .. 8.6 1998 <2 <0.5 7.3 1.7 1998 a,b 24.4 4.1 20.5 India 1994 36.7 30.5 35.0 1997 44.2 12.0 86.2 41.4 1997 a,b 37.8 3.5 33.5 Indonesia 1999 .. .. 27.1 1999 12.9 1.9 65.5 21.5 1999 a,b 31.7 4.0 26.7 Iran, Islamic Rep. .. .. .. .. .. .. .. .. .. .. Ireland .. .. .. .. .. .. .. 1987 c,d 35.9 2.5 27.4 Israel .. .. .. .. .. .. .. 1997 c,d 38.1 2.4 28.3 Italy .. .. .. .. .. .. .. 1995 c,d 27.3 3.5 21.8 Jamaica 2000 .. .. 18.7 1996 3.2 0.7 25.2 6.9 2000 a,b 37.9 2.7 30.3 Japan .. .. .. .. .. .. .. 1993 c,d 24.9 4.8 21.7 Jordan 1997 .. .. 11.7 1997 <2 <0.5 7.4 1.4 1997 a,b 36.4 3.3 29.8 Kazakhstan 1996 39.0 30.0 34.6 1996 <2 <0.5 15.3 3.9 1996 a,b 35.4 2.7 26.3 Kenya 1992 46.4 29.3 42.0 1994 26.5 9.0 62.3 27.5 1997 a,b 44.9 2.4 36.1 Korea, Rep. .. .. .. 1993 <2 <0.5 <2 <0.5 1993 a,b 31.6 2.9 24.3 Kuwait .. .. .. .. .. .. .. .. .. .. Kyrgyz Republic 1997 64.5 28.5 51.0 .. .. .. .. 1999 a,b 34.6 3.2 27.2 Lao PDR 1993 53.0 24.0 46.1 1997 26.3 6.3 73.2 29.6 1997 a,b 37.0 3.2 30.6 Latvia .. .. .. 1998 <2 <0.5 8.3 2.0 1998 c,d 32.4 2.9 25.9 Lebanon .. .. .. .. .. .. .. .. .. .. Lesotho 1993 53.9 27.8 49.2 1993 43.1 20.3 65.7 38.1 1986–87 a,b 56.0 0.9 43.4 Lithuania .. .. .. 1996 <2 <0.5 7.8 2.0 1996 a,b 32.4 3.1 25.6 Macedonia, FYR .. .. .. .. .. .. .. .. .. .. Madagascar 1993–94 77.0 47.0 70.0 1999 49.1 18.3 83.3 44.0 1999 a,b 38.1 2.6 28.6 Malawi 1990–91 .. .. 54.0 .. .. .. .. .. .. .. Malaysia 1989 .. .. 15.5 .. .. .. .. 1997 c,d 49.2 1.7 38.4 Note: For data comparability and coverage, see the technical notes. (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  Percentage National poverty lines International poverty line share of income or Population below the Population Poverty Population Poverty consumption poverty line (%) below gap at below gap at Survey Survey $1 a day $1 a day $2 a day $2 a day Gini Lowest Highest Economy year Rural Urban National year % % % % Survey year index 10% 10% Mali .. .. .. 1994 72.8 37.4 90.6 60.5 1994 a,b 50.5 1.8 40.4 Mauritania 1989–90 .. .. 57.0 1995 28.6 9.1 68.7 29.6 1995 a,b 37.3 2.5 28.4 Mexico 1988 .. .. 10.1 1998 15.9 5.2 37.7 16.0 1998 c,d 53.1 1.3 41.7 Moldova 1997 26.7 .. 23.3 1997 11.3 3.0 38.4 14.0 1997 c,d 40.6 2.2 30.7 Mongolia 1995 33.1 38.5 36.3 1995 13.9 3.1 50.0 17.5 1995 a,b 33.2 2.9 24.5 Morocco 1998–99 27.2 12.0 19.0 1990–91 <2 <0.5 7.5 1.3 1998–99 a,b 39.5 2.6 30.9 Mozambique .. .. .. 1996 37.9 12.0 78.4 36.8 1996–97 a,b 39.6 2.5 31.7 Myanmar .. .. .. .. .. .. .. .. .. .. Namibia .. .. .. 1993 34.9 14.0 55.8 30.4 .. .. .. Nepal 1995–96 44.0 23.0 42.0 1995 37.7 9.7 82.5 37.5 1995–96 a,b 36.7 3.2 29.8 Netherlands .. .. .. .. .. .. .. 1994 c,d 32.6 2.8 25.1 New Zealand .. .. .. .. .. .. .. .. .. .. Nicaragua 1993 76.1 31.9 50.3 .. .. .. .. 1998 a,b 60.3 0.7 48.8 Niger 1989–93 66.0 52.0 63.0 1995 61.4 33.9 85.3 54.8 1995 a,b 50.5 0.8 35.4 Nigeria 1992–93 36.4 30.4 34.1 1997 70.2 34.9 90.8 59.0 1996–97 a,b 50.6 1.6 40.8 Norway .. .. .. .. .. .. .. 1995 c,d 25.8 4.1 21.8 Pakistan 1991 36.9 28.0 34.0 1996 31.0 6.2 84.7 35.0 1996–97 a,b 31.2 4.1 27.6 Panama 1997 64.9 15.3 37.3 1998 14.0 5.9 29.0 13.8 1997 a,b 48.5 1.2 35.7 Papua New Guinea .. .. .. .. .. .. .. 1996 a,b 50.9 1.7 40.5 Paraguay 1991 28.5 19.7 21.8 1998 19.5 9.8 49.3 26.3 1998 c,d 57.7 0.5 43.8 Peru 1997 64.7 40.4 49.0 1996 15.5 5.4 41.4 17.1 1996 c,d 46.2 1.6 35.4 Philippines 1997 50.7 21.5 36.8 .. .. .. .. 1997 a,b 46.2 2.3 36.6 Poland 1993 .. .. 23.8 1998 <2 <0.5 <2 <0.5 1998 a,b 31.6 3.2 24.7 Portugal .. .. .. 1994 <2 <0.5 <2 <0.5 1994–95 c,d 35.6 3.1 28.4 Romania 1994 27.9 20.4 21.5 1994 2.8 0.8 27.5 6.9 1998 a,b 31.1 3.2 25.0 Russian Federation 1994 .. .. 30.9 1998 7.1 1.4 25.1 8.7 1998 a,b 48.7 1.7 38.7 Rwanda 1993 .. .. 51.2 1983–85 35.7 7.7 84.6 36.7 1983–85 a,b 28.9 4.2 24.2 Saudi Arabia .. .. .. .. .. .. .. .. .. .. Senegal 1992 40.4 .. 33.4 1995 26.3 7.0 67.8 28.2 1995 a,b 41.3 2.6 33.5 Sierra Leone 1989 76.0 53.0 68.0 1989 57.0 39.5 74.5 51.8 1989 a,b 62.9 0.5 43.6 Singapore .. .. .. .. .. .. .. .. .. .. Slovak Republic .. .. .. 1992 <2 <0.5 <2 <0.5 1992 c,d 19.5 5.1 18.2 Slovenia .. .. .. 1998 <2 <0.5 <2 <0.5 1998 c,d 28.4 3.9 23.0 South Africa .. .. .. 1993 11.5 1.8 35.8 13.4 1993–94 a,b 59.3 1.1 45.9 Spain .. .. .. .. .. .. .. 1990 c,d 32.5 2.8 25.2 Sri Lanka 1995–96 .. .. 25.0 1995 6.6 1.0 45.4 13.5 1995 a,b 34.4 3.5 28.0 Sweden .. .. .. .. .. .. .. 1992 c,d 25.0 3.7 20.1 Switzerland .. .. .. .. .. .. .. 1992 c,d 33.1 2.6 25.2 Syrian Arab Republic .. .. .. .. .. .. .. .. .. .. Tajikistan .. .. .. .. .. .. .. 1998 a,b 34.7 3.2 25.2 Tanzania 1993 49.7 24.4 41.6 1993 19.9 4.8 59.7 23.0 1993 a,b 38.2 2.8 30.1 Thailand 1992 15.5 10.2 13.1 1998 <2 <0.5 28.2 7.1 1998 a,b 41.4 2.8 32.4 Togo 1987–89 .. .. 32.3 .. .. .. .. .. .. .. Tunisia 1990 21.6 8.9 14.1 1995 <2 <0.5 10.0 2.3 1995 a,b 41.7 2.3 31.8 Turkey .. .. .. 1994 2.4 0.5 18.0 5.0 1994 a,b 41.5 2.3 32.3 Turkmenistan .. .. .. 1998 12.1 2.6 44.0 15.4 1998 a,b 40.8 2.6 31.7 Uganda 1993 .. .. 55.0 .. .. .. .. 1996 a,b 37.4 3.0 29.8 Ukraine 1995 .. .. 31.7 1999 2.9 0.6 31.0 8.0 1999 a,b 29.0 3.7 23.2 United Kingdom .. .. .. .. .. .. .. 1995 c,d 36.8 2.3 27.7 United States .. .. .. .. .. .. .. 1997 c,d 40.8 1.8 30.5 Uruguay .. .. .. 1989 <2 <0.5 6.6 1.9 1989 c,d 42.3 2.1 32.7 Uzbekistan .. .. .. 1993 3.3 0.5 26.5 7.3 1998 a,b 44.7 1.2 32.8 Venezuela, RB 1989 .. .. 31.3 1998 23.0 10.8 47.0 23.0 1998 c,d 49.5 0.8 36.5 Vietnam 1993 57.2 25.9 50.9 .. .. .. .. 1998 a,b 36.1 3.6 29.9 Yemen, Rep. 1992 19.2 18.6 19.1 1998 15.7 4.5 45.2 15.0 1998 a,b 33.4 3.0 25.9 Yugoslavia, Fed. Rep. .. .. .. .. .. .. .. .. .. .. Zambia 1993 .. .. 86.0 1998 63.7 32.7 87.4 55.4 1998 a,b 52.6 1.1 41.0 Zimbabwe 1990–91 31.0 10.0 25.5 1990–91 36.0 9.6 64.2 29.4 1995 a,b 50.1 2.0 40.4 a. Refers to expenditure shares by percentiles of population. b. Ranked by per capita expenditure. c. Refers to income shares by percentiles of population. d. Ranked by per capita income. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 3. Economic activity Agricultural Gross domestic productivity product Agr. Value added Household General External GDP per agricultural final gov’t. final Gross balance of implicit Avg. worker Value added as % of GDP cons. cons. capital goods and deflator Millions annual 1995 dollars Agricultural Industry Services expenditure expenditure formation services Avg. annual of dollars % growth % of GDP % of GDP % of GDP % of GDP % growth 2001 1990–2001 1988–90 1998–2000 2001 2001 2001 2001 2001 2001 2001 1990–2001 Albania 4,114 3.7 1,136 1,978 49 27 24 91 13 19 –23 34.4 Algeria 53,009 2.0 1,776 1,962 12 76 12 41 15 28 16 16.9 Angola 9,471 2.0 226 121 8 67 25 54 . .a 34 12 658.8 Argentina 268,773 3.7 7,284 10,246 5 28 68 71 14 16 –1 4.2 Armenia 2,012 –0.7 .. 5,477 26 34 40 99 9 19 –26 171.5 Australia 368,571 4.0 24,281 33,765 3 26 71 60 19 24 –2 1.6 Austria 188,742 2.1 15,575 28,523 2 33 65 57 20 24 0 1.9 Azerbaijan 5,692 2.7 .. 708 20 38 42 59 10 27 4 59.1 Bangladesh 46,652 4.9 251 296 23 25 52 79 5 23 –7 3.9 Belarus 12,070 –0.8 .. 1,985 16 42 42 70 14 18 –2 318.1 Belgium 227,618 2.1 29,807 55,874 2 27 72 54 21 22 3 1.9 Benin 2,269 4.8 397 586 38 15 47 80 12 20 –13 7.6 Bolivia 7,960 3.8 956 1,039 22 15 63 74 16 18 –8 7.9 Botswana 5,142 5.2 773 688 4 44 52 58 28 20 –6 8.9 Brazil 502,509 2.8 2,985 4,356 8 36 56 60 20 21 –1 168.1 Bulgaria 12,714 –1.5 3,413 6,252 14 28 58 69 19 17 –5 92.9 Burkina Faso 2,328 4.9 148 180 35 17 47 74 16 29 –18 3.5 Burundi 689 –2.2 183 141 50 19 31 90 14 7 –11 12.6 Cambodia 3,384 4.8 398 403 37 20 42 92 . .a 15 –7 21.7 Cameroon 8,591 2.1 842 1,104 46 21 33 69 11 18 2 4.8 Canada 677,178 3.0 25,362 36,597 .. .. .. 58 19 20 3 1.4 Central African Republic 978 2.1 381 469 55 21 25 78 11 14 –3 4.3 Chad 1,603 2.5 173 227 39 14 48 91 8 43 –42 6.7 Chile 63,545 6.4 4,853 5,712 11 34 56 63 12 23 1 6.8 China 1,159,017 10.0 227 321 15 52 33 48 12 39 1 6.2 Hong Kong, China 162,642 3.9 .. .. 0 14 85 58 10 28 5 3.3 Colombia 83,432 2.7 3,889 3,448 13 30 57 68 19 12 1 20.1 Congo, Dem. Rep. .. –5.1 248 252 .. .. .. .. .. .. .. 1,423.1 Congo, Rep. 2,751 –0.1 489 475 6 67 26 28 12 25 35 10.4 Costa Rica 16,156 5.1 3,721 5,140 9 29 62 70 14 18 –2 16.3 Côte d’Ivoire 10,411 3.1 937 1,097 24 22 54 74 9 10 7 8.4 Croatia 19,821 1.1 .. 8,839 8 32 59 60 21 26 –7 72.2 Czech Republic 56,424 1.1 .. 5,637 4 41 55 54 20 30 –4 10.6 Denmark 162,817 2.5 29,551 54,090 3 26 71 48 25 22 5 2.3 Dominican Republic 21,211 6.0 2,010 2,769 11 34 55 78 8 24 –10 9.1 Ecuador 17,982 1.7 1,489 1,773 11 33 56 68 10 25 –3 38.5 Egypt, Arab Rep. 97,545 4.6 997 1,240 17 34 49 72 10 23 –5 7.7 El Salvador 13,963 4.5 1,619 1,710 10 30 60 88 10 17 –15 6.9 Eritrea 681 3.1 .. .. 17 29 54 146 . .a 40 –85 10.4 Estonia 5,281 0.2 .. 3,698 6 28 66 55 21 29 –5 45.4 Ethiopia 6,366 4.9 .. 138 52 11 37 83 17 17 –16 6.1 Finland 121,987 3.0 23,997 36,557 4 34 62 50 21 20 10 2.0 France 1,302,793 1.8 30,641 53,785 3 26 71 55 23 21 1 1.5 Georgia 3,138 –5.6 .. .. 21 23 57 89 9 19 –16 279.0 Germany 1,873,854 1.5 16,878 29,553 1 31 68 58 19 23 0 1.9 Ghana 5,301 4.2 543 558 36 25 39 79 16 24 –18 26.6 Greece 116,347 2.3 10,525 13,400 8 24 68 71 15 22 –8 8.5 Guatemala 20,629 4.1 1,932 2,112 23 20 58 88 5 16 –9 9.9 Guinea 2,885 4.1 249 292 25 38 37 77 6 25 –8 5.0 Haiti 3,771 –0.4 430 334 28 20 51 104 . .a 11 –15 19.3 Honduras 6,386 3.1 855 979 18 32 51 66 13 35 –14 18.0 Hungary 52,361 1.9 5,133 5,016 .. .. .. 