W OR LD 19279 BA N 200 K 0 (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank Published for the World Bank Oxford University Press (c) The International Bank for Reconstruction and Development / The World Bank Oxford University Press OXFORD NEW YORK ATHENS AUCKLAND BANGKOK BOGOTA BUENOS AIRES CALCUTTA CAPE TOWN CHENNAI DAR ES SALAAM DELHI FLORENCE HONG KONG ISTANBUL KARACHI KUALA LUMPUR MADRID MELBOURNE MEXICO CITY MUMBAI NAIROBI PARIS SÃO PAULO SINGAPORE TAIPEI TOKYO TORONTO WARSAW and associated companies in BERLIN IBADAN © 1999 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. Published by Oxford University Press, Inc. 200 Madison Avenue, New York, N.Y. 10016 Oxford is a registered trademark of Oxford University Press. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Cover and chapter opener design by W. Drew Fasick, ULTRAdesigns. Illustration/calligraphy by Jun Ma. The cover and chapter openers depict the Chinese character for gate or door. Inside design and typesetting by Barton Matheson Willse & Worthington, Baltimore. Manufactured in the United States of America First printing August 1999 This volume is a product of the staff of the World Bank, and the judgments made herein do not necessarily reflect the views of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. ISBN 0-19-521125-1 clothbound ISBN 0-19-521124-3 paperback ISSN 0163-5085 Text printed on recycled paper that conforms to the American Standard for Permanence of Paper for Printed Library Material Z39.48-1984. (c) The International Bank for Reconstruction and Development / The World Bank Foreword T he World Development Report 1999/2000, instability that can erode years of hard- the 22nd in this annual series, addresses the earned gains. changing development landscape of the early Given their already present implications, 21st century. Development thinking has it is not surprising that globalization and lo- evolved into a broad pragmatism, realizing calization are a central preoccupation of pol- that development must move beyond eco- icymakers around the world. Globalization nomic growth to encompass important so- is praised for bringing new opportunities for cial goals—reduced poverty, improved qual- expanded markets and the spread of tech- ity of life, enhanced opportunities for better nology and management expertise, which in education and health, and more. Experience turn hold out the promise of greater pro- has also taught that sustainable progress to- ductivity and a higher standard of living. ward these goals requires integrated imple- Conversely, globalization is feared and con- mentation and must be firmly anchored in demned because of the instability and unde- processes that are open, participatory, and sired changes it can bring: to workers who inclusive. In the absence of a strong institu- fear losing their jobs to competition from tional foundation, the outcomes of good imports; to banks and financial systems and policy initiatives tend to dissipate. These even entire economies that can be over- lessons and insights are incorporated into the whelmed and driven into recession by flows Comprehensive Development Framework, of foreign capital; and, not least, to the recently initiated by the World Bank to ad- global commons, which are threatened in dress the challenges of development in a many ways with irreversible change. more holistic, integrated way by bringing in Localization is praised for raising levels aspects such as governance, legal institu- of participation and involvement, and pro- tions, and financial institutions, which were viding people with a greater ability to shape too often given short shrift earlier. the context of their own lives. By leading to Looking ahead, this report explores the decentralized government where more de- environment in which the major issues of cisions happen at subnational levels, closer the 21st century—poverty, population to the voters, localization can result in more growth, food security, water scarcity, cli- responsive and efficient local governance. mate change, cultural preservation—will be National governments may use a strategy faced. Many powerful forces, both glacial of decentralization to defuse civil strife or and fast-paced, are reshaping the develop- even civil war. However, when poorly de- ment landscape. These include innovations signed, decentralization can result in over- in technology, the spread of information burdened local governments without the and knowledge, the aging of populations, resources or the capacity to fulfil their basic the financial interconnectedness of the responsibilities of providing local infra- world, and the rising demands for political structure and services. It can also threaten and human rights. The report focuses in macroeconomic stability, if local govern- particular on two clusters of change—glob- ments, borrowing heavily and spending alization and localization—because of their unwisely, need to be bailed out by the na- immense potential impact. They open up tional government. unprecedented opportunities for growth This report seeks neither to praise nor to and development, but they also carry with condemn globalization and localization. them the threats of economic and political Rather it recognizes them as forces that  (c) The International Bank for Reconstruction and Development / The World Bank  bring new opportunities but also raise new or greater chal- need to focus on development strategies that are imple- lenges in terms of economic and political instability. Con- mented through mutual consent, whether through interna- taining this instability and providing an environment in tional agreements between countries, or through constitu- which a development agenda can be implemented to seize tional and institutional arrangements between different levels the opportunities will be a major institutional challenge in of government and components of civil society within a the coming decades. The discussion in the report focuses on country. At both the global and local levels, institutions based three main aspects of globalization: trade in goods and ser- on partnership, negotiation, coordination and regulation vices, international flows of capital, and global environmen- would provide the basis for sustainable development. tal issues, such as the dangers of climate change and destruc- Globalization and localization are not likely to disappear, tion of biodiversity. The focus of the discussion then shifts to or even to diminish in intensity. They are driven by power- three aspects of localization: the decentralization of political ful underlying forces like the new capabilities of information power to subnational levels of government, the movement of and communication technologies, and a rising sense among population and economic energy in developing countries to- people all over the world that they are entitled to participate ward urban areas, and the provision of essential public ser- openly in their government and society. As globalization vices in these growing cities of the future. brings distant parts of the world functionally closer together, In discussing the appropriate institutional responses to and localization multiplies the range of policy environments, the challenges and opportunities of globalization and local- it may well be that successful development policies will ization, the report draws on a vast array of national examples achieve results more quickly, while failed policies will have and cross-country empirical evidence, including both devel- their consequences exposed more quickly and painfully as opment success stories and episodes of failure. There is no well. In such a world, exploring the institutional responses to simple answer to dealing with globalization and localization. globalization and localization, and disseminating the insights Instead, the insights are rooted in pragmatic judgments broadly, offers enormous potential for advances in develop- about how the existing conditions of society will affect which ment strategy—advances that can be of great and lasting policy choices make sense, or how one sequence of policies is benefit to the poorest people of the world. preferable to another, or how certain policies can comple- ment and sustain each other. The commitments and actions of the national government remain central to any workable development strategy. However, the forces of globalization James D. Wolfensohn and localization imply that much of the institution-building President for development will be taking place at either the suprana- The World Bank tional or the subnational levels. In both cases, countries will August 1999 This report has been prepared by a team led by Shahid Yusuf and comprising Anjum Altaf, William Dillinger, Simon Evenett, Marianne Fay, Vernon Henderson, Charles Kenny, and Weiping Wu. The team was assisted by Mohammad Arzaghi and Stratos Safioleas. The work was carried out under the general direction of Joseph Stiglitz. Throughout the preparation of this report Lyn Squire provided valuable advice and contributions. Timothy Taylor was the principal editor. The team was advised by a distinguished panel of experts comprising Alberto Alesina, Masahiko Aoki, Richard Cooper, John Dixon, Barry Eichengreen, Jon Elster, Alan Harold Gelb, Harry Harding, Gregory K. Ingram, Christine Kessides, Jennie Litvack, Wallace Oates, Anthony J. Pellegrini, Guillermo Perry, David Satterthwaite, Paul Smoke, Paul Spray, T.N. Srinivasan, Jacques Thisse, and John Williamson. Many others inside and outside the World Bank provided helpful comments, wrote background papers and other contributions, and participated in consultation meetings. The preparation of some background papers and the convening of several workshops was supported by the Policy and Human Resources Development Fund financed by the Japanese Government and a grant from the Government of the United Kingdom Department for International Development. These contributors and participants are listed in the Bibliographical Note. The Development Data Group contributed to the data appendix and was responsible for the Selected World Devel- opment Indicators. Rebecca Sugui served as executive assistant to the team, and Maribel Flewitt, Leila Search, and Thomas A.J. Zorab as team assistants. Maria D. Ameal served as administrative officer. Book design, editing, and production were coordinated by the Production Services Unit of the World Bank’s Office of the Publisher. (c) The International Bank for Reconstruction and Development / The World Bank Contents Overview The frontiers of development thinking 2 Globalization and localization 4 Supranational issues 5 Subnational issues 8 Translating policies into actions 11 Introduction New Directions in Development Thinking Building on past development experiences 14 The many goals of development 18 The role of institutions in development 21 The record and outlook for comprehensive development 24 A changing world 28 Chapter 1 The Changing World International trade 33 International financial flows 34 International migration 37 Global environmental challenges 40 New political tendencies in developing countries 43 Emerging subnational dynamics 44 Urban imperatives 46 Implications for development policy 49 Chapter 2 The World Trading System: The Road Ahead How the global trading system benefits developing countries 52 WTO mechanisms for promoting and maintaining liberal trade regimes 53 Sustaining the momentum for trade reform 55 International trade and development policy: the next 25 years 60 Chapter 3 Developing Countries and the Global Financial System The gathering pace of international financial integration 70 Toward a more robust and diversified banking system 75 The orderly sequencing of capital account liberalization 79 Attracting foreign investment 81 Revitalizing international macroeconomic cooperation 84 Chapter 4 Protecting the Global Commons The link between national and global environmental issues 90 Moving from national to international action 93 The ozone treaties: a success story 94  (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  Climate change 97 Biodiversity 102 Exploiting the links between global environmental problems 103 Chapter 5 Decentralization: Rethinking Government What is at stake? 107 From centralized to decentralized governance 111 Balancing political power between central and local interests 112 The structure, functions, and resources of subnational governments 114 Making subnational governments accountable 121 Policies for the transition 122 What lessons for the future? 124 Chapter 6 Dynamic Cities as Engines of Growth What makes cities grow? 126 The national government’s role in urbanization 130 Local policies for urban economic growth 132 Chapter 7 Making Cities Livable The unfinished urban agenda 140 Learning from the past 142 Service provision in developing countries 144 Looking ahead 152 Chapter 8 Case Studies and Recommendations Making the most of trade liberalization: Egypt 157 Reforming weak banking systems: Hungary 160 Macromanagement under fiscal decentralization: Brazil 163 Improving urban living conditions: Karachi 166 Cultivating rural-urban synergies: Tanzania 169 The shifting development landscape at the dawn of the 21st century 172 Bibliographical Note 175 Appendix Selected Indicators on Decentralization, Urbanization, and the Environment 213 Selected World Development Indicators 223 Index 292 Boxes 1 Lessons from East Asia and Eastern Europe 17 2 Social capital, development, and poverty 18 3 Explaining power project outcomes in Sub-Saharan Africa 18 4 The Comprehensive Development Framework 21 5 A holistic approach to development in past World Development Reports 22 6 Institutions, organizations, and incentives 23 7 Trends in disease and health care 27 8 Sustainable development 28 9 The growing threat of water scarcity 29 1.1 The global macroeconomics of aging 35 1.2 The international Chinese network 40 (c) The International Bank for Reconstruction and Development / The World Bank   2.1 Regional trading arrangements and the global trading system: complements or substitutes? 54 2.2 Building technical expertise on trade policy: the Integrated Framework for Trade and Development in the Least-Developed Countries 58 2.3 Child labor: how much? how damaging? and what can be done? 62 3.1 A continuing role for aid 73 3.2 What causes financial contagion? 75 3.3 Subnational governments face commitment problems too 83 3.4 Mitigating the commitment problem: the role of the World Bank 83 4.1 Global environmental issues 88 4.2 Preserving the ocean commons: controlling overfishing 92 4.3 The Global Environment Facility 94 4.4 NGOs and efforts to preserve the international environment 96 4.5 Falling costs for renewable energy 98 4.6 Taxes and quotas to reduce emissions 99 4.7 Trade measures in international environmental agreements 104 5.1 Decentralization as the devolution of powers 108 5.2 South Africa and Uganda: unifying a country through decentralization 108 5.3 Bosnia-Herzegovina and Ethiopia: decentralization as a response to ethnic diversity 109 5.4 India: a decentralizing federation? 110 5.5 Decentralization in China 113 5.6 Financing intermediate tiers of government 118 5.7 The cart before the horse: decentralization in Russia 123 6.1 Cities and urban areas: some definitions 127 6.2 Rural-urban linkages 128 6.3 The dispersal of industry in Korea 129 6.4 Africa: urbanization without growth 130 6.5 City development and land markets 135 6.6 Regionalism and local economic development: lessons from Europe 137 6.7 Know thy economy: the importance of local economic information 138 7.1 A spatial mismatch: Jakarta’s kampung residents 147 7.2 Haiphong: partnering with consumers 148 7.3 Manila: a positive corporate image as an incentive to reduce pollution 151 7.4 Shenyang: social welfare in a struggling industrial city 152 7.5 Bangalore: citizens’ report cards 154 8.1 Five case studies 158 8.2 The Arab Republic of Egypt at a glance 159 8.3 Hungary at a glance 161 8.4 Brazil at a glance 163 8.5 Pakistan at a glance 167 8.6 Tanzania at a glance 170 Figures 1 Computers are linking the world 4 2 All but a few democracies have decentralized some political power 4 3 Trade is growing much faster than national income in developing countries 5 4 Countries are joining the WTO in increasing numbers 6 5 Private capital flows to developing countries have increased dramatically 7 6 There are more countries and more democracies 9 7 Urban population is growing—primarily in developing countries 10 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  8 The incomes of rich and poor countries continue to diverge 14 9 Investment alone cannot account for variation in growth 15 10 Infant mortality fell in most developing countries from 1980 to 1995, even where income did not increase 19 11 The number of poor people has risen worldwide, and in some regions the proportion of poor has also increased 25 12 Life expectancies have risen greatly in some countries, but others have suffered setbacks 26 1.1 Exports of commercial services have surged in most regions since 1990 34 1.2 An increasing number of developing countries is committed to trade reform 34 1.3 Nonperforming loans can account for up to 50 percent of all bank loans at the peak of a banking crisis 36 1.4 Resolving bank crises can cost up to 40 percent of GDP 37 1.5 Foreign direct investment was less volatile than commercial bank loans and total portfolio flows, 1992–97 37 1.6 Temperatures are rising as concentrations of greenhouse gases increase 41 1.7 More countries are becoming democratic 43 1.8 Most urban dwellers reside in developing countries 47 1.9 Asia and Africa are just beginning the urban transition 47 1.10 The largest increase in urban populations during 1980-2020 will occur in Africa and Asia 48 2.1 Foreign trade has increased in most developing countries since 1970 52 2.2 More of the world’s exports are covered by WTO disciplines, especially exports from developing countries 53 2.3 More regional trading arrangements (RTAs) came into force in the 1990s than ever before 54 2.4 Many developing countries started liberalizing before the end of the Uruguay Round 56 2.5 Equal players? African representatives at the WTO 57 2.6 The composition of many developing countries’ exports was transformed in just over 10 years 59 2.7 New users initiated an increasing number of antidumping suits during 1987–97 60 2.8 When filing antidumping investigations, industrial and developing countries target each other almost equally 61 2.9 Many countries bound their tariffs on agricultural products in the Uruguay Round at levels well above estimated actual tariffs in 1986–88 63 2.10 Exports of commercial services increased in every region from 1985 to 1997 64 3.1 Since 1980 net flows of foreign direct and portfolio investment to developing economies have grown enormously 70 3.2 Firms from developing countries are issuing more international debt than before 71 3.3 A growing pool of institutionally managed funds is invested abroad 71 3.4 A few developing countries received the lion’s share of FDI invested outside industrial countries in 1997 73 3.5 Bank intermediation typically accounts for a larger share of the financial sector in developing countries 76 4.1 Climate change jeopardizes crop yields, especially in developing countries 89 4.2 Atmospheric concentrations of ozone-depleting substances rose, then began to fall 95 4.3 A 1-meter rise in sea level would cut Bangladesh’s rice production approximately in half 100 4.4 Energy consumption in developing countries is forecast to outstrip industrial country consumption 100 4.5 High-income countries use energy more intensely than countries in low-income regions 101 5.1 Subnational expenditures are a small share of public expenditures, except in industrial countries and large federations 111 (c) The International Bank for Reconstruction and Development / The World Bank   5.2 Local governments never control a large share of public resources 112 6.1 Urbanization is closely associated with economic growth 126 6.2 Most of the world’s urban population lived in small and medium-size cities in 1995 128 6.3 Small cities had the fastest growing populations, and megacities the slowest, from 1970 to 1990 130 6.4 As countries develop, central governments’ share of public investment falls 133 7.1 Even low-income countries can achieve high levels of basic water and sanitation services 140 7.2 Housing affordability varies significantly at low levels of income 141 8.1 The population in Tanzania is increasingly urbanized 141 Tables 1.1 World foreign direct investment stock, 1997 38 1.2 Political and functional decentralization in large democracies, 1997 45 2.1 Reported antidumping actions by members of the GATT and WTO, 1987-97 60 2.2 Share of parts and components in exports, 1995 66 5.1 The structure of subnational governments in large democracies 116 5.2 Subnational borrowing controls in selected countries 119 7.1 Infant mortality rate, Bangladesh, 1990 142 Definitions and data notes Dollar figures are current U.S. dollars, unless oth- The countries included in regional and income erwise specified. Billion means 1,000 million; trillion groupings in this report are listed in the Classifica- means 1,000 billion. tion of Economies table at the end of the Selected World Development Indicators. Income classifica- The following abbreviations are used: tions are based on GNP per capita; thresholds for in- come classifications in this edition may be found in AIDS Acquired immune deficiency syndrome the Introduction to Selected World Development CDF Comprehensive Development Indicators. Group averages reported in the figures Framework and tables are unweighted averages of the countries FDI Foreign direct investment in the group unless noted to the contrary. GATT General Agreement on Tariffs The use of the word countries to refer to econ- and Trade omies implies no judgment by the World Bank GDP Gross domestic product about the legal of other status of a territory. The term GNP Gross national product developing countries includes low- and middle- NIE Newly industrialzing economy income economies and thus may include economies NGO Nongovernmental organization in transition from central planning, as a matter of OECD Organisation for Economic convenience. The term advanced countries may be Co-operation and Development used as a matter of convenience to denote the high- PPP Purchasing power parity income economies. WTO World Trade Organization (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank Overview T he development landscape is being to reducing poverty, these challenges transformed, presenting policymakers include issues of food security, water with new challenges at the global and scarcity, aging populations, cultural loss, local levels. This report charts the way and environmental degradation. forward by analyzing the contours of These challenges must be confronted the new landscape and distilling lessons even as many forces reshape the devel- from the past. It examines the unfold- opment terrain: innovations in technol- ing dynamic at the supranational and ogy, the spread of knowledge, the growth subnational levels. And it proposes new of population and its concentration in rules and structures to serve as a foun- cities, the financial integration of the dation for development policy in the world, and rising demands for political 21st century. and human rights. Some of these forces, Fifty years of development experi- like population growth, will work their ence have yielded four critical lessons. way gradually, giving policymakers time First, macroeconomic stability is an es- to respond. Others, such as financial con- sential prerequisite for achieving the tagion, could batter apparently healthy growth needed for development. Sec- economies without warning unless pre- ond, growth does not trickle down; de- emptive measures are in place. Some will velopment must address human needs give rise to challenges, like social wel- directly. Third, no one policy will trig- fare funding, that most nation-states can ger development; a comprehensive ap- cope with on their own. Others, such as proach is needed. Fourth, institutions global climate change, will be beyond matter; sustained development should the reach of any one state and will call be rooted in processes that are socially for international agreements. inclusive and responsive to changing If they are managed well, these forces circumstances. could revolutionize the prospects for These insights are central to how the development and human welfare. How- World Bank envisions its work in the ever, the same forces are also capable of 21st century and to the way in which it generating instability and human suf- proposes to tackle the principal devel- fering that are beyond the ability of in- opment challenges ahead. In addition dividual nation-states to remedy.       ⁄  This report views the changes that have been set in countries in the world, has had remarkable economic motion as contributing to—and as manifestations of— success with a development strategy that involves only a two phenomena: globalization and localization. Global- limited dose of market liberalization and privatization. ization, which reflects the progressive integration of the The failure of centrally planned economies to keep world’s economies, requires national governments to pace with their market-oriented counterparts has dem- reach out to international partners as the best way to onstrated clearly enough that planning entire economies manage changes affecting trade, financial flows, and the at the central government level is not a productive path global environment. Localization, which reflects the to long-term development. But the experiences of growing desire of people for a greater say in their govern- Japan, East Asia, and China make clear that it is possi- ment, manifests itself in the assertion of regional identi- ble for a country to have an interventionist government ties. It pushes national governments to reach down to re- and still enjoy extremely rapid economic growth over a gions and cities as the best way to manage changes period of decades. affecting domestic politics and patterns of growth. At Brazil also grew very rapidly in the 1960s, in part by both the supranational and subnational levels, institu- making widespread use of import-substitution policies. tions of governance, negotiation, coordination, and regu- These policies certainly appeared helpful to Brazil at the lation will play a critical role in promoting a new equilib- time—at a minimum, they did not prevent a surge of rium between and within countries—and in abetting the rapid growth—but this success does not mean that sim- creation of the stable environment that will make possi- ilar policies would make sense in other countries, or even ble the implementation of development programs. in Brazil three decades later. Similarly, certain policies that helped Japan develop in the 1950s and 1960s, gen- The frontiers of development thinking erated growth in East Asia in the 1970s and 1980s, and As the 20th century draws to a close, mainstream devel- sparked China’s economic boom in the 1980s and 1990s opment thinking has evolved toward a broad pragma- were specific to the time and place. They may not have tism. As with many subjects, a deeper understanding of worked well in other countries, nor are they likely to be development involves a recognition that sweeping be- appropriate in the opening decades of the 21st century. liefs are often incomplete, that layers of complexity are In any given country, progress depends on a constel- buried not far beneath the surface, and that wisdom is lation of factors, and on shifts in their configuration that often contingent on the particular conditions of time take place over time. What is required is to step beyond and place. In recent decades both experience and intel- the debates over the roles of governments and markets, lectual insight have pushed development thinking away recognizing that they need to complement each other, from debates over the role of states and markets, and the and to put to rest claims that any particular policy in- search for a single, overarching policy prescription. tervention—in education, health, capital markets, or Investment in physical and human capital, for exam- elsewhere—is the magic formula that will inspire devel- ple, should encourage economic growth, and as a gen- opment in all times and places. This shift in develop- eral rule, empirical evidence supports this proposition. ment thinking can be summarized in four propositions: But in a number of cases, high rates of investment and Sustainable development has many objectives. Raising education have not been enough to deliver rapid growth. per capita incomes is only one among many develop- A similar lesson holds true for industrial policies. Many ment objectives. Improving quality of life involves more countries decided, after experimenting with export sub- specific goals: better health services and educational op- sidies, that the subsidies enriched business owners but portunities, greater participation in public life, a clean did little to speed economic growth. They saw well- environment, intergenerational equity, and more. intended industrial subsidies turn into a costly form of Development policies are interdependent. No single corporate welfare, an expensive way of providing tax- development policy can make much of a difference in payer support for private jobs in a narrow range of in- an unfavorable policy regime. Countries need inte- dustries. Yet East Asian economies, making active use of grated policy packages and institutional environments export subsidies and credit allocation, experienced the that reward good outcomes, minimize perverse incen- most powerful sustained surge of economic development tives, encourage initiative, and facilitate participation. the world had seen in decades. And China, which alone Governments play a vital role in development, but there includes 40 percent of all the inhabitants of low-income is no simple set of rules that tells them what to do. Beyond   generally accepted rules, the role of government in the These items do not constitute an exhaustive list of economy varies, depending on capacity, capabilities, all the concerns development should embrace. Issues of the country’s level of development, external conditions, gender and equity are integral to every part of the and a host of other factors. framework. Moreover, as mentioned earlier, macroeco- Processes are just as important as policies. Outcomes nomic stability is a necessary condition for the success of policies based on consensual, participatory, and of development initiatives. How important each of transparent processes are more easily sustained. Institu- these concerns is to individual countries depends on the tions of good governance that embody such processes particulars of time and place. Every country will bene- are critical for development and should encompass fit from identifying and prioritizing its needs—an ex- partnerships among all elements of civil society. ercise that reveals the economic or governmental weak- nesses and institutional failures that stand in the way Creating new guidelines for development of full development. In light of these propositions, the World Bank is intro- ducing a comprehensive development framework to Building institutions and partnerships serve a number of purposes: to sharpen the focus on the Effective development requires partnerships among dif- major goals of development, to highlight the integrated ferent levels of government, the private sector, donor nature of policymaking, to emphasize the institutional groups, and civil society. A comprehensive strategy is processes required to sustain development, and to co- simply too demanding for any one level or area of gov- ordinate development efforts. ernment or for a single donor. National governments The framework underscores the growing realization need to provide the guidance that agencies and organi- that the many elements that make up the development zations require to coordinate their efforts to remove process must be planned together and coordinated in bottlenecks to development. order to obtain the best results—and sometimes in order A solid foundation of effective organizations and en- to arrive at any results at all. A school-building project abling institutions is a necessary precondition to devel- is a good example. Physically putting up the building is opment. In this context “institutions” are sets of rules only a start. Raising educational levels will depend on governing the actions of individuals and organizations, many other things, such as effective mechanisms for se- and the interaction of all relevant parties and the nego- lecting, training, and remunerating teachers adequately tiations among the participants. Specifically, countries and sufficient resources to buy enough textbooks and need institutions that strengthen organizations and pro- supplies. mote good governance, whether through laws and regu- What is true of a school-building project is also true lations or by coordinating the actions of many players, of privatization programs, social safety nets, and sus- as international treaties and public-private partnerships tainable energy programs. The complementarities be- do. Rule-based processes increase the transparency of tween projects and processes are vital to success. A com- policies designed to create desired outcomes and of or- prehensive framework makes these complementarities ganizations used to implement them. explicit by emphasizing the relationships among the The message of this report is that new institutional human, physical, sectoral, and structural aspects of responses are needed in a globalizing and localizing development. world. Globalization requires national governments to The human and physical aspects of development are seek agreements with partners—other national govern- well known. Sectoral aspects stress the importance of ments, international organizations, nongovernmental cross-cutting elements such as coordination, manage- organizations (NGOs), and multinational corpora- ment, and maintenance of an effective enabling envi- tions—through supranational institutions. Localization ronment for private business and community initia- requires national governments to reach agreements tives. Structural aspects focus on the need for good with regions and cities through subnational institutions governance, transparent decisionmaking, efficient legal on issues such as sharing responsibility for raising rev- and judicial processes, and sound regulatory systems. enues. Both globalization and localization often require This identification of rules and processes as a critical responses that are beyond the control of a single na- foundation for sustained development adds a new di- tional government. Yet national governments will re- mension to mainstream development thinking. main pivotal in shaping development policies in an en-      ⁄  vironment that circumscribes, constrains, and redefines Figure 2 their role. In an interconnected world in which coun- All but a few democracies have decentralized some political power tries may continue to fragment, development agendas must respond to both global and local imperatives. Number of countries Globalization and localization 75 Technological advances in communication have made it possible to know in an instant what is happening in a 50 household or factory or on a stock market half a world away. The growing importance of services and informa- tion in the world economy means that an increasing 25 proportion of economic value is weightless—that is, it can be transmitted over fiber-optic cable rather than 0 transported in a container ship. At the same time im- Without With provements in transportation networks and technology subnational elections subnational elections are reducing the costs of shipping goods by water, Note: Figure shows all countries classified as multiparty democracies ground, and air, and improvements in information for which data were available. Source: Freedom House, Freedom in the World, 1998; appendix technology have made it easier to manage the new in- table A.1. terconnections (figure 1). Multinational companies now rely on production chains that straddle many countries. Raw materials and components may come from two different countries and be assembled in an- Rising educational levels, technological innovations other, while marketing and distribution take place in that allow ideas to circulate, and the economic failure still other venues. Consumers’ decisions in, say, London of most centrally planned economies have all con- or Tokyo become information that has an almost im- tributed to the push for localization. National govern- mediate impact on the products that are being made— ments have responded to this push in various ways. and the styles that influence them—all over the globe. More countries have become democracies, and politi- cal participation through elections has expanded at Figure 1 both the national and subnational levels. National gov- Computers are linking the world ernments are increasingly sharing responsibilities and revenues with subnational levels of government that are Internet hosts worldwide (millions) closer to the people affected by policy decisions (figure 50 2). People are also forming NGOs to pursue objectives such as political reform, environmental protection, 45 gender equality, and better education. 40 Globalization and localization are terms that provoke 35 strong reactions, positive and negative. Globalization is 30 praised for the new opportunities it brings, such as ac- 25 cess to markets and technology transfer—opportunities 20 that hold out the promise of increased productivity and 15 higher living standards. But globalization is also feared 10 and often condemned because it sometimes brings in- 5 stability and unwelcome change. It exposes workers to competition from imports, which can threaten their 0 Jan. July Jan. July Jan. July Jan. July Jan. July Jan. July Jan. jobs; it undermines banks and even entire economies 1993 1994 1995 1996 1997 1998 1999 when flows of foreign capital overwhelm them. Source: Network Wizards, Internet Domain Survey, January 1999 Localization is praised for raising levels of participa- (www.nw.com.). tion in decisionmaking and for giving people more of a chance to shape the context of their own lives. By de-   centralizing government so that more decisions are to the best practices of foreign firms and to the demands made at subnational levels, closer to the voters, local- of discerning customers, encouraging greater efficiency. ization nourishes responsive and efficient governance. Trade gives firms access to improved capital inputs such But it can also jeopardize macroeconomic stability. as machine tools, boosting productivity as well. Trade Local governments that have borrowed heavily and encourages the redistribution of labor and capital to spent unwisely, for example, may have to be bailed out relatively more productive sectors. In particular, it has by the national government. contributed to the ongoing shift of some manufactur- This report does not praise or condemn globaliza- ing and service activities from industrial to developing tion and localization. Rather, it sees them as phenom- countries, providing new opportunities for growth. ena that no development agenda can afford to ignore. The creation of the World Trade Organization While national governments remain central to the de- (WTO) in 1995 built on the General Agreement on velopment effort, globalization and localization require Tariffs and Trade (GATT) and is the latest multilateral that they engage in essential institution-building at step toward creating an environment conducive to the both the supra- and subnational levels in order to cap- exchange of goods and services (figure 4). A number of ture the benefits of growth in the 21st century. other important measures must follow, so that the mo- mentum for reform is not lost. Supranational issues Future trade talks will require a forward-looking National governments will inevitably face frustrations agenda for broader trade liberalization if they are to repeat in dealing with globalization, and these frustrations will their past successes in opening markets. The Millennium be magnified for small developing economies. But such Round, which is scheduled to start in November 1999 countries stand to gain more from international trade under WTO auspices, may be the first test of such and finance than their larger counterparts, since they an agenda. Reducing trade barriers in agriculture and face tighter resource and market-size constraints. At the services should be high on the list of priorities. Trade in same time these economies may feel any disruption the agricultural products is one area that offers many devel- global economy generates far more intensely. An eco- oping economies real opportunities—if these opportu- nomic shock that may feel like only a ripple to an enor- nities are not blocked by trade barriers in wealthy coun- mous economy like the United States, or even to a rela- tries. Trade in services is another issue that must be tively large developing economy like Brazil, is a tidal wave for an economy the size of Ghana or Bangladesh. Figure 3 When it comes to environmental issues, national gov- Trade is growing much faster than national ernments can strike their own balance on domestic income in developing countries problems by, for example, determining how to apply Trade pollution standards to regions that lie entirely within (percentage of GDP) the country. But unless developing countries work 50 through international agreements, they have little abil- ity to address global environmental problems like the 45 threat of climate change. This report considers three di- Industrial countries 40 mensions of globalization: trade, financial flows, and environmental challenges. 35 Developing countries Trade 30 Foreign trade has grown more quickly than the world economy in recent years, a trend that is likely to con- 25 tinue (figure 3). For developing countries, trade is the 20 primary vehicle for realizing the benefits of globaliza- 1981 1983 1985 1987 1989 1991 1993 1995 1997 tion. Imports bring additional competition and variety Note: Trade is the sum of exports and imports of goods and to domestic markets, benefiting consumers, and exports services. enlarge foreign markets, benefiting businesses. But per- Source: World Bank, World Development Indicators, 1999. haps even more important, trade exposes domestic firms      ⁄  Figure 4 ucts are not sold below what is considered a “fair” price Countries are joining the WTO in increasing on domestic markets. But such rules can easily be turned numbers into barriers to imports, diluting market access and re- Number of member countries in GATT/WTO versing the gains from previous trade agreements. One 140 solution is to treat the pricing decisions of importers 134 and domestic firms according to the same criteria. Under 120 this approach only antitrust issues such as predation are 100 100 remedied directly. 85 80 Financial flows Financial flows across national borders have risen far 60 more quickly than trade in recent years. These capital 40 flows can be divided into foreign direct investment, for- eign portfolio investment, bank loans, and official de- 20 velopment flows. Foreign direct investment is made up 0 of flows intended to purchase a stake in the manage- 1980 1990 1999 ment of a company or factory. Foreign portfolio invest- Source: WTO, Annual Report, various years. ment includes purchases of “paper” assets like equities and bonds (below the threshold required to give own- ers managerial control of physical assets). The increase in foreign direct investment and portfolio flows is par- addressed. Driven by advances in information and com- ticularly striking (figure 5). munications technology, it is growing explosively— Flows of foreign capital offer substantial economic 25 percent between 1994 and 1997 alone. This type gains to all parties. Foreign investors diversify their risks of trade offers another opportunity for developing outside their home market and gain access to profitable countries, which can readily supply many sought-after opportunities throughout the world. Economies receiv- services. ing inflows of capital benefit in many ways. Initially, Countries must make greater use of WTO mechanisms. inflows raise the level of investment. When foreign di- For example, a country that wants to strengthen its com- rect investment is involved, management expertise, mitment to reducing (and maintaining) low trade barri- training programs, and important linkages to suppliers ers can “bind” its tariffs by incorporating the decision to and international markets often accompany the capi- lower them into its international obligations at the tal. Yet international capital flows, especially flows of WTO. The more countries view the WTO and interna- volatile short-term investments, also expose developing tional trade rules as mechanisms for advancing national countries to certain dangers. Among these are sharp goals (rather than as obstacles to self-determination), the changes in investor sentiment and waves of speculation greater will be the support for such institutions. that can upset exchange rate regimes, imperil banks and Public policies must take into account the plight of large firms, and wreak havoc on economies. Putting the workers displaced by the forces of trade. These policies genie of foreign capital back in the bottle is not possi- must address the concerns of displaced workers in gen- ble—and ultimately not desirable. But such capital eral, since many workers will blame foreign trade for comes with a challenge: to devise policies and institu- job losses and wage cuts whether it is responsible or tions that tip the balance so that capital mobility bene- not. Augmenting trade liberalization with labor market fits developing economies rather than injuring them. policies that ease workers’ adjustment to the effects of Governments of developing countries can begin this global trade will reduce pressure to close domestic mar- process by reforming their banking sectors and nurtur- kets to foreign goods. ing capital markets. The paucity of mutual and pension Governments must change policies that are still allowed funds and the weakness of stock and bond markets in under existing trade rules but that hinder rather than pro- developing countries make banks the primary providers mote trade. For example, antidumping laws are allowed of financial intermediation. Creating a robust banking under the WTO. They are intended to ensure that prod- regulatory framework offers a substantial economic   Figure 5 Countries can hedge these risks to some extent, but reg- Private capital flows to developing countries ulations are needed to moderate the demand for short- have increased dramatically term foreign borrowing in the first place. One such Net capital flows measure could require that a part of all capital inflows (billions of U.S. dollars) not intended to purchase productive physical assets be 160 set aside for a specified period, thereby raising the cost 140 of short-term borrowing from abroad. 1990 1998 In a world where financial markets continue to “go 120 global,” developing countries need to work toward be- 100 coming good homes for long-term foreign investment. 80 Building an investment-friendly environment requires a commitment to a transparent regime of investors’ 60 rights and regulations, a legal system that offers equal 40 treatment and protection to foreign and domestic in- vestors, sound macroeconomic fundamentals, and in- 20 vestment in human capital. When investment is inte- 0 grated into a well-functioning local economy, other Official Private debt Portfolio Foreign flows flows equity direct investors will always be ready to step in should one in- flows investment vestor decide to withdraw. Note: Private debt flows include bank loans and bonds. The Republic International institutions have a role to play in help- of Korea is included in the figures for developing countries. ing developing countries promote financial stability and Source: World Bank, Global Development Finance, 1999. investment. International banking agreements such as the Basle Accords can serve as models for local bank ac- counting standards. The International Monetary Fund payoff. An effective regulatory regime creates an envi- (IMF) can monitor economic performance and coordi- ronment that encourages prudent risk-taking. A regu- nate short-term relief for liquidity problems, dampen- latory structure for banking also sets out the conditions ing the severity of a financial crisis. Trade agreements for establishing banks, the services they can provide, the can help keep responses to financial shocks from turn- level of capital they must hold, and the amount of in- ing into a beggar-thy-neighbor cycle of protectionism. formation they must disclose. And a sound regulatory Regional and international talks on coordinating macro- framework specifies the prudential steps regulators are economic policies can seek ways to avoid actions that required to take if these standards are not met. favor one economy at the expense of its neighbors. Increased competition in the financial sector im- proves incentives for both banks and their customers. Global environmental challenges Competition increases as the domestic financial sector Just as a country’s economy can be swamped by global develops and securities, stock markets, and other inter- economic forces it has little power to control or deflect, mediaries begin playing larger roles. Allowing foreign its environment can be threatened by activities taking banks to enter a country, especially when their home place beyond its borders and its control. In some low- countries have sound regulatory systems in place, boosts income countries the threats may be severe enough to regulation by importing high-quality risk-management jeopardize further sustainable development. Climate standards, regulatory practices, and trained managers. changes, for example, could raise ocean levels, swamp- Banks in developing countries must balance two ing the homes of millions of people in low-lying coun- risks. Banks often raise short-term money on global tries like Bangladesh. Governments acting alone, and markets in one currency, such as U.S. dollars or Japan- even regional organizations, cannot respond effectively ese yen, and then loan that money out for longer peri- to this kind of environmental problem. The response ods in domestic currency. These banks run the risk of must be global. Industrial countries are responsible for losing their supply of short-term foreign money if the most of the existing global environmental problems— market dries up, as well as the risk of losing much of especially man-made greenhouse gases—but develop- the value of their assets if the exchange rate depreciates. ing countries are catching up rapidly. Their capacity to      ⁄  contribute to future environmental damage increases as on Climate Change created at the 1992 Rio Earth Sum- they grow. mit form a basis for moving forward. The Global En- The world has already seen one genuine environ- vironment Facility (GEF) is a joint initiative of the mental success story in the Montreal Protocol of 1987, United Nations Development Programme, the United which brought all countries together to address a com- Nations Environment Programme, and the World Bank. mon environmental threat. The Montreal Protocol at- The GEF provides grants and concessional funds to tempts to solve the problem of chlorofluorocarbon cover additional costs countries incur when a develop- emissions, which reduce ozone concentrations in the ment project also targets one or more of four global en- upper atmosphere. In the 1980s scientists realized that vironmental issues: climate change, biodiversity loss, allowing these emissions to continue unchecked would the pollution of international waters, and depletion of dangerously increase ultraviolet radiation in the higher the ozone layer. National governments can take a num- latitudes, raising rates of skin cancer and cataracts and ber of actions that improve domestic welfare while damaging the environment. Thanks to the Montreal helping preserve the global commons. Removing fuel Protocol and follow-on agreements, global interna- subsidies and improving public transportation, for ex- tional production of chlorofluorocarbons has fallen ample, not only are in the best interest of individual steeply, and global cooperation to reduce ozone deple- economies but also contribute to reducing global car- tion appears to be succeeding. bon dioxide emissions that affect other countries. The world faces a number of other pressing environ- Subnational issues mental problems that threaten the global commons. Per- haps the best known is climate change, which is associ- Even as globalization directs the attention of national ated with increasing emissions of carbon dioxide into the governments to events, forces, and ideas outside their atmosphere. Others include biodiversity loss, which is borders, localization highlights the opinions and aspira- occurring at an alarming rate; desertification; the deple- tions of groups and communities at home. Two aspects tion of fish stocks; the spread of persistent organic pol- of localization receive particular attention in this report: lutants; and threats to the ecology of Antarctica. decentralization and urbanization. The ozone success story provides a model for future international agreements on global environmental is- Political pluralism and decentralization sues. The scientific case for addressing the risk of en- Localization has generated political pluralism and self- vironmental damage needs to be made forcefully in determination around the world. One of its manifes- open and robust public debate. The world’s peoples and tations is the increase in the number of the world’s their governments must share the belief that the costs countries, which has climbed as regions win their inde- of environmental damage are heavy enough to justify pendence. Another is the change in countries’ choice of immediate action. Alternatives to current behavior governments. As recently as 25 years ago, less than one- must be technically feasible and reasonably inexpensive, third of the world’s countries were democracies. In the and all countries must be willing to participate in inter- late 1990s that proportion has risen to more than 60 national accords. Sometimes this willingness will come percent (figure 6). at a price, with high-income countries paying low- The ability of people to participate in making the de- income economies to comply with an agreement and cisions that affect them is a key ingredient in the process groups of signatories imposing penalties on countries of improving living standards—and thus in effective de- that fail to meet the standards the agreement sets. Fi- velopment. But political responses to localization, such nally, the standards themselves must be flexible, because as decentralization, can be successful or unsuccessful, very rarely is there a “one size fits all” solution to global depending on how they are implemented. The follow- problems. ing are several important lessons for governments to The conditions surrounding biodiversity and cli- consider when embarking upon decentralization. mate change suggest that reaching international agree- Decentralization is almost always politically motivated. ment on these issues will be more complex than it was Often its primary objective is to maintain political sta- with ozone depletion. But the international community bility and reduce the risk of violent conflict by bringing has already begun seeking solutions. The Convention a wide range of groups together in a formal, rule-bound on Biological Diversity and the Framework Convention bargaining process. Arguing about whether decentral-   ization should happen is largely irrelevant; the way it is hastily, and granting local elections is a relatively fast implemented will determine how successful it is. and easy step to take. But devising new regulatory rela- Devising a successful decentralization strategy is com- tionships between central and subnational governments plex because decisionmakers do not always fully control the is a slow and difficult task, as is the transfer of assets decentralization process. Decentralization requires chang- and staff from the central to the local level. Equally dif- ing the system of governance and establishing new po- ficult is the conversion of a system based on annual litical, fiscal, regulatory, and administrative institutions. budgetary transfers between units of a centralized ad- It involves not simply the decision to permit local elec- ministration to one based on the assignment of taxes tions but also a series of choices about electoral rules and expenditures at different levels of government. and party practices that will affect the options available National governments need to demonstrate their com- to voters. It involves more than a decision to devolve a mitment to the new rules of the intergovernmental rela- certain type of responsibility—for education, say—to tionship at the very outset. Precedents matter, because the local level. It requires deciding which level of gov- they affect expectations. One of the most important ernment will be responsible for financing education precedents a central government can set for newly de- (particularly in poor regions), which level will establish mocratized subnational governments is to keep the cen- curricula and develop instruction materials, and which tral budget constraint hard. Local governments must level will be responsible for the day-to-day management know that if they overspend, the national government of the schools, including hiring, promoting, and dis- will not bail them out and that local taxpayers and missing teachers. So that decentralization does not politicians will bear the burden of adjustment. occur at the expense of equity, it requires granting rev- enue sources to subnational authorities and designing a Urbanization system of intergovernmental fiscal transfers to comple- More and more of the world’s population is moving ment local resources. It demands rules governing sub- from rural to urban areas. Twenty-five years ago less national borrowing. And finally, it must include steps than 40 percent of the world’s population lived in to build the capability of subnational governments to urban areas; 25 years in the future this share could carry out their new responsibilities. reach nearly 60 percent. Of the urban dwellers of the The elements of reform must be synchronized. The po- future, nearly 90 percent will be living in developing litical impetus that is often behind decentralization countries. Half a century ago just 41 of the world’s 100 prompts central governments to make concessions largest cities were in developing countries. By 1995 that Figure 6 There are more countries . . . . . . and more democracies Democratic countries Number of countries in the world (percentage of total countries) 200 100 180 90 160 80 140 70 120 60 100 50 80 40 60 30 40 20 20 10 0 0 1974 1990 1998 1974 1990 1998 Source: Freedom House, Freedom in the World, 1998; Larry Diamond, “Is the Third Wave Over?” Journal of Democracy, 1996.      ⁄  number had risen to 64, and the proportion keeps ris- equate legal framework and a firm central government ing (figure 7). commitment against bailouts. Land use planning is an Some national governments tax rural areas or place important and useful tool, but the rules need to be spe- restrictions on the prices of rural products as a way of cific to local circumstances. supporting cities, on the grounds that such policies Countries do not need to wait until they become encourage a “modern” economy. Other governments, wealthy to improve urban services. Innovative institu- concerned about the growing population of urban poor, tional arrangements can result in much better service have tried to discourage rural-urban migration, some- provision, even at low levels of income. Recent trends times by requiring official permission to move to a city. in providing essential services point to the potential of Neither course of action has worked especially well. Pre- public-private partnerships. venting individuals from moving in response to incen- tives generally fails, as national governments have not Ⅲ Housing. Private developers, voluntary agencies, com- proven adept at deciding where households and firms munity organizations, and NGOs need to provide an should locate. Governments will be better off if they increased share. For its part, the public sector must pursue development policies that benefit both urban and focus on property rights, financing and subsidies, rural areas, recognize that the process of development building regulations, and trunk infrastructure. will spur urbanization over time, and plan accordingly. Ⅲ Water. Large cities are moving to private sector pro- Local governments can take steps to make their vision. Private concessions have already replaced pub- cities more hospitable venues for economic develop- lic providers in Buenos Aires, Jakarta, and Manila. ment. One important step is to maintain a sufficient The role of government is to regulate this industry level of investment in essential infrastructure, includ- and foster competition. ing water, sanitation, roads, telephones, electricity, and Ⅲ Sewerage. Governments are often unable to fund the housing. Increasingly, local governments are working heavy initial investment required for citywide solu- with the private sector, which has an important role to tions. But communities are managing, with the as- play in housing, on-site infrastructure, and municipal sistance of NGOs, to implement affordable solu- utilities. But municipalities will still be required to raise tions, providing a model for future efforts. substantial sums to finance capital investment, particu- Ⅲ Transportation. Public education and creative part- larly during the rural-urban transition. Private capital nerships can reduce air pollution. But the greatest markets are a promising source, but they require an ad- payoff is likely to come from channeling urban Figure 7 Urban population is growing—primarily in developing countries World urban population Location of the world’s 100 largest cities (percentage of total population) (number of cities) 60 80 59 50 47 60 In developing countries 40 38 30 40 In industrial countries 20 20 10 0 0 1975 2000 2025 1950 1995 2015 Source: United Nations Department of International Economics and Social Affairs, World Urbanization Prospects, 1998.   growth along public transit routes to create more ef- processes of globalization and localization will evolve. ficient transportation corridors. The report presents five case studies that illustrate how governments and organizations can capture some of the It is sometimes argued that poverty alleviation is benefits of these two phenomena and respond to poten- purely the national government’s responsibility. While tial disruptions. The studies cover trade liberalization the national government should play a prominent role in the Arab Republic of Egypt; the reform of Hungary’s in providing subsidies to the poor, many services that banking sector; Brazil’s efforts to structure the fiscal re- affect the poor most—water, health, education, and lationships between regional and national government; transportation—are best managed at the local level in efforts of community groups and local developers to ways that respond effectively to local needs. improve living standards in Karachi, Pakistan; and the creation of an urban-rural synergy in Tanzania. Translating policies into action The challenges for development are many: poverty, Globalization and localization offer exceptional oppor- hunger, ill health, lack of housing, and illiteracy, to name tunities, but can also have destabilizing effects. This re- a few. Much progress has been made, so that people in port identifies some steps governments can take, singly some regions such as East Asia are far better off than and together, to minimize potential crises. National they were several decades ago. Even in Sub-Saharan governments have a leading role, but international orga- Africa, where economic performance has been dismal in nizations, subnational levels of government (including recent decades, life expectancies and educational levels urban governments), the private sector, NGOs, and have risen. Still, the number of people living on less than donor organizations all play vital supporting parts. $1 per day is rising. This trend can be reversed, to the These organizations are building the institutions—the benefit of the world’s people, by harnessing the forces of formal and informal rules—that shape the way the globalization and localization in the 21st century. I n t ro d u c t i o n New Directions in Development Thinking T he principal goal of development policy Yet a consensus is emerging on the ele- is to create sustainable improvements in ments of future development policy. the quality of life for all people. While raising per capita incomes and con- Ⅲ Sustainable development has many ob- sumption is part of that goal, other ob- jectives. Insofar as raising per capita jectives—reducing poverty, expanding income improves people’s living stan- access to health services, and increasing dards, it is one among many develop- educational levels—are also important. ment objectives. The overarching aim Meeting these goals requires a compre- of lifting living standards encompasses hensive approach to development. a number of more specific goals: bet- The last half-century has been marked tering people’s health and educational by a mix of pessimism and optimism opportunities, giving everyone the about prospects for development. The chance to participate in public life, Green Revolution held out the prospect helping to ensure a clean environment, of overcoming the Malthusian threat, promoting intergenerational equity, and countries like India succeeded in and much more. achieving food security. But the world’s Ⅲ Development policies are interdepen- burgeoning population, combined with dent. When a policy does not work relatively slow growth in the productiv- well, what is involved may be more ity of food grains in the 1990s, is once than just the individual strategy. Poli- again raising fears of food shortages. cies require complementary measures Some development approaches, such as in order to work best, and a policy Brazil’s import-substitution policies, ap- failure can occur because these com- peared to work for a while but then plements are not in place. failed. The more recent downturn in the Ⅲ Governments play a vital role in devel- most remarkable economic success story opment, but there is no simple set of of all—East Asia—has raised new ques- rules that tells them what to do. There tions about development policies, as has is consensus that governments should the slow response to market reforms adhere to the policy fundamentals, shown by the economies in transition. but beyond that, the part the govern-       ⁄  ment plays depends on its capacity to make effective of the East Asian financial crisis that began in 1997. In decisions, its administrative capabilities, the coun- addition, East Asia is the only region of the world try’s level of development, external conditions, and a where incomes in low- and middle-income countries host of other factors. are converging toward incomes in richer countries. Ⅲ Processes are just as important as policies. Sustained de- Compared to this regional success, the broad picture velopment requires institutions of good governance of development outcomes is worrisome. The average per that embody transparent and participatory processes capita income of the poorest and middle thirds of all and that encompass partnerships and other arrange- countries has lost ground steadily over the last several ments among the government, the private sector, decades compared with the average income of the rich- nongovernmental organizations (NGOs), and other est third (figure 8). Average per capita GDP of the mid- elements of civil society. dle third has dropped from 12.5 to 11.4 percent of the richest third and that of the poorest third from 3.1 to The idea that development has multiple goals and 1.9 percent. In fact, rich countries have been growing that the policies and processes for meeting them are faster than poor countries since the Industrial Revolu- complex and intertwined has provoked an intense debate tion in the mid-19th century. A recent estimate suggests on the wisdom of traditional development thinking. that the ratio of per capita income between the richest This introduction draws on the threads of that debate to and the poorest countries increased sixfold between review perspectives and lessons from past development 1870 and 1985.1 Such findings are of great concern be- experiences. It emphasizes the need to reach beyond cause they show how difficult it is for poor countries to economics to address societal issues in a holistic fash- close the gap with their wealthier counterparts. ion. The chapter then turns to the role of institutions Standard economic theories predict that, other things in development and points to the institutional changes being equal, poor countries should grow faster than rich that will be necessary to ensure sustainable development ones. For instance, developing countries arguably have in the 21st century. While development still faces many an easier task in copying the new technology and pro- challenges, the opportunities waiting to be grasped in the duction processes that are central to economic develop- new century hold out just as many exciting prospects. ment than industrial countries have in generating them. Capital, expertise, and knowledge should flow from Building on past development experiences The evidence of recent decades demonstrates that while Figure 8 development is possible, it is neither inevitable nor easy. The incomes of rich and poor countries continue The successes have been frequent enough to justify a to diverge sense of confidence in the future. But while these suc- GDP per capita cesses may be replicable in other countries, the failure (1995 U.S. dollars) of many development efforts suggests that the task will 20,000 be a daunting one. One measure of development is per capita GDP, Upper third which is often correlated with other indicators of well- 15,000 being and so serves as a convenient starting point. The average level of per capita GDP in developing countries for which data are available rose at a rate of 2.1 percent 10,000 per year from 1960 to 1997—a growth rate that, if it kept rising, would double average per capita GDP every 35 years or so. 5,000 Middle third But such aggregate data invariably mask an array of Lower third variations across times and places. For example, the ᮡ growth rate of per capita income in developing coun- ᮡ 0 tries rose relatively quickly in the 1960s and 1970s and 1970 1975 1980 1985 1990 1995 leveled out in the 1980s. An optimist might see signs of a return to rapid growth in the first half of the 1990s, Source: World Bank, World Development Indicators, 1999. but such signs have been less apparent in the aftermath :       wealthier countries that have these resources in abun- Figure 9 dance to those developing economies in which they are Investment alone cannot account for variation in growth scarce—and where they should be even more productive. Both past and present development thinking has de- GDP growth rate voted much effort to uncovering explanations for why (percent) low-income countries have difficulties in following this 14 pattern.2 A number of studies show that low-income 12 countries can grow faster than high-income countries 10 (by about 2 percent per year), thus catching up gradu- 8 ally over time, if they implement an appropriate mix of 6 growth-enhancing policies.3 And increasing experience 4 with development outcomes is providing insight into 2 the complexity of the process and the multifaceted ap- 0 proach needed to achieve this growth.4 The complexity of the development process has long –2 been recognized. Arthur Lewis’s classic 1955 study The –4 0 5 10 15 20 25 30 35 40 45 Theory of Economic Growth includes chapters on profit Investment as a percentage of GDP incentives, trade and specialization, economic freedom, institutional change, the growth of knowledge, the ap- Note: Figure shows decade-average investment rates against decade-average GDP growth rates from 1950 to 1990 for a 160- plication of new ideas, savings, investment, population country sample. and output, the public sector and power, and politics.5 Source: Kenny and Williams 1999. But over the years, various development processes have been singled out as “first among equals” in terms of their impact. The conceptual frameworks for develop- equality, and reductions in poverty can proceed to- ment of the last 50 years, especially in their popular- gether, as they have in much of East Asia. Many policies ized versions, tended to focus too heavily on the search promote growth and equality simultaneously. For exam- for a single key to development. When a particular key ple, improving access to education builds human capi- failed to open the door to development in all times and tal and helps the poor, and providing land to poor farm- places, it was set aside in the search for a new one. ers increases not only equality but also productivity. The Development models popular in the 1950s and East Asian countries also showed that countries can have 1960s drew attention to the constraints imposed by high savings rates without high levels of inequality. limited capital accumulation and the inefficiency of re- Development theorists of the 1950s and 1960s also source allocation.6 This attention made increasing in- offered a wide variety of rationales explaining why open vestment (through either transfers from abroad or sav- economies and limited intervention would not suffice ings at home) a major objective. But the experience to spur growth. Many development economists focused of recent decades suggests that a focus on investment on planning as at least a partial solution to the prevail- misses other important aspects of the development ing problems of low investment and slow industrializa- process. Investment rates and growth rates for individ- tion, especially as memories of the Great Depression ual countries between 1950 and 1990 varied consider- made many policymakers skeptical about the virtues of ably (figure 9). Some low-investment countries grew unconstrained market forces. Two other factors seemed rapidly, while a number of high-investment countries to argue for an aggressive government role in develop- had low growth rates.7 Although investment is proba- ment: the U.S. government’s close management of pro- bly the factor that is most closely correlated with eco- duction during World War II, and the investment and nomic growth rates in these four decades, it does not GDP levels of the Soviet Union, which was then surg- fully explain them.8 ing forward under communism despite enormous Early theories of development, especially those asso- human costs. ciated with Simon Kuznets, also argued that inequality Over time, however, it became clear that while gov- generally increases during the early stages of develop- ernments do have a vital role in the development process, ment. Evidence from recent decades has not validated only a few governments have run state enterprises effi- these theories, and it now appears likely that growth, ciently. Returns to investment in the Soviet Union fell      ⁄  almost to zero. Governments padded public sector pay- health in previous decades, reduced distortions and rolls and the overstaffing, combined with inefficiency, greater austerity had become the central elements of the produced large deficits that imposed a fiscal burden and development agenda. diverted needed revenues. Concerns were also mounting The evidence of the last two decades continues to that governments of developing countries were making support the need for macro stability and sector reform. poor decisions in the macroeconomic sphere, leading to Once again, however, an exclusive focus on these issues problems such as inflation and the debt crises in Latin proved insufficient. Some countries followed policies America.9 of liberalization, stabilization, and privatization but In the late 1960s the attention of policymakers began failed to grow as expected. Several African countries im- to shift toward an emphasis on human capital, which is plemented sound macroeconomic policies but still often measured in terms of school enrollment (as a proxy reached an average growth rate of only 0.5 percent per for education) and life expectancy (as a proxy for health year.12 Low-inflation countries with small budget def- status). In the last two decades investment in human cap- icits face many alternate sources of economic instabil- ital has shown impressive results. Rates of return on pri- ity, including weak and inadequately regulated finan- mary education in low-income countries have been as cial institutions, as East Asia discovered. high as 23 percent per year.10 But like investment in The lessons of small versus big government perfor- physical capital, investment in health and education mance were also less clear than expected. In the Russ- alone does not guarantee development. In Sub-Saharan ian Federation the move from inefficient central plan- Africa, for example, life expectancy and school enroll- ning and state ownership to decentralized market ment rates have increased dramatically in recent decades, mechanisms, private ownership, and a profit orienta- but as a group the economies in the region have had slow tion should have increased output, perhaps in tandem and even negative growth since the early 1970s. with a slight increase in inequality. Instead, Russia’s By the 1980s the intellectual climate had shifted economy has shrunk by up to one-third, according to again. Confidence in government planning as a solu- some estimates, and income inequality has increased tion had diminished dramatically. The primary con- dramatically. Living standards have deteriorated along cerns of the day were, in fact, government-induced with GDP, and health indicators have worsened.13 price distortions (such as those associated with tariffs) Other countries intervened to a relatively large ex- and inefficiencies arising from government production. tent in markets and enjoyed rapid growth. The govern- Still, governments continued to be recognized as ments of East Asian countries failed to follow many of central to the development process. As the 1991 World the tenets of liberalization in the early stages of devel- Development Report noted: “[M]arkets cannot operate opment, yet their societies have been transformed in in a vacuum—they require a legal and regulatory frame- the last several decades.14 Even with a few years of zero work that only governments can provide. And markets or negative growth in the late 1990s, their per capita sometimes prove inadequate or fail altogether in other GDP at the turn of the century is many times what regimes as well. The question is not whether the state it was a half-century ago and far higher than those of or market should dominate: each has a unique role to countries that pursued alternative development strate- play.”11 At the same time, research was showing that the gies. The East Asian governments often pursued indus- market imperfections central to the discussion in the trial policies that promoted particular sectors. They in- 1950s and 1960s were (at least theoretically) more tervened in trade (although more to promote exports widespread than had been previously believed. How- than to inhibit imports). They regulated financial mar- ever, as a response to public sector inefficiency, policy kets, limiting the investment options available to indi- discussion nonetheless focused on market-conforming viduals, encouraging savings, lowering interest rates, solutions: eliminating government-imposed distortions and increasing the profitability of banks and firms.15 associated with protectionism, subsidies, and public Their policies placed heavy emphasis on education and ownership. A solution to the problem of excessive debt technology in order to close the knowledge gap with accumulation was also put forward that involved adjust- more advanced countries. More recently, China has ing the fiscal, monetary, and external imbalances ad- forged its own version of an East Asian–style develop- versely affecting price stability and growth. Like gov- ment path. Its transition strategy for replacing the cen- ernment intervention and investment in education and trally planned economy with a market-oriented regime :       has resulted in extraordinary gains for hundreds of mil- played a part in the most impressive development suc- lions of the poorest people in the world. cess story of the last 50 years—East Asia—undoubtedly The twists and turns of development policy and the contribute to growth and development in general: high nature of the successes and failures around the world il- savings, strong returns to investment, education, trade, lustrate the difficulty of interpreting the development and sound macroeconomic policy. At the same time, de- drama. The situations in which success and failure velopment failures point to the importance of institu- occur differ so much that it is sometimes not apparent tional structures, competition, and control of corrup- which lessons should be extracted or whether they can tion (box 1). be applied in other countries. For example, the role the Studies of World Bank projects illustrate the many government plays depends on a range of factors, in- elements necessary for successful development.16 These cluding administrative capacity, the country’s stage of studies show that projects in countries that adhere to development, and the external conditions it faces. the macroeconomic fundamentals of low inflation, lim- Despite the difficulty of drawing clearly applicable ited budget deficits, and openness to trade and finan- lessons from development history, current development cial flows are more successful than projects in closed thinking has been able to draw on country experiences countries with macroeconomic imbalances. But the to suggest a range of complementary policies. These projects need more than a stable macroeconomy in policies, if implemented together and in a way that takes order to succeed. For example, social projects are more into account the situations of individual countries, are likely to succeed if they emphasize beneficiary partici- likely to encourage development. Several factors that pation and are responsive to gender concerns. Studies Box 1 Lessons from East Asia and Eastern Europe The success of East Asia provides some notable lessons on Each of these points opens up a number of new issues. For successful development strategies. instance, the high savings rate might have been generated by personal preferences, government policies, or a combination Ⅲ Savings. All the East Asian countries had much higher sav- of the two. And while these countries invested their savings ings rates than other developing countries. From 1990 to well, many others do not. Nonetheless these elements of 1997, for example, gross domestic savings in the countries overtly successful policies point the way toward a partial de- of East Asia and Pacific were 36 percent of GDP, compared velopment agenda. with 20 percent in Latin America and the Caribbean and 17 Failures as well as successes can provide positive lessons percent in Sub-Saharan Africa. for development. Among the most recent (and sometimes Ⅲ Investment. The East Asian countries managed to invest spectacular) examples of such failures are Russia, some of the these savings productively, so that the return on capital in- economies in transition in Central and Eastern Europe, and vestment remained higher than in most other developing several East Asian countries affected by the economic and fi- countries (at least until the mid-1990s). nancial crisis of the mid-1990s. Their experiences point to Ⅲ Education. These economies invested heavily in educa- other factors that can influence economic growth, including tion—including female education. The investments paid off corporate and public governance and competition. in contributions to growth. Ⅲ Knowledge. The East Asian countries managed to narrow Ⅲ Legal frameworks. A sound legal framework helps ensure the knowledge gap with high-income countries by investing that managers and majority shareholders in the corporate heavily in science and engineering education and by encour- realm focus on building firms rather than on looting them. aging foreign direct investment. Ⅲ Corruption. Reducing corruption in the public sphere makes Ⅲ Global integration. The experience of East Asia’s economies a country more attractive to investors. Many privatization ef- shows that developing countries have a greater ability to forts have been racked by corruption, undermining confi- enter global markets for manufactured goods than many be- dence in both the government and the market economy. lieved possible several decades ago. The loans-for-shares scheme in Russia was so widely per- Ⅲ Macroeconomic policy. The East Asian countries imple- ceived as raising corruption to new heights that much of the mented sound macroeconomic policies that helped contain resulting wealth is considered illegitimate. inflation and avoid recessions. Indonesia and Thailand had Ⅲ Competition. Competition is essential. It encourages effi- positive real GDP growth from 1970 until 1996. Over that ciency and provides incentives for innovation, but monopo- same time period Malaysia and the Republic of Korea each lies may try to suppress it unless the government steps in. had only one year of negative real GDP growth.      ⁄  also find that government “ownership” of projects is a good policy framework (box 3). What is true of en- essential and that measures of government credibility ergy projects in Sub-Saharan Africa is equally true of are closely correlated with returns on the projects. In privatization programs. The outcome of privatization low-income countries stronger institutions are associ- projects is heavily dependent on governance structures, ated with a 20 percent increase in the likelihood that a macroeconomic and structural factors, the competitive- project will receive a “satisfactory” rating.17 The role of ness of the market, social sustainability, regulatory social capital in project success has also been high- regimes, corporate and commercial law, financial sec- lighted—indeed, it is hard to overemphasize the impor- tor reforms, and the state of business accounting.24 In tance of networks of trust and association for sustain- turn, what is true of power and privatization projects is able development (box 2). Finally, the studies emphasize just as true of efforts to create social safety nets, build the importance of coordinated development efforts schools, or improve the environment. among governments and donors.18 The many goals of development Overall, the impact of World Bank projects depends on a host of factors extraneous to the projects them- The World Bank’s experience with large dam projects selves. A recent review of World Bank energy projects highlights the importance of taking a broad view of in Sub-Saharan Africa offers some vivid examples of the outcomes of projects. In the 1950s and 1960s large these factors, including governance, human capital, and dams were almost synonymous with development. But more recent evidence of their effects on the environ- ment and on the welfare of groups displaced by con- Box 2 struction suggests that these projects must be handled Social capital, development, and poverty with great care if they are to have a positive impact on Social capital refers to the networks and relationships that both encourage trust and reciprocity and shape the quality Box 3 and quantity of a society’s social interactions.19 The level of social capital has a significant impact on a range of de- Explaining power project outcomes in velopment processes. For example: Sub-Saharan Africa Ⅲ In education, teachers are more committed, students Until the mid-1990s the record of World Bank power proj- achieve higher test scores, and school facilities are bet- ects in Sub-Saharan Africa was comparatively weak. Out ter used in communities where parents and citizens take of 44 such projects completed in the region between 1978 an active interest in children’s educational well-being.20 and 1996, 64 percent were rated satisfactory, compared Ⅲ In health services, doctors and nurses are more likely to with a worldwide average of 79 percent. A recent study show up for work and to perform their duties attentively analyzing the causes of this poor performance suggested where their actions are supported and monitored by citi- that a wide range of factors influenced project outcomes zen groups.21 and sector performance, including: Ⅲ In rural development, villages with higher social capital see greater use of credit and agrochemicals and more Ⅲ External factors, such as rising fuel prices, international village-level cooperation in constructing roads.22 interest rates, and terms-of-trade shocks Ⅲ Regulatory and legal structures, including lack of trans- Social capital serves as an insurance mechanism for the parency in regulatory processes poor who are unable to access market-based alternatives. Ⅲ Low technical capacity, especially a limited human re- It is therefore important to facilitate the formation of new source base networks in situations where old ones are disintegrating— Ⅲ Lack of private sector involvement, through either own- as, for example, during urbanization. ership or service contracting Social capital can have an important downside, how- Ⅲ Limited government ownership of reform processes ever. Communities, groups, or networks that are isolated, Ⅲ Weak coordination among donor agencies and little parochial, or counterproductive to society’s collective in- overall government direction. terests (for example, drug cartels) can actually hinder eco- nomic and social development.23 This has led some to This list indicates just how complex and intertwined make a distinction between vertical social capital (gener- the development process can be in practice. ally negative, as in gangs) and horizontal social capital (gen- erally positive, as in community associations). Source: Covarrubias 1999. :       sustainable development. They require a participatory Figure 10 approach that allows all the potential costs to be aired Infant mortality fell in most developing countries from 1980 to 1995, even where income openly and fully.25 This approach is appropriate for did not increase other projects as well. In order to be effective, all proj- ects must be implemented with an awareness of their Reduction in mortality rate for children under five social, civil, environmental, political, and international (percent) implications. 70 Similar lessons can be drawn from experience with 60 development at the macroeconomic level. While in- 50 creased income is clearly an important component of 40 an improved standard of living, its relationship to other 30 20 measures of well-being is complex. For example, those 10 living on less than $1 per day are five times more likely 0 to die before age five than those living on more than –10 $1.26 Nonetheless, recent studies suggest that rates of –20 economic growth over the last 30 years reveal little –30 about the rates of improvement in vital measures of de- –50 0 50 100 150 200 250 velopment such as political stability, education, life ex- Change in per capita GDP (percent) pectancy, child mortality, and gender equality. Reduc- Note: Calculations for this figure are based on a developing country tions in the mortality rate of children under the age of sample taken from the World Bank’s World Development Indicators, five, for example, appear to have little to do with the 1998. Per capita GDP was adjusted for purchasing power parity. speed of economic growth (figure 10). While economic Source: World Bank staff calculations. performance was poor in many developing countries in the 1980s and early 1990s, only one country in the sample used here (Zambia) saw an increase in infant the most effective way to obtain improved health out- and child mortality. comes may be direct spending that improves nutrition One likely reason for this weak relationship is that and discourages smoking, drugs, and alcohol, rather countries and communities place different priorities on than direct expenditures on health care. In some areas education and health. For example, public expenditures the most effective way to improve educational out- on health care are 63 percent of GDP in Latin Ameri- comes for children may not involve increased expendi- can and Caribbean countries and 5 percent of GDP in ture on books or teachers but instead may involve South Asia, but they account for just 2.7 percent of building a rural road or a bridge across a river to facili- GDP in Sub-Saharan Africa. Sri Lanka is often cited as tate access to schools. Countries that pay attention to an example of a poor country that has invested wisely such linkages may discover unexpected improvements in primary health care and has reaped the benefits. In in their indicators of human well-being. 1997 life expectancy averaged 59 years in the world’s Improving health is itself one clear case where tar- low-income countries, and infant deaths averaged 82 geting broad goals is likely to have dramatic spillover per 1,000 live births. But despite its low level of GDP effects. Studies suggest that as much as 30 percent of per capita, Sri Lanka’s life expectancy was 73 years, and the estimated per capita growth rate in the United infant mortality was just 14 per 1,000 live births—not Kingdom between 1870 and 1979 might be associated quite the levels of the high-income countries of the with improvements in health and nutritional status. world, but not far short either.27 Microstudies support such findings—in Indonesia, Further, like all development endeavors, achieve- for example, anemia reduced male productivity by 20 ments in health and education are interrelated, and percent.28 they may also affect other government programs. Improvement in gender equality is another impor- Countries that pursue egalitarian growth strategies— tant example of a development goal that reinforces for example, education or land reform—are more likely other elements of the development agenda. Low levels to perform well on indicators of human well-being. So, of education and training, poor health and nutritional      ⁄  status, and limited access to resources depress women’s Ⅲ Reduce extreme poverty by one-half quality of life throughout the developing world. And Ⅲ Ensure universal primary education and eliminate gender-based discrimination can also be significantly gender disparity in education deleterious to other elements of a sustainable develop- Ⅲ Reduce infant and child mortality by two-thirds and ment agenda. Women are a major part of the workforce maternal mortality by three-quarters, while provid- in developing countries—they comprise about 60 per- ing universal access to reproductive health services cent of Africa’s informal sector and 70 percent of the Ⅲ Implement worldwide national strategies for sustain- region’s agricultural labor, for example. Discrimination able development and reverse trends in the loss of en- reduces their productivity. Estimates from Kenya sug- vironmental resources. gest that if women had the same access to factors and inputs as men, the value of their output would increase These DAC development goals represent an important nearly 22 percent. Discrimination also has a negative step toward recognition of the need for a holistic ap- effect on a range of other development indicators. One proach. More recently, the World Bank has begun pi- study has found that a 10 percent increase in female lit- loting a strategy—the Comprehensive Development eracy rates reduces child mortality by 10 percent (in- Framework—to help operationalize a multifaceted de- creased male literacy had little effect).29 Throughout velopment agenda (box 4). the developing world, gains in the educational level of Economic history and lessons drawn from World women in the 1960–90 period might account for as Bank projects reinforce a number of conclusions. Sus- much as 38 percent of the decline in infant mortality tainable development is a multifaceted process, involv- over that time, and for 58 percent of the drop in total ing multiple instruments and goals. In some cases the fertility rate.30 Improving gender equality is likely to goals and instruments of successful development are produce dramatic results, and it is a goal that can be one and the same—as in the case of gender, health, and targeted at any level of development. education, for example. Strong interlinkages connect While the level of income is not necessarily cor- these goals, so that progress toward one is frequently de- related with a higher standard of living, economic pendent on progress toward others. The role of govern- growth is linked with some negative outcomes—par- ment and the participation of civil society are vital, as ticularly carbon dioxide and sulfur dioxide output and are the importance of sequencing and the complemen- the production of waste.31 This suggests the impor- tarities among development projects. These lessons tance of trade-offs in a comprehensive development point to the importance of identifying bottlenecks— strategy. Policymakers must sometimes make hard the economic or governmental weaknesses that stand in choices when a project or policy supports one develop- the way of a wide range of development objectives. Such ment goal while damaging the prospects for another. lessons are humbling and have come at great cost over Such trade-offs are not limited to those involving proj- the last 50 years. They alter the framework in which the ects with high economic returns and adverse environ- development enterprise should be approached, and they mental impacts. In education, for instance, primary cannot be ignored. schooling may offer the most benefits in terms of in- Since 1990 a number of World Development Reports creasing equity, but tertiary education may offer the have examined many of the elements of a broad-based most in terms of closing the knowledge gap with in- development strategy and have made recommendations dustrial countries. for improving the provision of structural, physical, Hence, development must pursue a range of out- human, and sectoral services (box 5). While some de- comes, such as equality, education, health, the environ- tails may have changed in light of recent experiences, ment, culture, and social well-being, among others. Fur- the tried and effective mechanisms for removing devel- thermore, the linkages between these outcomes—both opment bottlenecks presented in these reports remain positive and negative—need to be fully understood.32 In a useful starting point. All of the reports have discussed cooperation with the World Bank and the United Na- linkages among parts of the development process— tions, the Development Assistance Committee (DAC) poverty, education, health, gender issues, the environ- of the OECD has produced indicators that set global ment, and service provision, for instance. This report targets for the wider goals of development to be reached and future reports (specifically the 2000/2001 report by 2015 or earlier. These goals are to: on poverty) will continue that tradition, providing :       Box 4 The Comprehensive Development Framework The World Bank has been evolving the Comprehensive Devel- balance, encourage the efficient use of resources, and foster opment Framework (CDF) in an attempt to operationalize a transparency when trade-offs need to be made and comple- holistic approach to development. The framework is designed mentarities taken into account in the macroeconomic and to serve as both a planning and a management tool for coordi- social spheres. nating the responses aimed at overcoming bottlenecks and The proposed new framework is based on four areas of de- meeting development goals. Implementing this strategy in any velopment—structural, human, physical, and sectoral. country would involve consulting with and winning the support of a range of actors in civil society, as well as NGOs, donor Ⅲ Structural elements include honest, competent govern- groups, and the private sector. Under the overall direction of ments committed to the fight against corruption; strong the government, different agencies and organizations could co- property and personal rights laws supported by an efficient ordinate their efforts to overcome constraints on development. and honest legal and judicial system; a well-supervised fi- The framework could enable the government to develop a ma- nancial system that promotes transparency; and a strong so- trix of responsibilities in each area showing what each group cial safety net. must do to fight poverty and encourage growth.33 Ⅲ Human development includes universal primary education The CDF is designed to be a means of achieving greater ef- and strong secondary and tertiary systems, and a health sys- fectiveness in reducing poverty. It is based on the following tem that focuses on family planning and child care. principles: Ⅲ Physical concerns center around the efficient provision of water and sewerage; expanded access to reliable electric Ⅲ The country, not assistance agencies, should own its devel- power; access to road, rail, and air transportation and to tele- opment strategy, determining the goals, timing, and se- communications; preservation of the physical environment; quencing of its development programs. and a commitment to preserving cultural and historical sites Ⅲ Governments need to build partnerships with the private and artifacts that buttress indigenous cultures and values. sector, NGOs, assistance agencies, and the organizations Ⅲ Sectoral elements include an integrated rural development of civil society to define development needs and implement strategy, a strong urban management approach, and an en- programs. abling environment for the private sector. Ⅲ A long-term, collective vision of needs and solutions should be articulated that will draw sustained national support. The CDF does not seek to be exhaustive. A stable macro- Ⅲ Structural and social concerns should be treated equally economy, shaped by prudent fiscal and monetary policies, is an and contemporaneously with macroeconomic and financial essential backdrop to the development efforts the CDF pro- concerns. poses. This stable macroeconomic environment occupies the “other half of the balance sheet,” complementing the CDF. And It is important to note that the CDF is meant to be a com- the pressing issues of poverty, gender inequality, knowledge pass, not a blueprint. The way the principles are put into prac- and information gaps, and overpopulation are incorporated into tice will vary from country to country, depending on economic virtually all of its components. Gender, for example, is central and social needs and the priorities of the stakeholders in- to all aspects of a comprehensive framework. Additionally, each volved. Further, the CDF is only at the pilot stage and is very country is likely to have its own unique priorities that would much a work in progress. The mixed record of development need to be included in a matrix that evolves over time. The pri- programs in the past suggests the need for both caution in ap- ority each country gives to trade issues, the labor market, and plication and realism about expected results. Nonetheless, the employment concerns, for example, will depend on the condi- CDF might allow participants in a country’s development pro- tions specific to the economy and the results of a national dia- gram to think more strategically about the sequencing of poli- logue about development priorities and the programs needed cies, programs, and projects. It could help to improve sectoral to address them. practical advice on implementing the many strands of the importance of infrastructure provision. The report broad-based development. addresses environmental concerns at both the global This report extends past analysis in a number of and local levels. It also provides up-to-date lessons from ways. It looks at governance reform in the context of ur- experience and recommendations for successful devel- banization and decentralization. It discusses regulatory opment strategies. reform and examines financial systems in a global con- The role of institutions in development text. Human elements permeate the discussion of the impact of trade and the need for sustainable urban de- A strong network of effective organizations and en- velopment, and the section on urbanization emphasizes abling institutions is central to holistic development.      ⁄  Box 5 A holistic approach to development in past World Development Reports Macroeconomic policy and trade. World Development Re- Education. World Development Report 1998/99: Knowl- port 1991: The Challenge of Development laid out the impor- edge for Development suggested strategies for improving the tance of a stable macroeconomic framework and an open quality of education from the primary to tertiary levels by trade regime for development, a message that has been re- decentralizing, improving information flows, and targeting peated in reports since then. For example, World Development support. Report 1997 noted the role of the WTO in fostering world trade Infrastructure. World Development Report 1994: Infra- (a topic this report will discuss at greater length). structure for Development focused on the urgent need to make Government, regulation, and corruption. World Develop- the provision of infrastructure more efficient through commer- ment Report 1996: From Plan to Market pointed out the poten- cial management (public-private partnerships or privatization), tial economic consequences of corruption and looked at policies competition, and stakeholder involvement. The 1998/99 report that tend to increase or to mitigate its effects. Among other studied the role of reform and government support in improv- things, the report emphasized the need for a strong and inde- ing access to telecommunications. pendent judiciary and discussed methods for strengthening fi- Environment. World Development Report 1992: Develop- nancial systems in transition economies through banking reform ment and the Environment analyzed the linkages among eco- and the development of capital markets. It also examined mech- nomic policy, poverty, and environmental outcomes and dis- anisms that increase the effectiveness of government, includ- cussed methods of providing cost-effective interventions that ing expenditure control, budget management, and tax policy ensure sustainable development. It examined self-enforcing reform. World Development Report 1997: The State in a Chang- policies and standards, the role of local participation, and ing World further explored issues of government reform and improved know-how and technology. The 1998/99 report fo- regulation, looking at the institutions that are needed in a capa- cused on the links between information and environmental ble public sector, discussing restraints on corruption, and out- degradation. lining ways of bringing the state closer to the people. Rural strategy. In its study of poverty, the 1990 World De- Social safety nets. World Development Report 1990: velopment Report presented an effective strategy for improv- Poverty discussed the need for transfers and safety nets to ing access to government services for the rural poor. In partic- complement a market-oriented policy agenda that favors the ular it focused on providing social services and access to poor. It emphasized the importance of efficient targeting, dis- infrastructure, credit, and technology. cussed methods of improving formal social security systems, Private sector strategy. The 1996 World Development and suggested complementary mechanisms for food-based in- Report presented a framework for creating institutions to sup- terventions. World Development Report 1995: Workers in an port the private sector. It discussed the need for clearly de- Integrating World revisited these issues, addressing income fined property rights and laws governing corporations, con- security measures in the formal sector and methods of equip- tracts, competition, bankruptcy, and foreign investment and ping workers for change and of facilitating labor mobility. outlined methods of privatization. The 1997 World Develop- Health. World Development Report 1993: Investing in ment Report took another look at the roles of liberalization, reg- Health reviewed cost-effective mechanisms for providing gov- ulation, and industrial policy in fostering markets. ernment support for improved health care. The broad agenda Gender. World Development Report 1990 noted the covered female education and women’s rights, increased and high rates of return to women’s education and the role of retargeted expenditures, improved management, and decen- community-based health care and family planning services in tralized public-private partnerships. In all areas the mecha- ensuring safe motherhood. These issues were explored in the nisms included delivering information, providing protection 1993 report, which also discussed the broader agenda for against infectious diseases, and ensuring universal access to equality. Last year’s report emphasized the important role of essential clinical services. microcredit schemes for women. The term institutions, as it is used here, refers to sets of sure that contracts are enforced, property rights hon- formal and informal rules governing the actions of in- ored, bankruptcies settled, and competition maintained. dividuals and organizations and the interactions of par- The efficacy of markets, which are themselves insti- ticipants in the development process (box 6). The insti- tutions, depends on the strength of supporting institu- tutional infrastructure of an economy embraces two tions that help align the expectations of agents regarding primary areas. The first includes social capital and the procedures that govern their transactions. Institu- norms—the unwritten rules of behavior that allow co- tions affect the modes of participation and negotiation operation and dispute resolution, with low transaction among groups and, through their incentive effects, shape costs. The second includes formal legal rules, which en- the nature of agents’ reactions and responses. :       Box 6 ernment provides goods directly, it is often a monopoly Institutions, organizations, and incentives supplier. As such, it must not take advantage of its mo- nopoly position to provide a suboptimal level of service This report follows the notion of institutions introduced in to the public. Rather, it must structure itself in a way the new institutional economics, in which institutions are viewed as rules.34 Rules can be formal, taking the shape of that provides incentives for efficient production and for constitutions, laws, regulations, and contracts. Or they can ongoing gains in productivity. World Development Re- be informal, like values and social norms. Institutions simul- port 1997: The State in a Changing World outlined taneously enable and constrain the actions of individuals or methods of improving the operations of policymaking organizations. Institutional reforms specify new rules or and executive agencies that reduce opportunities for alter old ones with the intention of changing the behavior of individuals and organizations in desirable directions. For politicians and civil servants to exploit public ownership example, markets require social norms that offer at least a and control over supplies. The report found that coun- degree of respect for contract and property rights and a tries with stable governments, predictable methods of system of law that can quickly and inexpensively resolve changing laws, secure property rights, and a strong ju- disputes over such matters. Markets also require rules that eliminate unnecessary delays in processing cases and the diciary saw higher investment and growth than coun- biased decisions that make investors nervous about con- tries lacking these institutions. tributing to increased investment and growth. For this rea- A strong regulatory policy is of central importance son, judicial reform is a high priority for many countries. across a range of sectors. This year’s report discusses its Organizations themselves are characterized by internal rules that define for their members prerequisites for eligi- role in the provision of essential urban public services at bility, responsibilities, sanctions, and rewards. How effec- the local level, in resolving global environmental prob- tively and faithfully members pursue the organization’s lems, and in imparting stability to the financial sector. objectives depends on these rules. Large corporations are For example, without appropriate accounting and regu- continuously adapting their internal rules, centralizing some functions, decentralizing others, adding discretion where latory standards, neither bank depositors nor outside in- advantageous, and modifying the criteria for rewards vestors are well positioned to monitor the degree of risk when doing so is likely to improve performance. In many that banks take when making loans. Last year’s World countries civil service reforms that put in place internal Development Report noted that similar problems plague rules of monitoring and accountability are an important item on the policy agenda. Similar reforms of local institu- the relationship between investors and companies in tions aim to improve the delivery of urban services and stock markets more generally, and regulations address- their regulation. ing these information issues offer substantial benefits. Much remains to be learned about the determinants of institutional change. Institutions change slowly but con- Institutions and the provision stantly, either in response to shifts in outside circum- stances or as a result of group conflict and bargaining.35 of human development services Even so, it is possible to posit institutions that can help Governance institutions are also of primary importance stabilize the global economy and improve the prospects in determining how society addresses human develop- for development—and to suggest mechanisms that facili- ment. In the area of education, for example, the forces tate their implementation by aligning incentives with de- sirable outcomes. In this report institutional reform typi- of consumer choice that provide such strong incentives cally implies changing and specifying formal rules that for providers in other markets are limited in various determine the objectives and incentives for the behavior ways. Students and even parents are rarely in a position of individuals and organizations. to assess the quality and relevance of education, and it is costly for students to change schools. Institutional re- forms center around empowering teachers and schools A vital role for effective institutions of governance and improving access to information for both parents and regulation arises across the range of activities en- and students. In Minas Gerais, Brazil, the reforms im- compassed by a broad-based approach to develop- plemented since 1991 have increased school autonomy ment—including the structural, human, physical, and and parental participation and improved student evalu- sectoral elements included in the CDF.36 ation. These reforms, coupled with efforts to build ca- pacity and professional development in school staff, Institutions at the structural level have increased student test scores.37 A well-run civil service and an efficient judiciary are pre- Providing a social safety net that effectively targets requisites for efficient government action. When a gov- the poor requires efficiently designed programs that      ⁄  benefit those most in need. World Development Report banking. But innovative institutional structures can 1998/99 noted a study carried out in Jamaica which overcome this problem, as was discussed in World Devel- found that food stamps distributed through health clin- opment Report 1998/99. In Bangladesh the Grameen ics reach 94 percent of malnourished children. Over 30 Bank’s group lending program has given rural women ac- percent of the total benefits of targeted food stamps go cess to credit. The bank grants loans to members of a to the poorest 20 percent of Jamaican society, while group, who are held collectively responsible, creating in- universal food subsidies provide greater benefits to the centives for members to monitor each other. rich than the poor. Cities present a wide range of positive and negative externalities. They require efficient institutions if they Institutions and the provision of physical services are to benefit from the positive externalities associated A central feature of utilities and infrastructure is net- with agglomeration economies and mitigate the negative work externality—that is, the average costs of provid- externalities of congestion and environmental damage ing services tend to decrease and the usefulness of the that concentrated populations generate. World Develop- service tends to increase as the system grows. For exam- ment Report 1997: The State in a Changing World touched ple, a telephone network with only two connections is on the subject of efficient city government, and this re- costly to set up on a per-person basis and is of little use port examines the topic in greater detail. even to the two parties, since they can call only each Sustainable development is a complex task in which other. But a network with many connections costs less appropriate institutions will play a central role. But and provides greater benefits per user. Network exter- such institutions will not necessarily emerge sponta- nalities create situations that tend toward monopoly neously. Institutions grow and change over time, but ownership; in the absence of competition, however, the process of evolution does not necessarily produce firms often overcharge users and operate inefficiently. socially optimal institutions. Institutional change is The telecommunications sector needs regulation to en- more often the result of conflicts over the allocation of force competition, including rules requiring operators societal resources than of planning designed to maxi- to connect each other’s customers at an efficient price. mize social welfare. Thus, while institutions are central A well-run regulatory regime has had a dramatic effect to implementing broad-based development, under- on line rollout in Chile, for example, where a decade of standing which institutional changes will ensure sus- regulated competition has seen a tripling of the num- tainable development in the new century is equally im- ber of telephone lines per capita.38 portant. It requires having a clear conception not only The physical dimension of development concerns of the progress that has already been made but also of also includes the environment. Without some form of the challenges the new century will present. The next regulation, companies would not pay for the health and two sections look at these issues. environmental damage manufacturing processes inflict. The record and outlook for Individuals and organizations will often pollute indis- comprehensive development criminately if they are allowed to, leaving others to pay the costs. In some cases institutions can have a sizable What has been the record to date of development? And impact on pollution simply by collecting information what does the future hold? Answering these questions about what is happening and making that information involves looking at a range of indicators of economic, widely available. One model is the Indonesian Clean human, and environmental welfare. The evidence sug- Rivers program, which used to good advantage firms’ gests that while remarkable progress has been made in concern for their public image in inducing them to some areas, in others development has fallen behind. limit the release of pollutants. By publicizing informa- Current trends suggest that even the gains achieved tion on plant emissions, the program lowered the total could prove short-lived in the absence of new policies discharges of 100 participating plants by one-third be- and institutions. tween 1989 and 1994. Some parts of the developing world have enjoyed lev- els of growth high enough to reduce poverty in recent Institutions and sectoral issues decades. Even in parts of the world where poverty rates Rural areas often suffer because traditional formal mar- remain high, the percentage of the poorest—those liv- kets fail to provide them with adequate services such as ing on less than $1 per day (a frequently used poverty :       line)—has declined. In South Asia, for example, the pro- or less continues to increase. The worldwide total rose portion of the population below the poverty line declined from 1.2 billion in 1987 to 1.5 billion today and, if re- from 45.4 percent in 1987 to 43.1 percent in 1993. But cent trends persist, will reach 1.9 billion by 2015. the proportion is rising in some regions. In Latin Amer- With the recent East Asian crisis, poverty rates have ica it rose from 22.0 percent of the population in 1987 risen again, even in this successful developing region. If to 23.5 percent in 1993, and in Sub-Saharan Africa it in- the poverty level is set at $2 per day, Thailand is pro- creased from 38.5 percent to 39.1 percent (figure 11). jected to see poverty increase by 19.7 percent between The ongoing increase in population levels means 1997 and 2000.39 Inequality typically does not reverse that the absolute number of those living on $1 per day itself quickly, so that if average levels of income change, Figure 11 The number of poor people has risen worldwide, and in some regions the proportion of poor has also increased People living on less than $1 per day (millions) 600 515 480 500 464 1987 1993 446 400 300 219 180 200 91 110 100 2 15 10 11 0 East Asia Europe and Latin America Middle East South Asia Sub-Saharan and Pacific Central Asia and the and Africa Caribbean North Africa People living on less than $1 per day (percentage of total population) 50 45.4 45 43.1 1987 1993 38.5 39.1 40 35 30 28.8 26.0 25 23.5 22.0 20 15 10 4.7 4.1 5 3.5 0.6 0 East Asia Europe and Latin America Middle East South Asia Sub-Saharan and Pacific Central Asia and the and Africa Caribbean North Africa Source: World Bank, World Development Indicators, 1998.      ⁄  the number of individuals at the bottom—those in However, some of these gains are proving fragile. A poverty—will not move in tandem. An informal rule of number of factors—notably prolonged economic crises thumb is that a per capita growth rate of 3 percent or and slumps—have begun to erode previous advances in more is considered the minimum for reducing poverty life expectancy. In African countries burdened with slow rapidly.40 But the average long-term growth rate of de- economic growth and an increasing number of people veloping countries is below that level. Between 1995 with AIDS, life expectancy declined in 1997 to pre- and 1997 only 21 developing countries (12 of them in 1980 levels. Lower life expectancies are also apparent in Asia) met or exceeded this benchmark rate. Among the countries of the former Soviet Union and in Eastern 48 least-developed countries, only 6 exceeded it.41 Europe (figure 12). Measures of health and education offer another per- A number of other fundamental indicators, includ- spective on development and living standards. By and ing adequate calorie intake, reasonable shelter, and ac- large, income increases over the last 50 years have been cess to basic services, remain deeply unsatisfactory. Of accompanied by improvements in a variety of indica- the 4.4 billion people in developing countries, nearly tors of human well-being—life span, infant mortality, three-fifths lack basic sanitation; a third have no access and educational level. Even many low-income countries to clean water; a quarter lack adequate housing; and a with very slow economic growth have been able to man- fifth have no access to modern health services. About age some significant improvements in the quality of life 20 percent of children do not complete five years of of their citizens. In the group of low-income countries school, and a similar percentage does not receive enough as a whole, rates of infant mortality have fallen from calories and protein from their diet. 104 per 1,000 live births in 1970–75 to 59 in 1996, and Progress on countering infectious diseases over the life expectancy has risen by four months each year since last 40 years has been dramatic. While the worldwide 1970. Primary school enrollments have shown signifi- eradication of smallpox is perhaps the best-known suc- cant increases, and adult literacy has risen from 46 to cess, polio is also on the retreat. The last-known case of 70 percent. Gender disparities have narrowed, with the polio caused by wild poliovirus in the Western Hemi- average ratio of girls to boys in secondary school rising sphere was on August 23, 1991, and that in the western from 70:100 in 1980 to 80:100 in 1993. These trends Pacific was in March 1997. Sadly, the majority of African testify to the enormous gains that have been made in countries are still exposed to the poliovirus, as well as to the length and quality of life for billions of the poorest malaria and tuberculosis. New diseases such as AIDS people around the world.42 have also spread with alarming speed (box 7).43 In 1995 Figure 12 Life expectancies have risen greatly in some countries, but others have suffered setbacks Change in life expectancy, 1980–97 Life expectancy at birth (years) (percent) 1980 1997 25 22 21 Oman 60 73 20 18 17 Bangladesh 48 58 n 15 io Indonesia 55 65 t 15 ra de 10 Bolivia 52 61 n Fe e 10 sta bw Honduras 60 69 da ian da a kh ba bi an an 5 ss za m m Yemen, Rep. of 49 54 Rw Ug Ru Ka Za Zi 0 Russian Federation 67 67 0 an h ia ia as f .o Kazakhstan 67 65 es es liv ur Om –5 ep –3 lad on Bo nd Zimbabwe 55 52 ,R –5 ng d Ho In –10 en Ba Uganda 48 42 m Ye –15 –13 –13 Rwanda 46 40 –14 –20 Zambia 50 43 Source: World Bank, World Development Indicators, 1999. :       Box 7 Trends in disease and health care Standards of health profoundly influence economic perfor- emergence of multi-drug resistant (MDR) strains of the tu- mance and quality of life. The past 50 years have witnessed berculin bacillus and of plague,56 and strains of strep and enormous gains in medical science and health care in de- staph bacteria that are beginning to defy even the most veloping countries. However, on the threshold of a new powerful antibiotics such as vancomycin.57 In 1997, TB century, the epidemiological statistics present a mixed pic- caused 2.9 million deaths.58 Under conditions of poverty ture. Many infectious diseases are on the retreat because and crowding in urban areas, this toll could rise further, es- of improved sanitation, nutrition, drugs, and vaccines and pecially when health services are unable to cope. The prob- life expectancies are rising.44 Urbanization could decrease lem is not likely to be limited to low-income countries be- the incidence of waterborne and parasitic diseases if it im- cause in an integrated world in which population mobility is proves access to clean water and better sewerage. And the high, new pathogenic strains diffuse rapidly, turning local urban environment remains reliably inhospitable to certain outbreaks into global problems.59 The speed with which insect vectors.45 But the virulence of old infectious scourges new strains of influenza and cholera have spread through- such as tuberculosis (TB) and malaria has resisted modern out the world testifies to this aspect of globalization. science, and in recent years AIDS has emerged as a sizable At the national level, low- and middle-income countries cause of death and disability among adults in the 15–59 age will need to pursue a multi-track strategy, with the priori- group.46 Moreover, in middle- and many low-income coun- ties dictated by levels of income, financing, age profile, so- tries, the toll exacted by infectious diseases is increasingly cial circumstances, and organizational capacity. Preventive overshadowed by that of noncommunicable diseases such measures propagated by educational campaigns are likely as cancer, injuries, and neuro-psychiatric conditions. The fu- to be the most cost effective against HIV/AIDS, smoking, ture contribution of health to sustainable development will maternal complications, and conditions affecting children. depend on successful action on these several fronts.47 Simple but highly effective technologies such as vitamin A Using the concept of disability adjusted life years (DALY)— and zinc supplements60 and insecticide-treated bednets,61 which expresses years of life lost to premature death and propagated by well-designed campaigns, could be the most years lived with a disability—injuries account for 16 percent effective medium-term measures against malaria, whose of all DALYs, followed by psychiatric conditions (10 percent), diffusion to higher latitudes and altitudes could increase noncommunicable diseases (10 percent), and HIV/AIDS, TB, with climate change.62 and maternal conditions (7 percent). Major childhood condi- Controlling infectious diseases such as TB will require tions caused by diarrheal and respiratory infections and by a broader effort that embraces housing and the infrastruc- malaria comprise another major component of DALYs.48 ture of health services. In a decentralized milieu, this will As countries urbanize and further embrace automobil- require coordination between subnational entities, with ity, the risks from injury are likely to increase—road traffic some centralized oversight and funding. At the very least, accidents are already the ninth leading cause of DALYs a simpler, shorter duration regimen of drugs—as well as worldwide, and the fifth highest in industrial countries.49 the organization to identify the infected, administer treat- With longer life expectancies and older populations, many ment over a period of weeks, and keep track of patients— middle-income and some low-income countries will see an will be necessary for significant gains. 63 In the process of increase in the incidence of chronic diseases and psychi- treating diseases such as TB, medical personnel will have atric disorders. This will lead to rising expenses on diag- to husband the potency of available antibiotics through noses and curative treatment. Furthermore, the concentra- careful use, so as to contain the threat from resistant strains tion of populations in urban areas could exacerbate the of bacteria. spread of infectious diseases such as TB and HIV/AIDS, and Over the longer haul, the answer to many old and new possibly dengue fever, whose vector, the aedes mosquito, diseases, including possibly heart disease, could lie in new thrives in urban environments.50 DNA-based vaccines, better drugs that draw on advances In many developing countries, injuries, HIV/AIDS, and in genetic engineering, and ingenious new ways of target- TB51 could shave several points off the GDP growth rate by ing and destroying pathogens inside the body.64 But chronic winnowing the number of prime age adults. Together with conditions, injury, and poor mental health, which will be re- increased outlays on those suffering from chronic and psy- sponsible for a growing share of the DALYs, will be best chiatric ailments, these diseases could also substantially held in check by sustained educational efforts to influence raise expenditures on health care. Estimates of the effect living and eating habits and by controlling environmental of HIV/AIDS on the worst-hit African countries,52 where the hazards. rate of infection continues to spiral upward, suggest that Greater effort at the national level must be strongly re- potential GDP could be reduced by 10–15 percent over the inforced by well-orchestrated action at the international course of a decade by this one disease alone.53 level, with a coordinated division of labor among interna- Although research on vaccines that offer effective pro- tional organizations and other bodies. This will ensure both tection against HIV/AIDS 54 and malaria continues to move the requisite provision of public goods and the manage- forward (with encouraging progress in the case of the lat- ment of health-related externalities, whose likelihood has ter),55 in other areas ground is being lost because of the been greatly magnified by globalization.65      ⁄  alone more than 9 million children under the age of five Box 8 in developing countries died from preventable causes. Sustainable development Population growth is also connected with the suc- cess or failure of a sustainable development agenda. Any sustainable development agenda must be concerned with intergenerational equity—that is, with ensuring that Long-term projections show that the world’s popula- future generations have the same capability to develop as tion may level off around the middle of the 21st cen- the present generation. A development path is sustainable tury. But before it does, the number of people could only if it ensures that the stock of overall capital assets re- rise from the current level of 6 billion to more than 10 mains constant or increases over time. These assets in- clude manufactured capital (such as machines and roads), billion. This growth will pose difficult issues involving human capital (knowledge and skills), social capital (relation- education, worker training, cultural stability, retire- ships and institutions), and environmental capital (forests ment programs, political majorities, and much more. and coral reefs). The environment matters not just because In parts of the world with fragile ecological systems of its effect on psychic and noneconomic welfare but also because of its impact on production over the long term. that are already threatened by water stress and land Environmental sustainability is also closely connected degradation, increased population pressure could lead with intragenerational equity. While the wealthy consume to environmental catastrophes. Global food supplies more resources overall, the poor tend to rely more heavily will need to double over the next 35 years because of on the direct exploitation of natural resources than the population (and economic) growth. While food sup- rich. If they have no access to nonenvironmental re- sources—and so have limited capacity to adapt—they may plies have actually doubled in the last 25 years, agron- have no choice but to engage in unsustainable uses of en- omists warn that the next doubling will be far more vironmental resources. difficult—especially if it is to be environmentally sus- tainable. In Nepal, for instance, where population Source: Pearce and Warford 1993; Watson and others 1998. growth is reducing average farm size, farmers have been pushed into clearing and cropping hillsides in an at- tempt to maintain their income, and erosion is becom- ing an increasingly serious problem. Economic stagnation or collapse, new health crises, The doubling of food production will have to occur continued population growth, and a range of environ- at a time when 800 million people worldwide are al- mental issues all threaten the gains that have been made ready malnourished, 25 billion tons of topsoil are lost in the development agenda over the last half century and annually, and nearly three-quarters of the ocean’s fish will be a continuing challenge for development in the stocks are overexploited. The current costs of environ- new millennium. These issues will have to be faced in a mental damage, including such things as erosion and world that is very different from what it is today—a world the health and other effects of pollution, have been that will create a new set of challenges and opportunities. estimated at 5 percent or more of GNP worldwide—a figure that will increase rapidly if the world does not A changing world move toward a sustainable development agenda (box 8). The only thing that can be said with certainty about the Water scarcity also threatens the potential for contin- future is that it will differ from the present. Any list of ued improvements in the quality of life of the world’s the most significant changes that the world will undergo poorest people. Today, about one-third of the world is in the next few decades is to some degree arbitrary. How- living under moderate or severe water stress, with at least ever, such a list might include the following possibilities. 19 countries dependent on foreign sources for more than The spread of democracy. The proportion of countries 50 percent of their surface water. By 2050 the propor- that are considered democratic has more than doubled tion of people living at or above moderate water stress since 1974. In a worldwide shift, people are demand- could double (box 9). The great majority will be in de- ing a larger say in the way their governments are run. veloping countries where technical, financial, and man- In addition, demands for increased decentralization of agerial limitations will complicate attempts to respond.66 power often accompany democratic trends. Under conditions of water scarcity, agricultural yields will Urbanization. Agriculture accounts for a larger share fall as irrigation supplies dry up, and health will suffer as of production in low-income countries than it does in more people are reduced to using unsafe water sources high-income economies. In Sub-Saharan Africa, for ex- for drinking and washing. The potential for conflict over ample, agriculture today is about one-quarter of GDP— riparian rights among states is also likely to increase. not very different from the level of U.S. GDP in agri- :       Box 9 The growing threat of water scarcity Global population has doubled since 1940 but fresh water large part because of the imperatives of geography. Nearly 47 usage has risen fourfold. Estimates of the upper limit of usable percent of the land area of the world, excluding Antarctica, falls freshwater suggest a second quadrupling of world water use within international water basins shared by two or more coun- is unlikely.67 The prospect of water scarcity is very real with tries. There are 44 countries with at least 80 percent of their implications for regional peace, global food security, the growth total areas within international basins. And the number of river of cities, and the location of industries. The problem is exacer- and lake basins shared by two or more countries are now more bated by a very uneven distribution. Most available fresh water than 300. is found in industrial countries which have one-fifth of the Water shortages will be especially adverse for agriculture, world’s population. However, nearly all of the 3 billion increase which takes 70–80 percent of all available fresh water in the in global population expected by 2025 will be in developing world. Food security could be a casualty since the growth in countries where water is already scarce. food supply in recent decades has largely been fuelled by irri- Slowing population growth rates are providing some re- gation—both the expansion in area and productivity increases. prieve, dramatically lowering the projections of people who Under current best practice coefficients, it will take 17 percent will be living in countries subject to water stress or scarcity more water to feed the world’s population in 2025. But agri- (defined as fresh water resources of under 1,700 and 1,000 culture is already competing for available water resources with cubic meters per person per year, respectively) in 2050—from urban and industrial uses and the competition will only inten- 3.5 billion (more than tenfold the number in 1990) to 2 billion. sify with time. Although technological advances are making de- But, the problem of water scarcity is expected to get worse salination a feasible option for municipal and industrial usage before it gets better. Currently, only 166 million people in 18 in coastal areas, the costs remain much too high for agricul- countries are suffering from water scarcity, while almost 270 tural purposes. million more in 11 additional countries are considered water Preventing crises, regional disputes and their spillover ef- stressed. fects calls for a mix of economic and institutional measures. The consequences will be felt most acutely in arid and The growing competition for water indicates that there will be semi-arid areas, in rapidly growing coastal regions and in the benefits from treating and pricing it as an economic good. And megacities of the developing world. Urbanization will enlarge the geography of river basins makes unavoidable the effort to the claims on available supplies because of higher per capita search for cooperative arrangements. Both within and across water consumption in urban areas. Twenty-five years ago less countries, allocation and usage of water within a framework of than 40 percent of the world’s population lived in urban areas; clearly defined laws and policies and joint development of in- 25 years in the future this share could reach 60 percent. The frastructure for storing and distributing water would avoid eco- ability to supply safe, clean water and adequate sanitation, al- nomic inefficiencies associated with autarkic solutions. More ready stretched, will be severely tested. importantly, only strategies involving basin-wide rather than One major outcome, with regional and even global conse- national solutions will prove sustainable and advantageous for quences, is the greater likelihood of conflicts over water, in the majority of riparians. culture at the beginning of the 20th century. However, population will also shift as birthrates decline and life ex- two characteristics of economic development are work- pectancies increase. The transition will be particularly ing together to encourage migration away from rural rapid in the industrial world, where in 30 years one in areas and into cities: increased agricultural productivity four people will be over 65—up from one in seven (which allows fewer farmers to produce more food) and today.68 This shift will strongly influence global financial expanded economic opportunities in the manufacturing flows as an increasing number of retirees stop saving and and service sectors. The world’s urban population is set instead begin to draw down their accumulated assets. to rise by almost 1.5 billion people in the next 20 years, The revolution in information and communications and in developing countries the share of the population technology. Economic output has traditionally been visu- living in urban areas is likely to rise from one-half to alized as commodities and goods—wheat, coffee, shirts, about two-thirds by 2025. This growth will have a sig- or automobiles. This economic vision grows less accu- nificant effect on the political clout of cities and will rate each year. In industrial economies the service sector make getting policy right at the municipal level even has accounted for more than half of all output for more important than it is today. decades, and a similar shift toward services is under way Demographic pressures. The world’s population is likely in developing countries. The growing importance of ser- to increase by at least another 4 billion by 2050—a huge vices means that knowledge—how to do things, how to number of people who will need to be fed, sheltered, and communicate, how to work with other people—is be- absorbed into the workforce. The age composition of the coming ever more important, overshadowing the natural      ⁄  resource base. It means that investment in human capi- challenges for governments. Grasping the opportunities tal, including health and education, might become more and meeting the challenges requires building institutions urgent than investment in physical capital. It implies that that will shape and channel the forces of change to best economic output is becoming more “footloose,” since serve the cause of sustainable development. many services and information can be shipped over • • • phone wires or fiber-optic cable or even through the radio spectrum, increasing the range of choices for locat- Development thinking has followed a circuitous path ing production. Improved communications technol- over the last 50 years. At various times it has empha- ogy—and continued improvements in the efficiency of sized market failures and market successes, govern- international transport—have also facilitated the rapid ments as active interventionists or passive enablers, increase in global trade and financial flows.69 openness to trade, saving and investment, education, Threats to the environment. A number of environmen- financial stability, the spread of knowledge, macroeco- tal problems will become significant threats to sustainable nomic stability, and more. The list of policies accepted development if they are not addressed. Climate change as relevant to sustainable development is now longer from atmospheric concentrations of greenhouse gases and than it was even 10 years ago, and some of the em- the growing rate of global species extinction are two of phases have changed. Inflation remains a concern, for the most pressing, but others also demand attention, in- instance, but little evidence exists showing that low to cluding disease, water shortages, and land degradation. moderate rates of inflation have significantly adverse ef- This report argues that the changes the world is al- fects on growth. On the other hand, increasing recog- ready experiencing will greatly increase the importance nition is being given to the importance of strong finan- of global and local (or supra- and subnational) institu- cial institutions, and in the regulatory sphere the focus tions. In many cases the responses to economic, social, has shifted from deregulation to building an effective and environmental changes will require international regulatory framework. cooperation under enhanced or completely new insti- It would be presumptuous to predict which of tutional structures. At the same time, governments will these items will be high on policy agendas one or two increasingly decentralize, devolving greater power to decades from now. But even as the general understand- city and regional authorities. While the central author- ing of development grows and evolves, one lesson re- ities will continue to play an important role in coordi- mains. Understanding the process of development nating and enforcing cooperative outcomes, decisions requires acknowledging both its complexity and the affecting people’s lives will increasingly be taken at the context in which it operates. Simple solutions—invest- international and local levels. ments in physical and human capital, for instance, and The movement toward a globalized and localized unfettered markets—will not work in isolation. Gov- world with many more important players and voices ernments, the private sector, civil society, and donor or- from both above and below the national government ganizations need to work together in support of broad- level offers new opportunities for development and new based development. C h a 1 p t e r The Changing World P olicymakers in the 21st century will evolving into a highly integrated and find themselves pursuing development electronically networked system. So goals in a landscape that has been close are its ties that a retailer in one transformed economically, politically, country can describe the products and socially. Two main forces will be consumers want to producers in sev- shaping the world in which develop- eral other countries, setting in motion ment policy will be defined and imple- immediate revisions in design and pro- mented: globalization (the continu- duction. So closely interwoven are fi- ing integration of the countries of the nancial markets that exchange rates, world) and localization (the desire for interest rates, and stock prices are in- self-determination and the devolution timately linked, and the amount of of power). private capital circulating in financial At the end of the 20th century, markets dwarfs the resources of many globalization has already demonstrated countries. that economic decisions, wherever they At the same time that globalization is are made in the world, must take inter- gathering the world’s countries together, national factors into account. While the forces of localization are tilting the the movement of goods, services, ideas, balance of power within them. The de- and capital across national borders is mand for self-determination can take a not new, its acceleration in the last number of forms, including the replace- decade marks a qualitative break with ment of authoritarian or single-party the past.1 The world is no longer a rule by multiparty politics, greater au- collection of relatively autonomous tonomy of subnational political units, neighborhoods that are only margin- and the involvement of community ally connected (by trade, for example) groups and nongovernmental organi- and are generally immune to events zations (NGOs) in governance. Even in other neighborhoods. Information as private businesses consolidate to and ideas can be accessed in all corners gain leverage on the global market, of the globe at the push of a button. many countries are moving in the op- The international economic order is posite direction, fragmenting, ques-       ⁄  tioning established authority, and groping for mech- like climate change and preservation of biodiversity, anisms to coordinate their internal activities. these actions are but the first of many that must be At first glance, globalization and localization may taken to protect the global commons (see chapter 4). look like countervailing forces, but in fact they often The lack of consensus on many vital issues and the stem from the same source and reinforce each other. difficulties inherent in protracted negotiations stand For example, the same advances in information and in the way of meaningful international institution- communications technology that have been so im- building. portant in the spread of global economic forces often At the national level many countries are learning allow local groups to bypass central authorities in the which policies work well and which should be search for information, visibility, and even financing. avoided for the purposes of macroeconomic stability. Together, these global and local pressures are revolu- Many industrial economies learned of the potential tionizing traditional forms of centralized governance boom-and-bust dangers of capitalism from events and dramatically affecting development thinking. like the Great Depression and thus have put in place Development economics, born after World War a bevy of national policies and institutions. These II, came into being in an era when strong and au- policies seek to moderate economic volatility through tonomous states were the chief decisionmakers. But countercyclical macroeconomic actions designed to that autonomy is gradually being eroded. Markets, minimize the potential instability of capital flows; for instance, have developed the potential to disci- regulate the conduct of private agents; protect in- pline states, punish their mistakes and call their bluff. vestors, depositors, and consumers; disclose the in- Globalization has circumscribed the ability of many formation necessary to assess risks and make prudent central governments to raise revenues by taxing cor- decisions; and provide social insurance to ride out porations, which now have the option of moving part temporary crises. Such institutions have become in- or all of their economic activity to low-tax venues. As tegral parts of the capitalist system in industrial coun- central governments find themselves looking for tries, shaping expectations and fundamentally alter- other sources of revenue, regional and urban commu- ing private sector decisionmaking. As developing nities are coming together to assert their own inter- economies are increasingly exposed to the new global ests, putting yet more pressure on the traditional economy, they build similar institutions in their own forms of governance. The result has been new ways countries. They are likely to find that some poli- of thinking about how to manage the world’s econ- cies in particular offer exceptional payoffs. A stable omies and a corresponding need to create new insti- macroeconomic environment, a liberalized domestic tutions to do so. These institutions will be needed at business services sector, and a legal framework that three levels: supranational, national, and local. induces transparency and protects investors’ rights is At the supranational level some institutions for advantageous. Conversely, deficiencies in financial shaping and channeling the forces of globalization policies and business practices are a recipe for disas- are already in place. The World Trade Organization ter—a bitter lesson learned from the banking and (WTO), the Basle Accords, and the Montreal Proto- currency crises of the late 20th century. col, which affect, respectively, trade, banking sys- At the subnational level localization has led many tems, and the release of ozone-depleting chemical central governments to grant political, fiscal, and ad- agents worldwide, are representative of the kinds of ministrative powers to local governments. But arrange- institutions the world will need in the 21st century. ments aimed at maintaining workable intergovern- The events of the 1980s and 1990s have shown that mental relations have not kept pace with the speed of existing institutions are far from sufficient to address decentralization. Ideally, decentralization rests upon the economic and environmental issues of the future: effective institutions that determine voting proce- many more are needed. Economic catastrophes like dures, provide for the disclosure of information, en- the Latin American debt crises of the 1980s and the sure accountability at the local level, and define a sys- meltdown of the East Asian economies in the late tem of allocating resources and responsibilities to 1990s may continue to occur, perhaps in even more subnational authorities. But the devolution of central dramatic forms. And while countries have begun ini- government authority and functions has often pro- tiating responses to important environmental issues ceeded at its own pace—with elements not only of     learning by doing but of suffering by stumbling. stituent firms access to new markets and commercial There are vitually no good models of decentralization. relationships and facilitate technology transfer. Ad- Recent macroeconomic instability arising in part vances in information technology help to link firms from tensions between the central government and from developing countries into global production net- subnational entities (although the cases are very dif- works. General Electric, for instance, posts informa- ferent) in Brazil and Russia show how important har- tion on its components requirements on the Internet, monious relations between central and local authori- and firms from all over the globe bid to supply them. ties are in creating and sustaining market confidence. The tremendous growth of trade in services and, Governments at all levels have also begun to under- more recently, of electronic commerce is also a part stand the importance of due process and inclusive, of the new trade pattern. Exports of commercial ser- participatory, and consensual modes of public sector vices have been growing on every continent (particu- decisionmaking and resource allocation. larly Asia) throughout the 1990s (figure 1.1). This This chapter sets the stage for later chapters by change has its own special significance, as services are defining the forces of globalization and localization. frequently used in the production of goods and even It explores issues such as trade, capital flows, and the other services. Enhanced international competition implications of urban growth. And it looks at the in- in services means reductions in price and improve- stitutional changes that will be needed to keep de- ments in quality that will enhance the competitive- velopment sustainable, providing the groundwork ness of downstream industries. Both industrial and for a more detailed examination in later chapters of developing economies have much to gain by opening the report. their markets. Developing countries would derive large gains from an easing of barriers to agricultural International trade products and to labor-intensive construction and International trade flows are penetrating deeper into maritime services.3 Over the longer term, electronic the workings of developing economies, affecting the business will loom large as an area where expanding overall economic structure in general and income dis- opportunities for trade require an expanding frame- tribution, employment practices, and productivity work of rules.4 growth in particular.2 Trade in goods and services has Underpinning this surge in trade flows is the grow- grown twice as fast as global GDP in the 1990s, and ing commitment developing economies have shown the share attributable to developing countries has to liberalizing their trade regimes. Their resolution climbed from 23 to 29 percent. These aggregate has taken many forms: membership in the WTO numbers do not reflect the important compositional (110 of 152 developing countries were members in changes of the last 10 years, which offer developing 1999), participation in regional trade agreements, economies new opportunities for growth. Chapter 2 and unilateral reforms (figure 1.2).5 But this push for addresses these challenges and opportunities in depth. trade reform is meeting with increased resistance, es- The compositional shifts in trade have created a pecially in industrial economies, where adjustment to new pattern in the international exchange of goods, the competitive pressure of the international market- services, and ideas. Trade in components is one part place can be a painful process. Successful trade reform of that new pattern. “Sourcing” such components requires reallocating resources among economic from abroad is an increasingly common practice, and groups, and that adjustment can be costly for some. use of the Internet is sure to expand the process, en- Increasingly, governments are recognizing that suc- couraging entry by new producers throughout the de- cessful trade reform requires flexible labor market in- veloping world. While precise numbers are difficult stitutions, a point developed in chapter 2. Import- to come by, in the early 1990s one-third of all manu- competing firms are also resisting further trade reform factures trade (approximately $800 billion) involved by using antidumping laws to reverse the gains in parts and components. This type of trade has gener- market access previous reforms have secured. At least ated an ever-spreading web of global production net- 29 countries were applying such laws by 1997, and works that connect subsidiaries within transnational many more had them on their books. firms to unrelated designers, producers, and distribu- Although the 1990s saw impressive progress in tors of components. These networks offer their con- liberalizing trade regimes, sustaining that momen-      ⁄  Figure 1.1 Exports of commercial services have surged in most regions since 1990 Annual growth rate of exports of commercial services, 1990–97 (percent) 20 18 16 14 12 10 8 6 4 2 0 Industrial East Asia Europe and Latin America Middle East Sub-Saharan South Asia countries and Pacific Central Asia and the and Africa Caribbean North Africa Note: The annual growth rate for the European Union was zero in 1997. Figures for the European Union exclude intra–European Union trade. Source: WTO, Annual Report, various years. tum over the next 25 years will be more difficult. cial markets a top priority. Financial flows soared in The Millennium Round commencing in November the 1990s, spurred by the greater readiness of coun- 1999 will provide the international community with tries to liberalize capital account transactions. Even an opportunity to meet the challenge. For the devel- though they slipped in 1998, such flows are resum- oping countries it will be important to be fully en- gaged and to use the technical expertise at their dis- Figure 1.2 An increasing number of developing countries is posal to arrive at favorable outcomes in areas such as committed to trade reform liberalization of agricultural trade and of trade in those services of greatest relevance to their future de- Number of GATT/WTO member countries velopment. Recognizing that trade reform creates 140 both winners and losers (and more of the former Non-OECD members 120 than the latter) is the starting point. The real test will OECD members be persuading the winners to forgo some of their 100 gains in order to compensate influential losers who 80 110 could otherwise stymie the process of reform. 76 60 International financial flows 61 The financial crises of 1997–99 have put the grow- 40 ing interdependencies among countries in the spot- 20 light and led to their intense scrutiny. International 24 24 24 capital flows to developing countries, though still 0 1980 1990 1999 concentrated in a dozen or so host economies, are Note: Number of OECD members is for 1999. rapidly becoming a major force, making the effective Source: WTO, Annual Report, various years. development, regulation, and liberalization of finan-     ing their upward trend.6 Much has rightly been made the rate of growth of pension assets in the United of the technological developments in computing and Kingdom and the United States could drop to 6–7 telecommunications that are reducing transactions percent a year over the medium term, the projected costs. In addition, considerable attention has been value of global assets for 2002 is a hefty $13.7 tril- given to the possibility that hedge funds and the use lion. These resources will be aggressively seeking high of new derivatives instruments could increase the returns throughout the world. volatility of capital flows.7 At the same time, finan- Increases in demand for funds will match, if not cial innovation has done much to contain the newly exceed, any increase in supply. Some 85 percent of emergent risks and create a rich menu of investment the world’s people reside in developing countries, possibilities—another trend that will not be arrested, half of them in cities. Large numbers of them (close simply because the potential rewards are so attrac- to 1.5 billion in 2000) live on less than $1 a day (a tive.8 More significant, the supply of financial re- widely used poverty line). To modernize, industrial- sources will expand over the next two decades, fed by ize, and urbanize, developing countries will need pension and mutual funds in industrial societies (box huge injections of capital. Most of it will come from 1.1). The value of global pension assets rose from $6 domestic savings, but well-run developing countries trillion in 1992 to $9.7 trillion in 1997. Although offering solid returns can expect to supplement their Box 1.1 The global macroeconomics of aging The aging of populations in industrial countries and some in- country studies predict. The growing numbers of women join- dustrializing East Asian economies could seriously reduce the ing the labor force will partially offset the decreasing number international supply of capital by 2025. Three factors will de- of older male workers. And policies and institutions can nar- termine exactly how serious the reduction will be: the effect row (if not close) the savings gap in some industrial countries of aging and rising dependency ratios on household savings, and prevent shortages in developing economies. the ages at which people retire, and the coverage provided by Research on Japan and East Asia supports the view that social security systems. savings could decline as populations age but that diminishing The bleakest scenario projects a substantial drop in house- rates of investment will more than offset lower savings.10 Fur- hold savings in industrial and East Asian economies as the thermore, as fertility declines in South Asia and Central and number of people over 65 continues to climb. It points to an South America, dependency ratios will fall during the next two increasing tendency for people to retire in their fifties, as many decades, savings could climb—and countries in the region already do in Europe. And it indicates that unreformed, pay-as- could gradually become capital exporters, like Japan.11 In fact, you-go social security schemes will go bankrupt or at least some recent research on the United Kingdom and the United come under great pressure. States points to the likelihood of higher savings rates as baby The median age of the population in northeast Asia will rise boomers approach retirement in the next two decades.12 from 28 years to 36 years between 1995 and 2015.9 Just 12.5 Some sociologists believe that the retirement age will stop percent of the U.S. population and 11.8 percent of the Japan- falling in the next two decades and may even begin to climb.13 ese population were over 65 in 1990, proportions that will rise If it does—and there are plenty of opposing views that stress to 18.7 and 26.7 percent by 2025. Between 1990 and 2025 the attractiveness of retirement and the declining price of rapid aging will raise the share of the 65-plus cohort from 6 to recreation—a savings shortfall in industrial countries would be 13.3 percent in China and from 5 to 15 percent in the Republic a less pressing problem.14 But even if this favorable scenario of Korea. becomes more likely, industrial and developing countries with As countries begin to gray, the number of men between aging populations need to accelerate reform. Pay-as-you-earn the ages of 60 and 64 who are still in the labor force is drop- social security schemes in industrial countries will run out of ping precipitously. In the 35 years leading up to 1995 the per- money in the next two or three decades unless governments centage of men in this age group who were still working fell increase funding, shave benefits, and maintain or raise the age from 80 to 55 percent in the United States, from 80 to 20 per- of retirement. Raising contributions or cutting benefits will gen- cent in Italy, and from 70 to 15 percent in France. This scenario erate resistance, but this adjustment is unavoidable. The total suggests a severe global capital shortage that raises interest bill over the next 30 years for pensions and medical care for rates and depresses growth, trade, and commodity prices—a the aged is estimated at $64 trillion.15 To meet their share of gloomy prospect. these expenses, industrial countries need to create an institu- A second and much brighter scenario suggests that a sav- tional framework that minimizes the threat of inadequate sav- ings crunch can be avoided. Household surveys show that ings by ensuring that social security schemes are fully funded aging may not lead to the steep decline in savings some cross- and by discouraging early retirement.16      ⁄  savings with resources from all over the globe.17 De- have been badly bruised by financial turbulence asso- veloping countries are also the fastest-growing mar- ciated with the East Asian meltdown. In fact, the kets for the products of multinational corpora- costs of the crisis have been much higher than those tions.18 As these markets expand they will attract associated with other recent financial debacles (figure ever-greater amounts of foreign direct investment, 1.3). But the fiscal costs pale in comparison with the which provides jobs and managerial and technical forgone growth and increased poverty and inequality expertise, as well as capital. But the governments of these crises can create, especially in urban areas (fig- developing countries must take measures to attract ure 1.4).27 The East Asian crsis has abruptly pushed such investment, since it will not automatically find the issue of sequencing liberalization measures to the its way to them. In 1996 investors sent only one- top of the policy agenda. Several questions need to quarter of their money to the developing world.19 be answered. What role can capital controls play in The globalization of financial markets affects de- minimizing exposure to sudden changes in the senti- velopment because finance plays such an important ment of portfolio investors? Are controls on capital role in economic growth and industrialization.20 Fi- outflows desirable, or even possible? And, given the nancial globalization affects growth in two ways: by increasing number of international transactions of increasing the global supply of capital, and by pro- goods and services, how easily can these controls be moting domestic financial development that im- sidestepped? proves allocative efficiency, creates new financial Like earlier crises, the East Asian meltdown has instruments,21 and raises the quality of banking ser- enhanced the attractiveness of long-term capital in- vices.22 Competition comes not only from other vestment, with one difference. Until recently, govern- domestic banks but also from foreign banks and ments preferred debt to equity financing or to foreign from thriving nonbank financial intermediaries. Both direct investment, both because they did not want complement banks and, in the case of stock markets foreign interests controlling major segments of the and other monitoring agencies, enhance discipline by economy and because domestic owners of major cor- continuously assessing information on portfolios and porations feared losing control.28 The mood began performance.23 Moreover, experience suggests that shifting as countries recognized that foreign direct in- foreign financial institutions do not undermine do- vestment brings with it not only capital but also tech- mestic banking systems; they are rarely dominant and tend to exhibit a long-term commitment.24 Figure 1.3 The financial performance of emerging markets Nonperforming loans can account in the 1990s made capital account liberalization an for up to 50 percent of all bank loans at the peak of a banking crisis attractive option for developing countries. Markets seemed broadly stable and fairly disciplined, and Nonperforming loans as share of all bank loans many countries began to view the recommended (percent) sequence of liberalization (starting with the building Indonesia 49 of regulatory capabilities and the strengthening of banking and financial markets) as less important Thailand 44 than research had indicated it was.25 Several devel- Malaysia 22 oping countries, urged by the weight of opinion in some industrial countries, began loosening controls Chile 16 (1981–85) on inflows and outflows of capital, and while most Mexico retained some constraints, a few abolished all of (1994–95) 11 them.26 Furthermore, openness remained the most Brazil 9 popular option as containing outflows became in- (1994–96) creasingly difficult and the advantages of inflows 0 10 20 30 40 50 60 70 80 grew ever more evident. Source: Figures for Indonesia, Malaysia, and Thailand are from official The crisis in East Asia in 1997 made policymak- sources; figures for other countries are from Wall Street Journal, ers apprehensive about further financial globalization. December 9, 1998. Several of the most successful emerging economies     Figure 1.4 Figure 1.5 Resolving bank crises can cost Foreign direct investment was less volatile up to 40 percent of GDP than commercial bank loans and total portfolio flows, 1992–97 Resolution costs (percentage of GDP) Billions of U.S. dollars 50 120 0.8 Average flows (left axis) Coefficient of variation (right axis) 0.7 40 100 0.6 80 30 0.5 60 0.4 20 0.3 40 10 0.2 20 0 0.1 –8 le oi ico oi sia –9 re 94 . 