17300 IN A CHANGING WORLD LISHED THE WORLD BANK XF0RD RSITY PRESS Oxford University Press OXFORD NEW YORK TORONTO DELHI BOMBAY CALCUTtA MADPAS KARACHI KUALA LUMPUR SINGAPORE HONG KONG TOKYO NAIROBI DAR ES SALAAM CAPE TOWN MELBOURNE AUCKLAND and associated companies in BERLIN IBADAN © 1997 The International Bank for Reconstruction and Development I 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. Interior design and typesetting by Barton Matheson Willse & Worthington. Manufactured in the United States of America First printing June 1997 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 0-19-521115-4 clothbound ISBN 0-19-521114-6 paperback ISSN 0 163-5085 S Text printed on recycled paper that conforms to the American Standard for Permanence of Paper for Printed Library Material Z39.48-1984 FOREWORD DEVELOPMENT REPORT 1997, THE TWENTI- and technical inputs. It is also about the underlying, insti- WORLD eth in this annual series, is devoted to the role and tutional environment: the rules and customs that deter- effectiveness of the state: what the state should do, how it mine how those inputs are used. As this Report shows, should do it, and how it can do it better in a rapidly understanding the role the state plays in this environ- changing world. mentfor example, its ability to enforce the rule of law to These issues are high on the agenda in developing and underpin market transactionswill be essential to mak- industrial countries alike. For many, the lesson of recent ing the state contribute more effectively to development. years has been that the state could not deliver on its The pathways to an effective state are many and varied. promises: transition economies have had to make a wrench- The Report makes no attempt to offer a single recipe for ing shift toward the market economy, and much of state reforms worldwide. It does, however, provide a the developing world has had to face up to the failure of framework for guiding these efforts, in the form of a two- state-dominated development strategies. Even the mixed part strategy: economies of the industrialized world, in response to the failures of government intervention, have opted for a de- First, focus the state's activities to match its capability. cided shift in the mix in favor of market mechanisms. Many states try to do too much with few resources and Many have felt that the logical end point of all these little capability. Getting governments better focused on reforms was a minimalist state. Such a state would do no the core public activities that are crucial to develop- harm, but neither could it do much good. ment will enhance their effectiveness. The Report explains why this extreme view is at odds Second, over time, look for ways to improve the state's with the evidence of the world's development success capability by reinvigorating public institutions. The stories, be it the development of today's industrial eco- Report puts particular emphasis on mechanisms that nomies in the nineteenth century or the postwar growth give public officials the incentive to do their jobs better "miracles" of East Asia. Far from supporting a minimalist and to be more flexible, but that also provide restraints approach to the state, these examples have shown that to check arbitrary and corrupt behavior. development requires an effective state, one that plays a catalytic, facilitating role, encouraging and complement- Drawing from examples of successful and unsuccessful ing the activities of private businesses and individuals. states, and of state reform around the world, the Report Certainly, state-dominated development has failed. But so elaborates this two-part strategy and shows how it can be has stateless developmenta message that comes through implemented from a number of starting positions. Signif- all too clearly in the agonies of people in collapsed states icantly, although there is an enormous diversity of settings such as Liberia and Somalia. History has repeatedly and contexts, effective states clearly do have some com- shown that good government is not a luxury but a vital mon features. One is in the way government has set rules necessity. Without an effective state, sustainable develop- underpinning private transactions, and civil society more ment, both economic and social, is impossible. broadly. Another is in the way government has played by History and recent experience have also taught us that the rules itself, acting reliably and predictably and con- development is not just about getting the right economic trolling corruption. Building a more effective state to support sustainable the challenge for states is neither to shrink into insignifI- development and the reduction of poverty will not be cance, nor to dominate markets, but to start taking those easy. In any situation many people will have a vested small steps. interest in keeping the state as it is, however costly the results for the well-being of the country as a whole. Over- coming their opposition will take time and political effort. But the Report shows how opportunities for reform can open, and widen, with the help of careful sequencing of reforms and mechanisms to compensate losers. Even in James D. Wolfensohn the worst of situations, very small steps toward a more President effective state can have a large impact on economic and The World Bank social welfare. As we approach the twenty-first century, May 30, 1997 This Report has been prepared by a team led by Ajay Chhibber and comprising Simon Commander, Alison Evans, Harald Fuhr, Cheikh Kane, Chad Leechor, Brian Levy, Sanjay Pradhan, and Beatrice Weder. Valuable contributions were made by Jean-Paul Azam, Ed Campos, Hamid Davoodi, Kathleen Newland, Kenichi Ohno, Dani Rodrik, Susan Rose-Ackerman, Astri Suhrke, and Douglas Webb. The team was assisted by Ritu Basu, Gregory Kisunko, Une Lee, Claudia Sepulveda, and Adam Michael Smith. Stephanie Flanders was the principal editor. The work was carried out under the general direction of the late Michael Bruno, Lyn Squire, and Joseph Stiglitz. The team received useful advice from a distinguished panel of external experts comprising Masahiko Aoki, Ela Bhatt, Kwesi Botchwey, Peter Evans, Atul Kohli, Klaus Konig, Seymour Martin Lipset, Douglass North, Emma Rothschild, Graham Scott, and Vito Tanzi. Many others inside and outside the World Bank provided helpful comments, wrote background papers and other contributions, and participated in consultation meetings. These contributors and participants are listed in the Bibliographical Note. The International Economics Department contributed to the data appendix and was responsible for the Selected World Development Indicators. The production staff of the Report included Amy Brooks, Valerie Chisholm, Kathryn Kline Dahi, Joyce Gates, Stephanie Gerard, Jeffrey N, Lecksell, and Michael Treadway. Rebecca Sugui served as executive assistant to the team, and Daniel Atchison, Elizabete de Lima, Michael Geller, and Thomas Zorab as staff assistants. Maria Ameal served as administrative officer. This Report is dedicated to the memory of Michael Bruno, Senior Vice President and Chief Economist of the World Bank from 1993 to 1996, whose life's work, including his contributions to this and past editions of World Development Report, immeasurably advanced our understanding of development. Iv CONTENTS OVERVIEW 1 PART ONE RETHINKING THE STATE-THE WORLD OVER 1 The Evolving Role of the State 19 2 Refocusing on the Effectiveness of the State 29 PART TWO MATCHING ROLE TO CAPABILITY 3 Securing the Economic and Social Fundamentals 41 4 Fostering Markets: Liberalization, Regulation, and Industrial Policy 61 PART THREE REINVIGORATING INSTITUTIONAL CAPABILITY 5 Building Institutions for a Capable Public Sector 79 6 Restraining Arbitrary State Action and Corruption 99 7 Bringing the State Closer to People 110 8 Facilitating International Collective Action 131 FART FOUR REMOVING OBSTACLES TO CHANGE 9 The Challenge of Initiating and Sustaining Reforms 145 10 The Agenda for Change 157 Technical Note 168 Bibliographical Note 176 Appendix: Selected Indicators on Public Finance 195 SELECTED WORLD DEVELOPMENT INDICATORS 203 BOXES 1 The pathway to a more effective state 3 2 Credibility, investment, and growth 5 3 The regional agenda 14 VI WORLD DEVELOPMENT REPORT 1997 1.1 State and government: Some concepts 20 1.2 U.S. government action to support market development: Some examples 21 1.3 Evolution of the role of the state in India: The past fifty years 24 1.4 The economic rationale for state intervention and some definitions 26 2.1 Building the Internet: A contemporary example of fruitful public-private interaction 31 2.2 Measuring the state-its size, its policies, and its institutional capability 33 3.1 Weaknesses in fundamentals constrain firms the world over 42 3.2 Standing up to crime in Cali, Colombia 44 3.3 Contracting and the judicial system in Brazil 46 3.4 International track records on fiscal deficits and inflation 47 3.5 Commitment versus flexibility in the CFA zone 52 3.6 Private provision of social services: A historical perspective 54 3.7 The new Chilean unemployment insurance scheme 58 3.8 Reducing poverty in Indonesia-how social assistance complements broad-based growth 60 4.1 Mexico's deregulation czar 63 4.2 Six objections to privatization-and how to address them 64 4.3 How government supervision averted financial disaster in Malaysia 67 4.4 Telecommunications regulation in Jamaica 70 4.5 Environmental activism in Yokohama, Japan 71 4.6 Japan's postwar big push in metals industries 73 5.1 Laying bureaucratic foundations: The Northcote-Trevelyan reforms in the United Kingdom 80 5.2 Australia's mechanisms for transparent, competitive, and results-oriented policymaking 82 5.3 The mushrooming of Bangladesh's government 86 5.4 Vouchers and school choice 89 5.5 Contracting with NGOs for better schooling in Bolivia 90 5.6 Cultivating the best and the brightest: Mandarin versus open systems 94 5.7 Building worker dedication: Good government in Brazil's Ceará State 97 6.1 How popular participation improved property rights and dispute resolution in Peru 101 6.2 Urban political machines in the United States and their reform 105 6.3 Fighting corruption in Uganda 106 6.4 Hong Kong's independent commission against corruption 107 7.1 Public opinion and the state 111 7.2 Managing multiethnic societies in Malaysia and Mauritius 113 7.3 Does social capital matter? 115 7.4 Client surveys to motivate service improvements in India, Uganda, and Nicaragua 118 7.5 Does participation improve project performance? 119 7.6 Pitfalls in intergovernmental relations: The experiences of Brazil and China 125 7.7 Calculating fiscal equalization grants 127 8.1 The World Trade Organization-an international mechanism for bringing credibility to national policy 134 8.2 How international agricultural research benefits donors as well 137 8.3 The challenges of global climate change for international cooperation 138 8.4 Sharing the burden of environmental protection 139 8.5 How large the global peace dividend? 140 CONTENTS VII 9.1 Weighing the political costs and benefits of reform 146 9.2 The predatory state under the Duvalier dynasty in Haiti 149 9.3 Reform under external threat: The Meiji restoration in Japan 150 9.4 The Moncloa Pacts in Spain 155 9.5 Venezuela's 1989 reform program and its reversal 156 10.1 State collapse and beyond in Somalia 159 10.2 The economic underpinnings of conflict: The case of Liberia 160 TEXT FIGURES 1 The state has grown everywhere 2 2 A range of mechanisms can enhance state capability 7 3 Factors associated with corruption 8 4 Higher government employment often means lower government wages 10 5 Countries with good economic policies and stronger institutional capability grow faster 13 1.1 One world, many more states 21 1.2 Governments the world over have expanded since 1960 22 1.3 Transfers and interest payments have grown 23 2.1 The state, institutions, and economic outcomes 30 2.2 Good government helps explain the income gap between East Asia and Africa 32 2.3 Reliable institutions make for credible states 35 2.4 Perceived credibility and economic performance go hand in hand 37 3.1 The lawlessness syndrome 43 3.2 Negotiating with government officials can be arduous 43 3.3 Countries are discovering the advantages of the value added tax 48 3.4 Unlike Nigeria, Indonesia managed its recent oil windfall prudently 49 3.5 Well-designed budgetary institutions help avoid large deficits 51 3.6 In Vietnam, the benefits of hospital care are skewed toward the better-off 53 3.7 The balance of private and public education differs enormously worldwide 55 3.8 Pensions and other transfers have crept upward in the industrial countries 57 3.9 Housing subsidies in developing countries mostly fail to reach the needy 59 4.1 Bank crises are all too common and carry enormous fiscal cost 68 5.1 Guinea's policy goals and spending allocations do not add up 84 5.2 Decisionmaking in Ukraine is bogged down by overlapping responsibilities 85 5.3 Most firms rate government services poorly, but some services score higher than others 86 5.4 Three strategies for improving government service delivery 87 5.5 Recruiting and promoting on merit improve bureaucratic capability 92 5.6 Lack of meritocracy and poor pay in the Philippines' civil service have lowered capability 93 5.7 In Africa, as public employment has risen, wages have fallen 95 6.1 High and unpredictable corruption hurts investment 103 6.2 Some factors associated with corruption 104 7.1 The world has become much more democratic since 1980 112 7.2 Organizations at the interface of state, markets, and civil society 116 7.3 Vertical rules and horizontal incentives shape local government's capability 123 8.1 Many countries are loosening restraints on international capital 135 VIII WORLD DEVELOPMENT REPORT 1997 8.2 Refugees have been flooding Africa, Asia, and Europe 141 8.3 Poor policies nullify the effect of aid 142 9.1 Older workers will lose from reforming pensions, but the young will gain 147 9.2 Multiple veto points help countries resist pressure to expand welfare 148 TEXT TABLES 1.1 Functions of the state 27 3.1 Social insurance, social assistance, and poverty-targeted programs in developing countries: Characteristics and lessons 56 3.2 Implicit pension debt in selected countries 57 4. 1 Estimates of welfare gains from deregulation in the United States 62 4.2 The variety of regulatory experience 67 5.1 Mechanisms to improve service delivery 88 7.1 Changes in subnational finance in selected countries 121 7.2 Demand and supply characteristics of local and national public goods 122 7.3 Possible tax and expenditure assignments by level of government 128 7.4 Principles and best practices in grant design 129 7.5 Matching decentralization strategy to government capacity 130 9.1 Alignment of interest groups, political costs, and tactical sequencing of reform by reform type 145 9.2 Estimated efficiency gains from privatizing utilities in Argentina 147 9.3 First- and second-generation reforms 152 Definitions and data notes The countries included in regional and income Dollar figures are current U.S. dollars except groupings used in this Report (except those for the where otherwise specified. private sector survey) are listed in the Classification of Billion means 1,000 million; trillion means 1,000 Economies tables at the end of the Selected World billion. Development Indicators. Income classifications are based on GNP per capita; thresholds for income clas- The following abbreviations are used: sifications in this edition may be found in the Intro- duction to Selected World Development Indicators. CEE Central and Eastern Europe Group averages reported in the figures and tables are C's Commonwealth of Independent States unweighted averages of the countries in the group GDP Gross domestic product except where noted to the contrary. GNP Gross national product The use of the term "countries" to refer to eco- IMF International Monetary Fund nomies implies no judgment by the World Bank NGO Nongovernmental organization about the legal or other status of a territory. Statistics OECD Organization for Economic Cooperation reported for "developing countries" include eco- and Development nomies in transition from central planning except PPp Purchasing power parity where noted to the contrary. THE GLOBE, THE STATE IS IN THE SPOT- of experience since then is rather different: that the state is AROUND light. Far-reaching developments in the global eco- central to economic and social development, not as a nomy have us revisiting basic questions about government: direct provider of growth but as a partner, catalyst, and what its role should be, what it can and cannot do, and facilitator. how best to do it. What makes for an effective state differs enormously The last fifty years have shown clearly both the benefits across countries at different stages of development. What and the limitations of state action, especially in the pro- works in the Netherlands or New Zealand, say, may not motion of development. Governments have helped to work in Nepal. Even among countries at the same level of deliver substantial improvements in education and health income, differences in size, ethnic makeup, culture, and and reductions in social inequality. But government political systems make every state unique. But this very actions have also led to some very poor outcomes. And diversity enriches this Report's inquiry into why and how even where governments have done a good job in the past, some states do better than others at sustaining develop- many worry that they will not be able to adapt to the ment, eradicating poverty, and responding to change. demands of a globalizing world economy. Rethinking the state-the world over The new worries and questions about the state's role are many and various, but four recent developments have The world is changing, and with it our ideas about the given them particular impetus: state's role in economic and social development. Today's intense focus on the state's role is reminiscent of an earlier The collapse of command-and-control economies in the era, when the world was emerging from the ravages of former Soviet Union and Central and Eastern Europe World War II, and much of the developing world was just The fiscal crisis of the welfare state in most of the estab- gaining its independence. Then development seemed a lished industrial countries more easily surmountableand largely technicalchal- The important role of the state in the "miracle" eco- lenge. Good advisers and technical experts would formu- nomies of East Asia late good policies, which good governments would then The collapse of states and the explosion in humanitar- implement for the good of society. State-led intervention ian emergencies in several parts of the world. emphasized market failures and accorded the state a cen- tral role in correcting them. But the institutional assump- This Report shows that the determining factor behind tions implicit in this world view were, as we all realize these contrasting developments is the effectiveness of the today, too simplistic. Flexibility to implement the policies state. An effective state is vital for the provision of the devised by technocrats was accorded pride of place. goods and servicesand the rules and institutionsthat Accountability through checks and balances was regarded allow markets to flourish and people to lead healthier, as an encumbrance. happier lives. Without it, sustainable development, both In a few countries things have indeed worked out more economic and social, is impossible. Many said much the or less as the technocrats expected. But in many countries same thing fifty years ago, but then they tended to mean outcomes were very different. Governments embarked on that development had to be state-provided. The message fanciful schemes. Private investors, lacking confidence in 2 WORLD DEVELOPMENT REPORT 1997 public policies or in the steadfastness of leaders, held back. veloping world embraced state-dominated development Powerful rulers acted arbitrarily. Corruption became strategies. The result was a tremendous expansion in the endemic. Development faltered, and poverty endured. size and reach of government worldwide. State spending Over the last century the size and scope of government now constitutes almost half of total income in the estab- have expanded enormously, particularly in the industrial lished industrial countries, and around a quarter in devel- countries (Figure 1). The pre-World War II expansion was oping countries. But this very increase in the state's influ- driven by, among other factors, the need to address the ence has also shifted the emphasis from the quantitative to heavy toll on economic and social systems brought on by the qualitative, from the sheer size of the state and the the Great Depression. The postwar confidence in govern- scope of its interventions to its effectiveness in meeting ment bred demands for it to do more. Industrial eco- people's needs. nomies expanded the welfare state, and much of the de- As in the 1940s, today's renewed focus on the state's role has been inspired by dramatic events in the global economy, which have fundamentally changed the environ- Figure 1 The state has grown everywhere ment in which states operate. The global integration of economies and the spread of democracy have narrowed the Total government expenditure scope for arbitrary and capricious behavior. Taxes, invest- (percentage of GDP) ment rules, and economic policies must be ever more 50 responsive to the parameters of a globalized world econ- omy. Technological change has opened new opportunities 40 for unbundling services and allowing a larger role for mar- kets. These changes have meant new and different roles for 30 governmentno longer as sole provider but as facilitator and regulator. States have come under pressure even where 20 governments have previously seemed to perform well. Many industrial countries find themselves grappling with a welfare state that has grown unwieldy, and having to make 10 difficult choices about the services and benefits that people should expect government to provide. Marketsdomestic 0 About 1913 1920 1937 1960 1980 1990 1995 and globaland citizens vexed by state weaknesses have 1870 come to insist, often through grassroots and other non- OECD countries governmental organizations, on transparency in the con- duct of government, and on other changes to strengthen Central government expenditure the ability of the state to meet its assigned objectives. (percentage of GDP) The clamor for greater government effectiveness has 50 reached crisis proportions in many developing countries where the state has failed to deliver even such fundamental 40 public goods as property rights, roads, and basic health and education. There a vicious circle has taken hold: people 30 and businesses respond to deteriorating public services by avoiding taxation, which leads to further deterioration in 20 services. In the former Soviet Union and Central and East- ern Europe it was the state's long-term failure to deliver on 10 0 uIIIE 1960 1965 1970 1975 1980 1985 1990 its promises that led, finally, to its overthrow. But the col- lapse of central planning has created problems of its own. In the resulting vacuum, citizens are sometimes deprived of basic public goods such as law and order. At the limit, as in Afghanistan, Liberia, and Somalia, the state has some- Developing countries times crumbled entirely, leaving individuals and interna- tional agencies trying desperately to pick up the pieces. Note: Data for the OECD countries are for central and local government, including social security expenditures. See the A two-part strategy Technical Note for details. Source: Tanzi and Schuknecht 1995; OECD, various years; IMF, various years (b). How can we cut through the maze of questions and pres- sures now facing the world's states? No one-size-fits-all OVERVIEW 3 recipe for an effective state is suggested here. The range of differences among states is too enormous, as are their start- Box I The pathway to a more effective state ing points. Rather this Report provides a broad framework for addressing the issue of the state's effectiveness world- A more capable state can be a more effective state, wide. It points to a number of ways to narrow the grow- but effectiveness and capability are not the same ing gap between the demands on states and their capabil- thing. Capability, as applied to states, is the ability ity to meet those demands. Getting societies to accept a to undertake and promote collective actions effi- redefinition of the state's responsibilities will be one part cientlysuch as law and order, public health, and of the solution. This will include strategic selection of the basic infrastructure; effectiveness is a result of using collective actions that states will try to promote, coupled that capability to meet society's demand for those with greater efforts to take the burden off the state, by goods. A state may be capable but not very effective involving citizens and communities in the delivery of core if its capability is not used in society's interest. collective goods. The path to a more effective state, although not But reducing or diluting the state's role cannot be the linear, is likely to be a two-stage process. First, the end of the reform story. Even with more selectivity and state must focus what capability it has on those tasks greater reliance on the citizenry and on private firms, that it can and should undertake. As it does this, it meeting a broad range of collective needs more effectively can then focus on building additional capability. As will still mean making the state's central institutions work the figure illustrates, countries in Zone I pursue a better. For human welfare to be advanced, the state's broad range of activities in an unfocused manner capabilitydefined as the ability to undertake and promote despite little state capability, and their efforts prove collective actions efficientlymust be increased. ineffective. But countries cannot move to Zone III This basic message translates into a two-part strategy to overnightbuilding capability takes time. The path- make every state a more credible, effective partner in its way to greater effectiveness leads, first, to focusing on country's development: fundamental tasks and leveraging the state's limited capability through partnerships with the business Ivlatching the state's role to its capability is the first ele- community and civil society (Zone II). Countries ment in this strategy. Where state capability is weak, then can move gradually to Zone III by strengthen- how the state intervenesand whereshould be care- ing their capability over time. fully assessed. Many states try to do too much with few resources and little capability, and often do more harm than good. A sharper focus on the fundamentals would improve effectiveness (Box 1). But here it is a matter not just of choosing what to do and what not to do More effective but of how to do it as well. But capability is not destiny. Therefore the second ele- A ment of the strategy is to raise state capability by rein- vigorating public institutions. This means designing effective rules and restraints, to check arbitrary state actions and combat entrenched corruption. It means subjecting state institutions to greater competition, to increase their efficiency. It means increasing the perfor- mance of state institutions, improving pay and incen- Less tives. And it means making the state more responsive to effective people's needs, bringing government closer to the peo- Focused More ple through broader participation and decentralization. capable Unfocused Less Thus, the Report not only directs attention to refocus- capable ing the state's role, but also shows how countries might begin a process of rebuilding the state's capability. Matching role to capability do and what not to do is critical. But this also involves Matching role to capability is not a simple message of dis- choosing how to do thingshow to deliver basic ser- mantling the state. In some areas much greater focus is vices, provide infrastructure, regulate the economyand badly needed to improve effectiveness: choosing what to not just whether to do them at all. The choices here are 4 WORLD DEVELOPMENT REPORT 1997 many and must be tailored to the circumstances of each graphic areas. Marginalized from public discussion and country. excluded from the broader economy and society, such groups are fertile ground for violence and instability, as The first job of states: Getting the fundamentals right many parts of the world are increasingly learning. Five fundamental tasks lie at the core of every govern- Public policies and programs must aim not merely to ment's mission, without which sustainable, shared, deliver growth but to ensure that the benefits of market- poverty-reducing development is impossible: led growth are shared, particularly through investments in basic education and health. They must also ensure that Establishing a foundation of law people are protected against material and personal insecu- Maintaining a nondistortionary policy environment, rity. Where poverty and economic marginalization stem including macroeconomic stability from ethnic and social differences, policies must be care- Investing in basic social services and infrastructure fully crafted to manage these differences, as Malaysia and Protecting the vulnerable Mauritius have done. Protecting the environment. Government regulation is not the only answer to pollu- tion, An expanding toolkit of innovative and flexible Although the importance of these fundamentals has long incentives is now available to get polluters to clean up their been widely accepted, some new insights are emerging as to act. Although there is no substitute for meaningful regula- the appropriate mix of market and government activities in tory frameworks and information about the environment, achieving them. Most importanr, we now see that markets these new tools, which rely on persuasion, social pressure, and governments are complementary: the state is essential and market forces to help push for improved environmen- for putting in place the appropriate institutional founda- tal performance, can often succeed where regulation can- tions for markets. And government's credibilitythe pre- not. Countries are using some of these tools, with promis- dictability of its rules and policies and the consistency with ing results, in four areas: which they are appliedcan be as important for attracting private investment as the content of those rules and policies. Harnessing the power of public opinion A survey, specially commissioned for this Report, of Making regulation more flexible domestic entrepreneurs (formal and informal) in sixty- Applying self-regulatory mechanisms nine countries confirms what was already known anecdo- Choosing effective market-based instruments. tally: that many countries lack the basic institutional foundations for market development (Box 2). High levels Going beyond the basics.' The state need not be the of crime and personal violence and an unpredictable judi- sole provider ciary combine to produce what this Report defines as the There is a growing recognition that in many countries "lawlessness syndrome." Weak and arbitrary state institu- monopoly public providers of infrastructure, social services, tions often compound the problem with unpredictable, and other goods and services are unlikely to do a good job. inconsistent behavior. Far from assisting the growth of At the same time, technological and organizational innova- markets, such actions squander the state's credibility and tions have created new opportunities for competitive, pri- hurt market development. vate providers in activities hitherto confined to the public To make development stable and sustainable, the state sector. To take advantage of these new opportunitiesand has to keep its eye on the social fundamentals. Lawlessness better allocate scarce public capabilitygovernments are is often related to a sense of marginalization: indeed, beginning to separate the financing of infrastructure and breaking the law can seem the only way for the marginal- services from its delivery, and to unbundle the competitive ized to get their voices heard. Public policies can ensure segments of utility markets from the monopoly segments. that growth is shared and that it contributes to reducing Reformers are also moving to separate programs of social poverty and inequality, but only if governments put the insurance, designed to address the problems of health and social fundamentals high on their list of priorities. employment insecurity for all, from programs of social Too often, policies and programs divert resources and assistance, intended to help only the poorest in society. services from the people who need them most, The polit- COPING WITH HOUSEHOLD INSECURITY. It is now well ical clout of the more affluent in society sometimes leads established that the state can help households cope with governments to spend many times more on rich and mid- certain risks to their economic security: it can insure against dle-class students in universities than on basic education destitution in old age through pensions, against devastating for the majority and scholarships for the less well off. In illness through health insurance, and against job loss many regions poverty and inequality are often biased through unemployment insurance. But the idea that the against ethnic minorities or women, or disfavored geo- state alone must carry this burden is changing. Even in OVERVIEW 5 Box 2 Credibility, investment, and growth A survey of local entrepreneurs in sixty-nine countries reliability of the institutional framework (normalized shows that many states are performing their core func- here to the high-income OECD countries) as per- tions poorly: they are failing to ensure law and order, ceived by private entrepreneurswe call it credibility. protect property, and apply rules and policies pre- The other two panels show that, once differences in in- dictably. Investors do not consider such states credible, come and education and policy distortions have been and growth and investment suffer as a consequence. controlled for, there is a strong correlation between Firms were asked to rank each of several indicators countries' credibility rating and their record of growth on a scale from one (extreme problem) to six (no prob- and investment. The credibility ratings are based on lem). Averaging the answers, as the left panel does for investors' perceptions. But it is these perceptions that each world region, yields an overall indicator of the determine investment behavior. Growth of GDP per capita (percent per year) Credibility index 2.0 High-income OECD = 1.0 1.5 1.0 0.5 1.0 0.0 0.5 0.8 1.0 1.5 High Medium Low Level of credibility 0.6 Gross investment (percentage of GDP) 18 ______ 0.4 16 14 12 0.2 10 8 6 0.0 4 High- SouthMiddle CEE Latin Sub-CIS 2 income andEast and America Saharan OECD North South- and Africa 0 east Asa Africa Caribbean High Medium Low Level of credibility Note: The credibility index (left panel) is a summary indicator that combines the measures in Figure 2.3. Each bar in the two right panels is the average for a group of countries. The graphs are based on regressions for the period 1984-93 of GDP growth (thirty-two countries) and investment (thirty-three countriesi on the index, controlling for income, education, and policy distortion, South and Southeast Asia and Middle East and North Africa are each represented by only three economies. Source: World Bank staff calculations using data from the private sector survey conducted for this Report and Brunetti, Kisunko, and Weder, background papers. many industrial countries the welfare state is being re- Innovative solutions that involve businesses, labor, house- formed. Emerging economies from Brazil to China will be holds, and community groups are needed to achieve greater unable to afford even pared-down versions of the European security at lower cost. This is especially important for those system, especially with their rapidly aging populations. developing countries not yet locked into costly solutions. 