64 10 31 –4 18.3 India 477,555 5.9 343 397 24 27 48 68 11 24 –3 7.7 Indonesia 145,306 3.8 674 734 16 47 37 67 7 17 9 15.8 Iran, Islamic Rep. 118,868 3.6 2,838 3,756 19 26 54 60 15 20 5 25.8 Ireland 101,185 7.6 .. .. 4 36 60 49 14 23 14 3.6 Israel 110,386 5.1 .. .. .. .. .. 59 29 19 –7 10.0 Italy 1,090,910 1.6 13,916 24,827 3 30 68 60 18 20 1 3.6 Jamaica 7,784 0.6 1,027 1,559 6 31 63 66 16 29 –11 22.0 Japan 4,245,191 1.3 25,293 30,086 1 32 66 56 16 26 2 0.0 Jordan 8,829 4.8 1,810 1,422 2 25 73 80 24 22 –26 2.9 Kazakhstan 22,635 –2.8 .. 1,421 9 48 43 70 14 13 3 168.6 Kenya 10,419 2.0 264 225 21 19 60 85 11 13 –9 13.0 Korea, Rep. 422,167 5.7 7,159 12,374 4 41 54 61 10 27 2 4.5 Kuwait 37,783 3.2 .. .. .. .. .. 41 22 11 26 3.0 Kyrgyz Republic 1,525 –2.9 .. 1,583 38 27 35 65 20 15 0 95.2 Lao PDR 1,712 6.4 457 578 53 23 24 .. .. .. .. 28.4 Latvia 7,549 –2.2 .. 2,499 5 26 69 59 22 28 –9 42.0 Lebanon 16,709 5.4 . . 29,241 12 22 66 94 18 19 –31 15.1 Lesotho 789 3.9 595 540 20 46 34 91 27 33 –52 9.6 Lithuania 11,834 –2.3 .. 3,129 8 31 61 68 17 22 –6 63.3 Macedonia, FYR 3,445 –0.2 .. 4,095 11 31 58 71 24 19 –14 66.0 Madagascar 4,566 2.4 195 181 25 12 63 83 7 18 –8 17.8 Malawi 1,826 3.7 81 130 37 16 47 86 17 10 –14 33.0 Malaysia 87,540 6.5 5,680 6,519 8 50 42 45 12 24 18 3.6 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  Agricultural Gross domestic productivity product Agr. Value added Household General External GDP per agricultural final gov’t. final Gross balance of implicit Avg. worker Value added as % of GDP cons. cons. capital goods and deflator Millions annual 1995 dollars Agricultural Industry Services expenditure expenditure formation services Avg. annual of dollars % growth % of GDP % of GDP % of GDP % of GDP % growth 2001 1990–2001 1988–90 1998–2000 2001 2001 2001 2001 2001 2001 2001 1990–2001 Mali 2,629 4.1 252 285 38 26 36 78 13 20 –10 6.9 Mauritania 1,030 4.2 391 480 21 29 50 68 16 26 –10 6.2 Mexico 617,817 3.1 1,522 1,767 4 27 69 70 12 21 –2 18.2 Moldova 1,478 –8.4 .. 1,299 28 21 52 85 15 21 –21 103.1 Mongolia 1,049 1.2 1,125 1,300 30 17 53 67 19 30 –16 51.4 Morocco 33,733 2.5 1,847 1,785 16 32 53 62 18 25 –5 2.6 Mozambique 3,561 7.5 123 134 22 26 52 74 12 31 –18 28.5 Myanmar .. .. .. .. 60 9 31 87 . .a 13 0 25.2 Namibia 3,168 4.1 1,031 1,468 11 28 61 54 29 24 –7 9.4 Nepal 5,525 4.9 188 188 38 23 39 72 10 26 –8 7.8 Netherlands 374,976 2.8 34,080 53,819 3 27 70 50 23 22 5 2.1 New Zealand 48,277 2.9 22,341 27,106 .. .. .. 64 16 21 –1 1.5 Nicaragua .. 2.8 1,251 1,813 .. .. .. .. .. .. .. .. Niger 1,939 2.6 204 214 39 18 44 84 13 13 –10 5.8 Nigeria 41,237 2.5 499 672 30 46 25 67 13 23 –4 26.5 Norway 165,458 3.5 21,200 33,305 2 43 55 43 19 22 16 3.2 Pakistan 59,605 3.7 513 630 25 23 51 78 11 15 –4 9.7 Panama 10,170 3.8 2,192 2,632 7 17 77 61 15 30 –6 1.9 Papua New Guinea 2,959 3.6 666 767 26 42 32 64 14 19 3 7.3 Paraguay 6,926 2.0 3,261 3,508 21 27 52 83 10 22 –15 11.8 Peru 54,047 4.3 1,371 1,693 8 27 65 72 11 18 –1 23.3 Philippines 71,438 3.3 1,339 1,328 15 31 54 68 14 17 0 8.2 Poland 174,597 4.5 1,632 1,874 3 32 65 80 . .a 27 –7 21.4 Portugal 108,479 2.7 5,307 7,235 4 31 66 63 20 28 –12 5.1 Romania 39,714 –0.3 2,367 3,592 12 37 51 73 13 19 –5 90.7 Russian Federation 309,951 –3.7 .. 2,249 7 37 56 51 14 22 13 139.6 Rwanda 1,703 0.8 295 235 44 22 34 85 14 18 –16 13.1 Saudi Arabia 173,287 1.5 7,060 .. .. .. .. 33 27 16 24 2.2 Senegal 4,620 3.9 344 304 18 27 55 78 10 20 –8 4.2 Sierra Leone 749 –2.8 612 336 49 31 21 85 16 9 –9 27.1 Singapore 92,252 7.8 27,176 49,905 0 34 66 40 10 31 18 1.3 Slovak Republic 20,522 2.3 .. .. 4 30 66 55 17 35 –7 10.0 Slovenia 18,810 2.9 .. 31,539 3 38 58 55 21 28 –4 18.3 South Africa 113,274 2.1 3,586 3,866 3 31 66 63 19 15 3 9.3 Spain 577,539 2.6 16,127 21,824 4 31 66 59 17 26 –2 3.8 Sri Lanka 16,346 5.1 694 753 19 26 55 70 10 26 –7 9.1 Sweden 210,108 2.0 26,070 34,556 .. .. .. 50 26 18 5 2.0 Switzerland 247,362 0.9 .. .. .. .. .. 61 14 20 5 1.2 Syrian Arab Republic 17,938 5.5 2,056 2,890 24 30 46 63 13 19 5 6.4 Tajikistan 1,058 –8.7 .. 1,236 19 26 55 76 8 20 –4 202.3 Tanzania b 9,119 3.1 178 189 45 16 39 83 10 19 –11 20.0 Thailand 114,760 3.8 778 909 10 40 50 61 9 24 5 3.9 Togo 1,259 2.2 451 538 39 21 40 87 9 20 –16 6.6 Tunisia 20,035 4.7 2,228 3,083 12 29 59 62 14 28 –4 4.3 Turkey 147,627 3.3 1,847 1,878 15 27 58 69 13 15 2 74.1 Turkmenistan 5,962 –2.8 .. 1,229 27 50 23 34 16 40 10 328.0 Uganda 5,707 6.8 298 353 42 19 38 88 12 19 –20 11.3 Ukraine 37,588 –7.9 .. 1,345 15 41 45 59 18 19 3 220.9 United Kingdom 1,406,310 2.6 28,660 34,938 1 29 70 65 19 18 –2 2.8 United States 10,171,400 3.5 .. .. .. .. .. 68 14 21 –3 2.0 Uruguay 18,429 2.9 6,833 9,100 6 27 67 75 13 14 –1 27.7 Uzbekistan 11,270 0.0 .. 1,035 36 21 43 69 18 11 2 211.6 Venezuela, RB 124,948 1.5 4,449 5,143 5 50 45 68 8 19 5 42.8 Vietnam 32,903 7.6 181 240 24 37 39 69 6 27 –2 14.0 Yemen, Rep. 9,098 5.6 333 377 15 42 43 65 14 18 2 21.3 Yugoslavia, Fed. Rep. 10,883 0.7 .. .. .. .. .. 81 23 16 –20 54.4 Zambia 3,647 0.8 219 217 22 26 52 85 12 21 –17 48.1 Zimbabwe 9,057 1.8 295 366 18 24 58 72 19 8 1 28.4 World 31,283,839 t 2.7 w ..w 887 w 5w 31 w 64 w 61 w 17 w 22 w 1w Low income 1,083,360 3.4 361 417 23 32 45 70 11 20 –1 Middle income 5,097,044 3.4 633 802 10 38 52 59 15 25 1 Lower middle income 2,733,167 3.7 395 543 12 41 46 57 13 27 3 Upper middle income 2,360,861 3.1 .. 3,661 7 35 59 62 17 22 –1 Low & middle income 6,179,333 3.4 514 620 12 37 51 61 14 24 1 East Asia & Pacific 1,664,211 7.5 .. .. 14 48 37 53 11 31 4 Europe & Central Asia 986,652 –0.9 .. 2,099 10 34 56 60 15 21 4 Latin America & Carib. 1,943,350 3.1 2,595 3,165 7 34 59 64 17 21 –1 Middle East & N. Africa 652,277 3.0 .. .. .. .. .. 52 18 20 10 South Asia 615,308 5.5 345 401 24 26 49 70 10 23 –3 Sub-Saharan Africa 315,269 2.6 386 362 15 29 57 70 16 17 –2 High income 25,103,679 2.5 .. .. .. .. .. 61 17 22 0 a. Data on general government final consumption expenditure are not available separately; they are included in household final consumption expenditure. b. Data cover mainland Tanzania only. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 4. Trade, aid, and finance High Domestic Merchandise trade Manufactured technology Current Net private Foreign Official External debt credit exports imports exports exports account capital direct development provided % of total % of balance flows investment assistancea Total Present by banking Millions Millions merchandise manufactured Millions Millions Millions Dollars Millions value sector of dollars of dollars exports exports of dollars of dollars of dollars per capita of dollars % of GNI % of GDP 2001 2001 2000 2000 2001 2000 2000 2000 2000 2000 2001 Albania 280 1,210 82 1 –246 142 143 93 784 13 46.5 Algeria 20,050 9,700 2 4 .. –1,212 10 5 25,002 50 30.2 Angola 7,350 3,950 .. .. –414 1,206 1,698 23 10,146 137 –0.5 Argentina 26,655 20,311 32 9 –8,970 16,619 11,665 2 146,172 56 37.1 Armenia 335 890 43 5 –278 159 140 57 898 31 10.0 Australia 63,386 63,886 29 15 –15,330 .. 11,527 .. .. .. 94.0 Austria 70,293 73,857 83 14 –4,027 .. 9,066 .. .. .. 125.9 Azerbaijan 2,460 1,725 8 4 –73 175 130 17 1,184 20 5.7 Bangladesh 6,300 8,154 91 0 –1,284 269 280 9 15,609 20 38.7 Belarus 7,470 8,310 67 4 –273 123 90 4 851 8 19.2 Belgium b 188,862 181,705 b 78 b 10 13,037 .. 17,902 b .. .. .. 121.5 Benin 380 670 3 0 –97 30 30 38 1,598 45c 4.8 Bolivia 1,257 1,673 29 .. –464 923 733 57 5,762 34c 63.0 Botswana 2,310 2,360 .. .. 517 27 30 19 413 6 –75.2 Brazil 58,223 58,265 59 19 –23,208 45,672 32,779 2 237,953 39 59.2 Bulgaria 5,125 7,315 57 .. –701 1,114 1,002 38 10,026 82 22.0 Burkina Faso 210 540 .. .. –77 10 10 30 1,332 31c 16.3 Burundi 35 140 0 .. –33 12 12 14 1,100 97 32.7 Cambodia 1,531 1,476 .. .. –19 126 126 33 2,357 62 6.5 Cameroon 1,770 1,500 5 1 –171 –21 31 26 9,241 75 16.4 Canada 262,240 228,250 64 19 18,884 .. 62,758 .. .. .. 93.2 Central African Republic 150 120 .. .. 16 5 5 20 872 57 12.1 Chad 180 360 .. .. –660 14 15 17 1,116 42 12.5 Chile 17,665 17,184 16 3 –1,782 4,833 3,675 3 36,978 51 76.7 China 266,155 243,567 88 19 19,404 58,295 38,399 1 149,800 13 132.7 Hong Kong, China d 190,676 202,252 95 d 23 8,827 .. .. 1 .. .. 142.0 Colombia 12,414 12,947 34 7 –1,693 3,130 2,376 4 34,081 41 34.3 Congo, Dem. Rep. 420 310 .. .. .. 1 1 4 11,645 .. .. Congo, Rep. 2,395 940 .. .. .. 14 14 11 4,887 206 13.5 Costa Rica 5,010 6,564 66 .. –649 610 409 3 4,466 31 33.3 Côte d’Ivoire 3,850 2,860 14 .. 64 –47 106 22 12,138 117c 21.9 Croatia 4,659 9,044 73 8 –623 2,451 926 15 12,120 65 51.9 Czech Republic 33,370 36,505 88 8 –2,237 3,299 4,583 43 21,299 43 51.8 Denmark 51,812 45,551 64 21 4,102 .. 34,192 .. .. .. 56.7 Dominican Republic 5,550 8,870 .. .. –1,026 1,142 953 7 4,598 23 41.7 Ecuador 4,474 5,299 10 6 928 904 710 12 13,281 108 0.0 Egypt, Arab Rep. 4,095 13,060 37 0 –34 1,967 1,235 21 28,957 23 104.6 El Salvador 2,865 5,027 48 6 –418 338 185 29 4,023 29 42.3 Eritrea .. .. .. .. –325 35 35 43 311 26 .. Estonia 3,505 4,550 73 30 –353 485 387 47 3,280 66 48.0 Ethiopia 410 .. 