19 ep –9 R 82 hi 5 ng 88 oi ng 1 93 FY ng ex ng ne 4 R 0 0 19 C 19 'lv h –o M –o do 19 ia, d ec Foreign Foreign Equity Bonds Commercial on 97 In te Cz Cô bank ed direct portfolio 95 ac 19 loans 19 investment investment M Note: Resolution costs include the government’s direct costs as well Source: UNCTAD, World Investment Report, 1998. as quasi-fiscal costs such as exchange rate subsidies, as defined by the IMF. Source: Caprio and Klingebiel 1999. details how developing countries can reform their in- stitutions and policies to attract more foreign direct nology, market access, and organizational skills.29 investment. Studies of recent episodes of financial turmoil have The East Asian crisis also raised the issue of focused on the volatility of certain private financial whether coordinated macroeconomic and regulatory flows and the ways in which it helps create an unsta- actions could have averted or mitigated the crisis and ble environment and hurts economic development. lessened the contagion effects. A number of institu- An analysis of the period 1992–97 shows that foreign tional possibilities are explored in chapter 3. Closer direct investment was less volatile (as measured by the policy coordination among the principal economies coefficient of variation) than commercial bank loans in the Asia-Pacific region might have kept the ex- and foreign portfolio flows (figure 1.5). change rate and associated interest rate fluctuations In 1997 developing countries accounted for 30 within reasonable bounds, leading to earlier con- percent of the foreign direct investment stock, or certed action to contain the crisis.33 Coordination $1.04 trillion, 90 percent of which originated in in- and some degree of uniformity could also have dustrial countries (table 1.1). Five countries—Ar- extended to financial regulation. Regional and pos- gentina, Brazil, China, Mexico, and Poland—received sibly international bodies could have reviewed na- half the total for developing countries.30 Multina- tional banking practices to determine their com- tional corporations account for much of this invest- pliance with the basic prudential rules for banks ment. Their investment stimulates export-led growth established by the Basle Accords. (In principle, coun- in well-positioned economies through spillovers aris- tries can adopt unilateral or regional standards ing from the sourcing of their products, and distribu- higher than those in the Basle Accords, including the tion of their production facilities.31 Philips Electron- changes in the accords proposed in 1999.)34 ics, for instance, employs more workers in China than International migration in the Netherlands. Alliances between multinationals continue to fuse markets as corporations take advan- Along with goods, services, and investment, people tage of scale and scope economies and cope with the are crossing borders in record numbers. Each year be- rising cost of technological innovation.32 Chapter 3 tween 2 million and 3 million people emigrate, with      ⁄  Table 1.1 not temporary, and they extend beyond those who World foreign direct investment stock, 1997 moved. For example, the substantial displacement of people from southwestern Afghanistan caused consid- Amount erable damage to the pattern of cultivation practiced (billions of Percentage there.40 So many people left the area that the popula- Region U.S. dollars) of total tion fell below the levels needed to maintain the coun- World 3,455.5 100.0 try’s basic agricultural infrastructure. Industrial countries 2,349.4 68.0 Western Europe 1,276.5 36.9 Cross-border migration, combined with the “brain North America 857.9 24.8 drain” from developing to industrial countries, will Other industrial countries 215.1 6.2 be one of the major forces shaping the landscape of Developing countries 1,043.7 30.2 Argentina, Brazil, the 21st century, for at least three reasons. First, mi- and Mexico 249.2 7.2 gration is causing dramatic shifts in the demographic Other Latin America 126.2 3.7 profiles of both industrial and developing countries. China (includes Hong Kong) 244.2 7.1 Second, the movement of highly skilled people from Southeast Asiaa 253.1 7.3 the developing world affects low-income countries Other Asia 96.3 2.8 and recipient countries alike. Third, the international Africa 65.2 1.9 diasporas have tremendous business potential. Other developing countries 9.4 0.3 a. Indonesia, the Republic of Korea, Malaysia, Philippines, Singapore, In the next few decades many countries will see Taiwan (China), and Thailand. profound changes in their population growth rates Source: UNCTAD, World Investment Report, 1998. and demographic profiles. Indigenous populations are declining in most industrial and East European coun- tries, where fertility rates are low. But population growth rates remain high in Asia and Sub-Saharan the majority of them going to just four host countries: Africa, although they have begun to slow. Shortages the United States, Germany, Canada, and Australia, of agricultural land and urban unemployment are in that order.35 At the beginning of the 21st century, two important concerns, and through migration they more than 130 million people live outside the coun- could lead to problems for other economies. In Africa, tries of their birth, and that number has been rising parts of the Middle East, and South Asia, intense by about 2 percent a year. In relative terms the num- competition for jobs could create an additional incen- ber of migrants is a modest 2.3 percent of world pop- tive to emigrate. Shrinking and aging populations in ulation. But they are concentrated in just a few re- Europe, Japan, and the United States might also boost gions—North America, Western Europe, Oceania, the demand for migrant workers, as it did in Western and the Middle East.36 In North America and West- Europe between the mid-1950s and the mid-1970s.41 ern Europe the migrant stock grew at 2.5 percent a In a positive scenario, policy reforms in develop- year between 1965 and 1990, far outstripping growth ing countries, greater financial and trade integration, among indigenous populations. If Oceania is in- short-term migration generated by the liberalization cluded in this group, 1 in every 13 people living in of construction services, and increased possibilities these regions is foreign born.37 While the net benefits for emigrating to industrial countries could enable are positive for the receiving countries and for many low-income countries to cope with population pres- of the originating countries as well, the resulting eth- sures during the demographic transition. Foreign in- nic and labor market tensions in urban areas have led vestment and trade also have a role to play in devel- to tighter immigration restrictions in some countries. oping countries, where they accelerate growth, expand Conflict and natural disasters have dramatically in- employment opportunities, and thus reduce incen- creased the number of refugees. By 1975 a total of 2.5 tives to emigrate.42 This optimism must be tempered, million refugees had crossed national borders, but by however, by the fact that new, low-skilled migrants 1995 the total had risen to 23 million.38 To that fig- face serious hurdles as they enter labor markets in in- ure must be added the more than 20 million internally dustrial countries.43 displaced persons who have migrated within their In other, less attractive scenarios, globalization own countries.39 The effects of such displacement are slows, developing countries have less access to inter-     national capital and markets, and cross-border mi- verse migration, however. Among the most impor- gration becomes more difficult because industrial tant are encouraging emigrants to maintain their countries are reluctant to liberalize trade in services links with the home country and supplying informa- that entail short-term cross-border labor movement tion and advice both before and after their return. and adopt policies to significantly reduce immigra- The market for highly skilled workers will be- tion. While a few low-income countries may re- come even more globally integrated in the com- spond decisively—reducing fertility and promoting ing decades, and increasing returns to skilled peo- growth by mobilizing domestic resources and foster- ple might continue to favor spatial concentration. ing innovation—most are likely to experience ever- Knowledge workers will cross borders freely, facili- greater instability and slow income growth. tating the circulation of technology, inducing the A second concern is the emigration of skilled growth of technology-intensive industries (as in workers from developing economies, especially from Israel), and helping to create a truly global market- Africa and South Asia. A brain drain can impair a place of skills.48 Because development requires a developing country’s capacity to harness modern highly skilled workforce, primary and secondary ed- agricultural and industrial technology. Some coun- ucation will continue to be important. Countries tries in Sub-Saharan Africa, the Caribbean, Central unable or unwilling to create such a workforce, com- America, and South Asia have, in fact, lost one-third pete for skilled workers, and build a technology- of their skilled workers.44 But recent research also friendly environment will find themselves stuck on points to the benefits of outward migration. The the lower rungs of the income ladder. In order to most important of these is the money migrants send narrow the gap with rich countries, developing econ- to their countries of origin. These sums can be sub- omies must also put in place policies to nurture, stantial: foreign workers remit about $75 billion to through tertiary-level training, and effectively em- their home countries each year, 50 percent more ploy skilled workers, as the Republic of Korea and than total official development assistance.45 Those Taiwan (China) have done. Without such policies, remittances are used to support family members, or manufacturing and service activities with high value they may be invested (primarily in housing), thus added will not take root in the countries where the stimulating other expenditures.46 need is most urgent. Of all the potential overseas investors in a coun- A third facet of international migration in the 21st try, emigrants are likely to be the best informed century will be the expansion of far-flung diasporas about business and employment practices and legal from developing countries—another source of global norms. A country that adopts measures to enhance interconnection (box 1.2). Diasporas serve as infor- foreign direct investment and integrates with global mal channels for the flow of information, market in- production networks by maintaining low and pre- telligence, capital, and skills. They may supplement dictable trade barriers will find that doing so gener- formal channels that rely on market institutions, pro- ates additional benefits. If the quality and techno- viding a way for migrants to conduct transactions in logical sophistication of a developing country’s an atmosphere of trust. In this way they act to offset exports increases, highly skilled emigrants may de- information asymmetries and other market failures. cide to return. Several East Asian economies have Modern diasporas, like their Mediterranean prede- benefited from this reflux. cessors, expedite business transactions by resolving However, experience suggests that returning emi- monitoring problems, reducing opportunism, and grants do not always benefit a country. A study of building reputations and ethnic trust based on net- reverse migration in Turkey notes that only half of working.49 As migration continues, diasporas will ex- the returned migrants were economically active in pand, tying together regions and continents. Even if 1988.47 Of those, 90 percent were self-employed, governments attempt to slow the process, communi- and many had used funds saved abroad to establish cations, technology, and human relationships will new businesses. The few returnees who did have ed- maintain this trend. ucational qualifications found little demand for their Governments in South Asia, Central and South skills in Turkey. National governments can take a America, and Sub-Saharan Africa have made limited number of measures to increase the benefits of re- efforts to exploit the potential of overseas networks      ⁄  Box 1.2 The international Chinese network The Chinese diaspora embraces more than 50 million people, than the response from non-Chinese multinational corpora- commands enormous resources, and is a force behind the de- tions. Chinese investors have an edge over other investors, velopment of the Asia-Pacific region.50 This community, with who do not have an intimate knowledge of the region’s eco- its interlaced informal social and business ties and its formal nomic conditions and businesses. But an important caveat overseas Chinese associations, is a source of dynamism for must be added here. Subnational governments must partici- many East Asian economies. It has bridged market failures, pate in building rules-based institutions in order to encourage created markets where there were none, and helped emerg- continued investment. Only if every important level of govern- ing economies become competitive in a remarkably short time. ment enhances the predictability of its laws and regulations These successes are the result of collaboration with local com- will the countries of the Asia-Pacific region realize the many munities and governments in the region. The cross-hatching benefits the Chinese diaspora offers. of formal and informal business linkages will persist and per- In the meantime, ethnic Chinese have been making sub- haps eventually dwarf official linkages. stantial investments in Europe and the United States, particu- In addition to producing, assembling, and distributing larly in the computer industry. In 1997 Taiwanese (Chinese) goods through extensive manufacturing chains in the Asia- firms invested in 55 manufacturing projects throughout Eu- Pacific region, the Chinese diaspora has acquired widespread rope, 44 of them in the computer industry.51 The desire to be property holdings and citizenship rights. Citizenship rights are close to product and process development has also fueled an an elaborate mechanism for hedging against the risk of abrupt increase in the number of ethnic Chinese firms in California’s changes in economic conditions, political regimes, and regula- Silicon Valley. At the same time, other migrant communities tions. That risk falls when developing economies in the region are increasing their business and commercial orientation. The adopt rules-based regimes for trade, investment, and other South Asian diaspora, with a network reaching from Southeast policies, inducing members of the diaspora to shift their invest- Asia to the Middle East, the United Kingdom, and North Amer- ment portfolios toward direct investment and away from risk- ica, has a net worth of between $150 billion and 300 billion. Its hedging devices. potential remains to be tapped in the early 21st century. And Indeed, the diaspora’s response to emergent rules-based throughout the Americas, Hispanics are developing networks regimes could be much faster and, at least initially, stronger that profoundly affect industrial development and trade. to further development. The push to form partner- scientific evidence available on environmental prob- ships may come from local governments, as in lems commands the attention of governments and China, with the central authorities working to create the public alike. Moreover, along with globalization an environment conducive to such interaction. The has come a new recognition of a shared responsibility main hurdles in this process relate to openness and for the environment. Numerous organizations—in- regulation. As long as economies remain inwardly ternational, governmental, and nongovernmental— oriented, predisposed to regulating business activ- with a deep interest in this issue have appeared on the ities, and prone to arbitrary actions, diasporas may scene. These bodies have made full use of the United not be able to evolve into business networks that Nations system and the abilities of new communica- strengthen markets and prod development. In the tions technology to reach people all over the world.52 next few decades, however, countries with large and Climate change, the loss of biodiversity, and other growing emigrant communities scattered through- issues related to the global commons are slowly being out the world will have the opportunity to tap into recognized as problems that the community of na- the development potential of their diasporas. tions must take on collectively. Left unattended, they will worsen as the planet becomes more crowded and Global environmental challenges overpopulation puts increasing pressure on natural Environmental concerns have long been the subject resources. Many of these issues are closely linked to of international interest, in part because of the bur- the potential success of development efforts in poor geoning world population. But at the end of the 20th countries, and the growing awareness of these link- century, global concerns have acquired a new urgency. ages is part of the continuing shift in the development Over the past 20 years the content and quality of the perspective. Only 10 years ago the development com- discourse on the environment have been completely munity often brushed environmental concerns aside, transformed. By its sheer volume, the authoritative emphasizing instead the primacy of growth, stability,     and poverty reduction. Central to the discussion of the 118 years on record, even after the effects of El environmental sustainability at the start of the 21st Niño are filtered out.53 Satellite readings now con- century is the problem of how to devise mechanisms firm a similar elevation of temperatures in the upper that distribute the burdens of reform equally without atmosphere.54 Moreover, wintertime temperatures of discouraging the participation of every country that seawater north of 45º latitude have risen by 0.5° Cel- has the capacity to cause environmental damage. This sius since the 1980s. As a result, the incidence of sea challenge is particularly pressing because developing ice in the Grand Banks shipping lanes has declined, economies must sometimes balance environmental and in 1999, for the first time since the sinking of concerns with their people’s desire to advance eco- the Titanic in 1912, the International Ice Patrol did nomically. Chapter 4 examines the preconditions for not report a single iceberg south of 48º latitude. 55 international agreements that support environmental The concentration of carbon dioxide in the atmos- sustainability. Two areas in particular require con- phere has risen from 280 particles per million (ppm) certed international effort: climate change and biodi- in 1760 to 360 in 1990 and is expected to reach 600 versity loss. ppm in 2100. Average temperatures could then rise by some 2° Celsius.56 The source of the increase in Climate change carbon dioxide thus far, like the entire increase in at- Climate change is occurring at unprecedented rates mospheric chlorofluorocarbons responsible for de- because huge quantities of carbon dioxide, methane, pleting the ozone layer, is anthropogenic.57 and other greenhouse gases are being released into These facts are now widely accepted. Other infor- the atmosphere daily (figure 1.6). Global tempera- mation is less well understood: how severe fluctuations tures have been rising slowly since 1800. The 20th in weather will be in a warming world, how the effects century has been the warmest century in the past 600 of climate change on agriculture and living conditions years, and 14 of the warmest years since the 1860s will be distributed globally, the rapidity of the change, occurred in the 1980s and 1990s. Temperatures in and how people displaced by rising sea levels in coun- 1998 were higher than the mean temperatures for tries such as Bangladesh will be accommodated else- Figure 1.6 Temperatures are rising as concentrations of greenhouse gases increase Degrees Celsius Parts per million by volume 0.6 380 Temperature differences from average (left axis) 370 0.4 CO2 concentration (right axis) 360 350 0.2 340 330 0 320 310 –0.2 300 290 –0.4 280 270 –0.6 260 1856 1866 1876 1886 1896 1906 1916 1926 1936 1946 1956 1966 1976 1986 1996 Source: Carbon Dioxide Information Analysis Center (CDIAC).      ⁄  where. 58 These unknowns make precise measurement monitoring) countries with stronger legal structures of the economic impact of climate change very diffi- would wind up bearing inequitable burdens once cult. Nonetheless the impact will be huge, and it will treaties were ratified. be felt primarily in developing countries.59 The Rio Convention defined emissions levels for An international agreement to contain climate countries relative to their past history; thus, coun- change faces many hurdles. There is uncertainty about tries that polluted more were allowed to continue to the scale of possible benefits and the scope for adap- pollute more. For developing countries, this seemed tation. There is resistance to incurring the costs in- unjust: why allow industrial countries to emit higher volved in bringing about a drastic flattening of long- levels of emissions per capita (albeit lower levels of run trends. And there is concern about the difficulty emissions per unit GDP) simply because they had of monitoring compliance with emissions rules and done so historically? enforcing adherence through credible sanctions. Dis- While these equity issues were not effectively ad- tributional considerations also play a part. Industrial dressed at Kyoto, the Kyoto convention made an im- countries account for 60 percent of all energy-related portant step forward in trying to ensure efficiency carbon dioxide emissions, with the United States in reducing emissions, through the Clean Develop- alone responsible for 25 percent in 1998. But in the ment Mechanism. That scheme would allow indus- absence of corrective policies, developing countries trial countries to help developing countries lower will be emitting a higher proportion of all greenhouse their emissions, while granting the industrial coun- gases within 20 years, with China pulling ahead of the tries some “credit” for these lower emissions. Chap- United States by 2015.60 Until the early 1990s little ter 4 explores such issues and the possible course of progress was made in moving toward an agreement institution-building. that embraced the concerns of both developing and industrial countries. In particular, developing coun- Protecting biodiversity tries argued that because greenhouse gases were the The evidence on biodiversity loss is growing. The result of industrialization in wealthy countries, those United Nations Environment Programme (UNEP) countries must take responsibility for the ensuing estimates that about 22 million species exist at the problems. Developing countries also faced an uphill end of the 20th century. About 1.5 million have been battle to build the regulatory capacity to control the described. Some 7 million species, or more than four release of gases and other pollutants.61 This impasse times the number described, risk extinction in the has extended to other environmental issues, such as next 30 years. Among higher animals, three-fourths efforts to slow the loss of biodiversity. But some of the world’s bird species are declining, and some progress is being made on the international front to observers claim that nearly one-fourth of the world’s cope with climate change, with increased awareness mammal species are threatened with extinction.62 In of the long lags in arresting trends that are already ap- agriculture, crop plant varieties disappear every year, parent, such as the accumulation of greenhouse gases, but few of these crops are represented in world col- and a greater sensitivity to the risks they pose. lections of genetic material. Wild varieties are even There have been two international agreements to more poorly represented. Only 12 of the 38 base col- reduce emissions of greenhouse gases, in Rio in 1992 lections of rice listed in the International Board for and in Kyoto in 1997. In the first, industrial countries Plant Genetic Resources (IBPGR) directory of ge- undertook voluntary commitments to reduce their netic material include wild species, and only 5 col- level of emissions in 2000 to the level in 1990. The lections have long-term storage facilities.63 1997 agreement set more ambitious goals, and more The major causes of biodiversity loss are modern binding commitments (though it too lacked effective farming techniques, deforestation, and the destruc- enforcement measures). Developing countries have tion of wetland and ocean habitats, all of which are resisted entering into binding commitments, and closely linked to development activities. Of all the without their agreement several industrial countries, world’s countries, developing countries are richest in such as the United States, are reluctant to impose biodiversity, in part because many are in tropical binding commitments upon themselves. Moreover, climes. The pressure on these countries to protect bio- without adequate global enforcement (including diversity is severe. Just one-fifth of the earth’s original     forests remains in large, relatively natural ecosystems, Figure 1.7 or frontier forests. Seventy-six countries have lost all More countries are becoming democratic of their frontier forest, and 70 percent of what is left Democratic countries is found in just three countries—Brazil, Canada, and (percentage of total countries) Russia. Ninety percent of the remaining crop species 70 are in Africa, Asia, and Latin America.64 Despite these 60 worrying trends, however, international agreement on preserving biodiversity is only beginning. Nonethe- 50 less, the Convention on Biological Diversity and the Global Environment Facility are important first steps 40 in the process of preserving biodiversity. 30 New political tendencies in developing countries 20 Along with the wave of globalization in trade, fi- 10 nance, and environmental issues, another worldwide force is reshaping development efforts everywhere— 0 1974 1988 1993 1998 localization. Localization is the push to expand pop- ular participation in politics and to increase local Source: Diamond 1996; Freedom House, Freedom in the World, 1998. autonomy in decisionmaking. The impetus toward local autonomy stems in part from another global trend—urbanization. What are the main elements of the new local significance than elections. It will affect the process landscape? One is the replacement of authoritarian of institution-building and the types of policies that or single-party rule by plural politics and the increase are likely to be effective. With power decentralized in citizen participation through community groups and the central government less able to impose its and NGOs. Another is a growing demand for sub- own solutions, the demand for socially oriented poli- stantial power and autonomy on the part of subna- cies will grow. tional units. Central governments have responded to What will the push for increased citizen partici- this demand by devolving power and responsibility pation and plural politics mean for development? to local levels. Four changes are likely. First, vigorous political ac- Plural politics and broad-based popular participa- tivity involving many organized groups rooted in as- tion are rapidly becoming features of modern gover- sertive societies will substantially reduce the scope nance. The proportion of countries with some form for autonomous government action. The central of democratic government rose from 28 percent in government will have to engage and negotiate with 1974 to 61 percent in 1998. (figure 1.7). A majority society, field claims and pressures from diverse quar- of governments have made legally binding commit- ters, and seek legitimacy by winning public approval ments to respect the civil and political rights of their for its performance. There will be less room for close citizens. Thus far, 140 countries have ratified the In- business dealings, more calls for accountability, and ternational Covenant on Civil and Political Rights, a continuing move away from the authoritarianism and 42 have signed the optional protocol of the practiced in various parts of the world between the covenant, recognizing the authority of the United Na- 1960s and the 1980s. tions Human Rights Committee to consider claims However, this change has far-reaching implica- from those alleging violations of their rights. tions. Taking swift policy action to adjust to shocks As people’s participation in society grows, so does may become increasingly difficult. Efforts to pro- the number of organizations that give it voice. NGOs mote strategic sectors that also benefit special inter- and civic movements are on the rise, assuming an ests will come under close scrutiny—and not only ever-larger role in articulating people’s aspirations from the electorate. The business community, labor and pressuring governments to respond. This swell- unions, and others may serve as checks on the cen- ing of participation promises to be of even greater tral government’s authority. The Korean government      ⁄  learned of the power of these groups when it tried to goals by consensus. And more responsibility for de- push through a new labor code in 1997 and at- velopment may devolve to subnational governments. tempted to restructure large industrial conglomer- Fourth, participatory politics and limited interna- ates in 1998–99. tional labor mobility could increase calls for policies Recent events underscore the significance of the that address social dislocation.68 Until regional and new political landscape and the difficulties interna- global goverments begin coordinating policies to re- tional financial organizations face in negotiating duce the risks posed by shocks with potentially long- agreements that can have adverse consequences for term consequences, national governments will be re- segments of a country’s population, at least in the sponsible for buffering their populations against short term. The Russian and Brazilian parliaments extreme economic hardship. This situation will create refused to go along with international pledges made a number of dilemmas. In order to finance safety nets, by their presidents during the East Asian crisis. Their governments will have to adjust the composition of reaction prompted U.S. Treasury Secretary Robert public spending, possibly dampening growth in the Rubin to remark that when it comes to saving coun- short term. Attempts to mobilize additional resources tries from economic implosion, “the ultimate key is could meet with resistance from taxpayers who mis- not economics or finance, but politics.” Govern- trust the government’s ability to deliver services and ments must learn to muster support for strong poli- are accustomed to a culture of tax avoidance.69 Chap- cies, especially when those policies require sacrifices ters 5 and 6 review the preliminary evidence on mo- by present generations for the generations yet to bilizing financial and fiscal resources at the local level. come.65 What about the differences among countries? Second, political openness will highlight the dis- Economic and social instability of the kind that ex- closure of information and the emergence of the ists in Russia and Ukraine could dampen the desire private monitoring, regulatory, and information- for change. In Sub-Saharan Africa the small size of processing entities that are vital to a dynamic econ- the middle class, ethnic friction, and the region’s re- omy. These entities may include private mechanisms cent history of clientelistic politics hinder the spread for enforcing public laws like those suggested for the of pluralism and the pursuit of development objec- countries of the former Soviet Union.66 Authoritar- tives. Demographic pressure is testing African and ian regimes tend to become predatory and, except in Middle Eastern countries. In the next 20 years these rare cases, have not succeeded in creating efficient, countries will have to cope with large numbers of technocratic bureaucracies or in single-mindedly young people seeking jobs. Ethiopia’s population, for pursuing development. These failures are partly at- example, is likely to double to 120 million by 2030, tributable to the temptations offered by patronage, and already, more than half the people in Iran are political interference in the operations of public under the age of 25. To sustain economic growth, agencies, and the politicizing of bureaucrats at all political and social institutions will need to adapt levels. Plural politics and civic participation can re- rapidly to these changes. While the trend toward verse those forces, preventing the worst excesses of participatory politics is strong in the 1990s, the in- authoritarian systems. But good intentions are no stitutional reforms vital to future stability may not guarantee of rapid progress, as India’s situation illus- be keeping pace. trates.67 Whether opportunities exist for institu- Emerging subnational dynamics tional reforms that will encourage these reversals re- mains to be seen. As the 20th century draws to a close, people in sub- Third, participatory politics, by giving more voice national units such as provinces and states are de- to people, will hasten decentralization in some coun- manding the right to self-determination and self- tries. This trend is most likely to be felt in large coun- government. Such demands are part of the process tries and those with marked ethnic divisions and known as localization. They may originate in dissat- deeply rooted local identities. Strong urban regions isfaction with a central government, reluctance to could accelerate the redistribution of central author- subsidize other parts of a country, or conflict between ity to subnational entities, requiring the central gov- ethnic groups. Whatever its cause, localization gen- ernment to pursue major, long-term development erally results in the redistribution of power within a     country. It can, under certain circumstances (as in Table 1.2 southeastern Europe and Central Asia), lead to the Political and functional decentralization in large creation of new states.70 The number of countries has democracies, 1997 more than doubled in the last three decades, rising from 96 in 1960 to 192 in 1998. And the number of Decentralizing Decentralizing politically countries with fewer than 1 million people has almost politically only and functionally tripled, growing from 15 to 43. When accommo- Bangladesh Argentina Iran, Islamic Rep. Brazil dated in a democratic setting, localization involves a Kenya Colombia shift in the locus of decisionmaking, the structure Korea, Rep. of Ethiopia and quality of governance, and modes of policy im- Morocco Mexico Nepal Philippines plementation. For this reason it is expected to have a Nigeria Poland significant effect on the future of development. Pakistan a Russian Federation Romania South Africa Localization and decentralization Thailand Uganda Ukraine Governments have responded to demands for in- Venezuela creased self-government by sharing power with and Note: Sample includes all countries that had populations of 20 devolving authority to lower tiers of government. The million or more in 1997 and that introduced competitive multiparty elections at the subnational level between 1980 and 1995. action has been grudging at times, more forthcoming a. Local elections have not been held regularly in Pakistan, so local at others (especially when financially strapped central governments have mostly been run by administrators. Source: Freedom House, Freedom in the World, 1996; U.S. Central governments want to shed expenditure responsi- Intelligence Agency, The World Factbook, 1998; country-specific bilities). But the trend is clearly continuing, and the sources. numbers speak for themselves. In 1980 national elec- tions had taken place in 12 of the world’s 48 largest countries, and local elections had been held in 10 of number of very small countries such as Cape Verde, them. By 1998, 34 of these countries had held elec- Mauritius, and Swaziland. In Latin America, every tions at both the national and local levels. Half the country has elected mayors. Excluding Argentina, countries that decentralized politically also decentral- Brazil, Colombia, and Mexico (all large federations), ized major functional responsibilities (table 1.2).71 local governments account for 20 percent of govern- Poland has devolved responsibility for primary and ment expenditures in the countries for which data secondary education, for example, while the Philip- are available.74 pines has decentralized primary health care and local The end of the Cold War has been a key factor road maintenance. Decentralization often translates in the recent wave of decentralization. In the former into substantial increases in the subnational share of Soviet Union, dissolution of the party monopoly on public expenditure. In Mexico this share increased national political power has led subnational gov- from 11 percent in 1987 to 30 percent in 1996, and ernments to step up demands for increased local au- in South Africa from 21 to 50 percent.72 thority. In Eastern Europe the collapse of commu- Decentralization is not limited to large, wealthy nism removed the external military support that had countries. In the Middle East and North Africa, Jor- been propping up unpopular governments. Local dan, Lebanon, Morocco, and Tunisia all have elected governments rebounded—both in reaction to former local governments. In Europe and Central Asia, the regimes’ policy of forced centralization and as a bul- constitutions of Albania, Bosnia, Bulgaria, Croatia, wark against the return of authoritarianism. The de- Georgia, Hungary, Kazakhstan, Russia, Tajikistan, clining threat of a major international conflict, com- and Ukraine address the rights and responsibilities bined with increased openness to trade, has made the of subnational governments, although this does not advantages of being part of a large federation less at- automatically guarantee autonomy. The Baltics and tractive to smaller economies.75 the Kyrgyz Republic have also taken significant steps The end of the Cold War has had effects that are to strengthen local governments.73 In Africa, 25 of less direct but no less important in other regions. the 38 countries that held national elections in the In Latin America the declining threat of leftist vio- 1990s also had local elections—and that includes a lence (the initial grounds for military takeovers) has      ⁄  contributed to the demise of authoritarian regimes countable and efficient—and to lessen the risk that throughout most of the region. (In Peru, where the excessive borrowing and fiscal deficits at the subna- threat of leftist violence did not decline, most spend- tional level will cause economic instability.79 ing decisions continue to be made by the presidency.) Development will succeed in localized economies In Africa and parts of East Asia (Korea and the Philip- only if subnational units provide sound, effective pines), both the United States and Russia have ended governance. Good local governance gives people a their support for authoritarian governments. In a voice and incorporates rules that ensure the account- number of African countries, reduced external sup- ability of public employees. By providing citizens port combined with domestic economic collapse have with the opportunity to express their views, encour- undermined authoritarian governments’ ability to use aging them to monitor the workings of local govern- public spending to maintain the support of key inter- ment, and calling on them to participate, an effec- est groups. Ruling elites have been forced to concede tive system of governance creates a willingness to some power, though often they have relinquished just obey laws and pay taxes.80 Over time, sound gover- enough to permit them to retain their hold.76 nance builds trust and social capital. Whether in The pace of decentralization and the kinds of re- Tanzania or northern Italy, well-governed commu- forms that have been implemented vary from coun- nities are rich in social capital and adequately fur- try to country. India, a multiparty democracy at the nished with revenues. federal and state levels, has relatively weak local gov- In many cases local governments need to develop ernments (see box 5.4). China is still officially a cen- the skills and resources to provide high-quality services tralized state with a dominant party, though provin- that meet constituents’ demands.81 These services cial governors and mayors have had considerably (along with adequate fiscal and financial resources) are more autonomy and managerial authority since the what allow cities to perform well economically and to reforms initiated in 1978 (see box 5.5). In a number offer their residents a good standard of living. In most of countries decentralization has not resulted in the cases the private sector will need to be involved in center relinquishing much control.77 Ghana, Malawi, providing services, especially in countries with weak and Zambia have each created local councils, but the access to international capital resources.82 But encour- central government continues to direct almost all aging the private sector to participate requires govern- spending and management decisions. Similarly the ments to have in place regulatory structures that min- ruling national party in Tanzania holds almost all imize transactions costs and barriers to entry but also subnational offices. Pakistan convenes local elections, ensure that private providers deliver with respect to but so infrequently that the country has only rarely quality, cost, and volume. The good news is that in- had sitting, elected local governments.78 creasing administrative and regulatory capacity and improving governance will allow subnational units to Institutional challenges reap the benefits of market-led growth. Localization raises complex institutional and policy Urban imperatives issues that governments will have to resolve in com- ing decades. The expanded power of subnational At the beginning of the 21st century, half the world’s governments has implications not only for growth population will be living in areas classified as urban. and macroeconomic stability but also for gover- As recently as 1975 this share was just over a third, nance, coordination, and regulation. It will affect but by 2025 it will rise to almost two-thirds. The (and be affected by) the availability of international most rapid changes in urban demographics will financing, the delivery of public services, the man- occur in developing countries (figure 1.8). While the agement of social safety nets, and the reduced ability rate of urbanization has passed its peak in relatively to redistribute that could result in increasing inequal- high-income countries in Latin America, Eastern ity. Rules must be designed that apportion responsi- Europe, and the Middle East, the transition is just bility, manage relations among tiers of government, beginning in Asia and Africa (figure 1.9). Urban and strike an acceptable trade-off between central populations are expected to increase by almost 1.5 and local authority. The types of rules discussed in billion people in the next 20 years (figure 1.10). The chapter 5 are intended to help keep governments ac- speed of urbanization and the enormous numbers     Figure 1.8 Most urban dwellers reside in developing countries Urban population (billions) 5 Industrial 4 countries Developing countries 3 World 2 1 0 1950 1960 1970 1980 1990 2000 2010 2020 Source: UNDIESA, World Urbanization Prospects, 1998. involved make it one of the major development chal- The East Asian experience with sustained eco- lenges of the 21st century. nomic growth and successful rural development sug- Despite the challenges it presents, urbanization gests that the pattern can be repeated. Korea took should be a positive trend. In industrial countries eco- just 40 years to transform itself from a society that nomic growth and structural transformation accom- was 80 percent rural to one that is 80 percent urban. panied urbanization. As agricultural sectors modern- ized, mechanized, and became more efficient, the Figure 1.9 number of agricultural jobs declined. Workers went Asia and Africa are just beginning looking for jobs in nonagricultural industries, which the urban transition are generally located in areas with much higher pop- ulation densities than farming communities—that is, Urban population growth, 1996 (percent) in cities. Wealthy societies in the late 20th century are 6 four-fifths urban and derive less than 3 percent of Sub-Saharan Africa their GDP from agriculture, while in low-income 5 countries agriculture still accounts for 30 percent or East Asia and Pacific more of GDP. 4 South Asia Middle East Can this pattern repeat itself in developing coun- and Latin America 3 North Africa tries, given that urbanization and economic growth and the Caribbean in industrial countries took place over a fairly long 2 period and involved much smaller numbers of peo- Europe and ple? These economies were pioneers, their growth un- 1 Central Asia constrained by external standards and codes pertain- Industrial countries ing to labor, human rights, or the environment. The 0 20 30 40 50 60 70 80 conditions in which developing countries are at- Urban population (percentage of total) tempting the transition to urban societies are now dramatically different and, with continuing advances Source: World Bank, World Development Indicators, 1998. in communications technology, will continue to alter.      ⁄  Figure 1.10 The largest increase in urban populations during 1980–2020 will occur in Africa and Asia Urban population (billions) 1.2 1980 2000 1.0 2020 0.8 0.6 0.4 0.2 0.0 Middle East Europe Sub-Saharan Latin South Asia East Asia and and Africa America and North Africa Central Asia and the Pacific Caribbean Source: UNDIESA, World Urbanization Prospects, 1996. As it did, the share of agriculture in GDP fell from a can find. More than 700,000 people sleep on Mum- high of 37 percent in 1965 to 6 percent in 1996. But bai’s pavements.84 The growth of favelas in Rio de elsewhere, the link between urbanization and growth Janeiro and São Paulo is typical of what is happening has been less obvious. Famines, civil wars, ethnic in other large cities in the developing world.85 Increas- conflicts, stagnant agriculture, the absence of rural ing violence is linked to the growing inequality evi- development, or merely the removal of constraints dent in urban areas, most notably in Latin America, on mobility can push people to urban areas, which but also in South Asia.86 Immigration may exacerbate may lack the resources to absorb them productively. this trend.87 The well off live in fortified enclaves, Regions such as Sub-Saharan Africa and South Asia, abandoning entire neighborhoods to the poor—an where per capita GDP growth has been slow or flat, increasingly common characteristic of cities where the have had the most difficulty in absorbing urban new- decline in public services and life chances has created comers. The relationship between economic growth an increasingly differentiated urban environment.88 and urbanization is covered in detail in chapter 6. In many respects these patterns are a replay of the Urban centers are expected to offer better access decline in urban living conditions that occurred in than rural areas to such essentials as water and sew- Western Europe during the rapid industrialization erage and to health care and educational services. of the first half of the 19th century.89 The second The quality of urban living conditions has tradition- half of the century witnessed a remarkable turn- ally been reflected in reduced morbidity and infant around. Can historically similar reform strategies mortality rates and increased life expectancies. But help reverse the trends in developing countries, or since the mid-1980s the advantages urban areas (es- will urban decay become a permanent feature of mu- pecially big cities) have enjoyed have been declining. nicipal areas? In Sub-Saharan Africa mortality rates are nearly the The development community has long been aware same in rural areas and small cities—90 per 1,000— of the challenges rapid urbanization poses.90 But de- and rates in large Latin American cities have risen to centralization, globalization, and industrialization those of smaller urban areas.83 will heap new challenges on the old.91 Some of the Access to shelter in urban areas is, if anything, most important issues for the 21st century are sub- worsening. About 100 million people—including national borrowing for public infrastructure, the co- large numbers of children—have no permanent home ordination of interregional infrastructure, and the lo- and simply make use of whatever urban spaces they cation of lumpy investments. Competition for global     capital is another issue local authorities will have to patterns that encourage reliance on private automo- contend with, learning, as they do, that probusiness biles. Well-designed urban transportation systems policies must not take precedence over social welfare. not only affect land use but also improve growth National policies that inhibit the mobility of capital prospects by better integrating the urban labor mar- and labor are another significant issue. Governments ket. 95 Quality of life, which is often measured by will need to consider revising those policies in order the availability and efficiency of public services, is to promote efficient industrialization that allows also a major issue. 96 Inadequate investment in sew- firms in mature industries to relocate from large to ers and sanitation systems can create serious health small cities. problems. Weak land market institutions that fail to During the three decades of development be- clarify and strengthen tenure and ownership rights tween 1960 and 1990, the concentration of eco- can hurt the quality of shelter. Karachi, Pakistan, and nomic activities in urban areas coincided with the Lagos, Nigeria, which are experiencing a decline in rise of per capita GDP. But this trend was merely the volume and quality of infrastructural and social noted and did not leave a mark on policies or insti- services, are typical of cities where the public sector tutional design.92 Now that globalization and decen- is on the verge of collapse. These problems require tralization are reshaping geographic as well as eco- innovative solutions, including public-private part- nomic landscapes, the relationship between growth nerships, selective privatization of local functions, and urbanization can no longer be ignored. and community involvement in regulation. Rapid urbanization also has social and political im- Cities also face new industrial challenges. Ex- plications. The institutions, social capital, and poli- panded export opportunities and the emergence of tics that served a stable, dispersed rural population do “industry clusters” require careful planning to pro- not transfer well to cities. Much social capital is lost vide the necessary infrastructure and the skilled and needs to be replaced, reconstituted, and aug- workforce modern high-technology industries re- mented. The moral economy of a hierarchical rural quire.97 The poor quality of urban training facilities society, which provided a measure of insurance against has hindered the growth of such industries in Latin risks, needs to be replaced by urban safety nets, both America. But skills are only one part of the equation. formal and informal.93 Middle classes emerge and Industrialization in Kerala, which has the most edu- expand in cities and are subject to demonstration ef- cated workforce in India, is inhibited by labor mili- fects from industrial countries.94 Second- and third- tancy, land tenure constraints, and power shortages.98 generation urban residents often begin to organize Institutional and regulatory reform in domestic fi- and voice their demands with more assertiveness. nancial markets must reflect not just national and Governments need new political and social mecha- global imperatives but urban infrastructure and hous- nisms in order to meet rising expectations. ing investment requirements as well. Newly empow- ered state and local governments must find the means Implications for development policy to finance these investments.99 At the same time, they If governments do not establish the policies and in- will need to develop new measures to cope with un- stitutions needed to manage urbanization and pro- employment, poverty, and inequality. Avoiding urban vide complementary infrastructure, urban areas could poverty traps will require building industrial skills experience slow economic growth and social unrest, and creating a competitive urban economy. Skilled and valuable resources will be wasted. Mismanaged workers could move in search of employment, an im- cities with inadequate resources and ineffective polit- portant consideration because mobility is sometimes ical processes are unattractive to new industries. They the only recourse for workers in towns where the in- cannot raise the quality of life, and they do not build dustrial base is narrow and has been declining. Many human capital or attract fresh talent. In addition, un- urban areas in China, Eastern Europe, and the for- less governments provide a level playing field, small mer Soviet Union, for example, are saddled with fail- and medium-size cities cannot compete effectively ing industries and rising unemployment. with their larger urban counterparts for manufactur- Cities that want to compete for foreign direct in- ing activity. vestment need to meet world production standards. The absence of appropriate regulation in rapidly Being second best can also mean being saddled with growing urban areas can create inefficient land-use high unemployment and increasing poverty.100 The      ⁄  ease of international sourcing in a competitive mi- wide-ranging effects of demographic change and the lieu with many specialized suppliers has changed movement of populations across countries and to labor markets. Because tasks can be combined flexi- urban areas. It is also a world that confronts an bly, urban production centers worldwide are increas- ongoing shift in the climate and loss of biodiversity. ingly segmented between those that can meet the ex- These forces are modifying the roles and obligations acting standards of a global production system and of national governments. However, developing coun- those with too few skills to do so. This open produc- tries need not and should not be passive respondents tion environment mercilessly weeds out those cen- to these forces. The discussion in the chapters that fol- ters with below-par macroeconomic environments, low, along with the specific case studies presented in services, and labor-market flexibility. chapter 8, suggest a number of strong institutional Urban dwellers also need formal safety nets. In measures that can be taken to benefit from the oppor- rural villages kinship ties or patron-client relations tunities offered by these trends, and to limit the risks. often performed this function, but as people move to In this process, unilateral responses by national gov- cities, their ties to their home villages weaken. Provid- ernments will not suffice. Instead, there must be a con- ing basic services and some income insurance will be tinuing interplay of commitments and responsibilities. a priority for governments in the 21st century. Local Subnational governments will take on responsibilities authorities and communities may be required to take but will be monitored by the central government. The the initiative in helping build safety nets. But success central government will make commitments, which is predicated on organizational capacity, accountabil- will be monitored by both the subnational govern- ity, and trust. As with the Friendly Societies in Britain ments and relevant international organizations. Mar- in the early 20th century, private and community ket forces will play a central and vital role, both in provision can play a significant role, but only with ad- providing the engine of economic growth and in re- equate government support. Community efforts, sponding to the incentives and constraints decided abetted by responsive local governments, can supple- upon by different levels of government. Perhaps most ment government resources in other areas as well.101 important of all, the citizens of developing countries Developing countries enter the 21st century in a will partner with governments and nongovernmental world that is being transformed by the forces of glob- organizations and work through open and participa- alization and localization. It is a world subject to the tory institutions to shape their own future. C h a 2 p t e r The World Trading System: The Road Ahead M any aspects of globalization have captured trial to developing countries offers ample worldwide attention in the 1990s, in- opportunity to expand trade not only in cluding capital flows, migration, and en- goods, but also in services, which are be- vironmental issues. But for more than a coming increasingly tradable. In a few century, the driving force behind global- decades global trade in services may well ization has been the expansion of trade in exceed that in goods. goods and services. And throughout the Third, trade is intertwined with an- early decades of the 21st century, trade other element of globalization: the spread will continue to drive global integration, of international production networks. especially among developing countries. These networks break up sequential pro- Trade is important to developing duction processes, which traditionally countries for four reasons. First, it is have been organized in one location, and frequently the primary means of realiz- spread them across national borders. This ing the benefits of globalization. Coun- dynamic will result in further geographic tries win when they gain market access dispersion of production and increased for their exports and new technology trade among cities, regions, and coun- through international transfers, and tries. Increasingly, the fortunes of the when heightened competitive pressure new production venues are bound to- improves the allocation of resources. gether by trade. The rising share of imports and exports Fourth, the growth of trade is firmly in gross domestic product (GDP) for buttressed by international institutions Latin American and Southeast Asian of long standing. The World Trade Or- countries in 1980–97 attests to a grow- ganization (WTO), built on the legacy ing exposure to international trade (fig- of the General Agreement on Tariffs and ure 2.1). African economies have also Trade (GATT), is the latest step in creat- felt the effects of international trade for ing a commercial environment more some time. Although the continent’s conducive to the multilateral exchange share declined during the 1980s, it fell of goods and services. 2 The GATT and from a high starting point.1 WTO have served as the means of secur- Second, the continuing reallocation ing past gains through multilateral trade of manufacturing activities from indus- liberalization. But more important, the       ⁄  Figure 2.1 prices change, resources are reallocated to production Foreign trade has increased in most developing activities that raise national incomes. The tariff reduc- regions since 1970 tions implemented after the Uruguay Round raised national incomes by 0.3–0.4 percent.3 Second, much Imports and exports of goods and services (percentage of GDP) larger benefits accrue in the long run as economies ad- 90 just to technological innovations, new production struc- 80 1970s tures, and new patterns of competition. These gains will 70 1980s continue to be as important in the future as they have 1990s been in the past. 60 Trade liberalization has other powerful effects. First, 50 it strongly influences the way firms perform. The evi- 40 dence of its effects on domestic enterprises highlights 30 the benefits developing economies gain from access to 20 world markets. 10 Ⅲ Increased imports have been found to discipline do- 0 East Asia Latin South Sub-Saharan Middle mestic firms in Côte d’Ivoire, India, and Turkey by and Pacific America Asia Africa East and forcing incumbent firms to bring prices closer to and the North Africa Caribbean marginal costs, thereby reducing the distortions cre- ated by monopoly power.4 Note: Data are averages over each decade. Source: World Bank, World Development Indicators, 1999. Ⅲ Trade liberalization can permanently raise the produc- tivity of firms by providing access to up-to-date capi- tal equipment and high-quality intermediate inputs at WTO can function as the point of departure for future relatively low prices. Some firms in the Republic of rule-making to promote still greater openness to trade. Korea and Taiwan (China), for instance, raised produc- If trade is to continue expanding as rapidly as it has in tivity by diversifying their use of intermediate inputs.5 the past, and if it is to be of greater benefit to develop- Ⅲ Firms’ productivity levels also rise when businesses are ing countries, the international community must en- exposed to demanding international clients and the gage in further liberalization and institutional reforms. “best practices” of overseas competitors. Domestic This chapter starts by outlining how the global trad- firms may also benefit from the opportunity to re- ing system benefits developing countries, and review- engineer foreign firms’ products. Indeed, the dif- ing the impressive record of trade liberalization during ferences in the productivity levels of exporting and the last 15 years. However, the lack of attention given nonexporting firms often diminish once previously to the social consequences of reform has threatened a nonexporting firms begin selling products abroad, as backlash against trade, which has the potential to stall studies from Colombia, Mexico, Morocco, and Tai- this momentum toward reform. The chapter then de- wan (China) show.6 scribes how further trade liberalization in two sectors— agriculture and services—can especially benefit de- Second, trade liberalization can set off a chain of veloping countries. The rise of global production events that concentrates economic activity in a city or re- networks and cities will also have profound implica- gion.7 When costs fall as output rises, businesses have an tions for the world trading system—broadening partic- incentive to locate production activities in a few loca- ipation in the system and fusing its participants closer tions, laying the groundwork for “agglomerations” of together. The chapter ends by analyzing how the pace economic activity. As demand from overseas purchasers of and support for liberalized trade in developing coun- boosts output in these locations, average costs fall and tries will be affected by these developments. profits rise. The rising profits attract new firms that pro- duce similar goods and thus provide a new source of ag- How the global trading system benefits glomeration. The increase in final goods producers then developing countries encourages the entry of new intermediate input produc- Trade liberalization benefits economies in two impor- ers with products (such as nontradable services) tailored tant ways. First, when tariffs are lowered and relative specifically to the needs of the final goods producers. The    :     new inputs make the production of final goods yet more Figure 2.2 efficient, lowering costs and raising quality (and possibly More of the world’s exports are covered by WTO disciplines, especially exports revenues). Final goods production becomes still more from developing countries profitable, attracting more producers. The cycle contin- ues until it is curtailed by congestion—that is, when out- Share in total world exports put grows faster than the capacity of local infrastructure. These cumulative processes lead to the higher produc- 1982 11% tivity that characterizes urban areas (see chapter 6).8 23% WTO mechanisms for promoting and maintaining liberal trade regimes The international trading system owes its robust devel- opment to successful institutions that straddle interna- Developing country tional and national levels—for many decades the GATT members of GATT/WTO and now its successor, the WTO. An effective WTO 66% serves the interests of developing countries in four ways: Industrial country members of 1997 Ⅲ It facilitates trade reform. GATT/WTO 10% Ⅲ It provides a mechanism for settling disputes. 19% Nonmembers Ⅲ It strengthens the credibility of trade reforms. Ⅲ It promotes transparent trade regimes that lower transactions costs. These benefits explain the willingness of developing countries to join the WTO in increasing numbers. In 1987, 65 developing countries were GATT members.9 In 1999, 110 non-OECD countries were members of 71% the WTO, accounting for approximately 20 percent of world exports (figure 2.2).10 Source: WTO, Annual Report, 1997. Facilitating trade reform Countries benefit from unilateral reductions in their eralization. Growing numbers of industrial and devel- own barriers to imports. But in a classic dilemma for oping countries are signing regional trading arrange- policy reform, the costs of unilateral trade liberalization ments (RTAs), often, but not always, with neighboring are concentrated among a few import-competing inter- countries. Regional agreements have proliferated since ests, while the benefits are distributed thinly across 1990, covering not only trade in goods but also trade in many consumers. The would-be beneficiaries of trade services, investment regimes, and regulatory practices liberalization have little incentive to lobby against the (figure 2.3). This regionally based liberalization has in- opponents. The WTO exists to overcome this prob- creased intraregional trade and investment flows.11 In lem—that is, to facilitate trade reform by changing the some cases the regional concentration of trade has be- political equation to generate support for multilateral come pronounced. In 1992 trade among the members trade agreements. These agreements create a set of con- of the Andean Community—Bolivia, Colombia, Ecua- centrated “winners” in member states—the exporting dor, Peru, and Venezuela—was 2.7 times higher than firms, which benefit from lower tariffs in potential ex- their economies’ national incomes and geographic sepa- port markets, and which therefore have an incentive to ration would typically generate (box 2.1).12 oppose import-competing firms. To maximize the num- ber of winners, multilateral trade negotiations tend to Encouraging countries to resolve their disputes cover many sectors and countries. through negotiation Multilateral trade negotiations are not the only The dispute settlement mechanism of the WTO bene- means of tilting the political balance to favor trade lib- fits developing economies.13 Initially, members of the      ⁄  Figure 2.3 WTO undertake to settle disputes bilaterally. But if this More regional trading arrangements (RTAs) process fails, a dispute can be referred to an interna- came into force in the 1990s than ever before tional panel for adjudication. If the panel votes to up- Number of RTAs entering into force hold the complaint, it can recommend that the offend- ing measure be removed.14 If the country against which 90 82 the complaint has been lodged does not comply with 80 the panel’s ruling, the complainant can apply for per- 70 mission to retaliate by withdrawing trade concessions. In principle, the dispute settlement mechanism 60 makes it easier to enforce the numerous trade agree- 50 ments that fall under the WTO umbrella. But due to 39 the costs and expertise required to mount a case, and 40 the limited leverage gained by shutting a trade partner 30 out of a small market, more often than not the dispute 20 19 settlement mechanism is of greatest benefit either to 14 large developing countries or to several small countries 10 3 acting in concert. Still, in certain areas the mechanism 0 particularly benefits developing economies. For in- 1950–59 1960–69 1970–79 1980–89 1990–98 stance, many of the liberalizing measures affecting the Source: World Bank 1999g. textile trade that were agreed to during the Uruguay Round will be implemented in the first decade of the Box 2.1 Regional trading arrangements and the global trading system: complements or substitutes? The growing popularity of regional trading arrangements the global trading system. One response is to pursue further (RTAs) has ignited concerns that these agreements may un- multilateral trade liberalization to limit the margin of preference dermine the global trading system by discriminating against regional agreements create. Policymakers who believe that imports and investments from nonmembers. Critics of regional their country is suffering because of the rise of RTAs else- arrangements argue that this practice would violate a core prin- where thus have a further incentive to support multilateral ciple of the World Trade Organization (WTO): that all imports trade liberalization. from member states should face the same barriers to trade. A second response is to alter the WTO’s agreement on re- Furthermore, eliminating tariffs on imported goods from some gional trading arrangements to commit members to phase out countries but not others can be counterproductive. If imports any preferential market access within a certain time frame. from high-cost producers inside the agreement replace goods Such a provision ensures that preferential market access is from low-cost producers outside the agreement, the import- only a temporary feature of any regional initiative. To make this ing country will not only lose tariff revenue but will wind up approach more attractive to members of a regional initiative, with imports that cost nearly as much as before. they could be offered credit for the reduction in trade barriers, Supporters of RTAs maintain that these agreements have which could be used in future multilateral trade negotiations. enabled countries to liberalize trade and investment barriers to A third response is to negotiate a “model accession a far greater degree than multilateral trade negotiations allow. clause” for the principal types of RTAs. Such clauses contain Proponents also argue that regional agreements have gone a set of conditions nonmembers must meet in order to be- beyond trade liberalization, taking important steps toward har- come members. Meeting the conditions automatically triggers monizing regulations, adopting minimum standards for regula- a negotiation for accession to the regional agreement. These tions, and recognizing other countries’ standards and prac- clauses could also ensure that the trade barriers nonmembers tices—trends that enhance market access. Some empirical face do not rise when an RTA is established or when new evidence supports each view. Thus, a recent survey concluded members are admitted. that regional arrangements “seem to have generated welfare gains for participants, with small, possibly negative spillovers Source: Baldwin and Venables 1995; Bhagwati 1991; Fernan- onto the rest of the world.”15 dez and Portes 1998; Frankel 1997; Panagariya 1999; Pana- Should future research suggest that RTAs are having ad- gariya and Srinivasan 1997; Primo Braga, Safadi, and Yeats verse effects on the world trading system, the arrangements 1994; Schiff and Winters 1998; Serra and others 1998; Wei and will have to be aligned with the nondiscrimination principle of Frankel 1996; World Bank 1999g; Yeats 1996.    :     21st century.16 In this case the dispute resolution mech- those negotiations begin. The advantage of these in- anism can play a significant role in ensuring that devel- ducements was apparent in the Uruguay Round negoti- oping countries are still able to expand their textile ex- ations, when credit was given informally for such bind- ports. The dispute settlement mechanism can also be ings. Developing economies that bound substantial used to protect developing countries from the imposi- unilateral reforms received $1.50 of tariff concessions tion of banned market-closing measures, such as pres- for every $1 they offered, significantly more than the sure to agree to “voluntary” restraints on their exports, $1.10 received by countries that had not undertaken or the improper use of permitted market-closing mea- unilateral reforms.21 Codifying this informal system sures, such as the use of sanitary standards as a barrier would reduce uncertainty about the benefits of using to trade rather than a protection for public health.17 this commitment mechanism. Reinforcing the credibility of trade liberalization Promoting transparent trade policy regimes Countries that have a history of import substitution The WTO’s Trade Policy Review Mechanism, created in policies—that is, of imposing barriers to imports with 1989, is designed to enhance the transparency of trade the intention of producing the same goods domesti- policy regimes worldwide. Depending on a country’s cally—may want to signal that they have switched to a share of world trade, its trade policy regime is reviewed more liberal trade policy. In this case the WTO’s tariff- every two, four, or six years. Representatives from mem- binding option may prove particularly useful.18 A WTO ber states discuss the results of these reviews in a forum member can unilaterally reduce its trade barriers to some that provides a nonconfrontational atmosphere for dis- new level and then promise that future trade barriers to cussing trade practices.22 This process reduces the incen- imports from all other WTO members will be no higher tive for governments to adopt and retain trade policy than this new, lower level. This promise, known as a measures that contravene international rules, especially “binding,” is incorporated into the country’s obligations those countries with the largest shares of world trade. at the WTO. Binding reinforces the political will to Such mechanisms not only nudge governments to com- maintain a more liberal trade policy, even in the face of ply with WTO commitments but also lower tensions attempts by import-competing firms to reverse the re- among members. forms. If a country reneges on its obligations, WTO rules require that it offer compensation to trading part- Building technical capacity in trade matters ners whose interests have been adversely affected.19 in least-developed countries In the past 15 years, largely because of the environ- The growing number and complexity of the issues nego- ment created by the GATT and WTO, many develop- tiated at the WTO have prompted questions about the ing economies have unilaterally reduced their trade bar- adequacy of the technical expertise available to develop- riers. The trend toward outward-oriented trade policies ing countries in their national capitals and at their mis- is not confined to any one continent or region, and it sions in Geneva.23 In 1997 industrial countries deployed predates the completion of the Uruguay Round (figure an average of 6.8 officials to follow WTO activities in 2.4). For example, between 1988 and 1992 Kenya re- Geneva. Developing countries sent an average of 3.5 (fig- duced its average tariff rate from 41.7 to 33.6 percent. ure 2.5). Because they are not as well represented, devel- The credibility of such unilateral trade reforms plays a oping countries may have difficulty negotiating the most crucial role in their success. The private sector and inter- favorable trade agreements and using the dispute settle- national investors react less favorably to an announced ment mechanism effectively. To tackle this problem, the trade liberalization if they believe that the reforms are World Bank, in conjunction with other multilateral in- likely to be reversed at the first sign of import surges, cur- stitutions, has developed the Integrated Framework for rent account difficulties, or recession. Trade and Development in the Least-Developed Coun- Only a few countries have bound their unilateral tries. The aim of the framework is to prepare developing trade reforms, typically during a subsequent multilat- countries to participate effectively in the WTO (box 2.2). eral trade round.20 An additional incentive for binding Sustaining the momentum for trade reform unilateral reforms might be to give explicit credit in sub- sequent multilateral trade negotiations to developing The successful completion of the Uruguay Round of countries that “bind” their unilateral reforms before multilateral trade negotiations and the growing popu-      ⁄  Figure 2.4 Many developing countries started liberalizing before the end of the Uruguay Round Average tariff rate (percent) 50 1986b 45 1991–92 1980 1992 1988 40 35 1992 1993–94 1989 30 1985 1992 25 1988 1986 1994 1992c 1985 20 15 1989 1992 1993 1991 1992 10 5 0 Hungarya Bangladesh Indonesia Philippines Egypt, Kenya South Argentina Chile Mexico Arab Rep. of Africa Note: This figure focuses on pre-1994 tariff rates because reductions in the average tariff rate after 1994 could be due to the implementation of the Uruguay Round. a. Data are from the European Commission’s Market Access database. b. Trade-weighted average. c. Manufacturing-sector average. Source: Drabek and Laird 1998. larity of RTAs have created considerable momentum ports has changed, too, creating increased competition for integrating countries further into the global trading in manufactured products, especially in medium- and system. Policymakers in developing and industrial high-technology goods. For example, the share of high- countries now confront the task of maintaining this technology products exported by East Asian economies momentum. Concerns about the effects of trade have increased substantially between 1985 and 1996. Mean- received much attention in recent years, including wor- while, Latin American countries and India have shifted ries over inequality, poverty, the environment, and the their exports from resource-based manufactures to financing of social safety nets.24 Even though the em- low- and medium-technology exports (figure 2.6). The pirical evidence almost always fails to validate these quality of exports from the Czech Republic, Hungary, concerns, policymakers have become increasingly sen- Poland, and the former Yugoslavia in engineering, cloth- sitive to them. ing, textiles, and footwear products has also improved in the 1990s.26 Recent concerns about the pace of trade reform These heightened competitive pressures enhance Developing countries are indeed exporting more to their overall national welfare, but they are not well received industrial counterparts. As early as 1990, many indus- by import-competing firms. These firms are already trial countries had seen substantial increases in the ratio leading a reaction against trade liberalization in both of their merchandise imports to merchandise output, developing and industrial countries. In addition to lob- leading to even greater competition for sales in their bying policymakers, import-competing firms use an- markets.25 The composition of developing countries’ ex- tidumping laws—which are still permitted by WTO    :     Figure 2.5 Equal players? African representatives at the WTO IBRD 30413 Tu n i s i a Morocco Algeria For mer Libya Arab Rep. Spanish of Egypt Sahara Mauritania The Mali Niger Gambia Chad Eritrea Senegal Sudan Burkina Faso Guinea- Guinea Benin Bissau Nigeria Sierra Leone Côte Ethiopia d ’ I v o i r e G h a n a To g o Central African Liberia Republic Cameroon Somalia Equatorial Guinea S ã o To m é a n d P r í n c i p e Uganda Kenya Congo Gabon Dem. Rep. Rwanda of Congo Burundi Ta n z a n i a Seychelles Comoros Angola Mayotte (Fr) Malawi Number of trade officials in Zambia Geneva, location of the WTO: Mozambique Madagascar 6 and over Zimbabwe 4 to 5 Réunion Namibia Botswana (Fr) 1 to 3 0 Nonmember (1997) Swaziland The average number of trade of ficials South from OECD countries is 6.8. Africa Lesotho JULY 1999 Note: Data are for 1997. Mauritius, which is not pictured, had four officials in Geneva.      ⁄  Box 2.2 Building technical expertise on trade policy: the Integrated Framework for Trade and Development in the Least-Developed Countries The Integrated Framework for Trade and Development in the needs to produce multiyear programs of trade-related assis- Least-Developed Countries, a partnership among multilateral tance that will be presented at donor consultations on trade agencies and least-developed countries, provides assistance matters. For each participating country, this consultation will in integrating these countries into the global economy. The produce concrete pledges constituting a firm program of trade- framework was initiated by the 1996 World Trade Organization related assistance. (WTO) Ministerial Declaration, which asked WTO member Of the 48 least-developed countries, 40 have already countries to provide enhanced market access for the least- presented their needs assessments. Uganda has already im- developed countries. The declaration also requested that the plemented its program of trade-related assistance, and 16 multilateral institutions involved—the WTO, the World Bank, other countries have been preparing similar programs for the International Monetary Fund, the United Nations Develop- a 1999 donor consultation on trade matters. The discussion ment Programme, the United Nations Conference on Trade of the multiyear program at the Consultative Group meet- and Development, and the International Trade Center—provide ing in Kampala in December 1998 raised the profile of the an integrated framework for trade-related assistance. Integrated Framework. Several donors are prepared to sup- The framework includes initiatives to build infrastructure, port aspects of the program, including the U.S. Agency for streamline the business environment, ensure the efficiency International Development and the U.K.’s Department for and transparency of customs administration, increase govern- International Development. Country teams from multilateral ments’ capacity to develop effective trade policies, and en- agencies assist the least-developed countries whenever hance the private sector’s ability to identify and operate in requested. export markets. The framework also aims to enhance least- Uganda’s experience demonstrates just how much this developed countries’ participation in the WTO so that they can framework can contribute to a developing country. Uganda take a more active role in the day-to-day workings of the orga- presented its multiyear program of trade-related assistance at nization and help set the agenda for the next round of multilat- the Consultative Group in 1998. The World Bank’s resident eral negotiations. mission in Uganda created the operational process for the pro- In establishing the framework, the WTO invited each least- gram, using existing sector investment projects in education, developed country to submit a needs assessment for trade- health, and roads. A steering committee led by the trade min- related assistance, including for physical infrastructure, human istry reviewed the needs assessment, ranking items accord- resource development, and institutional capacity building. In ing to the country’s general priorities. The presence of donors their assessments of the major obstacles to trade expansion, and private sector representatives on the steering committee most countries identified supply-side constraints and a lack facilitated a consensus and ensured full financing of the priori- of technical capacity. The countries will update and rank their ties the program had identified. rules—to allege unfair trade practices by foreign com- dustrial countries (figure 2.8). Antidumping actions are petitors. A good is said to be dumped if its export price becoming a widespread phenomenon, diluting market is less than either the price in its home market or the access and the gains from trade liberalization.28 average cost of production. Antidumping laws enable The reaction against increased competition from countries to impose offsetting duties on the products of imports is not limited to antidumping suits. Concerns foreign firms found to be both dumping products on have been raised that rising import competition is ad- the domestic market and causing “material injury” to a versely affecting labor market outcomes and, in partic- domestic industry.27 ular, causing the widening income inequality observed Until the early 1990s the main users of these laws in some industrial economies. 29 These concerns have were Australia, Canada, the European Community (as led to calls to slow, halt, or even reverse trade liberal- it was then), New Zealand, and the United States. How- ization in industrial economies—actions that would di- ever, these countries have been joined by a number of rectly affect the number and size of export markets new users, primarily developing economies such as Ar- open to developing countries. gentina, Brazil, India, Korea, Mexico, and South Africa The link between increases in imports and rising in- (table 2.1). In the late 1980s developing countries initi- come inequality is highly controversial. With a few ex- ated less than 20 percent of all antidumping actions. By ceptions, empirical research has found that imports the late 1990s they accounted for around 50 percent from developing countries have relatively limited effects (figure 2.7). Developing countries have also become the on wages and employment in industrial countries.30 targets of antidumping actions at close to the rate of in- This research does not deny that income inequality is    :     Figure 2.6 increasing, but it does suggest that, because increased The composition of many developing countries’ trade is not a primary cause, erecting new trade barri- exports was transformed in just over 10 years ers is unlikely to solve this pressing problem.31 1985 1996 Sustaining reform by treating import competition Argentina 3% on a par with domestic competition 5% 12% The widening use of antidumping actions against for- 29% eign firms threatens to undermine one of the key bene- 49% fits of global trade rules: stable and predictable access 16% to foreign markets.32 Even though there is no economic 67% 19% rationale for doing so, antidumping laws treat the ef- fects of competition from foreign firms differently from 7% Brazil 9% those of competition from domestic firms. The parity 26% between foreign and domestic firms could be restored 27% 33% by an international agreement to eliminate antidump- ing laws and to apply national competition policy laws 33% to import competition. That is, if an antitrust issue 33% 32% exists—such as predation—deal with it, but otherwise China leave pricing decisions to individual firms. 6% 1% 21% 10% Sustaining reform by easing the adjustment 37% to trade liberalization 13% Supporters of trade liberalization should give greater at- tention to developing social safety nets and to educa- 56% 56% tion and retraining policies that facilitate labor market India adjustment to internal and external shocks.33 Augment- 4% 3% 13% ing trade liberalization policies with complementary 11% 40% labor market policies that ease adjustment will reinforce 31% social cohesion and help offset pressures to close do- mestic markets to foreign goods.34 Research into innovative public policies that reduce 46% 52% the costs of economic adjustment continues. “Income Malaysia insurance,” for instance, would compensate workers in 18% 31% the short term for part of any income they lose because of economic adjustment to liberalization. Such a pro- 53% 13% gram reduces the pain of job loss while preserving the 6% 60% incentive to look for employment. 35 However, there is 9% 10% little economic justification for treating workers af- fected by trade competition differently from workers Resource-based exports Medium-technology exports affected by domestic competition, macroeconomic Low-technology exports High-technology exports shocks, the adoption of new technology, or any other Note: Export groups are based on the use of scientists and engineers form of economic adjustment. Economic adjustment in production and on the amount of research and development activity policies should aim to reduce the adverse impact of all required. Resource-based exports are unfinished raw products. Low- technology exports are typically labor-intensive manufactures with low shocks, irrespective of their source. worker-skill requirements, such as textiles, garments, and footwear. Medium-technology exports are products that entail fast-moving produc- tion technologies and some design effort, such as automobiles, chemi- Sustaining reform by directly tackling labor cals, industrial machinery, and consumer electronics. High-technology exports are products that combine intensive use of highly skilled employ- conditions in developing countries ees with substantial research and development; examples include fine chemicals and pharmaceuticals, aircraft, and precision instruments. Labor practices in developing countries have received Source: Lall 1998. much publicity recently, thanks largely to the efforts of nongovernmental organizations (NGOs). Multinational      ⁄  Table 2.1 Reported antidumping actions by members of the GATT and WTO, 1987–97 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 New users 24 17 19 20 48 70 162 114 83 148 115 Traditional users 96 107 77 145 180 256 137 114 73 73 118 Note: Traditional users of antidumping laws are Australia, Canada, the European Community (and its successor, the European Union), New Zealand, and the United States. This classification is taken from the source. New users are Argentina, Brazil, India, the Republic of Korea, Mexico, and South Africa. Source: Miranda, Torres, and Ruiz 1998. corporations are particularly in the spotlight. Damaging performance is weak.37 Moreover, imposing trade sanc- reports have emerged of workers laboring for a fraction tions on imports from developing countries—especially of the minimum wage in industrial countries in facili- in labor-intensive industries—will lower wages and ties that fall far short of the safety standards of high- worsen working conditions in those countries, not im- income countries. This publicity has generated strong prove them. Better alternatives to imposing trade sanc- demands for incorporating international labor standards tions exist, including aid programs to improve labor into the WTO, with trade sanctions to enforce them.36 conditions. In addition, developing economies can take The debate on the merits of this proposal is intense, but steps themselves to improve the conditions of working the evidence that lower labor standards boost export people, including children (box 2.3). Sustaining reform by preserving the legitimacy Figure 2.7 of global trade rules New users initiated an increasing number of antidumping suits during 1987–97 The number of disputes among WTO members is likely to increase in the future, thanks to growing competition Reported antidumping actions by new users in the services and goods markets and the wider scope (percentage of total antidumping actions) 80 of multilateral trade rules. NGOs, subnational govern- ments, and even private sector firms will want to be 70 included as participants in the dispute settlement mech- anism.38 If this pressure is not handled well, the legiti- 60 macy of global trade rules will be called into question. A first step in maintaining the legitimacy of global 50 trade rules is to make more resources available for the 40 WTO to implement its dispute resolution mechanism. Several other reforms are also worth considering.39 Dis- 30 pute panels could be allowed to take evidence from groups other than governments so that all interested par- 20 ties can be heard. In addition, regular WTO ministerial meetings can review the ongoing case law that will re- 10 sult from the dispute resolution mechanism, resolving 0 the inconsistencies that disputes might reveal in the pro- 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 visions of WTO agreements. Note: Traditional users of antidumping laws are Australia, Canada, International trade and development policy: the European Community (and its successor, the European Union), New Zealand, and the United States. This classification is taken from the next 25 years the source. New users are Argentina, Brazil, India, the Republic of Korea, Mexico, and South Africa. International trade institutions and liberal trade policies Source: Miranda, Torres, and Ruiz 1998. are a means to an end. They boost trade in existing and new products, enhancing competition in markets, stim-    :     Figure 2.8 When filing antidumping investigations, industrial and developing countries target each other almost equally Antidumping investigations filed By industrial countries By developing countries By transition countries 4% 23% 31% 38% 37% 39% 32% 96% Against industrial countries Against developing countries Against transition countries Source: Miranda, Torres, and Ruiz 1998. ulating productivity, and fostering technology transfer. the case study on Tanzania in chapter 8 makes clear. All these developments in turn increase social welfare. The Uruguay Round of trade talks realized only a small The experience of the last 50 years demonstrates that part of the feasible gains from liberalized trade in agri- global trade rules enhance the benefits of unilateral trade culture because countries were reluctant to scale down liberalization by reinforcing incentives to lower trade barriers.42 Likely opportunities will arise from a variety barriers and avoid policies that constrain trade. of sources: changes in consumer habits, reductions in The global trade regime does face challenges (as dis- air transportation costs, advances in biotechnology, and cussed above) that must be tackled in order to make fur- the liberalization of global trade rules. ther gains. Should these challenges be overcome, what Rising consumer incomes and declining demand for are some of the growth-inducing possibilities? Four such frozen, canned, and other processed food are creating a possibilities are likely to be uppermost in the early need for high-value-added products rather than homo- decades of the 21st century: agricultural trade, foreign geneous bulk goods. Falling surface and air transporta- investment and trade in services, international produc- tion costs enable firms to supply new markets with fresh tion networks, and commerce arising from urban devel- products. By increasing the variety of available agricul- opment.40 Other possibilities have been discussed else- tural products, advances in biotechnology may become where. For example, devising the appropriate intellectual particularly relevant for developing countries whose cli- property rights regime for developing countries was dis- mates sustain only a narrow range of basic agricultural cussed in World Development Report 1998/99: Knowledge crops. These developments expand the range of poten- for Development.41 The World Bank is not alone in ana- tial exports as well as the markets to which products can lyzing these issues, as the OECD’s 1998 study Open be sold. But exports can be constrained if a country’s do- Markets Matter shows. mestic infrastructure and trade regulations do not per- mit speedy delivery. Fears about product safety that lead Stimulating trade in agricultural products to calls for banning imports of certain foods can also In developing countries, agriculture offers opportuni- constrain export growth. The long-standing dispute be- ties not only for expanding export trade but also for im- tween the European Union and the United States over proving the livelihoods of many rural populations, as hormones used in cattle feed is but one example of this      ⁄  Box 2.3 Child labor: how much? how damaging? and what can be done? In developing countries about 250 million children between attend school and work simultaneously may be the best ap- the ages of 5 and 14 work, at least 120 million of them full proach in rural areas. The school year must be carefully time. In Asia 61 percent of all children work full time; in Africa, scheduled in these areas in order not to conflict with the 32 percent; and in Latin America, 7 percent. Around 70 percent peak agricultural season, however. Reducing the cost of edu- of all child laborers are unpaid family workers. Fewer than 5 cation through subsidies, direct payments, and school feed- percent are employed in export-related production. The vast ing schemes also gives households an incentive to send chil- majority of children working in rural areas are engaged in agri- dren to school rather than to work. cultural activities, while urban children tend to work in services Ⅲ Providing support services to working children. These ser- and manufacturing. vices can include meals, basic literacy classes, and night Though official statistics suggest that more boys work than shelters. Since these programs usually concentrate on chil- girls, the main difference is that boys tend to work in more vis- dren working visibly on the street, their scope is somewhat ible types of employment (in factories, for instance), while girls limited. perform unpaid household tasks or work as domestics. When Ⅲ Raising public awareness. This approach covers a wide this difference is taken into account, boys and girls work in spectrum: improving the general awareness of hazards to similar proportions. The intensity of work boys and girls per- working children, raising parental awareness of the loss of form may differ, however, with girls working longer hours. This human capital associated with child labor, and involving em- fact is consistent with the common observation that girls in ployers, unions, and civil society in efforts to reduce child developing countries generally have lower school enrollment labor. rates than boys. Ⅲ Enforcing legislation and regulations. Most countries have Not all child labor is harmful. Working children who live in laws and regulations governing child labor, but enforcement a stable environment with their parents or under the protec- is weak. In fact stricter, across-the-board enforcement may tion of a guardian can benefit from informal education and job end up hurting those it intends to protect by reducing the training. Many working children are also studying, and their income of poor families and forcing children into more dan- wages help their siblings attend school. However, some forms gerous and hidden forms of employment. The alternative is of employment, in particular prostitution and forced or bonded to focus legislation on the most intolerable forms of child labor, involve working conditions that are hazardous to the chil- labor. A new International Labour Organization (ILO) conven- dren’s health, both physical and mental. tion targeting the worst forms of child labor—including slav- The rate of children’s participation in the labor force de- ery, prostitution, forced labor, bonded labor, and illegal and clines as a country’s per capita GDP rises. While as many as hazardous work—was adopted in June 1999. half of all children in the poorest countries work, the numbers begin falling rapidly as per capita GDP reaches around $1,200. Many other proposals for reducing child labor—including The incidence of child labor also tends to decline as educa- trade sanctions, consumer boycotts, social clauses and certifi- tional enrollment rises and school quality improves, although cation, and labeling schemes—are fraught with problems. For the cross-country variations in these relationships are large. example, exports produced in the formal sector are the prod- Policies that reduce child labor have strong support on ucts hit hardest by trade measures, and one effect can be to purely economic grounds. When children are sent to work at force workers (including child laborers) into the informal sec- very young ages for extended periods, they do not develop the tor, where working conditions are typically worse. Trade sanc- skills necessary to earn higher wages later in life, and society tions, which may be little more than a cover for the introduc- loses needed human capital. As adults these individuals have tion of protectionist measures, may be implemented in ways low productivity levels that become a drag on economic that have little to do with child labor. Finally, labeling schemes growth. and social clauses are often impossible to monitor. Several approaches to reducing child labor have been sug- The World Bank has taken steps to reduce harmful child gested. They are not mutually exclusive and probably work labor through its ongoing poverty reduction efforts and the best in combination. child labor program established in May 1998. The program is the focal point for Bankwide child labor activities and supports Ⅲ Reducing poverty. Poverty is a major cause of harmful child initiatives such as child labor reduction evaluations. It draws labor. In poor households, children’s wages may be essen- upon the international experience of labor experts from acade- tial to the family’s survival. Even though poverty reduction mia, nongovernmental organizations, and other multilateral and is a long-term process, programs that improve the earnings bilateral organizations such as the United Nations Children’s of the poor, address capital market constraints, and provide Fund (UNICEF) and the ILO. safety nets can help reduce child labor in the short term. Ⅲ Educating children. Increasing primary school enrollments Source: Fallon and Tzannatos 1998; Grootaert and Kanbur tends to decrease child labor. Making it easier for children to 1995; ILO 1993; World Bank 1999f.    :     problem. The debate over agricultural trade policy, then, Figure 2.9 is likely to encompass not just market access but meth- Many countries bound their tariffs on agricultural products in the Uruguay Round at levels well ods of production as well.43 above estimated actual tariffs in 1986–88 The Uruguay Round agreement on trade in agricul- tural products laid the foundation for future liberaliza- Tariffs on rice tion. Countries agreed to convert nontariff agricultural (percent) barriers into tariff barriers and to set their tariffs at or 250 Estimated actual tariff below a certain level (the “bound” tariff rate). Similar 200 in 1986–88 Uruguay Round maximums were agreed to for export subsidies and do- Base tariff mestic subsidies. The advantage of this approach is that 150 it converts a wide range of trade distortions into three ob- 100 servable trade policies, with maximum levels that can be 50 negotiated down over time.44 Unfortunately, many coun- tries took advantage of this opportunity to convert their 0 Mexico Colombia Venezuela Thailand Indonesia Czech nontariff barriers into extremely high maximum tariffs. Republic For three widely traded commodities—rice, coarse grains, Tariffs on coarse grains and sugar—many governments chose to set their maxi- (percent) mum permitted tariff in the Uruguay Round well above 300 Estimated actual tariff the actual tariff collected in 1986–88 (figure 2.9). 250 in 1986–88 There are several reasons why these tariffs are highly Uruguay Round 200 Base tariff damaging. First, by raising domestic prices above world 150 prices, they raise the cost of food to consumers. Sec- 100 ond, they increase costs for domestic food processing 50 firms, harming their export competitiveness. Third, the 0 artificial expansion of the domestic agricultural sector Mexico Colombia Venezuela South Indonesia Morocco increases the demand for resources, making them more Africa expensive for the rest of the economy.45 These econo- Tariffs on sugar mic costs must be added to those created by export sub- (percent) sidies for agriculture and the taxes that finance these 250 Estimated actual tariff in 1986–88 subsidies. Thus, the next round of trade negotiations 200 Uruguay Round should seek to negotiate substantial reductions both in 150 Base tariff agricultural trade barriers and in those market barriers 100 created by state-owned monopolies that trade in agri- 50 cultural products.46 Since agricultural trade barriers distort the allocation 0 Colombia Venezuela South Indonesia Morocco Czech of national resources, their removal will induce adjust- Africa Republic ments that may include migration from rural to urban Note: Figure shows selected commodities to which tariffication was areas. Moreover, reform may lead to fears about depen- applied and that are subject to safeguards. dence on foreign sources for food. Recognizing the dis- Source: Hathaway and Ingco 1996. location induced by trade reform reinforces the case for enhanced flexibility of domestic labor markets and for a robust social safety net. Furthermore, during 1996–97 particularly blunt instruments, imposing restrictions the World Bank provided loans to over 20 countries to on imports that go well beyond what is needed to pro- smooth the adjustments created by reform. In addition, tect human health.48 However, governments often have assistance was offered to countries facing food shortages legitimate concerns about protecting the well-being of and other agricultural emergencies.47 their citizens. The Agreement on Sanitary and Phy- Advances in biotechnology have introduced a new tosanitary Measures that resulted from the Uruguay factor into agricultural trade policy—sanitary and phy- Round seeks to strike a balance between these concerns tosanitary regulations. Sometimes these regulations are and unnecessary restrictions by ensuring that sanitary      ⁄  and phytosanitary regulations do not deliberately dis- During 1994–97, world exports of services grew by criminate against foreign suppliers. A core requirement more than 25 percent. Forecasts of the growth in U.S. is that domestic standards be based on scientific evi- trade in services suggest that this pace will resume in dence, and nothing prevents those standards from be- the early part of the 21st century, after the macroeco- ing above international norms.49 But even seemingly nomic effects of the East Asian crisis have abated. Much unobjectionable regulations based on scientific evi- of this growth will come from developing countries in dence can be disputed, and the implementation of the Asia and from Brazil, challenging the dominance of agreement will place further burdens on the WTO’s North American and European firms.52 In addition, the dispute settlement mechanism. Those hearing the cases rise of electronic commerce has created new possibili- may well have to assess each protagonist’s scientific case ties for trade in services. For example, a leading Ukrain- as well as the implications for international trade.50 ian manufacturer of wind turbines now contracts out all of its administrative and financial reporting to an ac- Liberalizing trade and foreign investment in services counting firm in southern England.53 Changes in technology, demand, and economic struc- The stakes in service liberalization are high because ture will make the exchange of services an increasingly most industries use services as inputs to production. important form of trade in the 21st century (figure Manufacturing industries need cheap and reliable ac- 2.10). Falling communication costs and the use of cess to global communication and transportation net- common international standards for some professional works to maintain export performance. With products services contributed to the large jump in service trade becoming increasingly time-sensitive—the result of that took place in the mid-1990s. Developing countries shorter product lives and the use of “just-in-time” pro- stand to gain considerably from the liberalization of duction—foreign buyers must be assured that a sup- trade in services, especially in labor-intensive sectors plier can deliver needed goods on time. Inefficient such as construction and maritime activities.51 The lib- transportation systems (see the case study on the Arab eralization of services will also promote competitive- Republic of Egypt in chapter 8) can prevent domestic ness in sectors that use services as inputs to production. industries from joining global production networks. Figure 2.10 Exports of commercial services increased in every region from 1985 to 1997 Exports of commercial services for selected regions (billions of U.S. dollars) 900 800 1985 1990 700 1997 600 500 400 300 200 100 0 Industrial Latin America Europe and Sub-Saharan Middle East East Asia South Asia countries and the Central Asia Africa and and Pacific Caribbean North Africa Source: WTO, Annual Report, 1996, 1998.    :     When service firms receive trade protection from competition: 25 percent in industrial countries and a foreign competition, they can raise the prices they paltry 7 percent in developing countries.56 charge to purchasers, which increases the purchasers’ Industrial countries tend to have more restrictions costs. In this case protecting service sectors effectively on services that require the temporary entry of people reduces any protection received by their purchasers— or the temporary establishment of businesses—for ex- as happened in Egypt in 1994, undermining industrial ample, construction services, which is one sector where performance in chemicals, crude petroleum and nat- developing countries have a comparative advantage.57 ural gas (where the services purchased accounted for 89 Looking forward, there is substantial room for the fur- percent of input costs), and iron and steel.54 ther liberalization of numerous service sectors in both The same core principle underlies trade policy re- developing and industrial economies. Since the com- forms in both services and goods. Measures that give petitiveness of these sectors differs across countries, ne- foreign firms increased access to domestic markets will gotiations that encompass a wide range of sectors, enhance competition, lower prices, raise quality, and rather than a few sectors in which one country (or improve social welfare. But trade policy for services group of countries) has a competitive advantage, offer must take into account an important issue that does the most room for trade-offs and mutually beneficial not affect trade in goods. Trade in services generally agreements.58 involves the movement of people or capital across na- tional boundaries, often in the form of new subsidia- Fusing domestic firms into ries. As a result, opening services to international com- global production networks petition may require changes in policies on border The fragmentation of production processes across inter- measures (as with tariffs), foreign direct investment (see national borders is an important new trend, particularly chapter 3), or migration, both temporary and perma- for developing economies. This “slicing up the value nent. Future trade negotiators, like those in the Uru- chain” involves separate stages of production being con- guay Round, face the challenge of refining global trade ducted in different countries.59 Declining communica- rules for services that take into account the interactions tion costs and improved transportation systems permit among these policies. just-in-time delivery and the coordination of production The Uruguay Round produced an agreement on re- across borders.60 Developing economies can expedite ducing barriers to trade in services, the General Agree- their integration into the new production systems by lib- ment on Trade in Services (GATS). The principal con- eralizing and improving their telecommunications and tribution of the agreement lies in the framework it transportation sectors. Global trade rules have fostered defines, which mandates the application of certain trade global production networks, and an associated rise in rules across service sectors. These include rules govern- intrafirm trade, by progressively lowering trade barriers ing most favored nation (MFN) treatment and prohi- and reducing the likelihood of unpredictable increases.61 bitions against certain restrictions on suppliers.55 The International trade data are useful indicators of the framework also defines four supply modes for services: rise in global production networks.62 More than half the cross-border, which does not require the physical move- exports of foreign affiliates of Japanese and U.S. firms ment of producer or consumer; movement of consumer go to other members of the firms’ production networks, to producer; permanent movement of the producer (in- and close to 40 percent of the parent firms’ exports go cluding establishing subsidiaries); and temporary move- to their foreign affiliates. In total, about one-third of ment of people. world trade in the mid-1990s took place within global But the agreement leaves a substantial amount of production networks. In certain industries the trend is room for future liberalization. The coverage of service even more impressive. In 1995 components accounted sectors and supply modes is limited. The agreement for more than one-third of all transportation and ma- covers only 47 percent of sectors (including the key chinery imports to Honduras, Indonesia, Mexico, the telecommunications and financial sectors) in industrial Philippines, and Thailand.63 Similarly, parts and com- countries and 16 percent in developing countries, with ponents accounted for more than one-third of total numerous exceptions. A revealing measure of the limits transportation and machinery exports from Barbados, of liberalization under this agreement is the percentage Brazil, the Czech Republic, Hong Kong (China), Nica- of service sectors that will experience full international ragua, and Taiwan (China) (table 2.2).      ⁄  Table 2.2 Share of parts and components in exports, 1995 Percentage of parts and components in: Economy Total exports Exports of manufactures Exports of transportation and machinery Singapore 18.2 21.7 27.8 Taiwan (China) 17.4 18.8 36.3 Malaysia 14.3 19.1 25.9 Hong Kong (China) 13.6 14.5 46.2 Mexico 13.0 16.8 24.9 Thailand 10.9 15.0 32.5 Barbados 10.9 18.5 61.6 Czech Rep. 10.6 13.0 36.2 Korea, Rep. of 10.0 11.0 19.1 Slovenia 7.7 8.6 24.5 Philippines 6.6 16.0 29.7 Brazil 6.4 12.1 33.9 China 6.0 7.2 28.8 Croatia 5.4 7.3 32.1 Nicaragua 5.0 24.6 81.6 Source: Yeats 1998. The creation of these global production networks, raise taxes on incomes that are less internationally mo- either as formal corporations or as part of ethnic dias- bile, such as labor. Such pressure could in turn under- poras (see chapter 1), helps foster an open trading sys- mine political support for open markets. Multinational tem. Their supporters can be expected to push for con- corporations may appear to be the primary beneficia- tinued liberalization on three main fronts. First, they ries of liberalization, while contributing little to the in- will argue for the removal of tariffs on parts and semi- frastructure that encourages production networks in finished goods because when these goods cross national the first place. borders several times, even small tariffs can accumulate In response to these concerns and others about the and undermine profitability. Second, proponents will environmental consequences of some types of produc- push for improvements in domestic and international tion and the competitive consequences of mergers by transportation systems because substandard communi- some of the largest corporations, multinationals may cation and transportation act as a tax on profitability.64 face more constraints on their activities. A farsighted ap- Third, the new production networks thrive on—indeed, proach would be for leading multinational corporations they expect—stable, predictable trade and investment to develop a code of practices on tax and environmental policies. For this reason alone, multinational corpora- measures that includes enforcement mechanisms simi- tions will support effective enforcement provisions in lar to those in international trade agreements. Alterna- regional and multinational trade agreements.65 tively, a long-term goal could be a unitary tax system Developing countries can benefit substantially from that distributes corporate tax revenues among countries their firms’ participation in global production net- according to a prearranged formula.67 works. However, they must also beware of possible ad- verse fiscal implications. A large portion of the trade Urban development, trade flows, these networks generate happens within firms that are and the world trading system able to realize profits in countries with low tax rates. The expected growth of cities is emphasized throughout Countries with high corporate tax rates may attract for- this report (especially in chapters 6 and 7) as a key fac- eign direct investment but will realize lower profits than tor shaping the future of developing economies. Urban they expected.66 The benefits of these networks to the growth, geographic and economic, will affect both trade economy are then partly offset by a smaller national flows and the international system governing them. One corporate tax base, resulting in increased pressure to challenge that has already been mentioned requires    :     accommodating more views in international trade not overburdened with regulations that prevent them forums—including those of urban policymakers—while from performing this function is essential to avoiding retaining the rights of national governments to initiate, permanent increases in unemployment. The speed at participate in, and conclude trade negotiations. But which information about profitable urban economic op- many other issues will arise as well. portunities reaches investors can increase cities’ capacity First, the economic strength of cities is built on ag- to absorb shocks. Again, capitalizing on this advantage glomeration economies, which enable producers to requires urban policies that ease the exit and entry of function more efficiently in proximity to a dense net- firms, including foreign firms. Ultimately, the rise of work of information, employees, suppliers, and cus- cities—especially cities that take measures to minimize tomers. These agglomeration economies can generate the damage wrought by external shocks—may quicken more specialized urban production structures. As a re- the pace of trade liberalization and the integration of de- sult, urban policymakers also have an interest in pre- veloping economies into the world trading system. serving market access abroad and at home—abroad for • • • their cities’ exports, and at home for intermediate in- puts that improve productivity and for consumption The impressive trade reforms developing countries have goods that may be cheaper elsewhere. The rise of cities undertaken in recent years have yielded substantial eco- as economic and political powers, then, is likely to re- nomic benefits. But sustaining the momentum of trade inforce support for an open world trading system. reform will be a key challenge for the next 25 years. The Second, to exploit agglomeration economies, cities continued liberalization of the agricultural and service will increasingly recognize the need to make progress in sectors, in particular, will deliver considerable benefits several policy areas, not just trade liberalization. For ex- to developing economies. ample, the effectiveness and cost of transportation and The social consequences of the new openness to trade communications services clearly affect cities’ capacity have been associated with a series of economic adjust- to import and export goods and services.68 Cities may ments, such as regional and sectoral disparities and in- become a force advocating the simultaneous negotia- ternal migration to cities. Labor market institutions, in- tion of liberalization in many sectors, counteracting the cluding schemes to enhance labor mobility and raise interests of producers who support a sector-by-sector skills, need to be strengthened in order to smooth the approach to negotiation. Because the number of poten- adjustment to trade reform. Policymakers must work to tial trade-offs across sectors in international trade nego- ensure that the considerable gains from trade reform are tiations is greater than those within sectors, cities may widely shared among the population, reassuring those offer increased support for broad-based trade liberaliza- who initially suffer from reform that their long-term tion in the WTO. welfare is secure. Third, while integration into the world trading sys- Maximizing the opportunities for development of- tem offers numerous opportunities for urban producers fered by expanding international trade will require a sta- and consumers, cities will have to bolster their capacity ble and predictable framework of institutions. Codify- to absorb external trade shocks, such as a collapse in ex- ing the rights, responsibilities, and policies of all parties port prices. The range of employment opportunities in in broad-based institutions will smooth the path of trade cities is wider than in rural areas (where production is liberalization and development reform over the next often concentrated in a few goods and services) and thus 25 years. The upcoming Millennium Round of trade helps absorb some of the effects of shocks on the labor negotiations provides an excellent opportunity to pur- market. However, ensuring that urban labor markets are sue such a wide-ranging approach to trade policy reform. C h a 3 p t e r Developing Countries and the Global Financial System T he 1990s saw a huge upsurge in flows of appeared—a huge pool of highly mobile private capital from industrial to devel- money channeled through mutual funds, oping countries. At the beginning of the pension funds, and wealthy individuals decade, private and official flows were that is ready to move across borders at a about the same, but only five years later moment’s notice in search of the highest private flows dwarfed official flows. Not short-term returns. since the late 19th century have inter- Countries that open themselves up national capital flows assumed such to these short-term capital flows are dis- prominence.1 But there are marked dif- covering that such investments have ferences between the movement of cap- their costs. Rapid changes in investor ital at the end of the 20th century and sentiment can cause enormous insta- the movement of capital a century ear- bility, particularly in developing econ- lier. These differences have important omies. This realization has led to a policy implications for developing coun- reexamination of the international eco- tries as they integrate into the global fi- nomic architecture, raising some im- nancial system. portant questions: Are the benefits of At the end of the 19th century capi- liberalizing capital accounts worth the tal flows financed infrastructure projects costs? Can developing countries find such as railroads and direct investment ways to capture the gains from finan- in foreign companies. A hundred years cial globalization without running such later foreign direct investment is chan- enormous risks, which often jeopardize neled primarily through multinational the poorest individuals? The policy re- corporations that are establishing plants sponse is to calibrate a sequential ap- and service operations throughout the proach to financial reform that both en- world. These investments bring with sures stability in developing countries them more than money. They open ac- and captures the benefits of integration cess to markets, make new technolo- into world capital markets. gies available, and provide workers with This chapter emphasizes the four training. But another type of capital has key components of that approach:       ⁄  Ⅲ Developing countries need to strengthen banking reg- developing and transition economies.3 Since 1980 the ulations and, where possible, build complementary amount of net foreign direct investment in developing and well-regulated securities markets, if the benefits countries has climbed more than twelvefold (figure of domestic financial liberalization are to materialize. 3.1).4 In contrast, net portfolio investment flows have Ⅲ While banking regulation is being strengthened, poli- been far more volatile throughout the 1990s, exceeding cies should be directed to reducing the demand for— $100 billion in 1993 and 1994 and falling considerably and volatility of—short-term foreign borrowing. since then. Ⅲ Further international cooperation in setting and im- Firms in developing and industrial countries alike plementing fiscal, monetary, and exchange rate poli- are raising more funds from international securities cies should be considered. markets. Multinational corporations are registering Ⅲ Long-term foreign investment should be attracted by their equity on more than one country’s stock exchange cultivating a healthy economic environment—in- and are raising funds from financial markets in differ- cluding investing in human capital, allowing domes- ent economies. Since 1993 the amount of outstanding tic markets to work without unnecessary distortions, international debt issued by all firms has risen by 75 and committing to a strong regime of investors’ percent, reaching $3.5 trillion in early 1998. Although rights and obligations—and not by offering subsidies financial and nonfinancial companies headquartered in or other inducements. industrial countries issue most of this debt, firms in countries such as Brazil, Mexico, and Thailand have The chapter examines the mixed record to date of also begun to tap the global market for capital—a path developing countries’ integration into the international others will surely follow (figure 3.2). financial system. It draws from a variety of experiences This rising number of international capital transac- to identify the principal benefits and risks of global fi- tions, together with the substantial growth in interna- nancial integration. Even more important, it proposes national and global responses that can further develop- ment goals without jeopardizing financial stability. Figure 3.1 Since 1980 net inflows of foreign direct The gathering pace of international and portfolio investment to developing financial integration economies have grown enormously Rapid improvements in technologies for collecting, Net inflows (billions of U.S. dollars) processing, and disseminating information, along with 160 the opening of domestic financial markets, the liberal- ization of capital account transactions, and increased 140 private saving for retirement, have stimulated financial Foreign direct investment innovation and created a multitrillion-dollar pool of in- 120 ternationally mobile capital. At the same time, consoli- dation in the global banking industry and competition 100 from nonbank financial institutions (including hedge Foreign portfolio investment and mutual funds) have lured new players to the inter- 80 national financial arena. These trends accelerated in the 1990s, expanding investment opportunities for savers 60 and offering borrowers a wide array of sources of capi- tal.2 The same trends can be expected to continue well 40 into the 21st century. 20 The growing pool of international financial capital 0 Over the last two decades, the financial markets of lead- 1970 1980 1990 1991 1992 1993 1994 1995 1996 1997 1998 ing industrial countries have melded into a global finan- cial system, permitting ever-larger amounts of capital Source: IMF, Balance of Payments Statistics Yearbook, 1998. to be allocated not only to their economies, but also to         Figure 3.2 Figure 3.3 Firms from developing countries are issuing A growing pool of institutionally managed funds more international debt than before is invested abroad Outstanding debt securities At home (billions of U.S. dollars) 1980 1995 Abroad 60 Brazil US$2 trillion US$20 trillion Hungary Mexico 5% 20% 50 South Africa Thailand 40 30 95% 80% 20 Source: IMF, International Capital Markets, 1998. 10 0 1993 1994 1995 1996 1997 1998 Liberalizing capital flows in developing Source: IMF, International Capital Markets, 1998. and transition economies The 1990s have seen a consistent trend toward more flexible exchange rate regimes and the liberalization of tional trade in goods and services, has increased turn- capital account transactions. The latter involves changes over on foreign exchange markets eightfold. In 1998 in policies toward different types of private capital the daily total stood at around $1.5 trillion, an amount flows, such as foreign direct investment, foreign bond equal to around one-sixth of the annual output of the and equity investment, and short-term borrowing from U.S. economy. Financial instruments with very similar abroad. Developing countries in Asia and the Western risks pay similar returns no matter where they are is- Hemisphere, and the transition economies, have moved sued, providing further evidence of the integration of toward having a single exchange rate, rather than try- national capital markets. The returns on these instru- ing to have one rate for those who are exchanging their ments varied widely across countries as recently as 10 currency because of foreign trade and an alternative rate or 20 years ago. for those who exchange currency in order to invest.5 Mutual funds, hedge funds, pension funds, insur- Old-style rules that used to require exporters to ex- ance companies, and other investment and asset man- change their earnings of foreign currency with the na- agers now compete with banks for national savings. Al- tion’s central bank have been relaxed by developing though thus far this phenomenon has been confined countries on every continent, particularly in the West- primarily to industrial economies, the consequences for ern Hemisphere and Eastern Europe. developing countries could be far-reaching. Institu- The speed and depth of capital account liberaliza- tional investors have taken advantage of the easing of tion have varied across countries, however. Most coun- restrictions in many industrial countries to diversify tries have moved toward capital account convertibility their portfolios internationally, enlarging the pool of as part of a wide-ranging, gradual economic reform financial capital potentially available to developing and program that includes measures to strengthen the fi- transition economies. In 1995 these investors con- nancial sector. But Argentina, the Baltic countries, trolled $20 trillion, 20 percent of it invested abroad. Costa Rica, El Salvador, Jamaica, the Kyrgyz Republic, This figure represents a tenfold increase in the funds Mauritius, Singapore, Trinidad and Tobago, and Vene- and a fortyfold increase in such investments since 1980 zuela have opened important parts of their capital ac- (figure 3.3). counts in one stroke.6      ⁄  In addition to moves toward capital account con- A proliferation of bilateral investment treaties rein- vertibility, other policies have made many developing forced these domestic reforms. Between 1990 and 1997 countries a more attractive destination for foreign in- developing countries were parties to 1,035 bilateral in- vestment: macroeconomic stabilization and structural vestment treaties, which protect the rights of foreign in- reforms, privatization policies, relaxed rules on foreign vestors and engender a regulatory environment that pro- direct investment, and lower interest rates in industrial motes investment. Other treaties also reduce investor countries. Rising confidence in the economic prospects exposure to double taxation by authorities in the home of developing countries in the 1990s was reflected in country of the investor and in the destination of the the fact that foreign direct investment accounted for a investment.11 Argentina, China, the Arab Republic of greater proportion of capital inflows, which signals a Egypt, the Republic of Korea, and Malaysia have signed commitment to invest over a longer time horizon than the most treaties, followed by Central and East European portfolio investments like equity holdings.7 countries. More recently, Latin American countries By 1997 approximately half of all capital flows to have also begun signing such treaties, starting, as is tra- developing countries was foreign direct investment.8 ditional, with their regional neighbors. By reinforcing These investments fell slightly in 1998 in response to commitments to stable national investment regimes, the East Asian crisis, a change that may prompt many these treaties are encouraging greater international in- countries to reevaluate their policies toward such in- vestment flows. In addition, these bilateral treaties are vestments—and the recommendations developed later being reinforced by a growing set of regional and sectoral in this chapter provide a framework for action. Devel- investment accords.12 oping countries are also becoming foreign investors A small group of developing countries has consis- themselves. In 1996 they invested $51 billion abroad, tently attracted most foreign investment (figure 3.4).13 raising their share of global foreign direct investment Brazil, Indonesia, Malaysia, Mexico, and Thailand have outflows to 15 percent. Like industrial countries, they been among the top 12 recipients in each of the past invest predominantly in economies in the same region three decades. China (including Hong Kong) joined this or continent. group in 1990 and by 1998 had received $265.7 billion Foreign direct investment in service industries ac- in foreign direct investment, making it the most sought- counts for close to two-thirds of such capital flows, after destination among developing countries. A few while the share of such investment in manufacturing African and Middle Eastern countries have been very has been falling. Although these aggregate figures con- successful in attracting foreign investment as well, but as ceal differences across countries, the shift toward ser- a group Africa and the Middle East have received less vices is significant. Traditionally, service industries have than 10 percent of foreign direct investment flows. In been less exposed to international trade and so lacked 1997 the stock of such investment in Africa was less than this stimulus to control costs, develop products, and in- 2 percent of the world total. For this reason many Sub- novate. Foreign direct investment offsets this deficiency Saharan countries will continue to rely on multilateral by enhancing the degree of competition in domestic and bilateral aid to finance investment projects (box 3.1). service markets and by transferring best practices from Although multinational corporations typically in- abroad (see chapter 2). In addition, firms in develop- vest in foreign countries in order to sell in domestic ing countries have become more involved in cross- markets or to create new bases for exporting, interna- border partnerships with foreign firms—joint ventures tional firms have long shown an interest in exploiting with or without equity stakes, franchises, licensing, and developing countries’ natural resources, including oil, subcontracting or marketing agreements. Since 1990 minerals, and lumber. Investment in natural resources more than 4,000 such agreements have been signed, is often enclave investment. It brings needed capital complementing the flows of foreign investment.9 into a country but offers few of the other benefits— The continuing liberalization of national regulatory new technologies, new markets, and increased human frameworks for foreign investment has fostered these capital—that are usually associated with manufactur- capital inflows and interfirm agreements. In 1997 at ing investment. In many cases, the economic activities least 143 nations had frameworks for foreign direct in- such investments entail are located in relatively remote vestment in place. Some 94 percent of the regulatory areas, far from other areas of economic activity. changes since 1990 have actually helped create more fa- The benefits to developing countries of foreign in- vorable environments for foreign direct investment.10 vestment in natural resource exploitation have been         Figure 3.4 Box 3.1 A few developing countries received A continuing role for aid the lion’s share of FDI invested outside industrial countries in 1997 Among least-developed countries, the smallest and most resource poor are the least likely to receive substantial pri- Worldwide stock of FDI in 1997 ($3,456 million) vate capital flows. These countries still need official aid flows to finance investments in health, education, the en- Other developing countries 10% vironment, and basic infrastructure. In 1998 net official China flows worldwide totaled approximately $51.5 billion. 6% Aid can be highly effective in promoting growth and re- ducing poverty. But aid is also a scarce resource that needs Brazil 4% to be used well, and using it well requires good decisions by governments and donors alike. Whether aid increases eco- 20% Mexico nomic growth, for instance, depends on a country’s policy 3% and institutional environment. Good macroeconomic man- Singapore agement, sound structural policies and public sector admin- 2% istration, and measures that increase equity are all impor- Indonesia 2% tant. They promote growth themselves, and they support Industrial the growth-enhancing effects of development assistance. countries Malaysia 70% 1% Development assistance, like so many other economic Saudi inputs, is subject to diminishing returns. Even countries Arabia with excellent policies are limited in their capacity to ab- 1% sorb such aid. Once official assistance reaches around 12 Argentina percent of GDP, its potential contribution to growth is usu- 1% ally exhausted. But few countries receive such high levels Source: UNCTAD, World Investment Report, 1998. of aid, so that only a country’s policy environment limits its capacity to absorb development assistance. While the governments of developing countries deter- mine the effectiveness of aid in the growth process, ambiguous, for several reasons. First, the benefits to a donors determine how effective aid is in global poverty re- duction. For it is donors, not recipient governments, that developing country may be smaller than GDP indica- decide which countries receive assistance. In making this tors initially suggest, as these indicators do not take into decision, donors need to keep in mind two factors: account the wealth the country loses when resources are extracted. Second, the resulting economic growth may Ⅲ The extent to which assistance will raise the growth rate, not be sustainable. In some cases the legacy may be a factor that depends on the policy and institutional envi- ronment and thus differs considerably across countries more negative than in others. If gold extraction tech- Ⅲ The existing level and distribution of income in the recip- nologies lace the surrounding environment with cya- ient country, since income growth in a country like Chile, nide, the costs of restoration can be enormous. In con- where poverty is low, tends to reduce poverty less than trast, companies can replant hardwood forests that have it would in a country with mass poverty, like India. been logged. Three-quarters of the world’s poor (those living on less The kinds of foreign direct investment that are most than $2 per day) now live in countries where the policy en- likely to provide useful benefits and sustainable, long- vironments are such that additional aid would raise the term growth are associated with manufacturing pro- growth rate. The challenge is to allocate the assistance ducer services. Unfortunately even those African coun- available in order to take advantage of the favorable cli- mate for growth. tries with a five-year record of good economic policies have found it difficult to attract this kind of investment, Source: Collier and Dollar 1998; World Bank 1998a, 1999i. in spite of evidence showing that the overall returns in these economies may be just as good as elsewhere. Financial interruptions to development: banking cies (or lack thereof ) toward foreign portfolio invest- and currency crises ment and short-term foreign borrowing.14 These kinds Even though it is widely accepted that developing coun- of flows have been closely linked with the financial and tries have substantially benefited from large inflows of currency market volatility of the late 1990s. Countries foreign direct investment, the far more controversial as- with high levels of short-term debt are vulnerable to sud- pect of capital account liberalization has concerned poli- den changes in investor sentiment. The resulting mas-      ⁄  sive shifts in the direction of flows are often too much ing countries must raise their rates. The higher costs are for even strong financial systems and are certain to have passed on to domestic borrowers, increasing the likeli- disastrous consequences for weaker ones. The economic hood of defaults. Second, many firms in developing crises resulting from such vacillations have imposed countries borrow from overseas banks. When such enormous costs on the countries involved—costs that borrowing is widespread, increases in interest rates in have affected not only borrowers but also huge numbers industrial countries create a common macroeconomic of innocent bystanders. In some cases workers have seen shock, leaving firms unable to repay their loans to unemployment soar and wages fall by one-fourth or domestic as well as to foreign banks.20 Balance sheets more.15 Small businesses with prudent levels of debt deteriorate even further when a jump in industrial have found themselves either cut off from access to countries’ interest rates leads to a depreciation in a de- credit or facing astronomical interest rates few can af- veloping country’s exchange rate, so that domestic bor- ford. Bankruptcies have soared, contributing to the eco- rowers need more domestic currency to repay their for- nomic havoc and destroying information and organiza- eign currency debts. tional capital that will not be recovered for years. Third, speculative attacks can seriously jeopardize In considering the risks inherent in the ebbs and the stability of a developing economy’s banking sys- flows of international capital, governments will want to tem.21 A speculative attack on a currency occurs when differentiate between liberalizing domestic financial in- foreign and domestic depositors suddenly shift their stitutions and liberalizing the capital account. Although funds out of domestic banks into foreign currency, they involve different policy instruments and pose dif- often leaving the domestic banking system facing a ferent risks, both types of liberalization can result in fi- bank run. These attacks take place because investors re- nancial instability if they are poorly managed. The past ceive new information that affects the attractiveness of two decades should leave no doubt about the heavy costs keeping money in a country. And financial contagion of global banking crises. Between 1977 and 1995, 69 tends to occur when a country’s economic characteris- countries faced banking crises so severe that most of tics resemble those of another country that is known to their bank capital was exhausted.16 Recapitalizing these be in severe macroeconomic difficulties (box 3.2).22 banks was extremely expensive, with budgetary costs Fears of a banking or currency run may be self- reaching approximately 10 percent of GDP in Malay- fulfilling, creating a macroeconomic crisis that would sia (1985–88) and 20 percent of GDP in Venezuela not otherwise have occurred.23 During the banking cri- (1994–99). These crises can retard the progress of eco- sis in Argentina in 1995, deposits fell by one-sixth in nomic growth for years. As the Mexican crisis of 1994 the first quarter of the year, and the central bank lost and the East Asian crisis of 1997–98 made clear, bank- $5 billion in reserves. The crisis was attributed in part ing and currency crises often come as a pair.17 to the collapse in confidence in Latin American finan- Liberalizing the capital account also influences do- cial markets that followed the Mexican crisis in Decem- mestic financial stability because portfolio investment ber 1994.24 The two recent financial crises in East Asia can be volatile.18 Latin America has seen its foreign cap- and Latin America suggest that geographic proximity ital flows rise and fall sharply. Net inflows were $60 bil- is an important determinant of financial contagion. lion in 1993, but in the wake of the Mexican crisis in “Institutional proximity,” or similarities in legal and 1995, net outflows reached $7.5 billion. Access to a regulatory systems, and exposure to the same shocks growing pool of global capital can mean more volatil- may also be factors. Countries thus have an interest in ity in emerging financial markets and greater exposure ensuring that the financial systems and macroeconomic to changes in sentiment by institutional investors in in- policies of neighboring countries do not increase the dustrial countries, too. Many empirical studies have likelihood of a financial crisis and induce contagion. demonstrated the sensitivity of portfolio flows of for- Potential spillovers across countries provide a com- eign capital to interest rates in industrial economies. pelling rationale for regional cooperation and coordi- Increases in interest rates in industrial countries raise nation in macroeconomic policy, banking standards, the probability of a banking crisis in developing and and the enforcement of bank regulations—a proposal transition countries, for three reasons.19 First, to retain explored later in this chapter.25 investments from industrial country investors who can Recent cross-country studies find that imposing cap- now realize higher returns at home, banks in develop- ital controls has little effect on economic growth.26 One         Box 3.2 What causes financial contagion? During a financial crisis elsewhere, contagion is said to have oc- through the attitudes of investors worldwide, is the response curred when a country succumbs to a financial crisis for rea- of mutual fund managers to country crises. Fund managers can sons other than a change in its fundamentals. The crises that spread financial volatility in several ways: began in Mexico during 1994 and Thailand in 1997 spread rapidly around the world. These crises had a major effect on fi- Ⅲ Emerging market fund managers often allocate their portfo- nancial markets, labor markets, and output in a range of other lios across different countries according to percentages countries in different regions—even half a world away. specified beforehand. When the value of investments in one What causes financial contagion?27 The series of events country drops, one manager’s response might be to sell could begin with a country that experiences a currency deval- stocks in other emerging markets to rebalance the portfo- uation, perhaps as the result of a combined bank and currency lio, depressing stock prices and putting pressure on curren- run by foreign investors. That country’s export goods become cies in all the countries in which the manager invests. cheaper for foreign consumers to buy, and other countries that Ⅲ Fund managers facing losses from investments in one coun- export the same goods find themselves at a competitive dis- try may have liquidity problems, forcing the sale of invest- advantage. The latter countries then come under pressure to ments in other markets. devalue their exchange rates. In 1997 and early 1998 many Ⅲ Investors, especially in emerging markets, find information feared that East Asian countries, in an attempt to shore up ex- on the prospects of a company or a country costly to collect. port sectors against regional competition, would engage in This difficulty encourages herd behavior: the disposal of rounds of “competitive devaluations” that would damage the stock by one investor is assumed to be based on news that economic prospects of every country involved. is not yet widely known, so other investors interpret this ac- These sorts of trade and exchange rate effects emanated tion as a signal to sell their own holdings. The lack of infor- from the Thai devaluation in 1997 and helped spread the East mation also encourages investors to take news of poor per- Asian crisis. But they cannot explain the depth or breadth of fi- formance in one emerging market as a signal that bad news nancial contagion. An alternative cause, which is disseminated is imminent in similar markets. plausible interpretation of this finding is that the bene- tal accumulation, productivity, and economic growth.30 fits of having access to a global pool of capital—like the This evidence and the frequent banking crises develop- opportunity for adding to investment capital or diversi- ing countries experience suggest that a robust banking fying risks—have been offset by the costs of the crises fi- regulatory framework offers substantial payoffs. Such a nancial liberalization causes. While cross-country regres- framework would ensure that bank managers and own- sions are always open to scrutiny, they do underscore the ers balance the costs and benefits of risk-taking behavior. difference between the evidence on the effects on eco- Striking the appropriate balance in designing bank nomic growth of trade liberalization and capital account regulations is difficult, however. Lax regulation raises liberalization. A wealth of studies exists on trade liberal- the risk that lending will move from the realm of mea- ization, all of them suggesting that it has many benefits, sured risk-taking to foolhardiness. But excessive bank but the evidence on capital account liberalization is regulation is likely to send funds flowing to the more much more mixed. The challenge is to devise policy and lightly regulated nonbank financial sector.31 This sec- institutional responses attractive enough to lure invest- tor is less likely to be associated with systemic failures ments that will have a significant positive impact on than banks, since severe bank failures lead to difficul- growth and, at the same time, to reduce the potential for ties with the payment mechanism. Yet this sector can costly financial crises. The rest of this chapter presents also breed financial instability, suggesting that at least an integrated program to do just that.28 some regulations may need to extend beyond the bank- ing system to other financial entities. Toward a more robust and diversified The growing complexity and diversity of banking banking system activities are straining bank regulatory resources every- Banking systems are especially important for raising and where, but especially in developing countries where allocating capital in developing countries, where the these resources are scarce. Private monitoring of banks banking sector typically accounts for a larger share of can complement formal regulations, and only a judi- total financial intermediation than it does in industrial cious combination of public and private oversight will economies (figure 3.5).29 Cross-country studies point to allow developing economies to reap all the possible the beneficial effects of a healthy banking sector on capi- benefits of financial liberalization.      ⁄  Figure 3.5 cerned about the security of their money must try to Bank intermediation typically accounts for a gauge the quality of their bank’s lending practices, larger share of the financial sector in developing which determine whether the bank is solvent enough to countries return deposits on demand. If many depositors—for Bank intermediation ratio, 1994 a good reasons or bad, based on good information or (percentage of total) poor—demand their deposits back at the same time, 100 Latin banks face a liquidity problem. When banks lend large 90 Asia America G-3 sums to each other, the resulting financial commitments can put pressure on a number of entities. If depositors 80 cannot differentiate between them, a run on one bank 70 may lead to runs on others, threatening the stability of 60 the entire financial system. To limit this possibility, gov- 50 ernments often insure deposits, guaranteeing depositors that they will get their money back and thereby reduc- 40 ing the incentive to start a bank run in the first place. 30 Central banks may also act as lenders of last resort to 20 help banks deal with short-term liquidity problems. 10 Deposit insurance has been criticized as contribut- ing to the fragility of the banking system, and without 0 the appropriate regulatory structure this can well be the an b Ma rea Th sia d an India Ind ina) ia lom le bia Ve xico ela l na es n azi an pa y es i tat Ch nti case. With deposit insurance, depositors simply put lay zu Ko Br ail (Ch Ja on Me dS ge ne rm of Co their money in the bank offering the highest return. A Ar Ge ite p. iw Un Re variant of Gresham’s law—with bad banks driving out Ta a. Ratio of the banking sector’s assets to the assets of all financial good banks—can occur; a bank that is willing to take institutions. greater risks with higher expected returns can offer b. The Universal banking system in Germany accounts for its very higher depositor rates; as funds flow to that bank, the high bank intermediation ratio. Source: World Bank 1997c. profitability of more conservative banks that invest in low-risk, low-return activities declines.34 Actually, the problem is not formal deposit insurance, as govern- In industrial countries an extensive legal and regula- ments will bail out any large bank because the risks of tory structure underpins banking operations. Laws systemic crisis are simply too great. Financial crises have protecting the rights of creditors permit banks to lend afflicted countries with and without formal deposit in- confidently and to collect deposits. Laws regulate bank- surance, as Sweden’s recent crisis bears testimony. In ruptcy and the recovery of assets and collateral, and ju- short, the moral hazard problem arises whenever there dicial proceedings implement such laws quickly and are large banks, and in most developing and transition impartially.32 Accounting and auditing standards help economies the concentration of banking activity is suf- in comparing investment projects and are prerequisites ficiently high that it is implausible that government for building efficient bond and stock markets. The rise would not intervene. of international bank lending increases the importance Not all deposit insurance schemes are alike, how- of global accounting standards.33 And because these le- ever.35 Some are more efficient than others, incorporat- gal and professional institutions take years to build, it ing practices that could usefully be emulated elsewhere. is important to begin constructing them now. In the Some governments limit deposit insurance coverage, meantime governments can develop regulatory frame- setting a ceiling on the size of deposits or the number works that address some of the special problems of of accounts that can be insured. Some collect premi- banking activities in developing countries. ums from all banks on a regular basis, rather than im- posing levies on surviving banks after a crisis. This last Why are deposits insured? practice is particularly pervasive, since leaving the sur- Banks borrow money on a short-term basis from depos- vivors to pick up the tab gives banks no incentive to itors and lend it out for longer periods. Depositors con- avoid collapse in the first place. Theoretically, deposit         insurance premiums can be linked to the risk level of a lending only to certain industries or regions. To the ex- bank’s portfolio or to the proportion of nonperforming tent that they prevent a bank from maintaining a well- loans. But to date, no government has tried this idea. diversified loan portfolio that balances risks in one industry or region against risks in others, such restric- Regulatory incentives to reduce risk-taking tions should be avoided. This concern is particularly A banking regulatory structure deals with many aspects important for banks that lend in only one geographic of bank operations: the requirements for setting up a region and where most borrowers are in the same bank, the services banks can provide, the levels of capi- industry. In such situations a collapse in prices that tal they must hold, the reserves they need to protect threatens the industry’s solvency will also affect the sol- themselves against nonperforming loans, and the liquid- vency of the banks. ity levels they must have to handle withdrawals. The reg- Two other challenges faced in designing appropriate ulatory structure defines the terms for disclosing a non- bank regulation are worth noting: competing jurisdic- performing loan, governs the portfolio composition of tions over banks, and close links between provincial banks, and specifies remedial measures in the event of banks and subnational governments.41 To avoid dupli- deteriorating loan portfolios or bank runs. As the num- cation of subnational and national regulatory resources, ber and variety of services banks offer increase, regula- subnational pressure for regulatory forbearance, and tors need to respond to the possibility that problems can the offer of implicit guarantees by subnational govern- occur simultaneously in many areas. ments, there is a strong case for executing bank regula- The reluctance of ever-hopeful regulators to control tion at the national level. risk-taking or to preemptively close banks with deterio- rating loan portfolios has made many banking crises Establishing private incentives to reduce risk-taking worse.36 For this reason, creating mechanisms that limit Private incentives that complement the framework of such “regulatory forbearance”—the term for putting off government regulation can help align the costs and tough actions in the hope that the bank will recover on benefits of the risks banks take. Banks can, for instance, its own—is another important step governments must periodically issue a special category of subordinated take to make bank regulations more effective.37 Some debt that is not guaranteed by the government. Since governments have already begun to remedy this prob- those holding subordinated debt lose their capital if a lem by insisting on independent audits of banks’ bal- bank defaults, they have a powerful incentive to moni- ance sheets, punishing failures to disclose nonperform- tor the riskiness of bank lending practices.42 But unlike ing debt in a timely fashion, and fining (or closing) holders of bank equity, holders of subordinated debt do banks that do not meet their capital adequacy require- not see higher returns if a bank increases its revenues ments. After its banking crisis in 1982, Chile intro- by making high-risk loans, since the market sets the ini- duced reforms specifically intended to reduce regulatory tial rate of return on subordinated debt.43 forbearance by increasing regulators’ autonomy and Banks wanting to reduce high interest payments to mandating public disclosure of the activities of both reg- those holding subordinated debt (especially because ulators and banks. Chilean law also proscribes links be- high interest rates send a signal to depositors and gov- tween insured banks and business conglomerates.38 ernment regulators) have an incentive to establish mon- The growing number of banking crises calls into itoring and disclosure practices that regularly report on question the merits of certain other government policies. the quality of the bank’s lending portfolio. Chile and For example, governments have tried to encourage lend- Argentina have adopted some of these practices.44 ing to targeted industries either by guaranteeing loans or by simply directing banks to make loans.39 Some Credible banking reform commentators on the East Asian crisis argue that these A new bank regulatory system may well face credibility initiatives have created implicit or explicit government problems, especially in countries with histories of di- guarantees.40 In these situations banks have little incen- rected government lending, regulatory corruption, and tive to carefully screen loan applications for favored proj- recurrent banking crises. Arm’s-length relationships be- ects, a lapse that often results in widespread default. tween regulators and regulated may well be a novel Banks are also sometimes restricted in the types of idea, along with the notion that strong interventions loans they can make. Often these restrictions permit occur automatically and without regulatory discretion      ⁄  when a bank fails to meet its legal obligations. Devel- in the form of additional pressure for appropriate risk- oping countries can improve the credibility of new taking by domestic banks. Admitting foreign banks is bank reforms by adopting and enforcing international no panacea, but if it is carefully timed and the economy banking standards. The various accords of the Banking can withstand the short-term disruptions, the benefits Regulations and Supervisory Practices Committee of can be considerable. the Bank for International Settlements, more widely Governments can foster the transfer of skills and best known as the “Basle Accords” or “Basle Standards,” can practices to their countries by allowing high-quality in- provide such standards. ternational banks with impeccable reputations to sup- Many argue that the current Basle Accords do not go ply domestic markets with financial services.47 This step far enough and in fact are now being revised.45 Critics requires governments to give foreign banks the right to say they do not do enough to discourage directed lend- establish themselves and to permit the immigration of ing, promote transparency (through the publication of skilled banking personnel. These international banks regulatory standards), or minimize the risks of regula- inevitably recognize that local bankers have a better tory discretion. The standards have also been criticized knowledge of the domestic economy, business practices, for recommending relatively low capital standards for and customs—and so offer them employment. Over developing countries that may face significant external time, local bankers will learn from the practices of the shocks.46 But developing countries can draw up a mem- international banks and acquire skills that they retain orandum of understanding with international finan- when they move back to domestic banks. cial bodies like the World Bank and the International The benefits of admitting foreign banks are not lim- Monetary Fund (IMF) adopting standards stronger ited to the transfer of skills and technology. Foreign than those in the Basle Accords. Or, given the risks of banks can stimulate competition, encouraging all banks regional contagion, neighboring countries can create to lower margins and overhead costs. A recent study of stronger voluntary banking standards for the region. the effects of foreign banks on the banking systems of Adopting internationally recognized banking stan- 80 countries found that in economies with relatively dards does more than just stabilize the banking system. large numbers of foreign banks, domestic banks have There are other payoffs, such as reduced borrowing lower expenses. However, domestic banks also have costs for domestic banks, which will be considered safe lower profitability.48 The findings suggest that the tim- risks. Realizing the payoffs is likely to require some ex- ing of foreign bank entry should be considered care- ternal monitoring of the country’s compliance with the fully. It would be highly undesirable if a rise in foreign new standards. For example, if a group of neighboring competition caused domestic banks to expand their countries agrees to a set of voluntary standards, the portfolio of high-risk loans in a desperate attempt to agreement can include a mechanism for periodically in- stave off default.49 vestigating compliance. This mechanism may be simi- Foreign banks are generally more diversified than lar to the Trade Policy Review Mechanism of the World domestic banks and can better withstand the effects of Trade Organization (WTO). An impartial body con- internal shocks. A severe macroeconomic downturn can ducts an investigation and, after a nonconfrontational push a domestic bank into default. But if a foreign bank discussion among the countries involved, publishes a has assets in healthy economies, a macroeconomic report on its findings. The country under investigation shock in the host country is likely to be less damaging. can produce a rejoinder that includes commitments for Of course, this benefit works only if the business cycles further reforms. These reports are available to investors, of the various countries differ. Economic shocks can be enabling them to better differentiate among countries. region-specific, continent-specific, or industry-specific. Ultimately, such a system reduces the likelihood of In such cases developing economies can expect little banking crises and financial contagion by inducing benefit from diversification if their foreign banks are countries to conform to higher banking standards. from the same region or continent or from countries with similar production structures. Another warning A role for foreign banks concerning the admission of foreign banks: events Allowing foreign banks to enter a country can disrupt abroad will affect the banks’ willingness to lend in the the domestic banking sector in the short term. But the new host country. For example, lower real estate and presence of foreign banks also offers long-term benefits stock prices in Japan in the 1990s led to reduced lend-         ing by Japanese banking subsidiaries in the United restraint in short-term foreign borrowing is a matter of States.50 In general, however, the risks posed by an un- government will. In the Mexican crisis, for example, diversified banking system overshadow this possibility. state entities were heavy foreign borrowers.54 Private A final benefit of admitting foreign banks is that the demand for short-term foreign debt should not be en- presence of these banks conserves on scarce administra- couraged with preferential tax treatment, as happened tive and bank regulatory resources in developing coun- in Thailand with borrowing through the Bangkok In- tries. Foreign banks are traditionally regulated by au- ternational Banking Facility. thorities in their home country. If foreign banks are A more aggressive way to limit short-term foreign allowed to take over domestic banks—or to buy domes- borrowing is to directly influence capital inflows.55 tic banks in privatization sales—regulatory responsibil- This discussion focuses on controls on inflows because ities are transferred abroad, and domestic regulators can controls on outflows are typically ineffective.56 One concentrate their resources on the remaining domestic method of circumventing outflow controls has multi- banks. This scenario highlights the need for a clear al- national firms selling goods to overseas parent compa- location of regulatory responsibilities across interna- nies at very low bookkeeping prices, transferring value tional borders. out of the country. Foreign investors wanting to cir- cumvent the controls also sometimes swap their funds The orderly sequencing of capital for the overseas assets of a domestic resident. account liberalization A scheme that provides disincentives for short-term Improving bank regulation would be an important pol- capital inflows has been in place in Chile since 1991.57 icy step even if world financial markets were not becom- This scheme imposes a one-year unremunerated reserve ing increasingly interconnected. However, the safety and requirement on all foreign inflows that do not increase security of a developing country’s banking system mat- the stock of physical capital, such as foreign loans, fixed ters even more in light of the volatility of international income securities, and equity investments. A portion of capital flows. The question then becomes one of find- any such inflows must be held in a non-interest-bearing ing a way to fit bank regulation into national strategies account for one year. The amount was initially set at 30 to liberalize the capital account. The macroeconomic percent, but it was lowered to 10 percent in June 1998 crises in Mexico and East Asia following the liquidation and subsequently to zero. The requirement remains on of short-term capital holdings by foreign investors has the statute book and can be reinstated, however. This rekindled interest in proposals for a measured, sequen- experience demonstrates that such a requirement can be tial approach to capital account liberalization.51 varied in order to stabilize the level of capital inflows. This discussion identifies a number of pitfalls devel- Rather than targeting specific types of capital inflows— oping countries face as they consider liberalizing their a measure investors can easily circumvent by rela- capital accounts. Each of these pitfalls must be side- beling—this scheme provides a sharp disincentive to stepped in order to minimize the risk of a financial cri- investing for less than one year.58 Empirical studies sis. Of course, developing countries differ substantially suggest that the effect of this tax has been to alter the in the nature of their legal institutions, corporate gover- composition of capital inflows toward less “footloose” nance practices, banking regulations, capital market de- foreign direct investment, although evidence on the velopment, and macroeconomic conditions.52 A unique overall impact on the level of capital inflows is mixed.59 recipe for sequencing capital account liberalization is Countries may be able to reduce their exposure to therefore unlikely to exist. Instead, the formula will vary changes in the sentiments of foreign portfolio investors across countries, dictated in part by how quickly coun- without banning such investment outright. Then, as tries can correct macroeconomic imbalances and enforce governments strengthen their bank regulation systems, credible financial regulations. they can gradually lower the nonremunerated deposit A key element of the sequential approach involves requirement. This approach reduces an economy’s vul- devising policies that control the demand for short- nerability to capital outflows by limiting certain of the term foreign debt.53 This type of foreign capital is the original inflows. most likely to flee, destabilizing the banking sector and In addition to modulating short-term foreign bor- the entire economy. Policies affecting short-term debt rowing, governments must decide how to treat foreign are best implemented before the inflows occur. In part, currency deposits in their domestic financial systems.      ⁄  Such deposits often account for a substantial percent- change rate regimes do provide different incentives to age of the broad money supply in developing countries potential borrowers of foreign short-term capital. In and in fact exceeded 30 percent in 18 countries in particular, a fixed exchange rate regime offers what 1995.60 While so-called dollarization undoubtedly has some interpret as an implicit guarantee to borrowers many implications for macroeconomic management, that they can ignore the risk of changes in the exchange the focus here is on its effects on financial stability and rate. Coupling fixed exchange rate regimes with deposit the implications for capital account liberalization.61 insurance is tantamount to relieving foreign depositors In a fractional reserve banking system, a rapid expan- of much of their credit risk.65 Such guarantees encour- sion of foreign currency deposits increases the liabilities age capital inflows, potentially exacerbating an econ- of domestic banks’ loan portfolios. The risk involved omy’s dependence on short-term foreign debt. More stems from the fact that the amount of net foreign cur- troublesome still, when investors call these guarantees rency in the economy is much lower than the total vol- into question, substantial capital outflows are likely. ume of foreign currency–denominated assets and lia- The exchange rate regime is then in jeopardy unless the bilities. Faced with a run on foreign currency deposits country has enough foreign reserves to cover the out- in the domestic banking system, the central bank may flows. Apparently, the preconditions for successfully come under pressure to act as lender of last resort and maintaining a fixed exchange rate are more stringent provide substantial loans in foreign currency to domes- than was previously thought. tic banks. 62 But these loans require the central bank to In contrast, flexible exchange rate regimes provide hold a relatively high level of costly foreign currency re- incentives for investors to take exchange rate risk into serves. In addition, the liquidation of foreign currency account and offer no protection against a fall in the ex- deposits may affect the exchange rate and the solvency change rate. As the experiences of Mexico in 1995, of domestic firms that have borrowed in foreign cur- Thailand in 1997, and Indonesia in 1998 show, the vi- rency. These factors argue for discouraging holdings of ability of a national banking system can be threatened foreign currency deposits in banking systems with rudi- when corporate borrowers face insolvency because a de- mentary regulatory oversight, by means of taxation or valuation of the national currency substantially in- higher bank capital-adequacy requirements. creases their foreign currency exposure. Financial crises Developing countries can also reduce the risk of fi- are certainly possible in flexible exchange rate regimes, nancial and economic crises from capital outflows by but these regimes create more incentives for investors maintaining high levels of foreign currency reserves.63 to take account of exchange rate movements than fixed The necessary level of reserves will depend on the coun- exchange rate regimes. Exchange rate regimes also dif- try’s level of international trade and on the amount of fer in the options available to policymakers when fac- footloose capital invested in the economy. Countries ing a surge of capital inflows—an issue discussed in the with enough reserves send a signal to investors, who World Bank’s Global Economic Prospects 1998/99. know they can convert their assets into foreign curren- The extent of macroeconomic instability and im- cies at the prevailing exchange rates. This knowledge balances suggests that other considerations are impor- reduces the risk that investors will all stampede for the tant in determining the appropriate pace of capital ac- exits at the same time because they fear a currency count liberalization. Although the consequences of crash.64 But accumulating reserves comes at a price. liberalization may depend on the exchange rate regime, Usually, domestic consumption and investment must removing barriers to capital flows at a time when a be limited so that exports exceed imports and the net massive inflow or outflow of funds seems likely is im- receipts are retained. Alternatively, reserves can be bor- prudent. For example, an outflow can be precipitated rowed by issuing long-term bonds, in which case the if capital account liberalization occurs during a period cost equals the difference between short-term and long- of high inflation, when domestic investors prefer sta- term interest rates. ble returns overseas. The choice of exchange rate regime is another im- The objective of a measured policy of sequential portant element affecting the sequencing of liberaliza- capital account liberalization is to gradually increase a tion. Of course, which exchange rate regime best serves national financial system’s tolerance for external disrup- a country’s interests depends on many considerations tions. While governments are building domestic capi- other than the regime’s compatibility with capital ac- tal market institutions (like bank regulation), they can count liberalization. However, different types of ex- also focus on ways of reducing exposure to changes in         the sentiments of holders of foreign debt instruments— vestment and technology, workers must be sufficiently so long as the methods chosen do not scare off too well educated—often with industry-specific skills— much long-term foreign investment. and able to continue to learn.69 And as foreign investors increasingly discriminate between regions and cities Attracting foreign investment within countries, the payoff to subnational govern- Long-term foreign investment will continue to provide ments of improving local systems of education and developing countries with important benefits. Public training increases still further. sector infrastructure projects will be in ever-greater de- mand in expanding cities, and governments and do- Liberalizing the trade policy regime mestic savers need not be the sole sources of financing. Foreign direct investment has a more profound impact In the private sphere the benefits of long-term for- on growth in countries that pursue policies promoting eign investment begin with the expansion of the host exports than it does in countries that follow import- country’s capital stock. However, since multinational substitution policies.70 The reason may be that foreign- corporations are responsible for most foreign direct in- owned companies aiming for global competitiveness vestment, there are other benefits as well. This invest- and international markets have a greater incentive to ment enhances competition in domestic markets, so re- bring in technology and training—with the accompa- sources are allocated more efficiently and domestic nying spillover benefits. In East Asian countries, foreign firms invest more. Foreign direct investment that in- direct investment has played an important role in bol- volves joint ventures or licensing arrangements between stering advanced manufacturing exports and output. In local and foreign firms often transfers technology66 and Korea, for example, foreign affiliates accounted for be- best practices to the host nation, stimulating produc- tween 65 and 73 percent of output in the electrical and tivity growth.67 (The importance of foreign direct in- electronics sector.71 vestment to Egypt and Tanzania is taken up in two case An open trade policy is also important for attracting studies in chapter 8.) foreign direct investment. Surveys of Japanese firms How can countries attract foreign investment? This which had decided to invest abroad found that a positive discussion presents several of the most effective meth- perception of policies governing such investments was a ods: adopting complementary human capital policies, strong determinant of plans to invest in a country and liberalizing the trade policy regime, avoiding induce- that low trade barriers made it more likely that multina- ments for foreign investors, creating a stable set of tional companies would enter a country.72 When first- rights and responsibilities for those investors, and de- rate information technology systems reinforce liberal veloping stock markets as alternative funding sources. market access, a country is further integrated into the world economy and becomes still more attractive as a Adopting complementary human capital policies destination for investment. A survey of international One recent study found that countries with low levels firms in Hong Kong (China), Singapore, and Taiwan of education and low rates of foreign direct investment (China) found that the presence of advanced infrastruc- grow much more slowly than countries with high edu- ture was the most important consideration in choosing cation rates and levels of inflow.68 Countries whose to locate regional headquarters and service and sourcing working populations have less than an average of five operations in a country, and the second most important months of secondary schooling and whose levels of for- factor in siting production. Foreign direct investment is eign investment are less than 0.1 percent of GDP have increasingly connected more with trading opportunities annual growth rates of less than 1 percent. But coun- than with local market exploitation.73 For example, the tries whose workers have an average of more than one huge increase in foreign direct investment in Mexico after year of secondary schooling and inflows worth more the North American Free Trade Agreement (NAFTA) than 0.2 percent of GDP enjoy, on average, annual came into force is evidence that the country is seen as a growth rates of 4.3 percent. Countries with high edu- desirable base for supplying the U.S. market. cational levels but low foreign direct investment, or Export-oriented development means that invest- with low educational levels but high foreign invest- ment decisions depend less on the scale of home mar- ment, do little better than countries that score low on kets, since firms are looking to sell in the global mar- both measures. These results may in part reflect the fact ketplace. Because multinational corporations are no that if labor is to facilitate continuous transfers of in- longer tied to domestic markets, they have more flexi-      ⁄  bility in choosing locations. Both points suggest that icy, which can be designed to induce foreign invest- stable and attractive economic policies have become ment. Chapter 8 describes Hungary’s successful efforts much more important. In fact, foreign direct invest- to attract foreign buyers for its formerly state-owned ment seems to be responding faster to economic factors banks. The second involves a country’s obligations than it has in the past.74 under the WTO’s General Agreement on Trade in Ser- vices. These obligations may include commitments to Avoiding inducements for foreign investors allow foreign firms access to certain domestic service Not all measures to attract foreign direct investment markets, as chapter 2 notes. have enhanced national welfare. In an assessment of Even if a nation implements sound macroeconomic 183 foreign direct investment projects in 30 countries policy, market liberalization measures, and clear legal over the past 15 years, one recent study found that be- rules, it is not always possible to ensure that successor tween 25 and 45 percent of projects had a negative net governments, including subnational governments and impact on national welfare. 75 This unwelcome and un- their agencies, will honor the commitments of their expected finding reflects the fact that foreign direct in- predecessors over the long term. This risk can limit the vestment is often accompanied by distortive policies. attractiveness of investments with high set-up costs and Such policies include requirements that producers use long payback periods, such as urban infrastructure proj- a specified number of domestic inputs; trade protection ects. The growing activities of subnational governments against imports that compete with the goods produced may exacerbate this problem (box 3.3). by foreign investors; financial inducements, subsidies, A dispute settlement mechanism can help resolve or tax holidays; and mandated joint ventures and tech- the issue of commitment. International arbitration is nology licensing arrangements. At least some of these often the preferred option. Arbitration clauses can be policies may encourage investment, but for society as a included in investment agreements with subnational whole the losses all too often outweigh the gains. Yet entities. In certain situations arbitration under the aus- another problem arises when urban centers and other pices of the International Centre for Settlement of In- subnational entities compete for investment, often en- vestment Disputes (ICSID) can be made available to gaging in inefficient beggar-thy-neighbor competition subnational governments that contract with foreign in- to provide public subsidies and incentives. National vestors. Almost 1,000 bilateral investment treaties and governments can play a role here in restricting the types 4 multilateral investment treaties contain clauses pro- of inducements that subnational governments can offer viding for binding arbitration under the ICSID. Some foreign investors. of the bilateral treaties explicitly state that their provi- sions cover acts and omissions of local governments in Creating a stable set of rights and responsibilities states signing the agreements. for foreign investors In the end, long-term investment agreements that National policies and regulatory institutions help fos- are balanced and mutually beneficial may be the most ter a climate conducive to foreign direct investment by lasting safeguards. Providing specialized training to in- multinational corporations. Taking steps to clearly de- crease local governments’ capacity to negotiate fair agree- fine the rights and obligations of multinational in- ments in the first place can advance this objective. The vestors is a start. Many developing countries are taking International Development Law Institute in Rome steps to create such legal frameworks and to simplify trains developing country lawyers to deal effectively bureaucratic procedures. This sort of institutional re- with foreign investors and lenders, and a number of form is especially attractive to investors considering in- World Bank initiatives also work to ameliorate this vesting in countries plagued by political risk and cor- commitment problem (box 3.4). ruption, since these practices are negatively associated The collapse of negotiations on a multilateral invest- with foreign direct investment.76 Countries that reduce ment agreement in 1998 suggests that a global treaty on red tape and bureaucratic delays not only make them- investment rules is still some way off. However, the num- selves more attractive to investment but help their own ber of bilateral and regional investment agreements and producers as well.77 treaties has increased. Signatories to these agreements re- Two other types of domestic regulations and com- alize that extending protections to foreign investors pro- mitments have particularly important ramifications for vides an incentive to cosignatories not to renege on long- foreign direct investment. The first is privatization pol- term deals with their own foreign investors. Since most         Box 3.3 Box 3.4 Subnational governments face commitment Mitigating the commitment problem: the role problems, too of the World Bank A U.S. company agreed to build the Dabhol Power Project, The World Bank has provided loans to host governments which would supply the Indian state of Maharashtra with to fund their obligations under political risk guarantees that 2,000 megawatts of power over a 20-year period.78 After are in turn issued to foreign investors. The Bank also of- the agreement was signed in 1993, the foreign investor fers lenders a guarantee that covers the risks of debt ser- began to incur heavy expenses for the construction of the vice defaults resulting from the failure of host govern- power station. The state government officials who signed ments to perform specified obligations in respect of the this contract lost the 1995 election, in which the invest- project. When issuing this guarantee, the Bank requires ment project had become a contentious political issue. that host governments sign a counterguarantee to reim- The new state government canceled the project, and only burse the Bank for any compensation the Bank pays the after 10 months of negotiations and several concessions foreign investor(s). Unless the host government plans to by the foreign investor was a new agreement signed. default on its obligations to the Bank (jeopardizing its en- Many argued that the original agreement was too gener- tire relationship with the World Bank Group), this counter- ous to the investor, and the fact that the company did not guarantee diminishes the government’s incentive to break abandon the project but instead chose to renegotiate its contractual obligations. offers some evidence for this view. With renegotiation, The Multilateral Investment Guarantee Agency (MIGA) the formal cost of construction fell from $1.3 million per provides foreign investors with insurance against losses megawatt to $0.9 million per megawatt.79 Canceling a proj- from war and civil disturbances, expropriations, and cur- ect the previous administration had agreed to was clearly rency inconvertibility. When a foreign investor cannot en- not the best way of attracting further foreign investment force a contract with a host government in that country’s to the sector. The investor reported that the delay cost ap- courts, MIGA can insure it against losses caused by the proximately $250,000 a day, and the international financial breach of contract. Between 1991 and March 1996, MIGA press gave the crisis extensive coverage. issued 30 contracts involving approximately $3.5 billion in This case shows how the proliferation of assertive sub- infrastructure projects. These contracts are in addition to national entities, which this report identifies as one of the those supplied by private insurers, which now offer con- chief political reactions to localization, can complicate the tracts for “breach of undertaking.” efforts of national governments to make binding commit- In 1992, at the request of the Development Commit- ments. If foreign investors cannot discriminate among tee, the World Bank Group issued a set of guidelines em- subnational entities in a given nation, the actions of one bodying commendable approaches to the legal framework entity may be seen as reflecting the behavior of all others. for the treatment of foreign investment. The guidelines This kind of spillover is a serious concern for national gov- cover the main areas dealt with in investment protection ernments keen on attracting foreign direct investment. treaties: the admission, treatment, and expropriation of foreign investments and the settlement of disputes be- tween governments and foreign investors. By their terms the guidelines are not binding and are intended to comple- foreign direct investment is intraregional—with devel- ment applicable international agreements. Moreover, by oping countries now investing substantial amounts their terms they are intended to apply to both states and abroad and so recognizing the need to protect their in- any of their constituent subdivisions. vestments—an even greater role for regional investment agreements is likely to emerge. When these investment accords include commit- tify them, and negotiate common guidelines for their ments to maintain domestic reforms, the reforms are use. Signatories can then negotiate additional con- more credible. Reversing the reforms once the accords straints later on, in much the same way as signatories to are signed would do more than wreak domestic havoc; international trade agreements have renegotiated tariff it would also invite retaliation by foreign governments. levels. These agreements also reduce incentives to en- NAFTA’s investment provisions in effect “locked in” gage in beggar-thy-neighbor policies to attract capital. Mexico’s domestic regulatory and institutional reforms. They allay fears that countries may be tempted to re- Similarly, the Mercado Común del Sur (MERCOSUR) duce environmental and other important protections in preferential trade agreement reinforced reforms in Brazil return for the promise of an investment project (the so- and Argentina and stimulated foreign direct investment called “race to the bottom” syndrome). from other countries, principally the United States.80 Regional foreign investment agreements can also Developing stock markets as alternative include constraints on the use of subsidies, tax induce- funding sources ments, and regulatory competition. The initial agree- Although foreign portfolio investment does not offer ment can identify accepted forms of favoritism, quan- the same opportunities for technology transfer and in-      ⁄  creased competition as foreign direct investment, it can on individual countries. The contributions of regional also be very useful to developing countries. Opening and global agreements to foreign direct investment and stock markets to foreign participation increases liquid- financial supervision have already been discussed. But a ity by deepening the pool of buyers and sellers. Price- corollary to the growing trend toward a globalized econ- earnings ratios rise as liquidity increases, making the omy exists. As economies become increasingly interde- market a far more attractive source of equity financ- pendent, the effects of national policy decisions spread, ing.81 As the stock market develops and strengthens, it with ramifications—including potentially disruptive benefits other parts of the financial sector as well as the ones—for other countries.85 Although the interdepen- wider economy—foreign direct investment accompa- dencies are typically strongest among neighboring coun- nies stock market purchases, for instance. Stock market tries, macroeconomic conditions in industrial economies development and banking development have a strong have distinct consequences for the rest of the world. positive relationship, as do stock market liquidity and Fluctuations in interest rate differentials between in- economic growth.82 dustrial countries alter the flow of capital to and from The potential volatility of a stock market is an on- developing countries, potentially destabilizing their fi- going concern. Many policies for reducing volatility in nancial systems. A variety of vehicles for international the banking sector can help reduce the volatility of cooperation could be considered that would enable in- bourses, however, and approaches to sequencing capi- dustrial countries to meet their own goals without buf- tal account liberalization can be applied to portfolio feting the outside world. equity flows as well. But as with other parts of the fi- The growing links among countries in the same nancial sector, the cause of stock market volatility is region also suggest a motivation for regional networks often a lack of reliable, up-to-date information. Accu- to prevent and fight financial crises.86 Because of the rate information from independent sources makes an growing trade and financial links among regional econ- emerging market attractive to foreign equity investors omies, one economy’s poor performance can profoundly and increases the stability of capital flows. Rules man- effect its neighbors. This fact argues for close monitor- dating the regular public reporting of financial posi- ing and mutual support among countries in the same tions in key areas such as investment, property and region. However, the growing strength of regional link- equipment, foreign currency operations, and long-term ages will cause national economic cycles within a region contracts reduce uncertainty.83 Financial markets de- to move more closely in phase. In this case the IMF’s velop best in the presence of legal codes that stress the function as an extraregional crisis management body rights of shareholders (especially minority holders) and will take on added importance, as countries in the same regulatory systems that encourage the disclosure of cor- region are likely to enter downturns together, reducing porate information.84 the resources they have available to help their regional During the next 25 years the flow of foreign invest- partners. ment to and from developing economies will increase One promising approach builds on the steps some substantially. Developing countries will have a growing countries are already taking toward regional economic interest in establishing secure and stable regimes that monitoring. The Association of Southeast Asian Nations protect their overseas investors—and that clearly delin- (ASEAN) agreed to implement an economic moni- eate their responsibilities. As the supply of capital grows, toring mechanism in November 1997. The mechanism subnational and central government entities will increase aims to monitor policies in “vulnerable” sectors, to im- their demands for capital to fund urban infrastructure prove economic policy coordination among members, projects. Developing economies can take action to attract and to assist members during a crisis.87 But doubts have and maximize the benefits of long-term foreign invest- been raised about this mechanism, with skeptics ques- ment by participating in regional agreements that en- tioning not only whether enough resources have been hance investor security and by maintaining stable macro- devoted to it, but whether governments will actually be economic, trade, and regulatory policies. willing to release timely information or to criticize each other’s domestic policies.88 This points to the difficulty Revitalizing international macroeconomic of sustaining cooperation in regional initiatives such as cooperation this and the Manila Framework. This sketch of international financial integration has de- When a regional grouping does establish a credible liberately avoided placing the entire burden of reform monitoring scheme to certify that members have im-         plemented commendable regulatory and macroeco- world. Since 1997, when the East Asian crisis began, the nomic practices, members can extend cooperation to world has learned that poorly managed financial liber- include pooling funds to deter speculative currency at- alization can lead to a protracted economic downturn tacks. This “seal of approval” helps investors differenti- and a renewed cycle of poverty. But the potential upside ate among member states. This pool of regional funds of international capital flows is enormous, as the posi- can be used to augment the national reserves of what tive contribution of foreign direct investment to boost- might otherwise become the “trigger economy” for a re- ing productivity in recipient countries demonstrates. gional crisis. If these additional reserves reduce the like- The discussion in this chapter has highlighted four lihood of a future devaluation of a country’s currency, essential and related measures for developing economies foreign and domestic investors will be less inclined to wishing to integrate into global financial markets. First, liquidate their portfolio investment in that country, even if an economy is completely isolated from foreign possibly preventing a currency run altogether. financial flows, the benefits of domestic financial liber- Countries can also explore opportunities for cooper- alization cannot be assured without strong banking reg- ation with regional partners during a financial crisis. ulation. Second, strengthening those regulations takes Crisis management accords can be signed in advance, years, and in the interim governments must develop providing investors with the expectation of a coordi- policies that reduce the volatility of short-term foreign nated response to shocks and helping allay the most inflows. Third, developing countries will want to in- pessimistic expectations. These accords can then serve crease their attractiveness to long-term foreign invest- as a framework for a coordinated fiscal policy of tax cuts ment. The rise of global production networks (dis- and spending increases that provides a safety net for cussed in chapter 2) shows that multinational firms are those most affected by shocks and stimulates the re- slicing up production processes, distributing them gional economy.89 The accords can also lay the ground- across economies. Large domestic markets are likely to work for commitments not to engage in competitive become less important to multinationals looking for devaluations or impair market access by raising existing new locations, creating opportunities for smaller devel- tariff and nontariff barriers. oping countries with suitable infrastructure and educa- tion. Finally, efforts to coordinate aspects of financial • • • and regulatory policies can be advantageous to devel- Internationally mobile capital is here to stay. Growing oping economies. Financial crises in developing coun- trade links, new communications technologies, and in- tries are not always homegrown. Fluctuating interest creasingly sophisticated financial products are making rate differentials between industrial countries have in- national borders more porous to financial flows. The creased the volatility of global capital flows, which can challenge facing policymakers in developing countries be ameliorated by policy coordination among industrial is how to navigate through this financially integrating countries. C h a 4 p t e r Protecting the Global Commons A t the end of the 20th century, environ- ple, is likely to raise sea levels, threat- mental problems are a matter of both na- ening island economies and low-lying tional and global concern. Many of them countries such as the Maldives and Ban- create spillovers that impose heavy costs gladesh. Climate change also jeopardizes not only on those close to the source of agricultural production in developing the problem but on society as a whole countries. The Russian Federation and and on future generations. Individual parts of Africa could see dramatic reduc- countries have strong economic and so- tions in their crop yields by 2050 (figure cial reasons for aggressively protecting 4.1). The overall impact of a doubling of their environments by creating incentives carbon dioxide in the atmosphere would to reduce and manage such spillovers.1 be to reduce the gross domestic product However, an important subset of envi- (GDP) of developing countries by an es- ronmental problems is global in scope. timated 2–9 percent (compared with Many countries have contributed to 1.0–1.5 percent of GDP in industrial these problems, and no individual coun- economies).2 Within developing coun- try can effectively address them by acting tries, the price of inaction is likely to alone. These are the problems of the fall particularly on the poorest, who have “global commons,” which will place all the fewest resources for responding to countries at risk if no collective action is climate change. And because of the con- taken. There are many such issues, in- centration of biodiverse areas in develop- cluding desertification, persistent organic ing countries, failure to preserve bio- pollutants, the fate of Antarctica, and the diversity would also disproportionately environmental health of the high seas affect poorer nations. and the seabed (box 4.1), but this chap- Despite the urgency and importance ter focuses on three in particular: ozone of environmental issues, building coop- depletion, global climate change, and eration to address global environmental threats to biodiversity. problems is not simple; it involves con- Effective responses to these problems tentious issues such as the division of re- are vital to the struggle for sustainable sponsibilities and differing capabilities to development. Climate change, for exam- respond. Industrial countries have cre-       ⁄  Box 4.1 Global environmental issues Beyond the three cases discussed in detail in this chapter— clear devices on the continent or in the surrounding seas. ozone depletion, climate change, and biodiversity protection— Since then, two conventions and one protocol to the treaty a range of other environmental issues calls for action on a have aimed to protect seals, the region’s unique marine living global scale. These issues include desertification and land resources, and the Antarctic environment in general.4 degradation, Antarctica, persistent organic pollutants (POPs), and the high seas and seabed. Persistent organic pollutants Twelve of these pollutants are currently the subject of interna- Desertification and land degradation tional negotiation. POPs are chemical substances used in a va- Today 900 million people in about 100 countries are affected riety of activities (including agricultural and industrial produc- by desertification and drought. By 2025 that number will dou- tion and disease control) that do not break down naturally and ble, and 25 percent of the earth’s land area will be degraded. that accumulate in the fatty tissues of animals at different lev- Land degradation, which is closely linked to issues of popula- els of the food chain. Because POPs are long-lasting and are tion, poverty, water use, and biodiversity, increases as grow- frequently able to travel long distances in the atmosphere, they ing numbers of people overexploit fragile ecosystems. have spread all over the world, even to areas where they have By mid-1998 almost 150 countries had ratified the United never been used. POPs harm both human and animal popula- Nations Convention to Combat Desertification. The convention tions—in humans, for example, they can cause cancer, dis- is a significant first step that will benefit millions of people if it eases of the immune system, and reproductive disorders. The is properly implemented. The convention’s thrust is not to set United Nations Environment Programme is leading the devel- up a separate program to counter desertification but to main- opment of a global, legally binding agreement to minimize the stream efforts toward this objective into a country’s overall de- release of POPs into the environment, with negotiations sched- velopment strategy, with the support of bilateral and multilat- uled to conclude in 2000.5 eral donors.3 The high seas and the seabed Antarctica The United Nations Convention on the Law of the Sea (UNCLOS), Since the negotiation of the Antarctic Treaty in 1959, countries which incorporated a number of earlier agreements, was adopted that had laid claim to territory on the continent have “frozen” in 1982 and entered into force in 1994. Beyond creating exclusive their claims. Under Article IV no signatory nation is allowed to economic zones (box 4.2), UNCLOS stipulates that states must assert its claims or make new ones. Furthermore, signatories take action to control marine pollution from both land-based are not allowed to deploy military units (except in support of sources and vessels at sea. It also sets up a global authority re- scientific missions), dump radioactive waste, or explode nu- sponsible for the environmental health of the seabed. ated much of the current stock of many transnational en- Even though industrial countries have played a dis- vironmental problems. In the pursuit of economic ad- proportionately large role in causing global environ- vance, they have destroyed much of their own biodiver- mental problems and should pay the lion’s share of the sity and have overexploited fisheries worldwide. They also costs of addressing them, developing countries are vital have the highest levels of energy use and thus bear the to any long-term solution to these problems and have overwhelming responsibility for the present level of man- accepted that they also have a role, under a system of made greenhouse gases in the atmosphere. At the same common but differentiated responsibilities.7 Develop- time, developing countries are unlikely to become ac- ing countries are already doing damage to the global tively involved in addressing global environmental prob- commons. Rain forests and coral reefs are rapidly being lems if the price is slower economic progress. The United destroyed in many developing countries. Urbanization, Nations Framework Convention on Climate Change industrialization, and growing numbers of automobiles and the Convention on Biological Diversity (both agreed worldwide mean yet more greenhouse gases in the at- at the 1992 Rio Earth Summit) specifically recognized mosphere. And overfishing has spread to seas controlled that economic and social development and poverty erad- by developing nations. Moreover, regardless of who has ication are developing countries’ overriding priorities.6 done the damage to the global commons, developing For this reason the need for flexible mechanisms that countries have a strong interest in ensuring that coop- transfer resources from rich to poor countries are central erative steps are taken to address these issues, which will to any solution of global environmental problems. have the greatest effect on their citizens. Figure 4.1 Climate change jeopardizes crop yields, especially in developing countries See inset below See inset below Estimated change in crop yields, 2050 Less than -4% -4% to -2% -2% to 0% Greater than 0% No data Note: Crops modeled are wheat, maize, and rice.     Source: Parry and Livermore 1997.       ⁄  Already, developing countries are taking steps to com- proves the economic efficiency and political viability of bat environmental degradation, including some environ- reforms designed to promote sustainable development. mental problems that have global implications. Ka- For example, the domestic environmental benefits of zakhstan and Uzbekistan have been taking measures to maintaining forest resources—including reduction of prevent rapid deforestation, and China has crafted an river sedimentation and soil erosion and preservation ambitious set of environmental plans based on the of water resources and fishing areas—greatly outweigh agenda that emerged from the Rio Earth Summit.8 any economic benefits that might be gained by trans- These efforts have overlapped with a growing movement forming the forest into poor-quality farmland. Simi- to tackle global environmental problems in a multi- larly, governments can justify preserving coral reefs national framework. Since the 1972 Stockholm Con- solely on the basis of their value to national econo- ference on the Human Environment, governments mies.10 Preservation, then, supports both the national have signed more than 130 environmental treaties, with environment and the national economy. But in both increasingly substantive regulatory provisions.9 These cases, efforts to protect national resources also benefit treaties have contributed to many positive developments, the global commons by preserving biodiversity and re- such as reduced water pollution in the Mediterranean ducing carbon dioxide output. and stronger protection for the Antarctic environment. Governments often take measures to promote eco- This chapter begins by discussing national initiatives nomic efficiency (on both the national and interna- aimed at improving the local economy or environment tional levels) that also reduce environmental degrada- that also have some role in slowing climate change and tion. Eliminating subsidies and tax credits for cutting biodiversity loss. Such initiatives illustrate the impor- timber and for building roads in forests is economically tance of the complementarities that can emerge from a advantageous. But this policy has another benefit: it sig- comprehensive development strategy. Policies designed nificantly reduces deforestation rates, preserving biodi- to improve economic efficiency, for instance, can some- versity and a valuable “carbon sink” that cuts carbon times have a significant and positive impact on rates of dioxide levels in the air.11 Similarly, doing away with en- deforestation or energy use. The chapter moves on to a ergy subsidies and imposing taxes on fuel reduces both discussion of the need for further international initia- global carbon dioxide emissions and local pollution tives that address regional and global environmental such as acid rain and smog. Studies in Mexico suggest problems. Although the measures employed to tackle that a 1 percent increase in gasoline prices is associated ozone depletion were based in part on circumstances with a 0.8 percent decline in gasoline consumption.12 particular to that case, they suggest guidelines for de- Eliminating energy subsidies could reduce carbon signing global measures that address the complex prob- emissions dramatically. If Western Europe and Japan lems of greenhouse gas emissions and biodiversity preser- abolished their coal production subsidies and their im- vation. The chapter concludes with a look at the linkages port restrictions on foreign coal by 2005, global carbon between biodiversity and greenhouse gas emissions, dioxide emissions would drop 5 percent. If the major pointing out how these links can be exploited to negoti- developing countries simultaneously raised the price of ate more effective international agreements. coal to international market levels, the combined effect would be an 8 percent reduction in global emissions.13 The link between national and global Removing subsidies is often difficult for political rea- environmental issues sons, but it is important to note that subsidies rarely Autonomous, self-interested state actions can improve benefit the most deserving, especially in the developing both the environment and economic performance, as world.14 For example, subsidizing the electricity bills of emphasized in World Development Report 1992. In rich consumers connected to the grid—or the gasoline some fortuitous cases, protecting the local environment of those who own cars—certainly does not help the will also contribute to addressing a global environmen- poor in developing countries. A recent World Bank tal problem. Exploiting these synergies is vital. Linking study found that in Malawi rich consumers receive actions that have short-term payoffs (such as control- $6.60 a year in electricity subsidies, while poor con- ling air pollution) to those with longer-term results sumers receive just $0.04. Of course, those not con- (such as controlling the release of carbon dioxide) im- nected to the electricity grid receive no subsidy at all.15      Even if the complete removal of subsidies is politi- that a carbon tax that applies to energy imports and cally impossible, there may be a strong case for better local sources of carbon-based energy such as coal mines targeting. The cost of protecting a German coal-mining and oil refineries might also be relatively easy to imple- job with per-ton subsidies reached $79,800 per job in ment, as only a limited number of industrial operations 1995.16 Much of the value of these subsidies went to the require monitoring. mine owners and operators, not to the workers. If the Another policy tool that can have a positive effect rationale for a subsidy is to protect jobs or workers’ in- on both the national and the global environment is the comes, a per-worker subsidy is a more efficient choice. imposition of market discipline on the exploitation of In Germany switching to a per-worker subsidy would natural resources.20 For example, making fishing quo- have raised the price of coal closer to market levels (re- tas tradable helps create a market that promotes the ef- ducing coal consumption) and decreased the overall cost ficient and sustainable use of fisheries resources (box of the subsidies while protecting the mine workers’ jobs 4.2). Market-based approaches are likely to be particu- and incomes. larly important in international environmental agree- Beyond national policies, local governments also ments, as discussed later in this chapter. have a role in countering global problems while tack- By removing or reforming subsidies, fostering mar- ling local issues. Automobile-related pollution does far kets, and confirming property rights, countries acting more damage in cities than in the countryside because alone can improve their own environments. To the of the high concentration of both cars and people in extent that these unilateral actions also reduce cross- urban areas. A recent U.S. study estimated that every border pollution and environmental damage, they im- gallon of gasoline consumed imposed a $0.10 cost on prove the welfare of other countries as well. But if such the country as a whole in terms of the damage caused actions are so advantageous, why have more countries by increased air pollution but that in Los Angeles the not taken them, and why are they not enough? amount can run as high as $0.62 per gallon.17 Such dif- Entrenched producer interests account for the po- ferentials suggest that local (and especially urban) gov- litical difficulty in removing subsidies. Even better- ernments have an important part to play in tackling targeted subsidies may meet resistance from workers. pollution issues (see chapter 7). By investing in effec- They may feel, for instance, that wage subsidies are de- tive public and nonmotorized transportation networks meaning to them in a way that price supports (which and providing people with the incentives to use them, are far less efficient) are not.21 This problem reinforces cities can reduce the economic and environmental costs a point made in chapter 2: that a primary policy con- of traffic congestion and motorized vehicle use. In the cern in the coming decades will be to help regional process, they also reduce greenhouse gas emissions.18 labor markets adjust to the economic changes caused Preserving the environment involves not only elimi- by reform. It also suggests that international agreements nating subsidies that encourage polluting activities and might play a role in stimulating domestic support for supporting more environmentally efficient alternatives environmental reform, much like the role the World but also ensuring that polluters pay for the environ- Trade Organization (WTO) assumes in encouraging mental damage they cause. These policies can fre- freer trade. quently be implemented in ways that help protect the But even if national-level environmental concerns global as well as the local environment and that mini- are fully addressed, international market failures call for mize the economic costs of environmental protection. an international response. Despite the sometimes posi- Carbon taxes, which are applied to energy sources ac- tive effects of national efforts on international well- cording to the amount of carbon dioxide they produce, being, a focus on local environmental issues frequently have been suggested as one way for industrial and de- leaves global concerns inadequately addressed. For ex- veloping countries to reduce greenhouse gas emissions. ample, catalytic converters can significantly reduce emis- Controversy often surrounds energy taxes. But propo- sions of local pollutants, cutting hydrocarbon emissions nents argue that such taxes sometimes have a broader by an average of 87 percent, carbon monoxide emis- base than other taxes commonly imposed in develop- sions by 85 percent, and nitrogen oxides by 62 percent. ing countries (such as those on trade) and so can be But depending on the type, these converters often have more economically efficient.19 Proponents also argue a minimal or negative effect on carbon dioxide output,      ⁄  Box 4.2 Preserving the ocean commons: controlling overfishing The imposition in the late 1970s of exclusive economic zones ever, is no panacea: it takes an estimated 5 kilograms of (EEZs) that stretch 200 miles from the coastlines of many oceanic fish reduced to fish meal to raise a single kilogram of countries has dramatically reduced the problem of fisheries as farmed shrimp, and the 300 to 1,000 kilograms of solid waste an international common pool resource, exploited by many and produced by each ton of farmed fish can cause problems with protected by none. Yet overfishing remains a significant issue. water quality, including overnutrification and algae blooms. But At the international level, regulating the stocks of migratory freshwater aquaculture at least can be made sustainable. fish that traverse the EEZs of several countries still presents For transnational or highly migratory stocks of fish or stocks problems. But since 90 to 95 percent of fish are found within that stray into the high seas, international agreements still play EEZs, such problems cannot account for global overfishing. In an important part in controlling overfishing. The 1995 United fact the most important causes of overfishing are national sub- Nations agreement on straddling fish stocks and highly migra- sidies, overcapacity in the fishing industry, and governments’ tory fish stocks struck a careful balance in determining the inability to enforce fishing limits in their economic zones. rights of coastal and distant-fishing states and strengthened In the underpatrolled waters off the coasts of some African the role of regional fishing organizations in controlling fishing nations, ships from both Europe and Asia fish illegally—and at on the high seas. Parties to regional agreements have been rates that cannot be sustained.22 But even legal fishing often given powers to board and inspect vessels from any nation, al- depletes local fish populations. Technological advances such though they have no power to impound the vessels or arrest as advanced sonar and drift nets have made large boats much the crews.23 Another regional solution is a register of foreign more effective. The Food and Agriculture Organization (FAO) vessels like that set up by the South Pacific Forum Fisheries estimates that the number of fishing boats more than doubled Agency. Ships must be on this register in order to obtain fish- between 1970 and 1990, reaching some 1.2 million (although ing licenses from any member country, and they can be re- many of these are small fishing boats). The European Union moved from the list for failure to pay fines.24 This type of co- alone has about 40 percent more boats than it needs to catch operation among states reduces the cost of enforcement. sustainable levels of fish. And as a result of overfishing, fish With the EEZs and the 1995 United Nations agreement in catches in recent years have not increased, despite the larger place, is broader international action needed to preserve fish- fleets. As stocks are exhausted, the fleets actually become eries? The United Nations Convention on the Law of the Sea less profitable. stipulates that countries have a duty to conserve fisheries Clearly, enforcing national rights, removing subsidies, and within their EEZs, although the obligations are not clearly implementing national programs to counter overfishing are spelled out.25 Some countries have apparently decided to allow very important. Some countries have introduced individual overfishing, thus placing a low value on future fish stocks. In- transferable quotas—tradable rights to land a percentage of ternational sanctions or transfers might change the incentives the annual catch—which, when well implemented, can ensure of the countries that continue to overfish. But for most devel- a sustainable catch for the most efficient fishermen. oping countries, support for more effective fisheries manage- Aquaculture may provide a technological solution for over- ment combined with voluntary sustainable fisheries labeling is fishing. While marine harvests still account for 80 percent likely to be more appropriate. A certification mechanism could of world seafood supplies, aquaculture is one of the fastest- also encourage sustainable fishing practices, an idea that has growing food production industries. Farmed fish production been taken up by the new Marine Stewardship Council. A fu- doubled between 1990 and 1996, reaching 26 million tons, and ture international agreement could also call for phasing out fish- output could reach 39 million tons by 2010. Aquaculture, how- ing subsidies, which clearly stimulate global overfishing. Aquaculture’s role in the production of fish and shellfish is growing Percentage of total production Millions of metric tons 30 125 Share of aquaculture (left axis) 120 Total catch, excluding aquaculture (right axis) 25 Total catch (right axis) 115 110 20 105 15 100 95 10 90 85 5 80 0 75 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Source: FAO, Fishery Statistics, various years; FAO, Aquacultural Production Statistics, 1999.      the source of so much concern because of its relation- recreational values (including tourism) of around $12 ship to climate change.26 per hectare per year. Estimates of the value of the vital Thus, national environmental policies are designed ecological functions of forests also overshadow agri- to benefit individual countries, not the rest of the world, cultural opportunity costs. These functions include and are likely to fall far short of global environmental watershed protection (around $10 per hectare per year), goals. For international cooperation in the environ- erosion prevention ($2–$28), fisheries protection (ap- mental arena to succeed, governments must consider proximately $14), and flood prevention ($2).28 These the implications of domestic environmental policy de- figures suggest that the most important method of pre- cisions for other countries. Recognition of the effect serving global biodiversity is to ensure that the func- that each nation’s policies might have on other nations’ tioning of markets and institutions at the national level welfare is an essential precondition for effective inter- reflects the value of the services ecosystems provide. national environmental cooperation. Technical assistance and knowledge transfer can sup- port this goal and are already a focus of international Moving from national to international action efforts to preserve biodiversity under the Global Envi- Every environmental issue involves a unique configura- ronment Facility (box 4.3). tion of scientific factors, stakeholders, costs, benefits, Nonetheless, at least some elements of biodiversity and policy implications. But all global environmental can be seen as nonexcludable and nonrivalrous, in com- problems have one thing in common: individual coun- mon with the atmosphere. Genetic material is arguably tries do not have sufficient incentives to act on them a global common resource, yet pharmaceutical compa- because countries cannot capture all the rewards of nies in industrial countries rarely pay for the genetic ma- doing so. In economic terminology, global environmen- terial they have extracted from plants in developing tal resources are public goods that are nonexcludable countries. A recent cost-benefit analysis of a preservation and nonrivalrous across borders. The atmosphere is a program for Cameroon’s Korup National Park rain for- particularly good example. No individual or group can est found that while many benefits of preserving the for- be prevented (excluded) from consuming or using the est could be captured at the national level, only around atmosphere. Furthermore, clean air does not benefit 10 percent of the genetic value of the forest’s biological one nation at the expense of others, so countries are not resources (including research material for pharmaceuti- rivals when it comes to consuming these goods. An op- cals, chemicals, and agricultural crop products) could be posite example is the sea: it can be divided into zones obtained by Cameroon through existing licensing struc- with boundaries that can be enforced, and at least in the tures and institutions. The rest would benefit others out- case of fishing, one nation’s use can be at the expense of side Cameroon. Furthermore, the study did not include another’s. the value of carbon storage (reducing carbon dioxide Biodiversity poses a slightly different problem from emissions) that forest preservation provides to the global that presented by the atmosphere. We cannot separate community. Carbon storage is both a useful example of what might be considered the global common resource the linkages among global environmental issues (since elements of biodiversity from the ecosystems in which preserving forests supports climate stability and slows they reside, and these are highly valuable at the national biodiversity loss) and another example of the nonrival- level. Forests and coral reefs both have usage values at rous, nonexcludable nature of some forest services.29 this level that far exceed any value that might be gained No system has ever been set up to pay for the “exis- by destroying them. A recent study in West Kaliman- tence value” of species in other countries—the value of tan, Indonesia, found that 95 percent of the forests in diversity independent of any expected economic re- the province have an agricultural opportunity cost of turns from factors such as genetic material or ecologi- less than $2 per hectare per year.27 This figure compares cal function. This scenario persists unchanged, even poorly with estimates of the benefits of forest preserva- though studies conducted in the United States suggest tion that can be captured at the national level. These a willingness to pay for the preservation of individual benefits include extractive values of minor forest prod- native species at prices that range from $2 to $150 per ucts (fruits, latex, medicines, and so on) that average household per year.30 around $70 per hectare per year, hunting and fishing When environmental resources have the features of values of between $1 and $16 per hectare per year, and a global public good, it becomes very difficult for pri-      ⁄  Box 4.3 half of the 20th century fall short of attaining the full The Global Environment Facility benefits of global cooperation. The Kyoto agreement is a solid first step away from “business as usual” and to- The Global Environment Facility (GEF) provides grants and ward adaptive management. However, calculations by concessional funds to cover the additional costs incurred when a development project also targets global environ- the Intergovernmental Panel on Climate Change (IPCC) mental objectives in four focus areas: biological diversity, show that emissions reductions well beyond the levels climate change, international waters, and depletion of the agreed at the Kyoto meeting would eventually be needed Earth’s ozone layer. The GEF is the interim financial mech- if governments wished to stabilize atmospheric concen- anism of both the Convention on Biological Diversity and trations of greenhouse gases at today’s levels over the very the United Nations Framework Convention on Climate Change. The GEF leverages its resources through cofi- long term.32 Specifically, a reduction in emissions of ap- nancing and cooperation with other donor groups and the proximately 60 percent from current levels would be re- private sector. quired for stabilization. At present, the members of the The GEF is involved in a range of innovative projects Organisation for Economic Co-operation and Develop- worldwide, including support for the management of pro- tected areas, conservation programs, biomass and energy ment (OECD) and the transition economies have agreed efficiency projects, solar home systems, and phaseout pro- to reductions of around 5 percent.33 This suggests that, grams for chlorofluorocarbons (CFCs). In the Czech Repub- if major climate change is to be avoided, there will at lic, for example, GEF support was central to the phaseout some point need to be an agreement with stricter emis- of production and use of ozone-depleting substances such as CFCs and their replacement with alternative tech- sions targets encompassing more countries. The situa- nologies. In a group of Caribbean countries, a GEF project tion is much the same with biological diversity loss. backed the implementation of the International Convention While the Convention on Biological Diversity provides for the Prevention of Pollution from Ships, which included a strong framework for future agreements, it has had lit- new legislation, regional cooperation among countries and with cruise lines, and improved port waste management tle effect on forestry practices and coral reef degradation. systems. Later in this chapter we discuss a project in Unsustainable forestry practices have slowed only mar- Poland designed to improve forest management systems. ginally since the convention was signed, and coral reef The GEF was never intended to cover all of the inter- degradation may have increased. national financing needs of global environmental pro- grams. As of September 1998, approximately seven years The rest of this chapter focuses on the conditions after its establishment, the GEF had allocated a total of and mechanisms that determine the success of interna- just under $2 billion—less than the maximum allowed for tional agreements designed to counter global environ- carbon credit transfers under the Kyoto Protocol. On the mental problems. International treaties are based on other hand, where it is involved, the GEF is playing an im- bargaining, financial incentives, and, under some cir- portant role in supporting a range of measures to ensure global environmental sustainability.31 cumstances, limited controls on trade and finance. In- ternational funding based on the kinds of transfer mechanisms discussed here can help resolve two of the vate market forces or national governments acting alone major problems that hold up such agreements: what to set prices for them that reflect their value, since any- kinds of environmental controls the agreements should one can use a nonexcludable good without paying for include, and who should pay for those controls. it and the cost of additional nonrivalrous users enjoy- The ozone treaties: a success story ing such a good is essentially zero. Because neither mar- kets nor national laws are likely to fully reflect the value Concern about declining ozone levels in the upper at- of public goods that are shared globally, only interna- mosphere gained worldwide attention in the early and tional agreements can fully protect these resources. But mid-1980s. Scenarios predicting huge increases in the the costs and benefits of protecting natural resources rates of skin cancer and cataracts were widespread. differ from country to country, as do the levels of re- Then, in 1987, the Montreal Protocol emerged as a co- sources available for countering environmental degra- operative effort to slow ozone depletion by reducing dation, creating a need for effective transfer mecha- output of chlorine and bromine ozone-depleting sub- nisms. The Global Environment Facility is one model stances. Twelve years later, thanks to the Protocol and for such transfers. follow-on agreements, concerns over ozone depletion Nonetheless, critics often argue that the agreements are largely behind us. Global production of CFCs has on biodiversity and climate change signed in the latter fallen dramatically, and atmospheric concentrations of      Figure 4.2 stances did not yet exist. So, although the Vienna talks Atmospheric concentrations of ozone-depleting created a framework for future agreements, they did substances first rose, then began to fall not contain a protocol limiting the use of CFCs. The discovery of an ozone hole over the Antarctic in the CFC-11 (parts per trillion) winter of 1985 pushed the issue into the news and 300 helped create consensus on the need for international action.36 Six months after the 1987 Montreal meetings, the International Ozone Trends Panel report heralded 250 the first occasion on which the link between CFCs and ozone depletion, along with evidence of depletion oc- curring over the populated mid- and high latitudes of 200 the Northern Hemisphere, was reported by the scien- tific community and accepted by policymakers from key CFC-producing countries.37 This stronger accep- 150 tance led to agreements being signed at the London meeting of the parties in 1990 that greatly accelerated the timetable for abandoning ozone-damaging chemi- 100 cals. This agreement and its successors covered 97 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 ozone-depleting chemicals—far more than the 8 cov- Note: Figure reports ground-level concentrations of CFC-11 at seven ered by the Montreal Protocol. monitoring stations (in Alaska, Hawaii, American Samoa, the South Pole, Canada, the continental United States, and Tasmania). When the first treaty limiting CFC production was Source: Elkins 1999. signed in Montreal in 1987, little or no evidence ex- isted that ozone had thinned anywhere but over the Antarctic—nor that CFCs had caused the ozone hole, nor that increased ultraviolet radiation was already these chemicals have not only stabilized but are be- starting to reach the earth.38 Montreal was the first sig- ginning to drop (figure 4.2).34 Not all the problems re- nificant treaty to accept the “precautionary principle,” lating to ozone have been resolved. The black market which holds that scientific uncertainty should not delay in CFCs, while declining, is estimated at 20,000 to an international policy response if the delay might re- 30,000 tons per year.35 But global cooperation to re- sult in irreversible damage.39 Nonetheless, growing sci- duce ozone depletion can be broadly declared a success. entific consensus on the costs of continued CFC pro- The key factors that allowed the negotiators in Mon- duction and possible substitutes was vital to the passage treal to reach a strong international agreement were: of the treaties. The process of coming to such a consen- sus was hastened by the Assessment Panel mechanism Ⅲ A consensus that the risks of ozone depletion as a re- created as part of the Montreal treaty. These interna- sult of CFCs and other substances containing chlo- tional panels of economic, scientific, and technical ex- rine and bromine had high costs, and that there was perts described the advancing status of scientific under- the technological and institutional ability to find standing and technical response options in the run-up cost-effective, environmentally benign substitutes. to meetings of the parties.40 Ⅲ The involvement of all parties with a significant role The high ratio of expected benefits to costs also to play in solving the problem, brought about by helped the passage of the CFC agreements. One reason using both payments and penalties, along with flexi- the costs were relatively low was that research into al- bility in setting conditions for meeting the treaty’s ternate technologies had been under way for some goals. time. In response to earlier public pressure, some coun- tries had begun introducing restrictions on CFCs in Consensus on high net benefits aerosol sprays in the late 1970s.41 The United States, At the time of the Vienna Convention for the Protec- which had begun regulating CFCs in 1977, banned all tion of the Ozone Layer in 1985, a consensus on the nonessential CFC aerosol sprays in 1978, giving CFC impact of chlorine and bromine ozone-depleting sub- producers time (and the incentive) to research alterna-      ⁄  tive production methods before all uses of CFCs were Box 4.4 banned.42 At the same time, the costs of policing com- NGOs and efforts to preserve pliance with CFC-reduction targets were relatively low the international environment because the production of CFCs was largely concen- Nonstate actors are playing an increasingly important role trated in a few countries and was controlled by rela- in the negotiations surrounding international agreements. tively few companies. This, combined with the large Groups such as nongovernmental organizations (NGOs) potential benefits of an international agreement to limit often make an enormous contribution by serving as con- CFC production, gave OECD countries a strong in- duits for information on the environmentally damaging ac- tivities of countries and governments. centive to negotiate. This was especially so given that The Montreal Protocol negotiations were open to rep- the threat of skin cancer as a result of exposure to in- resentatives from NGOs representing business and sci- creased ultraviolet radiation was far greater in OECD ence. The World Meteorological Organization, with the countries than elsewhere. United Nations Environment Programme (UNEP), played an important role in presenting numerous reports by the Nongovernmental organizations (NGOs) also played scientific community that illustrated the linkages between a role by helping to put pressure on governments to ne- chlorine- and bromine-containing substances and ozone de- gotiate deals. By raising public awareness of the possi- pletion.45 Outside the formal negotiating process, Friends bly catastrophic dangers of ozone depletion and the of the Earth UK led a boycott of CFC aerosol products that lasted until 1987. The boycott resonated with the public links with chlorine- and bromine-containing substances, and put pressure on the U.K. government to push for a NGOs worked with the scientific community to create strong treaty. popular support for an agreement (box 4.4). The role NGOs are also essential players in efforts to support of NGOs is in line with one of the themes of this re- best environmental practices and to discourage unsustain- able behavior. World Development Report 1998/99 cited port: that civil society can have an important place in the role of the West African Newsmedia and Development the international policymaking arena. Center, a regional NGO based in Benin, in disseminating environmental information through print and broadcast Global participation media.46 NGOs are also working with industry to create and advertise standards for areas such as fishing and A vital element in the success of the ozone treaties was forestry. NGOs and representatives from the timber trade the participation of all countries that produced or con- and forestry profession have formed the Forest Steward- sumed (or seemed likely to produce or consume) signif- ship Council, an international association aimed at promot- icant amounts of ozone-depleting substances—includ- ing sustainable forestry practices. The council’s interna- tional labeling scheme for forest products provides a ing developing countries. The post-Montreal consensus credible guarantee that the products bearing the labels on ozone damage served as a dramatic testimonial to the come from forests meeting the standards laid out in the importance of including developing countries in an council’s Principles and Criteria for Forest Stewardship. agreement. The World Resources Institute estimated that if Brazil, China, India, and Indonesia alone increased CFC production to the levels allowed in the Montreal fered a grace period of exclusion from controls on chlo- Protocol, global production of ozone-depleting sub- rine and bromine ozone-depleting substances. They stances would double from the 1986 base level. The im- would also have access to a fund set up to cover adjust- pact of such an increase on ozone levels would be pro- ment costs and finance technical assistance.47 The initial found.43 Not involving developing countries, especially fund introduced at the London meeting provided $160 in the more stringent targets set at London, would have million (paid for by OECD countries) and an additional also threatened the treaty with “leakage”—that is, com- $80 million if China and India signed the protocols.48 panies moving CFC factories from OECD sites to de- The Montreal agreement also banned international veloping countries with higher production limits. trade between signatories and nonsignatories of CFCs, But developing countries needed an incentive to agree products containing CFCs, and CFC technology. The to tighter restrictions. They feared that substitutes for significance of this provision was made clear when the ozone-depleting substances would be more expensive, threat of trade sanctions (combined with increased fund- and they felt in a poor position to bear such costs.44 ing from a number of OECD countries and the Global Questions of international equity took center stage. To Environment Facility) encouraged Russia to agree to secure their cooperation, developing countries were of- meet its commitments to phase out CFC production by      2000.49 Payments and trade mechanisms to support ducing carbon dioxide emissions in the Arab Republic compliance, along with flexibility in treaty restrictions, of Egypt and Zimbabwe by 20 percent would actually were vital in creating a strong global agreement. But the be negative, since the government would only have to payments and flexibility were possible and the trade remove inefficient subsidies—a net gain.53 sanctions credible only because eliminating CFCs would The benefits of efforts to prevent climate change will provide industrial countries with substantial net bene- become apparent only in the long term, while the costs fits. The potential benefits, plus the threat of sanctions of such mitigation must be paid today. And while con- gave these countries an incentive to sign the treaties, de- trolling climate change offers potentially significant ben- spite the financial burden the agreements imposed.50 efits, the costs of reducing carbon dioxide emissions are Finally, restrictions on CFC production were made also significant—far greater than the costs of controlling as flexible as possible. For example, Japan was recon- ozone-depleting substances. With climate change, then, ciled to the treaty despite a high reliance on CFC-113 the costs of prevention are higher and the relative scale for cleaning computer chips by a mechanism that set a of benefits is lower, especially for industrial countries. limit on total production of ozone-depleting chemicals While mechanisms such as carbon trading will reduce and allowed countries to use any combination of CFCs this disparity, it does suggest a reason for the greater within their overall limit.51 political complexity of negotiating strong greenhouse gas accords: unlike the relatively narrow range of activi- Climate change ties that affect the ozone layer, the major sources of Why have attempts to cut global greenhouse gas emis- greenhouse gas production are ubiquitous, including sions been less successful so far than efforts to halt pro- power generation, industrial energy use, transportation, duction of ozone-depleting substances? The contrast and farming.54 These activities account for a huge share between the progress that has been made in tackling of global GDP and are deeply entrenched in the pro- these two global environmental concerns highlights the duction structures of industrial and developing econ- importance of a consensus that actions to address the omies alike. problems have clear net benefits. Moreover, much of the technology required to make the switch to cleaner production methods is compara- Costs and benefits tively expensive, suggesting a greater economic and po- At the global level the benefits of stabilizing or reduc- litical burden in technology switching than in the case ing carbon emissions are potentially substantial. As of ozone-depleting substances. In the long term, renew- noted above, the IPCC estimates that a doubling of car- able energy sources may play a more important role in bon dioxide in the atmosphere would result in costs for production, but wind and solar energy are not yet fea- developing countries equal to 2–9 percent of GDP.52 sible economic substitutes for fossil fuels on a large The quantifiable costs are lower as a percentage of GDP scale. Even in areas where they are economically feasi- for industrial countries but are still around 1.0–1.5 per- ble today, market distortions and entry barriers limit cent of GDP. These estimates include only costs that their use (box 4.5). It should be noted, however, that can be easily quantified, omitting the effects of factors economic reform and funding for research could make (such as species extinction) for which it is hard to as- renewable energy sources more attractive. sign a monetary value. Indeed, increased support for research on new tech- While the benefits of controlling greenhouse gases nologies can lower the long-term costs of complying appear lower for industrial countries, estimates of the with stricter carbon emissions limits worldwide. Three costs of controlling emissions suggest the reverse—that of the most successful technologies supported by the costs are higher in industrial economies than in devel- U.S. Department of Energy—heat-reflecting windows, oping countries. Holding carbon dioxide output in the electronic ballasts for fluorescent lights, and variable- United States at 1990 levels until 2010 will reduce the capacity supermarket refrigeration units—are now sav- country’s GDP by an estimated 0.2–0.7 percent. Low- ing enough energy to justify the department’s entire ering output by 20 percent will cost 0.9–2.1 percent of efficiency research budget.55 Despite such remarkable GDP. The costs are certainly far lower for developing results, “efficiency and renewables” research received countries. One recent study suggests that the cost of re- only about 23 percent of the rapidly shrinking U.S.      ⁄  Box 4.5 Falling costs for renewable energy Renewable energy resources offer enormous potential for pro- Competition has reduced the cost of ducing electricity, particularly in developing countries, which wind-generated power in the United Kingdom often have an abundant supply of sun, water, wind, biomass, and other energy sources. This potential remains largely un- NFFO minimum and average bid prices tapped, mainly because of lack of familiarity with renewable for wind-generated power (cents per kilowatt-hour) energy technologies and because of their relatively high up- 20 front costs. But two trends indicate that the future may be 18 brighter for renewable energy sources in developing countries. 16 First, in certain niche areas, the costs of renewable energy are already competitive with conventional energy resources, 14 even at the low fossil fuel prices of the late 1990s. Conven- 12 tional power generation has two less costly competitors: mini- 10 hydropower sites and biomass cogeneration facilities. These 8 Average price facilities are located close to population centers or to transmis- 6 sion lines (into which they feed their power). A number of solar 4 Minimum price photovoltaic systems are feasible for off-grid power genera- 2 tion. These systems are most useful in rural areas far from the 0 main power grid and in sparsely populated areas where low 1991 1994 1998 demand makes the cost of extending the grid prohibitive. Second, it has become clear that creating competitive, Source: World Bank data. market-type conditions significantly reduces the costs of using renewable energy technologies. In Indonesia, once it became known that the World Bank and the GEF would finance a large clined dramatically, falling from an average of around 18 cents renewable power project, potential vendors began to cut prices per kilowatt-hour in 1991 to 5.1 cents per kilowatt-hour in 1998. to secure their position in the emerging market. Competition (Declining prices for fossil fuels during this period meant that also reduced the costs of wind-generated power under the the relative costs of renewable technology fell more slowly.) United Kingdom’s Non–Fossil Fuel Obligation (NFFO) scheme. Although technological progress is clearly essential to Under the NFFO, renewable energy projects are selected in reducing the costs of using renewable energy technologies, competitive bidding and receive an output subsidy financed by sector reform, including the removal of subsidies on fossil a levy on electricity generated with fossil fuels that applies to fuels and open competition, can also be an important factor. all electricity consumers. By November 1998, five NFFO bid- Whether managed, as in the United Kingdom, or spontaneous, ding rounds had taken place. As the figure shows, bid prices— as in Indonesia, reform has helped drive technological advance the lowest as well as the average bid—for wind energy de- and has encouraged the efficient use of technology. budget for energy research and development in 1997.56 alternatives requires emissions treaties, and treaties only Redirected and increased expenditure on research, bet- happen when costs of agreement are lower—seems ter coordinated at the international level, is a win-win likely to plague greenhouse gas negotiations for some international response to climate change. time to come. Governments can take several steps to encourage The long-term approach of reaching an interna- private sector investment in alternative energy research, tional agreement on reducing greenhouse gas output as well. Early and concrete moves toward carbon emis- might include agreements on common policies and sions limits will push firms to start looking at other en- measures, such as fuel efficiency standards for cars. But ergy sources. Turning from subsidizing carbon-based it is also likely to involve negotiating either an inter- fuels to taxing them instead (or raising such taxes grad- nationally coordinated tax or a system of quotas on ually) while offering support for research on alternative carbon emissions, which might be tradable between energy sources can change incentives. The recent com- countries. Either approach will face many practical dif- mitment of some of the world’s major oil companies ficulties (box 4.6). This is one more reason that the per- to reducing their carbon emissions is a hopeful sign ceived benefits of treaty making will have to rise far that the early negotiations on greenhouse gases have al- above the costs to create the flexibility needed to sign ready encouraged private sector responses that will such a tough agreement. lower the cost of future emissions compliance. Even so, In short, a number of reasons suggest that coming the chicken-and-egg problem—progress toward energy to an international agreement on greenhouse gas emis-      Box 4.6 Taxes and quotas to reduce emissions Two competing mechanisms are frequently suggested for use quotas could allow a more equitable distribution of the costs in a global agreement on reducing greenhouse gas output. The of treaty compliance than a tax agreement. Quotas could also first, which was used in the Kyoto agreement, sets caps on be used to transfer resources from industrial to developing each country’s output of greenhouse gases. Many economists countries. This “cap and trade” system does present prob- favor auctioning off emissions permits, up to the quantity of lems, however. Assigning quota allocations is not a simple the cap, that can be traded both within a country and across process. For example, the Kyoto Protocol is based on the as- national boundaries. The second main approach would involve sumption that countries will make broadly similar percentage implementing national carbon taxes at globally agreed levels. reductions, starting from 1990 levels. Future, more encom- The mechanics of incorporating either approach into an in- passing agreements will have difficulty with the assumption ternational agreement are complicated, however, because the of broadly equal reductions from treaty-start levels; develop- costs and benefits of reducing greenhouse gases vary consid- ing countries will find such reductions unacceptable because erably across countries. Energy taxes and energy efficiency they expect to consume more energy as they develop. A also differ vastly across countries, raising the question of how mixed approach would be needed that sets quotas according to set a baseline for either tax rates or output. to several factors, including present absolute output, output With a globally agreed tax on emissions applied equally, per capita, and level of development. Quotas could also be countries with low marginal benefits from emissions would based on a target rate of improvement in energy intensity (use spend more on abatement measures than those with high mar- of energy per unit of GDP). Further problems remain, however: ginal benefits. While the tax system would also generate healthy revenues for governments (and especially in the devel- Ⅲ If developing countries are to be enticed into the system, oping world, carbon taxes may be more efficient than the pres- the net quota trade would need to run from the developing ent tax regime), an equal tax regime would probably lead de- to the industrial world, which creates the potential for large veloping countries to abate more than those industrial countries economic transfers. The political viability of this transfer with higher marginal costs of abatement. To equalize the pain mechanism is questionable, however, since transfers would of reducing output, tax rates might have to vary across coun- be made without regard to the political and economic activi- tries. But that would create incentives for leakage, with high- ties of recipient countries.57 polluting industries moving to countries with the lowest tax Ⅲ What has been termed the problem of “low-hanging fruit” rather than reducing their greenhouse gas output. A global car- might also affect the carbon-trading mechanisms proposed bon tax agreement would also have to specify exactly which at Kyoto. This arises when developing countries have traded emissions were to be taxed. Certain emissions, such as those away the cheapest methods of reducing carbon emissions, from livestock, paddy fields, and wood-burning stoves, are re- and have to pay for more costly measures in order to meet garded as largely “untaxable.” These types of emissions differ their international obligations. dramatically across countries, which adds to the difficulties of Ⅲ Certifying that countries have met their obligations is likely allowing certain activities to be exempt from an emissions tax. to prove a major challenge, both in enforcing the Kyoto Finally, nations would have to agree whether the emissions agreements and beyond. As has been mentioned, many ac- taxes would be kept by each nation or shared to some extent tivities contribute in some way to climate change. And is- across nations. sues such as how to measure carbon sequestration (if that Under a global binding agreement on national emissions is to be included as part of the treaty mechanism) are still levels, the added flexibility of being able to negotiate national far from settled. sions as comprehensive as those on ozone will be a burning fossil fuels (releasing carbon dioxide) as more much more complex task. And it is not surprising that beneficial social priorities than reducing emissions of in many developing countries there are mixed feelings greenhouse gases. Thus while the AOSIS, for example, about controlling emissions. The Alliance of Small Is- has adopted unilateral carbon abatement measures, land States (AOSIS) and some other low-lying coun- until the coalition is larger, this first step is unlikely to tries such as Bangladesh give the most urgent priority translate into stronger multilateral accords.58 Nonethe- to curbing climate change, for understandable reasons. less, the world is moving toward increased global coop- A 1-meter increase in the sea level would force about eration on climate change. 70 million people to move and would have a dramatic effect on food security in Bangladesh (figure 4.3). But Increasing participation even most developing countries would still rank such Industrial countries are responsible for most of the in- activities as burning forests for agriculture (releasing crease in man-made greenhouse gases in the atmos- carbon dioxide and removing a carbon sink), raising phere. But before the middle of the next century, en- livestock and growing rice (releasing methane), and ergy consumption in developing countries is projected      ⁄  Figure 4.3 Figure 4.4 A 1-meter rise in sea level would cut Bangladesh’s Energy consumption in developing rice production approximately in half countries is forecast to outstrip industrial country consumption Gigatons oil equivalent BANGLADESH 20 Projected areas of inundation 18 Approximate 1 meter contour Developing countries 16 National capital Transition economies International boundary 14 OECD 12 10 8 6 INDIA 4 Dhaka Approximately half of 2 current rice production is in N 0 this area of Bangladesh. 1860 1880 1900 1920 1940 1960 1980 2000 2020 2040 2060 INDIA Note: Data for the period 2000–2060 show a scenario of future energy consumption based on current trends. Source: World Energy Council; World Bank. Ⅲ Current and historical emissions of greenhouse gases N MAR Bay of Bengal in developing countries are much lower than in ei- YA 0 25 50 75 Kilometers ther industrial or transition economies. Per capita 0 25 50 Miles M emissions are also likely to remain lower for the fore- seeable future. Source: World Bank 1998f. Ⅲ Industrial countries have greater economic, techni- cal, and institutional capacity to address the issue. Ⅲ The imperatives of social and economic development to be more than twice that of OECD countries, even argue for increasing energy use in developing countries. though per capita consumption will remain much lower, as it is now (figures 4.4 and 4.5). Developing The Kyoto Protocol encompasses transition econ- countries must be included in global greenhouse gas omies and involves developing countries through a sys- agreements, both because of the likelihood that they tem of limited and voluntary cooperation. Industrial will someday bear the responsibility for most green- countries can meet their commitments for lower emis- house gas emissions and because without their cooper- sions not only by reducing emissions within their coun- ation, any progress could be offset by leakages to devel- tries but also by trading obligations with countries that oping countries (box 4.6). For example, if a steel plant have committed to targets or by funding emissions re- tries to avoid emissions limits by moving its operations duction projects in developing countries. For transition from a relatively energy-efficient industrial country to economies that have agreed to emissions targets, the an energy-inefficient country not covered by an agree- treaty allows for commitment trading, while the Joint ment, total greenhouse gas output could rise.59 Implementation scheme enables industrial countries to But although drawing developing countries into acquire emissions trading permits in return for support- binding agreements on greenhouse gas emissions is ing emissions reduction projects in those economies.60 vital, industrial countries are still expected to take the After 2000, the Clean Development Mechanism may lead on such an agreement, for several reasons: allow industrial nations to buy project-based emissions (c) The International Bank for Reconstruction and Development / The World Bank      Figure 4.5 High-income countries use energy more intensively than countries in low-income regions 14 Total output (1,000 megatons 12 of CO2) Output per 10 capita (kg of CO2) 8 6 4 2 0 High-income East Asia and Europe and Latin America Middle East and South Asia Sub-Saharan countries Pacific Central Asia and the Caribbean North Africa Africa Source: World Bank, World Development Indicators, 1998. rights from developing countries that have not agreed problem of leakage. It could also create perverse incen- to binding emissions targets, with a portion of the pro- tives for carbon trading among industrial and develop- ceeds being used for administration costs and to help ing countries (see box 4.6). Still, it is an important first particularly vulnerable developing countries meet the step toward global involvement in the reduction of costs of adapting to climate change. greenhouse gas output.62 As noted, involving develop- These limited trading mechanisms should have a ing countries at some level and as early as possible is significant effect on the costs of emissions reductions. very important for controlling future greenhouse gas Estimates vary, but one model suggests that the mar- emissions. The demand for electric power in develop- ginal tax or quota price for the United States to meet ing countries is rising rapidly and is projected to climb the Kyoto target (93 percent of 1990 levels by 2012) by up to 300 percent between 1990 and 2010, outpac- would be about 72 percent lower if quota trading were ing by far the 20 percent rise expected in industrial allowed among industrial and transition economies. countries.63 Joint Implementation and the Clean De- Adding some key developing countries to the trading velopment Mechanism can be used to ensure that a network would reduce permit prices even further, to an significant proportion of the projected generating ca- estimated 12 percent of the autarky price.61 pacity in developing countries is based on low-carbon- The scale of trading—and thus of transfers among dependent technology.64 countries—is likely to be large. The OECD countries To further the goal of reducing greenhouse gas out- emit about 3 billion tons of carbon a year. The Kyoto put in developing countries, the World Bank has begun agreement alone will reduce the emissions these coun- a series of projects under the pilot phase of Activities Im- tries would have produced without the agreement by at plemented Jointly established at the Rio Summit. The least 30 percent. If carbon is valued at $23 a ton, and Ilumex project in Monterrey and Guadalajara in Mex- only half the reductions are met through quota trading, ico has replaced some 200,000 ordinary incandescent the global quota market will be worth $11.5 billion a light bulbs with compact fluorescent light bulbs. Be- year—more than the total U.S. aid budget. cause the new bulbs use far less energy than conventional In the long term, the Kyoto Protocol’s Clean Devel- lighting, power stations need to provide less electricity, opment Mechanism is not a full solution to the green- permanently reducing the demand for fuel. The project house gas problem, in part because it does not solve the should also help Mexico reach its own goals for reduc-      ⁄  ing emissions of sulfur dioxide and nitrogen oxides. In ects. In Poland, for example, the Forest Biodiversity Pro- Burkina Faso a sustainable energy management project tection Project has provided institutional support to the will promote solar power systems and kerosene cooking country’s environment ministry, funded pilot invest- stoves while supporting community-based sustainable ments in air- and soil-monitoring equipment and a for- forestry management and efficient carbonization tech- est gene bank, and supported farmers in the Bialowieza niques. The project will abate more than 300,000 tons Primeval Forest who are making the transition to “eco- of carbon emissions a year for just $2.5 million, or $8.30 logical agriculture.” In Algeria the El Kala National Park per ton of carbon.65 and Wetlands Management Project introduced actions to stop degradation within the complex and supported Biodiversity assessment activities that included surveys, studies, and The Convention on Biological Diversity signed at the public education programs aimed at bolstering long- Rio Earth Summit in 1992 has been ratified by 169 term preservation efforts. countries. Signatories to the convention are obliged to While such support may form the backbone of in- conserve and ensure the sustainable use of their own bi- ternational efforts to preserve biodiversity, the global ological diversity.66 Those countries with the greatest commons issues connected with existence value and ex- biodiversity are concentrated in the developing world. ploitation of genetic resources remain. The economics Only one of the eight countries that are home to the of these issues is complicated by disagreement on what largest number of native mammal species is industrial. exactly is being valued—whether it is the right of plants Of the countries with more than 10,000 species of or animals to exist, the material benefits that diversity higher plants, 18 out of 20 are developing countries, and offers, or the just the pleasure that the existence of many 12 of the 17 countries with more than 500 threatened living organisms brings to people. Even basic facts such species of higher plants are developing countries.67 De- as the total number of species on earth and the rate of veloping countries are thus key to meeting the goals set species extinction worldwide are not fully clear. The at Rio, and the Convention on Biological Diversity was UNEP’s Global Biodiversity Assessment estimates the passed with widespread support from these countries. number of species on the planet at 7 million–20 million Like the Framework Convention on Climate Change, and the expected loss of species over the next 25 years the Biodiversity Convention recognized economic and at between 140,000 and 5 million. Combining the social development as the top priorities for developing lower-bound estimates suggests that 2 percent of all countries. It also stated that the extent to which devel- species are at risk; combining the upper-bound esti- oping country parties would effectively implement their mates produces an estimate of 25 percent (although it commitments to preserve biodiversity would depend on should be noted that even the lower rate of extinction industrial country commitments related to financial re- is approximately 1,000 times the natural rate).68 sources and transfer of technology. Many of the benefits of preserving genetic material are also difficult to quantify in monetary terms. How The benefits of biodiversity and the costs is a dollar value to be placed on the rights of organisms of preservation to exist or on the pleasure people derive from their ex- As we have seen, the ecosystems (and the species) in which istence? Among the more quantifiable benefits is the genetic material resides provide valuable services at the medicinal use of genetic resources. The United Nations national level. For this reason, the primary role of inter- has estimated that medicines originally developed from national agencies and bilateral support in the area of bio- plant material are worth about $43 billion a year.69 The diversity should be to transfer knowledge and provide rosy periwinkle from Madagascar’s rain forest, for ex- technical assistance to help overcome national market fail- ample, provided a rare genetic trait that was used in de- ures and create national markets for ecological benefits. veloping pharmaceuticals to treat childhood leukemia. The GEF was chosen as the formal interim financ- Two of the drugs one company has developed from this ing mechanism for the Convention on Biological Di- plant have sales worth $100 million a year. (None of versity. Total GEF financing for biodiversity projects these proceeds, it should be noted, goes to Madagas- comes to over $800 million and has already been used car.)70 But even calculating the marginal benefit of the to support a range of technical and institutional proj- genetic material in a species is no easy task. The drugs      developed from plants must be collected, refined, tested, have already been explored for genetic material. Thus, and developed for the market, and sharing the profits while establishing limited property rights to genetic ma- along this value chain is clearly a complex issue. Ge- terial may encourage developing countries to participate netic materials are also likely to be present in more than in preservation efforts, they represent only a partial so- one species. This fact helps explain why estimates of the lution. If industrial countries feel that additional incen- marginal value of species existence (put another way, tives to preserve genetic material are required (to cover, the marginal value of preventing species extinction) are as it might be, the existence value of species, regardless so uncertain. They have been put at anywhere from $44 of their economic uses) the simplest method would be to $23.7 million for an untested species.71 to expand direct international support for this purpose. Biodiversity covers many different activities, includ- Expanding participation ing farming, forestry, coral reef protection, and others. Even if the value of genetic material is hard to estimate, This diversity calls for great flexibility in approach to- it is certainly true that the international community con- ward agreements on different biodiversity issues, at both tinues to exploit it without paying—a scenario that con- the regional and global levels.76 Technical and institu- stitutes a market failure. As a result, biodiversity may be tional support and flexible transfer payments are two undervalued in developing countries. An additional such approaches. Sanctions have also been used. When mechanism for promoting the preservation of genetic re- biodiverse habitats are exploited in order to produce a sources would be the extension of property rights to a tradable good—including tropical fish, tropical timber, country’s genetic material. This subject was raised at the and many of the animals covered by the Convention on Convention on Biological Diversity, but no agreement International Trade in Endangered Species (CITES)— was reached on what should be done about the situa- formal trade limits or certification schemes with strong tion.72 One model for resource transfer might be that of penalties for noncompliance can play an important role. Costa Rica’s private, nonprofit National Biodiversity In- One way to protect coral, for example, might be to ban stitute (INBio), which struck a deal with U.S.-based trade in fish caught by using cyanide, a significant pharmaceutical firm Merck and Company to help un- source of coral degradation. Similar incentives have derwrite INBio’s biodiversity prospecting plans.73 The been widely used in a range of environmental treaties. Merck deal will pay INBio $1.1 million plus royalties Although questions remain about the risk of overusing for any product Merck develops from Costa Rican re- trade measures to counter environmental threats (and sources. In return INBio provides Merck with samples thus using the environment as an excuse to strangle from all over Costa Rica. Ten percent of the up-front trade as a wealth-creating force), trade measures can be money and 50 percent of any royalties are allocated to a very effective method of pursuing environmental goals inventory, bioprospecting, and conservation.74 under some circumstances (box 4.7). A number of doubts have been raised about such Exploiting the links between global mechanisms. The Costa Rica–Merck agreement, for environmental problems example, does not involve enough resources to pay for significant increases in protected reserves. The scheme Climate change and biodiversity are not only serious is- might also not be widely replicable. One recent estimate sues in their own right but are also linked with each suggests that even in western Ecuador, one of the areas other and with a wide range of other environmental richest in endemic species, the per-hectare value of ge- concerns. Depending on the rate of climate change, netic material to drug companies is only about $20.75 forest species may be unable to adapt fast enough to Furthermore, by claiming royalties on products devel- avoid severe population declines.83 Aquatic ecosystems oped from plants and animals that may be found in such as mangroves and coral reefs adapt even more more than one country, INBio is effectively reducing the slowly.84 The loss of species and genetic material can incentive of neighboring countries to take similar mea- increase the vulnerability of ecosystems to other envi- sures to protect their genetic diversity. This problem is ronmental stresses, such as pollution.85 To complete the likely to be widespread: the rosy periwinkle was not en- circle, the destruction of forests has a dramatic impact demic to Madagascar, for example. Moreover, it is un- on climate change because forests release significant clear how such a scheme could work to protect areas that quantities of carbon dioxide as they burn.86      ⁄  Box 4.7 Trade measures in international environmental agreements The earliest environmental treaty to use trade measures was that free trade is considered a global good because it maxi- the International Convention Respecting Measures to be mizes welfare. If trade is instead causing serious environmen- Taken against the Phylloxera Vastatrix, which in 1881 banned tal damage, then, it can be argued, it must be limited. Trade trade in torn vines and dried shoots to prevent relocating the may be especially intertwined with certain environmental risks: plant louse to other vineyards.77 Other environmental agree- damage from relocation; insect infestation of previously unex- ments with trade measures include: posed crops from imported infected fruit; negative disposal ef- fects like those posed by imports of toxic waste; negative Ⅲ CITES, which allows trade in listed species or products made transport effects such as oil spills; and negative profit effects— from them with nonparties to the agreement only when as when trade ends up financing a decline in biodiversity.80 competent authorities in the nonparty country issue docu- Three recent international trade rulings suggest that inter- mentation comparable to that required of treaty members. national trade agreements allow little flexibility on the unilat- The convention also allows members to impose trade bans eral introduction of environmental trade bans: the WTO’s rul- on other members that do not comply with the restrictions. ing against U.S. laws banning the import of shrimp caught in In 1991 the CITES standing committee recommended that nets that also trap turtles, and two rulings by the General all trade with Thailand in flora and fauna species covered by Agreement on Tariffs and Trade (GATT) dispute panel on tuna the convention be stopped because of noncompliance. caught in nets that also trap dolphins. But the WTO ruling does Ⅲ The United Nations agreement on the Conservation and support the right of states to impose sanctions as part of an Management of Straddling Fish Stocks and Highly Migratory international treaty. Distinguishing between unilateral and mul- Fish Stocks, which allows members to prohibit the landing tilateral trade bans keeps environmental sanctions from being or transshipment of fish caught using methods that under- used as a cover for protectionist interests.81 mine the effectiveness of conservation and management Trade bans should also be limited to areas in which they measures.78 can be effective. The international ivory trade (as well as mis- Ⅲ The Montreal Protocol, which requires parties to ban the im- management of elephant stocks) must be seen as an impor- port of controlled ozone-depleting substances from non- tant reason for the drastic decline in elephant populations be- signatories unless the nonsignatories are found to be in full tween 1979 and 1989.82 In situations in which trade is the compliance with the protocol regime.79 dominant outlet for production, as it is for ivory, sanctions can have a major effect. Often, however, trade sanctions are too Trade measures can be an appropriate tool for addressing far from the source of the problem to be effective—which global environmental problems because they, like the problem, might limit the effectiveness of trade bans to counter green- are global in nature. They can also be justified on the grounds house gas emissions. Exploiting such links can greatly reduce the cost of international research, could lead to greater exploitation environmental protection. For example, whether land of this synergy between biodiversity preservation and use activities should be eligible under the Clean Devel- the prevention of climate change. opment Mechanism is an issue being decided by the par- These links across global environmental issues sug- ties to the Kyoto Protocol. But counting the preservation gest that the international community needs to move of developing country carbon sinks against emissions beyond simply negotiating separate agreements for each commitments could provide powerful synergies with environmental issue. As agreements such as CITES local environmental and biodiversity protection needs.87 demonstrate, treaties are often agreed to only because Costa Rica has already begun trading Certified Trad- complex problems are broken up into smaller units. able Offsets (CTOs)—carbon credits priced at $10 a But in some cases, agreements that cover many areas are ton—in ways that exploit such links. The profits from easier to negotiate because of the potential for trade- these credits are designed to support sustainable forestry offs or synergies between related issues. This pattern has practices on private land or to finance the conservation prevailed in multilateral trade negotiations, for exam- of land as national parks and bioreserves.88 So far, sus- ple, when countries that feel strongly about certain pro- tainable practices have been introduced on 3,000 farms visions have offered concessions in areas that concern covering 150,000 hectares. The bioreserve project has them less. Global environmental protection can also be conserved another 530,000 hectares.89 Although there hastened by improving coordination between treaty have been few early takers for the credits, Costa Rica’s and convention secretariats, including integration of experiences with the system, combined with continued meetings, scientific assessments, reporting require-      ments, publicity, training, and capacity-building efforts becoming clearer. Moreover, even taking preliminary and improved coordination under the UNEP. steps toward a partial agreement encourages private ac- tors to prepare for stricter agreements and thus lowers • • • the cost of future actions to resolve environmental con- The world’s countries have come far in cooperating to cerns. Consensus on biodiversity, climate change, and address global environmental issues, and the ozone ac- other global environmental issues will only expand over cords provide a model for future agreements. Although time. Furthermore, the growing understanding of link- drawing up international agreements on biodiversity ages among environmental concerns will create more and climate change that are as effective as the ozone opportunities to exploit both synergies and trade-offs, agreements has been difficult, the basic mechanics of helping to foster coalitions that support concerted successful international environmental agreements are global action. C h a 5 p t e r Decentralization: Rethinking Government P eople around the world are demanding This chapter argues that the success greater self-determination and influence of decentralization depends on its de- in the decisions of their governments— sign. It reviews developing countries’ a force this report has labeled localiza- experience with decentralization and tion. Some 95 percent of democracies shows that the stakes are high. Drawing now have elected subnational govern- on this experience, it offers guidelines ments, and countries everywhere—large for improving the political, fiscal, and and small, rich and poor—are devolv- administrative institutions of decentral- ing political, fiscal, and administrative ization. This advice is not only relevant powers to subnational tiers of govern- to countries that have already decentral- ment (box 5.1).1 But decentralization is ized. It can also help the many coun- often implemented haphazardly. Deci- tries now embarking on this path avoid sionmakers do not always fully control some of the major hurdles that have the pace or genesis of the decentrali- confronted their predecessors. zation process. Even when they do, What is at stake? models of decentralization are often ex- ported from one country to another The experience of the last 15 years without regard for local political tradi- shows that the devolution of powers tions, regulatory frameworks, or prop- affects political stability, public ser- erty rights. vice performance, equity, and macro- Decentralization itself is neither economic stability.2 good nor bad. It is a means to an end, often imposed by political reality. The Political stability issue is whether it is successful or not. A primary objective of decentralization Successful decentralization improves is to maintain political stability in the the efficiency and responsiveness of the face of pressures for localization. When public sector while accommodating a country finds itself deeply divided, potentially explosive political forces. especially along geographic or ethnic Unsuccessful decentralization threatens lines, decentralization provides an insti- economic and political stability and tutional mechanism for bringing oppo- disrupts the delivery of public services. sition groups into a formal, rule-bound       ⁄  Box 5.1 Box 5.2 Decentralization as the devolution of powers South Africa and Uganda: unifying a country through decentralization Decentralization entails the transfer of political, fiscal, and administrative powers to subnational units of government. South Africa and Uganda have adopted ambitious decen- A government has not decentralized unless the country tralization programs and, despite some difficulties with im- contains “autonomous elected subnational governments plementation, are emerging as two important models for capable of taking binding decisions in at least some policy devolving centralized power.9 The models operate in dif- areas.”3 Decentralization may involve bringing such gov- ferent contexts: a middle-income and predominantly urban ernments into existence. Or it may consist of expanding country (South Africa), and a low-income, predominately the resources and responsibilities of existing subnational rural country (Uganda). But both have the same goal: to re- governments. The definition encompasses many varia- unify the country. tions. India, for example, is a federal state, but the central South Africa. Apartheid fostered a dual structure of government has considerable power over subnational gov- government based on race. For whites, it promoted ac- ernments. Political power in China is officially centralized, countability, political involvement, and effective service but subnational units have substantial de facto autonomy delivery. But blacks, spatially segregated in “homelands” in what can be described as “decentralization Chinese and “townships” on the fringes of urban areas, had lim- style.” ited access to public goods and services. To reverse this Central governments can devolve their powers in other racial system, the new constitution provides for a compre- ways. Deconcentration increases the autonomy of staff in hensive decentralization policy, which the leadership has regional offices, while privatization moves responsibility been implementing. The racial jurisdictions were formally abolished along out of the public sector altogether. The policy implications with the system of apartheid. The country was subdivided differ. Deconcentration preserves the hierarchical relation- into 9 provinces, 5 metropolitan areas, and 850 munic- ship between field staff and the central government. Pri- ipalities, all racially mixed and with democratically elected vatization eliminates it altogether, introducing the profit governments. The central government retains primary fis- motive instead. Decentralization shifts the focus of ac- cal responsibility for expenditures that have a major redis- countability from the central government to constituents, tributive impact, such as health and education, but metro- usually through local elections. politan governments have been restructured to implement policies at the local level. Some difficulties remain—for ex- ample, how to divide responsibility for health and educa- tion between the central government and the provinces. But decentralization has succeeded in becoming one of bargaining process.4 In South Africa and Uganda decen- South Africa’s main instruments of unification. tralization has served as a path to national unity (box Uganda. The task President Museveni faced when he 5.2). In Sri Lanka it offers a potential political solution assumed power in 1985 was to reunite a country that had splintered into hostile factions during years of turmoil. The to the civil war. It is an instrument for deflating seces- broad-based politics of “resistance councils” and commit- sionist tendencies in Ethiopia and Bosnia and Herzego- tees that had been developed during the years of civil war vina (box 5.3). In Colombia centralized party elites re- helped pacify most parts of the country. This system— lied on decentralization to gain grassroots support, which entails giving power to the people of a village (the council) to freely choose their leaders (committees)— particularly in areas under rebel control.5 And Russia’s served as the basis for the local government policy en- transformation into a decentralized federal system can be shrined in the 1995 constitution. The 46 districts, which seen as a means of conceding enough power to regional are subdivided into smaller units down to the village level, interests to forestall their departure from the republic.6 have taken on substantial responsibilities for education, health, and local infrastructure. They now account for 30 percent of overall government spending. Public service performance Uganda still faces problems with implementing decen- The classic argument in favor of decentralization is that tralization. Limited local capacity and resistance from cen- it increases the efficiency and responsiveness of govern- tral ministries have hobbled the transfer of responsibilities. The revenues local governments control (primarily user ment.7 Locally elected leaders know their constituents charges and local taxes) have not increased as much as ex- better than authorities at the national level and so pected, and grants still account for 80 percent of local re- should be well positioned to provide the public services sources. Despite increased participation, local services and management have not become significantly more respon- local residents want and need. Physical proximity makes sive to local preferences—although this is now improving. it easier for citizens to hold local officials accountable Even with these difficulties, however, decentralization has for their performance.8 Finally, if the population is mo- been much more successful in maintaining national unity bile and citizens can “vote with their feet” by moving than the previous policies of centrally imposed controls. to another jurisdiction, decentralization can create :    Box 5.3 Bosnia and Herzegovina and Ethiopia: decentralization as a response to ethnic diversity Ethiopia and Bosnia and Herzegovina illustrate the tension be- self-determination within a common political and economic tween political imperatives and economic efficiency that community. The 1994 constitution, which establishes subna- emerges in countries with ethnic tensions. tional boundaries and mechanisms for intergovernmental fis- Bosnia and Herzegovina. The possibilities for instituting “ef- cal relations, stipulates that regions shall be formed on the ficient” federalism and equalization in an ethnically polarized basis of ethnic settlement patterns, language identity, and the society are limited. The Dayton Peace Agreement, which ad- consent of the people concerned. Subgroups within the mem- dressed the challenges of governing Yugoslavia’s successor ber states have the right to establish their own states, and states, had the potential to solidify relations among the three states have the right to secede from the federation. ethnic groups that ratified it. But the agreements had to com- The Ethiopian system differs from the Bosnian federalist promise on some key principles of fiscal federalism to reach a structure in one key way: in Ethiopia the central government politically acceptable solution. The Dayton agreement limited retains control of most tax revenues and has a strong redistrib- the state’s authority to international relations (including customs utive role. Central transfers consist of block grants determined and trade policies, debt service, and debt management), central according to population, development indicators, and revenue banking (through a currency board), and telecommunications performance. The poorer regions receive as much as 75 per- and national transport infrastructure. The national government’s cent of their revenues through these grants. But the capital, only revenues are now passport fees and transfers from its two Addis Ababa—which is the richest region—receives no central constituent entities, the Federation and the Republika Srpska. It government support. State-level spending is kept under con- has few spending powers and no redistributive functions. All trol by federal regulations on domestic borrowing and by a taxing powers belong to the two entities, which are responsi- block grant formula that reduces regional transfers in propor- ble for all other spending, including defense, pensions, health, tion to external borrowing and donor grant flows. States are and local roads. The entities are divided further into local gov- free to spend their block grants as they choose, subject only ernments that are responsible for education, housing, social to federal auditing. transfers, and public services. There are no cross-subsidies Ethiopia faces two challenges in its decentralization model. across the two entities and very few across local governments. One is to develop stronger state revenue sources to deflect The state faces challenges in carrying out even its minimal ethnic tension — especially resentment from ethnic groups in responsibilities, since it relies on transfers from the entity gov- richer regions that receive less in government transfers. The ernments. Moreover, since economic conditions differ sub- second is to strengthen local governments, which are respon- stantially across the country, large inequalities are likely to de- sible for delivering most services but do not have the neces- velop among and within the entities. sary capacity. Ethiopia. Ethiopia’s system of intergovernmental relations is designed to accommodate the rights of citizens to ethnic Source: Fox and Wallich 1997; World Bank 1999b. competition among local governments to better satisfy of decentralization makes all the difference. In Central citizens’ needs.10 America, decentralizing management responsibilities But evidence supporting these arguments is scanty— from the central government to provincial and local not because there is evidence to the contrary, but rather levels had little effect on the primary education sector. because the causal relationships are difficult to prove. But decentralizing management responsibility directly Governments perform a variety of functions under to the schools did improve educational performance.12 vastly different circumstances, which complicates com- Decentralization can also lower the quality of pub- parisons of performance in a country before and after lic services, as it has in Latin America and Russia.13 decentralization, or across countries between central- Conceding power to local governments is no guarantee ized and decentralized systems. Moreover, efficiency that all local interest groups will be represented in local and responsiveness can be hard to measure, and indica- politics. It may simply mean that power is transferred tors are seldom readily available.11 from national to local elites. In India, for instance, local How decentralization affects access to and quality of participation depends on social caste, and the poor have public services depends on the way it is designed and little influence.14 implemented. What local governments can achieve de- pends on the resources and responsibilities they are Equity granted and on the power of national governments to Whether decentralization exacerbates income differ- override their decisions, as happens in India (box 5.4) ences among regions or becomes a positive force in ef- and Zambia. Even within a particular sector, the mode forts to alleviate poverty depends on two factors. The      ⁄  Box 5.4 India: a decentralizing federation? India has a federal constitution that gives its states substantial states. States were not under any obligation to hold regular fiscal and regulatory powers. But three elements undercut local elections, and state-run agencies controlled most local those powers. First, the constitution also has strong unitary fea- functions, including urban planning and local infrastructure. tures, enabling the central government to dissolve state gov- Under the amendments states continue to define local govern- ernments and take over their administration. Second, central ments’ powers and resources and name their chief executive planning—which until recently governed India’s economy— officers. They also retain the power of supersession—that is, blunted the economic powers of states. Third, national parties the right to dissolve a local government and take over its pow- traditionally dominated subnational politics. Thus state bud- ers. However, the amendments suggest a list of local respon- getary outcomes were the result of centrally defined develop- sibilities for inclusion in state constitutions and call for the cre- ment policies and, in practice, state-level regulatory powers ation of state-level financial commissions to oversee fiscal had little meaning. relations between states and local governments. Most impor- The relative centralization of India’s federalism is changing, tant, states are required to hold elections within six months of however. The gradual weakening of central planning and the superseding a local government. growing strength of regional parties in national coalition govern- Implementing the amendments has been a slow process, ments are strengthening state governments and allowing them and some states have progressed more than others. With one a larger role in defining their development priorities. But most exception, all states have held local elections and are observ- states are having difficulties growing into their new role. Many ing the supersession rule. The proposed local functions are now are saddled with excessive debt and unsustainable wage and part of most states’ legislation, and a number of states have pension bills and have few incentives to mobilize their own re- set up finance commissions that have submitted recommen- sources. A few states, including Andhra Pradesh, Uttar Pradesh, dations. However, state governments have been slow to im- Orissa, and Haryana, are improving their financial situation and plement these recommendations and to enable local bodies to are making increasing use of the powers constitutionally execute the newly devolved functions. Recent assessments granted to them. show that Gujarat, Karnataka, Kerala, Madhya Pradesh, Maha- The trend toward greater decentralization in India was rein- rashtra, and West Bengal have made the most progress in de- forced in 1992 by the passage of the 73rd and 74th Amend- volving powers to local governments. ments, which offered constitutional recognition to local gov- ernments. Until then, the constitution had made no mention Source: Hemming, Mates, and Potter 1997; Mathur 1999; of local governments, which were effectively creatures of the Mohan 1999; World Bank 1998i. first is horizontal equity, or the extent to which subna- those of wealthier provinces, but expenditures are 59 tional governments have the fiscal capacity to deliver percent as a result of transfers from the central govern- an equivalent level of services to their population.15 ment.17 In Australia, Canada, and Germany grants guar- The second can be described as within-state equity, or antee a minimum level of per capita expenditures for the ability or willingness of subnational governments essential services in all regions. In other countries the to improve income distribution within their borders. goal is to ensure similar levels of service.18 A difficulty An additional complication springs from the fact that with equalization grants is that subnational govern- responsibilities for social services and direct income re- ments may differ in their willingness to raise taxes. Fur- distribution are typically shared across different tiers of thermore, the grants create an incentive for subnational government that have access to different sets of infor- authorities to understate their tax bases or relative mation and may have different objectives.16 wealth in order to maximize transfers.19 Horizontal equity. Tax bases vary substantially from Within-state equity. In most countries income in- region to region and city to city, but tax rates cannot. A equality is due more to differences among individuals local government with a relatively small tax base cannot within a state or province than to differences among the compensate by imposing much higher tax rates without states or provinces themselves.20 Providing poorer re- losing businesses and residents to jurisdictions with gions with additional resources, then, affects only one lower taxes. The costs of providing public services may aspect of the equity problem. Evidence from India also vary because of regional characteristics such as pop- and Indonesia shows that even dramatic redistribution ulation density and geographic location. To correct for across regions will have limited results unless targeting such variations, most decentralized fiscal systems in- is improved within regions themselves.21 This, in turn, clude equalization grants. In Vietnam the per capita tax depends on the ability and willingness of the local gov- revenues of low-income provinces are only 9 percent of ernment to engage in redistribution. :    Recent studies have shown that local officials and The result—seen in many Latin American countries— community groups are better placed to identify and is large central government deficits. More generally, reach the poor than central authorities. In Albania, for separating taxing and spending powers allows subna- example, local officials had considerable success in tional governments to incur only a fraction of the po- targeting the poor—far better than expected, given litical and financial costs of their expenditures, espe- available statistical information on income and family cially when most local resources are funded out of a characteristics.22 In Uzbekistan elected neighborhood common national pool of tax revenues. committees were able to increase both the efficiency The threat of macroeconomic instability is a serious and the cost-effectiveness of targeting.23 Their success issue only in countries where subnational governments suggests that local officials have access to social net- control substantial resources—usually, large federations works that help them identify the truly needy. But this or very decentralized wealthy countries (figures 5.1 and may not be the case in very large jurisdictions, such as 5.2).27 But even in these cases the evidence connect- China’s provinces. ing decentralization and macroeconomic instability is Subnational governments differ in their responsive- mixed. Several studies suggest that decentralization has ness to the needs of the poor. A recent review of an Ar- not undermined stability in the United States or in gentine social program that is funded by the central gov- Western European countries. In Latin America subna- ernment but implemented by provinces found that tional governments’ contribution to the national deficit poverty targeting varied substantially across provinces. was negligible in most countries, except federal ones.28 When reforms were introduced to improve the pro- From centralized to decentralized governance gram’s reach to the very poor, most of the improvements were due to reforms in intraprovincial targeting and bet- A decentralization program needs to be adapted to a ter national monitoring of provincial performance.24 country’s prevailing conditions. However, the experi- Similarly, in Bolivia it was only when decentralization ence of the past 15 years has yielded some universal gave communities more power to influence their local governments that the composition of local public expen- Figure 5.1 ditures shifted in favor of the poor.25 Subnational expenditures are a small share of Success in targeting the poor requires, therefore, a public expenditures, except in industrial combination of national and subnational efforts. In countries and large federations general, the bulk of the funding needs to remain a cen- Subnational expenditures tral government responsibility, but the better informa- (percentage of total government expenditures) tion available to local officials can be tapped by involv- 50 ing local governments in the delivery and management of social services. Central government needs to retain a 40 monitoring role, however, to ensure that redistributive goals are satisfied. 30 Macroeconomic stability 20 Decentralization, if handled poorly, can threaten macroeconomic stability.26 Fiscal decentralization re- 10 duces the central government’s control over public re- sources. The government of the Philippines, for exam- ple, is required to share nearly half its internal tax 0 Developing Transition Industrial Large revenue with subnational governments, limiting its countries economies countries federations (excl. transition (industrial and ability to adjust the budget in response to shocks. economies developing) and China) Deficit spending by local governments can also thwart central government efforts to cool the economy by re- Note: Graph shows median values rather than averages. Subnational straining public expenditure. expenditures are most recent available observations after 1990. Large federations are Argentina, Brazil, Canada, India, Mexico, the When revenues are decentralized before expenditure Russian Federation, and the United States. responsibilities, central governments are forced to Source: Appendix table A.1. maintain spending levels with a smaller resource base.      ⁄  Figure 5.2 erature on constitutionalism makes a strong case for Local governments never control a large share establishing the most fundamental of these rules— of public resources choosing the heads of state and government, electing members of the legislature, distributing power among Subnational and local taxes (percentage of total government taxes) branches of government—in a form that can be altered 40 only by exceptional majorities or complicated amend- 35 ment procedures.30 To be sustainable, such rules must Local taxes Subnational be “self-enforcing”—that is, all parties must believe 30 taxes they have more to gain by adhering to the rules than 25 they do by breaking them.31 20 Rules should be explicit, stable, and self-enforcing. But how should a country decide what their substance 15 should be? The answer involves three broad areas of 10 analysis: the division of national political power be- 5 tween national and subnational governments; the struc- ture, functions, and resources assigned to subnational 0 Developing Transition Industrial Large governments; and the electoral rules and other politi- countries economies countries federations cal institutions that bind local politicians to their (excl. transition (industrial and economies developing) constituents. and China) Note: Graph shows median values rather than averages. Local taxes Balancing political power between central and subnational taxes are most recent available observations after and local interests 1990. Large federations are Argentina, Brazil, Canada, India, Mexico, the Russian Federation, and the United States. The rules that govern relations between the central and Source: Appendix table A.1. subnational levels are almost always established at the national level, generally by the central government.32 Even when these rules are incorporated into constitu- lessons, which countries currently decentralizing can tions or treaties, they are still subject to renegotiation use to their advantage. One such lesson is the need for and to varying interpretations as to appropriate imple- a coherent set of rules to replace the hierarchical system mentation.33 The balance of powers between national of governance characteristic of centralized systems. and subnational governments will therefore depend on A major challenge of decentralization is to institu- the influence of regional interests on the national gov- tionalize the balance of power between the national and ernment. And the stability of this balance of powers local governments. This requires rules that both protect hinges upon the design of institutions that make it in and limit the rights of subnational governments. Such the interest of national and subnational political elites rules come in a variety of forms. Some are unwritten. to cooperate with each other. No law prohibits the United States government from providing relief to states in default, for example. Nor Moderating regional influence does Turkish law require the national government to on the national government bail out its defaulting municipalities. Yet in both cases The influence of subnational interests on the national these are well-established practices that influence the government depends on two factors. The first is the expectations of lenders and borrowers. way regional interests are incorporated in the national Making the rules of decentralization explicit and legislature, which determines a subnational govern- reasonably permanent reduces uncertainty and provides ment’s ability to pressure the national government to a common ground for all players in the political process. change rules. The second factor is the strength of the Informal, negotiation-based decentralization is difficult national executive, which influences the central govern- to manage, as illustrated by China’s experience (box ment’s ability to withstand such pressure. 5.5). Rules enable subnational governments to coordi- Regional interests and the legislature. Seats in parlia- nate a defense against an overassertive central govern- ment may be allocated to give equal representation to ment while restricting their ability to bargain.29 The lit- states or provinces, thereby favoring norms of territor- :    Box 5.5 Decentralization in China China is formally constituted as a unitary state, and the dominant government’s enterprise income tax) and transferred tax funds political party—the Chinese Communist Party—recommends that would otherwise have been shared with the central gov- candidates for the posts of governor and mayor for ratification ernment into local extrabudgetary accounts. by the People’s Congress.34 But political and economic power In 1994 new reforms created separate tax administrations has dispersed markedly in recent years, particularly since the for national and local taxes, a step that increased the central reforms of 1978. The number of posts controlled directly by government’s share of tax revenues but remains highly un- the central organization of the party declined from 13,000 to popular. Five years later the principle that taxes belong to the 5,000, and central planning has largely been abandoned. Under central government unless specifically assigned to localities is such conditions, local leaders have more incentives to gener- still widely contested at the local level. Further, subnational ate local economic prosperity than to follow some nationally governments continue to rely on extrabudgetary funds—some determined economic goal, and they have acquired substan- of them illegal—for the largest share of revenues. These tial autonomy in designing and implementing policy. China’s funds, combined with frequent (and also illegal) provincial central government can no longer unilaterally recapture the deficits, confer substantial fiscal independence on provincial powers it has conceded and may not even want to. In Febru- administrations. ary 1999, when a township elected its leader directly for the Decentralization Chinese-style does allow for considerable first time, the outcome was broadcast on national television, subnational autonomy. It creates incentives for local officials signaling official support for this event. to work toward local prosperity and has also been an effective China’s approach to decentralization relies on negotiations tool for instituting market reforms. But over time, the absence rather than rules to define relations between the central gov- of clear rules may threaten its successes. Decentralization has ernment and the four subnational tiers—provinces, prefec- accentuated a prereform tendency toward a fiefdom mentality tures/cities, counties, and towns. The allocation of responsibil- that hampers efforts to unify the national market and periodi- ities across tiers of government remains unclear except for cally threatens central control over macroeconomic stability. health and education, which are controlled by the provinces. Moreover, while administrative discretion has helped preserve On the revenue side, until the early 1990s local governments the momentum for growth and reform, it has also created op- were responsible for administering and collecting a large pro- portunities for rents that can be appropriated through financial portion of central government taxes, but their loyalty shifted corruption or political patronage. Official statistics show that away from the national government to the subnational level. by the end of 1998, 158,000 officials had been penalized by Provincial tax officers often used the tax administration sys- the Party’s Commission for Discipline Inspection, and corrup- tem to establish tax autonomy. They entered into direct nego- tion was one of the main themes of the National People’s Con- tiations with enterprises for payments (in lieu of the central gress, China’s parliament, in March 1999. ial representation over norms of population (or citizen) the regional governments themselves, as they are in representation. In bicameral systems the upper house Germany, India, Pakistan, and South Africa.36 In Rus- commonly gives equal weight to states and thus repre- sia provincial governors and prime ministers serve in sents regional interests in the national legislature. Sen- the upper house on an ex officio basis. In principle, ates in Argentina and Mexico award an equal number such explicit representation renders the upper house a of seats to each state or province regardless of popula- tool of regional governments. Again, however, the im- tion, giving small units of government disproportional pact on the national legislature depends on the powers voting power. In the Argentine senatorial elections, one granted to the upper house. vote in Tierra del Fuego is worth 180 votes in Buenos Finally, electoral arrangements matter. For example, Aires; in Mexico one vote in Baja California is worth electoral districts based on regional boundaries reinforce 31 votes in the state of Mexico. And in Brazil, senators political cleavages along regional lines. For this reason, representing less than 13 percent of the electorate con- the approach is seldom used.37 Among the large democ- trol 51 percent of the votes. In most bicameral coun- racies, Argentina, Brazil, Colombia, Italy, and Spain are tries, however, senates have limited powers, so the ef- the only ones defining legislative districts solely along re- fect of territorial representation is much greater if it is gional lines. Others rely on smaller subregional districts applied to the lower house.35 or have a number of legislators elected at large to repre- When members of the upper house are chosen to sent the whole country, rather than a specific region. represent regional interests, they can be elected directly The power of the executive. A central government’s by the people of that region, or they may be selected by ability to withstand regional pressure depends on the      ⁄  strength of the chief executive and on whether a clear Institutions can be designed to promote a common- majority emerges in parliament. Whether the executive ality of interests. The electoral system and the resulting is chosen by parliament or by direct popular vote (that party structure determine the degree to which the polit- is, whether the regime is parliamentary or presidential) ical system is nationally integrated. Holding national matters less than the powers of the executive in relation and local elections concurrently creates incentives to to the legislature. These powers include vetoes and the nurture the meaning of party labels and to develop ability to control the legislative agenda or to legislate nationwide parties.45 Legislative bodies that explicitly by decree. The degree to which the chief executive de- represent regions tend to promote integrated party struc- pends on the support of a political party is also a fac- tures.46 The executive authority of the central govern- tor. Thus the United States is a presidential regime, but ment relative to the regional government matters, since the constitutional division of powers forces presidents it determines whether the central government needs to to rely on their party’s support.38 govern by consensus or fiat. Similarly, the strength of the The electoral system also influences the vulnerability of chief executive’s powers determines the extent to which the executive to pressure groups. Proportional representa- the executive must rely on regional support. tion, which allocates seats in proportion to the share of There is no single best way to divide national politi- votes received by each party, tends to produce govern- cal power between central and subnational govern- ments that require a coalition of parties to govern.39 Such ments. Nor can a single constitutional provision ensure coalition governments are inherently less stable and more that central and subnational political elites will find it vulnerable to demands by interest groups than majoritar- in their interest to cooperate. But whatever system is ian governments.40 But proportional representation does adopted, it must not make the central government a allow disparate regional and ethnic interests to have a dis- prisoner of subnational interests. tinct voice in government. Combined with a parliamen- The structure, functions, and resources tary system of government, as in most Western European of subnational governments countries, proportional representation imposes the need to govern by consensus. Conventional political theory ad- The second major category of rules deals with the way vocates such a system for new democracies precisely be- subnational governments are structured, what they do cause it ensures a voice even for smaller groups, giving at each level, and how they are funded. These rules them a stake and presence in the new democracy rather need to be determined as a system, taking into account than shutting them out.41 But proportional representation the interactions among fiscal, political, and administra- combined with a presidential regime, as in Latin America, tive institutions. tends to produce executive-legislative deadlock.42 Structure and functions Creating incentives for national and subnational What is the best structure for subnational govern- governments to cooperate ments? The traditional approach of public finance For the balance of powers to be stable, a commonality economists to decentralization, known as “fiscal feder- of interests must develop between national and sub- alism,” calls for a subnational government structure national political elites.43 Political parties play a crucial with several tiers, with each tier delivering those ser- if often underestimated role in this process.44 In the vices that provide benefits to those residing in the ju- United States and Germany, national parties control the risdiction.47 Experience shows that this model, while state legislatures, whereas in Canada regional parties useful, has some limitations and that governments compete with each other in subnational elections. As a should seek instead to develop a regulatory framework result, Canada’s subnational politicians are often elected that allows for the sharing of responsibilities.48 on platforms explicitly framed in opposition to the na- The fiscal federalist framework and its practical limits. tional government and hardly ever move from provin- The fiscal federalist model identifies three roles for the cial to national elected office. In contrast, in the United public sector: macroeconomic stabilization, income re- States and Germany opposition between national and distribution, and resource allocation (in the presence subnational politicians is likely to be purely partisan, of market failure). The model assigns the stabilization and often national leaders in both countries start their role to the central government because it controls mon- careers as subnational elected leaders. etary policy and has more scope to use fiscal policy than :    subnational governments. The model also assigns in- Trends in mature decentralized countries suggest come redistribution to the center, since local attempts that costs are an important consideration. Most coun- at taxing the rich and redistributing wealth to the poor tries of the Organisation for Economic Co-operation would result in inefficient population movements— and Development (OECD) have a limited number of high-income groups would move to areas with low subnational tiers and jurisdictions (table 5.1). Some taxes, and low-income groups would concentrate in countries have recently been reducing the number of areas offering high benefits.49 subnational units, largely on grounds of efficiency and More recently the literature has recognized that cost.55 But in a number of developing countries, sub- while the central government should continue funding national governments are proliferating. In 1992 Mo- and designing redistribution efforts, local governments rocco increased the number of its municipalities from are often in a good position to implement and admin- 859 to 1,544 and made regions the third tier of subna- ister standardized national policies.50 In addition, local tional government.56 Even among very poor countries governments usually administer services that have im- such as Madagascar, Malawi, and Zambia, the trend is portant redistributive implications, such as primary toward a constant, if gradual, increase in the number health care, education, child care, housing, and public of local governments—perhaps in part because a block transportation. In poorer countries such services are grant available to each local government creates an in- often the only vehicle for providing in-kind transfers to centive to divide jurisdictions.57 poor households. Clarifying the allocation of functions and allowing for The fiscal federalist approach assigns a significant shared functions. Some services can be provided less ex- role to subnational government in allocating resources. pensively on a larger scale, or their benefits may spill This is because when the benefits of particular services over across districts. Providing these services centrally are largely confined to local jurisdictions, the appropri- creates economies of scale and captures externalities, ate levels and mix of services can be set to suit local pref- but at the cost of imposing a common policy on popu- erences. Local consumers can express their preferences lations with varied preferences and priorities.58 This by voting or by moving to other jurisdictions.51 In this trade-off, which is the basis of the fiscal federalist ap- respect, local politics can approximate the efficiencies proach, guides some of the choices that must be made of a market in the allocation of local public services. in allocating functions. The services central govern- This approach faces two practical obstacles, however. ments provide should benefit the entire economy or First, in developing countries where land and labor mar- exhibit substantial economies of scale—for example, kets may not function well and the democratic tradition national defense, external relations, monetary policy, or is in its infancy, it is not realistic to assume that people the preservation of a unified national market. Corre- can move easily between jurisdictions or make their spondingly, subnational units should provide local pub- voices heard through the political process.52 Second, es- lic goods. This model, which most established democ- tablishing separate tiers of government for each service racies have adopted, is also common to most countries is costly and poses serious coordination problems.53 that have recently decentralized, with the notable ex- The structure of subnational governments. The appro- ception of Bosnia and Herzegovina (see box 5.3). priate number of tiers of government and of jurisdic- Such responsibility-sharing arrangements are com- tions in each tier varies depending on a country’s phys- plex. But they work well when they are clear, when each ical characteristics, its ethnic and political makeup, and tier’s responsibilities are relatively well defined, and possibly its income level. But all countries face the same when the regulatory framework anticipates that local trade-off between representation and cost. The local governments are sometimes agents of the central gov- government of Midnapur in India may have difficulties ernment and sometimes principals acting on their own. managing local services in a way that is representative Without clarity and an appropriate regulatory frame- of the preferences of all its 8.3 million people. But very work, there can be no accountability. In South Africa small local governments—like those of Armenia, the the central government and the provinces have joint re- Czech Republic, Hungary, Latvia, and the Slovak Re- sponsibilities for health and education, but the exact public, which have an average population of less than responsibilities of each are not defined. The result is 4,000—are likely to use up most of their meager re- that provinces receive transfers to fund these services sources in fixed administrative costs.54 but use them for other purposes, knowing full well that      ⁄  Table 5.1 The structure of subnational governments in large democracies Country Intermediate Local Country Intermediate Local Industrial countries Kenya 39 county councils 52 municipal, town, and Canada 10 provinces, 4,507 municipalities urban councils 2 territories Korea, 6 special cities, 67 cities, 137 counties France 22 regions, 36,772 communes Rep. of 9 provinces 96 departments Malaysia 13 states 143 city, municipal, and Germany 13 states, 329 counties, district councils 3 city-states 115 county-free cities, Mexico 31 states, F.D. 2,412 municipios 14,915 municipalities Mozambique 10 provinces 33 municipalities Italy 22 regions, 8,100 municipalities 93 provinces Nepal 75 districts and 4,022 village panchayats town panchayats Japan 47 prefectures 655 cities, 2,586 towns Pakistan 4 provinces 15 municipal Spain 17 autonomous 50 provinces, corporations, 457 communities 8,097 municipalities municipal and town United Counties 540 rural districts, committees, 40 Kingdom metropolitan districts, cantonment boards, and London boroughs 4,683 union and district councils United 50 states, F.D. 39,000 counties and States municipalities, 44,000 Philippines 76 provinces 64 cities, special-purpose local 1,541 municipalities, authorities 41,924 barangays Poland 16 provinces, 2,489 gminas Other countries 307 poviats Argentina 23 provinces 1,617 municipios Russian 21 republics, 1,868 raions, Bangladesh — 4 city corporations, 129 Federation 17 territories or 650 first-tier cities, pourashavas (smaller autonomous areas, 26,766 secondary municipalities), 4,500 49 provinces cities, townships, union parishads (which (oblasts), 2 cities and villages group 85,500 villages) of federal status Brazil 27 states, F.D. 4,974 municipios South Africa 9 provinces 850 local authorities Colombia 32 departments, F.D. 1,068 municipalities Thailand 75 changwats, 6,397 districts, Ethiopia 9 regions, plus 550 woredas Bangkok 148 municipalities and 2 special city cities administrations, Turkey 74 provinces 2,074 municipalities 66 zones Uganda 45 districts, 950 subcounties, India 25 states, 3,586 urban local 13 municipalities 39 municipal divisions, 7 union territories bodies (95 municipal 51 town councils corporations, 1,436 municipal councils, Ukraine 24 regions (oblasts), 619 districts 2,055 nagar panchayats), 1 autonomous 234,078 rural local republic, bodies 2 municipalities Iran, 25 provinces 720 districts and Venezuela 23 states, F.D. 282 municipalities Islamic municipalities Republic of —Not applicable. F.D. Federal district. Source: Appendix table A.1. :    the central government will intervene to provide the some Indian states rely on the octroi (a tax on goods cir- needed service. culating across regional or municipal boundaries). Al- though efficient, such taxes are politically easier to im- Assigning and controlling resources pose, since their effects are hidden in the price of goods. The question of which tier of government controls which As a result, even mature decentralized democracies such resources is perhaps the thorniest issue of decentraliza- as Germany and the United States resort to them. Over- tion. The ability of subnational authorities to act in- all, subnational taxes are seldom a large share of subna- dependently of the central government depends on tional revenues (see figure 5.2), although there is scope, whether they have access to independent tax bases and particularly in developing countries, for improving local sources of credit.59 Experience provides two lessons in revenue collection.62 For intermediate levels of govern- this area. First, subnational governments need resources ment, the problem of matching taxes to the jurisdiction commensurate with their responsibilities. Second, sub- is even more complicated (box 5.6). national authorities must operate under firm budget con- The role of transfers. Since transfers account for a straints, so that they do not spend or borrow excessively large part of subnational finances everywhere, their de- in the expectation of a central government bailout.60 sign is a critical factor in the success of decentraliza- The guiding principle of revenue assignment is tion.63 Transfers are needed to fund the services local straightforward: finance should follow function. This governments provide on behalf of the central govern- is so not only because resources must be commensurate ment (while local revenues should ideally cover local with what they fund, but also because the type of rev- expenditures). And transfers are essential to ensure that enues used affects consumer behavior and results in dif- decentralization does not occur at the expense of equity, ferent patterns of incidence. User charges, such as bus particularly if the central government relies on pro- fares or water bills, affect the amounts consumed by grams administered at the subnational level to redistrib- users and are borne only by those who actually con- ute income or if there are large income differences sume the service. Overall, the appropriate structure of across districts. Finally, governments can use transfers subnational finance—the mix of user charges, taxes, to influence the sectoral pattern of local expenditure by and transfers—depends on the functions that have earmarking transfers or disbursing them in the form of been assigned to each tier of government. matching grants. Certain forms of taxation are appropriate for financ- Although transfers are almost always necessary, they ing local services with benefits that cannot be confined should not be so large as to eliminate the need for local to individual consumers, such as local roads. Such taxes taxes.64 Local taxes ensure that subnational governments must fall on the residents of the jurisdiction and must face, at least to some degree, the political consequences be direct—that is they must directly target individuals of their spending decisions. And political necessity or personal property so that their burden is local. Good sometimes imposes the need for relying heavily on local examples are the property tax, the personal income tax, taxes. Tax sharing was one of the more contentious is- and capitation or head tax. Indirect taxes such as the sues in the Yugoslav federation, where wealth differed value added tax (VAT) or corporate income tax, which greatly across ethnic groups and redistribution issues can be built into the price of the goods and passed on were embroiled in ethnic tensions. Similarly, the search to consumers outside the taxing jurisdictions, are not for a good regional tax is of paramount importance in generally appropriate as local taxes. Ethiopia, where regions are defined on the basis of eth- But direct taxation in developing countries often nic identity (see box 5.3). yields limited revenues. The income tax is of limited use Transfers have three variables.65 The first variable is where most of the economy operates informally. In the amount to be distributed. This can be fixed as a per- many countries the capitation tax, which was one of the centage of national taxes, or it can be an ad hoc deci- main forms of taxation in colonial times, is politically sion, sometimes to reimburse preapproved expendi- unacceptable. And the property tax, which requires tures. The second variable is the criteria for distributing good information systems, is usually poorly adminis- transfers among jurisdictions. In Argentina, for exam- tered.61 To compensate, most municipalities rely on ple, a predetermined formula is used to allocate a fixed various forms of business taxation. Jordan imposes a percentage of certain national taxes, whereas in India business license fee, Brazil has taxes on services, and the central government periodically determines, on the      ⁄  Box 5.6 their amount needs to be equivalent to the tax base they Financing intermediate tiers of government are replacing. In practice, however, most transfers take the form of block grants. This tendency may reflect a Intermediate tiers of government, such as states and search for administrative simplicity, or it may reflect the provinces, often have substantial responsibilities that can- not be funded solely through user fees.66 Yet direct taxes reluctance of subnational governments to accept any have limited yields in developing countries and tend to be restrictions on the use of transfers. In countries where allocated to local governments. Indirect taxes are generally subnational interests are well represented in national more appropriate for the national government, since the parliaments—France, Germany, Japan, and the United burden of such taxes can be passed on to consumers out- side the taxing jurisdiction (a problem referred to as tax Kingdom, for example—block grants account for the exporting). No perfect solution exists for the problem of bulk of intergovernmental transfers. financing the intermediate tier of government, and in prac- Some basic principles are applicable across all coun- tice large federal countries typically use a combination of tries and all types of transfers. Transfers should be de- two approaches. The first approach consists of granting exclusive rights termined as openly, transparently, and objectively as to a broad-based tax, such as an income tax or a value possible. They should be kept reasonably stable from added tax (VAT), to the intermediate tier. The income tax year to year so that local governments can plan their has the advantage of affecting only residents of the state budgets. And they should be distributed on the basis of or region, avoiding the tax-exporting problem, but is of lim- predetermined rules, which need to be kept as simple ited yield in poor countries. A VAT like that used in Brazil, Russia, and Ukraine provides substantial resources but as possible.67 Simplicity, transparency, and predictabil- raises issues of interstate smuggling and tax exporting. In ity would help eliminate one of the worst problems of fact, subnational VATs are so complex to administer that decentralization: the uncertainty and bargaining that they should only be considered in countries with efficient often plague intergovernmental fiscal relations. tax administrations. State corporate income taxes also pre- sent administrative difficulties, notably the problem of de- Controlling subnational debt. Subnational borrowing termining in which state a company has realized its profits. has emerged as one of the thorniest issues for decentral- The second approach is to share national taxes. This ization. In principle, it is a private transaction between can be implemented in a variety of ways. One is to let the borrower and lender. But the national government is states set a surcharge on a nationally administered and col- lected tax—which does present the advantage of making often drawn reluctantly into the transaction because state government bear at least part of the political burden of its responsibility for the stability of the financial sys- of a tax. Another is pure tax sharing, in which the central tem. As a result, subnational borrowing is almost always government remits a part of its tax revenues to the juris- subject to the assumption that the central government diction in which they were collected. Mexico, for example, imposes a national VAT which it redistributes to states on will fund a bailout if necessary—an assumption that the basis of what they would have received by imposing leads banks to lend to uncreditworthy local governments. this tax themselves. Argentina uses a similar system. Pure An alternative to the private financing of subnational tax sharing has no advantage over surcharges except for borrowing is for the central government to provide preserving a uniform tax rate. Revenue sharing, which re- lies on a formula for allocating the proceeds of a national long-term credit, lending either directly or through tax across different regions, is similar, although it can be intermediaries. In most countries—particularly those used to equalize revenues across jurisdictions regardless with shallow financial systems—this remains the prin- of their tax base. cipal source of subnational credit and largely dominates private financing. But the repayment record for cen- trally sponsored financial intermediation is poor (see basis of need, both the levels of the transfers and the chapter 6). Loan allocation tends to become politicized, method of distribution. The third variable concerns the while debt collection is often lax, with national taxpay- conditionalities imposed on the use of transfers. Trans- ers ultimately bearing the financial burden of bad loans. fers can be earmarked for specific uses, such as paying In general, however, private financing is either al- teachers’ salaries, or left unrestricted. ready the primary source of subnational credit or is Transfers should be designed according to their ob- meant to eventually replace central government financ- jectives. Those intended to finance functions that the ing. This requires developing means to protect the municipal government is performing on behalf of the central government and the national financial system central government should be earmarked. Transfers in- from exposure to excessive subnational debt. As shown tended as substitutes for local taxes should not, but in table 5.2, short of outright prohibition, four ap- :    Table 5.2 Subnational borrowing controls in selected countries Market Cooperative Administrative Rule-based Borrowing discipline control control control prohibited Overseas Domestic Overseas Domestic Overseas Domestic Overseas Domestic Overseas Domestic Industrial countries Australia • • Austria • • Belgium • • Canada • • Denmark • • Finland • • France • • Germany • • Greece • • Ireland • • Italy • • Japan • • Netherlands • • Norway • • Portugal • • Spain • • Sweden • • Switzerland • • United Kingdom • • United States • • Developing countries Argentina • • Bolivia • • Brazil • • Chile • • Colombia • • Ethiopia • • India • • Indonesia • • Korea, Rep. of • • Mexico • • Peru • • South Africa • • Thailand • • Transition economies Albania • • Armenia • • Azerbaijan • • Belarus • • Bulgaria • • China • • Estonia • • Georgia • • Hungary • • Kazakhstan • • Kyrgyz Republic • • Latvia • • Lithuania • • Poland • • Romania • • Russian Federation • • Slovenia • • Tajikistan • • Ukraine • • Uzbekistan • • Note: Classifications attempt to capture the predominant form of control. In most countries, the approach used involves a combination of several techniques. For detailed country-by-country explanatory notes, see Ter-Minassian and Craig (1997). Source: Ter-Minassian and Craig 1997.      ⁄  proaches are used to control subnational borrowing. ing of state debt in 10 years. Each debt crisis arose de- The first approach relies on market discipline; the sec- spite a blanket ceiling on subnational borrowing and a ond relies on cooperation between the central and sub- web of restrictions and controls on various forms of national governments to decide what constitutes an debt. Regulation, it seems, failed to withstand the pres- appropriate level of indebtedness; and the other two sure from strong regional interest groups. Even in in- directly regulate subnational borrowing. In practice, dustrial countries with sophisticated credit markets, countries use a combination of all four approaches. borrowing controls are subject to slippage.70 In the In principle, central governments can simply refuse United States, for example, regulations are less impor- to intervene in transactions between subnational gov- tant than market discipline. Bonds must be floated, and ernments and their creditors, relying on market disci- the federal government neither guarantees subnational pline to control subnational debt. This is the most im- debt nor bails out subnational governments.71 portant restraint on subnational borrowing in Canada, France, and Portugal, for example. But to be effective, Central regulation of subnational governments a laissez-faire approach requires that a number of con- Rules are needed to govern relationships among tiers of ditions hold—the most important being the credibility government. But central governments in decentralizing of the central government’s commitment not to inter- countries tend to compensate for their loss of direct con- vene.68 Establishing this credibility requires time, par- trol by stepping up their regulation of subnational gov- ticularly where bailouts have occurred in the past. It ernments. This tendency can be counterproductive if also requires avoiding situations in which the central central governments with only a limited knowledge of government would be forced to intervene—for exam- local conditions begin micromanaging local functions, ple, where a default threatens the national banking sys- or if they impose costs they are not prepared to finance. tem or the country’s international credit rating. Regu- Personnel matters are one area in which central regu- lation can help prevent such situations. lation is generally undesirable. Since wages are often a Some types of regulation are better than others.69 Di- very large part of local budgets, centrally mandated wage rect government controls, like annual limits on borrow- increases can cause a local fiscal crisis. The central regu- ing or administrative authorization for loans, are subject lation can prevent subnational governments from re- to political bargaining and are generally at odds with the sponding to local conditions by increasing or decreasing trend toward decentralization. Further, they may make staff size or by keeping wages at market levels. In Turkey it even more difficult for the central government to the central government creates the staff list for each mu- refuse to intervene and rescue a subnational government. nicipality, along with the corresponding salary scale. The But administrative controls are appropriate for external central government must approve any changes in a long borrowing because a subnational government’s behavior process that involves the Ministry of the Interior, the on the international market could have contagion effects state personnel organization, and the Council of Minis- on the ratings of other national borrowers and because ters. In Sri Lanka the central government determines the managing the external debt is part of the macroeconomic wage bill for provincial governments. responsibilities of a central government. If a central government is concerned about nepotism Rule-based controls like ceilings on debt-service ra- or overstaffing at local levels, it can address them in other tios or constraints on the type or purpose of borrowing ways. For example, it can provide suggested hiring levels are more transparent and less subject to political inter- and salary scales and require subnational governments ference. They function best when they set global limits to publish their employment rolls. But central govern- that mimic the markets—for example, by establishing ment involvement in personnel matters also reflects the ceilings on debt service as a share of revenues—and rely power of public sector unions and their ability to orga- on a global definition of what constitutes debt. De- nize nationally. This force has not been easy to counter, tailed regulations are hard to monitor and will encour- whether in developing or in industrial countries.72 age behavior aimed at circumventing them. Central government regulation remains appropriate Fundamentally, however, rules and controls will be in a wide range of other circumstances. When subna- ineffective unless accompanied by market discipline tional governments act as agents of the central govern- and a credible “no-bailout” pledge by the central gov- ment, regulation and monitoring are needed to enforce ernment. Brazil has just completed the third restructur- national mandates and standards. Even countries that :    have granted substantial autonomy to subnational gov- or province, minorities and low-income candidates are ernments require that centrally financed welfare pay- more likely to run and to win seats. In turn, the pres- ments be distributed according to criteria the central ence of such candidates shows minorities and the poor government establishes. Regulation is also essential to that they can play a role in the political decisionmaking ensure the validity of the local electoral process and to process and encourages them to mobilize and vote.73 address conflicts between units of subnational govern- The visibility of an election also influences partici- ment. But a free press, improved access to information, pation. In general, the more local an election is, the and the growth of democracy at subnational levels are lower the participation.74 As voter turnout drops, the decreasing the need for central regulation. Local inter- chances increase of narrowly focused special-interest est groups are increasingly able to monitor the perfor- groups gaining power. This problem suggests that there mance of local governments. is a trade-off between full representation, which re- quires small districts, and participation, which is en- Making subnational governments accountable couraged by the relatively high levels of visibility that The third major set of constitutional rules consists of come with elections in larger districts. those governing relations between local officials and Two measures can help increase visibility without re- their constituents. The degree to which local officials are quiring an increase in the size of local electoral jurisdic- accountable to their constituents determines whether tions. One is to hold concurrent local and national elec- decentralization produces the intended benefits—that tions, although this approach carries with it the risk is, more efficient and responsive services, and greater that national issues will overshadow local concerns. An- local self-determination. The process for electing gover- other is for the mayor or governor to be elected directly nors, mayors, and members of the subnational legisla- by the whole constituency, while state assembly mem- ture takes center stage in determining accountability. bers or municipal councilors are elected by district or But elections in and of themselves are not sufficient to neighborhood. Together, these measures help ensure ensure that local governments are truly responsive to higher voter participation and better representation people’s needs and wants. Three sets of complementary across social or income groups.75 measures should be pursued. First, electoral rules need Rules that promote effective governance. Effective gov- to encourage participation and representation and, at ernance requires stable coalitions and an executive with the same time, allow an effective majority to emerge. reasonably strong and clear powers. The probability Second, civil society can be drawn upon to complement that elections will produce a stable coalition is higher formal political processes. Finally, an effective local ad- with majority voting than with proportional represen- ministration needs to develop. tation, as explained earlier. Local governments com- posed of stable coalitions govern better than unstable Adopting effective electoral rules partnerships—for instance, they are better able to take Electoral rules affect whether local politics reflect the in- the measures needed to adjust to shocks.76 terests of the local population or are captured by local Separating the executive and the legislative branches elites. Of course, rules interact with certain characteris- of local government and electing the chief executive di- tics of civil society, such as education, access to informa- rectly may also yield more effective governance.77 May- tion, and the existence of groups that have a voice in ors elected directly are more likely than appointees to government. But making elections highly visible events, challenge the status quo. The vast majority of major facilitating participation, and demonstrating that votes municipal reforms around the world have been initi- matter will affect electoral outcomes in any society. ated by strong mayors. But too much authority concen- Rules to improve visibility, participation, and expected trated in the executive may not be appropriate, partic- payoffs. The size of electoral districts can influence the ularly in new democracies. The mayor of Moscow had outcome of an election. Electing council members by enough power to modify the city’s electoral laws against ward or neighborhood rather than at large ensures that the wishes of the legislative assembly. all geographically defined interest groups will have seats on the local council. This method also reduces the costs Harnessing civil society of running for office. Since candidates need to cam- A multitude of actors outside the public sector—grass- paign only in a single ward rather than in an entire city roots organizations, unions, universities, philanthropic      ⁄  foundations, user groups, nongovernmental organiza- pates. This was the case in the democratic revolutions tions (NGOs), and neighborhood associations—influ- of Africa, Eastern Europe, and Russia. Political parties ence public performance. Among other things, they can help maintain a continuing link between civil soci- can help hold local governments accountable. Such ety and government. Parties aggregate the demands of groups, known collectively as “civil society,” can also a dispersed population, represent political interests, re- complement local administration in the search for cruit and train new candidates for office, ensure elec- more responsive and effective governance. toral competition, and form governments. They can Civil society and formal political participation. How help organize minorities and the poor and facilitate can governments encourage the participation of civil so- their participation in the formal electoral process.84 ciety in governance? Much depends on the strength of Party systems thus improve legitimacy and governabil- community organizations and their ability to organize. ity by making the democratic process more inclusive, Local officials must also be willing to tap into these accessible, representative, and effective.85 groups. But examples abound of collaboration between civil society and local governments. In Colombia local Developing an effective local administration governments and community associations work to- Improving local services requires an effective local ad- gether to provide infrastructure for the poor. In Brazil, ministration. Even a well-meaning political team can- Chile, Mexico, and Venezuela many municipalities have not overcome incompetent administration. In fact, lack adopted participatory budgeting and hold open meet- of capacity at the local level and the need for a massive ings to consult the population on its priorities. Donors increase in skilled staff are the arguments most fre- everywhere have initiated projects to mobilize commu- quently invoked against decentralization. nity resources and encourage participation.78 Both central and local governments can take mea- The formal participation of civil society in public sures to improve the effectiveness of local administra- life has limits. Active civic organizations cannot be cre- tion.86 First, when a central government has decentral- ated in a vacuum but instead need to draw on local tra- ized responsibilities, it can also devolve the appropriate ditions. In Bolivia, for example, the neighborhood as- staff, as the Ugandan government did. Second, local sociations that report municipal mismanagement to the governments should be free to hire, fire, and offer ap- national senate are built on traditional customs.79 Fur- propriate incentive packages so that they can attract ca- ther, civic organizations are not always effective and pable local officials. Third, privatization can reduce the may only reflect the views of a narrow segment of the number of skilled administrators needed by local gov- population.80 But where civic organizations are weak, ernments, since the privatized services require only mon- local governments can use other mechanisms to give itoring and regulation rather than actual management. the public a voice, such as polling or collecting data While problems of capacity constraints are sur- from user groups.81 mountable, they deserve serious attention. Central gov- Civil society and political parties. Democratic revolu- ernments need to provide technical support to local gov- tions are often driven by a popular upsurge and the res- ernments as part of the process of decentralization. urrection of civil society. In Latin America’s move to- Decentralization itself, by giving subnational govern- ward greater democratization, trade unions, grassroots ments greater responsibilities and control over resources, movements, religious groups, intellectuals, and artists will then increase their incentives to invest further in supported each other’s efforts, coalescing into a whole their own administrative capabilities. that identifies itself as “the people.” 82 In a number of Policies for the transition African societies, popular respect gave religious leaders a status and influence that autocratic regimes could not Decentralization typically takes place during periods of ignore. And the activities of trade unions were crucial political and economic upheaval. Euphoria at the fall of in many countries. Strikes prompted by industrial griev- an authoritarian regime, an economic crisis that precip- ances, such as late payment of wages, against the govern- itates a regime’s collapse, the jockeying for power of new ment in its role as dominant employer rapidly exploded interest groups—all these conditions create an environ- into demands for political reform.83 ment in which a careful, rational, and orderly process Once democratic movements achieve their immedi- of decentralization is highly unlikely. Even when decen- ate goals, the civic energy that fueled them often dissi- tralization occurs in a less dramatic context, questions :    of strategy and timing still arise. The recent experiences Box 5.7 of decentralizing countries can help answer them. The cart before the horse: decentralization in Russia Synchronizing the elements of reform Under the Soviet system, subnational governments were The most compelling lesson of recent decentralization merely extensions of the central government under the experiences is that all elements of reform must be syn- authority of the Communist Party. The central government chronized. The political impetus behind decentraliza- controlled activities of national importance, such as trans- tion prompts central governments to make concessions portation and defense. The republics were responsible for hastily. Granting local elections is a step that can be light industries. Provinces (oblasts) were responsible for health care, housing, utilities, and education. Although taken rapidly. But making decentralization a success re- each tier of government was assigned a given tax base, quires taking a number of slow and difficult steps that the central government determined subnational budgets create new regulatory relationships between central and through central planning and closed-door negotiations. subnational governments, transfer assets and staff to Revenue sharing and intergovernmental transfers were local levels, and replace annual budgetary transfers with merely accounting devices used to bring each subnational budget into balance. a system of tax assignment and intergovernmental The party’s monopoly on power was officially abolished transfers. The recent history of decentralization illus- in 1990. Following the breakup of the Soviet Union in 1991, trates the dangers of not sequencing appropriately. a new constitution (adopted in 1993) declared Russia a Put expenditure and revenue rules in place before po- democratic federal state. The new constitution recognized litical liberalization. Russia liberalized politically while 89 subnational units (republics, autonomous regions, and oblasts) and mandated the election of governors (presi- the fiscal structure of the former Soviet system was still dents in the republics) and legislatures in each jurisdiction. in place (box 5.7). Subnational governments had his- However, Russia continued to struggle with its old sys- torically acted as tax collectors for both the provinces tem of intergovernmental fiscal relations for several years. and the central government. Once the provincial gov- Despite an attempt to establish a system based on sepa- rate tax assignments, subnational finances continued to ernments gained political autonomy, they began refus- depend on negotiations with Moscow. These talks soon ing to send tax revenues to the central government. Fis- became hostile, and the newly autonomous regional gov- cal relations stabilized only after 1994, when fixed rules ernments threatened to withhold the tax revenues they were established for dividing taxes among tiers of gov- owed to the federal government or to secede from the ernment. In contrast, Chile and Poland established fis- federation entirely if their demands were not met. Since 1994 Russia has been moving toward a rule-based cal rules before political liberalization and were spared system of intergovernmental fiscal relations. The 1994 re- a Russian-style fiscal crisis. forms divided revenues from each of the major taxes Decentralize a function and its corresponding revenue among central and regional governments and established a source simultaneously. Many African countries facing formula-based equalization system to assist poorer regions. economic collapse devolved a broad range of govern- However, the reforms did not entirely resolve the fiscal con- flicts between levels of government or settle the division of ment services to subnational governments without pro- responsibilities for social expenditures. Moreover, the fed- viding the necessary revenues. Not surprisingly, the eral government still runs considerable risk from potential quality of the decentralized services declined sharply. In defaults on loans to subnational governments. much of Latin America the opposite occurred: govern- ments decentralized revenues without offloading corre- Source: Freinkman 1998; Le Houerou 1996; Martinez- Vasquez 1998. sponding responsibilities. In Colombia central transfers to municipalities increased by 60 percent without a matching increase in responsibilities. Decentralize the needed management controls. Gov- subsequent decision to grant a major increase in salaries ernments have sometimes crippled local governments’ prompted a fiscal crisis at the provincial level that was ability to perform new functions by failing to decen- resolved only though the creation of a special compen- tralize management controls. In Colombia, for exam- sation fund.87 In Poland the public housing stock was ple, the central government continued to set the salaries transferred to municipal governments, but the central of public school teachers even after the management of government continues to control the rents. primary and secondary schools had ostensibly been de- The recent decentralization of education in Mexico centralized to the provinces. The central government’s followed a more balanced approach. The federal gov-      ⁄  ernment transferred full management responsibility for final outcomes is yet available.89 Nonetheless, some les- preschool, primary, and secondary education to the sons have emerged from recent experiences. Perhaps the state governments in 1992, along with funding that most important is that a system that is based on rules equaled spending on federal facilities in the previous produces better results than one that is not. Explicit year. Since then, funding has been based on a formula rules setting out the division of functional responsibili- that gradually shifts the distribution from its historic ties among levels of government reduce ambiguity and pattern to one that provides an equal amount per pupil increase political accountability. They also provide a across all states. The experience of the Philippines has framework within which interest groups can compete been similar.88 and negotiate without resorting to violence. Some rules work better than others. Revenues need Demonstrating the hard budget constraint to be decentralized at the same time as expenditures, so Central governments must demonstrate early on that that finance follows function. A “hands-off ” attitude they are committed to imposing a hard budget con- when subnational governments default on their loans straint on subnational governments. The mere possibil- may be more important in controlling debt than the ity of a central government bailout can prompt excess most comprehensive set of regulations and controls. spending and deficit financing at the subnational level. Ward-based local politics combined with direct elec- Brazil, where the federal government has assumed over tions for mayors and governors, and concurrent na- $100 billion in state debt, is a clear example (see chap- tional and local elections, improve participation and ter 8). Argentina, in contrast, succeeded in enforcing a representation. Subnational governments with multi- hard budget constraint. From the outset, the current ple tiers and many small units are likely to have high administration has refused to provide any significant administrative overhead costs. debt relief to the provincial governments. It has also Strategies to stop decentralization are unlikely to minimized its potential exposure in two ways. First, succeed, as the pressures to decentralize are beyond gov- provinces may not borrow directly from the federal trea- ernment control. The emergence of modern econ- sury. Second, loans that provincial banks make to their omies, the rise of an urban, literate middle class, and governments are not eligible for central bank discounts. the decline of both external and domestic military After the 1994 Mexican economic crisis temporarily threats have created nearly insurmountable pressure for dried up funding sources all over Latin America, Argen- a broader distribution of political power in Latin Amer- tina’s provincial governments were forced to adjust ica, Eastern Europe, Russia, and parts of East Asia. This rather than rely on federal relief. same pressure is likely to affect the rapidly urbanizing economies of South Asia and parts of Africa early in the What lessons for the future? 21st century. Rather than attempt to resist it, govern- Decentralization is a work in progress. Many experi- ments should face decentralization armed with lessons ments are under way, and only limited evidence on from countries that have gone before them. C h a 6 p t e r Dynamic Cities as Engines of Growth A s countries move through the develop- world, it also reinforces certain advan- ment process, agriculture declines as a tages of proximity. Firms competing in share of gross domestic product (GDP), the global economy (and their suppli- and manufacturing and services begin ers) still benefit considerably from ac- to dominate the economy. Goods and cess to a sizable pool of labor, materials, services are often produced most effi- services, and customers. As a result, glob- ciently in densely populated areas that alization is likely to contribute to fur- provide access to a pool of skilled labor, ther urbanization. This is particularly a network of complementary firms that true in developing countries, where act as suppliers, and a critical mass of access to the opportunities offered by customers. For this reason sustained globalization is much greater in cities. economic growth is always accompa- The growth of urban populations in nied by urbanization (figure 6.1). both large capital cities and smaller mu- Globalization and localization have nicipalities feeds demand for increased not diminished the importance—or the localization of political power. It puts pace—of the urbanization process. Glob- pressure on national institutions of gov- alization promotes economic growth, ernance and encourages them to take which is the driving force behind ur- the steps toward decentralization dis- banization. But communication and in- cussed in chapter 5. It makes the success formation technologies now allow firms of decentralization perhaps even more to market their goods in distant coun- important. When urban governments tries and to incorporate into their pro- have the power and ability to enact a duction chain firms located halfway development agenda, they can help the around the world. If globalization is citizens of their cities hook up with the lauded precisely because of its ability to global economy. These cities then be- make great distances seem much smaller, come reliable links in the global pro- why does urbanization remain such an duction chain and attractive destina- important trend? tions for foreign investment. Although globalization opens up new Urbanization is integral to develop- possibilities for linkages around the ment, but it also presents difficult chal-       ⁄  Figure 6.1 Urbanization is closely associated with economic growth Urban population (percentage of total population) 100 90 80 70 60 50 40 30 1970 20 1995 10 0 100 1,000 10,000 100,000 GDP per capita (1987 U.S. dollars) Note: Sample includes industrial and developing countries for which data are available. The figure shows progress from 1970 to 1995 in each country. GDP per capita is on a log scale. Source: World Bank, World Development Indicators, 1998. lenges. This chapter reviews the economic forces ining the urbanization process—the agglomerative underlying urbanization and discusses what national forces and locational inducements that shape cities—is governments can do—and should not do—if they want a useful way of identifying what role governments to foster urban economic growth. Chapter 7, in turn, should play. focuses on what makes cities livable, including essen- tial services like housing, sanitation, and infrastructure. Agglomeration economies—the source of urban efficiency What makes cities grow? Why is economic activity concentrated in urban areas, Healthy, dynamic cities are an integral part of sustained where land prices are often 50 to 100 times higher than economic growth (box 6.1).1 As countries develop, cities they are 30 or 40 miles away? Why do so many indi- account for an ever-increasing share of national income. viduals and firms settle in large metropolitan areas Urban areas generate 55 percent of gross national prod- where the cost of living is typically twice as high as it is uct (GNP) in low-income countries, 73 percent in in smaller urban areas?3 The answer must be that these middle-income countries, and 85 percent in high- costs are more than offset by the economic benefits income countries. The growth sectors of the economy— cities offer—benefits that are generally the result of ag- manufacturing and services—are usually concentrated in glomeration economies. cities, where they benefit from agglomeration economies Agglomeration increases the productivity of a wide and ample markets for inputs, outputs, and labor, and array of economic activities in urban areas. Productiv- where ideas and knowledge are rapidly diffused.2 ity rises with city size, so much so that a typical firm The way cities manage development, including the will see its productivity climb 5 to 10 percent if city size arrival of industries, goes far in determining the rate of and the scale of local industry double.4 Urban wages economic growth. Urban governments can foster eco- are also higher than rural wages—two to four times as nomic development, or they can slow it down. Exam- high in middle-income countries—reflecting the higher        Box 6.1 rate with distance.7 When firms are concentrated in Cities and urban areas: some definitions cities, transaction costs also fall, most notably the search costs involved in matching workers with employment This report uses the terms cities and urban areas inter- opportunities. changeably. The formal definition of urban areas describes them as concentrations of nonagricultural workers and Agglomeration economies come in various forms. nonagricultural production sectors. Most countries call set- Benefits that derive from firms locating close to firms tlements with 2,500–25,000 people urban areas. The defi- in the same industry are known as localization econ- nition varies from country to country and has changed over omies. Benefits that derive from proximity to many dif- time. If the criteria China used in its 1980 census had been applied to its 1990 census, the country’s urbanization rate ferent economic actors are known as urbanization for the 1980s would have been more than 50 percent—far economies. Evidence from Brazil and the Republic of more than the 26 percent produced by the more rigorous Korea shows the benefits of localization economies. If a approach used in 1990. A city has a certain legal status plant moves from a location shared by 1,000 workers (granted by the national or provincial government) that is employed by firms in the same industry to one with generally associated with specific administrative or local government structures. In most countries large urban 10,000 such workers, output will increase an average of areas are referred to as metropolitan areas because they 15 percent, largely because the pool of specialized work- encompass a geographic area of human settlement (that ers and inputs deepens.8 Whether an industry benefits may include legally defined cities) within which residents most from urbanization or localization economies de- share employment opportunities and sets of economic relations. pends on how innovative it is. New, dynamic industries are likely to locate in large urban centers where they can Source: Mills 1998; UNCHS 1996. benefit from the cross-fertilization provided by diverse actors. Older, mature industries concentrate in smaller, more specialized cities, where congestion costs are low productivity levels obtained from urban agglomeration and localization economies can be high. economies.5 A final benefit of agglomeration in large urban areas Urban areas have historically been more efficient is that these locales are less vulnerable to economic fluc- than rural areas because cities had markets for inputs tuations because of their diversified economic base. and outputs big enough to support good-sized plants Employment can flow from one sector to another, and thus could take advantage of economies of scale. keeping average unemployment low.9 The number and In smaller towns the economies of scale such plants variety of consumers offer firms some protection, al- provided were offset by high transportation costs to lowing them to apply the law of large numbers to in- consumers or from input sources. The relationship be- ventory management (a practice that results in substan- tween plant size and city size has all but disappeared, tial savings). For consumers, large cities provide a variety however. Transportation costs have also declined (and of services and shopping and entertainment opportu- become much less important) as services and light in- nities. Rural areas can tap into these benefits by build- dustries increasingly dominate the world economy. ing links to the urban sector (box 6.2). In a modern economy the benefit of the kind of proximity urban areas offer is that firms, regardless of Systems of cities size, are able to experience economies of scale and Although productivity is higher in large metropolitan scope. The presence of a common pool of labor, mate- areas, almost 65 percent of the world’s urban residents rials, and services allows large and small firms alike continue to live in small and medium-size cities (figure to profit from scale economies. Economies of scope 6.2). This pattern reflects the degree of agglomeration emerge when the presence of one activity makes carry- that works best for firms and industries and the kinds ing out a complementary activity cheaper by fostering of benefits agglomeration provides. Large metropolitan diversity in supply and specialization among firms.6 areas provide some firms with enough benefits to jus- Proximity also facilitates the diffusion of knowledge. tify the high labor and land costs. But other industries Firms operating in proximity to each other benefit from find smaller cities more lucrative bases. Economies can information spillovers, in some cases by observing what support a range of cities of different sizes and the ac- neighboring firms are doing. Evidence from patent companying variations in production patterns. And the citations shows that information flows actually deterio- effects of city size on workers are often minimal. A typ-      ⁄  Box 6.2 Figure 6.2 Rural-urban linkages Most of the world’s urban population lived in small and medium-size cities in 1995 Thinking on the links between urban and rural develop- ment has changed in the past 50 years. In the 1950s Share of total urban population (percent) urbanization was considered a desirable alternative to rural 70 overcrowding, particularly in densely populated areas 63.5 where the prospects of raising agricultural productivity seemed limited. Manufacturing was seen as a key to 60 growth. But manufacturing often failed to produce enough jobs for rural migrants to cities. As a result, governments 50 worried about the rising number of underemployed in large cities and sometimes tried to restrain rural-urban mi- gration—a policy that had the effect of lowering the mi- 40 grants’ welfare. In principle, urban and rural economies can enjoy a 30 symbiotic relationship. Cities benefit when agricultural pro- ductivity increases. Growing rural areas provide new, im- 21.4 portant markets for urban services and manufactured 20 goods. Mechanization and the use of fertilizers, pesticides, 15.1 and herbicides spur demand for these products. A boom 10 in commercial agriculture boosts demand for marketing, transportation, construction, and finance, which urban centers often provide. In Africa every $1 of additional out- 0 put in the agricultural sector generates an extra $1.50 of Small and Large cities Megacities output in the nonfarm sector. In Asia that figure is $1.80.10 medium-size cities Rural areas also benefit from the growth of cities. Nearby cities provide ready markets for agricultural prod- Note: Megacities are cities with populations over 5 million. Large ucts such as vegetables and dairy products and for rural cities are cities with populations between 1 million and 5 million. nonfarm output. Rural industries often supply parts and Medium-size cities are cities with populations between 0.5 million components to nearby urban manufacturers. Urbanization and 1 million. Small cities are cities with populations less than can also help raise rural productivity through technology 0.5 million. transfers, educational services, and training. Source: UNDIESA, World Urbanization Prospects, 1998. ical worker is generally as well off in a small city with The dynamics of city formation low wages and low living costs as a worker in a large The relationship between a country’s industrial organi- urban area where wages and living costs are as much as zation and its system of cities helps explain emerging 100 percent higher.11 patterns of urbanization. During the early stages of in- The biggest metropolitan areas provide a large, di- dustrialization in most developing countries, modern verse economic base for modern service and other in- industries—particularly in sectors that are influenced novative industries that derive important benefits from primarily by the location of consumers—often cluster such an environment. In contrast, small and medium- in one or two large metropolitan areas. The first site for size metropolitan areas tend to specialize in the produc- agglomeration is usually the national capital (Bangkok, tion of goods that are exported outside the city, focus- Bogotá, Jakarta, Mexico City, Seoul, and Suva, Fiji) or ing on a single standardized manufacturing or service a large city near the coast (Calcutta, São Paulo, and area such as primary metals, food processing, textiles, Shanghai). This clustering saves on scarce resources and pulp and paper, machinery, or transportation. By spe- helps industries cope with initial shortages of skilled cializing in one set of activities, smaller metropolitan labor, technical knowledge, business and financial ser- areas exploit localization economies while conserving vices, and modern telecommunications and transporta- on the congestion costs that affect larger cities. Special- tion infrastructure. For foreign investors and industrial ized cities grow with the economies of scale and local exporters, the national capital may be a prime location intermediate input linkages their activities generate, for entering the country and the best place to find mod- and with the size of regional markets and city-specific ern services. Capitals have the added advantage of prox- amenities. imity to government decisionmakers and regulators.12        As industrialization proceeds, manufacturing activi- Box 6.3 ties begin to move to smaller cities outside the capital. The dispersal of industry in Korea This shift occurs because congestion costs increase and because, to some extent, the benefits of agglomeration Urbanization in Korea has meant that the proportion of the population living in Seoul has grown steadily. But this decrease as production standardizes in mature plants. statement does not take into account the decline in The spread of effective telecommunications and trans- Seoul’s primacy in the country’s system of cities and its portation, the devolution of bureaucratic processes to manufacturing structure (see table). Seoul is growing, but local governments, and the opening of capital markets other Korean cities are growing faster. Even more dramatic is the exodus of manufacturing employment from metro- also encourage the movement of industries out of major politan Seoul to surrounding suburban areas. In 1970 three- cities (box 6.3). quarters of provincial manufacturing employment was in In the future, the forces of globalization, including metropolitan Seoul, but by 1993 the percentage had fallen trade liberalization and financial integration, will con- to one-third. Industry began moving out of Korea’s major metropolitan areas—Seoul, Pusan, and Taegu—and their tinue to reinforce the importance of urban agglomera- satellite cities in the mid-1980s. The share of other cities tion economies. Because international firms and in- and rural areas in national manufacturing employment rose vestors seek low-cost, accessible locations for their from 26 to 42 percent between 1983 and 1993. plants, localized production networks will be essential Policy changes were responsible for this trend. In the to a country’s global competitiveness.13 Manufacturing 1970s the government initiated policies designed to en- courage the decentralization of industry from metropolitan is placing increasing emphasis on high effective capital- Seoul. Key elements of these policies included financial in- labor ratios and light, high-tech materials, often in con- centives to relocate, direct relocation orders, and the con- nection with intermediate service inputs such as soft- struction of industrial parks. Despite the natural market ware, programming, and engineering services that can forces that were encouraging firms to leave Seoul (includ- ing high wages and rents), these initial policies had little im- be supplied at a distance. Sydney’s transformation into mediate effect. Strong government regulation and the as- a global city between 1971 and 1991 translated into a sociated red tape made plants unwilling to locate more 25 percent increase in employment creation as well as than a 45-minute drive from the capital. Within that zone, a radical shift toward financial and business services.14 only a few successful industrial parks existed.17 Ultimately, three developments sparked the move out Openness to the world economy will increase the of Seoul, Pusan, and Taegu. First, Korea liberalized its econ- volatility of urban economies and heighten competi- omy in the early 1980s, reducing the red tape tying indus- tion among cities within the same country. Cities that tries to Seoul. Second, the government reinstated local are able to exploit a comparative advantage in global government autonomy in 1988, enabling local authorities to hold elections and assess and collect taxes. Third, the tradables will thrive, but those that have depended on government invested heavily in communications infra- protected industries will struggle. structure and roads outside Seoul and Pusan—and has Technological change has enhanced agglomeration continued to do so. economies in the past and should continue to do so in The primacy of metropolitan Seoul the future. Commuter rail transportation, automobiles, (Seoul as a percentage of national total) and metropolitan highway systems have all contributed to urban growth in industrial economies during the 1960 1970 1980 1990 20th century. In the future, local human capital and National urban population 34 41 38 33 the accumulation of knowledge will also affect city National population 9 17 22 25 size. Estimates for 1940–90 suggest that an increase of Manufacturing .. .. 21 14 one standard deviation in the percentage of college- . . Not available. educated residents in a U.S. city is associated with a Source: Henderson, Lee, and Lee 1998; Henderson 1998. 20 percent increase in size, even after accounting for growth trends and specific city characteristics.15 Recent evidence suggests that telecommunications is a com- plement to, rather than a substitute for, face-to-face in- Most of the world’s urban population will remain in teraction.16 In a world of extraordinary technological small and medium-size cities, since they are growing gains, one of the most effective mechanisms for trans- faster than large urban areas (figure 6.3). But sizes mitting knowledge and conducting business may still are relative. In 1970 a medium-size city was defined as be geographic proximity. one with a population of anywhere from 250,000 to      ⁄  Figure 6.3 consisted of shifting resources from agriculture to fi- Small cities had the fastest growing populations, nance the expansion of “modern” economic sectors— and megacities the slowest, from 1970 to 1990 usually manufacturing—which were concentrated in Average annual population growth cities. Urban workers in the formal sector benefited (percent) from food and housing subsidies and government- 5.0 sponsored unemployment and pension schemes, while rural populations received low prices for their crops and 4.5 Small cities had little access to government support. Such mis- placed efforts are part of the reason Africa has seen 4.0 urbanization with very little economic growth (box 6.4). In other cases governments, alarmed at the growing 3.5 population of ill-housed and underemployed citizens Large cities living on the periphery of cities, have attempted to halt 3.0 Medium-size cities urbanization. In Indonesia squatters were rounded up and trucked back to the countryside. In China, the 2.5 Soviet Union, and Vietnam a system of permits re- stricted rural-urban migration. And in India industrial Megacities 2.0 Box 6.4 1.5 Africa: urbanization without growth 1970–75 1975–80 1980–85 1985–90 Note: Megacities are cities with populations over 5 million. Large Urbanization is typically associated with rising per capita cities are cities with populations between 1 million and 5 million. Medium-size cities are cities with populations between 0.5 million income. This pattern has held true in Europe, Latin Amer- and 1 million. Small cities are cities with populations less than ica, and—more recently—much of Asia. Africa has been 0.5 million. the exception. Source: UNDIESA, World Urbanization Prospects, 1998. Between 1970 and 1995 the average African country’s urban population grew by 4.7 percent annually, while its per capita GDP dropped by 0.7 percent a year. This nega- tive correlation between urbanization and per capita in- 500,000. Today a medium-size city is defined as one come is unique, even among poor countries and econ- with a population closer to a million. The same is true omies with low growth rates. Industrialization did not for large cities. In 1950 the average population of the accompany the boom in urban growth. Only 9 percent of world’s 100 largest cities was 2.1 million, but by 1990 Africa’s labor force is employed in industry, compared with 18 percent in Asia, which has seen comparable rates of it had reached more than 5 million. In 1800 it was only urbanization. Cities in Africa are not serving as engines of about 200,000.18 growth and structural transformation. Instead, they are The number of cities will also continue to grow. In part of the cause and a major symptom of the economic 1900 the United States had 75 metropolitan areas, and social crises that have enveloped the continent.19 which were defined as areas with a population of over Africa’s pattern of “urbanization without growth” is in part the result of distorted incentives that encouraged mi- 50,000. Today the number of metropolitan areas has grants to move to cities to exploit subsidies rather than in reached almost 350. As these urban centers grow, the response to opportunities for more productive employ- number of very large agglomerations will also increase. ment. African cities were the beneficiaries of food pricing In 1970 some 163 metropolitan areas worldwide had and trade policies that favored urban consumers over rural producers. While the structural adjustment programs initi- more than 1 million people. Today there are about 350 ated in the mid-1980s removed many of these distortions, such areas. Having more metropolitan areas in a coun- they have already contributed to excessive levels of rural- try means having more centers of political power that urban migration over prior decades. Worsening physical or feed the forces of localization and raise the stakes for economic security in rural areas may also be pushing the migration to the relative safety of cities. Over the years, good urban governance. wars and civil unrest have led millions in Angola, Liberia, and Mozambique to flee to cities. In Mauritania, Nouak- The national government’s role in urbanization chott’s population doubled during one drought year in the National governments have often tried to influence the mid-1980s. pace or location of urbanization. Often these efforts        firms were essentially prohibited from locating new export licenses and capital markets, affect the economic plants in or near large cities. life of firms. Central government bureaucrats like to keep Policies to stem urban population growth have tight control over the process of allocating licenses or largely failed. Indonesia’s effort to evict migrants did loans. But an overly centralized allocation process causes not succeed and was later abandoned. Substantial in- distortions when firms are deciding where to locate pro- ternal migration occurred in China, the Soviet Union, duction. Producers tend to locate in capital cities and and Vietnam despite controls on population move- other bureaucratic centers in order to be able to deal ef- ments. These efforts did, however, impose significant fectively with red tape.23 In the early 1980s Indonesia costs on both migrants and the economy. An over- liberalized capital and export-import markets, creating whelming body of evidence shows that when the poor new opportunities for small and medium-size firms. But migrate, they are responding efficiently to economic in- the dispensing functions remained highly centralized, centives—notably higher wages—and generally are bet- and the concentration of small and medium-size firms ter off after they move. Attempts to stop migration pre- in larger metropolitan areas increased.24 vent the poor from improving their economic situation The unhappy record of past government efforts to and can impose other costs on migrants. Limits on mi- prevent rural-urban migration or to steer urban growth gration to Dar es Salaam, for example, made the poor to particular locations leads to a straightforward con- more susceptible to extortion by corrupt officials.20 clusion: governments are not skilled at deciding where Governments have also distorted urban growth households and firms should locate. National govern- through their choice of locations for state-owned in- ments can perform a more useful function by working dustries and by creating special economic zones—deci- to provide an environment conducive to economic sions that are often influenced by political rather than growth regardless of location. Macroeconomic policies economic considerations. The state-owned portion of that promote price stability and national institutions the Brazilian iron and steel industry was placed near that enable firms and households to make binding con- politically influential São Paulo and Rio de Janeiro tracts may be the most important factors in creating a rather than near the source of raw materials in the state growth-oriented environment, and national govern- of Minas Gerais (where private iron and steel produc- ments can provide them.25 In matters of location the ers have chosen to locate). Brazil’s choice to put the ideal government policy is to provide a level playing heavily polluting iron and steel industries in the mid- field so that large and small cities and rural areas can dle of the country’s largest concentration of people compete fairly with each other. (Grande São Paulo) not only raised transportation costs Pursuing such a policy involves more than just elim- but had high human costs as well.21 inating subsidies and tax breaks, however. Many gov- Countries that set up special development zones of- ernment decisions have unavoidable spatial implica- fering relaxed tariffs encourage economic activity to set- tions, especially decisions on siting large-scale public tle in one privileged area at the expense of others. For infrastructure investments, military bases, and public example, if trade liberalization is introduced in the enterprises. As urbanization spreads within a country, coastal area of a country first, inland regions may find investments in public infrastructure must follow. In- themselves permanently disadvantaged. Such policies dustrial producers in remote cities and areas outside of foster dual societies, with cosmopolitan cities on the cities require interregional telecommunications, roads, coast and disadvantaged areas in the hinterland. The and electricity if they are to produce competitively, coastal cities that were the early beneficiaries of China’s move products to major markets, and communicate “open door” policy have maintained their advantage, with buyers and sellers. The national government plays even though their special status was abolished long a key role in determining whether and when such in- ago.22 Similarly, if the spread of technology or the lib- vestments take place. One difficulty is that centralized eralization of capital markets is confined to certain state-owned industries or established businesses may re- areas, these areas will have a permanent advantage over sist hinterland infrastructure investment for fear of others in the country. competition. Another complication may be that the Bureaucratic centralization is another, more subtle central government fails to understand the needs of form of the government-induced distortions that can in- hinterland areas. Industries in Korea began decentraliz- fluence the choice of new sites for production. Govern- ing in the late 1980s after the government made mas- ment regulations, especially rules governing import and sive investments in communications and transportation      ⁄  in regions outside urban centers and restored local gov- vate sources. 27 In industrial countries developers are ernment autonomy. frequently required to provide on-site infrastructure. In principle, a centralized government can create a These costs are incorporated into the price of finished level playing field for locational decisions. In practice, housing and are ultimately financed through the mort- however, resisting pressure to concentrate investment gage market. In developing countries poor and low- in the primary city requires institutional mechanisms income households have to finance housing from cur- that give other regions a voice in the allocation process. rent income, adding space and infrastructure as their Central governments are now under pressure to decen- means allow. In both cases capital is mobilized and al- tralize decisionmaking power and resources to sub- located independent of the government. The private national governments, as chapter 5 discusses. In a de- sector can also finance off-site costs of power, water, centralized system the central government’s role with and telecommunications. In fact, private firms are in- respect to urban development no longer involves elimi- creasingly signing contracts to build such infrastructure nating spatial biases in a centrally managed system of and in many instances agree as part of the deal to fi- investment allocation. Instead, the role of central gov- nance the future expansion or upgrading of networks. ernments is to provide the institutional structure for Publicly financed infrastructure will still be needed, decentralization and coordination across all levels of however. In the case of streets, cost recovery is difficult. government. In the case of social infrastructure, it is undesirable. Re- cent estimates for India suggest that urban investments Local policies for urban economic growth will require public funding equal to nearly 2 percent of If cities are to exploit the benefits of agglomeration, GDP—even though the private sector’s share of infra- they must provide an efficient and attractive place to do structure funding is expected to increase from its pres- business. This section focuses on three cross-sectoral ent level of 25 percent to 45 percent by 2006. elements of this strategy: financing for infrastructure Central or local? In most developing countries, central investment, land use policy, and municipal entrepre- governments have traditionally mobilized the resources neurship. Chapter 7 analyzes sector-specific policies for for public infrastructure through domestic taxation and water, sanitation, and housing. borrowing, forced savings schemes, external debt, and donor assistance. These funds have been spent directly Financing capital investment by central government ministries or government-owned Cities need to invest in infrastructure if they are to pro- enterprises. But pressure for decentralization is changing vide the basic services necessary for economic growth. this pattern to allow subnational politicians to make in- Pressure for investment will be particularly heavy dur- vestment decisions. Sound economic arguments exist for ing a country’s urban transition—the years of rapid pushing these infrastructure investment decisions to the urban population growth fueled by rural-urban migra- subnational level. Centrally determined spending can tion. In recent decades a boom in infrastructure spend- produce arbitrary allocations across cities and tends to ing has paralleled urban growth. Absorbing the 2.4 bil- sever the links among investment, operation, and main- lion new urban residents expected over the next 30 years tenance.28 In contrast, municipalities that have control will require further investment in housing, water and over investment decisions can respond to local priorities. sanitation, transportation, power, and telecommunica- High-income countries have apparently found this argu- tions. The need for these new infrastructure investments ment persuasive. The central government’s share of pub- comes on top of the backlog that already plagues the lic investment spending is generally below 50 percent in world’s cities. Providing universal coverage for water and countries with a per capita GDP of more than $5,000. sanitation alone in the cities of developing countries will Growth in GDP per capita is generally associated with a cost nearly 5 percent of those countries’ GDP.26 declining share of central government spending in pub- Public or private? Not all the necessary investment lic investment (figure 6.4). financing need come from government, as several alter- Local governments can finance their new responsi- native sources are available. Housing, which accounts bilities in several ways. Development fees, connection for about 30 percent of gross capital formation in many charges, and local tax revenue can all generate funds poor countries (including the on-site costs of water, that can be used for investment.29 While such resources sanitation, power, and access), is often funded by pri- can make a significant contribution to investment fi-        Figure 6.4 dures governing defaults and bankruptcies. Public dis- As countries develop, central governments’ closure guidelines and market intermediaries (such as share of public investment falls credit-rating agencies and bond insurers) help investors Share of central government expenditures in total process information on the risk of their investments. government expenditures on public investments And local governments have both well-established fi- (percent) nancial track records and the autonomy to respond to 100 ࡗ changing financial circumstances rather than simply 90 defaulting. ࡗ 80 In many developing countries, few of these condi- ࡗ tions exist. Long histories of macroeconomic instabil- ࡗ 70 ity make long-term financial commitments extremely ࡗ 60 ࡗ risky. Information on potential borrowers is unreliable. ࡗ ࡗ The legal framework needed to provide investors with 50 ࡗ recourse in cases of default is underdeveloped and often 40 ࡗ ࡗ untested. Municipal governments in these countries are ࡗ ࡗ ࡗ ࡗ ࡗ viewed—often correctly—as particularly unattractive ࡗ ࡗ 30 ࡗ ࡗ ࡗ borrowers because they lack the autonomy to raise rev- ࡗ 20 ࡗ ࡗ enues or reduce spending, particularly on personnel. ࡗ ࡗ ࡗ 10 ࡗ Moreover, local governments often have no credible po- ࡗ litical commitment to long-term financial obligations. 0 Under these conditions, even if long-term private capi- 0 5,000 10,000 15,000 20,000 25,000 30,000 tal is available, local governments generally can borrow GDP per capita (1987 U.S. dollars) only at a very high rate of interest, if at all. Note: Sample includes all developing and industrial countries for Despite these shortcomings, municipal bond mar- which calculations could be made of the share of central government kets are emerging in many developing countries. In investment in total government investment in the mid-1980s and mid-1990s. Latin America 52 municipalities and provinces accessed Source: IMF, Government Finance Statistics Yearbook, 1998; World capital markets between 1991 and 1998.31 Asia’s local Bank, World Development Indicators, 1998. bond market is estimated at $477 billion. All Czech cities with more than 100,000 people have issued mu- nicipal bonds, enabling the investment share of Czech nancing, particularly in slow-growing cities, they may municipalities to remain at more than 38 percent of not be enough to finance all infrastructure investments their budgets, despite deep cuts in central government at the peak of the urban transition. In this case debt fi- capital transfers. Standard and Poor’s has given Prague nancing may be required and can make financial sense. and Ostrava “A” ratings for foreign currency bonds. Roads, schools, and pipelines have long useful lives, Poland, Russia, South Africa, and Turkey also have mu- and debt spreads out the costs over their lifetimes. But nicipal bond markets. what options do local governments have for borrowing? Emerging municipal bond markets have an indiffer- The experience of industrial countries suggests two: ent track record. Much like the U.S. bond market in municipal bonds and municipal funds. the 19th century, the initial years have been marked by Municipal bonds. In the United States and Canada, defaults. Ankara and Istanbul have both defaulted on subnational governments rely on the bond market. their bond debt, and many Brazilian states have either Bond debt issued by subnational governments in the defaulted or had their debts taken over by the national two countries now totals more than $7.4 trillion.30 government (see the case study on Brazil in chapter 8). Bond financing is possible because both countries have However, governments are taking measures to increase well-developed capital markets, and their history of investor confidence. Poland, for example, is consider- macroeconomic stability has made private investors ing both legislation on a municipal bankruptcy law and willing to make the long-term financial commitments controls on the volume of subnational debt. infrastructure investment requires. Investors are famil- Municipal funds and banks. The other source of long- iar with and have confidence in the laws and proce- term financing in industrial countries is the munici-      ⁄  pal bank or the municipal development fund (MDF). Local governments, for their part, can improve their These have a long and successful history in Western Eu- attractiveness to borrowers by instituting accounting, rope. European MDFs (Crédit Local de France, the auditing, and disclosure practices that are compatible Spanish Banco de Crédito Local, and the British Public with international standards. They can also improve the Works Loans Board) were founded to address the un- quality of their collateral by allowing central govern- willingness of private capital markets to provide long- ments to deduct debt service directly from intergovern- term credit to small municipalities. In their initial years, mental transfers or by using a specific tax or other rev- many such funds were financed by the central govern- enue source to pay debt service. Loan contracts can ment. In effect, central governments used their excel- specify that debt service will receive priority, prohibit lent credit ratings to raise money cheaply in capital new borrowing backed by the same revenue source markets and then lent the proceeds to municipalities until the debt is retired, or both. Actions, however, are through MDFs. More recently, MDFs have sprung up more persuasive than words. The most convincing evi- throughout the developing world. dence a local government can offer potential lenders is Under an MDF the central government bears the ul- a long, unblemished credit history.33 timate risk of municipal default. Some governments have responded to this risk by behaving as diligent in- Land use vestors, insisting on prudent lending standards and Firms and households must be able to make efficient strict repayment schedules. When central governments decisions about where to locate within cities. Freedom do not impose such standards, levels of default are high. of mobility, or the lack of it, profoundly affects urban One mechanism for encouraging governments to act as economic growth. Agglomeration economies, by defin- prudent investors is to dilute their exposure with some ition, require proximity—firms to firms, households to private participation. Under Colombia’s FINDETER places of employment. The ability of firms and house- program, private banks originate all municipal loans holds to sort themselves into efficient location patterns and bear the full risk of default. The government func- requires an active real estate market in which land tions as a second-tier bank, providing liquidity without prices reflect the different economic values of various assuming risk. As a result the government is exposed sites (box 6.5). only to the risk that the originating bank itself will fail. Governments regulate the operation of land markets The Czech Republic operates a program along similar in several ways. The most extreme approach is to ban lines.32 And many of the European MDFs have shifted the real estate market entirely and make location deci- to market sources to finance their operations and are sions by fiat. Cities in the former Soviet Union and in now in the process of privatizing. Eastern Europe were laid out in this manner. In mar- Conditions in individual countries determine whether ket economies, zoning is the most common mechanism the bond or the bank approach makes more sense. Both for controlling land use. Zoning typically assigns vari- can operate simultaneously, as they do in the United ous uses—residential, retail, commercial, industrial, Kingdom. The challenge is not to choose between them, and mixed—to land in different parts of the city. It may but rather to establish an environment that gives local also dictate the intensity of use by imposing maximum governments the opportunity and incentive to become or minimum limits on lot sizes, floor space, or floor- worthy borrowers. Such an environment emphasizes a area ratios. Zoning is intended to coordinate private stable macroeconomy, a legal framework that defines the configurations of land use with the public portion of rights and remedies of lenders and borrowers, and the the market, where the roads and ports are. It is also creation of a supply of creditworthy borrowers. Central intended to minimize externalities across uses by, for governments especially need to concentrate on the legal example, isolating landfills from residential areas. framework affecting municipal borrowing, including Even zoning can be taken too far, however. If manu- bankruptcy procedures for municipalities. They need facturing is isolated from residential areas, commuting to take measures to forestall pressure for government becomes difficult and expensive for industrial workers. bailouts (see chapter 5). Finally, they need to do their part Excessively high standards for residential development to enhance municipal creditworthiness by stabilizing in- drive up housing costs and force low-income house- tergovernmental transfers and scaling back unfunded holds to locate far from job centers. Zoning can also be mandates and regulations that limit local governments’ too static. Cities change, but redrafting land use plans flexibility in making spending decisions. can be a slow process. In the mid-1970s Malaysia        Box 6.5 City development and land markets Most cities of the world have a common spatial pattern of eco- sence of geographic obstacles) less intense as the distance nomic activity. Most of the activity is densely packed near the from the city center grows. city center and declines with distance. Commercial activity ag- In Paris residential population density declines steadily with glomerates at the city center in skyscrapers because of scale distance from the city center. Land prices follow the same economies (from information exchanges and spillovers) and pattern. However, Moscow appears to violate the common low transaction and transportation costs. Public transportation pattern: its density gradient is upward, not downward. But systems and utilities also operate more efficiently in high- Moscow’s densities were determined not by market forces density areas. Some households, especially those without chil- but by planned allocations that did not recognize either the dren, cluster near the city center in high-rise apartments to min- benefits of central locations or the demand for them. imize commuting time to work and downtown entertainment. Market pricing is likely to change the pattern of land use in Land prices reflect these density patterns, decreasing as the socialist cities. The price gradient for land in Moscow, which distance from the city center increases. High land prices near was relatively flat in the first quarter of 1992, had already the city center mirror the many advantages of living there and begun to steepen two quarters later. Krakow, having opened the corresponding demand for office, housing, and retail space. land to market pricing somewhat earlier, has a considerably Low land prices further out reflect the comparative disadvan- steeper land price gradient. As market forces take hold, both tages of reduced benefits from economies of scale and the cities are likely to take on the steeply sloped density gradients long commuting times. Market forces thus tend to push cities of efficient Western cities, where economic activity clusters toward an efficient pattern of land use, one that is (in the ab- at the core areas. In Paris population densities fall as the distance from the city center increases; in Land prices in Moscow and Krakow are Moscow, densities increase beginning to look like those in the West Population density Relative price of land (people per hectare) 100 300 Moscow 90 Paris Moscow 250 80 Q1 1992 70 200 60 50 Moscow 150 Q4 1992 40 100 30 20 Krakow 50 1992 10 Paris 1995 0 0 0 5 10 15 20 25 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Distance from city center (kilometers) Distance from city center (kilometers) Source: Bertaud and Renaud 1997. Source: Bertaud and Renaud 1997. adopted the Town and Country Planning Act of Britain Governments also influence the location of eco- and Wales, imposing a rigid planning system developed nomic activity through their control over public land for a slow-growing country on a fast-growing economy. and transportation systems. Up to half of urban land is The impact was immediate. The supply of housing in in the public domain, including roads, highways, side- Kuala Lumpur became inelastic, and housing prices walks, parks, and public buildings and facilities. The climbed at two to three times the rate of economic way the government chooses to use the public portion growth, reaching five to six times the average annual in- of urban land determines the spatial configuration of a come. In Bangkok, where zoning regulations are more city: where industry locates, how congested the city is, liberal, housing prices are only two to three times the how dense neighborhoods are, and how the city will de- average annual income.34 velop. Cities expand through progressive additions of      ⁄  transportation corridors and ring roads that allow eco- creased competition within regions is causing cities to nomic activity to spread out in more or less concentric enter a race they cannot win, in which urban govern- circles. Failure to expand transportation facilities delays ments offering lavish and costly incentives to “footloose” the movement of people and industry from city centers investors force other local governments to follow suit. to suburbs, resulting in exceedingly dense core cities Such corporate welfare is estimated to cost several billion with poor living conditions and noncompetitive land dollars annually in the United States, where examples and wage costs. When Jakarta finally built toll roads abound of states and cities providing massive subsidies into the immediate surrounding countryside in the late that seldom lead to new jobs. In Philadelphia, Pennsyl- 1980s, population density in the city center fell from vania, the city and state have provided $426 million in 42,000 people per square kilometer in 1980 to 30,000 subsidies to Europe’s largest shipbuilder just to retain ex- in 1990. Meanwhile, the suburbs around Jakarta, where isting jobs. In Ohio one city government spent $156,000 wage costs were 25 percent lower than in the city cen- for each of the 180 jobs a General Motors plant created. ter, increased their share of the metropolitan area’s for- One intellectual justification for such subsidies is the mal manufacturing employment from 44 percent in infant industry or scale economy argument, which sup- 1985 to 65 percent in 1993. ports subsidizing a line of industrial activity until it Governments influence the efficiency of land use in achieves sufficient local scale to be viable. But if all cities a third way: through their role as a repository of claims in a region adopt this strategy and begin offering exces- to land ownership. Well-functioning land markets re- sive subsidies, they may well wind up with the same in- quire clear title arrangements and a well-kept land reg- dustrial base they would have had without the subsi- istry, so that ownership rights are clearly established and dies. An obvious policy solution is a national agreement all transactions are recorded. The lack of such arrange- to harmonize or cap subsidies. Although such agree- ments hinders private (re)development by jeopardizing ments are rare, they may become more common, given the gains developers and individuals expect when they the recent bad press on local subsidies in the United improve land. When a city has an informal sector where States and related debates in the European Union. land use rights are insecure, redevelopment becomes Even without regional agreements to limit industrial even more difficult. Finally, urban planners need up-to- incentives, international trade agreements are limiting date information on land use and transactions in order the scope for such incentives.36 The agreement on sub- to design and implement effective land use plans. sidies and countervailing measures adopted as part of the General Agreement on Tariffs and Trade (GATT) Municipal entrepreneurship in 1993 prohibits any domestic subsidy that could dis- In 1996 senior officials of the Indian state of Gujarat place imports in domestic markets or other countries’ went to the World Economic Forum and wooed the exports in international markets. Subsidies are defined chief executive officer of General Motors, convincing according to the benefits they confer and the geographic him that Gujarat was a suitable location for a plant. At area or industry they target. This agreement may keep the beginning of the 1980s two U.S. states had trade of- local governments from offering subsidies to specific in- fices abroad; by the end of the 1980s, 40 did. Today more dustries within their jurisdiction or using tax breaks to U.S. states have trade offices in Tokyo than in Washing- attract particular firms. By connecting local economies ton.35 In a world characterized by increased globalization more fully to the global economy, globalization may and urbanization, subnational governments are market- expand the ability of trade agreements to limit such ing their jurisdictions abroad, aiming to catalyze oppor- local industrial subsidies. Recent cases such as the one tunities for innovation and cooperation. Can city gov- brought against Nova Scotia, Canada, for incentives it ernments become strategic brokers that influence their offered to a tire plant show how the GATT agreement city’s—and even their country’s—position in the global has made state and local governments vulnerable to re- urban hierarchy? With appropriate planning and sup- taliatory actions initiated by foreign countries. port, the answer seems to be yes (box 6.6). In the debates over subsidizing industry, both politi- Some argue that cities need this kind of municipal en- cians and the public too often forget that the inputs most trepreneurship in order to seize the new opportunities relevant to economic development are often beyond the offered by globalization and localization and to cope control of local governments—labor costs and skills, nat- with the attendant challenges. But others fear that in- ural resources, climate, and energy prices.37 Business sur-        Box 6.6 Regionalism and local economic development: lessons from Europe The 1980s saw the demise—at least in Europe—of top-down tance to focus on the most vulnerable groups, increasing the industrial policies and their spatial correlate, regional economic resources available for economic development. To comple- development policies. By the early 1990s not a single national ment the centers, the government fostered partnerships in the industrial policy initiative could be identified in Europe, and na- same areas (and in rural areas) with a mandate to enhance the tionally determined regional policies were scarce. competitiveness of local firms by making residents more em- Two factors explain the demise of centrally issued regional ployable. Finally, with the support of the European Union, the policies. First, they had a record of picking industrial lame government created county enterprise boards that allocated ducks. Second, regional governments resented national poli- project grants locally using criteria set at the national level. cies aimed at their economies, complaining that local author- Despite some weaknesses, the Irish partnerships are gen- ities were rarely consulted. The result has been a drop in erally considered successful. The keys to their success are: spending on local development initiatives but greater regional input on how such funding is used. Ⅲ Their ability to draw directly on local resources, so that the The increased involvement of regions in development ini- experience of local businesspeople provides the foundation tiatives did result in some bidding wars to attract firms, but it for enterprise creation and the unemployed themselves set also led to strategic improvements. Ireland is a good example up programs targeting the jobless of these changes. The Irish program emerged from the na- Ⅲ Their ability to adapt the objectives and resources of state tional economic crisis of the mid-1980s, which was character- agencies to local needs ized by severe long-term unemployment and attendant social Ⅲ Their capacity for improving the targeting of social welfare— ills. The central government’s efforts to deal with the crisis and thus the cost-effectiveness of providing it. were clearly not working, and budgetary pressure was forcing a reconsideration of social policies. Part of the reason for the success of Ireland’s local partner- Out of this dilemma came the new Irish “social partner- ships is that they developed in a period of economic expan- ship,” which created decentralized centers for the unemployed sion. But their successes are proof that practical, positive area- managed by boards composed of representatives of local gov- based programming and public-private partnerships can work. ernments, training agencies, and the office of the prime minis- ter. The centers serve as vehicles for retargeting social assis- Source: Cooke and Morgan 1998; Sabel 1998. veys suggest that entrepreneurs care about operating firms and workers’ unions to identify needs and create costs and conditions most, followed by quality of life. potential partnerships. Arrangements among local gov- Transportation costs and wages are generally cited as the ernments, employers, and unions aimed at providing most important, followed by utility and occupancy costs. vocational training facilitate these efforts. In Penang, Among the public services that matter are transportation Malaysia, the Penang Skills Development Center brings and safety. Taxes matter only at the margin in choosing together representatives of industry, state and local gov- among similar locations.38 A municipality’s economic ernment, and academia to bridge the gap between for- development efforts should focus on efficiently provid- mal education and the job skills the area’s top investors ing the services it is responsible for and easing red tape require. Similarly, the Skill Development Councils of and excessive regulation. Karachi and Lahore (Pakistan), composed of provincial A possible role for municipal activism does remain, and federal government representatives, employers, and however. The efforts of local governments to promote workers’ representatives, are successful forums that serve industrial development can be successful and cost- as links between industry and training providers.39 effective if they focus on broad policies designed to What institutional arrangements are most likely form a critical mass for specific industries and not on to produce successful local development policies? Lead- firm-specific benefits. Sectorwide strategies are more ership is important, but it can emerge from many likely to create a competitive advantage because they sources, either private or public.40 A forum is needed “cluster” activities that can lead to agglomeration econ- within which the private and public sectors can com- omies. For example, local governments can develop municate with each other and define a common goal training initiatives adapted to local economic condi- or vision for a city. Such a forum requires the support tions and comparative advantages. France and Italy are of a common base of information (box 6.7). Different decentralizing vocational training on the theory that cities have different forums and institutional arrange- local governments are best suited to working with local ments that range from formal chambers of commerce      ⁄  Box 6.7 Know thy economy: the importance of local economic information A city can judge the appropriateness of regulations only if it has paucity of data, and each had spent precious time gathering in- reliable information on its economy and spatial organization. formation that was often outdated and not always consistent This lesson holds true whether the issue is deciding which or comparable. Unsurprisingly they found it difficult to place growth-hampering regulations to eliminate or which growth- their work in context and to establish cross-sectoral priorities.42 friendly regulations to implement during the urban transition. In most countries the needed information is available For example, the spatial organization implicit in a zoning plan is through completed censuses and surveys, and the amount of often hidden because zoning is usually the result of parcel-by- work required to compile the information is manageable and parcel negotiations. Few cities have an overall schematic zon- affordable. A modest investment of time and money supported ing map. When Krakow conducted an overall review of its city the collection of information for estimating and analyzing re- zoning plan, it found that while the stated objective was to pro- gional accounts in several West African regional capitals. Data mote a compact city with few suburbs, the plan’s constraints came from the national census office, trade bureaus, and ele- on land use and its tendency to reinforce existing land use pat- mentary surveys.43 New technology has made it easier and terns were actually blocking this goal. 41 cheaper to process data and understand its spatial implications. Regional analysis can help identify infrastructure investments If the information exists, why is it so difficult to access? that will improve integration between cities and nonurban areas, Most cities have local planning offices or economic bureaus increase access to national and global markets, and contribute to whose role is to collect and process statistical information regional prosperity. A regional economic analysis pointed out that about the city. But the census and survey data routinely col- for more than 30 years investment in Senegal’s river delta had lected at the national level are typically not available to local of- focused on rice farming—apparently because many believed fices, at least not in a readily usable form. In other cases local that rice farming was the source of the region’s growth. Yet rice offices collect basic demographic and production statistics. But farming has never generated more than 4 percent of the region’s these data are transmitted directly to the national capital and gross local product despite absorbing three times that amount in are not analyzed locally, either because local economic officers foreign aid in the 1990s. Meanwhile, the regional capital stag- do not have the skills or resources or because the city’s deci- nated (along with the region) because its port and local trans- sionmakers do not demand the information. portation infrastructure were never properly developed. The key is to establish a structure to ensure that local de- A common base of facts promotes a constructive debate velopment strategies and investment plans are based on good on municipal development and facilitates consensus around a information. Regions can contract out the tasks of analyzing local development strategy. Without a common factual base, and compiling data or develop partnerships with groups that the debate can be frustrating and inconclusive. At a conference can help collect the necessary information, such as local uni- organized by Durban, South Africa, to design an economic de- versities, national statistical institutes, chambers of com- velopment strategy, all the speakers had been hampered by the merce, and trade institutes. and municipal commissions to informal ad hoc com- in place, urban economies are likely to grow. National missions organized around a single vision or project. governments will find it best not to attempt to stop or Whatever its structure, the forum needs to have the direct internal migration, since such efforts inevitably powers and means necessary to collect and process reli- fail. Local governments can facilitate urban economic able information on the local economy. growth in their areas by investing in trunk infrastruc- ture and fostering an open land market. But the key • • • role of local government in economic development is In economic terms, what is good for a country is good to provide the basic infrastructure and public services for its cities. If the political, legal, and macroeconomic needed to create an attractive environment for both conditions for nationwide economic development are businesses and households. C h a 7 p t e r Making Cities Livable A s long as it is environmentally and so- private sector and by community initia- cially sustainable, economic growth will tives—a remarkable response that has in time lead to better living conditions. provided affordable housing and ser- But cities need not wait for the slow vices to millions of urban households. compounding of aggregate growth rates But unregulated and isolated com- to improve livability. With the appropri- munity initiatives cannot serve as the ate polices and institutions, many coun- building blocks for sustained citywide tries with low per capita incomes can improvements. Cities need to turn away enjoy considerably better service levels from an unsuccessful model that leaves (figures 7.1 and 7.2). In developing the most dynamic providers of essential countries with a relatively high per services outside the planning and im- capita GDP, the percentage of urban plementation framework to one that households with access to water and san- associates them in productive partner- itation services (two important indices ships. This step clearly requires chang- of the quality of urban life) is already rel- ing the rules so that partnerships are fa- atively high. Among countries with low cilitated and services provided in ways income levels, access to these services— that are guided by and respond to pub- as well to affordable housing—varies lic demand. widely. Clearly, some low-income coun- This chapter seeks to describe how tries have provided much better access an appropriate blend of policies and in- to essential services than others. stitutions can improve urban living con- Since the 1950s the dominant model ditions at various levels of economic for providing basic infrastructure and development. The chapter begins by re- services in developing countries has viewing the principal issues of urban liv- assigned primary responsibility to the ability, then briefly examines the history public sector. But this arrangement has of service provision (both public and left much to be desired in most low- private). The aim of this review is to income developing countries.1 The ser- identify the factors responsible for the vice gaps left by the public sector have public sector’s poor performance in pro- been filled largely by the unregulated viding essential services in developing       ⁄  Figure 7.1 countries. The discussion then turns to the roles of the Even low-income countries can achieve high public and private sectors and community initiatives in levels of basic water and sanitation services service provision. This analysis draws on recent experi- ence in a number of areas: housing, water supply, sani- Percentage of urban households with access to safe water 100 tation, transportation, and social protection. The chap- ter does not try to provide technical solutions for 90 sectoral problems. Instead it shows how an institutional 80 framework built on partnerships, inclusiveness, and in- formation sharing and responsive to demand holds gen- 70 uine promise for improving urban living conditions. 60 In meeting the urban challenges of the 21st century, 50 the most effective institutions and policy initiatives will exploit the opportunities globalization and localization 40 present. Globalization can provide the impetus for eco- 30 nomic growth, while successful localization can em- power communities to act as agents of change and give 20 rise to mechanisms that promote transparency and ac- 10 countability in public sector decisionmaking. For de- veloping countries ready to exploit them, these oppor- 0 0 1,000 2,000 3,000 4,000 5,000 tunities can have a lasting impact on the daily lives of GDP per capita (1995 U.S. dollars) millions of urban households. The unfinished urban agenda Cities provide their residents with chances for upward Percentage of urban households with sanitation mobility that are often absent in rural areas, and for 100 that reason urban areas act as magnets for rural mi- 90 grants.2 But living conditions for many of the most re- cent arrivals (as well as for other disenfranchised social 80 groups) have remained below acceptable thresholds, 70 even though urban living conditions have improved since World War II. Thus the urban agenda for im- 60 proved livability begins with reducing poverty and in- 50 equality. But it also includes creating a healthful urban 40 environment, minimizing crime and violence, estab- lishing a civil protection system, and making services 30 more accessible.3 20 Cities have often been overwhelmed by population growth, leaving them unable to provide sufficient basic 10 services. In 1994 at least 220 million urban dwellers 0 (13 percent of the developing world’s urban popula- 0 1,000 2,000 3,000 4,000 5,000 tion) lacked access to clean drinking water, and almost GDP per capita (1995 U.S. dollars) twice as many had no access even to the simplest la- trines. Roughly half of all solid waste went uncollected, Note: Safe water includes public taps within 200 meters that piling up on streets and in drains and contributing to offer adequate supplies for daily needs. Urban sanitation is the flooding and the spread of disease. Domestic and in- percentage of urban households with a connection to public sewers or with a household system such as a pit privy, septic tank, or dustrial effluents were being released into waterways communal toilet. with little or no treatment, often affecting the quality Source: World Bank, World Development Indicators, 1999. of water far beyond the city. The La Paz River flowing through Bolivia’s capital is still so polluted that down-     Figure 7.2 by more than 20 percent.6 When these facilities break Housing affordability varies significantly at low down or do not keep up with a city’s expanding popu- levels of income lation, the health hazards increase for a range of water- Rent-to-income ratio borne diseases and diseases spread by water-related 0.7 vectors (malaria and dengue fever being the most threat- Lahore, Pakistan ening). At any given time, close to half the urban popu- 0.6 lation in developing countries is suffering from one or more of these diseases.7 Airborne illnesses such as acute 0.5 respiratory infections and tuberculosis also spread faster in overcrowded urban residential quarters with inade- 0.4 quate ventilation (see box 7). Air pollution, which is closely associated with ur- 0.3 banization and industrialization in developing coun- tries, seriously impinges on the health of children and 0.2 adults alike. Pollution particularly affects those already suffering from malnutrition and infectious disease, 0.1 which lower their ability to resist chemical pollutants. For most children in the large cities of developing 0.0 countries, breathing the air may be as harmful as smok- 0 1,000 2,000 3,000 4,000 5,000 6,000 ing two packs of cigarettes a day. City income per capita (1993 U.S. dollars) Ⅲ In Delhi the incidence of bronchial asthma in the House price-to-income ratio 5–16 age group is 10–12 percent, and air pollution 40 is one of the major causes.8 Yerevan, Armenia Ⅲ A 1990 study of atmospheric lead pollution in 35 Bangkok estimated that 30,000 to 70,000 children 30 Podgorica, Yugoslavia, risked losing 4 or more IQ points because of high Fed. Rep. (Serbia/Montenegro) lead levels, and many more risked smaller reductions 25 in intelligence.9 Ⅲ China has 9 of the 10 cities with the highest counts 20 of total suspended particulates (TSPs). Industrial 15 and industrializing cities such as Jiaozou, Lanzhou, Taiyuan, Urumqi, Wanxian, and Yichang all have 10 mean annual concentrations of TSPs exceeding 500 5 micrograms per cubic meter. The World Health Or- ganization (WHO) puts acceptable levels at less than 0 100 micrograms per cubic meter.10 0 1,000 2,000 3,000 4,000 5,000 6,000 City income per capita (1993 U.S. dollars) Problems of inadequate infrastructure have eco- Source: UNCHS 1995. nomic as well as human costs. In Jakarta a poor resident typically pays 10 times more than a rich resident does for a liter of clean water and suffers 2 to 4 times more stream horticultural production has been curtailed.4 gastroenteritis, typhoid, and malaria.11 As traffic con- And the Pasig River that created the lush vegetation of tinues to clog the streets of most large cities in develop- Manila is now biologically dead.5 ing countries, the costs of traffic congestion grow. Esti- The lack of basic services continues to exact a high mated losses from traffic jams in Bangkok range from toll on human health. Epidemiological studies show $272 million to $1 billion a year, depending on how that improving access to water, drainage, and sanitation the value of time lost in traffic jams is computed.12 In facilities can reduce the incidence of diarrheal disease Seoul time losses from traffic congestion are estimated      ⁄  at $154 million.13 If China maintains its business-as- low-income countries. One argument holds that gov- usual response to air pollution, the health costs of urban ernments should withdraw as primary service providers residents’ exposure to TSPs will rise from $32 billion in and assume the role of enabler, relying increasingly on 1995 to nearly $98 billion in 2020.14 the private sector to deliver basic services.20 But the The poor suffer most from these problems. The locus public sector has successfully provided such services in of poverty is shifting to urban areas, yet cities can go industrial countries since the late 1800s. Why have only so far in addressing issues of income redistribution, publicly provided essential services been satisfactory in which often require central government action. On av- the one case and not in the other? erage, health indicators show that people are better off in cities than in rural areas, but the statistics mask in- Urban reform equalities within the urban population. Recent evidence Around 1850 European cities faced many of the same suggests that health conditions for the poor in many de- problems cities in developing countries face today. veloping cities are worse than in rural areas. In Bangla- Rural migrants were arriving in urban areas daily, in- desh, for example, reported infant mortality rates in creasing populations so precipitously that the supply of urban slums exceed rural rates (table 7.1).15 More than basic services could not keep up with demand. Urban 1.1 billion people—poor and rich alike—live in cities mortality rates were often far higher than those in the with levels of air pollution in excess of WHO standards. surrounding rural areas, in part because of epidemics But poor urban dwellers are likely to be exposed to ad- of diseases such as cholera. Public officials investigating ditional indoor air pollution from inadequate, badly the frequent epidemics associated the problems with ventilated cooking facilities and to further outdoor pol- the lack of decent sanitary conditions in the parts of the lution from industrial sites. The poorer areas of cities city where the new arrivals settled. A revolution in pub- are often adjacent to such sites, either because no one lic sanitation ensued, with cities investing heavily in else will live there or because the poor have no voice in housing and in water, sewerage, and drainage facilities. deciding where industries are located. 16 North American cities shared the experience of their Urban dwellers in poor districts of metropolitan European counterparts.21 areas suffer disproportionately because of crime and vi- These reforms succeeded for one important reason. olence, which increase alongside poverty and inequal- Wealthy residents of cities could not escape the effects ity.17 According to WHO, the global cost of injuries of unhealthy living conditions. Thus, although the risks from violence is almost $500 billion a year in medical were far worse in poor areas where structures such as ten- care and lost productivity.18 Estimates of the social costs ements abounded, wealthier urban residents could not of crime and violence range from about 2 percent of ignore the threat to their own well-being. 22 Their sup- GDP in Asia to 7.5 percent of GDP in Latin America.19 port, often in the form of influential political coalitions, affected the allocation of resources at both the national Learning from the past and subnational levels and helped direct public funds to Since the 1950s the common model of urban manage- urban areas in need of appropriate sanitary facilities. ment in developing countries has charged the public By the time rapid urbanization began to affect de- sector with planning and delivering basic services. But veloping countries, however, technological advances this model has failed to yield satisfactory outcomes in had altered the situation and weakened the impetus for public action, much to the disadvantage of the urban Table 7.1 poor. Advances in medicine, in particular, were making Infant mortality rate, Bangladesh, 1990 it possible for individuals to protect themselves against (per 1,000 live births) disease. Portable electricity generators and pumps had been developed that gave individual households access Urban slums National Rural Urban (1991) to light and water. More recently, filters and bottled Total 94 97 71 134 water have become available, mitigating (for those who Male 98 101 73 123 can afford them) the shortcomings of the public sys- Female 91 93 68 146 tem. Vacuum trucks and septic tanks permit house- Source: Harpham and Tanner 1995. holds to develop their own solutions to sanitation prob- lems. Air-conditioned residences, automobiles, and     offices block out the worst effects of air pollution. was not yet available. Given the situation, private “niche” Urban enclaves or suburbs and private security arrange- providers with an intimate knowledge of neighbor- ments partially insulate the wealthy from crime and hoods and customers were better able to match supply violence. And with time, the medical community has and demand. By the early 19th century, private water learned how to prevent the diseases of poverty from en- companies had been serving London for over 200 years. gulfing entire urban populations. The ability to provide Eight companies were operating in the city at the end for and protect oneself and one’s immediate family has of the century.26 become a given in modern urban life, undermining the Over time, however, people became dissatisfied with impetus to lobby for changes that will benefit society as private providers.27 Complaints centered on the lack of a whole. Individual action produces faster and more re- services in outlying areas, high prices, poor quality, and liable results and is more readily available to members political corruption. The introduction of flush toilets of politically influential groups—precisely those groups increased the amount of wastewater, polluting the local that once lobbied for action on a grander scale.23 water supply, and private companies proved reluctant As a result of these changes, cities around the world to invest in more distant water sources. As fire-fighting have been divided into those who can afford to supply technology changed, requiring more water at greater their own needs and those who cannot. Municipal gov- pressures, disagreements arose about how to supply ernments and public agencies often cater to one part of water to fight fires and who should pay for it.28 Courts a city and, at best, adopt a posture of benign neglect to- of law found it difficult to cope with the complex reg- ward the other, making the division even deeper. This ulatory problems that cropped up in these disputes.29 interpretation of urban history is supported by several At the same time, rising incomes led to much greater recent episodes in which concerted public action has oc- homogeneity in the demand for services such as gas, curred only when negative externalities spilled beyond water, sanitation, and electricity, eroding one advantage poor neighborhoods. Major initiatives in Calcutta were of having small niche providers. These providers also spurred by cholera outbreaks in the 1950s and 1960s, could not exploit the scale economies of networked ser- and more recent reforms in Surat and Ahmadabad, vices offered by regionally managed water resources, India, date from an outbreak of the plague in 1994. The reservoirs, and centralized facilities for treating waste- economic impact of the plague spread beyond the cities water. All these considerations led to a major shift in to threaten India’s national tourism industry. Those the way essential services were provided in the 20th same public sector agencies that were responsible for ne- century. Public or semiregulated, autonomous entities glecting their municipalities quickly began to focus on assumed responsibility for delivering basic services in solid waste collection and disposal. Their actions trans- industrial countries such as the United Kingdom and, formed Surat into India’s second-cleanest city.24 Such to a lesser extent, the United States. examples support the conclusion that the absence of in- Private provision is now making a significant come- fluential political lobbies for urban reform in develop- back in industrial countries. The United Kingdom un- ing countries is at least partly responsible for the lack of dertook major reforms in the 1980s, and a profound progress in providing decent services.25 change appears to be under way in Europe as the pri- vate and public sectors develop partnerships to fund Providing essential services privately and operate infrastructure projects.30 These partner- In the late 19th and early 20th centuries in England ships are in part the result of public expenditure con- and the United States, gas, water, canals, trolleys, high- straints imposed during the process leading up to the ways, and electricity were mostly provided privately. birth of the euro, the single European currency. But Eu- By 1890 private companies owned 57 percent of the rope’s shift to private infrastructure also reflects ad- waterworks in the United States. Municipalities often vances in regulatory capabilities, which were seriously arranged long-term contracts with these firms, primar- limited in the late 19th century. ily for financial reasons: cities lacked capital, and na- France’s experience illustrates the importance and tional subsidies were quite limited. At this early stage difficulty of regulating providers of basic services. of urban development, demand patterns varied widely France has a long history of private provision of public (especially among low-income homeowners, tenants, services. Its decentralized public-private system of mu- and home-based producers), and metering technology nicipal concessions developed during the 20th century      ⁄  has proved very successful. But the French experience sewerage, electricity, gas, and telecommunications. In also shows that such a system is not always easy to im- these cases its franchise is exclusive, and private provi- plement—and that it requires strong monitoring mech- sion is illegal. In other areas, such as housing, the pub- anisms. In the mid-1990s municipal water concessions lic sector establishes standards and regulations. were hit with allegations of corruption.31 Disputes When this broad mandate is executed well, the com- arose between municipalities and water concessionaires, bination of exclusive control and centralized manage- in part because of the uncertainty introduced by re- ment can theoretically yield economies of scale for peated legislative changes in the early 1990s and in part networked services. However, when it is not properly because of the number of unfavorable contracts in- carried out, it can generate severe problems. When the experienced municipalities had negotiated. As a result, public sector falls short, private companies and individ- private-public partnerships fell out of favor with elected uals begin offering water, transportation, accommoda- officials. The situation is changing, with two associa- tion, and other services on an ad hoc basis, outside the tions of local governments joining forces to create a reach of formal rules—a situation that creates many consulting agency, Service Public 2000, that will help dilemmas and inefficiencies. municipalities negotiate contracts and design regula- For many services, such as housing and water sup- tions. Several laws have also been passed since 1995 ply, the private sector is more than ready to respond to that require greater transparency and public disclosure demand, since providing these services can be prof- from concessionaires. These developments have sub- itable. But in many developing countries private firms stantially improved the situation and restored confi- cannot offer affordable housing without violating the dence in water concessions.32 building codes. More often than not, these codes are The history of urban services management in Buenos based on sophisticated engineering standards that are Aires is in some ways similar to France’s experience.33 inappropriate in a low-income country. Furthermore, In the late 19th century private companies operating in the private sector is unwilling to make long-term invest- a competitive market provided most infrastructure and ments when it is operating outside the law and is at the essential services, which compared well with what Eu- mercy of the public authorities. This scenario causes se- ropean cities enjoyed. Over time, however, politicians rious problems. Pushing basic services into an informal began to interfere in the regulatory process, causing ser- area of shadowy legality prevents investments large vice to suffer and, in the mid-20th century, providing enough to benefit from economies of scale. It also gives a justification for introducing centralized public man- rise to an underground economy in which the acquisi- agement. But the public sector was not up to the task. tion of state land, its subdivision, development, and Increasingly the demands of local users and the priori- settlement, and the provision of public services are all ties of the federally controlled utilities came into con- opaque and somewhat mysterious. flict, and once again the quality of service declined. At One of India’s best-known corruption fighters, K. J. the same time, the number of residents with no access Alphons, described the agency he worked for, the Delhi to services increased. Around 1990 the government Development Authority (DDA), as “the most corrupt began to replace public sector monopolies with private institution in the country.” Those who corrupt it, he monopoly providers. It is too early to evaluate the re- added, help illegal builders grab DDA land and then sults of this latest phase, but in order for private provi- build houses and shops that are sold to unwitting buy- sion to succeed, it will have to be effectively regulated. ers. Unauthorized buildings range from shanties for the Regulation is a particularly important issue in low- poor to shopping centers for the middle class to man- income developing countries, where regulatory mecha- sions for the rich, all established on government land nisms are still weak. under false pretenses, with political complicity. More- over, Alphons reported, nothing gets built, legal or ille- Service provision in developing countries gal, without a bribe.34 Many developing cities are ser- The public sector in developing countries has enjoyed viced in this fashion, with essential services available a broad mandate when it comes to urban areas. In only at a very heavy social cost. Karachi, Pakistan needs many cities the public sector owns most of the land. It an estimated 80,000 housing units each year, but be- is often the monopoly provider of many services, espe- tween 1987 and 1992 the authorities issued an average cially those based on physical networks: water supply, of only 26,700 building permits annually. The gap, of     course, is being filled in much the same way as it is in the pace of this revolution has been uneven. Commu- Delhi.35 Without reforms, the urban future of develop- nities are often unable to agree on a course of action be- ing countries will probably continue along these lines, cause of ethnic fragmentation or other divisions. Even with overcrowded squatter settlements, illegal subdivi- in India—which has been a democracy for more than sions, deteriorating environmental conditions, and half a century, has undergone constitutional decentral- costly service provision.36 ization, and has strong NGOs—progress has been hin- When confronted with a public provider that is un- dered by the lack of sufficient political pressure from responsive to demand but holds a franchise shutting out below and the absence of support from above.38 In private providers, households and businesses often re- addition, local governments often lack the technical sort to providing basic services like water and electric- and institutional capacity to form partnerships with ity themselves. This “self-provision” is a very inefficient community-based organizations. form of privatization. Typically, the small producer or This embryonic approach to urban management re- consumer cannot fully utilize the equipment that has quires strategic partnerships and reformed institutions been installed, cannot take advantage of economies of that are approved by both the public and private sectors. scale, and is unable to sell any surplus capacity in a mar- These partnerships also need to address citizens’ rights, ket that is, in any case, prevented from forming. Where security, participation, transparency, and accountability. technological advances have broken the link to physical Fully utilizing them may require redesigning national networks, as in telecommunications, private providers constitutions, as it did in Brazil and South Africa.39 De- have been able to establish markets that greatly benefit spite these issues, and even without wide-ranging re- consumers. But physical networks remain necessary in form, a growing number of examples are proving the ef- areas like water, sewerage, and electricity. fectiveness of the approach. In Karachi partnerships are In other situations when the private sector does not providing sanitation services for informal settlements. respond to demand for essential services, communities In Cali, Colombia, they are being used to combat crime have often organized themselves as providers. Such and violence. Such partnerships, which incorporate mu- arrangements are most common in the area of waste- nicipal governments and community-based organiza- water and solid waste disposal. Nongovernmental or- tions, with NGOs as intermediaries, can form the basis ganizations (NGOs) often play a key role in these ini- for new institutions. The following sections review ex- tiatives, providing technical input during the design periences in a number of specific sectors that demon- and implementation phases. This type of decentralized strate the potential of these partnerships. service provision has been successful in meeting the needs of many households. But municipal authorities Urban housing often do not integrate it into trunk infrastructure, ei- Public sector attempts to provide new housing for low- ther because the settlements are considered “irregular” income groups in developing countries have not met or because the community-provided infrastructure does with much success. Sometimes the locations chosen not conform to existing codes. Public sector proposals have been inappropriate, but more often building reg- for future citywide development often ignore the exis- ulations have priced the target populations out of the tence of functional community infrastructure that is al- market. In most developing economies formal building ready meeting the demands of households and repre- regulations are largely unrealistic, mandating oversized sents millions of dollars worth of private, unsubsidized plots and rights-of-way and setting standards for infra- investment. structure and building materials that result in structures These responses to inadequate public sector services low-income households cannot afford. Not surpris- suggest a new partnership-based model for service pro- ingly, the stock of housing complying with these regu- vision that incorporates the dynamism of the private lations has not been able to satisfy demand.40 The result sector and community groups into public planning. of this shortage is a proliferation of privately developed Models of this type are already being used in countries and quite illegal settlements in many cities throughout around the world, and because of their success they the developing world. Over half the urban population have been described as the “quiet revolution” in local in Turkey resides in such settlements, which are known governance.37 Latin American cities have been in the there as gecekondus. An equal number in Karachi live in vanguard, and the process is under way elsewhere. But katchi abadis (see chapter 8). And in São Paulo, Brazil,      ⁄  the proportion of the urban population living in fave- dies, building regulations, and trunk infrastructure.44 las is reported to have increased from 9 percent in 1987 The experience of the Russian Federation and the East to 19 percent in 1993.41 European countries suggests that infrastructure invest- The public sector has had much greater success when ment alone will not suffice to stimulate housing con- it has entered into partnerships with communities—for struction in the absence of an institutional framework instance, in order to upgrade slums. Some large upgrad- for mortgage financing and land property rights.45 The ing programs, such as Indonesia’s Kampung Improve- transition in the former socialist economies has been ment Programs (KIPs), have had national impact. KIPs disastrous for new housing construction, leading to sig- have been implemented in more than 500 urban areas nificant reductions in production and mismatches be- since 1968 and have benefited almost 15 million peo- tween supply and demand. ple. Other successful upgrading programs—including Only well-functioning land markets can provide those in the Aguablanca district of Cali and the El an adequate supply of housing, and maintaining these Mezquital settlement in Guatemala City, the Million markets is another task that deserves the attention of Houses Program in Sri Lanka, and others in Fortaleza, the public sector. Providing universal registration and Brazil; Sambizanga, Angola; and Amman and Aqaba, establishing clear property rights to all urban land will Jordan—show that such efforts reduce costs and subsi- require strengthening existing institutions. Ill-defined dies significantly, improve targeting, and provide secu- land rights render land useless and discourage the rede- rity of tenure.42 In order to succeed, however, these pro- velopment of entire portions of a city. But simply pro- grams require community and individual participation viding security of tenure creates incentives to improve and initiative. In Indonesia’s KIPs, for instance, resi- housing and infrastructure dramatically.46 To avoid dents generate requests for building materials based on adding to the backlog of problem housing and neigh- need and take responsibility for installing and con- borhoods, new developments must meet basic—but structing paths and drains. not excessive—compliance standards. For the sake of Housing is a private good, unlike infrastructure for the poor, developments must seek to overcome the “spa- services like water or sewerage, and is best provided tial mismatch” that occurs when informal neighbor- through market mechanisms except when social safety hoods are situated far from centers of economic activity measures justify public sector regulation. The enabling and thus from jobs. However, the task of formulating approach endorsed by the United Nations Global Shel- appropriate regulations without also creating opportu- ter Strategy for the Year 2000, which is likely to con- nities for rent-seeking by regulators remains a challenge tinue into the 21st century, calls for private developers if there is no pressure for accountability (box 7.1). and voluntary agencies, community organizations, and NGOs to provide a bigger share of housing.43 To re- Water duce costs and respond faster to changing demands, the Inefficient and inadequate public provision of water has UN strategy relies on market forces for many aspects of been a glaring problem in many developing cities. Pub- housing provision, including markets for land, build- lic utilities often do not know where half or more of ing materials, financing, and construction. Commu- their water goes. Many years of international assistance nity organizations, assisted by NGOs and public sector aimed at upgrading networks and building capacity in agencies, have a strong role to play in providing tech- cities like Manila have not improved the situation. nical advice and additional financing. The Community While 80 percent of high-income urban residents in the Mortgage Program in the Philippines is an example of developing world have a water supply connection, only a relatively successful housing program. Since 1988 it 18 percent of low-income residents do, though some has made loans in 33 cities through more than 300 proj- share water taps with neighbors. Those without access ects to allow communities to purchase the land they to safe water (like the low-income residents of Lima) live on. In the past five years the program has served an must buy from vendors at costs that are many times average of 10,000 families annually. those for piped city water.47 Studies of water vending With this approach the government’s role in hous- report similar cost differentials for small towns in many ing markets is to address areas in which private unregu- parts of the world.48 The results of this failure are every- lated markets do not work well. The public sector needs where evident in the developing world. Publicly pro- to focus on property rights, housing finance and subsi- vided water is often of such poor quality that residents     Box 7.1 A spatial mismatch: Jakarta’s kampung residents Land rights in Indonesia are complex, combining informal tradi- off, because Jakarta needs upscale, mixed-use land develop- tional rural processes with a modern registry system. Large ment in the harbor area. But the system of land rights prevents tracts of land in the Jakarta Utara harbor area, particularly in the this natural market exchange. low-income kampungs, have often been held by families for Since kampung residents typically lack secure titles to the some generations in traditional housing developments. Typi- lands their families have lived on for generations, they cannot cally, residents do not have a registered claim of ownership— sell their land to developers for new uses. They are literally they owned the land before titles were registered. They have trapped in the kampung areas. The result is a spatial mismatch possessory rights, so generally they cannot be displaced with- between business and employment opportunities in the sub- out some compensation. They can strengthen their claims to urbs and residents stuck in the inner city. Many workers must ownership by paying property taxes and having their claims rec- make a long commute to the suburbs each day, and many oth- ognized by kampung officials. But paying taxes can be difficult, ers remain under- or unemployed. The result is a is no-win sit- since some tax officials refuse to accept payments precisely uation for both workers and the city. to avoid strengthening residents’ ownership claims. Land with- To deal with the situation, the city government has pro- out a secure title changes hands among local residents at posed the Jakarta Water Development Program. To find space prices that are estimated to be 45 percent below the costs of for the needed mixed-use developments, the city will build out securely titled land of the same quality. into the existing harbor, a process requiring expensive and In a dynamic developing city, informal property rights fos- environmentally risky land reclamation. Kampung residents ter spatial mismatches and hinder urban redevelopment. In would be asked to yield their lands voluntarily in return for new Jakarta the pattern of industrial growth under globalization is public housing accommodation in the harbor area. But this plan moving low-skill manufacturing jobs to distant suburban loca- would only make the spatial mismatch worse. A more plausi- tions. Jakarta has also made street vending illegal, severely re- ble solution is to give traditional kampung residents full title to stricting the informal food-processing and -service industry. their land, allowing them to sell it and move to the suburbs to Many low-income residents would be financially better off sell- seek employment. With the money they receive for their land, ing their land and moving to the suburbs where jobs and busi- the residents would have the capital they need not ony to re- ness opportunities are located. The city would also be better locate but also to seek new business opportunities. must treat it before using it. Service is often intermit- tization. Aggregate private investments often exceed tent and water pressure low. And many households the full cost of an equivalent supply of public water, must spend money they can ill afford on bottled water even at the high construction rates public contractors just to meet their daily needs.49 charge.51 This kind of privatization is also environmen- As incomes rise, households in many cities are re- tally problematic because of the risk of contaminating sponding to poor water service by investing in private the shallow aquifers from which well water is drawn. systems that provide a continuous supply with ade- Finally, informal privatization makes proper manage- quate pressure to support modern showers, flush toi- ment of regional water resources impossible. lets, and washing machines. Gujranwala, a dynamic In urban neighborhoods a collective water supply secondary city in Pakistan with a population of more system is much more cost-effective than a widespread than 1 million, exemplifies the response to inadequate system of wells and pumps, even when high-quality water service. Just over half the city’s households have groundwater is easily accessible. Quite minimal scale access to the piped public water supply. Of this half, economies for a collective system ensure such an advan- two-thirds have made additional investments in stor- tage. Yet private piped supplies are often not allowed to age tanks and pumps to upgrade the level of service. compete with the public water monopoly. Households without access to the public supply, many Two approaches to resolving the water supply prob- of them low income, have installed manual or electric lem are available, both involving partnerships with the pumps to draw water from the shallow aquifer.50 These private sector. One involves replacing public service investments reveal a great deal of willingness to pay for providers with centralized private concessions, and reliable water service. They also suggest that much of some large cities (Buenos Aires, Manila, and Jakarta) the water supply has been informally privatized. are doing just that by signing contracts with interna- But having each household provide or upgrade its tional firms. This approach raises two questions, how- own supply of water is not an efficient form of priva- ever: whether a private monopoly provider will be more      ⁄  successful than the public sector at assessing and re- Box 7.2 sponding to the demands of low-income communities, Haiphong: partnering with consumers and whether the state can provide appropriate regula- tion. Côte d’Ivoire, where a private company operates A partnership with consumers helped Vietnam’s state- owned Haiphong Water Supply Company (HWSC) trans- the water utilities, provides a positive example. In Abid- form itself into a profit-making utility. The utility improved jan and other, smaller cities, SODECI—a private joint the system one ward at a time (a ward is the smallest unit venture between domestic and French firms—has as- of government administration). Within four years of enter- sumed responsibility for attracting investments and ing into the partnership, the HWSC was serving 68 per- cent of the urban population with metered, reliable, high- has maintained full cost recovery with its private con- pressure water. In the wards it served, it increased the tracts. Under a policy designed to provide low-income hours when water is available from 8 to 24 hours a day and households with direct access to water, 75 percent of tripled its rate of bill collection. SODECI’s domestic connections have been provided In each ward the HWSC opened suboffices that pro- vide a direct link to customers for meter reading, billing, without a connection charge.52 collection, and troubleshooting. By metering consumers Smaller cities may find that having private firms pro- and fining them for lack of payment, the company has vide water in a decentralized, competitive system offers created incentives for consumers to conserve water. It many advantages. In Paraguay the water market was has also improved service in some outlying wards where opened to private entrepreneurs, allowing them to the service was poorest, signaling its intentions to make future improvements throughout the city. legally drill wells and lay pipes in public streets. Busi- The suboffices are staffed by people from the commu- ness flourished, and an estimated 500 vendors (aguateros) nity and enjoy a close association with the neighborhood. now compete to supply households with water, with A set of publicly displayed objectives and a “water con- negligible water losses and full cost recovery.53 In cities tract” between HWSC and the consumers help to clarify the responsibilities of the offices. The HWSC is fostering that rely on regional water resources, this system gen- a sense of partnership between consumers and the erally succeeds only if the private providers purchase service provider, heightening mutual responsibility and water from a regional agency that carefully manages providing the community with a convenient venue for prices. In low-income areas with heterogeneous de- communicating its needs. The HWSC gives bonuses to employees for achieving clear targets, such as reducing mand patterns, this type of competitive privatization the quantity of unbilled water or increasing the percent- may be preferable to replacing the public monopoly age of bills collected. These targets serve as indicators of with a private monopoly, since small niche providers corporate performance and provide the staff with incen- interact much more closely with their customers.54 tives. They also help discourage the rent-seeking that often characterizes close relationships between consumers and Competitive markets also considerably reduce regula- local employees. tory problems. A natural process of consolidation and Ward water supply employees are monitored by their scale exploitation may ensue as the market matures and community, but they are also motivated to do well by the sorts out providers according to their efficiency and inherent opportunity and challenge of their discretionary, broadly defined, situation-responsive tasks. An employee performance. In both the privatization alternatives, contract and the temporal framework provided by meter public-private partnerships point the way forward. reading, billing, and collection give structure to their varied Partnerships with community organizations can also tasks. Monthly meetings with the ward People’s Commit- improve the performance of public water utilities. Com- tee and with HWSC headquarters reaffirm the ward of- munity participation has dramatically improved the fice’s responsibility to the HWSC and provide an opportu- nity to exchange ideas and suggestions with other wards. performance of the Haiphong Water Supply Company The Haiphong model is being evaluated for replication by in Vietnam (box 7.2). other city utilities. Sewerage Source: Coffee 1999. Piped sewerage is necessary in high-density urban areas, but the costs of providing access based on the standard engineering designs public agencies commonly adopt tance of many households to pay for a system beyond are high. The high-cost, centralized sewerage systems their homes, make these designs unworkable from the used throughout industrial countries are not feasible in start.55 For example, the immense up-front costs of developing cities that have no sewerage service at all. sewer systems led the World Bank to conclude that in The very high up-front costs of collecting and treating Jakarta, waterborne sewerage systems are unlikely to be wastewater at the city level, combined with the reluc- economically justifiable for any but the most wealthy     residential areas for the foreseeable future.56 The logic holds in 5,856 lanes have built sanitary pour-flush la- of this conclusion, which confuses economic justifi- trines, lane sewers, and more than 400 secondary sew- cation with the ability to cover costs, has been chal- ers to carry wastewater out of the neighborhood. The lenged.57However, the practical impact of aiming for costs were much lower than the costs of an equivalent an expensive, modern, centralized sewerage system has public sector project, and the system has been well been that monopoly public providers have failed to in- maintained for over 15 years. crease access at a satisfactory rate. Through this work the Orangi Research and Train- Full cost recovery, particularly from user fees, re- ing Institute has developed a concept for providing sew- mains virtually impossible with sewerage services. Under erage systems in which communities and the city or the “polluter pays” principle, all households should con- state are partners. Communities finance and build tribute to collection and treatment costs, but in prac- household latrines, lane sewers, and secondary sewers. tice it is difficult to collect such fees. If high fees are These three components are termed “internal develop- imposed, people seek informal solutions, and cheap ment,” and evidence shows that communities can fi- and easy methods of improper disposal and treatment nance and manage them with appropriate technical abound—all of them difficult to monitor and regulate. support and managerial guidance. But municipal or As a result the private sector, which would need to state governments or semiautonomous regional agen- build in accordance with existing engineering stan- cies must help with long collector sewers, trunks, and dards, has not entered this market in developing coun- treatment plants—the “external development” compo- tries in the same way that it has entered the market for nent. The cost ratio of internal to external development water supply. is typically about three to one. By adopting the partner- Yet certain communities wanting improved sanita- ship model, the government can use its limited funds tion have still managed to initiate affordable alterna- to increase coverage and save on maintenance costs as tives. Lesotho’s urban areas have had success with ven- well. Since 1987 the Orangi institute has worked with tilated improved pit latrines. Brazil’s northeastern cities communities in more than 45 other settlements in have used shallow small-bore sewer schemes, in which Karachi and in 7 other cities, and the model has proved condominial sewers run through all the households in to be relatively simple to replicate.60 a block. Wastewater is discharged from a single point Decentralized neighborhood and community-based into the main trunk line—an effective alternative to systems with shallow sewers and basic community connecting each household to the trunk. 58 Applied in treatment facilities lower unit costs significantly. The a number of Brazilian cities—including Brasilia and Orangi model would never have worked if the capital Recife—this design has lowered costs to affordable lev- costs per household were not low. This example has els. The experience highlights the importance of com- great relevance for other services. If incentives are cre- munity involvement and especially of intensive consul- ated that control costs, services become more afford- tation between public agency staff and residents when able, especially when they are combined with innova- projects are being designed and implemented.59 tive repayment procedures. Repayments for water and Community organizations, often with NGOs pro- sewer connections can be integrated into monthly bills, viding technical assistance, have also gone beyond the so that users repay capital costs over months or even a household and lane levels to address neighborhood few years. The willingness of households to pay for sew- sewerage problems. An unplanned low-income settle- erage increases when the sanitation system is technically ment in Karachi known as Orangi offers an example of adequate and thus acceptable to the users, as the suc- successful community cooperation. In 1980 this com- cess of Lesotho’s low-cost solution demonstrates. munity of almost 1 million had only bucket latrines or soak pits in which to dispose of human excreta, and Urban transportation only open drains to dispose of wastewater. The inci- Automobile use increases as incomes rise and employ- dence of disease was high, as were expenditures on ment is decentralized to outlying areas of a metropolis, medical care (which could have been avoided). Poor weakening mass transit systems.61 The major problems drainage was waterlogging the land, reducing property of urban transportation relate to traffic congestion, pol- values. The Orangi Pilot Project motivated, trained, lution from emissions, and the limited mobility of the and guided the community to build an underground poor. The appropriate policies for addressing these is- sewer system at its own cost. More than 88,000 house- sues require urban governments to optimize land use,      ⁄  manage traffic and demand for transportation, formu- Reducing air pollution is an important factor in late environmental policies and measures to mitigate making cities more livable. Inspecting all vehicles to en- congestion, improve fuel efficiency, and set up vehicle sure that they comply with emissions standards is not emissions control and inspection systems.62 feasible for most cities in developing countries because While public-private partnerships have proved help- of the expense involved and problems of enforcement. ful, the public sector plays a major role in the overall A more flexible institutional approach is needed. One planning of the transportation sector. Perhaps the great- possibility shifts the focus of such regulations to large est payoff is from integrated land use and transportation fleets of vehicles such as buses, which are easier to reg- planning. New roads open the doors to land develop- ulate (and which frequently emit large quantities of ment, and compact urban centers increase the possibili- pollutants). Cities can make compliance with vehicle ties for mass transit. Curitiba, Brazil, is a convincing ex- efficiency standards part of a contract with private bus ample of how integrated public planning can improve companies trying to establish routes. Random emis- accessibility at relatively low cost. By channeling urban sions testing is another approach. Quezon City, Philip- growth along mass transit routes, the city has reduced the pines, began such an inspection campaign in 1993 after use of private cars—despite having the second-highest a six-month education period. The owners of vehicles rate of per capita car ownership in Brazil. On a typical that failed the test (about 65 percent) were fined, had workday, more than 70 percent of commuters travel by their licenses taken away, and were given 24 hours to bus in the city. As a result, Curitiba’s gasoline use per have their vehicles fixed. More than 95 percent of ve- capita is 25 percent lower than that of eight comparable hicles passed the second test.65 Brazilian cities, and the city has one of Brazil’s lowest A creative and low-cost solution that relies on part- rates of ambient air pollution.63 nerships with large trucking firms has used the lure of Coordinating transportation and land use policies re- a positive corporate image as an incentive to stop pol- mains politically difficult in many developing countries, luting. This approach has yielded dividends in Manila although sooner or later such coordination may become (box 7.3). unavoidable. A start could be made in urban areas (such as Ho Chi Minh City, Vietnam) where motor vehicle Social protection ownership is still low, land remains available, and land Households need protection against crime and vio- use patterns are still evolving. lence, but they also need protection against income Even cities with high rates of automobile ownership shocks that impair their ability to sustain themselves. can develop efficient transportation alternatives that ac- Cities acting on their own cannot provide this type of commodate the needs of all social groups. Many cities long-term security. If a city enjoying economic growth have combined innovations in mass transit with effec- offers a strong safety net, it will attract low-income tive planning and controls for automobile use: Copen- households and individuals from nearby areas, swelling hagen; Curitiba; Freiburg, Germany; Hong Kong, China; the ranks of those receiving benefits and straining the Perth, Australia; Portland (Oregon), United States; Sin- local treasury. Conversely, if a city receives a severe eco- gapore; Surabaya, Indonesia; Toronto, Canada; and nomic shock that creates massive local unemployment, Zurich, Switzerland.64 Space for walking and cycling is its ability to help its unemployed is severely limited also consciously integrated into transportation planning (box 7.4). in some of these cities, such as Surabaya. In addition to Poverty must be addressed as a national issue, and improving housing and infrastructure, Surabaya’s Kam- most redistribution programs need to be financed pung Improvement Program has revamped alleyways through national transfers, as chapter 5 suggests. But and made them attractive with plantings and pedestrian policies and institutions operating (and typically de- zones. Privatizing and deregulating bus services have signed) at the local level by individual cities influence the improved the quality of service and reduced costs in quality of life and the health of the urban poor. In par- Colombo, Sri Lanka, and in New Zealand. Informal ticular, community-driven public work schemes—often transit services that cater to low- and middle-income nationally funded and locally designed—have emerged groups—such as jeepneys in Manila and kabu-kabus in as an effective means of enabling the poor to expand Lagos—can also be integrated into formal transporta- their income-earning potential. When designed as a pub- tion networks, improving safety and efficiency. lic guarantee of work with below-market wages, such     Box 7.3 poverty is limited. Such informal mechanisms can take Manila: a positive corporate image as an the form of food sharing, microfinancing, and the shar- incentive to reduce pollution ing of housing.67 Variants of microcredit programs can increase employment opportunities through both self- San Miguel Corporation, one of the largest business con- employment and wage employment. The Full Circle glomerates in the Philippines, took the lead in banning high-polluting vehicles from its premises. A pollution Fund in Chicago, United States, and the emergency control officer at one of the company’s breweries, the loan system (Mahila Milan) in Mumbai, India, have San Miguel Polo Brewery, began requiring suppliers and helped poor women generate incomes of their own. haulers to have their trucks’ emissions tested. Only those During a crisis, microcredit programs can also mitigate whose trucks passed the test were allowed to enter the plant premises and do business with the company. The the risk of permanent income losses by allowing people approved vehicles were given stickers and retested every to keep their productive assets. Such programs require six months. The San Miguel Corporation received much careful targeting, and clients must have a full under- positive publicity for this initiative and may actually have standing of the nature of the assistance.68 Successful increased its sales as a result. Good environmental prac- tices, it found, can be good marketing. programs can also strengthen social connections in urban When the program started in April 1993, nearly a third communities, since microcredit often relies on social of the vehicles tested failed to meet emissions standards. collateral in the form of peer pressure and support. Today, only 3 percent fail. The company has expanded the Poverty reduction programs are more likely to suc- program to all its plants and vehicles across the country, including vehicles belonging to employees. ceed when low-income groups successfully negotiate for Many other firms have followed San Miguel’s example. resources and room for autonomous action.69 Naga City, Corporate members of Philippine Business for Social Prog- south of metropolitan Manila, has developed an urban ress, the Management Association of the Philippines, and poverty program targeting those in informal settle- the Philippine Chamber of Commerce and Industry have banded together to establish the Center for Corporate Citi- ments. It relies on a partnership among communities, zenship, which is actively promoting the emissions pro- an NGO, the local government, and the national hous- gram. More than 100 companies have adopted it. These ing authority. Among other things, the program has companies have erected billboards at the entrances to their helped create land-swapping and land-sharing schemes plants and compounds proudly declaring that the areas that provide land and security of tenure for squatters. are “No Smoke-Belching Compounds.” Some companies (Pilipinas Shell, Far East Bank and Trust Company, and This unique local resource mobilization scheme con- Isuzu Zexel Corporation) have gone a step further, donating tributes to equity and helps with the provision of basic emissions-testing equipment to local government teams. services.70 Collective action enables the poor to lobby The approach has caught on with operators of public with municipal agencies for rights and services—and utility vehicles, who have signed agreements with the De- partment of Environmental and Natural Resources to field to help each other in times of temporary difficulties. only vehicles that meet emissions standards. For opera- When collective efforts occur, investments that improve tors and drivers, knowledge of the health effects of air pol- the delivery of services rise substantially, as they did in lution is key in convincing them to participate. Schools and the Wat Chonglom neighborhood in Bangkok.71 These residential subdivisions have also decided to implement the program, not only to manage their own microenviron- examples confirm the willingness and ability of the poor ments but also to help everyone breathe clean air. to invest in welfare-improving measures—and the po- tential of partnership arrangements. Reducing the incidence of crime and violence les- schemes can screen out the nonneedy and increase equal- sens another burden on the urban poor. Here again, the ity across households. They can also build infrastructure trend is toward community-based actions that involve of value to communities, especially when communi- community policing and citizen-police liaison commit- ties identify and determine what is needed. Targeted tees.72 One such initiative, Programa de Desarrollo, grant programs and the involvement of NGOs and Seguridad, y Paz (DESEPAZ) in Cali, has received community-based organizations are also important to worldwide attention. DESEPAZ has established munic- the success of such programs. Some successful examples ipal security councils that bring together government include Bolivia’s Emergency Social Fund, Chile’s Mini- officials and community leaders in public meetings in mum Employment Program, and Senegal’s AGETIP.66 each of Cali’s 20 districts. This process has generated Nongovernmental safety nets can also be useful tools, programs in law enforcement and public education. even though their effectiveness in addressing urban DESEPAZ is too recent for a rigorous evaluation, but the      ⁄  Box 7.4 Shenyang: social welfare in a struggling industrial city Shenyang is the central node of the industrial complex that China: the dramatic increase in village and household enter- covers China’s three northeastern provinces. The northeast prises, and the proliferation of foreign-funded businesses. As area is the most urbanized of China’s seven regions, an ag- a result, the area has missed out on product and organizational glomeration of cities and towns with tightly linked economies, diversification and still has an economic structure very similar all heavily dependent on state enterprises. When economic re- to that of the prereform period. An alternative to enterprise- forms began in 1979, the northeast was a showplace, with its based social welfare is taking shape but is years away from many heavy industries, model state enterprises, skilled, well- being fully operational. The system emerging in Shenyang and educated labor force, and a per capita income second only to in other cities is founded on a number of changes: that of Beijing, Tianjin, and Shanghai. But as the reforms enter their third decade, the state enterprises have become better Ⅲ Transferring social welfare administration to the municipal known for their losses than for their products. The region’s social security bureau high per capita income is steadily slipping, and unemployment Ⅲ Implementing joint financing of social insurance by employ- is spreading. ees, employers, and the municipal government, and even- The losses many of Shenyang’s state enterprises have sus- tually pooling risks at the provincial level tained in the past few years have debilitated the city’s social Ⅲ Revising the benefits schedule welfare system. State enterprises in Shenyang, as elsewhere, Ⅲ Gradually transferring social facilities such as schools and have always been responsible for the social welfare of their utilities to the municipal government employees and often of their families. The companies finance Ⅲ Privatizing the housing market. and administer old-age pensions, health care, and housing and in many cases provide ancillary services such as water sys- The administration of old-age pensions is moving to re- tems for both current and retired employees. They also run cently established social security bureaus, and joint financing schools and hospitals. Except when they are in dire straits, en- for pensions has been introduced. Responsibility for the xia- terprises are expected to keep their surplus employees on the gang employees is now divided among enterprises, the mu- books, provide them with a living allowance, help them find nicipal government, and the unemployment insurance fund, new jobs, and retrain them. In Liaoning Province—Shenyang with each paying a third. A system for pooling large medical is its capital city—unemployment in disguise, known as xia- expenses across enterprises is in place, and municipal-level gang, was estimated at 15 percent in 1997, or more than 1.8 health insurance along the lines of trial schemes in Jiujiang and million persons—more than four times the 440,000 workers Zhenjiang in the east is being introduced. Nondeductibles, co- who are formally unemployed. payments, and tight regulation of the cost of drugs and med- The enterprise-based social welfare system has been ical intervention have been adopted. The central government under stress for some time. It is now beginning to collapse is soon to unveil a national framework for municipal health in- under multiple pressures: a sharp deterioration in the financial surance schemes. position of state enterprises, new competition from other re- The immediate problem is that many enterprises cannot af- gions and imports, and the rising number of pensioners and ford to pay their social insurance contribution. Moreover, many surplus employees. Many enterprises are defaulting on old-age of the municipal governments that depend heavily on taxes pensions, living allowances to xiagang employees, reimburse- from local state enterprises face a fiscal squeeze because of ments of health care expenses, and sometimes also wages the eroding tax base. Safety net programs at the national level and salaries. Such defaults were the exception a few years are urgently needed. Shenyang has succeeded so far in pre- ago, but they are now widespread in Shenyang and even more venting destitution, but it has not been able to avoid economic so in small and medium-size cities in Liaoning. distress. The city is struggling to find a way to maintain a ro- The northeast has remained on the sidelines of two devel- bust social safety net while negotiating the path to a more di- opments spearheading the growth of the nonstate sector in versified economic structure. measures are reported to have produced results in Cali, tion in Karachi, water partnerships in Haiphong, envi- as well as in Medellín and Bogotá, where the initiative ronmental improvements in Surat, community polic- has been extended.73 ing in Cali—all represent remarkable achievements. The challenge now (and it is by no means out of reach) Looking ahead is to bring similar achievements to every city. The improvements in essential urban services discussed The success stories also reaffirm the importance for throughout this chapter offer hope and direction for cities of developing appropriate institutions that get the the future. Land use and transportation planning in most from the private sector, community-based organi- Curitiba, slum upgrades in Jakarta, community sanita- zations, and NGOs. A number of communities, like     Wat Chonglom in Thailand and Orangi in Karachi, hoods to organize themselves. Participatory budgeting is Pakistan, are fortunate to have solved some of their now in place in some 50 other Brazilian cities, and the problems through self-help (with guidance from NGOs) system is scheduled to be implemented in Buenos Aires and to have developed the confidence and cohesion to and Rosario, Argentina, and in Montevideo, Uruguay.76 interact with the municipality. The internal-external Involving the private sector in partnerships requires, approach to infrastructure provision demonstrated in as a starting point, modifying rules that inhibit the pri- Orangi is now a model for future partnerships. Such vate provision of services. Private water providers in partnerships point to some of the most valuable assets Paraguay provide a good example of the kind of action for cities: the capacity of civil and community organiza- that is needed. These vendors compete legally with the tions to identify local problems and their causes, to or- public water companies and with each other. They pay ganize and manage community initiatives, and to mon- commercial, corporate, and income taxes to the gov- itor the effectiveness of public or external inputs. ernment and operate within a clear set of rules. Many This self-generated community development process governments are now putting legislation in place to is a very slow one, however. The Orangi experience allow the private sector to invest in infrastructure, typi- identified four barriers that must be overcome: a psy- cally using a build-operate-transfer (that is, transfer to chological barrier created by the expectation that the the public sector) framework. The accumulating expe- municipal government should provide all services; an riences with such systems are generating model conces- economic barrier created by the high costs of conven- sion agreements that combine transparency, flexibility, tional infrastructure provision; a technical barrier that and provisions for fair arbitration. Results have been hampers the initiation of self-help activities; and a soci- forthcoming in the form of major international private ological barrier stemming from a lack of trust that mil- investments in water, electricity, and telecommunica- itates against collective action.74 tion infrastructure. Regulatory uncertainties still need For every Wat Chonglom and Orangi, there are to be reduced, but training programs for regulators thousands of communities, especially in smaller urban have begun to address this need. centers, where community development processes have To improve the accountability of service providers, not even been initiated. Cities need to be proactive in citizens and community representatives are becoming establishing formal but friendly institutional mecha- involved in performance monitoring through “voice nisms to encourage partnerships that will bring dy- mechanisms.”77 Even approaches as straightforward as namism to development. The much-appraised experi- a poll or survey of users’ views on services or the gather- ence of Porto Alegre, Brazil, offers an example of how ing of data from both users and service providers can such a process can be initiated.75 In Porto Alegre, a city sometimes offer an effective alternative to elaborate of 9.6 million, the mayor organized the division of the participatory arrangements. The public transparency city into 16 districts, each of which set up a popular that hard data generate can in turn encourage and mo- council made up of representatives of community asso- bilize citizen groups, creating pressure for reform. Citi- ciations. Two elected representatives from each district zens’ report cards on the performance of municipal council sit on the citywide council of representatives, agencies are beginning to show results in India (box and city hall officials are assigned to act as permanent 7.5). They are now spreading to other cities, including liaisons with the district representatives. Washington, D.C. The key institutional innovation in Porto Alegre is Successful urban development also requires strategic the municipal budget forum, where the council of rep- citywide or regional planning to guide trunk investments resentatives sets the agenda for municipal spending and identify the most appropriate locations for jobs, res- based on district priorities. The final decisions on pub- idences, and transportation. The process can help cities lic spending are made in a three-way meeting of city hall avoid the worst outcomes of unplanned growth. An officials, the council of representatives, and the chamber overall strategic plan needs to be followed by coherent of councillors (who are elected on a citywide basis). decentralized implementation that creates a substantial Once projects are selected, community representatives role for the private sector. This type of careful planning supervise their progress and monitor expenditures. The and implementation is particularly important in devel- opportunity to articulate community demands and vote oping megacities, some of which are larger than many on project selection creates an incentive for neighbor- countries. It is not an argument for the type of central      ⁄  Box 7.5 Bangalore: citizens’ report cards A “report card” on urban public services is an innovative way vice function. It also joined with the Bangalore Municipal Cor- to gather systematic feedback from citizens on the perfor- poration, which initiated experiments in such areas as waste mance of a city’s service providers. In 1993 in Bangalore, India, management, and created a forum of NGOs and public agen- local civic groups used a report card on services to nudge their cies to deal with key concerns. More recently the Karnataka monopolistic service providers into responding more effec- Electricity Board has formalized periodic dialogues with resi- tively to their customers. dents’ associations to improve its services in the city. Several A small group of people concerned about deteriorating pub- agencies have strengthened their systems for redressing con- lic services enlisted a market research agency to survey citi- sumer grievances. zens on the city’s services. The findings were used to create Of the eight agencies covered by the report card, four re- a report card that rated the performance of all the major public mained indifferent. But the service providers that mattered agencies. The report card was sent to the heads of all agen- most to the people did respond. The experiment has given the cies, and its findings were widely disseminated through the public a greater appreciation of the value of citizen feedback media. What started as an informal endeavor soon led to the and of how civil society can improve local governance. creation of a new nonprofit body, the Public Affairs Center, Whether the quality of services has improved, however, is a which has continued the work in different parts of India. difficult question. A small survey conducted a year ago showed The Bangalore experiment used separate surveys for that a majority of people perceived modest improvements in middle-class and slum households. Both surveys confirmed some services and in the responsiveness of agency staff to their that public dissatisfaction with the city’s services ran high. problems. But fewer than a third of respondents believed that Even the better-rated service providers received no more than corruption had declined. The problems are deeply rooted, and a 25 percent satisfaction rating. The worst, the Bangalore De- there are no quick fixes. Some 90 percent of respondents felt velopment Authority, received a mere 1 percent satisfaction that citizens’ groups were more active than before, a sure sign rating—but it won the highest rating for corruption. The ratings that public pressure on service providers will continue. received much media and public attention and were also dis- The Public Affairs Center has since prepared report cards cussed in public forums. on services in six other large cities of India, mostly in partner- The objective was to create public interest and awareness ship with NGOs and local civic groups. Report cards have also and to pressure service providers to respond positively to the been issued for specialized services such as hospitals and pub- citizen feedback. Not surprisingly, given their large bureaucra- lic transport. In all cases, citizens have used the report cards cies, these public agencies took some time to respond. The as a trigger for collective action to increase the responsiveness first to respond was the Bangalore Development Authority, of public agencies. which reviewed its internal systems for service delivery, intro- duced training for lower-level staff, and strengthened its ser- Source: Paul 1998. planning that led to the misallocation of public invest- terference, and a lack of transparency in dealings be- ments in Eastern Europe.78 Rather, it is based on the tween regulators and the firms they regulated. Here type of strategic planning that directed urban expansion again partnerships offer a promising institutional inno- along transportation corridors and made Curitiba a vation. The monitoring and verification of information model to emulate. The contribution of the Orangi Pilot can be contracted out to professional private sector Project sewerage investment in Karachi could have been firms, educational institutes, think tanks, or NGOs, all considerably enhanced if it had been part of an overall of which have reputations for independence to defend. city sewerage plan. To encourage public participation, Citizen involvement based on the public disclosure of the planning process needs to guarantee that all plans information can then provide a stimulus for providers will be disclosed before they are implemented and that to improve. This model of public performance audits— affected parties will have the right to lodge objections. in which the regulatory task is contracted out to rep- Many local governments in Japan have recently done just utable agencies and the public uses information to mo- that, enacting ordinances on information disclosure that tivate good behavior—holds great promise in developing make information on the environment easily available.79 countries. It has been successfully implemented for in- As the private sector and community organizations dustrial regulation in Indonesia and is to be used to reg- provide more services, the public sector needs to assume ulate the recently privatized water supply in Manila.80 a revised regulatory role. The traditional approach to The policies and institutional approaches described regulation suffered from industry influence, political in- in this chapter are intended to further the “quiet revo-     lution” in local governance that is already leading cities step is to initiate an empowerment process that en- and parts of cities to improve their livability. Many ables community-based groups to define their own of the innovative and successful programs suggest goals and options—and to assume responsibility for models of partnerships that can be institutionalized actions to achieve these goals. The growing movement and promoted. Such partnerships allow synergy and toward democratization and the decentralization of the combining of resources among the public sector, power and decisionmaking that are expected to char- international organizations, the voluntary and com- acterize the 21st century will help make this possibil- munity sector, individuals, and households. The next ity a reality. C h a 8 p t e r Case Studies and Recommendations C ountries around the world have been and Hungary, some of the reforms are al- initiating reforms aimed at integrating ready under way though more remains their economies into the global market- to be done. Pakistan and Tanzania are at place and devolving central power to an earlier stage in the reform process. local governments. This report has pre- Making the most of trade sented a series of policy prescriptions for liberalization: Egypt globalizing and localizing economies, and the five case studies included in this The number of regional trading arrange- chapter describe how some of those rec- ments has surged since 1990, and many ommendations can be put into practice. countries are now members of large free The examples differ, both because the trade areas or customs unions such as the type and extent of the reforms each European Union (EU) and Mercado country needs vary widely and because Común del Sur (MERCOSUR). Should each set of reforms has been imple- countries that are not members of a re- mented in a vastly different economic gional trading arrangement seek prefer- and policy environment. Context is par- ential access to their neighbors’ markets? ticularly important here, since the feasi- How does this option compare with uni- bility of reform depends on the political lateral or multilateral liberalization? This conditions in a country. Successful re- examination of Egypt’s trade policy form requires careful sequencing and options illustrates the trade-offs many the willingness to exploit sometimes developing countries face in choosing fleeting opportunities. whether to join a regional trade group The five cases discussed here—in the (box 8.2). And it demonstrates the im- Arab Republic of Egypt, Hungary, Bra- portance of some of the recommenda- zil, Pakistan, and Tanzania—represent a tions in chapters 2 and 3. regional sampling of fairly typical pol- Since the mid-1970s, Egypt has been icy situations (box 8.1). Each case study steadily liberalizing its trade policies, describes the policy setting, the recom- which has contributed to economic mended reform strategy, and the success growth. But the benefits from liberalized of new policies thus far. In Brazil, Egypt, trade have been stymied by domestic       ⁄  Box 8.1 It was fueled by sizable increases in foreign assistance, Five case studies remittances from Egyptians working abroad, and for- eign direct investment.2 This growth spurt ended in Making the most of trade liberalization: Egypt. This case 1986, largely because of a regional economic slowdown applies chapter 2’s proposals for trade reform, showing how international trade agreements can be used to caused by declining oil prices. The level of aggregate de- demonstrate commitment to freer trade. It also illustrates mand in the economy then fell further in the early some of the disadvantages of regional (as opposed to 1990s because of government spending cuts, an in- global) trade agreements and the kinds of domestic re- crease in real interest rates, and a drop in exports to the forms the Egyptian government will have to implement to take advantage of the opportunities offered by global trade. former Soviet Union and Eastern Europe. Per capita Reforming weak banking systems: Hungary. This case growth of real national output slowed from an average study deals with the financial sector reforms discussed in of 2.5–3.0 percent a year in 1989–91 to 0.4 percent in chapter 3. It demonstrates clearly that regulators need to 1992 and 1993. take prompt action when a bank violates specific guide- lines or procedures, as the report has argued. Taking steps The Egyptian government responded with an im- to reduce this so-called “regulatory forbearance” is the pressive program of economic reform. Fiscal tightening next major challenge facing Hungarian policymakers. reduced marginal tax rates and the government’s bud- Macromanaging under fiscal decentralization: Brazil. Building on the themes of decentralization and democratic get deficit.3 Monetary reforms included decontrolling subnational governance discussed in chapter 5, this case interest rates, devaluing and unifying exchange rates, study illustrates the need for carefully sequenced decen- reducing the growth of the money supply, and liberal- tralization. It also identifies the changes Brazil will need to izing the capital account. A 1991 law established a legal make in order for its newly centralized structure to function effectively, including establishing electoral rules, creating basis for privatization, and by September 1998, 113 of regulations to manage relations between national and sub- the initial 314 public enterprises originally targeted had national governments, and drafting rules for subnational been at least partially privatized. In the same year, the borrowing. Parliament ratified a law authorizing the privatization Improving urban living conditions: Karachi. The Karachi case study draws on chapter 7 to show how community of banks. groups and informal developers can complement the ef- Foreign investors were quick to react. In 1995 they forts of the public sector to provide essential services. put $400 million in foreign direct investment into Cultivating rural-urban synergies: Tanzania. The final Egypt, followed by $800 million in 1996 and around case study focuses on reforming foreign trade (chapter 2) and establishing policies that deal with urbanization and $1.2 billion in 1997. Half the foreign direct investment growth (chapter 6). It demonstrates how one country can is in manufacturing and 30 percent in banking. Tariff use international trade and urban-rural economic linkages revenues as a share of total imports fell from 25 percent to stimulate growth in both the urban and rural sectors. in 1985 to 17 percent by 1997, reflecting the coun- try’s increased openness to trade. As a result of these flows and trade reforms, real gross domestic product constraints, including an inefficient service sector, a slow- (GDP) grew by 5.1 percent in 1996 and by 5.9 percent moving government bureaucracy, and overcrowded ports in 1997. and transportation facilities. For some time, Egyptian in- dustrial goods have had duty-free access to European Red tape and inefficient services constrain exports markets, but Egypt is now considering signing an ex- Despite its reforms, Egypt has yet to take full advan- panded preferential trading arrangement with the EU.1 tage of the potential of trade liberalization. The coun- Such an agreement may reassure investors of Egypt’s try has many advantages to exploit in producing man- commitment to liberal trade policies, but—as explained ufactured exports, including a convenient location and in box 2.1—it would also mean that the pattern of wages that are one-tenth those in Israel or Tunisia. Egypt’s imports and exports will be shaped less by mar- Given these positive factors and its duty-free access to ket forces and more by the differences in tariff treatment European markets, the country was expected to in- between Europe and Egypt’s other trading partners. crease its manufactured exports rapidly.4 Manufactured exports (in 1992 prices) did increase, but slowly, rising Initial reforms from $1.4 billion in 1988 to an estimated $2.4 billion Economic growth in Egypt accelerated between 1975 in 1996—still only about 17 percent of total goods and and 1985 following the adoption of open-door policies. services export revenues.      