6 WORLD DEVELOPMENTREPORT 1997 EFFECTIVE REGULATION. Well-designed regulatory sys- Experience has shown that the way privatization is tems can help societies influence market outcomes for managed is terribly important to the end result. The key public ends. Regulation can help protect consumers, work- factors are transparency of process, winning the acquies- ers, and the environment. It can foster competition and in- cence of employees, generating broad-based ownership, novation while constraining the abuse of monopoly power. and instituting the appropriate regulatory reform. Where Thanks to regulatory reforms initiated in the early 1980s, privatization has been managed carefully, it is already Chile's telecommunications industry has enjoyed sustained showing positive results: in Chile, for example, and the private investment, increasing service quality and com- Czech Republic. Its importance in the strategy to foster petition, and declining prices. By contrast, until some markets may vary, but for many developing countries recent reform initiatives, dysfunctional regulation led the seeking to scale back an overextended state, privatization Philippine telecommunications industrylong privately must be kept on the front burner. A carefully managed ownedto underinvest. The result was poor and often privatization process brings very positive economic and high-priced service, imposing a high cost on citizens and fiscal benefits. other firms. Making the best use of the new options emerg- ing for private provision of infrastructure and social ser- Knowing the state's limits vices will also rely, often, on a good regulatory framework. The key to predictable and consistent implementation of INDUSTRIAL POLICY. When markets are underdevel- policy is a good fit between the state's institutional capa- oped, the state can sometimes reduce coordination prob- bilities and its actions, In well-developed states, adminis- lems and gaps in information and encourage market devel- trative capability is normally strong, and institutionalized opment. Many of today's oldest industrial economies used checks and balances restrain arbitrary action, even as they various mechanisms to spur the growth of markets in their provide government organizations the flexibility to pursue early stages of development. More recently, Japan, the their mandates. By contrast, states with weaker institu- Republic of Korea, and other countries in East Asia used a tions may need to err on the side of less flexibility and variety of mechanisms for market enhancement, in addi- more restraint. This can be done in two ways: tion to securing the economic, social, and institutional fun- damentals, Sometimes these interventions were quite elab- Through self-restricting rules, which precisely specify the orate: the highly strategic use of subsidies, for example. content of pohcy and lock it into mechanisms that are Other times they were less intrusive, taking the form of costly to reverse. Regional common-currency arrange- export promotion and special infrastructure incentives. But ments, such as the CFA currency zone in francophone the ability to choose wisely among these interventions and Africa or quasi currency boards as in Argentina, are use them effectively is critical; ill-considered trade, credit, examples of such mechanisms in the field of monetary and industrial policies can and have cost countries dearly. policy. "Take-or-pay" contracts with independent power Many developing countries pursued ill-thought-out activ- producers serve a similar function in utilities regulation. ist industrial policies, with poor results. Countries that Through working in partnership with firms and citi- have pursued an activist industrial policy successfully could zens. In industrial policy, for example, states can foster not have done so without strong institutional capability. private-to-private collaboration. In financial regulation MANAGING PRIVATIZATION. Carefully designed regula- they can give bankers an incentive to operate prudently. tions and other active government initiatives can enhance And in environmental regulation they can use the the growth of markets. But in many countries this can spread of information to encourage "bottom-up" citi- take time, as private initiative is held hostage to a legacy of zen initiatives. antagonistic state-market relations. And poorly perform- ing state enterprises are often a big drain on the state's Countries in transition face a special challenge: not finances. Privatization provides an obvious solution. In only are roles changing as a result of the adoption of general it is easier to sell off state assets once a supportive market-based systems; so are capabilities. Some transition environment for private sector development is in place. countries retain inherent capabilities in the form of quali- Economies such as China, Korea, and Taiwan (China) fled people and usable equipment, but they are not orga- have therefore opted not to give top priority to privatiza- nized to perform in their new roles. Sometimes islands of tion, but to allow the private sector to develop around the excellence are found in countries where overall effective- state sector. This option, however, may not be available ness has suffered. The task of improving effectiveness here where the fiscal burden is very high, and where the pres- is in some ways easier and in some ways more difficult: ence of poorly performing state enterprises impedes easier because capability does not start from a low base, much-needed overall restructuring of the economy. more difficult because rebuilding capability means chang- OVERVIEW 7 ing attitudes. Reform is not a matter of simply assigning that distort incentives and ultimately lead to poor out- people new responsibilities. comes. Three basic incentive mechanisms can be used, in a variety of settings, to combat these deeper problems and Reinvigorating state institutions improve capability (Figure 2): Acknowledging the state's existing, possibly meager capa- bilities does not mean accepting them for all time. The Effective rules and restraints second key task of state reform is to reinvigorate the state's Greater competitive pressure institutional capability, by providing incentives for public Increased citizen voice and partnership. officials to perform better while keeping arbitrary action in check. Effective rules and restraints Countries struggle to build the institutions for an Over the long term, building accountability generally calls effective public sector. One reason the task is so difficult for formal mechanisms of restraint, anchored in core state is political. Strong interests may develop, for example, to institutions. Power can be divided, whether among the maintain an inequitable and inefficient status quo, judicial, legislative, and executive branches of government whereas those who lose out from this arrangement may be or among central, provincial, and local authorities. The unable to exert effective pressure for change. broader the separation of powers, the greater the number But the problem of continued ineffectiveness, or of of veto points that can check arbitrary state action. But corruption, is not entirely political. Often politicians and multiple veto points are a double-edged sword: they can other public officials have strong incentives and a sincere make it as hard to change the harmful rules as the benefi- interest in improving public sector performance. But cial ones. managing a public bureaucracy is a complex business that In many developing countries legislative and judicial does not lend itself to clear, unambiguous solutions. In oversight of the executive is weak. The setting of goals and fact, building institutions for an effective public sector the links to the policies needed to achieve them are some- requires addressing a host of underlying behavioral factors times diffuse, legislatures suffer from limited information Figure 2 A range of mechanisms can enhance state capability Rules and restraints For example: Judicial independence Watchdog bodies Voice and partnership For example: ;Deliberation councilS Coproduction B WORLD DEVELOPMENT REPORT 1997 and capability, and judicial independence is compro- guarantees from international agencies, can serve as a mised. An independent judiciary is vital to ensure that the short-term substitute. legislative and executive authorities remain fully account- A major thrust of any effective strategy to reinvigorate able under the law, and to interpret and enforce the terms the public sector will be to reduce the opportunities for of a constitution. Writing laws is the easy part; they need corruption by cutting back on discretionary authority. to be enforced if a country is to enjoy the benefits of a Policies that lower controls on foreign trade, remove entry credible rule of law. These institutions of restraint take barriers for private industry, and privatize state firms in a time to establish themselves, but international commit- way that ensures competitionall of these will fight cor- ment mechanisms such as international adjudication, or ruption (Figure 3). Such reforms should not be half- Figure 3 Factors associated with corruption Index of corruption Index of corruption 5' 5 Low Medium High Low Medium High Policy distortion index Predictability of judiciary index Index of corruption Index of corruption 5 5 4 3 2 Low Medium High Low Medium High Merit-based recruitment index Ratio of civil service wages to manufacturing wages Note: Each index score is the average for a group of countries. See the Technical Note for details and definitions of the indexes. Higher values of the corruption index mean more corruption, and similarly for the other variables. The top left panel is based on a simple correlation for thirty- nine industrial and developing countries during 1984-93 (for the policy distortion index) and 1996 (for the corruption index). The top right panel is based on a regression using data from fifty-nine industrial and developing countries during 1996. The bottom left panel is based on a regression using data for thirty-five developing countries during 1970-90. The bottom right panel is based on a simple correlation for twenty industrial and developing countries in the late 1980s to the early 1990s; wage data are means. Source: World Bank staff calculations. OVERVIEW 9 hearted: reforms that open opportunities for private entry effectively and retain their competence even when all into closed sectors of the economy, but leave that entry to other institutions have declined. These agencies work well the discretion of public officials rather than establish open for all the reasons listed above. They are less subject to and competitive processes, also create enormous scope for political interference. They have limited but clear objec- corruption. Formal checks and balances can also help re- tives. They are given adequate resources and training. And duce official corruption, but they are seldom enough. their staff are usually better paid than their counterparts in Reforming the civil service, restraining political patron- other parts of government. age, and improving civil service pay have also been shown Cross-country evidence reveals that bureaucracies with to reduce corruption by giving public officials more more competitive, merit-based recruitment and promo- incentive to play by the rules. tion practices and better pay are more capable. In several Where corruption is deeply entrenched, more dramatic countries (Kenya, the Philippines) political appointments efforts will be needed to uproot it. These efforts should be run quite deep, whereas countries such as Korea have ben- focused on better monitoring of official actionboth by efited from reliance on highly competitive recruitment formal institutions and by individual citizensand pun- and a promotion system that explicitly rewards merit. ishment of wrongdoing in the courts. In Hong Kong Ongoing reforms in the Philippines are examining these (China, as of July 1, 1997), an independent commission issues in an effort to improve bureaucratic capability. By against corruption is one successful example of such an and large, countries in which broader checks and balances approach. Likewise, recent reforms in Uganda have incor- are weak need to rely more heavily on more transparent porated several elements of the anticorruption strategy and competitive systems. The experience of certain high- outlined here, with some encouraging results. The same performing East Asian economies also shows that meri- mechanisms could be applied around the globe: corrup- tocracy and long-term career rewards help build an esprit tion, despite claims to the contrary, is not culture specific. de corps, or a shared commitment to collective goals. This Reducing it will require a multipronged approach, which reduces the transactions costs of enforcing internal con- must include the private sector and civil society more straints and builds internal partnerships and loyalty. broadly. The briber has as much responsibility as the In many countries civil servants' wages have eroded as bribed; effective penalties on domestic and international a result of expanding public employment at lower skill business must be part of the solution. levels and fiscal constraints on the total wage bill (Figure 4). The result has been a significant compression of the Subjecting the state to more competition salary structure and highly uncompetitive pay for senior Governments can improve their capability and effective- officials, making it difficult to recruit and retain capable ness by introducing much greater competition in a variety staff. Some countries, such as Uganda, are undertaking of areas: in hiring and promotion, in policymaking, and far-reaching reforms to reduce overstaffing dramatically, in the way services are delivered. increase average pay, and decompress the salary structure. BOOSTING COMPETITION WITHIN THE CIVIL SERVICE. But in many countries these problems have yet to be Whether making policy, delivering services, or adminis- addressed. tering contracts, a capable, motivated staff is the lifeblood Moi COMPETITION IN THE PROVISION OF PUBLIC of an effective state. Civil servants can be motivated to GOODS AND SERVICES. In many developing countries ser- perform effectively through a combination of mechanisms vices are delivered badly or not at all. Politicians often to encourage internal competition: intervene in the day-to-day operations of public agencies, and managers have limited flexibility. There is limited A recruitment system based on merit, not favoritism accountability for results. And in many countries the pub- A merit-based internal promotion system lic sector has assumed a monopoly in delivery, eliminating Adequate compensation. pressures for better performance. Building an effective public sector in these circum- Starting in the nineteenth century, all of today's estab- stances will mean opening up core government institu- lished industrial countries used these principles to build tions, to improve incentives in areas that the public sector modern professional bureaucracies. More recently these has long monopolized. Dozens of countries through- principles have been applied in many countries in East out the Americas, Europe, and Asia have capitalized on Asia, which have transformed weak, corrupt, patronage- changes in technology and introduced competition in based bureaucracies into reasonably well functioning sys- telecommunications and electric power generation. This tems. But many developing countries do not even need to has resulted in lower unit costs and a rapid expansion of look overseas or to history for role models: they exist at service. Competition is also being enhanced by contract- home. Central banks, for example, often continue to work ing out services through competitive bids and auctions. 10 WORLD DEVELOPMENT REPORT 1997 ing public managers more flexibility will merely increase Figure 4 Higher government employment arbitrariness and corruption with no commensurate im- often means lower government wages provement in performance. And writing and enforcing contracts, particularly for complex outputs, require special- Average annual central government wages as a multiple of GDP per capita ized skills that are scarce in many developing countries. 12 These countries need first to strengthen rule-based compli- ance and financial accountability (as Argentina and Bolivia have done) within the public sector, provide greater clarity 10 of purpose and task, and introduce performance measure- ment (as in Colombia, Mexico, and Uganda). As output measurement and cx post controls on inputs are strength- 8 ened, agencies can be provided more flexibility in exchange for their greater accountability for results. 6 Bringing the state closer to people Governments are more effective when they listen to busi- 4 nesses and citizens and work in partnership with them in deciding and implementing policy. Where governments 2 lack mechanisms to listen, they are not responsive to peo- ple's interests, especially those of minorities and the poor, who usually strain to get their voices heard in the corridors 0 of power. And even the best-intentioned government is 1 2 3 4 unlikely to meet collective needs efficiently if it does not Government employment as a share know what many of those needs are. of population (percent) GIVING PEOPLE A VOICE. Partnership involves bringing Note: Data are for ninety-six industrial and developing countries the voice of the poor and of marginalized groups into the during various years in the early 1990s. See the Technical very center of the policymaking process. In many countries, Note. Source: Schiavo-Campo, de Tommaso, and Mukherjee, voice is distributed as unequally as income. Greater infor- background paper. mation and transparency are vital for informed public debate and for increasing popular trust and confidence in the statewhether in discussing expenditure priorities, This is a significant trend in industrial countries (the designing social assistance programs, or managing forests United Kingdom, Victoria State in Australia), but such and other resources. Client surveys (in India, Nicaragua, mechanisms are also being used to improve efficiency in and Tanzania) and citizen charters (in Malaysia) are pro- developing countries (for example, that of road mainte- viding new options for making voices heard. nance in Brazil). Faced with weak administrative capac- The best-established mechanism for giving citizens ity, some countries (Bolivia, Uganda) are also contracting voice is the ballot box. In 1974 only thirty-nine coun- out the delivery of social services to nongovernmental triesone in every four worldwidewere independent organizations. democracies. Today, 117 countriesnearly two of every There is a growing trend to set up focused, perfor- threeuse open elections to choose their leaders. But mance-based public agencies with more clarity of purpose periodic voting does not always mean the state is more and greater managerial accountability for outputs or out- responsive. Other mechanisms are needed to ensure that comes. New Zealand provides the most dramatic example the concerns of minorities and the poor are reflected in among the high-income countries. It broke up its con- public policies. Getting genuine intermediary organiza- glomerate ministries into focused business units, headed tions represented on policymaking councils is an impor- by chief executives on fixed-term, output-based contracts tant first step in articulating citizen interests in public with the authority to hire and fire and to bargain col- policymaking. Even more effective in local and provincial lectively. Singapore has long followed a broadly similar government, these organizations have recently become approach with its performance-based statutory boards. very active in developing countriesespecially where the Other developing countries are following suit, with state has performed poorly and where such organizations Jamaica, for example, establishing executive agencies are not suppressed. along the lines of the British model. BROADENING PARTICIPATION. Evidence is mounting But countries with inadequate controls and weak capac- that government programs work better when they seek the ity need to proceed with caution. For these countries, giv- participation of potential users, and when they tap the OVERVIEW 1 1 community's reservoir of social capital rather than work ment. The challenge is to find the right division of labor against it. The benefits show up in smoother implemen- between the center and the other tiers of government. tation, greater sustainability, and better feedback to gov- ernment agencies. Higher returns from water-borne sani- Strategic options for reform tation systems in Recife, Brazil; housing schemes for the Building a more responsive state requires working on poor in Port Elizabeth, South Africa; forest management mechanisms that increase openness and transparency, efforts in Gujarat State, India; and health care in Khar- increase incentives for participation in public affairs, and toum, Sudan, are all testament to the power of partner- where appropriate, lessen the distance between govern- shipthe participation of local people. This is in contrast ment and the citizens and communities it is intended to with top-down approaches, which often fail. serve. This yields four broad imperatives for policymakers: In successful countries policymaking has been embed- ded in consultative processes, which provide civil society, Where appropriate, ensure broad-based public discus- labor unions, and private firms opportunities for input sion of key policy directions and priorities. At a mini- and oversight. In East Asia public-private deliberation mum this includes making available information in the councilssuch as Korea's monthly export promotion public interest and establishing consultative mecha- meetings, Thailand's National Joint Public and Private nismssuch as deliberation councils and citizen com- Consultative Committee, and the Malaysian Business mitteesto gather the views and make known the pref- Councilhave provided mechanisms for feedback, infor- erences of affected groups. mation sharing, and coordination. Encourage, where feasible, the direct participation of DEVOLVING POWER, CAREFULLY. The typical developing users and other beneficiaries in the design, implementa- country has a more centralized government than the typical tion, and monitoring of local public goods and services. industrial country. But with some significant exceptions, Where decentralization is considered desirable, adopt a the past thirty years have seen a small shift in public spend- carefully staged and/or sectoral approach in priority ing power in developing countries from the national to areas. Introduce strong monitoring mechanisms and lower levels. The industrial economies have seen an oppo- make sure sound intergovernmental rules are in place to site trend, with spending power moving to the center. Nei- restrain arbitrary action at the central and the local level. ther of these observations, of course, takes into account the At the local level, focus on mechanismsand hori- decentralization implicit in recent market reforms, which zontal incentives in government's relations with the have clearly reduced the direct power and resources of cen- rest of the communitythat build accountability and tral government in a broad range of countries, competition. Decentralization is bringing many benefits in China, India, much of Latin America, and many other parts of Of course, a strategy of more openness and greater the world. It can improve the quality of government and decentralization has its dangers. The more numerous the the representation of local business and citizens' interests. opportunities for participation, the greater the demands And competition among provinces, cities, and localities that will be made on the state. This can increase the risk can spur the development of more-effective policies and of capture by vocal interest groups, or of gridlock. Bring- programs. But there are three big pitfalls to watch out for: ing government closer to some people must not result in taking it even further away from others. Equally, without Rising inequality. The gap between regions can widen clear-cut rules to impose restraints on different tiers of an issue of considerable concern in China, Russia, and government, and incentives to encourage local account- Brazil. Labor mobility provides a partial solution, but it ability, the crisis of governance that afflicts many central- is seldom easy, especially in ethnically diverse countries ized governments will simply be passed down to lower where migrants are not always welcome. levels. But there are some safe ways to start the ball rolling, Macroeconomic instability. Governments can lose con- including the use of communication and consensus build- trol of macroeconomic policy if local and regional fiscal ing to render reform intelligible to citizens and firms and indiscipline leads to frequent bailouts from the center, enhance its chances of success. as occurred in Brazil. Beyond national borders: Facilitating global Risk of local capture. A serious danger is that of local collective action governments falling under the sway of special interests, leading to misuse of resources and of the coercive Globalization is a threat to weak or capriciously governed power of the state. states. But it also opens the way for effective, disciplined states to foster development and economic well-being, These dangers show, once again, how central govern- and it sharpens the need for effective international coop- ment will always play a vital role in sustaining develop- eration in pursuit of global collective action. 12 WORLD DEVELOPMENT REPORT 1997 Embracing external competition how state policies (and international assistance) help The state still defines the policies and rules for those manage nascent conflict is needed in designing eco- within its jurisdiction, but global events and international nomic and social policy. agreements are increasingly affecting its choices. People Promoting global economic stability. Concern has been are now more mobile, more educated, and better in- growing about the potentially destabilizing effects of formed about conditions elsewhere. And involvement in large and rapid flows of portfolio capital, particularly the global economy tightens constraints on arbitrary state when a crisis in one country can spill over into other action, reduces the state's ability to tax capital, and brings markets. A variety of international mechanisms have much closer financial market scrutiny of monetary and been suggested to guard against such problems, and the fiscal policies. International Monetary Fund has recently created a "Globalization" is not yet truly globalit has yet to new facility to help members cope with sudden finan- touch a large chunk of the world economy. Roughly half cial crises. But prudent and responsive economic poli- of the developing world's people have been left out of the cies at home will be countries' best protection. Grow- much-discussed rise in the volume of international trade ing international labor mobility is also raising a host of and capital flows since the early 1980s. Governments' hes- issues requiring international collective action. itance to open up to the world economy is partly under- Protecting the environment. Urgent global environmen- standable. Joining the global economy, like devolving tal issues include climate change, loss of biodiversity, power from the center, carries risks as well as opportuni- and protection of international waters. International ties. For example, it can make countries more vulnerable collective action can help through better coordination, to external price shocks or to large, destabilizing shifts in greater public awareness, more effective technological capital flows. This makes the state's role all the more crit- transfer, and better national and local practices. ical, both in handling such shocks and in helping people Progress has been slow, however, raising the worry that and firms grasp the opportunities of the global market- it will take a major environmental catastrophe to goad place. But the difficulties should not be exaggerated, par- countries into concerted action. ticularly when laid against the risks of being left out of the Fostering basic research and the production of knowledge. globalization process altogether. Now being revitalized to meet renewed challenges in The cost of not opening up will be a widening gap in food production, the Consultative Group on Interna- living standards between those countries that have inte- tional Agricultural Research has shown how technology grated and those that remain outside. For lagging coun- can be developed and disseminated through interna- tries the route to higher incomes will lie in pursuing tional collective action. Similar consultative mecha- sound domestic policies and building the capability of the nisms need to be developed to tackle other pressing state. Integration gives powerful support to such poli- research problems in the domains of environmental ciesand increases the benefits from thembut it can- protection and health. not substitute for them. In that sense, globalization begins Making international development assistance more effec- at home. But multilateral institutions such as the World tive. To become more effective, foreign aid needs to be Trade Organization have an important role to play in pro- tied more closely to the policies of the recipient coun- viding countries with the incentive to make the leap. tries. A high priority for aid agencies is to systemati- cally channel resources to poor countries with good Promoting global collective action policies and a strong commitment to institutional rein- Global integration also gives rise to demands for states to vigoration. cooperate to combat international threats such as global warming. Economic, cultural, and other differences be- Removing obstacles to state reform tween countries can make such cooperation difficult The history of state reform in today's established indus- even, at times, impossible. But stronger cooperation is trial countries offers hopeand gives pauseto today's clearly needed for at least five major concerns that tran- developing countries. Until the last century many of the scend national borders: problems that now appear to have reduced the effective- ness of the state in the developing world were in plain Managing regional crises. The threat of nuclear war evidence in Europe, North America, and Japan. But the between the superpowers has given way to a mush- problems were addressed, and modern states with profes- rooming of smaller conflicts, entailing costly problems sional systems emerged. This gives us hope. But it also of refugee relief and rehabilitation. No solid interna- gives us pause, because institutional strengthening takes tional framework exists for managing these conflicts or time. The reforms of the Meiji restoration, which helping avoid them. A more integrated assessment of launched Japan onto the path of development, took al- OVERVIEW 13 most twenty-five years to take root. A more capable state can be built, but the process will be slow and will require Figure 5 Countries with good economic immense political commitment. It is urgent to act now. policies and stronger institutional capability Over the past fifteen years many governments have grow faster responded to internal and external pressure by launching Growth n GDP far-reaching reforms to improve their performance. Typi- per capita (percent cally, changes in macroeconomic policydealing with per year) exchange rates, fiscal policy, and trade policyhave come fastest. These reforms have political implications but do 3.0 not require the overhaul of institutions. They can be 2.5 undertaken quickly, often through decree, by a small group of competent technocrats. All it takes is the politi- 2.0 cal decision to make the change. But other state reforms, dealing with regulation, social 1.5 services, finance, infrastructure, and public works, cannot 1.0 be accomplished so rapidly because they involve changing institutional structures established for different purposes, 0.5 to fit different rules of the game. This kind of institutional reform involves wrenching changes in the way govern- 0.0 Low policy ment agencies think and act, and often a complete over- High capability distortion haul of long-established systems of patronage and corrup- Low capability High policy tion. But such change is absolutely essential if the distortion capability of the state is ever to improve. The two to- gethergood policies and more capable state institutions Note: Each growth rate is the average for a group of Countries. to implement themproduce much faster economic Results are based on a regression using panel data from ninety- development (Figure 5). four industrial and developing Countries for the period 1964-93, Comprehensive reform along these lines will take a controlling for education, income, and other variables. See the Technical Note for details. Source: Commander, Davoodi, and great deal of time and effort in many developing coun- Lee, background paper. tries, and the agenda varies considerably from region to region (Box 3). Reform will also encounter considerable political opposition. But reformers can make a good start by strengthening central agencies for strategic policy for- Neighbors, too, can be a powerful motivator for change. mulation, introducing more transparency and competi- There is a clear domino effect at work in the wave of tion, hiving off activities and agencies with easily specified reform sweeping East Asia, Latin America, and much of outputs, seeking more feedback from users about the Eastern Europe and the former Soviet Union. The threat delivery of services, and working with labor unions on of being left behind can goad countries to improve the programs that will enable workers to seek security in functioning of their bureaucracies. But research has yet to change rather than seek security against change. explain why some countries respond to crises and others do not. Why, for example, does popular tolerance of infla- When do reforms occur? tion seem to be much lower in Asia than in parts of Latin Deep distributional conflicts and constraints embedded in America? And why can some countries endure a long state institutions are at the heart of the explanation for so period of economic decline before responding, while oth- many countries' failure to reform. But they are not im- ers