10 0 –264 42 50 11 5,481 52 57.0 Finland 43,304 32,043 85 27 7,859 .. 9,125 .. .. .. 63.2 France 319,470 322,934 81 24 25,644 .. 43,173 .. .. .. 108.6 Georgia 290 810 .. .. –209 155 131 34 1,633 42 20.5 Germany 569,584 493,045 85 18 3,815 .. 189,178 .. .. .. 145.3 Ghana 1,700 3,030 15 14 –314 71 110 32 6,657 81c 40.8 Greece 8,764 27,359 50 9 –9,400 .. 1,083 .. .. .. 101.8 Guatemala 2,430 5,635 32 8 –1,049 178 230 23 4,622 23 15.4 Guinea 795 1,195 30 .. –155 63 63 21 3,388 80 9.4 Haiti 141 1,036 .. .. .. 13 13 26 1,169 17 31.5 Honduras 1,270 2,990 33 2 –204 301 282 70 5,487 54 34.7 Hungary 30,780 33,925 86 26 –1,097 1,721 1,692 25 29,415 63 49.5 India 43,877 50,533 79 4 –3,493 8,771 2,315 1 99,062 0 53.8 Indonesia 56,716 31,170 57 16 7,985 –11,210 –4,550 8 141,803 96 61.1 Iran, Islamic Rep. 26,350 16,450 7 2 12,645 –610 39 2 7,953 7 46.2 Ireland 83,437 50,924 86 48 –1,043 .. 22,778 .. .. .. 113.1 Israel 29,019 35,123 94 25 –1,730 .. 4,392 128 .. .. 86.5 Italy 241,257 233,727 88 9 1,591 .. 13,175 .. .. .. 99.6 Jamaica 1,283 3,409 73 0 –275 898 456 4 4,287 59 23.4 Japan 404,686 350,095 94 28 89,280 .. 8,227 .. .. .. 308.7 Jordan 2,230 4,945 69 8 –235 455 558 113 8,226 90 89.7 Kazakhstan 8,750 6,445 20 10 158 1,900 1,250 13 6,664 39 11.4 Kenya 1,775 2,890 21 4 –429 53 111 17 6,295 46 45.6 Korea, Rep. 150,653 141,116 91 35 8,617 13,215 9,283 –4 134,417 28 110.4 Kuwait 16,234 6,963 20 1 8,566 .. 16 1 .. .. 82.2 Kyrgyz Republic 560 475 20 5 –10 –65 –2 44 1,829 109 9.7 Lao PDR 320 437 .. .. 90 72 72 53 2,499 72 15.7 Latvia 2,030 3,535 56 4 –758 583 407 38 3,379 46 31.4 Lebanon 890 7,291 .. .. –3,984 2,028 298 45 10,311 61 201.9 Lesotho 260 750 .. .. –151 111 118 20 716 45 5.4 Lithuania 4,630 6,185 60 4 –574 799 379 28 4,855 43 16.0 Macedonia, FYR 1,170 1,630 66 1 –107 187 176 124 1,465 35 14.4 Madagascar 310 780 50 3 –260 83 83 21 4,701 79 16.1 Malawi 310 550 .. .. –524 45 45 43 2,716 89c 9.3 Malaysia 88,521 74,384 80 59 8,409 3,228 1,660 2 41,797 52 156.2 * Taiwan, China 122,902 107,243 95 39 9,316 .. .. 0 .. .. .. Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  High Domestic Merchandise trade Manufactured technology Current Net private Foreign Official External debt credit exports imports exports exports account capital direct development provided % of total % of balance flows investment assistance a Total Present by banking Millions Millions merchandise manufactured Millions Millions Millions Dollars Millions value sector of dollars of dollars exports exports of dollars of dollars of dollars per capita of dollars % of GNI % of GDP 2001 2001 2000 2000 2001 2000 2000 2000 2000 2000 2001 Mali 485 770 .. .. .. 76 76 33 2,956 58c 17.1 Mauritania 280 350 .. .. 30 3 5 80 2,500 126 0.5 Mexico 158,542 176,162 83 22 –17,683 11,537 13,286 –1 150,288 28 24.7 Moldova 565 915 33 3 –103 209 128 29 1,233 84 27.5 Mongolia 250 461 .. .. –52 27 30 91 859 59 12.5 Morocco 7,100 10,830 64 12 –352 –293 10 15 17,944 49 87.2 Mozambique 490 1,025 10 2 –764 138 139 50 7,135 32c 13.3 Myanmar 1,760 2,461 .. .. –651 188 255 2 6,046 .. 35.3 Namibia 1,630 1,720 .. .. 204 .. .. 86 .. .. 48.7 Nepal 646 1,235 77 0 –293 –4 4 17 2,823 27 43.2 Netherlands 229,830 207,858 70 35 12,405 .. 54,138 .. .. .. 157.5 New Zealand 13,754 13,335 28 10 –1,587 .. 3,209 .. .. .. 120.0 Nicaragua 606 1,776 8 5 –493 395 254 111 7,019 .. .. Niger 265 430 2 5 –216 13 15 19 1,638 58c 8.0 Nigeria 19,150 10,260 0 13 4,926 908 1,082 1 34,134 74 11.3 Norway 57,856 32,361 18 17 24,078 .. 5,882 .. .. .. 46.5 Pakistan 9,209 10,206 85 0 –1,946 –53 308 5 32,091 45 44.7 Panama 972 3,017 16 0 –499 947 603 6 7,056 77 114.9 Papua New Guinea 1,805 1,072 2 42 –74 128 130 54 2,604 66 24.2 Paraguay 972 2,145 19 3 –202 –16 82 15 3,091 39 29.3 Peru 7,140 8,656 20 3 –1,628 1,553 680 16 28,560 55 25.7 Philippines 33,589 31,373 92 59 4,503 2,459 2,029 8 50,063 64 58.6 Poland 35,500 50,035 80 3 –2,452 13,195 9,342 36 63,561 37 37.3 Portugal 23,730 37,660 85 5 –10,080 .. 6,227 .. .. .. 152.0 Romania 11,450 15,515 77 6 –1,359 1,900 1,025 19 10,224 27 12.4 Russian Federation 103,210 53,500 22 14 29,156 2,200 2,714 11 160,300 60 24.3 Rwanda 77 260 .. .. –73 14 14 38 1,271 41 12.6 Saudi Arabia 68,200 32,100 7 0 14,502 .. .. 1 .. .. 68.4 Senegal 1,080 1,510 30 13 –297 106 107 44 3,372 56 24.7 Sierra Leone 28 166 .. .. .. 1 1 36 1,273 132 52.1 Singapore 121,731d 115,961 86 63 21,797 .. 6,390 0 .. .. 89.6 Slovak Republic 12,630 14,765 85 4 –694 2,185 2,052 21 9,462 48 63.2 Slovenia 9,331 10,185 90 5 –66 .. 176 31 .. .. 49.5 South Africa 29,284 e 28,700 e 54 e 1 –166 2,736 961 11 24,861 19 81.0 Spain 110,830 144,467 78 8 –15,082 .. 36,023 .. .. .. 126.2 Sri Lanka 4,900 6,100 75 3 –1,042 262 173 14 9,065 44 45.1 Sweden 75,198 62,470 85 22 6,696 .. 22,125 .. .. .. 79.3 Switzerland 82,064 84,077 91 19 32,542 .. 17,902 .. .. .. 173.3 Syrian Arab Republic 5,410 4,290 8 1 1,062 107 111 10 21,657 128 27.0 Tajikistan 575 775 .. .. –61 64 24 23 1,170 100 .. Tanzania 780 1,660 15 6 –998 182 193 31 7,445 50 10.1 Thailand 64,223 60,190 76 32 6,195 –1,383 3,366 11 79,675 64 112.0 Togo 432 620 31 0 –127 30 30 15 1,435 85 20.5 Tunisia 6,615 9,505 77 3 –937 966 752 23 10,610 57 73.5 Turkey 31,220 40,455 81 5 –9,819 11,416 982 5 116,209 57 71.1 Turkmenistan 2,560 2,105 7 5 412 473 130 6 2,259 .. 30.7 Uganda 520 1,430 6 10 –889 231 220 37 3,408 16c 10.1 Ukraine 16,615 16,105 .. .. 1,402 927 595 11 12,166 38 23.8 United Kingdom 273,462 332,523 82 32 –25,107 .. 133,974 .. .. .. 142.3 United States 730,897 1,180,497 83 34 –417,440 .. 287,680 .. .. .. 160.6 Uruguay 2,088 3,119 42 2 –557 574 298 5 8,196 42 54.3 Uzbekistan 2,655 2,715 .. .. –28 18 100 8 4,340 31 .. Venezuela, RB 28,610 18,775 9 3 4,364 5,454 4,464 3 38,196 32 15.5 Vietnam 15,100 16,000 .. .. 507 581 1,298 22 12,787 36 39.5 Yemen, Rep. 4,130 2,450 1 0 1,107 –201 –201 15 5,615 48 2.9 Yugoslavia, Fed. Rep. 1,903 4,837 .. .. .. 0 0 107f 11,960g 142 .. Zambia 880 760 .. .. –553 191 200 79 5,730 162 51.4 Zimbabwe 1,770 1,540 28 2 .. 29 79 14 4,002 52 52.9 World 6,163,167 t 6,354,719 t 78 w 20 w . . s 1,167,337 s ..w ..s 152.5 w Low income 219,806 201,634 53 7 4,829 6,812 9 552,095 46.2 Middle income 1,326,040 1,266,001 59 13 207,538 150,572 8 1,798,508h 49.0 Lower middle income 705,744 661,566 61 13 92,195 61,925 7 909,722 91.3 Upper middle income 620,295 604,427 57 13 115,344 88,647 8 888,787h 46.2 Low & middle income 1,545,853 1,467,974 58 13 212,631 157,408 11 2,356,253 h 48.4 East Asia & Pacific 531,552 466,380 80 30 52,478 42,847 5 498,536 120.4 Europe & Central Asia 325,397i 327,066i 53i 10 45,446 28,495 23 499,344 37.8 Latin America & Carib. 346,471 375,183 48 16 97,305 75,088 10 774,419 38.3 Middle East & N. Africa 185,847 138,431 16 3 1,074 1,209 16 203,785 72.3 South Asia 65,213 77,339 80 4 9,254 3,093 3 164,375 51.4 Sub-Saharan Africa 91,356 83,102 36 8 7,074 6,676 20 215,794 47.2 High income 4,617,275 4,885,097 83 23 . . 1,009,929 .. .. 172.1 a. Regional aggregates include data for economies that are not specified elsewhere. World and income group totals include aid not allocated by country or region. b. Includes Luxembourg. c. Data are from debt sustainability analysis undertaken as part of the Heavily Indebted Poor Countries (HIPC) initiative. d. Includes re-exports. e. Data on total exports and imports refer South Africa only. Data on export commodity shares refer to the South African Customs Union (Botswana, Lesotho, Namibia, South Africa, and Swaziland). f. Aid to the states of the former Socialist Federal Republic of Yugoslavia that is not otherwise specified is included in regional and income group aggregates. g. Data are estimates and reflect borrowing by the former Socialist Federal Republic of Yugoslavia that are not yet allocated to the successor republics. h. Includes data for Gibraltar not included in other tables. i. Data include the intratrade of the Baltic states and the Commonwealth of Independent States. (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Table 1a. Key indicators for other economies PPP Population Gross national gross national Gross Under-5 Adult Carbon income (GNI)a income (GNI)b domestic Life mortality illiteracy dioxide Avg. density product expectancy rate rate emissions annual people per Millions per capita Millions per capita per capita at birth Per % of people Thousands Thousands % growth sq. km of dollars dollars of dollars dollars % growth Years 1,000 15 and above of tons 2001 1990–2001 2001 2001 2001 2001 2001 2000–2001 2000 2000 2000 1998 Afghanistan 27,248c 3.9 41.8 .. . .d .. .. .. 43 279 .. 1,037 American Samoa 65 .. 327.2 .. . .e .. .. .. .. .. .. 282 Andorra 67 .. 148.9 .. .. f .. .. .. 80 .. .. .. Antigua and Barbuda 68 0.6 155.7 621 9,070 676 9,870 –0.4 75 19 .. 337 Aruba 104 4.2 547.4 .. . .f .. .. .. .. .. .. 1,883 Bahamas, The 307 1.7 30.7 4,533 14,960 4,969 16,400 .. 69 22 5 1,792 Bahrain 714 3.2 1035.3 6,247 9,370 9,605 14,410 .. 73 11 12 18,688 Barbados 268 0.4 623.7 2,469 9,250 4,010 15,020 .. 75 18 .. 1,569 Belize 247 2.4 10.8 718 2,910 1,323 5,350 0.2 74 38 7 399 Bermuda 63 .. 1260.0 .. . .f .. .. .. .. .. .. 462 Bhutan 828 2.9 17.6 529 640 1,266 g 1,530 g 4.0 62 .. .. 