Box 8.2 The Arab Republic of Egypt at a glance Lower- Exports and imports Middle East middle- (millions of U.S. dollars) 15,000 Egypt, and North income Exports Poverty and social indicators a Arab Rep. Africa countries Imports 10,000 GNP per capita (U.S. dollars) b 1,180 2,060 1,230 Poverty (percentage of population 5,000 below $1 per day) 7.6 .. .. Urban population 0 (percentage of total population) 45 57 42 1991 1992 1993 1994 1995 1996 1997 Life expectancy at birth (years) 66 67 69 Key economic ratios 1976 1986 1996 1997 Gross domestic investment/GDP 28.4 23.7 16.6 17.7 Exports of goods and services/GDP 22.3 15.7 20.2 20.2 Gross domestic savings/GDP 16.7 13.8 10.8 13.0 1976–86 1987–97 1996 1997 1998–2002 Average annual GDP growth 7.1 4.0 5.0 5.5 5.2 . .Not available. a. Data shown are from the latest available year within the range 1991–97. GNP per capita figures are from 1997. b. Calculated using World Bank Atlas method. Source: World Bank, World Development Indicators, 1999. One reason for this sluggish growth is the inefficiency worth of imports a year; in Singapore the average is of services, which raises the price of inputs and transac- $666 million a year.6 tions costs to exporting firms and undermines their The government has begun to reduce bureaucratic competitiveness.5 For example, the four main Egyptian delays and charges and lower transportation costs.7 But ports (Damietta, Port Said, Dekheila, and Alexandria) further reform remains essential. For example, the cus- are essentially state monopolies, and their service charges toms system could be improved in a number of ways, are three times those of their closest competitors. Con- including the following: bringing in international in- tainer freight rates to Egyptian ports are generally 15 to spection firms; accepting valuations of imports based 20 percent higher than rates to other Mediterranean on invoices, rather than having the customs service ports, and air freight rates to and from northern Egypt- value items; focusing tests of imported goods on safety, ian cities are twice those to cities in Israel. which is a legitimate concern, not on quality, which can In addition, all trade transactions are subject to an be better judged by the ultimate buyer; and accepting onerous bureaucratic burden. A 10 percent sales tax is international standards of certification. applied to all commodities, including inputs to goods Local transportation networks also have to be produced for export, making it harder for firms to sell strengthened. Private competition should be intro- abroad at competitive prices. A process does exist for duced in port handling, a move that has reduced ship- refunding import tariffs on inputs to goods for export, ping charges by as much as 50 percent in Mexico and but it involves four forms, a letter, a permit, and two Chile. The build-operate-transfer contract offered to separate committee reviews. Imports also face delays, the private sector for the expansion of the Cairo inland as all goods must go through multiple clearance, licens- port of Athr al-Nabi and the construction of two new ing, and inspection procedures that impose a cost esti- specialized ports are encouraging steps in this direction. mated as equal to an extra 15 percent tariff. Each New projects are on the drawing board to improve road Egyptian customs official clears an average of $600,000 transportation, including an upgrade of the Mediter-      ⁄  ranean coastal road as part of the North African coastal measures taken to enhance competition in the service road (which will eventually link with Europe’s road net- sector, such as permitting foreign direct investment, can work via the Gibraltar crossing). A 113-kilometer improve the productivity of many industries further Greater Cairo Ring Road is also under construction, down the stream of production. Furthermore, to the ex- but there is still much room for improvement, as the tent that such an agreement reduces regulatory barriers high incidence of traffic fatalities indicates—44 deaths to Egyptian exports (because those exports now comply per 100,000 kilometers driven. with EU health, safety, and product standards), the ben- Nontrade constraints on foreign direct investment efits could be as much as 1.8 percent of Egyptian GDP.9 will also have to be eased. At the moment, entry into a Even further gains will accrue to Egypt if enhanced for- market requires government approval. Moreover, re- eign investment enables its firms to fuse into the global strictive labor laws make exit expensive, which discour- production networks of European firms. ages firms from entering markets in the first place. Sur- Only a comprehensive trade reform agenda that tack- veys of firms suggest that about 30 percent of managers’ les red tape and brings down barriers to trade and invest- time is devoted to coping with regulatory demands. Re- ments in goods and services will benefit Egypt. A broad moving these regulatory impediments, especially those preferential trade agreement with the EU would enable that discriminate against foreign investors, is crucial if Egypt to harmonize its domestic regulations with those a country is to increase investment rates, as chapters 3 of its major trading partner. But such an agreement is and 6 emphasize. no substitute for Egypt’s full participation in the forth- coming Millennium Round of World Trade Organiza- Further trade reform tion (WTO) negotiations, which holds out the promise In Egypt dissatisfaction with export performance has led of multilateral reform in services and agriculture. to renewed interest in trade reform. But entry into some Reforming weak banking systems: Hungary form of preferential agreement with the EU requires careful assessment. As noted earlier, Egyptian exporters The many banking crises in developing countries over have had duty-free access to EU markets for industrial the last several decades—with their deleterious conse- goods since the 1970s. Egypt is currently negotiating quences for poverty reduction, social stability, and a European-Mediterranean Agreement with the EU that growth—illustrate the importance of a sound regula- would seek to liberalize trade in other ways. However, tory framework for banks. The need becomes all the there are different types of preferential agreements with greater as capital flows move freely across national bor- the EU, not all of which would benefit Egypt. ders and as the number and complexity of financial in- A first option is for Egypt and the EU to eliminate struments available to banks expand. Making progress their tariffs on imports of goods from one another. toward a strong independent bank regulatory regime, Such an agreement could lead Egyptian importers to as described in chapter 3, should be a primary concern shift their purchases away from the most efficient for- for policymakers in developing countries. Hungary’s eign supplier to EU firms whose cost of supplying the progress points to several lessons of wider applicabil- Egyptian market is artificially lowered because they pay ity—and to the challenges facing countries that have no tariffs. Indeed, one analysis suggests that such an inherited state-run banking systems with substantial agreement could actually reduce Egyptian welfare by bad debts (box 8.3). the equivalent of 0.2 percent of GDP. In contrast, full In the last 10 years Hungary has dramatically trans- unilateral elimination by Egypt of such tariff barriers formed its banking sector. Once dominated by insol- would benefit Egypt.8 A preferential liberalization that vent government-owned institutions, the sector now is confined to tariffs on goods offers little to develop- has many privately owned banks and is oriented toward ing countries, especially when compared with unilat- serving a market economy. Hungary made this trans- eral elimination of tariffs on goods trade. formation as part of a radical restructuring of the econ- However, a preferential trade agreement that includes omy aimed at replacing socialist principles with a pri- liberalization in goods, harmonization of standards, and vate market system. greater access to service markets can offer substantial Hungary’s experience illustrates three recommenda- benefits to developing countries such as Egypt. As ser- tions from chapter 3. First, it demonstrates the need to vices are used extensively as inputs in the export sector, strengthen bank supervision and to insulate it from gov-      Box 8.3 Hungary at a glance Upper- Exports and imports Europe and middle- (millions of U.S. dollars) Central income 25,000 Exports Poverty and social indicators a Hungary Asia countries 20,000 Imports GNP per capita (U.S. dollars) b 4,430 2,320 4,520 15,000 Poverty (percentage of population 10,000 below $1 per day) 25 .. .. 5,000 Urban population 0 (percentage of total population) 66 67 73 1991 1992 1993 1994 1995 1996 1997 Life expectancy at birth (years) 70 69 70 Key economic ratios 1976 1986 1996 1997 Gross domestic investment/GDP 35.9 26.9 26.8 .. Exports of goods and services/GDP 38.8 39.6 38.9 .. Gross domestic savings/GDP 31.8 25.5 25.7 .. 1976–86 1987–97 1996 1997 1998–2002 Average annual GDP growth 2.4 –0.8 1.3 4.4 5.2 . .Not available. a. Data shown are from the latest available year within the range 1991–97. GNP per capita figures are from 1997. b. Calculated using World Bank Atlas method. Source: World Bank, World Development Indicators, 1999. ernment interference. The inability of Hungarian bank- result of the breakup of the Council of Mutual Eco- ing supervisors to take early action against banks with nomic Assistance (COMECON) and the collapse of deteriorating loan portfolios worsened the country’s the Soviet Union, Hungarian firms lost 60 percent of banking difficulties. Second, Hungary’s experience sup- their export market. Many enterprises were unable to ports the case for complementing regulatory reforms adjust to the competitive pressure of a liberalized im- with private sector monitoring of banks. Hungary port regime, which pitted them against both domestic strengthened its monitoring capabilities by reforming and foreign firms. As a result, enterprise arrears to the public deposit insurance scheme, improving corpo- banks skyrocketed, endangering the banking system. rate governance of banks, and mandating the issue of Hungary’s early attempts at bank reform were tenta- subordinated debt. Third, Hungary’s experience demon- tive.12 The government began by creating a two-tier strates that foreign participation in national banking structure in 1987, shifting the corporate banking busi- systems need not wait until domestic banks have been ness of the National Bank of Hungary to three newly strengthened. A recent analysis has suggested that for- formed commercial banks. The number of banks (ex- eign participation in transition economies’ banking sys- cluding deposit associations and innovation funds) ex- tems has tended to improve their performance.10 panded from 8 in 1986 to 30 in 1990 as a result of new entry and the conversion of small, specialized financial Initial reforms institutions into commercial banks. The market share When the Berlin Wall fell in 1989, Hungary was of the four largest commercial banks fell from 58 to slightly more advanced in banking reform than its East 48 percent between 1987 and 1990. But large institu- European neighbors. But the government still faced tions continued to dominate the banking sector. To- many of the same problems as they did. Most of the gether with the government-owned national savings banking sector was in public hands, and its assets were bank, the five largest banks accounted for 82 percent dominated by directed loans to state enterprises.11 As a of total assets in 1990.      ⁄  In 1991 the government introduced a new regula- 1994–98, and doubtful loans as a proportion of assets tory framework based on market-oriented principles.13 dropped from 20 to 3 percent in 1993–97. Margins on The 1991 banking act introduced prudential concepts, loans also began to fall with increased competition— Bank for International Settlements (BIS) regulations on from 7 to 5 percent in 1998. provisioning, and limits on exposure. The accounting But Hungary’s banking system faces continuing act introduced international accounting standards. The challenges. For example, problems remain in enforcing new bankruptcy code prohibited banks from simply regulations on domestically owned banks. Two such rolling over unpaid loans at maturity and forced them banks failed in 1998. One was the second biggest in to provision fully for their losses. Hungary; it appears that its management was largely While these reforms were enacted, they were not al- unconstrained by a dispersed local ownership, believed ways enforced. Nor did they address the immediate that it was too big for government to allow it to fail, problem of bank insolvency. Government-owned banks and so lent recklessly. Regulators were slow to act, de- were burdened by nonperforming loans, including many spite a bank run in February 1997. Rather than force inherited from the former regime and some more re- prompt corrective action, the government provided cent loans to state-owned enterprises. Under the terms cash infusions and suspended capital requirements. of two workout programs in 1991–92, the government Only in June 1998 was the management replaced and took over about $1 billion, or 90 percent, of the banks’ in-depth restructuring begun. nonperforming debt. Unfortunately, this debt relief was provided uncondi- Future reforms tionally. Banks receiving funds were not forced to mod- This episode, in which it took more than a year after a ernize, the same managers remained in place, and regu- bank run to restructure a bank, originated in part in lations were not enforced. As a result, bank managers legal impediments on the power of the supervisory au- continued to believe that the government was ready to thorities. The Basle Accords core principles suggest that provide unconditional relief to any bank in trouble. Not banking supervisors should have the legal authority to surprisingly, poor lending practices continued. issue and enforce the regulations necessary to maintain In 1994 the government decided to go one step fur- the soundness of the banking system. But in Hungary ther and privatize the banks. To make the banks salable, the Ministry of Finance—rather than banking supervi- it had to inject about 9 percent of GDP into the bank- sors—had exclusive power to issue regulations. More- ing system. Banks were recapitalized to meet BIS stan- over, the supervisory authority appeared constrained in dards by the end of 1995. In each troubled bank, loans its ability to take appropriate disciplinary actions. Be- were separated so that a core bank with a solid portfolio cause under current law disciplinary measures can only could be readied for privatization. Unlike the bailouts be taken on the basis of audited accounts, Hungary’s of 1991 and 1992, this plan stipulated that banks re- bank supervisors could not respond quickly to regula- ceiving state funds modernize their systems of control tory infractions. and operation, replenish the funds they held in liquid Strengthening the hand of banking supervisors will form against the risk of loan defaults, and adopt best help the stability of the banking system, but traditional practices in loan appraisal, risk assessment, and asset bank regulation may be insufficient by itself to forestall clarification. In some cases senior bank managers were excessive risk-taking by banks. As discussed in chapter replaced. 3, countries should consider how to complement gov- When privatization started in 1994, foreign banks ernment regulation by stimulating private sector moni- purchased many Hungarian banks. Between 1994 and toring of banks, through such steps as improving the 1998 foreign ownership in Hungary’s banking sector corporate governance of banks and mandating the issue increased from 15 to 60 percent, while direct state own- of subordinated debt.14 ership of the sector fell from 67 to 20 percent. Privati- If Hungary can take further steps to reduce regula- zation appeared to have the desired effect on bank per- tory forbearance and build a greater role for private sec- formance. MKB, the first large bank to be privatized, tor monitoring of banks, then the country will be well saw its income triple, the number of branches double, on its way to cultivating a first-class banking system. At and its staff shrink from 1,800 to 1,240. Returns on a fundamental level, Hungary has looked outward to bank assets increased from 0.5 to 1.0 percent in find solutions to its banking problems. It has recog-      nized the value of adopting and enforcing international tions: first, that the decentralization of revenues match banking standards, while increasingly resisting bank the decentralization of expenditures; second, that cen- bailouts to politically connected insiders. tral governments maintain a hard budget constraint in The Hungarian experience offers pointers for other their dealings with subnational governments; and third, transition countries, especially in Eastern Europe. Given that constitutional mandates, particularly electoral the central role that banks play in transforming both do- rules, be in place so that the first two measures can be mestic and international flows of savings into growth- enforced (box 8.4). enhancing investments, the payoff to a sound banking system will reach far beyond minimizing the risk and Formal decentralization costs of banking crises. In 1988 Brazil’s first postmilitary constitution sought to decentralize political power. Power at the federal Macromanagement under fiscal level is now divided among the executive, legislative, decentralization: Brazil and judicial branches. The president, who heads the ex- In the early decades of the 21st century, demands for ecutive branch, is elected by direct popular vote for a greater local political autonomy will mold the political four-year term. Congress has two houses—the Cham- structures of developing countries. Policymakers will ber of Deputies, in which each state receives a certain have to manage the process of reallocating rights and number of seats according to its population, and the obligations to different tiers of government. Brazil’s ex- Senate, in which each state has three senators. perience with decentralization, which resulted in a se- In principle, the constitution gives the president ries of intergovernmental fiscal crises, highlights the dif- considerable powers over the legislature. The president ficulty of managing the politics of fiscal decentralization has the exclusive right to initiate legislation in some in a period of democratic and economic transition. It policy areas, including those that create jobs or increase also confirms three of chapter 5’s policy recommenda- salaries in many parts of the public sector. The presi- Box 8.4 Brazil at a glance Latin Upper- Exports and imports America middle- (millions of U.S. dollars) 75,000 and the income Exports Poverty and social indicators a Brazil Caribbean countries Imports 50,000 GNP per capita (U.S. dollars) b 4,720 3,880 4,520 Poverty (percentage of population 25,000 below $1 per day) 17 .. .. Urban population 0 (percentage of total population) 80 74 73 1991 1992 1993 1994 1995 1996 1997 Life expectancy at birth (years) 67 70 70 Key economic ratios 1976 1986 1996 1997 Gross domestic investment/GDP 23.1 19.1 20.7 22.8 Exports of goods and services/GDP 7.0 8.8 7.1 6.2 Gross domestic savings/GDP 20.7 21.6 18.6 20.6 1976–86 1987–97 1996 1997 1998–2002 Average annual GDP growth 2.9 1.9 2.8 3.2 3.5 . .Not available. a. Data shown are from the latest available year within the range 1991–97. GNP per capita figures are from 1997. b. Calculated using World Bank Atlas method. Source: World Bank, World Development Indicators, 1999.      ⁄  dent alone prepares the annual budget and must seek to a set formula to states and municipal governments. congressional approval for it. The Congress is restricted The second involves the state value added tax (VAT), in the kinds of amendments it can propose to the bud- which state governments must share with the munici- get, and it cannot initiate programs or projects not in- palities in their jurisdictions. Consequently, the munic- cluded in the president’s budget.15 ipal share of net tax revenues after transfers increased In practice the president’s power is circumscribed by by roughly 40 percent in six years, rising from 12 per- the difficulty of marshaling support in a political sys- cent in 1987 to 17 percent in 1992.18 tem with so many parties (15 are represented in the Although the 1988 constitution emphasizes decen- Congress) and weak party discipline at the national tralization, it does strengthen the central government’s level. The electoral system, and particularly propor- control in one essential area: personnel. It defines the tional representation, are partly responsible for this rights of public sector employees at all three levels of multiplicity of parties. Candidates for the Chamber of government and provides employees with job and Deputies run at large in each state rather than facing salary security. Governments cannot dismiss redundant off in single-seat districts, so small parties must scour civil servants or reduce nominal salaries. The constitu- an entire state to obtain enough votes to win a seat or tion also gives public employees generous pension two. Strong state loyalties lead politicians to form al- rights, which have been a factor in subsequent fiscal liances in support of projects that will benefit their own crises, since labor costs are a significant share of subna- state, regardless of their party. Sitting state governors tional expenditures.19 These controls exemplify the command the loyalty of federal deputies, since the gov- problem of overregulating subnational governments de- ernor’s support is more useful in their campaigns than scribed in chapter 5. the president’s. Because of their influence over deputies and senators in their party, state governors can thwart State borrowing and the debt crisis or propel presidential designs.16 Decentralization in Brazil has resulted in a prolonged The constitution sets up a three-tier governmental macroeconomic crisis sparked by the growing indebt- structure consisting of the federal government, 26 states edness of the states.20 While the new constitution gives (plus a federal district with the status of a state), and the national Senate the power to deny all subnational about 5,500 municipalities. States elect their governors proposals for borrowing, the Senate has rarely done so. directly and have unicameral legislatures, with the mem- As a result, states and municipalities continue to bor- bers elected at large by proportional representation. This row from a wide variety of sources. They have issued structure is repeated at the municipal level, with may- bonds on the domestic market and have borrowed from ors elected directly and municipal councillors elected at domestic private commercial banks and various federal large. The constitution gives subnational governments intermediaries, including the federal housing and sav- broad but vaguely defined powers and creates no real ings bank and the federal development bank. All but boundary between them. It grants states “all powers not two of the 26 states own commercial banks from which otherwise prohibited to them by the constitution” and they have occasionally borrowed. More frequently, they municipalities “the power to provide services of local in- have forced these banks to lend to favored clients. States terest.” Since the constitution makes the municipal au- have also borrowed abroad, both from multilateral thorities the third tier of government, states have no agencies (which demand federal guarantees) and from power over the actions of the municipalities within their private lenders (which do not). jurisdictions. The debt crisis unfolded in three acts. The opening Although the constitution is vague about the divi- act was a legacy of the international debt crisis of the sion of responsibilities among levels of government, it 1980s, when states—along with the federal govern- divides up revenues very explicitly. It assigns specific tax ment—ceased servicing their debt to foreign creditors. bases to each level of government and creates a system Once the government and creditors at the national level of tax sharing that substantially redistributes revenue had reached an agreement, the federal government tried among both the levels of government and the regions.17 to induce the states to resume servicing their debt. In The tax-sharing system has two major components. 1989 the federal government agreed to transform the The first consists of fixed shares of the federal govern- accumulated arrears and remaining principal into a sin- ment’s two principal taxes—the income tax and the in- gle debt to the federal treasury, rescheduling $19 bil- dustrial products tax—which are distributed according lion on these terms.21      The second act, which began in the late 1980s, in- Negotiations to resolve the debt situation began in volved the states’ debts to federal financial institutions. mid-1995 with three parties: the federal Congress, the It was resolved by rescheduling roughly $28 billion in president and his economic team, and the states. But loans and transferring them to the federal treasury. But not until December 1997 did the first major debtor the federal government wrote an escape clause into the state, São Paulo, sign a binding agreement with the fed- agreement. If the ratio of states’ debt-servicing costs to eral government. The other major debtor states fol- their revenues rises above a threshold fixed by the Sen- lowed over the next nine months. In general, the agree- ate, the excess can be deferred and capitalized into the ments followed the pattern of the two previous debt outstanding stock of debt. By rescheduling the princi- agreements. Debt was rescheduled rather than written pal and placing a ceiling on debt-servicing costs, these off, and a debt service ceiling was imposed above which agreements considerably reduced the states’ immediate costs could be capitalized into the stock of debt. The burden. But the escape clause also made it seem that main innovation of the new debt agreements is a large the federal government was prepared to provide debt interest rate subsidy. Rather than requiring subnational relief to any state that required it. governments to pay the existing interest rate on federal The third act began in the early 1990s and revolved bonds, the federal government agreed to impose a fixed around defaults on state domestic bonds. Four large real interest rate of 6 percent. states do the most financing through bonds: São Paulo, With each debt workout, the federal government has Rio de Janeiro, Minas Gerais, and Rio Grande do Sul. tried to tighten the regulations on state borrowing. Traditionally, the states’ commercial banks underwrite States benefiting from debt rescheduling are required these bonds, which are ultimately sold to private banks to permit the federal government to deduct debt ser- and investors. They generally have a five-year maturity vice from intergovernmental transfers. New federal date, with interest due then. Ironically, the bond crisis lending to states currently in default is prohibited. The was precipitated by the considerable success of the gov- constitution has been amended to prohibit the issue of ernment’s stabilization plan, the Plano Real. The plan new state bonds until 2000, and the central bank does dramatically reduced inflation, so states could no longer not allow private banks to increase their holdings of count on inflation to reduce real salaries and pensions state debt. These federal regulations have not been over time.22 As a result, state governments soon found enough to forestall the most recent act of the debt cri- themselves with payrolls equal to 80 or 90 percent of sis that started in 1999, since most of the recent growth their revenues. in debt stems not from new borrowing but from the As state finances became more precarious, private capitalization of interest on existing debt. banks began increasing their interest rates and shorten- The macroeconomic effects of the rescheduling ing the length of time they would hold bonds. Ulti- agreements have been limited. Although the agree- mately, private banks declined to hold state debt at any ments lowered the interest rates the states pay, the fed- price. The states found themselves unable either to pay eral government continued to be the states’ creditor and or to reschedule their debts and sought relief from the to pay the actual cost of borrowing funds. The interest federal government, which authorized them to ex- rate paid by the public sector as a whole has not de- change their bonds for more readily marketable federal clined. The terms of the agreements, moreover, have bonds. But with the interest rate on federal bonds hov- not been enough to forestall the capitalization of inter- ering at 25–30 percent in real terms, the stock of bond est on debt owed to the federal government. State debt debt exploded by $12 billion in 1995 and by another has continued to grow, so the agreements have not re- $10.7 billion in 1996. At the end of 1996 the stock of duced the aggregate interest costs paid by the public state (and municipal) bond debt stood at $52 billion. sector. They have merely shifted more of the interest The heavy interest obligations on this growing stock of costs to the federal treasury. debt, combined with the states’ inability to reduce per- sonnel costs or raise revenues, has resulted in growing What can be done? state and municipal deficits. From a surplus of 0.7 per- Some aspects of the solution to this financial and inter- cent of GDP in 1992, the operational balance of state governmental crisis are not hard to identify. Initially, and municipal governments fell to a deficit of 2.3 per- the federal government must address the underlying cent of GDP in 1997—52 percent higher than the fed- source of the debt crisis by finding a way to control per- eral government deficit. sonnel costs, which consume 80–90 percent of current      ⁄  revenues. Reducing these costs will require eliminating Since the height of the debt crisis, Brazil has taken the controls on state personnel policies mandated in the several positive steps. In 1998 the Congress approved a 1988 constitution, so that states can dismiss redundant constitutional amendment that would allow states to dis- staff, negotiate salary reductions, adopt stricter criteria miss staff (provided their personnel spending exceeded a for retirement, and reduce pension benefits. certain percentage of state revenues). In 1999 the gov- The government must also act to eliminate the ex- ernment responded to one state’s much-publicized de- pectation of federal bailouts. The first bailout signaled fault on rescheduled debt by exercising its new authority states and their lenders that the federal government was to deduct the overdue debt service from federal-to-state ready to step in to rescue debt-ridden local govern- transfers. Later in the year the Congress opened debate ments. While some lenders may actually have believed on a proposal to change the electoral rules for the lower their borrowers were creditworthy, they also believed house, replacing the current system of proportional rep- that the federal government would make good on state resentation with one in which half the seats would be obligations if the stability of the financial system were filled from single-seat electoral districts. The first two of threatened or a breakdown of services in a major state these actions will go a long way toward providing states loomed. This implicit federal guarantee permitted with the means and incentive to respond to fiscal pres- states to continue to borrow well past the point at sures without resorting to default. The third, if it func- which they had the means to service their debts. tions as its advocates anticipate, may reduce party frag- The current federal regulations to limit subnational mentation and strengthen the government’s ability to borrowing are clearly not sufficient to counteract this resist appeals for bailouts. expectation. But the states cannot borrow unless some- Improving urban living conditions: Karachi one is willing to lend to them. If private lenders are con- vinced that the federal government will not bail out de- The explosive growth of urban populations in develop- faulting states, the lenders themselves will act as a source ing countries will challenge the capacity of society to im- of restraint.23 Convincing lenders that no federal bailout prove urban living conditions. This case study suggests will be forthcoming requires more than a statement how the recommendations of chapter 7 concerning the of intent, particularly given Brazil’s recent history of provision of municipal services can be translated into ac- bailouts. The federal government needs to demonstrate tion in Karachi, Pakistan’s major metropolis.25 Karachi its commitment by allowing a state government to de- is representative of many large cities in developing coun- fault and leaving the lender and the state to work out a tries where the public sector has had difficulty coping settlement. Once private lenders are persuaded that fi- with rapid urban growth. It shares many characteristics nancing subnational governments carries real risks, they with Bombay, Istanbul, Jakarta, and Lagos, though the are likely to restrain their lending despite the supplica- reasons for the difficulties in providing services differ tions of state governors. Establishing a constitutional re- from city to city. As chapter 7 recommends, in Karachi straint on the federal government’s ability to lend to the the public sector needs to tap the knowledge and dy- states will enhance the credibility of this policy.24 namism of the rest of society through partnerships with Softening federal mandates on subnational person- private enterprise, community groups, and nongovern- nel policies and hardening the budget constraint on sub- mental organizations (NGOs). With their support, the national borrowing will help to forestall future debt public sector can focus on those services only it can pro- crises. But ultimately, sustainable reform requires chang- vide, including land titling, appropriate building and ing the political circumstances that gave rise to these development regulation, and trunk infrastructure for policies. The distribution of power between the presi- water, sewerage, and roads. dent and the legislature needs to be reexamined, along with the electoral rules that result in such a high degree Karachi today of party fragmentation and lack of discipline. Several Karachi’s 11 million people account for around 8 per- measures discussed in chapter 5 are especially relevant cent of Pakistan’s total population and a quarter of its to the Brazilian case. To make it harder for interest urban population (box 8.5). The city grew rapidly after groups from the states to conspire against the whole, the the massive migration that followed the partition of office of the president must be strengthened, perhaps by British India in 1947, putting severe stress on the hous- requiring a supermajority to override a presidential veto. ing market.26 The public sector, which owned most of      Box 8.5 Pakistan at a glance Low- Exports and imports South income (millions of U.S. dollars) Poverty and social indicators a Pakistan Asia countries 15,000 Exports GNP per capita (U.S. dollars)b 490 390 350 Imports 10,000 Poverty (percentage of population below $1 per day) 34 .. .. 5,000 Urban population (percentage of total population) 35 27 28 0 Life expectancy at birth (years) 64 62 59 1991 1992 1993 1994 1995 1996 1997 Key economic ratios 1976 1986 1996 1997 Gross domestic investment/GDP 17.2 18.8 18.7 17.4 Exports of goods and services/GDP 10.7 12.3 16.5 16.2 Gross domestic savings/GDP 7.9 10.9 14.2 12.4 1976–86 1987–97 1996 1997 1998–2002 Average annual GDP growth 6.8 4.7 4.7 –0.4 5.8 . .Not available. a. Data shown are from the latest available year within the range 1991–97. GNP per capita figures are from 1997. b. Calculated using World Bank Atlas method. Source: World Bank, World Development Indicators, 1999. the land in and around Karachi, reserved for itself the unplanned settlements termed katchi abadis, in which dominant role in land development. Regulations on half the city’s people now live. The population of katchi land development drove up the cost of new housing by abadis has grown at an annual rate of 9 percent, nearly mandating large plot sizes, making generous allocations twice the city’s overall population growth rate of 4.8 for rights-of-way, setting high on-site infrastructure percent. Densification of existing housing in inner-city standards, and mandating costly building materials. areas and illegal construction in the suburbs have met This excessive regulation priced most households out of the rest of the housing gap. the market. Delays in extending trunk infrastructure— A supporting industry has emerged in connection roads, piped water, and sewerage—limited the supply with katchi abadis. Unregulated land developers obtain of land with access to those services and raised still fur- land—often in collusion with government develop- ther the prices of plots that already had them. These ment authorities—and divide it into plots that are sold constraints on the supply of housing interacted with to individual households. Such middlemen illegally ac- constraints on demand, especially the inability of low- quire at least 1,000 acres of state land every year in and middle-income households to obtain mortgage fi- Karachi and use it for informal housing. Water distri- nancing. The result has been an informal, unplanned, bution is controlled by the so-called “water mafia,” and unregulated system of housing development. which takes water from various water hydrants and dis- From 1970 to 1985 the informal sector managed an tributes it by truck. Even high-income areas regularly estimated 33 percent of all residential land conversion have water delivered by tankers. A 1,200-gallon tank and development in the metropolitan region and met sells for Rs. 200 ($3.40), and the price per unit in- more than 50 percent of the city’s housing needs. Al- creases for water sold in smaller quantities to house- though Karachi needs an estimated 80,000 housing holds unable to afford the money or storage space for units each year, between 1987 and 1992 an average of larger amounts. Over time, low-income neighborhoods only 26,700 building permits were issued yearly. The acquire trunk water connections by lobbying their rep- informal sector created about 28,000 units each year in resentatives on the municipal council or by collecting      ⁄  money and bribing public officials. Alternatively, the The path to reform supply can be arranged by the land developers, who What institutional changes and arrangements would create illegal connections to the public piped network. yield the most favorable outcomes, given Karachi’s con- Communities often collect money and lay internal dition today? As a key first step, the government needs water supply networks at their own cost. to recognize that what exists on the ground is not a tem- Katchi abadis also arrange their own wastewater dis- porary situation, but a reality. The katchi abadis repre- posal. A survey of 136 katchi abadis in Karachi com- sent the starting point for thinking about Karachi’s fu- prising 79,426 houses and 8,479 lanes shows that these ture. These vast community investments in housing and communities have laid sewer lines in 82 percent of the infrastructure are part of the city’s future, and wiping lanes at an estimated investment of Rs. 200 million them out in order to start again from scratch is simply ($3.4 million). In Orangi township, 88,211 houses in not feasible. Thus, any housing plan the government 5,856 lanes have built their latrines, lane sewers, and puts forward must take the informal communities into over 400 collector sewers with an investment of Rs. 74 account. million ($1.5 million). At public sector rates, this con- The government must also nurture—and ultimately struction might well have cost nearly 10 times as much. institutionalize—positive interactions among govern- A massive informal sector is far from the optimal ap- ment agencies, interest groups (formal and informal), proach to housing shortages. Because households ob- and communities. Currently, there is little trust among tain their land through irregular channels, they do not the various actors in the housing drama, especially be- own their primary asset and cannot use it as collateral tween the government and organizations representing to finance housing construction. The insecurity of their low-income households. These groups have the most ac- property rights undermines what should be a natural curate knowledge of housing conditions and are well incentive to invest in property and infrastructure. positioned to articulate residents’ needs. Working with Economies of scale do not exist for the delivery of es- them will help ensure that housing priorities are met, sential services because services are provided piecemeal but the groups need access to good information if they (and sometimes illegally). Facilities are often of ques- are to function effectively. An additional method of tionable quality because the informal sector does not reestablishing trust is to rationalize the overlapping re- have the necessary technical capacity. And the illegal sponsibilities of municipal, provincial, and federal agen- dumping of wastes and inadequate treatment of sewage cies in order to strengthen accountability at each level. lead to increasingly dangerous health conditions. Sew- Overregulation of the housing market has resulted age remains a particular problem in the informal set- in Karachi’s current unworkable system of housing pro- tlements, which often discharge it into open natural vision. This system needs to be replaced by one that in- drains. Community-built sewerage systems are seldom corporates legitimate private construction firms into integrated into official sewerage system plans. If they the market for low-income residents. Standards for sub- were, the costs would be dramatically reduced, the proj- dividing and building, for example, should be made ects would be completed in a fraction of the time it more realistic. While housing must meet public health takes to complete them now, and the poor, not the con- and safety requirements, it need not be so elaborate that tractors, would be the beneficiaries. it unnecessarily raises the price of housing out of the Governments have so far been indifferent, if not hos- reach of low-income people. tile, to the katchi abadis even though they house half The public sector, for its part, should confine its role the city’s population. The rationale is that katchi abadis in the formal system of housing production to areas in are a transitory phenomenon. Formal plans and projects which it has a comparative advantage. The first of these ignore the existing investment in these communities on areas is property rights. The government should pursue the assumption that the government will ultimately title adjudication and improve the administration of provide high-quality standard solutions. Community- property registration systems. The second area is trunk based organizations and NGOs have pressed for a infrastructure. Karachi needs new water and sewer lines change in policy, but official responsibility for housing and trunk roads that will connect the tertiary networks is fragmented among overlapping city, provincial, and already existing in the katchi abadis to existing public federal agencies, making concrete action difficult.27 infrastructure. The third area is housing credit. The      government can improve housing prospects for low- also experiencing the fastest increase in its urban popu- income residents by allowing them to apply for credit lation. The prognosis for the continent, having bright- collectively. Community groups able to make an accept- ened briefly in 1995–97, once again looks uncertain, able down payment on land can be an important source however. For the typical, predominantly rural African of infrastructure development. Once they have acquired economy such as Tanzania, globalization and urbaniza- title to the land, they can use this asset as collateral for tion open a small window of hope (box 8.6). How can infrastructure loans. Tanzania exploit these forces to galvanize its rural econ- These three measures can reduce the cost of new and omy and make it an engine of growth for a country tenured housing with access to essential services. But the whose GDP is currently rising at 3–4 percent per year? government must also address the current problems of the katchi abadis, possibly by adopting the model of de- Initial conditions velopment offered by the Orangi Pilot Project described Three-quarters of all Tanzanians live in rural areas, and in chapter 7.28 This model reduces the cost of internal agriculture accounts for over 50 percent of the country’s development to about 10 percent of conventional plan- GDP. Most farming is traditional land-extensive, low- ning costs and makes maintenance and operation feasi- input, subsistence agriculture. Agricultural production ble. It is socially acceptable and can be upgraded over has increased in recent years, largely because farmers are time. The city can design future infrastructure that in- cultivating more land (yields are low and have stagnated corporates community-built facilities into the overall for almost three decades). Manufacturing contributes a network. It can also provide technical advice to the bare 7 percent of GDP, a share that has declined over the informal contractors, perhaps through certification past two decades as tariff barriers have been removed and processes to upgrade their skills. the public sector has withdrawn from some loss-making For water supply, the government could consider production activities. The main activities are food pro- formalizing the de facto privatization that has already cessing, and manufacture of building materials and paper occurred. Rather than attempting to extend its water and packaging, largely for the domestic market. Tanza- networks into informal settlements, the Karachi water nia’s exports consist of unprocessed agricultural products authority might be better off considering competitive and minerals and have diversified little since the mid- wholesale water concessions. Paraguay’s experience 1980s. Export crops, which are produced mostly by shows that when small private providers are allowed to smallholders, account for only around 8 percent of agri- operate competitively in a stable environment, truckers cultural production, although sales of cut flowers are will eventually find it in their economic interest to rising. Apart from these items and commodities like cof- progress to piped distribution (chapter 7). In the mean- fee, tea, cashews, maize, cotton, and fish, the main for- time, however, private providers can better tailor their eign exchange earner is tourism, a significant source of services to the socioeconomic and physical characteris- income for the country. Over the medium term, exports tics of the neighborhoods they serve. of gold could overshadow income from cash crops. Over time, these measures can transform Karachi’s In 1998 Tanzania attracted $140 million a year in housing market. As the cost of formal housing declines, foreign direct investment, as against $70 million in the the proportion of households relying on the informal mid-1990s, most of it going into mining and the bal- system of housing production will fall. And as govern- ance into infrastructure for tourism. The privatization ments take a more supportive approach to katchi of banks and utilities is beginning to draw funds into abadis, the number of households lacking secure tenure some other sectors, such as telecommunications. Do- and trunk infrastructure will decline as well. mestic savings and public sector resource mobilization are modest, as they are in most African countries. But Cultivating rural-urban synergies: Tanzania investment, financed in part by international aid, is Of all the developing regions, Sub-Saharan Africa has fairly high relative to GDP. registered the weakest overall growth in the last 15 years. Tanzania’s urban population is growing at about 5 The area has become increasingly marginalized in the percent annually—a rapid but not unusual rate, given global economy, and its debt burden as a share of GDP the country’s relatively low level of urbanization (figure is now the heaviest of any region. Sub-Saharan Africa is 8.1). The six largest cities generate over one-third of      ⁄  Box 8.6 Tanzania at a glance Sub- Low- Exports and imports Saharan income (millions of U.S. dollars) Poverty and social indicators a Tanzania Africa countries 2,500 Exports GNP per capita (U.S. dollars) b 210 500 350 2,000 Imports Poverty (percentage of population 1,500 below $1 per day) 51 .. .. 1,000 Urban population 500 (percentage of total population) 24 32 28 Life expectancy at birth (years) 51 52 59 0 c 1991 1992 1993 1994 1995 1996 1997 Key economic ratios 1976 1986 1996 1997 Gross domestic investment/GDP .. 22.0 18.0 19.5 Exports of goods and services/GDP .. 7.8 21.5 23.2 Gross domestic savings/GDP .. 9.9 3.4 12.6 1976–86 1987–97 1996 1997 1998–2002 Average annual GDP growth .. 2.9 4.1 3.9 5.5 . .Not available. a. Data shown are from the latest available year within the range 1991–97. GNP per capita figures are from 1997. b. Calculated using World Bank Atlas method. c. Import data for 1997 are not available. Source: World Bank, World Development Indicators, 1999. GDP, with Dar es Salaam accounting for 17 percent. If knowledge, expanding access to markets for agricultural Tanzania is to achieve and maintain growth rates of 7–8 produce, and harnessing new biological, chemical, and percent over the next two decades (as it must to make mechanical inputs. Tanzania can adapt these approaches substantial progress against poverty), more growth must by taking the following steps: come from its cities. However, given the preponderance Step 1. Establish support networks that create trusting of the rural sector, overall economic performance will relationships between urban businesses and periurban and depend upon the multiplying of rural-urban linkages, rural producers. With over 70 percent of its rural in- the commercialization of agriculture, and the spread of come dependent on agriculture, Tanzania has great nonfarm activities. Currently, agricultural diversification scope for developing rural industry.30 Furthermore, and productivity are at low levels, and nonfarm incomes with only one-third of agricultural output currently are below the average for the Sub-Saharan Africa region. marketed at all, rural-urban linkages have much to con- But such development is most likely to flourish in the tribute to agricultural development. Tanzanian farmers periurban areas and then spread to the hinterland, deriv- lack information, infrastructure, transportation, and ing impetus from markets and agglomeration economies credit because of the small size and subsistence orienta- in cities.29 tion of farming activities.31 In rural areas that are rela- tively close to cities, however, proximity to markets and Urban-rural partnerships information can help overcome these problems. Mar- How would an urban-rural partnership work? The ex- ket transactions must operate against a backdrop of as- perience of other low-income agricultural economies, surances that most of the time, obligations will be met, such as China, Indonesia, and Vietnam, suggests four bills will be paid, goods will be delivered, and trans- ways of improving the links between rural and urban actions costs will be manageable. Formal legal and in- areas that can help raise rural productivity. These in- surance contracts are one mechanism for providing clude employing new technical and organizational these assurances. But ties of ethnicity, religion, and kin-      Figure 8.1 system. Good roads are particularly needed in the im- The population in Tanzania mediate vicinity of the major urban areas in order to fa- is increasingly urbanized cilitate the integration of cities and the surrounding Urban population countryside. Only 12 percent of Tanzania’s roads are in (percentage of total population) good condition, with the rest imposing excess vehicle 30 operating costs equivalent to a third of export earnings in 1990.34 The implications for the development of mar- 25 keted crops are severe: in the past, when rising prices have resulted in a spurt in production, crops could not 20 be moved because of inadequate transportation. As a re- sult, farmers had difficulties selling their produce, and 15 production quickly fell back again.35 Improved transportation and communications are 10 important not only because they will strengthen ties within Tanzania but also because they will link the 5 country more closely with the global economy. To maintain close contact with foreign markets—and to 0 1960 1965 1970 1975 1980 1985 1990 1995 ship and receive goods on exacting schedules—Tanza- nia’s businesses need a well-managed infrastructure that Source: World Bank, World Development Indicators, 1999. keeps handling and user charges at seaports and airports to a minimum.36 The same is true for international tele- communications rates. For Tanzanian exporters to com- ship are a source of social capital and support flexible pete effectively against suppliers in South Africa and production arrangements, subcontracting, and out- South Asia, the country’s infrastructure must deliver sourcing schemes (see box 2). Such social networks comparable services and charge similar rates. This re- flourish in eastern Nigeria along an axis that includes quires the private sector to play a large role in building Aba, Nnewi, and Onitsha. In Nnewi, for example, and operating the transportation, communication, and members of the Igbo community have established a electricity infrastructure. motor parts industry that relies on ethnic ties to reduce Step 3. Improve channels for agricultural and indus- transactions costs.32 trial research and extension services to bring new technol- Intermediaries with strong rural connections, play a ogy to the rural economy. The diffusion of technology, by large part in building such networks. Much depends on private businesses, government research institutes, and periurban-urban social relationships and on the will- the media, is vital to raising Tanzania’s agricultural pro- ingness of urban business groups to reach out to the ductivity, incomes, and demand for nonfarm products. surrounding rural areas.33 In such cities as Arusha and It can also nurture processing and industrial activities Moshi, an entrenched business elite—the Chaggas— in the belts around cities, where the potential returns may already possess a local network, along with ade- from these activities are highest and most visible.37 quate financing, that could serve as a basis for expan- Specialized agricultural extension services managed sion. Likewise, Asian communities in Dar es Salaam, from the top down have not worked well in East Lindi, and elsewhere could enlarge their marketing net- Africa.38 But experience in other countries suggests that works among the periurban villages. But a deepening extension services can be made more effective. They of formal institutions that safeguard rights would sup- must be client driven, customized to the needs of par- plement informal arrangements. ticular groups, and capable of delivering the newest Step 2. Build infrastructure. A modern economy de- technologies.39 They must operate in areas with sound pends on efficient surface transportation and telecom- infrastructure, especially roads and electricity, ready munications, which link rural producers, service pro- access to modern agricultural inputs, and good market viders such as freight transporters and marketing firms, access, for example, in the vicinity of cities. These and urban businesses. In transportation, the govern- efforts must also be directed to those groups that are ment’s imperative is to appreciably strengthen the road most likely to be innovative—that is, to groups that      ⁄  have the educational level to exploit the opportunities The entrepreneurship, knowledge, and capital from new technologies present. this reverse migration have helped these countries find Providing effective research and extension services new overseas markets. But reverse migration is at best a requires an understanding of the rural economy around partial solution. Tanzania must expand its secondary cities, where the incentive to innovate is greatest. The education facilities and rebuild its tertiary education services can then be oriented to the emerging periurban and research facilities. Tanzanian universities have been commercial agriculture with industrial linkages and can drained of talent and barely partake in the international evolve with the changing economic environment. Try- commerce in ideas and research.44 Strengthening the ing to provide such services everywhere in the country scientific culture and the competitiveness of universi- will probably have a negligible payoff. But concentrat- ties are necessary steps to rapid and sustainable devel- ing what capacity there is in areas where the scope for opment in a globalizing environment where technol- rural-urban synergy is greatest might create the dynamic ogy is one of the principal drivers.45 For example, the clusters of growth that Tanzania so badly needs. promise of transgenic technology to enhance the yield Step 4. Exploit the advantages of global business and in- and disease resistance of staple crops, such as maize, will tellectual linkages. Tanzania does have an indigenous only be realized by building the research base and business community connected to the Middle East, biosafety regulatory capacity and by enforcing rules to Europe, and South Asia (see chapter 1). But a history of protect breeders’ rights. Without these, there is limited government constraints on private sector activity has prospect that the country will participate as an in- channeled much of the energies of this community into formed stakeholder in the biotechnology business.46 trade, wholesaling, and retailing activities.40 Whether Laying the foundation for this new strategy requires the objective is to encourage local businesses to expand political initiative. Tanzania’s leaders must change the and diversify or to make Tanzania more attractive to for- climate of opinion in the country, building a consen- eign investors, effective constitutional and legal rules are sus among local and foreign businesses and following needed to safeguard property rights, enforce contracts, up their views with credible institutions. The govern- and reduce state interference.41 A free press can support ment can signal its commitment to change by more ac- these legal measures by acting as an agent of restraint tively pursuing privatization and transparent reforms and by helping enforce accountability of public as well in the banking industry. as private bodies. The basic framework exists in Tanza- Policies that support macroeconomic stability, liber- nia but lacks credibility in the eyes of investors, who re- alize the market, and build human capital should pro- cently ranked Tanzania as one of the world’s most cor- vide some of the conditions for future development rupt countries.42 in Tanzania. But Tanzania must do more in order to The progressive lowering of trade barriers will im- experience the kind of development that will substan- prove urban entrepreneurs’ access to the equipment, in- tially reduce poverty. The government needs to estab- puts, and technology they need to build competitive lish strong political and legal institutions that reduce business operations. But openness embraces more than the risk to local and foreign investors of doing business free trade. It also involves subscribing to rules govern- in the country. It must also invest in urban and periur- ing commercial codes, contracts, and individual rights. ban infrastructure, especially transportation and com- In this deeper sense, openness can reinforce the govern- munication. Finally, by fostering openness the govern- ment’s assurances to the business community on prop- ment can help create competitive markets, spread erty rights and contractual agreements. It can also stim- knowledge, and build human capital. ulate private domestic investment and increase the flow The shifting development landscape at the dawn of foreign direct investment in industry.43 of the 21st century A secure and open business environment is likely to lead to a reflux of skills from abroad and to generate the In 1990 many in the development community and incentives that encourage individuals to acquire tech- elsewhere expected the raw vitality of market capital- nical expertise. Several East Asian economies, after ex- ism to lift billions of people out of poverty into a new periencing a brain drain from the 1960s through the era of sustainable development. These expectations early 1980s, established open and flexible business en- have not been borne out. Some nations have achieved vironments that drew back many of those who had left. remarkable progress. But with nearly 1.5 billion people      living on less than $1 a day and more than 2 billion on The recognition that the health of a country’s bank- less than $2 a day, the task is very far from completed. ing system can alter investor perceptions of the health Development policy is being rethought. The World of neighboring countries’ banks has transformed the Bank’s Comprehensive Development Framework calculus of international cooperation. Yet, as chapter 3 (CDF), this report, and the 2000/2001 report on poverty pointed out, merely adopting common banking stan- alleviation strive for better understanding of all aspects of dards without adequately enforcing them is unlikely to development—drawing from experience and the forces restrain excessive risk-taking. Countries must not only that will shape the development landscape to provide adopt sound banking standards but must also devise in- pointers for the future course of development policy. stitutional structures to ensure that these standards are This reexamination has been wide ranging. It suggests enforced by insulating bank regulators from external that no one element of development is above all others; pressures. They also need to consider establishing a sys- no one future trend is all-encompassing; no one institu- tem of regional surveillance of banking practices, as is tional or policy initiative is likely to be a panacea. taking place in the Manila Framework. This report argues that two forces will significantly Some of the starkest examples of growing interde- alter the development landscape in the early decades of pendencies appear in the discussion of the global com- the 21st century, with implications for how the devel- mons in chapter 4. Although slow progress has been opment agenda can be tackled, who the relevant actors achieved in negotiating an agreement to substantially will be, and what forms their interactions are likely reduce greenhouse gas emissions, the growing recogni- to take: tion of the linkages among international environmen- tal problems suggests that better policies and new in- Ⅲ Globalization—the continued integration of the stitutions will be needed. world’s economies through expanding flows of goods, A second outcome of the forces of globalization and services, capital, labor, and ideas and through collec- localization is a more crowded development scene. Nation- tive action by countries to address global environ- states will less and less be the sole agents for develop- mental problems ment. Instead, countries will increasingly act through Ⅲ Localization—the increasing demands of local com- multinational agreements, and through their interac- munities for greater autonomy, which will be bol- tions with multinational corporations, nongovernmen- stered by the growing concentration of developing tal actors, and subnational entities, especially cities. countries’ populations in urban centers. Institutions and norms will have to evolve to define re- lations between these new actors and nation-states. This These forces interact in many ways. The urban cen- will have implications for the way global accords are ne- ters discussed in chapters 6 and 7 have much to gain gotiated, for how governments within a country man- from the open world trading system of chapter 2 and age central-local relations, and for how enduring part- the global capital flows of chapter 3. Foreign direct in- nerships are established within cities. vestment (chapter 3) will play a significant role in pro- In describing the challenge of localization, chapter 5 viding the needed urban services (chapter 7). Many offers suggestions for avoiding the pitfalls highlighted in more such connections have been drawn throughout the the case study of Brazil. Chapter 7 points to the impor- report. These disparate yet interrelated forces pose many tant role partnerships play in revitalizing cities and im- challenges for the development process, each requiring proving the quality of life for urban residents. Each of a different type of institutional and policy response. these developments requires new institutions that will ac- commodate the growing number of development actors. Three central implications for development policy Despite the new supranational and subnational chal- First, these forces underscore the growing global in- lenges, governments will remain central players in the terdependence across space, time, and issues. The rapid development process. National governments may well spread of financial contagion from East Asia to the undertake fewer functions, ceding responsibility to Russian Federation and Latin America in 1997–98 other entities. But they remain the linchpin that holds stands as a compelling testimonial to the growing in- the institutions of governance together. They alone can terdependencies that can undermine economic growth define the constitutional rules within their borders and and increase poverty. shape their relations with each other.      ⁄  The greater mobility of capital and labor between gary in this chapter, developing countries can use pre- and within countries—and the potential competition existing and accepted global standards to guide and sup- among national, subnational, and urban governments port the credibility of their reforms. for scarce resources—underlies the third implication of These institutions do not arise in a vacuum, and due these forces: the rewards for successful development strate- regard must be given to how rules are negotiated and gies, and the punishments for failure, are likely to be greater enforced. Whether the concern is global or local, far- and experienced more quickly than in the past. For exam- sighted policymakers have to induce the participation ple, urban centers that provide stable property rights of every actor with the capacity to enhance or reduce and an environment conducive to the accumulation of the collective welfare. No doubt some government en- social and human capital can attract more foreign in- tities will be tempted to “hold up” negotiations to press vestment and more skilled migrants. The consequences for greater benefits from such agreements. However, of delays and partial, half-hearted attempts at reform, such tactics are likely to prove less and less successful: giving little thought to building long-term credibility, growing interdependencies will create linkages across will become apparent much more immediately than in issues, and pariahs will be shut out of the benefits of the past, as discussed in chapters 2, 3, and 6. cooperation on many fronts. These institutions, once established, will evolve in The central role of institutions response to numerous factors: the changing needs of This report’s focus on the institutions of governance members, technological advances, growing or receding does not diminish the key role government policymak- consensus among experts, and pressures from non- ing plays in development. Fostering the administrative members. Such institutions will also have to be robust and analytical capacity to formulate, innovate, and im- enough to handle rapid changes in sentiment—abetted plement policies will remain essential to promoting de- by improvements in communications, disseminating velopment in the future. new information faster to greater numbers of con- However, government policies alone will not suffice. cerned parties. Responding to these new forces of globalization and lo- The last 10 years have been a mixed blessing for de- calization requires robust mediating institutions, espe- veloping countries. East Asian nations surrendered some cially when countries commit to take actions in a crisis, previous gains in a crisis with substantial human and eco- separately or collectively. Institutions serve to balance nomic costs. Large parts of Africa have had yet another the diverse interests of society and to determine how the lost decade. No one wants to see these experiences re- forces of development distribute their benefits and ad- peated. We have learned from the past, and we have a vantages, and their costs and risks. Fortunately, devel- better sense of the forces that will mold the development oping countries do not need to create all of these insti- landscape over the coming decades. Globalization and tutional structures from scratch; in many cases, they can localization are transforming many aspects of the human build upon existing international agreements and inter- experience—so many that only a comprehensive, multi- nationally recognized standards. Countries can use the layered response of policy and institutional reform will procedures of the WTO to enhance the credibility of be adequate. If we fail to meet this challenge, we will con- their unilateral trade reforms by binding reforms into demn the world’s poor to a cycle of instability, hunger, their multilateral commitments, as discussed in chapter and despair. By seizing the opportunities presented at 2. By moving toward international banking standards, the dawn of the 21st century, together we can turn our as discussed in chapter 3 and in the case study of Hun- dream into a reality—a world free of poverty. Bibliographical Note T his report has drawn on a wide range of Bahl, J. Michael Bamberger, Suman World Bank documents and on numer- Bery, Sofia Bettencourt, Amar Bhatta- ous outside sources. World Bank sources charya, Richard Bird, Ilona Blue, Clem- include ongoing research as well as ens Boonekamp, François Bourguignon, country economic, sector, and project Nicolette L. Bowyer, John Briscoe, Lynn work. These and other sources are listed R. Brown, Robin Burgess, Shahid Javed alphabetically by author and organiza- Burki, William Byrd, Jerry Caprio, Rich- tion in two groups: background papers ard Carey, Gonzalo Castro, Herman Ce- and references. The background papers sar, Ajay Chhibber, Kenneth Chomitz, were commissioned for this report and Paul Collier, Maureen Cropper, Angus are available on the World Development Deaton, Julia Devlin, Samir El Daher, Report 1999/2000 World Wide Web A. Charlotte de Fontaubert, Dipak Das- page (http://www.worldbank. org/wdr/). gupta, Alan Deardorff, Shantayanan In addition, some background papers Devarajan, Hinh Truong Dinh, Simeon will be made available through the Pol- Djankov, Gunnar Eskeland, François icy Research Working Paper series or as Falloux, Caroline Farah, Charles Fein- other World Bank publications, and the stein, J. Michael Finger, Louis Forget, rest will be available through the World Per Fredriksson, David Freestone, Lev Development Report office. The views Freinkman, Caroline Freund, Christo- they express are not necessarily those of pher Gibbs, Marcelo Giugale, Steve Gor- the World Bank or of this report. man, Vincent Gouarne, Elisea G. Gozun, In addition to the principal sources Emma Grant, Angela Griffin, Jeffrey listed, many persons, both inside and Gutman, Kirk Hamilton, Sonia Ham- outside the World Bank, provided advice mam, Trudy Harpham, Nigel Harris, and guidance. Valuable comments and Arif Hasan, Ian Graeme Heggie, Jesko S. contributions were provided by Richard Hentschel, Bernard Hoekman, Gordon Ackermann, James Adams, Ehtisham Hughes, David Hummels, Athar Hus- Ahmad, Junaid Ahmad, Tauheed Ah- sain, Zahid Hussain, Roumeen Islam, med, Jock Anderson, Lance Anelay, Emmanuel Jimenez, Ian Johnson, Bar- Preeti Arora, Jehan Arulpragasam, Roy bara Kafka, Ravi Kanbur, Kamran Khan,       ⁄  Anupam Khanna, Homi Kharas, Bona Kim, Daniela Theresa Moyo, Fidelis Mtatifikolo, Charles Mutalemwa, Klingebiel, Tufan Kolan, Mihaly Kopanyi, Nicholas Crispin Mwanyika, Mthuli Ncube, Dominique Njin- Kraft, Kathie Krumm, Donald Larson, Kyu Sik Lee, keu, Temitope Oshikoya, T. Ademole Oyejide, Hajji Danny Leipziger, Robert Litan, Peter Lloyd, Millard Semboja, Joseph Semboja, and Moshe Syrquin; in Long, Susan Loughhead, Patrick Low, Michael Lyons, London (NGOs): T. Allen, Tamsyn Barton, Teddy Brett, Dorsati Madani, Antonio Magalhaes, Catherine Mann, Ashvin Dayal, Nicholas Fenton, Alistair Fraser, Caro- Manuel Marino, Keith Maskus, Douglas Massey, line Harper, A. Hussain, Duncan McLaren, Richard Subodh Mathur, Alexander F. McCalla, Kathleen B. McNally, Claire Melamet, Brian O’Riordan, and David McCollom, Dominique van der Mensbrugghe, Patrick Woodward; in Paris: Scott Barrett, Jean Claude Ber- Messerlin, Jonathan Michie, Steven Miller, Pradeep thélemy, François Bourguignon, Richard Carey, Won Mitra, Gobind Nankani, Benno Ndulu, Vikram Nehru, Hho Cho, Daniel Cohen, Jean-Marie Cour, A. De Eric Neumayer, Kenneth Newcombe, Ian Newport, Palma, Zdenek Drabek, Patrice J. Dufour, Richard Francis Ng, Judy O’Connor, W. Paatii Ofosu-Amaah, Eglin, John Hawkins, Pierre Jacquet, Ad Koekkoek, Alexandra Ortiz, Tracey Osborne, Kyran O’Sullivan, Patrick Messerlin, Robert F. Owen, Pier Carlo Padoan, Samuel K. E. Otoo, Berk Ozler, John Page, Stefano Hubert Prevot, Rémy Prud’homme, Thérèse Pujolle, Pagiola, Eul Yong Park, Antonio Parra, Odil Tunali Regine Qualman, Helmut Reisen, Karl Sauvant, Paul Payton, Guy Pfeffermann, Rachel Phillipson, Robert Spray, Rolph Van Der Hoeven, Daniel Voizot, and Picciotto, Gunars Platais, Lant Pritchett, Tom Prusa, Soogil Young; in São Paulo: Suman Bery, Shahid Javed Rudolf V. Van Puymbroeck, Regine Qualmann, Navaid Burki, Claudia Dutra, Javier Fraga, Carlos Langoni, Qureshi, S. K. Rao, Martin Ravallion, Gordon Redding, Marcos Mendes, Gobind Nankani, Fernando Rezende, Don Reisman, J. David Richardson, Gabriel Roth, Arun Richard Webb, Jorge Wilheim, and Juan Zapata; in Sanghvi, Barbara Santos, Karl Sauvant, Norbert Schady, Singapore: Shankar Acharya, Kym Anderson, Masahisa Sergio Schmulker, Jitendra J. Shah, Zmarak Shalizi, Fujita, Utis Kaothien, Ya-Yeow Kueh, Rajiv Lall, Howard Shatz, Martin Slough, Stephen Smith, Richard Rakesh Mohan, Romeo Ocampo, Phang Sock Yong, Stren, Maria Stuttaford, Hiroaki Suzuki, Hans Helmut Mohd. Haflah Piei, Peter J. Rimmer, David Satterth- Taake, Vito Tanzi, Stanley Taylor, Brigida Tuason, waite, Guo Shuqing, Victor Sit, Lyn Squire, Augustine P. Zafiris Tzannatos, Anthony Venables, David Vines, Tan, Douglas Webster, John Wong, and Chia Siow Yue; Michael Walton, Robert Watson, Dana Weist, George in Tokyo: Kengo Akizuki, Masahiko Aoki, Judith Ban- T. West, Matthew Westfall, Debbie Wetzel, John Whal- ister, Sang-Chuel Choe, Ryo Fujikura, Yukiko Fuka- ley, David Wheeler, Alex Wilks, Fahrettin Yagci, Kei-Mu gawa, Shun’Ichi Furukawa, Fan Gang, Kazumi Goto, Yi, Roberto Zagha, Akbar Zaidi, Anders Zeijlon, and Naomi Hara, Yujiro Hayami, Akiyoshi Horiuchi, Shi- Heng-Fu Zou. geru Ishikawa, Yoshitsugu Kanemoto, Hisakazu Kato, Other valuable assistance was provided by Nobuko Tetsuo Kidokoro, Fukunari Kimura, Naohiro Kitano, Aoki, Pansy Chintha, Meta de Coquereaumont, Kate Fu-Chen Lo, Katsuji Nakagane, Shuzo Nakamura, Kat- Hull, Keiko Itoh, Mika Iwasaki, Mani Jandu, Polly sutoshi Ohta, Takashi Onishi, M. G. Quibria, John M. Means, Boris Pleskovic, Jean Gray Ponchamni, F. Halsey Quigley, Kunio Saito, Yuji Suzuki, Kazuo Takahashi, Rogers, Bruce Ross-Larson, and Tomoko Hagimoto. Junichi Yamada, Toru Yanagihara, and Yue-Man Yeung; A wide range of consultations was undertaken for NGOs in Tokyo: Takeo Asakura, Kazuko Aso, Ienari this report. We wish to thank Scott Barrett, Chia Siow Dan, Francisco P. Flores, Yoko Kitazawa, Satoru Mat- Yue, David Currie, Patrice Dufour, Riccardo Faini, sumoto, Setsuko Matsumoto, Sakoe Saito, Kiyotake Carlos Fortin, Norman Gall, Morris Goldstein, E. Takahashi, Michiko Takahashi, Kunio Takaso, and Monty Graham, Katherine Hagen, Andrew Hughes Yoshiko Wakayamwa. Hallett, Gudrun Kochendorfer, Patrick Low, Martin The team consulted with the Asian Development Mayer, Andrew Rogerson, Jagdish Saigal, Robert Bank (ADB), the Bank for International Settlements Skidelsky, and Alfredo Sfeir-Younis; in Dar es Salaam: (BIS), the European Bank for Reconstruction and De- James Adams, Haidiri Amani, Patrick Asea, Melvin velopment (EBRD), the Organisation for Economic Ayogu, Enos Bukuku, Sumana Dhar, Augustin F. Fosu, Co-operation and Development (OECD), the Interna- T. Ayme Gogue, Mats Harsmar, Satu Kahkonen, Brian tional Labour Organization (ILO), the United Nations Kahn, Louis Kasekende, Frederick Kilby, A. K. Maziku, Conference on Trade and Development (UNCTAD),    the World Trade Organization (WTO), the United 32. Thomas 1999. Nations Development Programme (UNDP), the Inter- 33. World Bank 1999a; Dollar 1999. 34. North 1997. national Monetary Fund (IMF), the German Develop- 35. A complete political economy model needs to include ment Forum, the Commonwealth Secretariat, the U.K. an agent with the authority to establish or alter the rules under Department for International Development (DFID), which any game is played. In the national context the govern- the Overseas Economic Cooperation Fund (Japan), the ment is such an agent. It is neither a benevolent despot maxi- Foundation for Advanced Studies on International De- mizing a societal welfare function nor a neutral umpire, but a velopment (Japan), and the Institute of Southeast Asian privileged, self-interested agent in the game. Moreover, govern- ment consists of layers of agents who are responsible to each Studies (Singapore). other and to outside constituencies in differing ways. For a gen- Endnotes eral treatment of rule makers, see Altaf (1983). 36. The examples in this and the following four paragraphs Introduction are drawn from the last two World Development Reports (World 1. Pritchett 1997. Bank 1998m; World Bank 1997d), unless otherwise noted. 2. Among some of the more widely cited papers on why and 37. Burki and Perry 1998. how convergence is not happening are Bernard and Durlauf 38. Burki and Perry 1998. 1996; Easterly and Levine 1997; Pritchett 1997, 1998; Quah 39. Ravallion and Chen 1998. 1993; and Sachs and Warner 1997b. 40. UNDP 1998. 3. Some papers that find strong evidence of conditional con- 41. Ravallion and Chen 1998. vergence are: Barro 1991; Mankiw, Romer, and Weil 1992; 42. World Bank 1998. Sachs 1996; and Sala-i-Martin 1997. Caselli, Esquivel and 43. WHO 1999. Lefort (1996) suggest the convergence rate to country-specific 44. Jamison 1993. steady states could be even faster than the cross-country rate of 45. See Ridley (1997) on future disease threats in an urban two percent per year. environment. 4. See Aziz and Wescott (1997) on the need for macro pol- 46. Walt 1998. icy complimentarity and Stiglitz (1998a) on the need for a 47. Fredland (1998) explores the wider political and psycho- broader approach involving a range of elements. logical consequences of AIDS for the course of development in 5. Lewis 1955. Sub-Saharan Africa. 6. Stiglitz 1998b. 48. WHO 1999; Marsh 1998. 7. Devarajan, Easterly, and Pack 1999. 49. WHO 1999. 8. Levine and Renelt 1992. 9. Easterly and Fischer 1995. 50. There are now between 50–100 million cases of dengue 10. Psacharopoulos 1994. fever worldwide (Rigau-Perez and others 1998). See Howson, 11. World Bank 1991. Fineberg, and Bloom (1998). 12. Buckley 1999. 51. Mortality tends to be higher in HIV-positive patients in- 13. Stiglitz 1999b. fected by the tuberculin bacillus (Del Ano and others 1999). On 14. Stiglitz 1996. South Africa see Millard (1998). 15. Stiglitz and Uy 1996. 52. Although some 70 percent of the nearly 47 million cases 16. Evans and Bataille 1997; Isham, Narayan, and Pritchett of HIV are in Africa, the disease is also spreading rapidly in South 1994; World Bank 1991, 1997d. and Southeast Asia with the poor being most affected. See Tibai- 17. Buckley 1999. juka 1997; Lancet 1996; New England Journal of Medicine 1996; 18. Evans and Battaile 1997. Financial Times, “Toll from AIDS Heaviest Among the World’s 19. This box is drawn largely from the World Bank’s Poor.” June 24, 1998; Caron 1999; Financial Times, “AIDS May Social Capital Web Page (http://www.worldbank.org/poverty/ Kill Half South African Youth.” October 10/11 1998. The in- scapital/index.htm). tergenerational effects of AIDS are described starkly in UNESCO 20. Coleman and Hoffer 1987; Francis and others 1998. (1999), which notes that by 2000, 13 million children in Sub- 21. Drèze and Sen 1995. Saharan Africa will have lost one or both parents to the disease. 22. Narayan and Pritchett 1997. 53. See The Economist (1999f) and AIDS Analysis (1998) on 23. Portes and Landolt 1996. loss of life expectancy in several African countries. 24. Evans and Bataille 1997. 25. Buckley 1999. 54. This is proving to be an exceedingly by complex task with 26. WHO 1999. many leads being pursued. See The Economist 1998b; Financial 27. UNDP 1998. Times, “Simple to Identify, Difficult to Destroy.” July 16, 1998. 28. WHO 1999. 55. Good 1999; Business Week 1998. 29. World Bank 1999k. 56. New England Journal of Medicine 1997. 30. WHO 1999. 57. Nearly two thirds of all infections caused by staphylocco- 31. Easterly 1999. cus aureus in Europe are methicillin-resistant and also resistant      ⁄  to most other antibiotics. Howson, Fineberg, and Bloom (1998); $3.2 billion to $51.5 billion (Financial Times, “Bank Loans Cut Lancet 1998; New England Journal of Medicine 1998; Cohen to Emerging Economies.” May 31, 1999; The Economist 1999e). 1992. Walsh (1999) describes the attempt to modify vancomycin 7. The cost of a three-minute transatlantic call dropped from to cope with new superbugs. See also Business Week (1999). $31.58 in 1970 to less than $1 in 1998. Computer use is in- 58. Oxford Analytica, “Africa: HIV/AIDS Concentration.” creasing at a dramatic pace as more and more people gain access December 1, 1998. TB is not just a serious problem for low- to the Internet. Access speeds themselves have risen from 14.4 income countries and is also spreading in middle-income coun- kilobytes to 10 megabytes per second. Financial Times, “Bank- tries such as Russia, see Feschbach (1999); Farmer (1999). er’s Black Hole.” July 21, 1999. 59. In Southeast Asia alone, 10–15 million people cross 8. Feldstein 1998. borders each year. Oxford Analytica, “Southeast Asia: Spreading 9. Eberstadt 1998. Diseases.” July 15, 1998; Guerrant 1998. 10. Higgins and Williamson 1997; Horioka 1990; Kosai, 60. The Economist 1998a. Saito, and Yashiro 1998. 61. Curtis and Kanki 1998. 11. Average fertility rate in Mexico has fallen from 7.0 in the 62. Ambio 1995; The New York Times, “Malaria, A Swamp mid-1960s to 2.5 in the late 1990s (The New York Times, “Smaller Dweller Finds a Hillier Home.” July 21, 1999. Families to Bring Big Change in Mexico.” June 8, 1999). 63. See John and others (1998) on surveillance procedures; 12. Attanasio and Banks 1998. WHO 1999. 13. Giddens 1998. 64. Harvard Working Group 1994; Ewald and Cochran 14. Costa 1998. 1999. On the infectious causes of many malignancies, includ- 15. Peterson 1999. ing cancers, see Parsonnet (1999). 16. Deaton 1998. 65. Jamison, Frank, and Kaul 1998; Walt 1998; Howson, 17. Vamvakidis and Wacziarg 1998. Fineberg, and Bloom 1998; The Economist 1998c. 18. UNCTAD 1998. 66. Watson and others 1998. 19. OECD 1998. Much of the foreign direct investment in 67. This box is based on Gardner-Outlaw and Engelman developing countries goes to fewer than 20 East Asian and Latin (1997); Rogers and Lydon (1994); Seckler and others (1998); American economies (Fry 1995). and “World Day for Water, 22 March 1999,” a news release 20. Until the early 1990s, reducing financial repression was issued jointly by UNEP and United Nations University. viewed as a way to support growth but was not considered as 68. Peterson 1999. important as other factors. New research is changing that per- 69. Commentators sometimes downplay the role of technol- ception. Financial deepening, including the development of ogy in globalization, pointing out the limited extent of global well-functioning stock markets, is now seen as contributing sub- communications just a hundred years ago. In fact, U.S. exports stantially to future growth, principally by improving the alloca- in the late 1900s are only 1 percent higher as a percentage of tion of resources. The relationship to growth is stronger in some GDP than they were in the late 1800s, and international capital regions than in others, and the evidence quite plausibly suggests movements are a smaller percentage of output than they were in that well-regulated banking systems facilitate growth. The avail- the 1880s (International Herald Tribune, May 23, 1999). But ability of equity financing and venture capital closely parallels the absolute levels are clearly much larger. Trade has a much industrial progress and is especially important to small and broader base that involves a far larger percentage of manufac- medium-size enterprises seeking to exploit new technologies and tures and services, and finance includes more short-term invest- to industries that depend heavily on external funding. Of course ment that relies on highly sophisticated information technology. the causation could run both ways (Fry 1995). Data for five in- dustrial countries between 1870 and 1929 show that financial Chapter 1 intermediation spurs output. These results buttress the findings 1. Analyses of the recent growth in trade flows (Baldwin and for more recent periods, as well as Gerschenkron’s, which were Martin 1999) and capital flows (Bordo, Eichengreen, and Kim based on his historical studies of European countries (Ger- 1998; Obstfeld and Taylor 1999) have identified factors that schenkron 1962; Gregorio and Guidotti 1995; Levine 1997, were also present during the period of globalization before 707; Levine and Zervos 1998c; Rajan and Zingales 1998; World War I and, more important, factors that are unique to Rousseau and Wachtel 1998). the end of the 20th century. 21. Fry 1995. 2. Recent work by Frankel and Romer (1999) brings out 22. Levine 1997. strongly the relationship between trade and income growth. In 23. Flannery 1998; Knight 1998. fact, their results suggest that a 1 percentage point increase in 24. Dobson and Jacquet 1998. the trade-to-GDP ratio raises income per capita by between 0.5 25. Eichengreen and others 1999; McKinnon 1991. and 2 percent. 26. Goldstein 1998. 3. Anderson 1999. 27. UNDP 1999. 4. The Economist 1999d; Oxford Analytica, “East Asia, Elec- 28. Bennell 1997. tronic Commerce.” June 1, 1999. 29. Far Eastern Economic Review 1998; Urata 1996. 5. Kleinknecht and der Wengel 1998. 30. Foreign direct investment accounted for 14 percent of 6. Bank lending to developing countries dropped by $75 bil- financing for fixed assets in China between 1979 and 1997, out lion in 1998, but official development assistance increased by of a total of $220 billion (Guo 1998).    31. Prahalad and Liebenthal 1998. 10 times the global average (Financial Times, “Stormy Forecast.” 32. The number of interfirm technology arrangements re- June 3, 1999). corded by UNCTAD rose from an annual average of 300 in the 57. Apart from methyl chloride and methyl bromide, whose mid-1980s to 600 in the mid-1990s. Those involving firms concentrations have remained roughly constant, the rising bur- from developing countries have quadrupled from 10 per year to den of chlorofluorocarbons is entirely traceable to human activ- 40 (UNCTAD 1998; Kobrin 1997). ity and began rising in the early 20th century (Butler 1999). The 33. McKinnon 1998. relationship between atmospheric gases and warming was first 34. In its attempt to upgrade the 1988 accord, the Basle discerned by the great French mathematician Jean-Baptiste Committee has proposed a number of changes, including the Fourier in an article published in 1824 (Christianson 1999). use of both ratings by external agencies to determine banks’ cap- 58. While higher carbon dioxide concentrations could en- ital ratios and internally defined credit ratings of certain banks hance plant growth and increase efficiency of water use, changes (Financial Times, “Radical Banking Reforms Announced.” June in tissue chemistry will render some plants less palatable. Heat 4, 1999). and water stress on vegetation will offset some of the gains from 35. Between 1990 and 1997 the United States alone admit- an increase in the concentration of carbon dioxide (Science ted close to a million immigrants each year. See Population Ref- 1997, 496). erence Bureau (1999). 59. One alarming possibility is the melting and eventual dis- 36. Oceania encompasses Australia, New Zealand, and the appearance of Himalayan glaciers in about 40 years, leading ini- South Pacific islands. tially to flooding and then to the drying up of the rivers feeding 37. Zlotnik 1998. the Indian plains (New Scientist 1999). Although current re- 38. Kane 1995. search suggests that climate change should not significantly dis- 39. U.S. Committee for Refugees 1996. rupt the U.S. economy or agriculture, the impact on some re- 40. Cohen and Deng 1998. gions of the country could be quite severe (Lewandrowski and Schimmelpfennig 1999). 41. For a recent examination of European migration poli- 60. The flooding in Bangladesh, Central America, and cies see Faini (1998). His analysis of wage differences and de- China in 1998 prompted Argentina and Kazakhstan to call for mographic pressures suggests that West European countries are a cap on developing countries’ greenhouse gas emissions that more likely to see migration from North Africa than from East- would be incorporated into the Kyoto Agreement at the Buenos ern Europe. See also Dervis and Shafiq (1998). For a recent ex- Aires meetings in November 1998. ploration of the impact of demographic changes on migration, 61. Sims (1999) describes the regulatory problems con- see Teitelbaum and Winter (1998). fronting China. 42. Borjas 1998. 62. Flavin 1997. 43. Bohning and de Beiji (1995) examine the effects of poli- 63. Prescott-Allen 1995. cies that attempt to facilitate the entry of migrant workers into 64. Madeley 1995a. labor markets. Faini (1998) assesses the difficulties faced by mi- 65. Speech at the World Economic Forum 1999, quoted in grants as they assimilate into European countries. As they do, the New York Times, February 7, 1999. intense ethnic rivalries can develop that have little to do with 66. Hay and Shleifer 1998. the level of unemployment or the degree to which migrants dis- 67. Root 1998. place existing workers in service or low-skilled jobs. 68. Garrett 1998; Rodrik 1998b. 44. Carrington and Detragiache 1998. 69. Alesina 1998. 45. From IMF, International Financial Statistics, cited in 70. Boniface 1998. However, Pegg (1999) argues that the Russell and Teitelbaum (1992) and Taylor and others (1996). likelihood of new states appearing is low. Regions within coun- 46. See the evidence assembled on this point in Taylor and tries prefer autonomy to secession, and, as in the case of Soma- others (1996). In addition, see the evidence on remittances from liland, external recognition of sovereignty is granted reluctantly. Asian migrants to the Gulf States in Amjad (1989). 71. Panizza (1999) finds that fiscal decentralization is posi- 47. Castles 1998. tively correlated with ethnic fractionalization and level of 48. The influx of immigrants from Russia to Israel in the first democracy (as well as with country size and per capita income). half of the 1990s encouraged the growth of moderately skill- 72. IMF 1997, 1998c. intensive sectors. See Gandal, Hanson, and Slaughter (1999). 73. Wetzel and Dunn 1998. 49. Greif 1998. 74. Gavin and Perotti 1997. 50. Redding 1998; Skeldon 1998. 75. Alesina and Spolaore 1997. The end of the Cold War 51. Redding 1998. and the social history of these countries largely explain the ap- 52. Meyer and others 1997. pearance of 22 new countries between 1991 and 1998. 53. The New York Times, “Human Influences on Climate 76. Wiseman 1997. Are Becoming Clearer.” June 29, 1999. 77. Research suggests that the absence of any significant de- 54. Nature 1998. volution of authority is related in part to the quality of gover- 55. Science 1999. nance at different levels (Huther and Shah 1998). 56. Around this long-run average there is likely to be a lot 78. Shah 1997. of variation. Already over the past 30 years the average winter- 79. The future course of decentralization will depend upon time temperature in northern latitudes is 4–5°C higher, up to the experiences with ongoing experimentation in countries with      ⁄  both centralized and more diffuse political regimes (Willis, Gar- 101. Fujikura 1998. In order to act on environmental con- man, and Hoggard, 1999). cerns, the community needs the support of laws and govern- 80. Tyler 1997. ment regulations that provide access to information on pollu- 81. A fairly typical study of a sample of municipalities in tion. Without such access, individuals, groups, and NGOs are Latin America indicates that building the capacity of local orga- handicapped. Even in the European Union and Japan, obtain- nizations is a function of leadership and community participa- ing such information is often difficult, and the situation is much tion (Fiszbein 1997). worse in developing countries (New Scientist 1998). 82. Verdier 1998. 83. Brockerhoff and Brennan 1998. Chapter 2 84. Khilnani 1997. 1. Drabek and Laird 1998. 85. Lloyd-Sherlock 1997. Although a current and reliable 2. This is not to suggest that the WTO is the only interna- global headcount of the urban poor is unavailable for develop- tional institution committed to facilitating the expansion of in- ing countries, scattered evidence suggests that the number is sig- ternational trade. The International Monetary Fund and the nificantly higher than the estimated 300 million for 1988 (Had- World Bank share these goals and have designed programs to dad, Ruel, and Garrett 1998). achieve them. These institutions are also taking steps to pro- 86. Bourguignon 1998; Fajnzylber, Lederman, and Louyza mote interagency coordination, including the formulation of a 1998. “coherence” approach to policymaking. The foundations of this 87. Tonry 1997. approach are laid out in the “Report of the Managing Director 88. Caldeira 1996. of the International Monetary Fund, the President of the World 89. Szreter 1997. Bank, and the Director General of the World Trade Organiza- 90. World Bank 1979. tion on Coherence,” October 2, 1998. 91. Begg 1999. Globalization and the advances in informa- 3. Srinivasan 1998. François, McDonald, and Nordström tion technology sharpen the competition between cities. (1996) found that the static impact of the Uruguay Round 92. Krugman (1998a) notes the neglect of spatial issues in on developing countries raised national incomes by 0.3 per- economics texts. However, geographers have not neglected spa- cent. Harrison, Rutherford, and Tarr (1996) found gains of tial issues and locational decisions. And during the 1960s and 0.38 percent. 1990s, economists made notable contributions to the field of 4. Foroutan 1996; Harrison 1994; Krishna and Mitra 1998; urban development. But recent advances in economic model- Levinsohn 1993. In contrast to evidence cited here, the impact ling have raised the profile of urban economics and its impor- of trade on firms’ performance through economies of scale, ex- tance to economists more generally (Boddy 1999; Martin 1999). ternal economies of scale, and learning by doing is relatively 93. Scott 1976. Elster (1989) strikes a similar note when he weak (see Tybout 1998). discusses how social norms might be weakened in modern soci- 5. Feenstra and others 1997. ety because of mobility, the ephemeral nature of interaction, 6. Aw and Batra 1998; Clerides, Lach, and Tybout 1998. and the pace of change. Although the public provision of safety 7. See chapter 6; Fujita, Krugman, and Venables 1999; nets for the poor to replace informal kinship- or patron-based Glaeser 1998; Puga 1998; Venables 1998. insurance schemes has been widely discussed, creating viable 8. Quigley 1998. schemes is and will remain a considerable challenge. 9. Bolbol (1999), among others, points to the benefits for 94. Crystal 1997. Arab countries of joining the rule-based WTO. 95. Satterthwaite 1996. Curitiba, Brazil, is a famous exam- 10. As of February 10, 1999. See the World Trade Organi- ple of efficient transport planning and land use. But the virtual zation’s internet site (www.wto.org) for the latest information absence of other success stories is testimony to the difficulties on membership. confronting municipal regulatory agencies. Prudhomme and 11. Kleinknecht and der Wengel 1998. Lee (1998), show that urban sprawl and the speed with which 12. See Frankel (1997). This result must be interpreted care- trips can be made strongly affect the nature of the labor market. fully, as increased interregional trade flows may have encour- 96. An analysis of urban economic growth in Australia, mea- aged policymakers to sign the RTA, in turn generating more in- sured by the increase in the price of labor, finds that growth is traregional trade. The very fact that establishing causality is dependent on the quality of life, as reflected by the availability difficult gives some credence to the notion that RTAs are asso- of community services, local administration, public amenities, ciated with greater interregional trade flows. the degree of congestion, and the initial level of human capital 13. Rodrik 1994. (Bradley and Gans 1998). 14. The nation named in the complaint can appeal the panel’s 97. ILO 1998. See also Porter (1998). decision, and the case moves to an appellate body for review. 98. Mani 1996. 15. Baldwin and Venables 1995. 99. Despite two decades of decentralization, urban authori- 16. See Finger and Schuknecht (1999) for evidence on the ties have only a meager amount to invest. Most are still largely minuscule amount of textile liberalization that has occurred dependent on national (or higher levels of subnational) govern- since the end of the Uruguay Round. ments and international agencies (Satterthwaite 1996). 17. As could happen with measures justified on phytosani- 100. Cohen 1998; Kremer 1993; ILO 1998. tary grounds. See Hertel, Bach, Dimaranan, and Martin (1996);    Hertel, Martin, Yanagishima, and Dimaranan (1996); Krueger 36. Graham 1996; Rodrik 1997. (1998); Srinivasan (1998); Thomas and Whalley (1998); and 37. Maskus 1997. Trela (1998). 38. Countries are attaching increasing importance to the dis- 18. Rodrik 1994. pute settlement mechanisms. The substantial attention given to 19. Usually this compensation is in the form of enhanced international trade disputes over imports of genetically modi- access to other markets in the same economy, a practice that is fied foods, bananas, and beef into the European Union and likely to have detrimental effects on import-competing firms in magazine imports into Canada highlights this attention. those markets. Anticipating such compensation, these firms are 39. Ostry 1997, 1998. likely to oppose businesses that favor reversing the reforms in 40. Anderson (1999) provides a succinct account of these the first place. And if a country reverses its reforms without of- and other issues for trade reform. fering compensation, it may well find itself a defendant in a 41. Ryan (1998) provides an in-depth analysis of the intel- complaint brought under the WTO’s dispute settlement mech- lectual property rights issue too. anism. Such a flagrant breach of WTO obligations is likely to 42. Hoekman and Anderson 1999. result in sanctions against the offender, often in the form of re- 43. Josling 1998a, 1998b. duced market access. Again, foreseeing this chain of events, the 44. Laird 1997. exporters will apply pressure not to reverse the reforms in the 45. Josling 1998a. first place. In both cases, binding lower trade barriers into the 46. Hoekman and Anderson 1999; Ingco and Ng 1998. country’s WTO obligations sharpens the incentives of domestic 47. World Bank 1998o. parties who have an interest in preserving the improved access 48. James and Anderson 1998; Roberts and DeRemer 1997. to domestic markets and signals to the private sector that the 49. However, this agreement has already come under pres- trade reforms are there to stay. sure with disagreements between the European Union and the 20. Finger and Winters 1998. United States over genetically modified crops and food, in par- 21. Fung and Ng 1998. ticular beef. Some have gone so far as to advocate banning in- 22. Keesing 1998. ternational trade in genetically modified crops, in direct con- 23. Michalopoulos 1999; Short 1999. travention of WTO rules. See Financial Times, “Genetically 24. For a careful study of the differential effects of trade Modified Trade Wars.” February 18, 1999, and The Economist flows on the concentrations of a number of pollutants, see (1999a). Antweiler, Copeland, and Taylor (1998). See also box 4.7. 50. Kerr (1999) assesses the prospects for future agricultural 25. Feenstra 1998. trade disputes, pointing to the increased burdens that they will 26. EBRD 1998. place on the WTO’s Dispute Settlement Mechanism. 27. These laws are sanctioned by a WTO agreement (Jack- 51. For a recent analysis of the effects of introducing more son 1997, 1998; Financial Times, “Developing World Leads in competition into the maritime sector, see François and Wooton Anti-dumping.” October 29, 1998). (1999). 28. For a series of case studies of the detrimental effects of 52. Deardorff and others 1998. This potential development antidumping actions see Finger (1993) and Lawrence (1998). reinforces the argument that Western firms can expect to face The recent surge in antidumping cases in Europe and the growing competition from developing countries in most areas United States against Asian steel producers highlights the effect of international trade. antidumping laws can have on market access. As Asian coun- 53. Cairncross 1997. tries try to stabilize their output levels, they find themselves 54. Hoekman and Djankov 1997b. faced with export restrictions that hinder their efforts. Tharakan 55. In addition to the traditional disciplines of most favored (1999) presents proposals for the reform of antidumping laws, nation treatment and national treatment, there were disciplines while Horlick and Sugarman (1999) offer proposals to reform on market access that in effect prohibited nations from using six the application of these laws to “nonmarket” economies. particular restrictions on service suppliers. See Hoekman and 29. Burtless and others 1998; Cohen, D. 1998; The Econo- Primo Braga (1997) for details. mist, 1999b; Hufbauer and Kotschwar 1998; Rodrik 1997, 56. Hoekman 1996. 1998a; Williamson 1998. 57. Krueger 1998. 30. For a careful survey of this debate see Cline (1997). An- 58. Negotiations in the telecommunications and financial derson and Brenton (1998) offer a more recent analysis of the service sectors have been completed since the signing of the effect of trade and technology on income inequality in the Uruguay Round. While useful, sector-specific negotiations only United States. permit countries to make trade-offs of commitments within 31. Aghion and Williamson (1998) provide a conceptual each sector rather than across sectors, where the differences in and empirical analysis of the effect of globalization on income national cost levels and the gains from liberalization are likely inequality and growth. to be greater. See WTO (1998a) for an account of the financial 32. Prusa 1997. services agreement concluded in December 1997. 33. For a recent survey of the costs of adjusting to trade re- 59. For a qualitative discussion of the importance of these form see Matusz and Tarr (1998) and UNDP (1999). networks for trade see Krugman (1995), and Feenstra (1998). 34. Lawrence 1996. Hummels, Ishii, and Yi (1999) carefully document the rise of 35. Burtless and others 1998. trade in vertically differentiated products. Deardorff (1998) pro-      ⁄  vides a novel theoretical treatment of production fragmentation 14. The case for liberalizing international capital flows was across borders. laid out succinctly by the IMF’s Deputy Managing Director 60. Cairncross 1997. Stanley Fischer in a speech in September 1997. See “Financial 61. Graham 1996. Instability,” Oxford Analytica, November 4, 1998. For a careful 62. UNCTAD 1997. discussion of the effects of international capital mobility on the 63. Yeats 1998. efficiency of the worldwide allocation of resources, see Cooper 64. For recent estimates of the effects of infrastructure and (1999). transportation on trade flows, see Bougheas, Demetriades, and 15. World Bank 1998b. Morgenroth (1999). 16. This discussion draws on Caprio and Klingebiel (1996), 65. Graham 1996; Vernon 1998. Demirgüç-Kunt and Detragiache (1998), Eichengreen and Rose 66. For an analysis of how corporate tax rates and rules on (1998), Goldstein (1998), Goldstein and Turner (1996), and the repatriation of profits affect firms’ location decisions, see World Bank (1998h). Mutti and Grubert (1998). 17. Eichengreen 1999. 67. Vernon 1998. 18. “Financial Instability,” Oxford Analytica, November 4, 68. See chapters 6 and 7 for a more extensive discussion of 1998. urban issues. 19. Eichengreen and Rose 1998. 20. Widespread borrowing is more likely when inadequate Chapter 3 corporate governance systems undermine monitoring and other 1. Bordo, Eichengreen, and Irwin 1999. measures to restrict risk-taking. 2. Recent evidence suggests that investors’ appetite for high- 21. Kaminsky and Reinhart 1998; Calvo 1999. risk investment vehicles remains strong. Despite the 1998 col- 22. Goldstein and Hawkins 1998. lapse of Long-Term Capital Management, which was at the 23. Radelet and Sachs 1998. heart of the hedge fund crisis, estimates show that the total 24. Leipziger 1998. amount invested in risky hedge funds in early 1999 was less than 25. By extension, global contagion provides a rationale for 2 percent lower than it was a year earlier. See The Economist global banking standards (Goldstein 1997). (1999c). 26. Rodrik 1998c. However, the evidence presented in Quinn 3. This section draws on various editions of the IMF’s Inter- (1997) points to the positive effect of capital account liberaliza- national Capital Markets. Mussa and Richards (1999) present a tion on growth. detailed overview of the magnitude and composition of capital 27. For a detailed analysis of the nature, causes, and conse- flows in the 1990s. quences of financial contagion, see IMF (1999b). Evidence of 4. Foreign direct investment (FDI) involves investments in the correlation between key financial variables, thought to be a companies that account for more than 10 percent of the recipi- central feature of contagion, is marshaled in Wolf (1999). ent company’s assets. In contrast, foreign portfolio investment 28. For a comprehensive list of all of the measures promoted (FPI) refers to purchases of foreign liquid financial assets. In to enhance the stability of the international financial system, see practice, the distinction is not clear cut, since financial transac- IMF (1999a). tions may involve elements of both. However, portfolio invest- 29. The relative importance of capital markets and bank in- ment is presumed to have higher liquidity and to be more “foot- loose” than FDI. For a further discussion of the definition and termediation is determined in part by national policies. See measurement of foreign investment, see Lipsey (1999). FDI and Berthélemy and Varoudakis (1996). FPI should not be confused with short-term borrowing from 30. Levine 1997, 1998. abroad. 31. Stiglitz 1999a. 5. See Eichengreen and Mussa (1998). 32. Levine (1998) presents cross-country evidence of the im- 6. For a detailed account of the pace of capital account lib- portance of creditors’ rights and other legal institutions in facil- eralization, see Quirk and Evans (1995) and recent editions of itating banking sector development. Demirgüç-Kunt and De- the IMF’s Annual Report on Exchange Rate Arrangements and Ex- tragiache (1998) find that the impact of domestic financial change Restrictions. liberalization on the probability of a banking crisis is greater in 7. This discussion of foreign direct investment draws on IFC nations with widespread corruption, inefficient bureaucracies, (1998), Knight (1998), Mallampally and Sauvant (1999), and and little respect for the rule of law. See also G-22 Committee UNCTAD (1998). (1998b). 8. The canonical survey on these issues is found in Caves 33. “Financial Infrastructure,” Oxford Analytica, November (1996), chapter 7. See also Oxley and Yeung (1998). 9, 1998. 9. UNCTAD 1998. 34. Dewatripont and Tirole 1994. 10. UNCTAD 1998. 35. Garcia 1996, 1998; Lindgren and Garcia 1996. 11. Mallampally and Sauvant (1999) report that by 1997 36. Kane 1998. there were 1,794 double-taxation treaties in effect. 37. Litan 1998. 12. UNCTAD 1996, 1998. 38. Calomiris 1997. 13. For a detailed analysis of the geographic distribution of 39. See G-22 Committee (1998a). foreign investment, see Lipsey (1999). 40. Krugman 1998b.    41. For an account of the difficulties caused by provincial (1999); Eichengreen and Fishlow (1998); Hernández and banks in Argentina’s bank crisis in 1995, see Leipziger (1998). Schmidt-Hebbel (1999); and Valdes-Prieto and Soto (1996). For 42. It has recently been argued that banking activities have a critical survey of the effects of capital controls, see Dooley become so complex that senior bank executives are unlikely to (1996). know (and in some cases to understand) the consequences of 60. Baliño, Bennett, and Borensztein (1999). their employees’ actions (Financial Times, “Too Much on Their 61. For a lengthy discussion of the effects of dollarization Plate.” February 4, 1999). Holders of subordinated debt may on developing countries, see Baliño, Bennett, and Borensztein be similarly disadvantaged. (1999). 43. Evanoff 1998; Calomiris 1997, 1999. 62. Such a run might be caused by an increase in foreign in- 44. Calomiris 1997. terest rates. 45. Financial Times, “G7 Offers Shelter from Storm.” Feb- 63. Calomiris (1999) has proposed changing the IMF’s role ruary 22, 1999. The Group of Seven industrial nations set up a to include a discount window lending facility that would pro- “financial stability forum” in February 1999 designed to vide liquidity to qualified countries. See also Feldstein (1999). strengthen surveillance and supervision of the international fi- 64. The need for interventions can be seen in another way. nancial system. Central bankers, finance ministry officials, and Recent studies have suggested that the best predictor of a crisis supervisory officials will serve on this committee. Initially, is the ratio of short-term foreign indebtedness to reserves. Thus, members will come from Group of Seven countries. if a firm borrows more short-term money abroad, its govern- 46. Goldstein 1997, 1998. High capital requirements come ment—if it wishes to maintain a prudential stance—must set at a cost—the opportunity cost of forgone lending that lower aside more funds in reserves, typically holding these reserves as capital requirements could sustain. U.S. Treasury bills or similar instruments from other industrial 47. For an extensive discussion of the pros and cons of economies. the admission of foreign banks, see Caprio (1998). See also 65. Eichengreen 1998. Calomiris 1999 and EBRD 1998. 66. Borensztein, De Gregorio, and Lee 1998; UNCTAD 1998. Of course, foreign direct investment is not the only source 48. Claessens, Demirgüç-Kunt, and Huizinga 1998. of technology transfer—patent licensing schemes were used in 49. Hellman, Murdock, and Stiglitz 1998; Stiglitz 1999a. Korea and Japan in the early stages of the countries’ develop- 50. Peek and Rosengren 1997. ment. See Kim and Ma (1997). 51. World Bank 1998h; Eichengreen 1998, 1999; Eichen- 67. Wacziarg 1998. green and Mussa 1998; Johnston, Darbar, and Echeverria 1997; 68. Borensztein, De Gregorio, and Lee 1998. McKinnon 1991. 69. Berthélemy, Dessus, and Varoudakis 1997. 52. Harwood 1997; Johnston 1997. 70. De Mello 1997. 53. A theoretical and emprical analysis by Rodrik and Ve- 71. This evidence is taken from Kozul-Wright and Rowthorn lasco (1999) concludes that measures to restrict short-term bor- (1998). rowing are desirable. 72. Kinoshita and Mody 1997. 54. Feldstein 1999; Eichengreen and Mussa 1998; McKin- 73. De Mello 1997. non and Pill 1998. 74. Kozul-Wright and Rowthorn 1998. 55. Two recent studies of the East Asian and Latin Ameri- 75. See Moran (1999) for detailed case studies of each of these can experience with capital controls are broadly supportive of policy instruments. Moran did find that information dissemina- their use (Le Fort and Budnevich 1998; Park and Song 1998). tion and purely promotional initiatives by developing countries Dornbusch (1998) explores the pros and cons of various restric- are effective tools for boosting foreign direct investment. tions on capital inflows, pointing out situations in which such 76. Gastanaga, Nugent, and Pashamova 1998. controls improve economic performance. 77. World Bank 1997d. 56. Caprio 1998; Eichengreen 1998. In 1997 Malaysia in- 78. Vernon 1998. troduced controls on both short-term capital inflows and cer- 79. Oxford Analytica, “Energy Investment.” February 1, tain capital outflows. Although it is too soon to determine the 1996. overall impact of these measures, since 1997 foreign direct in- 80. Blomström and Kokko 1997. vestment in Malaysia has not been affected any more adversely 81. Claessens and Rhee 1994; Demirgüc-Kunt and Levine than in neighboring countries that did not impose controls. 1995; Levine and Zervos 1998a. 57. Johnston, Darbar, and Echeverria 1997; Reinhart and 82. Levine 1997; Levine and Zervos 1998a, 1998b. Reinhart 1998; Velasco and Cabezas 1998; United Kingdom 83. Saudagaran and Diga 1997. 1998; Oxford Analytica, “Financial Regulation.” December 29, 84. Levine 1997, 1999. Strong shareholder rights may also 1998. play a role in reducing excess short-term foreign borrowing by 58. An alternative is to permanently raise reserve require- company managers. ments on foreign deposits or capital adequacy requirements on 85. Bryant 1995; Eichengreen and Kenen 1994; Sachs and foreign borrowings. McKibbin 1991. 59. The evidence in Edwards (1998a) suggests that these con- 86. Rajan 1998. trols had a temporary effect on interest rate differentials between 87. ADBI 1998. Chile and overseas markets. For information on Chilean capital 88. Rajan 1998. controls, see Chumacero, Laban, and Larrain (1996); Cooper 89. Bergsten 1998.      ⁄  Chapter 4 mile, hydrocarbon emissions by 95 percent, and carbon monox- 1. World Bank 1992b. ide emissions by 98 percent. If the car does not have a converter, 2. Pearce and others 1996. the reductions are even greater (Ornusal and Gautam 1997). 3. World Bank 1998g. 19. See Goulder (1994), however, who suggests that carbon 4. Antarctica Project 1999. taxes would not be efficient in the United States because they 5. Watson and others 1998. would be likely to replace income taxes—which are even more 6. Imber 1996; Porter and others 1998. broadly based. 7. Grossman and Krueger 1995. 20. World Bank 1999d. 8. For information on Kazakhstan and Uzbekistan, see World 21. Elster 1988; Schlicht 1985. Bank (1998l). For China, see Chinese State Council (1994). 22. Madeley 1995b. Since 1994 senior Chinese officials have reiterated their con- 23. de Fontaubert 1996. cerns about environmental issues, and the government has in- 24. Peterson 1993. vested considerable resources in protecting air and water. But 25. Rose and Crane 1995. much remains to be done (see World Bank 1997a). Agenda 21, 26. French 1997; Ornusal and Gautam 1997. the principal agreement to emerge from the Rio Earth Summit, 27. Chomitz and Kumari 1998. committed national leaders to action programs organized under 28. Lampietti and Dixon 1995. Clearly, these numbers are the following six themes: quality of life, efficient use of natural very rough estimates that will vary dramatically from forest to resources, protection of the global commons, management of forest. human settlements, waste management, and sustainable eco- 29. Perrings 1995. nomic growth. See Flavin (1997) and World Bank (1997b). 30. Lampietti and Dixon 1995. It should be noted that these 9. Wapner 1995; Zurn 1998. species were all “prominent”—grizzly bears, whooping cranes, and bald eagles—rather than different types of beetle. 10. Cesar 1998. 31. World Bank 1998g; Porter and others 1998. GEF fund- 11. Below-market sales of timber concessions constitute an- ing for ozone projects amounted to nearly $126 million by mid- other subsidy that affects the rate of deforestation. In 1990 the 1999, according to World Bank sources. Indonesian government sold timber concessions at prices far 32. The Kyoto meeting is officially termed the Third Con- below prevailing market prices, capturing only 17 percent of the ference of the Parties to the United Nations Framework Con- value of the trees and costing the treasury more than $2.1 bil- vention on Climate Change. lion in forgone revenue. By contrast, the Brazilian government’s 33. No limits on emissions in developing countries (other decision in 1988 to cancel tax credits for ranchers who cleared than the economies in transition) were set at the Kyoto meeting. land slowed deforestation in the Amazon significantly—and 34. However, the ozone hole over Antarctica continues to saved the government money. Subsidies for building roads af- grow (The Sciences 1997). fect deforestation because access roads markedly increase the 35. WRI 1998. probability that a forest will be converted to agricultural use. 36. French 1997; Miller 1995. Chapter 5 discusses the complexities of government support for 37. Seaver 1997. infrastructure investments, but one thing that is clearly impor- 38. Barrett 1998a, 1998b. tant is to take the environmental impact of such decisions into 39. French 1997. account (Roodman 1997). 40. UNEP 1999. 12. Eskeland and Feyzioglu 1994. 41. Barrett 1998a, 1998b. 13. Anderson and McKibbin 1997. 42. Sell 1996; Seaver 1997. 14. The transition economies of Eastern Europe and Asia 43. Sell 1996. have shown that it is politically possible to phase out these sub- 44. Sell 1996. sidies fairly rapidly. China’s subsidies for coal, which is the 45. Seaver 1997. source of 70 percent of the country’s energy, fell from 61 to 46. World Bank 1998m. 11 percent over 1984–95, cutting government costs from 47. Barkin and Shambaugh 1996. $25 billion in 1990–91 to $10 billion in 1995–96 (Watson and 48. Miller 1995. others 1998). 49. Barrett 1998a, 1998b. 15. World Bank 1998f. More recent estimates from Malawi 50. Barrett 1998a, 1998b. suggest that middle-income customers may receive as much as 51. Seaver 1997. $180 in annual subsidies. 52. Pearce and others 1996. 16. Roodman 1997. 53. Hourcade 1996. 17. World Bank 1996a. 54. Sell 1996. 18. It should be noted that poorly maintained diesel buses 55. Roodman 1997. can be a serious source of pollutants. Even so, making the switch 56. The United States is only indicative of a wider trend; from private cars to buses nearly always reduces urban air pol- public sector energy R&D expenditure fell sixfold in the United lution dramatically. In Mexico City, for example, taking a bus Kingdom and fourfold in Germany and Italy between 1984 and instead of driving a car equipped with a catalytic converter re- 1994 (President’s Committee of Advisors on Science and Tech- duces nitrogen oxide emissions by 40 percent per passenger- nology Panel on Energy Research and Development 1997).    57. To enforce an agreement, whether based on taxes or on ments—although it should be noted that this situation devel- quotas, it might be necessary to charge offsetting, or even puni- oped before the Rio Convention (Munson 1995; Miller 1995). tive, charges on imports from countries that have failed to make 73. Sell 1996. or live up to commitments for greenhouse gas reductions. This 74. Miller 1995. suggests the possible use of fines (such as those contemplated in 75. Simpson, Sedjo, and Reid 1996. This figure is low. the European Union for violations of the fiscal stabilization pact) While many thousands, if not millions, of species are repre- or economic sanctions. But it would be difficult to deny imports sented in each hectare, dividing the total number of endemic related to greenhouse gas emissions without, in effect, prohibit- species types by the total number of hectares in western Ecuador ing trade with the offending country, since carbon dioxide–pro- produces a small number of endemic species per hectare. ducing energy is required for virtually all production. Calculat- 76. Regional agreements (such as the Joint Comprehensive ing the optimal punitive tariff also turns out to be very complex, Environmental Action Program for the Baltic, which supports and present multilateral trade rules do not allow trade restric- information flows, technical assistance, and environmental fund- tions based on how a product was made. As mentioned earlier, ing) can also play an important role in preserving genetic biodi- the credibility of trade sanctions is strongly linked with the costs versity and habitats (Freestone 1999). and benefits of treaty compliance, and in this case many coun- 77. Charnovitz 1996. tries might find their costs of imposing sanctions to be larger 78. Freestone and Makuch 1998. than the benefits of enforcing treaty compliance (Stiglitz 1997; 79. Charnovitz 1996. Barrett 1998c; World Bank 1998d). 80. Charnovitz 1996. 58. Sell 1996. 81. The Economist 1998d; Howse and Trebilcock 1996. 59. Stiglitz 1997. 82. Howse and Trebilcock 1996. 60. World Bank 1998d. 83. In Costa Rica there is evidence that species in high- 61. Cooper 1998. altitude forests are dying out because climate change is lifting 62. Stiglitz 1997. cloud cover above the forests (see, for example, Holmes 1999). 63. World Bank 1998k. 84. Watson and others 1998. 64. Trading mechanisms are not free from controversy. At 85. Watson and others 1998. Kyoto some developing countries opposed trading, seeing it as 86. WRI 1998. a mechanism for wealthy countries to buy their way out of emis- 87. World Bank 1998e. sions restrictions and transfer those limits to poor countries, 88. This program falls under the auspices of the Kyoto Pro- where such limits would interfere with development (Anderson tocol’s Clean Development Mechanism. 1998). 89. Goodman 1998. 65. World Bank 1998d. 66. Watson and others 1998. Chapter 5 67. Calculated from World Bank (1999i). This is clearly a 1. Subnational elections are held in 71 out of 75 multiparty poor way to measure stocks of biodiversity. Many of these plants democracies for which data were available. The total number of and animals will be present in more than one country, and many multiparty democracies in the world, as classified by Freedom of the animals are not threatened with extinction. The point re- House, is 117. See appendix table A.1 for details on decentral- mains, however, that the majority of species left on the planet ization and Freedom House (1998) for multiparty democracy resides in developing countries. classification. 68. Heywood 1995. 2. Decentralization and devolution are used synonymously 69. Madeley 1995a. throughout the chapter. 70. Miller 1995. 3. Smith 1996. See also Dahl (1986) and Stepan (1999) on 71. Simpson, Sedjo, and Reid 1996. the relation between democracy and decentralization. In a strict 72. The convention contains only vague language on pay- sense, only a constitutional democracy can credibly guarantee ment for genetic resources. Article 15 states that contracting that the prerogatives of subunits will be respected. parties will share in “a fair and equitable way the results of re- 4. Treisman 1998. search and development and the benefits arising from commer- 5. Hommes 1996. cial and other utilization of genetic resources . . . [on] mutually 6. Litvack 1994. agreed terms,” without defining a framework for the terms or 7. Musgrave and Musgrave 1973; Oates 1972; Tiebout 1956. the words “fair and equitable.” Industrial countries have encour- 8. Ostrom, Schroeder, and Wynne 1993. aged the notion that biodiversity is a global good that should 9. Junaid Ahmad contributed to the writing of this box, not be assigned to nations as property, while at the same time which is also based on Ablo and Reinikka (1998) and a note by arguing that companies should be able to patent products they Paul Smoke. develop from plants and animals. Without some kind of reform, 10. Breton 1996. resource transfers for such drugs will continue to run from de- 11. A number of studies are available, though they tend to veloping to industrial countries rather than the other way. De- focus on a particular sector within a country (King and Ozler spite the value of the drugs developed from the rosy periwinkle, 1998; Ablo and Reinikka 1998) or a particular tier of govern- for instance, Madagascar still receives nothing in royalty pay- ment within a country (Faguet 1998; World Bank 1995b).      ⁄  They all support the notion that the consequences of decentral- The article also reported that similar protests against corrupt or ization depend on the way it is designed and implemented. abusive officials were occurring around the country. Informa- 12. King and Ozler 1998. tion on corruption is from the Financial Times, March 5, 1999, 13. Burki, Perry, and Dillinger 1999. “Officials Arrested over Chinese Fraud Scandal” and “Tentacles 14. Litvack, Ahmad, and Bird 1998. of Corruption May Threaten the State.” 15. Ahmad and Craig 1997. 35. Public spending has been found to be biased in favor of 16. Local governments generally have access to a more com- the least populous regions in Brazil and Argentina, which have plete set of information about both the population’s preferences territorial representation in both houses. In contrast, public and their own resources and performance. This information spending per capita does not vary significantly across Mexican gives local authorities an advantage in delivering the appropri- and U.S. states, where territorial representation is applied only ate mix of services. But the fact that the central government may to the Senate (Gibson, Calvo, and Falleti 1999). not share this information complicates the task of monitoring 36. This was also the practice in the United States until 1913 the local government’s performance and establishing the true and in Argentina until 1994. need for financial assistance. For a discussion of these issues and 37. In Brazil, for example, it is estimated that roughly 40 the means of overcoming them, see Ravallion (1999a, 1999b) percent of senators have been governors and that many senators and Burgess (1998). aspire to be governors. Further, in the 1991–94 legislature, ap- 17. Bird and Rodriguez 1999. proximately 35 percent of the sitting deputies exhibited a pref- 18. Ahmad and Craig 1997. erence or actually gave up their seat for a state-level post. In this 19. For means of resolving this issue, see Ravallion (1999b). context, national parliamentarians are more likely to care about 20. Ravallion 1999a. pleasing their constituents and the governor of their state than 21. Ravallion 1999b. about the national good (Stepan 1999). 22. Alderman 1998. 38. Ordeshook and Shvetsova 1997. 23. World Bank 1999h. 39. In contrast, plurality or first-past-the-post systems virtu- 24. Ravallion 1999b. ally guarantee parliamentary majorities (Lijphart 1994). 25. Faguet 1998. 40. Carey 1997. Even when coalition partners have enough 26. Tanzi 1996. power to block change, they may not have enough leverage to 27. China, which is neither wealthy nor a federation, is a effect positive change on their own (Alesina and Perotti 1997; notable exception. Subnational entities are responsible for a Roubini and Sachs 1989). Evidence for both Latin America and large share of the tax collection and for expenditures (box 5.5). Europe shows that the central government’s ability to respond Comparable data from China were not available for figures 5.1 decisively to shocks, restrain expenditures, and contain the size and 5.2. of government is lower in countries with proportional represen- 28. Gavin and Perotti 1997 (Latin America); McKinnon tation. In Europe, however, budgetary rules seem to help over- 1997 (United States); Spahn 1998 (Western Europe). For fur- come these problems (Hallerberg and von Hagen 1997; Stein, ther discussions of macroeconomic stability and decentraliza- Talvi, and Grisanti 1998). tion, see Fornasari, Webb, and Zou (1999); McLure (1999); 41. Gamble and others 1992. See Lijphart (1994) for a full Prud’homme (1995); Sewell (1996); Shah (1998); Tanzi (1996); discussion. and Wildasin (1997). For a discussion on decentralization and 42. Lijphart 1994. growth, see Davoodi and Zou (1998); Xie, Zou, and Davoodi 43. This section is based on Ordeshook and Shvetsova (1997). (1999). For the relation between decentralization and the size 44. For a discussion of this point in the Latin American con- of government see Jin and Zou (1998); Persson and Tabellini text, see Willis, Garman, and Haggard (1999). (1994); Quigley and Rubinfeld (1997). For an overall review of 45. In Yugoslavia, the first competitive elections were held decentralization and growth see Martinez-Vasquez and McNab at the subnational level and were won by regional and ethnic (1997). nationalist parties. The civil wars occurred before there was a 29. de Figueiredo and Weingast 1998. nationwide election. In the founding election in Nigeria in 30. Linz and Stepan 1997; Elster and Slagsrad 1993. 1959, there were virtually no elected representatives of nation- 31. Weingast 1995. wide parties, a situation that directly contributed to the escala- 32. In “bottom-up federations” like the European Union tion of ethnic tensions and the civil war over the Biafran seces- and the United States, the constituent members decide upon sionist attempt (Stepan 1999). the initial set of rules. Such federations tend to generate a much 46. The integration of party structures can be built into the weaker center than top-down ones. See de Figueiredo and political system in a variety of ways. In Germany, for example, Weingast (1998). the upper house, which represents subnational interests, has very 33. Ordeshook and Shvetsova 1997. limited powers. But half of the lower house is elected with the 34. This box is based on Bahl (1999b); Lall and Hofman use of regional lists that are controlled by the same parties that (1994); Qian and Weingast (1997); Wong (1998); World Bank elect candidates to Länder (state) positions (Ordeshook and (1995a). The Washington Post, February 27, 1999, reported in Shvetsova 1997). “China Praises Sichuan Election” on the local election that took 47. Oates 1972; Tiebout 1956; Musgrave and Musgrave place in Buyun (Sichuan Province) after the people there had 1973. forced the township’s leader out of office for governing badly. 48. Donahue 1997.    49. Musgrave 1997. 71. Stotsky and Sunley 1997. 50. Hemming and Spahn 1998. 72. In the United Kingdom each municipal council has the 51. This model is the traditional Tiebout (1956) one of authority to set its salary scales, but 90 percent of them partici- “voice” and “exit.” pate in collective bargaining with the national public employ- 52. For a discussion of these assumptions and their relevance ees’ union. In Germany subnational governments are required to the fiscal federalist model, see Oates (1998). by statute to adhere to agreements negotiated jointly with the 53. Manning 1998; Fay and others 1998. federal government and public employees unions. 54. Wetzel and Dunn 1998. 73. Smith 1996. More generally, the expected payoffs affect 55. The number of school districts in the United States de- participation. Individuals must believe that the benefits they will creased markedly in the 1950s as jurisdictions tried to put to- receive will exceed the costs of their time, labor, and money gether groups of students large enough to run grade-differentiated (Hirschman 1970; North 1990; Ostrom, Schroeder, and Wynne primary schools. Germany has reduced the number of municipal- 1993). ities (Gemeinden) by half. The United Kingdom has eliminated 74. Galeotti 1992. a tier of subnational government in Scotland, Wales, and the met- 75. Bridges 1997; Hawley 1970. ropolitan areas of England. With the exception of France, the low- 76. Poterba 1994. est tier of subnational government in large OECD countries has 77. Dahl 1971. an average population of about 5,000–7,000. In Japan, however, 78. Stren 1998. this figure can run as high as 39,000, and in the United Kingdom, 79. The 1994 Popular Participation Law formalized the role it can reach 109,000. of community organizations as watchdogs at the municipal 56. Vaillancourt 1998. level, granting them the right to report suspected wrongdoing 57. The number of municipios in Brazil increased from 3,000 to the Senate (Campbell 1998). to nearly 5,000 in the 15 years following the return of democ- 80. Tendler 1997; Vivian 1994; Zaidi 1999. racy. The new constitution of the Philippines recognizes not only 81. World Bank 1992a. 1,605 cities and municipalities but also nearly 42,000 neighbor- 82. O’Donnell, Schmitter, and Whitehead 1986. hood organizations (barangays) as units of local government. 83. Wiseman 1997. 58. For a discussion of these trade-offs in the context of the 84. Boeninger 1992; Przeworski and Limongi 1997. European Union, see Alesina and Wacziarg (1998). 85. Diamond 1996. 59. Diamond 1999. 86. Bird and Vaillancourt 1999. 60. See Wildasin (1997) for a further discussion on the 87. Dillinger and Webb 1999a. topic. 88. Under the revised local government code effective in 61. The property tax is potentially one of the best sources of 1992, central government agencies were required to transfer to revenues for local governments. For a discussion of problems subnational units specific activities (including agricultural exten- and possible reforms of the property tax, see Dillinger (1992). sion, forest management, local hospital operations, primary 62. For further discussion of subnational tax reform see Bahl health care programs, local roads, water supply, and communal and Linn (1992), Bird (1999), McLure (1999), Norregaard irrigation infrastructure). To finance these costs local govern- (1997), and Vehorn and Ahmad (1997). See also Bird, Ebel, and ments received a larger share of national tax revenues. In the first Wallich (1995) and Bird and Vaillancourt (1999) for country- year of implementation, the code required the central government specific examples and Inman and Rubinfeld (1996) for a more to provide funds to cover the cost of personnel devolved to the theoretical treatment. local governments in addition to the increase in revenue sharing. 63. The Scandinavian countries, which have allocated sub- 89. A number of countries seem to be faring reasonably well, stantial taxing powers to their local governments, are a rare ex- but none has emerged as an unmitigated success. ception. See Litvack, Ahmad, and Bird (1998). 64. Diamond 1999. 65. This section is largely based on Bahl and Linn (1992). Chapter 6 66. This box is based on McLure (1999); Bird and Gendron 1. Hohenberg 1998. (1997); and Inman and Rubinfeld (1996). For a discussion of 2. Glaeser and Rappaport 1998. subnational VATs see Bird and Gendron (1997). 3. Richardson (1987) provides evidence from Brazil, France, 67. Bahl and Linn 1992; Bahl 1999a. Peru, and the United States on the high cost of city living. 68. See Ter-Minassian and Craig (1997) for a more detailed 4. Shukla 1996. discussion. 5. Mazumdar 1986; Mills and Becker 1986. 69. Ter-Minassian and Craig 1997. 6. Krugman 1993; Quigley 1998. 70. In the United States, states that have formal controls on 7. Dumais, Ellison, and Glaeser 1997; Glaeser 1997; Jaffe, borrowing have lower debt levels on average (Poterba 1994) but Trajtenberg, and Henderson 1993. they are equally prone to serious fiscal crisis (Von Hagen 1991). 8. Henderson 1998; Henderson, Lee, and Lee 1998. In addition, these controls are self-imposed and not the result 9. Lucas 1998. of federal government mandates—that is, they have been vol- 10. Brown and McCalla 1998. untarily written by state assemblies into state constitutions 11. Rousseau 1995; Thomas 1980. rather than imposed by the central government. 12. Ades and Glaeser 1995.      ⁄  13. Gertler 1997. 9. WRI 1996. 14. Yeates 1997. 10. WRI-WHO 1999. 15. Black and Henderson 1998. 11. World Bank 1994. 16. Gaspar and Glaeser 1998. 12. WRI 1996. 17. Choe and Kim 1999. 13. UNDP 1998. 18. UNCHS 1996. 14. World Bank 1997a. 19. Tarver 1995. 15. WRI 1996; Harpham and Tanner 1995. For recent evi- 20. Lucas 1998; Mills 1998; Tacoli 1998. dence on urban differentials in the United States, see Claudio 21. Henderson 1998. and others (1999). 22. Head and Ries 1995. 16. WRI 1996; Haddad, Ruel, and Garrett 1999. 23. Ades and Glaeser 1995. 17. Other determinants are important as well, including cul- 24. Henderson and Kuncoro 1996. tural and political alienation, ethnic conflicts, and media vio- 25. Gertler 1997. lence. See also Bourguignon (1998). 26. This calculation assumes per capita costs of $150 for 18. Zaidi 1998. water, $300 for sewers. It also assumes that 30 percent of the 19. Bourguignon 1998. A recent estimate for South Africa urban population has no access to potable water and that 40 places the costs of crime and violence at 6 percent (at least) of percent lack access to sewers. GDP figure from World Bank the country’s GDP (Business Times, February 14, 1999). (1998l). 20. World Bank 1994. 27. Mayo and Angel 1993. 21. Rosen 1993. 28. Mohan 1999. 22. Rosen 1993. 29. The subject of local tax revenue sources is beyond the 23. In many cities (Bogotá, Karachi, Manila, and Taipei, for scope of the WDR. A key reference is Bahl and Linn (1992). instance) the rich remain vulnerable to crime and kidnappings. Property taxes remain the key revenue source for many cities be- Even the most stringent security measures are unable to guar- cause they are relatively easy to collect, although collections are antee personal safety. See Simon Romero, “Cashing in on Secu- often incomplete and discriminatory. In theory land taxation is rity Worries,” The New York Times, July 24, 1999. a nondistortionary source of funding, but in practice assessing 24. G. Shah 1997. pure land values is difficult. All land has been improved to some 25. Chaplin 1999. degree, tempting governments to overassess. 26. Tynan and Cowen 1998. 30. Dailami and Leipziger 1998. 27. Foreman-Peck and Millward 1994. 31. AB Assesores 1998; Freire, Huertas, and Darche 1998. 28. Anderson 1988. 32. Peterson and Hammam 1997. 29. Shugart 1997. 33. Peterson and Hammam 1997; Dailami and Leipziger 30. Financial Times, April 29, 1999. 1998. 31. A report issued by the French Auditor’s Office (Cour des 34. Private communication with S. Mayo (Lincoln Insti- Comptes) for the water and wastewater sector in January 1997 tute), 1998. reported a lack of transparency in a number of instances and 35. Colgan 1995. found that in some cases private participants appeared to be pre- 36. Colgan 1995. venting good information from reaching elected officials. The 37. Markusen 1998. report did conclude that water services overall were delivered 38. Miranda and Rosdil 1995; Bradbury, Kodrzycki, and satisfactorily. See Shugart (1997) for a discussion. 32. “Gestion de l’eau: renégociations en chaîne des contrats Tannenwald 1997. avec les groupes privés,” Les Echos, March 25, 1999. 39. ILO 1998. 33. Pirez 1998. 40. Markusen 1998. 34. Quoted in Root (1998). 41. Bertaud and others 1997. 35. See chapter 8 for a case study of Karachi. 42. Cour 1998a. 36. Hardoy and Satterthwaite 1990. 43. Cour 1998b. 37. Campbell 1998; Stren 1998. 38. Root 1998. Chapter 7 39. Harpham and Stuttaford 1999. 1. World Bank 1994. 40. Mayo and Angel 1993. 2. Upward mobility in Karachi, Pakistan, is discussed in 41. Hasan 1997a; Leitman and Baharoglu 1998; Lloyd- Altaf and others (1993). A case study of Karachi is presented in Sherlock 1997. chapter 8. 42. Espinosa and López Rivera 1994; UNCHS 1996; World 3. Kessides 1998; Evans 1998. Bank 1996b. 4. Brown and McCalla 1998. 43. Buckley and Mayo 1989; UNCHS 1996. Although the 5. Douglass 1992. UN strategy formally endorsed the enabling approach, strong 6. WRI 1996. An earlier WHO study (1986) estimated the advocates appeared much earlier. The 1979 World Development reduction at 40–50 percent. Report proposed an urban housing strategy for developing coun- 7. WHO 1995. tries that would focus on encouraging the private sector to im- 8. Chhabra and others 1998. prove the housing supply (World Bank 1979).    44. Reforms should include the following: developing prop- crime control: in 1989 koban officers were responsible for 73 erty rights and expanding land registration, streamlining the reg- percent of all arrests and 76 percent of all the theft cases that ulatory process and regulations to reduce housing costs, encour- were solved. See Zaidi (1998). aging greater competition in housing construction, providing 73. Ayres 1997. trunk infrastructure at full cost recovery, fostering the develop- 74. Hasan 1998. ment of mortgage financing systems, and, particularly, improv- 75. Conger 1999. ing access to credit and targeting subsidies (Mayo and Angel 76. Participatory budgeting has also been introduced in cities 1993; UNCHS 1996). in Mexico and Venezuela. See Campbell (1998) and Coelho 45. Strong, Reiner, and Szyrmer 1996; Struyk 1997. (1996). 46. Gilbert and Gugler 1992; Hasan 1997a. 77. World Bank 1992a. 47. WRI 1996. 78. Anderson 1998. 48. Whittington, Lauria, and Mu 1991. 79. Fujikura 1999. 49. The Wall Street Journal, “Populist Perrier? Nestle Pitches 80. Afsah, Laplante, and Wheeler 1997. Bottled Water to World’s Poor.” June 18, 1999. 50. Atlaf 1994a. 51. Altaf 1994b. Chapter 8 1. Hoekman and Djankov 1996. 52. World Bank 1994. 2. World Bank 1998c. 53. Solo 1999. 3. Egypt benefited significantly from its participation in the 54. Porter 1996; Cowen and Tynan 1999. Gulf War (1990–91) and subsequently received substantial debt 55. Porter 1996. relief from the United States and others. 56. World Bank 1993a. 57. Porter 1996. 4. World Bank 1998c. 58. Blackett 1994; World Bank 1994; WRI 1996. 5. Hoekman and Djankov 1997a; World Bank 1998c. 59. On condominial sewers, see Watson (1995) and World 6. Kenny 1999. Bank (1992b). But because households often convert from dry 7. Hoekman, Konan, and Maskus 1998. latrines to pour-flush systems without connecting to proper drain 8. Konan and Maskus 1997. fields, these systems may discharge into open street drains. Such 9. Hoekman and Konan 1999. patterns have been documented in Gujranwala, Pakistan; Kumasi, 10. EBRD 1998. Ghana; and Ouagadougou, Burkina Faso. See Altaf (1994a); Altaf 11. Long and Kopanyi 1998; Vittas and Neal 1992. and Hughes (1994); Whittington and others (1993). 12. Abel and Szakadat 1997–98. 60. Hasan 1998. 13. Long and Kopanyi 1998. 61. Ingram 1998. 14. Calomiris 1997. 62. Kitano 1998. 15. Souza 1996. 63. Rabinovitch 1992; WRI 1996. 16. Mendes 1999. 64. Copenhagen is an example of a city that has reduced au- 17. The state governments are assigned a value-added tax tomobile dependency by revitalizing downtown housing and (VAT), which they assess and collect directly. As the highest- street life and restricting parking in the city center. Freiburg’s yielding revenue source in Brazil, the VAT gives the states an improvements to public transport have focused on extending independent power base, particularly in the wealthy southeast, and upgrading its light rail system, which uses buses as feeders. where it is the principal source of state revenues. Perth has had limited success trying to discourage automobile 18. Afonso 1992; Rezende 1995. use by integrating bus services with the newly constructed elec- 19. Mainwaring 1997. tric rail system. See UNCHS (1996). 20. Two of the 26 states, Bahia and Ceará, have since un- 65. WRI 1996. dertaken substantial adjustment and reform. See Dillinger and 66. Burgess 1999; Frigenti and Harth 1998; Graham 1994. Webb (1999). 67. Haddad, Ruel, and Garrett 1999. 21. Dillinger 1997. 68. World Bank 1999e. 22. After the plan was introduced in mid-1994, annual in- 69. Mitlin and Satterthwaite 1998. flation (as measured by the INPC index) fell from 929 percent 70. UNCHS 1996. in 1994 to 22 percent in 1995, 9 percent in 1996, 4.3 percent 71. Douglass 1992; Evans 1998. in 1997, and 2.5 percent in 1998. 72. Japan’s neighborhood police stations, or kobans, provide 23. Ter-Minassian and Craig 1997. a model of effective community policing. A vast network of 24. An option worth exploring is prohibiting all government community-based crime control organizations operating in lending to subnational governments. Argentina and Colombia, neighborhoods, schools, and workplaces is credited with reduc- for example, leave subnational financing entirely to the private ing crime rates in Japan, which are very low and falling. The sector; a practice that so far has forestalled any claims for fed- Crime Prevention Associations have 540,000 local liaison units. eral debt relief. Officers are required to visit every family and business in their 25. The case study is based on Hasan, Zaidi, and Younus neighborhoods at least twice a year and to provide many com- 1998. munity services, such as helping to organize newsletters, meet- 26. Mahmood 1999. ings, and sports events. The koban system is highly effective in 27. Zaidi 1997.      ⁄  28. Hasan 1997b. able African countries to raise yields as well as enlarge their ex- 29. World Bank 1999j. Food processing, beverage prepara- port market share of processed nuts. tion, and trading activities have begun to multiply in the periur- ban villages, helping to supplement household incomes (Baker Background papers 1999). Barrett, Scott. “Facilitating International Cooperation.” 30. Brautigam 1997. Bourguignon, Francois. “Crime as a Social Cost of Poverty and 31. World Bank 1999j. Inequality: A Review Focusing on Developing Countries.” 32. Lele and Christiansen 1989. Only 20 percent of farm- Burgess, Robin. “Social Protection, Globalization, and Decen- land in Tanzania is in farm blocks of more than 10 hectares. In tralization.” Kenya 43 percent of farmland is in units of over 200 hectares Castles, Stephen. “Impacts of Emigration on Countries of (Tomich, Kilby, and Johnston 1995). 33. Buckley 1997. Origin.” 34. EIU 1998. Choe, Sang-Chuel, and Won Bae Kim. “Globalization and Ur- 35. Carr 1993. More generally in Tanzania, food crops are banization in Korea.” still given priority over cash crops because of the fear that it will Cooper, Richard N. “International Approaches to Global Cli- be impossible to sell cash crops and buy food at the end of the mate Change.” growing season. Deaton, Angus. “Global and Regional Effects of Aging and of 36. On average, Sub-Saharan African countries pay freight Demographic Change.” charges on their exports that are 20 percent higher than those Fay, Marianne. “How Many Tiers? How Many Jurisdictions? A paid by exporters in East Asia. Additional costs abound in Tan- Review of Decentralization Structures across Countries.” zania because of delays in customs clearance, inefficient cargo Glaeser, Edward L., and Jordan Rappaport. “Cities and handling facilities, and high forwarding charges (Hertel, Mas- Governments.” ters, and Elbehri 1998). The building of a cargo center with re- Henderson, Vernon. “Urbanization In Developing Countries.” frigeration facilities at Nairobi’s Jomo Kenyatta Airport will Hohenberg, Paul M. “Urban Systems and Economic Develop- provide the capacity for handling 160,000 tons per year of ment in Historical Perspective: The European Long Term horticultural and other produce. This will be helpful for Tan- and Its Implications.” zanian producers, but more capacity at Arusha’s airport or at Hufbauer, Gary, and Barbara Kotschwar. “The Future Course Dar es Salaam might be more advantageous ( Financial Times, of Trade Liberalization.” “Kenyan Air Cargo Capacity Boosted.” June 2, 1999). Hughes Hallett, A. J. “Policy Co-ordination: Globalization or 37. Islam 1997. Localization in International Monetary Arrangements?” 38. Gautam and Anderson 1998. Litan, Robert. “Toward a Global Financial Architecture for the 39. Tendler 1997. 21st Century.” 40. The contrast with East Asia is instructive. In China’s Fu- Lucas, Robert E. B. “Internal Migration and Urbanization: Re- jian and Guandong provinces as well as in Taiwan (China), the cent Contributions and New Evidence.” “relational networks” created in the hinterland of towns have Malpezzi, Stephen. “The Regulation of Urban Development: served as the basis of highly successful industrial clusters that Lessons from International Experience.” make effective use of subcontracting arrangements and reach Mohan, Rakesh. “Financing of Sub-National Public Investment out to international markets. See Hayami (1998) and the dis- in India.” cussion in chapter 1. Satterthwaite, David, and Diana Mitlin. “Urban Poverty: Some 41. After reviewing constraints on growth in the manufac- Thoughts about Its Scale and Nature and about Responses turing sector, Tybout (1998) concludes that the primary prob- to It by Community Organizations, NGOs, Local Govern- lems revolve around uncertainty about policies and demand, ments and National Agencies.” weak laws, and corruption. Smoke, Paul. “Strategic Fiscal Decentralization in Developing 42. Transparency International 1998. Countries: Issues and Cases.” 43. Bennell 1997. Srinivasan, T. N. “Think Globally, Act Locally! Development 44. Yudkin 1999; Nature 1999. Policy at the Turn of the Century.” 45. Cole and Phelan (1999) associate the research output of Stren, Richard. “Urban Governance and Politics in a Global a country with its wealth and with the number of research sci- Context: The Growing Importance of the Local.” entists, but also with a culture that attaches a high value to sci- entific achievement and with competition among universities References to attract talent of the highest caliber. 46. See Wambugu (1999) and Lipton (1999). Between 1994 The word processed describes informally reproduced works that and 1998, the acreage devoted to transgenic crops worldwide may not be commonly available through libraries. rose from 4 million acres to 70 million acres. But little of this was in Africa, where the yields of staples such as potatoes and AB Assesores. 1998. “Sub-Sovereign Capital Market Transac- maize are the lowest for any region. Cramer (1999) draws at- tions in Latin America: Six Case Studies.” World Bank Joint tention to the importance of research on cashew nut tree vari- Program on Sub-Sovereign Capital Markets. World Bank, eties, grafting techniques, and processing methods so as to en- Washington, D.C. Processed.    Abel, I., and L. Szakadat. 1997–98. “Bank Restructuring in ———. 1994a. “Household Demand for Improved Water and Hungary.” Acta Oeconomica 49(1–2): 157–90. Sanitation in a Large Secondary City: Findings from a Study Ablo, Emmanuel, and Ritva Reinikka. 1998. “Do Budgets in Gujranwala, Pakistan.” Habitat International 18(1). Really Matter? Evidence from Public Spending on Educa- ———. 1994b. “The Economics of Household Response to In- tion and Health in Uganda.” Policy Research Working adequate Water Supplies: Evidence from Pakistan.” Third Paper 1926. World Bank, Africa Region, Washington, World Planning Review 16(1). D.C. Altaf, Mir A., and Jeffrey A. Hughes. 1994. “Measuring the De- ADBI (Asian Development Bank Institute). 1998. “Executive mand for Improved Urban Sanitation Services: Results of a Summary of Workshop on Economic Monitoring of Finan- Contingent Valuation Study in Ouagadougou, Burkina cial Systems in East and Southeast Asia.” Tokyo. Processed. Faso.” Urban Studies 31(10). Ades, Alberto, and E. Glaeser. 1995. “Trade and Circuses: Ex- Altaf, Mir A., A. Ercelawn, K. Bengali, and A. Rahim. 1993. plaining Urban Giants.” Quarterly Journal of Economics “Poverty in Karachi: Incidence, Location, Characteristics, 110(1): 195–258. and Upward Mobility.” Pakistan Development Review 32(2). Afonso, José Roberto Rodrigues. 1992. “Federalismo Fiscal e Ambio. 1995. “Malaria and Malaria Potential Transmission to Reforma Institutional: Falácias, Conquistas e Descentraliza- Climate.” 24(6): 200–07. ção.” Discussion Paper 3. Centro de Estudos de Politicas Amjad, Rashid. 1989. “To the Gulf and Back: Studies on the Publicas, Rio de Janeiro. Economic Impact of Asian Labour Migration.” International Afsah, Shakeb, Benoit Laplante, and David Wheeler. 1997. Labour Organisation, Geneva. “Regulation in the Information Age: Indonesian Public In- Anderson, Bob, and Paul Brenton. 1998. “The Dollar, Trade, formation Program for Environmental Management.” Technology, and Inequality in the USA.” National Institute World Bank, Development Research Group, Washington, Economic Review 166 (October): 78–86. D.C. Processed. Anderson, J. 1998. “The Kyoto Protocol on Climate Change: Aghion, Philippe, and Jeffrey G. Williamson.1998. Growth, In- Background, Unresolved Issues and Next Steps.” Resources equality and Globalization. Cambridge: Cambridge Univer- for the Future, Washington, D.C. Processed. sity Press. Anderson, Kym. 1999. “The WTO Agenda for the New Mil- Ahmad, Ehtisham, and Jon Craig. 1997. “Intergovernmental lennium.” The Economic Record 75(228). Transfers.” In Fiscal Federalism in Theory and Practice, edited Anderson, Kym, and Warwick McKibbin. 1997. “Reducing by Teresa Ter-Minassian. Washington, D.C.: International Coal Subsidies and Trade Barriers: Their Contribution to Monetary Fund. Greenhouse Gas Abatement.” World Bank, Washington, Ahmad, Junaid K. 1999. “Decentralizing Borrowing Powers.” D.C. Processed. Poverty Reduction and Economic Management Network Anderson, Letty. 1988. “Fire and Disease: The Development of (PREM) Notes 15. World Bank, Washington, D.C. Water Supply Systems in New England, 1879–1900.” In Ahmad, Junaid K., and Charles E. McLure, Jr. 1994. “Intergov- Technology and the Rise of the Networked City in Europe and ernmental Fiscal Relations in South Africa: A Case Study of America, edited by Joel A. Tarr and Gabriel Dupuy. Phila- Policy-Induced Dysfunction.” World Bank, Washington, delphia: Temple University Press. D.C. Processed. Anderson, Mats. 1998. “Improving Urban Quality of Life in AIDS Analysis. 1998. “World Population Profile Reveals Bleak Europe and Central Asia.” World Bank, Europe and Cen- Data.” 8(3/June): 1–2. tral Asia Region (ECA) Urban Sector, Washington, D.C. Alderman, Harold. 1998. “Do Local Officials Know Something Processed. We Don’t? Decentralization of Targeted Transfers in Alba- Antarctica Project. 1999. “The Antarctic Treaty System.” nia.” World Bank, Washington, D.C. Processed. Washington, D.C. Available online at http:// www.asoc.org/. Alesina, Alberto. 1998. “Too Large and Too Small Govern- Antweiler, Werner, Brian R. Copeland, and M. Scott Taylor. ments.” Conference on Economic Policy and Equity. Inter- 1998. “Is Free Trade Good for the Environment?” Working national Monetary Fund, Washington, D.C. Paper 6707. National Bureau of Economic Research, Cam- Alesina, Alberto, and Robert Perotti. 1997. “Fiscal Adjust- bridge, Mass. ments in OECD Countries: Composition and Macroeco- Attanasio, Orazio, and James Banks. 1998. “Trends in House- nomic Effects.” IMF Staff Papers 44(2): 210–48. Washing- hold Saving Don’t Justify Tax Incentives to Book Saving.” ton, D.C. Economic Policy 27(October). ———. 1998. “Economic Risk and Political Risk in Fiscal Aw, Bee-Yan, and Geetra Batra. 1998. “Technological Capabil- Unions.” The Economic Journal 108: 989–1008. ity and Firm Efficiency in Taiwan (China).” World Bank Alesina, Alberto, and E. Spolaore. 1997. “On the Number and Economic Review 12(1): 59–80. Size of Nations.” Quarterly Journal of Economics 112: Ayres, Robert L. 1997. Crime and Violence as Development Is- 1027–56. sues in Latin America and the Caribbean. World Bank, Latin Alesina, Alberto, and Romain Wacziarg. 1998. “Is Europe Going America and the Caribbean Studies, Washington, D.C. Too Far?” Massachusetts Institute of Technology. Processed. Processed. Altaf, Mir A. 1983. “The Strategic Implications of Varying En- Aziz, Jahangir, and Robert F. Wescott. 1997. “Policy Comple- vironments, Aspects of Decisionmaking under Instability.” mentarities and the Washington Consensus.” Working Paper Ph.D. diss., Stanford University. Processed. 97/118. International Monetary Fund, Washington, D.C.      ⁄  Bahl, Roy W. 1999a. “Intergovernmental Transfers in Devel- Bertaud, Alain, and Bertrand Renaud. 1997. “Socialist Cities oping and Transition Countries: Principles and Practice.” without Land Markets.” Journal of Urban Economics 41: School of Political Studies, Georgia State University, At- 137–51. lanta. Processed. Bertaud, Alain, Robert Buckley, Margret Thalwitz, and Cracow ———. 1999b. Fiscal Policy in China: Taxation and Intergov- Real Estate Institute. 1997. “Cracow in the Twenty-first ernmental Fiscal Relations. San Francisco: The 1990 Institute. Century: Princes or Merchants?” Paper presented to Lincoln Bahl, Roy W., and Johannes F. Linn. 1992. Urban Public Finance Institute Conference on Land Prices, Land Information Sys- in Developing Countries. New York: Oxford University Press. tems, and the Market for Land Information. Cambridge. Baker, Jonathan. 1999. “Rural-Urban Links and Economic Dif- World Bank, Washington, D.C. Processed. ferentiation in Northwest Tanzania.” African Rural and Berthélemy, Jean-Claude, and Aristomène Varoudakis. 1996. Urban Studies 3(1): 25–48. “Policies for Economic Take-off.” Policy Brief 12: 1–32. Or- Baldwin, Richard E., and Philippe Martin. 1999. “Two Waves ganisation for Economic Co-operation and Development, of Globalization: Superficial Similarities, Fundamental Dif- Paris. ferences.” Working Paper 6904. National Bureau of Eco- Berthélemy, Jean-Claude, Sébastien Dessus, and Aristomène nomic Research, Cambridge, Mass. Varoudakis. 1997. “Capital humain, ouverture extérieure et Baldwin, Richard E., and Anthony J. Venables. 1995. “Regional croissance: estimation sur données de panel d’un modèle a Economic Integration.” In Handbook of International Eco- coefficients variables.” Policy Brief 121: 1–32. Organisation nomics, edited by Gene Grossman and Kenneth Rogoff, 3: for Economic Co-operation and Development, Paris. 1597–1643. Amsterdam: Elsevier Science B.V. Bhagwati, Jagdish. 1991. The World Trading System at Risk. Baliño, Tomás J. T., Adam Bennett, and Eduardo Borensztein. Princeton: Princeton University Press. 1999. “Monetary Policy in Dollarized Economies.” Occa- Bird, Richard M. 1999. “Rethinking Tax Assignment: The sional Paper 171. International Monetary Fund, Washing- Need for Better Subnational Taxes.” International Mone- ton, D.C. tary Fund, Washington, D.C. Processed. Barkin, Samuel, and George Shambaugh. 1996. “Common- Bird, Richard M., and Pierre-Pascal Gendron. 1997. “Dual Pool Resources and International Environmental Politics.” VATs and Cross-Border Trade: Two Problems, One Solu- Environmental Politics 5(3): 429–47. tion?” International Tax and Public Finance 5: 429–42. Baron, James. N., and Michael T. Hannan. 1994. “The Impact Bird, Richard M., and Edgard R. Rodriguez. 1999. “Decentral- of Economics on Contemporary Sociology.” Journal of Eco- ization and Poverty Alleviation: International Experience nomic Literature 32: 1111–46. and the Case of the Philippines.” Department of Econom- Barrett, Scott. 1998a. “Facilitating International Cooperation.” ics, University of Toronto. Processed. Background Paper for World Development Report 1999/ Bird, Richard M., and François Vaillancourt, eds. 1999. Fiscal 2000. World Bank, Washington, D.C. Decentralization in Developing Countries. Cambridge: Cam- ———. 1998b. “Montreal v. Kyoto: International Cooperation and the Global Environment.” Prepared for the Office of De- bridge University Press. velopment Studies, UNDP (United Nations Development Bird, Richard M., Robert D. Ebel, and Christine I. Wallich, Programme) Project on Global Public Goods. Processed. eds. 1995. Decentralization of the Socialist State . Washing- ———. 1998c. “The Credibility of Trade Sanctions in Inter- ton, D.C.: World Bank. national Environmental Agreements.” London Business Black, D., and J. V. Henderson. 1998. “Urban Evolution in the School, London. Processed. USA.” Department of Economics, Brown University, Provi- Barro, Robert. 1991. “Economic Growth in a Cross Section of dence R.I. Processed. Countries.” Quarterly Journal of Economics 106: 407–43. Blackett, Isabel C. 1994. “Low-Cost Urban Sanitation in Barth, James R., Gerard C. Caprio, and Ross Levine. 1999. “Fi- Lesotho.” Water and Sanitation Discussion Paper Series 10. nancial Regulation and Performance: Cross-Country Evi- UNDP–World Bank Water and Sanitation Program. World dence.” Policy Research Working Paper 2037. World Bank, Bank, Washington, D.C. Washington, D.C. Blomström, Magnus, and Ari Kokko. 1997. “Regional Integra- Begg, Iain. 1999. “Cities and Competitiveness.” Urban Studies tion and Foreign Direct Investment: A Conceptual Frame- 36(5–6). work and Three Cases.” Policy Research Working Paper Begum, Shamshad, and A. F. M. Shamsuddin. 1998. “Exports 1750. World Bank, Washington, D.C. and Economic Growth in Bangladesh.” Journal of Develop- Boddy, Martin. 1999. “Geographical Competitiveness: A Cri- ment Studies 35(1/October): 89–114. tique.” Urban Studies 36(5–6). Bennell, Paul. 1997. “Foreign Direct Investment in Africa: Boeninger, Edgardo. 1992. “Governance and Development: Rhetoric and Reality.” SAIS Review (Summer/Fall): 127–40. Issues and Constraints.” In Proceedings of the World Bank An- Bergsten, C. Fred. 1998. “A New Strategy for the Global Cri- nual Conference on Development Economics, edited by Lawrence sis.” International Economics Policy Brief. Institute for In- Summers and Shekhar Shah. Washington, D.C.: World Bank. ternational Economics, Washington, D.C. Bohning W. R., and R. Zegers de Beiji. 1995. “The Integration Bernard, Andrew, and Steven Durlauf. 1996. “Interpreting of Migrant Workers in the Labour Market: Policies and Tests of the Convergence Hypothesis.” Journal of Economet- Their Impact.” International Migration Papers 8: 1–59. In- rics 71: 161–73. ternational Labour Office, Geneva.    Bolbol, Ali A. 1999. “Arab Trade and Free Trade: A Preliminary Buckley, Robert. 1999. 1998 Annual Review of Development Analysis.” International Journal of Middle Eastern Studies 31: Effectiveness. Washington, D.C.: World Bank. 3–17. Buckley, Robert, and Stephen Mayo. 1989. “Housing Policy in Boniface, Pascal. 1998. “The Proliferation of States.” The Wash- Developing Economies: Evaluating the Macroeconomic Im- ington Quarterly 21(3). pacts.” Review of Urban and Regional Development Studies Bordo, Michael D., Barry Eichengreen, and Douglas A. Irwin. 2(27). 1999. “Is Globalization Today Really Different Than Glob- Burgess, Robin. 1998. “Social Protection, Globalization, and De- alization a Hundred Years Ago?” Paper for the Brookings In- centralisation.” Department of Economics, London School stitution Trade Policy Forum on Governing in a Global of Economics. Processed. Economy. Washington D.C., April 15–16. ———. 1999. “Social Protection, Globalization and Decen- Bordo, Michael D., Barry Eichengreen, and Jongwoo Kim. tralisation.” Background Paper for World Development Re- 1998. “Was There Really an Earlier Period of International port 1999/2000. World Bank, Washington, D.C. Financial Integration Comparable to Today?” Working Burki, Shahid Javed, and Guillermo E. Perry. 1998. Beyond the Paper 6738. National Bureau of Economic Research, Cam- Washington Consensus: Institutions Matter. World Bank Latin bridge, Mass. American and Caribbean Studies: Viewpoints. Washington, Borensztein, Eduardo, José De Gregorio, and Johng-wha Lee. D.C.: World Bank. 1998. “How Does Foreign Direct Investment Affect Eco- Burki, Shahid Javed, Guillermo E. Perry, and William nomic Growth?” Journal of International Economics 45(1): Dillinger. 1999. Beyond the Center: Decentralizing the State. 115–35. World Bank Latin American and Caribbean Studies: View- Borjas, George J. 1998. “Economic Research on the Determi- points. Washington, D.C.: World Bank. nants of Immigration: Lessons for the European Union.” Burtless, Gary, Robert Z. Lawrence, Robert E. Litan, and Department of Economics, Harvard University, Cambridge, Robert J. Shapiro. 1998. Globaphobia. Washington, D.C.: Mass. Processed. The Brookings Institution Press. Bougheas, Spiros, Panicos O. Demetriades, and Edgar L. W. Business Week. 1998. “A Fresh Shot at Malaria.” September 21. Morgenroth. 1999. “Infrastructure, Transport Costs and ———. 1999. “Fresh Strains of Unzappable Germs.” August 2. Trade.” Journal of International Economics 47: 169–89. Butler, James H. 1999. “A Record of Atmospheric Halocarbons Bourguignon, François. 1998. “Crime as a Social Cost of during the Twentieth Century from Polar Air.” Nature Poverty and Inequality: A Review Focusing on Developing 339(June 24): 749–55. Countries.” Background Paper for World Development Re- Cairncross, Frances. 1997. The Death of Distance: How the Com- port 1999/2000. World Bank, Washington, D.C. munications Revolution Will Change Our Lives. Cambridge: Bradbury, L. Katherine, Yolanda K. Kodrzycki, and Robert Harvard Business School Press. Tannenwald. 1997. “The Effects of State and Local Public Caldeira, Teresa P. R. 1996. “Building up Walls: The New Pat- Policies on Economic Development: An Overview.” New tern of Spatial Segregation in São Paulo.” International So- England Economic Review March/April: 1–12. cial Science Journal (147/March). Bradley, Rebecca, and Joshua S. Gans. 1998. “Growth in Aus- Calomiris, Charles. 1997. The Postmodern Bank Safety Net: tralian Cities.” Economic Record 74: 266–78. Lessons from Developed and Developing Countries. Washing- Brautigam, Deborah. 1997. “Substituting for the State: Institu- ton, D.C.: American Enterprise Institute. tions and Industrial Development in Eastern Nigeria.” World ———. 1999. “How to Invent a New IMF.” The International Development 25(7): 1081–93. Economy (January/February): 32ff. Breton, Albert. 1996. Competitive Governments. Cambridge: Calvo, Sarah. 1999. “Reducing Vulnerability to Speculative At- Cambridge University Press. tacks.” Poverty Reduction and Economic Management Net- Bridges, Amy. 1997. Morning Glories: Municipal Reform in the work (PREM) Economic Policy Notes 16. World Bank, Southwest. Princeton: Princeton University Press. Washington, D.C. Brockerhoff, Martin, and E. Brennan. 1998. “The Poverty of Campbell, Tim E. 1998. “The Quiet Revolution: The Rise of Cities in Developing Regions.” Population and Development Political Participation and Local Government with Decen- Review 24(1/March). tralization in Latin America and the Caribbean.” World Brown, Lynn, and Alex F. McCalla. 1998. “Global Urbaniza- Bank, Washington, D.C. Processed. tion Trends: Implications for Food Systems and Food Ser- Caprio, Gerard C. 1998. “International Financial Integration: vices.” World Bank, Washington, D.C. Processed. Pitfalls and Possibilities.” World Bank, Washington, D.C. Brunetti, Aymo. 1997. Politics and Economic Growth: A Cross- Processed. Country Data Perspective. Development Centre Studies. Paris: Caprio, Gerard C., and Daniela Klingebiel. 1996. “Bank Insol- Organisation for Economic Co-operation and Development. vencies: Cross-Country Experience.” Policy Research Work- Bryant, Ralph. 1995. International Coordination of National ing Paper 1620. World Bank, Washington, D.C. Stabilization Policies. Washington, D.C.: The Brookings ———. 1999. “Table of Episodes of Major Bank Insolvencies.” Institution Press. World Bank, Washington, D.C. Processed. Buckley, Graeme. 1997. “Microfinance in Africa: Is It Either Carey, John M. 1997. “Institutional Designs and Party Sys- the Problem or the Solution?” World Development 25(7): tems.” In Consolidating the Third Wave Democracies. Balti- 1063–80. more: The Johns Hopkins University Press.      ⁄  Caron, Mary. 1999. “The Politics of Life and Death.” World Claudio, L., L. Tulton, J. Doucette, P. J. Landrigan. 1999. Watch (May/June): 30–38. “Socioeconomic Factors and Asthma Hospitalization Rates Carr, Stephen. 1993. Improving Cash Crops in Africa: Factors In- in New York City.” Asthma 36(4): 343–50. fluencing the Productivity of Cotton, Coffee, and Tea Grown Clerides, Sofronis, Saul Lach, and James Tybout. 1998. “Is by Smallholders. Technical Paper 216. Washington, D.C.: Learning-by-Exporting Important? Micro-Dynamic Evi- World Bank. dence from Colombia, Mexico and Morocco.” Quarterly Carrington, William J., and Enrica Detragiache. 1998. “How Journal of Economics 113: 903–47. Big Is the Brain Drain?” Working Paper 98/102. Interna- Cline, William R. 1997. Trade and Income Distribution. Wash- tional Monetary Fund, Washington, D.C. ington, D.C.: Institute for International Economics. Caselli, Francesco, G. Esquivel, and F. Lefort. 1996. “Re- Coelho, Magda Prates. 1996. “Urban Governance in Brazil.” In opening the Convergence Debate: A New Look at Cross- Cities and Governance: New Directions in Latin America, Asia Country Growth Empirics.” Journal of Economic Growth 1 and Africa, edited by Patricia L. McCarney. Toronto: Uni- (September): 363–89. versity of Toronto Press. Castles, Stephen. 1998. “Impacts of Emigration on Countries Coffee, Joyce Elena. 1999. “Innovations in Municipal Service of Origin.” Background Paper for World Development Report Delivery: The Case of Vietnam’s Haiphong Water Supply 1999/2000. World Bank, Washington, D.C. Company.” Master’s diss. Massachusetts Institute of Tech- Caves, Richard E. 1996. Multinational Enterprise and Economic nology, Cambridge, Mass. Processed. Analysis, 2nd ed. Cambridge: Cambridge University Press. Cohen, Barney. 1998. “The Emerging Fertility Transition in Cesar, Herman. 1998. “Indonesian Coral Reefs: A Precious but Sub-Saharan Africa.” World Development 26. Threatened Resource.” In Coral Reefs: Challenges and Oppor- Cohen, Daniel. 1998. The Wealth of the World and the Poverty tunities for Sustainable Management, edited by Maria Hatzi- of Nations. Cambridge, Mass.: MIT Press. olos, Anthony Hooten, and Martin Fodor. Washington, Cohen, Mitchell L. 1992. “Epidemiology of Drug Resistance: D.C.: World Bank. Implications for a Post Microbial Era.” Science. August. Chaplin, Susan E. 1999. “Cities, Sewers, and Poverty: India’s Cohen, Roberta, and Francis M. Deng. 1998. Masses in Flight: Politics of Sanitation.” Environment and Urbanization 11(1): The Global Crisis of Internal Displacement. Washington, D.C.: The Brookings Institution Press. 145–58. Cole, Stephen, and Thomas J. Phelan. 1999. “The Scientific Charnovitz, Steve. 1996. “Trade Measures and the Design of Productivity of Nations.” Minerva 37(1): 1–23. International Regimes.” Journal of Environment and Devel- Coleman, James, and Thomas Hoffer. 1987. Public and Private opment 5(2): 168–96. High Schools: The Impact of Communities. New York: Basic Chhabra, S. K., C. K. Gupta, P. Chhabra, and S. Rajpal. 1998. Books. “Prevalence of Bronchial Asthma in Schoolchildren of Colgan, Charles S. 1995. “International Regulation of State and Delhi.” Journal of Asthma 35(3). Local Subsidies.” Economic Development Quarterly 9(2). Chinese State Council. 1994. “A White Paper on Chinese Pop- Collier, Paul, and David Dollar. 1998. “Aid Allocation and ulation, Environment and Development in the Twenty-first Poverty Reduction.” World Bank, Washington, D.C. Century, Adopted at the Sixteenth Regular Meeting of the Processed. State Council.” March 25. Translated excerpts reprinted in Conger, Lucy. 1999. “Porto Alegre: Where the Public Controls Chinese Environment and Development, 7(4): 74–95. the Purse Strings.” Urban Age 6(4): 4–5. Choe, Sang-Chuel, and Won Bae Kim. 1999. “Globalization Connelly, James. 1996. “Review of Making Nature, Shaping and Urbanization in Korea.” Background Paper for World Culture: Plant Biodiversity in Global Context.” Environ- Development Report 1999/2000. World Bank, Washington, mental Politics 5(4): 770–1. D.C. Cooke, Philip, and Kevin Morgan. 1998. The Associational Chomitz, K., and K. Kumari. 1998. “The Domestic Benefits of Economy. Firms, Regions, and Innovation. Oxford: Oxford Tropical Forests: A Critical Review.” World Bank Research University Press. Observer 13(1): 13–35. Cooper, Richard N. 1998. “International Approaches to Global Christianson, Gale E. 1999. Greenhouse. New York: Walker and Climate Change.” Background Paper for World Development Company. Report 1999/2000. World Bank, Washington, D.C. Chumacero, Romulo, Raul Laban, and Felipe Larrain. 