take action much sooner? mutable. Ultimately, change comes when the incentives Often the analysis of winners and losers yields a predic- to throw out the old policies and old institutional arrange- tion of whenor at least whetherreforms will be ments become stronger than the incentives to keep them. undertaken. Reforms have little appeal if the winners can- An economic crisis or an external threat, or the arrival of not compensate the losers. Even when the potential gains a new government with fewer vested interests in the old are enough to allow for compensation, reform can be system, may provide the impetus for reform. But reform hard to achieve because the gains are spread over many can be delayed if those in power stick with outdated poli- people, whereas the losers, although smaller in number, are cies because it is in their (or their allies') interest to do so. powerful and articulate. A further problem is that the ben- And the delay can sometimes be painfully long, as in Haiti efits are often realized in the future, whereas the losses are under the Duvaliers, or Zaire today. immediate. Yet sometimes conditions have deteriorated so 14 WORLD DEVELOPMENT REPORT 1997 Box 3 The regional agenda The key features and challenges of improving the effec- I The job of reorienting the state toward the task of tiveness of the state in the various developing regions "steering, not rowing" is far from complete in Cen- are summarized below. These are of necessity broad tral and Eastern Europe. But most countries have generalizations, and each region includes several coun- made progress and are on the way to improving tries whose experiences are very different. capability and accountability. Low state capability in many countries of the ('om- Many countries in Sub-Saha ran Africa are suffering monwealth of Independent States is a serious and from a crisis of statehooda crisis of capability. An mounting obstacle to further progress in most areas urgent priority is to rebuild state effectiveness of economic and social policy. Reorientation of the through an overhaul of public institutions, reasser- state is still at an early stage, and a host of severe tion of the rule of law, and credible checks on abuse problems have emerged from a general lack of ac- of state power. Where the links between the state, countability and transparency. the private sector, and civil society are fragile and In Latin America, decentralization of power and of underdeveloped, improving the delivery of public spending, coupled with democratization, has dra- and collective services will require closer partner- matically transformed the local political landscape, in ships with the private sector and civil society. what some have called a "quiet revolution." A new The capability of the state in most East Asian coun- model of government is emerging in the region. But tries cannot be considered a problem. But states' greater emphasis is also needed on reform of the legal ability to change in response to the new challenges system, the civil service, and social policies. facing the region will play a critical role in their con- In the Middle East and North Africa, unemployment is tinued economic success. by far the greatest economic and social problem and The main issue in South Asia is overregulation, both makes government downsizing especially difficult. Be- a cause and an effect of bloated public employment cause the political and social difficulties of reform are and the surest route to corruption. Regulatory sim- considerable, although not insurmountable, a promis- plification and public enterprise reform, and the ing approach might be to begin decentralizing selected resulting contraction of the role of the state, will be services, and focus on reforming state enterprises, complex and politically difficult. while preparing the ground for wider-ranging reforms. far that the winners far outnumber the losers. Then reform fected by reform (which may not always be the poorest in can produce immediate economic and political gains. society) can help secure their support. Although compen- sation may be costly in the short run, it will pay off in the How can reforms be sustained? long run. Deep-seated differences and mutual suspicions Reform-oriented political leaders and elites can speed re- among groups can also delay reform. There are no quick form by making decisions that widen people's options, fixes for removing age-old enmities, but social pacts, such articulate the benefits clearly, and ensure that policies as Spain's Moncloa Pacts and Benin's National Economic are more inclusive. In recent years farsighted political Conference, can help. leaders have transformed the options for their people International agencies can encourage and help sustain through decisive reform. They were successful because reform in four ways. First, they can provide important they made the benefits of change clear to all, and built technical advice on what to do and what not to do. This coalitions that gave greater voice to often-silent beneficia- assistance is often invaluable, especially for smaller states ries. They also succeededand this is crucialbecause that lack the resources to handle all the technical issues they spelled out a longer-term vision for their society, internally. But it must be complemented by local ex- allowing people to see beyond the immediate pain of pertise, to adapt reforms to local conditions and institu- adjustment. Effective leaders give their people a sense of tions. The World Trade Organization plays a major role owning the reformsa sense that reform is not something in trade reform, the World Health Organization on health imposed from without. issues, and the International Labour Organisation on Reforming the state requires cooperation from all labor legislation and employment policy. Second, interna- groups in society. Compensation of groups adversely af- tional agencies can provide a wealth of cross-country OVERVIEW 16 experience on a wide range of issues. Often staffed by peo- Instances of state collapse are both extreme and ple from all over the world, they can bring in experts from unique, but they are growing. As the Report elaborates, different backgrounds. Third, the financial assistance no simple generalizations about their causes or effects can these agencies provide can help countries endure the early, be made, nor, for that matter, are there any easy solutions painful period of reform until the benefits kick in. Fourth, to their reconstruction; each case brings its own challenges they can provide a mechanism for countries to make for countries, their neighbors, and the international sys- external commitments, making it more difficult to back- tem. The consequences, however, are almost uniformly track on reforms. If the history of development assistance borne by ordinary people, illustrating once again how teaches anything, however, it is that external support can fundamental an effective, responsive state is to the long- achieve little where the domestic will to reform is lacking. term health and wealth of society. The quest for a more effective state even in the estab- Good government is not a luxuryit is a vital lished industrial countries suggests that the returns to in- necessity for development cremental improvements are high. This is especially true The approach of the twenty-first century brings great in countries where the effectiveness of the state is low. promise of change and reason for hope. In a world of Over time, even the smallest increases in the capability of dizzying changes in markets, civil societies, and global the state have been shown to make a vast difference to forces, the state is under pressure to become more effec- the quality of people's lives, not least because reforms tive, but it is not yet adapting rapidly enough to keep tend to produce their own virtuous circle. Small improve- pace. Not surprisingly, there is no unique model for ments in the state's effectiveness lead to higher standards change, and reforms will often come slowly because they of living, in turn paving the way for more reforms and fur- involve a fundamental rethinking of the roles of institu- ther development. tions and the interactions between citizens and govern- A tour of the world's economies in 1997 would turn ment. But the issues raised in this Report are now an inte- up countless examples of these virtuous circles in action. gral part of the rethinking of the state in many parts of the But it would provide equally plentiful evidence of the world and are on the agenda of the international organi- reverse: countries and regions caught in vicious cycles of zations that assist them. poverty and underdevelopment set in train by the chronic People living with ineffective states have long suffered ineffectiveness of the state. Such cycles can all too easily the consequences in terms of postponed growth and social lead to social violence, crime, corruption, and instability, development. But an even bigger cost may now threaten all of which undermine the state's capacity to support states that postpone reforms: political and social unrest development or even to function at all. The crucial chal- and, in some cases, disintegration, exacting a tremendous lenge facing states is to take those steps, both small and toll on stability, productive capacity, and human life. large, toward better government that set economies on the The enormous cost of state collapse has naturally turned upward path, using the two-part framework suggested in attention to prevention as a preferable and potentially this Report. Reform of state institutions is long, difficult, less costly course of actionbut there are no shortcuts. and politically sensitive. But if we now have a better sense Once the spiral into collapse has occurred, there are no of the size of the reform challenge, we are also much more quick fixes. aware of the costs of leaving things as they are. \RT o RETHNKNG THE STTE THE WORLD OVER HROUGHOUT HISTORY, NOTIONS OF THE STATE'S ROLE HAVE SHIFTED DRAMAT- ically. For m; of this century people looked to government to do morein some cases a great deal more. But during the past fifteen years the pendulum has been swinging again, forcing the world to look at government from a range of conflicting perspectives. The end of the Cold War and the collapse of command-and-control economies, the fiscal crises of welfare states, the dra- matic succes! of some East Asian countries in accelerating economic growth and reducing poverty, and the crisis of failed states in parts of Africa and else- whereall of these have challenged existing conceptions of the state's place in the world and its potential contribution to human welfare Governments are also having to respond to the rapid diffusion of technol- ogy, growing demographic pressures, increased environmental concerns, greater global integration of markets, and a shift to more democratic forms of govern- ment. And amid all these pressures remain the formidableand persistent challenges of reducing poverty and fostering sustainable development. It is not surprising, then, that countries are again putting the state under scrutiny, asking what government's role ought to be and, critically, how that role should be played. This Report explores why and how some states have 18 WORLD DEVELOPMENT REPORT 1997 been more effective than others at playing a catalytic and sustainable role in economic development and the eradication of poverty This part of the Report provides a broad historical and conceptual introduction to the issues (in Chap- ter 1) and examines the empirical evidence of the impact of state policies and institutions on development (in Chapter 2). It conveys three principal messages: I Developmenteconomic, social, and sustainablewithout an effective state is impossible. It is increasingly recognized that an effective statenot a min- imal oneis central to economic and social development, but more as part- ner and facilitator than as director. States should work to complement mar- kets, not replace them. I A rich body of evidence shows the importance of good economic policies (including the promotion of macroeconomic stability), well-developed human capital, and openness to the world economy for broad-based, sustainable growth and the reduction of poverty. But as our understanding of the ingredients of development improves, a deeper set of questions emerges why have some societies pursued these actions with greater success than others, and how, precisely, did the state contribute to these differing outcomes? I The historical record suggests the importance of building on the relative strengths of the market, the state, and civil society to improve the state's effectiveness This suggests a two-part strategy of matching the role of the state to its capability, and then improving that capability These are the sub- ject, respectively, of Parts Two and Three. THE EVOLVING ROLE OF THE STATE AGO, A CANADIAN FARMER AND AN recall Thomas Hobbes' insight, in his 1651 treatise ACENTURY Ivorian probably felt little connection with their Leviathan, that life without an effective state to preserve governments, and none at all to each other. Government order is "solitary, poor, nasty, brutish and short." affected their lives oniy to the extent that it provided a This Report aims to show how every state, regardless of handful of classic public goods, such as law and order and its point of departure, can improve its effectiveness and basic infrastructure, and collected taxes from them. move ever further away from this doomsday scenario. Today the state has expanded dramaticallyand the Toward that end, this chapter begins with a reminder of world has shrunk. The same farmers' descendants send how we got here. Surveying the history of the state from their children to government-run schools, receive medical its early beginnings, it shows how notions of the state's treatment from publicly supported clinics, rely on an role have evolved to produce, in both industrial and devel- array of publicly provided services, and may benefit from oping countries, a dramatic expansion of the state and, government price controls on the seed and fertilizer they more recently, a change of emphasis from the quantity of buy, or the wheat or coffee they sell. These later genera- government to the quality. The chapter then lays out a tions of Canadians and Ivorians are therefore likely to be simple framework for rethinking the state, introducing a much more concerned than their ancestors were about the two-part strategy for greater state effectiveness that the effectiveness of government and the checks and balances rest of the Report explores. The message is that the state on its decisions. And they are likely to be much more can rise to the challenges it faces, but only by, first, match- aware of how their own government's performance com- ing what it tries to do to what it can do, and second, pares with others'. Vastly expanded communications, working to increase the number of things it can do capa- trade and investment, radio and television, friends and rel- bly by reinvigorating public institutions. atives traveling to foreign lands as tourists or migrant Where the state began laborersall of these give Canadians and Ivorians today a much better idea of how the government services they From earliest times human beings have banded together receive measure up to those in other countries. The state's into larger associations, starting with household and kin- behavior, and the consequences of that behavior, are being ship groups and extending through to the modern state. scrutinized like never before. For states to exist, individuals and groups have to cede This scrutiny might lead to better government. But if authority in key areas, such as defense, to a public agency. states are unable to respond constructively to the chal- That agency has to possess coercive power over all other lenges they face, the result could simply be further erosion organizational forms within a designated territory. of the state's credibility, as the gap between what the state States have come in all shapes and sizes, depending can do, and what people ask it to do, widens even further. on a mix of factors including culture, natural endow- The terminal phase of this process is visible in the recent ments, opportunities for trade, and distribution of power. agonies of Angola, Somalia, and Zaire. The state collapses The ancient Athenian state, for example, was under- from within, leaving citizens bereft of even the most basic pinned by slavery and colonial spoils. Further east, elabo- conditions of a stable existence: law and security, trust in rate state structures were built from early times on the contracts, and a sound medium of exchange. These crises basis of state ownership of land or, in Mugha! India and 20 WORLD DEVELOPMENT REPORT 1997 Box 1.1 State and government: Some concepts State, in its wider sense, refers to a set of institutions that of offices, which is more narrow in conception, and the possess the means of legitimate coercion, exercised over relationship between government and the governed. a defined territory and its population, referred to as soci- The first classification is based on the relationship be- ety. The state monopolizes rulemaking within its terri- tween the executive and the legislature. In a parliamen- tory through the medium of an organized government. tary system the executive's continuance in office de- The term government is often used differently in dif- pends on its maintaining the support of the legislature. ferent contexts. It can refer to the process of governing, Members of the executive are commonly also members to the exercise of power. It can also refer to the existence of the legislature. A prime minister may be the most of that process, to a condition of "ordered rule." "Gov- powerful member of the executive, but important deci- ernment" often means the people who fill the positions sions within the executive are usually made collectively of authority in a state. Finally, the term may refer to the by a group of ministers. In a presidential system the manner, method, or system of governing in a society: to executive's position is independent of the legislature. the structure and arrangement of offices and how they Members of the executive are not normally also mem- relate to the governed. While keeping these distinctions bers of the legislature, and ultimate decisionmaking in mind, we also use the terms state and government col- authority within the executive lies with one person, the loquially and sometimes interchangeablyas they are president. often used in discussion and writing around the world. The second classification concentrates on the distri- Government is normally regarded as consisting of bution cf power between levels of government. In a three distinct sets of powers, each with its assigned unitary state, all authority to make laws is vested in one role. One is the legislature, whose role is to make the supreme legislature whose jurisdiction covers the whole law. The second is the executive (sometimes referred country. Local legislatures may exist, but only with the to as "the government"), which is responsible for im- sufferance of the national legislature. In a federal state, plementing the law. The third is the judiciary, which is local legislatures are guaranteed at least a measure of responsible for interpreting and applying the law. autonomous decisionmaking authority. In a confedera- Classifications of government are many but have tion, a group of sovereign states combine for specified tended to concentrate on two criteria: the arrangement purposes, but each state retains its sovereignty. imperial China, highly developed systems of administra- of the public and private spheres have not. Whether in tion and tax collection. The combination of public own- Niccolô Machiavelli's The Prince, Kautiliya's Arthashastra, ership of land and a complex bureaucracy long impeded Confucius' writings, or Ibn Khaldoun's The Muqad- the emergence of modern, market-based economies in dimah, the discussion has revolved around the mutual these regions. rights and obligations of states and citizens. Almost all Yet despite this diversity of origins, states over time came these traditions have included a role for the state in pro- to acquire several common and defining features world- viding basic public goods (although the weight accorded wide. Modern states have a consolidated territory and pop- to public, as opposed to private, goals has varied consid- ulation, and within these they play a centralizing and coor- erably). Using public resources to provide critical public dinating role. Sovereign authority commonly encompasses goods and to raise private productivity is nothing new. separate judicial, legislative, and executive functions (Box Beyond these minimal functions, however, there has 1.1). Since the eighteenth century, through conquest and been much less agreement on the appropriate role of the colonization, nation-states have incorporated most of the state in promoting development. Seventeenth-century world into their own mutually exclusive territories. As em- mercantilists saw a major role for the state in guiding pires disintegrated and minority groups established claims trade. Not until Adam Smith wrote The Wealth ofNations to statehood, the number of nations increased sharply. in the late eighteenth century was it generally recognized Membership in the United Nations jumped from 50 inde- that the market was the best instrument for realizing pendent countries in 1945 to 185 in 1996 (Figure 1.1). growth and improving welfare. The state, on this view, was best held to certain core functionsproviding public Modest beginnings goods such as defense, ensuring the security of persons The configuration of states has varied widely across conti- and property, educating the citizenry, and enforcing con- nents and centuries, but arguments over the proper roles tractsdeemed essential for the market to flourish. THE EVOLVING ROLE OF THE STATE 21 Figure 1.1 One world, many more states Box 1.2 U.S. government action to support market development: Some examples Number of United Nations member countries 200 The United States is the country that produced and believes in the dictum that "That government is 180 best that governs least." Whereas in many parts of 160 the world the role of the government has evolved gradually, the United States was founded on a rev- 140 olution. The framers of the Constitution explicitly asked, What should be the role of the government? 120 Yet even in the United States, where laissez-faire and distrust of government are central to the fram- 100 ing of the state's role in society, government action 80 has often proved critical to the growth and devel- opment of markets. For example: 60 The global telecommunications industry has its 40 roots in U.S. government support for the first 20 telegraph line between Washington and Balti- more in the early 184 Os. 0 The enormous increase in agricultural productiv- 1945 1949 1959 1969 1979 1989 1996 ity in the nineteenth and twentieth centuries can be traced to the federally supported program of Note: Data are as of the end of the year. Source: United Nations research and extension services dating from the data. Morrill Act of 1863. The Northwest Ordinances of 1785 and 1787 committed the government to supporting educa- tion, and to devoting the revenues from the sale of But even then, state intervention went on to play a certain lands to that purpose. And in 1 863 the vital, catalytic role in the development and growth of mar- federal government helped establish the public kets in Europe, Japan, and North America. In the United university system. States, where state involvement in the economy has histor- In 1863, in the midst of the Civil War, the Con- ically been more limited than in Europe or Japan, govern- gress recognized the need for a national financial ment was instrumental in constructing the first telegraph system and passed the National Banking Act, line, which spurred development of the telecommunica- establishing the first nationwide bank supervi- tions industry, and in agricultural research and extension, sory agency. In later years the government cre- which stimulated productivity gains (Box 1.2). ated the Federal Reserve System (the U.S. central In the nineteenth century the state's role in redistribut- bank) as well as a series of public financial inter- ing income was still quite limited. Redistribution in mediaries. Europe came mainly through private charity and other The interstate highway system and federal support voluntary action. Tax systems were usually restricted to for the establishment of railroads are cases of vital customs, excise, monopoly, and commodity taxes. Income public involvement in transport infrastructure taxation, which had been introduced in France and Britain that helped the development of markets in the by the end of the eighteenth century, was not a major United States. source of revenue. The first faint stirrings of the modern welfare state were seen in Germany, at the end of the nine- teenth century, where Chancellor Otto von Bismarck in- troduced the first nationwide systems of social insurance. the Russian Revolution of 1917, which led to the aboli- tion there of most private property and put the state in The expanding state in industrial countries. control, through central planning, of all economic activ- States remained small by modern standards until well into ity. The second was the Great Depression of the 1930s, this century. A series of dramatic events in the aftermath which caused such economic devastation in the noncom- of World War I marked the turning point. The first was munist world that states were spurred to experiment with 22 WORLD DEVELOPMENT REPORT 1997 countercyclical policies to restore economic activity. The half decades between 1960 and 1995, governments in the third event, unleashed by World War II, was the rapid industrial countries swelled to twice their starting size breakup of European empires. That geopolitical change (Figure 1.2), with much of the expansion driven by as well as the clamor for social insurance in the industrial increases in transfers and subsidies. economiesushered in fifty years of policy debate focused Indeed, by the early 1990s it was fair to say that most around a more activist role for government. industrial-country governments spent more time moving The postwar paradigm coalesced around three basic money around the economy in the form of transfers and themes, all of which commanded broad, if not uniform, subsidies than they spent providing traditional public agreement. This three-pillared consensus remained largely goods. Spending on defense and on law and order had undisturbed until the first oil price shock of 1973. First was shrunk to some 10 percent of general-government outlays, the need to provide welfare benefits to those suffering from while over half of all tax revenues were transferred to indi- transitory loss of income or other deprivation. Second was vidual beneficiaries (Figure 1.3). Demographics accounted the desirability of a mixed public-private economy, which for some of the shift, as aging populations forced an would often mean nationalizing a range of strategic indus- increase in outlays for pensions and health care. But tries. Third was the need for a coordinated macroeconomic national preferences also made a difference. Thus, from a policy, on the grounds that the market alone could not point of rough equivalence in 1960, the Swedish state deliver stable macroeconomic outcomes that were consis- grew to nearly twice the size of that in the United States tent with individuals' objectives. In time, the goals of by 1995, in terms of both spending as a share of income macroeconomic policy were made explicit: full employ- and public employment as a share of the population. ment, price stability, and balance of payments equilibrium. States thus took on new roles and expanded existing And in developing countries ones. By mid-century the range of tasks performed by Governments in developing countries were also reaching public institutions included not only wider provision of into new areas. They, too, grew dramatically in the second infrastructure and utilities, but also much more extensive half of the twentieth century (Figure 1.2). Initially, much support for education and health care, In the three-and-a- of the growth came from state and nation building after Figure 1.2 Governments the world over have expanded since 1960 Central government spending (percentage of GDP) 40 1960-64 35 1970-74 1980-84 1990-94 30 25 - a- 20 15 10 5 0 OECD countries All developing Sub-Saharan East Asia and South Asia Latn America Middle East and countries Africa Pacific and Caribbean North Africa Note: Data are in Current national prices. Data for Middle East and North Africa for L960-64 are unavailable. Source: IMF, various years Ia) and (b); World Bank data. THE EVOLVING ROLE OF THE STATE 23 Figure 1.3 Transfers and interest payments have grown Composition of central government spending (percentage of total) OECD Sub-Saharan East Asia and South Asia Latin America Middle East countries Africa Pacific and caribbean and North Africa 100 80 60 Consumption Investment 40 20 I Interest, transfers, and subsidies 0 I I I Note: Data are averages for the years indicated. Source: IMF, various years )a). the collapse of colonialism. The ups and downs in in- the state a central role in correcting them. Centralized ternational commodity markets also had an impact. planning, corrective interventions in resource allocation, Resource-rich economies such as Mexico, Nigeria, and and a heavy state hand in infant-industry development Venezuela tended to use the revenue bonanzas from the were part and parcel of this strategy. Economic national- oil price rises of the 1970s and other commodity booms ism was added to the mix, to be promoted through state to expand their public sectors, sometimes with reckless enterprises and encouragement of the indigenous private abandon. Oil-importing countries, under the imperative sector. By the 1960s states had become involved in virtu- of fiscal austerity, were less able to expand their spending. ally every aspect of the economy, administering prices and Perhaps more significant than these forces has been the increasingly regulating labor, foreign exchange, and finan- shift in thinking about the role of the state over the past cial markets. fifty years. Most developing countries in Asia, the Middle By the 1970s the costs of this strategy were coming East, and Africa came out of the colonial period with a home to roost. The oil price shocks were a last gasp for strong belief in state-dominated economic development. state expansion. For the oil exporters they created a The state would mobilize resources and people and direct bonanza, which many threw into even greater expansion them toward rapid growth and the eradication of social of state programs. As long as resources were flowing in, injustice. State control of the economy, following the the institutional weaknesses stayed hidden. The oil- example of the Soviet Union, was central to this strategy. importing countries, for their part, got caught on a tread- (How it was followed in one country, India, is described null of heavy borrowing of recycled petrodollars to keep in Box 1.3.) Many Latin American, Middle Eastern, and the state growing. The costs of this development strategy African countries also followed this postwar pattern of were suddenly exposed when the debt crisis hit in the state-dominated, import-substituting industrialization. 