386 Bosnia and Herzegovina 4,060 –0.8 79.6 5,037 1,240 .. .. 3.8 73 18 .. 4,686 Brunei 345 2.7 65.5 .. .. f .. .. .. 76 11 8 5,488 Cape Verde 454 2.6 112.6 596 1,310 2,211g 4,870g 0.4 69 48 26 121 Cayman Islands 35 .. 134.6 .. . .f .. .. .. .. .. .. 289 Channel Islands 149 0.4 768 .. . .f .. .. .. 79 .. .. .. Comoros 572 2.6 256.5 217 380 922g 1,610g –0.5 61 80 44 70 Cuba 11,222 0.5 102.2 .. .. h .. .. .. 76 9 3 24,881 Cyprus 761 1.0 82.3 9,361 12,370 15,734 g 20,780 g .. 78 9 3 5,918 Djibouti 644 2.9 27.8 572 890 1,369 2,120 –0.3 46 178 35 366 Dominica 73 0.1 97.6 224 3,060 369 5,040 –5.0 76 .. .. 84 Equatorial Guinea 469 2.6 16.7 327 700 2,644 5,640 –1.3 51 167 17 253 Faeroe Islands 45 .. 32.1 .. . .f .. .. .. .. .. .. 641 Fiji 824 1.0 45.1 1,755 2,130 4,231 5,140 1.1 69 21 7 721 French Polynesia 239 1.7 65.2 4,064 17,290 5,486 g 23,340 g .. 73 13 .. 561 Gabon 1,261 2.7 4.9 3,990 3,160 6,890 5,460 0.0 53 89 .. 2,820 Gambia, The 1,341 3.3 134.1 440 330 2,319g 1,730g 2.7 53 .. 63 227 Greenland 56 .. 0.2 .. .. f .. .. .. .. .. .. 528 Grenada 99 0.5 291.2 368 3,720 666 6,720 –5.7 72 17 .. 183 Guam 157 1.4 285.4 .. . .f .. .. .. 78 10 .. 4,111 Guinea-Bissau 1,226 2.4 43.6 202 160 872 g 710 g –2.0 45 211 62 846 Guyana 766 0.4 3.9 641 840 2,870g 3,750g 0.8 63 73 2 1,649 Iceland 284 1.0 2.8 8,201 28,880 8,470 29,830 1.9 80 6 .. 2,083 Iraq 23,750 2.5 54.3 .. . .h .. .. .. 61 121 44 82,378 Isle of Man 75 .. 131 .. . .e .. .. .. .. .. .. .. Kiribati 93 2.3 127.2 77 830 .. .. –0.8 62 .. .. 22 Korea, Dem. Rep. 22,384 1.0 185.9 .. .. d .. .. .. 61 90 .. 226,149 Liberia 3,216 2.5 33.4 .. . .d .. .. .. 47 185 46 353 Libya 5,410 2.1 3.1 .. . .e .. .. .. 71 32 20 36,448 Liechtenstein 32 .. 200.0 .. . .f .. .. .. .. .. .. .. Luxembourg 444 1.4 172 18,550 41,770 21,352 48,080 3.8 77 6 .. 7,678 Macao, China 443 1.6 .. 6,385 i 14,580 i 7,967 g 18,190 g .. 79 .. 6 1,630 Maldives 283 2.6 942.5 578 2,040 1,277g 4,520g 4.5 68 34 3 330 Malta 392 0.8 1225.0 3,559 9,120 6,448 g 16,530 g .. 78 8 8 1,803 Marshall Islands 53 .. 292 115 2,190 .. .. –0.7 65 .. .. .. Mauritius 1,198 1.1 590.1 4,592 3,830 12,468 10,410 6.1 72 20 15 1,726 Mayotte 145 .. 388 .. .. e .. .. .. .. .. .. .. Micronesia, Fed. Sts. 120 2.0 171 258 2,150 .. .. –0.9 68 31 .. .. Monaco 32 .. 16,410 .. . .f .. .. .. .. .. .. .. Netherlands Antilles 217 1.2 271.0 .. . .f .. .. .. 76 16 3 7,753 New Caledonia 216 2.3 11.8 3,203 15,060 4,641 21,820 .. 73 11 .. 1,746 Northern Mariana Islands 72 .. 151 .. .. f .. .. .. .. .. .. .. Oman 2,452 3.7 11.5 .. .. e .. .. .. 74 22 28 20,270 Palau 20 .. 42.4 131 6,730 .. .. –1.0 70 .. .. 242 Puerto Rico 3,950 1.0 445.4 .. . .e .. .. .. 76 .. 6 17,567 Qatar 598 1.9 54.3 .. .. f .. .. .. 75 19 19 46,772 Samoa 171 0.6 60.5 260 1,520 933g 5,450g 9.3 69 .. 20 132 San Marino 27 .. 450.0 .. . .f .. .. .. 80 .. .. .. São Tomé and Principe 151 2.5 157.5 43 280 .. .. 0.8 65 62 .. 77 Seychelles 82 1.5 183.1 573 7,050 .. .. .. 72 14 .. 198 Solomon Islands 432 2.8 15.4 253 580 726g 1,680g –11.5 69 27 .. 161 Somalia 9,089 2.2 14.5 .. .. d .. .. .. 48 195 .. 0 St. Kitts and Nevis 41 –0.2 114.1 283 6,880 482 11,730 1.6 71 .. .. 103 St. Lucia 158 1.5 259.2 628 3,970 822 5,200 –4.8 71 19 .. 198 St. Vincent and the Grenadines 116 0.7 297.1 312 2,690 608g 5,250g –1.3 73 20 .. 161 Sudan 31,687 2.2 13.3 10,346 330 51,135 1,610 4.0 56 .. 42 3,597 Suriname 420 0.4 2.7 709 1,690 1,389g 3,310g 1.2 70 31 .. 2,139 Swaziland 1,068 3.0 62.1 1,388 1,300 5,006g 4,690g –0.6 46 119 20 399 Tonga 101 0.4 139.9 154 1,530 .. .. 2.5 71 23 .. 117 Trinidad and Tobago 1,310 0.7 255.3 7,249 5,540 11,887 9,080 4.3 73 19 6 22,396 United Arab Emirates 2,976 4.4 35.6 .. . .f .. .. .. 75 10 24 88,198 Vanuatu 201 2.8 16.5 212 1,050 546 2,710 –6.1 68 40 .. 62 Virgin Islands (U.S.) 122 1.5 360.0 .. . .f .. .. .. 78 11 .. 11,706 West Bank and Gaza 3,091 4.1 .. 4,177 1,350 .. .. –15.5 72 26 .. .. Note: Figures in italics are for years other than those specified. a. Preliminary World Bank estimates calculated using the World Bank Atlas method. b. Purchasing power parity; see the Technical Notes. c. Estimate does not account for recent refugee flows. d. Estimated to be low income ($745 or less). e. Estimated to be upper middle income ($2,976 to $9,205). f. Estimated to be high income ($9,206 or more). g. The estimate is based on regression; others are extrapolated from the latest International Comparison Programme benchmark estimates. h. Estimated to be lower middle income ($746 to $2,975). i. Refers to GDP and GDP per capita. (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  Classification of economies by region and income, FY2003a East Asia and the Pacific Latin America and the Caribbean South Asia High income OECD American Samoa UMC Antigua and Barbuda UMC Afghanistan LIC Australia Cambodia LIC Argentina UMC Bangladesh LIC Austria China LMC Barbados UMC Bhutan LIC Belgium Fiji LMC Belize LMC India LIC Canada Indonesia LIC Bolivia LMC Maldives LMC Denmark Kiribati LMC Brazil UMC Nepal LIC Finland Korea, Dem. Rep. LIC Chile UMC Pakistan LIC France Lao PDR LIC Colombia LMC Sri Lanka LMC Germany Malaysia UMC Costa Rica UMC Greece Marshall Islands LMC Cuba LMC Sub-Saharan Africa Iceland Micronesia, Fed. Sts. LMC Dominica UMC Angola LIC Ireland Mongolia LIC Dominican Republic LMC Benin LIC Italy Myanmar LIC Ecuador LMC Botswana UMC Japan Palau UMC El Salvador LMC Burkina Faso LIC Korea, Rep. Papua New Guinea LIC Grenada UMC Burundi LIC Luxembourg Philippines LMC Guatemala LMC Cameroon LIC Netherlands Samoa LMC Guyana LMC Cape Verde LMC New Zealand Solomon Islands LIC Haiti LIC Central African Republic LIC Norway Thailand LMC Honduras LMC Chad LIC Portugal Tonga LMC Jamaica LMC Comoros LIC Spain Vanuatu LMC Mexico UMC Congo, Dem. Rep. LIC Sweden Vietnam LIC Nicaragua LIC Congo, Rep. LIC Switzerland Panama UMC Côte d’Ivoire LIC United Kingdom Europe and Central Asia Paraguay LMC Equatorial Guinea LIC United States Albania LMC Peru LMC Eritrea LIC Armenia LIC Puerto Rico UMC Ethiopia LIC Other high income Azerbaijan LIC St. Kitts and Nevis UMC Gabon UMC Andorra Belarus LMC St. Lucia UMC Gambia, The LIC Aruba Bosnia and Herzegovina LMC St. Vincent and the Ghana LIC Bahamas, The Bulgaria LMC Grenadines LMC Guinea LIC Bahrain Croatia UMC Suriname LMC Guinea-Bissau LIC Bermuda Czech Republic UMC Trinidad and Tobago UMC Kenya LIC Brunei Estonia UMC Uruguay UMC Lesotho LIC Cayman Islands Georgia LIC Venezuela, RB UMC Liberia LIC Channel Islands Hungary UMC Madagascar LIC Cyprus Isle of Man UMC Middle East and North Africa Malawi LIC Faeroe Islands Kazakhstan LMC Algeria LMC Mali LIC French Polynesia Kyrgyz Republic LIC Djibouti LMC Mauritania LIC Greenland Latvia UMC Egypt, Arab Rep. LMC Mauritius UMC Guam Lithuania UMC Iran, Islamic Rep. LMC Mayotte UMC Hong Kong, China Macedonia, FYR LMC Iraq LMC Mozambique LIC Israel Moldova LIC Jordan LMC Namibia LMC Kuwait Poland UMC Lebanon UMC Niger LIC Liechtenstein Romania LMC Libya UMC Nigeria LIC Macao, China Russian Federation LMC Malta UMC Rwanda LIC Monaco Slovak Republic UMC Morocco LMC São Tomé and Principe LIC Netherlands Antilles Tajikistan LIC Oman UMC Senegal LIC New Caledonia Turkey LMC Saudi Arabia UMC Seychelles UMC Northern Mariana Islands Turkmenistan LMC Syrian Arab Republic LMC Sierra Leone LIC Qatar Ukraine LIC Tunisia LMC Somalia LIC San Marino Uzbekistan LIC West Bank and Gaza LMC South Africa LMC Singapore Yugoslavia, Fed. Rep. LMC Yemen, Rep. LIC Sudan LIC Slovenia Swaziland LMC Taiwan, China Tanzania LIC United Arab Emirates Togo LIC Virgin Islands (U.S.) Uganda LIC Zambia LIC Zimbabwe LIC This table classifies all World Bank member economies, and all other economies with populations of more than 30,000. Economies are divided among income groups according to 2001 GNI per capita, calculated using the World Bank Atlas method. The groups are: low income (LIC), $745 or less; lower middle income (LMC), $746–2,975; upper middle income (UMC), $2,976–9,205; and high income, $9,206 or more. a. Official World Bank Classification for fiscal year (ending on 30 June). Source: World Bank data. (c) The International Bank for Reconstruction and Development / The World Bank Technical Notes T hese technical notes discuss the sources and methods methods, coverage, practices, and definitions differ widely; used to compile the indicators included in this edi- and cross-country and intertemporal comparisons involve tion of Selected World Development Indicators. The complex technical and conceptual problems that cannot be notes follow the order in which the indicators appear in the unequivocally resolved. Data coverage may not be complete tables. Note that the Selected World Development Indica- for economies experiencing problems, such as those deriving tors uses terminology in line with the 1993 System of Na- from internal or external conflicts, affecting the collecting tional Accounts (SNA). For example, in the 1993 SNA gross and reporting of data. For these reasons, although the data national income replaces gross national product. See the tech- are drawn from the sources thought to be most authoritative, nical notes for tables 1 and 3 for other examples. they should be construed only as