1996. ———. 1999. “Should Capital Controls Be Banished?” De- “What Determines Capital Inflows: An Empirical Analysis partment of Economics, Harvard University. Processed. for Chile.” Universidad Católica de Chile, Santiago. Unpub- Costa, L. Dora. 1998. The Evolution of Retirement. Chicago: lished manuscript. University of Chicago Press. Claessens, Stijn, and Moon-Whoan Rhee. 1994. “The Effects Cour, Jean-Marie. 1998a. “Draft Proposal for a Data System of Barriers on Equity Investments in Developing Countries.” for the Durban Metropolitan Area.” Club du Sahel. OECD Policy Research Working Paper 1263. World Bank, Wash- (Organisation for Economic Co-operation and Develop- ington, D.C. ment), Paris. Processed. Claessens, Stijn, Asli Demirgüç-Kunt, and Harry Huizinga. ———. 1998b. “First Lessons from the Ecolog Program.” Note 1998. “How Does Foreign Entry Affect the Domestic Bank- from the Club du Sahel Secreteriat. OECD (Organisation ing Market?” World Bank, Washington, D.C. Processed. for Economic Co-operation and Development), Paris.    Covarrubias, Alvaro. 1999. “Lending for Electric Power in Sub- Demirgüç-Kunt, Asli, and Ross Levine. 1995. “Stock Market Saharan Africa.” World Bank, Washington, D.C. Processed. Development and Financial Intermediaries: Stylized Facts.” Cowen, Penelope Brook, and Nicola Tynan. 1999. “Reaching World Bank, Washington, D.C. Processed. the Urban Poor with Private Infrastructure.” Finance, Pri- Dervis, Kemal, and Nemat Shafiq. 1998. “The Middle East and vate Sector, and Infrastructure Network Viewpoint 188. North Africa: A Tale of Two Futures.” Middle East Journal World Bank, Washington, D.C. 52(4): 505–16. Coyle, Diane. 1998. The Weightless World. Cambridge, Mass.: Devarajan, Shantayanan, William Easterly, and Howard Pack. MIT Press. 1999. “Is Investment in Africa Too Low or Too High?” Cramer, Christopher. 1999. “Can Africa Industrialize by Pro- World Bank, Washington, D.C. Processed. cessing Primary Commodities? The Case of Mozambican Dewatripont, Mathias, and Jean Tirole. 1994. The Prudential Cashew Nuts.” World Development 27(7): 1247–66. Regulation of Banks. Cambridge: MIT Press. Crystal, David. 1997. English as a Global Language. Cambridge: Diamond, Larry. 1996. “Is the Third Wave Over?” Journal of Cambridge University Press. Democracy 7(3). Curtis, Valerie, and Bernadette Kanki. 1998. “Bednets and ———. 1999. Developing Democracy: Toward Consolidation. Malaria.” Africa Health. May. Baltimore: Johns Hopkins University Press. Dahl, Robert A. 1971. Polyarchy: Participation and Opposition. Dillinger, William. 1992. “Urban Property Tax Reform”. New Haven: Yale University Press. Urban Management Program Working Paper 1. World ———. 1986. “Federalism and the Democratic Process.” In Bank, Washington, D.C. Democracy, Identity and Equality. Oslo: Norwegian Univer- ———. 1997. “Brazil’s State Debt Crisis: Lessons Learned.” sity Press. Latin America and the Caribbean Region Economic Notes. Dailami, Mansoor, and Danny Leipziger. 1998. “Infrastructure World Bank, Washington, D.C. Project Finance and Capital Flows: A New Perspective.” Dillinger, William, and Steven B. Webb. 1999a. “Decentraliza- World Bank, Washington D.C. tion and Fiscal Management in Colombia.” Policy Research Davoodi, Hamid, and Heng-fu Zou. 1998. “Fiscal Decentral- Working Paper 2122. World Bank, Washington, D.C. ization and Economic Growth: A Cross-Country Study.” ———. 1999b. “Fiscal Management in Federal Democracies: World Bank, Policy Research Department, Washington, Argentina and Brazil.” Policy Research Working Paper D.C. Processed. 2121. World Bank, Washington, D.C. de Figueiredo, Rui J. P. Jr., and Barry R. Weingast. 1998. “Self- Dobson, Wendy, and Pierre Jacquet. 1998. Financial Services Enforcing Federalism: Solving the Two Fundamental Liberalization in the WTO. Washington, D.C.: Institute for Dilemmas.” Department of Political Science, Stanford Uni- International Economics. versity, Stanford, Calif. Processed. Dollar, David. 1999. “The Comprehensive Development de Fontaubert, A. Charlotte. 1996. “The United Nations Con- Framework and Recent Development Research.” World ference on Straddling Fish Stocks and Highly Migratory Fish Stocks: Another Step in the Implementation of the Law Bank, Washington, D.C. Processed. of the Sea Convention.” Living Resources. Donahue, John D. 1997. “Tiebout? Or Not Tiebout? The Mar- De Mello, Luiz R. 1997. “Foreign Direct Investment in Devel- ket Metaphor and America’s Devolution Debate.” Journal oping Countries and Growth: A Selective Survey.” Journal of Public Economics 11: 73–82. of Development Studies 34: 1–34. Dooley, Michael P. 1996. “A Survey of Literature on Controls Deardorff, Alan V. 1998. “Fragmentation in Simple Trade over International Capital Transactions.” IMF Staff Papers Models.” Department of Economics, University of Michi- 43: 639–87. International Monetary Fund, Washington, gan, Ann Arbor. Processed. D.C. Deardorff, Alan V., Saul H. Hymans, Robert M. Stern, and Dornbusch, Rudi. 1998. “Cross-Border Payments Taxes and Chong Xiang. 1998. “The Economic Outlook for U.S. Alternative Capital Account Regimes.” In Capital Account Trade in Services, 1999–2001.” Department of Economics, Regimes and the Developing Countries, edited by G. K. University of Michigan, Ann Arbor. Processed. Helleiner. New York: St. Martin’s Press. Deaton, Angus. 1998. “Global and Regional Effects of Ag- Douglass, Mike. 1992. “The Political Economy of Urban ing and of Demographic Change.” Background Paper for Poverty and Environmental Management in Asia: Access, World Development Report 1999/2000. Princeton University. Empowerment and Community-Based Alternatives.” Envi- Processed. ronment and Urbanization 4(2). Deininger, Klaus, and Lyn Squire. 1996. “A New Data Set Drabek, Zdenek, and Sam Laird. 1998. “The New Liberalism: Measuring Income Inequality.” World Bank Economic Re- Trade Policy Developments in Emerging Markets.” Journal view 10(September): 565–91. of World Trade 32(5): 241–69. Del Amo, Julia, and others. 1999. “Does Tuberculosis Acceler- Drèze, Jean, and Amartya Sen. 1995. India: Economic Develop- ate the Progression of HIV Disease? Evidence from Basic ment and Social Opportunity. New York: Oxford University Science and Epidemiology.” AIDS 13(10). Press. Demirgüç-Kunt, Asli, and Enrica Detragiache. 1998. “Finan- Dumais, Guy, Glenn Ellison, and Edward L Glaeser. 1997. cial Liberalization and Financial Fragility.” Working Paper “Geographic Concentration as a Dynamic Process.” Na- 98/83. International Monetary Fund, Washington, D.C. tional Bureau of Economic Research, Cambridge, Mass.      ⁄  Easterly, William. 1999. “Life during Growth.” World Bank, EIU (Economist Intelligence Unit). 1998. “Country Profile: Washington, D.C. Processed. Tanzania.” London. Easterly, William, and Stanley Fischer. 1995. “The Soviet Eco- Ekpo, Akpan H. and John E. U. Ndebbio. 1998. “Local Gov- nomic Decline.” World Bank Economic Review 9(Septem- ernment Fiscal Operations in Nigeria.” Research Paper 73. ber): 341–71. African Economic Research Consortium, Nairobi. Easterly, William, and Ross Levine. 1997. “Africa’s Growth Elkins, James. 1999. “Chlorofluorocarbons (CFCs).” In The Tragedy: Policies and Ethnic Divisions.” Quarterly Journal Chapman and Hall Encyclopedia of Environmental Science, of Economics 112(November): 1203–50. edited by David Alexander and Rhodes Fairbridge. New Eberstadt, Nicholas. 1998. “Asia Tomorrow, Gray and Male.” York: Chapman and Hall. National Interest 53(Fall). Elster, Jon. 1988. “Is There (or Should There Be) a Right to EBRD (European Bank for Reconstruction and Development). Work?” In Democracy and the Welfare State, edited by A. 1997. Transition Report 1997: Economic Performance and Guttman. Princeton: Princeton University Press. Growth. London. ———. 1989. The Cement of Society. Cambridge: Cambridge ———. 1998. Transition Report 1998: Economic Performance University Press. and Growth. London. Elster, Jon, and Rune Slagsrad. 1993. Constitutionalism and The Economist. 1998a. “Lost without a Trace.” August 1. Democracy. Cambridge: Cambridge University Press. ———. 1998b. “Recipes for an AIDS vaccine.” July 14. Eskeland, Gunnar, and Tarhan Feyzioglu. 1994. “Is Demand ———. 1998c. “Repositioning the WHO.” May 9. for Polluting Goods Manageable? An Econometric Study of ———. 1998d. “Turtle Soup.” October 17. Car Ownership and Use in Mexico.” Policy Research Work- ———. 1999a. “Seeds of Discontent.” February 20. ing Paper 1309. World Bank, Washington, D.C. ———. 1999b. “Throwing Sand in the Gears.” January 30. Espinosa, Lair, and Oscar A. López Rivera. 1994. “UNICEF’s ———. 1999c. “Trimmed, Not Axed.” February 27. Urban Basic Services Program in Illegal Settlements in ———. 1999d. “A Survey of Business and the Internet.” June 26. Guatemala City.” Environment and Urbanization 6 (2). ———. 1999e. “International Aid.” July 3. Evanoff, Douglas D. 1998. “Global Banking Crises: Common- ———. 1999f. “Global Disaster.” January 2. alities, Mistakes, and Lessons.” In Preventing Bank Crises: Edwards, Sebastian. 1998a. “Capital Flows, Real Exchange Lessons from Recent Global Bank Failures, edited by Gerard C. Rates, and Capital Controls: Some Latin American Experi- Caprio. Washington, D.C.: Federal Reserve Bank of Chicago ences.” University of California, Los Angeles. Unpublished and the World Bank Economic Development Institute. manuscript. Evans, Peter. 1998. “Looking for Agents of Urban Livability in ———. 1998b. “Openness, Productivity, and Growth: What a Globalized Political Economy.” University of California, Do We Really Know?” The Economic Journal 108(March): Berkeley. Draft. 383–98. Evans, Alison, and William Bataille. 1997. Annual Review of De- Eichengreen, Barry. 1998. “International Economic Policy in velopment Effectiveness. Washington, D.C.: World Bank. the Wake of the Asian Crisis.” Working Paper C98-102. Faguet, Jean-Paul. 1998. “Decentralization and Local Govern- University of California, Berkeley. ment Performance: Improving Public Service in Bolivia.” ———. 1999. Toward a New International Financial Archi- Discussion Paper 999. Centre for Economic Performance, tecture. Washington, D.C.: Institute for International London. Economics. Faini, Riccardo. 1998. “European Migration Policies in Ameri- Eichengreen, Barry, and Albert Fishlow. 1998. “Contending can Perspective.” In Transatlantic Economic Relations in the with Capital Flows: What Is Different about the 1990s?” In Post–Cold War Era, edited by Barry Eichengreen. New York: Capital Flows and Financial Crises, edited by Miles Kahler. Council on Foreign Relations Press. Ithaca: Cornell University Press. Fajnzylber, Pablo, Daniel Lederman, and Norman Louyza. Eichengreen, Barry, and Peter B. Kenen. 1994. “Managing the 1998. Determinants of Crime Rates in Latin America and the World Economy under the Bretton Woods System: An World: An Empirical Assessment. World Bank Latin America Overview.” In Managing the World Economy, edited by Peter and the Caribbean Studies. Washington, D.C.: World Bank. B. Kenen. Washington, D.C.: Institute for International Fallon, Peter, and Zafiris Tzannatos. 1998. “Child Labor: Economics. Issues and Directions for the World Bank.” World Bank, Eichengreen, Barry, and Michael Mussa. 1998. “Capital Ac- Washington, D.C. Processed. count Liberalization: Theoretical and Practical Aspects.” Oc- FAO (Food and Agriculture Organization of the United Na- casional Paper 172. International Monetary Fund, Wash- tions). 1990. Fishery Statistics Catch and Landing 1988. ington, D.C. Rome. Eichengreen, Barry, and Andrew K. Rose. 1998. “Staying Afloat ———. 1998a. Aquacultural Production Statistics 1987–96. When the Wind Shifts: External Factors and Emerging- Rome. Market Banking Crises.” Working Paper 6370. National ———. 1998b. Fishery Statistics Catch and Landing 1996. Bureau of Economic Research, Cambridge, Mass. Rome. Eichengreen, Barry, and others. 1999. “Liberalizing Capital ———. 1999a. Aquacultural Production Statistics. Rome. Movements: Some Analytical Issues.” Economic Issues (Feb- ———. 1999b. Fishery Statistics Catch and Landing 1997. ruary), International Monetary Fund, Washington, D.C. Rome.    Far Eastern Economic Review. 1998. December 24. Fox, William, and Christine Wallich. 1997. “Fiscal Federalism Farmer, Paul. 1999. “TB Superbugs: The Coming Plague on in Bosnia-Herzegovina.” Policy Research Working Paper All Our Houses.” Natural History No. 4. 1714. World Bank, Washington, D.C. Fay, Marianne, Darfy Chaponda, Helen Mbao, and Winnie Francis, Paul A., and others. 1998. Hard Lessons: Primary Mulongo. 1998. “A Review of Local Institutions in Zam- Schools, Community, and Social Capital in Nigeria. Techni- bia.” World Bank, Washington, D.C. Processed. cal Paper 420. Washington, D.C.: World Bank. Feenstra, Robert C. 1998. “Integration of Trade and Disinte- François, Joseph F., Bradley McDonald, and Håkan Nord- gration of Production in the Global Economy.” Journal of ström. 1996. “The Uruguay Round: A Numerically Based Economic Perspectives 12(4): 31–50. Qualitative Assessment.” In The Uruguay Round and the De- Feenstra, Robert C., Dorsati Madani, Tzu-Han Yang, and Chi- veloping Countries, edited by Alan Winters and William Yuan Liang. 1997. “Testing Endogenous Growth in South Martin. Cambridge: Cambridge University Press. Korea and Taiwan.” Working Paper 6028. National Bureau François, Joseph F., and Ian Wooton. 1999. “Trade in Interna- of Economic Research, Cambridge, Mass. tional Transport Services: The Role of Competition.” Feldstein, Martin. 1998. “International Capital Flows: Intro- Center for Economic Policy Research. European Research duction.” Remarks at the National Bureau of Economic Re- Workshop in International Trade. Bergen, Norway, June search Conference on International Capital Flows. Wood- 24–27. stock, Vermont, October 17–18. Frankel, Jeffrey. 1997. Regional Trading Blocs in the World Eco- ———. 1999. “A Self-Help Guide for Emerging Markets.” nomic System. Washington, D.C.: Institute for International Foreign Affairs 78(2): 93–109. Economics. Fernandez, Raquel, and Jonathan Portes. 1998. “Returns to Frankel, Jeffrey A., and David Romer. 1999. “Does Trade Regionalism: An Analysis of Nontraditional Gains from Cause Growth?” American Economic Review 89(3): 379–98. Regional Trade Agreements.” World Bank Economic Review Fredland, Richard A. 1998. “Aids and Development: An Inverse 12(2): 197–220. Correlation?” The Journal of Modern African Studies 36(4): Feshback, Murray. 1999. “Dead Souls.” Atlantic (January): 547–68. 26–27. Freedom House. 1990. Freedom in the World: The Annual Sur- Finger, J. Michael, ed. 1993. Antidumping: How It Works and vey of Political Rights and Civil Liberties. New York. Who Gets Hurt. Ann Arbor: University of Michigan Press. ———. 1991. Freedom in the World: The Annual Survey of Finger, J. Michael, and Ludger Schuknecht. 1999. “Implement- Political Rights and Civil Liberties. New York. ing the Uruguay Round Market Access Agreements.” World ———. 1992. Freedom in the World: The Annual Survey of Bank, Washington, D.C. Political Rights and Civil Liberties. New York. Finger, J. Michael, and L. Alan Winters. 1998. “What Can the ———. 1993. Freedom in the World: The Annual Survey of WTO Do for Developing Countries?” In The WTO as an Political Rights and Civil Liberties. New York. International Organization, edited by Anne O. Krueger. ———. 1994. Freedom in the World: The Annual Survey of Chicago: University of Chicago Press. Political Rights and Civil Liberties. New York. Fischer, Stanley. 1999. “On the Need for an International ———. 1995. Freedom in the World: The Annual Survey of Lender of Last Resort.” Speech delivered to a joint luncheon Political Rights and Civil Liberties. New York. of the American Economic Association and the American ———. 1996a. Freedom in the World: The Annual Survey of Finance Association. New York, January 3. Political Rights and Civil Liberties. New York. Fiszbein, Ariel. 1997. “The Emergence of Local Capacity: ———. 1996b. Freedom Review 27 (January–February). New Lessons from Colombia.” World Development 25: 1029–43. York. Flannery, Mark J. 1998. “Using Market Information in Pruden- ———. 1998. Freedom in the World: The Annual Survey of tial Bank Supervision: A Review of the U.S. Empirical Political Rights and Civil Liberties. New York. Evidence.” Journal of Money, Credit, and Banking 30(3): Freeman, R., and D. Lindauer. 1998. “Why Not Africa?” World 273–305. Bank, Washington, D.C. Processed. Flavin, C. 1997. “The Legacy of Rio.” In State of the World, Freestone, D. 1999. Review of The Implementation and Effec- edited by Lester Brown. New York: W. W. Norton. tiveness of International Environmental Commitments, edited Foreman-Peck, James, and Robert Millward. 1994. Public and by D. Victor, K. Raustiala, and E. Skolnikoff. World Bank, Private Ownership of British Industry 1829–1990. Oxford: Washington, D.C. Processed. Clarendon Press. Freestone, D., and Z. Makuch. 1998. “The New International Fornasari, Francesca, Steven B. Webb, and Heng-Fu Zou. Environmental Law of Fisheries: The 1995 United Nations 1999. “Decentralized Spending and Central Government Straddling Stocks Agreement.” Yearbook of International En- Deficits: International Evidence.” World Bank, Washing- vironmental Law 7. New York. ton, D.C. Processed. Freinkman, Lev. 1998. “Russian Federation: Subnational Bud- Foroutan, Faezeh. 1996. “Turkey, 1976–85: Foreign Trade, In- geting in Russia: Preempting a Potential Crisis.” World dustrial Productivity, and Competition.” In Industrial Evo- Bank, Washington, D.C. Processed. lution in Developing Countries, edited by Mark J. Roberts Freire, Maria E., Marcela Huertas, and Benjamin Darche. 1998. and James Tybout. New York: Oxford University Press. “Subnational Access to the Capital Markets: The Latin      ⁄  American Experience.” World Bank, Washington, D.C. Much Difference Do They Make?” World Development 26: Processed. 1299–1314. French, Hilary. 1997. “Learning from the Ozone Experience.” Gautam, Madhur, and Jock R. Anderson. 1998. “Returns to In State of the World, edited by Lester Brown. New York: T&V Extension in Kenya: Some Alternative Findings.” W. W. Norton. World Bank, Operations Evaluations Department. Wash- Frigenti, Laura, and Alberto Harth. 1998. “Local Solutions to ington, D.C. Processed. Regional Problems: The Growth of Social Funds and Public Gavin, Michael, and Roberto Perotti. 1997. “Fiscal Policy in Works and Employment Projects in Sub-Saharan Africa.” Latin America.” National Bureau of Economic Research World Bank, Washington, D.C. Processed. (NBER) Macroeconomics Annual. Cambridge, Mass. Fry, Maxwell J. 1995. Money, Interest and Banking in Economic Gerschenkron, Alexander. 1962. Economic Backwardness in His- Development, 2nd ed. Baltimore: The Johns Hopkins Uni- torical Perspective. Cambridge, Mass.: Belknap Press. versity Press. Gertler, Meric S. 1997. “Globality and Locality: The Future of Fuente, A. 1995. “The Empirics of Growth and Convergence: ‘Geography’ and the Nation-State.” In Pacific Rim Develop- A Selective Review.” Discussion Paper 1275. Center for Eco- ment: Integration and Globalisation in the Asia-Pacific Econ- nomic Policy Research, London. omy, edited by Peter Rimmer. Canberra City, Australia: Fujikura, Ryo. 1998. “Public Participation in Urban Environ- Aussie Print. mental Management in Japan.” Paper for World Develop- Gibson, Edward L., Ernesto F. Calvo, and Tulia G. Falleti. ment Report 1999/2000, Tokyo Workshop. Processed. 1999. “Reallocative Federalism: Overrepresentation and ———. 1999. “Public Participation in Urban Environmental Public Spending in the Western Hemisphere.” Department Management in Japan.” Background Paper for World Devel- of Political Science, Northwestern University. Processed. opment Report 1999/2000. World Bank, Washington, D.C. Giddens, Anthony. 1998. Conversations by Anthony Giddens. Fujita, Masahisa, P. R. Krugman, and A. J. Venables. 1999. The Cambridge, United Kingdom: Polity Press. Spatial Economy: Cities, Regions and International Trade. Gilbert, Alan, and Josef Gugler. 1992. Cities, Poverty and De- Cambridge, Mass.: MIT Press. velopment. New York: Oxford University Press. Fung, K. C., and Francis Ng. 1998. “What Do Trade Negotia- Glaeser, Edward L. 1997. “Learning in Cities.” Discussion tors Negotiate About? Some Evidence from the Uruguay Paper 1814: 1–23. Harvard Institute of Economic Research, Round.” Working Paper 412. Department of Economics, Cambridge, Mass. University of California, Santa Cruz. ———. 1998. “Are Cities Dying?” Journal of Economic Perspec- Galeotti, Gianluigi. 1992. “Decentralization and Political tives 12(2): 139–60. Rents.” In Local Government Economics in Theory and Prac- Glaeser, Edward L., and Jordan Rappaport. 1998. “Cities and tice, edited by David King. London: Routledge. Governments.” Background Paper for World Development Gamble, John King, Zachary T. Irwin, Charles M. Redenius, Report 1999/2000. World Bank, Washington, D.C. and James W. Weber. 1992. Introduction to Political Science. Englewood Cliffs, N.J.: Prentice Hall. Glaeser, Edward L., Hedi D. Kallal, José A. Scheinkman, and Gandal, Neil, Gordon H. Hanson, and Matthew J. Slaughter. Andrei Shleifer. 1992. “Growth in Cities.” Journal of Eco- 1999. “Rybczynski Effects and Adjustment to Immigration nomic Perspectives 12(2): 1126–53. in Israel.” CEPR (Centre for Economic Policy Research) Goldstein, Morris. 1997. The Case for an International Banking Workshop in International Trade. June 24–27. Standard. Washington, D.C.: Institute for International Gang, Fan. 1999. “Impacts of Globalization on the Developing Economics. Countries: The Case of China.” Background Paper for ———. 1998. “The Case for International Banking Stan- World Development Report 1999/2000, Tokyo Workshop. dards.” In Preventing Bank Crises: Lessons from Recent Global Garcia, Gillian G. 1996. “Deposit Insurance: Obtaining the Bank Failures, edited by Gerard C. Caprio. Washington, Benefits and Avoiding the Pitfalls.” Working Paper 96/83. D.C.: Federal Reserve Bank of Chicago and the World Bank International Monetary Fund, Washington, D.C. Economic Development Institute. ———. 1998. “Deposit Insurance.” In Preventing Bank Crises: Goldstein, Morris, and John Hawkins. 1998. “The Origins of Lessons from Recent Global Bank Failures, edited by Gerard C. the Asian Financial Turmoil.” Discussion Paper 9805. Re- Caprio. Washington, D.C.: Federal Reserve Bank of Chicago serve Bank of Australia, Canberra. and the World Bank Economic Development Institute. Goldstein, Morris, and Philip Turner. 1996. “Banking Cri- Gardner-Outlaw, Tom, and Robert Engelman. 1997. “Easing ses in Emerging Economies: Origins and Policy Options.” Scarcity: A Second Update.” Population Action Interna- Economic Papers 46. Bank for International Settlements, tional, Washington, D.C. Geneva. Garrett, Geoffrey. 1998. Partisan Politics in the Global Economy. Good, Michael F. 1999. “Tying the Conductor’s Arms.” Na- Cambridge: Cambridge University Press. ture. July 15. Gaspar, Jess, and Edward L. Glaeser. 1998. “Information Tech- Goodhart, Charles, Philipp Hartmann, David Llewellyn, Lil- nology and the Future of Cities.” Journal of Urban Econom- iana Rojas-Suarez, and Steven Weisbrod. 1998. Financial ics 43(136). Regulation. London: Routledge. Gastanaga, Victor M., Jeffrey B. Nugent, and Bistra Pashamova. Goodman, A. 1998. “Carbon Trading Up and Running.” To- 1998. “Host Country Reforms and FDI Inflows: How morrow Magazine (May/June).    Goulder, Lawrence H. 1994. “Energy Taxes: Traditional Effi- Uruguay Round and the Developing Countries, edited by Alan ciency Effects and Environmental Implications.” Tax Policy Winters and William Martin. Cambridge: Cambridge Uni- and the Economy 8:105–58. versity Press. Graham, Carol. 1994. “Safety Nets, Politics and the Poor: Harvard Working Group. 1994. “The Emergence of New Dis- Transition to Market Economies.” The Brookings Institu- eases.” American Scientist 82(1): 52–60. tion, Washington, D.C. Harwood, Alison. 1997. “Financial Reform in Developing Graham, Edward. 1996. Global Corporations and National Gov- Countries.” In Sequencing? Financial Strategies for Develop- ernments. Washington, D.C.: Institute for International ing Countries, edited by Alison Harwood and Bruce L. R. Economics. Smith. Washington, D.C.: The Brookings Institution Press. Gregorio, José De, and P. E. Guidotti. 1995. “Financial Devel- Hasan, Arif. 1997a. Urban Housing Policies and Approaches in a opment and Economic Growth.” World Development 23(3): Changing Asian Context. Karachi: City Press. 443–48. ———. 1997b. Working with Government. Karachi: City Press. Greif, Avner. 1998. “Historical and Comparative Institutional ———. 1998. Community Initiatives: Four Case Studies from Analysis.” American Economic Review, Papers and Proceed- Karachi. Karachi: City Press. ings 88: 80–84. Hasan, Arif, Akbar Zaidi, and Muhammad Younis. 1998. Back- Grootaert, Christiaan, and Ravi Kanbur. 1995. “Child Labor: ground Note on Karachi Prepared for World Development A Review.” Policy Research Working Paper 1454. World Report 1999/2000. Washington, D.C. Bank, Washington, D.C. Hathaway, Dale E., and Merlinda D. Ingco. 1996. “Agricul- Grossman, Gene M., and Alan B. Krueger. 1995. “Economic tural Liberalization and the Uruguay Round.” In The Growth and the Environment.” Quarterly Journal of Eco- Uruguay Round and the Developing Countries, edited by Alan nomics 110: 353–77. Winters and William Martin. Cambridge: Cambridge Uni- G-22 (Group of Twenty-two) Committee. 1998a. “Report of versity Press. the Working Group on Strengthening Financial Systems.” Hawley, Willis D. 1970. Nonpartisan Elections and the Case of Washington, D.C. Informal publication. Party Politics. New York: Wiley Press. ———. 1998b. “Report of the Working Group on Trans- Hay, Jonathan R., and Andrei Shleifer. 1998. “Private Enforce- parency and Accountability.” Washington, D.C. Informal ment of Public Laws: A Theory of Legal Reform.” American publication. Economic Review, Papers and Proceedings, 88: 398–407. Guerrant, Richard L. 1998. “Why America Must Care about Hayami, Yujiro. 1998. “Toward a New Model of Rural-Urban Tropical Medicine: Threats to Global Health and Security Linkages under Globalization.” Background Paper for World from Tropical Infectious Diseases.” American Journal of Development Report 1999/2000, Tokyo Workshop. Tropical Medicine and Hygiene 59(1): 3–16. Head, Keith, and John Ries. 1995. “Inter-City Competition for Guo, Shuqing. 1998. “Globalization and China’s Economy.” Foreign Investment: Static and Dynamic Effects of China’s Background Paper for World Development Report 1999/2000. Incentive Areas.” Journal of Urban Economics 40(July 1996): World Bank, Washington, D.C. Processed. 38–60. Haddad, Lawrence, Marie T. Ruel, and James L. Garrett. 1999. Hellman, Thomas, Kevin Murdock, and Joseph E. Stiglitz. “Are Urban Poverty and Undernutrition Growing? Some 1998. “Liberalization, Moral Hazard in Banking, and Pru- Newly Assembled Evidence.” Discussion Paper 63, Food dential Regulaton: Are Capital Requirements Enough?” Consumption and Nutrition Division, International Food Graduate School of Business, Stanford University, Calif. Policy Research Institute. Hemming, Richard, and Paul Bernard Spahn. 1998. “European Hallerberg, Mark, and Jürgen von Hagen. 1997. “Electoral In- Integration and the Theory of Fiscal Federalism.” In Macro- stitutions, Cabinet Negotiations, and Budget Deficits in the economic Dimensions of Public Finance: Essays in Honour of European Union.” Georgia Institute of Technology, Atlanta. Vito Tanzi, edited by Bario Blejer and Teresa Ter-Minassian. Processed. Washington, D.C.: International Monetary Fund. Hardoy, Jorge, and David Satterthwaite. 1990. “The Future Hemming, Richard, Neven Mates, and Barry Potter. 1997. City.” In The Poor Die Young, edited by J. E. Hardoy, “India.” In Fiscal Federalism in Theory and Practice, edited S. Cairncross, and D. Satterthwaite. London: Earthscan. by Teresa Ter-Minassian. Washington, D.C.: International Harpham, Trudy, and M. Stuttaford. 1999. “Health, Gover- Monetary Fund. nance and the Environment.” In Sustainability in Cities in Henderson, J. Vernon. 1998. Urban Development: Theory, Fact Developing Countries: Theory and Practice at the Millennium, and Illusion. New York: Oxford University Press. edited by Cedric Pugh. London: Earthscan. Henderson, J. Vernon, and Ari Kuncoro. 1996. “Industrial Harpham, Trudy, and M. Tanner, eds. 1995. Urban Health Centralization in Indonesia.” World Bank Economic Review in Developing Countries: Progress and Prospects. London: 10: 513–40. Earthscan. Henderson, J. Vernon, T. Lee, and J-Y Lee. 1998. “Externali- Harrison, Ann. 1994. “Productivity, Imperfect Competition, ties, Location, and Industrial Deconcentration in a Tiger and Trade Reform: Theory and Evidence.” Journal of Inter- Economy.” Department of Economics, Brown University. national Economics (36): 53–73. Processed. Harrison, Glenn W., Thomas F. Rutherford, and David G. Hernández, Leonardo, and Klaus Schmidt-Hebbel. 1999. “Cap- Tarr. 1996. “Quantifying the Uruguay Round.” In The ital Controls in Chile: Effective? Efficient? Endurable?” Paper      ⁄  presented at World Bank/International Monetary Fund/ Hommes, Rudolf. 1996. “Conflicts and Dilemmas of Decen- World Trade Organization Conference on Capital Flows, Fi- tralization.” In Annual World Bank Conference on Develop- nancial Crises, and Policies. Washington, D.C., April 15–16. ment Economics 1995, edited by Michael Bruno and Boris Hertel, Thomas W., William A. Masters, and Aziz Elbehri. Pleskovic. Washington, D.C.: World Bank. 1998. “The Uruguay Round and Africa: A Global General Horioka, C. Y. 1990. “Why Is Japan’s Household Saving So Equilibrium Analysis.” Journal of African Economies 7(2): High? A Literature Survey.” Journal of Japanese and Interna- 208–34. tional Economics 4(1): 49–92. Hertel, Thomas W., Christian F. Bach, Betina Dimaranan, and Horlick, Gary N., and Steven A. Sugarman. 1999. “Antidump- Will Martin. 1996. “Growth, Globalization, and Gains from ing Policy as a System of Law.” In Trade Rules in the Mak- the Uruguay Round.” Policy Research Working Paper 1614. ing, edited by Miguel Rodriguez Mendoza, Patrick Low, and World Bank, Washington, D.C. Barbara Kotschwar. Washington, D.C.: The Brookings In- Hertel, Thomas W., Will Martin, Koji Yanagishima, and Betina stitution Press. Dimaranan. 1996. “Liberalizing Manufactures Trade in a Hourcade, J. 1996. “A Review of Mitigation Cost Studies.” In Changing World Economy.” The Uruguay Round and the Climate Change 1995, Economic and Social Dimensions of Developing Countries, edited by Alan Winters and William Climate Change: Contribution of Working Group III to the Martin. Cambridge: Cambridge University Press. Second Assessment Report of the Intergovernmental Panel on Heywood, V. H., ed. 1995. Global Biodiversity Assessment. Cam- Climate Change, edited by James Bruce, Hoesung Lee, and bridge: Cambridge University Press. Erik Haites. Cambridge: Cambridge University Press. Higgins, Matthew, and Jeffrey G. Williamson. 1997. “Age Howse, Robert, and Michael Trebilcock. 1996. “The Fair Structure Dynamics in Asia and Dependence on Foreign Trade–Free Trade Debate: Trade, Labour and the Environ- Capital.” Population and Development Review 23(2): 261–93. ment.” International Review of Law and Economics 16: 61–79. Hirschman, Alberto. 1970. Exit, Voice and Loyalty. Cambridge, Howson, Christopher P., Harvey V. Fineberg, and Barry R. Mass.: Harvard University Press. Bloom. 1998. “The Pursuit of Global Health: The Rele- Hoekman, Bernard. 1996. “Assessing the General Agreement vance of Engagement for Developed Countries.” Lancet on Trade in Services.” In The Uruguay Round and the Devel- 351(21/February). oping Countries, edited by Alan Winters and William Mar- Hufbauer, Gary, and Barbara Kotschwar. 1998. “The Future tin. Cambridge: Cambridge University Press. Course of Trade Liberalization.” Background Paper for Hoekman, Bernard, and Kym Anderson. 1999. “Developing World Development Report 1999/2000. World Bank, Wash- Country Agriculture and the New Trade Agenda.” World ington, D.C. Bank, Washington, D.C. Processed. Hughes Hallett, A. J. 1998. “Policy Coordination: Globaliza- Hoekman, Bernard, and Simeon Djankov. 1996. “The Euro- tion or Localization in International Monetary Arrange- pean Union’s Mediterranean Free Trade Initiative.” The ments?” Background Paper for World Development Report World Economy 19(4): 387–406. 1999/2000. World Bank, Washington, D.C. ———. 1997a. “Effective Protection and Investment Incen- Hummels, David. 1998. “Data on International Transportation tives in Egypt and Jordan during the Transition to Free Costs: A Report Prepared for the World Bank.” Background Trade with Europe.” World Development 25(2): 281–91. Paper for World Development Report 1999/2000. World ———. 1997b. “Towards a Free Trade Agreement with the Bank, Washington, D.C. European Union: Issues and Policy Options for Egypt.” In Hummels, David, Jun Ishii, and Kei-Mu Yi. 1999. “The Na- Regional Partners in Global Markets: Limits and Possibilities ture and Growth of Vertical Specialization in World Trade.” of the Euro-Med Agreements, edited by Ahmed Galal and University of Chicago Graduate School of Business and the Bernard Hoekman. Centre for Egyptian Policy Research/ Federal Reserve Bank of New York. Processed. Egyptian Centre for Economic Studies, London. Hunter, Brian, ed. 1998. Statesman’s Yearbook: A Statistical, Po- Hoekman, Bernard, and Denise Konan. 1999. “Deep Integra- litical and Economic Account of the States of the World for the tion, Nondiscrimination, and Euro-Mediterranean Free Year 1998–99. London: Macmillan. Trade.” World Bank, Washington, D.C. Huther, Jeff, and Anwar Shah. 1998. “Applying a Simple Mea- Hoekman, Bernard, and Carlos A. Primo Braga. 1997. “Protec- sure of Good Governance to the Debate on Fiscal Decen- tion and Trade in Services: A Survey.” Open Economies Re- tralization.” Policy Reaearch Working Paper 1894. World view 8: 285–308. Bank, Washington, D.C. Hoekman, Bernard, Denise Konan, and Keith Maskus. 1998. IFC (International Finance Corporation). 1998. Foreign Direct “An Egypt–United States Free Trade Agreement: Economic Investment. Lessons of Experience 5. Washington, D.C. Incentives and Effects.” Discussion Paper 1882. Centre for ILO (International Labour Office). 1993. World Labour Report. Economic Policy Research, London. Geneva. Hohenberg, Paul M. 1998. “Urban Systems and Economic De- ILO (International Labour Organisation). 1998. “Employabil- velopment: The European Long Term and Its Implications.” ity in the Global Economy: How Training Matters.” World Background Paper for World Development Report 1999/2000. Employment Report 1998–99. Geneva. World Bank, Washington, D.C. Imber, Mark. 1996. “The Environment and the United Na- Holmes, Robert. 1999. “Head in the Clouds.” New Scientist tions.” In The Environment and International Relations, 162(2185). edited by John Vogler and Mark Imber. London: Routledge.    IMF (International Monetary Fund). Various years. Annual Country Study.” Development Research Group, World Report on Exchange Arrangements and Exchange Restrictions . Bank, Washington, D.C. Processed. Washington, D.C. John, T. Jacob, Reuben Samuel, Vinohar Balraj, and Rohan ———. 1997. Government Finance Statistics. Washington, John. 1998. “Disease Surveillance at the District Level: A D.C. Model for Developing Countries.” Lancet 352:(4/July). ———. 1998a. Balance of Payments Statistics Yearbook. Wash- Johnston, R. Barry. 1997. “The Speed of Financial Sector Re- ington, D.C. forms: Risks and Strategies.” In Sequencing? Financial Strate- ———. 1998b. Government Finance Statistics. Washington, gies for Developing Countries, edited by Alison Harwood and D.C. Bruce L. R. Smith. Washington, D.C.: The Brookings In- ———. 1998c. International Capital Markets. Washington, stitution Press. D.C. Johnston, R. Barry, Salim M. Darbar, and Claudia Echeverria. ———. 1999a. “A Guide to Progress in Strengthening the 1997. “Sequencing Capital Account Liberalization: Lessons Architecture of the International Financial System” (http:// from the Experiences in Chile, Indonesia, Korea, and Thai- www.imf.org/external/np/exr/facts/arch.htm). land.” Working Paper 97/157. International Monetary ———. 1999b. World Economic Outlook Database. Washing- Fund, Washington, D.C. ton, D.C. Josling, Timothy. 1998a. “Agricultural Trade Policy: Complet- Ingco, Merlinda, and Francis Ng. 1998. “Distortionary Effects ing the Reform.” In Launching New Global Trade Talks: An of State Trading in Agriculture: Issues for the Next Round Action Agenda, Special Report 12, edited by Jeffrey J. Schott. of Multilateral Trade Negotiations.” Policy Research Work- Washington, D.C.: Institute for International Economics. ing Paper 1915. World Bank, Washington, D.C. ———. 1998b. “Agricultural Trade Policy: Completing the Ingram, Gregory K. 1998. “Patterns of Metropolitan Develop- Reform.” Institute for International Economics. Washing- ment: What Have We Learned?” Urban Studies 35(7). ton, D.C. Inman, Robert P., and Daniel L. Rubinfeld. 1996. “Designing Kaminsky, Graciela L., and Carmen M. Reinhart. 1998. “Fi- Tax Policy in Federalist Economies: An Overview.” Journal nancial Crises in Asia and Latin America: Then and Now.” of Public Economics 60: 307–34. American Economic Review Papers and Proceedings 88(2): Intergovernmental Panel on Climate Change (IPCC). 1995. 444–48. Climate Change 1995: Economic and Social Dimensions. Kane, Edward J. 1998. “Understanding and Preventing Bank Working Group I, Contribution to the Second Assessment Re- Crises.” In Preventing Bank Crises: Lessons from Recent Global port of the Intergovernmental Panel on Climate Change. Cam- Bank Failures, edited by Gerard C. Caprio. Washington, bridge: Cambridge University Press. D.C.: Federal Reserve Bank of Chicago and the World Bank Isham, Jonathan, Deepa Narayan, and Lant Pritchett. 1994. Economic Development Institute. “Does Participation Improve Performance? Empirical Evi- Kane, Hal. 1995. “What’s Driving Migration?” Worldwatch dence from Project Data.” Policy Research Working Paper (January/February). 1357. World Bank, Washington, D.C. Keesing, Donald B. 1998. Improving Trade Policy Reviews in the Islam, Nurul. 1997. “The Nonfarm Sector and Rural Develop- World Trade Organization. Washington, D.C.: Institute for ment: Review of Issues and Evidence.” Food, Agriculture, International Economics. and the Environment Discussion Paper 22. International Kenny, Charles. 1999. “Telecommunications and Competitive Food Policy Research Institute, Washington D.C. Cities.” In World Urban Economic Development, edited by Jackson, John. 1997. The World Trading System, 2nd edition. Elizabeth Cooper. London: World Markets Research Center. Cambridge, Mass.: MIT Press. Kenny, Charles J., and David Williams. 1999. “What Do ———. 1998. The World Trade Organization: Constitution and We Know about Economic Growth? or Why Don’t We Jurisprudence. The Royal Institute of International Affairs, Know Very Much?” Lady Margaret Hall, Oxford Univer- London. sity. Processed. Jaffe, Adam B., Manuel Trajtenberg, and Rebecca Henderson. Kerr, William A. 1999. “International Trade in Transgenic 1993. “Geographic Localization of Knowledge Spillovers as Food Products: A New Focus for Agricultural Trade Dis- Evidenced by Patent Citation.” Quarterly Journal of Econom- putes.” The World Economy 22(2): 245–59. ics 108: 577–98. Kessides, Christine. 1998. “A Strategic View of Urban and Local James, S., and Kym Anderson. 1998. “On the Need for More Government Issues: Implications for the Bank.” World Bank, Economic Assessment of Quarantine Policies.” Australian Washington, D.C. Processed. Journal of Agricultural and Resource Economics 41(4/Decem- Khilnani, Sunil. 1997. The Idea of India. New York: Farrar, ber): 525–44. Straus, and Giroux. Jamison, Dean T. 1993. “Investing in Health.” Finance and De- Kilgour, D. Marc, and Ariel Dinar. 1995. “Are Stable Agree- velopment (September): 2–5. ments for Sharing International River Waters Now Possi- Jamison, Dean T., Julio Frenk, and Felicia Kaul. 1998. “Inter- ble?” Policy Research Working Paper 1474. World Bank, national Collective Action in Health Objectives, Functions Washington, D.C. and Rationale.” Lancet 351(14/February). Kim, Hyung-Ki, and Jun Ma. 1997. “The Role of Government Jin, Jing, and Heng-fu Zou. 1998. “The Effects of Fiscal in Acquiring Technological Capability: The Case of the Decentralization on the Sizes of Governments: A Cross- Petrochemical Industry in East Asia.” In The Role of Govern-      ⁄  ment in East Asian Economic Development, edited by Masahiko ———. 1995. “Growing World Trade: Causes and Conse- Aoki, Hyung-ki Kim, and Mashiro Okuno-Funiwara. New quences.” Brookings Papers on Economic Activity 1: 327–77. York: Oxford University Press. ———. 1998a. “Space: The Final Frontier.” Journal of Eco- King, Elizabeth M., and Berk Ozler. 1998. “What’s Decentral- nomic Perspectives 12(2): 161–74. ization Got to Do with Learning? The Case of Nicaragua’s ———. 1998b. “What Happened to Asia?” On Paul Krug- School Autonomy Reform.” Impact Evaluation of Educa- man’s web page (http:/web.mit.edu/people/krugman/index. tion Working Paper Series 9. World Bank, Washington, html ). D.C. Laird, Sam. 1997. “Issues for the Forthcoming Multilateral Ne- Kinoshita, Yuko, and Ashoka Mody. 1997. “The Usefulness of gotiations in Agriculture.” WTO (World Trade Organiza- Private and Public Information for Foreign Investment De- tion), Geneva. cisions.” Policy Research Working Paper 1733. World Bank, Lall, Rajiv, and Bert Hofman. 1994. “Decentralization and Washington, D.C. Government Deficit in Chile.” World Bank, Washington, Kitano, Naohiro. 1998. “Analysis of Spatial Organization and D.C. Processed. Transportation Demand in an Expanding Urban Area Using Lall, Sanjaya. 1998. “Exports of Manufactures by Developing Centrographic Methods: Sendai 1972–1992.” Paper for Countries: Emerging Patterns of Trade and Location.” Ox- World Development Report 1999/2000, Tokyo Workshop. ford Review of Economic Policy 14: 54–73. World Bank, Washington, D.C. Lampietti, Julian, and John Dixon. 1995. “To See the Forest Kleinknecht, Alfred, and Jan der Wengel. 1998. “The Myth of for the Trees: A Guide to Non-Timber Forest Benefits.” Economic Globalization.” Cambridge Journal of Economics Paper No. 13. Environment Department, World Bank, 22: 637–47. Washington, D.C. Knight, Malcolm. 1998. “Developing Countries and the Glob- Lancet. 1996. “HIV Epidemic in India. Opportunity to Learn alization of Financial Markets.” Working Paper 98/105. In- from the Past.” 347 (18/May): 1349–50. ternational Monetary Fund, Washington, D.C. ———. 1998. “Action Against Antibiotic Resistance.” 351 Kobrin, Stephen J. 1997. “The Architecture of Globalization: (2/May). State Sovereignity in a Networked Global Economy.” In Law, Lisa. 1997. “Cebu and Ceboom: The Political Place of Governments, Globalization and International Business, edited Globalisation in a Philippine City.” In Pacific Rim Develop- by John H. Dunning. New York: Oxford University Press. ment: Integration and Globalisation in the Asia-Pacific Econ- Konan, Denise, and Keith E. Maskus. 1997. “A Computable omy, edited by Peter Rimmer. Canberra City, Australia: General Equilibrium Analysis of Egyptian Trade Liberaliza- Aussie Print. tion Scenarios.” In Regional Partners in Global Markets: Lim- Lawrence, Robert Z. 1996. “Current Economic Policies: Social its and Possibilities of the Euro-Med Agreements, edited by Implications over the Longer Term.” In Social Cohesion and Ahmed Galal and Bernard Hoekman. Centre for Egyptian the Globalising Economy: What Does the Future Hold ? Paris: Policy Research/Egyptian Centre for Economic Studies, London. Organisation for Economic Co-operation and Development. Kono, Masamichi, and Ludger Schuknecht. 1999. “Financial ———, ed. 1998. Brookings Trade Forum 1998. Washington, Services Trade, Capital Flows, and Financial Stability.” D.C.: The Brookings Institution Press. Paper presented at the World Bank/International Monetary Le Fort, V. Guillermo, and Carlos Budnevich. 1998. “Capital Fund/World Trade Organization Conference on Capital Account Relations and Macroeconomic Policy: Two Latin Flows, Financial Crises, and Policies. Washington, D.C., American Experiences.” In Capital Account Regimes and the April 15–16. Developing Countries, edited by G. K. Helleiner. New York: Kosai, Yutaka, Jun Saito, and Nashiro Yashiro. 1998. “Declin- St. Martin’s Press. ing Population and Sustained Economic Growth: Can They Le Houerou, Philippe. 1996. “Fiscal Management in Russia.” Coexist?” American Economic Review. 88(2): 412–16. World Bank, Washington, D.C. Processed. Kozul-Wright, Richard, and Robert Rowthorn. 1998. “Spoilt Leipziger, Danny M. 1998. “The Argentine Banking Crisis: Ob- for Choice? Multinational Corporations and the Geography servations and Lessons.” In Preventing Bank Crises: Lessons of International Production.” Oxford Review of Economic from Recent Global Bank Failures, edited by Gerard C. Policy 14(2): 74–92. Caprio. Washington, D.C.: Federal Reserve Bank of Chicago Kremer, Michael. 1993. “O-Ring Theory of Economic Devel- and the World Bank Economic Development Institute. opment.” Quarterly Journal of Economics (108): 551–75. Leitmann, Josef, and Deniz Baharoglu. 1998. “Informal Rules! Krishna, Pravin, and Devahish Mitra. 1998. “Trade Liberaliza- Using Institutional Economics to Understand Service Pro- tion, Market Discipline, and Productivity Growth: New Ev- vision in Turkey’s Spontaneous Settlements.” Journal of De- idence from India.” Journal of Development Economics 56: velopment Studies 34(5). 447–62. Lele, Uma J., and Robert E. Christiansen. 1989. “Markets, Mar- Krueger, Anne O. 1998. “The Developing Countries and the keting Boards, and Cooperatives in Africa: Issues in Adjust- Next Round of Multilateral Trade Negotiations.” Stanford ment Policy.” World Bank, Washington, D.C. Processed. University, Calif. Processed. Levine, Ross. 1997. “Financial Development and Economic Krugman, Paul. 1993. “On the Number and Location of Growth: Views and Agenda.” Journal of Economic Literature Cities.” European Economic Review 37: 293–28. 35: 688–726.    ———. 1998. “The Legal Environment, Banks, and Long-Run Lucas, Robert. 1998. “Internal Migration and Urbanization: Economic Growth.” Journal of Money, Credit, and Banking Recent Contributions and New Evidence.” Background 30(3): 596–613. Paper for World Development Report 1999/2000. World ———. 1999. “Napoleon, Bourses and Growth: With a Focus Bank, Washington, D.C. on Latin America.” Paper delivered at the IRIS Market Aug- Madeley, John. 1995a. “Biodiversity: A Matter of Extinction.” menting Government Conference. Washington, D.C., Panos Media Briefing 17. Panos Institute, London. Processed. March 26–27. ———. 1995b. “Fish: A Net Loss for the Poor.” Panos Media Levine, Ross, and David Renelt. 1992. “Sensitivity Analysis of Briefing 15. Panos Institute, London. Processed. Cross-Country Growth Regressions.” American Economic Mahmood, Saman. 1999. “Shelter within My Reach: Medium- Review 82(September): 942–63. Rise Apartment Housing for the Middle Income Group in Levine, Ross, and Sara Zervos. 1998a. “Capital Control Liber- Karachi, Pakistan.” Master’s diss., Massachusetts Institute alization and Stock Market Development.” World Develop- of Technology, Cambridge, Mass. Processed. ment 26(7): 1169–83. Mainwaring, Scott. 1997. “Multipartism, Robust Federalism, ———, ed. 1998b. Brookings Trade Forum 1998. Washington, and Presidentialism in Brazil.” In Presidentialism and Democ- D.C.: The Brookings Institution Press. racy, edited by Scott Mainwaring and Matthew Soburt ———. 1998c. “Stock Markets, Banks and Economic Growth.” Shogart. New York: Cambridge University Press. American Economic Review 88(3): 537–54. Mallampally, Padma, and Karl P. Sauvant. 1999. “Foreign Di- Levinsohn, James. 1993. “Testing the Imports-as-Market- rect Investment in Developing Countries.” Finance and De- Discipline Hypothesis.” Journal of International Economics velopment 36(1): 34–37. 35: 1–22. Mani, Sunil. 1996. “Economic Liberalization and Kerala’s In- Lewandrowski, Jan, and David Schimmelpfennig. 1999. “Eco- dustrial Sector.” Economic and Political Weekly (August nomic Implications of Climate Change for U.S. Agriculture: 24–31). Assessing Recent Evidence.” Land Economics 75(1): 39–57. Mankiw, N. Gregory, David Romer, and David Weil. 1992. Lewis, W. Arthur. 1955. The Theory of Economic Growth. “Contribution to the Empirics of Economic Growth.” Quar- Reprint, New York: Harper Torchbooks, 1970. terly Journal of Economics 107: 407–37. Lijphart, Arend. 1994. Electoral Systems and Party Systems: A Manning, Nick. 1998. “Unbundling the State: Autonomous Study of Twenty-Seven Democracies 1945–1990. New York: Agencies and Service Delivery.” World Bank, Washington, Oxford University Press. D.C. Processed. Lindgren, Carl-Johan, and Gillian Garcia. 1996. “Deposit In- Markusen, Ann. 1998. “What Distinguishes Success Among surance and Crisis Management.” Monetary and Exchange Second Tier Cities?” Presentation at World Development Re- Affairs Department Operation Paper 96/3. International port 1999/2000, July Workshop. World Bank, Washington, Monetary Fund, Washington, D.C. D.C. Linz, Juan J., and Alfred Stepan. 1997. “Toward Consolidated Democracies.” In Consolidating the Third Wave Democracies, Marsh, Kevin. 1998. Malaria Disaster in Africa. Lancet 352 edited by Larry Diamond, Marc F. Plattner, Yun-han Chu, (19/September): 924. and Hung-mao Tien. Baltimore: The Johns Hopkins Uni- Martin, Ron. 1999. “The New Geographical Turn in Econom- versity Press. ics.” Cambridge Journal of Economics (January). Lipsey, Robert E. 1999. “The Role of Foreign Direct Investment Martinez-Vasquez, Jorge. 1998. “Fiscal Decentralization in the in International Capital Flows.” Working Paper Series 7094. Russian Federation: Major Trends and Issues.” School of Po- National Bureau of Economic Research, Cambridge, Mass. litical Studies, Georgia State University, Atlanta. Processed. Lipton, Michael. 1999. “Saving Undernourished Lives.” Finan- Martinez-Vasquez, Jorge, and Robert M. McNab. 1997. “Fis- cial Times, Letters to the Editor. June 2. cal Decentralization, Economic Growth and Democratic Litan, Robert E. 1998. “Toward a Global Financial Architec- Governance.” School of Political Studies, Georgia State Uni- ture for the 21st Century.” Background Paper for World De- versity, Atlanta. Processed. velopment Report 1999/2000. World Bank, Washington, Maskus, Keith E. 1997. “Should Core Labor Standards Be Im- D.C. posed through International Trade Policy?” Policy Research Litvack, Jennie. 1994. “Regional Demands and Fiscal Federal- Working Paper 1817. World Bank, Washington, D.C. ism.” In Russia and the Challenge of Fiscal Federalism, edited Mathur, Om Prakash. 1999. “Decentralization in India: A Re- by Christine Wallich. World Bank, Washington, D.C. port Card.” National Institute of Public Finance and Policy, Litvack, Jennie, Junaid Ahmad, and Richard Bird. 1998. “Re- New Delhi. thinking Decentralization in Developing Countries.” World Matusz, Steven, and David Tarr. 1998. “Adjusting to Trade Bank, Washington, D.C Policy Reform.” World Bank, Washington, D.C. Processed. Lloyd-Sherlock, Peter. 1997. “The Recent Appearance of Fave- Mayo, Stephen K., and Shlomo Angel. 1993. Housing: Enabling las in São Paulo City: An Old Problem in a New Setting.” Markets To Work. World Bank Policy Paper. Washington, Latin American Studies 16(3). D.C. Long, Millard, and Mihaly Kopanyi. 1998. “Hungary: Finan- Mazumdar, Dipak. 1986. “Rural-Urban Migration in Develop- cial Sector Development.” World Bank, Washington, D.C. ing Countries.” In Handbook of Regional and Urban Econom- Processed. ics, edited by Peter Nijkamp. New York: North-Holland.      ⁄  McKinnon, Ronald I. 1991. The Order of Economic Liberaliza- Musgrave, Richard A. 1997. “Devolution, Grants, and Fiscal tion: Financial Control in the Transition to a Market Econ- Competition.” Journal of Public Economics 11: 65–72. omy. Baltimore: The Johns Hopkins University Press. Musgrave, Richard A., and Peggy Musgrave. 1973. Public Fi- ———. 1997. “Monetary Regimes, Government Borrowing nance in Theory and Practice. New York: McGraw Hill. Constraints and Market Preserving Federalism: Implications Mussa, Michael, and Anthony Richards. 1999. “Capital Flows for EMU.” Stanford University, Calif. Processed. in the 1990s before and after the Asian Crisis.” Paper pre- ———. 1998. “The IMF. The East Asian Currency Crisis and sented at the World Bank/International Monetary Fund/ the World Dollar Standard.” Paper presented at the Ameri- World Trade Organization Conference on Capital Flows, Fi- can Economics Association meetings. Chicago, January 3–5. nancial Crises, and Policies. Washington, D.C., April 15–16. McKinnon, Ronald I., and Huw Pill. 1998. “International Mutti, John, and Harry Grubert. 1998. “The Significance of In- Overborrowing: A Decomposition of Credit and Currency ternational Tax Rules for Sourcing Income: The Relation- Risks.” World Development 26(7): 1267–82. ship between Income Taxes and Trade Taxes.” In Geography McLure, Charles E., Jr. 1999. “Intergovernmental Fiscal Rela- and Ownership as Bases for Economic Accounting, edited by tions and Local Financial Management.” World Bank, Robert E. Baldwin, Robert E. Lipsey, and J. David Richard- Washington, D.C. Processed. son. National Bureau of Economic Research (NBER) Stud- Mendes, Marcos Jose. 1999. Incentivos Eleitorais e Desequilibrio ies in Income and Wealth, vol. 59. Fiscal de Estados e Municipios. São Paulo: Instituto Fernand Narayan, Deepa, and Lant Pritchett. 1997. “Cents and Socia- Braudel de Economia Mundial. bility: Household Income and Social Capital in Rural Tan- Meyer, John W., David J. Frank, Ann Hironaka, Evan Schofer, zania.” Policy Research Working Paper 1796. World Bank, and Nancy B. Tuma. 1997. “The Structuring of a World Washington, D.C. Environmental Regime, 1870–1990.” International Organi- Nature. 1998. “Falling Satellites, Rising Temperatures.” August. zation 51(4): 623–51. ———. 1999. “Scientific Societies Build Better Nations.” June Michalopoulos, Constantine. 1999. “Developing Countries’ 17. Participation in the World Trade Organization.” The World New England Journal of Medicine. 1996. “The March of AIDS Economy 22(1): 117–44. through Asia.” 335(5). Millard, F. J. C. 1998. South Africa: A Physician’s View. Lancet ———. 1997. “Multi-drug Resistance in Plague.” 337(10). 351(7/March): 748–49. ———. 1998. “The Global Threat of Multi Drug Resistant Miller, Marian. 1995. The Third World in Global Environmen- Tuberculosis” 338(23). tal Politics. Boulder, Colo.: Lynne Reinner. New Scientist. 1998. “Dirty Secrets.” August 29. Mills, Edwin S. 1998. “Internal Functioning of Urban Areas.” Ng, Francis, and Alexander Yeats. 1997. “Open Economies Kellogg Graduate School of Management, Northwestern Work Better! Did Africa’s Protectionist Policies Cause Its University. Processed. Marginalization in World Trade?” World Development 25: Mills, Edwin S., and Charles Becker. 1986. Studies in Indian 889–975. Urban Development. New York: Oxford University Press. Norregaard, John. 1997. “Tax Assignment.” In Fiscal Federal- Miranda, Jorge, Raul A. Torres, and Mario Ruiz. 1998. “The ism in Theory and Practice, edited by Teresa Ter-Minassian. International Use of Anti-Dumping: 1987–1997.” World Washington, D.C.: International Monetary Fund. Trade Organization, Geneva. Processed. North, Douglass C. 1990. Institutions, Institutional Change and Miranda, Rowan, and Donald Rosdil. 1995. “From Boosterism Economic Performance. Cambridge: Cambridge University to Qualitative Growth.” Urban Affairs Review 30(6). Press. Mitlin, Diana, and David Satterthwaite. 1998. “Urban Poverty: ———. 1997. “Prologue.” In The Frontiers of the New Institu- Some Thoughts about Its Scale and Nature and about Re- tional Economics, edited by John N. Drobak and J. V. C. sponses to It by Community Organizations, NGOs, Local Nye. San Diego: Academic Press. Governments and National Agencies.” Paper for World De- Oates, Wallace. 1972. Fiscal Federalism. New York: Harcourt velopment Report 1999/2000, Singapore Workshop. World Brace Jovanovich. Bank, Washington, D.C. ———. 1998. “An Essay on Fiscal Federalism.” Journal of Eco- Mohan, Rakesh. 1999. “Financing of Sub-National Public In- nomic Literature. Forthcoming. vestment in India.” Paper for World Development Report Obstfeld, Maurice, and Alan M. Taylor. 1999. Global Capital 1999/2000, Singapore Workshop. World Bank, Washing- Markets: Integration, Crisis, and Growth. Cambridge: Cam- ton, D.C. bridge University Press. Moran, Theodore, H. 1999. Foreign Direct Investment and De- O’Donnell, Guillermo, Philippe C. Schmitter, and Laurence velopment: The New Policy Agenda for Developing Countries Whitehead. 1986. Transitions from Authoritarian Rule. Bal- and Economies in Transition. Washington, D.C.: Institute timore: The Johns Hopkins University Press. for International Economics. OECD (Organisation for Economic Co-operation and Devel- Munson, Abby. 1995. “The United Nations Convention on Bi- opment). 1997. “Managing across Levels of Government.” ological Diversity.” In The Earthscan Reader in Sustainable Paris. Development, edited by John Kirby, Phil O’Keefe, and Lloyd ———. 1998. Open Markets Matter: The Benefits of Trade and Timberlake. London: Earthscan. Investment Liberalisation. Paris.    Ordeshook, Peter C., and Olga Shvetsova. 1997. “Federalism and Peterson, George E., and Sonia Hammam. 1997. “Building Constitutional Design.” Journal of Democracy 8(1): 28–42. Local Credit Systems.” World Bank, Washington D.C. Ornusal, B., and S. Gautam. 1997. Vehicular Air Pollution: Ex- Processed. perience from Seven Latin American Urban Centers. Techni- Peterson, M. J. 1993. “International Fisheries Management.” cal Paper 373. Washington, D.C.: World Bank. In Institutions for the Earth: Sources of Effective International Ostrom, Elinor, Larry D. Schroeder, and Susan G. Wynne. Environmental Protection, edited by Peter Haas, Robert Keo- 1993. Institutional Incentives and Sustainable Development. hane, and Marc Levy. Cambridge, Mass.: MIT Press. Boulder, Colo.: Westview Press. Peterson, Peter G. 1999. Gray Dawn: How the Coming Age Ostry, Sylvia. 1997. The Post–Cold War Trading System: Who’s Wave Will Transform America—and the World. New York: on First? Chicago: University of Chicago Press. Times Books. ———. 1998. “Reinforcing the WTO.” Occasional Paper 56. Pirez, Pedro. 1998. “The Management of Urban Services in the Group of Thirty, Washington, D.C. City of Buenos Aires.” Environment and Urbanization 10(2). Oxley, Joanne, and Bernard Yeung. 1998. “Industrial Location Population Reference Bureau. 1999. “Immigration to the Growth and Government Activism: The Changing Eco- United States.” Population Bulletin 54(2). nomic Landscape.” School of Business Administration, Uni- Porter, G., R. Clemencon, W. Ofusu-Amaah, and M. Phillips. versity of Michigan, Ann Arbor. Processed. 1998. “Study of GEF’s Overall Performance.” Global Envi- Panagariya, Arvind. 1998. “The Regionalism Debate: An Over- ronment Facility, Washington, D.C. view.” Department of Economics, University of Maryland, Porter, Michael E. 1998. “Location, Clusters, and the ‘New’ College Park. Processed. Microeconomics of Competition.” Business Economics 33 ———. 1999. “The Regionalism Debate: An Overview.” The (January): 7–13. World Economy 22(4): 477–512. Porter, Richard C. 1996. The Economics of Water and Waste: A Panagariya, Arvind, and T. N. Srinivasan. 1997. “The New Re- Case Study of Jakarta, Indonesia. Aldershot, U.K.: Avebury. gionalism: A Benign or Malign Growth?” Department of Portes, Alejandro, and Patricia Landolt. 1996. “The Downside Economics, University of Maryland, College Park. Processed. of Social Capital.” The American Prospect 26(May/June): Panizza, Ugo. 1999. “On the Determinants of Fiscal Decentral- 18–21. ization: Theory and Evidence.” Journal of Public Economics. Poterba, James M. 1994. “State Responses to Fiscal Crises: The Forthcoming. Effects of Budgetary Institutions and Politics.” Journal of Po- Park, Yung Chul, and Chi-Young Song. 1998. “Capital Inflows litical Economy 102(4): 799–821. and Macroeconomic Policy in Sub-Saharan Africa.” In Cap- Prahalad, C. K., and Kenneth Liebenthal. 1998. “The End of ital Account Regimes and the Developing Countries, edited by Corporate Imperialism.” Harvard Business Review (July/ G. K. Helleiner. New York: St. Martin’s Press. August). Parry, M., and M. Livermore. 1997. Climate Change and Its Im- Prescott-Allen, Robert. 1995. “Conservation of Wild Genetic pacts. London: Crown Publishers. Parsonnet, Julie, ed. 1999. Microbes and Malignancy. New York: Resources.” In The Earthscan Reader in Sustainable Develop- Oxford University Press. ment, edited by John Kirby, Phil O’Keefe, and Lloyd Tim- Paul, Samuel. 1998. “Making Voice Work: The Report Card berlake. London: Earthscan. on Bangalore’s Public Service.” Policy Research Working President’s Committee of Advisors on Science and Technology Paper 1921. World Bank, Washington, D.C. Panel on Energy Research and Development. 1997. “Report Pearce, D., and others. 1996. “The Social Costs of Climate to the President on Federal Energy Research and Develop- Change.” In Climate Change 1995, Economic and Social Di- ment for the Challenges of the Twenty-first Century.” mensions of Climate Change: Contribution of Working Group Washington, D.C. III to the Second Assessment Report of the Intergovernmental Primo Braga, Carlos A., Raed Safadi, and Alexander Yeats. Panel on Climate Change, edited by James Bruce, Hoesung 1994. “NAFTA’s Implications for East Asian Exports.” Pol- Lee, and Erik Haites. Cambridge: Cambridge University Press. icy Research Working Paper 1351. World Bank, Washing- Pearce, David W., and Jeremy J. Warford. 1993. World With- ton, D.C. out End: Economics, Environment and Sustainable Develop- Pritchett, Lant. 1997. “Divergence, Big Time.” Journal of Eco- ment. New York: Oxford University Press. nomic Perspectives 11(3/Summer). Peek, Joe, and Eric S. Rosengren. 1997. “The International ———. 1998. “Patterns of Economic Growth: Hills, Plateaus, Transmission of Financial Shocks: The Case of Japan.” Mountains and Plains.” Policy Research Working Paper American Economic Review (87): 495–505. 1947. World Bank, Washington, D.C. Pegg, Scott. 1999. “The Nonproliferation of States: A Reply to Prud’homme, Rémy. 1995. “The Dangers of Decentralization.” Pascal Boniface.” Washington Quarterly 22(2): 139–47. World Bank Research Observer 10(2): 201–20. Perrings, Charles. 1995. “The Economic Value of Diversity.” Prusa, Thomas J. 1997. “The Trade Effects of U.S. Antidump- In Global Biodiversity Assessment, edited by V. H. Heywood. ing Actions.” Working Paper 5440. National Bureau of Eco- Cambridge: Cambridge University Press. nomic Research, Cambridge, Mass. Persson, Torsten, and Guido Tabellini. 1994. “Does Decentral- Przeworski, Adam, and Fernando Limongi. 1993. “Political ization Increase the Size of Government?” European Eco- Regimes and Economic Growth.” Journal of Economic Per- nomic Review 38: 765–73. spectives 7: 51–69.      ⁄  ———. 1997. “Modernization: Theories and Facts.” World Reinhart, Carmen M., and Vincent Raymond Reinhart. 1998. Politics 49 (January):155–83. “Some Lessons for Policy Makers Who Deal With the Psacharopoulos, George. 1994. “Returns to Investment in Educa- Mixed Blessing of Capital Inflows.” In Capital Flows and Fi- tion: A Global Update,” World Development 22(9): 1325–43. nancial Crises, edited by Miles Kahler. Ithaca: Cornell Uni- Puga, Diego. 1998. “Urbanization Patterns: European vs. Less- versity Press. Developed Countries.” Journal of Regional Science 38: 231–52. Rezende, Fernando. 1995. “Descentralização e Desenvolvi- Putnam, Robert D. 1993. Making Democracy Work. Princeton: mento: Problemas Atuals do Financiamento das Politicas Princeton University Press. Publicas.” Ministry of Development, Industry, and Com- Qian, Yingyi, and Barry R. Weingast. 1997. “China’s Transi- merce, Government of Brazil. Processed. tion to Markets: Market-Preserving Federalism, Chinese Richardson, Harry W. 1987. “The Costs of Urbanization: A Style.” Policy Reform 1: 149–85. Four-Country Comparison.” Economic Development and Quah, Danny. 1993. “Empirical Cross-Section Dynamics in Cultural Change 35: 561–80. Economic Growth.” European Economic Review 37: 426–34. Ridley, Matt. 1997. The Future of Disease. London: Phoenix. Quigley, John M. 1998. “Urban Diversity and Economic Rigau-Perez, Jose and others. 1998. “Dengue and Dengue Growth.” Journal of Economic Perspectives 12: 127–38. Haemorrhagic Fever.” Lancet. 352 (19/September). Quigley, John M., and Daniel L. Rubinfeld. 1997. “Federalism Roberts D., and K. DeRemer. 1997. “Overview of Foreign as a Device for Reducing the Budget of the Central Govern- Technical Barriers to U.S. Agricultural Exports.” Staff Paper ment.” Burch Working Paper B96-11. Burch Center for Tax 9705. ERS, U.S. Department of Agriculture, Washington, Policy and Public Finance, University of California, Berkeley. D.C. Quinn, Dennis. 1997. “The Correlation of Change in Interna- Rodrik, Dani. 1994. “Developing Countries after the Uruguay tional Financial Regulation.” American Political Science Re- Round.” Discussion Paper Series 1084. Centre for Eco- view 91: 700–36. nomic Policy Research, London. Quirk, Peter J., and Owen Evans. 1995. “Capital Account Con- ———. 1997. Has Globalization Gone Too Far? Washington, vertibility: Review of Experience and Implications for IMF D.C.: Institute for International Economics. Policies.” Occasional Paper 131. International Monetary ———. 1998a. “Symposium on Globalization in Perspective: Fund, Washington, D.C. An Introduction.” Journal of Economic Perspectives 12(4). Rabinovitch, Jonas. 1992. “Curitiba: Towards Sustainable 3–8. Urban Development.” Environment and Urbanization 4(2). ———. 1998b. “Where Did All the Growth Go? External Rabinovitch, Jonas, and Josef Leitman 1996. “Urban Planning Shocks, Social Conflict and Growth Collapses.” Discussion in Curitiba.” Scientific American (March). Paper 1789. CEPR (Centre for Economic Policy Research), Radelet, Steven, and Jeffrey D. Sachs. 1998. “The Onset of the London. East Asian Financial Crisis.” Harvard Institute for Interna- ———. 1998c. “Who Needs Capital-Account Convertibility?” tional Development, Cambridge, Mass. Symposium Paper to appear in Princeton Essays in Interna- Rajan, Raghuram G., and Luigi Zingales. 1998. “Financial De- tional Finance. Princeton University. pendence and Growth.” American Economic Review 88(3): Rodrik, Dani, and Andrés Velasco. 1999. “Short-Term Capital 559–85. Flows.” Paper prepared for the 1999 Annual World Bank Rajan, Ramkishen. 1998. “Regional Initiatives in Response to Conference on Development Economics. World Bank, the East Asian Crisis.” Institute of Policy Studies, Singapore. Washington, D.C. Processed. Rao, M. Govinda. 1999. “India: Intergovernmental Fiscal Rela- Rogers, Peter, and Peter Lydon, eds. 1994. Water in the Arab tions in a Planned Economy.” In Fiscal Decentralization in World: Perspectives and Progress. Cambridge: Harvard Uni- Developing Countries, edited by Richard M. Bird and François versity Press. Vaillancourt. Cambridge: Cambridge University Press. Roodman, David. 1997. “Reforming Subsidies.” In State of the Ravallion, Martin. 1998. “Reaching Poor Areas in a Federal Sys- World 1997, edited by Lester Brown. New York: W. W. tem.” Development Research Group, World Bank, Wash- Norton. ington, D.C. Processed. Root, Hilton L. 1998. “A Liberal India: The Triumph of Hope ———. 1999a. “Are Poorer States Worse at Targeting Their over Experience.” Asian Survey (38)5: 510–33. Poor?” World Bank, Washington, D.C. Processed. Rose, Greg, and Sandra Crane. 1995. “The Evolution of Inter- ———. 1999b. “Monitoring Targeting Performance When national Whaling Law.” In The Earthscan Reader in Sustain- Decentralized Allocations to the Poor are Unobserved.” De- able Development, edited by John Kirby, Phil O’Keefe, and velopment Research Group. Policy Research Working Paper Lloyd Timberlake. London: Earthscan. 2080. World Bank, Washington, D.C. Rosegrant, Mark. 1997. Water Resources in the Twenty-first Cen- Ravallion, Martin, and Shaohua Chen. 1998. “Poverty Reduc- tury: Challenges and Implications for Action. Washington, tion and the World Bank Progress in Fiscal 1998.” Harvard D.C.: International Food Policy Research Institute. Institute for International Development, Cambridge, Mass. Rosen, George. 1993. A History of Public Health. Baltimore: Forthcoming. The Johns Hopkins University Press. Redding, S. Gordon. 1998. “International Network Capitalism Roubini, Nouriel, and Jeffrey D. Sachs. 1989. “Political and Chinese Style.” Background Paper for World Development Economic Determinants of Budget Deficits in the Industrial Report 1999/2000. World Bank, Washington, D.C. Democracies.” European Economic Review 33: 903–38.    Rousseau, Marie-Paule. 1995. “Les Parisiens sont surproduc- Seckler, David, Upali Amarasinghe, David Molden, Radhika de tifs.” Etudes foncières (68): 13–18. Silva, and Randolph Barker. 1998. “World Water Demand and Rousseau, Peter L., and Paul Wachtel. 1998. “Financial Inter- Supply, 1990 to 2025: Scenarios and Issues.” Research Report mediation and Economic Performance: Historical Evidence 19. International Water Management Institute, Colombo, from Five Industrialized Countries.” Journal of Money, Sri Lanka. Credit, and Banking 30(4): 657–78. Sell, Susan. 1996. “North-South Environmental Bargaining: Russell, Sharon Stanton, and Michael S. Teitelbaum. 1992. In- Ozone, Climate Change and Biodiversity.” Global Gover- ternational Migration and International Trade. Discussion nance 2: 97–118. Paper 160. Washington, D.C.: World Bank. Serra, Jaime, Guillermo Aguilar, José Cordoba, Gene Gross- Ryan, Michael P. 1998. Knowledge Diplomacy: Global Competi- man, Carla Hills, John Jackson, Julius Katz, Pedro Noyola, tion and the Politics of Intellectual Property. Washington, and Michael Wilson. 1998. Reflections on Regionalism: Re- D.C.: The Brookings Institution Press. port of the Study Group on International Trade. Washington, D.C.: The Brookings Institution Press. Sabel, Charles. 1989. “Flexible Specialization and the Re- Sewell, David. 1996. “The Dangers of Decentralization Accord- emergence of Regional Economics.” In Reversing Industrial ing to Prud’homme: Some Further Aspects.” World Bank Re- Decline, edited by Paul Hirst and Jonthan Zeitlan. Oxford: search Observer 11(1). Oxford University Press. Shah, Anwar. 1997. “Federalism Reform Imperatives, Restruc- ———. 1998. “Local Development in Ireland; Partnership, In- turing Principles and Lessons for Pakistan.” Pakistan Devel- novation, and Social Justice.” Organisation for Economic opment Review 36(4 Part II Winter): 499–536. Co-operation and Development, Paris. ———. 1998. “Fiscal Federalism and Macroeconomic Gover- Sachs, Jeffrey D. 1996. “Growth in Africa: It Can Be Done.” nance.” Policy Research Working Paper 2005. World Bank, The Economist. June 29. Washington, D.C. Sachs, Jeffrey D., and Andrew Warner. 1997a. “Fundamental Shah, Ghanshyam. 1997. Public Health and Urban Develop- Sources of Long-run Growth.” American Economic Review ment: The Plague in Surat. New Delhi: Sage Publications. 87(2): 184–88. Sharma, Narenda, and others. 1996. African Water Resources: ———. 1997b. “Sources of Slow Growth in African Econ- Challenges and Opportunities for Sustainable Development. omies.” Journal of African Economies 6 (October): 335–76. Technical Paper 331. Washington, D.C.: World Bank. Sachs, Jeffrey D., and Warwick J. McKibbin. 1991. Global Short, Clare. 1999. “Future Multilateral Trade Negotiations: A Linkages. Washington, D.C.: The Brookings Institution ‘Development Round’?” Speech presented at United Na- Press. tions Conference on Trade and Development, Geneva. Sala-i-Martin, Xavier. 1997. “I Just Ran Four Million Regres- Shugart, Chris. 1997. “Decentralization and the Challenges of sions.” Working Paper Series 6252. National Bureau of Eco- Regulation for Local-Level Public Services in Central and nomic Research, Cambridge, Mass. Eastern Europe.” Extended version of a commentary given Samuels, David J. 1998. “Institutions of Their Own Design? at an EBRD seminar on Commercial Infrastructure Chal- Democratization and Fiscal Decentralization in Brazil, lenges in Transition Economies. April 12. 1975–95.” Paper presented at the American Political Science Shukla, Vibhotti. 1996. Urbanization and Economic Growth. Association. Boston. Processed. Delhi: Himalaya Publishing House. Satterthwaite, David. 1996. “Sustainable Cities or Cities that Simpson, R., R. Sedjo, and J. Reid. 1996. “Valuing Biodiver- Contribute to Sustainable Development.” Urban Studies sity for Use in Pharmaceutical Research.” Journal of Political 39(4). Economy 104: 163–85. Sims, Holly. 1999. “One Fifth of the Sky: China’s Environmen- Saudagaran, Shahrokh, and Joselita G. Diga. 1997. “Financial tal Stewardship.” World Development 27(7): 1227–45. Reporting in Emerging Capital Markets: Characteristics and Skeldon, R. 1998. Migration and Development: A Global Policy Issues.” Accounting Horizons 11(2): 41–64. Perspective. Harlow, U.K.: Addison Wesley Longman. Schiff, Maurice, and L. Alan Winters. 1998. “Dynamics and Smith, B. C., 1996. “Sustainable Local Democracy.” Public Ad- Politics in Regional Integration Arrangements: An Introduc- ministration and Development 16: 164–78. tion.” World Bank Economic Review 12(2): 177–96. Solo, Tova Maria. 1999. “Small-Scale Entrepreneurs in the Schlicht, E. 1985. “The Emotive and Cognitive View of Jus- Urban Water and Sanitation Market.” Environment and Ur- tice.” Institute for Advanced Studies, Princeton University. banization 11(1): 117. Processed. Solo, Tova, and Suzanne Snell. 1998. “Water and Sanitation Science. 1997. “Human Domination of Earth’s Ecosystems.” Services for the Urban Poor: Small-Scale Providers—Profiles July 25. and Typology.” UNDP–World Bank Water and Sanitation ———. 1999. “Lack of Icebergs Another Sign of Global Program, Washington, D.C. Draft. Warming.” July 2. Souza, Celina. 1996. “Redemocratization and Decentralization Scott, James C. 1976. The Moral Economy of the Peasant. New in Brazil: The Strength of the Member States.” Development Haven: Yale University Press. and Change 27: 529–55. Seaver, Brenda. 1997. “Stratospheric Ozone Protection: IR Spahn, Paul Bernd. 1998. “Decentralized Government and Theory and the Montreal Protocol on Substances that De- Macroeconomic Control.” University of Frankfurt am plete the Ozone Layer.” Environmental Politics 6(3): 31–67. Main, Frankfurt. Processed.      ⁄  Srinivasan, T. N. 1998. Developing Countries and the Multilat- ceedings of the Annual World Bank Conference on Development eral Trading System. Boulder, Colo.: Westview Press. Economics 1995. World Bank, Washington, D.C. Stein, Ernesto, Ernesto Talvi, and Alejandro Grisanti. 1998. Tarver, James D. 1995. Urbanization in Africa: A Handbook. “Institutional Arrangements and Fiscal Performance: The London: Greenwood Press. Latin American Experience.” Working Paper 6358. National Taylor J. Edward, Joaquin Arango, Graeme Hugo, Aki Bureau of Economic Research, Cambridge, Mass. Kouaouci, Douglas S. Massey, and Adela Pellegrino. 1996. Stepan, Alfred. 1999. “Toward a New Comparative Analysis of “International Migration and National Development.” Pop- Democracy and Federalism: Demos Constraining and ulation Index 62(2/Summer): 181–212. Demos Enabling Federations.” In Arguing Comparative Pol- Teitelbaum, Michael S., and Jay Winter. 1998. A Question of itics, by Alfred Stephan. Oxford: Oxford University Press. Numbers. New York: Hill and Wang. Forthcoming. Tendler, Judith. 1997. Good Government in the Tropics. Balti- Stiglitz, Joseph E. 1996. “Some Lessons from the East Asian more: The John Hopkins University Press. Miracle.” World Bank Research Observer 11(2). Ter-Minassian, Teresa. 1997. “Intergovernmental Fiscal Rela- ———. 1997. “Stepping Towards Balance: Addressing Global tions in a Macroeconomic Perspective: An Overview.” In Climate Change.” Speech delivered at the Conference on Fiscal Federalism in Theory and Practice, edited by Teresa Environmentally and Socially Sustainable Development. Ter-Minassian. Washington, D.C.: International Monetary Washington D.C., October 6. Fund. ———. 1998a. “More Instruments and Broader Goals: Mov- Ter-Minassian, Teresa, and Jon Craig. 1997. “Control of Sub- ing toward the Post-Washington Consensus.” The 1998 national Government Borrowing.” In Fiscal Federalism in WIDER Annual Lecture. Helsinki, January 7. Theory and Practice, edited by Teresa Ter-Minassian. Wash- ———. 1998b. “Towards a New Paradigm for Development: ington, D.C.: International Monetary Fund. Strategies, Policies, and Processes.” Prebisch Lecture. United Tharakan, P. K. M. 1999. “Is Anti-Dumping Here to Stay?” Nations Conference on Trade and Development, Geneva. The World Economy 22(2): 179–206. ———. 1999a. “Principles of Financial Regulation: A Dy- The Sciences. 1997. “The Hole the World Is Watching.” July namic, Portfolio Approach.” World Bank, Washington, 25. D.C. Processed. Thomas, Harmon, and John Whalley, eds. 1998. Uruguay ———. 1999b. “Two Principles for the Next Round: Or, How Round Results and the Emerging Trade Agenda: Quantitative- to Bring Developing Countries in from the Cold.” Speech Based Analyses from the Development Perspective. Geneva: delivered in Stockholm, Sweden, April 12. United Nations Conference on Trade and Development. Stiglitz, Joseph E., and Marilou Uy. 1996. “Financial Markets, Thomas, Vinod. 1980. “Spatial Differences in the Cost of Public Policy, and the East Asian Miracle.” The World Bank Living.” Journal of Urban Economics 8: 108–22. Research Observer 11(2). ———. 1999. “Revisiting the Challenge of Development.” Stotsky, Janet G., and Emil M. Sunley. 1997. “United States.” World Bank, Washington, D.C. Processed. In Fiscal Federalism in Theory and Practice, edited by Teresa Tibaijuka, Anna Kajumulo. 1997. “AIDS and Economic Wel- Ter-Minassian. Washington, D.C.: International Monetary fare in Peasant Agriculture. Case Studies from Kagabiro Vil- Fund. lage, Kagera Region, Tanzania.” World Development 25(6): Stren, Richard. 1998. “Urban Governance and Politics in a 963–75. Global Context: The Growing Importance of the Local.” Tiebout, Charles. 1956. “A Pure Theory of Local Expendi- Background Paper for World Development Report 1999/ tures.” Journal of Political Economy 64(5): 416–24. 2000. World Bank, Washington, D.C. Tomich, Thomas P., Peter Kilby, and Bruce F. Johnston. 1995. Strong, Ann L., Thomas A. Reiner, and Janusz Szyrmer. 1996. Transforming Agrarian Economies: Opportunities Seized, Op- Transitions in Land and Housing: Bulgaria, the Czech Repub- portunities Missed. Ithaca: Cornell University Press. lic, and Poland. New York: St. Martin’s Press. Tonry, Michael. 1997. Ethnicity, Crime and Immigration. Struyk, Raymond J., ed. 1997. Restructuring Russia’s Housing Sec- Chicago: University of Chicago Press. tor: 1991–1997. Washington, D.C.: The Urban Institute. Transparency International. 1998. “1998 Corruption Percep- Summers, Robert, and Alan W. Heston. 1988. “A New Set of tion Index.” Berlin. International Comparisons of Real Product and Prices: Esti- Treisman, Daniel. 1998. After the Deluge: Regional Crisis and mates for 130 Countries: 1950–1985.” International Asso- Political Consolidation in Russia. Ann Arbor: University of ciation for Research in Income and Wealth, Philadelphia. Michigan Press. Szreter, Simon. 1997. “The Politics of Public Health in Nine- Trela, Irene. 1998. “Phasing Out the MFA in the Uruguay teenth Century Britain.” Population and Development Review Round: Implications for Developing Countries.” In Uruguay 3(4/December). Round Results and the Emerging Trade Agenda, edited by Tacoli, Cecilia. 1998. “Rural-Urban Interactions: A Guide to Harmon Thomas and John Whalley. New York: United the Literature.” Environment and Urbanization 10(1/April). Nations. Tanzi, Vito. 1995. Taxation in an Integrating World. Washing- Tsur, Yacov, and Ariel Dinar. 1995. “Efficiency and Equity ton, D.C.: The Brookings Institution Press. Considerations in Pricing and Allocating Irrigation Water.” ———. 1996. “Fiscal Federalism and Decentralization: A Re- Policy Research Working Paper 1460. World Bank, Wash- view of Some Efficiency and Macroeconomic Aspects.” Pro- ington, D.C.    Tybout, James. 1998. “Manufacturing Firms in Developing Vehorn, Charles L., and Ehtisham Ahmad. 1997. “Tax Admin- Countries: How Well They Do and Why?” Policy Research istration.” In Fiscal Federalism in Theory and Practice , edited Working Paper 1965. World Bank, Washington, D.C. by Teresa Ter-Minassian. Washington, D.C.: International Tyler, Tom. 1997. “Citizen Discontent with Legal Procedures: Monetary Fund. A Social Science Perspective on Civil Procedure Reform.” Velasco, Andres, and Pablo Cabezas. 1998. “Alternative Re- American Journal of Comparative Law 45(4). sponses to Capital Inflows: A Tale of Two Countries.” In Tynan, Nicola, and Tyler Cowen. 1998. “The Private Provision Capital Flows and Financial Crises, edited by Miles Kahler. of Water in 18th and 19th Century London.” Department Ithaca: Cornell University Press. of Economics, George Mason University. Draft. Processed. Venables, Anthony. 1998. “Cities and Trade: External Trade U.S. Central Intelligence Agency. 1998. The World Factbook. and Internal Geography in Developing Economies.” World Washington, D.C. Bank, Washington, D.C. Processed. U.S. Committee for Refugees. 1996. World Refugee Survey. Verdier, David. 1998. “Domestic Responses to Capital Market Washington, D.C. Internationalization under the Gold Standard 1870–1914.” UNCHS (United Nations Center for Human Settlements). International Organization 52(1). 1995. Global Urban Indicators Database. Nairobi. Vernon, Raymond. 1998. In the Hurricane’s Eye: The Troubled ———. 1996. An Urbanizing World: Global Report on Human Prospects of Multinational Enterprises. Cambridge, Mass.: Settlements 1996. Oxford: Oxford University Press. Harvard University Press. UNCTAD (United Nations Conference on Trade and Devel- Vittas, Dimitri, and Craig Neal. 1992. “Competition and Ef- opment). 1996. World Investment Report 1996. New York ficiency in Hungarian Banking.” Policy Research Work- and Geneva. ing Paper 1010, World Bank, Washington, D.C. ———. 1997. World Investment Report 1997: Transnational Vivian, Jessica. 1994. “NGOs and Sustainable Development in Corporations, Market Structure and Competition Policy. New Zimbabwe: No Magic Bullets.” Development and Change York and Geneva. 25(1): 167–93. ———. 1998. World Investment Report 1998. New York and von Hagen, Jürgen. 1991. “A Note on the Empirical Effective- ness of Formal Fiscal Restraints.” Journal of Public Econom- Geneva. ics 44: 199–210. UNDIESA (United Nations Department of International Eco- Wacziarg, Romain. 1998. “Measuring the Dynamic Gains from nomic and Social Affairs). 1996. World Urbanization Trade.” Policy Research Working Paper 2001. World Bank, Prospects. New York. Washington, D.C. ———. 1998. World Urbanization Prospects. New York. Wallensteen, P., and A. Swain. 1997. “Comprehensive Assess- UNDP (United Nations Development Programme). 1998. ment of the Water Resources of the World.” Stockholm En- Human Development Report. New York: Oxford University vironment Institute, Stockholm. Press. Wallerstein, Immanuel. 1974. The Modern World System. New ———. 1999. Human Development Report. New York: Oxford York: Academic Press. University Press. Walsh, Christopher. 1999. “Deconstructing Vancomycin.” Sci- UNEP (United Nations Environment Programme). 1999. Syn- ence. April 16. thesis of the Reports of the Scientific, Environmental Effects, and Walt, Gill. 1998. “Globalization of International Health.” Technological and Economic Assessment Panels of the Montreal Lancet 351(7/February). Protocol. Nairobi. Wambugu, Florence. 1999. “Why Africa Needs Agricultural UNESCO (United Nations Education, Scientific, and Cultural Biotech.” Nature. July 1. Organization). 1999. Progress of Nations. New York. Wapner, Paul. 1995. “The State and Environmental Chal- United Kingdom, Commonwealth Secretariat. 1998. “Report lenges: A Critical Exploration of Alternatives to the State of the Expert Group on Protecting Countries against Desta- System.” Environmental Politics 4(1): 44–69. bilizing Effects of Volatile Capital Controls.” London. Watson, Gabrielle. 1995. “Good Sewers Cheap? Agency- Urata, Shujiro. 1996. “Trade Liberalization and Productivity Customer Interactions in Low-Cost Urban Sanitation in Growth in Asia: Introduction and Major Findings.” Devel- Brazil.” World Bank, Washington, D.C. Processed. oping Economies 32(4). Watson, Robert Tony, John Dixon, Stephen Hamburg, An- Vaillancourt, François. 1998. “Morocco and Tunisia: Financ- thony Janetos, and Richard Moss. 1998. “Protecting Our ing Local Governments—The Impact of Infrastructure Fi- Planet, Securing Our Future: Linkages among Global Envi- nance.” In Fiscal Decentralization in Developing Countries, ronmental Issues and Human Needs.” United Nations En- edited by Richard Bird and François Vaillancourt. Cam- vironment Programme, Nairobi; U.S. National Aeronautics bridge: Cambridge University Press. and Space Administration, Washington, D.C.; World Bank, Valdes-Prieto, Salvador, and Marcelo Soto. 1996. “New Selec- Washington, D.C. tive Capital Controls in Chile: Are They Effective?” Univer- Wei, Shang-Jin, and Jeffrey Frankel. 1996. “Can Regional Blocs sidad Católica de Chile, Santiago. Unpublished manuscript. Be Stepping Stones to Global Free Trade?” Review of Inter- Vamvakidis, Athanasios, and Roman Wacziarg. 1998. “Devel- national Economics and Finance 5(4). oping Countries and the Feldstein-Horioka Puzzle.” Work- Weingast, Barry R. 1995. “The Economic Role of Political Insti- ing Paper 98/2. International Monetary Fund, Washington, tutions: Market-Preserving Federalism and Economic Devel- D.C. opment.” Journal of Law, Economics, and Organization 11(1).      ⁄  Wetzel, Deborah, and Jonathan Dunn. 1998. “Decentralization ———. 1993b. World Development Report 1993: Investing in in the ECA Region: Progress and Prospects.” World Bank, Health. New York: Oxford University Press. Washington, D.C. Processed. ———. 1994. World Development Report 1994: Infrastructure Whittington, Dale, D. T. Lauria, and X. Mu. 1991. “A Study for Development. New York: Oxford University Press. of Water Vending and Willingness to Pay for Water in Onit- ———. 1995a. China: Macroeconomic Stability in a Decentral- sha, Nigeria.” World Development 19(2/3). ized Economy. A World Bank Country Study. Washington, Whittington, Dale, D. T. Lauria, A. M. Wright, K. Choe, J. D.C. Hughes, and V. Swarna. 1993. “Household Demand for Im- ———. 1995b. Local Government Capacity in Colombia: Be- proved Sanitation Services in Kumasi, Ghana: A Contingent yond Technical Assistance. A World Bank Country Study. Valuation Study.” Water Resources Research 29(6). Washington, D.C. WHO (World Health Organization). 1986. “Intersectoral Ac- ———. 1995c. World Development Report 1995: Workers in an tion for Health.” Geneva. Integrating World. New York: Oxford University Press. ———. 1995. World Health Report 1995: Bridging the Gaps. ———. 1996a. Sustainable Transport: Priorities for Policy Re- Geneva. form. Washington, D.C.: World Bank. ———. 1999. World Health Report 1999. Geneva. ———. 1996b. “The Hashemite Kingdom of Jordan: Housing Wildasin, David E. 1997. “Externalities and Bailouts.” Policy Finance and Urban Sector Reform Project.” Staff Appraisal Research Working Paper 1843. World Bank, Washington, Report. Washington, D.C. D.C. ———. 1996c. World Development Report 1996: From Plan to Wildavsky, A. 1995. But Is It True? Cambridge, Mass.: Harvard Market. New York: Oxford University Press. University Press. ———. 1997a. Clear Water, Blue Skies: China’s Environment Williamson, Jeffrey G. 1998. “Globalization, Labor Markets in the New Century. Washington, D.C. and Policy Backlash in the Past.” Journal of Economic Per- ———. 1997b. Five Years after Rio: Innovations in Environmen- spectives 12(4): 51–72. tal Policy. Environmentally Sustainable Development Stud- Willis, Eliza, Christopher da C. B. Garman, and Stephan Hag- ies and Monographs Series 18. Washington, D.C. gard. 1999. “The Politics of Decentralization in Latin Amer- ———. 1997c. Private Capital Flows to Developing Countries. ica.” Latin American Research Review 34(1): 7–56. New York: Oxford University Press. Wiseman, John H. 1997. “The Rise and Fall and Rise (and ———. 1997d. World Development Report 1997: The State in a Changing World. New York: Oxford University Press. Fall?) of Democracy in Sub-Saharan Africa.” In Democrati- ———. 1998a. Assessing Aid: What Works, What Doesn’t, and zation, edited by David Potter, David Goldblatt, Margaret Why. New York: Oxford University Press. Kiloh, and Paul Lewis. Cambridge: Polity Press. ———. 1998b. East Asia: The Road to Recovery. Washington, Wolf, Holger. 1999. “International Asset Price and Capital D.C. Flow Comovements during Crisis: The Role of Contagion, ———. 1998c. Egypt in the Global Economy: Strategic Choices Demonstration Effects and Fundamentals.” Paper presented for Savings, Investments, and Long-Term Growth. MENA at the World Bank/International Monetary Fund/World Economic Studies. Washington, D.C. Trade Organization Conference on Capital Flows, Finan- ———. 1998d. Environmental and Socially Sustainable Devel- cial Crises, and Policies. World Bank, Washington, D.C. opment Website (http://www-esd.worldbank.org ). Processed. ———. 1998e. “Forest Carbon Action Plan: World Bank’s Wolfensohn, James D. 1998. “The Other Crisis.” Address to Program to Improve the Understanding of the Climate Ben- the Board of Governors of the World Bank. Washington, efits of Forestry and Land Management.” Washington, D.C. D.C., October 6. Processed. Wong, Christine P. 1998. “A Note on the Outcomes of the ———. 1998f. “Fuel for Thought. A New Environmental 1994 Fiscal Reforms.” World Bank Resident Mission in Strategy for the Energy Sector.” Washington, D.C. China, Note 3. January 23. World Bank. ———. 1998g. GEF Programs at (http://www.gefweb.org/ Woolcock, Michael. 1998. “Social Capital and Economic De- OPERPORT/PROGLIST.PDF ). velopment: Toward a Theoretical Synthesis and Policy ———. 1998h. Global Economic Prospects 1998/99. Washing- Framework.” Theory and Society (27)2. ton, D.C. World Bank. 1979. World Development Report 1979. New York: ———. 1998i. “India: Urban Infrastructure Services Report. Oxford University Press. Draft Country Study.” Washington, D.C. ———. 1990. World Development Report 1990: Poverty. New ———. 1998j. “Meeting the Challenge for Rural Energy and York: Oxford University Press. Development.” Washington, D.C. ———. 1991. World Development Report 1991: The Challenge ———. 1998k. “Questions and Answers on the World Bank of Development. New York: Oxford University Press. and Climate Change.” Washington, D.C. Processed. ———. 1992a. Governance and Development. World Bank, ———. 1998l. World Development Indicators 1998. Washing- Washington, D.C. ton, D.C. ———. 1992b. World Development Report 1992: Development ———. 1998m. World Development Report 1998: Knowledge and the Environment. New York: Oxford University Press. for Development. New York: Oxford University Press. ———. 1993a. “Indonesia: Environment and Development: ———. 1998n. New Opportunities for Development: The Deser- Challenges for the Future.” Washington, D.C. Processed. tification Convention. Washington, D.C.    ———. 1998o. “Agricultural Policy Reform and the Least De- ———. 1996. Annual Report. Geneva. veloped and Net Importing Countries.” Washington, D.C. ———. 1997. Annual Report. Geneva. Processed. ———. 1998a. Annual Report. Geneva. ———. 1999a. “A Proposal for a Comprehensive Development ———. 1998b. “Financial Services.” Background Note by the Framework.” Processed. Secretariat. Geneva. ———. 1999b. “Ethiopia Regionalization Study.” Report Xie, Danyang, Heng-fu Zou, and Hamid Davoodi. 1999. “Fis- 188898-ET. Washington, D.C. cal Decentralization and Economic Growth in the United ———. 1999c. Global Development Finance 1999. Washing- States.” Journal of Urban Economics 45: 1–12. ton, D.C. Yanagihara, Toru. 1998. “Regional Policy Coordination in ———. 1999d. “Greening Industry: New Roles for Commu- Asia.” Background Paper for World Development Report nities, Markets and Governments.” Washington, D.C. 1999/2000. Processed. ———. 1999e. “Poverty Reduction and the World Bank Yeates, Noel R. 1997. “Creating a Global City: Recent Changes Progress in Fiscal 1998.” Washington, D.C. Processed. to Sydney’s Economic Structure.” In Pacific Rim Develop- ———. 1999f. “The Child Labor Program.” Human Develop- ment: Integration & Globalization in the Asia-Pacific Econ- ment Network. Washington, D.C. omy, edited by Peter Rimmer. Canberra City, Australia: ———. 1999g. “Trade Blocs and Beyond.” Policy Research Aussie Print. Report. Washington, D.C. Draft. Yeats, Alexander J. 1996. “Does Mercosur’s Trade Performance ———. 1999h. “Uzbekistan Structural Policy Review: An Eco- Raise Concerns about the Effects of Regional Trade nomic Report.” Washington, D.C. Arrangements?” Policy Research Working Paper 1729. ———. 1999i. World Development Indicators 1999. Washing- World Bank, Washington, D.C. ton, D.C. Yeats, Alexander J. 1998. “Just How Big is Global Production ———. 1999j. “Tanzania: Peri-Urban Development in the Sharing?” Policy Research Working Paper 1871. World African Mirror.” Report No. 19526-TA. Washington, D.C. Bank, Washington, D.C. ———. 1999k. “Africa Regional Gender Action Plan.” Wash- Yudkin, John S. 1999. “Tanzania: Still Optimistic after All ington, D.C. Processed. These Years?” Lancet 353(May): 1519–21. WRI (World Resources Institute). 1996. The Urban Environ- Zaidi, S. Akbar. 1997. “Politics, Institutions, and Poverty: The ment 1996–97. A joint publication by WRI, UNEP, Case of Karachi.” Economic and Political Weekly 32(5). UNDP, and IBRD. New York: Oxford University Press. ———. 1998. “Urban Safety and Crime Prevention.” UNCHS ———. 1998. World Resources 1998–99: Environmental Change Regional Symposium on Urban Poverty. Fukuoka, Japan, and Human Health. A joint publication by WRI, UNEP, October 27–29. UNDP, and IBRD. New York: Oxford University Press. ———. 1999. “NGO Failure and the Need to Bring Back the WRI (World Resources Institute) and WHO (World Health State.” Journal of International Development. Forthcoming. Organization). 1999. “An International Comparative Study Zlotnik, Hania. 1998. “International Migration 1965–96. of Air Pollution Health Risks in Major Urban Areas in De- An Overview.” Population and Development Review 24(3): veloped and Developing Countries.” Washington, D.C. 429–68. Processed. Zurn, Michael. 1998. “The Rise of International Environmen- WTO (World Trade Organization). 1995. Annual Report. tal Politics: A Review of Current Research.” World Politics Geneva. 50: 617–49. A p p e n d i x Selected Indicators on Decentralization, Urbanization, and the Environment T he following data presentation conven- Structures across Countries,” by Mari- tions are used in this section. anne Fay, a World Development Report Italics indicate data for years or peri- 1999/2000 background paper. The data ods other than those specified. The clos- were compiled from a variety of sources, est available year is shown instead. including the Area Handbook Series Aggregates of ratios are generally cal- published by the Federal Research Divi- culated as weighted averages of the ratios sion of the Library of Congress; the CIA (indicated by the letter w), using the World Factbook 1998; The Statesman’s value of the denominator as a weight. Yearbook 1998–99; Local Finance in the The letter t denotes totals where Fifteen Countries of the European Union, missing values are imputed. published by DEXIA in 1997; The Di- The letter s denotes totals where rectory of Local Government Systems in missing values are not imputed. Africa, published in 1998 by the Munic- The symbol – means not applicable. ipal Development Program; “Decentral- The symbol . . means not available. ization in the ECA Region: Progress and Prospects,” by Deborah Wetzel and Table A.1. Decentralization Jonathan Dunn, a World Bank paper Data on national and subnational gov- (1998); and various World Bank coun- ernment revenue and expenditure are try reports. This information was cross- from the electronic edition of the Gov- checked with World Bank country teams ernment Finance Statistics Yearbook (GFS) and country diplomatic representatives of the International Monetary Fund. in Washington. Data on subnational elections, on tiers of Share of subnational government in elected subnational governments, and on total public expenditure is calculated the number of jurisdictions are from from information in the GFS. It is the “How Many Tiers? How Many Jurisdic- ratio of expenditure by subnational (in- tions? A Review of Decentralization termediate and local) governments to       ⁄  total expenditure by all levels of government. Both cur- ple, Tehran, Islamic Republic of Iran, contains 6.8 mil- rent and capital transfers among levels of government are lion people in the 700-square-kilometer core of the city, excluded to prevent double counting. Using the IMF’s but the greater metropolitan area covers 2,100 square GFS codes, the formula is [CII local + (CIII – C3.2 – kilometers and is home to more than 10 million. Thus, C7.1.1) intermediate]/[CII local + (CIII – C3.2 – depending on which boundaries are used, Tehran’s pop- C7.1.1) intermediate + (CIII – C3.2 – C7.1.1) consoli- ulation can vary from 11 percent to 18.5 percent of the dated central government]. Share of subnational gov- total population of Iran. ernment in total tax revenue is calculated from infor- When urban boundaries are redefined in the world’s mation in the GFS. It is the ratio of tax revenue (GFS more populous countries, such as China or India, it can code AIV) collected by subnational governments to total significantly alter the estimate of the world’s urban pop- tax revenue collected by all levels of government. The ulation. In the mid-1990s, for example, when China’s 1990 figure is in italics if the data were not available for State Statistical Bureau reclassified many of the coun- 1990 and were replaced by data from the year, in the pe- try’s hundreds of towns as cities, it more than doubled riod 1988–92, closest to 1990 for which data were avail- the measured share of China’s urban population. At the able. The 1997 figure is in italics if the data were not end of 1996 about 43 percent of the country’s popula- available for 1997 and were replaced by data for the most tion was considered urban, compared with only 20 per- recent year for which data were available in the period cent in 1994. Estimates by international organizations 1993–97. The entry for either column under subna- such as the United Nations and the World Bank indi- tional elections is “Yes” if the most recent data indicate cate that 47 percent of the world’s population is urban, that elections are held at that level and that an elected but using the new figures for China would suddenly in- government is currently in place. “No+” indicates that, crease that share to more than half. Because the esti- although the legislature is elected, a nominated executive mates in the table are based on national definitions of head (for example, a mayor or governor) holds signifi- what constitutes an urban area, cross-country compar- cant powers. Number of elected subnational tiers in- isons should be made with caution. dicates the number of tiers of currently sitting elected Aggregate measures for regions and income groups government below the central or federal government. include all 210 economies for which data are available. It excludes subnational governments headed by an ap- Urban population is the combined midyear popu- pointed executive who holds significant powers. Num- lation of all areas defined as urban in each country, as ber of jurisdictions indicates, for each tier of sub- reported to the United Nations. Urban population by national elected government, the number of separate size of city shows a breakdown of the urban population jurisdictions at that tier. At the intermediate level, it in- according to city size. Population share of largest city dicates the number of states (in federations), provinces, is the percentage share of the urban population living or province equivalents; at the local level, it indicates the in the country’s largest metropolitan area. This is a mea- number of municipalities or equivalent local govern- sure of concentration of the urban population. Access ments. Comparisons should be made with care, as the to sanitation in urban areas is the share of the urban size and functions of subnational governments vary from population served by connections to public sewers or country to country, and even within countries. other systems such as pit privies, pour-flush latrines, septic tanks, communal toilets, and similar facilities. Table A.2. Urbanization Table A.3. Urban living conditions Data on urban population are from the United Na- tions’ World Urban Prospects: The 1996 Revision. Total Data are from the Global Urban Indicators database of population figures are World Bank estimates. Data on the Urban Indicators Programme of the United Nations access to sanitation in urban areas are from the World Centre for Human Settlements (UNCHS). The table Health Organization. The table includes those econ- shows selected indicators and cities from the UNCHS omies with populations exceeding 1 million for which data set, which covers 46 key urban indicators and 237 data are available for at least 5 of the 11 indicators, in- cities. Cities are included in the table if data for at least cluding the most recent data on access to sanitation. 6 indicators were available out of the 11 shown. Estimates of the population of a city or metropoli- The data should be used with care. Countries may tan area depend on the boundaries chosen. For exam- use different data collection methods and definitions,   making comparisons misleading. Also, the sample is bi- tricity and fossil fuel production are from the Interna- ased toward smaller cities. Data are available only for tional Energy Agency. Data on biodiversity are from 1993, so no inferences can be made about conditions the World Conservation Monitoring Center’s Biodi- improving or worsening. versity Data Sourcebook 1994 and the World Conser- Urban area refers to the city proper, the suburban vation Union’s (IUCN) 1997 IUCN Red List of Threat- fringe, and any other built-up, thickly settled areas ened Animals and 1997 IUCN Red List of Threatened lying outside but adjacent to the city boundaries. Plants. Data on fisheries are from the Yearbook of Fish- Urban population refers to the population of the ery Statistics, volume 82, published by the Food and urban agglomeration, a contiguous inhabited territory Agriculture Organization (FAO), supplemented by defined without regard to administrative boundaries. data that the FAO makes available electronically to the Average household income is the average of house- World Bank. Data selection is based on availability and hold incomes by quintile. Household income is income on the global significance of each economy on these of all household members from all sources, including measures. Economies are included if their carbon diox- wages, pensions or benefits, business earnings, rents, ide emissions exceed 2 percent of the world total, fos- and the value of any business or subsistence products sil fuel production is over 50 million metric tons, the consumed (for example, foodstuffs). Income differen- number of threatened bird and mammal species ex- tial is the ratio of the average household income in the ceeds 100, or the marine fish catch is over 10 million top quintile to that in the bottom quintile. House metric tons. The aggregate measures by income level price–income ratio is the average house price divided and region include all economies (out of a maximum by the average household income. Crowding is mea- of 210) for which data are available and aggregation sured as the median floor area of usable living space per is possible. person. Work trips by public transportation is the Carbon dioxide emissions refers to emissions stem- percentage of trips to work made by bus or minibus, ming from the burning of fossil fuels and the manufac- tram, or train. Other means of transport commonly ture of cement. It includes carbon dioxide produced used in developing countries, such as taxis, ferries, rick- during consumption of solid, liquid, and gas fuels and shaws, or animals, are not included. Travel time to gas flaring. Electricity production is measured at the work is the average time in minutes, for all modes, for terminals of all alternator sets at the power station. The a one-way trip to work. Households with sewerage percentage from fossil fuel is the share produced by oil, connection is the percentage of households with a con- petroleum products, coal, and natural gas. Fossil fuel nection to sewerage. Households with regular waste production is total production of all types of fossil collection is the percentage of households served by fuels, converted to metric tons of crude oil of equiva- regular waste collection, whether household-by-house- lent energy content. Mammal and bird species ex- hold collection or regular “dumpster” group collection. cludes whales and includes birds within wintering It does not include households that transport their own ranges of countries. Higher plant species refers to garbage to a local dump. Households with access to native vascular plant species. The number of species potable water is the percentage of households with ac- threatened is the number classified by the IUCN as en- cess to potable water within 200 meters of the dwelling, dangered, vulnerable, rare, indeterminate, previously where potable water is water that is free from contami- endangered but now stabilized, or insufficiently known. nation and safe to drink without further treatment. Annual marine catch is the total catch of fish taken for all purposes (commercial, industrial, recreation, and Table A.4. Environment subsistence) by all types and classes of fishing units Data on carbon dioxide emissions are from the Carbon (individual fishermen, fishing vessels, etc.) from the wa- Dioxide Information Analysis Center, which is spon- ters of the Atlantic, Indian, and Pacific Oceans and sored by the U.S. Department of Energy. Data on elec- their adjacent seas.      ⁄ Table A.1. Decentralization Fiscal decentralization Electoral decentralization Share of subnational government (%) Subnational electionsa No. of elected No. of jurisdictions In total public expenditure In total tax revenue Intermediateb Localc subnational tiers Intermediateb Localc Economy 1990 1997 1990 1997 1999 1999 1999 1999 1999 Albania .. 24.9 .. 0.9 No Yes 1 .. 374 Algeria .. .. .. .. No+ No+ 0 48 1,552 Angola .. .. .. .. No No 0 .. .. Argentina 46.3 43.9 38.2 41.1 Yes Yes 2 24 1,617 Armenia .. 5.1d .. 3.3d No Yes 1 .. 931 Australia 50.9 47.9 20.0 22.7 Yes Yes 2 8 900 Austria 31.9 32.2 21.7 20.7 Yes Yes 2 9 2,353 Azerbaijan .. .. .. .. No No 0 .. .. Bangladesh .. .. .. .. No Yes 1e .. 4,642 Belarus 30.6 32.5 29.4 23.7 No No+ 0 .. 179 Belgium 11.9 11.8 4.5 5.4 Yes Yes 2 10 589 Benin .. .. .. .. No No 0f .. 77 Bolivia 17.7 36.3 15.1 19.1 No+ Yes 1 9 312 Bosnia and Herzegovina .. .. .. .. Yes Yes 3g 2 137 Botswana 7.9 3.8 0.1 0.6 No Yes 1 .. 17 Brazil 35.3 36.5 30.9 31.3 Yes Yes 2 28 5,581 Bulgaria 18.9 15.7 22.4 11.8 No Yes 1 .. 294 Burkina Faso .. .. .. .. Yes Yes 2 45 250 Burundi .. .. .. .. No No 0 .. .. Cambodia .. .. .. .. No No 0h .. .. Cameroon .. .. .. .. No Yes 1 .. 336 Canada 58.7 49.4 49.5 43.5 Yes Yes 2 12 4,507 Central African Republic .. .. .. .. No Yes 1 .. 174 Chad .. .. .. .. No No 0 .. .. Chile 7.2 8.5 6.4 7.0 No Yes 1 .. 340 China .. 55.6 .. 51.4 No No 0 .. .. Colombia .. .. .. .. Yes Yes 2 33 1,068 Congo, Dem. Rep. .. .. .. .. No No 0 .. .. Costa Rica 3.0 2.8 2.3 2.3 No No+ 0i .. 496 Côte d’Ivoire .. .. .. .. No+ Yes 1 50 196 Croatia .. 12.1 .. 7.5 Yes Yes 2 21 543 Cuba .. .. .. .. Yes Yes 2 15 169 Czech Republic .. 21.3 .. 12.3 No Yes 1 .. 5,768 Denmark 54.8 54.5 31.1 31.5 Yes Yes 2 16 275 Dominican Republic 1.6 2.6 0.5 0.2 No Yes 1 .. 90 Ecuador .. .. .. .. Yes Yes 2 21 1,079 Egypt, Arab Rep. .. .. .. .. No No+ 0 .. 199 El Salvador .. .. .. .. No Yes 1 .. 262 Eritrea .. .. .. .. No+ Yes 1j 6 .. Estonia 34.8 22.4 26.5 14.2 No Yes 1 .. 254 Ethiopia 1.5 .. 1.6 .. Yes Yes 2 11 910 Finland 46.5 41.2 25.9 27.6 No Yes 1 .. 455 France 18.7 18.6 9.7 10.8 Yes Yes 3 22 36,559 Georgia .. .. .. .. No Yes 1 or 2 .. 4,000 Germany 40.2 37.8 28.9 28.8 Yes Yes 3 16 16,121 Ghana .. .. .. .. No Yes 1 .. 110 Greece .. .. .. .. Yes Yes 2 13 5,922 Guatemala 10.1 10.3 1.3 1.7 No Yes 1 .. 324 Guinea .. .. .. .. No Yes 1 .. 33 Haiti .. .. .. .. No Yes 1 .. 133 Honduras .. .. .. .. No Yes 1 .. 293 Hungary 20.6 23.7 7.6 8.9 Yes Yes 2 20 3,153 India 51.1 53.3 33.8 36.1 Yes Yes 2 32 237,687k Indonesia 13.1 14.8 2.9 2.9 No No 0 .. .. Iran, Islamic Rep. 4.9 .. 8.4 .. No Yes 1 .. 720 Iraq .. .. .. .. No No 0 .. .. Ireland 27.9 30.7 2.5 2.4 Yes Yes 3 8 80 Israel 12.7 15.1 6.9 6.2 No Yes 1 .. 273 Italy 22.8 25.4 3.6 6.5 Yes Yes 3 20 8,104 Japan .. .. 37.8 .. Yes Yes 2 47 3,233 Jordan .. .. .. .. No Yes 1 .. 669 Kazakhstan .. .. .. .. No+ No+ 0 16 303 Kenya 4.4 3.5 2.2 1.9 No Yes 1 .. 168 Korea, Dem. Rep. .. .. .. .. No No 0 .. .. Korea, Rep. .. .. .. .. Yes Yes 2 15 204 Kyrgyz Republic .. .. .. .. No+ Yes 1 7 61 Lao PDR .. .. .. .. No No 0 .. .. Latvia .. 25.8 .. 15.8 No+ Yes 1 33 566 Lebanon .. .. .. .. No No 0 .. .. Libya .. .. .. .. No Yes 1 .. 1,500 Lithuania 30.4 22.6 14.4 16.2 No+ Yes 1 10 56 Madagascar .. .. .. .. No Yes 1l .. 1,391 Malawi .. .. .. .. No No 0m .. .. Malaysia 20.2 19.1 3.7 2.4 No+ No 0 13 143 Mali .. .. .. .. No Yes 1 .. 279 (c) The International Bank for Reconstruction and Development / The World Bank   Fiscal decentralization Electoral decentralization Share of subnational government (%) Subnational electionsa No. of elected No. of jurisdictions In total public expenditure In total tax revenue Intermediateb Localc subnational tiers Intermediateb Localc Economy 1990 1997 1990 1997 1999 1999 1999 1999 1999 Mexico 17.8 26.1 19.0 20.6 Yes Yes 2 32 2,418 Moldova .. .. .. .. No+ Yes 1 3 35 Morocco .. .. .. .. No+ Yes 1 65 1,547 Mozambique .. .. .. .. Yes Yes 2 10 33 Myanmar .. .. .. .. No No 0 .. .. Nepal .. .. .. .. Yes Yes 2 75 4,053 Netherlands 29.0 26.1 3.4 4.1 Yes Yes 2 12 572 New Zealand 9.3 10.8 6.9 6.3 Yes Yes 3 12 155 Nicaragua 3.5 9.6 2.5 8.3 No Yes 1 .. 143 Niger .. .. .. .. No+ No+ 0 32 150 Nigeria .. .. .. .. Yes Yes 2 31 589 Norway 36.7 37.4 20.9 19.6 No Yes 1 .. 435 Pakistan .. .. .. .. No+ No+ 0n 4 5,195 Papua New Guinea .. .. .. .. No Yes 1 .. 284 Paraguay 1.9 2.6 0.8 2.0 Yes Yes 2 17 212 Peru 9.8 24.4 1.2 2.1 No Yes 1 .. 1,808 Philippines 6.5 .. 4.0 .. Yes Yes 2 76 1,541 Poland .. 22.0 21.3 9.6 Yes Yes 3 o 16 2,489 Portugal 8.7 11.6 3.6 5.9 No Yes 2p .. 275 Romania 15.4 13.3 12.8 9.2 No+ Yes 1 41 2,948 Russian Federation .. 37.6 .. 40.0 Yes Yes 3 90 2,000 Rwanda .. .. .. .. No No+ 0 .. 143 Saudi Arabia .. .. .. .. No .. 0 .. .. Senegal .. .. .. .. No+ No+ 0 10 99 Sierra Leone .. .. .. .. No Yes 1 .. 204 Slovak Republic .. .. .. .. No Yes 1 .. 2,834 Slovenia .. .. .. .. No Yes 1 .. 192 South Africa 20.7 49.8 5.5 5.3 Yes Yes 2 9 840 Spain 34.3 35.0 13.3 13.8 Yes Yes 3 17 8,082 Sri Lanka .. .. .. .. No+ Yes 1 9 238 Sudan .. .. .. .. Noq Yes 1 .. 615 Sweden 39.8 36.2 28.2 31.4 Yes Yes 2 24 286 Switzerland 51.2 49.3 37.0 35.5 Yes Yes 2 26 3,000 Syrian Arab Republic .. .. .. .. No Yes 1 .. 300 Tajikistan .. .. .. .. No+ No+ 0r 3 41 Tanzania .. .. .. .. No Yes 1 .. 101 Thailand 7.5 9.6 4.4 5.5 No Yes 1 .. 149s Togo .. .. .. .. No Yest 1 .. 30 Tunisia .. .. .. .. No Yes 1 .. 257 Turkey .. .. .. .. No+ Yes 1 80 2,801 Turkmenistan .. .. .. .. No No 0 .. .. Uganda .. .. .. .. Yes Yes 2 58 1,040 Ukraine .. .. .. .. No+ Yes 1 27 619 United Kingdom 29.0 27.0 5.9 3.6 Yes Yes 1 or 2 135 319 United States 42.0 46.4 33.8 32.9 Yes Yes 3 51 70,500 Uruguay .. .. .. .. No Yes 1 .. 19 Uzbekistan .. .. .. .. No+ u No+ u 0 14 281 Venezuela .. .. .. .. Yes Yes 2 24 330 Vietnam .. .. .. .. No No 0 .. .. Yemen, Rep. .. .. .. .. No .. 0 .. .. Zambia .. .. .. .. No Yes 1 .. 72 Zimbabwe 13.5 .. 3.4 .. No Yes 1 .. 80 a. “No+” indicates that, although the legislature is elected, a nominated executive head (for example, a mayor or governor) holds significant powers. b. State, province, region, department, or other elected entity between the local and the national government. c. Municipality or equivalent. d. Subnational fiscal data come from World Bank country data and staff calculations. e. The 1996 Local Government Commission recommended a four-tier subnational government system composed (from the bottom up) of approximately 85,000 villages; 4,633 unions, and municipalities; 460 thanas and upazilas; and 64 zilas. Parliament has passed the upazila council bill, and elections are scheduled for 1999; the zila council bill had not been passed as of June 1999. Elected local government currently exists only at the municipal level, composed of 4,500 union parishads in rural areas, 129 pourashavas, or smaller municipalities, and 4 city corporations. f. A law passed in 1998 allows for elections at the commune level, but elections have not yet taken place. g. Bosnia and Herzegovina is divided into a federation and the Republika Srpska, with two substate levels within the federation (10 cantons and 73 municipalities), but only one in the Republika Srpska (64 municipalities). h. Local elections are planned for late 1999 or early 2000. A law is being drafted to define the powers and responsibilities of elected commune councils. i. Heads of local government are currently appointed, although this is slated to change in 1999. j. Villages elect representatives, who represent them at the district level and in turn elect a provincial parliament. The provincial governor is appointed by the head of state. Eritrea is in the process of changing its constitution, which could modify this system. k. Local government consists of 3,609 urban local bodies and, in rural areas, 474 zila parishads, which wield some authority over the 5,906 panchayats samithis, which in turn have some authority over the 227,698 gram panchayats. It is therefore not strictly correct to aggregate these into one level of local authority. l. A 1998 revision of the constitution allows for six provinces and an unspecified number of regions, in addition to the existing municipalities. Only the municipalities currently have sitting elected governments. m. Malawi has a local government administration, but no elected local government has been in place for several years. Local elections are expected in October 1999. n. Local elections have been held infrequently, and local governments are established by provincial governments. o. The three tiers are the 16 gminas, 368 powiats, and 2,365 municipalities. p. Portugal also has 4,207 submunicipalities as a second tier of elected local government. q. At the intermediate level the country is divided into 26 states, some of which have elected governors, whereas others have nominated governors. r. The assemblies of the oblasts (provinces) and rains (districts) are elected, but their heads are nominated by the president. At the jamoat, or community, level, the local governing authority is elected at a general meeting of the residents. s. Thailand currently has elected municipal governments governing 149 cities. In addition there are 1,050 sanitary districts, which provide services in densely populated areas outside cities. Each is governed by a board composed of appointed and elected members; 983 of these districts will soon be upgraded to municipality status. There are up to 7,823 tambon administrative organizations, which provide basic services in rural areas and are governed by elected assemblies and appointed executives. The 1997 constitution mandates that executives and councils of local authorities be largely elected. These changes are expected to be completed by October 1999, in which case the country would still have only one tier of elected local government but close to 8,955 fully elected local governments. t. Not all mayors are elected; about 10 are appointed. u. Appointed khokims (governors or mayors) exercise almost unlimited power in oblasts and rayons, with quasi-elected councils having very limited authority. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table A.2. Urbanization Urban population by size of city Population share Access to sanitation % of total urban population of largest city in urban areas Urban population % of urban % of urban 750,000— Millions % of total population <750,000 3 million >3 million population population Economy 1980 1997 1980 1997 1995 1995 1995 1980 1995 1982 1995 Albania 0.9 1.3 34 38 .. .. .. .. .. .. 97 Algeria 8.1 16.8 43 57 76 0 24 25 24 95 .. Angola 1.5 3.8 21 32 39 61 0 63 61 27 71 Argentina 23.3 31.6 83 89 51 11 39 43 39 76 80 Armenia 2.0 2.6 66 69 50 50 0 51 50 .. .. Australia 12.6 15.7 86 85 32 24 44 26 23 .. .. Austria 4.9 5.2 65 64 60 40 0 42 40 .. .. Azerbaijan 3.3 4.3 53 56 56 44 0 48 44 .. 67 Bangladesh 9.8 24.1 11 19 45 16 39 33 39 20 41 Belarus 5.4 7.4 56 72 76 24 0 24 24 .. .. Belgium 9.4 9.9 95 97 89 11 0 13 11 .. .. Benin 0.9 2.3 27 40 .. .. .. .. .. 45 60 Bolivia 2.4 4.8 46 62 53 47 0 30 28 51 77 Bosnia and Herzegovina 1.5 1.0 36 42 .. .. .. .. .. .. 71 Botswana 0.1 1.0 15 65 .. .. .. .. .. 79 91 Brazil 80.5 130.1 66 80 56 14 30 16 13 33 74 Bulgaria 5.4 5.7 61 69 79 21 0 20 21 .. .. Burkina Faso 0.6 1.8 9 17 48 52 0 44 52 38 .. Burundi 0.2 0.5 4 8 .. .. .. .. .. 90 71 Cameroon 2.7 6.5 31 46 59 41 0 19 22 .. 73 Canada 18.6 23.3 76 77 46 20 34 16 19 .. .. Chad 0.8 1.6 19 23 45 55 0 40 55 .. 74 Chile 9.0 12.3 81 84 59 0 41 41 41 79 95 China 192.3 390.7 20 32 60 19 21 6 4 .. 68 Hong Kong, China 4.6 6.2 91 95 1 0 99 100 99 .. .. Colombia 18.2 29.4 64 74 53 14 33 20 22 96 70 Congo, Dem. Rep. 7.8 13.7 29 29 60 6 34 28 34 8 53 Congo, Rep. 0.7 1.6 41 60 33 67 0 67 67 17 15 Costa Rica 1.0 1.7 43 50 45 55 0 61 55 100 100 Côte d’Ivoire 2.9 6.3 35 45 52 48 0 44 48 13 .. Croatia 2.3 2.7 50 57 63 37 0 28 37 72 71 Cuba 6.6 8.5 68 77 73 27 0 29 27 .. 92 Czech Republic 6.5 6.8 64 66 82 18 0 18 18 .. .. Denmark 4.3 4.5 84 85 70 30 0 32 30 .. .. Dominican Republic 2.9 5.1 51 63 8 27 65 50 65 72 89 Ecuador 3.7 7.2 47 60 54 46 0 30 27 79 70 Egypt, Arab Rep. 17.9 27.2 44 45 44 5 51 38 37 95 .. El Salvador 1.9 2.7 42 46 52 48 0 39 48 89 89 Ethiopia 4.0 9.7 11 16 72 28 0 30 28 .. .. Finland 2.9 3.3 60 64 67 33 0 22 33 100 100 France 39.5 44.0 73 75 70 8 22 23 22 .. .. Gabon 0.2 0.6 34 52 .. .. .. .. .. .. 79 Georgia 2.6 3.2 52 59 58 42 0 42 42 .. .. Germany 64.7 71.3 83 87 49 28 23 10 9 .. .. Ghana 3.4 6.6 31 37 73 27 0 30 27 47 61 Greece 5.6 6.3 58 60 34 16 50 54 50 .. .. Guatemala 2.6 4.2 37 40 43 57 0 29 57 73 91 Guinea 0.9 2.1 19 31 19 81 0 65 81 54 24 Guinea-Bissau 0.1 0.3 17 23 .. .. .. .. .. 21 32 Haiti 1.3 2.5 24 33 36 64 0 55 64 42 43 Honduras 1.2 2.7 35 45 60 40 0 33 40 22 91 Hungary 6.1 6.7 57 66 69 31 0 34 31 .. .. India 158.8 264.1 23 27 59 18 23 5 6 25 46 Indonesia 32.9 74.8 22 37 73 14 13 18 13 30 88 Iran, Islamic Rep. 19.4 36.6 50 60 57 23 20 26 20 90 86 Iraq 8.5 16.5 66 75 55 17 28 39 28 30 85 Ireland 1.9 2.1 55 58 56 44 0 48 44 .. .. Israel 3.4 5.3 89 91 61 39 0 41 39 .. 100 Italy 37.6 38.4 67 67 66 15 19 14 11 .. .. Jamaica 1.0 1.4 47 55 .. .. .. .. .. 92 99 Japan 89.0 98.9 76 78 50 8 42 25 28 .. .. Jordan 1.3 3.2 60 73 61 39 0 49 39 91 91 Kazakhstan .. 9.6 54 60 87 13 0 .. 13 .. .. Kenya 2.7 8.7 16 30 77 23 0 32 23 75 69 Korea, Dem. Rep. 10.1 14.2 57 62 82 18 0 18 18 100 100 Korea, Rep. 21.7 38.3 57 83 29 28 43 2 2 100 100 Kuwait 1.2 1.8 90 97 29 71 0 67 71 100 100 Kyrgyz Republic 1.4 1.8 38 39 .. .. .. .. .. 78 87 Lao PDR 0.4 1.1 13 22 .. .. .. .. .. .. 13 Latvia 1.7 1.8 68 73 50 50 0 49 50 .. 90 Lebanon 2.2 3.7 74 88 48 52 0 55 52 94 .. Lesotho 0.2 0.5 13 26 .. .. .. .. .. 22 76 Libya 2.1 4.5 69 86 41 59 0 38 40 100 90 Madagascar 1.6 3.9 18 28 75 25 0 29 25 8 64 Malawi 0.6 1.5 9 14 .. .. .. .. .. 88 82 (c) The International Bank for Reconstruction and Development / The World Bank   Urban population by size of city Population share Access to sanitation % of total urban population of largest city in urban areas Urban population % of urban % of urban 750,000— Millions % of total population <750,000 3 million >3 million population population Economy 1980 1997 1980 1997 1995 1995 1995 1980 1995 1982 1995 Malaysia 5.8 11.9 42 55 89 11 0 16 11 .. 94 Mali 1.2 2.9 19 28 65 35 0 40 35 90 .. Mexico 44.8 69.6 66 74 55 15 30 31 25 77 93 Moldova 1.6 2.3 40 53 66 34 0 .. .. .. 96 Morocco 8.0 14.5 41 53 68 9 23 26 23 85 97 Mozambique 1.6 6.0 13 36 59 41 0 47 41 51 68 Myanmar 8.1 11.7 24 27 65 0 35 27 35 34 44 Namibia 0.2 0.6 23 38 .. .. .. .. .. .. 78 Nepal 0.9 2.4 7 11 .. .. .. .. .. 5 34 Netherlands 12.5 13.9 88 89 84 16 0 8 8 .. 100 New Zealand 2.6 3.2 83 86 69 31 0 30 31 .. .. Nicaragua 1.6 3.0 53 63 59 41 0 41 41 35 88 Nigeria 19.1 48.7 27 41 73 3 23 23 23 30 82 Norway 2.9 3.2 71 74 .. .. .. .. .. .. 100 Oman 0.3 1.8 32 79 .. .. .. .. .. 60 98 Pakistan 23.2 45.4 28 35 42 23 35 22 23 48 60 Panama 1.0 1.5 50 56 34 66 0 62 66 99 99 Papua New Guinea 0.4 0.7 13 17 .. .. .. .. .. 51 95 Paraguay 1.3 2.7 42 54 57 43 0 52 43 66 20 Peru 11.2 17.5 65 72 60 0 40 39 40 67 78 Philippines 18.1 41.1 37 56 73 3 24 33 24 .. .. Poland 20.7 24.9 58 64 66 20 14 16 14 .. .. Portugal 2.9 3.6 29 37 47 53 0 46 53 .. .. Puerto Rico 2.1 2.8 67 74 52 48 0 51 48 .. .. Romania 10.9 12.8 49 57 83 17 0 18 17 .. 81 Russian Federation 97.0 112.9 70 77 73 14 13 8 8 .. .. Saudi Arabia 6.2 16.8 66 84 69 31 0 16 17 100 .. Senegal 2.0 4.0 36 45 53 47 0 47 47 87 .. Singapore 2.3 3.1 100 100 0 0 100 100 100 85 100 Slovenia 0.9 1.0 48 52 .. .. .. .. .. 90 100 South Africa 13.3 20.2 48 50 36 64 0 12 11 .. 78 Spain 27.2 30.2 73 77 75 12 14 16 14 .. .. Sri Lanka 3.2 4.2 22 23 .. .. .. .. .. .. 81 Sudan 3.7 9.2 20 33 73 27 0 31 27 70 79 Sweden 6.9 7.4 83 83 69 31 0 20 21 .. .. Switzerland 3.6 4.4 57 62 79 21 0 20 21 .. .. Syrian Arab Republic 4.1 7.9 47 53 47 53 0 34 28 58 97 Tajikistan 1.4 2.0 34 32 .. .. .. .. .. .. 83 Tanzania 2.7 8.0 15 26 62 38 0 30 24 93 97 Thailand 7.9 12.5 17 21 45 0 55 59 55 50 98 Togo 0.6 1.4 23 32 .. .. .. .. .. 34 76 Trinidad and Tobago 0.7 0.9 63 73 .. .. .. .. .. 100 97 Tunisia 3.3 5.8 52 63 69 31 0 35 31 64 100 Turkey 19.5 45.7 44 72 63 18 19 23 19 .. .. Turkmenistan 1.3 2.1 47 45 .. .. .. .. .. .. 70 Uganda 1.1 2.7 9 13 60 40 0 42 40 40 60 Ukraine 30.9 36.1 62 71 73 27 0 7 8 .. 70 United Arab Emirates 0.7 2.2 71 85 59 41 0 31 41 93 .. United Kingdom 50.0 52.7 89 89 71 15 15 15 15 .. .. United States 167.6 204.8 74 77 44 27 29 9 8 .. .. Uruguay 2.5 3.0 85 91 54 46 0 49 46 59 56 Uzbekistan 6.5 9.9 41 42 76 24 0 28 24 .. 46 Venezuela 12.0 19.7 79 86 58 26 16 21 16 57 74 Vietnam 10.3 15.0 19 20 67 9 25 27 25 .. .. Yemen, Rep. 1.7 5.7 20 35 .. .. .. .. .. .. 40 Yugoslavia, FR (Serb./Mont.) 4.5 6.1 46 58 80 20 0 24 20 .. .. Zambia 2.3 4.1 40 44 66 34 0 23 34 56 66 Zimbabwe 1.6 3.8 22 33 60 40 0 39 40 100 99 World 1,748.2 s 2,676.0 s 39 w 46 w 59 w 19 w 22 w 18 w 17 w .. .. Low income 307.7 577.7 22 28 59 21 20 16 19 29 56 Middle income 824.3 1,389.9 37 49 62 18 19 19 16 .. 77 Lower middle income 559.0 966.2 31 42 64 18 18 16 14 .. 75 Upper middle income 265.4 423.7 62 74 58 20 22 24 20 .. .. Low and middle income 1,132.1 1,967.7 31 40 61 19 20 18 17 .. .. East Asia & Pacific 288.4 578.0 21 33 64 16 20 13 9 .. 74 Europe & Central Asia 240.1 317.7 56 67 71 20 9 15 15 .. .. Latin America & Carib. 233.8 366.5 65 74 55 17 28 27 25 60 80 Middle East & N. Africa 83.7 161.9 48 58 58 20 22 31 27 81 .. South Asia 198.5 345.5 22 27 56 19 25 9 11 27 48 Sub-Saharan Africa 87.6 198.0 23 32 62 30 9 28 30 .. .. High income 616.1 708.4 75 76 53 20 27 17 16 .. .. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table A.3. Urban living conditions Income differential Work trips Households with Average Ratio of House Crowding by public Travel Sewerage Regular Access to Urban Urban household top to price– m2 of floor transpor- time to connec- waste potable area population income bottom income space per tation work tion collection water km2 Thousands Dollars quintile ratio person % Minutes % % % Economy City 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 Armenia Yerevan 215 1,223 1,407 28.4 39.0 13.0 98 52 93 81 98 Australia Melbourne 1,148 3,023 30,216 12.0 3.6 55.0 16 25 99 100 100 Azerbaijan Baku 2,300 .. 977 8.7 13.0 12.9 80 57 79 .. 100 Bangladesh Dhaka 1,194 7,500 478 6.9 5.0 2.7 .. .. 44 50 .. Tangail 32 155 228 6.9 8.0 1.2 .. 15 .. .. 51 Benin Cotonou 88 559 2,745 6.0 1.6 5.9 .. 60 1 25 60 Porto Novo 50 183 1,479 6.1 3.4 5.5 .. 40 1 25 76 Bolivia Santa Cruz de la Sierra 165 742 3,786 7.6 2.6 .. 60 25 22 100 87 La Paz 51 726 3,787 11.7 1.2 .. 51 35 58 92 90 El Alto 58 442 1,786 7.2 1.4 .. .. 25 20 95 86 Cochabamba 68 425 4,035 8.3 2.6 .. 46 17 47 95 71 Botswana Gaborone .. 473 .. .. 7.2 12.5 42 20 33 98 100 Brazil Rio de Janeiro 1,255 5,554 12,087 20.3 2.5 18.9 67 51 87 88 98 Recife .. 1,503 815 28.7 2.2 15.5 70 40 38 95 95 Curitiba .. 1,352 1,091 16.1 5.7 21.0 72 30 75 95 97 Brasilia .. .. 12,087 20.3 3.0 17.3 .. 49 74 95 89 Bulgaria Sofia .. 1,294 .. .. 5.8 16.7 75 35 98 95 100 Burkina Faso Ouagadougou 170 716 2,622 3.3 8.5 12.2 .. 22 .. 40 75 Bobo-Dioulasso 67 284 2,379 9.1 10.2 12.0 .. 15 .. 30 81 Burundi Bujumbura 100 278 1,823 17.0 1.9 5.8 .. 30 29 41 93 Cameroon Douala 144 1,094 .. .. 4.6 10.0 11 45 3 60 83 Yaounde .. 923 677 .. 3.9 12.6 6 50 3 44 85 Canada Toronto .. 4,236 49,791 9.5 3.9 41.1 30 23 100 100 100 Central African Republic Bangui 163 471 .. .. 6.2 11.2 .. 45 1 25 45 Chile Santiago .. 4,820 8,043 16.6 2.4 14.4 54 36 92 95 98 China Hefei .. 3,809 2,080 13.8 .. 11.0 0 .. 57 .. 100 Qingdao .. 2,121 1,165 1.8 .. 11.1 .. 11 .. 100 Foshan 32 385 3,354 3.2 .. 16.3 0 .. 100 .. 100 Colombia Bogotá 482 5,314 7,120 14.7 3.1 8.8 75 39 99 94 97 Congo, Dem. Rep. Kinshasa 591 4,566 2,241 6.7 .. .. 61 120 3 0 70 Côte d’Ivoire Abidjan 369 2,462 2,827 7.9 7.2 7.2 49 90 45 70 62 Bouake .. 439 1,820 9.5 5.6 7.4 10 35 .. 35 28 Croatia Zagreb .. 868 4,354 5.9 11.0 22.1 52 26 80 100 90 Cuba Havana .. 2,176 .. .. 2.1 16.0 58 42 85 100 85 Camaguey 155 296 .. .. .. 18.7 6 30 46 93 71 Cienfuegos 44 131 .. .. 1.5 19.2 0 30 70 97 100 Pinar del Rio 28 129 .. .. 3.7 21.0 0 80 48 100 93 Czech Republic Prague 496 1,214 .. .. 11.9 26.0 67 57 94 100 100 Denmark Copenhagen 2,863 .. 29,320 14.0 3.1 44.0 27 22 100 100 100 Djibouti Djibouti .. .. 6,856 12.0 3.7 13.1 19 22 15 65 69 Ecuador Guayaquil 178 1,773 5,406 12.1 2.0 15.6 50 45 55 70 85 Quito 178 1,615 .. .. 2.4 8.6 0 .. 93 89 .. Egypt, Arab Rep. Cairo 420 14,524 1,658 6.1 4.9 13.0 58 60 91 65 98 Gharbeya .. 383 1,656 6.1 3.9 13.3 32 30 91 45 99 Assiout 10 322 1,721 6.7 3.1 14.0 29 25 30 25 93 El Salvador San Salvador 163 1,343 4,320 12.7 2.7 6.6 0 .. 80 46 91 Santa Ana 18 142 2,998 10.6 3.2 8.1 0 .. 57 90 82 San Miguel .. 132 3,420 13.2 4.3 9.7 0 .. 46 99 56 Estonia Tallinn 185 468 .. .. 3.6 21.3 0 27 95 99 100 France Paris 2,586 9,319 20,899 14.7 4.3 30.0 40 35 98 100 100 Marseille 351 800 14,640 5.2 .. .. 0 25 99 99 100 Strasbourg 78 388 15,942 9.7 .. .. 0 15 98 100 100 Gambia, The Banjul .. 479 230 8.1 4.8 11.5 60 40 13 35 74 Georgia Tbilisi 204 1,295 .. .. .. 16.2 98 70 100 52 100 Germany Cologne 405 1,006 .. .. .. 34.0 17 .. 99 100 100 Duisburg 233 536 .. .. 7.9 32.1 21 .. 100 100 100 Leipzig 151 481 .. .. .. 33.0 33 .. 95 100 100 Wiesbaden 204 266 .. .. .. 37.0 23 .. 100 100 100 Erfurt 268 213 .. .. 5.1 29.1 32 .. 95 100 100 Ghana Accra 411 1,718 403 .. 8.0 6.2 47 45 12 60 86 Kumasi .. 758 822 2.9 17.8 5.8 55 20 12 11 57 Tamale 22 193 682 1.9 17.4 5.2 45 18 6 5 38 Greece Athens .. 1,464 .. .. 3.1 29.0 34 53 95 90 100 Guatemala Guatemala City .. 1,327 2,760 76.7 9.0 8.0 53 40 .. 53 64 Guinea Conakry .. 1,308 .. .. 6.4 6.5 26 55 17 50 75 Hungary Budapest .. 320 5,621 9.2 7.7 29.4 66 40 90 100 100 India Mumbai . . 12,810 1,504 6.7 3.5 3.5 79 33 51 90 96 Delhi 624 8,957 1,196 11.4 7.0 6.9 53 44 40 77 92 Chennai 612 5,651 1,184 8.0 7.0 6.2 42 22 37 90 60 Bangalore .. 4,472 1,224 6.5 10.8 9.5 46 18 35 96 81 Lucknow .. 1,804 992 7.5 4.6 5.5 1 23 30 74 88 Varanasi 104 1,078 928 7.8 5.1 4.5 21 22 41 88 85 Mysore .. 701 1,236 6.4 7.5 11.8 13 20 60 60 90 Bhiwandi 26 572 .. .. 0.3 2.4 8 15 15 40 86 Gulbarga .. 330 1,028 7.6 3.5 6.1 8 11 14 74 90 Tumkur .. 194 809 6.1 4.9 7.4 21 8 .. 50 86 Hubli-Dharbad .. .. 1,114 7.1 3.6 6.2 37 22 37 89 89 (c) The International Bank for Reconstruction and Development / The World Bank   Income differential Work trips Households with Average Ratio of House Crowding by public Travel Sewerage Regular Access to Urban Urban household top to price– m2 of floor transpor- time to connec- waste potable area population income bottom income space per tation work tion collection water km2 Thousands Dollars quintile ratio person % Minutes % % % Economy City 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 Indonesia Jakarta . . 13,048 2,460 6.6 9.9 15.0 38 82 .. 84 93 Bandung .. 1,819 1,625 5.8 12.0 13.1 0 29 27 97 86 Medan .. 1,810 1,674 4.5 5.5 13.9 44 30 19 19 94 Semarang .. 1,076 1,351 6.0 5.4 12.0 14 25 .. 69 88 Banjarmasin .. .. 1,474 4.4 4.0 6.4 12 37 .. 70 94 Surabaya .. .. 1,970 8.1 8.6 11.5 23 23 .. 87 99 Jordan Amman .. .. 12,813 13.9 6.5 15.4 14 31 79 100 100 Kazakhstan Almaty .. 1,173 .. .. 7.2 14.5 43 35 88 83 100 Kenya Mombasa 234 382 .. .. 1.9 5.9 31 27 2 40 95 Nairobi 64 333 .. .. 1.8 15.6 0 64 .. 63 45 Latvia Riga .. 1,026 .. .. .. 19.4 57 27 97 85 100 Liberia Monrovia .. 697 .. .. 24.0 14.0 75 60 1 0 20 Lithuania Vilnius .. 670 .. .. 5.4 16.2 49 25 94 95 100 Malawi Blantyre .. 403 .. .. 8.3 8.3 39 44 8 20 80 Lilongwe .. 220 .. .. 4.2 6.6 5 31 12 .. 80 Mali Bamako 267 .. .. .. 3.7 3.2 12 40 2 95 53 Mauritania Nouakchott 72 576 1,481 8.9 6.4 10.0 45 50 4 15 68 Moldova Chisinau 131 662 1,055 9.7 13.0 15.0 48 25 86 83 100 Mongolia Ulaanbaatar 3,542 .. 317 3.2 37.7 9.2 85 29 51 .. 49 Morocco Rabat .. 1,345 7,514 8.1 6.8 10.0 0 .. 95 90 100 Mozambique Maputo .. .. 414 4.9 .. 12.0 13 .. 23 37 73 Namibia Windhoek 69 142 11,618 15.2 6.0 43.0 0 20 75 93 98 Netherlands Amsterdam 202 724 21,687 5.2 3.5 38.3 0 22 100 100 100 New Zealand Auckland .. 942 25,900 8.1 4.4 40.0 6 .. 98 .. .. Niger Niamey 224 505 1,369 13.2 7.3 7.7 17 27 .. 25 77 Nigeria Lagos 959 5,968 492 18.2 10.0 5.5 54 85 2 8 75 Ibadan 2,937 1,941 415 50.0 6.8 9.0 40 40 .. 40 70 Kano 123 1,510 340 6.9 3.2 2.8 56 .. 25 38 16 Onitsha 9 .. 623 18.5 .. 12.0 53 33 .. 38 95 Pakistan Lahore .. 5,150 3,298 7.7 16.0 1.2 16 25 74 50 90 Paraguay Asunción 67 949 5,496 8.8 5.3 4.7 31 60 10 79 58 Peru Lima .. 6,232 1,109 .. 9.2 25.7 65 35 69 57 87 Trujillo 45 509 .. .. 3.8 15.2 74 30 71 48 98 Philippines Manila .. .. 5,318 8.4 .. 34.1 40 120 80 85 94 Poland Warsaw .. .. 3,021 3.1 5.4 18.2 0 34 91 97 100 Romania Bucharest .. 2,350 .. .. 6.8 12.9 65 78 90 86 98 Russian Federation Kostroma .. .. 2,357 5.1 5.1 17.8 65 21 91 90 100 Moscow .. .. 4,040 7.6 17.0 19.7 85 62 100 100 100 Nizhny Novgorod .. .. 2,459 4.6 6.4 17.1 78 35 95 100 100 Novgorod .. .. 2,865 5.9 7.3 16.3 44 30 96 99 100 Ryazan .. .. 2,348 6.9 8.9 16.2 88 25 92 99 100 Senegal Dakar .. 1,801 3,008 17.0 3.0 8.1 53 45 25 75 92 Kaolack .. 187 1,488 20.9 .. .. 13 27 3 .. 56 Ziguinchor .. 155 1,150 22.0 .. .. 27 20 2 .. 30 Mbour .. 101 2,192 15.9 .. .. 20 31 2 .. 79 Sierra Leone Freetown 82 395 370 11.4 .. 10.0 0 .. 1 .. 53 Slovak Republic Bratislava 2,144 651 3,984 5.1 5.6 22.3 72 34 96 100 100 Slovenia Ljubljana 275 316 11,729 6.1 .. .. 1 22 99 99 100 Maribor 738 185 9,314 6.2 .. .. 41 28 58 90 100 Sri Lanka Colombo .. 2,190 436 3.4 .. 18.7 74 35 60 94 98 Sudan Khartoum 249 826 .. .. .. 21.9 63 42 3 12 55 Sweden Stockholm 309 .. 30,840 4.5 4.6 40.0 37 35 100 100 100 Tanzania Arusha .. .. 564 4.1 5.0 5.0 61 30 16 .. 60 Dar es Salaam .. .. 564 4.1 5.0 4.5 48 30 6 25 60 Mwanza 94 .. .. .. 5.0 4.0 24 30 8 15 74 Togo Lome 288 802 .. .. 3.5 12.0 30 30 .. 37 .. Tunisia Tunis .. 1,684 4,032 6.0 5.2 12.0 0 45 73 61 96 Uganda Kampala 202 840 .. .. 2.3 4.0 45 23 9 20 87 United Arab Emirates Dubai 604 594 26,564 22.8 .. .. 0 18 60 100 100 United Kingdom Hertfordshire 1,604 1,000 28,270 10.9 6.0 34.8 7 27 100 100 100 Glasgow .. 618 7,329 1.8 4.5 .. 39 .. 99 .. 99 Bedfordshire .. 539 32,080 10.9 3.0 34.6 10 .. 93 98 98 Cardiff 137 306 .. .. 2.9 17.5 13 .. 100 100 100 United States New York .. .. 39,256 14.8 6.3 .. 51 37 99 .. 100 Vietnam Hanoi 47 .. 32,966 3.4 10.4 5.8 0 .. 40 45 100 Yemen, Rep. Sana’a .. .. 183 .. 17.0 4.0 0 15 12 51 60 Yugoslavia FR Belgrade 765 1,318 .. .. 16.0 19.4 0 35 71 86 99 (Serb./Mont.) Novi Sad 290 232 .. .. 30.0 21.8 60 21 93 95 100 Nis 150 214 .. .. 17.4 19.7 61 25 84 87 92 Zambia Lusaka .. .. 867 14.0 6.5 6.9 65 20 36 .. 60 Zimbabwe Harare .. .. 754 5.0 9.8 8.0 48 56 93 100 97 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table A.4. Environment Climate change Biodiversity Carbon dioxide Electricity Mammal and bird Fisheries Fossil fuel emissions production production species Higher plant species Annual marine Million % of Billion % from Thousand Number Number catch metric tons world kWh fossil fuel metric tons Number threatened Number threatened Thousand metric tons Economy 1996 1996 1996 1996 1996 1997 1997 1997 1997 1970 1996 Algeria 94.3 0.4 21 99 115,700 284 23 3,164 141 25 103 Argentina 129.9 0.6 70 56 68,249 1,217 68 9,372 247 163 925 Australia 306.6 1.4 177 90 182,819 901 103 15,638 2,245 54 128 Bolivia 10.1 0.0 3 36 4,290 1,590 51 17,367 227 0 1 Brazil 273.4 1.2 290 5 49,577 1,886 174 56,215 1,358 407 545 Canada 409.4 1.8 571 21 293,525 619 12 3,270 278 1,127 443 Chile 48.8 0.2 31 45 2,684 387 34 5,284 329 1,101 7,270 China 3,363.5 14.9 1,080 81 874,408 1,494 165 32,200 312 2,076 10,087 Colombia 65.3 0.3 45 20 56,817 2,054 99 51,220 712 16 103 Costa Rica 4.7 0.0 5 14 .. 805 27 12,119 527 5 16 Cuba 31.2 0.1 13 92 1,221 168 22 6,522 888 86 57 Denmark 56.6 0.3 54 95 16,005 239 5 1,450 2 1,184 1,578 Ecuador 24.5 0.1 9 32 20,100 1,690 81 19,362 824 81 484 Egypt, Arab Rep. 97.9 0.4 58 81 57,621 251 26 2,076 82 25 105 Germany 861.2 3.8 551 65 93,004 315 13 2,682 14 851 198 Greece 80.6 0.4 42 90 7,765 346 23 4,992 571 84 138 Guatemala 6.8 0.0 4 21 740 708 12 8,681 355 1 1 India 997.4 4.4 435 82 193,816 1,239 148 16,000 1,236 941 2,420 Indonesia 245.1 1.1 67 83 172,364 1,955 232 29,375 264 732 2,868 Iran, Islamic Rep. 266.7 1.2 91 92 219,538 463 34 8,000 2 16 237 Italy 403.2 1.8 239 80 22,129 324 17 5,599 311 295 261 Jamaica 10.1 0.0 6 93 .. 137 11 3,308 744 9 9 Japan 1,167.7 5.2 1,003 59 6,327 382 62 5,565 707 7,229 4,587 Kazakhstan 173.8 0.8 59 88 61,923 .. 30 .. 71 .. 0 Kenya 6.8 0.0 4 9 .. 1,203 67 6,506 240 8 4 Korea, Dem. Rep. 254.3 1.1 35 36 18,107 .. 26 2,898 4 445 1,599 Korea, Rep. 408.1 1.8 223 66 2,228 161 25 2,898 66 521 1,729 Kuwait .. .. 25 100 112,600 41 4 234 0 3 6 Libya 40.6 0.2 18 100 77,617 167 13 1,825 57 6 34 Madagascar 1.2 0.0 .. .. .. 307 74 9,505 306 11 71 Malaysia 119.1 0.5 51 90 66,757 787 76 15,500 490 243 921 Mauritius 1.7 0.0 .. .. .. 31 14 750 294 5 17 Mexico 348.1 1.5 163 72 195,899 1,219 100 26,071 1,593 212 981 Netherlands 155.2 0.7 85 92 71,543 246 9 1,221 1 200 380 New Zealand 29.8 0.1 36 21 8,965 160 47 2,382 211 40 453 Nigeria 83.3 0.4 15 63 105,266 955 35 4,715 37 78 212 Norway 67.0 0.3 104 0 198,023 297 7 1,715 12 2,896 2,475 Panama 6.7 0.0 4 37 .. 950 27 9,915 1,302 46 162 Peru 26.2 0.1 17 22 6,972 1,882 110 18,245 906 12,468 9,441 Philippines 63.2 0.3 37 63 523 548 135 8,931 360 784 1,561 Poland 356.8 1.6 141 98 97,962 311 16 2,450 27 447 388 Portugal 47.9 0.2 34 54 60 270 20 5,050 269 453 237 Puerto Rico 15.8 0.1 .. .. .. 121 14 2,493 223 2 2 Russian Federation 1,579.5 7.0 846 68 889,367 897 69 .. 214 .. 3,787 Saudi Arabia 267.8 1.2 98 100 474,997 232 20 2,028 7 17 39 South Africa 292.7 1.3 198 93 113,023 843 49 23,420 2,215 1,205 560 Spain 232.5 1.0 173 43 10,981 360 29 5,050 985 1,235 967 Sri Lanka 7.1 0.0 5 28 .. 338 25 3,314 455 86 204 Tanzania 2.4 0.0 2 12 3 1,138 63 10,008 436 20 39 Thailand 205.4 0.9 87 91 21,951 881 79 11,625 385 946 2,462 Turkey 178.3 0.8 95 57 16,018 418 29 8,650 1,876 168 578 Ukraine 397.3 1.8 182 51 57,293 .. 25 .. 52 .. 381 United Arab Emirates 81.8 0.4 20 100 148,818 92 7 .. 0 40 105 United Kingdom 557.0 2.5 346 70 242,852 280 6 1,623 18 1,028 781 United States 5,301.0 23.4 3,652 69 1,386,112 1,078 85 19,473 4,669 1,575 3,580 Venezuela 144.5 0.6 75 29 188,822 1,486 46 21,073 426 98 367 Vietnam 37.6 0.2 17 100 17,470 748 85 10,500 341 407 412 World 22,653.9 t 100.0 w 13,621 t 62 w .. .. .. .. .. 46,462 t 75,144 t Low income 1,448.1 6.4 672 72 .. .. .. .. .. .. .. Middle income 10,068.9 44.4 4,447 69 .. .. .. .. .. 22,657 48,358 Lower middle income 7,512.7 33.2 3,041 72 .. .. .. .. .. 18,360 35,282 Upper middle income 2,556.2 11.3 1,407 61 .. .. .. .. .. 4,297 13,075 Low and middle income 11,517.0 50.8 5,119 69 .. .. .. .. .. 25,531 53,749 East Asia & Pacific 4,309.5 19.0 1,379 81 .. .. .. .. .. 6,003 20,646 Europe & Central Asia 3,412.7 15.1 1,780 68 1,308,476 .. .. .. .. .. .. Latin America & Carib. 1,209.1 5.3 810 32 .. .. .. .. .. 14,752 20,592 Middle East & N. Africa 988.6 4.4 380 93 1,089,769 .. .. .. .. 567 1,567 South Asia 1,125.1 5.0 509 79 .. .. .. .. .. .. .. Sub-Saharan Africa 472.1 2.1 261 79 .. .. .. .. .. .. 2,266 High income 11,136.9 49.2 8,503 58 .. .. .. .. .. 20,931 21,396 (c) The International Bank for Reconstruction and Development / The World Bank Selected World Development Indicators (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank Contents Introduction to Selected World Development Indicators 226 Tables World View Table 1 Size of the economy 230 Table 2 Quality of life 232 People Table 3 Population and labor force 234 Table 4 Poverty 236 Table 5 Distribution of income or consumption 238 Table 6 Education 240 Table 7 Health 242 Environment Table 8 Land use and agricultural productivity 244 Table 9 Water use, deforestation, and protected areas 246 Table 10 Energy use and emissions 248 Economy Table 11 Growth of the economy 250 Table 12 Structure of output 252 Table 13 Structure of demand 254 Table 14 Central government finances 256 Table 15 Balance of payments, current account, and international reserves 258 States and Markets Table 16 Private sector finance 260 Table 17 Role of government in the economy 262 Table 18 Power and transportation 264 Table 19 Communications, information, and science and technology 266 Global Links Table 20 Global trade 268 Table 21 Aid and financial flows 270 Table 1a Key indicators for other economies 272 Technical Notes 273 Data Sources 289 Classification of Economies by Income and Region, 1999 290  (c) The International Bank for Reconstruction and Development / The World Bank Introduction to Selected World Development Indicators S elected World Development Indicators velopment Indicators is complemented by provides a core set of standard indica- a separately published CD-ROM data- tors drawn from the World Bank’s de- base that gives access to over 1,000 data velopment databases. The layout of the tables and 500 time-series indicators for 21 tables retains the tradition of pre- 227 countries and regions. senting comparative socioeconomic Organization of Selected World data for more than 130 economies for Development Indicators the most recent year for which data are available and for an earlier year. An Tables 1 and 2, World View, offer an additional table presents basic indi- overview of key development issues: cators for 78 economies with sparse How rich or poor are the people in each data or with populations of less than economy? What is their real level of 1.5 million. welfare as reflected in child malnutri- The indicators presented here are a tion and mortality rates? What is the selection from more than 500 included life expectancy of newborns? What per- in World Development Indicators 1999. centage of adults is illiterate? Published annually, World Development Tables 3 to 7, People, show the rate Indicators reflects a comprehensive view of progress in social development dur- of the development process. Its opening ing the past decade. Data on population chapter reports on the record of and the growth, labor force participation, and prospects for social and economic prog- income distribution are included. Mea- ress in developing countries, measured sures of well-being such as expenditure against six international goals. Its five on health care, school enrollment ratios, main sections recognize the contribution and gender differences in access to edu- of a wide range of factors: human capi- cational attainment are also provided. tal development, environmental sustain- Tables 8 to 10, Environment, bring ability, macroeconomic performance, together key indicators on land use and private sector development, and the glo- agricultural output, water resources, en- bal links that influence the external en- ergy consumption, and carbon dioxide vironment for development. World De- emissions.  (c) The International Bank for Reconstruction and Development / The World Bank      Tables 11 to 15, Economy, present information on the 1999 several large countries changed classification, re- structure and growth of the world’s economies, includ- sulting in significant changes in the income and regional ing government finance statistics and a summary of the aggregates. For example, the Republic of Korea, previ- balance of payments. ously classified as a high-income economy, now falls in Tables 16 to 19, States and Markets, look at the roles the upper-middle-income group; therefore data for of the public and the private sector in creating the neces- Korea are also included in the aggregates for developing sary infrastructure for economic growth. These tables pre- countries in East Asia and Pacific. Revisions to estimates sent information on private investment, stock markets, of China’s GNP per capita have caused that economy to and the economic activities of the state (including mili- be reclassified as low-income. The following changes are tary expenditure), as well as a full table of indicators on also reflected: South Africa moved from upper-middle- information technology and research and development. to lower-middle-income; Indonesia and the Solomon Tables 20 and 21, Global Links, contain information Islands from lower-middle- to low-income; Grenada on trade and financial flows, including aid and lending and Panama from lower-middle- to upper-middle- to developing countries. income; and Albania and Bosnia and Herzegovina from Because the World Bank’s primary business is provid- low-income to lower-middle-income. ing lending and policy advice to its low- and middle- Data sources and methodology income members, the issues covered in these tables focus mainly on these economies. Where available, in- Socioeconomic and environmental data presented here formation on the high-income economies is also pro- are drawn from several sources: primary data collection vided for comparison. Readers may wish to refer to na- by the World Bank, member country statistical publi- tional statistical publications and publications of the cations, research institutes such as the World Resources Organisation for Economic Co-operation and Develop- Institute, and international organizations such as the ment and the European Union for more information United Nations and its specialized agencies, the Inter- on the high-income economies. national Monetary Fund, and the Organisation for Economic Co-operation and Development (see the Classification of economies Data Sources following the Technical Notes for a com- As in the rest of the report, the main criterion used in plete listing). Although international standards of cov- the Selected World Development Indicators to classify erage, definition, and classification apply to most sta- economies and broadly distinguish stages of economic tistics reported by countries and international agencies, development is GNP per capita. Economies are classi- there are inevitably differences in coverage, currentness, fied into three categories according to income. The and the capabilities and resources devoted to basic data classification used in this edition has been updated to collection and compilation. For some topics, compet- reflect the World Bank’s current operational guide- ing sources of data require review by World Bank staff lines. The GNP per capita cutoff levels are as follows: to ensure that the most reliable data available are pre- low-income, $760 or less in 1998; middle-income, sented. In some instances, where available data are $761 to $9,360; and high-income, $9,361 and above. deemed too weak to provide reliable measures of levels A further division at GNP per capita $3,030 is made and trends or do not adequately adhere to international between lower-middle-income and upper-middle- standards, the data are not shown. income economies. Economies are further classified by The data presented are generally consistent with region. See the table on Classification of Economies at those in World Development Indicators 1999. However, the end of this volume for a list of economies in each data have been revised and updated wherever new infor- group (including those with populations of less than mation has become available. Differences may also re- 1.5 million). flect revisions to historical series and changes in method- From time to time an economy’s classification is re- ology. Thus data of different vintages may be published vised because of changes in the above cutoff values or in in different editions of World Bank publications. Read- the economy’s measured level of GNP per capita. When ers are advised not to compile data series from different such changes occur, aggregates based on those classifi- publications or different editions of the same publica- cations are recalculated for the past period so that a con- tion. Consistent time-series data are available on World sistent time series is maintained. Between 1998 and Development Indicators 1999 CD-ROM. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  All dollar figures are in current U.S. dollars unless Data for Jordan refer to the East Bank only, unless otherwise stated. The various methods used to convert otherwise noted. from national currency figures are described in the In 1991 the Union of Soviet Socialist Republics was Technical Notes. formally dissolved into 15 countries: Armenia, Azerbai- jan, Belarus, Estonia, Georgia, Kazakhstan, the Kyrgyz Summary measures Republic, Latvia, Lithuania, Moldova, the Russian Fed- The summary measures at the bottom of each table are eration, Tajikistan, Turkmenistan, Ukraine, and Uzbek- either totals (indicated by t if the aggregates include istan. Whenever possible, data are shown for the indi- estimates for missing data and nonreporting countries, vidual countries. or by an s for simple sums of the data available), Data for the Republic of Yemen refer to that coun- weighted averages (w), or median values (m) calculated try as constituted from 1990 onward; data for previous for groups of economies. Data for the countries ex- years refer to the former People’s Democratic Republic cluded from the main tables (those presented in table of Yemen and the former Yemen Arab Republic, unless 1a) have been included in the summary measures, otherwise noted. where data are available, or by assuming that they fol- Whenever possible, data are shown for the individ- low the trend of reporting countries. This gives a more ual countries formed from the former Yugoslavia: consistent aggregated measure by standardizing coun- Bosnia and Herzegovina, Croatia, the Former Yugoslav try coverage for each period shown. Where missing in- Republic of Macedonia, Slovenia, and the Federal Re- formation accounts for a third or more of the overall public of Yugoslavia (Serbia and Montenegro). estimate, however, the group measure is reported as not Technical notes available. The section on “Statistical methods” in the Technical Notes provides further information on aggre- Because data quality and intercountry comparisons are gation methods. Weights used to construct the aggre- often problematic, readers are encouraged to consult gates are listed in the technical notes for each table. the Technical Notes, the table on Classification of Econ- omies by Income and Region, and the footnotes to the Terminology and country coverage tables. For more extensive documentation see World The term country does not imply political indepen- Development Indicators 1999. The Data Sources section dence but may refer to any territory for which authori- at the end of the Technical Notes lists sources that con- ties report separate social or economic statistics. Data tain more comprehensive definitions and descriptions are shown for economies as they were constituted in of the concepts used. 1998, and historical data are revised to reflect current For more information about the Selected World De- political arrangements. Throughout the tables, excep- velopment Indicators and the World Bank’s other sta- tions are noted. tistical publications, please contact: As of July 1, 1997, China resumed its exercise of sovereignty over the Special Administrative Region of Information Center, Development Data Group Hong Kong. Data for Hong Kong, China, are shown The World Bank on a separate line following the entry for China and are 1818 H Street, N.W. included in the aggregates for high-income economies. Washington, D.C. 20433 Data for China do not include data for Taiwan, China, Hotline: (800) 590-1906 or (202) 473-7824 unless otherwise noted. Fax: (202) 522-1498 Data are shown separately whenever possible for the E-mail: info@worldbank.org countries formed from the former Czechoslovakia: the World Wide Web: http://www.worldbank.org/wdi Czech Republic and the Slovak Republic. Data are shown separately for Eritrea whenever pos- To order World Bank publications, e-mail your re- sible; in most cases prior to 1992, however, they are in- quest to books@worldbank.org, or write to World cluded in the data for Ethiopia. Bank Publications at the address above, or call (202) Data for Germany refer to the unified Germany, un- 473-1155. less otherwise noted. (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 1. Size of the economy Surface Population Gross national product (GNP) GNP per capita GNP measured at PPPa area density Population Thousands People per Billions Avg. annual Avg. annual Billions Per capita Millions of sq. km sq. km of dollars Rank growth rate (%) Dollars Rank growth rate (%) of dollars Dollars Rank Economy 1998 1996 1998 1998b 1998 1997–98 1998b 1998 1997–98 1998 1998 1998 Albania 3 29 123 2.7 137 .. 810 144 .. .. .. .. Algeria 30 2,382 13 46.5 50 7.3 1,550 116 5.0 131.4c 4,380c 104 Angola 12 1,247 10 4.1 121 7.9 340 178 4.8 10.1c 840c 197 Argentina 36 2,780 13 324.1 17 4.0 8,970 55 2.7 368.5 10,200 64 Armenia 4 30 135 1.8 156 .. 480 162 .. .. .. .. Australia 19 7,741 2 380.6 14 3.8 20,300 24 2.6 377.5 20,130 25 Austria 8 84 98 217.2 21 3.4 26,850 12 3.2 183.9 22,740 16 Azerbaijan 8 87 91 3.9 125 9.4 490 161 8.1 14.3 1,820 157 Bangladesh 126 144 965 44.0 52 5.0 350 175 3.4 137.7 1,100 188 Belarus 10 208 49 22.5 61 .. 2,200 102 .. .. .. .. Belgium 10 33 311 259.0 19 2.9 25,380 15 2.7 239.7 23,480 12 Benin 6 113 54 2.3 142 4.5 380 173 1.5 7.5 1,250 182 Bolivia 8 1,099 7 7.9 94 4.7 1,000 138 2.3 22.4 2,820 140 Botswana 2 582 3 5.6 107 5.5 3,600 82 3.5 13.0 8,310 70 Brazil 166 8,547 20 758.0 8 0.0 4,570 72 –1.4 1,021.4 6,160 88 Bulgaria 8 111 75 10.1 84 .. 1,230 131 .. .. .. .. Burkina Faso 11 274 39 2.6 140 6.3 240 196 3.8 11.0c 1,020c 191 Burundi 7 28 256 0.9 173 4.6 140 206 2.2 4.1 c 620 c 207 Cambodia 11 181 61 3.0 135 –0.1 280 191 –2.3 13.3 1,240 184 Cameroon 14 475 31 8.7 89 6.7 610 156 3.8 25.9 1,810 158 Canada 31 9,971 3 612.2 9 6.1 20,020 26 5.1 735.6 24,050 9 Central African Republic 3 623 6 1.0 170 4.5 300 186 2.6 4.5 c 1,290 c 181 Chad 7 1,284 6 1.7 160 .. 230 197 .. .. .. .. Chile 15 757 20 71.3 42 8.0 4,810 71 6.5 191.1 12,890 53 China 1,239 9,597d 133 928.9 7 7.4 750 149 6.5 3,983.6 3,220 129 Hong Kong, China 7 1 6,755 158.3e 24 –5.1 23,670e 21 –7.8 147.1 22,000 18 Colombia 41 1,139 39 106.1 35 5.6 2,600 95 3.7 306.0 7,500 76 Congo, Dem. Rep. 48 2,345 21 5.3 108 4.0 110 209 0.7 36.4 c 750 c 200 Congo, Rep. 3 342 8 1.9 151 11.9 690 153 8.9 4.0 1,430 174 Costa Rica 4 51 69 9.8 85 4.7 2,780 93 3.1 23.3 6,620 86 Côte d’Ivoire 14 322 46 10.1 83 5.7 700 152 3.6 25.0 1,730 161 Croatia 5 57 82 20.7 63 .. 4,520 73 .. .. .. .. Czech Republic 10 79 133 51.8 48 .. 5,040 69 .. .. .. .. Denmark 5 43 125 176.4 23 3.0 33,260 6 2.6 126.4 23,830 11 Dominican Republic 8 49 171 14.6 78 6.5 1,770 109 4.6 38.8 4,700 99 Ecuador 12 284 44 18.6 70 2.1 1,530 119 0.2 56.3 4,630 100 Egypt, Arab Rep. 61 1,001 62 79.2 40 5.1 1,290 127 3.3 192.5 3,130 132 El Salvador 6 21 292 11.2 81 3.6 1,850 107 1.4 17.3 2,850 139 Eritrea 4 118 38 0.8 176 –4.0 200 202 –6.7 3.7 950 193 Estonia 1 45 34 4.9 112 .. 3,390 87 .. .. .. .. Ethiopia 61 1,104 61 6.1 104 –0.8 100 210 –3.2 30.8 500 208 Finland 5 338 17 124.3 30 5.2 24,110 19 4.8 104.5 20,270 23 France 59 552 107 1,466.2 4 3.2 24,940 17 2.9 1,312.0 22,320 17 Georgia 5 70 78 5.1 109 .. 930 139 .. .. .. .. Germany 82 357 235 2,122.7 3 –0.4 25,850 13 –0.4 1,708.5 20,810 20 Ghana 18 239 81 7.2 98 4.6 390 171 1.9 29.8c 1,610c 168 Greece 11 132 82 122.9 31 3.7 11,650 47 3.4 137.2 13,010 52 Guatemala 11 109 100 17.7 72 4.8 1,640 115 2.1 44.0 4,070 107 Guinea 7 246 29 3.8 127 4.3 540 159 1.9 12.5 1,760 160 Haiti 8 28 277 3.1 134 3.0 410 167 1.1 9.6c 1,250c 182 Honduras 6 112 55 4.5 117 3.9 730 151 1.0 13.2 2,140 154 Hungary 10 93 110 45.6 51 .. 4,510 74 .. .. .. .. India 980 3,288 330 421.3 11 6.1 430 165 4.2 1,660.9 1,700 163 Indonesia 204 1,905 112 138.5 28 –14.8 680 154 –16.2 568.9 2,790 141 Iran, Islamic Rep. 62 1,633 38 109.6 33 .. 1,770 109 .. .. .. .. Ireland 4 70 53 67.5 43 9.0 18,340 27 8.5 67.5 18,340 30 Israel 6 21 290 95.2 36 1.9 15,940 32 –0.4 103.4 17,310 33 Italy 58 301 196 1,166.2 6 2.3 20,250 25 2.2 1,163.4 20,200 24 Jamaica 3 11 238 4.3 118 –1.1 1,680 113 –1.9 8.3 3,210 130 Japan 126 378 335 4,089.9 2 –2.6 32,380 7 –2.8 2,928.4 23,180 14 Jordan 5 89 51 6.9 100 0.3 1,520 120 –2.5 14.8 3,230 128 Kazakhstan 16 2,717 6 20.6 64 –2.6 1,310 126 –2.0 53.4 3,400 126 Kenya 29 580 51 9.7 86 1.5 330 180 –0.9 33.1 1,130 187 Korea, Rep. 46 99 470 369.9 15 –6.3 7,970 59 –7.1 569.3 12,270 55 Kuwait 2 18 105 .. .. .. . .f .. .. .. .. .. Kyrgyz Republic 5 199 24 1.6 162 4.2 350 175 2.8 10.3 2,200 152 Lao PDR 5 237 22 1.6 163 4.0 330 180 1.4 6.5c 1,300c 180 Latvia 2 65 39 5.9 105 .. 2,430 98 .. .. .. .. Lebanon 4 10 412 15.0 77 4.3 3,560 84 2.7 25.9 6,150 89 Lesotho 2 30 68 1.2 168 –3.1 570 158 –5.4 4.8 c 2,320 c 147 Lithuania 4 65 57 9.0 88 5.6 2,440 97 5.9 15.9 4,310 105 Macedonia, FYR 2 26 79 2.6 139 2.9 1,290 127 2.2 7.4 3,660 116 Madagascar 15 587 25 3.8 128 4.8 260 193 1.6 13.1 900 194 Malawi 11 118 112 2.1 144 1.8 200 202 –0.7 7.7 730 203 Malaysia 22 330 68 79.8 39 –6.3 3,600 82 –8.4 155.1 c 6,990 c 79 Note: For data comparability and coverage, see the Technical Notes. Figures in italics are for years other than those specified. Rankings are based on 210 economies, including the 78 listed in Table 1a. See Technical Notes. (c) The International Bank for Reconstruction and Development / The World Bank      WORLD VIEW Surface Population Gross national product (GNP) GNP per capita GNP measured at PPPa area density Population Thousands People per Billions Avg. annual Avg. annual Billions Per capita Millions of sq. km sq. km of dollars Rank growth rate (%) Dollars Rank growth rate (%) of dollars Dollars Rank Economy 1998 1996 1998 1998b 1998 1997–98 1998b 1998 1997–98 1998 1998 1998 Mali 11 1,240 9 2.6 138 5.3 250 194 2.2 7.7 720 204 Mauritania 3 1,026 2 1.0 171 5.2 410 167 2.4 4.2c 1,660c 165 Mexico 96 1,958 50 380.9 13 4.8 3,970 76 3.0 785.8c 8,190c 71 Moldova 4 34 130 1.8 158 .. 410 167 .. .. .. .. Mongolia 3 1,567 2 1.0 172 4.9 400 170 3.2 3.9 1,520 170 Morocco 28 447 62 34.8 56 0.8 1,250 130 –1.0 86.8 3,120 133 Mozambique 17 802 22 3.6 130 11.3 210 199 9.2 14.5c 850c 196 Myanmar 44 677 68 .. .. .. . .g .. .. .. .. .. Namibia 2 824 2 3.2 131 1.2 1,940 106 –1.2 8.2c 4,950c 94 Nepal 23 147 160 4.8 114 2.2 210 199 –0.1 24.9 1,090 189 Netherlands 16 41 463 388.7 12 3.3 24,760 18 2.7 339.3 21,620 19 New Zealand 4 271 14 55.8 46 1.4 14,700 36 0.5 60.1 15,840 40 Nicaragua 5 130 40 .. .. .. .. g .. .. 8.6 c 1,790 c 159 Niger 10 1,267 8 1.9 150 4.3 190 204 0.8 8.4 830 198 Nigeria 121 924 133 36.4 55 1.1 300 186 –1.7 99.7 820 199 Norway 4 324 14 152.1 25 2.4 34,330 4 1.8 107.6 24,290 8 Pakistan 132 796 171 63.2 44 5.0 480 162 2.5 204.9 1,560 169 Panama 3 76 37 8.5 91 3.8 3,080 90 2.0 19.2 6,940 81 Papua New Guinea 5 463 10 4.1 120 2.3 890 140 0.0 12.4c 2,700c 142 Paraguay 5 407 13 9.2 87 0.2 1,760 111 –2.1 19.0 3,650 117 Peru 25 1,285 19 61.1 45 .. 2,460 96 .. .. .. .. Philippines 75 300 252 78.9 41 0.1 1,050 135 –2.1 265.6 3,540 122 Poland 39 323 127 150.8 26 5.4 3,900 79 5.4 260.7 6,740 83 Portugal 10 92 109 106.4 34 3.9 10,690 51 3.8 143.1 14,380 45 Romania 22 238 98 31.3 58 –5.6 1,390 125 –5.3 89.3 3,970 109 Russian Federation 147 17,075 9 337.9 16 –6.6 2,300 101 –6.3 579.8 3,950 110 Rwanda 8 26 329 1.9 155 9.9 230 197 7.1 5.6 690 206 Saudi Arabia 21 2,150 10 .. .. .. . .h .. .. .. .. .. Senegal 9 197 47 4.8 115 6.0 530 160 3.1 15.4 1,710 162 Sierra Leone 5 72 68 0.7 181 –0.7 140 206 –2.9 1.9 390 210 Singapore 3 1 5,186 95.1 37 1.5 30,060 9 –0.4 90.5 28,620 5 Slovak Republic 5 49 112 20.0 66 .. 3,700 80 .. .. .. .. Slovenia 2 20 99 19.4 67 .. 9,760 52 .. .. .. .. South Africa 41 1,221 34 119.0 32 0.6 2,880 92 –1.2 288.7c 6,990c 79 Spain 39 506 79 553.7 10 3.7 14,080 39 3.7 631.5 16,060 38 Sri Lanka 19 66 290 15.2 76 .. 810 144 .. .. .. .. Sweden 9 450 22 226.9 20 3.5 25,620 14 3.5 172.5 19,480 27 Switzerland 7 41 180 284.8 18 2.1 40,080 3 1.8 189.1 26,620 7 Syrian Arab Republic 15 185 83 15.6 75 4.4 1,020 136 1.8 45.8 3,000 136 Tajikistan 6 143 43 2.1 143 .. 350 175 .. .. .. .. Tanzania 32 945 36 6.7i 101 3.2 210i 199 0.6 15.9 490 209 Thailand 61 513 120 134.4 29 –7.7 2,200 102 –8.5 357.1 5,840 91 Togo 4 57 82 1.5 164 –1.0 330 180 –3.5 6.2 1,390 176 Tunisia 9 164 60 19.2 69 5.5 2,050 105 3.9 48.3 5,160 93 Turkey 63 775 82 200.5 22 .. 3,160 89 .. .. .. .. Turkmenistan 5 488 10 .. 136 .. .. g .. .. .. .. .. Uganda 21 241 105 6.7 102 5.8 320 185 2.9 24.5c 1,170c 185 Ukraine 50 604 87 42.7 53 .. 850 142 .. .. .. .. United Kingdom 59 245 244 1,263.8 5 2.0 21,400 22 1.9 1,218.6 20,640 22 United States 270 9,364 29 7,921.3 1 3.7 29,340 10 2.8 7,922.6 29,340 3 Uruguay 3 177 19 20.3 65 6.6 6,180 67 5.8 31.2 9,480 67 Uzbekistan 24 447 58 20.9 62 3.0 870 141 1.2 69.8 2,900 138 Venezuela 23 912 26 81.3 38 –0.4 3,500 85 –2.4 190.4 8,190 71 Vietnam 78 332 238 25.6 60 4.0 330 180 2.8 131.0 1,690 164 Yemen, Rep. 16 528 31 4.9 110 7.3 300 186 4.6 12.1 740 202 Zambia 10 753 13 3.2 132 –1.8 330 180 –4.0 8.3 860 195 Zimbabwe 12 391 30 7.1 99 –0.4 610 156 –2.2 25.2 2,150 153 World 5,897 s 133,567 s 45 w 28,862.2 t 1.5 w 4,890 t 0.1 w 36,556.8 t 6,200 w Low income 3,515 42,695 85 1,843.7 3.8 520 2.1 7,475.1 2,130 Excl. China & India 1,296 29,810 45 493.5 –3.9 380 –5.9 1,821.3 1,400 Middle income 1,496 58,789 26 4,419.6 –0.4 2,950 –1.5 8,315.8 5,560 Lower middle income 908 36,729 25 1,557.4 –1.5 1,710 –2.6 3,709.4 4,080 Upper middle income 588 22,060 27 2,862.1 0.2 4,860 –1.1 4,606.3 7,830 Low and middle income 5,011 101,484 50 6,263.3 1.0 1,250 –0.5 15,790.8 3,150 East Asia & Pacific 1,817 16,384 114 1,801.6 –1.1 990 –2.2 6,179.5 3,400 Europe & Central Asia 473 24,208 20 1,038.8 .. 2,190 .. 2,005.5 4,240 Latin America & Carib. 502 20,462 25 1,977.6 2.5 3,940 0.8 3,401.5 6,780 Middle East & N. Africa 285 11,000 26 585.6 .. 2,050 .. 1,203.3 4,220 South Asia 1,305 5,140 273 555.5 5.9 430 3.9 2,100.4 1,610 Sub-Saharan Africa 628 24,290 27 304.2 2.2 480 –0.4 900.6 1,430 High income 885 32,082 29 22,599.0 1.6 25,510 1.1 20,766.0 23,440 a. Purchasing power parity; see the Technical Notes. b. Preliminary World Bank estimates calculated using the World Bank Atlas method. c. Estimate based on regression; others are extrapolated from the latest International Comparison Programme benchmark estimates. d. Includes Taiwan, China. e. GNP data refer to GDP. f. Estimated to be high income ($9,361 or more). g. Estimated to be low income ($760 or less). h. Estimated to be upper middle income ($3,031 to $9,360). i. Data refer to mainland Tanzania only. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 2. Quality of life Access to Growth of sanitation private consumption Prevalence Adult in urban per capita of child Life expectancy illiteracy rate areas Avg. annual growth malnutrition Under-5 at birth % of people Urban % of rate (%) 1980–97 % of children mortality rate Years 15 and above population urban pop. Distribution- under age 5 Per 1,000 1997 1997 % of total with access Economy Uncorrected corrected 1992–97a 1980 1997 Males Females Males Females 1980 1998 1995 Albania .. .. .. 57 40 69 75 .. .. 34 38 97 Algeria –1.8 –1.2 13 139 39 69 72 27 52 43 58 .. Angola –7.8 .. 35 261 209 45 48 .. .. 21 33 71 Argentina .. .. 2 38 24 70 77 3 4 83 89 80 Armenia .. .. .. .. .. 70 77 .. .. 66 69 .. Australia 1.7 1.1 0 13 7 76 81 .. .. 86 85 .. Austria 2.0 1.6 .. 17 7 74 81 .. .. 65 65 .. Azerbaijan .. .. 10 .. 23 67 75 .. .. 53 57 .. Bangladesh 2.1 1.5 56 211 104 58 58 50 73 11 20 41 Belarus –3.5 –2.7 .. .. .. 63 74 0 2 56 73 .. Belgium 1.6 1.2 .. 15 7 73 80 .. .. 95 97 .. Benin –0.7 .. 29 214 149 52 55 52 79 27 41 60 Bolivia 0.1 0.0 8 170 96 60 63 9 23 46 63 77 Botswana 2.3 .. 27 94 88 46 48 28 23 15 68 91 Brazil 0.5 0.2 6 .. 44 63 71 16 16 66 80 74 Bulgaria –0.6 –0.4 .. 25 24 67 74 1 2 61 69 .. Burkina Faso 0.3 .. 33 .. 169 44 45 70 89 9 17 .. Burundi –0.8 .. 38 193 200 41 44 46 64 4 8 .. Cambodia .. .. 38 330 147 53 55 .. .. 12 22 .. Cameroon –1.5 .. .. 173 78 55 58 21 35 31 47 .. Canada 1.3 0.9 .. 13 8 76 82 .. .. 76 77 .. Central African Republic –1.5 .. 23 .. 160 43 47 44 70 35 40 .. Chad 0.0 .. 39 235 182 47 50 .. .. 19 23 74 Chile 3.8 1.7 1 35 13 72 78 5 5 81 84 95 China 7.7 4.5 16 65 39 68 71 9 25 20 33 68 Hong Kong, China 5.2 .. .. .. .. 76 82 4 12 91 95 .. Colombia 1.2 0.5 8 58 30 67 73 9 9 64 74 70 Congo, Dem. Rep. –4.5 .. 34 210 148 49 52 .. .. 29 30 53 Congo, Rep. 0.2 .. 24 125 145 46 51 15 30 41 61 .. Costa Rica 0.8 0.4 5 29 15 74 79 5 5 43 51 100 Côte d’Ivoire –2.3 –1.5 24 170 140 46 47 49 66 35 45 .. Croatia .. .. 1 23 10 68 77 1 4 50 57 71 Czech Republic .. .. 1 19 8 71 78 .. .. 64 66 .. Denmark 1.7 1.3 .. 10 6 73 78 .. .. 84 86 .. Dominican Republic –0.2 –0.1 6 92 47 69 73 17 18 51 64 89 Ecuador –0.2 –0.1 17 101 39 68 73 7 11 47 61 70 Egypt, Arab Rep. 2.0 1.3 15 175 66 65 68 35 60 44 45 95 El Salvador 2.9 1.5 11 120 39 67 73 20 26 42 46 89 Eritrea .. .. 44 .. 95 49 52 .. .. 14 18 12 Estonia –2.2 –1.3 .. 25 13 64 76 .. .. 70 74 .. Ethiopia –0.4 .. 48 213 175 42 44 59 71 11 17 .. Finland 1.4 1.1 .. 9 5 73 81 .. .. 60 64 100 France 1.7 1.1 .. 13 6 74 82 .. .. 73 75 .. Georgia .. .. .. .. 21 69 77 .. .. 52 60 .. Germany .. .. .. 16 6 74 80 .. .. 83 87 .. Ghana 0.2 0.1 27 157 102 58 62 23 43 31 37 75 Greece 1.8 .. .. 23 9 75 81 2 5 58 60 .. Guatemala 0.1 0.0 27 .. 55 61 67 26 41 37 40 91 Guinea 1.0 0.5 24 299 182 46 47 .. .. 19 31 24 Haiti .. .. 28 200 125 51 56 52 57 24 34 43 Honduras –0.2 –0.1 18 103 48 67 72 29 30 35 46 91 Hungary –0.1 –0.1 .. 26 12 66 75 1 1 57 66 .. India 2.7 1.9 53 177 88 62 64 33 61 23 28 .. Indonesia 4.5 3.0 34 125 60 63 67 9 20 22 38 88 Iran, Islamic Rep. 0.2 .. 16 126 35 69 70 19 34 50 61 86 Ireland 2.7 1.8 .. 14 7 73 79 .. .. 55 58 .. Israel 3.3 2.1 .. 19 8 76 79 2 7 89 91 100 Italy 2.2 1.5 .. 17 7 75 82 1 2 67 67 .. Jamaica 2.2 1.3 10 39 14 72 77 19 10 47 55 99 Japan 2.9 .. .. 11 6 77 83 .. .. 76 79 .. Jordan –1.2 –0.7 10 48 35 69 73 8 18 60 73 .. Kazakhstan .. .. 8 .. 29 60 70 .. .. 54 61 .. Kenya 0.9 0.4 23 115 112 51 53 13 28 16 31 .. Korea, Rep. 7.0 .. .. 18 11 69 76 1 4 57 84 100 Kuwait .. .. 11 35 13 74 80 17 23 90 97 100 Kyrgyz Republic .. .. 11 .. .. 63 71 .. .. 38 40 .. Lao PDR .. .. 40 200 .. 52 55 .. .. 13 22 .. Latvia .. .. .. 26 19 64 75 0 1 68 74 90 Lebanon .. .. 3 .. 32 68 72 9 22 74 89 .. Lesotho –2.8 –1.2 16 168 137 55 57 29 7 13 26 76 Lithuania .. .. .. 24 13 66 77 0 1 61 74 .. Macedonia, FYR .. .. .. 69 17 70 75 .. .. 53 61 .. Madagascar –2.4 –0.2 34 216 158 56 59 .. .. 18 28 .. Malawi 0.6 .. 30 265 224 43 43 27 57 9 15 94 Malaysia 3.1 1.6 20 42 14 70 75 10 19 42 56 94 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      WORLD VIEW Access to Growth of sanitation private consumption Prevalence Adult in urban per capita of child Life expectancy illiteracy rate areas Avg. annual growth malnutrition Under-5 at birth % of people Urban % of rate (%) 1980–97 % of children mortality rate Years 15 and above population urban pop. Distribution- under age 5 Per 1,000 1997 1997 % of total with access Economy Uncorrected corrected 1992–97a 1980 1997 Males Females Males Females 1980 1998 1995 Mali –0.4 .. 40 .. 235 49 52 57 72 19 29 61 Mauritania 0.1 0.1 23 175 149 52 55 51 72 27 55 44 Mexico 0.1 0.0 14 74 38 69 75 8 12 66 74 93 Moldova .. .. .. .. 24 63 70 1 3 40 54 96 Mongolia .. .. 12 .. 68 64 67 .. .. 52 62 100 Morocco 1.6 1.0 10 152 67 65 69 41 67 41 54 97 Mozambique –2.3 .. 26 223 201 44 47 43 75 13 38 68 Myanmar .. .. 43 134 131 59 62 11 21 24 27 56 Namibia –3.0 .. 26 114 101 55 57 19 22 23 39 .. Nepal 2.1 1.3 47 180 117 58 57 44 79 7 11 74 Netherlands 1.6 1.1 .. 11 7 75 81 .. .. 88 89 .. New Zealand 0.9 .. .. 16 7 75 80 .. .. 83 87 .. Nicaragua –2.6 –1.3 12 143 57 66 71 37 37 53 64 88 Niger –2.6 –1.7 43 320 .. 45 50 78 93 13 20 .. Nigeria –4.7 –2.6 39 196 122 52 55 31 49 27 42 82 Norway 1.5 1.2 .. 11 6 76 81 .. .. 71 74 100 Pakistan 2.0 1.4 38 161 136 61 63 45 75 28 36 75 Panama 1.6 0.7 6 36 26 72 76 8 10 50 57 99 Papua New Guinea –1.1 –0.5 30 100 82 57 59 .. .. 13 17 .. Paraguay 1.8 0.7 .. 61 28 68 72 6 9 42 55 20 Peru –0.5 –0.3 8 126 52 66 71 6 16 65 72 78 Philippines 0.7 0.4 30 81 41 67 70 5 6 37 57 88 Poland 0.9 0.6 .. .. 12 69 77 0 0 58 65 .. Portugal 3.1 .. .. 31 8 71 79 6 12 29 37 .. Romania 0.3 0.2 6 36 26 65 73 1 3 49 57 .. Russian Federation .. .. 3 .. 25 61 73 0 1 70 77 .. Rwanda –1.1 –0.8 29 .. 209 39 42 29 44 5 6 .. Saudi Arabia .. .. .. 85 28 69 72 19 38 66 85 .. Senegal –0.7 –0.3 22 190 110 51 54 55 75 36 46 68 Sierra Leone –3.2 –1.2 .. 336 286 36 39 .. .. 24 35 .. Singapore 4.9 .. .. 13 6 73 79 4 13 100 100 .. Slovak Republic .. .. .. 23 .. 69 77 .. .. 52 60 .. Slovenia .. .. .. 18 6 71 79 0 0 48 52 100 South Africa –0.3 –0.1 9 91 65 62 68 15 17 48 50 78 Spain 2.2 1.5 .. 16 7 75 82 2 4 73 77 .. Sri Lanka 2.8 1.9 38 48 19 71 75 6 12 22 23 81 Sweden 0.7 0.5 .. 9 5 77 82 .. .. 83 83 .. Switzerland 0.6 0.4 .. 11 6 76 82 .. .. 57 62 .. Syrian Arab Republic 1.0 .. 13 73 38 67 71 13 43 47 54 .. Tajikistan .. .. .. .. 36 66 71 1 2 34 33 .. Tanzania 0.0 0.0 31 176 136 47 49 18 38 15 26 .. Thailand 5.5 2.9 .. 58 38 66 72 3 7 17 21 98 Togo –0.5 .. 19 175 138 48 50 31 62 23 32 76 Tunisia 1.0 0.6 9 100 33 68 71 22 44 52 64 100 Turkey 2.5 .. 10 133 50 67 72 8 26 44 73 .. Turkmenistan .. .. .. .. 50 62 69 .. .. 47 45 .. Uganda 1.7 1.0 26 180 162 43 42 25 47 9 14 60 Ukraine .. .. .. .. 17 62 73 .. .. 62 72 .. United Kingdom 2.6 1.8 .. 14 7 75 80 .. .. 89 89 .. United States 1.9 1.1 1 15 .. 73 79 .. .. 74 77 .. Uruguay 2.4 .. 4 42 20 70 78 3 2 85 91 56 Uzbekistan .. .. 19 .. 31 66 72 .. .. 41 42 .. Venezuela –0.8 –0.4 5 42 25 70 76 7 8 79 87 74 Vietnam .. .. 45 105 40 66 71 5 11 19 20 .. Yemen, Rep. .. .. 29 198 137 54 55 36 79 20 36 40 Zambia –3.7 –2.0 24 149 189 43 43 17 33 40 44 66 Zimbabwe 0.3 .. 16 108 108 51 54 6 12 22 34 .. World 3.1 w 2.1 w 125 w 79 w 65 w 69 w 18 w 33 w 39 w 46 w ..w Low income 3.9 2.7 151 97 62 64 22 42 21 31 .. Excl. China & India 0.4 .. 178 130 55 58 30 47 21 31 .. Middle income 1.2 .. .. 42 66 72 10 16 55 66 .. Lower middle income .. .. .. 47 65 71 11 18 50 58 .. Upper middle income 1.5 .. .. 34 67 74 9 13 63 77 83 Low and middle income 3.3 2.2 137 83 63 67 19 34 32 41 .. East Asia & Pacific 6.8 4.0 81 46 67 70 9 22 22 35 75 Europe & Central Asia .. .. .. 30 64 73 2 6 56 68 .. Latin America & Carib. 0.5 0.2 .. 41 66 73 12 14 65 75 80 Middle East & N. Africa 0.7 .. 137 62 66 68 27 50 48 58 .. South Asia 2.5 1.8 180 100 62 63 36 63 22 27 .. Sub-Saharan Africa –2.1 .. 189 147 49 52 34 50 23 33 .. High income 2.1 1.2 15 7 74 81 .. .. 75 76 .. a. Data are for the most recent year available within the period. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 3. Population and labor force Population Labor force Total Avg. annual Aged 15–64 Total Avg. annual Female Children aged 10–14 Millions growth rate (%) Millions Millions growth rate (%) % of labor force % of age group Economy 1980 1998 1980–90 1990–98 1980 1998 1980 1998 1980–90 1990–98 1980 1998 1980 1998 Albania 2.7 3.4 2.3 0.3 2 2 1 2 3.1 0.6 39 41 4 1 Algeria 18.7 30.0 4.7 2.6 9 18 5 10 7.1 4.9 21 26 7 1 Angola 7.0 12.0 5.4 3.8 4 6 3 6 4.7 3.4 47 46 30 27 Argentina 28.1 36.1 2.5 1.5 17 22 11 14 3.0 2.2 28 32 8 4 Armenia 3.1 3.8 2.0 1.0 2 2 1 2 2.5 1.3 48 48 0 0 Australia 14.7 18.8 2.4 1.3 10 13 7 10 3.5 1.6 37 43 0 0 Austria 7.6 8.1 0.7 0.7 5 6 3 4 1.1 1.0 41 40 0 0 Azerbaijan 6.2 7.9 2.5 1.4 4 5 3 3 2.2 1.6 48 44 0 0 Bangladesh 86.7 125.6 3.7 1.9 44 71 41 64 4.5 3.3 42 42 35 29 Belarus 9.6 10.2 0.6 0.0 6 7 5 5 0.4 0.0 50 49 0 0 Belgium 9.8 10.2 0.4 0.3 6 7 4 4 0.6 0.7 34 41 0 0 Benin 3.5 6.0 5.4 3.3 2 3 2 3 4.8 3.3 47 48 30 27 Bolivia 5.4 7.9 3.9 2.7 3 4 2 3 4.5 2.7 33 38 19 13 Botswana 0.9 1.6 5.4 2.9 0 1 0 1 5.4 2.9 50 46 26 16 Brazil 121.7 165.9 3.1 1.6 70 108 47 76 4.7 2.3 28 35 19 15 Bulgaria 8.9 8.2 –0.7 –0.8 6 6 5 4 –0.9 –0.8 45 48 0 0 Burkina Faso 7.0 10.7 4.3 2.7 3 5 4 5 3.4 2.1 48 47 71 48 Burundi 4.1 6.6 4.7 2.7 2 3 2 4 4.5 2.7 50 49 50 49 Cambodia 6.5 10.7 5.0 3.1 4 6 4 6 4.6 3.1 55 52 27 24 Cameroon 8.7 14.3 5.0 3.2 5 8 4 6 4.8 3.5 37 38 34 24 Canada 24.6 30.6 2.2 1.4 17 21 12 17 3.0 1.6 40 45 0 0 Central African Republic 2.3 3.5 4.1 2.4 1 2 .. .. .. .. .. .. 39 30 Chad 4.5 7.4 5.0 3.5 2 3 2 4 4.6 3.5 43 45 42 38 Chile 11.1 14.8 2.8 1.8 7 10 4 6 4.5 2.5 26 33 0 0 China 981.2 1,238.6 2.3 1.2 586 836 540 743 3.2 1.5 43 45 30 10 Hong Kong, China 5.0 6.7 2.8 2.3 3 5 2 3 3.4 2.5 34 37 6 0 Colombia 28.4 40.8 3.6 2.2 16 25 9 18 6.3 3.2 26 38 12 6 Congo, Dem. Rep. 27.0 48.2 5.8 3.6 14 24 12 20 5.3 3.6 45 43 33 29 Congo, Rep. 1.7 2.8 5.1 3.2 1 1 1 1 4.9 2.9 42 43 27 26 Costa Rica 2.3 3.5 4.3 2.1 1 2 1 1 5.7 2.8 21 31 10 5 Côte d’Ivoire 8.2 14.5 5.7 3.1 4 8 3 6 5.7 3.9 32 33 28 20 Croatia 4.6 4.6 0.0 –0.6 3 3 2 2 0.4 0.0 40 44 0 0 Czech Republic 10.2 10.3 0.1 –0.1 6 7 5 6 0.8 0.7 47 47 0 0 Denmark 5.1 5.3 0.3 0.4 3 4 3 3 0.9 0.2 44 46 0 0 Dominican Republic 5.7 8.3 3.7 2.1 3 5 2 4 5.2 3.2 25 30 25 15 Ecuador 8.0 12.2 4.2 2.4 4 7 3 5 6.0 3.6 20 27 9 5 Egypt, Arab Rep. 40.9 61.4 4.1 2.3 23 37 14 23 4.6 3.0 27 30 18 10 El Salvador 4.6 6.1 2.8 2.4 2 4 2 3 4.9 3.9 27 36 17 15 Eritrea 2.4 3.9 4.9 3.0 .. 