1980s and oil prices plunged. This belief was reinforced by the popularity of state The collapse of the Soviet Unionby then no longer activism worldwide. The Great Depression was seen as a an attractive modelsounded the death knell for a devel- failure of capitalism and markets, while state interven- opmental era. Suddenly, government failure, including tionsthe Marshall Plan, Keynesian demand manage- the failure of publicly owned firms, seemed everywhere ment, and the welfare stateseemed to record one success glaringly evident. Governments began to adopt policies after another. The new interventionist credo had its coun- designed to reduce the scope of the state's interven- terpart in the development strategy of the day, adopted by tion in the economy. States curbed their involvement in many developing countries at independence, which em- production, prices, and trade. Market-friendly strategies phasized the prevalence of market failures and accorded took hold in large parts of the developing world. The 24 WORLD DEVELOPMENT REPORT 1997 Box 1.3 Evolution of the role of the state in India: The past fifi:y years When India became independent in 1947, income per ment schemes, but only a small fraction of the rising capita had been stagnating for haifa century, and mod- subsidies actually reached the poor. Competition ern industry was minimal. between political parties drove subsidies up at every The Nehru years, 1947-64. India's first prime min- election. The resulting large fiscal deficits (8.4 percent ister, Jawaharlal Nehru, saw industrialization as the key of GDP in 1985) contributed to a rising current to alleviating poverty, and a powerful state with a account deficit. India's foreign exchange reserves were planned economy as essential if the country was to virtually exhausted by mid-1991, when a new govern- industrialize rapidly, accelerate public saving and in- ment headed by Narasimha Rao came to power. vestment, and reduce the role of foreign trade and The reform phase, 1991 to the present. Rising interest achieve self-sufficiency. Unlike many East Asian coun- payments on India's foreign debt meant that neither tries, which used state intervention to build strong pri- the central government nor the state governments vate sector industries, India opted for state control over could continue to finance both subsidies and heavy key industries. Believing the potential of agriculture public investment. The former won out, and the gov- and exports to be limited, Indian governments taxed ernment began to woo private and foreign investment. agriculture by skewing the terms of trade against it and Thus, impending bankruptcy drove the reform process emphasizing import substitution. They saw technical and changed the state's role from that of principal education as vital for industrialization. investor to that of facilitator of entrepreneurship. This Garibi hatao, 1966-77 Under Prime Minister shift was expected to free up government finances for Indira Gandhi, two major shifts took place in the role more social spending, but in practice the fiscal crunch of the state. First, the neglect of agriculture was re- prevented a significant increase. versed through state activism in subsidizing new seeds Rao's government abolished most industrial and and fertilizers, agricultural credit, and rural electrifica- import licensing, devalued the rupee, drastically re- tion. The green revolution took off, and by the mid- duced import tariffs, liberalized the financial sector 1970s India was self-sufficient in grain. The second and foreign investment, and allowed private invest- shift was the tightening of state control over every ment in areas previously reserved for the government. aspect of the economy. Under the slogan of garibi The new coalition government that came to power hatao ("abolish poverty"), banks were nationalized, in 1996 has by and large sustained these reforms. trade was increasingly restricted, price controls were And the 1997 budget takes very positive steps in that imposed on a wide range of products, and foreign direction. investment was squeezed. The state achieved a stran- Thus the old national consensus on socialism has glehold on the economy. Yet growth of gross domestic given way over the course of a few years to a new product (GDP) failed to accelerate, remaining during consensus on liberalization. But formidable challenges this period at 3.5 percent a year. remain. Most parties agree on the need for reform, The spending boom and rising fiscal deficits, 1977-91. yet no party is eager to retrench surplus labor, close Between 1977 and 1991, most stringent controls on unviable factories, or reduce subsidies. The reforms imports and industrial licensing were gradually relaxed, so far are a positive step but must be extended stimulating industrial growth. The government and accelerated if India is to catch up with the East expanded antipoverty schemes, especially rural employ- Asian tigers. pendulum had swung from the state-dominated develop- health, and infrastructure as often asor more than ment model of the 1960s and 1970s to the minimalist they cut low-priority programs, bloated civil service rolls, state of the l980s. and money-losing enterprises. Cuts came primarily in cap- As often happens with such radical shifts in perspective, ital budgets and, in Africa, in operating and maintenance countries sometimes tended to overshoot the mark. Efforts outlays, further reducing the efficiency of investment. The to rebalance government spending and borrowing were result, seen most starkly in Africa, the former Soviet uncoordinated, and the good was as often cut as the bad. Union, and even parts of Latin America, was neglect of the To meet their interest obligations, countries mired in debt state's vital functions, threatening social welfare and erod- squeezed critically important programs in education, ing the foundations for market development. THE EVOLVING ROLE OF THE STATE 25 The consequences of an overzealous rejection of gov- power can help resolve problems of collective action that ernment have shifted attention from the sterile debate of would otherwise reduce the social benefits of insurance state versus market to a more fundamental crisis in state markets, or prevent mutually complementary private effectiveness. In some countries the crisis has led to out- investments from being made, to take just two examples. right collapse of the state. In others the erosion of the state's At the same time, however, the state confronts unique capability has led nongovernmental and people's organiza- challenges both in clarifying its objectives and in ensuring tionscivil society more broadlyto try to take its place. that its employees pursue them. First, even though elec- In their embrace of markets and rejection of state activism, tions and other political mechanisms help mediate many have wondered whether the market and civil society between citizens and the state, citizens' mandates can could ultimately supplant the state. But the lesson of a half- remain vagueand powerful special interests continually century's thinking and rethinking of the state's role in try to direct the focus of government in their favor. Sec- development is more nuanced. State-dominated develop- ond, monitoring performance is difficult in many govern- ment has failed, but so will stateless development. Devel- ment activities such as primary education, environmental opment without an effective state is impossible. protection, and preventive health care. This can make it hard to set standards or put other mechanisms in place to Rethinking the state: A framework ensure accountability. Both problems can lead to state A central difficulty in redefining the state's role is that the bureaucracies being granted enormous room for discre- ground beneath governments' feet is always changing. If tion. When that happens, state officials at all levels may we consider how global economic and social forces have pursue their own agendas rather than the society's. Haiti changed prevailing notions of the state, it is clear that it under the Duvaliers provides a vivid example of where the retains a distinctive role in providing the public goods that use of arbitrary public power for personal profit, rather promote economic and social development. And market than for the social good, can lead. failures continue to offer powerful economic arguments In many countries the voluntary sector has stepped in for state intervention (Box 1.4). But changes in technol- to address some of the gaps in collective goods and ser- ogy are transforming the nature of market failure: in infra- vices left by market and by government failure. The vol- structure, for example, technology has created new scope untary sector brings its own strengths to the table, but also for competition in telecommunications and electric power its own weaknesses. It does a lot of good in increasing generation. And many of the most successful examples of public awareness, voicing citizens' concerns, and deliver- development, recent and historical, entail states working ing services. Local self-help organizations are sometimes in partnership with markets to correct their failures, not the preferred providers of local public goods and services, replacing them. because of their closeness to local concerns. But their con- Equity also remains a central concern of the state. New cern is often for certain religious or ethnic groups and not evidence, especially from East Asia, shows that the famil- society as a whole, their accountability is limited, and iar tradeoff between growth and equity is not inevitable, as their resources are often constrained. The challenge, then, was once thought. Appropriately designed policies in basic for the state is to build on the relative strengths of private education and health care can reduce poverty and increase markets and the voluntary sector while taking into equity while promoting economic growth. Neglecting account and improving its own institutional capability. these social fundamentals of development can be fatal. But All these considerations point to a two-part strategy to the mere fact of market failure, and other problems of improve the state's ability to enhance economic and social inequality and insecurity, does not mean that only the welfare. The first task is to match the state's role to its state canor shouldresolve these problems. The state's existing capabilityto establish the institutional rules and coercive authority within its boundaries gives it unique norms that will enable the state to provide collective goods strengths in seeking to address these concerns, but also and services efficiently. The second is to reinvigorate the unique weaknesses. Governments must keep a firm eye on state's capability through rules, partnerships, and compet- both in deciding whether, and how, to respond. itive pressures outside and within the state. The state's unique strengths are its powers to tax, to prohibit, to punish, and to require participation. The Matching role to capability: What states do and state's power to tax enables it to finance the provision of how they do it public goods. Its power to prohibit and punish enables it Part Two of this Report discusses the first part of the strat- to protect personal safety and property rights. And its egy: matching the state's role to its capability, to improve power to require participation enables it to minimize free the effectiveness and efficiency of public resource use. It riding: those who would reap the benefits of public goods advocates ways in which states can provide the fundamen- can be made to pay their share of the costs. The same tals for development, especially where capability is low 26 WORLD DEVELOPMENT REPORT 1997 Box 1.4 The economic rationale for state intervention and sonie definitions Market failure and the concern for equity provide the crease prices and profits. Governments have addressed economic rationale for government intervention. But this problem by regulating private monopolists or there is no guarantee that any such intervention will providing the good or service themselves. Changes in benefit society. Government failure may be as com- technology have created new scope for competition in mon as market failure. The challenge is to see that the services once considered natural monopolies, such as political process and institutional structures get the telecommunications and power generation. incentives right, so that their interventions actually Incomplete markets and imperfect or asymmetric info improve social welfare. mation are pervasive problems and can result in ineffi- Ma rket failure refers to the set of conditions under cient outcomes. Markets are incomplete whenever which a market economy fails to allocate resources effi- they fail to provide a good or service even though the ciently. There are many sources of market failure and cost would be less than what individuals are willing to many degrees of failure. The implications for the role pay. Imperfect information on the part of consumers of the state and the form of public intervention can be can lead to systematic undervaluation of some ser- quite different in each case. vices, such as primary education or preventive health Public goods are goods that are nonrival (consump- care. Asymmetry of informationwhen suppliers tion by one user does not reduce the supply available know more than consumers, or vice versacan lead to for others) and nonexciudable (users cannot be pre- excessive or supplier-induced demand, for example in vented from consuming the good). These characteris- the provision of medical care. Problems of adverse tics make it infeasible to charge for the consumption of selection and moral hazard can lead to the failure of public goods, and therefore private suppliers will lack insurance markets. Adverse selection occurs when buyers the incentive to supply them. National public goods, of a service tend to impose higher-than-average costs on such as defense, benefit an entire country; local public the service provider, or when sellers are able to exclude goods, such as rural roads, benefit a smaller area. Pri- such high-cost customers. Health insurance provides vate goods are those that are both rival and excludable, an example: those who are more likely to need care are common property goods are nonexcludable but rival (an more likely to buy insurance, and more likely to be example is groundwater irrigation), and club goods are turned down by insurers. Moral hazard is present when nonrival but excludable (examples are interurban high- persons carrying insurance have an incentive to cause or ways and toll roads). allow the insured-against event to happen. An example Externalities arise when the actions of one person or is the tendency of health care consumers to seek, as well firm hurt or benefit others without that person or as providers to provide, more treatment than they need firm paying or receiving compensation. Pollution is an when a third party, the insurer, is paying most of the example of a negative externality, which imposes Un- cost. Governments have sought to address these prob- compensated costs on society; the broader benefit to lems by ensuring widespread coverage and holding society at large of a literate population is a positive ex- down costs. They have done this by either regulating ternality of primary education. Governments can curb private insurance, financing or mandating social insur- negative and promote positive externalities through ance, or providing health care themselves. regulation, taxation or subsidy, or outright provision. Equity may prompt state intervention even in the A natural monopoly occurs when the unit cost of absence of market failure. Competitive markets may providing a good or service to an additional user distribute income in socially unacceptable ways. Persons declines over a wide range of output, reducing or elim- with few assets may be left with insufficient resources to inating the scope for competition. But left to operate achieve acceptable living standards. Government action freely, monopoly providers can restrict output to in- may be required to protect the vulnerable. (Chapter 3). Chapter 4 looks at more demanding state from activities that will not be undertaken at all without functions, such as regulation and industrial policy, and state intervention to activities in which the state plays an shows how getting the right Lit between roles and capabili- activist role in coordinating markets or redistributing assets: ties is vital for improving the state's effectiveness. Table 1. presents a framework for thinking about these issues. It Countries with low state capability need to focus first on classifies the functions of government along a continuum, basic functions: the provision of pure public goods such THE EVOLVN6 ROLE OF THE STATE 27 Table 1.1 Functions of the state Addressing market failure Improving equity Minimal functions Defense tipoverty programs Law and order Disaster relief Propert :hts Macroecono nagement h Intermediate g ing functions nonopo, .rmati. Basic educati rtyregu urance (he distributive pen Environment Antitrust ife. pensio Family allowan protection :;ncial regula Unemployment er protectio ins Activist ting private Redis functions Fostering markets sset redistribution ster initia as property rights, macroeconomic stability, control of olies. The design of regulation needs to fit the capabil- infectious diseases, safe water, roads, and protection of ity of state regulatory agencies and the sophistication of the destitute. In many countries the state is not even pro- markets, and give greater emphasis to personal respon- viding these. Recent reforms have emphasized economic sibility. fundamentals. But social and institutional (including Although the state still has a central role in ensuring the legal) fundamentals are equally important to avoid social provision of basic serviceseducation, health, infra- disruption and ensure sustained development. structureit is not obvious that the state must be the Going beyond these basic services are the intermediate only provider, or a provider at all. The state's choices functions, such as management of externalities (pollu- about provision, financing, and regulation of these ser- tion, for example), regulation of monopolies, and the vices must build on the relative strengths of markets, provision of social insurance (pensions, unemployment civil society, and state agencies. benefits). Here, too, the government cannot choose In protecting the vulnerable, countries need to distin- whether, but only how best to intervene, and govern- guish more clearly between insurance and assistance. ment can work in partnership with markets and civil Insurance, against cyclical unemployment for example, society to ensure that these public goods are provided. aims to help smooth households' income and con- States with strong capability can take on more-activist sumption through a market economy's inevitable ups functions, dealing with the problem of missing markets and downs. Assistance, such as food-for-work programs by helping coordination. East Asia's experience has or bread subsidies, seeks to provide some minimum renewed interest in the state's role in promoting mar- level of support to the poorest in society. kets through active industrial and financial policy. Reinvigo rating the state's capabi/ity Matching role to capability involves not only what the Reinvigorating the state's capabilitythe second, equally state does but also how it does it. Rethinking the state also vital part of the reform strategyis the subject of Part means exploring alternative instruments, existing or new, Three. Its theme is that such improvements are possible that can enhance state effectiveness. For example: only if the incentives under which states and state institu- tions operate are changed. Improving capability is not In most modern economies the state's regulatory role is easy. The modest successes, and many failures, of techni- now broader and more complex than ever before, cov- cal assistance efforts over the decades underscore that it is ering such areas as the environment and the financial a matter of changing the incentives that determine behav- sector, as well as more traditional areas such as monop- ior as much as it is one of training and resources. The key 2S WORLD DEVELOPMENT REPORT 1997 is to Find rules and norms that create incentives for state A remote, imperious state, whose deliberations are not agencies and officials to act in the collective interest, while transparent, is much more likely to fall into the downward restraining arbitrary action. This can be achieved through: spiral of arbitrary rule and decreasing effectiveness. Chap- ter 7 looks at the benefits of making government reach Rules and restraints. Mechanisms for enforcing the rule out to people and grant them a greater role in deciding of law, such as an independent judiciary, are critical and implementing policy. And it shows how, carefully foundations for sustainable development. Along with pursued, decentralizationthe transfer of powers and appropriate separation of powers and the presence of resources to lower levels of governmentcan support this watchdog bodies, they also restrain arbitrary behavior. effort. Finally, Chapter 8 provides a reminder that collec- Competitive pressure. Competitive pressure can come tive action increasingly involves looking across national from within the state bureaucracy, through recruitment borders. The chapter examines the various ways in which of civil servants on the basis of merit. It can come from national governments can and must cooperate to meet the domestic private sector, through contracting out demands that, although felt at home, can only be for services and allowing private providers to compete addressed effectively at the international level. directly with public agencies. Or it can come from the Strategic options: Initiating and sustaining reforms international marketplace, through trade and through the influence of global bond markets on fiscal decisions. This two-part strategy for improving state effectiveness Voice and partnership. The means to achieve trans- is far easier said than accomplished. The difficult job for parency and openness in modern society are many and reformers will be not only devising the right kind of variedbusiness councils, interaction groups, and con- reforms but combating the deep-seated opposition of sumer groups, to name a few. Institutional working those with a vested interest in the old ways. Matching arrangements with community groups can contribute role to capability means shedding some roles, including to greater state effectiveness by giving citizens a greater some that benefit powerful constituencies. Proponents of voice in the formulation of government's policies. And a more capable state will quickly discover that it is in partnerships between levels of government and with many people's interest to keep it weak. Nevertheless, international bodies can help in the provision of local politicians have an incentive to undertake reforms if they and global public goods. result in net gains to important constituencies. Windows of opportunity occasionally open in response to crisis or All three mechanisms are a recurrent theme of Part external threat, and effective political leadership is skilled Three, which starts (in Chapter 5) by looking at the basic at devising strategies for building consensus or compen- building blocks of a more effective public sector. The sating losers. emphasis there is on rules and forms of competition to Part Four explores the challenge of initiating and sus- enhance the three basics: policymaking, service delivery, taining reforms of the state (Chapter 9). Its central argu- and the lifeblood of the public sector, the civil service. But ment is that constraints on reform are largely political and history tells us that rebuilding public trust in govern- institutional. Hence fundamental institutional reform is mentand therefore its capabilitywill involve putting likely to be long term, but reform opportunities arise, or restraints on arbitrary action. These issues are taken up in can be created, and these must be seized. Finally, Chapter Chapter 6, which analyzes the checks and balances in the 10 lays out the prospects for change and the reform constitutional structure of the state and the best ways to agenda for each developing region. The message is that control arbitrary behavior and corruption. reform will be difficult and must be tailored to its cir- A third layer, which supports the other two, is efforts cumstances, but the special challenge of' collapsed states to open up the government and make it more responsive. provides a salutary reminder of the risks of failure. REFOCUSING ON THE EFFECTIVENESS OF THE STATE Men are powerless to secure the ture; institutions alone fix the destinies of nations. Napoleon I, Imperial séance (June 7, 1815) IFTY YEARS AGO, WORLD WAR H HAD ENDED AND edge and ideas and increases the speed with which they are F- reconstruction was under way in much of Europe, absorbed, disseminated, and used. Since the 1980s the the Soviet Union, and Japan. Many developing countries focus has shifted to the role of sound policies in explain- were starting to emerge from colonialism, and the future ing why countries accumulate human and physical capital seemed full of promise. The difficulties of economic de- at different rates. This, in turn, has led to yet another shift velopment were not yet haunting us. Improving people's of focus, to the quality of a country's institutions. New, lives looked so achievable, a simple matter of applying the more complex questions have emerged. What institu- right ideas, technical expertise, and resources. And so it tional arrangements best allow markets to flourish? What provedin some cases. But in others progress was mea- is the role of the state both as a direct agent (mostly in the ger. Despite five decades of effort, enormous disparities provision of services) and as a shaper of the institutional remain in the quality of life of people around the world. context in which markets function? How do policies and Indeed, by some measures the gap between rich and poor institutions interact in development? has widened. The answers to these questions are central to our Explanations for these huge international differences in understanding of the deeper sources of differences in living standards have changed over the years. For cen- development outcomesand of why the response to eco- turies, access to natural resourcesland and minerals nomic reform often varies so widely from one country to was considered the prerequisite for development. Much of another. They help explain, for example, why investment Africa, Asia, and the Americas was colonized to acquire and economic activity have revived more strongly follow- these resources, and countries went to war over them. ing the embrace of the market in Poland than in Russia. Gradually, however, the thinking changed, and physical They also help explain why many countries in Africa and capitalmachines and equipmentwas held to be the Latin America have yet to see much of the improvement key to development. "Industrialized" became synonymous in the quality of life they were promised when they with "developed." But around the middle of this century embarked on their economic reforms a decade ago. economic theorists realized that even this was too simplis- The state has much to do with whether countries tic. Embodied in machines and equipment was technol- adopt the institutional arrangements under which markets ogyknowledge and ideas. But no one could explain in can flourish. Not only is the state the arbiter of rules; simple terms why technology developed better and faster through its own economic activity it shapes the environ- in some parts of the world than others. ment for business and the rest of the economy. For good Other factors, such as human capital, have since at- or ill, the state sets the tone. tracted much attention as possible solutions to the puzzle. This chapter makes the empirical case for shifting the Investment in human capital both leads to new knowl- focus of our thinking about development toward the 30 WORLD DEVELOPMENT REPORT 1997 quality of a country's institutions and the capability of the explain much of the difference between countries in statefor bringing institutions into the mainstream of rates of return on development projects. our dialogue about development. That case is supported The state, institutions, and economic outcomes by three new sets of findings: What does the state do? For one thing, it sets the formal First, panel data analyzed for this Report, covering ruleslaws and regulationsthat are part and parcel of a thirty years and ninety-four industrial and developing country's institutional environment (Figure 2.1). These countries, show that policies and institutional capabil- formal rules, along with the informal rules of the broader ity matter for economic growth and for other indicators society, are the institutions that mediate human behavior. of the quality of life, such as infant mortality. But the state is not merely a referee, making and enforc- Second, and taking the analysis a step further, are the ing the rules from the sidelines; it is also a player, indeed results of a survey, specially commissioned for this often a dominant player, in the economic game. Every Report, of over 3,600 domestic firms in sixty-nine day, state agencies invest resources, direct credit, procure countries (including local affiliates of international goods and services, and negotiate contracts; these actions firms). These results, too, provide strong evidence that have profound effects on transactions costs and on eco- institutional capabilityor the lack of ithas a major nomic activity and economic outcomes, especially in de- impact on growth and investment. veloping economies. Played well, the state's activities can The third set of findings explores how institutional accelerate development. Played badly, they will produce capability affects not just the environment for business, stagnation or, in the extreme, economic and social dis- but also the overall setting for a country's development. integration. The state, then, is in a unique position: not Using the results from the survey on institutional capa- only must it establish, through a social and political bility, we show that these cross-country differences help process, the formal rules by which all other organizations Figure 2.1 The state, institutions, and economic outcomes V Informal rules 4 and norms Formal rules Behavior of state agencies Technology Contracts Economic outcomes REFOCUSINO ON THE EFFECTIVENESS OF THE STATE 31 must abide; as an organization itself, it, too, must abide by By providing the institutional infrastructureproperty those rules. rights, peace, law and order, and rulesthat encour- Examples of the power of the state to improve the ages efficient long-term investment, and quality of people's lives are not hard to find. From the By ensuring the provision of basic education, health care, clean water and sanitation systems of ancient Rome to the and the physical infrastructure required for economic elimination of smallpox in this century, public actions in activity, and by protecting the natural environment. the areas of health and sanitation have achieved repeated breakthroughs in public health. And states have long Yet history also teaches that the state can do enormous played a vital role in stimulating lasting development harm: gains by providing infrastructure, security, and a stable macroeconomy. The Internet is oniy the latest in a long The wrong kind of rules can actively discourage the line of remarkable scientific and technical advances made creation of wealth. For example, the state may penalize possible by early and significant public support (Box 2.1). private wealth by distorting pricesthrough an overval- Distilling the lessons of centuries, we see that the state ued currency, for example, or by creating agricultural can improve development outcomes in a number of ways: marketing boards that tax farmers' output and give them little in return. By providing a macroeconomic and a microeconomic Even if the rules themselves are benign, they may be environment that sets the right incentives for efficient applied by public organizationsand their employees economic activity in harmful fashion. They may, for example, impose Box 2.1 Building the Internet: A contemporary example of fruitful public-private interaction The precursor of what we today call the Internet was finance for a high-speed backbone infrastructure for launched in the United States in 1969. Then called the system, the NSF made grants available to universi- ARPANET, the system comprised just four intercon- ties to encourage them to form regional networks that nected computers. By mid-1996, however, the Inter- would feed into the system. But the networks were also net was accessible in 174 countries and on all seven told that they would have to become self-sustaining. continents, linking together nearly 13 million host The private sector's involvement has deepened over computer systems. By 2000 that number could well be time. The NSF encouraged commercial carriers of 100 million. electronic mail to link to the Internet. Companies also ARPANET owed its existence to the economics of began to create their own backbone facilities, and the defense research in the 1960s. Its original purpose was number of firms supplying access to the Internet mul- to link government computers in far-flung locations tiplied. These trends were accelerated by the creation and so avoid duplication of what were then quite costly and rapid growth of the multimedia part of the Inter- computing resources. In 1968 the U.S. Department of netthe World Wide Web. Developed at the labora- Defense invited proposals from 140 private companies tories of the European Organization for Nuclear to design and build the first four interface message Research in Switzerlandanother publicly supported processors, or routers. With these in place, public con- agencythe Web drew in talent from universities and tracts with leading universities then led to develop- firms, leading to yet another explosion in use: from ment of the crucial set of protocols that could link 130 sites in July 1993 to over 230,000 in June 1996. diverse computer networks. It was these protocols that In 1995 NSFNET was replaced by a fully commer- later made the Internet possible. cial system. Major telephone companies now provide Complementary to this public financial support not only backbone facilities but also Internet access to has been the partnership of academia, business, and their customers. Cable and direct broadcast satellite government led by the U.S. National Science Founda- companies are also entering the market. The public tion (NSF). Initially this partnership primed the con- sector is still involved in some advanced research, but its nection of university computer science departments, focus has shifted to such questions as how to ensure equi- but its influence soon expanded. NSFNET replaced table access (for example, through pricing rules), freedom ARPANET in 1990. Besides providing the critical of expression, protection from fraud, and privacy. 