indicating trends and char- acterizing major differences among economies rather than Sources offering precise quantitative measures of those differences. The data published in the Selected World Development In- Also, national statistical agencies tend to revise their histori- dicators are taken from World Development Indicators 2002. cal data, particularly for recent years. Thus, data of different Where possible, however, revisions reported since the clos- vintages may be published in different editions of World ing date of that edition have been incorporated. In addition, Bank publications. Readers are advised not to compile such newly released estimates of population and gross national in- data from different editions. Consistent time series are avail- come (GNI) per capita for 2001 are included in table 1. able from the World Development Indicators 2002 CD-ROM. The World Bank draws on a variety of sources for the sta- Ratios and growth rates tistics published in the World Development Indicators. Data on external debt are reported directly to the World Bank by For ease of reference, the tables usually show ratios and rates developing member countries through the Debtor Report- of growth rather than the simple underlying values. Values ing System. Other data are drawn mainly from the United in their original form are available from the World Develop- Nations and its specialized agencies, from the International ment Indicators 2002 CD-ROM. Unless otherwise noted, Monetary Fund (IMF), and from country reports to the growth rates are computed using the least-squares regression World Bank. Bank staff estimates are also used to improve method (see Statistical methods below). Because this method currentness or consistency. For most countries, national ac- takes into account all available observations during a period, counts estimates are obtained from member governments the resulting growth rates reflect general trends that are not through World Bank economic missions. In some instances unduly influenced by exceptional values. To exclude the ef- these are adjusted by staff to ensure conformity with inter- fects of inflation, constant price economic indicators are national definitions and concepts. Most social data from na- used in calculating growth rates. Data in italics are for a year tional sources are drawn from regular administrative files, or period other than that specified in the column heading— special surveys, or periodic censuses. up to two years before or after for economic indicators and For more detailed notes about the data, please refer to the up to three years for social indicators, because the latter tend World Bank’s World Development Indicators 2002. to be collected less regularly and change less dramatically over short periods. Data consistency and reliability Constant price series Considerable effort has been made to standardize the data, but full comparability cannot be assured, and care must be An economy’s growth is measured by the increase in value taken in interpreting the indicators. Many factors affect data added produced by the individuals and enterprises operating availability, comparability, and reliability: statistical systems in that economy. Thus, measuring real growth requires esti- in many developing economies are still weak; statistical mates of GDP and its components valued in constant prices.  (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  The World Bank collects constant price national accounts (See the section on statistical methods below for further dis- series in national currencies and recorded in the country’s cussion of the Atlas method.) original base year. To obtain comparable series of constant GNI per capita is GNI divided by midyear population. price data, it rescales GDP and value added by industrial ori- It is converted into current U.S. dollars by the Atlas method. gin to a common reference year, currently 1995. This process The World Bank uses GNI per capita in U.S dollars to clas- gives rise to a discrepancy between the rescaled GDP and the sify economies for analytical purposes and to determine bor- sum of the rescaled components. Because allocating the dis- rowing eligibility. crepancy would give rise to distortions in the growth rate, it PPP Gross national income, which is GNI converted is left unallocated. into international dollars using purchasing power parity (PPP) conversion factors, is included because nominal ex- Summary measures change rates do not always reflect international differences The summary measures for regions and income groups, pre- in relative prices. At the PPP rate, one international dollar sented at the end of most tables, are calculated by simple ad- has the same purchasing power over domestic GNI that the dition when they are expressed in levels. Aggregate growth U.S. dollar has over U.S. GNI. PPP rates allow a standard rates and ratios are usually computed as weighted averages. comparison of real price levels between countries, just as con- The summary measures for social indicators are weighted by ventional price indexes allow comparison of real values over population or subgroups of population, except for infant time. The PPP conversion factors used here are derived from mortality, which is weighted by the number of births. See the most recent round of price surveys conducted by the In- the notes on specific indicators for more information. ternational Comparison Programme, a joint project of the For summary measures that cover many years, calcula- World Bank and the regional economic commissions of the tions are based on a uniform group of economies so that the United Nations. This round of surveys, completed in 1996 composition of the aggregate does not change over time. and covering 118 countries, is based on a 1993 reference Group measures are compiled only if the data available for a year. Estimates for countries not included in the survey are given year account for at least two-thirds of the full group, derived from statistical models using available data. as defined for the 1995 benchmark year. As long as this cri- PPP GNI per capita is PPP GNI divided by midyear terion is met, economies for which data are missing are as- population. sumed to behave like those that provide estimates. Readers Gross domestic product (GDP) per capita growth is should keep in mind that the summary measures are esti- based on GDP measured in constant prices. Growth in GDP mates of representative aggregates for each topic and that is considered a broad measure of the growth of an economy. nothing meaningful can be deduced about behavior at the GDP in constant prices can be estimated by measuring the country level by working back from group indicators. In ad- total quantity of goods and services produced in a period, dition, the estimation process may result in discrepancies be- valuing them at an agreed set of base year prices, and sub- tween subgroup and overall totals. tracting the cost of intermediate inputs, also in constant Table 1. Key indicators of development prices. See the section on statistical methods for details of the least-squares growth rate. Population is based on the de facto definition, which counts Life expectancy at birth is the number of years a new- all residents, regardless of legal status or citizenship, except born infant would live if patterns of mortality prevailing at for refugees not permanently settled in the country of asy- its birth were to stay the same throughout its life. lum, who are generally considered part of the population of Under-5 mortality rate is the probability that a child the country of origin. born in the indicated year will die before reaching age 5, if Average annual population growth rate is the exponen- the child is subject to current age specific mortality rates. The tial rate of change for the period (see the section on statisti- probability is expressed as a rate per 1,000. cal methods below). Adult illiteracy rate is the percentage of persons aged 15 Population density is midyear population divided by and above who cannot, with understanding, read and write land area. Land area is a country’s total area excluding areas a short, simple statement about their everyday life. The defi- under inland bodies of water and coastal waterways. Density nition here is based on the concept of functional literacy: a is calculated using the most recently available data on land person’s ability to use reading and writing skills effectively area. in the context of his or her society. Measuring literacy using Gross national income (GNI—formerly gross national such a definition requires census or sample survey measure- product or GNP), the broadest measure of national income, ments under controlled conditions. In practice, many coun- measures total value added from domestic and foreign tries estimate the number of illiterate adults from self- sources claimed by residents. GNI comprises gross domestic reported data or from estimates of school completion rates. product (GDP) plus net receipts of primary income from Because of these differences in method, comparisons across foreign sources. Data are converted from