2 1 2 4.7 3.0 47 47 44 39 Estonia 1.5 1.4 –0.3 –1.2 1 1 1 1 –0.3 –0.9 51 49 0 0 Ethiopia 37.7 61.3 4.9 2.6 20 31 17 26 4.4 1.9 42 41 46 42 Finland 4.8 5.2 0.8 0.5 3 3 2 3 1.0 0.2 47 48 0 0 France 53.9 58.8 0.9 0.5 34 38 24 26 1.1 0.8 40 45 0 0 Georgia 5.1 5.4 0.7 –0.1 3 4 3 3 0.5 –0.1 49 47 0 0 Germany 78.3 82.1 0.5 0.5 52 56 38 41 0.9 0.5 40 42 0 0 Ghana 10.7 18.5 5.4 3.1 6 10 5 9 5.4 3.1 51 51 16 13 Greece 9.6 10.5 0.9 0.5 6 7 4 5 1.9 1.2 28 37 5 0 Guatemala 6.8 10.8 4.6 3.0 3 6 2 4 5.2 3.4 22 28 19 15 Guinea 4.5 7.1 4.6 3.0 2 4 2 3 4.0 2.7 47 47 41 33 Haiti 5.4 7.6 3.6 2.4 3 4 3 3 2.9 2.0 45 43 33 24 Honduras 3.6 6.2 5.5 3.3 2 3 1 2 6.3 4.5 25 31 14 8 Hungary 10.7 10.1 –0.6 –0.3 7 7 5 5 –0.6 0.3 43 45 0 0 India 687.3 979.7 3.5 2.0 394 596 302 431 3.5 2.7 34 32 21 13 Indonesia 148.3 203.7 3.2 1.9 83 130 58 98 5.2 3.1 35 40 13 9 Iran, Islamic Rep. 39.1 61.9 4.6 1.9 20 36 12 19 4.6 2.3 20 26 14 4 Ireland 3.4 3.7 0.8 0.7 2 2 1 2 1.8 2.2 28 34 1 0 Israel 3.9 6.0 4.3 3.5 2 4 1 3 5.8 4.9 34 41 0 0 Italy 56.4 57.6 0.2 0.2 36 39 23 25 1.2 0.5 33 38 2 0 Jamaica 2.1 2.6 1.9 1.0 1 2 1 1 3.3 1.8 46 46 0 0 Japan 116.8 126.3 0.8 0.3 79 87 57 68 1.8 0.9 38 41 0 0 Jordan 2.2 4.6 7.4 5.2 1 3 1 1 9.3 6.2 15 23 4 0 Kazakhstan 14.9 15.7 0.5 –0.6 .. 10 .. 8 .. –0.3 48 47 0 0 Kenya 16.6 29.3 5.7 3.1 8 15 8 15 6.5 4.0 46 46 45 40 Korea, Rep. 38.1 46.4 2.0 1.1 24 33 16 23 4.0 2.3 39 41 0 0 Kuwait 1.4 1.9 3.1 –1.8 1 1 0 1 3.9 –2.9 13 31 0 0 Kyrgyz Republic 3.6 4.7 2.6 1.0 2 3 2 2 2.6 1.6 48 47 0 0 Lao PDR 3.2 5.0 4.4 3.0 2 3 .. .. .. .. .. .. 31 26 Latvia 2.5 2.4 –0.4 –1.3 2 2 1 1 –0.8 –1.6 51 50 0 0 Lebanon 3.0 4.2 3.4 2.1 2 3 1 1 5.3 3.4 23 29 5 0 Lesotho 1.3 2.1 4.3 2.6 1 1 1 1 4.0 2.9 38 37 28 22 Lithuania 3.4 3.7 0.8 –0.1 2 2 2 2 0.6 –0.1 50 48 0 0 Macedonia, FYR 1.9 2.0 0.6 0.8 1 1 1 1 1.3 1.1 36 41 1 0 Madagascar 8.9 14.6 5.0 3.2 5 8 4 7 4.6 3.2 45 45 40 35 Malawi 6.2 10.5 5.3 3.1 3 5 3 5 4.9 2.8 51 49 45 34 Malaysia 13.8 22.2 4.8 2.8 8 13 5 9 5.5 3.5 34 37 8 3 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      PEOPLE Population Labor force Total Avg. annual Aged 15–64 Total Avg. annual Female Children aged 10–14 Millions growth rate (%) Millions Millions growth rate (%) % of labor force % of age group Economy 1980 1998 1980–90 1990–98 1980 1998 1980 1998 1980–90 1990–98 1980 1998 1980 1998 Mali 6.6 10.6 4.7 3.2 3 5 3 5 4.3 2.9 47 46 61 53 Mauritania 1.6 2.5 4.9 3.2 1 1 1 1 4.5 3.5 45 44 30 23 Mexico 67.6 95.9 3.5 2.0 35 59 22 38 5.4 3.1 27 33 9 6 Moldova 4.0 4.3 0.7 –0.2 3 3 2 2 0.3 0.1 50 49 3 0 Mongolia 1.7 2.6 4.4 2.2 1 2 1 1 5.0 3.1 46 47 4 2 Morocco 19.4 27.8 3.6 2.1 10 17 7 11 4.4 2.8 34 35 21 4 Mozambique 12.1 16.9 3.4 2.6 6 9 7 9 2.8 2.3 49 48 39 33 Myanmar 33.8 44.4 2.7 1.3 19 29 17 24 3.1 1.9 44 43 28 24 Namibia 1.0 1.7 4.8 3.0 1 1 0 1 4.3 2.6 40 41 34 20 Nepal 14.5 22.9 4.5 2.8 8 13 7 11 4.1 2.8 39 40 56 44 Netherlands 14.2 15.7 1.0 0.7 9 11 6 7 2.7 1.0 32 40 0 0 New Zealand 3.1 3.8 2.0 1.7 2 2 1 2 3.7 2.3 34 45 0 0 Nicaragua 2.9 4.8 5.0 3.2 1 3 1 2 6.6 4.8 28 35 19 13 Niger 5.6 10.1 6.0 3.9 3 5 3 5 5.3 3.3 45 44 48 45 Nigeria 71.1 121.3 5.3 3.3 36 64 29 49 5.1 3.3 36 36 29 25 Norway 4.1 4.4 0.8 0.6 3 3 2 2 1.6 1.2 41 46 0 0 Pakistan 82.7 131.6 4.6 2.8 44 71 29 49 5.2 3.2 23 28 23 17 Panama 2.0 2.8 3.5 2.0 1 2 1 1 5.3 3.1 30 35 6 3 Papua New Guinea 3.1 4.6 4.0 2.6 2 3 2 2 4.0 2.6 42 42 28 18 Paraguay 3.1 5.2 5.1 3.0 2 3 1 2 5.1 3.4 27 30 15 7 Peru 17.3 24.8 3.6 2.0 9 15 5 9 5.4 3.2 24 31 4 2 Philippines 48.3 75.1 4.4 2.6 27 44 19 32 5.2 3.3 35 38 14 7 Poland 35.6 38.7 0.8 0.2 23 26 19 20 0.6 0.8 45 46 0 0 Portugal 9.8 10.0 0.2 0.1 6 7 5 5 0.8 0.4 39 44 8 2 Romania 22.2 22.5 0.1 –0.5 14 15 11 11 –0.3 –0.1 46 44 0 0 Russian Federation 139.0 146.9 0.6 –0.1 95 101 76 78 0.2 0.1 49 49 0 0 Rwanda 5.2 8.1 4.5 2.2 3 4 3 4 5.1 2.7 49 49 43 42 Saudi Arabia 9.4 20.7 7.9 3.9 5 12 3 7 8.9 3.5 8 15 5 0 Senegal 5.5 9.0 4.9 3.0 3 5 3 4 4.7 3.0 42 43 43 30 Sierra Leone 3.2 4.9 4.1 2.8 2 2 1 2 3.5 2.8 36 37 19 15 Singapore 2.3 3.2 3.3 2.2 2 2 1 2 4.1 2.0 35 39 2 0 Slovak Republic 5.0 5.4 0.8 0.3 3 4 2 3 1.6 1.1 45 48 0 0 Slovenia 1.9 2.0 0.4 –0.1 1 1 1 1 0.4 0.2 46 46 0 0 South Africa 27.6 41.3 4.0 2.3 16 25 10 16 4.3 2.3 35 38 1 0 Spain 37.4 39.3 0.5 0.2 23 27 14 17 2.2 1.2 28 37 0 0 Sri Lanka 14.7 18.8 2.4 1.4 9 12 5 8 3.9 2.4 27 36 4 2 Sweden 8.3 8.9 0.6 0.5 5 6 4 5 1.2 0.5 44 48 0 0 Switzerland 6.3 7.1 1.2 0.8 4 5 3 4 2.4 1.1 37 40 0 0 Syrian Arab Republic 8.7 15.3 5.6 3.3 4 8 2 5 6.6 4.8 24 26 14 4 Tajikistan 4.0 6.1 4.3 2.0 2 3 2 2 4.1 2.8 47 44 0 0 Tanzania 18.6 32.1 5.5 3.3 9 17 9 16 5.5 3.0 50 49 43 38 Thailand 46.7 61.1 2.7 1.4 26 42 24 37 4.1 2.1 47 46 25 15 Togo 2.6 4.5 5.3 3.4 1 2 1 2 4.6 3.1 39 40 36 28 Tunisia 6.4 9.4 3.8 2.0 3 6 2 4 5.2 3.5 29 31 6 0 Turkey 44.5 63.5 3.6 1.8 25 42 19 30 4.9 3.3 36 37 21 22 Turkmenistan 2.9 4.7 5.0 3.6 2 3 1 2 5.5 4.3 47 46 0 0 Uganda 12.8 20.9 4.9 3.5 6 10 7 10 4.3 3.0 48 48 49 45 Ukraine 50.0 50.3 0.0 –0.5 33 34 27 25 –0.5 –0.5 50 49 0 0 United Kingdom 56.3 59.1 0.5 0.4 36 38 27 30 0.9 0.4 39 44 0 0 United States 227.2 270.0 1.7 1.1 151 177 109 138 2.3 1.4 41 46 0 0 Uruguay 2.9 3.3 1.2 0.8 2 2 1 1 2.4 1.1 31 41 4 2 Uzbekistan 16.0 24.1 4.1 2.3 9 14 6 10 4.6 3.3 48 47 0 0 Venezuela 15.1 23.2 4.3 2.5 8 14 5 9 5.9 3.6 27 34 4 1 Vietnam 53.7 77.6 3.7 2.3 28 47 26 40 4.3 2.3 48 49 22 8 Yemen, Rep. 8.5 16.5 6.6 4.7 4 8 2 5 7.6 5.6 33 28 26 20 Zambia 5.7 9.7 5.2 3.1 3 5 2 4 5.2 3.4 45 45 19 16 Zimbabwe 7.0 11.7 5.1 2.6 3 6 3 5 5.3 2.6 44 44 37 28 World 4,429.9 s 5,896.5 s 2.9 w 1.6 w 2,586 s 3,697 s 2,028 s 2,847 s 3.4 w 2.0 w 39 t 41 t 20 w 13 w Low income 2,508.6 3,514.7 3.4 2.0 1,423 2,155 1,206 1,759 3.8 2.3 40 41 28 17 Excl. China & India 840.0 1,296.4 4.3 2.6 442 723 364 585 4.7 3.0 40 41 29 24 Middle income 1,132.1 1,496.4 2.8 1.5 658 950 465 658 3.5 2.0 37 39 10 6 Lower middle income 695.0 908.3 2.7 1.4 404 572 292 397 3.1 1.7 40 40 9 4 Upper middle income 437.1 588.1 3.0 1.6 254 379 173 261 4.1 2.4 33 36 11 9 Low and middle income 3,640.7 5,011.1 3.2 1.8 2,080 3,105 1,672 2,417 3.7 2.2 39 40 23 14 East Asia & Pacific 1,397.5 1,817.1 2.6 1.5 820 1,204 718 1,026 3.6 1.8 43 44 26 10 Europe & Central Asia 425.8 473.4 1.1 0.2 265 315 207 236 1.3 0.6 47 46 3 4 Latin America & Carib. 360.3 501.9 3.3 1.9 201 313 130 212 4.9 2.7 28 34 13 9 Middle East & N. Africa 173.7 285.1 5.0 2.6 91 165 54 94 5.5 3.4 24 27 14 5 South Asia 902.6 1,305.3 3.7 2.1 508 778 392 574 3.8 2.8 34 33 23 16 Sub-Saharan Africa 380.7 628.3 5.0 3.0 195 330 170 275 4.8 3.0 42 42 35 30 High income 789.2 885.5 1.2 0.7 505 592 357 430 1.9 1.1 38 43 0 0 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 4. Poverty International poverty lines National poverty lines Population Poverty Population Poverty below gap at below gap at Population below the Population below the $1 PPP $1 PPP $2 PPP $2 PPP poverty line (%) poverty line (%) a day a a day a a day a a daya Economy Survey year Rural Urban Total Survey year Rural Urban Total Survey year % % % % Albania 1994 28.9 .. .. 1996 . . 19.6 .. .. .. .. .. Algeria 1988 16.6 7.3 12.2 1995 30.3 14.7 22.6 1995 <2 .. 17.6 4.4 Angola .. .. .. .. .. .. .. .. .. .. Argentina 1991 .. . . 25.5 .. .. .. .. .. .. .. Armenia .. .. .. .. .. .. .. .. .. .. Australia .. .. .. .. .. .. .. .. .. .. Austria .. .. .. .. .. .. .. .. .. .. Azerbaijan 1995 .. . . 68.1 .. .. .. .. .. .. .. Bangladesh 1991–92 46.0 23.3 42.7 1995–96 39.8 14.3 35.6 .. .. .. .. Belarus 1995 .. . . 22.5 .. .. .. 1993 <2 .. 6.4 0.8 Belgium .. .. .. .. .. .. .. .. .. .. Benin 1995 .. . . 33.0 .. .. .. .. .. .. .. Bolivia .. .. .. .. .. .. .. .. .. .. Botswana .. .. .. .. .. . . 1985–86 33.0 12.4 61.0 30.4 Brazil 1990 32.6 13.1 17.4 .. .. .. 1995 23.6 10.7 43.5 22.4 Bulgaria .. .. .. .. .. .. 1992 2.6 0.8 23.5 6.0 Burkina Faso .. .. .. .. .. .. .. .. .. .. Burundi 1990 .. . . 36.2 .. .. .. .. .. .. .. Cambodia 1993–94 43.1 24.8 39.0 1997 40.1 21.1 36.1 .. .. .. .. Cameroon 1984 32.4 44.4 40.0 .. .. .. .. .. .. .. Canada .. .. .. .. .. .. .. .. .. .. Central African Republic .. .. .. .. .. .. .. .. .. .. Chad 1995–96 67.0 63.0 64.0 .. .. .. .. .. .. .. Chile 1992 .. . . 21.6 1994 .. .. 20.5 1992 15.0 4.9 38.5 16.0 China 1994 11.8 <2 8.4 1996 7.9 <2 6.0 1995 22.2 6.9 57.8 24.1 Hong Kong, China .. .. .. .. .. .. .. .. .. .. Colombia 1991 29.0 7.8 16.9 1992 31.2 8.0 17.7 1991 7.4 2.3 21.7 8.4 Congo, Dem. Rep. .. .. .. .. .. .. .. .. .. .. Congo, Rep. .. .. .. .. .. .. .. .. .. .. Costa Rica .. .. .. .. .. .. 1989 18.9 7.2 43.8 19.4 Côte d’Ivoire .. .. .. .. .. .. 1988 17.7 4.3 54.8 20.4 Croatia .. .. .. .. .. .. .. .. .. .. Czech Republic .. .. .. .. .. .. 1993 3.1 0.4 55.1 14.0 Denmark .. .. .. .. .. .. .. .. .. .. Dominican Republic 1989 27.4 23.3 24.5 1992 29.8 10.9 20.6 1989 19.9 6.0 47.7 20.2 Ecuador 1994 47.0 25.0 35.0 .. .. .. 1994 30.4 9.1 65.8 29.6 Egypt, Arab Rep. .. .. .. .. .. . . 1990–91 7.6 1.1 51.9 15.3 El Salvador 1992 55.7 43.1 48.3 .. .. .. .. .. .. .. Eritrea .. .. .. .. .. .. .. .. .. .. Estonia 1994 14.7 6.8 8.9 .. .. .. 1993 6.0 1.6 32.5 10.0 Ethiopia .. .. .. .. .. . . 1981–82 46.0 12.4 89.0 42.7 Finland .. .. .. .. .. .. .. .. .. .. France .. .. .. .. .. .. .. .. .. .. Georgia .. .. .. .. .. .. .. .. .. .. Germany .. .. .. .. .. .. .. .. .. .. Ghana 1992 34.3 26.7 31.4 .. .. .. .. .. .. .. Greece .. .. .. .. .. .. .. .. .. .. Guatemala .. .. .. .. .. .. 1989 53.3 28.5 76.8 47.6 Guinea .. .. .. .. .. .. 1991 26.3 12.4 50.2 25.6 Haiti 1987 .. . . 65.0 1995 81.0 .. .. .. .. .. .. Honduras 1992 46.0 56.0 50.0 .. .. .. 1992 46.9 20.4 75.7 41.9 Hungary 1993 .. . . 25.3 .. .. .. 1993 <2 .. 10.7 2.1 India 1992 43.5 33.7 40.9 1994 36.7 30.5 35.0 1994 47.0 12.9 87.5 42.9 Indonesia 1987 16.4 20.1 17.4 1990 14.3 16.8 15.1 1996 7.7 0.9 50.4 15.3 Iran, Islamic Rep. .. .. .. .. .. .. .. .. .. .. Ireland .. .. .. .. .. .. .. .. .. .. Israel .. .. .. .. .. .. .. .. .. .. Italy .. .. .. .. .. .. .. .. .. .. Jamaica 1992 .. . . 34.2 .. .. .. 1993 4.3 0.5 24.9 7.5 Japan .. .. .. .. .. .. .. .. .. .. Jordan 1991 .. . . 15.0 .. .. .. 1992 2.5 0.5 23.5 6.3 Kazakhstan 1996 39.0 30.0 34.6 .. .. .. 1993 <2 .. 12.1 2.5 Kenya 1992 46.4 29.3 42.0 .. .. .. 1992 50.2 22.2 78.1 44.4 Korea, Rep. .. .. .. .. .. .. .. .. .. .. Kuwait .. .. .. .. .. .. .. .. .. .. Kyrgyz Republic 1993 48.1 28.7 40.0 .. .. .. 1993 18.9 5.0 55.3 21.4 Lao PDR 1993 53.0 24.0 46.1 .. .. .. .. .. .. .. Latvia .. .. .. .. .. .. 1993 <2 .. <2 .. Lebanon .. .. .. .. .. .. .. .. .. .. Lesotho 1993 53.9 27.8 49.2 .. .. . . 1986–87 48.8 23.8 74.1 43.5 Lithuania .. .. .. .. .. .. 1993 <2 .. 18.9 4.1 Macedonia, FYR .. .. .. .. .. .. .. .. .. .. Madagascar .. .. .. .. .. .. 1993 72.3 33.2 93.2 59.6 Malawi 1990–91 .. . . 54.0 .. .. .. .. .. .. .. Malaysia 1989 .. . . 15.5 .. .. .. 1995 4.3 0.7 22.4 6.8 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      PEOPLE International poverty lines National poverty lines Population Poverty Population Poverty below gap at below gap at Population below the Population below the $1 PPP $1 PPP $2 PPP $2 PPP poverty line (%) poverty line (%) a day a a day a a day a a daya Economy Survey year Rural Urban Total Survey year Rural Urban Total Survey year % % % % Mali .. .. .. .. .. .. .. .. .. .. Mauritania 1990 .. . . 57.0 .. .. .. 1988 31.4 15.2 68.4 33.0 Mexico 1988 .. . . 10.1 .. .. .. 1992 14.9 3.8 40.0 15.9 Moldova .. .. .. .. .. .. 1992 6.8 1.2 30.6 9.7 Mongolia 1995 33.1 38.5 36.3 .. .. .. .. .. .. .. Morocco 1984–85 32.6 17.3 26.0 1990–91 18.0 7.6 13.1 1990–91 <2 .. 19.6 4.6 Mozambique .. .. .. .. .. .. .. .. .. .. Myanmar .. .. .. .. .. .. .. .. .. .. Namibia .. .. .. .. .. .. .. .. .. .. Nepal 1995–96 44.0 23.0 42.0 .. .. .. 1995 50.3 16.2 86.7 44.6 Netherlands .. .. .. .. .. .. .. .. .. .. New Zealand .. .. .. .. .. .. .. .. .. .. Nicaragua 1993 76.1 31.9 50.3 .. .. .. 1993 43.8 18.0 74.5 39.7 Niger 1989–93 66.0 52.0 63.0 .. .. .. 1992 61.5 22.2 92.0 51.8 Nigeria 1985 49.5 31.7 43.0 1992–93 36.4 30.4 34.1 1992–93 31.1 12.9 59.9 29.8 Norway .. .. .. .. .. .. .. .. .. .. Pakistan 1991 36.9 28.0 34.0 .. .. .. 1991 11.6 2.6 57.0 18.6 Panama .. .. .. .. .. .. 1989 25.6 12.6 46.2 24.5 Papua New Guinea .. .. .. .. .. .. .. .. .. .. Paraguay 1991 28.5 19.7 21.8 .. .. .. .. .. .. .. Peru 1994 67.0 46.1 53.5 1997 64.7 40.4 49.0 .. .. .. .. Philippines 1994 53.1 28.0 40.6 1997 51.2 22.5 37.5 1994 26.9 7.1 62.8 27.0 Poland 1993 .. . . 23.8 .. .. .. 1993 6.8 4.7 15.1 7.7 Portugal .. .. .. .. .. .. .. .. .. .. Romania 1994 27.9 20.4 21.5 .. .. .. 1992 17.7 4.2 70.9 24.7 Russian Federation 1994 .. . . 30.9 .. .. .. 1993 <2 .. 10.9 2.3 Rwanda 1993 .. . . 51.2 .. .. . . 1983–85 45.7 11.3 88.7 42.3 Saudi Arabia .. .. .. .. .. .. .. .. .. .. Senegal 1991 40.4 16.4 33.4 .. .. . . 1991–92 54.0 25.5 79.6 47.2 Sierra Leone 1989 76.0 53.0 68.0 .. .. .. .. .. .. .. Singapore .. .. .. .. .. .. .. .. .. .. Slovak Republic .. .. .. .. .. .. 1992 12.8 2.2 85.1 27.5 Slovenia .. .. .. .. .. .. 1993 <2 .. <2 .. South Africa .. .. .. .. .. .. 1993 23.7 6.6 50.2 22.5 Spain .. .. .. .. .. .. .. .. .. .. Sri Lanka 1985–86 45.5 26.8 40.6 1990–91 38.1 28.4 35.3 1990 4.0 0.7 41.2 11.0 Sweden .. .. .. .. .. .. .. .. .. .. Switzerland .. .. .. .. .. .. .. .. .. .. Syrian Arab Republic .. .. .. .. .. .. .. .. .. .. Tajikistan .. .. .. .. .. .. .. .. .. .. Tanzania 1991 .. . . 51.1 .. .. .. .. .. .. .. Thailand 1990 .. . . 18.0 1992 15.5 10.2 13.1 1992 <2 .. 23.5 5.4 Togo 1987–89 .. . . 32.3 .. .. .. .. .. .. .. Tunisia 1985 29.2 12.0 19.9 1990 21.6 8.9 14.1 1990 3.9 0.9 22.7 6.8 Turkey .. .. .. .. .. .. .. .. .. .. Turkmenistan .. .. .. .. .. .. 1993 4.9 0.5 25.8 7.6 Uganda 1993 .. . . 55.0 .. .. . . 1989–90 69.3 29.1 92.2 56.6 Ukraine 1995 .. . . 31.7 .. .. .. 1992 <2 .. <2 .. United Kingdom .. .. .. .. .. .. .. .. .. .. United States .. .. .. .. .. .. .. .. .. .. Uruguay .. .. .. .. .. .. .. .. .. .. Uzbekistan .. .. .. .. .. .. .. .. .. .. Venezuela 1989 .. . . 31.3 .. .. .. 1991 11.8 3.1 32.2 12.2 Vietnam 1993 57.2 25.9 50.9 .. .. .. .. .. .. .. Yemen, Rep. 1992 19.2 18.6 19.1 .. .. .. .. .. .. .. Zambia 1991 88.0 46.0 68.0 1993 .. .. 86.0 1993 84.6 53.8 98.1 73.4 Zimbabwe 1990–91 .. . . 25.5 .. .. . . 1990–91 41.0 14.3 68.2 35.5 a. 1985 prices. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 5. Distribution of income or consumption Percentage share of income or consumption Economy Survey year Gini index Lowest 10% Lowest 20% Second 20% Third 20% Fourth 20% Highest 20% Highest 10% Albania .. .. .. .. .. .. .. .. Algeria 1995a,b 35.3 2.8 7.0 11.6 16.1 22.7 42.6 26.8 Angola .. .. .. .. .. .. .. .. Argentina .. .. .. .. .. .. .. .. Armenia .. .. .. .. .. .. .. .. Australia 1989c,d 33.7 2.5 7.0 12.2 16.6 23.3 40.9 24.8 Austria 1987c,d 23.1 4.4 10.4 14.8 18.5 22.9 33.3 19.3 Azerbaijan .. .. .. .. .. .. .. .. Bangladesh 1992a,b 28.3 4.1 9.4 13.5 17.2 22.0 37.9 23.7 Belarus 1995c,d 28.8 3.4 8.5 13.5 17.7 23.1 37.2 22.6 Belgium 1992c,d 25.0 3.7 9.5 14.6 18.4 23.0 34.5 20.2 Benin .. .. .. .. .. .. .. .. Bolivia 1990 c,d 42.0 2.3 5.6 9.7 14.5 22.0 48.2 31.7 Botswana .. .. .. .. .. .. .. .. Brazil 1995c,d 60.1 0.8 2.5 5.7 9.9 17.7 64.2 47.9 Bulgaria 1992c,d 30.8 3.3 8.3 13.0 17.0 22.3 39.3 24.7 Burkina Faso 1994a,b 48.2 2.2 5.5 8.7 12.0 18.7 55.0 39.5 Burundi .. .. .. .. .. .. .. .. Cambodia .. .. .. .. .. .. .. .. Cameroon .. .. .. .. .. .. .. .. Canada 1994c,d 31.5 2.8 7.5 12.9 17.2 23.0 39.3 23.8 Central African Republic .. .. .. .. .. .. .. .. Chad .. .. .. .. .. .. .. .. Chile 1994c,d 56.5 1.4 3.5 6.6 10.9 18.1 61.0 46.1 China 1995c,d 41.5 2.2 5.5 9.8 14.9 22.3 47.5 30.9 Hong Kong, China .. .. .. .. .. .. .. .. Colombia 1995c,d 57.2 1.0 3.1 6.8 10.9 17.6 61.5 46.9 Congo, Dem. Rep. .. .. .. .. .. .. .. .. Congo, Rep. .. .. .. .. .. .. .. .. Costa Rica 1996c,d 47.0 1.3 4.0 8.8 13.7 21.7 51.8 34.7 Côte d’Ivoire 1988a,b 36.9 2.8 6.8 11.2 15.8 22.2 44.1 28.5 Croatia .. .. .. .. .. .. .. .. Czech Republic 1993 c,d 26.6 4.6 10.5 13.9 16.9 21.3 37.4 23.5 Denmark 1992c,d 24.7 3.6 9.6 14.9 18.3 22.7 34.5 20.5 Dominican Republic 1989c,d 50.5 1.6 4.2 7.9 12.5 19.7 55.7 39.6 Ecuador 1994a,b 46.6 2.3 5.4 8.9 13.2 19.9 52.6 37.6 Egypt, Arab Rep. 1991a,b 32.0 3.9 8.7 12.5 16.3 21.4 41.1 26.7 El Salvador 1995 c,d 49.9 1.2 3.7 8.3 13.1 20.5 54.4 38.3 Eritrea .. .. .. .. .. .. .. .. Estonia 1995c,d 35.4 2.2 6.2 12.0 17.0 23.1 41.8 26.2 Ethiopia 1995a,b 40.0 3.0 7.1 10.9 14.5 19.8 47.7 33.7 Finland 1991c,d 25.6 4.2 10.0 14.2 17.6 22.3 35.8 21.6 France 1989 c,d 32.7 2.5 7.2 12.7 17.1 22.8 40.1 24.9 Georgia .. .. .. .. .. .. .. .. Germany 1989c,d 28.1 3.7 9.0 13.5 17.5 22.9 37.1 22.6 Ghana 1997a,b 32.7 3.6 8.4 12.2 15.8 21.9 41.7 26.1 Greece .. .. .. .. .. .. .. .. Guatemala 1989 c,d 59.6 0.6 2.1 5.8 10.5 18.6 63.0 46.6 Guinea 1994a,b 40.3 2.6 6.4 10.4 14.8 21.2 47.2 32.0 Haiti .. .. .. .. .. .. .. .. Honduras 1996c,d 53.7 1.2 3.4 7.1 11.7 19.7 58.0 42.1 Hungary 1993c,d 27.9 4.1 9.7 13.9 16.9 21.4 38.1 24.0 India 1994a,b 29.7 4.1 9.2 13.0 16.8 21.7 39.3 25.0 Indonesia 1996c,d 36.5 3.6 8.0 11.3 15.1 20.8 44.9 30.3 Iran, Islamic Rep. .. .. .. .. .. .. .. .. Ireland 1987c,d 35.9 2.5 6.7 11.6 16.4 22.4 42.9 27.4 Israel 1992c,d 35.5 2.8 6.9 11.4 16.3 22.9 42.5 26.9 Italy 1991c,d 31.2 2.9 7.6 12.9 17.3 23.2 38.9 23.7 Jamaica 1991a,b 41.1 2.4 5.8 10.2 14.9 21.6 47.5 31.9 Japan .. .. .. .. .. .. .. .. Jordan 1991a,b 43.4 2.4 5.9 9.8 13.9 20.3 50.1 34.7 Kazakhstan 1993c,d 32.7 3.1 7.5 12.3 16.9 22.9 40.4 24.9 Kenya 1994a,b 44.5 1.8 5.0 9.7 14.2 20.9 50.2 34.9 Korea, Rep. .. .. .. .. .. .. .. .. Kuwait .. .. .. .. .. .. .. .. Kyrgyz Republic 1993c,d 35.3 2.7 6.7 11.5 16.4 23.1 42.3 26.2 Lao PDR 1992a,b 30.4 4.2 9.6 12.9 16.3 21.0 40.2 26.4 Latvia 1995c,d 28.5 3.3 8.3 13.8 18.0 22.9 37.0 22.4 Lebanon .. .. .. .. .. .. .. .. Lesotho 1986–87a,b 56.0 0.9 2.8 6.5 11.2 19.4 60.1 43.4 Lithuania 1993c,d 33.6 3.4 8.1 12.3 16.2 21.3 42.1 28.0 Macedonia, FYR .. .. .. .. .. .. .. .. Madagascar 1993 a,b 46.0 1.9 5.1 9.4 13.3 20.1 52.1 36.7 Malawi .. .. .. .. .. .. .. .. Malaysia 1989c,d 48.4 1.9 4.6 8.3 13.0 20.4 53.7 37.9 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      PEOPLE Percentage share of income or consumption Economy Survey year Gini index Lowest 10% Lowest 20% Second 20% Third 20% Fourth 20% Highest 20% Highest 10% Mali 1994a,b 50.5 1.8 4.6 8.0 11.9 19.3 56.2 40.4 Mauritania 1995a,b 38.9 2.3 6.2 10.8 15.4 22.0 45.6 29.9 Mexico 1995 c,d 53.7 1.4 3.6 7.2 11.8 19.2 58.2 42.8 Moldova 1992 c,d 34.4 2.7 6.9 11.9 16.7 23.1 41.5 25.8 Mongolia 1995a,b 33.2 2.9 7.3 12.2 16.6 23.0 40.9 24.5 Morocco 1990–91a,b 39.2 2.8 6.6 10.5 15.0 21.7 46.3 30.5 Mozambique .. .. .. .. .. .. .. .. Myanmar .. .. .. .. .. .. .. .. Namibia .. .. .. .. .. .. .. .. Nepal 1995–96 a,b 36.7 3.2 7.6 11.5 15.1 21.0 44.8 29.8 Netherlands 1991c,d 31.5 2.9 8.0 13.0 16.7 22.5 39.9 24.7 New Zealand .. .. .. .. .. .. .. .. Nicaragua 1993a,b 50.3 1.6 4.2 8.0 12.6 20.0 55.2 39.8 Niger 1995a,b 50.5 0.8 2.6 7.1 13.9 23.1 53.3 35.4 Nigeria 1992–93a,b 45.0 1.3 4.0 8.9 14.4 23.4 49.3 31.4 Norway 1991c,d 25.2 4.1 10.0 14.3 17.9 22.4 35.3 21.2 Pakistan 1996a,b 31.2 4.1 9.4 13.0 16.0 20.3 41.2 27.7 Panama 1995 c,d 57.1 0.7 2.3 6.2 11.3 19.8 60.4 43.8 Papua New Guinea 1996 a,b 50.9 1.7 4.5 7.9 11.9 19.2 56.5 40.5 Paraguay 1995 c,d 59.1 0.7 2.3 5.9 10.7 18.7 62.4 46.6 Peru 1996c,d 46.2 1.6 4.4 9.1 14.1 21.3 51.2 35.4 Philippines 1994a,b 42.9 2.4 5.9 9.6 13.9 21.1 49.6 33.5 Poland 1992a,b 27.2 4.0 9.3 13.8 17.7 22.6 36.6 22.1 Portugal .. .. .. .. .. .. .. .. Romania 1994c,d 28.2 3.7 8.9 13.6 17.6 22.6 37.3 22.7 Russian Federation 1996 a,b 48.0 1.4 4.2 8.8 13.6 20.7 52.8 37.4 Rwanda 1983–85a,b 28.9 4.2 9.7 13.2 16.5 21.6 39.1 24.2 Saudi Arabia .. .. .. .. .. .. .. .. Senegal 1991a,b 53.8 1.0 3.1 7.4 12.1 19.5 57.9 42.3 Sierra Leone 1989a,b 62.9 0.5 1.1 2.0 9.8 23.7 63.4 43.6 Singapore .. .. .. .. .. .. .. .. Slovak Republic 1992c,d 19.5 5.1 11.9 15.8 18.8 22.2 31.4 18.2 Slovenia 1993 c,d 29.2 4.0 9.3 13.3 16.9 21.9 38.6 24.5 South Africa 1993–94a,b 59.3 1.1 2.9 5.5 9.2 17.7 64.8 45.9 Spain 1990 c,d 32.5 2.8 7.5 12.6 17.0 22.6 40.3 25.2 Sri Lanka 1990a,b 30.1 3.8 8.9 13.1 16.9 21.7 39.3 25.2 Sweden 1992c,d 25.0 3.7 9.6 14.5 18.1 23.2 34.5 20.1 Switzerland 1982c,d 36.1 2.9 7.4 11.6 15.6 21.9 43.5 28.6 Syrian Arab Republic .. .. .. .. .. .. .. .. Tajikistan .. .. .. .. .. .. .. .. Tanzania 1993a,b 38.2 2.8 6.8 11.0 15.1 21.6 45.5 30.1 Thailand 1992a,b 46.2 2.5 5.6 8.7 13.0 20.0 52.7 37.1 Togo .. .. .. .. .. .. .. .. Tunisia 1990 a,b 40.2 2.3 5.9 10.4 15.3 22.1 46.3 30.7 Turkey .. .. .. .. .. .. .. .. Turkmenistan 1993c,d 35.8 2.7 6.7 11.4 16.3 22.8 42.8 26.9 Uganda 1992–93a,b 39.2 2.6 6.6 10.9 15.2 21.3 46.1 31.2 Ukraine 1995c,d 47.3 1.4 4.3 9.0 13.8 20.8 52.2 36.8 United Kingdom 1986c,d 32.6 2.4 7.1 12.8 17.2 23.1 39.8 24.7 United States 1994 c,d 40.1 1.5 4.8 10.5 16.0 23.5 45.2 28.5 Uruguay .. .. .. .. .. .. .. .. Uzbekistan .. .. .. .. .. .. .. .. Venezuela 1995c,d 46.8 1.5 4.3 8.8 13.8 21.3 51.8 35.6 Vietnam 1993a,b 35.7 3.5 7.8 11.4 15.4 21.4 44.0 29.0 Yemen, Rep. 1992a,b 39.5 2.3 6.1 10.9 15.3 21.6 46.1 30.8 Zambia 1996a,b 49.8 1.6 4.2 8.2 12.8 20.1 54.8 39.2 Zimbabwe 1990a,b 56.8 1.8 4.0 6.3 10.0 17.4 62.3 46.9 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 6. Education Public expenditure Net enrollment ratioa Percentage of cohort on education % of relevant age group reaching grade 5 Expected years of schooling % of GNP Primary Secondary Males Females Males Females Economy 1980 1996 1980 1996 1980 1996 1980 1996 1980 1996 1980 1995 1980 1995 Albania .. 3.1 .. 102 .. .. .. 81 .. 83 .. .. .. .. Algeria 7.8 5.1 81 94 31 56 90 94 85 95 9 11 6 10 Angola .. .. .. .. .. .. .. .. .. .. 8 .. 7 .. Argentina 2.7 3.5 .. .. .. .. .. .. .. .. .. .. .. .. Armenia .. 2.0 .. .. .. .. .. .. .. .. .. .. .. .. Australia 5.5 5.6 102 95 70 92 .. .. .. .. 12 16 12 16 Austria 5.4 5.6 87 100 .. 88 .. .. .. .. 11 14 11 14 Azerbaijan .. 3.3 .. .. .. .. .. .. .. .. .. .. .. .. Bangladesh 1.5 2.9 .. .. .. .. 18 .. 26 .. .. .. .. .. Belarus .. 6.1 .. 85 .. .. .. .. .. .. .. .. .. .. Belgium 6.0 3.2 97 98 .. 99 .. .. .. .. 14 16 13 15 Benin .. 3.2 .. 63 .. .. 59 64 62 57 .. .. .. .. Bolivia 4.4 5.6 79 .. 16 .. .. .. .. .. 9 .. 8 .. Botswana 6.0 10.4 76 81 14 45 85 87 88 93 7 11 8 11 Brazil 3.6 5.5 80 90 14 20 .. .. .. .. 9 .. 9 .. Bulgaria 4.5 3.3 96 92 73 74 .. .. .. .. 11 12 11 13 Burkina Faso 2.2 1.5 15 31 .. .. 77 74 74 77 2 3 1 2 Burundi 3.4 3.1 20 .. .. .. 100 .. 96 .. 3 5 2 4 Cambodia .. 2.9 .. 98 .. .. .. .. .. .. .. .. .. .. Cameroon 3.6 2.9 .. .. 15 .. 70 .. 70 .. 8 .. 6 .. Canada 6.9 7.0 .. 95 .. 93 .. .. .. .. 15 17 15 18 Central African Republic .. .. 56 .. .. .. 63 .. 50 .. .. .. .. .. Chad .. 2.4 .. 46 .. 6 .. 62 .. 53 .. .. .. .. Chile 4.6 3.1 .. 88 .. 58 94 100 97 100 .. 12 .. 12 China 2.5 2.3 .. 102 .. .. .. 93 .. 94 .. .. .. .. Hong Kong, China 2.4 2.9 95 90 61 71 98 .. 99 .. 12 13 12 13 Colombia 2.4 4.4 .. 85 .. 50 36 70 39 76 .. .. .. .. Congo, Dem. Rep. 2.6 .. .. 54 .. 17 .. .. .. .. .. 7 .. 4 Congo, Rep. 7.0 6.2 96 .. .. .. 81 40 83 78 .. .. .. .. Costa Rica 7.8 5.3 89 91 39 43 77 86 82 89 10 .. 10 .. Côte d’Ivoire 7.2 5.0 .. 55 .. .. 86 77 79 71 .. .. .. .. Croatia .. 5.3 .. 82 .. 66 .. .. .. .. .. 12 .. 12 Czech Republic .. 5.4 .. 91 .. 87 .. .. .. .. .. 13 .. 13 Denmark 6.8 8.2 96 99 88 87 99 100 99 99 14 15 14 15 Dominican Republic 2.2 2.0 .. 81 .. 22 .. .. .. .. .. 11 .. 11 Ecuador 5.6 3.5 .. 97 .. .. .. 84 .. 86 .. .. .. .. Egypt, Arab Rep. 5.7 4.8 .. 93 .. 68 92 .. 88 .. .. 11 .. 9 El Salvador 3.9 2.2 .. 78 .. 21 46 76 48 77 .. 10 .. 10 Eritrea .. 1.8 .. 30 .. 16 .. 73 .. 67 .. 5 .. 4 Estonia .. 7.3 .. 87 .. 83 .. 96 .. 97 .. 12 .. 13 Ethiopia 3.1 4.0 .. 28 .. .. .. 57 .. 53 .. .. .. .. Finland 5.3 7.6 .. 99 .. 93 .. 100 .. 100 .. 15 .. 16 France 5.0 6.1 100 100 79 94 .. .. .. .. 13 15 13 16 Georgia .. 5.2 .. 87 .. 71 .. .. .. .. .. 10 .. 10 Germany .. 4.8 .. 100 .. 87 .. .. .. .. .. 15 .. 15 Ghana 3.1 .. .. .. .. .. .. .. .. .. .. .. .. .. Greece 2.0 3.0 96 90 .. 87 99 .. 98 .. 12 14 12 14 Guatemala 1.8 1.7 59 .. 13 .. .. 52 .. 47 .. .. .. .. Guinea .. .. .. 37 .. .. 59 .. 41 .. .. .. .. .. Haiti 1.5 .. .. .. .. .. 33 .. 34 .. .. .. .. .. Honduras 3.2 3.6 78 90 .. .. .. .. .. .. .. .. .. .. Hungary 4.7 4.7 95 97 .. 87 96 .. 97 .. 9 12 10 13 India 3.0 3.4 .. .. .. .. .. .. .. .. .. .. .. .. Indonesia 1.7 1.4 88 97 .. 42 .. .. .. .. .. 10 .. 10 Iran, Islamic Rep. 7.5 4.0 .. 90 .. 69 .. .. .. .. .. .. .. .. Ireland 6.3 5.8 100 100 78 86 .. .. .. .. 11 14 11 14 Israel 7.9 7.2 .. .. .. .. .. .. .. .. .. .. .. .. Italy .. 4.7 .. 100 .. .. 99 100 99 100 .. .. .. .. Jamaica 7.0 7.4 96 .. 64 .. 91 .. 91 .. .. 11 .. 11 Japan 5.8 3.6 101 103 93 98 100 .. 100 .. 13 14 12 14 Jordan 6.6 7.3 .. .. .. .. 92 .. 94 .. 12 .. 12 .. Kazakhstan .. 4.7 .. .. .. .. .. .. .. .. .. .. .. .. Kenya 6.8 6.6 91 .. .. .. 60 .. 62 .. .. .. .. .. Korea, Rep. 3.7 3.7 104 92 70 97 94 100 94 100 12 15 11 14 Kuwait 2.4 5.7 85 54 .. 54 .. .. .. .. .. 9 .. 9 Kyrgyz Republic .. 5.7 .. 95 .. .. .. .. .. .. .. .. .. .. Lao PDR .. 2.5 .. 72 .. 18 .. .. .. .. .. 8 .. 6 Latvia 3.3 6.5 .. 90 .. 79 .. .. .. .. .. 11 .. 12 Lebanon .. 2.5 .. 76 .. .. .. .. .. .. .. .. .. .. Lesotho 5.1 7.0 67 70 13 17 50 .. 68 .. .. 8 .. 9 Lithuania .. 5.6 .. .. .. 80 .. .. .. .. .. .. .. .. Macedonia, FYR .. 5.6 .. 95 .. 51 .. 95 .. 95 .. 10 .. 10 Madagascar 4.4 1.9 .. 61 .. .. .. 49 .. 33 .. .. .. .. Malawi 3.4 5.5 43 68 .. .. 48 .. 40 .. .. .. .. .. Malaysia 6.0 5.2 .. 102 .. .. 97 .. 97 .. .. .. .. .. 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      PEOPLE Public expenditure Net enrollment ratioa Percentage of cohort on education % of relevant age group reaching grade 5 Expected years of schooling % of GNP Primary Secondary Males Females Males Females Economy 1980 1996 1980 1996 1980 1996 1980 1996 1980 1996 1980 1995 1980 1995 Mali 3.7 2.2 20 28 .. .. 48 87 42 82 .. .. .. .. Mauritania .. 5.1 .. 57 .. .. .. 61 .. 68 .. .. .. .. Mexico 4.7 4.9 .. 101 .. 51 .. 85 .. 86 .. .. .. .. Moldova .. 9.7 .. .. .. .. .. .. .. .. .. .. .. .. Mongolia .. 6.4 .. 81 .. 53 .. .. .. .. .. 6 .. 8 Morocco 6.1 5.3 62 74 20 .. 79 79 78 77 .. .. .. .. Mozambique 4.4 .. .. 40 .. 6 .. 52 .. 39 5 4 4 3 Myanmar 1.7 1.2 .. .. .. .. .. .. .. .. .. .. .. .. Namibia 1.3 9.1 .. 91 .. 36 .. .. .. .. .. .. .. .. Nepal 1.8 2.8 .. .. .. .. .. .. .. .. .. .. .. .. Netherlands 7.6 5.2 93 99 81 91 94 .. 98 .. 14 16 13 15 New Zealand 5.8 7.3 .. 100 81 97 94 .. 94 .. 14 16 13 17 Nicaragua 3.4 3.7 70 78 23 27 40 .. 47 .. 8 9 9 9 Niger 3.1 .. 21 25 4 6 74 72 72 74 .. .. .. .. Nigeria 6.4 0.9 .. .. .. .. .. .. .. .. .. .. .. .. Norway 6.5 7.5 98 99 84 96 100 100 100 100 13 15 13 15 Pakistan 2.0 3.0 .. .. .. .. .. .. .. .. .. .. .. .. Panama 4.9 4.6 89 .. 46 .. 74 .. 79 .. 11 .. 11 .. Papua New Guinea .. .. .. .. .. .. .. .. .. .. .. .. .. .. Paraguay 1.5 3.9 89 91 .. 38 59 .. 58 .. .. 9 .. 9 Peru 3.1 2.9 86 91 .. 53 78 .. 74 .. 11 13 10 12 Philippines 1.7 2.2 94 101 45 60 68 .. 73 .. 10 11 11 11 Poland .. 5.2 98 95 71 85 .. .. .. .. 12 13 12 13 Portugal 3.8 5.5 99 104 .. 78 .. .. .. .. .. 14 .. 15 Romania 3.3 3.6 .. 95 .. 73 .. .. .. .. .. 12 .. 11 Russian Federation 3.5 4.1 .. 93 .. .. .. .. .. .. .. .. .. .. Rwanda 2.7 .. 59 .. .. .. 55 .. 59 .. .. .. .. .. Saudi Arabia 4.1 5.5 49 61 21 42 82 87 86 92 7 9 5 8 Senegal .. 3.5 37 58 .. .. 89 89 82 81 .. .. .. .. Sierra Leone 3.5 .. .. .. .. .. .. .. .. .. .. .. .. .. Singapore 2.8 3.0 99 .. .. .. 100 .. 100 .. 11 .. 11 .. Slovak Republic .. 4.9 .. .. .. .. .. .. .. .. .. .. .. .. Slovenia .. 5.8 .. 95 .. .. .. .. .. .. .. .. .. .. South Africa .. 7.9 .. .. .. 51 .. 72 .. 79 .. 13 .. 13 Spain 2.3 4.9 102 105 74 .. 95 .. 94 .. 13 15 12 16 Sri Lanka 2.7 3.4 .. .. .. .. 92 83 91 84 .. .. .. .. Sweden 9.0 8.3 .. 102 .. 98 98 98 99 97 12 14 13 15 Switzerland 4.8 5.3 .. .. .. .. .. .. .. .. 14 15 13 14 Syrian Arab Republic 4.6 4.2 90 91 39 38 93 93 88 94 11 10 8 9 Tajikistan .. 2.2 .. .. .. .. .. .. .. .. .. .. .. .. Tanzania .. .. 68 48 .. .. 89 .. 90 .. .. .. .. .. Thailand 3.4 4.1 .. .. .. .. .. .. .. .. .. .. .. .. Togo 5.6 4.7 .. 85 .. .. 59 .. 45 .. .. .. .. .. Tunisia 5.4 6.7 82 98 23 .. 89 90 84 92 10 .. 7 .. Turkey 2.2 2.2 .. 96 .. 50 .. .. .. .. .. 11 .. 9 Turkmenistan .. .. .. .. .. .. .. .. .. .. .. .. .. .. Uganda 1.2 2.6 .. .. .. .. 82 .. 73 .. .. .. .. .. Ukraine 5.6 7.2 .. .. .. .. .. .. .. .. .. .. .. .. United Kingdom 5.6 5.4 100 100 79 92 .. .. .. .. 13 16 13 17 United States 6.7 5.4 .. 95 .. 90 .. .. .. .. 14 15 15 16 Uruguay 2.3 3.3 .. 93 .. .. .. 97 .. 99 .. .. .. .. Uzbekistan .. 8.1 .. .. .. .. .. .. .. .. .. .. .. .. Venezuela 4.4 5.2 82 84 14 22 .. 86 .. 92 .. 10 .. 11 Vietnam .. 2.6 95 .. .. .. .. .. .. .. .. .. .. .. Yemen, Rep. .. 6.5 .. 52 .. .. .. .. .. .. .. .. .. .. Zambia 4.5 2.2 77 75 .. 17 88 .. 82 .. .. 8 .. 7 Zimbabwe 6.6 8.3 .. .. .. .. 82 78 76 79 .. .. .. .. World 4.0 m 4.8 m .. ..w .. ..w ..w ..w ..w ..w Low income 3.2 3.9 .. .. .. .. .. .. .. .. Excl. China & India 3.4 .. .. .. .. .. .. .. .. .. Middle income 4.0 5.1 .. .. .. .. .. .. .. .. Lower middle income 4.2 5.3 .. .. .. .. .. .. .. .. Upper middle income 4.0 5.0 .. 94 .. 43 .. .. .. .. Low and middle income 3.5 4.1 .. .. .. .. .. .. .. .. East Asia & Pacific 2.5 2.3 .. 101 .. .. .. 93 .. 94 Europe & Central Asia .. 5.4 .. 92 .. .. .. .. .. .. Latin America & Carib. 3.8 3.7 .. 91 .. 33 .. .. .. .. Middle East & N. Africa 5.0 5.3 .. 85 .. 61 88 .. 84 .. South Asia 2.0 3.0 .. .. .. .. .. .. .. .. Sub-Saharan Africa 4.1 4.3 .. .. .. .. .. .. .. .. High income 5.6 5.4 .. 97 .. 90 .. .. .. .. a. Net enrollment ratios exceeding 100 indicate discrepancies between estimates of the school-age population and reported enrollment data. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 7. Health Access to Access to Infant Contraceptive Total Maternal Public expenditure safe water sanitation mortality rate prevalence rate fertility rate mortality rate on health % of population % of population Per 1,000 % of women Births Per 100,000 % of GDP with access with access live births aged 15–49 per woman live births Economy 1990–97a 1982 1995 1982 1995 1980 1997 1990–98a 1980 1997 1990–97a Albania 2.5 92 76 .. 58 47 26 .. 3.6 2.5 28b Algeria 3.3 77 .. .. .. 98 32 51 6.7 3.6 140b Angola 3.9 28 32 18 15 154 125 .. 6.9 6.8 1,500c Argentina 4.3 55 65 69 75 35 22 .. 3.3 2.6 100c Armenia 3.1 .. .. .. .. 26 15 .. 2.3 1.5 21b Australia 5.8 99 99 99 .. 11 5 .. 1.9 1.8 9c Austria 5.7 99 .. .. .. 14 5 .. 1.6 1.4 10c Azerbaijan 1.1 .. .. .. 36 30 20 .. 3.2 2.1 44b Bangladesh 1.2 40 84 4 35 132 75 49 6.1 3.2 850c Belarus 5.2 100 .. .. .. 16 12 .. 2.0 1.2 22b Belgium 6.7 98 .. .. .. 12 6 .. 1.7 1.6 10c Benin 1.7 14 72 10 24 116 88 16 7.0 5.8 500d Bolivia 3.8 53 70 36 41 118 66 45 5.5 4.4 370d Botswana 1.8 77 70 36 55 71 58 .. 6.1 4.3 250c Brazil 1.9 75 69 24 67 70 34 77 3.9 2.3 160d Bulgaria 3.5 85 .. .. .. 20 18 .. 2.0 1.1 20b Burkina Faso 4.7 35 .. 5 .. 121 99 8 7.5 6.6 930c Burundi 1.0 23 58 52 48 122 119 .. 6.8 6.3 1,300c Cambodia 0.7 .. 13 .. .. 201 103 .. 4.7 4.6 900c Cameroon 1.0 36 41 36 40 94 52 16 6.4 5.3 550c Canada 6.3 97 99 60 95 10 6 .. 1.7 1.6 6c Central African Republic 2.0 16 23 19 45 117 98 14 5.8 4.9 700d Chad 1.6 31 24 14 21 123 100 4 6.9 6.5 840d Chile 2.3 86 91 67 81 32 11 .. 2.8 2.4 65b China 2.1 .. 83 .. .. 42 32 85 2.5 1.9 95e Hong Kong, China 2.3 .. .. .. .. 11 5 .. 2.0 1.3 7c Colombia 2.9 91 75 68 59 41 24 72 3.9 2.8 100c Congo, Dem. Rep. 0.2 .. .. .. .. 112 92 .. 6.6 6.4 870c Congo, Rep. 1.8 .. .. 40 9 89 90 .. 6.3 6.1 890c Costa Rica 6.0 93 100 95 97 20 12 .. 3.7 2.8 55c Côte d’Ivoire 1.4 20 72 17 51 108 87 11 7.4 5.1 810d Croatia 8.4 70 63 67 61 21 9 .. .. 1.6 12b Czech Republic 6.4 100 .. .. .. 16 6 69 2.1 1.2 2b Denmark 5.1 .. .. .. 100 8 6 .. 1.5 1.8 9c Dominican Republic 1.8 49 73 66 80 76 40 64 4.2 3.0 110c Ecuador 2.0 58 55 57 53 74 33 57 5.0 3.0 150c Egypt, Arab Rep. 1.7 90 84 70 70 120 51 48 5.1 3.2 170c El Salvador 2.4 51 53 62 77 84 32 53 4.9 3.2 300c Eritrea 1.1 .. 7 .. .. 91 62 8 .. 5.8 1,000c Estonia 5.8 .. .. .. .. 17 10 .. 2.0 1.2 52b Ethiopia 1.6 4 26 .. 8 155 107 4 6.6 6.5 1,400c Finland 5.7 95 98 100 100 8 4 .. 1.6 1.9 11c France 7.7 98 100 .. .. 10 5 .. 1.9 1.7 15c Georgia 0.6 .. .. .. .. 25 17 .. 2.3 1.5 19b Germany 8.1 90 .. .. .. 12 5 .. 1.4 1.4 22c Ghana 2.9 .. 65 26 32 94 66 20 6.5 4.9 740c Greece 5.3 85 .. .. .. 18 7 .. 2.2 1.3 10c Guatemala 1.7 58 67 54 67 84 43 32 6.3 4.5 190d Guinea 1.2 20 55 12 14 185 120 2 6.1 5.5 880d Haiti 1.2 38 39 19 26 123 71 18 5.9 4.4 600d Honduras 2.8 50 77 32 82 70 36 50 6.5 4.3 220c Hungary 4.5 87 .. .. .. 23 10 .. 1.9 1.4 14b India 0.7 54 85 8 16 115 71 41 5.0 3.3 440d Indonesia 0.7 39 65 30 55 90 47 57 4.3 2.8 390d Iran, Islamic Rep. 1.7 50 90 60 81 87 32 73 6.7 2.8 120c Ireland 5.1 97 .. .. .. 11 5 60 3.2 1.9 10c Israel 0.3 100 99 .. 100 15 7 .. 3.2 2.7 7c Italy 5.3 99 .. .. .. 15 5 .. 1.6 1.2 12c Jamaica 2.5 96 93 91 74 21 12 65 3.7 2.7 120c Japan 5.7 99 96 99 100 8 4 .. 1.8 1.4 18b Jordan 3.7 89 98 76 98 41 29 53 6.8 4.2 150c Kazakhstan 2.5 .. .. .. .. 33 24 59 2.9 2.0 53b Kenya 1.9 27 45 44 45 75 74 38 7.8 4.7 650c Korea, Rep. 2.3 83 83 100 100 26 9 .. 2.6 1.7 30b Kuwait 3.5 100 100 100 100 27 12 .. 5.3 2.9 20b Kyrgyz Republic 2.9 .. 81 .. .. 43 28 60 4.1 2.8 32b Lao PDR 1.3 .. 51 .. 32 127 98 .. 6.7 5.6 660c Latvia 3.5 .. .. .. .. 20 15 .. 2.0 1.1 15b Lebanon 3.0 92 94 59 97 48 28 .. 4.0 2.5 300c Lesotho 3.7 18 62 12 .. 119 93 23 5.5 4.8 610c Lithuania 5.0 .. .. .. .. 20 10 .. 2.0 1.4 13b Macedonia, FYR 6.2 .. .. .. .. 54 16 .. 2.5 1.9 22b Madagascar 1.4 31 16 .. 34 119 94 19 6.6 5.8 500d Malawi 2.3 32 60 60 64 169 133 22 7.6 6.4 620d Malaysia 1.4 71 89 75 94 30 11 .. 4.2 3.2 34b 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      PEOPLE Access to Access to Infant Contraceptive Total Maternal Public expenditure safe water sanitation mortality rate prevalence rate fertility rate mortality rate on health % of population % of population Per 1,000 % of women Births Per 100,000 % of GDP with access with access live births aged 15–49 per woman live births Economy 1990–97a 1982 1995 1982 1995 1980 1997 1990–98a 1980 1997 1990–97a Mali 2.0 .. 48 21 37 184 118 7 7.1 6.6 580d Mauritania 1.8 37 64 .. 32 120 92 .. 6.3 5.5 800c Mexico 2.8 82 95 57 76 51 31 .. 4.7 2.8 110c Moldova 6.2 .. 56 .. 50 35 20 74 2.4 1.6 23b Mongolia 4.3 100 54 50 .. 82 52 .. 5.3 2.6 65c Morocco 1.2 32 57 50 68 99 51 50 5.4 3.1 370 f Mozambique 4.6 9 24 10 23 145 135 6 6.5 5.3 1,100c Myanmar 0.4 20 60 20 43 109 79 .. 4.9 2.4 580c Namibia 4.1 .. 60 .. 42 90 65 29 5.9 4.9 220d Nepal 1.2 11 59 0 23 132 83 .. 6.1 4.4 1,500c Netherlands 6.2 100 99 .. 100 9 5 .. 1.6 1.5 12c New Zealand 5.9 87 90 88 .. 13 7 .. 2.0 1.9 25c Nicaragua 5.3 50 62 27 59 84 43 44 6.3 3.9 160c Niger 1.6 37 48 9 17 150 118 8 7.4 7.4 590d Nigeria 0.2 36 50 .. 57 99 77 6 6.9 5.3 1,000c Norway 6.2 99 100 .. 100 8 4 .. 1.7 1.9 6c Pakistan 0.8 38 62 16 39 127 95 24 7.0 5.0 340c Panama 4.7 82 84 81 90 32 21 .. 3.7 2.6 55c Papua New Guinea 2.8 .. 31 .. 25 78 61 26 5.8 4.3 370c Paraguay 1.8 23 39 49 32 50 23 51 5.2 3.8 190d Peru 2.2 53 66 48 61 81 40 64 4.5 3.2 280c Philippines 1.3 65 83 57 77 52 35 48 4.8 3.6 210d Poland 4.8 82 .. .. .. 26 10 .. 2.3 1.5 5b Portugal 4.9 66 82 .. .. 24 6 .. 2.2 1.4 15c Romania 2.9 77 62 .. 44 29 22 57 2.4 1.3 41b Russian Federation 4.1 .. .. .. .. 22 17 34 1.9 1.3 53b Rwanda 1.9 .. .. .. 94 128 124 21 8.3 6.2 1,300c Saudi Arabia 6.4 91 93 76 86 65 21 .. 7.3 5.9 18b Senegal 1.2 44 50 .. .. 117 70 13 6.8 5.6 510d Sierra Leone 1.6 24 34 13 .. 190 170 .. 6.5 6.1 .. Singapore 1.5 100 100 85 100 12 4 .. 1.7 1.7 10c Slovak Republic 6.1 .. .. 43 51 21 9 .. 2.3 1.4 8b Slovenia 7.1 .. 98 80 98 15 5 .. 2.1 1.3 5b South Africa 3.6 .. 59 .. 53 67 48 69 4.6 2.8 230c Spain 5.8 99 .. .. 97 12 5 .. 2.2 1.1 7c Sri Lanka 1.4 37 70 66 75 34 14 .. 3.5 2.2 30b Sweden 7.2 100 .. .. .. 7 4 .. 1.7 1.7 7c Switzerland 7.1 100 100 .. .. 9 5 .. 1.5 1.5 6c Syrian Arab Republic .. 71 88 45 71 56 31 40 7.4 4.0 180b Tajikistan 2.4 .. 69 .. 62 58 30 .. 5.6 3.5 58b Tanzania 1.1 52 49 .. 86 108 85 18 6.7 5.5 530d Thailand 2.0 66 89 47 96 49 33 .. 3.5 1.7 200c Togo 1.6 35 55 14 41 110 86 .. 6.6 6.1 640c Tunisia 3.0 72 90 46 80 69 30 60 5.2 2.8 170c Turkey 2.7 69 .. .. .. 109 40 .. 4.3 2.5 180c Turkmenistan 1.2 .. 60 .. 60 54 40 .. 4.9 3.0 44b Uganda 1.9 16 42 13 67 116 99 15 7.2 6.6 550 f Ukraine 3.9 .. 55 .. 49 17 14 .. 2.0 1.3 30b United Kingdom 5.7 100 100 .. 96 12 6 .. 1.9 1.7 9b United States 6.6 100 73 98 .. 13 7 76 1.8 2.0 12c Uruguay 1.9 83 89 59 61 37 16 .. 2.7 2.4 85c Uzbekistan 3.3 .. 57 .. 18 47 .. .. 4.8 3.3 24b Venezuela 1.0 84 79 45 72 36 21 .. 4.2 3.0 120b Vietnam 1.1 .. 47 30 60 57 29 75 5.0 2.4 105b Yemen, Rep. 1.3 .. 39 .. 19 141 96 21 7.9 6.4 1,400c Zambia 2.9 48 53 47 51 90 113 26 7.0 5.6 650d Zimbabwe 1.7 10 77 5 66 80 69 48 6.4 3.8 280d World 2.5 w ..w 75 w ..w ..w 80 w 56 w 3.7 w 2.8 w Low income 1.0 .. 74 .. .. 98 69 4.3 3.2 Excl. China & India .. 37 55 21 45 114 84 6.0 4.4 Middle income 2.4 74 .. .. .. 59 33 3.7 2.5 Lower middle income 2.2 .. .. .. .. 61 38 3.6 2.5 Upper middle income 3.0 78 79 52 75 57 27 3.7 2.4 Low and middle income 1.8 .. 75 .. .. 87 60 4.1 2.9 East Asia & Pacific 1.8 .. 77 .. .. 55 37 3.0 2.1 Europe & Central Asia 3.9 .. .. .. .. 41 23 2.5 1.7 Latin America & Carib. 2.6 73 75 46 68 60 32 4.1 2.7 Middle East & N. Africa 2.3 69 .. 62 .. 95 49 6.2 3.7 South Asia 0.8 50 81 9 20 119 77 5.3 3.5 Sub-Saharan Africa 1.7 .. 47 .. 47 115 91 6.6 5.5 High income 6.0 98 .. .. .. 12 6 1.8 1.7 a. Data are for the most recent year available within the period. b. Official estimate. c. UNICEF-WHO estimate based on statistical modeling. d. Indirect estimate based on a sample survey. e. Based on a survey covering 30 provinces. f. Based on a sample survey. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 8. Land use and agricultural productivity Agricultural Agricultural productivity Land under machinery Agr. value added permanent Tractors per agricultural crops Irrigated land Arable land per thousand worker Food production index % of land area % of cropland Hectares per capita agricultural workers 1995 dollars 1989–91 = 100 Economy 1980 1996 1979–81 1994–96 1979–81 1994–96 1979–81 1994–96 1979–81 1995–97 1979–81 1995–97 Albania 4.3 4.6 53.0 48.4 0.22 0.18 15 10 1,193 1,717 .. .. Algeria 0.3 0.2 3.4 6.9 0.37 0.27 27 43 1,411 1,903 69.7 118.2 Angola 0.4 0.4 2.2 2.1 0.41 0.27 4 3 .. 117 91.9 130.1 Argentina 0.8 0.8 5.8 6.3 0.89 0.72 132 190 12,195 13,833 94.9 121.9 Armenia .. 3.5 .. 43.7 .. 0.15 .. 64 .. 4,477 .. 82.3 Australia 0.0 0.0 3.5 4.8 2.97 2.68 751 698 20,880 29,044 91.5 126.9 Austria 1.2 1.1 0.2 0.3 0.20 0.18 945 1,492 9,761 15,474 92.3 100.0 Azerbaijan .. 4.6 .. 50.0 .. 0.21 .. 31 .. 847 .. 55.6 Bangladesh 2.0 2.5 17.1 39.1 0.10 0.07 0 0 181 221 79.2 106.0 Belarus .. 0.7 .. 1.9 .. 0.60 .. 131 .. 3,461 .. 58.9 Belgium 0.4 0.5 0.1 0.1 .. .. 917 1,130 .. .. 88.4 114.4 Benin 4.0 4.1 0.3 0.5 0.39 0.26 0 0 302 504 63.4 129.5 Bolivia 0.2 0.2 6.6 3.7 0.35 0.27 4 4 .. .. 71.0 126.7 Botswana 0.0 0.0 0.5 0.3 0.44 0.25 9 21 619 647 87.6 104.2 Brazil 1.2 1.4 3.3 4.9 0.32 0.32 31 51 2,047 3,931 69.5 122.2 Bulgaria 3.2 1.8 28.3 18.7 0.43 0.48 66 61 2,754 4,351 105.3 68.3 Burkina Faso 0.1 0.1 0.4 0.7 0.39 0.34 0 0 134 159 62.6 122.4 Burundi 10.1 12.9 0.7 1.3 0.22 0.13 0 0 177 139 80.5 96.4 Cambodia 0.4 0.6 4.9 4.5 0.30 0.37 0 0 .. 407 51.1 124.8 Cameroon 2.2 2.3 0.2 0.3 0.68 0.45 0 0 834 958 83.2 118.7 Canada 0.0 0.0 1.3 1.6 1.86 1.53 824 1,683 .. .. 79.9 112.7 Central African Republic 0.1 0.1 .. .. 0.81 0.59 0 0 396 439 79.9 122.7 Chad 0.0 0.0 0.2 0.4 0.70 0.48 0 0 155 212 90.6 117.5 Chile 0.3 0.4 29.6 32.6 0.36 0.25 43 44 2,612 5,211 71.5 128.7 China 0.4 1.2 45.1 37.0 0.10 0.10 2 1 162 296 61.0 155.8 Hong Kong, China 1.0 1.0 43.8 28.6 0.00 0.00 0 0 .. .. 97.4 56.7 Colombia 1.4 2.4 7.7 23.4 0.13 0.05 8 7 1,926 2,890 76.0 110.8 Congo, Dem. Rep. 0.3 0.4 0.1 0.1 0.25 0.16 0 0 270 285 71.9 104.9 Congo, Rep. 0.1 0.1 0.7 0.6 0.07 0.05 2 1 391 470 80.3 114.5 Costa Rica 4.4 4.8 12.1 23.8 0.12 0.08 22 23 3,159 4,627 73.0 128.4 Côte d’Ivoire 7.2 13.5 1.0 1.0 0.24 0.21 1 1 1,074 1,005 70.9 119.2 Croatia .. 2.2 .. 0.2 .. 0.24 .. 14 .. 7,144 .. 57.7 Czech Republic .. 3.1 .. 0.7 .. 0.30 .. 148 .. .. .. 81.9 Denmark 0.3 0.2 14.5 20.3 0.52 0.45 973 1,088 21,321 46,621 83.2 102.5 Dominican Republic 7.2 11.4 11.7 13.7 0.19 0.17 3 3 1,839 2,454 85.1 109.1 Ecuador 3.3 5.2 19.4 8.1 0.20 0.14 6 7 1,206 1,764 76.6 136.9 Egypt, Arab Rep. 0.2 0.5 100.0 100.0 0.06 0.05 4 10 721 1,163 68.4 129.8 El Salvador 8.0 10.5 14.8 14.2 0.12 0.11 5 5 2,013 1,705 90.8 109.5 Eritrea .. 0.8 .. 5.4 .. 0.12 .. 1 .. .. .. 102.3 Estonia .. 0.4 .. .. .. 0.76 .. 475 .. 3,342 .. 49.3 Ethiopia .. 0.6 .. 1.6 .. 0.20 .. 0 .. .. 90.2 .. Finland .. .. .. .. 0.54 0.49 721 1,301 16,995 28,296 92.8 92.4 France 2.5 2.1 4.6 8.2 0.32 0.31 737 1,189 14,956 34,760 93.7 103.6 Georgia .. 4.7 .. 42.2 .. 0.14 .. 28 .. 1,838 .. 74.6 Germany 1.4 0.7 3.7 3.9 0.15 0.14 624 954 .. 19,930 91.0 90.9 Ghana 7.5 7.5 0.2 0.1 0.18 0.16 1 1 663 533 73.5 147.7 Greece 7.9 8.4 24.2 33.8 0.30 0.28 120 267 8,804 12,611 91.2 98.4 Guatemala 4.4 5.1 5.0 6.5 0.19 0.14 3 2 2,110 1,902 69.9 114.0 Guinea 0.9 1.2 12.8 10.9 0.11 0.09 0 0 .. 262 96.5 129.2 Haiti 12.5 12.7 7.9 9.7 0.10 0.08 0 0 578 407 105.5 90.5 Honduras 1.8 3.1 4.1 3.6 0.44 0.30 5 7 697 1,018 88.2 104.7 Hungary 3.3 2.4 3.6 4.2 0.47 0.47 59 54 3,389 4,655 91.0 76.8 India 1.8 2.4 22.8 32.0 0.24 0.17 2 5 253 343 68.4 117.1 Indonesia 4.4 7.2 16.2 15.0 0.12 0.09 0 1 610 745 63.5 122.4 Iran, Islamic Rep. 0.5 1.0 35.5 38.0 0.36 0.30 17 39 2,533 3,831 60.9 136.8 Ireland 0.0 0.0 .. .. 0.33 0.37 606 978 .. .. 83.3 106.2 Israel 4.3 4.2 49.3 45.3 0.08 0.06 294 336 .. .. 85.7 114.1 Italy 10.0 9.1 19.3 24.9 0.17 0.14 370 867 9,994 19,001 101.5 99.7 Jamaica 5.5 6.1 13.6 14.0 0.08 0.07 9 11 892 1,294 86.0 117.9 Japan 1.6 1.0 62.6 62.7 0.04 0.03 209 593 15,698 28,665 94.2 96.9 Jordan 0.4 1.0 11.0 18.2 0.14 0.08 48 42 1,447 1,634 55.3 157.3 Kazakhstan .. 0.1 .. 6.9 .. 2.04 .. 106 .. 1,477 .. 68.5 Kenya 0.8 0.9 0.9 1.5 0.23 0.15 1 1 262 230 67.7 102.9 Korea, Rep. 1.4 2.0 59.6 60.7 0.05 0.04 1 34 3,957 10,962 77.9 119.1 Kuwait .. .. .. .. 0.00 0.00 3 14 .. .. 98.9 139.3 Kyrgyz Republic .. 2.7 .. 76.8 .. 0.23 .. 44 .. 2,917 .. 123.8 Lao PDR 0.1 0.2 15.4 20.3 0.21 0.17 0 0 .. 526 71.2 112.4 Latvia .. 0.5 .. .. .. 0.68 .. 284 .. 3,125 .. 49.8 Lebanon 8.9 12.5 28.3 28.4 0.07 0.05 28 77 .. .. 57.8 117.6 Lesotho .. .. .. .. 0.22 0.17 6 6 498 319 89.4 104.4 Lithuania .. 0.9 .. .. .. 0.79 .. 239 .. 2,907 .. 69.8 Macedonia, FYR .. 1.9 .. 9.4 .. 0.31 .. 323 .. 1,528 .. 95.9 Madagascar 0.9 0.9 21.5 35.0 0.28 0.19 1 1 198 180 82.1 105.3 Malawi 0.9 1.1 1.3 1.6 0.20 0.16 0 0 100 122 91.2 105.3 Malaysia 11.6 17.6 6.7 4.5 0.07 0.09 4 23 3,279 6,267 55.4 124.0 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      ENVIRONMENT Agricultural Agricultural productivity Land under machinery Agr. value added permanent Tractors per agricultural crops Irrigated land Arable land per thousand worker Food production index % of land area % of cropland Hectares per capita agricultural workers 1995 dollars 1989–91 = 100 Economy 1980 1996 1979–81 1994–96 1979–81 1994–96 1979–81 1994–96 1979–81 1995–97 1979–81 1995–97 Mali 0.0 0.0 2.9 2.3 0.31 0.37 0 1 225 241 79.7 118.7 Mauritania 0.0 0.0 22.8 10.3 0.14 0.20 1 1 301 439 86.1 103.2 Mexico 0.8 1.1 20.3 23.1 0.34 0.27 16 20 1,482 1,690 84.9 120.6 Moldova .. 12.5 .. 14.1 .. 0.41 .. 82 .. 1,473 .. 58.3 Mongolia 0.0 0.0 3.0 6.1 0.71 0.54 32 22 727 1,085 88.2 81.6 Morocco 1.1 1.9 15.2 13.0 0.38 0.33 7 10 1,117 1,593 55.9 94.9 Mozambique 0.3 0.3 2.1 3.4 0.24 0.19 1 1 .. 76 99.2 119.5 Myanmar 0.7 0.9 10.4 15.9 0.28 0.22 1 1 .. .. 87.8 133.5 Namibia 0.0 0.0 0.6 0.8 0.64 0.51 10 11 876 1,235 107.4 118.8 Nepal 0.2 0.4 22.5 30.6 0.16 0.13 0 0 162 187 65.1 113.5 Netherlands 0.9 1.0 58.5 61.5 0.06 0.06 561 646 21,663 43,836 87.0 106.1 New Zealand 3.7 6.4 5.2 8.9 0.80 0.43 619 451 .. .. 90.8 120.3 Nicaragua 1.5 2.4 6.0 3.3 0.39 0.54 6 7 1,334 1,407 117.9 123.7 Niger 0.0 0.0 0.7 1.4 0.62 0.53 0 0 222 190 101.4 118.4 Nigeria 2.8 2.8 0.7 0.7 0.39 0.27 1 1 370 541 57.7 134.2 Norway .. .. .. .. 0.20 0.22 824 1,251 17,044 31,577 91.8 99.7 Pakistan 0.4 0.7 72.7 80.2 0.24 0.17 5 12 392 585 66.4 130.5 Panama 1.6 2.1 5.0 4.9 0.22 0.19 27 20 2,122 2,463 85.6 102.5 Papua New Guinea 0.9 1.1 .. .. 0.01 0.01 1 1 717 827 86.1 106.8 Paraguay 0.3 0.2 3.4 3.0 0.52 0.45 14 25 2,506 3,295 60.6 116.7 Peru 0.3 0.4 32.8 41.8 0.19 0.16 5 3 1,349 1,619 78.4 131.5 Philippines 14.8 14.4 14.0 16.7 0.09 0.07 1 1 1,348 1,379 86.4 120.6 Poland 1.1 1.2 0.7 0.7 0.41 0.37 112 277 .. 1,647 87.9 84.8 Portugal 7.8 8.2 20.1 21.7 0.25 0.22 72 203 .. 5,574 71.9 99.8 Romania 2.9 2.4 21.9 31.4 0.44 0.41 39 80 .. 3,170 112.7 100.5 Russian Federation .. 0.1 .. 4.0 .. 0.88 .. 122 .. 2,540 .. 69.5 Rwanda 10.3 12.2 0.4 0.3 0.15 0.13 0 0 307 201 89.7 76.9 Saudi Arabia 0.0 0.1 28.9 38.7 0.20 0.20 2 11 2,167 10,507 31.0 90.8 Senegal 0.0 0.1 2.6 3.1 0.42 0.27 0 0 341 321 74.5 109.1 Sierra Leone 0.7 0.8 4.1 5.4 0.14 0.11 0 1 368 404 84.5 94.7 Singapore 9.8 0.0 .. .. 0.00 0.00 3 16 13,937 39,851 154.3 37.9 Slovak Republic .. 2.7 .. 13.4 .. 0.28 .. 100 .. 3,347 .. 74.4 Slovenia .. 2.7 .. 0.7 .. 0.12 .. 2,762 .. 26,006 .. 100.9 South Africa 0.7 0.7 8.4 8.1 0.45 0.38 90 69 2,465 3,355 92.8 97.5 Spain 9.9 9.8 14.8 17.7 0.42 0.39 200 513 .. 12,022 82.1 99.4 Sri Lanka 15.9 15.5 28.3 29.2 0.06 0.05 8 9 648 732 98.4 113.0 Sweden .. .. .. .. 0.36 0.32 715 931 .. .. 100.2 95.1 Switzerland 0.5 0.6 6.2 5.9 0.06 0.06 494 616 .. .. 95.8 96.2 Syrian Arab Republic 2.5 3.9 9.6 20.4 0.60 0.33 29 65 .. .. 94.5 136.7 Tajikistan .. 0.5 .. 80.6 .. 0.14 .. 37 .. .. .. 67.9 Tanzania 1.0 1.0 3.8 4.6 0.12 0.10 1 1 .. 159 76.8 97.2 Thailand 3.5 6.6 16.4 23.2 0.35 0.29 1 7 630 928 79.9 107.2 Togo 6.6 6.6 0.3 0.3 0.76 0.50 0 0 345 510 77.0 129.9 Tunisia 9.7 13.1 4.9 7.5 0.51 0.32 30 39 1,743 2,750 67.6 108.3 Turkey 4.1 3.2 9.6 15.4 0.57 0.40 38 57 1,852 1,835 75.8 106.3 Turkmenistan .. 0.1 .. 87.8 .. 0.32 .. 83 .. .. .. 108.7 Uganda 8.0 8.8 0.1 0.1 0.32 0.26 0 1 .. 326 70.5 107.7 Ukraine .. 1.8 .. 7.5 .. 0.65 .. 92 .. 2,259 .. 69.9 United Kingdom 0.3 0.2 2.0 1.8 0.12 0.10 726 871 .. .. 91.6 100.5 United States 0.2 0.2 10.8 12.0 0.83 0.67 1,230 1,452 .. 34,727 94.7 113.7 Uruguay 0.3 0.3 5.4 10.7 0.48 0.39 171 172 6,822 9,384 86.8 128.8 Uzbekistan .. 0.9 .. 81.6 .. 0.20 .. 59 .. 2,085 .. 100.7 Venezuela 0.9 1.0 3.6 5.2 0.19 0.12 50 58 4,041 4,931 79.6 114.0 Vietnam 1.9 3.8 24.1 29.6 0.11 0.07 1 4 .. 226 64.0 132.7 Yemen, Rep. 0.2 0.2 19.9 31.3 0.16 0.09 3 2 .. 305 75.0 115.5 Zambia 0.0 0.0 0.4 0.9 0.89 0.59 3 2 331 226 74.2 95.6 Zimbabwe 0.3 0.3 3.1 4.6 0.36 0.27 7 7 307 316 82.1 94.8 World 0.9 w 1.0 w 17.8 w 18.8 w 0.24 w 0.24 w 19 w 20 w ..w ..w 76.0 w 128.2 w Low income 0.9 1.3 25.5 28.9 0.18 0.15 2 3 .. 339 69.3 137.5 Excl. China & India 1.0 1.3 16.2 19.4 0.23 0.18 .. .. .. .. 74.9 123.1 Middle income 1.2 1.0 15.8 14.1 0.23 0.36 25 46 .. .. 80.1 118.4 Lower middle income 1.3 0.8 22.9 14.9 0.16 0.39 17 34 .. .. .. .. Upper middle income 1.1 1.3 10.3 12.4 0.34 0.30 37 71 .. .. 78.6 116.7 Low and middle income 1.0 1.1 21.9 21.5 0.20 0.21 5 8 .. 601 72.0 132.6 East Asia & Pacific 1.5 2.6 37.0 35.5 0.11 0.11 2 2 .. .. 67.1 152.9 Europe & Central Asia 3.2 0.4 11.6 9.9 0.14 0.61 .. 102 .. 2,272 .. .. Latin America & Carib. 1.1 1.3 11.6 13.3 0.33 0.28 25 34 .. .. 80.7 121.0 Middle East & N. Africa 0.4 0.7 25.8 35.0 0.29 0.21 12 24 .. .. 70.5 128.5 South Asia 1.5 1.9 28.7 38.9 0.23 0.16 2 5 189 380 70.8 119.1 Sub-Saharan Africa 0.7 0.8 4.0 4.3 0.32 0.26 3 2 269 355 80.2 119.6 High income 0.5 0.5 9.8 11.1 0.46 0.41 520 877 .. .. 93.1 105.2 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 9. Water use, deforestation, and protected areas Freshwater Access to safe water Annual Nationally resources % of population deforestation protected areas Cu. meters Annual freshwater withdrawals with access 1990–95 1996 per capita Billion % of total % for % for % for 1995 Square Avg. annual Thousand % of total Economy 1996 cu. ma resourcesa agricultureb industryb domestic useb Urban Rural kilometers % change square km land area Albania 16,785c 0.2d 0.4c 76 18 6 97 70 0 0.0 0.8 2.9 Algeria 463c 4.5 33.2c 60e 15e 25e .. .. 234 1.2 58.9 2.5 Angola 15,782 0.5 0.3 76e 10e 14e 69 15 2,370 1.0 81.8 6.6 Argentina 27,861c 27.6d 2.8c 73 18 9 71 24 894 0.3 46.6 1.7 Armenia 2,136c 3.8 47.0c 72 15 13 .. .. –84 –2.7 2.1 7.4 Australia 18,508 14.6d 4.3 33 2 65 .. .. –170 0.0 563.9 7.3 Austria 11,187c 2.4 2.6c 9 58 33 .. .. 0 0.0 23.4 28.3 Azerbaijan 4,339c 15.8 47.9c 74 22 4 .. .. 0 0.0 4.8 5.5 Bangladesh 19,065 c 22.5 1.0 c 96 1 3 49 .. 88 0.8 1.0 0.8 Belarus 1,841 c 3.0 15.9 c 19 49 32 .. .. –688 –1.0 8.6 4.1 Belgium 1,227 c 9.0 72.2 c 4 85 11 .. .. 0 0.0 0.8 .. Benin 4,451 c 0.2 0.6 c 67 e 10 e 23 e 82 69 596 1.2 7.8 7.1 Bolivia 38,625 1.2 0.4 85 5 10 88 43 5,814 1.2 156.0 14.4 Botswana 9,589c 0.1 0.7c 48e 20e 32e 100 77 708 0.5 105.0 18.5 Brazil 42,459c 36.5 0.5c 59 19 22 80 28 25,544 0.5 355.5 4.2 Bulgaria 24,663c 13.9 6.8c 22 76 3 .. .. –6 0.0 4.9 4.4 Burkina Faso 1,671 0.4 2.2 81e 0e 19e .. .. 320 0.7 28.6 10.5 Burundi 559 0.1 2.8 64e 0e 36e .. .. 14 0.4 1.4 5.5 Cambodia 47,530c 0.5 0.1c 94 1 5 20 12 1,638 1.6 28.6 16.2 Cameroon 19,231 0.4 0.1 35 e 19 e 46 e .. 30 1,292 0.6 21.0 4.5 Canada 95,785c 45.1 1.6c 12 70 18 .. .. –1,764 –0.1 921.0 10.0 Central African Republic 41,250 0.1 0.0 74 e 5 e 21 e 20 25 1,282 0.4 51.1 8.2 Chad 6,011c 0.2 0.4c 82e 2e 16e 48 17 942 0.8 114.9 9.1 Chile 32,007 16.8 d 3.6 89 5 6 99 47 292 0.4 141.3 18.9 China 2,282 460.0 16.4 87 7 6 .. .. 866 0.1 598.1 6.4 Hong Kong, China .. .. .. .. .. .. .. .. .. .. 0.4 40.4 Colombia 26,722 5.3 0.5 43 16 41 90 32 2,622 0.5 93.6 9.0 Congo, Dem. Rep. 21,816c 0.4 0.0c 23e 16e 61e 89 26 .. .. 101.9 4.5 Congo, Rep. 307,283c 0.0 0.0c 11e 27e 62e .. 11 416 0.2 15.4 4.5 Costa Rica 27,425 1.4d 1.4 89 7 4 100 99 414 3.0 7.0 13.7 Côte d’Ivoire 5,468 0.7 0.9 67e 11e 22e .. .. 308 0.6 19.9 6.3 Croatia 12,879 .. .. .. .. .. 75 41 0 0.0 3.7 6.6 Czech Republic 5,649 2.7 4.7 2 57 41 .. .. –2 0.0 12.2 15.8 Denmark 2,460 c 1.2 9.2 c 43 27 30 .. .. 0 0.0 13.7 32.3 Dominican Republic 2,467 3.0 14.9 89 6 5 88 55 264 1.6 12.2 25.2 Ecuador 26,305 5.6 1.8 90 3 7 81 10 1,890 1.6 119.3 43.1 Egypt, Arab Rep. 966 c 55.1 94.5 c 86 e 8 e 6 e 95 74 0 0.0 7.9 0.8 El Salvador 3,197 1.0 d 5.3 89 4 7 82 24 38 3.3 0.1 0.5 Eritrea 2,332c .. . .c .. .. .. .. .. 0 0.0 5.0 5.0 Estonia 12,071c 3.3 18.8c 3 92 5 .. .. –196 –1.0 5.1 12.1 Ethiopia 1,841 2.2 2.0 86e 3e 11e .. .. 624 0.5 55.2 5.5 Finland 21,985c 2.2 1.9c 3 85 12 .. .. 166 0.1 18.2 6.0 France 3,029c 37.7 21.3c 15 69 16 100 100 –1,608 –1.1 58.8 10.7 Georgia 8,291 c 4.0 8.9 c 42 37 21 .. .. 0 0.0 1.9 2.7 Germany 2,084c 46.3 27.1c 20 70 11 .. .. 0 0.0 94.2 27.0 Ghana 2,958 c 0.3 d 0.6 c 52 e 13 e 35 e 88 52 1,172 1.3 11.0 4.8 Greece 5,289c 5.0 9.1c 63 29 8 .. .. –1,408 –2.3 3.1 2.4 Guatemala 11,028 0.7 d 0.6 74 17 9 97 48 824 2.0 18.2 16.8 Guinea 32,661 0.7 0.3 87 e 3 e 10 e 55 44 748 1.1 1.6 0.7 Haiti 1,468 0.0 0.4 68 8 24 38 39 8 3.4 0.1 0.4 Honduras 9,259 c 1.5 2.7 c 91 5 4 91 66 1,022 2.3 11.1 9.9 Hungary 11,817 c 6.8 5.7 c 36 55 9 .. .. –88 –0.5 6.3 6.8 India 2,167c 380.0d 18.2c 93 4 3 .. 82 –72 0.0 142.9 4.8 Indonesia 12,625 16.6 0.7 76 11 13 87 57 10,844 1.0 192.3 10.6 Iran, Islamic Rep. 1,339c 70.0d 85.8c 92 2 6 98 82 284 1.7 83.0 5.1 Ireland 13,657c 0.8d 1.6c 10 74 16 .. .. –140 –2.7 0.6 0.9 Israel 377c 1.9 84.1c 79e 5e 16e 100 95 0 0.0 3.1 15.0 Italy 2,903 c 56.2 33.7 c 59 27 14 .. .. –58 –0.1 21.5 7.3 Jamaica 3,250 0.3d 3.9 86 7 7 .. .. 158 7.2 0.0 0.0 Japan 4,338 90.8 16.6 50 33 17 .. .. 132 0.1 25.5 6.8 Jordan 198c 0.5d 51.1c 75 3 22 .. .. 12 2.5 3.0 3.4 Kazakhstan 8,696 c 37.9 27.6 c 79 17 4 .. .. –1,928 –1.9 73.4 2.7 Kenya 1,056c 2.1 6.8c 76e 4e 20e .. .. 34 0.3 35.0 6.1 Korea, Rep. 1,438 27.6 41.7 46 35 19 93 77 130 0.2 6.8 6.9 Kuwait 11 0.5 2,700.0 60 2 37 100 100 0 0.0 0.3 1.7 Kyrgyz Republic 2,509 11.0 94.9 95 3 2 .. .. 0 0.0 6.9 3.6 Lao PDR 55,679 1.0 0.4 82 10 8 .. .. .. .. 0.0 0.0 Latvia 13,793c 0.7 2.1c 14 44 42 92 .. –250 –0.9 7.8 12.6 Lebanon 941c 1.3e,d 33.1c 68 4 28 .. .. 52 7.8 0.0 0.0 Lesotho 2,597 0.1 1.0 56 e 22 e 22 e 64 60 0 0.0 0.1 0.3 Lithuania 6,531 c 4.4 18.2 c 3 90 7 .. .. –112 –0.6 6.5 10.0 Macedonia, FYR .. .. .. .. .. .. .. .. 2 0.0 1.8 7.1 Madagascar 23,819 16.3 4.8 99 e 0 e 1 e .. .. 1,300 0.8 11.2 1.9 Malawi 1,814 c 0.9 4.8 c 86 e 3 e 10 e 97 52 546 1.6 10.6 11.3 Malaysia 21,046 9.4d 2.1 47 30 23 100 86 4,002 2.4 14.8 4.5 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      ENVIRONMENT Freshwater Access to safe water Annual Nationally resources % of population deforestation protected areas Cu. meters Annual freshwater withdrawals with access 1990–95 1996 per capita Billion % of total % for % for % for 1995 Square Avg. annual Thousand % of total Economy 1996 cu. ma resourcesa agricultureb industryb domestic useb Urban Rural kilometers % change square km land area Mali 9,718c 1.4 1.4c 97e 1e 2e 56 20 1,138 1.0 45.3 3.7 Mauritania 4,632c 1.6d 14.3c 92 2 6 87 41 0 0.0 17.5 1.7 Mexico 3,788 77.6 d 21.7 86 8 6 .. .. 5,080 0.9 71.0 3.7 Moldova 397 c 3.7 216.4 c 23 70 7 98 18 0 0.0 0.4 1.2 Mongolia 9,677 0.6 2.2 62 27 11 100 68 0 0.0 161.3 10.3 Morocco 1,088 10.9 36.5 92 e 3 e 5 e 97 20 118 0.3 3.2 0.7 Mozambique 12,989c 0.6 0.3c 89 2e 9e .. 40 1,162 0.7 47.8 6.1 Myanmar 24,651 4.0 0.4 90 3 7 78 50 3,874 1.4 1.7 0.3 Namibia 28,042c 0.3 0.5c 68e 3e 29e .. .. 420 0.3 106.2 12.9 Nepal 7,616 2.7 1.6 95 1 4 61 59 548 1.1 11.1 7.8 Netherlands 5,767c 7.8 8.7c 34 61 5 .. .. 0 0.0 2.4 7.1 New Zealand 532 2.0 100.0 44 10 46 .. .. –434 –0.6 63.3 23.6 Nicaragua 37,420 0.9d 0.5 54 21 25 93 28 1,508 2.5 9.0 7.4 Niger 3,317c 0.5 1.5c 82e 2e 16e 70 44 0 0.0 96.9 7.6 Nigeria 2,375c 3.6 1.3c 54e 15e 31e 80 39 1,214 0.9 30.2 3.3 Norway 89,008c 2.0 0.5c 8 72 20 100 100 –180 –0.2 93.7 30.5 Pakistan 3,256c 155.6d 37.2c 97 2 2 85 56 550 2.9 37.2 4.8 Panama 52,961 1.3 0.9 77 11 12 99 73 636 2.1 14.2 19.1 Papua New Guinea 177,963 0.1 0.0 49 22 29 .. .. 1,332 0.4 0.1 0.0 Paraguay 61,750c 0.4 0.1c 78 7 15 70 6 3,266 2.6 14.0 3.5 Peru 1,641 6.1 15.3 72 9 19 91 31 2,168 0.3 34.6 2.7 Philippines 4,393 29.5d 9.1 61 21 18 91 81 2,624 3.5 14.5 4.9 Poland 1,454c 12.3 21.9c 11 76 13 .. .. –120 –0.1 29.1 9.6 Portugal 6,998c 7.3 10.5c 48 37 15 .. .. –240 –0.9 5.9 6.4 Romania 9,222c 26.0 12.5c 59 33 8 .. .. 12 0.0 10.7 4.6 Russian Federation 30,168c 117.0 2.6c 23 60 17 .. .. 0 0.0 516.7 3.1 Rwanda 798 0.8 12.2 94e 2e 5e 79 44 4 0.2 3.6 14.6 Saudi Arabia 120 17.0d 709.2 90 1 9 .. .. 18 0.8 49.6 2.3 Senegal 4,482c 1.4 3.5c 92e 3e 5e 90 44 496 0.7 21.8 11.3 Sierra Leone 33,698 0.4 0.2 89e 4e 7e 58 21 426 3.0 0.8 1.1 Singapore 193 0.2d 31.7 4 51 45 .. .. 0 0.0 0.0 0.0 Slovak Republic 5,720 1.8 5.8 .. .. .. .. .. –24 –0.1 10.5 21.8 Slovenia .. .. .. .. .. .. .. .. 0 0.0 1.1 5.5 South Africa 1,231c 13.3 26.6c 72e 11e 17e 90 33 150 0.2 65.8 5.4 Spain 2,398c 30.8 32.6c 62 26 12 .. .. 0 0.0 42.2 8.4 Sri Lanka 2,329 6.3d 14.6 96 2 2 88 65 202 1.1 8.6 13.3 Sweden 20,340c 2.9 1.6c 9 55 36 .. .. 24 0.0 36.2 8.8 Switzerland 7,054c 1.2 2.4c 4 73 23 100 100 0 0.0 7.1 18.0 Syrian Arab Republic 859c 14.4 112.6c 94 2 4 96 79 52 2.2 0.0 0.0 Tajikistan .. 12.6 .. 88 7 5 .. .. 0 0.0 5.9 4.2 Tanzania 2,842c 1.2 1.3c 89e 2e 9e .. .. 3,226 1.0 138.2 15.6 Thailand 2,954c 31.9 17.8c 90 6 4 94 88 3,294 2.6 70.7 13.8 Togo 2,762c 0.1 0.8c 25e 13e 62e 82 41 186 1.4 4.3 7.9 Tunisia 447c 3.1 74.5c 89e 3e 9e 100 76 30 0.5 0.4 0.3 Turkey 2,246c 31.6 22.1c 72e 11e 16e .. .. 0 0.0 10.7 1.4 Turkmenistan 3,950c 22.8 123.9c 91 8 1 .. .. 0 0.0 19.8 4.2 Uganda 3,248c 0.2 0.3c 60 8 32 60 36 592 0.9 19.1 9.6 Ukraine 4,556c 34.7 15.0c 30 54 16 .. .. –54 –0.1 9.0 1.6 United Kingdom 1,203 11.8 16.6 3 77 20 100 100 –128 –0.5 50.6 20.9 United States c 9,259 467.3 18.9 c 42 e 45 e 13 e .. .. –5,886 –0.3 1,226.7 13.4 Uruguay 37,966c 0.7d 0.5c 91 3 6 99 .. 4 0.0 0.5 0.3 Uzbekistan 5,476c 82.2 63.4c 84 12 4 .. .. –2,260 –2.7 8.2 2.0 Venezuela 57,821c 4.1d 0.3c 46 11 43 79 79 5,034 1.1 319.8 36.3 Vietnam 4,902 28.9 7.7 78 9 13 .. .. 1,352 1.4 9.9 3.0 Yemen, Rep. 255 2.9 71.5 92 .. .. 74 14 0 0.0 0.0 0.0 Zambia 12,284c 1.7 1.5c 77e 7e 16e 66 37 2,644 0.8 63.6 8.6 Zimbabwe 1,744c 1.2 6.1c 79e 7e 14e .. .. 500 0.6 30.7 7.9 World 8,338 w 69 w 22 w 9w ..w . . w 101,724 s 0.3 w 8,542.7 s 6.6 w Low income 5,214 90 5 5 .. .. 49,332 0.7 2,439.4 5.9 Excl. China & India .. .. .. .. .. .. .. .. .. .. Middle income 14,950 67 22 11 .. .. 64,086 0.3 2,809.9 4.8 Lower middle income 11,573 67 24 9 .. .. 21,162 0.2 1,563.6 4.3 Upper middle income .. 68 19 13 .. .. 42,924 0.5 1,246.3 5.7 Low and middle income 8,095 80 13 7 .. .. 113,418 0.4 5,249.3 5.3 East Asia & Pacific .. 82 10 8 .. .. 29,956 0.8 1,102.2 6.9 Europe & Central Asia 13,255 54 36 10 .. .. –5,798 –0.1 768.0 3.2 Latin America & Carib. 27,386 77 11 12 83 36 57,766 0.6 1,456.3 7.3 Middle East & N. Africa 1,045 90 4 6 .. .. 800 0.9 242.0 2.2 South Asia 4,085 94 3 3 84 84 1,316 0.2 213.0 4.5 Sub-Saharan Africa 8,565 85 4 10 74 32 29,378 0.7 1,467.8 6.2 High income .. 39 46 15 .. .. –11,694 –0.2 3,293.4 10.8 a. Refers to any year from 1980 to 1997, unless otherwise noted. b. Unless otherwise noted, percentages are estimated for 1987. c. Includes river flows from other countries. d. Data refer to estimates for years before 1980 (see World Development Indicators, 1999). e. Data refer to years other than 1987 (see World Development Indicators, 1999). (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 10. Energy use and emissions Commercial energy use Thousand metric Per capita GDP per unit Net energy imports Carbon dioxide emissions tons of Kg of oil Avg. annual of energy use % of commercial Total Per capita oil equivalent equivalent % growth 1995 $ per kg energy use Million metric tons Metric tons Economy 1980 1996 1980 1996 1980–96 1980 1996 1980 1996 1980 1996 1980 1996 Albania 3,049 1,188 1,142 362 –7.8 0.8 2.2 –12 9 4.8 1.9 1.8 0.6 Algeria 12,410 24,150 665 842 1.0 2.5 1.8 –440 –381 66.2 94.3 3.5 3.3 Angola 4,538 6,017 647 532 –1.2 .. 0.9 –149 –573 5.3 5.1 0.8 0.5 Argentina 41,868 58,921 1,490 1,673 0.7 5.7 5.0 7 –27 107.5 129.9 3.8 3.7 Armenia 1,070 1,790 346 474 –4.9 5.2 1.7 –18 59 .. 3.7 .. 1.0 Australia 70,372 100,612 4,790 5,494 0.9 3.3 3.7 –22 –88 202.8 306.6 13.8 16.7 Austria 23,450 27,187 3,105 3,373 0.9 7.1 8.7 67 71 52.2 59.3 6.9 7.4 Azerbaijan 15,002 11,862 2,433 1,570 –5.6 .. 0.3 1 –21 .. 30.0 .. 4.0 Bangladesh 14,920 23,928 172 197 0.9 1.3 1.7 11 10 7.6 23.0 0.1 0.2 Belarus 2,385 24,566 247 2,386 7.5 .. 0.8 –8 87 .. 61.7 .. 6.0 Belgium 46,100 56,399 4,682 5,552 1.4 4.6 4.9 84 79 127.2 106.0 12.9 10.4 Benin 1,363 1,920 394 341 –1.0 0.9 1.1 11 –2 0.5 0.7 0.1 0.1 Bolivia 2,335 3,633 436 479 0.0 2.3 1.9 –84 –44 4.5 10.1 0.8 1.3 Botswana .. .. .. .. .. .. .. .. .. 1.0 2.1 1.1 1.4 Brazil 108,997 163,374 896 1,012 1.0 4.7 4.4 43 31 183.4 273.4 1.5 1.7 Bulgaria 28,673 22,605 3,235 2,705 –2.0 0.4 0.5 73 54 75.3 55.3 8.5 6.6 Burkina Faso .. .. .. .. .. .. .. .. .. 0.4 1.0 0.1 0.1 Burundi .. .. .. .. .. .. .. .. .. 0.1 0.2 0.0 0.0 Cambodia .. .. .. .. .. .. .. .. .. 0.3 0.5 0.0 0.0 Cameroon 3,687 5,000 426 369 –1.1 1.7 1.7 –58 –100 3.9 3.5 0.4 0.3 Canada 193,000 236,170 7,848 7,880 0.3 2.1 2.5 –7 –51 420.9 409.4 17.1 13.7 Central African Republic .. .. .. .. .. .. .. .. .. 0.1 0.2 0.0 0.1 Chad .. .. .. .. .. .. .. .. .. 0.2 0.1 0.0 0.0 Chile 9,525 20,456 855 1,419 3.7 2.8 3.1 41 62 27.9 48.8 2.5 3.4 China 593,109 1,096,800 604 902 2.6 0.3 0.7 –3 0 1,476.8 3,363.5 1.5 2.8 Hong Kong, China 5,681 12,190 1,127 1,931 4.4 10.0 12.0 99 100 16.3 23.1 3.2 3.7 Colombia 19,127 31,393 672 799 1.0 2.4 2.6 5 –113 39.8 65.3 1.4 1.7 Congo, Dem. Rep. 8,706 13,799 322 305 –0.2 1.0 0.5 0 1 3.5 2.3 0.1 0.1 Congo, Rep. 845 1,205 506 457 –0.6 1.5 1.9 –370 –854 0.4 5.0 0.2 1.9 Costa Rica 1,527 2,248 669 657 0.7 3.7 4.0 50 67 2.5 4.7 1.1 1.4 Côte d’Ivoire 3,662 5,301 447 382 –0.6 2.3 2.0 34 10 4.6 13.1 0.6 0.9 Croatia .. 6,765 . . 1,418 .. .. 2.8 .. 42 .. 17.5 .. 3.7 Czech Republic 46,910 40,404 4,585 3,917 –1.7 .. 1.3 9 22 .. 126.7 .. 12.3 Denmark 19,734 22,870 3,852 4,346 0.8 6.8 8.2 95 23 62.9 56.6 12.3 10.7 Dominican Republic 3,464 5,191 608 652 0.1 2.2 2.5 62 72 6.4 12.9 1.1 1.6 Ecuador 5,191 8,548 652 731 0.2 2.4 2.1 –126 –156 13.4 24.5 1.7 2.1 Egypt, Arab Rep. 15,970 37,790 391 638 2.6 1.8 1.6 –114 –58 45.2 97.9 1.1 1.7 El Salvador 2,540 4,058 554 700 1.0 2.9 2.4 25 36 2.1 4.0 0.5 0.7 Eritrea .. .. .. .. .. .. .. .. .. .. .. .. .. Estonia .. 5,621 . . 3,834 .. .. 0.9 .. 31 .. 16.4 .. 11.2 Ethiopia 11,157 16,566 296 284 –0.1 0.4 0.4 5 6 1.8 3.4 0.0 0.1 Finland 25,413 31,482 5,316 6,143 1.1 3.7 4.1 73 57 54.9 59.2 11.5 11.5 France 190,111 254,196 3,528 4,355 1.6 6.1 6.1 76 49 482.7 361.8 9.0 6.2 Georgia 4,474 1,576 882 291 –5.8 2.7 2.1 –5 55 .. 3.0 .. 0.5 Germany 360,441 349,552 4,603 4,267 –0.5 .. 7.0 48 60 .. 861.2 .. 10.5 Ghana 4,071 6,657 379 380 0.4 1.0 1.0 19 16 2.4 4.0 0.2 0.2 Greece 15,960 24,389 1,655 2,328 2.5 5.7 4.8 77 64 51.7 80.6 5.4 7.7 Guatemala 3,754 5,224 550 510 0.0 2.9 2.9 33 23 4.5 6.8 0.7 0.7 Guinea .. .. .. .. .. .. .. .. .. 0.9 1.1 0.2 0.2 Haiti 2,099 1,968 392 268 –2.8 1.5 1.4 11 19 0.8 1.1 0.1 0.1 Honduras 1,877 2,925 526 503 –0.2 1.4 1.4 30 40 2.1 4.0 0.6 0.7 Hungary 28,895 25,470 2,699 2,499 –0.8 1.6 1.8 48 50 82.5 59.5 7.7 5.8 India 242,024 450,287 352 476 1.9 0.6 0.8 8 13 347.3 997.4 0.5 1.1 Indonesia 59,561 132,419 402 672 3.5 1.3 1.6 –116 –66 94.6 245.1 0.6 1.2 Iran, Islamic Rep. 38,918 89,340 995 1,491 3.2 1.4 1.1 –116 –147 116.1 266.7 3.0 4.4 Ireland 8,484 11,961 2,495 3,293 2.0 4.0 5.9 78 71 25.2 34.9 7.4 9.6 Israel 8,609 16,185 2,220 2,843 2.6 5.1 5.6 98 96 21.1 52.3 5.4 9.2 Italy 138,629 161,140 2,456 2,808 1.3 6.0 6.8 86 82 371.9 403.2 6.6 7.0 Jamaica 2,378 3,718 1,115 1,465 2.3 1.3 1.1 91 85 8.4 10.1 4.0 4.0 Japan 346,491 510,359 2,967 4,058 2.4 9.3 10.5 88 80 920.4 1,167.7 7.9 9.3 Jordan 1,714 4,487 786 1,040 0.6 2.2 1.5 100 96 .. .. .. .. Kazakhstan 76,799 43,376 5,163 2,724 –4.9 .. 0.5 0 –44 .. 173.8 .. 10.9 Kenya 9,791 13,279 589 476 –1.1 0.6 0.7 19 15 6.2 6.8 0.4 0.2 Korea, Rep. 43,756 162,874 1,148 3,576 8.1 3.1 3.0 72 86 125.2 408.1 3.3 9.0 Kuwait 9,564 13,859 6,956 8,167 0.7 2.4 1.7 –884 –712 .. .. .. .. Kyrgyz Republic 1,717 2,952 473 645 4.1 .. 1.2 –27 51 .. 6.1 .. 1.3 Lao PDR .. .. .. .. .. .. .. .. .. 0.2 0.3 0.1 0.1 Latvia .. 4,171 . . 1,674 .. 16.0 1.5 54 76 .. 9.3 .. 3.7 Lebanon 2,483 4,747 827 1,164 1.7 .. .. 93 96 6.2 14.2 2.1 3.5 Lesotho .. .. .. .. .. .. .. .. .. .. .. .. .. Lithuania 11,701 8,953 3,428 2,414 –4.0 .. 0.8 95 53 .. 13.8 .. 3.7 Macedonia, FYR .. .. .. .. .. .. .. .. .. .. 12.7 .. 6.4 Madagascar .. .. .. .. .. .. .. .. .. 1.6 1.2 0.2 0.1 Malawi .. .. .. .. .. .. .. .. .. 0.7 0.7 0.1 0.1 Malaysia 11,128 41,209 809 1,950 6.0 2.9 2.3 –50 –69 28.0 119.1 2.0 5.6 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      ENVIRONMENT Commercial energy use Thousand metric Per capita GDP per unit Net energy imports Carbon dioxide emissions tons of Kg of oil Avg. annual of energy use % of commercial Total Per capita oil equivalent equivalent % growth 1995 $ per kg energy use Million metric tons Metric tons Economy 1980 1996 1980 1996 1980–96 1980 1996 1980 1996 1980 1996 1980 1996 Mali .. .. .. .. .. .. .. .. .. 0.4 0.5 0.1 0.0 Mauritania .. .. .. .. .. .. .. .. .. 0.6 2.9 0.4 1.2 Mexico 98,904 141,384 1,464 1,525 0.2 2.3 2.1 –51 –51 251.6 348.1 3.7 3.8 Moldova .. 4,601 . . 1,064 .. .. 0.6 106 99 .. 12.1 .. 2.8 Mongolia .. .. .. .. .. .. .. .. .. 6.8 8.9 4.1 3.6 Morocco 4,778 8,822 247 329 2.1 4.5 4.2 82 90 15.9 27.9 0.8 1.0 Mozambique 8,386 7,813 693 481 –2.0 0.2 0.3 –2 7 3.2 1.0 0.3 0.1 Myanmar 9,430 12,767 279 294 0.3 .. .. –1 7 4.8 7.3 0.1 0.2 Namibia .. .. .. .. .. .. .. .. .. .. .. .. .. Nepal 4,663 6,974 322 320 0.1 0.5 0.7 3 9 0.5 1.6 0.0 0.1 Netherlands 65,000 75,797 4,594 4,885 0.9 4.4 5.4 –11 3 152.6 155.2 10.8 10.0 New Zealand 9,251 16,295 2,972 4,388 2.9 4.7 3.8 41 17 17.6 29.8 5.6 8.0 Nicaragua 1,562 2,391 535 525 –0.1 1.3 1.0 42 37 2.0 2.9 0.7 0.6 Niger .. .. .. .. .. .. .. .. .. 0.6 1.1 0.1 0.1 Nigeria 52,846 82,669 743 722 –0.1 0.4 0.4 –181 –106 68.1 83.3 1.0 0.7 Norway 18,819 23,150 4,600 5,284 1.2 5.1 6.7 –196 –799 90.4 67.0 22.1 15.3 Pakistan 25,479 55,903 308 446 2.3 1.0 1.1 18 26 31.6 94.3 0.4 0.8 Panama 1,865 2,280 957 853 –0.3 2.8 3.6 72 67 3.5 6.7 1.8 2.5 Papua New Guinea .. .. .. .. .. .. .. .. .. 1.8 2.4 0.6 0.5 Paraguay 2,094 4,285 672 865 1.5 2.8 2.1 23 –56 1.5 3.7 0.5 0.7 Peru 11,700 13,933 675 582 –1.2 4.1 4.3 –25 11 23.6 26.2 1.4 1.1 Philippines 21,212 37,992 439 528 1.1 2.7 2.1 50 55 36.5 63.2 0.8 0.9 Poland 124,806 108,411 3,508 2,807 –2.0 0.9 1.2 2 6 456.2 356.8 12.8 9.2 Portugal 10,291 19,148 1,054 1,928 4.5 6.8 5.6 86 87 27.1 47.9 2.8 4.8 Romania 64,694 45,824 2,914 2,027 –2.9 0.6 0.7 19 32 191.8 119.3 8.6 5.3 Russian Federation 764,349 615,899 5,499 4,169 –3.6 0.5 0.5 2 –54 .. 1,579.5 . . 10.7 Rwanda .. .. .. .. .. .. .. .. .. 0.3 0.5 0.1 0.1 Saudi Arabia 35,357 92,243 3,773 4,753 0.4 3.0 1.4 –1,408 –415 130.7 267.8 14.0 13.8 Senegal 1,921 2,588 347 302 –0.7 1.6 1.8 46 39 2.8 3.1 0.5 0.4 Sierra Leone .. .. .. .. .. .. .. .. .. 0.6 0.4 0.2 0.1 Singapore 6,054 23,851 2,653 7,835 8.1 4.6 3.8 100 100 30.1 65.8 13.2 21.6 Slovak Republic 20,810 17,449 4,175 3,266 –1.8 .. 1.1 84 72 .. 39.6 .. 7.4 Slovenia 4,313 6,167 2,269 3,098 1.0 .. 3.1 62 55 .. 13.0 .. 6.5 South Africa 65,355 99,079 2,370 2,482 –0.4 1.7 1.4 –12 –29 211.3 292.7 7.7 7.3 Spain 68,583 101,411 1,834 2,583 2.8 5.7 5.6 77 68 200.0 232.5 5.3 5.9 Sri Lanka 4,493 6,792 305 371 0.7 1.5 2.0 29 38 3.4 7.1 0.2 0.4 Sweden 40,984 52,567 4,932 5,944 0.9 4.5 4.5 61 39 71.4 54.1 8.6 6.1 Switzerland 20,861 25,622 3,301 3,622 0.8 12.1 12.0 66 59 40.9 44.2 6.5 6.3 Syrian Arab Republic 5,348 14,541 614 1,002 2.4 1.7 1.2 –78 –132 19.3 44.3 2.2 3.1 Tajikistan 1,650 3,513 416 594 5.1 .. 0.5 –20 62 .. 5.8 .. 1.0 Tanzania 10,280 13,798 553 453 –1.1 .. 0.3 8 5 1.9 2.4 0.1 0.1 Thailand 22,740 79,987 487 1,333 7.3 2.3 2.2 51 45 40.0 205.4 0.9 3.4 Togo .. .. .. .. .. .. .. .. .. 0.6 0.8 0.2 0.2 Tunisia 3,900 6,676 611 735 1.4 2.7 2.9 –79 6 9.4 16.2 1.5 1.8 Turkey 31,314 65,520 704 1,045 2.6 2.8 2.8 45 59 76.3 178.3 1.7 2.8 Turkmenistan 7,948 12,164 2,778 2,646 –10.5 .. 0.3 –1 –168 .. 34.2 .. 7.4 Uganda .. .. .. .. .. .. .. .. .. 0.6 1.0 0.1 0.1 Ukraine 97,893 153,937 1,956 3,012 1.0 .. 0.5 –12 49 .. 397.3 .. 7.8 United Kingdom 201,299 234,719 3,574 3,992 0.8 4.0 4.8 2 –14 583.8 557.0 10.4 9.5 United States 1,811,650 2,134,960 7,973 8,051 0.4 2.7 3.4 14 21 4,575.4 5,301.0 20.1 20.0 Uruguay 2,637 2,955 905 912 0.2 5.8 6.4 75 65 5.8 5.6 2.0 1.7 Uzbekistan 4,821 42,406 302 1,826 7.0 .. 0.5 4 –12 .. 95.0 .. 4.1 Venezuela 35,026 54,962 2,321 2,463 –0.4 1.7 1.4 –280 –253 89.6 144.5 5.9 6.5 Vietnam 19,348 33,750 360 448 0.7 .. 0.7 7 –14 16.8 37.6 0.3 0.5 Yemen, Rep. 1,424 2,936 167 187 0.6 .. 1.3 96 –519 .. .. .. .. Zambia 4,551 5,790 793 628 –1.7 0.7 0.6 8 7 3.5 2.4 0.6 0.3 Zimbabwe 6,511 10,442 929 929 0.4 0.7 0.7 13 16 9.6 18.4 1.4 1.6 World 6,954,847 t 9,317,404 t 1,622 w 1,684 w 2.9 w 3.1 w 3.2 w ..w . . w 13,640.7 t 22,653.9 t 3.4 w 4.0 w Low income 1,153,366 2,063,558 480 640 3.9 .. .. –14 –9 2,126.1 5,051.8 0.9 1.5 Excl. China & India 318,233 516,471 433 486 3.7 .. 0.8 .. .. 302.0 690.9 0.4 0.6 Middle income 2,030,275 2,588,365 1,852 1,801 5.0 2.4 1.7 –35 –33 2,804.5 6,871.5 3.3 4.8 Lower middle income 1,368,743 1,537,541 2,040 1,763 7.4 1.7 1.0 –13 –20 1,150.1 4,194.9 2.6 4.8 Upper middle income 661,532 1,050,824 1,557 1,861 2.8 2.8 2.6 –98 –65 1,654.4 2,676.6 4.0 4.7 Low and middle income 3,183,641 4,651,923 910 998 4.5 1.4 1.3 –32 –28 4,930.6 11,923.3 1.5 2.5 East Asia & Pacific 812,075 1,621,801 588 925 4.6 .. .. .. .. 1,958.5 4,717.5 1.4 2.7 Europe & Central Asia 1,339,527 1,287,193 3,349 2,739 7.6 .. 0.8 7 –13 886.9 3,412.7 .. 7.4 Latin America & Carib. 376,913 557,686 1,062 1,163 2.4 3.5 3.2 –24 –35 848.5 1,209.1 2.4 2.5 Middle East & N. Africa 146,215 337,073 842 1,244 5.1 2.2 1.6 –577 –225 493.6 986.9 3.0 3.9 South Asia 301,578 543,884 334 441 3.9 0.7 0.9 10 15 392.4 1,125.1 0.4 0.9 Sub-Saharan Africa 207,332 304,286 720 670 2.3 .. .. .. .. 350.7 472.1 0.9 0.8 High income 3,771,206 4,665,482 4,792 5,346 1.6 4.1 5.0 27 24 8,710.2 10,730.6 12.3 12.3 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 11. Growth of the economy Average annual % growth Gross GDP Agriculture Industry Services Exports of goods Gross domestic domestic product implicit deflator value added value added value added and services investment Economy 1980–90 1990–98 1980–90 1990–98 1980–90 1990–98 1980–90 1990–98 1980–90 1990–98 1980–90 1990–98 1990–98 Albania 1.5 1.8 –0.4 58.1 1.9 8.1 2.1 –9.7 –0.4 4.7 .. 18.3 26.9 Algeria 2.7 1.2 8.1 21.4 4.6 2.6 2.3 –2.0 3.6 4.8 4.1 3.0 –0.8 Angola 3.7 –0.4 5.9 921.1 0.5 –4.3 6.4 3.6 1.8 –5.7 3.7 5.9 12.6 Argentina –0.4 5.3 389.8 10.0 0.7 2.1 –1.3 4.6 0.0 3.9 3.8 9.3 12.5 Armenia .. –10.3 .. 482.8 .. 0.2 .. –18.1 .. –10.8 .. 2.3 –10.9 Australia 3.4 3.6 7.3 1.8 3.3 1.1 2.9 2.5 3.7 4.4 6.9 8.1 5.4 Austria 2.2 2.0 3.3 2.5 1.1 –0.7 1.9 1.3 2.5 2.2 4.9 4.0 2.6 Azerbaijan .. –10.5 .. 316.5 .. –2.7 .. 4.2 .. 9.9 .. 19.5 108.8 Bangladesh 4.3 4.8 9.5 3.6 2.7 1.5 4.9 7.0 5.0 5.2 7.7 13.7 7.0 Belarus .. –6.1 .. 561.4 .. –5.9 .. –7.8 .. –3.8 .. –23.3 –12.2 Belgium 2.0 1.6 4.4 2.3 2.0 1.6 2.2 0.7 1.8 1.3 4.3 4.4 –0.3 Benin 2.9 4.6 1.3 10.1 5.5 5.2 3.0 4.0 1.4 4.3 –2.4 3.3 4.6 Bolivia –0.2 4.2 327.9 10.0 .. .. .. .. .. .. 1.0 6.7 6.9 Botswana 10.3 4.8 13.6 10.3 3.3 0.1 10.2 3.1 11.7 7.1 10.6 4.9 2.0 Brazil 2.7 3.3 284.0 347.3 2.8 3.1 2.0 3.2 3.3 3.4 7.5 5.6 3.9 Bulgaria 3.4 –3.3 1.8 109.5 –2.1 –3.1 5.2 –5.5 4.5 –0.6 –3.5 2.3 –12.8 Burkina Faso 3.6 3.5 3.3 6.6 3.1 3.4 3.8 3.1 4.6 3.2 –0.4 –0.8 4.1 Burundi 4.4 –3.2 4.4 12.2 3.1 –2.4 4.5 –7.8 5.6 –2.9 3.4 0.1 –16.1 Cambodia .. 5.5 .. 37.8 .. 2.2 .. 10.7 .. 7.6 .. .. .. Cameroon 3.4 0.6 5.6 6.1 2.1 5.0 5.9 –3.3 2.1 0.0 5.9 –1.5 –1.6 Canada 3.3 2.2 4.5 1.6 1.2 1.2 3.1 1.8 3.6 1.8 6.3 9.0 1.5 Central African Republic 1.4 1.5 7.9 5.4 1.6 3.5 1.4 0.2 1.0 –1.3 –1.2 14.3 –5.4 Chad 3.7 4.6 2.9 7.3 2.3 5.4 8.1 0.0 7.7 –0.5 6.5 3.7 18.6 Chile 4.2 7.9 20.7 9.4 5.9 5.2 3.5 6.8 2.9 7.7 6.9 9.8 13.9 China 10.2 11.1 5.9 9.8 5.9 4.3 11.1 15.4 13.7 9.3 11.5 14.9 13.4 Hong Kong, China 6.9 4.4 7.7 6.4 .. .. .. .. .. .. 14.4 9.5 8.9 Colombia 3.6 4.2 24.8 21.7 2.9 1.6 5.0 2.9 2.8 4.9 7.5 6.8 13.6 Congo, Dem. Rep. 1.6 –5.1 62.9 1,423.1 2.5 2.9 0.9 –11.7 1.3 –15.2 9.6 –5.5 –3.5 Congo, Rep. 3.3 1.0 0.5 7.1 3.4 1.6 5.2 0.2 2.1 1.4 5.1 4.9 4.1 Costa Rica 3.0 3.7 23.6 17.4 3.1 2.8 2.8 3.3 3.1 4.3 6.1 8.7 2.8 Côte d’Ivoire 0.7 3.5 2.8 8.7 0.3 2.4 4.4 5.1 –0.3 3.5 1.9 4.5 18.0 Croatia .. –1.0 .. 218.1 .. –4.4 .. –8.2 .. –3.9 .. 0.9 1.2 Czech Republic 1.7 –0.2 1.5 17.1 .. .. .. .. .. .. .. 7.0 5.0 Denmark 2.3 2.8 5.6 1.7 3.1 1.7 2.9 1.9 2.6 1.4 4.3 3.7 0.1 Dominican Republic 3.1 5.5 21.6 10.6 0.4 3.6 3.6 6.1 3.5 5.6 4.5 20.4 11.8 Ecuador 2.0 2.9 36.4 32.7 4.4 2.7 1.2 3.7 1.7 2.5 5.4 4.4 4.2 Egypt, Arab Rep. 5.4 4.2 13.7 9.7 2.7 2.9 5.2 4.2 6.6 4.1 5.2 4.3 4.2 El Salvador 0.2 5.3 16.3 9.1 –1.1 1.3 0.1 5.4 0.7 6.3 –3.4 13.2 7.7 Eritrea .. 5.2 .. 10.1 .. .. .. .. .. .. .. 4.7 .. Estonia 2.2 –2.1 2.3 75.5 .. –4.3 .. –5.9 .. 0.5 .. .. –3.6 Ethiopia a 1.1 4.9 4.6 7.9 0.2 2.8 0.4 6.5 3.1 6.4 2.4 9.0 15.4 Finland 3.3 2.0 6.8 1.8 –0.2 0.2 3.3 2.1 4.1 –0.1 2.2 9.2 –5.5 France 2.3 1.5 6.0 1.7 2.0 0.4 1.1 0.1 3.0 1.6 3.7 4.1 –2.0 Georgia 0.4 –16.3 1.9 1,033.2 .. .. .. .. .. .. .. .. .. Germany b 2.2 1.6 .. 2.2 1.7 0.8 1.2 .. 2.9 2.5 .. 2.8 0.8 Ghana 3.0 4.2 42.1 28.6 1.0 2.8 3.3 4.4 5.7 5.6 2.5 10.2 2.8 Greece 1.8 2.0 18.0 10.6 –0.1 2.0 1.3 –0.5 2.7 1.8 7.2 4.0 3.4 Guatemala 0.8 4.2 14.6 11.5 1.2 2.2 –0.2 4.3 0.9 5.0 –1.8 7.2 3.7 Guinea .. 5.0 .. 5.9 .. 4.4 .. 1.6 .. 7.8 .. 2.6 5.7 Haiti –0.2 –2.5 7.5 25.3 –0.1 –4.9 –1.7 –2.7 0.9 –0.7 1.2 –4.4 1.8 Honduras 2.7 3.6 5.7 20.8 2.7 3.2 3.3 3.8 2.5 3.8 1.1 2.2 9.1 Hungary 1.3 –0.2 8.9 22.8 1.7 –3.8 0.2 1.1 2.1 0.3 3.6 4.9 7.3 India 5.8 6.1 8.0 7.5 3.1 3.4 7.0 6.7 6.9 7.9 5.9 12.4 5.9 Indonesia 6.1 5.8 8.5 12.5 3.4 2.8 6.9 9.9 7.0 7.2 2.9 8.6 4.4 Iran, Islamic Rep. 1.7 4.0 14.4 32.5 4.5 4.8 3.3 3.8 –1.0 6.0 6.9 2.4 –0.8 Ireland 3.2 7.5 6.6 1.9 .. .. .. .. .. .. 9.0 12.6 1.9 Israel 3.5 5.4 101.1 10.9 .. .. .. .. .. .. 5.5 8.6 8.9 Italy 2.4 1.2 10.0 4.4 0.1 1.3 2.0 0.8 2.8 1.1 4.1 7.5 –1.9 Jamaica 2.0 0.1 18.6 29.5 0.6 2.3 2.4 –0.4 1.8 0.2 5.4 0.0 6.0 Japan 4.0 1.3 1.7 0.4 1.3 –2.0 4.2 0.2 3.9 2.0 4.5 3.9 0.2 Jordan 2.5 5.4 4.3 3.3 6.8 –3.1 1.7 6.8 2.0 5.3 5.9 7.8 4.4 Kazakhstan .. –6.9 .. 329.9 .. –12.7 .. –10.2 .. 2.1 .. –0.3 –15.3 Kenya 4.2 2.2 9.1 15.0 3.3 1.2 3.9 2.0 4.9 3.5 4.3 2.7 4.3 Korea, Rep. 9.4 6.2 6.1 5.1 2.8 2.1 12.1 7.5 9.0 7.8 12.0 15.7 6.3 Kuwait 1.3 .. –2.8 .. 14.7 .. 1.0 .. 2.1 .. –2.3 .. .. Kyrgyz Republic .. –7.3 .. 157.8 .. –1.2 .. –12.0 .. –7.2 .. –1.8 8.6 Lao PDR .. 6.7 .. 12.2 3.4 4.5 6.1 11.9 3.4 6.7 .. .. .. Latvia 3.5 –8.5 0.0 87.7 2.3 –10.8 4.3 –15.9 3.2 –0.2 .. –0.6 –25.1 Lebanon .. 7.7 .. 24.0 .. 3.2 .. 2.1 .. 2.6 .. 15.6 18.4 Lesotho 4.4 7.2 13.8 7.7 2.2 6.0 7.1 9.2 4.6 6.2 4.1 11.1 11.1 Lithuania .. –5.2 .. 111.5 .. –1.4 .. –10.1 .. –0.4 .. .. .. Macedonia, FYR .. –0.1 .. 44.8 .. 1.9 .. –4.6 .. –0.6 .. 0.6 2.1 Madagascar 1.1 1.3 17.1 22.1 2.5 1.5 0.9 1.5 0.3 1.5 –1.7 1.3 0.4 Malawi 2.5 3.9 14.6 32.8 2.0 8.9 2.9 1.3 3.6 0.1 2.5 4.7 –8.0 Malaysia 5.3 7.7 1.7 4.5 3.8 2.0 7.2 10.8 4.2 8.8 10.9 13.2 10.8 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      ECONOMY Average annual % growth Gross GDP Agriculture Industry Services Exports of goods Gross domestic domestic product implicit deflator value added value added value added and services investment Economy 1980–90 1990–98 1980–90 1990–98 1980–90 1990–98 1980–90 1990–98 1980–90 1990–98 1980–90 1990–98 1990–98 Mali 0.9 3.7 4.5 9.2 3.3 3.3 4.3 7.6 1.9 2.2 4.8 9.2 1.5 Mauritania 1.8 4.2 8.4 5.9 1.7 5.0 4.9 3.4 0.4 4.6 3.6 –2.3 4.0 Mexico 0.7 2.5 72.1 19.8 0.8 1.4 1.1 3.2 0.6 2.4 7.0 14.7 2.4 Moldova 3.0 –14.1 .. 222.5 .. –7.1 .. –13.0 .. –19.9 .. 6.4 –21.9 Mongolia 5.4 0.1 –1.6 78.4 1.4 1.9 6.7 –2.0 5.8 1.2 .. .. .. Morocco 4.2 2.1 7.1 3.8 6.7 0.3 3.0 3.2 4.2 2.1 6.8 6.6 1.3 Mozambique –0.1 5.7 38.3 41.3 6.6 4.8 –4.5 8.5 8.1 5.3 –6.8 14.8 8.9 Myanmar 0.6 6.3 12.2 24.2 0.5 5.0 0.5 10.1 0.8 6.4 1.9 8.8 13.0 Namibia 0.9 3.5 13.9 9.5 1.8 2.9 –1.2 3.3 1.5 3.6 –0.1 5.4 4.1 Nepal 4.6 4.8 11.1 9.0 4.0 2.3 8.7 7.3 3.9 9.6 3.9 16.8 6.0 Netherlands 2.3 2.6 1.6 2.1 3.4 3.7 1.6 1.2 2.6 2.3 4.5 4.5 0.6 New Zealand 1.8 3.2 10.8 1.7 3.8 2.2 1.1 3.7 1.9 3.5 4.0 5.8 8.8 Nicaragua –2.0 4.1 422.6 67.7 –5.8 8.7 2.1 –4.8 –2.8 2.0 –7.8 10.6 9.8 Niger –0.1 1.9 1.9 6.8 1.7 2.2 –1.7 1.7 –0.7 1.6 –2.9 –0.2 4.4 Nigeria 1.6 2.6 16.7 38.6 3.3 2.9 –1.1 1.2 3.7 3.6 –0.3 5.2 8.0 Norway 2.8 3.9 5.6 1.8 –0.2 4.5 3.3 5.6 2.7 3.1 5.2 5.9 4.1 Pakistan 6.3 4.1 6.7 11.2 4.3 3.8 7.3 5.0 6.8 4.6 8.4 3.2 2.7 Panama 0.5 4.3 1.9 2.4 2.5 1.7 –1.3 6.3 0.7 4.1 –0.6 0.7 12.9 Papua New Guinea 1.9 5.7 5.3 6.7 1.8 4.1 1.9 8.9 2.0 4.3 3.3 10.6 8.2 Paraguay 2.5 2.8 24.4 14.5 3.6 2.9 0.3 3.1 3.1 2.6 12.2 7.3 3.6 Peru –0.3 5.9 231.3 33.7 2.7 5.5 –0.9 7.1 –0.7 4.9 –1.6 8.2 11.3 Philippines 1.0 3.3 14.9 8.5 1.0 1.5 –0.9 3.6 2.8 3.8 3.5 11.0 4.4 Poland 1.8 4.5 53.8 27.0 –0.7 –1.6 –1.3 4.7 2.8 3.0 4.5 12.3 10.6 Portugal 3.1 2.3 18.0 5.8 .. –0.4 .. 0.5 .. 2.3 8.7 4.8 2.2 Romania 0.5 –0.6 2.5 113.3 .. –0.2 .. –0.8 .. –0.2 .. 6.1 –8.3 Russian Federation .. –7.0 .. 235.3 .. –6.9 .. –8.1 .. –4.7 .. 2.0 –14.8 Rwanda 2.2 –3.3 4.0 18.4 0.5 –5.2 2.5 –0.6 5.5 –2.9 3.4 –9.8 –3.9 Saudi Arabia 0.0 1.6 –4.9 1.0 13.4 0.7 –2.3 1.5 1.3 2.0 .. .. .. Senegal 3.1 3.0 6.5 6.1 2.8 1.4 4.3 4.0 2.8 3.1 3.7 2.3 2.2 Sierra Leone 0.3 –4.7 64.0 32.5 3.1 1.5 1.7 –7.8 –2.8 –3.1 2.1 –9.4 –13.3 Singapore 6.6 8.0 2.2 2.5 –6.2 2.1 5.4 8.8 7.5 8.4 10.8 13.3 9.8 Slovak Republic 2.0 0.6 1.8 12.6 1.6 –0.4 2.0 –6.5 0.8 8.1 .. 12.1 2.1 Slovenia .. 1.4 .. 32.3 .. 0.2 .. 0.8 .. 3.8 .. –2.3 9.0 South Africa 1.2 1.6 14.9 9.8 2.9 2.7 0.0 0.9 2.4 1.8 1.9 5.1 3.4 Spain 3.0 1.9 9.3 4.2 .. –2.5 .. –0.4 .. –13.1 5.7 10.4 –1.5 Sri Lanka 4.0 5.3 11.0 9.8 2.2 1.5 4.6 6.5 4.7 6.3 4.9 9.0 5.8 Sweden 2.3 1.2 7.4 2.3 1.5 –1.9 2.8 –0.7 2.6 –0.1 4.3 7.6 –3.2 Switzerland 2.0 0.4 3.4 1.7 .. .. .. .. .. .. 3.5 1.6 –0.9 Syrian Arab Republic 1.5 5.9 15.3 8.9 –0.6 .. 6.6 .. 0.1 .. 7.3 5.4 8.3 Tajikistan .. –16.4 .. 394.3 .. .. .. .. .. .. .. .. .. Tanzania c .. 2.9 .. 24.3 .. 3.7 .. 1.8 .. 2.3 .. 10.9 –2.3 Thailand 7.6 7.4 3.9 4.8 3.9 3.1 9.8 9.0 7.3 7.1 14.1 11.1 6.5 Togo 1.7 2.3 4.8 8.8 5.6 4.5 1.1 2.6 –0.3 0.2 0.1 0.8 12.6 Tunisia 3.3 4.4 7.4 4.8 2.8 1.7 3.1 4.5 3.5 5.2 5.6 5.1 3.1 Turkey 5.4 4.1 45.2 79.3 1.3 1.1 7.8 5.0 4.4 4.1 .. 12.1 4.2 Turkmenistan .. –9.6 . . 1,074.2 .. .. .. .. .. .. .. .. .. Uganda 3.2 7.4 104.0 15.3 2.1 3.6 5.0 13.3 2.8 8.3 1.8 16.1 10.0 Ukraine .. –13.1 .. 591.0 .. –21.4 .. –16.4 .. –8.6 .. –3.2 –15.4 United Kingdom 3.2 2.2 5.7 3.0 .. .. .. .. .. .. 3.9 5.5 1.4 United States 3.0 2.9 4.2 2.2 .. 2.0 .. 4.3 .. 1.9 4.7 8.1 5.8 Uruguay 0.4 3.9 61.3 40.4 0.0 4.2 –0.2 1.2 0.8 5.1 4.3 8.0 8.3 Uzbekistan .. –1.9 .. 355.1 .. –1.6 .. –5.0 .. –0.9 .. .. .. Venezuela 1.1 2.0 19.3 49.7 3.0 1.1 1.6 3.5 0.4 0.5 2.8 5.4 3.9 Vietnam 4.6 8.6 210.8 19.7 4.3 5.1 .. 13.3 .. 8.8 .. 27.7 28.4 Yemen, Rep. .. 3.8 .. 24.2 .. 4.3 .. 6.4 .. 1.0 .. 6.9 8.8 Zambia 1.0 1.0 42.2 63.5 3.6 –4.9 0.8 –4.7 –0.4 8.9 –3.4 2.0 12.1 Zimbabwe 3.6 2.0 11.6 22.4 3.1 3.4 3.2 –1.0 3.1 3.1 4.3 8.9 4.5 World 3.2 w 2.4 w 2.7 w 1.2 w ..w 2.1 w ..w 2.0 w 5.2 w 6.4 w 2.3 w Low income 6.6 7.3 4.1 3.5 7.8 11.0 8.0 7.3 5.9 11.1 9.9 Excl. China & India 4.1 3.6 3.0 2.5 4.6 5.9 5.0 4.7 2.7 7.0 5.2 Middle income 2.6 1.9 2.6 –0.2 2.5 1.6 2.7 2.7 6.1 7.5 1.9 Lower middle income .. –1.3 .. –2.2 .. –2.8 .. 0.4 .. 2.8 –4.0 Upper middle income 2.7 3.9 2.5 1.9 2.5 4.4 2.7 4.0 7.6 11.5 5.9 Low and middle income 3.5 3.3 3.4 1.7 3.7 4.2 3.7 3.7 6.1 8.4 4.2 East Asia & Pacific 8.0 8.1 4.4 3.5 9.5 11.5 8.8 7.9 9.6 14.0 10.6 Europe & Central Asia .. –4.3 .. –6.3 .. –5.5 .. –1.4 .. 3.9 –7.5 Latin America & Carib. 1.6 3.7 2.1 2.6 1.2 3.7 1.6 3.4 5.4 9.3 5.7 Middle East & N. Africa 2.0 3.0 5.5 1.7 0.6 2.2 2.1 3.6 .. .. .. South Asia 5.7 5.7 3.2 3.2 6.8 6.5 6.5 7.1 6.6 10.5 5.7 Sub-Saharan Africa 1.8 2.2 2.5 2.6 0.9 1.2 2.4 2.1 2.4 4.6 4.2 High income 3.1 2.1 .. 0.3 .. 1.5 .. 1.8 5.1 6.1 1.7 a. Data prior to 1992 include Eritrea. b. Data prior to 1990 refer to the Federal Republic of Germany before unification. c. Data cover mainland Tanzania only. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 12. Structure of output Gross domestic product Value added as a % of GDP Millions of dollars Agriculture Industry Manufacturing Services Economy 1980 1998 1980 1998 1980 1998 1980 1998 1980 1998 Albania .. 2,460 34 63 45 18 .. .. 21 19 Algeria 42,345 49,585 10 12 54 47 9 9 36 41 Angola .. 6,648 .. 14 .. 54 .. 5 .. 32 Argentina 76,962 344,360 6 7 41 37 29 25 52 56 Armenia .. 1,628 .. 41 .. 36 .. 25 .. 23 Australia 160,110 364,247 5 3 36 26 19 14 58 71 Austria 78,539 212,069 4 1 36 30 25 20 60 68 Azerbaijan .. 4,127 .. 19 .. 44 .. 9 .. 36 Bangladesh 17,430 42,775 34 23 24 28 18 18 42 49 Belarus .. 22,629 .. 14 .. 44 .. 37 .. 42 Belgium 119,938 247,076 2 1 34 27 21 18 64 72 Benin 1,405 2,322 35 39 12 14 8 8 52 47 Bolivia 2,750 8,558 .. 16 .. 33 .. 4 .. 52 Botswana 1,105 5,690 11 4 45 46 5 5 44 51 Brazil 234,873 778,292 11 8 44 36 33 23 45 56 Bulgaria 20,040 10,085 14 23 54 26 .. 18 32 50 Burkina Faso 1,709 2,581 33 32 22 28 16 21 45 40 Burundi 920 949 62 49 13 19 7 11 25 32 Cambodia .. 3,089 .. 51 .. 15 .. 6 .. 34 Cameroon 6,741 8,736 31 42 26 22 10 11 43 36 Canada 266,002 598,847 4 .. 38 .. 19 .. 58 .. Central African Republic 797 1,057 40 55 20 18 7 9 40 27 Chad 1,033 1,603 45 39 9 15 .. 12 46 46 Chile 27,572 78,025 7 8 37 35 22 17 55 57 China 201,687 960,924 30 18 49 49 41 37 21 33 Hong Kong, China 28,495 166,554 1 0 32 15 24 7 67 85 Colombia 33,399 91,108 22 13 35 38 26 19 43 49 Congo, Dem. Rep. 14,922 6,964 25 58 33 17 14 .. 42 25 Congo, Rep. 1,706 1,961 12 12 47 50 7 8 42 39 Costa Rica 4,815 10,252 18 14 27 22 19 16 55 64 Côte d’Ivoire 10,175 11,041 26 25 20 23 13 19 54 52 Croatia .. 19,081 .. .. .. .. .. .. .. .. Czech Republic 29,042 52,035 7 .. 63 .. .. .. 30 .. Denmark 67,791 174,272 5 .. 29 .. 20 .. 66 .. Dominican Republic 6,631 15,489 20 12 28 33 15 17 52 56 Ecuador 11,733 19,766 12 12 38 34 18 22 50 54 Egypt, Arab Rep. 22,912 78,097 18 17 37 33 12 26 45 50 El Salvador 3,574 12,148 38 13 22 28 16 22 40 59 Eritrea .. 650 .. 9 .. 30 .. 16 .. 61 Estonia .. 5,462 .. 5 .. 27 .. 17 .. 67 Ethiopiaa 5,179 6,568 56 .. 12 .. 8 .. 32 .. Finland 51,306 125,673 10 4 40 34 28 25 51 62 France 664,596 1,432,902 4 2 34 26 24 19 62 72 Georgia .. 5,244 24 32 36 23 28 18 40 45 Germany .. 2,142,018 .. 1 .. .. .. 24 .. 44 Ghana 4,445 7,501 58 37 12 25 8 8 30 38 Greece 48,613 120,304 14 .. 25 .. 16 .. 61 .. Guatemala 7,879 19,281 25 21 22 19 17 13 53 60 Guinea .. 3,615 .. 22 .. 35 .. 4 .. 42 Haiti 1,462 2,815 .. 31 .. 20 .. .. .. 48 Honduras 2,566 4,722 24 23 24 30 15 18 52 47 Hungary 22,186 45,725 19 6 47 34 .. 25 34 60 India 186,439 383,429 38 25 24 30 16 19 39 45 Indonesia 78,013 96,265 24 16 42 43 13 26 34 41 Iran, Islamic Rep. 92,664 .. 18 .. 32 .. 9 .. 50 .. Ireland 20,080 80,880 .. .. .. .. .. .. .. .. Israel 21,885 100,031 .. .. .. .. .. .. .. .. Italy 449,913 1,171,044 6 3 39 31 28 20 55 66 Jamaica 2,652 6,607 8 7 38 35 17 16 54 58 Japan 1,059,254 3,783,140 4 .. 42 .. 29 .. 54 .. Jordan 3,962 7,015 8 3 28 25 13 13 64 72 Kazakhstan .. 21,029 .. 10 .. 27 .. 12 .. 63 Kenya 7,265 11,083 33 29 21 16 13 10 47 55 Korea, Rep. 62,803 297,900 15 6 40 43 28 26 45 51 Kuwait 28,639 30,373 0 .. 75 .. 6 .. 25 .. Kyrgyz Republic .. 1,704 .. 46 .. 24 .. 18 .. 30 Lao PDR .. 1,753 .. 52 .. 21 .. 16 .. 27 Latvia .. 5,527 12 7 51 31 46 21 37 62 Lebanon .. 17,073 .. 12 .. 27 .. 17 .. 61 Lesotho 369 792 24 11 29 42 7 17 47 47 Lithuania .. 10,517 .. 14 .. 40 .. 26 .. 46 Macedonia, FYR .. 2,201 .. 12 .. 27 .. 0 .. 61 Madagascar 4,042 3,749 30 31 16 14 .. 11 54 56 Malawi 1,238 1,643 44 39 23 19 14 15 34 41 Malaysia 24,488 71,302 22 12 38 48 21 34 40 40 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      ECONOMY Gross domestic product Value added as a % of GDP Millions of dollars Agriculture Industry Manufacturing Services Economy 1980 1998 1980 1998 1980 1998 1980 1998 1980 1998 Mali 1,787 2,695 48 45 13 21 7 6 38 34 Mauritania 709 971 30 24 26 30 .. 9 44 45 Mexico 223,505 393,224 8 5 33 27 22 20 59 68 Moldova .. 1,872 .. 31 .. 35 .. 28 .. 34 Mongolia .. 1,043 15 33 33 28 .. .. 52 40 Morocco 18,821 33,514 18 16 31 30 17 17 51 54 Mozambique 3,526 3,959 48 34 30 18 .. 10 22 48 Myanmar .. .. 47 59 13 10 10 7 41 31 Namibia 2,262 3,108 11 10 55 34 9 14 34 56 Nepal 1,946 4,479 62 40 12 22 4 10 26 38 Netherlands 171,861 382,487 3 .. 32 .. 18 .. 64 .. New Zealand 22,395 54,093 11 .. 31 .. 22 .. 58 .. Nicaragua 2,144 1,971 23 34 31 22 26 16 45 44 Niger 2,509 2,048 43 41 23 17 4 6 34 42 Nigeria 64,202 41,353 21 32 46 41 8 5 34 27 Norway 63,419 145,896 4 2 35 32 15 11 61 66 Pakistan 23,690 63,895 30 25 25 25 16 17 46 50 Panama 3,810 9,218 9 7 19 17 11 9 72 76 Papua New Guinea 2,548 4,639 33 28 27 36 10 9 40 36 Paraguay 4,579 8,571 29 25 27 22 16 15 44 53 Peru 20,658 64,122 10 7 42 38 20 22 48 55 Philippines 32,500 65,096 25 17 39 32 26 22 36 52 Poland 57,068 148,863 .. 4 .. 26 .. 17 .. 70 Portugal 28,729 106,650 .. .. .. .. .. .. .. .. Romania .. 34,843 .. 15 .. 36 .. 25 .. 48 Russian Federation .. 446,982 .. 9 .. 42 .. .. .. 49 Rwanda 1,163 2,082 50 34 23 23 17 16 27 43 Saudi Arabia 156,487 125,840 1 6 81 45 5 10 18 49 Senegal 2,986 4,836 19 17 15 23 11 15 66 59 Sierra Leone 1,199 647 33 44 21 24 5 6 47 32 Singapore 11,718 85,425 1 0 38 35 29 24 61 65 Slovak Republic .. 19,461 .. 5 .. 33 .. .. .. 62 Slovenia .. 18,201 .. 5 .. 39 .. 29 .. 57 South Africa 78,744 116,730 7 4 50 38 23 24 43 57 Spain 213,308 551,923 .. 3 .. .. .. 18 .. .. Sri Lanka 4,032 15,093 28 22 30 26 18 17 43 52 Sweden 125,557 224,953 4 .. 34 .. 23 .. 63 .. Switzerland 107,474 264,352 .. .. .. .. .. .. .. .. Syrian Arab Republic 13,062 17,899 20 .. 23 .. .. .. 56 .. Tajikistan .. 1,990 .. .. .. .. .. .. .. .. Tanzania b .. 7,917 .. 46 .. 14 .. 7 .. 40 Thailand 32,354 153,909 23 11 29 40 22 29 48 49 Togo 1,136 1,510 27 42 25 21 8 9 48 37 Tunisia 8,742 22,041 14 14 31 28 12 18 55 58 Turkey 68,824 189,878 26 15 22 28 14 18 51 57 Turkmenistan .. 4,397 .. .. .. .. .. .. .. .. Uganda 1,244 6,653 72 43 4 18 4 9 23 39 Ukraine .. 49,677 .. 12 .. 40 .. 6 .. 48 United Kingdom 537,389 1,357,429 2 2 43 31 27 21 55 67 United States 2,709,000 8,210,600 3 2 33 27 22 18 64 71 Uruguay 10,132 20,155 14 8 34 27 26 18 53 64 Uzbekistan .. 14,194 .. 28 .. 30 .. 13 .. 42 Venezuela 69,417 105,756 5 4 46 43 16 17 49 52 Vietnam .. 24,848 .. 26 .. 31 .. .. .. 43 Yemen, Rep. .. 4,318 .. 18 .. 49 .. 11 .. 34 Zambia 3,884 3,352 14 16 41 30 18 12 44 55 Zimbabwe 6,679 5,908 16 18 29 24 22 17 55 58 World 10,939,459 t 28,854,043 t 7w 5w 38 w ..w 25 w 20 w 56 w 61 w Low income 801,498 1,811,106 31 21 38 41 27 29 30 38 Excl. China & India 451,756 451,051 29 25 32 33 13 18 39 42 Middle income 2,303,442 4,420,845 13 9 41 36 25 21 46 56 Lower middle income .. 1,704,528 .. 12 .. 36 .. .. .. 52 Upper middle income 1,165,003 2,816,378 11 7 42 35 26 22 47 57 Low and middle income 3,106,342 6,251,315 18 12 40 37 25 23 42 51 East Asia & Pacific 503,834 1,688,394 24 15 42 45 31 31 33 41 Europe & Central Asia .. 1,137,953 .. 11 .. 34 .. .. .. 55 Latin America & Carib. 782,173 2,076,540 10 8 40 34 29 22 50 58 Middle East & N. Africa .. .. 10 .. 53 .. 9 .. 37 .. South Asia 237,343 517,654 36 25 24 29 16 19 40 46 Sub-Saharan Africa 270,391 316,517 18 17 39 34 16 19 43 50 High income 7,936,460 22,560,624 3 2 37 .. 25 19 59 65 a. Data prior to 1992 include Eritrea. b. Data cover mainland Tanzania only. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 13. Structure of demand Percentage of GDP Private General government Gross domestic Gross domestic Exports of goods Resource consumption consumption investment saving and services balance Economy 1980 1998 1980 1998 1980 1998 1980 1998 1980 1998 1980 1998 Albania 56 103 9 11 35 12 35 –13 23 12 0 –25 Algeria 43 56 14 11 39 27 43 33 34 29 4 6 Angola .. 48 .. 39 .. 25 .. 13 .. 57 .. –12 a Argentina 76 78 .. 3 25 22 24 19 5 9 –1 –2 Armenia .. 116 .. 13 .. 9 .. –29 .. 20 .. –38 Australia 59 63 18 17 25 20 24 21 16 21 –2 –1 Austria 55 57 18 20 29 24 27 23 36 42 –2 –1 Azerbaijan .. 90 .. 11 .. 34 .. –1 .. 25 .. –35 Bangladesh 86 80 2 4 22 21 13 15 4 14 –9 –6 Belarus .. 59 .. 19 .. 26 .. 22 .. 60 .. –4 Belgium 64 63 18 15 22 18 19 22 57 73 –3 5 Benin 96 81 9 10 15 16 –5 9 23 24 –20 –8 Bolivia 67 75 14 15 17 19 19 9 25 15 2 –9 Botswana 46 40 20 25 37 25 34 35 50 45 –2 10 Brazil 70 67 9 14 23 21 21 19 9 7 –2 –2 Bulgaria 55 70 6 12 34 12 39 17 36 61 5 6 Burkina Faso 95 77 10 12 17 26 –6 11 10 14 –23 –14 Burundi 91 90 9 11 14 8 –1 –1 9 6 –14 –9 Cambodia .. 87 .. 9 .. 16 .. 4 .. 30 .. –12 Cameroon 69 71 10 9 21 18 22 20 28 27 1 2 Canada 53 58 22 21 23 18 25 21 28 41 2 2 Central African Republic 94 84 15 12 7 14 –9 4 25 16 –16 –9 Chad 100 92 4 7 3 19 –9 1 17 17 –12 –18 Chile 71 72 12 6 21 27 17 22 23 25 –4 –5 China 51 44 15 13 35 39 35 43 6 22 0 5 Hong Kong, China 60 60 6 9 35 30 34 30 90 125 –1 0 Colombia 70 77 10 9 19 18 20 14 16 17 1 –4 Congo, Dem. Rep. 82 83 8 8 10 8 10 9 16 24 0 2 Congo, Rep. 47 59 18 14 36 35 36 26 60 63 0 –9 Costa Rica 66 63 18 13 27 27 16 24 26 43 –10 –3 Côte d’Ivoire 63 65 17 11 27 18 20 24 35 43 –6 6 Croatia .. 66 .. 30 .. 15 .. 3 .. 42 .. –11 Czech Republic .. 51 .. 20 31 34 .. 28 .. 58 .. –5 Denmark 56 .. 27 .. 18 .. 16 .. 33 .. –2 .. Dominican Republic 77 72 8 10 25 26 15 19 19 32 –10 –7 Ecuador 60 68 15 15 26 21 26 17 25 25 0 –4 Egypt, Arab Rep. 69 80 16 10 28 19 15 10 31 17 –12 –9 El Salvador 72 86 14 10 13 17 14 5 34 24 1 –12 Eritrea .. 81 .. 48 .. 41 .. –29 .. 20 .. –70 Estonia .. 62 .. 21 .. 26 .. 17 .. 76 .. –9 Ethiopiab 79 77 14 14 13 20 7 9 11 16 –6 –11 Finland 54 53 18 22 29 17 28 25 33 40 –1 9 France 59 61 18 19 24 17 23 20 22 24 –1 3 Georgia 56 95 13 9 29 7 31 –4 .. 12 2 –11 Germany .. 58 .. 20 .. 21 .. 22 .. 27 .. 2 Ghana 84 77 11 10 6 23 5 13 8 27 –1 –10 Greece 62 75 12 14 33 19 27 11 16 15 –6 –9 Guatemala 79 88 8 5 16 14 13 7 22 17 –3 –7 Guinea .. 74 .. 7 .. 22 .. 19 .. 22 .. –3 Haiti 82 97 10 7 17 10 8 –4 22 8 –9 –15 Honduras 70 62 13 13 25 30 17 25 36 42 –8 –5 Hungary 61 63 10 10 31 27 29 27 39 45 –2 0 India 73 71 10 11 20 23 17 18 6 12 –3 –5 Indonesia 51 63 11 7 24 31 38 31 34 28 14 0 Iran, Islamic Rep. 53 .. 21 .. 30 .. 26 .. 13 .. –3 .. Ireland 67 53 19 14 27 18 14 33 48 76 –13 15 Israel 53 62 40 29 22 22 7 9 44 32 –16 –13 Italy 61 61 15 16 27 17 24 22 22 27 –3 4 Jamaica 64 54 20 21 16 34 16 24 51 49 0 –9 Japan 59 .. 10 .. 32 .. 31 .. 14 .. –1 .. Jordan 79 68 29 25 37 27 –8 6 40 50 –44 –21 Kazakhstan .. 75 .. 12 .. 16 .. 13 .. 34 .. –3 Kenya 62 72 20 15 29 18 18 13 28 26 –11 –5 Korea, Rep. 64 55 12 11 32 35 24 34 34 38 –7 –1 Kuwait 31 47 11 28 14 13 58 25 78 53 44 12 Kyrgyz Republic .. 82 .. 16 .. 18 .. 2 .. 35 .. –16 Lao PDR .. 81 .. 7 .. 29 .. 11 .. 24 .. –17 Latvia 59 67 8 23 26 20 33 10 .. 50 7 –10 Lebanon .. 98 .. 15 .. 28 .. –13 .. 11 .. –40 Lesotho 133 121 26 22 43 49 –59 –43 20 33 –102 –91 Lithuania .. 67 .. 20 .. 28 .. 14 .. 50 .. –14 Macedonia, FYR .. 83 .. 12 .. 20 .. 4 .. 45 .. –15 Madagascar 89 89 12 6 15 13 –1 5 13 21 –16 –8 Malawi 70 80 19 14 25 18 11 5 25 33 –14 –13 Malaysia 51 42 17 11 30 32 33 47 58 118 3 15 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      ECONOMY Percentage of GDP Private General government Gross domestic Gross domestic Exports of goods Resource consumption consumption investment saving and services balance Economy 1980 1998 1980 1998 1980 1998 1980 1998 1980 1998 1980 1998 Mali 87 77 12 13 15 21 0 10 15 24 –14 –11 Mauritania 68 80 25 13 36 22 7 7 37 40 –29 –15 Mexico 65 68 10 8 27 26 25 24 11 31 –2 –2 Moldova .. 74 .. 26 .. 24 .. 0 .. 53 .. –24 Mongolia 44 60 29 16 63 23 27 24 21 68 –36 1 Morocco 68 65 18 16 24 22 14 18 17 28 –10 –3 Mozambique 98 90 12 9 6 21 –11 1 11 12 –16 –20 Myanmar 82 88 . .a . .a 21 13 18 12 9 1 –4 –1 Namibia 47 56 17 26 29 19 37 19 76 63 8 0 Nepal 82 82 7 9 18 21 11 9 12 23 –7 –11 Netherlands 61 60 17 14 22 20 22 26 51 56 0 7 New Zealand 62 63 18 14 21 22 20 22 30 29 –1 1 Nicaragua 82 84 20 13 17 28 –2 3 24 41 –19 –25 Niger 75 84 10 13 28 10 15 3 25 16 –14 –7 Nigeria 56 77 12 11 21 20 31 12 29 23 10 –8 Norway 47 48 19 20 28 23 34 32 43 41 6 7 Pakistan 83 77 10 10 18 17 7 13 12 16 –12 –4 Panama 52 57 18 18 28 27 31 25 51 36 2 –2 Papua New Guinea 61 44 24 23 25 37 15 33 43 56 –10 –4 Paraguay 76 73 6 10 32 21 18 17 15 45 –13 –4 Peru 57 68 11 12 29 25 32 20 22 12 3 –5 Philippines 67 73 9 13 29 25 24 15 24 56 –5 –11 Poland 67 65 9 16 26 24 23 20 28 25 –3 –4 Portugal 65 65 13 18 34 24 21 17 25 31 –13 –9 Romania 60 77 5 10 40 20 35 13 35 24 –5 –7 Russian Federation .. 67 .. 10 .. 20 .. 24 .. 27 .. 3 Rwanda 83 96 12 11 16 10 4 –7 14 5 –12 –17 Saudi Arabia 22 35 16 30 22 20 62 35 71 45 41 14 Senegal 85 75 20 10 12 20 –5 15 27 32 –17 –5 Sierra Leone .. 93 21 8 .. 8 .. –1 18 22 –10 –9 Singapore 53 39 10 10 46 37 38 51 215 .. –9 14 Slovak Republic .. 49 .. 22 .. 35 .. 28 .. 56 .. –7 Slovenia .. 57 .. 20 .. 24 .. 23 .. 57 .. –1 South Africa 50 61 13 22 28 16 36 17 36 29 8 1 Spain 66 62 13 16 23 21 21 21 16 26 –2 1 Sri Lanka 80 72 9 10 34 24 11 17 32 36 –23 –7 Sweden 51 52 29 26 21 15 19 21 29 44 –2 7 Switzerland 62 61 12 14 29 20 25 24 35 40 –3 4 Syrian Arab Republic 67 70 23 11 28 29 10 18 18 29 –17 –11 Tajikistan .. .. .. .. .. .. .. .. .. .. .. .. Tanzania c .. 85 .. 9 .. 16 .. 6 .. 16 .. –10 Thailand 65 54 12 10 29 35 23 36 24 47 –6 1 Togo 54 81 22 11 28 14 23 7 51 34 –5 –7 Tunisia 62 61 14 15 29 25 24 24 40 42 –5 –2 Turkey 77 68 12 12 18 25 11 19 5 25 –7 –6 Turkmenistan .. .. .. .. .. .. .. .. .. .. .. .. Uganda 89 84 11 10 6 15 0 6 19 10 –7 –10 Ukraine .. 62 .. 22 .. 20 .. 16 .. 41 .. –4 United Kingdom 59 64 22 21 17 16 19 15 27 29 2 0 United States 64 68 17 16 20 18 19 16 10 12 –1 –1 Uruguay 76 81 12 7 17 13 12 12 15 22 –6 –1 Uzbekistan .. 57 .. 22 .. 23 .. 22 .. 22 .. –1 Venezuela 55 78 12 6 26 16 33 16 29 17 7 –1 Vietnam .. 70 .. 9 .. 29 .. 21 .. 46 .. –8 Yemen, Rep. .. 76 .. 22 .. 22 .. 2 .. 34 .. –19 Zambia 55 84 26 11 23 14 19 5 41 29 –4 –9 Zimbabwe 68 63 19 17 17 21 14 20 23 45 –3 –2 World 61 w 63 w 15 w 16 w 25 w 20 w 24 w 21 w 20 w 25 w –1 w 0w Low income 60 57 12 12 28 30 28 32 12 19 0 0 Excl. China & India 65 70 11 9 22 24 24 20 25 27 2 –3 Middle income 63 63 12 14 26 24 25 23 22 22 –1 –2 Lower middle income .. 65 .. 14 .. 23 .. 22 .. 28 .. –2 Upper middle income 64 68 11 11 25 23 25 21 20 19 –1 –2 Low and middle income 62 65 12 12 27 25 26 24 19 21 –1 –1 East Asia & Pacific 56 52 13 11 32 36 31 37 21 34 –1 1 Europe & Central Asia .. 65 .. 14 .. 23 .. 21 .. 31 .. –2 Latin America & Carib. 68 70 10 10 24 22 22 20 12 14 –2 –2 Middle East & N. Africa 45 .. 18 .. 27 .. 38 .. 42 .. 11 .. South Asia 76 73 9 10 21 22 15 17 8 13 –5 –5 Sub-Saharan Africa 59 67 14 17 24 18 26 15 33 30 2 –3 High income 60 63 16 17 25 19 24 19 20 24 –1 0 a. General government consumption figures are not available separately; they are included in private consumption. b. Data prior to 1992 include Eritrea. c. Data cover mainland Tanzania only. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 14. Central government finances Percentage of GDP Percentage of total expenditureb Current Current Current Capital Overall Goods Social tax revenue nontax revenue expenditure expenditure deficit/surplusa and services servicesc Economy 1980 1997 1980 1997 1980 1997 1980 1997 1980 1997 1980 1997 1980 1997 Albania .. 16.6 .. 4.6 .. 25.5 .. 5.5 .. –9.0 .. 26.2 .. 33.0 Algeria .. .. .. .. .. .. .. .. .. .. .. .. .. .. Angola .. .. .. .. .. .. .. .. .. .. .. .. .. .. Argentina 10.4 11.2 5.2 1.1 18.2 12.7 0.0 1.1 –2.6 –1.3 57.1 21.4 28.6 63.6 Armenia .. .. .. .. .. .. .. .. .. .. .. .. .. .. Australia 19.5 23.2 2.2 1.9 21.1 25.3 1.5 0.9 –1.5 0.4 21.1 21.9 45.5 60.7 Austria 31.2 34.4 2.6 2.8 33.3 38.6 3.3 3.1 –3.3 –4.1 25.6 24.2 70.0 65.7 Azerbaijan .. .. .. .. .. .. .. .. .. .. .. .. .. .. Bangladesh 5.7 .. 2.7 .. .. .. .. .. 1.8 .. .. .. 20.1 .. Belarus .. 29.4 .. 2.4 .. 28.9 .. 5.1 .. –1.9 .. 29.0 .. 45.7 Belgium 41.2 43.0 1.8 1.1 45.9 45.9 4.2 2.4 –8.0 –3.2 22.2 18.6 60.2 .. Benin .. .. .. .. .. .. .. .. .. .. .. .. .. .. Bolivia .. 15.0 .. 1.9 .. 18.4 .. 3.5 .. –2.3 .. 37.7 .. 53.6 Botswana 21.9 14.7 8.0 29.5 20.3 28.5 9.5 6.8 –0.1 8.4 40.5 46.8 30.6 42.7 Brazil 17.8 .. 4.8 .. 18.6 .. 1.6 .. –2.4 .. 16.1 .. 32.3 .. Bulgaria .. 25.2 .. 6.8 .. 30.9 .. 2.6 .. 2.1 .. 32.9 .. 42.0 Burkina Faso 10.4 .. 1.2 .. 9.8 .. 2.3 .. 0.2 .. 69.4 .. 30.1 .. Burundi 13.2 12.7 0.8 1.0 11.5 17.3 10.9 3.7 –3.9 –5.5 39.2 55.2 .. 23.0 Cambodia .. .. .. .. .. .. .. .. .. .. .. .. .. .. Cameroon 14.9 9.4 1.3 3.6 10.5 11.4 5.2 1.1 0.5 0.2 54.7 52.6 25.4 21.0 Canada 16.0 .. 2.5 .. 20.8 .. 0.2 .. –3.5 .. 20.9 .. 43.8 .. Central African Republic 15.0 .. 1.5 .. 18.5 .. 1.3 .. –3.5 .. 66.0 .. 28.6 .. Chad .. .. .. .. .. .. .. .. .. .. .. .. .. .. Chile 25.6 18.9 6.4 3.8 25.3 17.3 2.7 3.5 5.4 1.9 40.2 28.8 57.6 66.2 China .. 4.9 .. 0.6 .. .. .. .. .. –1.6 .. .. .. 2.6 Hong Kong, China .. .. .. .. .. .. .. .. .. .. .. .. .. .. Colombia 10.3 .. 1.7 .. 10.4 .. 4.1 .. –1.8 .. 35.2 .. 44.1 .. Congo, Dem. Rep. 8.3 4.9 1.1 0.4 9.9 8.0 2.4 0.3 –0.8 0.0 65.2 94.5 22.1 1.6 Congo, Rep. 27.0 .. 8.3 .. 21.8 .. 17.7 .. –5.2 .. .. .. 23.0 .. Costa Rica 16.8 23.5 1.0 3.2 21.3 27.6 5.2 2.9 –7.4 –3.9 52.2 47.1 62.4 59.6 Côte d’Ivoire 21.1 .. 1.7 .. 19.1 .. 9.0 .. –10.8 .. .. .. .. .. Croatia .. 42.8 .. 2.6 .. 41.2 .. 5.5 .. –0.5 .. 47.9 .. 62.6 Czech Republic .. 32.7 .. 1.2 .. 32.6 .. 3.3 .. –1.1 .. 14.3 .. 71.3 Denmark 30.7 33.7 4.0 5.1 35.9 40.0 2.7 1.5 –2.6 –1.9 21.3 18.9 56.3 54.5 Dominican Republic 11.1 13.9 3.2 1.2 11.4 9.0 5.2 6.3 –2.6 –0.3 49.5 36.5 35.4 41.5 Ecuador 12.3 .. 0.5 .. 11.9 .. 2.3 .. –1.4 .. 28.2 .. 43.9 .. Egypt, Arab Rep. 28.8 21.5 15.2 13.9 39.5 27.7 10.8 6.6 –11.7 0.9 34.1 31.2 20.9 31.6 El Salvador 11.1 10.4 0.5 0.8 11.7 10.5 2.8 2.3 –5.7 –0.6 49.8 55.0 34.3 37.7 Eritrea .. .. .. .. .. .. .. .. .. .. .. .. .. .. Estonia .. 30.1 .. 3.4 .. 28.7 .. 2.9 .. 2.4 .. 42.0 .. 57.5 Ethiopia d 12.8 11.9 3.5 5.2 18.0 18.1 3.3 7.1 –3.1 –4.5 85.9 52.4 19.6 30.8 Finland 25.1 28.4 2.1 5.1 25.2 38.5 3.0 1.6 –2.2 –6.3 20.4 17.6 50.3 53.6 France 36.7 39.2 2.9 2.6 37.4 44.6 2.1 2.0 –0.1 –3.5 30.1 23.6 69.4 .. Georgia .. .. .. .. .. .. .. .. .. .. .. .. .. .. Germany .. 26.7 .. 5.0 .. 32.1 .. 1.3 .. –1.4 33.9 31.6 68.8 .. Ghana 6.4 .. 0.5 .. 9.8 .. 1.1 .. –4.2 .. 47.3 .. 35.1 .. Greece 22.6 20.6 2.7 2.4 25.7 28.5 4.6 4.3 –4.1 –8.5 44.3 29.3 51.2 36.8 Guatemala 8.7 8.7 0.7 0.7 7.3 6.8 5.1 2.3 –3.4 –1.0 46.6 53.1 29.8 .. Guinea .. .. .. .. .. .. .. .. .. .. .. .. .. .. Haiti 9.3 .. 1.3 .. 13.9 .. 3.5 .. –4.7 .. 81.6 .. .. .. Honduras 13.6 .. 0.9 .. .. .. .. .. .. .. .. .. .. .. Hungary 44.8 32.5 8.6 4.7 48.7 38.5 7.5 4.1 –2.8 –2.6 19.4 18.6 26.7 43.2 India 9.0 10.8 1.8 3.3 10.8 14.7 1.4 1.7 –6.0 –4.9 20.4 20.5 5.5 8.7 Indonesia 20.2 14.7 1.0 2.3 11.7 8.7 10.4 6.0 –2.3 1.2 23.7 27.9 11.8 36.2 Iran, Islamic Rep. 6.9 6.7 14.7 17.8 27.7 15.7 8.0 7.6 –13.8 1.4 57.3 55.8 36.7 41.1 Ireland 30.9 32.4 3.9 1.6 40.4 34.4 4.6 3.7 –12.5 –1.4 17.3 18.1 49.3 60.3 Israel 44.9 36.8 7.3 5.8 69.7 45.1 2.9 3.4 –16.2 0.4 46.2 35.0 25.7 59.9 Italy 29.3 42.2 2.5 2.5 37.8 45.4 2.2 2.5 –10.8 –3.1 17.1 18.5 48.8 .. Jamaica 27.8 .. 1.2 .. .. .. .. .. –15.5 .. .. .. .. .. Japan 11.0 .. 0.6 .. 14.8 .. 3.6 .. –7.0 .. 12.6 .. .. .. Jordan 14.0 22.4 4.0 6.3 25.9 28.0 12.1 7.0 –9.3 –1.4 39.5 60.0 23.0 44.7 Kazakhstan .. .. .. .. .. .. .. .. .. .. .. .. .. .. Kenya 19.2 23.4 2.8 3.7 19.4 25.6 5.9 3.4 –4.5 –0.9 52.9 44.5 30.3 29.6 Korea, Rep. 15.5 18.6 2.2 2.9 14.8 14.7 2.4 4.1 –2.2 –1.4 38.6 21.6 22.0 27.8 Kuwait 2.7 1.2 86.6 .. 18.9 35.8 8.9 5.8 58.7 .. 40.5 .. 24.0 .. Kyrgyz Republic .. .. .. .. .. .. .. .. .. .. .. .. .. .. Lao PDR .. .. .. .. .. .. .. .. .. .. .. .. .. .. Latvia .. 29.2 .. 3.3 .. 30.6 .. 1.5 .. 0.9 .. 30.6 .. 58.3 Lebanon .. 14.1 .. 3.3 .. 29.4 .. 8.5 .. –20.6 .. 30.8 .. 17.2 Lesotho 29.4 30.7 4.8 7.0 32.9 26.5 12.4 13.3 –7.4 1.0 50.0 54.3 22.8 .. Lithuania .. 25.4 .. 1.0 .. 25.0 .. 2.4 .. –1.9 .. 44.9 .. 50.2 Macedonia, FYR .. .. .. .. .. .. .. .. .. .. .. .. .. .. Madagascar 12.9 8.5 0.3 0.2 .. 10.5 .. 6.8 .. –1.3 .. 24.6 .. 16.5 Malawi 16.6 .. 2.5 .. 18.0 .. 16.6 .. –15.9 .. 32.4 .. 14.2 .. Malaysia 23.5 19.4 2.8 4.2 19.2 15.5 9.9 4.6 –6.0 3.0 33.5 40.5 26.8 42.5 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      ECONOMY Percentage of GDP Percentage of total expenditureb Current Current Current Capital Overall Goods Social tax revenue nontax revenue expenditure expenditure deficit/surplusa and services servicesc Economy 1980 1997 1980 1997 1980 1997 1980 1997 1980 1997 1980 1997 1980 1997 Mali 8.7 .. 0.8 .. 11.2 .. 1.7 .. –4.2 .. 43.8 .. 20.7 .. Mauritania .. .. .. .. .. .. .. .. .. .. .. .. .. .. Mexico 13.9 12.8 1.1 2.5 11.7 13.7 5.0 1.9 –3.0 –0.2 30.2 25.9 42.0 50.1 Moldova .. .. .. .. .. .. .. .. .. .. .. .. .. .. Mongolia .. 17.0 .. 4.8 .. 16.0 .. 3.7 .. –6.0 .. 24.1 .. .. Morocco 20.4 23.8 2.9 4.7 22.8 26.1 10.3 7.2 –9.7 –4.4 46.6 48.5 27.0 26.9 Mozambique .. .. .. .. .. .. .. .. .. .. .. .. .. .. Myanmar 9.6 4.0 6.4 2.9 12.1 4.7 3.8 5.4 1.2 –3.2 .. .. 26.5 18.9 Namibia .. .. .. .. .. .. .. .. .. .. .. .. .. .. Nepal 6.6 8.9 1.3 1.8 .. .. .. .. –3.0 –4.1 .. .. 15.6 25.9 Netherlands 44.1 42.7 5.3 3.0 48.2 46.0 4.6 1.7 –4.6 –1.7 15.3 15.4 62.9 63.9 New Zealand 30.7 31.2 3.5 2.7 35.9 31.4 2.4 0.9 –6.7 4.0 27.1 52.7 57.0 76.5 Nicaragua 20.3 23.9 2.4 1.5 24.8 22.3 5.7 10.9 –6.8 –0.6 59.6 28.8 33.2 .. Niger 12.3 .. 2.2 .. 9.5 .. 9.1 .. –4.8 .. 29.1 .. 24.8 .. Nigeria .. .. .. .. .. .. .. .. .. .. .. .. .. .. Norway 33.7 32.5 3.5 9.2 32.5 35.1 2.0 1.7 –1.7 5.1 17.9 20.3 36.8 50.2 Pakistan 13.3 12.9 2.9 3.1 14.5 19.9 3.1 2.8 –5.7 –7.9 36.6 50.0 .. .. Panama 18.6 15.9 6.8 10.1 25.0 23.9 5.5 3.5 –5.2 –0.7 49.7 49.2 39.6 64.0 Papua New Guinea 20.6 .. 2.4 .. 29.2 .. 5.2 .. –1.9 .. 56.4 .. 27.2 .. Paraguay 9.8 .. 0.9 .. 7.5 .. 2.4 .. 0.3 .. 57.2 .. 33.6 .. Peru 15.8 14.0 1.3 1.7 15.0 13.1 4.4 2.4 –2.4 0.3 44.7 38.0 .. .. Philippines 12.5 17.0 1.5 2.0 9.9 16.3 3.4 2.2 –1.4 0.1 52.2 51.1 20.8 26.5 Poland .. 35.2 .. 3.4 .. 39.3 .. 1.9 .. –1.4 .. 25.3 .. 71.4 Portugal 24.1 31.1 1.9 3.1 28.7 36.2 4.4 5.3 –8.4 –2.3 32.0 40.8 46.0 .. Romania 10.1 24.4 35.2 2.1 29.8 29.1 15.0 2.9 0.5 –3.9 11.3 30.1 18.8 49.0 Russian Federation .. 17.9 .. 1.1 .. .. .. .. .. –4.5 .. .. .. 31.1 Rwanda 11.0 .. 1.8 .. 9.3 .. 5.0 .. –1.7 .. 56.8 .. .. .. Saudi Arabia .. .. .. .. .. .. .. .. .. .. .. .. .. .. Senegal 21.0 .. 1.5 .. 22.5 .. 1.9 .. 0.9 .. 71.6 .. 36.8 .. Sierra Leone 13.6 10.2 1.5 0.3 19.6 13.4 5.0 4.3 –11.8 –6.0 .. 39.0 .. .. Singapore 17.5 15.9 7.8 8.3 15.6 11.6 4.5 5.0 2.1 11.6 47.6 36.7 24.1 23.2 Slovak Republic .. .. .. .. .. .. .. .. .. .. .. .. .. .. Slovenia .. .. .. .. .. .. .. .. .. .. .. .. .. .. South Africa 20.5 27.5 3.0 2.0 19.1 32.5 3.0 1.3 –2.3 –3.8 39.9 29.2 .. .. Spain 22.1 28.3 1.9 2.0 23.6 34.9 2.9 1.9 –4.2 –6.0 37.6 16.4 64.8 49.2 Sri Lanka 19.1 16.2 1.1 2.3 24.7 20.7 16.6 5.0 –18.3 –4.5 30.4 39.5 23.6 33.6 Sweden 30.1 36.9 4.9 5.1 37.5 43.2 1.8 1.1 –8.1 –1.3 15.8 14.0 58.2 58.1 Switzerland 17.2 21.1 1.4 1.6 17.9 25.3 1.3 1.0 –0.2 –1.2 27.1 26.5 63.6 70.6 Syrian Arab Republic 10.5 16.5 16.3 6.7 30.3 14.3 17.9 9.4 –9.7 –0.2 .. .. 17.6 18.2 Tajikistan .. .. .. .. .. .. .. .. .. .. .. .. .. .. Tanzania .. .. .. .. .. .. .. .. .. .. 51.8 .. 21.9 .. Thailand 13.2 16.1 1.2 1.9 14.4 11.0 4.4 7.7 –4.9 –0.9 53.3 49.8 28.0 39.1 Togo 27.0 .. 4.3 .. 23.7 .. 8.9 .. –2.0 .. 51.9 .. 39.9 .. Tunisia 23.9 24.8 6.9 4.8 22.1 25.9 9.4 6.7 –2.8 –3.1 38.3 37.9 34.2 46.6 Turkey 14.3 15.2 3.7 3.1 15.5 24.7 5.9 2.2 –3.1 –8.4 46.6 32.7 23.8 19.0 Turkmenistan .. .. .. .. .. .. .. .. .. .. .. .. .. .. Uganda 3.1 .. 0.1 .. 5.4 .. 0.8 .. –3.1 .. .. .. 23.5 .. Ukraine .. .. .. .. .. .. .. .. .. .. .. .. .. .. United Kingdom 30.6 33.4 4.6 2.8 36.4 39.6 1.8 2.1 –4.6 –5.3 30.2 29.6 43.7 51.7 United States 18.5 19.8 1.7 1.5 20.7 21.0 1.3 0.7 –2.8 –0.3 28.3 22.2 48.8 53.5 Uruguay 21.0 27.9 1.2 2.3 20.1 30.0 1.7 1.7 0.0 –1.3 46.7 28.8 61.1 74.6 Uzbekistan .. .. .. .. .. .. .. .. .. .. .. .. .. .. Venezuela 18.9 17.5 3.4 6.4 14.9 17.4 4.0 3.4 0.0 2.2 41.9 22.8 .. .. Vietnam .. .. .. .. .. .. .. .. .. .. .. .. .. .. Yemen, Rep. .. 13.3 .. 24.5 .. 33.6 .. 5.6 .. –2.6 .. 39.0 .. 19.4 Zambia 23.1 17.1 1.8 1.5 33.0 14.3 4.0 7.1 –18.5 0.7 45.8 39.0 17.4 29.8 Zimbabwe 15.4 .. 3.9 .. 26.5 .. 1.4 .. –8.8 .. 55.3 .. 28.5 .. a. Includes grants. b. Total expenditure includes lending minus repayments. c. Refers to education, health, social security, welfare, housing, and community amenities. d. Data prior to 1992 include Eritrea. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 15. Balance of payments, current account, and international reserves Millions of dollars Goods and services Net current Current account Gross international Exports Imports Net income transfers balance reserves Economy 1980 1997 1980 1997 1980 1997 1980 1997 1980 1997 1980 1998 Albania 378 222 371 809 4 50 6 265 16 –272 .. 382 Algeria 14,128 14,779 12,311 8,568 –1,869 –2,523 301 .. 249 .. 7,064 8,452 Angola .. 5,223 .. 5,389 .. –826 .. 3,841 .. 3,266 .. 206 Argentina 9,897 29,382 13,182 34,968 –1,512 –4,190 23 347 –4,774 –9,429 9,297 24,856 Armenia .. 330 .. 952 .. 102 .. 217 .. –303 .. 328 Australia 25,755 83,703 27,089 81,891 –2,688 –14,132 –425 –270 –4,447 –12,591 6,366 16,144 Austria 26,650 88,266 29,921 91,446 –528 –122 –66 –1,695 –3,865 –4,996 17,725 25,208 Azerbaijan .. 1,150 .. 2,101 .. –33 .. 45 .. –939 .. 447 Bangladesh 885 5,096 2,545 7,677 14 –91 802 1,770 –844 –902 331 1,936 Belarus .. 8,306 .. 9,103 .. –79 .. 78 .. –798 .. 339 Belgiuma 70,498 185,415 74,259 173,865 61 6,287 –1,231 –3,898 –4,931 13,939 27,974 21,013 Benin 226 524 421 673 8 –38 151 .. –36 .. 15 261 Bolivia 1,030 1,362 833 2,049 –263 –266 60 248 –6 –705 553 1,130 Botswana 645 3,030 818 2,365 –33 –145 55 201 –151 721 344 6,025 Brazil 21,869 60,256 27,826 79,817 –7,018 –16,091 144 1,812 –12,831 –33,840 6,875 43,902 Bulgaria 9,302 6,277 7,994 5,730 –412 –357 58 237 954 427 .. 3,127 Burkina Faso 210 298 577 654 –3 –33 322 .. –49 .. 75 373 Burundi .. 96 .. 139 .. –12 .. 60 .. 4 105 70 Cambodia .. 896 .. 1,252 .. –43 .. 188 .. –210 .. 324 Cameroon 1,880 2,443 1,829 2,041 –628 –609 83 87 –495 –121 206 1 Canada 74,977 247,438 70,259 236,225 –10,764 –20,913 –42 439 –6,088 –9,261 15,462 24,023 Central African Republic 201 171 327 241 3 –17 81 .. –43 .. 62 146 Chad 71 271 79 563 –4 –2 24 .. 12 .. 12 120 Chile 5,968 20,608 7,052 22,218 –1,000 –2,975 113 528 –1,971 –4,057 4,128 16,014 China 23,637 207,251 18,900 166,754 451 –15,923 486 5,144 5,674 29,718 10,091 152,843 Hong Kong, China 25,585 228,877 27,017 231,485 .. .. .. .. –1,432 –2,608 . . 89,620 Colombia 5,328 15,861 5,454 18,784 –245 –3,371 165 612 –206 –5,682 6,474 8,397 Congo, Dem. Rep. .. .. .. .. .. .. .. .. .. .. 380 83 Congo, Rep. 1,021 1,800 1,025 1,368 –162 –664 –1 –20 –167 –252 93 1 Costa Rica 1,195 4,478 1,661 4,666 –212 –202 15 136 –664 –254 197 1,064 Côte d’Ivoire 3,577 4,927 4,145 3,693 –553 –849 –706 –350 –1,826 35 46 855 Croatia .. 8,199 .. 11,402 .. –83 .. 852 . . –2,434 .. 2,816 Czech Republic .. 29,868 .. 32,713 .. –791 .. 365 . . –3,271 . . 12,625 Denmark 21,989 63,680 21,727 57,971 –1,977 –3,635 –161 –1,190 –1,875 883 4,347 15,881 Dominican Republic 1,271 7,060 1,919 7,780 –277 –795 205 1,352 –720 –163 279 507 Ecuador 2,887 6,000 2,946 5,787 –613 –1,347 30 391 –642 –743 1,257 1,739 Egypt, Arab Rep. 6,246 16,171 9,157 18,296 –318 884 2,791 4,146 –438 2,905 2,480 18,824 El Salvador 1,214 2,706 1,170 3,885 –62 –87 52 1,363 34 96 382 1,748 Eritrea .. 201 .. 583 .. –3 .. 364 .. –21 .. .. Estonia .. 3,609 .. 4,142 .. –146 .. 117 .. –562 .. 813 Ethiopiab 569 1,017 782 1,683 7 –43 80 259 –126 –450 262 520 Finland 16,802 48,228 17,307 37,976 –783 –2,736 –114 –852 –1,403 6,664 2,451 10,271 France 153,197 365,342 155,915 319,781 2,680 2,693 –4,170 –8,780 –4,208 39,474 75,592 73,773 Georgia .. 622 .. 1,192 .. 35 .. 196 .. –339 .. 192 Germany c 224,224 590,984 225,599 558,835 914 –2,436 –12,858 –32,487 –13,319 –2,774 104,702 108,265 Ghana 1,210 1,655 1,178 2,640 –83 –131 81 576 30 –541 330 457 Greece 8,122 14,863 11,145 25,601 –273 –1,632 1,087 7,510 –2,209 –4,860 3,607 18,501 Guatemala 1,731 3,187 1,960 4,193 –44 –224 110 607 –163 –624 753 1,397 Guinea .. 741 .. 834 .. –114 .. 116 .. –91 .. 122 Haiti 306 218 481 810 –14 –14 89 463 –101 –138 27 83 Honduras 942 2,191 1,128 2,511 –152 –212 22 260 –317 –272 159 824 Hungary 9,671 24,514 9,152 25,067 –1,113 –1,426 63 997 –531 –982 .. 9,348 India 11,265 44,102 17,378 59,236 356 –2,507 2,860 11,830 –2,897 –5,811 12,010 30,647 Indonesia 23,797 63,238 21,540 62,830 –3,073 –6,332 250 1,034 –566 –4,890 6,803 23,606 Iran, Islamic Rep. 13,069 23,251 16,111 18,072 606 –410 –2 463 –2,438 5,232 12,783 .. Ireland 9,610 61,447 12,044 51,711 –902 –9,708 1,204 1,956 –2,132 1,984 3,071 9,527 Israel 8,668 30,320 11,511 38,810 –757 –2,791 2,729 6,266 –871 –5,014 4,055 22,674 Italy 97,298 310,550 110,265 261,884 1,278 –11,202 1,101 –4,040 –10,587 33,424 62,428 53,880 Jamaica 1,363 3,192 1,408 4,005 –212 –193 121 624 –136 –382 105 682 Japan 146,980 478,542 156,970 431,094 770 55,739 –1,530 –8,834 –10,750 94,354 38,919 222,443 Jordan 1,181 3,572 2,417 5,186 36 –209 1,481 1,852 281 29 1,745 1,988 Kazakhstan .. 7,611 .. 8,279 .. –315 .. 75 .. –909 .. 1,965 Kenya 2,007 2,994 2,846 3,771 –194 –232 157 632 –876 –377 539 783 Korea, Rep. 19,815 164,920 25,152 171,300 –512 –2,455 536 667 –5,312 –8,167 3,101 52,100 Kuwait 21,857 16,041 9,823 12,876 4,847 6,277 –1,580 –1,507 15,302 7,935 5,425 4,678 Kyrgyz Republic .. 676 .. 817 .. –65 .. 68 .. –139 .. 188 Lao PDR .. 417 .. 715 .. –19 .. 91 .. –225 .. 117 Latvia .. 2,871 .. 3,348 .. 55 .. 77 .. –345 .. 800 Lebanon .. 1,557 .. 8,053 .. 380 .. 2,635 . . –3,481 7,025 9,210 Lesotho 90 267 475 1,080 266 318 175 .. 56 .. 50 575 Lithuania .. 5,224 .. 6,237 .. –198 .. 230 .. –981 .. 1,463 Macedonia, FYR .. 1,330 .. 1,862 .. –34 .. 290 .. –275 .. 335 Madagascar 516 755 1,075 1,032 –44 –109 47 210 –556 –153 9 171 Malawi 313 672 487 1,269 –149 –96 63 .. –260 .. 76 273 Malaysia 14,098 92,897 13,526 91,521 –836 –5,074 –2 –1,094 –266 –4,792 5,755 26,236 * Taiwan, China 21,495 139,396 22,361 132,739 48 2,391 –95 –1,327 –913 7,721 4,055 94,246 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      ECONOMY Millions of dollars Goods and services Net current Current account Gross international Exports Imports Net income transfers balance reserves Economy 1980 1997 1980 1997 1980 1997 1980 1997 1980 1997 1980 1998 Mali 262 642 519 896 –17 –51 150 126 –124 –178 26 403 Mauritania 253 407 449 414 –27 –46 90 76 –133 22 146 206 Mexico 22,622 121,831 27,601 122,424 –6,277 –12,108 834 5,247 –10,422 –7,454 4,175 31,863 Moldova .. 1,024 .. 1,431 .. 63 .. 76 .. –267 .. 144 Mongolia 475 624 1,272 588 –11 –5 0 77 –808 39 .. 103 Morocco 3,233 9,510 5,207 10,627 –562 –1,175 1,130 2,205 –1,407 –87 814 4,638 Mozambique 399 500 844 1,005 22 –113 56 283 –367 –359 .. 608 Myanmar 539 1,439 806 2,415 –48 –64 7 430 –307 –610 409 382 Namibia .. 1,726 .. 1,908 .. 54 .. 322 .. 193 .. 260 Nepal 224 1,295 365 1,855 13 5 36 95 –93 –460 272 800 Netherlands 90,380 216,530 91,622 193,107 1,535 4,686 –1,148 –6,123 –855 21,985 37,549 31,155 New Zealand 6,403 18,224 6,934 18,269 –538 –5,444 96 336 –973 –5,153 365 4,204 Nicaragua 495 863 907 1,609 –124 –222 124 367 –411 –601 75 355 Niger 617 300 956 441 –33 –21 97 31 –276 –152 132 53 Nigeria 27,071 15,994 20,014 14,213 –1,304 –3,145 –576 1,916 5,178 552 10,640 4,329 Norway 27,264 63,213 23,749 52,286 –1,922 –1,391 –515 –1,424 1,079 8,112 6,746 18,947 Pakistan 2,958 9,956 5,709 14,677 –281 –2,167 2,163 3,213 –868 –3,675 1,568 1,626 Panama 3,422 8,316 3,394 8,649 –397 –419 40 160 –329 –592 117 954 Papua New Guinea 1,029 2,557 1,322 2,407 –179 –310 184 61 –289 –99 458 211 Paraguay 701 4,343 1,314 4,960 –4 87 0 47 –618 –483 783 784 Peru 4,631 8,356 3,970 10,842 –909 –1,602 147 681 –101 –3,407 2,804 9,882 Philippines 7,235 40,365 9,166 50,477 –420 4,681 447 1,080 –1,904 –4,351 3,978 10,789 Poland 16,061 39,717 17,842 46,367 –2,357 –1,129 721 2,035 –3,417 –5,744 574 27,383 Portugal 6,674 32,339 10,136 40,684 –608 –245 3,006 6,713 –1,064 –1,877 13,863 21,606 Romania 12,087 9,853 13,730 12,448 –777 –322 .. 579 –2,420 –2,338 2,511 3,793 Russian Federation .. 102,196 .. 90,065 . . –9,200 .. –362 .. 2,569 . . 12,043 Rwanda 165 152 319 488 2 –16 104 260 –48 –93 187 169 Saudi Arabia 106,765 64,939 55,793 52,399 526 3,156 –9,995 –15,439 41,503 257 26,129 8,843 Senegal 807 1,281 1,215 1,557 –98 –62 120 166 –386 –200 25 431 Sierra Leone 275 91 471 160 –22 11 53 26 –165 –127 31 44 Singapore 24,285 156,252 25,312 144,168 –429 3,906 –106 –1,187 –1,563 14,803 6,567 74,928 Slovak Republic .. 10,959 .. 12,367 .. –124 .. 173 . . –1,359 .. 3,240 Slovenia .. 10,450 .. 10,631 .. 131 .. 88 .. 37 .. 3,639 South Africa 28,627 35,440 22,073 34,626 –3,285 –2,602 239 –143 3,508 –1,931 7,888 5,508 Spain 32,140 148,357 38,004 142,478 –1,362 –6,396 1,646 3,003 –5,580 2,486 20,473 60,881 Sri Lanka 1,293 5,514 2,197 6,569 –26 –165 274 832 –655 –388 283 1,998 Sweden 38,151 100,989 39,878 84,779 –1,380 –6,174 –1,224 –2,736 –4,331 7,301 6,996 15,457 Switzerland 48,595 120,696 51,843 107,187 4,186 13,566 –1,140 –3,360 –201 23,714 64,748 65,158 Syrian Arab Republic 2,477 5,661 4,531 5,092 785 –504 1,520 499 251 564 828 .. Tajikistan .. 772 .. 808 .. –68 .. 20 .. –84 .. .. Tanzania 748 1,200 1,384 1,961 –14 –124 129 341 –521 –544 20 599 Thailand 7,939 72,415 9,996 72,437 –229 –3,480 210 479 –2,076 –3,024 3,026 29,537 Togo 550 709 691 836 –40 7 86 .. –95 .. 85 118 Tunisia 3,262 8,081 3,766 8,644 –259 –863 410 785 –353 –640 700 1,856 Turkey 3,621 52,004 8,082 56,536 –1,118 –3,013 2,171 4,866 –3,408 –2,679 3,298 20,568 Turkmenistan .. 1,691 .. 1,532 .. .. .. .. .. 43 .. .. Uganda 329 825 441 1,651 –7 –17 –2 322 –121 –521 3 725 Ukraine .. 20,355 .. 21,891 .. –644 .. 845 . . –1,335 .. 793 United Kingdom 146,072 375,033 134,200 375,128 –418 18,171 –4,592 –7,773 6,862 10,304 31,755 38,830 United States 271,800 937,434 290,730 1,043,473 29,580 –9,487 –8,500 –39,849 2,150 –155,375 171,413 146,006 Uruguay 1,526 4,256 2,144 4,450 –100 –208 9 81 –709 –321 2,401 2,587 Uzbekistan .. 3,980 .. 4,417 .. –175 .. 29 .. –583 .. .. Venezuela 19,968 25,120 15,130 18,282 329 –2,031 –439 –123 4,728 4,684 13,360 14,729 Vietnam .. 11,485 .. 13,465 –72 –602 17 713 –775 –1,870 .. 1,986 Yemen, Rep. .. 2,522 .. 3,005 .. –636 .. 1,254 .. 135 .. 1,010 Zambia 1,609 1,321 1,765 1,270 –205 –543 –155 .. –516 .. 206 69 Zimbabwe 1,610 3,059 1,730 3,692 –61 –405 31 .. –149 .. 419 310 World 2,291,841 t 6,886,726 t 2,323,396 t 6,763,911 t Low income 100,391 410,532 125,802 407,224 Excl. China & India 79,559 159,077 101,300 181,262 Middle income 509,704 1,282,683 470,588 1,335,448 Lower middle income 197,222 476,598 208,570 496,002 Upper middle income 312,785 805,648 267,306 838,247 Low and middle income 632,929 1,693,448 596,880 1,742,630 East Asia & Pacific 105,229 661,970 110,191 640,933 Europe & Central Asia .. 347,889 .. 371,154 Latin America & Carib. 114,161 337,037 129,051 377,410 Middle East & N. Africa 180,284 177,797 130,208 155,923 South Asia 17,314 66,540 28,820 90,646 Sub-Saharan Africa 87,905 100,807 81,894 106,398 High income 1,680,398 5,195,331 1,732,925 5,022,907 a. Includes Luxembourg. b. Data prior to 1992 include Eritrea. c. Data prior to 1990 refer to the Federal Republic of Germany before unification. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 16. Private sector finance Interest rate spread Domestic credit Private investment Stock market (lending minus provided by % of gross domestic capitalization No. of listed deposit rate) the banking sector fixed investment Millions of dollars domestic companies Percentage points % of GDP Economy 1980 1997 1990 1998 1990 1997 1990 1998 1990 1998 Albania .. .. .. .. .. .. 2.1 7.2 .. 53.2 Algeria 67.4 72.5 .. .. .. .. .. .. 74.7 42.9 Angola .. 88.0 .. .. .. .. .. 8.1 .. 14.3 Argentina .. 94.2 3,268 45,332 179 136 .. 3.1 32.4 28.3 Armenia .. 53.7 .. 16 .. 59 .. 23.5 62.3 8.2 Australia 73.5 81.8 107,611 696,656 1,089 1,219 4.5 3.4 103.5 89.4 Austria .. .. 11,476 35,724 97 101 .. 3.8 123.0 131.9 Azerbaijan .. 96.5 .. .. .. .. .. .. 57.2 13.5 Bangladesh 57.7 67.8 321 1,034 134 202 4.0 5.6 24.1 32.8 Belarus .. .. .. .. .. .. .. 12.7 .. 17.7 Belgium .. .. 65,449 136,965 182 138 6.9 4.2 70.9 147.9 Benin .. 59.5 .. .. .. .. 9.0 .. 22.4 7.0 Bolivia .. 58.1 .. 344 .. 11 18.0 26.6 30.7 67.1 Botswana 60.4 44.6 261 724 9 12 1.8 4.8 –46.4 –74.5 Brazil 89.8 88.7 16,354 160,887 581 536 .. .. 89.8 53.3 Bulgaria 85.9 .. .. 992 .. 15 8.9 10.3 118.5 30.0 Burkina Faso .. 52.4 .. .. .. .. 9.0 .. 13.7 13.4 Burundi 8.1 30.9 .. .. .. .. .. .. 24.5 25.4 Cambodia .. 68.9 .. .. .. .. .. 10.5 .. 7.7 Cameroon 77.8 93.7 .. .. .. .. 11.0 17.0 31.2 16.8 Canada 87.4 86.3 241,920 567,635 1,144 1,362 1.3 1.6 85.8 99.0 Central African Republic 46.5 42.2 .. .. .. .. 11.0 17.0 12.9 10.6 Chad .. .. .. .. .. .. 11.0 17.0 10.9 9.8 Chile .. 80.9 13,645 51,866 215 295 8.6 5.3 73.0 65.6 China 43.4 49.1 2,028 231,322 14 764 0.7 2.6 90.0 120.0 Hong Kong, China 85.1 .. 83,397 413,323 284 658 3.3 2.4 156.3 146.8 Colombia 58.2 59.1 1,416 13,357 80 189 8.8 9.7 35.9 45.7 Congo, Dem. Rep. 42.4 64.4 .. .. .. .. .. .. 25.3 .. Congo, Rep. .. 66.5 .. .. .. .. 11.0 17.0 29.1 21.8 Costa Rica 61.3 80.0 475 820 82 114 11.4 9.7 29.9 46.1 Côte d’Ivoire 53.2 70.2 549 1,818 23 35 9.0 .. 44.5 28.1 Croatia .. 59.6 .. 3,190 1 77 499.3 11.1 .. 46.4 Czech Republic .. .. .. 12,045 .. 276 .. 4.7 .. 74.2 Denmark .. .. 39,063 93,766 258 237 6.2 4.8 63.0 61.2 Dominican Republic 68.4 83.0 .. 140 .. 6 15.2 8.0 31.5 33.1 Ecuador 59.7 82.9 69 1,527 65 41 –6.0 10.4 17.2 45.9 Egypt, Arab Rep. 30.1 68.4 1,765 24,381 573 650 7.0 3.7 106.8 95.5 El Salvador 44.8 77.0 .. 499 .. 59 3.2 4.7 32.0 40.8 Eritrea .. 53.8 .. .. .. .. .. .. .. .. Estonia .. 74.4 .. 519 .. 22 .. 8.6 65.0 31.6 Ethiopia .. 56.6 .. .. .. .. 3.6 4.5 50.4 44.1 Finland .. .. 22,721 73,322 73 124 4.1 3.3 84.3 57.4 France .. .. 314,384 674,368 578 683 6.0 3.3 106.1 103.3 Georgia .. 84.0 .. .. .. .. .. .. .. .. Germany .. .. 355,073 825,233 413 700 4.5 6.1 108.5 145.8 Ghana .. 46.4 76 1,384 13 21 .. .. 13.2 27.7 Greece 51.5 .. 15,228 79,992 145 230 8.1 7.9 73.3 56.3 Guatemala 63.8 80.4 .. 139 .. 7 5.1 11.1 17.4 16.1 Guinea .. 68.5 .. .. .. .. 0.2 .. 5.4 6.8 Haiti .. 51.0 .. .. .. .. .. 10.6 32.9 25.8 Honduras 62.1 72.2 40 .. 26 119 8.3 12.1 40.9 28.5 Hungary .. .. 505 14,028 21 49 4.1 3.2 82.6 .. India 55.1 68.7 38,567 105,188 6,200 5,843 .. .. 50.6 48.2 Indonesia .. 60.5 8,081 21,224 125 282 3.3 –6.9 45.5 57.9 Iran, Islamic Rep. .. .. 34,282 15,123 97 263 .. .. 62.1 .. Ireland .. .. .. 24,135 .. 83 5.0 5.8 57.3 100.2 Israel .. .. 3,324 39,628 216 640 12.0 5.2 106.2 82.3 Italy .. .. 148,766 344,665 220 235 7.3 4.7 90.1 93.6 Jamaica .. .. 911 2,139 44 49 6.6 19.1 34.8 42.7 Japan .. .. 2,917,679 2,216,699 2,071 2,387 3.4 2.1 266.8 137.4 Jordan 51.3 84.0 2,001 5,838 105 139 2.2 3.2 110.0 93.2 Kazakhstan .. .. .. .. .. .. .. .. .. 9.1 Kenya 54.7 61.8 453 2,024 54 58 5.1 11.1 52.9 51.7 Korea, Rep. 86.0 .. 110,594 114,593 669 776 0.0 2.0 56.9 84.1 Kuwait .. .. .. 25,880 .. 74 0.4 2.6 217.6 92.3 Kyrgyz Republic .. 94.9 .. 5 .. 27 .. 37.7 .. 19.1 Lao PDR .. .. .. .. .. .. 2.5 11.5 5.1 16.4 Latvia .. 89.2 .. 382 .. 50 .. 9.0 .. 15.2 Lebanon .. 79.3 .. 2,904 .. 9 23.1 6.9 132.6 134.9 Lesotho .. 81.8 .. .. .. .. 7.4 9.3 27.4 –27.2 Lithuania .. 88.2 .. 1,074 .. 607 .. 6.2 .. 11.7 Macedonia, FYR .. 91.2 .. .. .. .. .. 9.4 .. 20.7 Madagascar .. 46.9 .. .. .. .. 5.3 15.6 26.2 13.9 Malawi 21.4 27.7 .. .. .. .. 8.9 18.6 17.8 6.5 Malaysia 62.6 73.0 48,611 107,104 282 708 1.3 2.1 77.9 162.4 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      STATES AND MARKETS Interest rate spread Domestic credit Private investment Stock market (lending minus provided by % of gross domestic capitalization No. of listed deposit rate) the banking sector fixed investment Millions of dollars domestic companies Percentage points % of GDP Economy 1980 1997 1990 1998 1990 1997 1990 1998 1990 1998 Mali .. 60.8 .. .. .. .. 9.0 .. 13.7 14.4 Mauritania .. 78.3 .. .. .. .. 5.0 .. 54.7 2.4 Mexico 57.0 81.5 32,725 91,746 199 198 .. 14.9 36.6 34.8 Moldova .. 86.2 .. .. .. .. .. 9.1 62.8 26.7 Mongolia .. .. .. 54 .. 434 .. 15.7 68.5 13.3 Morocco .. 70.4 966 15,676 71 49 0.5 .. 42.9 81.7 Mozambique .. 43.7 .. .. .. .. .. .. 15.6 2.4 Myanmar 20.6 55.0 .. .. .. .. 2.1 4.0 44.7 34.4 Namibia 42.0 62.2 21 689 3 13 10.6 7.8 19.5 53.9 Nepal 60.2 65.8 .. 200 .. 98 2.5 5.1 33.4 35.9 Netherlands 85.1 86.3 119,825 468,736 260 201 8.4 3.4 107.4 131.5 New Zealand 69.2 87.2 8,835 90,483 171 190 4.4 4.4 74.3 104.5 Nicaragua .. 38.6 .. .. .. .. 12.5 10.9 206.6 141.0 Niger 20.1 45.3 .. .. .. .. 9.0 .. 16.2 9.3 Nigeria .. 44.0 1,372 2,887 131 182 5.5 13.1 23.7 14.2 Norway 70.3 .. 26,130 66,503 112 196 4.6 0.7 67.4 62.4 Pakistan 36.1 65.4 2,850 5,418 487 781 .. .. 50.9 50.9 Panama .. 83.3 226 2,175 13 21 3.6 4.1 52.7 92.9 Papua New Guinea 58.6 84.9 .. .. .. .. 6.9 4.0 35.8 35.7 Paraguay 85.1 67.5 .. 389 .. 60 8.1 14.0 14.9 33.4 Peru 75.6 84.7 812 11,645 294 248 2,335.0 15.7 16.2 22.0 Philippines 69.0 .. 5,927 35,314 153 221 4.6 4.7 23.2 69.8 Poland .. 86.6 144 20,461 9 143 462.5 6.3 19.5 38.6 Portugal .. .. 9,201 62,954 181 148 7.8 3.9 71.8 108.0 Romania .. .. .. 1,016 .. 76 .. .. 79.7 24.2 Russian Federation .. 76.6 244 20,598 13 208 .. 24.7 .. 35.6 Rwanda .. 18.0 .. .. .. .. 6.3 .. 17.1 12.1 Saudi Arabia .. .. 48,213 42,563 59 70 .. .. 14.4 .. Senegal 58.1 70.1 .. .. .. .. 9.0 .. 33.8 21.8 Sierra Leone .. .. .. .. .. .. 12.0 16.7 26.3 52.1 Singapore 75.6 .. 34,308 106,317 150 303 2.7 2.8 60.9 85.4 Slovak Republic .. .. .. 965 .. 872 .. 4.9 .. 71.8 Slovenia .. 90.4 .. 2,450 24 26 142.0 5.5 36.8 35.8 South Africa 50.8 72.9 137,540 170,252 732 642 2.1 5.3 102.7 83.4 Spain .. .. 111,404 290,383 427 384 5.4 2.1 110.9 114.9 Sri Lanka 77.4 77.6 917 1,705 175 239 –6.4 –7.0 43.1 32.2 Sweden .. 79.7 97,929 272,730 258 245 6.8 4.0 145.5 80.9 Switzerland .. .. 160,044 575,338 182 216 –0.9 3.4 179.0 177.2 Syrian Arab Republic 36.1 .. .. .. .. .. .. .. 56.6 38.5 Tajikistan .. .. .. .. .. .. .. .. .. .. Tanzania .. 83.8 .. .. .. .. .. 18.9 39.2 13.5 Thailand 68.1 67.7 23,896 34,903 214 431 2.2 3.8 91.1 159.5 Togo 28.3 85.0 .. .. .. .. 9.0 .. 21.3 24.9 Tunisia 46.9 49.3 533 2,268 13 34 .. .. 62.5 53.3 Turkey .. 78.5 19,065 33,646 110 257 .. .. 25.9 34.1 Turkmenistan .. .. .. .. .. .. .. .. .. .. Uganda .. 63.6 .. .. .. .. 7.4 9.5 17.7 7.0 Ukraine .. 0.0 .. 570 .. .. .. 32.2 83.2 24.7 United Kingdom 70.0 87.0 848,866 1,996,225 1,701 2,046 2.2 2.7 123.0 129.3 United States 86.5 85.9 3,059,434 11,308,779 6,599 8,851 .. .. 114.6 162.8 Uruguay .. 72.1 .. 212 36 16 76.6 42.8 60.1 41.3 Uzbekistan .. .. .. 465 .. 4 .. .. .. .. Venezuela 51.4 43.6 8,361 7,587 76 91 0.5 11.3 37.4 17.5 Vietnam .. 79.7 .. .. .. .. .. 5.3 15.9 22.6 Yemen, Rep. .. 63.2 .. .. .. .. .. .. 62.0 35.7 Zambia .. 60.1 .. 705 .. 6 9.4 18.7 67.8 63.5 Zimbabwe 87.3 88.7 2,395 1,310 57 64 2.9 13.0 41.7 62.7 World ..w ..w 9,398,391 s 23,540,720 s 29,189 s 40,394 s 125.2 w 126.2 w Low income 47.7 55.2 54,588 387,184 7,211 8,948 60.0 86.0 Excl. China & India .. 63.0 16,021 52,352 1,011 2,341 38.1 37.4 Middle income .. 82.4 430,570 1,404,501 4,914 9,193 57.9 52.9 Lower middle income .. 70.5 176,701 524,675 2,455 4,433 .. 57.5 Upper middle income 69.1 87.6 253,869 879,826 2,459 4,760 54.1 51.8 Low and middle income .. 73.3 485,158 1,791,685 12,125 18,141 58.5 65.3 East Asia & Pacific 57.3 66.9 197,109 426,006 1,443 3,624 70.9 108.6 Europe & Central Asia .. 75.3 19,065 243,096 110 2,711 .. 32.9 Latin America & Carib. 70.0 84.0 78,470 608,395 1,748 2,238 59.1 41.9 Middle East & N. Africa .. .. 5,265 125,286 817 1,328 54.3 .. South Asia 53.8 68.6 42,655 143,250 6,996 7,163 48.3 47.2 Sub-Saharan Africa 52.0 67.3 142,594 245,652 1,011 1,077 57.5 45.5 High income .. .. 8,913,233 21,749,035 17,064 22,253 140.0 140.4 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 17. Role of government in the economy Highest marginal tax ratea Subsidies and other Value added by state- Composite Institutional Individual current transfers owned enterprises Military expenditure ICRG Investor On income Corporate % of total expenditure % of GDP % of GNP risk ratinga credit ratinga % over (dollars) % Economy 1985 1997 1985–90 1990–96 1985 1995 February 1999 March 1999 1998 1998 1998 Albania .. 48 .. .. 5.3 1.1 60.5 10.7 .. .. .. Algeria .. .. .. .. 2.5 3.2 52.8 25.2 .. .. .. Angola .. .. .. .. 19.9 3.0 46.5 11.5 .. .. .. Argentina 59 58 2.7 1.3 3.8 1.7 76.3 42.7 33 120,000 33 Armenia .. .. .. .. .. 0.9 61.0 .. .. .. .. Australia 63 69 .. .. 2.7 2.5 80.0 74.3 47 32,404 36 Austria 58 59 .. .. 1.3 0.9 84.8 88.7 50 55,564 34 Azerbaijan .. .. .. .. .. 2.8 56.0 .. 40 1,850 32 Bangladesh .. .. 3.1 3.4 1.7 1.7 66.0 25.0 .. .. .. Belarus .. 54 .. .. .. 0.8 59.8 11.9 .. .. .. Belgium 56 59 2.8 .. 3.1 1.7 80.5 83.5 55 65,547 39 Benin .. .. .. .. 2.2 1.2 .. 16.3 .. .. .. Bolivia 27 40 13.9 13.8 3.3 2.3 67.5 28.0 .. .. 25 Botswana 29 31 5.6 5.6 2.5 5.3 82.0 53.5 30 21,008 15 Brazil 42 .. 7.6 8.0 0.8 1.7 61.5 37.4 25 19,459 15 Bulgaria .. 37 .. .. 14.1 2.8 75.5 28.6 40 7,232 30 Burkina Faso 9 .. .. .. 1.9 2.9 65.5 18.8 .. .. .. Burundi .. 11 7.3 .. 3.0 4.4 .. .. .. .. .. Cambodia .. .. .. .. .. 3.1 .. .. .. .. .. Cameroon 14 13 18.0 8.5 1.9 .. 63.5 18.1 60 12,345 39 Canada 60 .. .. .. 2.2 1.7 82.8 83.0 29 41,370 38 Central African Republic .. .. 4.1 .. 1.8 .. .. .. .. .. .. Chad 2 .. .. .. 2.0 3.1 .. .. .. .. .. Chile 51 52 14.4 8.1 4.0 3.8 74.0 61.8 45 6,748 15 China .. .. .. .. 4.9 2.3 75.5 57.2 45 12,077 30 Hong Kong, China .. .. .. .. .. .. 76.3 61.8 20 11,688 17 Colombia 48 .. 7.0 .. 1.6 2.6 57.3 44.5 35 38,764 35 Congo, Dem. Rep. 7 2 .. .. 1.2 0.3 39.5 11.1 50 13,167 .. Congo, Rep. .. .. 15.1 .. 4.0 2.9 50.0 9.7 .. .. .. Costa Rica 33 23 8.1 .. 0.7 0.6 76.3 38.4 25 15,746 30 Côte d’Ivoire .. .. .. .. .. .. 67.3 24.3 10 3,950 35 Croatia .. 38 .. .. .. 10.5 70.8 39.0 .. .. .. Czech Republic .. 74 .. .. .. 2.3 76.5 .. 40 23,750 35 Denmark 57 64 .. .. 2.3 1.8 86.0 84.7 58 .. 34 Dominican Republic 17 17 .. .. 1.2 1.4 72.0 28.1 25 16,176 25 Ecuador .. .. 10.2 .. 2.8 3.7 61.5 25.5 25 66,226 25 Egypt, Arab Rep. 31 25 .. .. 12.8 5.7 69.0 44.4 32 13,749 40 El Salvador 11 20 1.8 .. 5.7 1.1 76.8 31.2 30 22,857 25 Eritrea .. .. .. .. .. .. .. .. .. .. .. Estonia .. 47 .. .. .. 1.1 73.0 42.8 26 .. 26 Ethiopia 7 13 .. .. 6.7 2.2 57.8 16.2 .. .. .. Finland 67 65 .. .. 1.7 2.0 86.5 82.2 38 56,450 28 France 64 65 11.2 .. 4.0 3.1 81.8 90.8 .. .. 33 Georgia .. .. .. .. .. 2.4 .. 10.9 .. .. .. Germany 55 58 .. .. .. .. 82.8 92.5 53 66,988 30 Ghana 10 .. 8.5 .. 1.0 1.4 62.8 29.5 35 7,269 35 Greece 35 22 11.5 .. 7.0 5.5 76.3 56.1 45 55,923 35 Guatemala 14 8 1.9 2.1 1.6 1.3 68.3 27.2 25 29,221 30 Guinea .. .. .. .. .. 1.5 60.5 15.4 .. .. .. Haiti 43 .. .. .. 1.5 2.9 55.0 11.2 .. .. .. Honduras .. .. 5.5 .. 3.5 1.4 58.8 19.8 30 75,758 15 Hungary 69 55 .. .. 7.2 1.5 77.8 55.9 42 5,394 18 India 44 38 13.4 13.4 3.5 2.4 63.3 44.5 40 5,059 40 Indonesia 24 21 14.5 .. 2.4 1.8 48.5 27.9 30 8,938 30 Iran, Islamic Rep. 13 15 .. .. 7.7 2.6 66.3 27.7 54 173,227 12 Ireland 57 60 .. .. 1.7 1.3 87.5 81.8 46 14,493 32 Israel 33 48 .. .. 20.3 9.6 64.8 54.3 50 57,387 36 Italy 57 57 .. .. 2.2 1.8 80.8 79.1 46 181,801 37 Jamaica 1 .. .. .. 0.9 0.8 71.3 28.0 25 2,215 33 Japan 52 .. .. .. 1.0 1.0 83.3 86.5 50 230,592 38 Jordan 14 11 .. .. 15.5 7.7 73.8 37.3 .. .. .. Kazakhstan .. .. .. .. .. 0.9 69.0 27.9 40 .. 30 Kenya 18 18 11.6 .. 2.3 2.3 63.8 24.1 33 384 35 Korea, Rep. 38 49 10.3 .. 5.0 3.4 74.5 52.7 40 56,529 28 Kuwait 26 20 .. .. 5.7 11.6 73.5 56.5 0 .. 6 Kyrgyz Republic .. .. .. .. .. .. .. .. 40 250 30 Lao PDR .. .. .. .. 7.4 4.2 .. .. .. .. .. Latvia .. 61 .. .. .. 0.9 71.0 38.0 25 .. 25 Lebanon .. 13 .. .. .. 3.7 55.3 31.9 .. .. .. Lesotho 5 9 .. .. 5.3 1.9 .. .. .. .. .. Lithuania .. 41 .. .. .. 0.5 73.5 .. 33 .. 29 Macedonia, FYR .. .. .. .. .. 3.3 .. .. .. .. .. Madagascar .. 8 .. .. 1.9 0.9 66.0 .. .. .. .. Malawi 7 .. 4.3 .. 2.0 1.6 61.8 20.4 38 1,969 38 Malaysia 13 24 .. .. 3.8 3.0 70.8 51.0 30 38,961 28 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      STATES AND MARKETS Highest marginal tax ratea Subsidies and other Value added by state- Composite Institutional Individual current transfers owned enterprises Military expenditure ICRG Investor On income Corporate % of total expenditure % of GDP % of GNP risk ratinga credit ratinga % over (dollars) % Economy 1985 1997 1985–90 1990–96 1985 1995 February 1999 March 1999 1998 1998 1998 Mali 8 .. .. .. 2.9 1.8 66.5 15.4 .. .. .. Mauritania .. .. .. .. 6.9 3.2 .. .. .. .. .. Mexico 21 43 6.7 4.9 0.7 1.0 66.3 46.0 35 25,492 34 Moldova .. .. .. .. .. 2.1 54.5 .. .. .. .. Mongolia .. 42 .. .. 8.3 2.4 66.3 .. .. .. .. Morocco 15 12 16.8 .. 6.0 4.3 72.3 43.2 44 6,203 35 Mozambique .. .. .. .. 9.9 5.4 58.5 17.9 .. .. .. Myanmar .. .. .. .. .. .. 55.0 18.7 .. .. .. Namibia 29 .. .. .. .. 2.1 77.8 .. 35 16,461 35 Nepal .. .. .. .. 1.1 0.9 .. 24.4 .. .. .. Netherlands 69 72 .. .. 3.0 2.1 87.8 91.7 60 51,373 35 New Zealand 51 38 .. .. 2.0 1.3 77.5 73.1 33 19,922 33 Nicaragua 11 25 .. .. 17.4 2.2 47.8 11.6 30 18,083 30 Niger .. .. 5.1 .. 0.8 1.2 54.8 .. .. .. .. Nigeria 9 .. .. .. 1.5 .. 56.3 16.8 25 1,600 28 Norway 68 69 .. .. 3.1 2.7 88.3 86.8 28 6,835 28 Pakistan 15 8 .. .. 6.2 6.1 53.5 20.4 .. .. .. Panama 17 25 7.6 7.6 2.0 1.4 72.3 39.9 30 200,000 15 Papua New Guinea 16 .. .. .. 1.5 1.4 67.0 30.4 47 57,803 15 Paraguay 23 .. 4.8 4.5 1.1 1.4 63.0 31.3 0 .. 30 Peru 11 36 6.4 5.7 6.7 1.7 66.3 35.0 30 50,036 30 Philippines 7 18 2.3 2.2 1.4 1.5 73.0 43.3 34 12,464 34 Poland 75 62 .. .. 10.2 2.3 80.5 56.7 40 14,372 36 Portugal 45 37 15.1 .. 2.9 2.6 82.0 76.1 40 34,186 37 Romania 27 50 .. .. 6.9 2.5 57.8 31.2 45 3,672 38 Russian Federation .. .. .. .. .. 11.4 49.8 20.0 35 8,587 35 Rwanda .. .. .. .. 1.7 5.2 .. .. .. .. .. Saudi Arabia .. .. .. .. 22.7 13.5 69.0 54.4 0 .. 45 Senegal .. .. 6.9 .. 2.8 1.6 63.0 21.7 50 20,821 35 Sierra Leone 5 24 .. .. 0.8 6.1 29.5 6.3 .. .. .. Singapore 10 8 .. .. 5.9 4.7 87.5 81.3 28 238,095 26 Slovak Republic .. .. .. .. .. 3.0 77.8 41.3 42 31,576 40 Slovenia .. .. .. .. .. 1.5 79.5 58.4 .. .. .. South Africa 31 48 14.9 .. 3.8 2.2 68.8 45.8 45 20,576 35 Spain 55 66 .. .. 2.4 1.6 79.5 80.3 48 69,216 35 Sri Lanka 16 20 .. .. 2.9 4.6 63.8 33.3 30 4,862 35 Sweden 64 71 .. .. 3.0 2.8 83.5 79.7 31 27,198 28 Switzerland .. 66 .. .. 2.4 1.6 87.3 92.7 13 46,382 45 Syrian Arab Republic .. .. .. .. 21.8 7.2 71.5 23.0 .. .. .. Tajikistan .. .. .. .. .. 3.7 .. .. .. .. .. Tanzania 22 .. 12.9 .. 3.8 1.8 58.8 18.3 35 13,405 30 Thailand 8 7 .. .. 4.2 2.5 67.0 46.9 37 84,836 30 Togo 11 .. .. .. 2.6 2.3 60.8 16.6 .. .. .. Tunisia 29 29 .. .. 3.6 2.0 72.8 50.3 .. .. .. Turkey 41 47 6.5 5.1 4.6 4.0 56.0 36.9 45 59,259 25 Turkmenistan .. .. .. .. .. 1.7 .. .. .. .. .. Uganda .. .. .. .. 2.0 2.3 63.0 20.3 30 4,316 30 Ukraine .. .. .. .. .. 2.9 59.0 17.2 40 10,754 30 United Kingdom 55 56 3.6 2.8 5.1 3.0 81.3 90.2 40 44,580 31 United States 49 60 .. .. 6.1 3.8 82.8 92.2 40 278,450 35 Uruguay 43 61 5.0 .. 2.9 2.4 73.0 46.5 0 .. 30 Uzbekistan .. .. .. .. .. 3.8 .. .. .. .. .. Venezuela 31 48 22.3 .. 2.1 1.1 62.8 34.4 .. .. .. Vietnam .. .. .. .. 19.4 2.6 60.3 27.8 50 5,695 25 Yemen, Rep. .. 33 .. .. .. .. 62.8 .. .. .. .. Zambia .. 15 32.2 .. .. 2.8 59.8 16.1 30 1,212 35 Zimbabwe 37 .. 10.8 11.3 5.7 4.0 52.0 26.5 40 3,578 38 World 28 m 32 m 5.2 w 2.8 w 67.7 m 35.3 m Low income .. .. 59.3 19.8 Excl. China & India .. .. 59.0 18.5 Middle income 23 33 69.5 36.3 Lower middle income 19 25 67.9 29.2 Upper middle income 38 49 73.0 42.9 Low and middle income .. .. 64.0 28.6 East Asia & Pacific .. .. 67.8 38.0 Europe & Central Asia .. .. 65.9 33.8 Latin America & Carib. 23 33 67.0 33.7 Middle East & N. Africa .. 13 70.5 34.3 South Asia 16 20 62.8 25.7 Sub-Saharan Africa .. .. 60.8 18.5 High income 55 59 83.4 80.8 a. This copyrighted material is reprinted with permission from the following data providers: PRS Group, 6320 Fly Road, Suite 102, P.O. Box 248, East Syracuse, N.Y. 13057; Institutional Investor, Inc., 488 Madison Avenue, New York, N.Y. 10022; PricewaterhouseCoopers, 1177 Avenue of the Americas, New York, N.Y. 10036. Prior written consent from the original data providers cited must be obtained for third-party use of these data. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 18. Power and transportation Electric power Goods transported Goods Transmission and by road transported by rail Air passengers Consumption per capita distribution losses Paved roads Millions of Ton-km per $ million carried Kilowatt-hours % of output % of total ton-km hauled of GDP (PPP) Thousands Economy 1980 1996 1980 1996 1990 1997 1990 1997 1990 1997 1996 Albania 1,083 904 4 52 .. 30 1,195 80 85,396 5,523 13 Algeria 265 524 11 18 67 69 14,000 .. 25,161 .. 3,494 Angola 67 61 25 28 25 25 .. .. .. .. 585 Argentina 1,170 1,541 13 18 29 29 .. .. 36,412 .. 7,913 Armenia 2,729 905 10 38 99 100 1,533 479 .. .. .. Australia 5,393 8,086 10 7 35 39 .. .. 82,122 .. 30,075 Austria 4,371 5,952 6 6 100 100 13,300 16,600 89,362 78,423 4,719 Azerbaijan 2,440 1,822 14 22 .. .. 3,287 497 .. .. 1,233 Bangladesh 16 97 35 30 7 12 .. .. 8,032 .. 1,252 Belarus 2,455 2,476 9 16 96 98 22,128 9,065 1,297,626 624,045 843 Belgium 4,402 6,878 5 5 81 80 32,100 42,800 46,189 31,976 5,174 Benin 36 48 220 87 20 20 .. .. .. .. 75 Bolivia 226 371 10 12 4 6 .. .. 37,118 .. 1,784 Botswana .. .. .. .. 32 24 .. .. .. .. 104 Brazil 974 1,660 12 17 10 9 .. .. 56,068 .. 22,012 Bulgaria 3,349 3,577 10 13 92 92 13,823 483 360,291 210,161 718 Burkina Faso .. .. .. .. 17 16 .. .. .. .. 138 Burundi .. .. .. .. 18 7 .. .. .. .. 9 Cambodia .. .. .. .. 8 8 .. 1,200 .. .. .. Cameroon 167 171 7 20 11 13 .. .. 33,209 34,023 362 Canada 12,329 15,129 9 7 35 35 54,700 71,473 433,765 .. 22,856 Central African Republic .. .. .. .. .. .. 144 60 .. .. 75 Chad .. .. .. .. 1 1 .. .. .. .. 93 Chile 876 1,864 12 9 14 14 .. .. 15,882 5,998 3,622 China 253 687 8 7 .. .. .. .. 671,824 364,633 51,770 Hong Kong, China 2,167 5,013 11 14 100 100 .. .. .. .. .. Colombia 561 922 16 22 12 12 6,227 .. 2,400 .. 8,342 Congo, Dem. Rep. 147 130 8 3 .. .. .. .. 32,198 .. 178 Congo, Rep. 94 207 1 0 10 10 .. . . 144,851 .. 253 Costa Rica 860 1,349 0 12 15 17 2,243 3,070 .. .. 918 Côte d’Ivoire 192 174 7 16 9 10 .. .. 15,791 13,486 179 Croatia 0 2,291 .. 16 80 82 2,458 470 190,170 86,593 727 Czech Republic 3,595 4,875 7 8 100 100 .. 43,088 .. 207,099 1,394 Denmark 4,245 6,113 7 5 100 100 9,400 9,400 19,119 14,518 5,892 Dominican Republic 433 608 21 25 45 49 .. .. .. .. 30 Ecuador 361 616 14 21 13 19 2,638 3,558 .. .. 1,925 Egypt, Arab Rep. 380 924 13 0 72 78 31,400 31,500 23,310 .. 4,282 El Salvador 293 516 13 13 14 20 .. .. .. .. 1,800 Eritrea .. .. .. .. 19 22 .. .. .. .. .. Estonia 3,433 3,293 5 19 52 51 4,510 2,773 516,391 536,100 149 Ethiopia 16 18 8 1 15 15 .. .. 2,467 .. 743 Finland 7,779 12,979 6 4 61 64 26,300 24,100 99,052 68,994 5,598 France 3,881 6,091 7 6 .. 100 137,000 158,200 49,908 39,109 41,253 Georgia 1,910 1,020 16 23 94 94 7,370 98 .. .. 152 Germany 5,005 5,596 4 5 99 99 245,700 281,300 .. 39,350 40,118 Ghana 426 275 0 0 20 24 .. .. 6,811 .. 197 Greece 2,064 3,395 7 7 92 92 12,600 12,800 6,395 1,913 6,396 Guatemala 212 364 6 13 25 28 .. .. .. .. 300 Guinea .. .. .. .. 15 17 .. .. .. .. 36 Haiti 41 34 26 54 22 24 .. .. .. .. .. Honduras 225 350 14 27 21 20 .. .. .. .. .. Hungary 2,335 2,814 10 13 50 43 1,836 770 247,428 104,327 1,563 India 130 347 18 18 47 46 .. .. 248,469 176,217 13,395 Indonesia 44 296 19 12 46 46 .. .. 8,619 .. 17,139 Iran, Islamic Rep. 491 1,142 10 20 .. 50 .. .. 40,223 .. 7,610 Ireland 2,528 4,363 10 9 94 94 5,100 5,500 14,322 9,132 7,677 Israel 2,826 5,081 5 4 100 100 .. .. 16,663 11,947 3,695 Italy 2,831 4,196 9 7 100 100 177,900 197,600 20,795 18,420 25,839 Jamaica 482 2,108 17 11 64 71 .. .. .. .. 1,388 Japan 4,395 7,083 4 4 69 74 274,444 305,510 11,603 8,664 95,914 Jordan 387 1,187 19 10 100 100 .. .. 78,625 47,242 1,299 Kazakhstan 0 2,865 .. 15 55 83 44,775 6,481 5,042,201 .. 568 Kenya 92 126 16 16 13 14 .. .. 75,496 .. 779 Korea, Rep. 841 4,453 6 5 72 74 31,841 74,504 40,875 24,826 33,003 Kuwait 4,749 12,808 10 0 73 81 .. .. .. .. 2,133 Kyrgyz Republic 1,556 1,479 6 33 90 91 5,627 350 .. .. 488 Lao PDR .. .. .. .. 24 14 120 .. .. .. 125 Latvia 2,664 1,783 26 47 13 38 5,853 800 1,209,517 1,114,210 276 Lebanon 789 1,651 10 13 95 95 .. .. .. .. 775 Lesotho .. .. .. .. 18 18 .. .. .. .. 17 Lithuania 2,715 1,785 12 11 82 89 7,019 8,622 915,522 545,100 214 Macedonia, FYR 0 2,443 .. .. 59 64 1,708 1,210 .. .. 287 Madagascar .. .. .. .. 15 12 .. .. .. .. 542 Malawi .. .. .. .. 22 19 .. .. 14,881 10,003 153 Malaysia 630 2,078 9 11 70 75 .. .. 16,313 9,416 15,118 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      STATES AND MARKETS Electric power Goods transported Goods Transmission and by road transported by rail Air passengers Consumption per capita distribution losses Paved roads Millions of Ton-km per $ million carried Kilowatt-hours % of output % of total ton-km hauled of GDP (PPP) Thousands Economy 1980 1996 1980 1996 1990 1997 1990 1997 1990 1997 1996 Mali .. .. .. .. 11 12 .. .. 53,882 .. 75 Mauritania .. .. .. .. 11 11 .. .. .. .. 235 Mexico 846 1,381 11 15 35 37 108,884 165,000 64,884 53,917 14,678 Moldova 1,495 1,314 8 23 87 87 6,305 780 .. .. 190 Mongolia .. .. .. .. 10 3 1,871 . . 1,324,119 .. .. Morocco 223 408 10 4 49 52 2,638 2,086 72,108 55,523 2,301 Mozambique 370 76 0 0 17 19 .. 110 .. .. 163 Myanmar 31 58 22 36 11 12 .. .. .. .. 335 Namibia .. .. .. .. 11 8 .. .. 308,833 139,137 237 Nepal 13 39 29 28 38 42 .. .. .. .. 755 Netherlands 4,057 5,555 4 4 88 90 22,900 27,600 12,779 9,751 17,114 New Zealand 6,269 8,420 13 11 57 58 .. .. 51,927 .. 9,597 Nicaragua 303 256 14 28 11 10 .. .. .. .. 51 Niger .. .. .. .. 29 8 .. .. .. .. 75 Nigeria 68 85 36 32 30 19 .. .. 3,009 .. 221 Norway 18,289 23,487 9 8 69 74 7,940 11,838 .. .. 12,727 Pakistan 125 333 29 23 54 58 352 84,174 43,586 26,582 5,375 Panama 828 1,140 13 18 32 34 .. .. .. .. 689 Papua New Guinea .. .. .. .. 3 4 .. .. .. .. 970 Paraguay 233 914 6 7 9 10 .. .. .. .. 261 Peru 502 598 13 15 10 10 .. .. 7,486 .. 2,328 Philippines 353 405 2 17 0 0 .. .. .. .. 7,263 Poland 2,470 2,420 10 13 62 66 49,800 95,500 475,103 284,381 1,806 Portugal 1,469 3,044 12 10 .. .. 10,900 11,200 13,976 13,598 4,806 Romania 2,434 1,757 6 12 51 51 13,800 22,400 507,379 231,838 913 Russian Federation 4,706 4,165 8 9 74 .. 300 138 2,725,816 .. 22,117 Rwanda .. .. .. .. 9 9 .. .. .. .. 9 Saudi Arabia 1,356 3,980 9 8 41 43 .. .. 4,634 4,206 11,706 Senegal 97 103 11 16 27 29 .. .. 51,209 .. 155 Sierra Leone .. .. .. .. 11 8 .. .. .. .. 15 Singapore 2,412 7,196 5 4 97 97 .. .. .. .. 11,841 Slovak Republic 3,817 4,450 8 6 99 99 4,180 3,779 . . 297,426 63 Slovenia 4,089 4,766 8 6 72 83 3,440 1,775 142,879 112,529 393 South Africa 3,213 3,719 8 8 30 42 .. .. 430,594 337,153 7,183 Spain 2,401 3,749 9 9 74 99 151,000 186,700 22,427 15,984 27,759 Sri Lanka 96 203 15 17 32 40 19 30 5,926 .. 1,171 Sweden 10,216 14,239 9 7 71 77 26,500 31,200 127,826 103,299 9,879 Switzerland 5,579 6,919 7 7 .. .. 10,400 13,000 .. .. 10,468 Syrian Arab Republic 354 755 18 0 72 23 .. .. 48,075 29,655 599 Tajikistan 2,217 2,292 7 12 72 83 .. .. .. .. 594 Tanzania 50 59 14 12 37 4 .. .. 77,466 91,623 224 Thailand 279 1,289 10 9 55 98 .. .. 14,869 .. 14,078 Togo .. .. .. .. 21 32 .. .. .. .. 75 Tunisia 379 674 12 11 76 79 .. .. 58,795 53,343 1,371 Turkey 439 1,161 12 17 .. 25 .. 139,789 30,838 17,747 8,464 Turkmenistan 1,720 1,020 12 11 74 81 .. .. .. .. 523 Uganda .. .. .. .. .. .. .. .. 12,582 11,567 100 Ukraine 3,598 2,640 8 10 94 95 79,668 20,532 2,109,937 1,411,737 1,151 United Kingdom 4,160 5,198 8 9 100 100 136,300 153,900 17,191 .. 64,209 United States 8,914 11,796 9 7 58 61 1,073,100 1,439,532 360,699 361,911 571,072 Uruguay 977 1,605 15 20 74 90 .. .. 10,455 16,125 504 Uzbekistan 2,085 1,657 9 9 79 87 .. .. .. .. 1,566 Venezuela 2,037 2,498 12 20 36 39 .. .. .. .. 4,487 Vietnam 50 177 18 19 24 25 .. .. 13,526 16,352 2,108 Yemen, Rep. 59 99 6 26 9 8 .. .. .. .. 588 Zambia 1,016 560 7 11 17 .. .. .. 73,728 56,426 235 Zimbabwe 990 765 14 7 14 47 .. .. 274,759 196,429 654 World 1,576 w 2,027 w 8w 8w 39 m 44 m 1,389,943 s Low income 188 433 12 12 17 19 103,110 Excl. China & India 155 218 14 19 17 18 37,945 Middle income 1,585 1,902 9 12 52 51 238,360 Lower middle income 1,835 1,771 8 11 54 51 102,609 Upper middle income 1,188 2,106 10 13 52 47 135,751 Low and middle income 633 886 9 12 29 30 341,470 East Asia & Pacific 260 724 8 9 24 12 143,204 Europe & Central Asia 2,925 2,795 8 11 77 83 46,014 Latin America & Carib. 854 1,347 12 16 22 26 76,275 Middle East & N. Africa 483 1,162 10 9 67 50 37,484 South Asia 116 313 19 19 38 41 22,445 Sub-Saharan Africa 444 439 9 10 17 16 16,049 High income 5,783 8,121 8 6 86 92 1,048,473 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 19. Communications, information, and science and technology Scientists and High- Per 1,000 people Internet engineers technology hosts in R&D exports No. of patent Daily Television Telephone Mobile Personal Per 10,000 Per million % of mfg. applications fileda newspapers Radios sets main lines telephones computers people people exports 1996 Economy 1996 1996 1997 1997 1997 1997 January 1999 1985–95 1997 Residents Nonresidents Albania 34 235 161 23 1 .. 0.30 .. 1 1 18,761 Algeria 38 239 67 48 1 4.2 0.01 .. 22 48 150 Angola 12 54 91 5 1 0.7 0.00 .. .. .. .. Argentina 123 677 289 191 56 39.2 18.28 671 15 .. .. Armenia 23 5 218 150 2 .. 1.01 .. .. 162 20,268 Australia 297 1,385 638 505 264 362.2 420.57 3,166 39 9,196 34,125 Austria 294 740 496 492 144 210.7 176.79 1,631 24 2,506 75,985 Azerbaijan 28 20 211 87 5 .. 0.21 .. .. 165 16,470 Bangladesh 9 50 7 3 0 .. .. .. 0 70 156 Belarus 174 290 314 227 1 .. 0.70 2,339 .. 701 20,347 Belgium 160 792 510 468 95 235.3 162.39 1,814 23 1,356 59,099 Benin 2 108 91 6 1 0.9 0.02 177 .. .. .. Bolivia 55 672 115 69 15 .. 0.78 250 9 17 106 Botswana 27 155 27 56 0 13.4 4.18 .. .. 5 56 Brazil 40 435 316 107 28 26.3 12.88 168 18 2,655 29,451 Bulgaria 253 531 366 323 8 29.7 9.05 .. .. 318 22,235 Burkina Faso 1 32 6 3 0 0.7 0.16 .. .. .. .. Burundi 3 68 10 3 0 .. 0.00 32 .. 1 4 Cambodia .. 127 124 2 3 0.9 0.06 .. .. .. .. Cameroon 7 162 81 5 0 1.5 0.00 .. 3 .. .. Canada 159 1,078 708 609 139 270.6 364.25 2,656 25 3,316 45,938 Central African Republic 2 84 5 3 0 .. 0.00 55 0 .. .. Chad 0 249 2 1 0 .. 0.00 .. .. .. .. Chile 99 354 233 180 28 54.1 20.18 .. 19 189 1,771 China .. 195 270 56 10 6.0 0.14 350 21 11,698 41,016 Hong Kong, China 800 695 412 565 343 230.8 122.71 98 29 41 2,059 Colombia 49 565 217 148 35 33.4 3.93 .. 20 87 1,172 Congo, Dem. Rep. 3 98 43 1 0 .. 0.00 .. .. 2 27 Congo, Rep. 8 124 8 8 0 .. 0.00 .. 16 .. .. Costa Rica 91 271 403 169 19 .. 9.20 .. 14 .. .. Côte d’Ivoire 16 157 61 9 2 3.3 0.16 .. .. .. .. Croatia 114 333 267 335 27 22.0 12.84 1,978 19 259 356 Czech Republic 256 806 447 318 51 82.5 71.79 1,159 13 623 24,856 Denmark 311 1,146 568 633 273 360.2 526.77 2,647 27 2,452 72,151 Dominican Republic 52 177 84 88 16 .. 5.79 .. 23 .. .. Ecuador 70 342 294 75 13 13.0 1.26 169 12 7 354 Egypt, Arab Rep. 38 316 127 56 0 7.3 0.31 458 7 504 706 El Salvador 48 461 250 56 7 .. 1.33 19 16 3 64 Eritrea .. 101 11 6 0 .. 0.00 .. .. .. .. Estonia 173 680 479 321 99 15.1 152.98 2,018 24 12 21,144 Ethiopia 2 194 5 3 0 .. 0.01 .. 0 3 .. Finland 455 1,385 534 556 417 310.7 1,058.13 2,812 26 3,262 61,556 France 218 943 606 575 99 174.4 82.91 2,584 31 17,090 81,418 Georgia .. 553 473 114 6 .. 1.27 .. .. 289 21,124 Germany 311 946 570 550 99 255.5 160.23 2,843 26 56,757 98,338 Ghana 14 238 109 6 1 1.6 0.10 .. .. .. 33 Greece 153 477 466 516 89 44.8 48.81 774 12 434 52,371 Guatemala 31 73 126 41 6 3.0 0.83 99 13 2 102 Guinea .. 47 41 3 0 0.3 0.00 .. .. .. .. Haiti 3 55 5 8 0 .. 0.00 .. .. 3 6 Honduras 55 409 90 37 2 .. 0.16 .. 4 10 126 Hungary 189 697 436 304 69 49.0 82.74 1,033 39 832 24,147 India .. 105 69 19 1 2.1 0.13 149 11 1,660 6,632 Indonesia 23 155 134 25 5 8.0 0.75 .. 20 40 3,957 Iran, Islamic Rep. 24 237 148 107 4 32.7 0.04 521 .. .. .. Ireland 153 703 455 411 146 241.3 148.70 1,871 62 925 52,407 Israel 291 530 321 450 283 186.1 161.96 .. 33 1,363 12,172 Italy 104 874 483 447 204 113.0 58.80 1,325 15 8,860 71,992 Jamaica 64 482 323 140 22 4.6 1.24 8 67 .. .. Japan 580 957 708 479 304 202.4 133.53 6,309 38 340,861 60,390 Jordan 45 287 43 70 2 8.7 0.80 106 26 .. .. Kazakhstan 30 384 234 108 1 .. 0.94 .. .. 1,024 20,064 Kenya 9 108 19 8 0 2.3 0.23 .. 11 15 39,034 Korea, Rep. 394 1,037 341 444 150 150.7 40.00 2,636 39 68,446 45,548 Kuwait 376 688 491 227 116 82.9 32.80 .. 4 .. .. Kyrgyz Republic 13 115 44 76 0 .. 4.04 703 24 126 20,179 Lao PDR 4 139 4 5 1 1.1 0.00 .. .. .. .. Latvia 246 699 592 302 31 7.9 42.59 1,189 15 197 21,498 Lebanon 141 892 354 179 135 31.8 5.56 .. .. .. .. Lesotho 7 48 24 10 1 .. 0.09 .. .. 2 37,043 Lithuania 92 292 377 283 41 6.5 27.48 .. 21 101 21,249 Macedonia, FYR 19 184 252 204 6 .. 2.56 .. .. 53 18,934 Madagascar 4 192 45 3 0 1.3 0.04 11 2 7 20,800 Malawi 3 256 2 4 0 .. 0.00 .. 3 3 39,031 Malaysia 163 432 166 195 113 46.1 21.36 87 67 .. .. 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      STATES AND MARKETS Scientists and High- Per 1,000 people Internet engineers technology hosts in R&D exports No. of patent Daily Television Telephone Mobile Personal Per 10,000 Per million % of mfg. applications fileda newspapers Radios sets main lines telephones computers people people exports 1996 Economy 1996 1996 1997 1997 1997 1997 January 1999 1985–95 1997 Residents Nonresidents Mali 1 49 10 2 0 0.6 0.00 .. .. .. .. Mauritania 1 150 89 5 0 5.3 0.06 .. .. .. .. Mexico 97 324 251 96 18 37.3 11.64 213 33 389 30,305 Moldova 59 720 302 145 1 3.8 1.17 1,539 9 290 20,245 Mongolia 27 139 63 37 1 5.4 0.08 943 2 114 20,882 Morocco 26 241 160 50 3 2.5 0.20 .. 27 90 237 Mozambique 3 39 4 4 0 1.6 0.08 .. 8 .. .. Myanmar 10 89 7 5 0 .. 0.00 .. .. .. .. Namibia 19 143 32 58 8 18.6 15.79 .. .. .. .. Nepal 11 37 4 8 0 .. 0.07 .. 0 .. .. Netherlands 305 963 541 564 110 280.3 358.51 2,656 44 4,884 61,958 New Zealand 223 1,027 501 486 149 263.9 360.44 1,778 11 1,421 26,947 Nicaragua 32 283 190 29 2 .. 1.47 214 38 .. .. Niger 0 69 26 2 0 0.2 0.02 .. .. .. .. Nigeria 24 197 61 4 0 5.1 0.03 15 .. .. .. Norway 593 920 579 621 381 360.8 717.53 3,678 24 1,550 25,628 Pakistan 21 92 65 19 1 4.5 0.23 54 4 16 782 Panama 62 299 187 134 6 .. 2.66 .. 14 31 142 Papua New Guinea 15 91 24 11 1 .. 0.25 .. .. .. .. Paraguay 50 182 101 43 17 .. 2.18 .. 4 .. .. Peru 43 271 143 68 18 12.3 1.91 625 10 52 565 Philippines 82 159 109 29 18 13.6 1.21 157 56 163 2,634 Poland 113 518 413 194 22 36.2 28.07 1,299 12 2,414 24,902 Portugal 75 306 523 402 152 74.4 50.01 1,185 11 105 71,544 Romania .. 317 226 167 9 8.9 7.42 1,382 7 1,831 22,139 Russian Federation 105 344 390 183 3 32.0 10.04 3,520 19 18,138 28,149 Rwanda 0 102 .. 3 0 .. 0.00 24 .. .. .. Saudi Arabia 59 319 260 117 17 43.6 0.15 .. 29 27 810 Senegal 5 141 41 13 1 11.4 0.21 .. 55 .. .. Sierra Leone 5 251 20 4 0 .. 0.03 .. .. .. .. Singapore 324 739 354 543 273 399.5 210.02 2,728 71 215 38,403 Slovak Republic 185 580 401 259 37 241.6 33.27 1,821 15 201 22,865 Slovenia 206 416 353 364 47 188.9 89.83 2,544 16 301 21,686 South Africa 30 316 125 107 37 41.6 34.67 938 .. .. .. Spain 99 328 506 403 110 122.1 67.21 1,210 17 2,689 81,294 Sri Lanka 29 210 91 17 6 4.1 0.29 173 .. 50 21,138 Sweden 446 907 531 679 358 350.3 487.13 3,714 34 7,077 76,364 Switzerland 330 969 536 661 147 394.9 315.52 .. 28 2,699 75,576 Syrian Arab Republic 20 274 68 88 0 1.7 0.00 .. 1 .. .. Tajikistan 20 .. 281 38 0 .. 0.12 709 .. 32 19,570 Tanzania 4 278 21 3 1 1.6 0.04 .. .. .. .. Thailand 65 204 234 80 33 19.8 3.35 119 43 203 4,355 Togo 4 217 19 6 1 5.8 0.24 .. .. .. .. Tunisia 31 218 182 70 1 8.6 0.07 388 11 46 128 Turkey 111 178 286 250 26 20.7 4.30 261 9 367 19,668 Turkmenistan .. 96 175 78 0 .. 0.55 .. .. 66 18,948 Uganda 2 123 26 2 0 1.4 0.05 .. .. .. 38,497 Ukraine 54 872 493 186 1 5.6 3.13 3,173 .. 3,640 22,862 United Kingdom 332 1,445 641 540 151 242.4 240.99 2,417 41 25,269 104,084 United States 212 2,115 847 644 206 406.7 1,131.52 3,732 44 111,883 111,536 Uruguay 116 610 242 232 46 21.9 46.61 688 8 25 182 Uzbekistan 3 452 273 63 0 .. 0.10 1,760 .. 914 21,088 Venezuela 206 471 172 116 46 36.6 3.37 208 10 182 1,822 Vietnam 4 106 180 21 2 4.6 0.00 308 .. 37 22,206 Yemen, Rep. 15 64 273 13 1 1.2 0.01 .. 0 .. .. Zambia 14 121 80 9 0 .. 0.31 .. .. 6 93 Zimbabwe 18 96 29 17 1 9.0 0.87 .. 6 30 181 World ..w 380 w 280 w 144 w 40 w 58.4 w 75.22 w Low income .. 147 162 32 5 4.4 0.17 Excl. China & India 13 133 59 16 1 .. 0.23 Middle income 75 383 272 136 24 32.4 10.15 Lower middle income 63 327 247 108 11 12.2 4.91 Upper middle income 95 469 302 179 43 45.5 19.01 Low and middle income .. 218 194 65 11 12.3 3.08 East Asia & Pacific .. 206 237 60 15 11.3 1.66 Europe & Central Asia 99 412 380 189 13 17.7 13.00 Latin America & Carib. 71 414 263 110 26 31.6 9.64 Middle East & N. Africa 33 265 140 71 6 9.8 0.25 South Asia .. 99 69 18 1 2.1 0.14 Sub-Saharan Africa 12 172 44 16 4 7.2 2.39 High income 286 1,300 664 552 188 269.4 470.12 a. Other patent applications filed in 1996 include those filed under the auspices of the African Intellectual Property Organization (75 by residents, 20,863 by nonresidents), the African Regional Industrial Property Organization (10 by residents, 20,347 by nonresidents), the European Patent Office (38,546 by residents, 48,068 by nonresi- dents), and the Eurasian Patent Organization (39 by residents, 18,055 by nonresidents). The original information was provided by the World Intellectual Property Organization (WIPO). The International Bureau of WIPO assumes no liability or responsibility with regard to the transformation of these data. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 20. Global trade Exports of Imports of Merchandise exports commercial services Merchandise imports commercial services Millions of dollars Mfg. % of total Millions of dollars Millions of dollars Mfg. % of total Millions of dollars Economy 1983 1998a 1983 1997 1983 1997 1983 1998a 1983 1997 1983 1997 Albania .. 135 .. 99 13 52 .. 650 .. .. 22 93 Algeria 12,480 9,380 1 4 649 .. 10,399 9,080 71 63 2,251 .. Angola 1,822 4,222 .. .. 129 226 983 2,332 .. .. 628 1,738 Argentina 7,836 25,227 16 33 1,405 2,941 4,504 31,402 75 88 2,026 6,104 Armenia .. 235 .. .. .. .. .. 875 .. .. .. .. Australia 20,113 55,949 15 27 3,954 18,360 21,458 64,678 69 80 6,735 18,385 Austria 15,427 61,717 85 89 9,343 29,213 19,423 68,260 71 81 5,662 28,371 Azerbaijan .. 545 .. .. .. .. .. 1,075 .. .. .. .. Bangladesh 725 3,778 66 .. 164 266 2,165 6,710 41 .. 329 1,184 Belarus .. 7,016 .. .. .. .. .. 8,509 .. .. .. .. Belgiumb 51,939 171,703 68 76 9,589 33,431 55,313 158,843 58 74 9,119 31,606 Benin 67 195 .. .. 43 .. 318 613 .. .. 83 .. Bolivia 755 1,103 1 16 95 180 577 1,983 70 81 229 379 Botswana 635 2,942 .. .. 84 145 735 2,261 .. .. 186 339 Brazil 21,899 50,992 39 53 1,648 6,765 16,801 60,980 34 74 3,734 17,612 Bulgaria 12,140 4,275 .. 61 1,059 1,308 12,290 4,980 .. 50 598 1,153 Burkina Faso 58 327 10 .. .. .. 291 735 53 .. .. .. Burundi 80 86 .. .. .. .. 182 121 .. .. .. .. Cambodia 15 330 .. .. .. 150 180 660 .. .. .. 182 Cameroon 976 1,860 .. 8 408 242 1,224 1,358 .. 63 703 485 Canada 76,749 214,298 51 62 8,284 29,290 64,789 205,038 73 80 11,869 35,944 Central African Republic 80 174 .. 33 11 .. 77 232 .. 60 91 .. Chad 105 202 .. .. .. .. 157 240 .. .. .. .. Chile 3,830 14,895 7 14 756 3,592 3,085 18,828 48 73 1,116 3,854 China 22,151 183,757 55 85 2,466 24,516 21,323 140,165 70 77 1,840 30,063 Hong Kong, China c 22,454 174,145 89 93 6,267 38,179 24,409 188,745 73 87 3,696 23,209 Colombia 3,001 10,890 18 30 819 4,053 4,963 15,840 70 79 1,214 4,171 Congo, Dem. Rep. 1,131 530 .. .. .. .. 470 460 .. .. .. .. Congo, Rep. 640 1,600 9 .. 71 45 648 550 79 .. 715 553 Costa Rica 873 4,066 26 23 264 1,490 988 4,676 66 85 249 1,135 Côte d’Ivoire 2,090 4,183 11 .. 376 577 1,839 2,817 57 52 919 1,186 Croatia .. 4,541 .. 72 .. 3,994 .. 8,383 .. 73 .. 1,972 Czech Republic .. 26,360 .. 85 .. 7,033 .. 28,820 .. 79 .. 5,305 Denmark 16,053 47,047 55 61 5,018 15,105 16,266 45,795 60 73 4,425 14,936 Dominican Republic 785 903 20 .. 451 2,071 1,471 4,716 40 .. 292 956 Ecuador 2,348 4,133 1 8 297 689 1,487 5,496 84 71 469 1,089 Egypt, Arab Rep. 3,215 3,908 12 38 2,955 9,096 10,275 13,600 63 62 2,509 5,813 El Salvador 735 1,263 21 39 127 276 892 3,112 61 67 238 354 Eritrea .. .. .. .. .. .. .. .. .. .. .. .. Estonia .. 3,208 .. 65 .. 1,314 .. 4,750 .. 71 .. 649 Ethiopiad .. 551 .. .. 119 318 .. 1,100 .. .. 220 378 Finland 12,518 42,360 74 83 2,489 7,097 12,826 31,945 59 73 2,429 8,180 France 94,943 307,031 70 76 33,380 80,269 105,907 287,210 56 76 24,694 62,086 Georgia .. 195 .. .. .. .. .. 1,095 .. .. .. .. Germany e 169,417 539,689 84 86 23,285 74,722 152,877 466,619 .. 68 34,714 118,144 Ghana 1,158 1,550 0 .. 35 152 1,248 1,680 28 .. 91 395 Greece 4,413 9,709 48 51 2,812 9,224 9,500 23,470 52 72 1,304 4,196 Guatemala 1,159 2,550 24 31 43 542 1,126 4,619 74 68 244 627 Guinea 488 730 .. .. .. 70 267 1,000 .. .. .. 204 Haiti 166 133 .. .. .. .. 441 606 .. .. .. .. Honduras 672 1,580 9 20 80 328 803 2,417 67 72 154 359 Hungary 8,770 22,940 61 46 583 4,825 8,555 25,820 59 66 447 3,634 India 9,148 33,210 52 73 3,167 8,679 14,061 42,850 53 51 3,622 12,277 Indonesia 21,152 48,840 6 42 546 6,792 16,352 27,420 62 73 4,228 16,214 Iran, Islamic Rep. 19,950 13,150 .. .. 478 743 18,320 13,000 .. .. 4,110 2,899 Ireland 8,592 63,252 62 80 1,092 6,020 9,159 43,681 67 79 1,351 15,032 Israel 5,108 23,282 80 92 2,671 8,338 9,574 29,130 59 76 3,136 10,867 Italy 72,877 240,869 85 89 17,435 71,729 79,808 213,995 41 67 13,570 70,146 Jamaica 718 1,352 14 26 520 1,428 1,494 3,025 49 65 384 1,146 Japan 146,965 387,965 96 95 19,560 68,136 126,437 280,531 21 54 33,540 122,079 Jordan 580 1,750 46 .. 1,102 1,717 3,036 3,910 54 .. 911 1,241 Kazakhstan .. 5,410 .. .. .. 833 .. 4,300 .. .. .. 1,081 Kenya 876 2,053 15 25 359 764 1,334 3,273 52 64 295 731 Korea, Rep. 24,446 133,223 91 87 3,662 25,439 26,192 93,345 51 61 3,369 29,037 Kuwait 11,504 9,700 19 4 679 1,513 7,373 8,200 83 80 2,896 4,302 Kyrgyz Republic .. 605 .. 38 .. .. .. 835 .. 48 .. .. Lao PDR 41 359 .. .. .. .. 150 648 .. .. .. .. Latvia .. 1,812 .. 61 .. 1,027 .. 3,189 .. 62 .. 637 Lebanon 691 716 .. .. .. .. 3,661 7,060 .. .. .. .. Lesotho 31 170 .. .. 23 .. 485 980 .. .. 30 .. Lithuania .. 3,755 .. 60 .. 1,020 .. 6,025 .. 66 .. 850 Macedonia, FYR .. .. .. .. .. .. .. .. .. .. .. .. Madagascar 263 215 9 36 40 253 387 477 61 73 122 280 Malawi 229 530 6 .. 29 .. 311 760 71 .. 128 .. Malaysia 14,130 73,275 25 76 1,743 14,868 13,198 58,540 70 82 3,872 17,363 * Taiwan, China 25,094 109,890 89 96 2,342 17,021 20,308 104,240 51 73 3,626 24,112 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      GLOBAL LINKS Exports of Imports of Merchandise exports commercial services Merchandise imports commercial services Millions of dollars Mfg. % of total Millions of dollars Millions of dollars Mfg. % of total Millions of dollars Economy 1983 1998a 1983 1997 1983 1997 1983 1998a 1983 1997 1983 1997 Mali 165 518 .. .. 28 62 353 811 .. .. 154 324 Mauritania 292 448 .. .. 20 19 240 380 .. .. 170 197 Mexico 25,559 117,505 37 80 3,749 11,214 10,896 128,940 96 83 4,300 11,813 Moldova .. 680 .. .. .. .. .. 1,075 .. .. .. .. Mongolia 560 418 .. 10 55 47 852 443 .. 65 43 87 Morocco 2,006 7,295 40 35 774 2,203 3,592 10,270 44 48 476 1,267 Mozambique 132 200 .. 20 .. .. 636 760 .. 66 .. .. Myanmar 378 866 .. .. 56 .. 268 2,053 .. .. 70 .. Namibia 846 1,400 .. .. .. 356 921 1,600 .. .. .. 494 Nepal 94 402 41 77 107 795 464 1,716 63 32 83 216 Netherlands 64,684 198,212 49 67 13,133 48,529 61,652 184,148 52 68 13,824 43,812 New Zealand 5,414 12,114 20 29 1,315 3,905 5,333 12,501 68 82 1,749 4,893 Nicaragua 429 610 8 33 36 124 826 1,553 66 67 101 229 Niger 299 268 .. .. .. .. 324 424 .. .. .. .. Nigeria 10,357 10,360 0 .. 402 786 12,254 9,900 54 .. 2,211 4,694 Norway 17,997 39,645 29 23 6,988 14,256 13,497 36,193 74 77 7,102 14,460 Pakistan 3,077 8,370 63 86 668 1,463 5,329 9,170 51 52 847 2,413 Panama 321 712 9 18 976 1,382 1,412 3,097 60 71 517 1,154 Papua New Guinea 813 2,142 2 .. 62 436 1,120 1,697 50 .. 314 747 Paraguay 269 1,021 7 17 134 .. 546 3,050 62 65 149 .. Peru 3,015 5,550 8 15 649 1,447 2,548 10,050 60 61 892 2,190 Philippines 4,890 29,330 52 85 1,516 15,130 7,977 31,960 60 74 1,598 14,073 Poland 11,580 26,300 64 72 1,990 8,969 10,600 48,020 52 77 1,783 5,681 Portugal 4,599 23,503 72 84 1,427 7,523 8,240 35,082 50 73 1,131 6,148 Romania 10,160 8,295 .. 79 727 1,398 7,640 11,820 .. 67 726 1,998 Russian Federation .. 73,900 .. 23 .. 13,898 .. 59,500 .. 46 .. 19,082 Rwanda 121 88 .. .. 18 42 279 299 .. .. 86 151 Saudi Arabia 45,861 38,800 .. 9 4,151 4,484 39,197 23,700 84 73 16,424 13,927 Senegal 618 924 .. .. 200 364 1,025 1,189 .. .. 253 405 Sierra Leone 119 17 25 .. 17 71 160 91 37 .. 40 79 Singaporec 21,833 109,846 49 84 7,733 30,379 28,158 101,496 55 82 3,747 19,422 Slovak Republic .. 10,665 .. 76 .. 2,151 .. 12,965 .. 60 .. 2,062 Slovenia .. 9,120 .. 90 .. 2,032 .. 10,100 .. 76 .. 1,439 South Africa 18,508 26,322 18 43 2,669 4,882 15,813 29,268 63 64 3,360 6,050 Spain 19,734 109,037 69 69 11,252 43,570 29,193 132,789 37 68 4,825 24,264 Sri Lanka 1,066 4,770 28 .. 282 850 1,820 5,970 55 .. 396 1,270 Sweden 27,446 84,455 77 78 6,191 17,584 26,098 67,637 63 76 6,166 19,462 Switzerland 25,592 78,741 91 93 8,230 25,615 29,192 80,017 73 84 4,625 14,132 Syrian Arab Republic 1,923 3,916 15 10 384 1,366 4,542 3,900 46 68 698 1,302 Tajikistan .. 560 .. .. .. .. .. 725 .. .. .. .. Tanzania 364 674 .. 9 106 460 832 1,454 .. 64 162 706 Thailand 6,368 53,575 31 71 1,733 15,619 10,287 41,800 60 77 1,845 17,126 Togo 163 237 25 .. 58 .. 282 373 58 .. 112 .. Tunisia 1,850 5,746 44 78 921 2,427 3,107 8,333 64 75 483 1,014 Turkey 5,728 26,140 46 75 1,917 19,193 9,235 46,400 43 72 1,073 8,085 Turkmenistan .. 650 .. .. .. .. .. 1,015 .. .. .. .. Uganda 372 557 .. .. .. .. 377 1,312 .. .. .. .. Ukraine .. 12,825 .. .. .. 4,937 .. 14,746 .. .. .. 2,268 United Kingdom 91,619 272,692 63 83 27,060 91,928 100,080 316,077 65 81 20,962 71,265 United States 205,639 682,977 65 80 51,040 231,896 269,878 944,586 60 78 39,590 152,448 Uruguay 1,045 2,848 29 37 255 1,465 788 3,842 41 76 455 903 Uzbekistan .. 3,940 .. .. .. .. .. 4,205 .. .. .. .. Venezuela 13,937 17,200 2 11 1,035 1,290 6,419 15,600 67 69 2,636 5,213 Vietnam 616 8,980 .. .. .. .. 1,526 11,015 .. .. .. .. Yemen, Rep. 701 2,481 .. .. .. .. 3,101 1,901 .. .. .. .. Zambia 836 901 .. .. 79 .. 851 807 .. .. 321 .. Zimbabwe 1,135 2,508 .. 27 124 .. 1,205 3,092 .. 77 409 .. World 1,757,216 t 5,414,844 t 66 w 78 w 356,892 t 1,326,312 t 1,755,569 t 5,358,567 t 57 w 73 w 377,843 t 1,307,618 t Low income 88,785 334,896 42 75 10,869 51,538 102,719 295,254 63 71 21,228 85,092 Excl. China & India .. .. .. .. 5,457 18,068 .. .. .. .. 17,369 44,337 Middle income 410,520 953,662 41 64 57,320 230,847 381,036 1,018,458 60 71 87,836 247,297 Lower middle income .. 329,691 .. .. 27,570 101,056 205,214 370,345 .. .. 35,868 103,897 Upper middle income 225,563 622,990 48 72 30,088 130,233 184,578 647,211 60 73 51,234 143,661 Low and middle income 493,984 1,288,084 42 66 68,072 282,785 482,412 1,313,145 61 71 108,707 332,063 East Asia & Pacific 97,271 537,234 52 78 12,292 105,518 101,854 411,054 62 73 17,773 128,602 Europe & Central Asia .. 249,450 .. 51 .. 77,726 .. 309,720 .. 64 .. 59,655 Latin America & Carib. 99,355 270,876 25 50 14,268 44,471 74,429 337,406 63 77 21,329 63,390 Middle East & N. Africa 118,705 103,782 .. 16 14,926 30,412 123,259 113,156 68 .. 38,488 36,039 South Asia 14,868 50,743 53 75 4,457 12,396 25,032 67,304 52 52 5,329 17,494 Sub-Saharan Africa 49,231 84,706 12 .. 6,603 13,026 51,878 86,534 59 .. 14,347 25,133 High income 1,274,830 4,124,433 72 81 288,345 1,043,005 1,278,838 4,040,845 56 74 271,116 977,279 a. WTO 1998 figures are based on preliminary estimates made in early 1999; for many countries, the estimates are based on incomplete preliminary data and are subject to revision. b. Includes Luxembourg. c. Includes reexports. d. Data prior to 1992 include Eritrea. e. Data prior to 1990 refer to the Federal Republic of Germany before unification. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 21. Aid and financial flows External debt Millions of dollars Total Official development assistance Present value Net private capital flows Foreign direct investment Millions of dollars % of GNP Dollars per capita % of GNP Economy 1990 1997 1990 1997 1990 1997 1997 1990 1997 1990 1997 Albania 31 47 0 48 349 706 22 3 51 0.5 6.7 Algeria –424 –543 0 7 27,877 30,921 65 10 8 0.4 0.6 Angola 237 –24 –335 350 8,594 10,160 206 29 37 3.3 10.2 Argentina –203 19,834 1,836 6,645 62,233 123,221 38 6 6 0.1 0.1 Armenia 0 51 0 51 41 666 26 1 45 0.1 9.7 Australia .. .. 7,465 8,737 .. .. .. .. .. .. .. Austria .. .. 653 2,354 .. .. .. .. .. .. .. Azerbaijan .. 658 .. 650 .. 504 10 0 23 0.0 5.0 Bangladesh 70 118 3 135 12,768 15,125 20 19 8 6.9 2.3 Belarus 173 169 7 200 189 1,162 5 18 4 0.5 0.2 Belgium .. .. .. .. .. .. .. .. .. .. .. Benin 1 3 1 3 1,292 1,624 46b 57 39 15.0 10.7 Bolivia 3 812 27 601 4,275 5,247 51 b 85 92 12.0 9.2 Botswana 77 95 95 100 563 562 9 117 81 4.2 2.4 Brazil 562 43,377 989 19,652 119,877 193,663 23 1 3 0.0 0.1 Bulgaria –42 569 4 498 10,890 9,858 96 2 25 0.1 2.2 Burkina Faso 0 0 0 0 834 1,297 30b 38 35 12.3 15.6 Burundi –5 1 1 1 907 1,066 58 49 19 24.1 12.6 Cambodia 0 200 0 203 1,854 2,129 52 17 36 13.0 12.1 Cameroon –125 16 –113 45 6,679 9,293 93 39 36 4.2 5.9 Canada .. .. 7,581 7,132 .. .. .. .. .. .. .. Central African Republic 0 6 1 6 698 885 53 86 27 17.2 9.3 Chad –1 15 0 15 524 1,026 35 55 31 19.9 14.3 Chile 2,098 9,637 590 5,417 19,227 31,440 43 8 9 0.4 0.2 China 8,107 60,828 3,487 44,236 55,301 146,697 15 2 2 0.6 0.2 Hong Kong, China .. .. .. .. .. .. .. 7 1 0.1 0.0 Colombia 345 10,151 500 5,982 17,222 31,777 27 3 7 0.3 0.2 Congo, Dem. Rep. –24 1 –12 1 10,270 12,330 215 24 4 10.5 3.2 Congo, Rep. –100 9 0 9 4,953 5,071 247 104 99 9.9 14.7 Costa Rica 23 104 163 57 3,756 3,548 34 78 –1 4.4 0.0 Côte d’Ivoire 57 –91 48 327 17,251 15,609 141b 59 31 7.5 4.7 Croatia .. 2,397 .. 388 .. 6,842 36 0 10 0.0 0.2 Czech Republic 876 1,818 207 1,286 6,383 21,456 40 1 10 0.0 0.2 Denmark .. .. 1,132 2,792 .. .. .. .. .. .. .. Dominican Republic 130 401 133 405 4,372 4,239 27 16 9 1.7 0.5 Ecuador 183 829 126 577 12,109 14,918 72 16 15 1.7 0.9 Egypt, Arab Rep. 698 2,595 734 891 32,947 29,849 28 104 32 12.4 2.5 El Salvador 8 61 2 11 2,148 3,282 25 68 51 7.4 2.7 Eritrea .. 0 .. 0 .. 76 4 .. 33 .. 14.8 Estonia 104 347 82 266 58 658 14 10 44 0.3 1.4 Ethiopiaa –45 28 12 5 8,634 10,078 131 21 11 15.8 10.1 Finland .. .. 812 2,128 .. .. .. .. .. .. .. France .. .. 13,183 23,045 .. .. .. .. .. .. .. Georgia 21 50 0 50 79 1,446 20 0 46 0.0 4.7 Germany .. .. 2,532 –344 .. .. .. .. .. .. .. Ghana –5 203 15 130 3,873 5,982 57 b 38 28 9.8 7.2 Greece .. .. 1,005 984 .. .. .. 4 .. 0.0 .. Guatemala 44 166 48 90 3,080 4,086 21 23 29 2.7 1.7 Guinea –1 –23 18 1 2,476 3,520 65 49 55 10.9 10.1 Haiti 8 3 8 3 889 1,057 21 27 44 5.8 11.8 Honduras 77 124 44 122 3,724 4,698 86 93 51 16.4 7.0 Hungary –308 2,605 0 2,079 21,276 24,373 52 6 16 0.2 0.4 India 1,872 8,307 162 3,351 83,717 94,404 18 2 2 0.4 0.4 Indonesia 3,235 10,863 1,093 4,677 69,872 136,174 62 10 4 1.6 0.4 Iran, Islamic Rep. –392 –303 –362 50 9,020 11,816 9 2 3 0.1 0.2 Ireland .. .. 627 2,727 .. .. .. .. .. .. .. Israel .. .. 101 2,706 .. .. .. 294 204 2.7 1.2 Italy .. .. 6,411 3,700 .. .. .. .. .. .. .. Jamaica 92 377 138 137 4,671 3,913 90 117 28 7.4 1.8 Japan .. .. 1,777 3,200 .. .. .. .. .. .. .. Jordan 254 61 38 22 8,177 8,234 110 275 104 23.8 6.8 Kazakhstan 117 2,158 100 1,321 35 4,278 19 7 8 0.4 0.6 Kenya 124 –87 57 20 7,056 6,486 49 51 16 14.8 4.6 Korea, Rep. 1,056 13,069 788 2,844 46,976 143,373 33 1 –3 0.0 0.0 Kuwait .. .. .. 20 .. .. .. 3 1 0.0 0.0 Kyrgyz Republic 0 50 0 50 4 928 39 5 52 1.1 14.1 Lao PDR 6 90 6 90 1,768 2,320 53 44 71 20.6 19.5 Latvia 43 559 29 521 65 503 8 1 33 0.0 1.5 Lebanon 12 1,070 6 150 1,779 5,036 32 71 58 7.5 1.6 Lesotho 17 42 17 29 396 660 35 83 46 13.9 7.4 Lithuania –3 637 0 355 56 1,540 15 1 27 0.0 1.1 Macedonia, FYR .. 8 .. 15 .. 1,542 75 .. 75 .. 6.9 Madagascar 7 13 22 14 3,701 4,105 85 35 59 13.7 24.3 Malawi 2 1 0 2 1,558 2,206 45 b 59 34 28.8 13.7 Malaysia 769 9,312 2,333 5,106 15,328 47,228 48 26 –11 1.1 –0.3 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      GLOBAL LINKS External debt Millions of dollars Total Official development assistance Present value Net private capital flows Foreign direct investment Millions of dollars % of GNP Dollars per capita % of GNP Economy 1990 1997 1990 1997 1990 1997 1997 1990 1997 1990 1997 Mali –8 15 –7 15 2,467 2,945 73b 58 44 20.5 18.7 Mauritania 6 2 7 3 2,096 2,453 169 122 102 25.8 23.9 Mexico 8,253 20,533 2,634 12,477 104,431 149,690 37 2 1 0.1 0.0 Moldova 0 257 0 60 39 1,040 52 2 15 0.3 3.5 Mongolia 28 16 2 7 350 718 49 134 99 .. 26.7 Morocco 341 1,303 165 1,200 24,458 19,321 53 44 17 4.2 1.4 Mozambique 35 37 9 35 4,653 5,991 135b 76 58 45.6 29.6 Myanmar 153 180 161 80 4,695 5,074 .. 4 1 .. .. Namibia .. .. 29 137 .. .. .. 91 102 5.0 5.0 Nepal –8 12 6 23 1,640 2,398 25 23 19 11.8 8.3 Netherlands .. .. 12,352 9,012 .. .. .. .. .. .. .. New Zealand .. .. 1,735 2,650 .. .. .. .. .. .. .. Nicaragua 21 157 0 173 10,708 5,677 244b 101 90 39.0 22.7 Niger 9 –12 –1 2 1,726 1,579 56b 52 35 16.5 18.6 Nigeria 467 1,285 588 1,539 33,440 28,455 72 3 2 1.0 0.5 Norway .. .. 1,003 3,545 .. .. .. .. .. .. .. Pakistan 182 2,097 244 713 20,663 29,664 38 10 5 2.7 1.0 Panama 127 1,443 132 1,030 6,678 6,338 88 42 47 2.0 1.5 Papua New Guinea 204 143 155 200 2,594 2,272 41 109 78 13.5 7.8 Paraguay 67 273 76 250 2,104 2,052 20 14 24 1.1 1.3 Peru 59 3,094 41 2,030 20,064 30,496 45 19 20 1.3 0.8 Philippines 639 4,164 530 1,222 30,580 45,433 51 20 9 2.9 0.8 Poland 71 6,787 89 4,908 49,366 39,889 27 35 17 2.4 0.5 Portugal .. .. 2,610 1,713 .. .. .. .. .. .. .. Romania 4 2,274 0 1,215 1,140 10,442 29 11 9 0.6 0.6 Russian Federation 5,562 12,453 0 6,241 59,797 125,645 27 2 5 0.0 0.2 Rwanda 6 1 8 1 712 1,111 33 43 75 11.6 32.0 Saudi Arabia .. .. 1,864 –1,129 .. .. .. 3 1 0.0 0.0 Senegal 42 44 57 30 3,732 3,671 56 112 49 14.9 9.7 Sierra Leone 36 4 32 4 1,151 1,149 89 18 27 9.1 16.0 Singapore .. .. 5,575 8,631 .. .. .. –1 0 0.0 0.0 Slovak Republic 278 1,074 0 165 2,008 9,989 48 1 13 0.0 0.3 Slovenia .. .. .. 321 .. .. .. .. 49 .. 0.5 South Africa .. 3,610 .. 1,725 .. 25,222 19 .. 12 .. 0.4 Spain .. .. 13,984 5,556 .. .. .. .. .. .. .. Sri Lanka 54 574 43 430 5,863 7,638 35 43 19 9.2 2.3 Sweden .. .. 1,982 9,867 .. .. .. .. .. .. .. Switzerland .. .. 4,961 5,506 .. .. .. .. .. .. .. Syrian Arab Republic 18 69 71 80 17,068 20,865 114 58 13 6.0 1.2 Tajikistan 0 20 0 20 10 901 34 2 17 0.4 5.0 Tanzaniac 5 143 0 158 6,447 7,177 77 48 31 30.3 13.9 Thailand 4,399 3,444 2,444 3,745 28,165 93,416 61 14 10 0.9 0.4 Togo 0 –6 0 0 1,275 1,339 59 74 29 16.4 8.4 Tunisia –122 903 76 316 7,691 11,323 58 48 21 3.3 1.1 Turkey 1,782 12,221 684 805 49,424 91,205 43 21 0 0.8 0.0 Turkmenistan .. 847 .. 85 .. 1,771 59 2 2 0.1 0.4 Uganda 16 179 0 180 2,583 3,708 31b 42 41 16.2 12.8 Ukraine 369 1,419 0 623 551 10,901 21 6 4 0.3 0.4 United Kingdom .. .. 32,518 37,007 .. .. .. .. .. .. .. United States .. .. 47,918 93,448 .. .. .. .. .. .. .. Uruguay –192 632 0 160 4,415 6,652 32 17 17 0.7 0.3 Uzbekistan 40 435 40 285 60 2,760 11 3 6 0.3 0.5 Venezuela –126 6,282 451 5,087 33,170 35,541 41 4 1 0.2 0.0 Vietnam 16 1,994 16 1,800 23,270 21,629 78 19 13 4.2 4.1 Yemen, Rep. 30 –138 –131 –138 6,345 3,856 56 37 23 9.3 7.3 Zambia 194 79 203 70 7,265 6,758 136 62 65 16.0 16.7 Zimbabwe 85 32 –12 70 3,247 4,961 52 36 29 4.2 4.1 World ..s ..s 192,662 s 400,394 s ..s ..s 14 w 11 w 1.4 w 0.7 w Low income 14,819 88,685 5,732 59,509 473,398 669,626 15 11 4.3 2.9 Excl. China & India 4,840 19,551 2,083 11,922 334,380 428,525 .. .. .. .. Middle income 28,091 210,049 18,697 103,786 998,783 d 1,645,941 d 13 9 1.0 0.5 Lower middle income .. .. .. .. .. .. 13 10 1.5 0.9 Upper middle income .. .. .. .. .. .. 10 5 0.3 0.1 Low and middle income 42,910 298,734 24,429 163,295 1,472,181 d 2,315,567 d 14 11 1.5 0.9 East Asia & Pacific 18,720 104,257 11,135 64,284 286,061 654,551 6 4 1.0 0.5 Europe & Central Asia 7,695 49,875 1,097 22,314 221,028 390,579 19 15 0.9 0.5 Latin America & Carib. 12,411 118,918 8,188 61,573 475,366 703,669 12 13 0.5 0.3 Middle East & N. Africa 622 7,899 2,711 5,240 182,399 192,378 45 19 2.3 1.0 South Asia 2,174 11,110 464 4,662 129,899 154,946 5 3 1.5 0.8 Sub-Saharan Africa 1,288 6,674 834 5,222 177,428 219,445 40 26 10.7 5.0 High income .. .. 168,233 237,099 601e 1,034e .. .. .. .. a. Data prior to 1992 include Eritrea. b. Data are from debt sustainability analyses undertaken as part of the Heavily Indebted Poor Countries (HIPC) Debt Initiative. Present value estimates for these countries are for public and publicly guaranteed debt only. c. GNP data refer to mainland Tanzania only. d. Includes data for Gibraltar not included in other tables. e. Data refer only to Malta. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄ Table 1a. Key indicators for other economies Gross national Carbon Surface Population product (GNP) GNP per capita GNP measured Life Adult dioxide at PPPa area density Avg. annual Avg. annual expectancy illiteracy emissions Population Thousands People per Millions growth growth Millions Per capita at birth % of people Thousands Thousands of sq. km sq. km of dollars rate (%) Dollars rate (%) of dollars (dollars) Years 15 and above of tons Economy 1998 1996 1998 1998b 1997–98 1998b 1997–98 1998 1998 1997 1997 1996 Afghanistan 25,761 652.1 40 .. .. .. c .. .. .. 45 67 1,176 American Samoa 63 0.2 315 .. .. . .d .. .. .. .. .. 282 Andorra 65 0.5 144 .. .. . .e .. .. .. .. .. .. Antigua and Barbuda 67 0.4 152 555 2.1 8,300 1.2 631 9,440 75 .. 322 Aruba 94 0.2 495 .. .. .. e .. .. .. .. .. 1,517 Bahamas, The 294 13.9 29 .. 3.0 . .e 1.4 3,073 10,460 74 4 1,707 Bahrain 641 0.7 929 4,912 2.1 7,660 –1.3 8,787 13,700 73 14 10,578 Barbados 266 0.4 618 2,096 2.5 7,890 2.2 3,257 12,260 76 .. 835 Belize 236 23.0 10 615 0.8 2,610 –1.8 927 3,940 75 .. 355 Bermuda 63 0.1 1,260 .. .. .. e .. .. .. .. .. 462 Bhutan 759 47.0 16 .. .. .. c .. .. .. 61 .. 260 Bosnia and Herzegovina .. 51.1 .. .. .. . .g .. .. .. .. .. 3,111 Brunei 314 5.8 60 .. .. .. e .. .. .. 76 10 5,071 Cape Verde 412 4.0 102 437 4.5 1,060 1.7 1,216f 2,950f 68 29 121 Cayman Islands 36 0.3 138 .. .. ..e .. .. .. .. .. 282 Channel Islands 149 0.3 478 .. .. . .e .. .. .. 78 .. .. Comoros 531 2.2 238 196 1.0 370 –1.5 787 f 1,480 f 60 45 55 Cuba 11,103 110.9 101 .. .. .. g .. .. .. 76 4 31,170 Cyprus 753 9.3 82 .. .. ..e .. .. .. 78 4 5,379 Djibouti 653 23.2 28 .. .. . .g .. .. .. 50 .. 366 Dominica 74 0.8 98 222 0.2 3,010 0.2 291 3,940 76 .. 81 Equatorial Guinea 432 28.1 15 647 36.0 1,500 32.5 1,900 4,400 50 20 143 Faeroe Islands 42 1.4 42 .. .. .. e .. .. .. .. .. 630 Fiji 827 18.3 45 1,745 –4.2 2,110 –5.7 2,962 3,580 73 8 762 French Guiana 163 90.0 2 .. .. .e .. .. .. .. .. 920 French Polynesia 228 4.0 62 .. .. . .e .. .. .. 72 .. 561 Gabon 1,181 267.7 5 4,664 5.7 3,950 3.2 7,865 6,660 52 .. 3,690 Gambia, The 1,216 11.3 122 413 5.0 340 2.0 1,743f 1,430f 53 67 216 Greenland 56 341.7 0 .. .. .. e .. .. .. 68 .. 509 Grenada 96 0.3 283 305 1.6 3,170 0.9 454 4,720 72 .. 161 Guadeloupe 431 1.7 255 .. .. . .d .. .. .. 77 .. 1,513 Guam 149 0.6 271 .. .. . .e .. .. .. 77 .. 4,078 Guinea-Bissau 1,161 36.1 41 186 –28.9 160 –30.4 872 750 44 66 231 Guyana 857 215.0 4 660 –1.5 770 –2.6 2,302 2,680 64 2 953 Iceland 274 103.0 3 7,675 5.1 28,010 4.1 6,256 22,830 79 .. 2,195 Iraq 22,347 438.3 51 .. .. . .g .. .. .. 58 .. 91,387 Isle of Man 73 0.6 122 .. .. . .d .. .. .. .. .. .. Kiribati 85 0.7 117 101 15.2 1,180 12.6 297 3,480 60 .. 22 Korea, Dem. Rep. 23,171 120.5 192 .. .. .. g .. .. .. 63 .. 254,326 Liberia 2,969 111.4 31 .. .. . .c .. .. .. 47 52 326 Libya 5,330 1,759.5 3 .. .. . d .. .. .. 70 24 40,579 Liechtenstein 32 0.2 200 .. .. . .e .. .. .. .. .. .. Luxembourg 427 2.6 161 18,587 4.2 43,570 3.0 15,962 37,420 76 .. 8,281 Macao 455 0.0 22,763 .. .. .. e .. .. .. 78 8 1,407 Maldives 262 0.3 874 323 .. 1,230 .. .. .. 67 4 297 Malta 378 0.3 1,180 3,564 4.1 9,440 3.4 f 5,138 13,610 f 77 9 1,751 Marshall Islands 62 0.2 310 .. –4.4 1,540 .. .. .. .. .. .. Martinique 397 1.1 374 .. .. . .e .. .. .. 79 3 2,023 Mauritius 1,159 2.0 571 4,288 4.5 3,700 3.5 10,899 9,400 71 17 1,744 Mayotte 126 0.4 315 .. .. . .d .. .. .. .. .. .. Micronesia, Fed. Sts. 113 0.7 155 203 –3.1 1,800 –4.8 .. .. 67 .. .. Monaco 32 0.0 1,600 .. .. .. e .. .. .. .. .. .. Netherlands Antilles 213 0.8 266 .. .. . .e .. .. .. 75 4 6,430 New Caledonia 206 18.6 11 .. .. . .e .. .. .. 73 .. 1,751 Northern Mariana Islands 70 0.5 140 .. .. . .e .. .. .. .. .. .. Oman 2,322 212.5 11 .. .. .. d .. .. .. 73 33 15,143 Palau 19 0.5 32 .. .. .. d .. .. .. 71 .. 245 Puerto Rico 3,857 9.0 435 .. .. . .d .. .. .. 75 7 15,806 Qatar 742 11.0 67 .. .. . .e .. .. .. 74 20 29,121 Reunion 687 2.5 275 .. .. . .e .. .. .. 75 14 1,561 Samoa 176 2.8 62 180 1.8 1,020 0.5 607 3,440 69 .. 132 São Tomé and Principe 142 1.0 148 40 2.5 280 0.2 192 1,350 64 .. 77 Seychelles 79 0.5 175 507 –1.3 6,450 –2.5 827 10,530 71 .. 169 Solomon Islands 415 28.9 15 311 –7.0 750 –9.7 862f 2,080f 70 .. 161 Somalia 9,076 637.7 14 .. .. . .c .. .. .. 47 .. 15 St. Kitts and Nevis 41 0.4 113 250 3.7 6,130 3.7 324 7,940 70 .. 103 St. Lucia 160 0.6 263 546 3.7 3,410 3.0 738 4,610 70 .. 191 St. Vincent and the Grenadines 113 0.4 290 274 2.3 2,420 1.6 463 4,090 73 .. 125 Sudan 28,347 2,505.8 12 8,221 5.0 290 2.7 38,602 1,360 55 47 3,473 Suriname 413 163.3 3 685 2.7 1,660 2.4 .. .. 70 .. 2,099 Swaziland 988 17.4 57 1,384 1.8 1,400 –1.3 3,540 3,580 60 23 341 Tonga 99 0.8 137 167 –1.0 1,690 –1.8 381 3,860 70 .. 117 Trinidad and Tobago 1,317 5.1 257 5,835 6.3 4,430 5.5 8,854 6,720 73 2 22,237 United Arab Emirates 2,671 83.6 32 48,666 –5.7 18,220 –8.9 52,659f 19,720f 75 25 81,843 Vanuatu 182 12.2 15 231 2.1 1,270 –0.4 574f 3,160f 65 .. 62 Virgin Islands (U.S.) 118 0.3 348 .. .. . .e .. .. .. 77 .. 12,912 West Bank and Gaza 2,673 6.2 411 .. .. . .g .. .. .. 71 .. .. Yugoslavia, FR (Serb./Mont.) 10,640 102.2 104 .. .. . .g .. .. .. 