32 WORLD DEVELOPMENT REPORT 1997 huge transactions costs, in the form of red tape or bribery, on entrepreneurs setting up new businesses or Figure 2.2 Good government helps explain the restructuring old ones. income gap between East Asia and Africa But potentially the largest source of state-inflicted dam- GDP per capita age is uncertainty. If the state changes the rules often, (thousands of 1985 international dollars) or does not clarifr the rules by which the state itself will 4 behave, businesses and individuals cannot be sure today Contributors to East Asia's what will be profitable or unprofitable, legal or illegal, additional tomorrow. They will then adopt costly strategies to in- growth: sure against an uncertain futureby entering the infor- mal economy, for example, or sending capital abroad Policies all of which impede development. Economic growth and the state State size and capability Government's enormous impact on development is well illustrated by the contrasting economic performance of Unexplained developing countries in Sub-Saharan Africa and East Asia. Initial income gap In 1960 incomes per capita in much of East Asia were only a little higher than in Africa. Governments in the two Sub-Saharan Africa regions were also similar in size, although not in compo- 0 sition: African governments were already spending more 1964 1992 on consumption, primarily on public employment. By the mid-1990s, however, incomes in East Asia were more Note: Results are based on an instrumental variable regression than five times those in Africa. And government con- for the period 1964-93, controlling for initial income, education, and other variables. "Policies" includes the effects of openness sumption in Africa, relative to GDP, had ballooned to to trade and investment, lack of price distortions, relative one-and-a-half times that in East Asia. The sources of this absence of black market exchange rate premiums, and high divergence are complex, but it is widely believed that the levels of education and investment. See the Technical Note. Source: World Bank staff calculations based on data in superior performance of the state in East Asiathe limits Commander, Davoodi, and Lee, background paper. it set on its own growth, the soundness of the policies it adopted, and the effectiveness with which it delivered ser- vicesmade a powerful contribution to the growing gap in the quality of life experienced by the average citizen institutional environment for behavior: the incentive between these two parts of the world (Figure 2.2). structure to which economic agents respond. The private In considering the effect of government size on growth, sector's ability to function will depend critically on the it is useful to distinguish between public consumption and reliability and effectiveness of institutions such as the rule public investment (Box 2.2). Where government con- of law and the protection of property rights. None of sumption spending is very high, it has generally been these benefitsand costsof the quality of government found to be a drag on growth: a net tax on society with few are ever likely to appear in the national accounts. corresponding benefits. Conversely, certain types of pub- The analysis in this section tries to show this distinc- lic investment spending, particularly investment in infra- tion between what the state does and how well it does it, structure, have tended to exert a positive effect on growth, by reporting on both policy content and institutional in part by raising the returns to private investment. Com- capability. Figure 5 in the Overview showed the effect of plicating the picture is the fact that some public con- both factors on income growth over the last three decades sumptionteachers' salaries, for example, or purchases of across a large sample of industrial and developing coun- medicinecan affect people's lives for the better, and even tries. In countries with weak state capability and poor raise the efficiency of investment. Cutting consumption policies, income per capita grew only about half a percent indiscriminately to boost equally indiscriminate invest- per year. In contrast, in countries with strong capability ment is clearly not the answer. and good policies, income per capita grew at an average But even sophisticated measures of the size of the rate of about 3 percent per year. Over a thirty-year period, government only tell part of the story. As noted above, these differences in income growth have made a huge dif- governments also play a leading role in setting the broader ference to the quality of people's lives. A country with an REFOCUSING ON THE EFFECTIVENESS OF THE STATE 33 Box 2.2 Measuring the stateits size, its policies, and its institutional capability A common measure of government size is the ratio of available only as an aggregate of public and private government expenditure to the economy's total expen- investment. To facilitate cross-country comparisons diture or total output. But such data are generally not over time, these ratios are translated into international comprehensive, and coverage of public enterprises is or purchasing-power-parity (PPP) valuesa not en- especially sketchy in many developing countries. This tirely innocuous transformation, particularly for low- measure of size also tends to ignore important off- income countries where much of government con- budget items. Government expenditure itself can be sumption is labor intensive. For these countries, using broken down into consumption and investment. Gov- international prices markedly increases the government ernment consumptionwhich mostly consists of the consumption ratio. public wage billgives a narrow but more precise A more inclusive picture of the economic presence indicator of consumers' current benefits from govern- of government requires a measure that captures key ment spending. Transfers, such as pensions or disabil- government interventions through policy and institu- ity benefits, can be included in government expendi- tions, in addition to fiscal interventions. We summa- ture, but transfers only redistribute resources. Further rize a government's policy stance over time through an complicating matters, nominal and real ratios for index that combines three key indicators: the openness expenditure will vary significantly over time. An alter- of the economy (the share of trade in GDP), overvalu- native measure of government size that avoids these ation of the currency (the black market exchange rate), problems is government employment, but this, too, and the gap between local and international prices. We has its drawbacks. For example, it ignores changes in also attempt to evaluate the quality of a key compo- the productivity of government workers. nent of government, its bureaucracy. This evaluation is The results reported in this chapter use data on real drawn from survey responses by foreign investors (in government consumption, because the concern is the next section we evaluate the responses of local mainly with how the division of output across public investors) that focus on the amount of red tape in- and private goods affects performance. Information on volved in any transaction, the regulatory environment, physical investment is also used, but this is normally and the degree of autonomy from political pressure. average income per capita of $600 in 1965 (in interna- ers cannot afford to focus solely on improving policies; tional PPP dollars), with distorted policies and weak insti- they must also look for ways to strengthen the institu- tutional capability, would after thirty years have reached tional environment those policies have to work within. an average income of only about $678 at 1965 prices. On Important though income growth is, it is only one the other hand, a country with strong institutional capa- of several measures of well-being. Our interest in the bility and good policies would have more than doubled its wide range of factors that make people better or worse average income, to $1,456 at 1965 prices. Many countries off suggests that countries' performance should also be in East Asia have done even better than that. judged by other standards of well-being, such as infant Good policies by themselves can improve results. But mortality. High-quality government institutions lower the benefits are magnified where institutional capability is infant mortality by improving outcomes for a given also higher where policies and programs are imple- amount of social spending. Thus, the capability of the mented more efficiently and where citizens and investors state has an important role in the quality of human life have greater certainty about government's future actions. generallynot simply the pace of income growth. This Thus, good policies such as those being pursued more explains why countries at the same income level can have recently by many countries in Latin America and Africa widely disparate quality-of-life indicatorswhy Sri would increase growth in income per capita by around 1.4 Lanka, for example, had an infant mortality rate of only percent per year. Such a country starting with an average 18 per 1,000 live births, whereas some countries with income of $600 in 1965 would see it rise to around $900 higher incomes per capita had substantially higher rates: after thirty years. But it would rise even higher with good 67 per 1,000 live births in Egypt, and 68 per 1,000 in policies and strong institutions. The lesson is that reform- Morocco, for example. The amount of social spending as 34 WORLD DEVELOPMENT REPORT 1997 well as the care with which services are delivered makes a To this end a large-scale survey of the private sector huge difference. was conducted for this Report. The aim was to capture the full array of uncertainties that entrepreneurs face and Understanding institutional capability better: The private investor's view to build an overall measure of the credibility of rules in a given country. The responses showed that in many coun- As this chapter has already stressed, the mark of a capable tries private investors give the state very poor marks for statebesides its ability to facilitate collective actionsis credibility indeed. its ability to set the rules that underpin markets and permit them to function. Although private arrangements can Credibility: How private investors perceive the state sometimes supplement formal property and contract rights, The private sector survey covered sixty-nine countries and they can only take the development of markets so far. Gov- over 3,600 firms. Entrepreneurs were asked for their sub- ernments, of course, have to do more than establish sound jective evaluation of different aspects of the institutional rules of the game; they also have to make sure those rules framework in their country, including security of property are enforced consistently and that private actorsbusiness, rights, predictability of rules and policies, reliability of the labor, trade associationscan have confidence that the judiciary, problems with corruption and discretionary rules will not be changed overnight. States that change the power in the bureaucracy, and disruptions due to changes rules frequently and unpredictably, announce changes but in government. fail to implement them, or enforce rules arbitrarily will lack Sometimes the source of uncertainty is the instability credibility, and markets will suffer accordingly. of the rules to which firms are subject. Two key indicators How good are governments at providing credible rules included in the survey were: that will nurture the development of markets? Hard evi- dence is difficult to come by. To begin with, credibility is Predictability of rulemaking: the extent to which entrepre- tricky to measure: it depends as much on perceptions as neurs have to cope with unexpected changes in rules and on hard facts. At first glance, for example, one would policies about which they have had no say. think that the number of times a country has changed its government might be a good indicator of the degree of The survey showed that entrepreneurs in some parts of uncertainty about market rules, and therefore of the gov- the world live in constant fear of policy surprises. In the ernment's credibility. Yet businesses in Thailand generally Commonwealth of Independent States (CIS) almost considered their environment to be relatively stable, de- 80 percent of entrepreneurs reported that unpredict- spite numerous coups and changes in government. By the able changes in rules and policies seriously affected same token, the environment for business can be highly their business. In Central and Eastern Europe (CEE), volatile and unpredictable even if the government does Latin America, and Sub-Saharan Africa around 60 per- not change. Peruvian entrepreneurs reported severe credi- cent of entrepreneurs voiced the same complaint. By bility problems in the 1980s because rules were being contrast, in the industrial countries and in South and drawn up hastily, implemented by presidential decree, Southeast Asia, only about 30 percent of respondents and often overturned soon thereafter. considered this a problem for their business (top left Measures of corruption might seem another good sig- panel of Figure 2.3). A large part of the unpredict- nal of government credibility. But simple estimates of cor- ability of rule changes came from companies' having ruption, like measures of political instability, may not little or no role in the state's decisionmaking process; capture entrepreneurs' concerns. Some forms of corrup- indeed, they may not even be informed of important tion entail large uncertainties and risks, whereas others rule changes before they take place. This problem may be more predictable and act more like speed money. appeared to be particularly severe in the CIS, CEE, and In the words of one entrepreneur, "There are two kinds of Sub-Saharan Africa, whereas Asian entrepreneurs (even corruption. The first is one where you pay the regular small ones) considered themselves well informedeven price and you get what you want. The second is one where better informed, in fact, than their industrial-country you pay what you have agreed to pay and you go home counterparts did. Unsurprisingly, perhaps, the survey and lie awake every night worrying whether you will get it also revealed that small companies tend to have less or if somebody is going to blackmail you instead." The knowledge of, and involvement in, the drafting of new best way to understand the problems holding back private regulations and were therefore more subject to policy sector development is to ask entrepreneurs directly. surprises. REFOCUSINO ON THE EFFECTIVENESS OF THE STATE 35 Figure 2.3 Reliable institutions make for credible states Survey respondents reporting dissatisfaction with: Unpredictable changes in laws and policies Unstable government Percent Percent 90 90 80 80 70 70 60 60 50 50 40 40 30 20 10 0 HIC SSEA MENA CEE LAC SSA CIS 30 20 10 0 11 HIG SSEA MENA GEE Ii LAG SSA GIS Insecurity of property Unreliable judiciary Percent Percent 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 HIC SSEA MENA GEE LAC SSA CIS HIG SSEA MENA GEE LAG SSA GIS Corruption Percent 90 80 H IC High-income OEGD 70 SSEA South and Southeast Asia 60 M ENA Middle East and North Africa 50 GEE Gentral and Eastern Europe 40 LAG Latin America and Garibbean 30 SSA Sub-Saharan Africa 20 GIS Commonwealth of 10 Independent States 0 HIC SSEA MENA GEE LAG SSA CIS Note: Results are from a survey of over 3,600 firms in sixty-nine industrial and developing countries conducted in 1996. Regions are listed from left to right according to their overall credibility index (see Box 2 in the Overview). SSEA and MENA are represented by only three countries each. See the Technical Note on the private sector survey for details. Source: Brunetti, Kisunko, and Weder, background paper (b). Perceptions of political stability.' whether changes in gov- Entrepreneurs in many regions felt that the institu- ernment (constitutional or unconstitutional) are usually tional framework was not well enough entrenched to accompanied by far-reaching policy surprises that could withstand changes in government without serious dis- have serious effects on the private sector. ruption. In the CIS, Africa, and the Middle East over 35 WORLD DEVELOPMENT REPORT 1997 60 percent of entrepreneurs said that they constantly cent, in Asia about 30 percent, and in the CIS over 60 feared government changes and the painful policy sur- percent (bottom panel of Figure 2.3). Furthermore, prises that tended to go with them (top right panel of over half the respondents worldwide did not regard a Figure 2.3). bribe as a guarantee that the promised service would be delivered, and many lived in fear that they would sim- Uncertainty may relate less to the rules themselves ply be asked for more by another official. than to the way they are enforced. The relevant indicators here were: Lack of credibility reduces investment, growth, and the return on development projects Crime against persons and property whether entrepreneurs When the private sector does not believe that the state felt confident that the authorities would protect them will enforce the rules of the game, it responds in a variety and their property from criminal actions, and whether of ways, all of which worsen economic performance. An theft and other forms of crime represented serious prob- unreliable judiciary forces entrepreneurs to rely on infor- lems for business. mal agreements and enforcement mechanisms. A corrupt bureaucracy that is allowed too much discretion creates Private entrepreneurs in many countries complained of incentives to seek economic rents rather than in produc- the lack of even the most basic institutional infrastruc- tive activity. A generalized environment of crime and ture for a market economy. Across the globe, crime insecurity of property rights prompts entrepreneurs to and theft were listed as serious problems, which sub- enlist the help of private security agents, or forces them to stantially increased the cost of doing business. A com- pay organized crime for "protection"if it does not put plete institutional vacuum seems to prevail in some them off going into business altogether. countries, leading to crime, violence, and a generalized Investment suffers because entrepreneurs choose not to insecurity of property rights. In Latin America, Sub- commit resources in highly uncertain and volatile environ- Saharan Africa, the CIS, and GEE almost 80 percent of ments, especially if those resources will be difficult to entrepreneurs reported a lack of confidence that the recover should the business environment turn unfavorable. authorities would protect their person and property Where even the most basic types of property are not pro- from criminals (middle left panel of Figure 2.3). tected, investors move their resources to other countries, or invest them in projects that offer lower returns but require Reliability ofjudicial enforcement whether the judiciary less capital commitment. Thus, trade and services may sur- enforces rules arbitrarily, and whether such unpre- vive even in low-credibility environments, but manufac- dictability presents a problem for doing business. turing and, especially, high-technology projects are un- likely to flourish. A similar distortion occurs when highly A well-functioning judiciary is a central pillar of the talented people choose to become tax inspectors or cus- rule of law. Unfortunately, in many countries it seems toms officials rather than train to become engineers. to be the exception rather than the rule. In developing Therefore, credibility affects not only the level of invest- countries over 70 percent of entrepreneurs said that ment in physical and human capital but also its quality. As judicial unpredictability was a major problem in their a consequence, in a low-credibility environment, growth business operations (middle right panel of Figure 2.3). suffers. Disturbingly, in most regions entrepreneurs felt that The top two panels of Figure 2.4 show how credibility these problems had increased over the last ten years. relates to investment and growth in the countries surveyed for the period 1985-95. After controlling for other eco- Freedom from corruption: whether it is common for pri- nomic variables, countries with high credibility had invest- vate entrepreneurs to have to make irregular additional ment rates significantly higher than countries with low payments to get things done, and whether, after paying a credibility; a shift from a low- to a high-credibility envi- bribe, they have to fear blackmail by another official. ronment makes a substantial difference in growth as well. Low credibility may also help explain why many countries The survey confirmed that corruption is an impor- do not see the expected private sector response after imple- tantand widespreadproblem for investors. Over- menting stabilization and structural adjustment programs. all, more than 40 percent of entrepreneurs reported Finally, the credibility of rules affects not only the having to pay bribes to get things done as a matter of business environment, but also the environment for the course. In industrial countries the figure was 15 per- implementation of development projects. The same fac- REFOCUSING ON THE EFFECTIVENESS OF THE STATE 37 Figure 2.4 Perceived credibility and economic performance go hand in hand Gross investment (percentage of GDP) Growth of GOP per capita (percent per year) 35 30 25 20 15 10 5 0 1 2 3 4 5 6 Index of credibility Index of credibility Average rate of return on World Bank- financed projects (percent per year) 40 S 30 I S Note: The top two panels are based on regressions, for thirty- 20 S three (top left panel) or thirty-two (top right panel) of the S countries examined in Figure 2.3 for the period 1984-93, . controlling for income, education, and policy distortion. The . p S bottom panel is based on a regression, for 312 World Bank- S financed projects in thirty countries over varying periods, 10 controlling for terms-of-trade changes and policy distortion. See the Technical Note for details. Source: World Bank staff S calculations using data from Brunetti, Kisunko, and Weder, 0 background paper (a). 2 3 4 5 6 Index of credibility torscrime, corruption, uncertainty about policy, and even more widespread in many public projects than in the judicial behavioraffect the outcome for all such proj- private sector. As a result, many projects are delayed be- ects. One reason is that these concerns are part and parcel cause of cost overruns. of any contractual environment. If corruption affects the The bottom panel of Figure 2.4 shows the correlation private sector, it is likely to affect the outcome of devel- between government credibility and rates of return for 312 opment projects as well. A second reason is that many development projects in thirty countries. On average, in public projects are implemented by private contractors countries with a low-credibility environment rates of re- who, in an environment of weak institutions, are subject turn are substantially higher than in countries with a high- to the same behavioral problems that affect private firms. credibility environment. These results take account of The contractor is awarded a project, pays off corrupt offi- differences in economic policies and other project- and cials, and gets more projects regardless of how the first country-specific factors. The lesson, once again, is that turns out. Pilfering, theft, and enforcement problems are institutions make an enormous difference to development 38 WORLD DEVELOPMENT REPORT 1997 outcomes. Napoleon's insight, cited at the beginning of tainties that have kept markets weak and underdeveloped. this chapter, is as true today as it was in 1815. Countries suffering from such an institutional vacuum risk postponing economic and social development indefi- Strategic options: Refocusing on the state's nitely. There is also a danger that dissatisfaction with the institutional capability statewhether expressed through social protest, capital A clearer understanding of the institutions and norms flight, or the ballot boxwill undermine economic pros- embedded in markets shows the folly of thinking that pects even further. development strategy is a matter of choosing between the The state's capabilityits ability to deliver collective state and the market. As this brief review of the evidence goods efficiently-is central to providing a viable institu- on the relationship between institutions and development tional framework for development. As we have seen, many has confirmed, the two are inextricably linked. Countries developing countries are starting out from a very low base need markets to grow, but they need capable state institu- indeed in this regard. But the state's ability to provide the tions to grow markets. institutional support that development requires can be Reformers the world over need to apply this lesson by improved over time, through matching the state's role to refocusing attention on institutional capability. The task its capability, and then rebuilding that capability by focus- is particularly urgent in many developing countries, where ing on the incentives that drive the behavior of the state. weak and arbitrary governments are feeding the uncer- We turn to these issues in Parts Two and Three. AS PART ONE HIGHLIGHTED, AN EFFECTIVE STATE PROVIDES VITAL IN e NTS for development. This part of the Report argues that governments will achieve better results by being realistic in what they set out to accomplish They must strive to match what they doand how they do itto their instit 1 capa- bilities, not to some idealized model Where government has a longprd t - re, seeking a be " match between the state's role and its capability cnrsound like a recipe fo an- ding the state altogether But market development without a functioning state is not an option. Rather, as Chapter 3 explains, the point is to prioritize. In many countries the state is still not securing the economic and social funda- mentals a foundation of lawfulness, a benign (and stable) policy e nment, basic social services, and some protection of the vulnera$le At th time, * it is overproviding many goods and services that private markets and voluntary initiative could deliver instead For development to proceed, such governments need to go back to the basics Chapter 4 explores how governments can fin4the right match between role and capability in a second area of policy where the state's behavior will inevitably make a large difference to developmt outcomes regulation, hberarn, and 40 WORLD DEVELOPMENT REPORT 1997 industrial policies designed to foster markets. Markets and society need and ben- efit from effective regulation of certain activities But many states are stifling pri- vate sector development by overregulating or, often, monopolizing large chunks of the economy by attempting complex strategic interventions in industry that are beyond their institutional capabilities Deregulation, privatization, and less demanding approaches to regulation and industrial support in these circum- stances will deliver large and immediate payoffs. These chapters carry four basic messages for policymakers: a States at all levels of institutional capability should respect, nurture, and take advantage of private and voluntary initiative and competitive markets. States with weak institutional capabilities should focus on providing the pure public goods and services that markets cannot provide (and that vol- untary collective initiatives underprovide), as well as goods and services with large positive externalities, such as property rights, safe water, roads, and basic education. Credibility is vital for success States with weak institutional capabilities should also focus on the tools for policymaking and implementation that give firms and citizens confidence that state officials and organizations will not act arbitrarily and will live within their fiscal means. Matching role to capability is a dynamic process. As institutional capability develops, states can take on more difficult collective initiatives (initiatives to foster markets, for example), and use efficient but difficult-to-manage tools for collective action, such as sophisticated regulatory instruments SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS STATES MUST TAILOR THEIR AMBITIONS TO do much to put long-stagnant economies back on track. WEAK their capability, yet some tasks are inescapable. We address the various approaches to government's role The challenge, addressed in this chapter, is to find ways for in effective environmental protection in Chapter 4. stateseven states with relatively weak capabilityto get Establishing a foundation of law and property rights those basic government tasks right. Sustainable, shared, poverty-reducing development has five crucial ingredients: Markets rest on a foundation of institutions. Like the air we breathe, some of the public goods these institutions A foundation of law provide are so basic to daily economic life as to go unno- A benign policy environment, including macroeco- ticed. Only when these goods are absent, as in many nomic stability developing countries today, do we see their importance Investment in people and infrastructure for development. Without the rudiments of social order, Protection of the vulnerable underpinned by institutions, markets cannot function. Protection of the natural environment. The lawlessness syndrome The importance of these fundamentals for develop- Markets cannot develop far without effective property ment has long been widely accepted. But as is shown rights. And property rights are only effective when three below, new insights are emerging on the appropriate mix conditions are fulfilled. The first is protection from theft, of market and government activities for achieving them. It violence, and other acts of predation. The second is pro- is now much clearer that markets and governments are tection from arbitrary government actionsranging from complementary, that government action can be vital in unpredictable, ad hoc regulations and taxes to outright laying the institutional foundations for markets. Also corruptionthat disrupt business activity. These two are much clearer is that faith in governments' ability to sus- the most important. Unhappily, as Figure 3.1 makes evi- tain good policies can be as important for attracting pri- dent, and as the regional patterns in Figure 3.2 highlight, vate investment as the policies themselves. in many countries neither is in place. The third condition The track record of developing countries in managing is a reasonably fair and predictable judiciary. This is a tall the fundamentals has been mixed. Many countries in East order indeed for countries in the earliest stages of devel- Asiaplus others elsewhere such as Botswana, Chile, and opment, yet firms in more than half the countries sur- Mauririushave done a good job. But others have not. As veyed considered it a major problem. Box 3.1 reveals, private firms in many developing regions The absence of these critical supports for property are seriously constrained by the absence of such basic state rights gives rise to what this Report terms the lawlessness functions as the protection of private property. Institu- syndrome. Firms in twenty-seven of sixty-nine countries tional impediments are largely to blame and will be hard surveyedincluding more than three-fourths of those in to overcome. Yet windows of opportunity for reform can the CIS, and about half in Latin America and Africa (but open and widen even in the most inhospitable of settings. none in the OECD)are subject to this triple curse on And a major theme of this chapter is that even a modest markets: corruption, crime, and an unpredictable judi- shift in policy priorities in favor of the bare essentials can ciary that offers little prospect of recourse. 42 WORLD DEVELDFMENTREFDRT 1997 Box 3.1 Weaknesses in fundamentals constrain firms the world over In many countries the fundamentals needed to allow Policy-related problemsnotably regarding taxa- firms to go about creating wealth are not in place. tion and the operation of financial marketsalso The survey of businesspeople described in Chapter 2 tend to rank high (except in Latin America). But it asked firms to rank the relative importance of eight dis- is impossible to tell from the survey results alone tinct obstacles to economic activity, to identify which whether these widespread perceptions reflect the aspects of government action most need improving. As universal desire of firms to pay lower taxes and to the table below shows: borrow more at lower interest rates, or whether they are symptomatic of fundamental policy shortcom- Obstacles associated with uncertain property rights ings. vIore telling is the perception in countries of and dealing with arbitrarinesscorruption and the C1S that policy instability is a major constraint. crimerank among the top three everywhere except Poor infrastructure emerges as the leading constraint among the high-income countries of the Organization in South Asia and the Middle East and North for Economic Cooperation and Development. Regu- Africa, and as one of the top three constraints in lation does not emerge directly as a major obstacle. Latin America and Sub-Saharan Africa. Firms' rankings of obstacles to doing business (Worst = 1) Sub-Saharan Latin America East and Middle East and High-income Obstacle Africa and Caribbean South Asia North Africa CIS CEE OECD Property rights Corruption 1 1 3 2 3 3 5 Crime and theft 5 3 8 8 4 6 6 Regulation 8 8 7 7 8 8 4 Policy Taxes 2 5 2 3 1 1 1 Financing 6 4 5 4 5 2 2 Inflation 4 7 4 6 6 4 8 Policy instability 7 6 6 5 2 7 7 Public investment Poor infrastructure 3 2 1 1 7 5 3 Source: Private sector survey conducted for this Report. Corruption emerged from the survey as a major prob- America, for example, between 1980 and 1991 the mur- lem. Its consequences often do not end with paying off der rate rose from 12,8 per 100,000 people in 1980 to officials and getting on with business. Arbitrary govern- 21.4 per 100,000 in 1991, with increases evident in vir- ment entangles firms in a web of time-consuming and tually all countries and subregions. economically unproductive relations. More than half of Much remains to be learned about how to reverse law- senior managers in firms surveyed in the CISbut only lessness among private citizens. But the solution is likely about 10 percent of those in the OECD countries to embrace many of the reform priorities highlighted by reported spending more than 15 percent of their time this Report, including better protection of the vulnerable negotiating with government officials over laws and regu- and stronger overall capability of state institutions. A lations (Figure 3.2). The burden of red tape is less in other community's descent into lawlessness can evoke a sense of developing countries, but still consistently worse than in helplessness among the law-abiding. But as Box 3.2 de- the OECD countries. Chapter 6 examines in some detail scribes, a recent initiative in Cali, Colombia, has shown how the scourge of corruption can be tamed. that, even under the most difficult of circumstances, civic The high ranking by CIS firms of the two other ele- action can start a reversal from despair to hope. ments of the lawlessness syndromecrime and judicial unpredictabilitypartly reflects the unique institutional More complex institutional underpinnings vacuum created by the rejection of central planning in the Containing lawlessness is necessary to secure property transition economies. Yet indicators from other regions rights, but it may not be sufficient. Information and coor- suggest that institutional decay is widespread. In Latin dination problems can also impede development by under- SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 43 mining markets and property rights, a problem often found in low-income countries. Figure 3.1 The lawlessness syndrome Information problems occur because people and firms inevitably have limited information and understanding, Number of countries where one or more aspects of lawlessness are serious obstacles to doing business or because the rules of the game are unclear. The scope of property rightsincluding the right to use an asset, to permit or exclude its use by others, to collect the income generated by the asset, and to sell or otherwise dispose of itmay not be well defined. People and firms may lack knowledge of profit opportunities, or of the probity of potential business partners. The costs of seeking out such information decline as markets thicken and their support- ing institutions develop, making economies more infor- mation intensive. In developing countries, however, the costs of learning can be high. Coordination of economic activity is difficult because None of predictabi self-interested people and firms behave strategicallythey the three 'iciary generally are willing to share information only when they 21 do not lose by doing so. The presence of moral hazard the risk that other parties might opportunistically renege Note: A country falls within a circle if half or more of its on agreementshinders firms from taking advantage of responding firms scored the problem as 1 or 2 (on a scale of opportunities for mutual gain. As markets develop, insti- 1 to 6, with lower values indicating greater severity), and within tutional arrangements evolve to facilitate cooperation an intersection if at least 40 percent of firms scored both problems as 1 or 2 and the average of the two percentages among firms. Again, however, in developing countries exceeded 50 percent. Source: Private sector survey conducted where those institutions are underdeveloped, such cooper- for this Report. ation can be difficult to achieve. Spot markets can emerge even when information and enforcement mechanisms are weak, since the fact that the exchange is simultaneous makes it more difficult to cheat. symbiotic development of markets and institutions: new But for other transactions the costs of providing adequate industries create demand for more-complex institu- information and enforcement mechanisms to enable busi- tions, which in turn enable the industry to develop fur- ness to proceed can be formidable. ther. Consider the example of mining in the "Wild Well-functioning institutions can reduce these trans- West" Nevada territory of the nineteenth-century United actions costs. History provides abundant examples of the States. Figure 3.2 Negotiating with government officials can be arduous Percentage of firms devoting more than 15 percent of managers' time to negotiating with officials CIS 52 Latin America and Caribbean 38 Middle East and North Africa 38 Sub-Saharan Africa 37 CEE 31 South and Southeast Asia High-income OECD 0 10 20 30 40 50 60 Source: Private sector survey conducted for this Report. 