national currency countries—and even over time within countries—should be to current U.S. dollars using the World Bank Atlas method. made with caution. This involves using a three-year average of exchange rates to Carbon dioxide emissions (CO2) measures those emis- smooth the effects of transitory exchange rate fluctuations. sions stemming from the burning of fossil fuels and the man- (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  ufacture of cement. These include carbon dioxide produced the World Bank, are used because they take into account the during consumption of solid, liquid, and gas fuels and from local prices of goods and services not traded internationally. gas flaring. But PPP rates were designed not for making international The Carbon Dioxide Information Analysis Center (CDI- poverty comparisons but for comparing aggregates from na- AC), sponsored by the U.S. Department of Energy, calcu- tional accounts. As a result, there is no certainty that an in- lates annual anthropogenic emissions of CO2. These calcu- ternational poverty line measures the same degree of need or lations are derived from data on fossil fuel consumption, deprivation across countries. based on the World Energy Data Set maintained by the Past editions of the World Development Indicators and the UNSD, and from data on world cement manufacturing, Selected World Development Indicators used PPPs from the based on the Cement Manufacturing Data Set maintained Penn World Tables. Because the Penn World Tables up- by the U.S. Bureau of Mines. Each year the CDIAC recal- dated to 1993 are not yet available, this year’s edition (like culates the entire time series from 1950 to the present, in- last year’s) uses 1993 consumption PPP estimates produced corporating its most recent findings and the latest corrections by the World Bank. The international poverty line, set at $1 to its database. Estimates exclude fuels supplied to ships and a day in 1985 PPP terms, has been recalculated in 1993 PPP aircraft engaged in international transportation because of terms at about $1.08 a day. Any revisions in the PPP of a the difficulty of apportioning these fuels among the coun- country to incorporate better price indexes can produce dra- tries benefiting from that transport. matically different poverty lines in local currency. Problems also exist in comparing poverty measures within Table 2. Poverty and income distribution countries. For example, the cost of living is typically higher Survey year is the year in which the underlying data were in urban than in rural areas. (Food staples, for example, tend collected. to be more expensive in urban areas.) So the urban monetary Rural poverty rate is the percentage of the rural popula- poverty line should be higher than the rural poverty line. But tion living below the rural poverty line. Urban poverty rate it is not always clear that the difference between urban and is the percentage of the urban population living below the rural poverty lines found in practice properly reflects the dif- urban poverty line. National poverty rate is the percentage ference in the cost of living. In some countries the urban of the total population living below the national poverty line. poverty line in common use has a higher real value—mean- National estimates are based on population weighted sub- ing that it allows the purchase of more commodities for con- group estimates from household surveys. sumption—than does the rural poverty line. Sometimes the Population below $1 PPP a day and $2 PPP a day are difference has been so large as to imply that the incidence of the percentages of the population living on less than $1.08 a poverty is greater in urban than in rural areas, even though day and $2.15 a day at 1993 international prices (equivalent the reverse is found when adjustments are made only for dif- to $1 and $2 in 1985 prices adjusted for purchasing power ferences in the cost of living. As with international compar- parity). Poverty rates are comparable across countries, but as isons, when the real value of the poverty line varies, it is not a result of revisions in PPP exchange rates, they cannot be clear how meaningful such urban-rural comparisons are. compared with poverty rates reported in previous editions The problems of making poverty comparisons do not end for individual countries. there. More issues arise in measuring household living stan- Poverty gap at $1 PPP a day and Poverty gap at $2 dards. The choice between income and consumption as a PPP a day is the mean shortfall below the poverty line welfare indicator is one issue. Income is generally more diffi- (counting the non-poor as having zero shortfall), expressed cult to measure accurately, and consumption accords better as a percentage of the poverty line. This measure reflects the with the idea of the standard of living than does income, depth of poverty as well as its incidence. which can vary over time even if the standard of living does International comparisons of poverty data entail both not. But consumption data are not always available, and conceptual and practical problems. Different countries have when they are not there is little choice but to use income. different definitions of poverty, and consistent comparisons There are still other problems. Household survey question- between countries can be difficult. Local poverty lines tend naires can differ widely, for example, in the number of dis- to have higher purchasing power in rich countries, where tinct categories of consumer goods they identify. Survey more generous standards are used than in poor countries. Is quality varies, and even similar surveys may not be strictly it reasonable to treat two people with the same standard of comparable. living—in terms of their command over commodities—dif- Comparisons across countries at different levels of devel- ferently because one happens to live in a better-off country? opment also pose a potential problem, because of differences Can we hold the real value of the poverty line constant across in the relative importance of consumption of nonmarket countries, just as we do when making comparisons over time? goods. The local market value of all consumption in kind Poverty measures based on an international poverty line (including consumption from own production, particularly attempt to do this. The $1 a day standard, measured in 1985 important in underdeveloped rural economies) should be in- international prices and adjusted to local currency using pur- cluded in the measure of total consumption expenditure. chasing power parities (PPPs), was chosen for the World Similarly, the imputed profit from production of nonmarket Bank’s World Development Report 1990: Poverty because it is goods should be included in income. This is not always typical of the poverty lines in low-income countries. PPP ex- done, though such omissions were a far bigger problem in change rates, such as those from the Penn World Tables or surveys before the 1980s. Most survey data now include val- (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  uations for consumption or income from own production. able. For further details on the estimation method for low- Nonetheless, valuation methods vary. For example, some and middle-income economies see Ravallion and Chen surveys use the price in the nearest market, while others use (1996). the average farm gate selling price. Because the underlying household surveys differ in Whenever possible, consumption has been used as the method and in the type of data collected, the distribution in- welfare indicator for deciding who is poor. When only dicators are not strictly comparable across countries. These household income was available, average income has been problems are diminishing as survey methods improve and adjusted to accord with either a survey-based estimate of become more standardized, but achieving strict comparabil- mean consumption (when available) or an estimate based on ity is still impossible. consumption data from national accounts. This procedure Two sources of noncomparability should be noted. First, adjusts only the mean, however; nothing can be done to cor- the surveys can differ in many respects, including whether rect for the difference in Lorenz (income distribution) curves they use income or consumption expenditure as the living between consumption and income. standard indicator. The distribution of income is typically Empirical Lorenz curves are weighted by household size, more unequal than the distribution of consumption. In ad- so they are based on percentiles of population, not house- dition, the definitions of income used usually differ among holds. In all cases the measures of poverty have been calcu- surveys. Consumption is usually a better welfare indicator, lated from primary data sources (tabulations or household particularly in developing countries. Second, households dif- data) rather than existing estimates. Estimation from tabula- fer in size (number of members) and in the extent of income tions requires an interpolation method; the method chosen sharing among members. And individuals differ in age and was