72 .. .. a. Purchasing power parity; see the Technical Notes. b. Calculated using the World Bank Atlas method. c. Estimated to be low income ($760 or less). d. Estimated to be upper middle income. ($3,031 to $9,360). e. Estimated to be high income ($9,361 or more). f. The estimate is based on regression; others are extrapolated from the latest International Comparison Programme benchmark estimates. g. Estimated to be lower middle income ($761 to $3,030). (c) The International Bank for Reconstruction and Development / The World Bank Technical Notes T hese technical notes discuss the sources and coverage, practices, and definitions differ methods used to compile the 149 indicators in- widely; and cross-country and intertemporal cluded in this edition of Selected World Devel- comparisons involve complex technical and opment Indicators. The notes follow the order conceptual problems that cannot be unequiv- in which the indicators appear in the tables. ocally resolved. For these reasons, although the data are drawn from the sources thought Sources to be most authoritative, they should be con- The data published in the Selected World De- strued only as indicating trends and character- velopment Indicators are taken from World izing major differences among economies Development Indicators 1999. Where possible, rather than offering precise quantitative mea- however, revisions reported since the closing sures of those differences. Also, national sta- date of that edition have been incorporated. tistical agencies tend to revise their historical In addition, newly released estimates of pop- data, particularly for recent years. Thus, data ulation and gross national product (GNP) per of different vintages may be published in dif- capita for 1998 are included in table 1. ferent editions of World Bank publications. The World Bank draws on a variety of Readers are advised not to compile such data sources for the statistics published in the World from different editions. Consistent time series Development Indicators. Data on external debt are available on the World Development Indi- are reported directly to the World Bank by cators 1999 CD-ROM. developing member countries through the Ratios and growth rates Debtor Reporting System. Other data are drawn mainly from the United Nations and its For ease of reference, the tables usually show specialized agencies, from the International ratios and rates of growth rather than the sim- Monetary Fund (IMF), and from country re- ple underlying values. Values in their original ports to the World Bank. Bank staff estimates form are available on the World Development are also used to improve currentness or consis- Indicators 1999 CD-ROM. Unless otherwise tency. For most countries, national accounts noted, growth rates are computed using the estimates are obtained from member govern- least-squares regression method (see “Statisti- ments through World Bank economic mis- cal methods” below). Because this method sions. In some instances these are adjusted by takes into account all available observations staff to ensure conformity with international during a period, the resulting growth rates re- definitions and concepts. Most social data flect general trends that are not unduly influ- from national sources are drawn from regular enced by exceptional values. To exclude the administrative files, special surveys, or periodic effects of inflation, constant-price economic census inquiries. The Data Sources section fol- indicators are used in calculating growth rates. lowing the Technical Notes lists the principal Data in italics are for a year or period other international sources used. than that specified in the column heading— up to two years before or after for economic Data consistency and reliability indicators, and up to three years for social in- Considerable effort has been made to stan- dicators because the latter tend to be collected dardize the data, but full comparability can- less regularly and change less dramatically over not be assured, and care must be taken in in- short periods. terpreting the indicators. Many factors affect Constant-price series data availability, comparability, and reliabil- ity: statistical systems in many developing An economy’s growth is measured by the in- economies are still weak; statistical methods, crease in value added produced by the individ-  (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  uals and enterprises operating in that economy. Thus, measur- age annual growth rate of GNP per capita is calculated from ing real growth requires estimates of GDP and its components constant-price GNP per capita in national currency units. GNP valued in constant prices. The World Bank collects constant- measured at PPP is GNP converted to U.S. dollars by the pur- price national accounts series in national currencies and recorded chasing power parity (PPP) exchange rate. At the PPP rate, one in the country’s original base year. To obtain comparable series dollar has the same purchasing power over domestic GNP that of constant-price data, it rescales GDP (and value added) by in- the U.S. dollar has over U.S. GNP; dollars converted by this dustrial origin to a common reference year, currently 1995. This method are sometimes called international dollars. process gives rise to a discrepancy between the rescaled GDP and GNP, the broadest measure of national income, measures the sum of the rescaled components. Because allocating the dis- total value added from domestic and foreign sources claimed by crepancy would give rise to distortions in the growth rate, the residents. GNP comprises gross domestic product (GDP) plus discrepancy is left unallocated. net receipts of primary income from nonresident sources. The World Bank uses GNP per capita in U.S. dollars to classify Summary measures economies for analytical purposes and to determine borrowing The summary measures for regions and income groups, pre- eligibility. When calculating GNP in U.S. dollars from GNP sented at the end of most tables, are calculated by simple addi- reported in national currencies, the World Bank follows its Atlas tion when they are expressed in levels. Aggregate growth rates conversion method. This involves using a three-year average of and ratios are usually computed as weighted averages. The sum- exchange rates to smooth the effects of transitory exchange rate mary measures for social indicators are weighted by population fluctuations. (See “Statistical methods” below for further dis- or subgroups of population, except for infant mortality, which cussion of the Atlas method.) Note that growth rates are calcu- is weighted by the number of births. See the notes on specific lated from data in constant prices and national currency units, indicators for more information. not from the Atlas estimates. For summary measures that cover many years, calculations Because nominal exchange rates do not always reflect inter- are based on a uniform group of economies so that the compo- national differences in relative prices, table 1 also shows GNP sition of the aggregate does not change over time. Group mea- converted into international dollars using PPP exchange rates. sures are compiled only if the data available for a given year ac- PPP rates allow a standard comparison of real price levels be- count for at least two-thirds of the full group, as defined for the tween countries, just as conventional price indexes allow com- 1987 benchmark year. As long as this criterion is met, econ- parison of real values over time. The PPP conversion factors omies for which data are missing are assumed to behave like used here are derived from the most recent round of price sur- those that provide estimates. Readers should keep in mind that veys conducted by the International Comparison Programme, a the summary measures are estimates of representative aggregates joint project of the World Bank and the regional economic for each topic and that nothing meaningful can be deduced commissions of the United Nations. This round of surveys, about behavior at the country level by working back from group completed in 1996 and covering 118 countries, is based on a indicators. In addition, the weighting process may result in dis- 1993 reference year. Estimates for countries not included in the crepancies between subgroup and overall totals. survey are derived from statistical models using available data. Rankings are based on 210 economies and include the 78 Table 1. Size of the economy economies with sparse data or populations of less than 1.5 mil- Population is based on the de facto definition, which counts all lion from table 1a. Range estimates for GNP and GNP per residents, regardless of legal status or citizenship, except for capita have been used to rank many of these 78 economies— refugees not permanently settled in the country of asylum, who such as Liechtenstein and Luxembourg, which rank first and are generally considered part of the population of the country second respectively for GNP per capita. of origin. The indicators shown are midyear estimates (see the Table 2. Quality of life technical note for table 3). Surface area is a country’s total area, including areas under Growth of private consumption per capita is the average an- inland bodies of water and coastal waterways. nual rate of change in private consumption divided by the Population density is midyear population divided by land midyear population. (See the definition of private consumption area. Land area is a country’s total area excluding areas under in the Technical Note to table 13.) The distribution-corrected inland bodies of water and coastal waterways. Density is calcu- growth rate is 1 minus the Gini index (see the Technical Note lated using the most recently available data on land area. to table 5) multiplied by the annual rate of growth of private Gross national product (GNP) is the sum of value added by consumption. Improvements in private consumption per capita all resident producers, plus any taxes (less subsidies) not included are generally associated with a reduction in poverty, but where in the valuation of output, plus net receipts of primary income the distribution of income or consumption is highly unequal, (employee compensation and property income) from nonresi- the poor may not share in the improvement. The relationship dent sources. Data are converted from national currency to cur- between the rate of poverty reduction and the distribution of rent U.S. dollars by the World Bank Atlas method (see “Statisti- income or consumption, as measured by an index such as the cal methods” below). Average annual growth rate of GNP is Gini index, is complicated. But Ravallion and Chen (1997; see calculated from constant-price GNP in national currency units. Data Sources) have found that the rate of poverty reduction is, GNP per capita is GNP divided by midyear population. It is on average, proportional to the distribution-corrected rate of converted into current U.S. dollars by the Atlas method. Aver- growth of private consumption. (c) The International Bank for Reconstruction and Development / The World Bank      Prevalence of child malnutrition is the percentage of chil- counting occur even in high-income economies; in developing dren under age 5 whose weight for age is less than minus 2 stan- countries such errors may be substantial because of limits on dard deviations from the median of the reference population, transportation, communication, and the resources required to which is based on children from the United States, who are as- conduct a full census. Moreover, the international comparabil- sumed to be well nourished. Weight for age is a composite in- ity of population indicators is limited by differences in the con- dicator of both weight for height (wasting) and height for age cepts, definitions, data collection procedures, and estimation (stunting). Estimates of child malnutrition are from the WHO. methods used by national statistical agencies and other organi- Under-5 mortality rate is the probability that a child born zations that collect population data. The data in table 3 are pro- in the indicated year will die before reaching age 5, if the child vided by national statistical offices or by the United Nations is subject to current age-specific mortality rates. The probabil- Population Division. ity is expressed as a rate per 1,000 children. Average annual population growth rate is the exponential Life expectancy at birth is the number of years a newborn rate of change for the period (see “Statistical methods” below). infant would live if patterns of mortality prevailing at its birth Population aged 15–64 is a commonly accepted measure of were to stay the same throughout its life. the number of people who are potentially economically active. Age-specific mortality data such as infant and child mortal- In many developing countries, however, children under age 15 ity rates, along with life expectancy at birth, are probably the work full or part time, and in some high-income economies best general indicators of a community’s current health status many workers postpone retirement past age 65. and are often cited as overall measures of a population’s welfare Total labor force comprises people who meet the definition or quality of life. The main sources of mortality data are vital established by the International Labour Organization (ILO) for registration systems and direct or indirect estimates based on the economically active population: all people who supply labor sample surveys or censuses. Because civil registers with relatively for the production of goods and services during a specified pe- complete vital registration systems are fairly uncommon, esti- riod. It includes both the employed and the unemployed. Al- mates must be obtained from sample surveys or derived by ap- though national practices vary, in general the labor force includes plying indirect estimation techniques to registration, census, or the armed forces and first-time jobseekers but excludes home- survey data. Indirect estimates rely on estimated actuarial (“life”) makers and other unpaid caregivers and workers in the informal tables, which may be inappropriate for the population con- sector. Data on the labor force are compiled by the ILO from cerned. Life expectancy at birth and age-specific mortality rates census or labor force surveys. Despite the ILO’s efforts to en- are generally estimates based on the most recently available cen- courage the use of international standards, labor force data are sus or survey; see the Primary data documentation table in not fully comparable because of differences among countries, and World Development Indicators 1999. sometimes within countries, in definitions and methods of col- Adult illiteracy rate is the percentage of persons aged 15 and lection, classification, and tabulation. The labor force estimates above who cannot, with understanding, read and write a short, reported in table 3 were calculated by applying activity rates from simple statement about their everyday life. Literacy is difficult the ILO database to the World Bank’s population estimates to to define and to measure. The definition here is based on the create a labor force series consistent with those estimates. This concept of functional literacy: a person’s ability to use reading procedure sometimes results in estimates that differ slightly from and writing skills effectively in the context of his or her society. those published in the ILO’s Yearbook of Labour Statistics. Measuring literacy using such a definition requires census or Average annual labor force growth rate is calculated using sample survey measurements under controlled conditions. In the exponential end-point method (see “Statistical methods” practice, many countries estimate the number of illiterate adults below). from self-reported data or from estimates of school completion Females as a percentage of the labor force shows the extent rates. Because of these differences in method, comparisons to which women are active in the labor force. Estimates are from across countries—and even over time within countries—should the ILO database. These estimates are not comparable interna- be made with caution. tionally because in many countries large numbers of women as- Urban population is the share of the population living in sist on farms or in other family enterprises without pay, and areas defined as urban in each country. countries use different criteria to determine the extent to which Access to sanitation in urban areas is the percentage of the such workers are to be counted in the labor force. urban population served by connections to public sewers or Children aged 10–14 in the labor force is the share of that household systems such as pit privies, pour-flush latrines, septic age group that is working or seeking work. Reliable estimates of tanks, communal toilets, or other such facilities. child labor are difficult to obtain. In many countries child labor is illegal or officially presumed not to exist and is therefore not Table 3. Population and labor force reported or included in surveys or recorded in official data. Data Total population includes all residents regardless of legal status are also subject to underreporting because they do not include or citizenship, except for refugees not permanently settled in the children engaged in agricultural or household activities with country of asylum, who are generally considered part of the pop- their families. ulation of their country of origin. The indicators shown are Table 4. Poverty midyear estimates. Population estimates are usually based on national censuses. Intercensal estimates are interpolations or ex- Survey year is the year in which the underlying data were trapolations based on demographic models. Errors and under- collected. (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  Rural population below the national poverty line is the per- are not always available, and when they are not, there is little centage of the rural population living below the rural poverty line choice but to use income. There are still other problems. House- determined by national authorities. Urban population below the hold survey questionnaires can differ widely, for example in the national poverty line is the percentage of the urban population number of distinct categories of consumer goods they identify. living below the urban poverty line determined by national Survey quality varies, and even similar surveys may not be authorities. Total population below the national poverty line is strictly comparable. the percentage of the total population living below the national Comparisons across countries at different levels of develop- poverty line. National estimates are based on population- ment also pose a potential problem because of differences in the weighted subgroup estimates from household surveys. relative importance of consumption of nonmarket goods. The Population below $1 PPP a day and Population below $2 local market value of all consumption in kind (including con- PPP a day are the percentages of the population living at those sumption from a household’s own production, particularly im- levels of consumption or income at 1985 prices, adjusted for portant in underdeveloped rural economies) should be included purchasing power parity. in the measure of total consumption expenditure. Similarly, the Poverty gap at $1 PPP a day and Poverty gap at $2 PPP a imputed profit from production of nonmarket goods should be day are calculated as the average difference between the poverty included in income. This is not always done, although such line and actual income or consumption for all poor households, omissions were a far bigger problem in surveys before the 1980s expressed as a percentage of the poverty line. This measure re- than today. Most survey data now include valuations for con- flects the depth of poverty as well as its prevalence. sumption or income from own production. Nonetheless, valua- International comparisons of poverty data entail both con- tion methods vary: for example, some surveys use the price at ceptual and practical problems. Different countries have differ- the nearest market, whereas others use the average farmgate sell- ent definitions of poverty, and consistent comparisons between ing price. countries using the same definition can be difficult. National The international poverty measures in table 4 are based on poverty lines tend to have greater purchasing power in rich the most recent PPP estimates from the latest version of the countries, where more generous standards are used than in poor Penn World Tables (National Bureau of Economic Research countries. 1997; see Data Sources). However, any revisions in the PPP International poverty lines attempt to hold the real value of conversion factor of a country to incorporate better price in- the poverty line constant between countries. The standard of $1 dexes can produce dramatically different poverty lines in local a day, measured in 1985 international prices and adjusted to currency. local currency using PPP conversion factors, was chosen for Whenever possible, consumption has been used as the welfare World Development Report 1990: Poverty because it is typical of indicator for deciding who is poor. When only household income poverty lines in low-income economies. PPP conversion factors is available, average income has been adjusted to accord with ei- are used because they take into account the local prices of goods ther a survey-based estimate of mean consumption (when avail- and services that are not traded internationally. However, these able) or an estimate based on consumption data from national ac- factors were designed not for making international poverty com- counts. This procedure adjusts only the mean, however; nothing parisons but for comparing aggregates in the national accounts. can be done to correct for the difference between the Lorenz (in- As a result, there is no certainty that an international poverty come distribution) curves for consumption and income. line measures the same degree of need or deprivation across Empirical Lorenz curves were weighted by household size, countries. so they are based on percentiles of population, not of house- Problems can arise in comparing poverty measures within holds. In all cases the measures of poverty have been calculated countries as well as between them. For example, the cost of food from primary data sources (tabulations or household data) staples—and the cost of living generally—are typically higher rather than existing estimates. Estimates from tabulations re- in urban than in rural areas. So the nominal value of the urban quire an interpolation method; the method chosen is Lorenz poverty line should be higher than the rural poverty line. But it curves with flexible functional forms, which have proved reli- is not always clear that the difference between urban and rural able in past work. poverty lines found in practice properly reflects the difference Table 5. Distribution of income or consumption in the cost of living. For some countries the urban poverty line in common use has a higher real value—meaning that it allows Survey year is the year in which the underlying data were poor people to buy more commodities for consumption—than collected. does the rural poverty line. Sometimes the difference has been Gini index measures the extent to which the distribution of so large as to imply that the incidence of poverty is greater in income (or, in some cases, consumption expenditure) among urban than in rural areas, even though the reverse is found when individuals or households within an economy deviates from a adjustments are made only for differences in the cost of living. perfectly equal distribution. The Gini index measures the area Other issues arise in measuring household living standards. between the Lorenz curve (described in the technical note to The choice between income and consumption as a welfare indi- table 4) and a hypothetical line of absolute equality, expressed cator is one. Incomes are generally more difficult to measure ac- as a percentage of the maximum area under the line. As defined curately, and consumption accords better with the idea of a here, a Gini index of zero would represent perfect equality, and standard of living than does income, which can vary over time an index of 100 would imply perfect inequality (one person or even if the standard of living does not. But consumption data household accounting for all income or consumption). (c) The International Bank for Reconstruction and Development / The World Bank      Percentage share of income or consumption is the share lation. Enrollment data are based on annual enrollment surveys, that accrues to deciles or quintiles of the population ranked by typically conducted at the beginning of the school year. They do income or consumption. Percentage shares by quintiles may not not reflect actual attendance or dropout rates during the school add up to 100 because of rounding. year. Problems affecting cross-country comparisons of enroll- Data on personal or household income or consumption ment data stem from inadvertent or deliberate misreporting of come from nationally representative household surveys. The age and from errors in estimates of school-age populations. Age- data in the table refer to different years between 1982 and 1997. sex structures from censuses or vital registration systems, the pri- Footnotes to the survey year indicate whether the rankings are mary sources of data on school-age populations, are commonly based on income or consumption. Distributions are based on subject to underenumeration, especially of young children. percentiles of population, not of households. Where the origi- Percentage of cohort reaching grade 5 is the share of stu- nal data from the household survey were available, they have dents enrolled in primary school who eventually reach fifth been used to directly calculate the income or consumption grade. Because tracking data for individual students are not shares by quintile. Otherwise, shares have been estimated from available, aggregate student flows from one grade to the next are the best available grouped data. estimated using data on average promotion, repetition, and The distribution indicators have been adjusted for household dropout rates. Other flows, caused by new entrants, reentrants, size, providing a more consistent measure of income or con- grade skipping, migration, or school transfers during the school sumption per capita. No adjustment has been made for differ- year, are not considered. This procedure, called the recon- ences in the cost of living in different parts of the same country structed cohort method, makes three simplifying assumptions: because the necessary data are generally unavailable. For further that dropouts never return to school; that promotion, repeti- details on the estimation method for low- and middle-income tion, and dropout rates remain constant over the entire period economies, see Ravallion (1996; see Data Sources). in which the cohort is enrolled; and that the same rates apply to Because the underlying household surveys differ in method all pupils enrolled in a given grade, regardless of whether they and in the type of data collected, the distribution indicators are previously repeated a grade. not strictly comparable across countries. These problems are di- Expected years of schooling is the average number of years minishing as survey methods improve and become more stan- of formal schooling that a child is expected to receive, includ- dardized, but strict comparability is still impossible. The income ing university education and years spent in repetition. It may distribution and Gini indexes for the high-income economies also be interpreted as an indicator of the total educational re- are directly calculated from the Luxembourg Income Study sources, measured in school years, that a child will require over database. The estimation method used here is consistent with the course of his or her “lifetime” in school. that applied to developing countries. Data on education are compiled by the United Nations Ed- The following sources of noncomparability should be noted. ucational, Scientific, and Cultural Organization (UNESCO) First, the surveys can differ in many respects, including whether from official responses to surveys and from reports provided by they use income or consumption expenditure as the living stan- education authorities in each country. Because coverage, defini- dard indicator. Income is typically more unequally distributed tions, and data collection methods vary across countries and than consumption. In addition, the definitions of income used over time within countries, data on education should be inter- in surveys are usually very different from the economic definition preted with caution. of income (the maximum level of consumption consistent with Table 7. Health keeping productive capacity unchanged). Consumption is usu- ally a much better welfare indicator, particularly in developing Public expenditure on health consists of recurrent and capital countries. Second, households differ in size (number of members) spending from government (central and local) budgets, external and in the extent of income sharing among members. Individu- borrowings and grants (including donations from international als differ in age and in consumption needs. Differences between agencies and nongovernmental organizations), and social (or countries in these respects may bias distribution comparisons. compulsory) health insurance funds. Because few developing countries have national health accounts, compiling estimates of Table 6. Education public health expenditure is complicated in countries where Public expenditure on education is the percentage of GNP ac- state, provincial, and local governments are involved in health counted for by public spending on public education plus subsi- care financing. Such data are not regularly reported and, when dies to private education at the primary, secondary, and tertiary reported, are often of poor quality. In some countries health ser- levels. It may exclude spending on religious schools, which play vices are considered social services and so are excluded from a significant role in many developing countries. Data for some health sector expenditures. The data on health expenditure in countries and for some years refer to spending by the ministry table 7 were collected by the World Bank as part of its health, of education of the central government only and thus exclude nutrition, and population strategy. No estimates were made for education expenditures by other central government ministries countries with incomplete data. and departments, local authorities, and others. Access to safe water is the percentage of the population with Net enrollment ratio is the number of children of official reasonable access to an adequate amount of safe water (includ- school age (as defined by the education system) enrolled in pri- ing treated surface water and untreated but uncontaminated mary or secondary school, expressed as a percentage of the num- water, such as from springs, sanitary wells, and protected bore- ber of children of official school age for those levels in the popu- holes). In urban areas the source may be a public fountain or (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  standpipe located not more than 200 meters from the residence. water, including land irrigated by controlled flooding. Arable In rural areas the definition implies that household members do land includes land defined by the Food and Agriculture Orga- not have to spend a disproportionate part of the day fetching nization (FAO) as land under temporary crops (double-cropped water. An “adequate” amount of safe water is that needed to sat- areas are counted once), temporary meadows for mowing or for isfy metabolic, hygienic, and domestic requirements, usually pasture, land under market or kitchen gardens, and land tem- about 20 liters per person per day. The definition of safe water porarily fallow. Land abandoned as a result of shifting cultiva- has changed over time. tion is not included. Access to sanitation is the percentage of the population with The comparability of land use data from different countries disposal facilities that can effectively prevent human, animal, is limited by variations in definitions, statistical methods, and and insect contact with excreta. Suitable facilities range from the quality of data collection. For example, countries may de- simple but protected pit latrines to flush toilets with sewerage. fine land use differently. The FAO, the primary compiler of To be effective, all facilities must be correctly constructed and these data, occasionally adjusts its definitions of land use cate- properly maintained. gories and sometimes revises earlier data. Because the data thus Infant mortality rate is the number of infants who die be- reflect changes in data-reporting procedures as well as actual fore reaching 1 year of age, expressed per 1,000 live births in a changes in land use, apparent trends should be interpreted with given year (see the discussion of age-specific mortality rates in caution. the technical note to table 2). Agricultural machinery refers to wheel and crawler tractors Contraceptive prevalence rate is the percentage of women (excluding garden tractors) in use in agriculture at the end of who are practicing, or whose sexual partners are practicing, any the calendar year specified or during the first quarter of the fol- form of contraception. It is usually measured for married lowing year. women aged 15–49 only. Contraceptive prevalence includes all Agricultural productivity refers to agricultural value added methods: ineffective traditional methods as well as highly effec- per agricultural worker, measured in constant 1995 U.S. dol- tive modern methods. Unmarried women are often excluded lars. Agricultural value added includes that from forestry and from the surveys, and this may bias the estimate. The rates are fishing. Thus interpretations of land productivity should be obtained mainly from demographic and health surveys and con- made with caution. To smooth annual fluctuations in agricul- traceptive prevalence surveys. tural activity, the indicators have been averaged over three years. Total fertility rate is the number of children who would be Food production index covers food crops that are consid- born to a woman if she were to live to the end of her childbear- ered edible and that contain nutrients. Coffee and tea are ex- ing years and bear children in accordance with current age- cluded because, although edible, they have no nutritive value. specific fertility rates. Data are from vital registration systems The food production index is prepared by the FAO, which ob- or, in their absence, from censuses or sample surveys. Provided tains data from official and semiofficial reports of crop yields, that the censuses or surveys are fairly recent, the estimated rates area under production, and livestock numbers. Where data are are considered reliable. As with other demographic data, inter- not available, the FAO makes estimates. The index is calculated national comparisons are limited by differences in data defini- using the Laspeyres formula: production quantities of each com- tion, collection, and estimation methods. modity are weighted by average international commodity prices Maternal mortality ratio is the number of women who die in the base period and summed for each year. The FAO’s index during pregnancy or childbirth, per 100,000 live births. Mater- may differ from those of other sources because of differences in nal mortality ratios are difficult to measure because health in- coverage, weights, concepts, time periods, calculation methods, formation systems are often weak. Classifying a death as mater- and use of international prices. nal requires a cause-of-death attribution by medically qualified Table 9. Water use, deforestation, staff, based on information available at the time of death. Even and protected areas then, some doubt may remain about the diagnosis in the ab- sence of an autopsy. In many developing countries, causes of Freshwater resources consists of internal renewable resources, death are assigned by nonphysicians and often attributed to “ill- which include flows of rivers and groundwater from rainfall in defined causes.” Maternal deaths in rural areas often go unre- the country and river flows from other countries. Freshwater re- ported. The data in table 7 are official estimates from adminis- sources per capita are calculated using the World Bank’s popu- trative records, survey-based indirect estimates, or estimates lation estimates. derived from a demographic model developed by the United Data on freshwater resources are based on estimates of runoff Nations Children’s Fund (UNICEF) and the WHO. In all cases into rivers and recharge of groundwater. These estimates are the standard errors of maternal mortality ratios are large, and based on different sources and refer to different years, so cross- this makes the indicator particularly unsuitable for monitoring country comparisons should be made with caution. Because changes over a short period. they are collected intermittently, the data may hide significant variations in total renewable water resources from one year to Table 8. Land use and agricultural productivity the next. These annual averages also obscure large seasonal and Land under permanent crops is land cultivated with crops that interannual variations in water availability within countries. occupy the land for long periods and do not need to be re- Data for small countries and countries in arid and semiarid planted after each harvest, excluding trees grown for wood or zones are less reliable than those for larger countries and coun- timber. Irrigated land refers to areas purposely provided with tries with more rainfall. (c) The International Bank for Reconstruction and Development / The World Bank      Annual freshwater withdrawals refers to total water with- underestimate of the extent and number of protected areas. The drawals, not counting evaporation losses from storage basins. It data do not include sites protected under local or provincial law. also includes water from desalination plants in countries where Data on protected areas are compiled from a variety of these are a significant source of water. Withdrawal data are for sources by the World Conservation Monitoring Centre, a joint single years between 1980 and 1997 unless otherwise indicated. venture of the United Nations Environment Programme, the Caution is advised in comparing data on annual freshwater with- World Wide Fund for Nature, and the World Conservation drawals, which are subject to variations in collection and esti- Union. Because of differences in definitions and reporting prac- mation methods. Withdrawals can exceed 100 percent of renew- tices, cross-country comparability is limited. Compounding able supplies when extraction from nonrenewable aquifers or these problems, the data available cover different periods. Des- desalination plants is considerable or when there is significant ignating land as a protected area does not necessarily mean, reuse of water. Withdrawals for agriculture and industry are total moreover, that protection is in force. withdrawals for irrigation and livestock production and for di- Table 10. Energy use and emissions rect industrial use (including withdrawals for cooling thermo- electric plants), respectively. Withdrawals for domestic uses in- Commercial energy use refers to apparent consumption, which clude drinking water, municipal use or supply, and use for is equal to indigenous production plus imports and stock public services, commercial establishments, and homes. For changes, minus exports and fuels supplied to ships and aircraft most countries sectoral withdrawal data are estimated for engaged in international transportation. The International En- 1987–95. ergy Agency (IEA) and the United Nations Statistical Division Access to safe water refers to the percentage of people with (UNSD) compile energy data. IEA data for nonmembers of the reasonable access to an adequate amount of safe drinking water Organisation for Economic Co-operation and Development in their dwellings or within a convenient distance of their (OECD) are based on national energy data that have been ad- dwellings. Information on access to safe water, although widely justed to conform with annual questionnaires completed by used, is extremely subjective, and such terms as “adequate” and OECD member governments. UNSD data are compiled pri- “safe” may have very different meanings in different countries, marily from responses to questionnaires sent to national gov- despite official WHO definitions. Even in industrial countries, ernments, supplemented by official national statistical publica- treated water may not always be safe to drink. Although access tions and by data from intergovernmental organizations. When to safe water is equated with connection to a public supply sys- official data are not available, the UNSD bases its estimates tem, this does not take account of variations in the quality and on the professional and commercial literature. The variety of cost (broadly defined) of the service once connected. Thus cross- sources affects the cross-country comparability of data. country comparisons must be made cautiously. Changes over Commercial energy use refers to domestic primary energy use time within countries may result from changes in definitions or before transformation to other end-use energy sources (such as measurements. electricity and refined petroleum products). It includes energy Annual deforestation refers to the permanent conversion of from combustible renewables and waste. All forms of commer- forest area (land under natural or planted stands of trees) to cial energy—primary energy and primary electricity—are con- other uses, including shifting cultivation, permanent agricul- verted into oil equivalents. To convert nuclear electricity into oil ture, ranching, settlements, and infrastructure development. equivalents, a notional thermal efficiency of 33 percent is as- Deforested areas do not include areas logged but intended for sumed; for hydroelectric power, 100 percent efficiency is assumed. regeneration or areas degraded by fuelwood gathering, acid pre- GDP per unit of energy use is the U.S. dollar estimate of cipitation, or forest fires. Negative numbers indicate an increase real gross domestic product (at 1995 prices) per kilogram of oil in forest area. equivalent of commercial energy use. Estimates of forest area are from the FAO’s State of the Net energy imports is calculated as energy use less produc- World’s Forests 1997, which provides information on forest tion, both measured in oil equivalents. A minus sign indicates cover as of 1995 and a revised estimate of forest cover in 1990. that the country is a net exporter of energy. Forest cover data for developing countries are based on country Carbon dioxide emissions measures those emissions stem- assessments that were prepared at different times and that, for ming from the burning of fossil fuels and the manufacture of reporting purposes, had to be adapted to the standard reference cement. These include carbon dioxide produced during con- years of 1990 and 1995. This adjustment was made with a de- sumption of solid, liquid, and gas fuels and from gas flaring. forestation model designed to correlate forest cover change over The Carbon Dioxide Information Analysis Center (CDIAC), time with certain ancillary variables, including population sponsored by the U.S. Department of Energy, calculates annual change and density, initial forest cover, and ecological zone of anthropogenic emissions of carbon dioxide. These calculations the forest area under consideration. are derived from data on fossil fuel consumption, based on the Nationally protected areas refers to totally or partially pro- World Energy Data Set maintained by the UNSD, and from tected areas of at least 1,000 hectares that are designated as na- data on world cement manufacturing, based on the Cement tional parks, natural monuments, nature reserves, wildlife sanc- Manufacturing Data Set maintained by the U.S. Bureau of tuaries, protected landscapes and seascapes, or scientific reserves Mines. Each year the CDIAC recalculates the entire time series with limited public access. The indicator is calculated as a per- from 1950 to the present, incorporating its most recent find- centage of total area. For small countries whose protected areas ings and the latest corrections to its database. Estimates exclude may be smaller than 1,000 hectares, this limit will result in an fuels supplied to ships and aircraft engaged in international (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  transportation because of the difficulty of apportioning these industry, manufacturing, and services value added). Since 1968 fuels among the countries benefiting from that transport. the United Nations’ System of National Accounts (SNA) has called for estimates of GDP by industrial origin to be valued at Table 11. Growth of the economy either basic prices (excluding all indirect taxes on factors of pro- Gross domestic product is gross value added, at purchasers’ duction) or producer prices (including taxes on factors of pro- prices, by all resident and nonresident producers in the econ- duction, but excluding indirect taxes on final output). Some omy plus any taxes and minus any subsidies not included in the countries, however, report such data at purchasers’ prices—the value of the products. It is calculated without deducting for de- prices at which final sales are made—and this may affect esti- preciation of fabricated assets or for depletion or degradation of mates of the distribution of output. Total GDP as shown in this natural resources. Value added is the net output of a sector after table is measured at purchasers’ prices. GDP components are adding up all outputs and subtracting intermediate inputs. The measured at basic prices. industrial origin of value added is determined by the Interna- Among the difficulties faced by compilers of national ac- tional Standard Industrial Classification (ISIC), revision 2. counts is the extent of unreported economic activity in the in- The GDP implicit deflator reflects changes in prices for all formal or secondary economy. In developing countries a large final demand categories, such as government consumption, cap- share of agricultural output is either not exchanged (because it ital formation, and international trade, as well as the main com- is consumed within the household) or not exchanged for ponent, private final consumption. It is derived as the ratio of money. Financial transactions also may go unrecorded. Agricul- current- to constant-price GDP. The GDP deflator may also be tural production often must be estimated indirectly, using a calculated explicitly as a Laspeyres price index in which the combination of methods involving estimates of inputs, yields, weights are base-period quantities of output. and area under cultivation. Agriculture value added corresponds to ISIC divisions The output of industry ideally should be measured through 11–13 and includes forestry and fishing. Industry value added regular censuses and surveys of firms. But in most developing comprises the following sectors: mining (ISIC divisions 10–14), countries such surveys are infrequent and quickly go out of date, manufacturing (ISIC divisions 15–37), construction (ISIC di- so many results must be extrapolated. The choice of sampling vision 45), and electricity, gas, and water supply (ISIC divisions unit, which may be the enterprise (where responses may be 40 and 41). Services value added corresponds to ISIC divisions based on financial records) or the establishment (where produc- 50–96. tion units may be recorded separately), also affects the quality Exports of goods and services represents the value of all of the data. Moreover, much industrial production is organized goods and market services provided to the rest of the world. In- not in firms but in unincorporated or owner-operated ventures cluded is the value of merchandise, freight, insurance, travel, not captured by surveys aimed at the formal sector. Even in large and other nonfactor services. Factor and property income (for- industries, where regular surveys are more likely, evasion of ex- merly called factor services), such as investment income, inter- cise and other taxes lowers the estimates of value added. Such est, and labor income, is excluded, as are transfer payments. problems become more acute as countries move from state con- Gross domestic investment consists of outlays on additions trol of industry to private enterprise because new firms go into to the fixed assets of the economy plus net changes in the level business and growing numbers of established firms fail to re- of inventories. Additions to fixed assets include land improve- port. In accordance with the SNA, output should include all ments (fences, ditches, drains, and so on); plant, machinery, and such unreported activity as well as the value of illegal activities equipment purchases; and the construction of buildings, roads, and other unrecorded, informal, or small-scale operations. Data railways, and the like, including commercial and industrial on these activities need to be collected using techniques other buildings, offices, schools, hospitals, and private dwellings. In- than conventional surveys. ventories are stocks of goods held by firms to meet temporary In sectors dominated by large organizations and enterprises, or unexpected fluctuations in production or sales. data on output, employment, and wages are usually readily Growth rates are annual averages calculated using constant- available and reasonably reliable. But in the service sector the price data in local currency. Growth rates for regional and in- many self-employed workers and one-person businesses are come groups are calculated after converting local currencies to sometimes difficult to locate, and their owners have little incen- U.S. dollars at the average official exchange rate reported by the tive to respond to surveys, let alone report their full earnings. IMF for the year shown or, occasionally, using an alternative Compounding these problems are the many forms of economic conversion factor determined by the World Bank’s Develop- activity that go unrecorded, including the work that women and ment Data Group. Methods of computing growth rates and the children do for little or no pay. For further discussion of the alternative conversion factors are described under “Statistical problems encountered in using national accounts data see Srini- methods” below. For additional information on the calculation vasan (1994) and Heston (1994) in Data Sources. of GDP and its sectoral components, see the technical note to Table 13. Structure of demand table 12. Private consumption is the market value of all goods and Table 12. Structure of output services, including durable products (such as cars, washing ma- Gross domestic product represents the sum of value added by chines, and home computers), purchased or received as income all producers in the economy (see the technical note to table 11 in kind by households and nonprofit institutions. It excludes pur- for a more detailed definition and for definitions of agriculture, chases of dwellings but includes imputed rent for owner- (c) The International Bank for Reconstruction and Development / The World Bank      occupied dwellings. In practice, it may include any statistical dis- curring when ownership changes hands.) This issue has assumed crepancy in the use of resources relative to the supply of resources. greater significance with the increasing globalization of interna- Private consumption is often estimated as a residual, by sub- tional business. Neither customs nor balance of payments data tracting from GDP all other known expenditures. The result- capture the illegal transactions that occur in many countries. ing aggregate may incorporate fairly large discrepancies. When Goods carried by travelers across borders in legal but unreported private consumption is calculated separately, the household sur- shuttle trade may further distort trade statistics. veys on which a large component of the estimates is based tend Resource balance is the difference between exports of goods to be one-year studies with limited coverage. Thus the estimates and services and imports of goods and services. quickly become outdated and must be supplemented by price- Table 14. Central government finances and quantity-based statistical estimating procedures. Compli- cating the issue, in many developing countries the distinction Current tax revenue comprises compulsory, unrequited, nonre- between cash outlays for personal business and those for house- payable receipts collected by central governments for public hold use may be blurred. purposes. It includes interest collected on tax arrears and penal- General government consumption includes all current ties collected on nonpayment or late payment of taxes. It is spending for purchases of goods and services (including wages shown net of refunds and other corrective transactions. and salaries) by all levels of government, excluding most gov- Current nontax revenue includes requited, nonrepayable re- ernment enterprises. It also includes most expenditure on na- ceipts for public purposes, such as fines, administrative fees, or tional defense and security, some of which is now considered entrepreneurial income from government ownership of prop- part of investment. erty, and voluntary, unrequited, nonrepayable current govern- Gross domestic investment consists of outlays on additions ment receipts other than from governmental sources. This cate- to the fixed assets of the economy plus net changes in the level gory does not include grants, borrowing, repayment of previous of inventories. For the definitions of fixed assets and invento- lending, or sales of fixed capital assets or of stocks, land, or in- ries see the technical note to table 11. Under the revised (1993) tangible assets, nor does it include gifts from nongovernmental SNA guidelines, gross domestic investment also includes capital sources for capital purposes. Together, tax and nontax revenue outlays on defense establishments that may be used by the gen- make up the current revenue of the government. eral public, such as schools and hospitals, and on certain types Current expenditure includes requited payments other than of private housing for family use. All other defense expenditures for capital assets or for goods or services to be used in the pro- are treated as current spending. duction of capital assets. It also includes unrequited payments Investment data may be estimated from direct surveys of en- for purposes other than permitting the recipients to acquire cap- terprises and administrative records or based on the commodity ital assets, compensating the recipients for damage or destruc- flow method, using data from trade and construction activities. tion of capital assets, or increasing the financial capital of the The quality of public fixed investment data depends on the recipients. Current expenditure does not include government quality of government accounting systems, which tend to be lending or repayments to the government, or government ac- weak in developing countries; measures of private fixed invest- quisition of equity for public policy purposes. ment—particularly capital outlays by small, unincorporated en- Capital expenditure is spending to acquire fixed capital as- terprises—are usually very unreliable. sets, land, intangible assets, government stocks, and nonmili- Estimates of changes in inventories are rarely complete but tary, nonfinancial assets. Also included are capital grants. usually include the most important activities or commodities. Overall deficit/surplus is current and capital revenue and In some countries these estimates are derived as a composite official grants received, less total expenditure and lending minus residual along with aggregate private consumption. According repayment. to national accounts conventions, adjustments should be made Goods and services expenditure comprises all government for appreciation of the value of inventories due to price changes, payments in exchange for goods and services, including wages but this is not always done. In economies where inflation is and salaries. high, this element can be substantial. Social services expenditure comprises expenditure on health, Gross domestic saving is the difference between GDP and education, housing, welfare, social security, and community total consumption. amenities. It also covers compensation for loss of income to the Exports of goods and services represents the value of all sick and temporarily disabled; payments to the elderly, the per- goods and services (including transportation, travel, and other manently disabled, and the unemployed; family, maternity, and services such as communications, insurance, and financial ser- child allowances; and the cost of welfare services such as care of vices) provided to the rest of the world. Data on exports and im- the aged, the disabled, and children. Many expenditures rele- ports are compiled from customs returns and from balance of vant to environmental protection, such as pollution abatement, payments data obtained from central banks. Although data on water supply, sanitation, and refuse collection, are included in- exports and imports from the payments side provide reasonably distinguishably in this category. reliable records of cross-border transactions, they may not ad- Data on government revenues and expenditures are collected here strictly to the appropriate valuation and timing definitions by the IMF through questionnaires distributed to member gov- of balance of payments accounting or, more important, corre- ernments, and by the OECD. In general, the definition of gov- spond with the change-of-ownership criterion. (In conventional ernment excludes nonfinancial public enterprises and public balance of payments accounting, a transaction is recorded as oc- financial institutions (such as the central bank). Despite the (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  IMF’s efforts to systematize and standardize the collection of services, income, and current transfers. The capital and financial public finance data, statistics on public finance are often incom- account records capital transfers; the acquisition or disposal of plete, untimely, and noncomparable. Inadequate statistical cov- nonproduced, nonfinancial assets (such as patents); and transac- erage precludes the presentation of subnational data, making tions in financial assets and liabilities. Gross international reserves cross-country comparisons potentially misleading. are recorded in a third set of accounts, the international invest- Total central government expenditure as presented in the ment position, which records the stocks of assets and liabilities. IMF’s Government Finance Statistics Yearbook is a more limited The balance of payments is a double-entry accounting sys- measure of general government consumption than that shown tem that shows all flows of goods and services into and out of in the national accounts because it excludes consumption ex- an economy; all transfers that are the counterpart of real re- penditure by state and local governments. At the same time, the sources or financial claims provided to or by the rest of the IMF’s concept of central government expenditure is broader world without a quid pro quo, such as donations and grants; than the national accounts definition because it includes gov- and all changes in residents’ claims on, and liabilities to, non- ernment gross domestic investment and transfer payments. residents that arise from economic transactions. All transactions Central government finances can refer to one of two ac- are recorded twice: once as a credit and once as a debit. In prin- counting concepts: consolidated or budgetary. For most coun- ciple, the net balance should be zero, but in practice the ac- tries central government finance data have been consolidated counts often do not balance. In these cases a balancing item, into one account, but for others only budgetary central govern- called net errors and omissions, is included in the capital and fi- ment accounts are available. Countries reporting budgetary data nancial account. are noted in the Primary data documentation table in World De- Discrepancies may arise in the balance of payments because velopment Indicators 1999. Because budgetary accounts do not there is no single source for balance of payments data and no necessarily include all central government units, the picture they way to ensure that data from different sources are fully consis- provide of central government activities is usually incomplete. tent. Sources include customs data, monetary accounts of the A key issue is the failure to include the quasi-fiscal operations banking system, external debt records, information provided by of the central bank. Central bank losses arising from monetary enterprises, surveys to estimate service transactions, and foreign operations and subsidized financing can result in sizable quasi- exchange records. Differences in recording methods—for exam- fiscal deficits. Such deficits may also result from the operations ple, in the timing of transactions, in definitions of residence and of other financial intermediaries, such as public development fi- ownership, and in the exchange rate used to value transac- nance institutions. Also missing from the data are governments’ tions—contribute to net errors and omissions. In addition, contingent liabilities for unfunded pension and insurance plans. smuggling and other illegal or quasi-legal transactions may be unrecorded or misrecorded. Table 15. Balance of payments, current account, The concepts and definitions underlying the data in table 15 and international reserves are based on the fifth edition of the IMF’s Balance of Payments Goods and services exports and goods and services imports to- Manual. That edition redefined as capital transfers some trans- gether comprise all transactions between residents of a country actions previously included in the current account, such as debt and the rest of the world involving a change in ownership of forgiveness, migrants’ capital transfers, and foreign aid to ac- general merchandise, goods sent for processing and repairs, non- quire capital goods. Thus the current account balance now more monetary gold, and services. accurately reflects net current transfer receipts in addition to Net income refers to compensation earned by workers in an transactions in goods, services (previously nonfactor services), economy other than the one in which they are resident, for work and income (previously factor income). Many countries still performed and paid for by a resident of that economy, and in- maintain their data collection systems according to the concepts vestment income (receipts and payments on direct investment, and definitions in the fourth edition. Where necessary, the IMF portfolio investment, other investment, and receipts on reserve converts data reported in earlier systems to conform with the assets). Income derived from the use of intangible assets is fifth edition (see the primary data documentation table in World recorded under business services. Development Indicators 1999). Values are in U.S. dollars con- Net current transfers consists of transactions in which resi- verted at market exchange rates. dents of an economy provide or receive goods, services, income, Table 16. Private sector finance or financial items without a quid pro quo. All transfers not con- sidered to be capital transfers are current transfers. Private investment covers gross outlays by the private sector Current account balance is the sum of net exports of goods (including private nonprofit agencies) on additions to its fixed and services, income, and current transfers. domestic assets. When direct estimates of private gross domes- Gross international reserves comprises holdings of mone- tic fixed investment are not available, such investment is esti- tary gold, special drawing rights, reserves of IMF members held mated as the difference between total gross domestic investment by the IMF, and holdings of foreign exchange under the con- and consolidated public investment. No allowance is made for trol of monetary authorities. The gold component of these re- the depreciation of assets. Because private investment is often serves is valued at year-end London prices ($589.50 an ounce estimated as the difference between two estimated quantities— in 1980 and $287.80 an ounce in 1998). domestic fixed investment and consolidated public invest- The balance of payments is divided into two groups of ac- ment—private investment may be undervalued or overvalued counts. The current account records transactions in goods and and subject to errors over time. (c) The International Bank for Reconstruction and Development / The World Bank      Stock market capitalization (also called market value) is the Military expenditure for members of the North Atlantic sum of the market capitalizations of all firms listed on domestic Treaty Organization (NATO) is based on the NATO defini- stock exchanges, where each firm’s market capitalization is its tion, which covers military-related expenditures of the defense share price at the end of the year times the number of shares out- ministry (including recruiting, training, construction, and the standing. Market capitalization, presented as one measure used purchase of military supplies and equipment) and other min- to gauge a country’s level of stock market development, suffers istries. Civilian-related expenditures of the defense ministry are from conceptual and statistical weaknesses such as inaccurate excluded. Military assistance is included in the expenditure of reporting and different accounting standards. the donor country. Purchases of military equipment on credit Number of listed domestic companies is the number of do- are recorded at the time the debt is incurred, not at the time of mestically incorporated companies listed on stock exchanges at payment. Data for other countries generally cover expenditures the end of the year, excluding investment companies, mutual of the ministry of defense; excluded are expenditures on public funds, and other collective investment vehicles. order and safety, which are classified separately. Interest rate spread, also known as the intermediation mar- Definitions of military spending differ depending on gin, is the difference between the interest rate charged by banks whether they include civil defense, reserves and auxiliary forces, on short- and medium-term loans to the private sector and the police and paramilitary forces, dual-purpose forces such as mili- interest rate offered by banks to resident customers for demand, tary and civilian police, military grants-in-kind, pensions for time, or savings deposits. Interest rates should reflect the respon- military personnel, and social security contributions paid by one siveness of financial institutions to competition and price incen- part of government to another. Official government data may tives. However, the interest rate spread may not be a reliable omit some military spending, disguise financing through extra- measure of a banking system’s efficiency, to the extent that in- budgetary accounts or unrecorded use of foreign exchange re- formation about interest rates is inaccurate, that banks do not ceipts, or fail to include military assistance or secret imports of monitor all bank managers, or that the government sets deposit military equipment. Current spending is more likely to be re- and lending rates. ported than capital spending. In some cases a more accurate es- Domestic credit provided by the banking sector includes timate of military spending can be obtained by adding the value all credit to various sectors on a gross basis, with the exception of estimated arms imports and nominal military expenditures. of credit to the central government, which is net. The banking This method may understate or overstate spending in a particu- sector includes monetary authorities, deposit money banks, and lar year, however, because payments for arms may not coincide other banking institutions for which data are available (includ- with deliveries. ing institutions that do not accept transferable deposits but do The data in table 17 are from the U.S. Arms Control and incur such liabilities as time and savings deposits). Examples of Disarmament Agency (ACDA). The IMF’s Government Finance other banking institutions include savings and mortgage loan Statistics Yearbook is a primary source for data on military spend- institutions and building and loan associations. ing. It uses a consistent definition of defense spending based on In general, the indicators reported here do not capture the the United Nations’ classification of the functions of govern- activities of the informal sector, which remains an important ment and the NATO definition. The IMF checks data on de- source of finance in developing economies. fense spending for broad consistency with other macroeconomic data reported to it, but it is not always able to verify their accu- Table 17. Role of government in the economy racy and completeness. Moreover, country coverage is affected Subsidies and other current transfers includes all unrequited, by delays or failure to report data. Thus most researchers sup- nonrepayable transfers on current account to private and public plement the IMF’s data with independent assessments of mili- enterprises and the cost to the public of covering the cash oper- tary outlays by organizations such as ACDA, the Stockholm ating deficits on sales to the public by departmental enterprises. International Peace Research Institute, and the International In- Value added by state-owned enterprises is estimated as sales stitute for Strategic Studies. However, these agencies rely heav- revenue minus the cost of intermediate inputs, or as the sum of ily on reporting by governments, on confidential intelligence these enterprises’ operating surplus (balance) and their wage estimates of varying quality, on sources that they do not or can- payments. State-owned enterprises are government-owned or not reveal, and on one another’s publications. -controlled economic entities that generate most of their rev- Composite ICRG risk rating is an overall index taken from enue by selling goods and services. This definition encompasses the International Country Risk Guide and based on 22 compo- commercial enterprises directly operated by a government de- nents of risk. The PRS Group’s International Country Risk partment and those in which the government holds a majority Guide collects information on each component, groups these of shares directly or indirectly through other state enterprises. It components into three major categories (political, financial, and also includes enterprises in which the state holds a minority of economic), and calculates a single risk assessment index ranging shares, if the distribution of the remaining shares leaves the gov- from 0 to 100. Ratings below 50 indicate very high risk and ernment with effective control. It excludes public sector activ- those above 80 very low risk. Ratings are updated monthly. ity—such as education, health services, and road construction Institutional Investor credit rating ranks, from 0 to 100, the and maintenance—that is financed in other ways, usually from probability of a country’s default. A high number indicates a the government’s general revenue. Because financial enterprises low probability of default. Institutional Investor country credit are of a different nature, they have generally been excluded from ratings are based on information provided by leading interna- the data. tional banks. Responses are weighted using a formula that gives (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  more importance to responses from banks with greater world- kilometers traveled per million dollars of GDP measured in PPP wide exposure and more sophisticated country analysis systems. terms. Air passengers carried includes passengers on both do- Risk ratings may be highly subjective, reflecting external per- mestic and international passenger routes. ceptions that do not always capture a country’s actual situation. Data for most transportation industries are not internation- But these subjective perceptions are the reality that policymak- ally comparable, because unlike demographic statistics, national ers face in the climate they create for foreign private inflows. income accounts, and international trade data, the collection of Countries not rated favorably by credit-risk rating agencies typ- infrastructure data has not been standardized internationally. ically do not attract registered flows of private capital. The risk Data on roads are collected by the International Road Federa- ratings presented here are not endorsed by the World Bank but tion (IRF) and data on air transportation by the International are included for their analytical usefulness. Civil Aviation Organization. National road associations are the Highest marginal tax rate is the highest rate shown on primary source of IRF data; in countries where such an associa- the schedule of tax rates applied to the taxable income of indi- tion is absent or does not respond, other agencies are contacted, viduals and corporations. The table also presents the income such as road directorates, ministries of transportation or public threshold above which the highest marginal tax rate applies for works, or central statistical offices. As a result, the compiled data individuals. are of uneven quality. Tax collection systems are often complex, containing many Table 19. Communications, information, and exceptions, exemptions, penalties, and other inducements that science and technology affect the incidence of taxation and thus influence the decisions of workers, managers, entrepreneurs, investors, and consumers. Daily newspapers is the number of copies distributed of news- A potentially important influence on both domestic and inter- papers published at least four times a week, per thousand peo- national investors is the tax system’s progressivity, as reflected ple. Radios is the estimated number of radio receivers in use for in the highest marginal tax rate on individual and corporate in- broadcasts to the general public, per thousand people. Data on come. Marginal tax rates on individuals generally refer to em- these two indicators are obtained from statistical surveys by the ployment income. For some countries the highest marginal tax United Nations Educational, Scientific, and Cultural Organi- rate is also the basic or flat rate, and other surtaxes, deductions, zation (UNESCO). In some countries, definitions, classifica- and the like may apply. tions, and methods of enumeration do not entirely conform to UNESCO standards. For example, some countries report news- Table 18. Power and transportation paper circulation as the number of copies printed rather than Electric power consumption per capita measures the produc- the number distributed. In addition, many countries impose tion of power plants and combined heat and power plants less radio license fees to help pay for public broadcasting, discour- distribution losses and their own use. Electric power transmis- aging radio owners from declaring ownership. Because of these sion and distribution losses measures losses occurring between and other data collection problems, estimates of the number of sources of supply and points of distribution, and in distribution newspapers and radios vary widely in reliability and should be to consumers, including pilferage. interpreted with caution. The IEA collects data on electric power production and con- Television sets is the estimated number of sets in use, per sumption from national energy agencies and adjusts those data thousand people. Data on television sets are supplied to the In- to meet international definitions, for example, to account for ternational Telecommunication Union (ITU) through annual establishments that, in addition to their main activities, gener- questionnaires sent to national broadcasting authorities and in- ate electricity wholly or partly for their own use. In some coun- dustry associations. Some countries require that television sets tries self-production by households and small entrepreneurs is be registered. To the extent that households do not register substantial because of their remoteness or because public power some or all of their sets, the number of registered sets may un- sources are unreliable, and these adjustments may not ade- derstate the true number of sets in use. quately reflect actual output. Telephone main lines counts all telephone lines that con- Although power plants’ own consumption and transmission nect a customer’s equipment to the public switched telephone losses are netted out, electric power consumption includes con- network, per thousand people. Mobile telephones refers to users sumption by auxiliary stations, losses in transformers that are of portable telephones subscribing to an automatic public mo- considered integral parts of those stations, and electricity pro- bile telephone service using cellular technology that provides ac- duced by pumping installations. Where data are available, con- cess to the public switched telephone network, per thousand sumption covers electricity generated by all primary sources of people. The ITU compiles data on telephone main lines and energy: coal, oil, gas, nuclear, hydroelectric, geothermal, wind, mobile phones through annual questionnaires sent to telecom- tide and wave, and combustible renewables. Neither production munications authorities and operating companies. The data are nor consumption data capture the reliability of supplies, includ- supplemented by annual reports and statistical yearbooks of ing the frequency of outages, breakdowns, and load factors. telecommunications ministries, regulators, operators, and in- Paved roads are roads that have been sealed with asphalt or dustry associations. similar road-building materials. Goods transported by road is Personal computers is the estimated number of self-contained the volume of goods transported by road vehicles, measured in computers designed to be used by a single person, per thousand millions of metric tons times kilometers traveled. Goods trans- people. Estimates by the ITU of the number of personal comput- ported by rail measures the tonnage of goods transported times ers are derived from an annual questionnaire, supplemented by (c) The International Bank for Reconstruction and Development / The World Bank      other sources. In many countries mainframe computers are used their R&D intensity, and the top 10 SIC groups (as classified at extensively, and thousands of users may be connected to a single the three-digit level) were designated high-technology industries. mainframe computer; in such cases the number of personal com- To translate Davis’s industry classification into a definition puters understates the total use of computers. of high-technology trade, Braga and Yeats (1992) used the con- Internet hosts are computers connected directly to the world- cordance between the SIC grouping and the Standard Interna- wide network; many computer users can access the Internet tional Trade Classification (SITC), revision 1, classification pro- through a single host. Hosts are assigned to countries on the basis posed by Hatter (1985). In preparing the data on high-technology of the host’s country code, though this does not necessarily indi- trade, Braga and Yeats considered only SITC groups (classified cate that the host is physically located in that country. All hosts at the four-digit level) that had a high-technology weight above lacking a country code identification are assigned to the United 50 percent. Examples of high-technology exports include air- States. Because Network Wizards (the source of these data at craft, office machinery, pharmaceuticals, and scientific instru- http://www.nw.com) changed the methods used in its Internet ments. This methodology rests on the somewhat unrealistic as- domain survey beginning in July 1998, the data shown here are sumption that using U.S. input-output relations and trade not directly comparable with those published last year. The new patterns for high-technology production does not introduce a survey is believed to be more reliable and to avoid the problem bias in the classification. of undercounting that occurs when organizations restrict down- Number of patent applications filed is the number of doc- load access to their domain data. Nevertheless, some measure- uments, issued by a government office, that describe an inven- ment problems remain, and so the number of Internet hosts tion and create a legal situation in which the patented invention shown for each country should be considered an approximation. can normally only be exploited (made, used, sold, imported) by, Scientists and engineers in R&D is the number of people or with the authorization of, the patentee. The protection of in- trained to work in any field of science who are engaged in pro- ventions is limited in time (generally 20 years from the filing fessional research and development activity (including adminis- date of the application for the grant of a patent). Information trators), per million people. Most such jobs require completion on patent applications filed is shown separately for residents and of tertiary education. nonresidents of the country. Data on patents are from the UNESCO collects data on scientific and technical workers World Intellectual Property Organization, which estimates that and R&D expenditure from its member states, mainly from of- at the end of 1996 about 3.8 million patents were in force in ficial replies to UNESCO questionnaires and special surveys, as the world. well as from official reports and publications, supplemented Table 20. Global trade by information from other national and international sources. UNESCO reports either the stock of scientists and engineers or Merchandise exports shows the f.o.b. (free on board) value, in the number of economically active persons qualified to be sci- U.S. dollars, of goods provided to the rest of the world. Mer- entists and engineers. Stock data generally come from censuses chandise imports shows the c.i.f. (cost plus insurance and and are less timely than measures of the economically active freight) value, in U.S. dollars, of goods purchased from the rest population. UNESCO supplements these data with estimates of the world. Manufactured exports and imports refers to com- of the number of qualified scientists and engineers by counting modities in SITC sections 5 (chemicals), 6 (basic manufactures), the number of people who have completed education at ISCED 7 (machinery), and 8 (miscellaneous manufactured goods), ex- (International Standard Classification of Education) levels 6 and cluding division 68 (nonferrous metals) and group 891 (arms 7. The data on scientists and engineers, normally calculated in and ammunition). Commercial services comprises all trade in terms of full-time equivalent staff, cannot take into account the services, including transportation, communication, and busi- considerable variations in the quality of training and education. ness services, excluding government services, which comprise High-technology exports consists of goods produced by in- services associated with government sectors (such as expendi- dustries (based on U.S. industry classifications) that rank among tures on embassies and consulates) and with regional and inter- a country’s top 10 in terms of R&D expenditure. Manufactured national organizations. exports are those commodities in the Standard International Data on merchandise exports and imports are derived from Trade Classification (SITC), revision 1, sections 5–9 (chemi- customs records and may not fully conform to the concepts and cals and related products, basic manufactures, manufactured ar- definitions contained in the fifth edition of the IMF’s Balance of ticles, machinery and transport equipment, and other manufac- Payments Manual. The value of exports is recorded as the cost tured articles and goods not elsewhere classified), excluding of the goods delivered to the frontier of the exporting country division 68 (nonferrous metals). for shipment—the f.o.b. value. Many countries collect and re- Industry rankings are based on a methodology developed by port trade data in U.S. dollars. When countries report in local Davis (1982; see Data Sources). Using input-output techniques, currency, the value is converted at the average official exchange Davis estimated the technology intensity of U.S. industries in rate for the period. The value of imports is generally recorded as terms of the R&D expenditure required to produce a certain the cost of the goods when purchased by the importer plus the manufactured good. This methodology takes into account direct cost of transport and insurance to the frontier of the importing R&D expenditure by final producers as well as indirect R&D ex- country—the c.i.f. value. Data on imports of goods are derived penditure by suppliers of intermediate goods used in producing from the same sources as data on exports. In principle, world ex- the final good. Industries, classified on the basis of the U.S. Stan- ports and imports should be identical. Similarly, exports from dard Industrial Classification (SIC), were ranked according to an economy should equal the sum of imports by the rest of the (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  world from that economy. But differences in timing and defini- use of IMF credit, and short-term debt. Short-term debt in- tion result in discrepancies in reported values at all levels. cludes all debt having an original maturity of one year or less The data in this table were compiled by the World Trade and interest in arrears on long-term debt. Present value of ex- Organization (WTO). Data on merchandise trade come from ternal debt is the sum of short-term external debt plus the dis- the IMF International Financial Statistics Yearbook, supple- counted sum of total debt service payments due on public, pub- mented by data from the COMTRADE database maintained licly guaranteed, and private nonguaranteed long-term external by the United Nations Statistical Division and from national debt over the life of existing loans. publications for countries that do not report to the IMF. Data Data on the external debt of low- and middle-income on trade in manufactures come from the COMTRADE data- economies are gathered by the World Bank through its Debtor base. Where data were not available from the WTO, World Reporting System. World Bank staff calculate the indebtedness Bank staff estimated the shares of manufactures using the most of developing countries using loan-by-loan reports submitted by recent information available from the COMTRADE database. these countries on long-term public and publicly guaranteed Wherever available, WTO reports merchandise trade data on borrowing, along with information on short-term debt collected the basis of the general system of trade, which includes goods by the countries or from creditors through the reporting systems imported for reexport. Two economies, Hong Kong (China) of the Bank for International Settlements and the OECD. These and Singapore, with substantial levels of reexports are noted in data are supplemented by information on loans and credits from the table. Goods transported through a country en route to an- major multilateral banks and loan statements from official lend- other are not included. Data on trade in commercial services are ing agencies in major creditor countries, and by estimates from drawn from the IMF Balance of Payments database, supple- World Bank country economists and IMF desk officers. In ad- mented by national publications from countries that do not re- dition, some countries provide data on private nonguaranteed port to the IMF. debt. In 1996, 34 countries reported their private nonguaran- teed debt to the World Bank; estimates were made for 28 addi- Table 21. Aid and financial flows tional countries known to have significant private debt. Net private capital flows consists of private debt and nondebt The present value of external debt provides a measure of fu- flows. Private debt flows include commercial bank lending, ture debt service obligations that can be compared with such in- bonds, and other private credits; nondebt private flows are for- dicators as GNP. It is calculated by discounting debt service (in- eign direct investment and portfolio equity investment. Foreign terest plus amortization) due on long-term external debt over direct investment is net inflows of investment to acquire a last- the life of existing loans. Short-term debt is included at its face ing management interest (10 percent or more of voting stock) value. Data on debt are in U.S. dollars converted at official ex- in an enterprise operating in an economy other than that of the change rates. The discount rate applied to long-term debt is de- investor. It is the sum of equity capital flows, reinvestment of termined by the currency of repayment of the loan and is based earnings, other long-term capital flows, and short-term capital on the OECD’s commercial interest reference rates. Loans from flows as shown in the balance of payments. the International Bank for Reconstruction and Development The data on foreign direct investment are based on balance and credits from the International Development Association are of payments data reported by the IMF, supplemented by data discounted using a reference rate for special drawing rights, as on net foreign direct investment reported by the OECD and of- are obligations to the IMF. When the discount rate is greater ficial national sources. The internationally accepted definition than the interest rate of the loan, the present value is less than of foreign direct investment is that provided in the fifth edition the nominal sum of future debt service obligations. of the IMF’s Balance of Payments Manual. The OECD has also Official development assistance (ODA) consists of disburse- published a definition, in consultation with the IMF, Eurostat ments of loans (net of repayments of principal) and grants made (the Statistical Office of the European Communities), and the on concessional terms by official agencies of the members of the United Nations. Because of the multiplicity of sources and dif- Development Assistance Committee (DAC) and certain Arab ferences in definitions and reporting methods, more than one countries to promote economic development and welfare in re- estimate of foreign direct investment may exist for a country, cipient economies listed by DAC as developing. Loans with a and data may not be comparable across countries. grant element of more than 25 percent are included in ODA, Foreign direct investment data do not give a complete pic- as are technical cooperation and assistance. Also included are ture of international investment in an economy. Balance of pay- aid flows (net of repayments) from official donors to the transi- ments data on foreign direct investment do not include capital tion economies of Eastern Europe and the former Soviet Union raised in the host economies, which has become an important and to certain higher-income developing countries and territo- source of financing for investment projects in some developing ries as determined by DAC. These flows are sometimes referred countries. There is also increasing awareness that foreign direct to as “official aid” and are provided under terms and conditions investment data are limited because they capture only cross- similar to those for ODA. Data for aid as a share of GNP are border investment flows involving equity participation and omit calculated using values in U.S. dollars converted at official ex- nonequity cross-border transactions such as intrafirm flows of change rates. goods and services. For a detailed discussion of the data issues The data cover bilateral loans and grants from DAC coun- see volume 1, chapter 3, of World Debt Tables 1993–94. tries, multilateral organizations, and certain Arab countries. They Total external debt is debt owed to nonresidents repayable do not reflect aid given by recipient countries to other develop- in foreign currency, goods, or services. It is the sum of public, ing countries. As a result, some countries that are net donors publicly guaranteed, and private nonguaranteed long-term debt, (such as Saudi Arabia) are shown in the table as aid recipients. (c) The International Bank for Reconstruction and Development / The World Bank      The data do not distinguish among different types of aid where pn and p1 are the last and first observations in the period, (program, project, or food aid; emergency assistance; peacekeep- n is the number of years in the period, and ln is the natural log- ing assistance; or technical cooperation), each of which may arithm operator. This growth rate is based on a model of con- have a very different effect on the economy. Technical coopera- tinuous, exponential growth between two points in time. It does tion expenditures do not always directly benefit the recipient not take into account the intermediate values of the series. Note economy to the extent that they defray costs incurred outside also that the exponential growth rate does not correspond to the the country for salaries and benefits of technical experts and for annual rate of change measured at a one-year interval which is overhead of firms supplying technical services. given by (pn – pn-1)/pn-1. Because the aid data in table 21 are based on information from donors, they are not consistent with information recorded The Gini index by recipients in the balance of payments, which often excludes The Gini index measures the extent to which the distribution all or some technical assistance—particularly payments to expa- of income (or, in some cases, consumption expenditure) among triates made directly by the donor. Similarly, grant commodity individuals or households within an economy deviates from a aid may not always be recorded in trade data or in the balance perfectly equal distribution. A Lorenz curve plots the cumula- of payments. Although estimates of ODA in balance of pay- tive percentages of total income received against the cumulative ments statistics are meant to exclude purely military aid, the dis- percentage of recipients, starting with the poorest individual or tinction is sometimes blurred. The definition used by the coun- household. The Gini index measures the area between the try of origin usually prevails. Lorenz curve and a hypothetical line of absolute equality, ex- pressed as a percentage of the maximum area under the line. Statistical methods Thus a Gini index of zero represents perfect equality, where an This section describes the calculation of the least-squares growth index of 100 percent implies maximum inequality. rate, the exponential (end-point) growth rate, the Gini index, The World Bank employs a numerical analysis program, and the World Bank’s Atlas methodology for calculating the POVCAL, to estimate values of the Gini index; see Chen, Datt, conversion factor used to estimate GNP and GNP per capita in and Ravallion (1993; see Data Sources). U.S. dollars. World Bank Atlas method Least-squares growth rate In calculating GNP and GNP per capita in U.S. dollars for cer- Least-squares growth rates are used wherever there is a suffi- tain operational purposes, the World Bank uses a synthetic ex- ciently long time series to permit a reliable calculation. No change rate commonly called the Atlas conversion factor. The growth rate is calculated if more than half the observations in a purpose of the Atlas conversion factor is to reduce the impact of period are missing. exchange rate fluctuations in the cross-country comparison of The least-squares growth rate, r, is estimated by fitting a lin- national incomes. ear regression trendline to the logarithmic annual values of the The Atlas conversion factor for any year is the average of a variable in the relevant period. The regression equation takes country’s effective exchange rate with the G-5 countries (or al- the form ternative conversion factor) for that year and those for the two preceding years, after adjusting for differences in rates of infla- ln X t = a + bt, tion between the country and the G-5 countries. A country’s effective exchange rate is an average of its exchange rates with a which is equivalent to the logarithmic transformation of the selection of other countries, usually weighted by the country’s compound growth equation, trade with those countries. The G-5 (Group of Five) countries are France, Germany, Japan, the United Kingdom, and the Xt = Xo (1 + r )t . United States. A country’s inflation rate is measured by its GNP deflator. The inflation rate for the G-5 countries is mea- In this equation, X is the variable, t is time, and a = log Xo and sured by changes in the SDR deflator. (Special drawing rights, b = ln (1 + r ) are the parameters to be estimated. If b* is the or SDRs, are the IMF’s unit of account.) The SDR deflator is least-squares estimate of b, the average annual growth rate, r, is calculated as a weighted average of the G-5 countries’ GDP de- obtained as [exp(b* )–1] and is multiplied by 100 to express it flators in SDR terms. The weights are determined by the as a percentage. amount of each currency included in one SDR unit. Weights The calculated growth rate is an average rate that is repre- vary over time because the currency composition of the SDR sentative of the available observations over the entire period. It and the relative exchange rates for each currency both change. does not necessarily match the actual growth rate between any The SDR deflator is calculated in SDR terms first and then two periods. converted to U.S. dollars using the SDR-to-dollar Atlas con- version factor. Exponential growth rate This three–year averaging smooths annual fluctuations in The growth rate between two points in time for certain demo- prices and exchange rates for each country. The Atlas conver- graphic data, notably labor force and population, is calculated sion factor is then applied to a country’s GNP. The resulting from the equation GNP in U.S. dollars is divided by the country’s midyear popu- lation for the latest of the three years to derive its GNP per r = ln (pn /p1)/n, capita. When official exchange rates are deemed to be unreli- (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  able or unrepresentative during a period, an alternative estimate year t, pt is the GNP deflator for year t, ptS$ is the SDR deflator of the exchange rate is used in the Atlas formula (see below). in U.S. dollar terms for year t, Yt$ is the Atlas GNP in U.S. dol- The following formulas describe the computation of the lars in year t, Yt is current GNP (local currency) for year t, and Atlas conversion factor for year t: Nt is the midyear population for year t. 1  p ptS $   p ptS $   Alternative conversion factors et* = et − 2  t 3  / S $   pt − 2 pt − 2  + et −1  t / S $   pt −1 pt −1  + et    The World Bank systematically assesses the appropriateness of official exchange rates as conversion factors. An alternative con- version factor is used when the official exchange rate is judged and for calculating GNP per capita in U.S. dollars for year t: to diverge by an exceptionally large margin from the rate effec- tively applied to domestic transactions of foreign currencies and Yt$ = (Yt / Nt ) / et* traded products. This is the case for only a small number of countries (see the primary data documentation table in World where et* is the Atlas conversion factor (units of national cur- Development Indicators 1999 ). Alternative conversion factors are rency to the U.S. dollar) for year t, et is the average annual ex- used in the Atlas method and elsewhere in the Selected World change rate (units of national currency to the U.S. dollar) for Development Indicators as single-year conversion factors. (c) The International Bank for Reconstruction and Development / The World Bank Data Sources ACDA (Arms Control and Disarmament Agency). 1997. World Military National Bureau of Economic Research. 1997. Penn World Tables Mark 5.6. Expenditures and Arms Transfers 1996. Washington, D.C. http://nber.harvard.edu/pwt56.html. Ahmad, Sultan. 1992. “Regression Estimates of Per Capita GDP Based on OECD (Organisation for Economic Co-operation and Development). Purchasing Power Parities.” Policy Research Working Paper 956. World 1989. Geographical Distribution of Financial Flows to Developing Coun- Bank, International Economics Department, Washington, D.C. tries. Paris. ———. 1994. “Improving Inter-Spatial and Inter-Temporal Comparabil- ———. 1997a. National Accounts 1960–1995. Vol. 1, Main Aggregates. ity of National Accounts.” Journal of Development Economics 4:53–75. Paris. Ball, Nicole. 1984. “Measuring Third World Security Expenditure: A Re- ———. 1997b. National Accounts 1960–1995. Vol. 2, Detailed Tables. search Note.” World Development 12(2):157–64. Paris. Bos, Eduard, My T. Vu, Ernest Massiah, and Rodolfo Bulatao. 1994. World ———. 1998. Development Co-operation: 1997 Report. Paris. Population Projections 1994–95. Baltimore, Md.: Johns Hopkins Uni- PRS Group. 1999. International Country Risk Guide. February. East Syra- versity Press. cuse, N.Y. Braga, C.A. Primo, and Alexander Yeats. 1992. “How Minilateral Trading PricewaterhouseCoopers. 1998a. Corporate Taxes: A Worldwide Summary. Arrangements May Affect the Post-Uruguay Round World.” World New York. Bank, International Economics Department, Washington, D.C. ———. 1998b. Individual Taxes: A Worldwide Summary. New York. Chen, Shaohua, Gaurav Datt, and Martin Ravallion. 1993. “Is Poverty In- Ravallion, Martin. 1996. “What Can New Survey Data Tell Us about the creasing in the Developing World?” Policy Research Working Paper. Recent Changes in Living Standards in Developing and Transitional World Bank, Washington, D.C. Economies?” World Bank, Policy Research Department, Washington, Council of Europe. Various years. Recent Demographic Developments in D.C. Europe and North America. Strasbourg: Council of Europe Press. Ravallion, Martin, and Shaohua Chen. 1997. “Can High-Ineqaulity Devel- Davis, Lester. 1982. Technology Intensity of U.S. Output and Trade. Wash- oping Countries Escape Absolute Poverty?” Economic Letters 56: 51–57. ington, D.C.: U.S. Department of Commerce. Srinivasan, T.N. 1994. “Database for Development Analysis: An Over- Eurostat (Statistical Office of the European Communities). Various years. view.” Journal of Development Economics 44(1):3–28. Demographic Statistics. Luxembourg. UNCTAD (United Nations Conference on Trade and Development). Var- FAO (Food and Agriculture Organization). 1997. State of the World’s Forests ious years. Handbook of International Trade and Development Statistics. 1997. Rome. Geneva. ———. Various years. Production Yearbook. FAO Statistics Series. Rome. UNESCO (United Nations Educational, Scientific, and Cultural Organiza- Happe, Nancy, and John Wakeman-Linn. 1994. “Military Expenditures tion). Various years. Statistical Yearbook. Paris. and Arms Trade: Alternative Data Sources.” IMF Working Paper UNICEF (United Nations Children’s Fund). 1999. The State of the World’s Children 1999. Oxford, U.K.: Oxford University Press. 94/69. International Monetary Fund, Policy Development and Review UNIDO (United Nations Industrial Development Organization).1996. In- Department, Washington, D.C. ternational Yearbook of Industrial Statistics 1996. Vienna. Hatter, Victoria L. 1985. U.S. High-Technology Trade and Competitiveness. United Nations. 1968. A System of National Accounts: Studies and Methods. Washington, D.C.: U.S. Department of Commerce. Series F, no. 2, rev. 3. New York. Heston, Alan. 1994. “A Brief Review of Some Problems in Using National ———. 1985. National Accounts Statistics: Compendium of Income Distri- Accounts Data in Level of Output Comparisons and Growth Studies.” bution Statistics. New York. Journal of Development Economics 44:29–52. ———. 1997. World Urbanization Prospects: The 1996 Revision. New York. ICAO (International Civil Aviation Organization). 1998. Civil Aviation ———. Various years. Energy Statistics Yearbook. New York. Statistics of the World: 1997. ICAO Statistical Yearbook. 22nd ed. ———. Various issues. Monthly Bulletin of Statistics. New York. Montreal. ———. Various years. National Income Accounts. Statistics Division. New IEA (International Energy Agency). 1998a. Energy Statistics and Balances of York. Non-OECD Countries 1995–96. Paris. ———. Various years. Statistical Yearbook. New York. ———. 1998b. Energy Statistics of OECD Countries 1995–96. Paris. ———. Various years. Update on the Nutrition Situation. Administrative IFC (International Finance Corporation). 1998. Emerging Stock Markets Committee on Coordination, Subcommittee on Nutrition. Geneva. Factbook 1998. Washington, D.C. ———. Various years. Population and Vital Statistics Report. New York. ILO (International Labour Organization). Various years. Yearbook of Labour U.S. Bureau of the Census. 1996. World Population Profile 1996. Washing- Statistics. Geneva: International Labour Office. ton, D.C.: U.S. Government Printing Office. ———. 1995a. Labour Force Estimates and Projections, 1950–2010. Geneva. WHO (World Health Organization). Various years. World Health Statistics. ———. 1995b. Estimates of the Economically Active Population by Sex and Geneva. Age Group and by Main Sectors of Economic Activity. Geneva. ———. Various years. World Health Statistics Report. Geneva. IMF (International Monetary Fund). 1986. A Manual on Government WHO and UNICEF. 1996. Revised 1990 Estimates on Maternal Mortality: Finance Statistics. Washington, D.C. A New Approach. Geneva. ———. 1993. Balance of Payments Manual. 5th ed. Washington, D.C. World Bank. 1993a. Purchasing Power of Currencies: Comparing National ———. Various years. Direction of Trade Statistics Yearbook. Washington, Incomes Using ICP Data. Washington, D.C. D.C. ———. 1993b. World Debt Tables 1993–94. Washington, D.C. ———. Various years. Government Finance Statistics Yearbook. Washing- ———. 1998. Global Development Finance 1998. Washington, D.C. ton, D.C. ———. 1999. World Development Indicators. Washington, D.C. ———. Various years. International Financial Statistics Yearbook. Washing- World Resources Institute, UNEP (United Nations Environment Pro- ton, D.C. gramme), and UNDP (United Nations Development Programme). Institutional Investor, 1999. New York. (March). 1994. World Resources 1994–95: A Guide to the Global Environment. IRF (International Road Federation). 1998. World Road Statistics 1998. New York: Oxford University Press. Geneva. World Resources Institute, in collaboration with UNEP (United Nations ITU (International Telecommunication Union). 1998. World Telecommu- Environment Programme), and UNDP (United Nations Development nication Development Report. Geneva. Programme). 1998. World Resources 1998–99: A Guide to the Global En- Luxembourg Income Study. LIS database. http://lissy.seps.lu/index. htm. vironment. New York: Oxford University Press.  (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  Classification of Economies by Income and Region, 1999 Sub-Saharan Africa Europe and Central Asia East and Asia Eastern Middle East and North Africa Income Southern West East Asia South Europe and Rest of Middle North group Subgroup Africa Africa and Pacific Asia Central Asia Europe East Africa Americas Angola Benin Cambodia Afghanistan Armenia Yemen, Rep. Haiti Burundi Burkina Faso China Bangladesh Azerbaijan Honduras Comoros Cameroon Indonesia Bhutan Kyrgyz Nicaragua Congo, Dem. Central Korea, Dem. India Republic Rep. African Rep. Nepal Moldova Eritrea Republic Lao PDR Pakistan Tajikistan Ethiopia Chad Mongolia Turkmenistan Kenya Congo, Rep. Myanmar Lesotho Côte d'Ivoire Solomon Madagascar Gambia, The Islands Malawi Ghana Vietnam Low- Mozambique Guinea income Rwanda Guinea- Somalia Bissau Sudan Liberia Tanzania Mali Uganda Mauritania Zambia Niger Zimbabwe Nigeria São Tomé and Principe Senegal Sierra Leone Togo Djibouti Cape Verde Fiji Maldives Albania Iran, Islamic Algeria Belize Namibia Equatorial Kiribati Sri Lanka Belarus Rep. Egypt, Arab Bolivia South Africa Guinea Marshall Bosnia and Iraq Rep. Colombia Swaziland Islands Herzegovina Jordan Morocco Costa Rica Micronesia, Bulgaria Syrian Arab Tunisia Cuba Fed. Sts. Georgia Republic Dominica Papua New Kazakhstan West Bank Dominican Guinea Latvia and Gaza Republic Philippines Lithuania Ecuador Samoa Macedonia, El Salvador Lower Thailand FYRa Guatemala Tonga Romania Guyana Vanuatu Russian Jamaica Federation Paraguay Ukraine Peru Uzbekistan St. Vincent Yugoslavia, and the Fed. Rep.b Grenadines Middle- Suriname income Botswana Gabon American Croatia Isle of Man Bahrain Libya Antigua and Mauritius Samoa Czech Turkey Lebanon Barbuda Mayotte Korea, Rep Republic Oman Argentina Seychelles Malaysia Estonia Saudi Barbados Palau Hungary Arabia Brazil Poland Chile Slovak Grenada Republic Guadeloupe Mexico Upper Panama Puerto Rico St. Kitts and Nevis St. Lucia Trinidad and Tobago Uruguay Venezuela Subtotal: 157 26 23 23 8 26 2 10 5 34      Classification of Economies by Income and Region, 1999 (continued) Sub-Saharan Africa Europe and Central Asia East and Asia Eastern Middle East and North Africa Income Southern East Asia and Europe and Rest of Middle North group Subgroup Africa West Africa Pacific South Asia Central Asia Europe East Africa Americas Australia Austria Canada Japan Belgium United States New Zealand Denmark Finland France Germany Greece Iceland Ireland OECD Italy Luxembourg Netherlands Norway Portugal Spain Sweden Switzerland United High- Kingdom income Réunion Brunei Slovenia Andorra Israel Malta Aruba French Channel Kuwait Bahamas, Polynesia Islands Qatar The Non-OECD Guam Cyprus United Arab Bermuda Hong Kong, Faeroe Emirates Cayman Chinac Islands Islands Macao Greenland French New Liechtenstein Guiana Caledonia Monaco Martinique N. Mariana Netherlands Islands Antilles Singapore Virgin Taiwan, Islands China (U.S.) Total: 211 27 23 35 8 27 27 14 6 44 a. Former Yugoslav Republic of Macedonia. b. Federal Republic of Yugoslavia (Serbia/Montenegro). c. On July 1, 1997, China resumed its sovereignty over Hong Kong. Source: World Bank data. For operational and analytical purposes, the World development or that other economies have reached a Bank’s main criterion for classifying economies is gross preferred or final stage of development. Classification by national product (GNP) per capita. Every economy is income does not necessarily reflect development status. classified as low-income, middle-income (subdivided This table classifies all World Bank member econ- into lower-middle and upper-middle), or high-income. omies with populations of more than 30,000. Econ- Other analytical groups, based on geographic regions omies are divided among income groups according to and levels of external debt, are also used. 1998 GNP per capita, calculated using the World Bank Low-income and middle-income economies are Atlas method. The groups are: low-income, $760 or sometimes referred to as developing economies. The use less; lower-middle-income, $761–$3,030; upper-middle- of the term is convenient; it is not intended to imply income, $3,031–$9,360; and high-income, $9,361 or that all economies in the group are experiencing similar more. Index References to boxes are noted by b, to figures by f, and to tables by t. Afghanistan, emigration, 38 capital account convertibility, 71 foreign currency deposits and liability, 80 Africa. See also North Africa; Sub-Saharan decentralization of government, 45t, government incentives to reduce Africa; specific countries 113–14 risk-taking, 77 authoritarian governments, 46 foreign direct investment, 37, 73f loan screening, 77 banking crisis, 37f government structure, 116t municipal development funds (MDFs), biodiversity, 43 investment treaties, 72 134 civil society and political parties, 122 private infrastructure, 144 private incentives to reduce risk-taking, decentralization of government, 45–46, tax sharing, 118, 118b 77 123 trade reform, 56f, 59f reform, 77–78 demographic shifts, 44 Armenia, government structure, 115 case study of Hungary, 158, 160–63, foreign direct investment, 72 ASEAN (Association of Southeast Asian 161b infectious diseases, 26 Nations), 85 regulation, 75–79 life expectancy, 26, 26f Asia. See also Central Asia; East Asia; South rural women, loans to, 24 macroeconomics, 16 Asia; specific countries subnational issues, 77 trade, 51 biodiversity, 43 subordinated debt and, 77 urbanization, 47, 47f, 48f, 130, 130b bond market, 133 Bankruptcies, 74 water scarcity, 28b chart by region and income, 290–91 municipal, 133–34 WTO membership, 57f child labor, 62b Barbados, components and parts exports, Aging population, 1, 29, 35, 36b, 38 trade in services, 65 66, 66t Agreement on Sanitary and Phytosanitary urbanization, 47, 47f, 48f Basle Accords, 7, 32, 37–38, 78 Measures, 64 Association of Southeast Asian Nations Biodiversity, 8, 32, 42–43, 90, 93–94, Agriculture, 27–28 (ASEAN), 85 102–5 chart by country, 244–45, 250–53 Australia chart by country, 222–23 crop variations, 42 antidumping laws, 58 Block grants, 118 environmental issues, 89f, 93 immigration trends, 38–40 Bolivia food supplies. See Food shortage public services and horizontal equity, Andean Community trade, 53 GDP percent, 28, 47 110 life expectancy, 26f map of changing crop yields, 89f local governments, 111, 122 trade, 6, 61, 63–64, 63f Balance of payments, chart by country, water pollution, 140–41 water scarcity and, 29b 258–59 Bond market, 133 AIDS, 26, 27b Balance of power, 111–14 Bosnia and Herzegovina, decentralization Air pollution. See also Environmental issues Bangladesh of government, 108, 109b urban areas, 141, 150, 151b banking sector, 24 “Brain drain” from developing countries, Albania, poverty reduction, 111 decentralization of government, 45t 38–39 Algeria, natural resources preservation, 102 flooding due to climate change, 42, 87, Brazil Alliance of Small Island States (AOSIS), 99 99–100, 100f antidumping laws, 58 Alphons, K.J., 144 government structure, 116t banking sector, 37f, 76f Andean Community trade, 53 infant mortality, 142, 142t biodiversity, 43 Antarctica, 8, 88b, 90, 95 life expectancy, 26f components and parts exports, 66, 66t Antibiotics, 27b trade reform, 56f decentralization of government, 45t, Antidumping laws, 6, 34, 58–59, 60f, 60t, Bank for International Settlements. See 113–14, 158, 163–66, 163b 61f Basle Accords educational reforms, 23 Antitrust issues, 6, 52, 59 Banking Regulations and Supervisory firms from and international debt, 70, AOSIS (Alliance of Small Island States), 99 Practices Committee. See Basle 71f Aquaculture, 92b Accords foreign direct investment, 37, 72, 73f Arab Republic of Egypt. See Egypt Banking sector government structure, 33, 116t, 153 Arbitration. See Dispute resolution competition, 35 import-substitution policies and, 2, 13 Argentina deposit insurance, 76–77 localization economies, 127 antidumping laws, 58 developing countries, 7, 35, 70, 73–79 politics, 44, 122 banking sector, 74, 76f, 77 diversification, 75–79 state debt, 119, 124, 133, 164–65 budgetary constraint on local financial contagion, 74, 75b taxes, 117 governments, 124 foreign banks. See Foreign banks technology exports, 59f  (c) The International Bank for Reconstruction and Development / The World Bank   trade in services, 65 foreign direct investment, 37, 72, 73f Costa Rica urbanization issues, 131, 145–46, investment treaties, 72 capital account convertibility, 71 149–50, 153 ozone depletion, 96 Certified Trading Offsets (CTOs), 104 Britain. See United Kingdom rapid economic growth, 2, 16–17 National Biodiversity Institute (INBio), Building codes, 144 social welfare, 152 103 technology exports, 59f Côte d’Ivoire, 37f, 52, 148 Cameroon and forests’ biological resources, total suspended particulates (TSPs), Crime. See Violence 93 141–42 Croatia, components and parts exports, 66t Canada urbanization issues, 130–31, 141 CTOs (Certified Trading Offsets), 104 antidumping laws, 58 Chlorofluorocarbons (CFCs), 8, 42, 94b, Cultural differences biodiversity, 43 95–97 Comprehensive Development bond market, 133 Cholera, 143 Framework and, 21b chart by region and income, 290–91 CITES (Convention on International decentralization of government and, 109b decentralization issues, 114 Trade in Endangered Species), 103, loss of, 1 GATT case for offering industrial 104b Currency crises, 73–75, 75b incentives, 136 Civil society and accountability of local Czech Republic government structure, 116t governments, 122 banking crisis, 37f immigration trends, 38–40 Climate change, 41–42, 41f, 87, 94, bond market, 133 public services and horizontal equity, 103–5. See also Carbon dioxide components and parts exports, 66, 66t 110 emissions environmental issues, 94b Capital account liberalization, 79–81 chart by country, 222–23 government structure, 115 Capital flows. See International capital global attempts to control, 1, 8, 30, 32, municipal development funds (MDFs), flows 97–102 134 Carbon dioxide emissions, 8, 41–42, 42f, map of changing crop yields, 89f trade reform, 58 87, 90, 93. See also Climate change Coal production subsidies, 90–91 chart by country, 248–49, 272 Colombia DAC (Development Assistance Committee economic growth and, 20 Andean Community trade, 53 of the OECD), 20 global attempts to control, 97–102 banking sector, 76f Dam projects of World Bank, 18, 18b trading mechanisms, 101 decentralization of government, 45t, Decentralization of government, 4, 4f, Caribbean. See Latin America 108, 114, 116t, 123 8–10, 28, 32–33, 43–44, 107–24. See CDF. See Comprehensive Development FINDETER program, 134 also Subnational governments Framework (World Bank) trade, 52 balance of power, 111–14 Central Asia. See also specific countries violence, 145, 151–52 block grants, 118 chart by region and income, 290–91 Communications case study of Brazil, 158, 163–66, 163b decentralization of government, 45 chart by country, 266–67 chart by country, 216–17 energy consumption, 101f institutions providing civil society and, 122 poverty, 25f telecommunications, 24 data indicators, 213–14 trade in services, 34f technological advances, 4, 29–30 devolution of powers, as, 108b urbanization, 47f, 48f trade via electronic communication, 33 elections, 121–23 CFCs. See Chlorofluorocarbons Community resources, 50 ethnic diversity and, 109b Child labor, 62b Competition executive power, 114 chart by country, 234–35 antitrust issues, 6, 59 fiscal issues, 115, 117–19, 123 Child mortality, 19f. See Infant mortality banking sector, 35 incentives for national and subnational Chile economic growth and, 17b cooperation, 114 banking sector, 37f, 76f, 77 trade, 52, 59 large democracies, 45, 45t decentralization of government, 123 water supply, 148 legislative allocations and regional disincentives for short-term capital Components and parts exports, 33, 65–67, interests, 113 inflows, 79 66t local administration, effectiveness, 122 political participation and civil society, Comprehensive Development Framework political parties, 122 122 (CDF) (World Bank), 20, 21b, 173 political stability and, 107–8 telephone service, 24 Consumption, chart by country, 238–39 poverty reduction and, 109–11 trade reform, 56f Contraceptive use, chart by country, public service performance and, 108–9 China 242–43 regional influence on national aging population, 36b Convention on Biological Diversity, 8, 43, government, 113 carbon dioxide emissions, 42 88, 94, 94b, 102–3 resource control, 117–19 components and parts exports, 66t Convention on International Trade in sequencing of events, 123 decentralization of government, 46, Endangered Species (CITES), 103, shared functions, 115, 117 108b, 111–12, 113b 104b smaller countries, 45 development strategy, 2, 16–17 Coral reef protection, 103 structures and functions of subnational diaspora, 39–40, 40b Corruption, 17b, 21b governments, 114–21 environmental issues, 90 World Development Reports, 22b taxation, 117–19, 118b (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  Decentralization of government—continued past experiences as starting point, 14–18 information and technology revolution, threat to macroeconomic stability, 111 processes’ and institutions’ importance, effect, 29–30 transfers of revenues, 117–18 3, 11, 14 institutional reforms, 23 transition policies, 123–24 sustainable development agenda, 28, 28b social capital and, 18b Deconcentration and central governments, urbanization and, 49–50 trade liberalization and, 59, 67 108b Diasporas, 39, 40b, 66 vocational training, 137 Deforestation. See Forests and deforestation Diseases, 26, 27b, 141–43. See also Health World Development Reports, 22b Democracies issues Egypt increase in number of, 9f Displaced workers, 6 environmental issues, 97 Democracies, increase in number of, 8–9, Dispute resolution investment treaties, 72 9f, 28, 43, 43f investment agreement provisions, 82 trade reform, 56f, 65, 157–60, 159b Demographic shifts trade complaints, 54–55, 61 Elections. See Decentralization of Africa, 44 Distribution of income or consumption, government; Subnational aging of population, 29, 35, 36b chart by country, 238–39 governments immigration’s effect, 38 Diversification of banking sector, 75–79 Electronic commerce, 33, 65 Middle East, 44 Diversity. See Cultural differences Emerging market funds, 75b urbanization, 46–47, 47f Dollarization, 80 Emissions. See Carbon dioxide emissions Deposit insurance, 76–77 Drugs, 27b, 103 Employment issues Derivative instruments, 34 child labor, 62b Desertification, 8, 87, 88b East Asia. See also specific countries displaced workers, 6 Developing countries. See also specific aging population, effect, 36b labor conditions, 60 countries authoritarian governments, 46 labor costs, 137 antidumping complaints, 61f economic downturn, 13, 16, 17b, 32, labor force, chart by country, 234–35 banking sector, 7, 35, 70, 73–75, 76f 35–37, 74, 75b trade liberalization, 59, 67 biodiversity, 43, 102 economic growth, 2, 11, 14–16, 17b, 48 unemployment, 49–50, 137b child labor, 62b energy consumption, 101f urbanization and, 128, 150–52 climate change, 42 export subsidies and, 2 Endangered species. See Biodiversity emigration of skilled workers, 38–39 poverty, 25, 25f Energy. See also Environmental issues environmental issues, 88, 96, 100 return of emigrants, 39 chart by country, 248–49, 264–65 exchange rates, 71 technology exports, 56 Comprehensive Development financial reform, 69–85 trade in services, 34f Framework and, 21b foreign direct investment in, 37, 38t, trade reform and foreign investment, 81 consumption patterns, 100, 100f, 101f 69–75, 73b urbanization, 47f, 48f falling costs of renewable energy, 98b foreign investment by, 72 Eastern Europe. See also specific countries research, 98 GDP per capita, 14 decentralization of government, 45 subsidies, effect on environment, 90–91 gender-based discrimination, 19–20 economic problems, 17b World Bank projects, 18, 18b, 98b international capital flows, 7f. See housing, 146 Entrepreneurship on municipal level, International capital flows land use, 134 136–38 macroeconomics and, 16, 111f life expectancy, 26, 26f Environmental issues, 1, 8, 20, 30, 32, municipal development funds (MDFs), Economic growth 40–43, 87–105, 88b. See also 134 chart by country, 250–51 Biodiversity; Carbon dioxide poverty, 26 urbanization and, 125–38, 126f emissions public services, 144–55 Economy chart by country, 222–23, 244–49 subnational debts, 120t chart by country, 230–31, 250–59 Comprehensive Development subnational expenditures, 111 chart by region and income, 290–91 Framework and, 21b trade growth, 5–6, 5f, 33, 51–53, 52f, government’s role, chart by country, DAC goal, 20 59, 64–65 262–63 data indicators, 215 urbanization, 10–11, 10f, 46, 47f, 142, Ecuador fishing and overfishing, 91, 92b 144–55 Andean Community trade, 53 institutions that provide physical services Development Assistance Committee genetic material and biodiversity, 103 and, 24 (DAC) of the OECD, 20 Education, 11, 16, 26 international measures needed, 93–94 Development policy, 2–4, 13–30, 172–74 chart by country, 240–41 lead pollution, 141 future outlook, 24–30 child labor and, 62b population growth and, 27–28 government intervention, effect, 15–16 Comprehensive Development protected areas, chart by country, government’s role, 2, 13 Framework and, 21b 246–47 guidelines proposed for, 3 DAC goal, 20 subsidies to industries, effect, 90–91 interdependence of policies, 2–3, 13–14, decentralization of Mexican education, sustainable development agenda and, 28, 173 124 28b investment correlated with growth, 15, 15f East Asia, 16 taxation to pay for damage, 91 objectives of sustainable development, 2, foreign direct investment and, 81 trade and, 67, 104b 13, 18–21 health outcomes and, 19 treaties and conventions, 90, 94–97, 104b (c) The International Bank for Reconstruction and Development / The World Bank   urbanization and, 141 land use in urban areas, 135b Governmental institutions, 23 World Development Reports, 22b, 90, 96b private infrastructure, 143–44 World Development Reports, 22b Equity issues. See Gender equality; vocational training, 137 Great Britain. See United Kingdom Inequality Fund managers’ methods to avoid spread of Greenhouse gases. See Climate change Ethiopia financial crises, 75b Gross domestic product (GDP) decentralization of government, 45t, agriculture and, 28, 47 108, 109b GATS. See General Agreement on Trade in chart by country, 250–57 government structure, 116t Services growth rate, 15, 15f population growth, 44 GATT. See General Agreement on Tariffs per capita, 14, 14f, 16 tax sharing, 117 and Trade Gross national product (GNP), key Europe. See also specific countries GDP. See Gross domestic product indicator for economy, chart by agricultural trade disputes, 63 GEF. See Global Environment Facility country, 272 antidumping laws, 58 Gender equality, 19–20, 26 Growth of the economy. See Economic capital account convertibility, 71 CDF goal, 21b growth chart by region and income, 290–91 DAC goal, 20 coal production subsidies, 90–91 education, chart by country, 240–41 Health issues energy consumption, 101f labor force, chart by country, 234–235 chart by country, 242–43 executive power, 114 World Development Reports, 22b Comprehensive Development municipal development funds (MDFs), General Agreement on Tariffs and Trade Framework and, 21b 134 (GATT), 5, 51, 53, 55 expenditures on, 19 poverty, 25f antidumping actions, 60t improvements, 26 social partnerships, 137b environmental rulings by dispute panel, infectious diseases, 26, 141–43 trade in services, 34f 104 information and technology revolution, urbanization, 47f, 48f subsidies and incentives, 136 effect, 29–30 Exchange rates, 71 trade reform, commitment to, 35f institutions and sectoral issues involving, Executive power, 114 General Agreement on Trade in Services 24 Exports. See Trade (GATS), 65, 82 social capital and, 18b Germany trends, 27b Finance. See also Banking sector; banking sector, 76f urban growth and, 141 Investment; Taxation coal production subsidies, 91 World Development Reports, 22b chart by country, 256–57 decentralization issues, 114 Hedge funds, 34 Financial contagion, 74, 75b government structure, 116t HIV. See AIDS Financial flows, 6–7, 7f, 34–38. See also immigration trends, 38–40 Homeless people, 48 International capital flows legislative representation, 113 Honduras chart by country, 270–71 public services and horizontal equity, 110 components and parts imports, 66 developing countries and, 69–85 taxes, 117 life expectancy, 26f institutionally managed funds, 71, 71f Global Biodiversity Assessment (UNEP), Hong Kong. See also China Financial integration of world, 1, 69–75 102 components and parts exports, 66, 66t Fishing and overfishing, 8, 91, 92b Global commons, 87–105. See also Horizontal equity and public services, chart by country, 222–23 Environmental issues 110 Food shortage, 1, 13, 27–28, 64 Global environment. See Climate change Housing, 10–11, 48 chart by country, 232–33 Global Environment Facility (GEF), 8, 43, affordable housing, 139, 141f, 144–46 food stamps and food distribution, 23–24 94b, 98b, 102 chart by country, 220–21 Foreign aid, 73b Global integration, 17b prices, 135 chart by country, 270–71 Global production networks and Human capital. See Education; Life Foreign banks, 35, 74, 78–79 international trade, 51, 65–67, 66t expectancy Foreign direct investment, 6, 7f, 36–38, Global warming. See Climate change Human needs and development, 1, 3 37f, 38t, 69–75 Globalization, 173–74. See also development services provided by attracting, 81–84 Supranational issues institutions, 23–24 chart by country, 270–71 advantages and disadvantages, 4–5 Human rights, 1, 43 net flows to developing countries, 70f, defined, 2 Hungary 73b factors contributing to, 4 banking reform, 158, 160–63, 161b regional foreign investment agreements, financial markets, 34–38 firms from and international debt, 71f 83–84 partnerships and institutions required, 3 government structure, 115 service industries, 72 trade and, 51–68. See also Trade trade reform, 56f, 58 Forests and deforestation, 90, 93–94 trend of, 31 chart by country, 246–47 urban challenges and, 140 IBPGR (International Board for Plant Framework Convention on Climate Government finances. See also Taxation Genetic Resources), 42 Change, 8, 88, 94b, 102 chart by country, 256–57 ICSID (International Centre for Settlement France Government intervention and economic of Investment Disputes), 82 government structure, 116t growth, 15–16, 20, 21b Illiteracy, chart by country, 232–33, 272 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  ILO (International Labour Organization), subnational debts, 120t International trade. See Trade 62b subnational expenditures, 111 Internet, 4f, 33 IMF. See International Monetary Fund trade in services, 34f chart by country, 266–67 Import-substitution policies, 13 trade reform, 56, 58 Investment, 2. See also Foreign direct Imports. See Trade urbanization, 47f investment INBio (National Biodiversity Institute Industrial sector, 51, 65 attracting foreign investment, 81–84 (Costa Rica)), 103 chart by country, 250–53 capital flows. See International capital Income insurance, 50. See also Social Inequality. See also Gender equality flows welfare decentralization of government and, chart by country, 270–71 Income tax. See Taxation 109–11 East Asia, 17b Incomes economic growth and, 15 economic growth correlated with, 15, distribution of income, chart by country, horizontal equity, 110 15f 238–39 income inequality and import increases, foreign direct investment, 7f map of world, 229 59 foreign portfolio investment, 6, 7f, 70f rich vs. poor countries, 14–15, 14f within-state equity, 110–11 urban growth and capital investment, India Infant mortality, 19–20, 19f, 26, 48, 142, 132–34 air pollution, 141 142t IPCC (Intergovernmental Panel on Climate antidumping laws, 58 chart by country, 232–33, 242–43 Change), 94, 97 banking sector, 76f Infectious diseases, 26, 27b. See also Health Iran citizens’ report cards, 153, 154b issues decentralization of government, 45t corruption, 144–45 Inflation, 30 government structure, 116t food supply, 13 Information Ireland and social partnerships, 137b government structure, 115, 116t chart by country, 266–67 Irrigation. See Water health issues in urban areas, 143 economic value, 4 Italy industrial challenges, 49 revolution in, 29–30 government structure, 116t investment projects, 83 Infrastructure. See also Privatization; specific legislative representation, 114 legislative representation, 113 type of service vocational training, 137 local governments, 46, 109, 110b publicly financed, 132–33 ozone depletion, 96 World Development Reports, 22b Jamaica politics, 44 Institutionally managed funds and financial capital account convertibility, 71 power projects, 98b flows, 71, 71f food stamps, 23–24 taxes and revenues, 117–18 Institutions, 3–4, 11, 21–24, 23b, 174 Japan trade, 52, 56, 59f governmental, 23 aging population, effect, 36b urbanization issues, 131–32, 136, 141, human development services, 23–24 banking sector, 76f 145, 153, 154b infrastructure and physical services, 24 coal production subsidies, 90 within-state equity, 110 national, 32 foreign banks in U.S., 79 Indonesia subnational, 32–33, 46 foreign investments by Japanese firms, banking sector, 37f, 76f, 80 supranational, 32 81 Clean Rivers program, 24 Integrated Framework for Trade and government structure, 116t components and parts imports, 66 Development in the Least-Developed ozone depletion, 97 environmental issues, 93 Countries (World Bank), 56, 58b rapid economic growth, 2 foreign direct investment, 72, 73f Intergovernmental Panel on Climate Jordan health status, 19 Change (IPCC), 94, 97 housing, 146 Kampung Improvement Programs International Board for Plant Genetic taxes, 117 (KIPs), 146 Resources (IBPGR), 42 land use, 147b International capital flows, 6, 7f, 35, 70–73 Kazakhstan and life expectancy, 26f life expectancy, 26f benefits of liberalization, 75 Kenya trade reform, 56f chart by country, 270–71 decentralization of government, 45t transportation, 136 International Centre for Settlement of gender-based discrimination, 20 urbanization issues, 130–31, 141, 146, Investment Disputes (ICSID), 82 government structure, 116t 147b, 149 International Convention for the Prevention trade reform, 56f within-state equity, 110 of Pollution from Ships, 94b Knowledge, 1. See also Information Industrial countries. See also specific countries International Covenant on Civil and Korea, Republic of aging population, 36b Political Rights, 43 aging population, 36b antidumping complaints, 61f International Development Law Institute antidumping laws, 58 energy-related issues, 100, 100f, 101f (Rome), 83 banking sector, 37f, 76f foreign direct investment stock, 38t International Labour Organization (ILO), central government’s loss of authority, government structure, 116t 62b 44 infrastructure, 132 International Monetary Fund (IMF), 7 components and parts exports, 66t institutionally managed funds invested banking standards memorandum of decentralization of government, 45t abroad, 71, 71f understanding, 78 government structure, 116t (c) The International Bank for Reconstruction and Development / The World Bank   industry’s dispersal, 129b, 131–32 Macedonia and banking crisis, 37f MIGA (Multilateral Investment Guarantee investment treaties, 72 Macroeconomics Agency), 83b localization economies, 127 African countries and, 16 Migration trends, 38–40 trade, 52 Comprehensive Development Military expenditures, chart by country, training of skilled workers, 39 Framework and, 21b 262–63 urbanization, 48 developing countries and, 16 Millennium Round, 6, 34 Kuznets, S., 15 East Asia and, 16, 17b Monopolies. See Antitrust issues Kyoto Convention, 42, 94, 99b, 101, 104 investment reform, 84–85 Montreal Protocol, 8, 32, 95, 96b, 104b stability, 1, 3, 16, 33 Morocco Labor issues. See Employment issues stability and decentralization of decentralization of government, 45t Land use, 134–36, 135b, 138b, 147b, 151 government, 111 trade, 52 chart by country, 244–45 World Development Reports, 22b Mortality rates. See Infant mortality; Life Latin America. See also specific countries Madagascar and government structure, 115 expectancy biodiversity, 43 Malawi and government structure, 115 Mozambique and government structure, bond market, 133 Malaysia 116t capital account convertibility, 71 banking sector, 37f, 74, 76f Multilateral Investment Guarantee Agency chart by region and income, 290–91 components and parts exports, 66t (MIGA), 83b debt crisis, 16, 32, 111 foreign direct investment, 72, 73f Multinational corporations, 35, 37 decentralization of government, 45, 109, government structure, 116t foreign direct investment, 69–70, 72, 81 111, 123 land use, 135 trade policies and, 66, 81 emigration of skilled workers, 39 trade, 59f Municipalities. See Urbanization energy consumption, 101f vocational training, 137 environmental issues, 94b Manufacturing. See Industrial sector NAFTA (North American Free Trade executive power, 114 Marine pollution, 88b Agreement), 82–83 health care expenditures, 19 Marine Stewardship Council, 92b National governments industrial challenges, 49 MDFs (municipal development funds), 134 policy trends, 32 international capital flows, 74 Medical care. See Health issues privatization and, 108b investment treaties, 72 Memorandum of understanding, banking regional influence on, 113 poverty, 25, 25f standards, 78 types of, 4 services provided through partnerships, Mercado Común del Sur (MERCOSUR), urbanization and, 130–32 145 83 Natural resources protection, 91. See also trade, 51, 56 Merck and Company, funding of Costa Biodiversity; Fishing and overfishing; trade in services, 34f Rican National Biodiversity Institute Forests and deforestation training needs, 49 (INBio), 103 chart by country, 246–47 urbanization, 47f, 48f MERCOSUR (Mercado Común del Sur), international investment and, 72–73 Latvia and government structure, 115 83 Nepal Lead pollution, 141 Mexico agriculture, 28 Legal frameworks, necessity of, 17b, 21b, 23 antidumping laws, 58 decentralization of government, 45t Legislative allocations and regional banking crisis, 35, 37f, 74, 75b, 79–80 government structure, 116t interests, 113 banking sector, 76f New Zealand, antidumping laws, 58 Lewis, A., 15 components and parts trade, 66, 66t NGOs. See Nongovernmental Life expectancy, 11, 16, 19, 26, 26f, 48 decentralization of government, 45, 45t, organizations chart by country, 232–33, 272 113, 124 Nicaragua, components and parts exports, DAC goal, 20 energy projects with World Bank, 102 66, 66t demographic shift, 29 firms from and international debt, 70, 71f Nigeria Living conditions in cities. See foreign direct investment, 37, 72, 73f, 82 decentralization of government, 45t Urbanization government structure, 116t infrastructure failure, 49 Loans from banks. See Banking sector political participation and civil society, Nongovernmental organizations (NGOs), Local governments. See Subnational 122 3–4, 43 governments tax sharing, 118b civil society and, 122 Localization. See also Subnational trade, 52, 56f environmental role, 96, 96b governments Middle East. See also specific countries housing assistance, 146 advantages and disadvantages, 5 chart by region and income, 290–91 public service providers, as, 145 defined, 2 decentralization of government, 45 sewerage systems, 149 economies, 127 demographic shifts, 44 North Africa. See also specific countries factors contributing to, 4 energy consumption, 101f chart by region and income, 290–91 partnerships and institutions required, foreign direct investment, 72 decentralization of government, 45 3–4 life expectancy, 26f energy consumption, 101f trend of, 31–33, 43–46, 173–74 poverty, 25f poverty, 25f urban challenges and, 140 trade in services, 34f trade in services, 34f Long-term foreign investment, 7, 70, 81 urbanization, 47f, 48f urbanization, 47f, 48f (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  North American Free Trade Agreement Poor countries. See also Developing Reverse migration, 39 (NAFTA), 82–83 countries Rich countries, growth of incomes, 14–15, growth of incomes, 14–15, 14f 14f OECD. See Organisation for Economic water scarcity, 29b Rio Earth Summit, 8, 42, 88, 90, 102 Co-operation and Development Popular participation in government, 43. Roads. See Transportation Oman and life expectancy, 26f See also Democracies Romania, decentralization of government, Organisation for Economic Co-operation Population growth 45t and Development (OECD), 20, 94, 96 urban areas, 128f, 130f, 140 RTAs. See Regional trade arrangements Ozone depletion and treaties, 8, 32, 42, world-wide, 1, 27 Rules. See Institutions 94–97, 95f Poverty reduction, 1, 24–26, 25f. See also Rural development Food shortage case study of Tanzania, 158, 169–72, Pacific. See East Asia chart by country, 236–37 170b Pakistan child labor and, 62b Comprehensive Development government, 45t, 46, 113, 116t Comprehensive Development Framework and, 21b infrastructure failure, 49 Framework and, 21b institutional structures and, 24 job training, 137 DAC goal, 20 social capital and, 18b sewerage systems and Orangi Research decentralization of government and, vs. urban growth, 10, 28, 130, 140. See and Training Institute, 149, 153, 109–11 also Urbanization 168–69 economic growth and, 15 World Development Reports, 22b urban conditions, 141f, 147, 158, local government’s responsibility, 11 Russian Federation 166–69, 167b social capital and, 18b biodiversity, 43 Participatory politics. See Decentralization urban areas and, 49–50, 142–43, bond market, 133 of government 150–52 decentralization of government, 45, 45t, Partnerships, 3–4, 40 World Development Reports, 22b 108–9, 123, 123b Comprehensive Development Power projects. See Energy economic problems, 16, 17b, 44 Framework and, 21b Power projects of World Bank, 18b government structure, 33, 116t, 123 cross-border partnerships with firms in Private sector finance, chart by country, housing, 146 developing countries, 72 260–61 life expectancy, 26, 26f urban services, 139, 145 Privatization, 18, 18b, 143–45, 153 ozone depletion, 97 Patents, chart by country, 266–67 foreign direct investments and, 82 politics and elections, 44, 113 Pension assets. See Aging population local governments and, 122, 142 private enforcement of laws, 44 Persistent organic pollutants, 87, 88b national governments and, 108b social problems, 44 Peru “self-provision,” 145 Andean Community trade, 53 water supply, 143–44, 147–48, 153 Sanctions governmental authority, 46 Production networks and international biodiversity protections, 103 Pharmaceutical drugs, 27b, 103 trade, 51, 65–67, 66t trade violations, 60, 62b Philippines Productivity Sanitary and phytosanitary regulations, 64 Community Mortgage Program, 146 output, chart by country, 252–53 Sanitation. See Sewerage components and parts trade, 66, 66t trade and, 52 Saudi Arabia, foreign direct investment, decentralization of government, 45, 45t, Public services 73f 111 decentralization of government, 108–10 Savings government structure, 116t developing countries, 144–55 East Asia, 17b trade reform, 56f horizontal equity, 110 pension assets. See Aging population urban poverty program, 151 shift to private sector. See Privatization Science. See Technology vehicle emission inspections, 150, 151b urban areas, 139–44 Services sector water pollution, 141 chart by country, 250–55 Poland Quality of life. See Urbanization economic growth in, 29–30 biodiversity project, 102 economic value, 4 bond market, 133 Regional economic development, 137b foreign direct investment, 72 decentralization of government, 45, 45t, Regional foreign investment agreements, trade, 6, 33, 34f, 64–65, 64f 123 83–84 chart by country, 268–69 foreign direct investment, 37 Regional trade arrangements (RTAs), 53, Sewerage government structure, 116t 54b, 54f, 56, 66 chart by country, 220–21, 232–33, trade reform, 58 Reproductive health services 242–43 Polio, 26 Comprehensive Development Comprehensive Development Political parties, 122 Framework and, 21b Framework and, 21b Political pluralism. See Decentralization of DAC goal, 20 urban areas and, 11, 49, 140, 140b, 142, government Republic of Korea. See Korea, Republic of 148–49 Political rights, 1, 43–44 Responsive processes and development, 1, 3 Sex equality. See Gender equality Pollution. See Air pollution; Environmental Retirement age, 36b Short-term foreign investment, 7, 69–70, issues; Sewerage Revenue sharing, 118b 79 (c) The International Bank for Reconstruction and Development / The World Bank   Singapore trade in services, 34f firms from and international debt, 70, capital account convertibility, 71 urbanization, 47f, 48, 48f 71f components and parts exports, 66t water scarcity, 29b foreign direct investment, 72 foreign direct investment, 73f Subnational governments, 3, 8–11, 32–33, government structure, 116t Slovak Republic, government structure, 115 44–46. See also Decentralization of land use, 135 Slovenia, components and parts exports, 66t government poverty, 25 Slums. See Housing accountability, 121–23 urban transportation problems, 141 Smallpox, 26 banking sector, 77 The Theory of Economic Growth, 15 Social capital, 18, 18b, 22 central regulation of, 119, 121 Trade, 51–68. See also General Agreement urbanization and, 49 debts and borrowing controls, 119, 120t, on Tariffs and Trade (GATT); World Social welfare, 1, 50, 150–52 124 Trade Organization (WTO) Socially inclusive processes and electoral rules, 121–22 agriculture, 6, 61, 63–64 development, 1, 21b expenditures, 111f, 118 benefits of liberalization, 52–53 SODECI, 148 foreign direct investments, 82, 83b chart by country, 250–51, 254–55, “Sourcing,” 33 horizontal equity, 110 258–59, 268–69 South Africa incentives for national and subnational competition, 52, 59 antidumping laws, 58 cooperation, 114 components and parts exports, 33, bond market, 133 personnel issues, 119, 121 65–67, 66t decentralization of government, 45, 45t, structures and functions, 114–21 developing countries, 51–53, 52f 107, 108b taxes and revenues, 112f, 117–19, 118b, dispute resolution, 54–55, 61 firms from and international debt, 71f 123 employment issues, 59, 67 government structure, 116t within-state equity, 110–11 environmental issues, 67, 104b legislative representation, 113 Subsidies final goods production, 52–53 trade reform, 56f chart by country, 262–63 growth, 5–6, 5f urban management, 145 coal production, 90–91 import-substitution policies, 2, 13 South America. See also specific countries export, 2 Integrated Framework for Trade and Andean Community trade, 53 municipalities giving to overseas Development in the Least-Developed chart by region and income, 290–91 industries, 136 Countries (World Bank), 56, 58b South Asia. See also specific countries Supranational issues, 3, 5–8, 32 production networks, 65–67, 66t diaspora, 40, 40b productivity and, 52 emigration of skilled workers, 39 Taiwan (China) reduction of trade barriers, 55 energy consumption, 101f banking sector, 76f reform, 33–34, 53–54, 56–61, 56f health care expenditures, 19 components and parts exports, 66, 66t case study of Egypt, 157–60, 159b poverty, 25, 25f computer industry investments in foreign investments and, 81–82 trade, 51 Europe and U.S., 40b regional trade arrangements (RTAs), 53, trade in services, 34f trade, 52 54b, 54f, 56, 66 urbanization, 47f, 48, 48f training of skilled workers, 39 sanctions, effectiveness, 60, 62b Soviet Union. See now Russian Federation Tanzania and urban-rural synergies, 158, services. See Services sector economic growth, 15–16 169–72, 170b “sourcing,” 33 land use, 134, 135b Tariffs. See Trade subsidies, 2 urbanization issues, 130–31 Taxation tariff binding, 55, 63, 63f Spain chart by country, 256–57, 262–63 tax implications, 66–67 government structure, 116t decentralized governments, 117–19, 123 technology exports, 59f legislative representation, 114 environmental taxes, 91 transparent trade policy regimes and, 55 Sri Lanka horizontal equity and, 110 trends, 33–34, 51–68 government structure, 119 international investment, 72 urban growth and, 67, 131, 136 health care expenditures, 19 international trade profits, 66–67 World Development Reports, 22b housing programs, 146 location choices for businesses, 137 Training. See Education Stock markets, 84 subnational and local taxes, 112f, Transition economies Sub-Saharan Africa. See also specific 117–19, 118b, 123 subnational debts, 120t countries tax sharing, 117, 118b subnational expenditures, 111t agriculture, 28 Technology subnational taxes, 112f chart by region and income, 290–91 chart by country, 266–67 Transportation diaspora, 40 innovations, 1, 29–30, 33 chart by country, 220–21, 264–65 emigration of skilled workers, 39 trade, 56, 59f, 81 Comprehensive Development energy consumption, 101f Telecommunications. See Communications Framework and, 21b energy projects of World Bank, 18, 18b Thailand reduced costs, 4 health care expenditures, 19 air pollution, 141 urban areas, 11, 136–37, 149 life expectancy and education, 11, 16 banking sector, 37f, 75b, 76f, 79–80 Treaties politics, 44 components and parts trade, 66, 66t environmental, 90, 94–97, 95f, 104b poverty, 25, 25f decentralization of government, 45t foreign investors’ rights, 72, 83 (c) The International Bank for Reconstruction and Development / The World Bank      ⁄  TSPs (total suspended particulates), chart by country, 218–19, 232–33 Water 141–42 citizens’ report cards, 153, 154b chart by country, 220–21, 242–43, Tuberculosis (TB), 27b, 141 data indicators, 214 246–47 Turkey definitions of terminology, 127b Comprehensive Development bond market, 133 dynamics of city formation, 128–30 Framework and, 21b government structure, 116t, 119 economic analysis, importance of, 138b partnerships with community return of emigrants, 39 economic growth and, 125–38, 126f organizations, 148, 148b trade, 52 entrepreneurship on municipal level, private providers of public water supply, urbanization issues, 145 136–38 143–44, 147–48, 153 housing. See Housing scarcity, 1, 28, 29b Uganda land use. See Land use urban areas, 11, 140f, 143–44, 146–48 decentralization of government, 45t, living conditions, 139–55 Welfare. See Social welfare 108, 108b case study of Pakistan, 158, 166–69, Women. See Gender equality government structure, 116t 167b World Bank life expectancy, 26f chart by country, 220–21, 232–33 banking standards memorandum of Ukraine data indicators, 214–15 understanding, 78 decentralization of government, 45, 45t local policies for economic growth, Comprehensive Development economic and social instability, 44 132–38 Framework, 20, 21b, 173 government structure, 116t municipal bankruptcies, 133–34 energy projects in Mexico, 102 Unemployment, 49–50, 137b, 150, 152b. municipal bonds, 133 foreign investment role, 83b See also Employment issues municipal development funds (MDFs), Integrated Framework for Trade and United Kingdom 134 Development in the Least-Developed government structure, 116t national government’s role, 130–32 Countries, 56, 58b health status, 19 population and size of cities, 128f, 130f, projects and successful development, municipal development funds and 140 17–18, 18b bonds, 134 poverty. See Poverty reduction World Development Reports since 1990, 20, Non-Fossil Fuel Obligation (NFFO), production site choice, 131 22b, 90 98b public services, 139–44 World Trade Organization (WTO), 5–6, pension assets, 35 rural-urban linkages, 127, 128b 32, 51 private provision of basic services, 143 case study of Tanzania, 158, 169–72, African membership, 57f United Nations (UN) 170b antidumping actions, 60t Convention on the Law of the Sea sanitation issues. See Sewerage benefits of, 53–56 (UNCLOS), 88b systems of cities, 127–28 developing countries’ representation, Convention to Combat Desertification, trade and, 67, 131, 136 55–56, 57f 88b transportation issues, 11, 136–37, dispute resolution, 54–55, 61 Environment Programme (UNEP), 42, 149–50 environmental ruling, 104 88b, 96b, 102, 105 water issues. See Water labor conditions, 60 environmental issues, 42, 92b, 104b Uruguay Round, 52, 55–56, 61, 63–65, number of member countries, 6f, 33, Global Shelter Strategy for the Year 63f 35f, 53 2000, 146 Utilities. See Public services reduction of trade barriers, 55 Human Rights Committee, 43 Uzbekistan and poverty reduction, 111 regional trade arrangements and, 54b United States scope of, 53f aging population, 35, 36b Value added tax (VAT). See Taxation tariff binding, 55 antidumping laws, 58 Venezuela Trade Policy Review Mechanism, 55 banking sector, 76f Andean Community trade, 53 trade reform, 53–54 bond market, 133 banking sector, 74, 76f transparent trade policy regimes and, 55 carbon dioxide emissions, 42, 97 capital account convertibility, 71 WTO. See World Trade Organization chart by region and income, 290–91 decentralization of government, 45t decentralization issues, 114 government structure, 116t Yemen and life expectancy, 26f executive power, 114 political participation and civil society, Yugoslavia government structure, 116t 122 tax sharing, 117 immigration trends, 38–40 Vienna Convention for the Protection of trade reform, 58 institutionally managed funds invested the Ozone Layer, 95 abroad, 71, 71f Vietnam Zambia metropolitan areas, 130 taxation and horizontal equity, 110 government structure, 115 pension assets, 35 urbanization issues, 130–31, 141, 150 life expectancy, 26f state marketing offices overseas, 136 water supply, 148, 148b local governments, 109 taxes, 117 Violence, 48, 142, 151–52 Zimbabwe Urbanization, 10–11, 10f, 28–29, 46–50, environmental issues, 97 125–55 Wages, 60, 62b, 137 life expectancy, 26f agglomeration economies, 126–27 Waste disposal. See Environmental issues Zoning, 134 (c) The International Bank for Reconstruction and Development / The World Bank DISTRIBUTORS CHINA The Middle East Observer INDONESIA Chinese Corporation for Promotion and 41, Sherif Street Pt. Indira Limited OF WORLD BANK Humanities 15, Ding Hui Dong Li, Kun Lan Hotal Cairo Tel: (20 2) 392 6919 Jalan Borobudur 20 PO Box 181 GROUP Haidian District 100036 Beijing Fax: (20 2) 393 9732 E-mail: mafouda@meobserver.com.eg Jakarta 10320 Tel: (62 21) 390 4290 PUBLICATIONS Tel: (86 10) 88117711 Fax: (86 10) 88129871 URL: www.meobserver.com.eg Fax: (62 21) 390 4289 Prices and credit terms vary from E-mail: wangjiang99@yahoo.com FINLAND IRAN country to country. Please con- Akateeminen Kirjakauppa Kowkab Publishers China Book Import Centre P.O. Box 128 P.O. BOX 19575-511 sult your local distributor before Ms. Zhou Rui FIN-00101 Helsinki Tehran placing an order. P.O. Box 2825 Tel: (358 9) 121 4418 Tel: (98 21) 258 3723 Beijing Fax: (358 9) 121 4435 Fax: (98 21) 258 3723 E-mail: akatilaus@akateeminen.com E-mail: kowkabpub@tavana.net ARGENTINA China Financial & Economic Publishing URL: www.akateeminen.com World Publications SA House Ketab Sara Co. Publishers Av. Cordoba 1877 8, Da Fo Si Dong Jie FRANCE P.O. Box 15745-733 1120 Buenos Aires Beijing Editions Eska; DJB/Offlib Tehran 15117 Tel: (54 11) 815 8156 Tel: (86 10) 6401 7365 12, rue du Quatre-Septembre Tel: (98 21) 8716104 Fax: (54 11) 815 8156 Fax: (86 10) 6401 7365 75002 Paris Fax: (98 21) 8712479 E-mail: wpbooks@infovia.com.ar Tel: (33 1) 42 86 58 88 E-mail: ketab-sara@neda.net.ir COLOMBIA Fax: (33 1) 42 60 45 35 AUSTRALIA, PAPUA NEW GUINEA, Infoenlace Ltda./An IHS Group Company E-mail: offilib@offilib.fr IRELAND FIJI, SOLOMON ISLANDS, VANUATU, Calle 72 No. 13-23 - Piso 3 URL: www.offilib.fr Government Supplies Agency AND SAMOA Edificio Nueva Granada 4-5 Harcourt Road D.A. Information Services P.O. Box (A.A) 34270 GERMANY Dublin 2 648 Whitehorse Road Santafé de Bogotá, D.C. UNO-VERLAG Tel: (353 1) 661 3111 Mitcham 3132 Tel: (57 1) 255 8783 Poppelsdorfer Allee 55 Fax: (353 1) 475 2670 Victoria, Australia Fax: (57 1) 248 0808 D-53115 Bonn URL: www.opw.ie/govt.htm Tel: (61 3) 9210 7777 E-mail: infoenlace@gaitana.interred.net.co Tel: (49 228) 949020 Fax: (61 3) 9210 7788 Fax: (49 228) 217492 ISRAEL E-mail: service@dadirect.com.au CôTE D’IVOIRE E-mail: unoverlag@aol.com Yozmot Literature Ltd. URL: www.dadirect.com.au Centre d’Edition et de Diffusion Africaines URL: www.uno-verlag.de P.O. Box 56055 (CEDA) 3 Yohanan Hasandlar St. AUSTRIA 04 B.P. 541 GHANA Tel Aviv 61560 Gerold and Co. Abidjan 04 Epp Books Services Tel: (972 3) 5285 397 Weihburggasse 26 Tel: (225) 24 6510 Post Office Box 44 Fax: (972 3) 5285 397 A-1010 Wien Fax: (225) 25 0567 TUC Tel: (43 1) 512 47310 E-mail: info@ceda-ci.com Accra R.O.Y. International Fax: (43 1) 512 473129 URL: www.ceda-ci.com Tel: (233 21) 778843 P.O. Box 13056 E-mail: buch@gerold.telecom.at Fax: (233 21) 779099 Tel Aviv 61130 CYPRUS E-mail: epp@africaonline.com.gh Tel: (972 3) 649 9469 BANGLADESH Center for Applied Research Fax: (972 3) 648 6039 Micro Industries Development Assistance 6, Diogenes Street, Engomi GREECE E-mail: royil@netvision.net.il Society (MIDAS) P.O. Box 2006 Papasotiriou S.A., URL: www.royint.co.il House 5, Road 16 Nicosia International Technical Bookstore Dhanmondi R/Area Tel: (357 2) 59 0730 35, Stournara Str. Palestinian Authority/Middle East Dhaka 1209 Fax: (357 2) 66 2051 106 82 Athens Index Information Services Tel: (880 2) 326427 E-mail: ttzitzim@sting.cycollege.ac.cy Tel: (30 1) 364 1826 P.O.B. 19502 Jerusalem Fax: (880 2) 811188 Fax: (30 1) 364 8254 Tel: (972 2) 6271219 CZECH REPUBLIC E-mail: gprekas@papasotiriou.gr Fax: (972 2) 6271634 BELGIUM USIS, NIS Prodejna Jean de Lannoy Havelkova 22 HAITI ITALY Av. du Roi 202 130 00 Prague 3 Culture Diffusion Licosa Libreria Commissionaria Sansoni 1060 Brussels Tel: (42 2) 2423 1486 Mr. Yves Clément Jumelle S.P.A. Tel: (32 2) 538 5169 Fax: (42 2) 2423 1114 5, Rue Capois Via Duca di Calabria 1/1 Fax: (32 2) 538 0841 E-mail: pospisilovaj@usiscr.cz C.P. 257 50125 Firenze E-mail: jean.de.lannoy@infoboard.be URL: : www.usiscr.cz Port-au-Prince Tel: (39 55) 645 415 Tel: (509) 23 9260 Fax: (39 55) 641 257 BOSNIA AND HERZEGOVINA DENMARK Fax: (509) 23 4858 E-mail: licosa@ftbcc.it Book Trading Company “Sahinpasic” Samfundslitteratur URL: www.ftbcc.it/licosa Marsala Tita 29/II Rosenoerns Alle 11 HONG KONG, CHINA; MACAU 71000 Sarajevo DK-1970 Frederiksberg C Asia 2000 Ltd. JAMAICA Tel: (387 71) 64 48 56 Tel: (45 35) 351942 Sales & Circulation Department Ian Randle Publishers Ltd Fax: (387 71) 64 48 56 Fax: (45 35) 35 7822 302 Seabird House 206 Old Hope Road E-mail: tajib@btcsahinpasic.com E-mail: ck@sl.cbs.dk 22-28 Wyndham Street, Central Kingston 6 URL: www.btcsahinpasic.com URL: www.sl.cbs.dk Hong Kong, China Tel: (876) 927 2085 Tel: (852) 2530 1409 Fax: (876) 977 0243 BRAZIL ECUADOR Fax: (852) 2526 1107 E-mail: irpl@colis.com Publicacões Tecnicas Internacionais Ltda. Libri Mundi—Libreria Internacional E-mail: sales@asia2000.com.hk Rua Peixoto Gomide, 209 Juan Leon Mera 851 URL: www.asia2000.com.hk JAPAN 01409 Sao Paulo, SP P.O. Box 17-01-3029 Eastern Book Service (EBS) Tel: (55 11) 259 6644 Quito HUNGARY 3-13 Hongo 3-chome, Bunkyo-ku Fax: (55 11) 258 6990 Tel: (593 2) 521606 Euro Info Service Tokyo 113 E-mail: postmaster@pti.uol.br Fax: (593 2) 504209 Margitszgeti Europa Haz Tel: (81 3) 3818 0861 URL: www.uol.br E-mail: librimu1@librimundi.com.ec H-1138 Budapest Fax: (81 3) 3818 0864 Tel: (36 1) 350 80 24; 350 80 25 E-mail: orders@svt-ebs.co.jp CANADA CODEU Fax: (36 1) 350 90 32 URL: www.svt-ebs.co.jp Renouf Publishing Co. Ltd. Ruiz de Castilla 763, Edif. Expocolor E-mail: euroinfo@mail.matav.hu 5369 Canotek Road Primer piso, Of. #2 KENYA Ottawa, Ontario K1J 9J3 Quito INDIA Legacy Books Tel: (613) 745-2665 Tel: (593 2) 507-383 Allied Publishers Ltd. Loita House Fax: (613) 745-7660 Fax: (593 2) 507-383 751 Mount Road P.O. Box 68077 E-mail: order.dept@renoufbooks.com E-mail: codeu@impsat.net.ec Madras 600 002 Nairobi URL: www.renoufbooks.com Tel: (91 44) 852 3938 Tel: (254 2) 330853 EGYPT, ARAB REPUBLIC OF Fax: (91 44) 852 0649 Fax: (254 2) 330854 Al Ahram Distribution Agency E-mail: allied.mds@smb.sprintrpg.ems. E-mail: legacy@form-net.com Al Galaa Street vsnl.net.in Cairo Tel: (20 2) 578 60 83 Fax: (20 2) 578 68 33 (c) The International Bank for Reconstruction and Development / The World Bank Africa Book Service (E.A.) Ltd. Oasis Official SINGAPORE; TAIWAN, CHINA, THAILAND Mr. Talat Lone P.O. Box 3627 MYANMAR; BRUNEI Central Books Distribution Co. Ltd. Quaran House, Mfangano Street Wellington Hemisphere Publishing Services 306 Silom Road P.O. Box 45245 Tel: (64 4) 4991551 Golden Wheel Building Bangkok 10500 Nairobi Fax: (64 4) 499 1972 41 Kallang Pudding Road, #04-03 Tel: (66 2) 233-6930-9 Tel: (254 2) 223 641 E-mail: oasis@actix.gen.nz Singapore 349316 Fax: (66 2) 237-8321 Fax: (254 2) 330 272 URL: www.oasisbooks.co.nz Tel: (65) 741 5166 Fax: (65) 742 9356 TRINIDAD & TOBAGO AND KOREA, REPUBLIC OF NIGERIA E-mail: info@hemisphere.com.sg THE CARIBBEAN Dayang Books Trading Co. University Press Plc Systematics Studies Ltd. International Division Three Crowns Building Jericho SLOVENIA St. Augustine Shopping Center 954-22, Bangbae-Dong, Socho-ku Private Mail Bag 5095 Gospodarski vestnik Publishing Group Eastern Main Road Seoul Ibadan Dunajska cesta 5 St. Augustine Tel: (82 2) 582 3588 Tel: (234 22) 411356 1000 Ljubljana Tel: (868) 645 8466 Fax: (82 2) 521 8827 Fax: (234 22) 412056 Tel: (386 61) 133 83 47 Fax: (868) 645 8467 E-mail: dico3@chollian.net E-mail: suike@hotmail.com Fax: (386 61) 133 80 30 E-mail: tobe@trinidad.net E-mail: repansekj@gvestnik.si Eulyoo Publishing Co., Ltd. PAKISTAN URL: www.gvestnik.si/EUROPA/index.htm UGANDA 46-1, Susong-Dong Oxford University Press Gustro Limited Jongro-Gu 5 Bangalore Town, Sharae Faisal SOUTH AFRICA, BOTSWANA P.O. Box 9997 Seoul P.O. Box 13033 For single titles: Madhvani Building Tel: (82 2) 734 3515 Karachi 75350 Oxford University Press Southern Africa Plot 16/4, Jinja Road Fax: (82 2) 732 9154 Tel: (92 21) 446307; 449032; 440532 P.O. Box 12119 Kampala E-mail: eulyoo@chollian.net Fax: (92 21) 4547640;449032 N1 City 7463 Tel: (256 41) 251467 E-mail: usmanm@oup.net.pk Cape Town Fax: (256 41) 251468 LEBANON Tel: (27 21) 595 4400 E-mail: gus@swiftuganda.com Librairie du Liban Pak Book Corporation Fax: (27 21) 595 4430 P.O. Box 11-9232 Aziz Chambers 21 E-mail: oxford@oup.co.za UKRAINE Beirut Queen’s Road LIBRA Publishing House Tel: (961 9) 217 944 Lahore For subscription orders: Ms. Sophia Ghemborovskaya Fax: (961 9) 217 434 Tel: (92 42) 636 3222; 636 0885 International Subscription Service 53/80 Saksahanskoho Str. E-mail: hsageh@cyberia.net.lb Fax: (92 42) 636 2328 P.O. Box 41095 252033, Kiev 33 URL: www.librairie-du-liban.com.lb E-mail: pbc@brain.net.pk Craighall Tel: (7 44) 227 62 77 Johannesburg 2024 Fax: (7 44) 227 62 77 MALAYSIA Mirza Book Agency Tel: (27 11) 880 1448 University of Malaya Cooperative 65, Shahrah-e-Quaid-e-Azam Fax: (27 11) 880 6248 UNITED KINGDOM Bookshop, Limited Lahore 54000 E-mail: iss@is.co.za Microinfo Ltd. P.O. Box 1127, Jalan Pantai Baru Tel: (92 42) 7353601 P.O. Box 3, Omega Park 59700 Kuala Lumpur Fax: (92 42) 576 3714 SPAIN Alton Tel: (60 3) 7565 000 E-mail: merchant@brain.net.pk Mundi-Prensa Libros, s.a. Hampshire GU34 2 PG Fax: (60 3) 755 4424 Castello 37 Tel: (44 1420) 86 848 E-mail: umkoop@tm.net.my PERU 28001 Madrid Fax: 44 1420) 89 889 Editorial Desarrollo SA Tel: (34 91) 436 37 00 E-mail: wbank@ukminfo.demon.co.uk MEXICO Apartado 3824 Fax: (34 91) 575 39 98 URL: www.microinfo.co.uk INFOTEC Ica 242, OF. 106 E-mail: libreria@mundiprensa.es Av. San Fernando No. 37 Lima 1 URL: www.mundiprensa.es The Stationery Office Col. Toriello Guerra Tel: (51 14) 285 380 51 Nine Elms Lane 14050 Mexico D.F. Fax: (51 14) 286 628 Mundi-Prensa Barcelona London SW8 5DR Tel: (52 5) 624 2800 Consell de Cent No. 391 Tel: (44 171) 873-8372 Fax: (52 5) 624 2822 PHILIPPINES 08009 Barcelona Fax: (44 171) 873-8242 E-mail: infotec@rtn.net.mx International Booksource Center, Inc. Tel: (34 3) 488 3492 E-mail: chris.allen@theso.co.uk URL: www.rtn.net.mx 1127-A Antipolo St. Fax: (34 3) 487 7659 URL: www.tsonline.co.uk Barangay, Venezuela E-mail: barcelona@mundiprensa.es Mundi-Prensa Mexico, S.A. de C.V. Makati City VENEZUELA c/Rio Panuco, 141 - Colonia Cuauhtemoc Tel: (63 2) 896 6501 SRI LANKA, THE MALDIVES Tecni-Ciencia Libros, S.A. 06500 Mexico DF Fax: (63 2) 896 6497 Lake House Bookshop Sr. Luis Fernando Ramirez, Director Tel: (52 5) 533 56 58 P.O. Box 244 Centro Cuidad Comercial Tamanaco Fax: (52 5) 514 67 99 POLAND 100, Sir Chittampalam Gardiner Mawatha Nivel C-2 E-mail: 1015245.2361@compuserve.com International Publishing Service Colombo 2 Caracas Ul. Piekna 31/37 Tel: (94 1) 32 104 Tel: (58 2) 959 5547 NEPAL 00 677 Warsaw Fax: (94 1) 432 104 Fax: (58 2) 959 5636 Everest Media International Services (P.) Tel: (48 2) 628 6089 E-mail: LHL@sri.lanka.net E-mail: lfrg001@ibm.net Ltd. Fax: (48 2) 621 7255 GPO Box 5443 E-mail: books%ips@ikp.atm.com.pl SWEDEN ZAMBIA Kathmandu URL: www.ipscg.waw.pl/ips/export Wennergren-Williams Informations Service University Bookshop, University of Zambia Tel: (977 1) 416 026 AB Great East Road Campus Fax: (977 1) 224 431 PORTUGAL P.O. Box 1305 P.O. Box 32379 Livraria Portugal S-171 25 Solna Lusaka NETHERLANDS Apartado 2681 Tel: (46 8) 705 9750 Tel: (260 1) 252576 De Lindeboom/Internationale Rua Do Carmo 70-74 Fax: (46 8) 27 0071 Fax: (260 1) 253952 Publikaties b.v. 1200 Lisbon E-mail: mail@wwi.se E-mail: hunene@admin.unza.zm M.A. de Ruyterstraat 20A Tel: (351 1) 347 4982 7482 BZ Haaksbergen Fax: (351 1) 347 0264 SWITZERLAND ZIMBABWE Tel: (31 53) 574 0004 Librarie Payot S.A. Academic and Baobab Books (Pvt.) Ltd. Fax: (31 53) 572 9296 ROMANIA Service Institutionnel 4 Conald Road E-mail: lindeboo@worldonline.nl Compani De Librarii Bucuresti s.a. Côtes-de-Montbenon 30 Graniteside URL: home.worldonline.nl/~lindeboo Str. Lipscani nr. 26, sector 3 1002 Lausanne P.O. Box 567 Bucharest Tel: (41 21) 341 3229 Harare NEW ZEALAND Tel: (40 1) 313 9645 Fax: (41 21) 341 3235 Tel: (263 4) 755 035 EBSCO NZ Ltd. Fax: (40 1) 312 4000 E-mail: institutionnel@payot-libraire.ch Fax: (263 4 ) 759 052 Private Mail Bag 99914 E-mail: Academic@Africaonline.Co.Zw New Market RUSSIAN FEDERATION ADECO Van Diermen Editions Techniques Auckland Izdatelstvo << Ves Mir >> Ch. de Lacuez 41 Tel: (64 9) 524 8119 Moscow 101831 CH-1807 Blonay Fax: (64 9) 524 8067 Tel: (7 95)917 8749 Tel: (41 21) 943 2673 E-mail: WGent%ess-nz.ebsco@iss. Fax: (7 95)917 9259 Fax: (41 21) 943 3605 ebsco.com E-mail: ozimarin@glasnet.ru URL: www.vesmir.tsx.org (c) The International Bank for Reconstruction and Development / The World Bank World Development Report Entering the 21st Century The theme of this 22nd edition of the World Development global environmental agreements, the pace of regional Report is the changing development landscape in the growth and the scope of urban development. 21st century. The Report proposes new rules The Report focuses on two forces 200 5000 and structures on which to build a of change: the integration of the 4500 comprehensive approach to devel- world economy and the increasing 160 4000 opment policy in the 21st century. demand for self-government which 3500 The challenges remain great, but will affect responses to key issues the opportunities available in the Billions in US Dollars 120 Number of Countries 3000 such as poverty reduction, climate new century hold out prospects 2500 change, and water scarcity. These for a better future. The World tidal forces of globalization and 80 2000 Development Report 1999/2000 localization will require nation 1500 provides invaluable guidance for states to sustain a dynamic equilibri- 40 1000 decisionmakers in the next century. um with international and subna- 500 It gives an in-depth analysis of tional partners. The nature of this evolving development concerns 0 0 equilibrium will have far reaching 1974 1990 1991 1992 1993 1994 1995 1996 1997 and explains how current views implications for the gains from trade Total Number Total Share Gross Private must adapt to meet the challenges of Countries Democratic Capital Flows and capital flows, the fruitfulness of of the 21st century. Use this convenient order form to reserve your copy of the World Development Report 1999/2000-Entering the 21st Century: The Changing Development Landscape. You may also call 703-661-1580 or 800-645-7247 or e-mail books@worldbank.org I YES, please send my copy of World Development Title Stock # Price Qty. Total US$ Report 1999/2000-Entering the 21st Century: World Development Report 1999/2000- D61125 $50 The Changing Development Landscape. Entering the 21st Century: The Changing Development Landscape Hardcover. (ISBN 0-19-521125-1) (Please Print) World Development Report 1999/2000- D61124 $26 Entering the 21st Century: The Changing Development Landscape Name ______________________________________ Paperback. (ISBN 0-19-521124-3) * Shipping and Handling charges are $8.00 per order. If a purchase order Subtotal Title ________________________________________ is used, actual shipping will be charged. Shipping and handling* Organization _________________________________ For air mail delivery outside the United States, charges are $13.00 for one item plus $6.00 for each additional item. Total US$ Address_____________________________________ ___________________________________________ PAYMENT METHOD Expiration date______________/_______________ Orders from individuals must be accompanied by City ________________________________________ payment or credit card information. Name ____________________________________ Credit cards accepted only for orders addressed to as it appears on the card (required for all credit card charges) State _______________________________________ the World Bank. Please do not send cash. Signature __________________________________ I Check no. _______________________________ Zip/Postal Code_______________________________ in the amount of $__________________ is Institutional customers in the U.S. only: enclosed. When ordering directly from the World I Bill me. Please include purchase order. Country _____________________________________ Bank, make check payable in U.S. funds drawn on a U.S. bank to: The World Bank. Please send your Mail order to: Phone ______________________________________ check with your order. World Bank Publications P.O. Box 960, Herndon, VA 20172-0960, USA, I Charge $__________________ to my: or Fax to 703-661-1501. Fax ________________________________________ I American Express I Mastercard I Visa Order by phone: 703-661-1580 or 800-645-7247. Order online: www.worldbank.org/publications E-mail ______________________________________ I Card no. ________________________________ Questions? E-mail us at books@worldbank.org World Bank Publications Visit our website at www.worldbank.org/publications (c) The International Bank for Reconstruction and Development / The World Bank (c) The International Bank for Reconstruction and Development / The World Bank G lobalization and localization—the integration of the world economy and the increasing demand for local autonomy—are two of the most important forces shaping development as we enter the 21st century. These forces will affect trade, capital flows, the global environment, decentralization, and urban development, and will require countries to seek a dynamic equilibrium at both international and subnational levels. This 22nd edition of the World Development Report analyzes how these forces could reshape the international landscape in the new millennium. It proposes rules and structures on which to build a more effective, comprehensive approach to development; provides valuable insight into how current viewpoints can be adapted to fit evolving development con- cerns; and offers guidance for decisionmakers, researchers, and others with an interest in development. World Development Report 1999/2000 also includes Selected World Development Indicators, an essential refer- ence on recent trends in development. 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. O Telephone: 202 477 1234 R Facsimile: 202 477 6391 L Telex: MCI 64145 WORLDBANK D MCI 248423 WORLDBANK Internet: www.worldbank.org B E-mail: books@worldbank.org A N 0 K 200 Our dream is a (c) The International Bank for Reconstruction and Development / The World Bank