44 WORLD DEVELOPMENT REPORT 1997 Box 3.2 standing up to crime in Cali, Colombia One of the centers of the illegal world trade in cocaine, Public education campaigns promoted tolerance Cali saw its homicide rate leap from 23 per 100,000 and respect for the rights of others. Community citizens in 1983 to over 100 per 100,000 in the early leaders were trained in peaceful dispute settlement; 1990s. Many murders could be directly attributed to children were encouraged to join a Friends of Peace drug trafficking, but many more seemed due to a program; humorous TV commercials aimed to re- spreading culture of violence. Fed up with the growing educate citizens to follow the rules of everyday life, lawlessness, in 1992 the city elected as mayor a re- such as obeying traffic signals, or waiting in line to spected physician who put combating violent crime at board a bus. the center of his political platform. Public services were directed at reducing inequities. Within months the new mayor had mounted a Primary and secondary schools were expanded in major anticrime initiative, the Program for the Devel- depressed areas of the city; water, light, and sewer- opment of Safety and Peace. Starting from the princi- age services were introduced into squatter areas; and ple that prevention should take precedence over repres- youth centers and enterprise development programs sionand after an exhaustive analysis of the patterns worked to bring teenage gang members back into of crimethe program worked to combat crime across society's mainstream. a variety of fronts: Catalysts of violent crime were directly confronted. The city banned the carrying of handguns on certain Organizations of civic order were upgraded. Special high-risk weekends, and sales of alcohol were re- education and housing programs were established for stricted late at night and during holidays. police officers, and improvements were made in the quality of services (including legal aid and conciliation In 1995, after seven consecutive years of increase (to a services) available in the frontline inspectorate offices peak of over 120 murders per 100,000 people), Cali's where citizens file complaints of criminal action. murder rate finally began to decline. In the l850s a few hundred miners worked a forty- The progress of land titling in Thailand is a more con- square-mile area of seemingly marginal value. Only loosely temporary illustration of how the formal specification of tied to the U.S. polity, they operated under entirely property rights can unleash "locked-up" assets and accel- unwritten and informal ownership agreements. The dis- erate private sector-led development. Thailand has issued covery of the gold- and silver-bearing Comstock Lode in more than 4 million title deeds since 1985, in two land the late 1850s precipitated a flood of prospectors. Within titling projects. A third project to title another 3.4 million five months the new miners had established a formal min- parcels is under way. Land is an ideal form of collateral, so ing camp government, which enacted written rules on pri- possession of secure title has improved access to formal vate holdings and enforced them through a permanent credit. Three years after the first titles were issued, Thai claim recorder and an ad hoc miners' court. farmers who had received titles had increased their bor- By 1861 the surface ore was exhausted and miners rowing from the formal sector by 27 percent. By enhanc- resorted to subsurface mininga substantially more ex- ing security of tenure, title to land can boost investment pensive and capital-intensive undertaking. With more at in land improvements (irrigation, fencing, destumping). stake financially and with disputes over underground min- Newly titled Thai farmers increased their use of inputs by ing rights increasing in complexity, the miners pressured 10 to 30 percent, their rate of capital formation by 30 to for, and won, creation of a formal territorial government 67 percent, and their investment in land improvements by with a more extensive judiciarysubsidized in part by the 37 to 100 percent. Even after adjusting for other factors, U.S. Congress. productivity on titled land was between 12 and 27 percent By 1864, with mining production still expanding, the higher than on untitled land. territorial judicial system was overwhelmed by a massive Not every country is in a position to achieve such case load, which could have taken up to four years to results. In Thailand certain background conditions played clear. At the end of that year Nevada was admitted to the an important role. First, formal credit markets were union as a state, and within a year some important judi- already well developed, and lack of formal title (and hence cial rulings resolved disputes over subsurface rights. Prop- of collateral) was the only reason why many farmers could erty rights stabilized, and legal uncertainty ended. not get loans. By contrast, in a number of African coun- SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 45 tries with weak credit markets, no measurable impact of pribumi) business community used its extended Southeast titling on borrowing and investment could be discerned. Asian network to kickstart exports of garments and furni- Second, the Thai titling projects took place against a back- ture. A World Bank survey showed that over 90 percent drop of land disputes that threatened security of tenure of the initial export marketing contacts of non-pribumi but could no longer be adequately resolved through tradi- firms were made through private business connections. tional mechanisms. This is not always the case. Indeed, Indonesia's non-Chinese exporters relied much more on where land is cultivated individually but owned commu- initial support from public agencies. nally, strengthening traditional, community-based systems Even when parties cannot rely on social enforcement, of land administration could increase security at a fraction mechanisms of information sharing can allow quite com- of the cost of establishing individual titles. This is a par- plex transactions to take place. Box 3.3 shows how in ticularly attractive option where communities can switch Brazil, for example, sophisticated credit information sys- to individual titles once the efficiency gains from allowing tems have developed to enable firms to bypass some of the sales to outsiders, and from being able to collateralize land problems created by a predictable, but cumbersome, judi- for borrowing, outweigh the benefits associated with com- cial system. munal tenure. But some complex transactions can proceed even with Focus on the foundations simple judicial systems. A well-functioning judiciasy is an Taken together, the evidence presented here offers reasons important asset, which developing countries would do for hopeand a major challenge. The hope comes from well to build up. As Chapter 6 details, creating a workable the fact that simple institutions can do much to facili- formal judicial system from scratch can be slow and diffi- tate market-based economic development. The challenge cult. But the best should not become an enemy of the comes from the recognition that so many countries better. Even less-than-perfect judicial systems that are presently lack even the most basic underpinnings of mar- cumbersome and costly can help sustain credibility. What kets. The first priority in such economies must be to lay matters is not so much that judicial decisionmaking be the initial building blocks of lawfulness: protection of life fast but that it be fair and predictable. And for that to hap- and property from criminal acts, restraints on arbitrary pen, judges must be reasonably competent, the judicial action by government officials, and a judicial system that system must keep judges from behaving arbitrarily, and is fair and predictable. legislatures and executives need to respect the indepen- Once a foundation of lawfulness is in sight, the focus dence and enforcement capability of judiciaries. can turn to the ways in which specific parts of the legal Without a well-developed judicial system, firms and cit- system can buttress property rights. The legal terrain is izens tend to find other ways of monitoring contracts and vast, ranging from land titling and the collateralization of enforcing disputes. These can often make quite complex movable property to laws governing securities markets, private transactions possible. In the early Middle Ages, for the protection of intellectual property, and competition example, European merchants devised their own sophisti- law. However, reforms in these areasespecially the more cated legal code, the lex mercatoria, to govern commercial sophisticated oneswill yield fruit only where institu- transactions; the code helped revive long-distance trade. A tional capabilities are strong. In many countries, more widespread alternative to legal mechanisms is social basic challenges remain to be met first. enforcement, based on long-term personal relationships. Sustaining a benign policy environment Cheating is deterred, not by the law, but by the "long shadow of the future": both parties pass up the one-time Property rights are the foundation for market-led growth gains from cheating in expectation of the larger gains from and poverty reduction. But much more is needed. Firms a long-term business relationship. The extended family need an environment that induces them to allocate re- plays just this role in supporting business transactions in sources efficiently, to improve productivity, and to inno- many Latin American countries. Although family size lim- vate. And unless firms are confident that policies will re- its the number and variety of possible transactions, families main reasonably stable over time, they will fail to invest, find ways to, in effect, expand their membership, for exam- and growth will lag. ple through marriage among business families or "adop- This section reviews international experience with tion" of trading partners as godfathers, uncles, and aunts. some key policies that support development. It highlights The extensive business networks created by Chinese some institutional reasons why countries find it so diffi- clans, some of which have global teach, are another exam- cult to put good policies in placeand the increasing ple of social enforcement at work. Against the backdrop risks, in a mote integrated world, of pursuing bad policies. of sound economic policies in large parts of East Asia, The emphasis throughout is on finding ways in which these networks have been very successful in generating countries with different institutional capabilities can lock wealth for their members. Indonesia's Chinese (non- in good policies. 46 WORLD DEVELOPMENT REPORT 1997 Box 3.3 Contracting and the judicial system in Brazil The Brazilian judicial system is exceedingly cumber- ers and buyers of garments finds its way into court some from a firm's perspective. A complex maze of laws only one in every 2,600 does so in Chile, and one in may apply to an otherwise simple business transaction. every 20,000 in Peru. In 1981, for example, getting an export license took A second reason why firms shrug off the slowness of 1,470 separate legal actions involving thirteen govern- the judicial system is that (as in all private market ment ministries and fifty agencies. The legal process is economies) private institutional arrangements have also exceedingly slow, primarily because of a complex evolved to restrain opportunism in business dealings, appeals procedure. Yet surprisingly, when asked to while bypassing court proceedings. We cite three evaluate the relative importance of a diverse array of examples. First, Brazilian firms readily provide short- constraints on doing business, firms assigned a low term credit even to new customers with whom they ranking to problems associated with the legal system. have had no prior dealings; they base their confidence One reason is that, cumbersome though it is, on a well-developed credit information system (back- Brazil's judicial system nonetheless seems to provide a stopped by a juridically sanctioned mechanism for secure backdrop of judicial recourse for business trans- publicizing information on people who fail to pay their actions. Most firms report that the judiciary is reason- debts). Second, although it is difficult to claim pledged ably fair and predictable, and they do on occasion turn property when loans are not repaid, under Brazilian to it: two-thirds of a sample of Brazilian firms have dis- law leased property can be reclaimed much more read- agreed with a government official and sought to have a ilyso Brazilians make liberal use of leasing arrange- ruling changed; 60 percent have taken the government ments. Third, for some simple financial transactions, to court, and over 80 percent would do so again. Sim- special judicial mechanisms allow the usual proceed- ilarly, one in every 1,000 transactions among produc- ings to be bypassed. Good policies promote growth signals and incentives for economic agents to accumulate The past few decades have yielded a rich crop of lessons resources, use them efficiently, and innovate. Over time, about the kinds of economic policies that support develop- as we saw in Chapter 2, getting these basics right can have ment. The East Asian miracle shows how government and a dramatic effect on living standards. the private sector can cooperate to achieve rapid growth The relationship between growth and macroeconomic and shared development. The recent recovery of some stabiliiy is well known. Empirical work has shown that Latin American economies, breaking out of a long history high rates of inflation (above single digits) adversely affect of inflation and into renewed growth, has further con- growth. High inflation creates uncertainty about the re- firmed the power of market liberalization, budget restraint, turns on saving and investment, thus creating a disincen- and credibility-enhancing institutions. Africa, especially tive for capital accumulation. Inflation also makes it diffi- south of the Sahara, has been slower in joining this move- cult to maintain a stable but competitive exchange rate, ment, with the exception of a few countries such as Mauri- impeding the country's ability to exploit the benefits of tius and Botswana. But several moreCôte d'Ivoire since openness and creating wage volatility. the devaluation of the CFA franc, Uganda more recently As Box 3.4 shows, governments around the world find have embarked on promising new development paths. it difficult to achieve the strong fiscal and monetary disci- Analyses of these and other experiences consistently pline required for economic stability. Maintaining such find a core set of policies that appear to be essential for policies is harder still. But reforming governments will not growth: inspire the confidence necessary to generate growth unless people believe the new discipline will be sustained. We Providing macroeconomic stability discuss below a range of institutional arrangements that Avoiding price distortions can help inspire such confidence. Liberalizing trade and investment. Limiting price distortions is an essential element of good policies, because price distortions impede growth. They These policies help position an economy to benefit from can discourage necessary investment, divert effort into un- competitive market forces. These forces provide the right productive activity, and encourage inefficient use of SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 47 Box 3.4 International track records on fiscal deficits and inflation As the figure shows, fiscal deficits in the industrial contrast, in the first half of the 1990s neither the countries as a whole rose progressively for two decades African nor the Middle Eastern countries have been starting in the early I 960s, stabilized briefly in the late able to follow through on deficit reductions achieved 1980s, and then began to grow again. Persistently high in the second half of the 1980s. deficits have boosted public debt (even before un- Inflation rates have varied across regions even more funded pension liabilities are included) from about 40 than have fiscal deficits. The inflationary episode of the percent of GDP in 1980 to 70 percent in 1995. De- 1970s and early 1980s spread quickly around the veloping countries in the aggregate have shown con- world. The cooldown of inflation that started in the siderable improvement in fiscal discipline, although industrial countries in the early 1980s has begun to with substantial variation. Fiscal deficits started falling take hold, but with a lag. In the developing countries in the early I980s, mainly because of expenditure cuts. inflation began to moderate in the early 1990s, but not However, this aggregate picture reflects mainly suc- everywhere. In some developing regions, inflation rates cesses in Asia and Latin America, where sustained and are showing signs of convergence toward those of the dramatic deficit reductions have been achieved. By industrial countries. Fiscal deficits Inflation Percentage of GDP Percent per year 6 120 5 100 4 80 Developing countries 3 60 2 40 20 Industrial 1 countries 0 0 1961-65 1971-75 1981-85 1991-95 1961-65 1971-75 1981-85 1991-95 Percentage of GDP Percent per year 8 300 Latin America and Caribbean Latin America 7 and Caribbean Europe and 250 6 - Central Asia Sub-Saharan 5 200 Africa 4 150 Europe and 3 Central Asia 2 100 Middle East and N. Africa 1 Middle East 50 - E. and S. Asia 0 - and N. Africa 1 0 1961-65 1971-75 1981-85 1991-95 1961-65 1971-75 1981-85 1991-95 Source: IMF, various years (a). 48 WORLD DEVELOPMENT REPORT 1997 distortion by cutting explicit taxes and, often, correcting Figure 3.3 Countries are discovering the overvaluations. The 1994 devaluation of the CFA franc advantages of the value added tax (see Box 3.5) significantly reduced the bias against agri- Number of countries using the value added tax culture among virtually all the franc zone countries that had not reformed earlier. 90 Harder to detect, but also widespread, are price distor- 80 tions in labor and capital markets. Legal minimum wages, for instance, may be set too high, unintentionally making 70 it more difficult for unskilled and low-wage workers to 60 find jobs in the formal economy. Similarly, the price of capitalthe interest rateis sometimes kept falsely high 50 through heavy taxation of financial transactions or high 40 reserve requirements. When the authorities respond to borrowers' complaints by clamping a lid on lending rates, 30 or by handing out subsidies to investors, yet another layer 20 of distortion is added to the price system. Maintaining liberal trade, capital market, and invest- 10 ment regimes is also essential for growth. As Chapter 8 0 details, many countries have recently moved toward 1960s 1970s 1980s 1990s greater openness. Open markets offer opportunities for citizens and businesses by increasing access to supplies, equipment, technology, and finance. Trade linkages with Source: Tanzi 1995b. the world economy also help domestic prices adjust to global market conditions, so that prices reflect the scarcity values of goods and services. And improved incentives sources. Price distortions come in different forms, de- and opportunities allow entrepreneurs to use resources pending on their historical origins. The most common, more efficiently. however, involve discrimination against agriculture, over- Recent changes in the way developing countries raise valuation of currencies, unrealistic wages, and hidden taxes tax revenues show how increased global integration can or subsidies on the use of capital. affect domestic policies. Internationalization of business African agriculture illustrates vividly how price distor- and relentless competition for foreign investmentplus tions can undermine economic development. Agriculture the presence of tax havens and low-tax jurisdictions accounts for about 35 percent of Africa's GDP, 40 percent imply that countries cannot hope to tax corporate or per- of exports, and 70 percent of employment. Yet histori- sonal income at rates much higher than global norms and cally, African farmers have faced high rates of both explicit still attract investment. And a growing worldwide consen- and implicit agricultural taxation. Explicit taxes (notably sus for lower national trade barriers has put pressure on on agricultural exports) were high because administrative the collection of border taxes, historically a major source weaknesses precluded raising adequate revenue from other of tax revenue for developing countries. (As a group, sources. Implicit taxes were high because pro-urban and developing countries still derive about 30 percent of their pro-industry policies combined with high levels of import revenue from trade taxes.) With increasing integration, protection resulted in currencies being seriously overval- the share of trade taxes in the total revenue of developing ued in real effective terms. In addition, in some countries countries may be expected to fall further. public sector monopolies raised border prices well above With these new constraints on traditional sources of those at the farm gate, absorbing much of the difference in revenue, many countries are turning toward consumption- in-house expenditures. The combination of high explicit based taxes such as the value added tax (VAT). Indeed, the taxes and overvalued currencies contributed to alarming combination of its revenue potential and pressures on declines in Sub-Saharan Africa's agricultural growth rates: other sources of revenue has led to dramatic growth in the from an annual average of 2.2 percent in 1965-73 to 1.0 number of countries using the VAT (Figure 3.3). percent in 1974-80 and 0.6 percent in 198 1-85. A liberal and open trade regime is also a powerful dis- Since the mid-1980s many African countries have cipline on the other elements of economic policy. More- made great strides in reversing the long-standing bias open economies are more exposed to external risks, mak- against agriculture. By the early 1990s two-thirds of a sam- ing it more costly for governments to pursue inconsistent ple of twenty-seven countries had reduced the degree of policies. Consequently, economies in which trade looms SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 49 relatively large tend to have lower fiscal deficits than those Favorable surprises can cause as much trouble as where it does not. The need to comply with the rules and adverse ones. The prudent response to a positive eco- conventions of international treaties will be another spur nomic shock is to set aside part of the windfall for future to good behavior. use. When the 1990-91 Gulf War pushed up oil prices, An economy without sound policies is unable to Nigeria used its windfall oil revenue to expand spending engage fully in international trade and investment. But (Figure 3.4). So in spite of the large increases in revenue, being part of an integrating world economy also carries Nigeria's fiscal deficit actually rose in 1990. When oil new risks. Where markets for goods and capital are open, prices and revenue fell in 1991, spending remained at the the state has a hard time suppressing the consequences of new higher levels. By contrast, Indonesia responded to its monetary indiscipline. If it prints too much money, the oil windfall with fiscal discipline, explicitly budgeting a foreign exchange market will quickly expect higher infla- reserve fund to keep the increase in expenditure below the tion, and the local currency will depreciate. This market increase in revenue and maintain budget balance. feedback causes domestic interest rates to rise, and with them the government's financing costs. Good policies are Good policies are hard to achieve needed to cope with the risks of capital flight, volatile Although the recipe for good policies is well known, too arbitrage activity, and sharp movements in commodity many countries still fail to take it to heart, and poor per- prices. Box 3.4 summarized some differences in how formance persists. This often signals the presence of polit- countries have responded to the new global environment. ical and institutional incentives for maintaining "bad" Foreign capital inflows also impose discipline on poli- policies. cymakers. Inflows tend to make the currency appreciate in Policies that are bad from a development perspective real terms, and they can affect competitiveness and are often highly effective at channeling benefits to politi- domestic saving. They can also be seriously destabilizing cally influential groups. Many macroeconomic prob- because they respond quickly to short-run financial tur- lemsinflation, exchange rate misalignmentare in fact bulence. Recent experience suggests that this turbulence covert ways of levying unexpected taxes on the private sec- can be contagious, spilling over to other countries and tor or of redistributing economic benefits. Similarly, a even other regions in ways not necessarily commensurate broad array of microeconomic restrictions on the opera- with the change in risk. Countries experiencing sizable tion of marketsimport restrictions, local monopoly capital inflows may need to run positive fiscal balances, using these precautionary savings as a hedge against the possibility of sudden capital outflows. Capital inflows also have major implications for exchange rate policy; fixed exchange rates, for example, are unlikely to be a workable Figure 3.4 Unlike Nigeria, Indonesia managed option if a country is vulnerable in financial markets. In its recent oil windfall prudently short, the quality of a government's management of the Percentage of 1989 revenue economy is critical. 450 The risk of capital flight and financial turmoil is vividly Nigeria Indonesia illustrated by Mexico's experience in 1994-95. An impor- 400 tant reason for the loss of confidence there was an over- 350 valued peso, maintained despite very large current account deficits. As foreign exchange reserves fell below the domes- 300 Expenditure tic monetary base late in 1994, the authorities failed to 250 bring about the necessary monetary contraction. More- consistent policies could have limited the loss of confidence. 200 Revenue - Expenditure An open economy is also exposed to price shocks aris- 150 ing from world markets. Energy and food prices are par- 100 - venue ticularly volatile and can affect a country's external pay- ments and fiscal positions. Exchange rates and interest 50 Oil shock Oil shock rates are also volatile. Prudence calls for anticipating ad- 0 verse shocks (a sharp price increase for importers, a price 1989 1990 1991 1992 1989 1990 1991 1992 drop for exporters) by not borrowing excessively and maintaining scope for new borrowing and by holding ade- Source: IMF, various years (b). quate foreign exchange reserves, and in the medium term by establishing a more diversified economic base. 50 WORLD DEVELDPMENTREPDRT 1997 privileges, regulatory red tapeserve to shelter powerful suggests that long-run goals are better served by sticking incumbent firms or other favored segments of society. to self-imposed restraints and living with the rigidities The political system in some countries has a built-in they inflict. The examples here concern fiscal and mone- tendency toward chronic budget deficits. Legislators tary policy; further examples in the field of regulation are exchange favors, each promising to vote for benefits to the discussed in Chapter 4. other's constituents, without specifying how these bene- FISCAL POLICY. Many macroeconomic disturbances fits will be paid for. So fiscal deficits rise. start life as fiscal imbalances. Recent research suggests that When revenues fall short and politicians have little changing the institutional features of the budgeting process stomach for cutting spending, governments have to choose can improve fiscal performance significantly. either to levy or raise taxes that are desirable from an effi- Increasing the transparency of budgeting is particularly ciency standpoint, or impose hidden taxes such as the so- important. Although society as a whole will lose from bud- called inflation taxthe tax on real incomes that comes getary ambiguity, it can be a boon for politicians, blurring from financing government spending with debased cur- the cost of favors to special interests, for example, or rency. The latter course is often the easier. Increasing for- understating the long-run costs of short-term profligacy. mal tax collection requires an efficient and honest tax When budgets are not transparent, "creative accounting" administration. Achieving that may first require deep practices, such as off-budget spending and overoptimistic structural reform of fiscal administration. A change in the revenue and growth projections, become all too easy. VAT rate might take a vote of parliament, implying delays Needless to say, all of these gimmicks make it harder to and political compromise. But an increase in the inflation control spending. tax might involve no more than a ministerial order to the How budgets are formulated and approved is also central bank. important. The evidence suggests, for example, that it Even when intentions are good, governments may matters whether a country takes a hierarchical approach to sometimes be forced to use hidden taxes like the inflation budgetinggiving considerable power over departmental taxalthough they recognize that in the long run this spending totals to the finance ministryor one that is brings huge costs and undermines credibility. How does a more collegial. In principle, the hierarchical approach government with a history of inflationary financing con- ought to foster greater fiscal discipline by enabling more vince potential bondholders that it will not inflate its way "top-down" control of spending and limiting the scope out of its obligations this time, or simply default? How for legislators to expand the budget piecemeal. can it convince trade union members that it will not cut A recent study of twenty Latin American countries their real income by raising the cost of living? If it cannot, suggests that moves toward more transparent, hierarchical investors will protect themselves by demanding a higher budgeting could deliver improved restraint (Figure 3.5). interest rate on government debt, and workers will protect It found that budget deficits tended to be higher among themselves by demanding bigger raises. Their doubts may countries that used collegial and nontransparent ap- then become self-fulfilling: the government could be proaches to budget preparation. Countries with the least forced to bring about the inflation that these private transparent and least hierarchical systems ran public agents expect, by loosening monetary policy and allowing deficits averaging 1.8 percent of GDP. The middle third real wages or interest rates to rise. ran an average budget surplus of 1.1 percent, while those These perverse but powerful institutional incentives with the highest combined hierarchy-transparency scores can make policy reform very difficult. And even if reforms had budget surpluses, on average, of 1.7 percent. These are initiated, the skepticism of businesses, workers, and results highlight that countries looking to improve their consumers may be borne out by events, unless the gov- aggregate fiscal management need to scrutinize not just ernment can communicate the seriousness of its intent. their balance sheets, but also the institutional environ- ment that shapes the incentives to spend. Locking in good policies MONETARY POLICY. A well-functioning, independent Once reforms are announced, their lasting success may central bank can effectively reduce the threat of politically depend on designing and implementing policies in ways motivated monetary expansion while maintaining some that credibly signal that the government will not renege flexibility to accommodate unavoidable outside shocks. on its promises. A number of possible lock-in mechanisms Many countries seeking credibility for their monetary pol- are available, all with the same basic logic: to provide icy have chosen the model of central bank independence. checks that restrain any impulse to depart from an- In many cases this enthusiasm sprang from evidence nounced commitments. If institutional capabilities are that OECD countries with independent central banks strong enough to allow some flexibility to adapt rapidly to generally had lower rates of inflation than otherswith unexpected events, so much the better. If not, experience no slowdown in growth. But attempts to find a similar SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 51 pattern in developing countries have yielded mixed results, depending on how central bank independence is Figure 3.5 Well-designed budgetary defined. Russia's move to central bank independence in institutions help avoid large deficits the early 1990s, for example, did not seem to restrain that Primary deficit (percentage of GDP) country's inflation. This more complex story for develop- 6 ing countries suggests that monetary restraint through central bank independence cannot simply be manufac- tured by fiat. It may require a prior foundation of checks Bolivia, Honduras and balances on arbitrary action by public officials. Choosing a conservative central bank governor, one El Salvador Trinidad and Tobago who is more opposed to inflation than society in general, may be one way for developing countries to reap the ben- , Argentina .Guatemala efits of central bank independence while containing the risks. Another way is to assign the bank only instrument Peru. Costa Rica ' Uruguay Paraguay ,Colombia independencethe day-to-day setting of policy to achieve \Barbados Dominican a certain goalwhile leaving the choice of the goal itself , Panama Republic to the political authorities. A third option is to establish a contract for the central bank governor that provides for Ecuador ,,Venezuela some penalty for deviating from an announced inflation Mexico , target. This mimics the effect of employing a conservative central banker without relying on subjective judgments Brazil , Jamaica about the person holding the position. Chile , The mixed success of independent central banks in restraining inflation raises the possibility that some devel- oping countries may simply be unable to put in place Low High mechanisms that credibly signal monetary restraint and Index of the quality of budget institutions at the same time maintain the capacity to respond flexibly to outside shocks, For these countries the choice may Note: Budget deficits are averages over the period 1980-92. The index of budgetary institutions is constructed from be between commitment through rigid mechanisms and responses to a survey of country budget directors and has ten no commitment whatsoever. A variety of inflexible ap- components. Source: Alesina, background paper. proaches have been tried: Argentina, in breaking away from a long tradition of These hard-line approaches represent a high-stakes inflation, enacted a currency convertibility law in April race against time. By raising the cost of policy reversal, 1991 that essentially turns the central bank into a quasi such policies contribute to a belief that the government currency board. The money stock must be fully backed will hold fast. In time, however, some exogenous shock by foreign exchange. will be strong enoughor, perhaps, political opposition Many Latin American countries switched to a fixed to some side effect of the policy will be strong enough nominal exchange rate to anchor prices and coordinate to demand a reconsideration. At that point, countries that private sector expectations. A fixed rate precludes the have won the race against time will already have put in use of devaluation to accommodate short-run external