Lorenz curves with flexible functional forms, which have consumption needs. Differences among countries in these proved reliable in past work. respects may bias comparisons of distribution. Gini index measures the extent to which the distribution World Bank staff have made an effort to ensure that the of income (or, in some cases, consumption expenditure) data are as comparable as possible. Whenever possible, con- among individuals or households within an economy devi- sumption has been used rather than income. The income ates from a perfectly equal distribution. A Lorenz curve plots distribution and Gini indexes for high-income countries are the cumulative percentages of total income received against calculated directly from the Luxembourg Income Study the cumulative number of recipients, starting with the poor- database using an estimation method consistent with that ap- est individual or household. The Gini index measures the area plied for developing countries. between the Lorenz curve and a hypothetical line of absolute Table 3. Economic activity equality, expressed as a percentage of the maximum area under the line. Thus a Gini index of zero represents perfect Gross domestic product is gross value added, at purchasers’ equality, while an index of 100 implies perfect inequality. prices, by all resident producers in the economy plus any Percentage share of income or consumption is the taxes and minus any subsidies not included in the value of share that accrues to subgroups of population indicated by the products. It is calculated without deducting for depreci- deciles or quintiles. ation of fabricated assets or for depletion or degradation of Inequality in the distribution of income is reflected in the natural resources. Value added is the net output of an indus- percentage shares of either income or consumption accruing try after adding up all outputs and subtracting intermediate to segments of the population ranked by income or con- inputs. The industrial origin of value added is determined sumption levels. The segments ranked lowest by personal in- by the International Standard Industrial Classification come receive the smallest share of total income. The Gini (ISIC) revision 3. The World Bank conventionally uses the index provides a convenient summary measure of the degree U.S. dollar and applies the average official exchange rate re- of inequality. ported by the International Monetary Fund for the year Data on personal or household income or consumption shown. An alternative conversion factor is applied if the offi- come from nationally representative household surveys. The cial exchange rate is judged to diverge by an exceptionally data in the table refer to different years between 1985 and large margin from the rate effectively applied to transactions 2000. Footnotes to the survey year indicate whether the in foreign currencies and traded products. rankings are based on per capita income or consumption. Gross domestic product average annual growth rate is Each distribution is based on percentiles of population— calculated from constant price GDP data in local currency. rather than of households—with households ranked by in- Agricultural productivity refers to the ratio of agricul- come or expenditure per person. tural value added, measured in constant 1995 U.S. dollars, Where the original data from the household survey were to the number of workers in agriculture. available, they have been used to directly calculate the in- Value added is the net output of an industry after adding come (or consumption) shares by quintile. Otherwise, shares up all out-puts and subtracting intermediate inputs. The in- have been estimated from the best available grouped data. dustrial origin of value added is determined by the Interna- The distribution indicators have been adjusted for house- tional Standard Industrial Classification (ISIC) revision 3. hold size, providing a more consistent measure of per capita Agriculture value added corresponds to ISIC divisions income or consumption. No adjustment has been made for 1–5 and includes forestry and fishing. spatial differences in cost of living within countries, because Industry value added comprises mining, manufacturing, the data needed for such calculations are generally unavail- construction, electricity, water, and gas (ISIC divisions 10–45). (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  Services value added correspond to ISIC divisions 50–99. Merchandise imports show the c.i.f. value of goods (the Household final consumption expenditure (private cost of the goods including insurance and freight) purchased consumption in previous editions) is the market value of all from the rest of the world valued in U.S. dollars. Data on goods and services, including durable products (such as cars, merchandise trade come from the World Trade Organiza- washing machines, and home computers), purchased by tion (WTO) in its annual report. households. It excludes purchases of dwellings but includes Manufactured exports comprise the commodities in imputed rent for owner-occupied dwellings. It also includes Standard Industrial Trade Classification (SITC) sections 5 payments and fees to governments to obtain permits and li- (chemicals), 6 (basic manufactures), 7 (machinery and trans- censes. Here, household consumption expenditure includes port equipment), and 8 (miscellaneous manufactured goods), the expenditures of nonprofit institutions serving house- excluding division 68. holds, even when reported separately by the country. In prac- High technology exports are products with high R&D tice, household consumption expenditure may include any intensity. They include high-technology products such as statistical discrepancy in the use of resources relative to the in aerospace, computers, pharmaceuticals, scientific instru- supply of resources. ments, and electrical machinery. General government final consumption expenditure Current account balance is the sum of net exports of (general government consumption in previous editions) in- goods and services, net income, and net current transfers. cludes all government current expenditures for purchases of Net private capital flows consist of private debt and goods and services (including compensation of employees). nondebt flows. Private debt flows include commercial bank It also includes most expenditures on national defense and lending, bonds, and other private credits; nondebt private security, but excludes government military expenditures that flows are foreign direct investment and portfolio equity are part of government capital formation. investment. Gross capital formation (gross domestic investment in Foreign direct investment is net inflows of investment previous editions) consists of outlays on additions to the to acquire a lasting management interest (10 percent or more fixed assets of the economy plus net changes in the level of of voting stock) in an enterprise operating in an economy inventories and valuables. Fixed assets include land improve- other than that of the investor. It is the sum of equity capi- ments (fences, ditches, drains, and so on); plant, machinery, tal, re-investment of earnings, other long-term capital, and and equipment purchases; and the construction of buildings, short-term capital, as shown in the balance of payments. roads, railways, and the like, including commercial and in- Data on the current account balance, private capital flows, dustrial buildings, offices, schools, hospitals, and private and foreign direct investment are drawn from the IMF’s Bal- dwellings. Inventories are stocks of goods held by firms to ance of Payments Statistics Yearbook and International Finan- meet temporary or unexpected fluctuations in production or cial Statistics. sales, and “work in progress.� According to the 1993 SNA Official development assistance or official aid from the net acquisitions of valuables are also considered capital high-income members of the Organisation for Economic formation. Co-operation and Development (OECD) are the main External balance of goods and services is exports of source of official external finance for developing countries, goods and services less imports of goods and services. Trade but official development assistance (ODA) is also disbursed in goods and services comprise all transactions between resi- by some important donor countries that are not members of dents of a country and the rest of the world involving a OECD’s Development Assistance Committee (DAC). DAC change in ownership of general merchandise, goods sent for has three criteria for ODA: it is undertaken by the official processing and repairs, non-monetary gold, and services. sector; it promotes economic development or welfare as a The GDP implicit deflator reflects changes in prices for main objective; and it is provided on concessional terms, all final demand categories, such as government consump- with a grant element of at least 25 percent on loans. tion, capital formation, and international trade, as well as the Official development assistance comprises grants and main component, private final consumption. It is derived loans, net of repayments, that meet the DAC definition of as the ratio of current to constant price GDP. The GDP de- ODA and are made to countries and territories in part I of flator may also be calculated explicitly as a Paasche price the DAC list of aid recipients. Official aid comprises grants index in which the weights are the current period quantities and ODA-like loans, net of repayments, to countries and ter- of output. ritories in part II of the DAC list of aid recipients. Bilateral National accounts indicators for most developing countries grants are transfers in money or in kind for which no repay- are collected from national statistical organizations and cen- ment is required. Bilateral loans are loans extended by gov- tral banks by visiting and resident World Bank missions. Data ernments or official agencies that have a grant element of at for high-income economies come from the Organization for least 25 percent and for which repayment is required in con- Economic Co-operation and Development data files. vertible currencies or in kind. Total external debt is debt owed to nonresidents re- Table 4. Trade, aid, and finance payable in foreign currency, goods, or services. It is the sum Merchandise exports show the f.o.b. value of goods pro- of public, publicly guaranteed, and private non-guaranteed vided to the rest of the world valued in U.S. dollars. long-term debt, use of IMF credit, and short-term debt. (c) The International Bank for Reconstruction and Development / The World Bank �������� ����� ����������� ����������  Short-term debt includes all debt having an original matu- riod. It does not necessarily match the actual growth rate be- rity of one year or less and interest in arrears on long-term tween any two periods. debt. Present value of debt is the sum of short-term external Exponential growth rate debt plus the discounted sum of total debt service payments The growth rate between two points in time for certain de- due on public, publicly guaranteed, and private nonguaran- mographic data, notably labor force and population, is cal- teed long-term external debt over the life of existing loans. culated from the equation The main sources of external debt information are reports to the World Bank through its Debtor Reporting System r = ln (pn /p1 )/n, from member countries that have received World Bank loans. Additional information has been drawn from the files where pn and p1 are the last and first observations in the of the World Bank and the IMF. Summary tables of the ex- period, n is the number of years in the period, and ln is the ternal debt of developing countries are published annually natural logarithm operator. This growth rate is based on a in the World Bank’s Global Development Finance. model of continuous, exponential growth between two Domestic credit provided by banking sector includes points in time. It does not take into account the intermedi- all credit to various sectors on a gross basis, with the excep- ate values of the series. Note also that the exponential growth tion of credit to the central government, which is net. The rate does not correspond to the annual rate of change mea- banking sector includes monetary authorities, deposit money sured at a one-year interval which is given by banks, and other banking institutions for which data are available (including institutions that do not accept transfer- (pn – pn–1)/pn–1. able deposits but do incur such liabilities as time and savings deposits). Examples of other banking institutions include The Gini index savings and mortgage loan institutions and building and loan The Gini index measures the extent to which the distribu- associations. Data are from the IMF’s International Finance tion of income (or, in some cases, consumption expenditure) Statitics. among individuals or households within an economy devi- ates from a perfectly equal distribution. A Lorenz curve plots Statistical methods the cumulative percentages of total income received against This section describes the calculation of the least-squares the cumulative number of recipients, starting with the poor- growth rate, the exponential (endpoint) growth rate, and the est individual or household. The Gini index measures the World Bank’s Atlas methodology for calculating the conver- area between the Lorenz curve and a hypothetical line of ab- sion factor used to estimate GNI and GNI per capita in U.S. solute equality, expressed as a percentage of the maximum dollars. area under the line. Thus a Gini index of zero represents per- fect equality, and an index of 100 percent implies perfect Least-squares growth rate inequality. Least-squares growth rates are used wherever there is a suffi- ciently long time series to permit a reliable calculation. No World Bank Atlas method growth rate is calculated if more than half the observations In calculating GNI and GNI per capita in U.S. dollars for in a period are missing. certain operational purposes, the World Bank uses the Atlas The least-squares growth rate, r, is estimated by fitting a conversion factor. The purpose of the Atlas conversion fac- linear regression trendline to the logarithmic annual values tor is to reduce the impact of exchange rate fluctuations in of the variable in the relevant period. The regression equa- the cross-country comparison of national incomes. tion takes the form The Atlas conversion factor for any year is the average of a country’s exchange rate (or alternative conversion factor) ln Xt = a + bt, for that year and its exchange rates for the two preceding years, adjusted for the difference between the rate of infla- which is equivalent to the logarithmic transformation of the tion in the country, and through 2000, that in the G-5 coun- compound growth equation, tries (France, Germany, Japan, the United Kingdom, and the United States). For 2001, these countries include the Euro Xt = Xo (1 + r)t . Zone, Japan, the United Kingdom, and the United States. A country’s inflation rate is measured by the change in its In this equation, X is the variable, t is time, and a = log Xo GDP deflator. and b = ln (1 + r ) are the parameters to be estimated. If b* The inflation rate for G-5 countries (through 2000), or is the least-squares estimate of b, the average annual growth the Euro Zone, Japan, the United Kingdom, and the United rate, r, is obtained as [exp(b* )–1] and is multiplied by 100 States (for 2001), representing international inflation, is to express it as a percentage. measured by the change in the SDR deflator. (Special draw- The calculated growth rate is an average rate that is rep- ing rights, or SDRs, are the IMF’s unit of account.) The resentative of the available observations over the entire pe- SDR deflator is calculated as a weighted average of the G-5 (c) The International Bank for Reconstruction and Development / The World Bank  ����� ����������� ������  countries’ (through 2000, and the Euro Zone, Japan, the Yt$ = (Yt / Nt ) / et* United Kingdom, and the United States for 2001) GDP de- flators in SDR terms, the weights being the amount of each where et* is the Atlas conversion factor (national currency to country’s currency in one SDR unit. Weights vary over time the U.S. dollar) for year t, et is the average annual exchange because both the composition of the SDR and the relative rate (national currency to the U.S. dollar) for year t, pt is the exchange rates for each currency change. The SDR deflator GDP deflator for year t, pt S$ is the SDR deflator in U.S. dol- is calculated in SDR terms first and then converted to U.S. lar terms for year t, Yt $ is the Atlas GNI per capita in U.S. dollars using the SDR to dollar Atlas conversion factor. The dollars in year t, Yt is current GNI (local currency) for year Atlas conversion factor is then applied to a country’s GNI. t, and Nt is the midyear population for year t. The resulting GNI in U.S. dollars is divided by the midyear population to derive GNI per capita. Alternative conversion factors When official exchange rates are deemed to be unreliable The World Bank systematically assesses the appropriateness or unrepresentative of the effective exchange rate during a of official exchange rates as conversion factors. An alterna- period, an alternative estimate of the exchange rate is used tive conversion factor is used when the official exchange rate in the Atlas formula (see below). is judged to diverge by an exceptionally large margin from The following formulas describe the calculation of the the rate effectively applied to domestic transactions of for- Atlas conversion factor for year t: eign currencies and traded products. This applies to only a   p  small number of countries, as shown in Primary data docu- 1 ptS $   p ptS $  et* = et − 2  t S $  + et −1  t S $  + et  mentation table in World Development Indicators 2002. Al- 3    pt − 2 pt − 2   pt −1 pt −1    ternative conversion factors are used in the Atlas methodol- ogy and elsewhere in the Selected World Development and the calculation of GNI per capita in U.S. dollars for year t: Indicators as single-year conversion factors. (c) The International Bank for Reconstruction and Development / The World Bank