place more flexible approaches to monetary restraint, or shocks. But as Mexico discovered to its dismay in 1994, will have won enough credibility that adapting the strat- a fixed nominal exchange rate can become dangerously egy will not be interpreted as a reversal. destabilizing when capital inflows or domestic policies Investing in people and infrastructure pull the real exchange rate out of line. I Most of the francophone countries of Africa affiliated Well-functioning markets are usually the most efficient themselves with the CFA franc zone and its suprana- means of providing the goods and services an economy tional central bank, Central bank advances to a member needsbut not always. In particular, markets undersup- government are limited to 20 percent of tax revenues ply a range of collective goodspublic goods, and private collected the previous year. This prevents countries from goods that have important spillover benefits for society at substituting the inflation tax for conventional taxes (Box large. Generally these are goods that have a significant 3.5). But the same mechanism can also provoke defla- impact on the quality of life: clean air and safe water, basic tion if growth turns negative, as happened in the 1980s. literacy and public health, and low-cost transportation 52 WORLD DEVELOPMENT REPORT 1997 Box 3.5 Commitment versus flexibility in the CFA zone The CFA franc zone of West and Central Africa is hit by a pair of external shocks: a real appreciation of the both a currency union and a monetary standard: the French franc against the dollar, which led directly to a CFA franc is convertible to French francs at a fixed real appreciation of the CFA franc; and a dramatic drop nominal exchange rate. France established the zone in the prices of some members' major exports. The fixed after World War II to oversee monetary and financial exchange rate ruled out adjustment through a nominal policies in its African colonies, and France continues to devaluation. Inflation remained low in this period, but play a central role in its operation. at the cost of stagnant growth. The very factors that had In exchange for France's guarantee of convertibility, contributed to the credibility and stability of the CFA member countries surrender the right to print new cur- zone now made it extremely difikult to devalue the rency. Policy changes require multilateral negotiations CFA franc. By the early 1990s, however, a consensus among the member states and France. Short of with- was finally reached that a devaluation was necessary. drawing completely from the zone, a single country A 50 percent devaluation was announced in Janu- cannot unilaterally renege on its commitment. ary 1994. Its dramatic size signaled that the devalua- Compared with similarly endowed neighbors, zone tion was a once-and-for-all measure. Thus its benefits members experienced lower average inflation and faster could be reaped without undermining the future cred- growth throughout the 1970s and early 1980s. By the ibility of the fixed exchange rate. Indications to date second half of the 1980s, however, certain costs of zone suggest that the devaluation has proved largely success- membership had become apparent. The CFA zone was ful on both counts. and communications. They are also goods whose provi- As World Development Report 1994 highlighted, pub- sion can dramatically affect the welfare and life prospects lic investment in infrastructure boosts private activity in of the poorest in society. developing and industrial countries alike. A study of eighty-five districts in thirteen Indian states found that Public investments in health, education, and infrastructure lower transport costs led to considerable agricultural yield high returns expansion by making it easier for farmers to get their Access to safe water and the control of infectious disease goods to market. More broadly, competing for new export are public goods and services with large externalities that markets requires high-quality infrastructure, to transport will be underprovided, or not provided at all, by the pri- goods large distances at lowest cost. vate sector. Infectious diseases still account for a large pro- portion of deaths in developing countries, and the poor Yet public resources often do not go to these suffer the most. Nearly 1 billion people in the developing high-return investments world lack access to clean water, and 1.7 billion have no The world over, too few resources are devoted to vital sanitation. Water-borne diseases such as cholera, typhoid, basic services. Governments spend roughly $1 per capita and paratyphoid remain a pervasive threat in many devel- on public health, against a minimum requirement of oping countries, especially for the poor. Evidence from $4 per capita. About 130 million primary-school-age chil- Malaysia shows that traditional public health interven- dren-60 percent of them girlswere not enrolled in tions, such as immunizations and provision of safe water, schools in 1990. Half the children in Africa do not go to can make a significant difference in rates of illness and school. Girls, the rural poor, and children from linguistic death, especially in infant mortality. and ethnic minorities are less likely to be in school than Returns to education are especially high at the primary others. level, because universal basic literacy yields large external- Part of the problem is misallocation of resources across ities to society. Educating girls, for example, is linked to sectorsamong defense, state enterprises, and social ser- better health for women and their children and to lower vices, for example. In many developing countries, state fertility rates. Many attribute a good part of the East Asian enterprises produce goods that private markets could sup- countries' economic success to their unwavering commit- ply; the funds these enterprises absorb could be better ment to public funding for basic education as the corner- spent on public goods. Turkey's state-owned coal-mining stone of economic development. company lost $3.5 billion between 1990 and 1996. Tan- SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 53 zania's central government spent one-and-a-half times what it spent on public health to subsidize money-losing Figure 3.6 In Vietnam, the benefits of hospital state enterprises. In low-income countries state enter- care are skewed toward the better-off prises' losses averaged 2.3 percent of GDP between 1978 and 1991. Share of total spending Another part of the problem is the misallocation of Middle Second-richest quintile resources within sectors. Spending on infrastructure and quintile 22% 21% social services tends to be concentrated in areas where markets and private spending can meet most needs urban hospitals, clinics, universities, and transport Richest rather than on essential public goods. These expenditures quintile often benefit the rich disproportionately, while the poor 30% receive only a small fraction. For example, governments often try to finance the Second- poorest Poorest entire range of health care services. Yet public health inter- quintile quintile ventions directed at improving the health status of large 16% 11% sections of the population, including the poor, warrant a Note: Each slice represents the share of total spending directed higher priority. Most curative health care is a (nearly) pure toward that quintile in 1993. Source: World Bank 1995f. private goodif government does not foot the bill, all but the poorest will find ways to pay for care themselves. This may explain why the public provision of clinical care ser- vices had no effect on health status in Malaysia, where Niaking better use ofpublic resources people have the option of using private clinical services. To focus public resources more efficiently on providing Although some governments are beginning to spend collective goods and services, countries will need to reallo- more on primary and secondary education, higher edu- cate expenditures and learn to use their resources more cation is still heavily subsidized relative to other tiers. efficiently. In many countries this will take both political Whereas the Republic of Korea, for example, allocates 84 and institutional change. The vital first step in institu- percent of its education budget to basic schooling, tional change is a readiness to embrace a pluralistic Venezuela allocates just 31 percent. Thirty-five percent approach to delivery: to permit private participation while of Bolivia's education budgetbut only 11 percent of focusing direct public involvement on genuinely collective Indonesia'sis allocated to higher education. The tilt goods and services (although, as discussed below, govern- toward higher education is most acute in Africa, where ments might also choose to subsidize needy groups' con- public spending is about forty-four times greater per sumption of goods even when the returns are wholly pri- student in higher education than in primary schools. At vate). Viewed against the common postwar presumption the extremein Tanzaniathe ratio was 238 to 1. that infrastructure and social services are the exclusive This emphasis on clinical health services and higher domain of public monopolies, pluralistic approaches education entrenches social inequities. Evidence from might seem radical and untested. In fact, private and com- Vietnam confirms that wealthier groups benefit dispro- munity participation in infrastructure and social services portionately from hospital care: the richest fifth of the has a long historical pedigree (Box 3.6). population are estimated to enjoy some 30 percent of the Only in the twentieth century did governments, first in benefits of hospital spending, while the poorest fifth get Europe and later elsewhere, become important providers only 11 percent (Figure 3.6). of services, in extreme cases excluding the private sector Government decisions about what kind of services to altogether. This transition to a more pervasive govern- supply are not the only reason why the benefits of public ment role evolved differently for different services and in spending are unequally distributed. Differences in different countries, giving rise to wide variation in pat- demand, especially those related to gender, are also impor- terns of financing and delivery within and across income tant. In Côte d'Ivoire, for instance, almost two-thirds of groups. Among low-income countries, for example, the public spending on education goes to boys. In Pakistan, private share of total education expenditure ranges from boys benefit from about one-and-a-half times as much around 20 percent in Sri Lanka to around 60 percent in public spending on their education as do girls. Often the Uganda and Vietnam (Figure 3.7). The breakdown of relative disadvantage of girls is even greater in poorer health spending shows similar variation. In Latin America households, reflecting differences in demand in these the private share ranges widely: from one-third of total households for education for girls and boys. health expenditures in Ecuador to 43 percent in Mexico 54 WORLD DEVELOPMENTREPDRT 19g7 Box 3.6 Private provision of social services: A historical perspective Only in the twentieth century did the state assume an borhood doctors. Not until the first antibiotics were important role in providing social services such as edu- mass-produced after World War II did Western medi- cation and health care. The ability of the state to pro- cine begin to benefit large groups of people. In devel- vide these services has varied, however, resulting in dif- oping countries, increased urbanization and industrial- ferent public-private mixes. ization led to the formation of labor groups, which Today's modern education systems were founded on organized themselves to provide health insurance privateoften religiousinitiatives. From the Islamic through "sickness funds" or pressed for publicly schools in Indonesia and West Africa to the Hindu gurus financed social insurance systems. By 1950 sixteen Latin in India, the Christian churches in most of Europe, and American countries had enacted laws to provide health the village teachers of China, private religious schools insurance to selected groups, but only two African and have been teaching children for centuries. In general, four Asian nations had done so. however, education was a privilege of the elites. Mass The International Conference on Primary Health public education is a nineteenth-century invention, orig- Care, held in Alma-Ata, Kazakstan, in 1979, pro- inating in Europe and North America and spreading to claimed health a "basic human right" and urged gov- former colonies after independence. Significant public ernments to take "responsibility for the health of their investment led to expanding public sector enrollments, people." Several governments in developing countries accompanied in several countries by a shrinking role for created national health systems that purport to provide private schools. In Malawi, for example, enrollment in free medical care to the entire population. These efforts private primary schools went from 77 percent of the total met with mixed success, and the private sector ex- in 1965 to 10 percent in 1979. Elsewhere the inability of panded to fill the void. In Malaysia, for example, physi- governments to keep up with demand or overcome dis- cians in private practice rose from 43 percent of the satisfaction with public school quality led to an increase total in 1975 to 90 percent of the total in 1990. But in private school enrollments. large parts of the population still lack access to basic ser- Historically, most medical services were privately vices, while others rely chiefly on private providers paid provided by midwives, traditional healers, and neigh- out of pocket. and 57 percent in Brazil. Eighty percent of health expen- be delivered privately (sometimes, as we see in Chapter ditures in Thailand are private. 4 for utilities, taking advantage of these new opportuni- In many settings unbundling the delivery of infrastruc- ties may require new regulatory arrangements) ture and social services can help achieve a better match To increase the transparency of the uses to which pub- between roles and capabilities. In bundled systems of deliv- lic money is being put (much harder when many diverse ery a diverse array of activitiesprivate and collective, sub- activities are bundled together within a monopoly pub- sidized and unsubsidized, competitive and monopolistic lic provider). are all undertaken by a single public provider. When services are unbundled, it becomes possible: Yet organizational changes will not do it all. Perhaps the most important change in the incentive environment To distinguish between activities that could be financed is to empower users themselves with "voice"not only to and delivered entirely through private markets, and work in partnership with providers where localized infor- those that have important collective elementsand to mation is key to efficient delivery, but also to monitor begin to shed the former providers' performance and to enforce, through the polit- To distinguish between those collective activities whose ical process, a commitment to quality. How this can be delivery should remain in public hands, and those achieved is the subject of Chapter 7. whose financing should be public and their delivery privatewith vouchers, contracting, and similar mech- Protecting the vulnerable anisms providing the bridge between the public and the Over the long term, rapid growth and investment in peo- private sector (Chapter 5 explores these options in ple will cut poverty dramatically. Yet regardless of a coun- more detail) try's income leveland regardless of the gains accruing to To take advantage of new opportunities for competi- the economy as a wholesome citizens will be left be- tion among the array of goods and services that can now hind, and others will suffer temporary hardship. This sec- SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 55 Figure 3.7 The balance of private and public education differs enormously worldwide Private education spending (percentage of total education spending) Austria Israel United States High-income Finland economies Switzerland South Africa Iran, Islamic Rep. of Cyprus Middle-income economies Colombia Indonesia Sri Lanka Kenya Low-income Uganda economies Vietnam Haiti 0 10 20 30 40 50 60 70 80 90 100 Note: Data are for various years between 1988 and 1993. Source: Psacharopoulos and Nguyen 1997. tion examines how states have wrestled with the challenge vulnerable in industrial societies. These three programs, of protecting the vulnerable. pensions especially, absorb a rapidly increasing share of national income, and rich countries around the world are A wide variety ofprotective measures revisiting some aspects of their welfare programs (Figure Table 3.1 offers a glimpse of the rich variety of initiatives 3.8). Even Sweden, where the commitment to the welfare governments have tried to protect the vulnerable in devel- state remains firm, and which has an unrivaled record in oping countries. All of these initiatives fall into one of two eradicating poverty, has embarked on wide-reaching broad categories: reforms to find a better balance between the social bene- fits and the heavyoften invisibleeconomic costs. Pension, unemployment, and other social insurance In Central and Eastern Europe and the former Soviet programs aim to support people whofor reasons of Union the state has traditionally provided a wide range of age, the business cycle, or other circumstancesare social services. Before their transition to the market these outside the wage economy for some part of their lives. states offered comprehensive benefits, but they differed Programs of social assistance aim to help the poorest in from those in industrial market economies in four respects. society, those who are barely able to support themselves. First, because the system was premised on full employment guaranteed by the state, there was no unemployment insur- In industrial countries the universal welfare state, ance. Second, social protection focused on those (such as which has influenced welfare programs around the world, the old and the disabled) who could not work. Third, ben- has blurred this distinction. Most of the main transfer efits were decentralized at the firm level. And fourth, in- programspensions, unemployment insurance, family kind subsidies (housing, energy) played an important role. assistance began during the 1930s and 1940s in re- With an unprecedented economic contraction and sponse to the Great Depression and World War II, and tight budgets accompanying transition, some countries in following the realization that the elderly were especially Central and Eastern Europe and the former Soviet Union 56 WORLD DEVELOPMENT REPORT 1997 Table 3.1 Social insurance, social assistance, and poverty-targeted programs in developing countries: Characteristics and lessons Coverage and Program type regional patterns Design issues and lessons Positive stories Pensions Nearly universal in transi- Actuarial imbalances, even in some countries with Innovative schemes in tion countries, very low in young populations, threaten macroeconomic stabil- Argentina and Chile Sub-Saharan Africa, ity, especially in transition countries, Brazil, and medium to high in Latin Uruguay. Transition countries need to Increase America. Pay-as-you-go pensionable age. Separate redistribution from schemes dominate. insurance. Family Included in middle- to Family size correlates highly with poverty in the assistance high-income countries as Central Asian republics but not in Eastern Europe part of social insurance. It and the rest of the CIS. Poverty incidence deter- is universal at the enter- mines the degree of progressivity. Where inci- prise level in transition dence is low, means testing is crucial to contain- economies. ing cost. Social Limited in transition coun- More suitable to countries with relatively low Chile's family subsidy assistance tries, rare in Asia, nonexis- poverty incidence. and old-age social (cash) tent in Latin America and assistance pension Africa. General price subsidies Open-ended price subsidies are fiscally unsustain- Tunisia's price subsidy Food dominate in Africa and the able, distortionary, and regressive. Leakages can reform, which reduced subsidies Middle East. Quantity be prevented by innovative targeting. Nutrition pro- costs by 2 percent of rationing is prevalent in grams are more cost-effective than quantity rations GOP and improved tar- South Asia. Food-for-work or general subsidies. Programs that set work geting; 1993 Food for schemes are used in Latin requirements are more cost--effective than rations. Education Program in America. Countries are Political economy often entails an urban bias. Bangladesh shifting toward food stamp and targeted programs. Housing Prevalent in transition Often regressive. Urban poor are best protected by Chile's one-time subsi- subsidies economies, mostly on- increasing and encouraging low-cost housing pro- dies for housing pur- budget; less prevalent in duction. Community organizations and coopera- chase on the private other regions, mostly off- tives have been more successful at targeting. Sub- market budget. sidies in the former Soviet Union complicate functioning of housing and labor markets. Energy Prevalent in transition In Asia, Africa, and Latin America gasoline subsi- subsidies countries and oil-producing dies largely benefit the nonpoor. They are also countries, such as somewhat regressive in transition countries Venezuela. because of their importance in the consumption basket of the nonpoor. EIim:nation of subsidies would affect the urban poor. Public works The Maharashtra Employ- Provide both insurance and assistance. They are India's Maharashtra ment Guarantee scheme appropriate in areas where poverty is transient scheme; Rorea's intro- in India and social funds and there is scope for unskilled labor-intensive duction and cancella- in Africa and Latin America projects. The program wage should not exceed the tion of work program are funded domestically prevailing market wage. In-kind payments attract and by international more women. donors. Credit-based Prevalent everywhere, Main problem is the inability to borrow in the Grameen Bank in programs especially in Africa, South absence of collateral. Programs should subsidize Bangladesh Asia, and Latin America. transactions costs but not interest rates, use local groups instead of direct targeting programs, orga- nize beneficiaries, and incorporate incentives to both borrowers and lenders to enforce repayment. Incorporate saving as a necessary component. Source: Adapted from World Bank 1996e. SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 57 Latin America, reflects the shift to nonfood aid and greater Figure 3.8 Pensions and other transfers have cooperation with nongovernmental organizations (NGOs) crept upward in the industrial countries and community-based groups in the delivery of targeted assistance. Labor-intensive public works programs have Percentage of GDP risen in popularity, particularly in South Asia and Africa. 12 In many countries, social insurance and assistance pro- grams have failed to achieve their objective of protecting 10 the vulnerable. Often they have resulted instead in trans- fers of resources to elite groups, sometimes with fiscally 8 destabilizing consequences. New approaches are begin- ning to emerge for both insurance and assistance. We examine each in turn. 6 Pensions Social insuranceoptions and hazards 4 The generosity of social insurance programs has sometimes Unemployment wrought havoc with long-term fiscal policy. As Table 3.2 2 insurance suggests, in many countries the liabilities implicit in indi- miIy a viduals' accrued pension rights far outweigh any reasonable 0 measure of the government's tax-raising capacity. 1965 1970 1975 1980 1985 1990 Demographic changes partly explain these ballooning pension liabilities. Aging populations account for more Note: See the Technical Note for details. Source: ILO, various than half of the expansion of pension and other welfare years; OECD 1996. benefits in the OECD countries over a recent thirty-year period. Ukraine and Hungary, too, have older popula- tions, which partly account for their high implicit pension are beginning to realize that this system of universal cov- debt. Demographic pressures on pension programs are erage is no longer affordable and must be replaced by likely to intensify especially rapidly in some developing more-targeted programs. Cash transfers as a percentage of countries. China's over-60 population will double from 9 GDP are high. But adapting the welfare system to the new to 1 8 percent of the total in thirty yearsa transition that conditions is proving politically difficult. In Poland trans- took a century in France and Britain. fers doubled from 9 percent of GDP in 1988 to 18 per- cent in 1993. In contrast to the OECD countries, the vast majority Table 3.2 Implicit pension debt in selected of developing countries have created "oasis" social insur- countries ance systems, which grant family benefits and pensions to (percentage of GDP) formal sector workers and civil servants. The size of this Implicit pension debt oasis increases with income per capita. It covers 6 percent Country of governments of the labor force in Sub-Saharan Africa, 23 percent in Uruguay 296 Asia, and 38 percent in Latin America. Formal unem- Hungary 213 Brazil 187 ployment insurance is rare, but the use of the public sec- Ukraine 141 tor as employer of last resort is a form of disguised unem- Turkey 72 ployment insurance. China 63 Developing countries have also experimented with a Cameroon 44 Peru 37 variety of social assistance measures for meeting the basic Congo 30 needs of the poorest. These have ranged from programs Venezuela 30 that bundle cash assistance and insurance, to price subsi- Senegal 27 dies (food, housing, energy) and labor-intensive public Mali 13 Ghana 9 works (Table 3.1). The design of social assistance pro- Burkina Faso 6 grams has often been heavily influenced by international Note: Data are for various years between 1990 and 1996 and aid. The prevalence of food aid from the United States in are net present values calculated at a discount rate of 4 the 1950s and 1960s, for example, led to the adoption of percent. many food-for-work programs, particularly in South Asia. Source: Kane and Palacios 1996. The emergence of social funds in the 1980s, especially in 58 WORLD DEVELOPMENT REPORT 1997 economy (generally the targets of insurance programs). Box 3.7 The new Chilean unemployment Experience suggests that failure to make this distinction is insurance scheme virtually certain to undermine both the fiscal viability of insurance programs (because the "insured" can lobby for Chile has a severance pay scheme but no system unfunded benefits) and the impact of assistance programs of unemployment insurance. The government has (because nontargeted groups are likely to capture re- drafted a law to create one (called PROTAC). The sources intended for the poor). design of this scheme departs from models prevail- With insurance clearly distinguished from assistance, ing in the OECD countries. The proposed scheme states can bring private participation and competition into tries to circumvent the disincentives to work often insurance systems previously dominated by public associated with unemployment insurance. It would monopolies. This can be done in several ways: create individual accounts to which workers and employers would jointly contribute 4.4 percent of The redistributive component of pensions can be Un- the worker's salary. These accounts would accumu- bundled from the saving component through a manda- late up to five months of salary and would be pri- tory multipillar system, with the saving pillar fully vately managed, possibly by the same institutions funded, privately managed, and publicly regulated. that now manage Chile's private pensions. A laid- Redistribution can be accomplished through a flat pub- off worker would receive severance pay of one lic pension (as in Argentina). month of salary per year of service, to a maximum States can introduce mandatory savings accounts for of five years, and could make up to five monthly unemployment insurance, as well as pensions (Box 3.7 withdrawals from his or her individual account describes a Chilean initiative along these lines). while unemployed. Workers who became unem- Companies and individuals can be allowed to choose ployed following a resignation would only be enti- between public and private providers, as in Japan, Sri tled to the monthly withdrawals. The state would Lanka, and the United Kingdom. regulate these unemployment insurance accounts Management of the assets of public insurance programs and guarantee a minimum unemployment benefit. can be contracted out to the private sector (as in As with pensions, this minimum guarantee would Malaysia). be provided only after funds in the account are States can enlist independent professionals, rather than exhausted. The individual account would thus act political appointees, for the boards of trustees of public as a deductible. programs. Of course, private provision of social insurance is only workable if financial markets are well enough developed But at their root the problems go beyond demograph- so that private intermediaries can readily match these ics. Civil servants in many countries view their pensions as long-term liabilities with long-term assets. Yet even in an entitlement, rather than a form of savings: they make poor regions such as Sub-Saharan Africa, thin capital limited contributions to a retirement scheme but receive a markets need not be a bar to the development of private full salary as pension after thirty to thirty-five years of ser- pension funds. Given an appropriateand enforceable vice. More generally, influential constituencies success- legal framework for financial sector development, coun- fully lobby for transfers from the budget, which they are tries could set up regional equity markets. This is a par- unwilling to see scaled back even in the face of a severe ticularly attractive option for countries of the CFA zone, fiscal crunch. Or, as in some African countries, public which share a common currency. Already some equity bureaucracies direct toward themselves resources intended markets in Sub-Saharan Africa compare favorably in for social insurance or for vulnerable groups. terms of market capitalization with those in Latin Ameri- Whatever the cause of these problems, unless social can countries that have recently privatized their pension insurance can be put on a sounder financial footing, either systems (such as Peru). the programs will collapse, or countries will be plunged into deep fiscal crisis, or both. An essential first step Sustainable approaches to social assistance toward reform is for governments to distinguish between Unlike social insurance, which can be self-financing, social the goals of insurance and those of assistanceespecially assistance requires direct expenditure of public funds. Bal- in developing countries where the gap is often vast ancing the objectives of poverty alleviation and fiscal pru- between the poorest citizens (generally the targets of assis- dence is thus vital to success. (Table 3.1 summarized the tance programs) and those who participate in the formal wide variety of approaches that have been tried.) In the past SECURING THE ECONOMIC AND SOCIAL FUNDAMENTALS 59 the debate was primarily over the relative merits of broad- The challenge of sustaining programs of social assis- based subsidies and means-tested programs. Today, the tance is political as well as fiscal: since the marginalized limitations of both have become more apparent. poor are politically weak almost everywhere, in times of Because means-tested programs (in which benefits are fiscal belt tightening even prudently designed programs set according to the recipient's income) are administra- risk losing support. Self-targeted programsespecially tively demanding, they are likely to achieve their goal at those that impose reciprocal obligations on recipients reasonable cost only in countries with strong institutional seem more politically resilient than those targeted more capability. But broad-based subsidies have also lost their narrowly, but they too are vulnerable. At its root, then, appeal: they are expensive and relatively inefficient at the taskexplored in Chapter 7-is to find ways of giv- reducing poverty. Housing and infrastructure subsidies, ing voice to the concerns of the poor, enabling them to for example, turn out to benefit higher-income house- become more effective advocates of their own interests. holds disproportionately (Figure 3.9). Food subsidies Strategic options: Doing better on the fundamentals can be more effective if they are targeted toward items consumed primarily by the poor. Tunisia has effectively Each of the four sets of economic and social fundamentals moved from a nontargeted to a targeted program by elim- poses distinctive challenges, but all have some challenges inating all subsidies on goods consumed disproportion- in common. ately by the nonpoor and, for those food products still First, prioritization is vital. As this chapter shows, in all subsidized, by differentiating product lines through dif- too many countries the state still does not provide the full ferences in packaging and the use of generic ingredients. complement of core public goods and services: a founda- These reforms have reduced the cost of food subsidies tion of lawfulness, a stable macroeconomy, the rudiments from 4 percent of GDP in the mid-1980s to 2 percent of public health, universal primary education, adequate by 1993, while still maintaining a food safety net for transport infrastructure, and a minimal safety net. At the the poor. same time states are overproviding a wide variety of goods With both means-tested and more broad-based assis- and services that private markets could supply in their tance programs increasingly in question, attention has stead. Especially in countries with weak institutional capa- shifted to self-targeted approaches. One approach is to bilities, the need is therefore urgent to focus the state's focus delivery on those localities, urban and rural, with role on the fundamentals. disproportionate numbers of poor residents. Another is to Second, skillful use of private, competitive markets and set the level of benefits low and build in some kind of quid voluntary activity can support development while sharply pro quo. Food-for-work programs incorporate these fea- reducing the burden on states with weak institutional tures. So, too, do lending programs for microenterprises capabilities. Market-led growth in a supportive incentive in poor communities. Box 3.8 illustrates how Indonesia, environment is fundamental. Additionally, markets can which has made huge strides in reducing poverty through provide a variety of private goods and services that in broad-based growth, is initiating a variety of self-targeted many countries have somehow wandered into the domain programs in an effort to eliminate poverty by 2005. of public provision, such as higher education, curative Figure 3.9 Housing subsidies in developing countries mostly fail to reach the needy Percentage of subsidies reaching households below the median ncome South Asia Latin America and Caribbean Sub-Saharan Africa Middle East and North Africa East Asia High-income OECD 0 10 20 30 40 50 60 Note: Data are for 1991-92. Source: World Bank 1996e. 60 WORLD DEVELOPMENT REPORT 1997 Box 3.8 Reducing poverty in Indonesiahow social assistance complements broad-based growth Indonesia's rapid and broad-based growth has had a 20,000 villagesthe poorest one-third of all spectacular effect on poverty reduction. Between 1970 Indonesian villagesto be used as seed capital for and 1990 the proportion of the population living income-generating activities. The program is com- below the official poverty line declined from 56 to 15 bined with public works programs. percent; other indicators of welfare, such as infant The Prosperous Family program, launched in 1996, mortality, showed similar improvement. The govern- aims to improve the conditions of families living in ment has now set itself the ambitious target of eradi- non-IDT villages, and whose living standards are cating absolute poverty within the next decade. The below a certain level, through small grants and sub- challenge is that the remaining poor are concentrated sidized credit. in isolated pockets of poverty with poor natural Under the Transmigration Program about 750,000 resource endowments, low population densities, and families, or over 3.6 million people, have been reset- other socioeconomic characteristics that make them tled at government expense from overpopulated Java difficult to reach. Several targeted interventions have to less populated outer islands. The program aims to been initiated in recent years, including the following: address landlessness as a cause of poverty and provide new settlers with agricultural land and other benefits. The Inpres Desa Tertinggal (IDT) program, launched The Kampung Improvement Program is targeted at in 1994, is directed at villages that the country's improving the provision of social services and infra- development has left behind. The program distrib- structure to densely populated, low-income, urban utes grants totaling $200 million per year among neighborhoods. health services, and pensions and other forms of insurance. Finally, states should seek ways to enhance the credibil- In a range of other areasusing social funds for poverty ity of their actions. In the short run, while weak domestic alleviation, enhancing the quality of primary education, institutions are being reinforced, stronger ties with external encouraging participation by NGOs and communities actorsfor example, through stabilization programs with reform can greatly improve service delivery. Countries the IMFcan help governments signal their commit- with weak public institutions should assign high priority ment. In the long run, however, as Part Three explores in to finding ways to use markets and involve private firms depth, the vital challenge is to build homegrown commit- and other nongovernmental providers in service delivery. ment mechanisms, rooted in domestic institutions. FOSTERING MARKETS: LIBERALIZATION, REGULATION, AND INDUSTRIAL POLICY EW DJSPUTE THE CENT1t&L ROLE OF THE STATE IN Many countries with weak institutional capability are F- securing the economic and social fundamentals saddled by their history with governments whose reach is discussed in Chapter 3. There is much less agreement, overextended; for them, privatization and market liberal- however, about the state's precise role in regulation ization is a key part of the policy agenda. As capability and industrial policy. A counterpart to the rise of state- develops, public organizations and officials will be able to dominated development strategies in the early postwar take on more challenging collective initiatives, to foster years was a dramatic expansion in government regulation markets and to make increasing use of efficientbut dif- in many countries. As countries have liberalized, those ficult to manageregulatory tools. aspects of the regulatory framework that have proved Privatizing and liberalizing markets in counterproductive are being abandoned. But govern- overextended states ments are learning that market reforms and fast-changing technology pose their own regulatory challenges. States Interest has revived in finding ways for the government to cannot abandon regulation. The task, rather, is to adopt work with the private sector in support of economic devel- approaches to regulation that fit not merely the shifting opment, and to provide regulatory frameworks supportive demands of the economy and society but, critically, the of competitive markets. Yet in all too many countries, state country's existing institutional capability. and market remain fundamentally at odds. Private initia- Attention to the proper match between the state's tive is still held hostage to a legacy of antagonistic relations role and its institutional capability helps reconcile some with the state. Rigid regulations inhibit private initiative. seemingly clashing prescriptions for state action. Many, And state enterprises, often buttressed by monopoly privi- for example, would argue that, in complex industries such leges, dominate economic terrain that could more fruit- as telecommunications, regulators ought to have consider- fully be given over to competitive markets. At the extreme, able flexibility in devising and implementing market rules. a mass of inefficient state enterprises blocks private Yet where institutional capability remains weak, the scope dynamism entirely, even as it imposes an unmanageable for flexible initiatives is limited; the focus should instead fiscal and administrative burden on the rest of the public be on winning credibility with firms and citizens, convinc- sector. In such countries the first step toward increasing ing them that the state will follow through on its com- the state's effectiveness must be to reduce its reach. mitments and will refrain from arbitrary and capricious The recent economic performance of such countries as action. China and Poland provides dramatic evidence of the ben- The same applies even more forcefully to more inter- efits of shrinking the state in former centrally planned ventionist policiesthose aimed at not merely laying the economies. But relaxing government's grip, whether that foundations of industrial development but actively accel- grip is maintained through public ownership or regula- erating it. In principle, there seems to be room for gov- tion, can also yield large dividends in more mixed eco- ernment to play such a role. But in practice its scope for nomies. It can: doing so turns out to rely heavily on a range of stringent institutional conditions being fulfilled. Except where role Free up public resources for high-p riorily activities. and capability have been skillfully matched, activist indus- Diverting subsidies away from money-losing state trial policy has often been a recipe for disaster. enterprises and toward basic education would have 62 WORLD DEVELOPMENT REPORT gg Table 4.1 Estimates of welfare gains from deregulation in the United States (billions of dollars) Gains to Gains to Total Further potential Industry consumers producers gains gains Airlines 8.8-14.8 4.9 13.7-19.7 4.9 Railroads 7.2-9.7 3.2 10.4-12.9 0.4 Trucking 15.4 4.8 10.6 0.0 Telecommunications 0.7-1.6 0.7-1.6 11.8 Cable television 0.4-1.3 0.4-1.3 0.4-0.8 Brokerage 0.1 0.1 0.0 0.0 Natural gas 4.1 Total 32.6-43 3.2 35.8-46.2 21.6-22.0 .Not available. Source: Winston 1993. increased central government education expenditures and other objectives preferably at most secondary in by 50 percent in Mexico, 74 percent in Tanzania, and importance 160 percent in Tunisia. Clarification of the criteria to be used in assessing Pave the way to better, cheaper services. Divestiture of which regulations are useful, which should be dis- state assets had positive effects in all but one of twelve carded, and which should be strengthened to com- carefully studied cases in Chile, Malaysia, Mexico, and plement privatization the United Kingdom. The benefits came in the form of Preparation of financial statements and public bud- increased productivity and investment as well as more gets (including information on borrowing from efficient pricing. Deregulation in five hitherto tightly banks) to assess which state enterprises are money- regulated sectors in the United States had by 1990 losers and uncover the reasons for their losses yielded gains of $40 billion (Table 4.1). In Argentina, Specification of open and competitive mechanisms liberalizing harbor terminals in Buenos Aires led to an (such as auctions) for divesting state enterprises. 80 percent reduction in fees. Unlock opportunities for private sector development. Such efforts have an added rationale. Often they will Excessive regulation can inhibit market entry, fuel the show whether or not a country is truly ready for reform growth of informal activity, and even create new indus- whether key political actors want reform and find it polit- tries solely devoted to helping firms navigate the regu- ically feasible to translate that desire into action. If politi- latory maze. Eliminating these excesses enables markets cal will is lacking, further efforts will be wasted. Indeed, to function more flexibly, at lower transactions costs. they may prove counterproductive if interpreted as another in a long line of arbitrary shifts in policy. The challenges of scaling back the overextended state With the initial preparation done, the second phase of are as much political and institutional as they are techni- reform is to put in place a business environment that sup- cal. Success relies on the ability to proceed with reform in ports competitive private markets. Such an environment the face of opposition from powerful groups who benefit includes rules of the game that facilitate entry and com- from the status quo. Chapter 9 examines how reforms in petition, and a complementary institutional, legal, and general can most effectively be initiated and sustained. regulatory framework that can undergird property rights Here we focus more narrowly on programs of market lib- and markets, including (notably) financial markets. eralization and privatization. The economic advantages of early reform of the busi- Initiatives to foster market liberalization and privatiza- ness environmenteven before privatizationare sub- tion can be segmented into three overlapping phases: pre- stantial. One advantage is that fostering external and paring for reform, establishing an enabling business envi- domestic competition ensures that many of the benefits of ronment, and privatizing (or liquidating) state enterprises. privatization will be passed on to consumers, rather than Transparency is the vital ingredient as governments begin simply result in a transfer from public coffers to private to prepare for reform. Ideally, transparent preparation monopolies. Otherwise the latter are likely to become includes: powerful, entrenched interests, willing and able to stifle subsequent efforts to introduce more competition into the An explicit statement of the main objectiveto un- economy. A second advantage is that, if clear regulatory leash a competitive market economywith fiscal structures are in place, bidders will have a better idea of FOSTERING MARKETS: LIBERALIZATION, REGULATION, AND INDUSTRIAL POLICY 63 the economic potential of companies being privatized A strategy of "growing out" of state dominance appears the risk premium will be lowerand government will to have worked in some East Asian economies. But else- receive higher bids. where economic and political considerations will favor More broadly, liberalization of the business environ- keeping privatization on the front burner. Delay imposes ment can be a powerful catalyst, setting off a virtuous three major economic costs. First, money-losing state enter- spiral whereby each reform makes the next one easier. The prises may continue to drain money from the public coffers stronger the business environment, the greater the range (or from banks in the form of never-to-be-repaid "loans"). of opportunities and supports available to entrepreneurs, Unless such losses can be contained, the resulting fiscal bureaucrats, and workersand thus the weaker the polit- instability can undermine an entire reform program. Sec- ical opposition to dismantling dysfunctional rules and ond, anticipating privatization down the road, managers agencies and liquidating or privatizing state enterprises. and workers in state enterprises can be tempted to steal the The challenge is finding a way to set this virtuous spiral in company's most valuable assets while the going is good. motion. For at the outset those who prosper under the Third, poorly performing state enterprises may obstruct dysfunctional system will have much to lose, while the liberalization and restructuring in other sectors. In Zambia eventual winners are unlikely to have reached the critical market liberalization created opportunities for smallholder mass needed to lobby for their own interests, Box 4.1 farms to expand production and exports of cotton. But describes how Mexico was able to overcome initial resis- before being exported, cotton must be processed, and for tance to the rollback of regulatory controls. some years after liberalization virtually all the country's Because it takes time for the business environment to processors were under the control of a monopoly state become supportiveand because privatization becomes enterprise. Once the sector was restructured, the pace at easier as the environment improvesreformers may be which farmers and businesses took advantage of new mar- tempted to give privatization a backseat. This is precisely ket opportunities picked up dramatically. the approach adopted by China and, in earlier years, Given the importance of keeping privatization on the by the Republic of Korea and Taiwan (China). In the front burner, its sequencing in relation to liberalization early 1960s, state enterprises accounted for about half of thus poses some difficult dilemmas. On the one hand, manufacturing production in Taiwan (China) and one- privatization will yield greater economic benefits, and quarter in Korea. By the mid-i 980s their share had fallen impose fewer hardships on society, if it is preceded by to about 10 percent in both economiesnot as a result of liberalization and regulatory reform. On the other hand, privatization, but because of the rapid expansion of their the longer privatization is delayed, the more entrenched private sectors. management of state enterprises can become. Box 4.2 Box 4.1 Mexico's deregulation czar In 1988 the president of Mexico appointed a "deregu- Unequivocal presidential support, signaling to both lation czar." Each month this official reported directly bureaucrats and citizens the need to comply with to the president and his economic council of ministers. the czar's decisions Every business in Mexico, large or small, had equal The fact that his decisions could be overruled only access to the czar's office to complain about burden- at the highest level of government some rules and regulations. When the office received a The setting of tough penalties for officials who complaint, it was obliged to find out why the rule failed to implement the rulings existed, how it interacted with other regulations, and The time limit, which ensured quick and visible whether it should continue in effect. The office oper- results ated under a strict timetable: if it did not act to main- The czar's staff, who were skilled in the economic tain, revise, or abolish the disputed rule within forty- consequences of regulations, their interactions with five days, the rule was annulled automatically. other regulations, and their administrative require- The work of the deregulation czar over his first four mentsno one person can effectively carry out a years is widely credited with greatly accelerating Mex- government-wide program of deregulation ico's reforms. It provided struggling private business- Finally, the fact that the czar won credibility with people with an effective, responsive champion at the officials and with the public by giving a fair hearing highest level of government. The factors behind this to the powerless and the influential alike, and setting success include: a consistent record of impartiality. 64 WORLD DEVELOPMENT REPORT 1997 Box 4.2 Six objections to privatizationand how to address them "We can't throw public sector workers into the street. It's "Our localprivate sector is too weak. Without state enter- wrongand they won't stand for it." prises, our economy will grind to a halt. Winning the acquiescence of employees is essential Certainly, privatization is easier if a well-function- to successful privatization. Some countries have given ing market economy, including financial markets, is shares to employees or privatized through employee already in place. Thus, a key complement (and, if and management buyouts. Others have offered gener- appropriate, antecedent) to privatization is market lib- ous severance pay. Privatization becomes easier as eralization, perhaps accompanied by the activist initia- countries develop programs to protect the vulnerable, tives to foster markets described later in this chapter. of the kind described in Chapter 3. Even so, in most settings it is precisely the heavy hand of the overextended state that is restraining private "Privatization is just another way for powerful politicians activity--the objection confuses cause with effect. and businessmen to scratch each other's backs, and get rich at the expense of the people." "All that privatization will do is replace a public monop- oly with a private monopoly." Process matters. Privatization must be based on Regulatory reform is another important accompani- competitive bidding, with the criteria for selecting ment to privatization: deregulation to remove artificial buyers carefully specified in advance. And it all should monopoly privileges, and development of a regulatory be done in the open, in full view of the media and system that credibly restrains the abuse of economic citizens. power in noncompetitive markets. "Our citizens won t accept our handing over precious "Why put ourselves through this trauma? Let just man- national assets to foreign (or local) fat cats." age our state enterprises better. Broad-based ownership can help win popular sup- True, if governments are willing to put hard budget port for privatization. One approach, adopted in the constraints in place, to allow competition from private Czech Republic, Russia, and Mongolia, is to distribute firms, and to give managers appropriate incentives, privatization vouchers to citizens to be redeemed for the performance of state enterprises can improve. The shares. Another approach, adopted in Argentina, Chile, sad reality is that, although some committed govern- and the United Kingdom, is to make an initial public ments have reformed their state enterprises in the short offering of shares to citizens at attractive prices. Both term, making these reforms stick is much harder. approaches can be designed to make room for a strong World Development Report 1983 spotlighted a number strategic partner with the incentive and expertise to of well-performing state enterprises around the world; effectively restructure the enterprise. by 1993 a majority of these had sunk into decline. describes how reformers opting to push ahead with priva- state needs to own iess, and although there is no good eco- tization have tried to contain the risks. nomic reason for state ownership to persist in tradable- goods industries, there is no single "correct" stage in the Rolling back overextended states: Two central lessons reform program to start privatizing. The appropriate timing Experience worldwide with attempts to scale back overex- will depend on the dynamics of reform in each country. tended states suggests that success contains two vital ingre- dients. First is a commitment to competitive markets Better regulation and an accompanying willingness to eliminate obstacles to Skillful regulation can help societies influence market out- their operation. Market liberalization enables new entrants comes to achieve public purposes. It can protect the envi- to create jobs and wealth. It also eases the difficulties of ronment. It can also protect consumers and workers from privatization while increasing the potential economic the effects of information asymmetries: the fact that gains. The second lesson is that, although the overextended banks, for example, know much more about the quality of FOSTERING MARKETS: LIBERALIZATION, REGULATION, AND INDUSTRIAL POLICY 65 their portfolios than do depositors, or the fact that busi- they can continue to attract depositsand even aggres- ness managers may know more about health and safety sively pursue them by offering favorable interest rates. risks in production or consumption than do workers or Failing banks often engage in ever-more-reckless gambles consumers. Regulation can also make markets work more to salvage their position, throwing good deposits after efficiently by fostering competition and innovation and bad, and driving up their losses before the inevitable crash. preventing the abuse of monopoly power. And more And third, because banks' balance sheets can be difficult broadly, it can help win public acceptance of the fairness to interpret, especially because a rising share of their port- and legitimacy of market outcomes. folios may now be taken up with derivatives and other With economic liberalization, many areas of regulation new financial instruments that are hard to monitor. have been recognized as counterproductive, and wisely This information asymmetry can be destabilizing. abandoned. Yet in some areas the traditional rationales for Depositors, fearing for the safety of their funds, might regulation remain, and market liberalization and privati- rush to withdraw them when they begin to hear stories zation have themselves brought new regulatory issues to about troubled banks. Bank failures tend to be conta- the fore. The challenge, illustrated here with reference to gious. When one insolvent bank goes under, nervous three important regulatory domainsbanking, utilities, depositors may start runs on others. As liquidity drains and the environmentis not to abandon regulation alto- out of the system, even solvent banks may be forced to gether. Instead it is to find regulatory approaches in each close. And a systemwide run can have severe macroeco- country that match both its needs and its capabilities. nomic consequences. For all these reasonsthe difficul- ties in assessing a bank's financial health, the adverse Some new rationales for regulation spillover and distributional effects of bank failures FINi\NCE: FROM CONTROLS TO PRUDENTIAL REGULATION. banks' behavior needs to be tempered by regulatory and Our understanding of financial sector development has other public actions, outlined later in this section. changed dramatically over the past decade. We now know UTILITIES: REGULATION WITH COMPETITION. For util- that the depth of a country's financial sector is a powerful ities, too, regulation has taken on renewed prominence. predictor and driver of development. Just as important, we Here, however, the reason is revolutionary technological know that the control-oriented regulation widely adopted and organizational change, not just conscious shifts in in the early postwar yearsdirecting subsidized credit to policy. The argument for utility regulation used to be favored activities at very negative real interest rates, limiting straightforward. Utilities were natural monopolies. Con- the sectoral and geographic diversification of financial sequently, unless they were regulated, private utility oper- intermediariesmay often work against financial deepen- ators would act as monopolists, restricting output and ing. The near-universal response has been to move away raising prices, with harmful consequences for economy- from controls over the structure of financial markets wide efficiency and income distribution. Today, changes and their allocation of finance, and embark on a process of in technology have created new scope for competition, liberalization. but would-be competitors may need special reassurance Yet liberalization, in the financial sector is not the same from regulators before entering. as deregulation. The case for regulating banking is as com- In telecommunications, dozens of countries through- pelling as ever. Only the purpose has changed, from chan- out the Americas, Europe, and Asiaplus a few in Africa, neling credit in preferred directions to safeguarding the including Ghana and South Africahave introduced health of the financial system. competition in long-distance, cellular, and value added The banking system needs effective prudential controls (fax, data transmission, videoconferencing) services. A few because banks are different. Without appropriate regula- countriesChile and El Salvador, for exampleare even tion, outsiders will be less able to judge for themselves a exploring options for competition in local fixed-link net- bank's financial health than that of a nonfinancial com- works. Electric power generation (but not transmission or pany. Why? First, because outstanding loans are banks' distribution) is also now viewed as an arena for competi- primary assets. So long as banks receive interest on their tion. In China, Indonesia, Malaysia, and the Philippines, loans, outside observers may well judge their portfolios to private investors are adding generating capacity through be healthy, even if (unknown to the observers) the bor- independent power projects, alleviating acute shortages rowers lack the resources to repay the principal or, worse, and enabling private finance to fill the gap left by short- are effectively bankrupt and are only keeping up the inter- falls in public resources. est payments by taking out new loans. Second, because In this new environment the degree of natural monop- unlike many companies, banks can be hopelessly insolvent oly has been drastically reduced (although perhaps not without running into a liquidity crisis. So long as insol- eliminated entirely). But regulation is still crucial, for two vent bankers can disguise their condition to outsiders, reasons. First, it can facilitate competition. Consider the 66 WORLD DEVELOPMENT REPORT 1997 problem of interconnection. By failing for more than a ambiguity and negotiated outcomes rather than pre- decade to establish workable rules to allow different net- dictable and consistent implementation. Poor communi- works to connect with one another, Chile's telecommuni- ties daily confront a dismal bargain, borrowing immediate cations regulators seriously obstructed competition, leaving survival against long-term environmental degradation. Pri- dominant incumbent firms in control of how the system vate firms weigh the predictable costs and the benefits of evolved. After numerous court disputes a multicarrier sys- complying with well-defined environmental regulations tem was introduced in 1994: customers can now choose against the prospect of cutting costs by avoiding regulation their long-distance provider. Within months, six new altogether. Consequently, politicians may often conclude providers had entered the market, and the price of long- that environmental inaction (perhaps veiled behind the distance calling had dropped by half. Similar interconnec- appearance of activism) is the politically expedient course. tion problems can arise in the electric power industry when In this climate of ambiguity, as later sections will show, generators supply customers through common-carrier purely technocratic approaches to environmental regula- transmission lines. This is an issue that Argentina, among tion have little hope of success. Especially in developing others, has had to grapple with in the wake of privatization. countries where the institutional foundations for regulation A second reason for improved regulation is that com- are weak, the potential for successfully containing the envi- petition may not suffice to insure private investors against ronmental hazards of unfettered private markets may be "regulatory risk": the danger that decisions by regulators greater with approaches that rely at least as much on pub- or other public agencies will impose new and costly lic information and citizen participation as on formal rules. demands some time down the line. A utility's assets are unique to its business, and nonredeployable in other uses. Where capability is strong, regulation can raise credibility This means that utilities will be willing to operate as long and efficiency as they can recover their working costs. That, in turn, So how should states respond to continually changing, makes them peculiarly vulnerable to administrative expro- and often conflicting, regulatory demands? Three princi- priationas when, for example, regulators set prices ples are key. First, different ways of regulating have differ- below long-run average cost. Consequently, countries ent costs and benefits, which countries should assess ex- without a track record of respecting property rights may plicitly before proceeding. Second, this assessment should fail to attract private investors into utilities, regardless of also incorporate the administrative dimension: some any commitment to competition in utility markets. As the forms of regulation are intensive in their requirements for next sections show, a well-designed mechanism that com- information, whereas others require much less (or much mits the regulator to a clearly defined course of action can more easily monitorable) information; likewise, some offer the reassurance that potential investors need. regulatory approaches depend on command-and-control, THE ENVIRONMENT: BALANCING SCIENCE, ECONOM- others more on market-like mechanisms. In general, ICS, AND CITIZEN PRESSURE. Economists have long recog- information-light and market-like approaches are easier to nized pollution to be a negative externality. Without some implement, and often at least as efficient. Third, states dif- form of regulatory protection, the environment can fer markedly both in their institutional capabilities and in become an innocent victim of bad business practices. Buy- the structure of their economies. Their approaches to reg- ers seek goods that are attractively priced, and producers ulation should reflect these differences. seek ways of providing these goods at lower cost to them- We begin to show how these principles can be applied selves than their competitors can provide them. Unless in practice by considering some "best-case" scenarios: the there is some countervailing incentive, the temptation to range of regulatory options for banking, utilities, and the cut corners by producing in a cheaper but environmen- environment that only work well with strong institutions. tally "dirtier" way can be great. These institution-intensive approaches combine three Even countries with strong institutions find environ- central elements (Table 4.2): mental regulation immensely challenging. Noxious fumes, poisoned water, earsplitting noiseand their conse- Relying on public administrators to manage complex quencesare easy to spot. But the costs of many other technical problems forms of environmental damage are diffuse, and may be Giving regulators considerable flexibility to respond to invisible even to those closest to the source of pollution, changing circumstances who may suffer serious long-term effects. Polluting emis- Using an array of checks and balances to restrain arbi- sions can also be tricky to measure. And the environmen- trary behavior by regulatory agencies and build their tal consequences may depend heavily on the demographic credibility. and ecological features of the surrounding area. A further complication is that the political incentives of BANK SUPERVISION. Banking sector regulation around community, business, and political stakeholders can foster the world tends to be institution intensive. Later sections FOSTERING MARKETS: LIBERALIZATION, REGULATION, AND INDUSTRIAL POLICY 67 Table 4.2 The variety of regulatory experience Utilities regulation Environmental regulation Financial regulation lnstitutionintensive Price-cap regulation, with the Precise rules (command-and- Detailed regulation options regulator setting the price control or, preferably, monitored by competent, adjustment factor incentive based) established impartial supervisory by the regulatory agency or authorities (possibly Regulation by independent legislature including some deposit commission, with public insurance) hearings Institution-light options Regulation based on simple Bottom-up regulatory Incentives structured so that rules, embodied in approaches: public bankers and depositors have transaction-specific legal information, local initiatives a substantial stake in agreements and enforceable to strengthen citizens' voice, maintaining bank solvency domestically or through an and initiatives by local international mechanism authorities discuss some new ideas for maintaining the solvency of that their owners have something to lose in the event of banks where supervisory agencies are weak. In many failure. Authorities should also be required to consider countries, however, formal supervision remains a vital the qualifications and track record of proposed owners bulwark. The idea behind it is that well-designed regula- and managers. tion, monitored and enforced by competent supervisory Restraints on insider lending. Restrictions on lending to authorities, can overcome the information asymmetries bank insiders can cut down on fraudulent loans. Simi- inherent in banking, and detector at least contain larly, many countries also limit a bank's lending to a potentially ruinous banking crises (Box 4.3). Key ele- single client (commonly to a maximum of 15 to 25 ments of such systems include: percent of the bank's capital); this prevents any one client from becoming "too big to fail," prompting the Capital adequacy and entry criteria. Minimum capital bank to make unsound loans solely to keep that client requirements impose discipline on banks by ensuring afloat. Box 4.3 How government supervision averted financial disaster in Malaysia In 1985 a sudden fall in world commodity prices billion in assers, but subject to much less supervision reversed Malaysia's decade-long boom. The Malaysian than the commercial bankswere in severe distress. stock index, which had surged from 100 in 1977 to 427 Bank supervisors at Bank Negara, Malaysia's central in early 1984, fell below 200 by early 1986; the value of bank, devised a series of complex rescue packages for prime commercial property in Kuala Lumpur fell by the three ailing commercial banks and the twenty-four even more. Banks, which had moved heavily into real cooperatives. All told, losses as a result of the bank- estate lending in the boom years, faced the specter of ing crisis amounted to 4.7 percent of Malaysia's 1986 rising nonperforming loans and doubtful debts. gross national product (GNP). Because Malaysia had maintained a fairly high Malaysia's experience underscores the value of good degree of banking supervision, provisioning for non- supervision. Losses in the tightly supervised banking performing loans rose rapidly: from 3.5 percent of total sector amounted to only 2.4 percent of depositsfar lending in 1984 to 14.5 percent by 1988. Even so, less than the 40 percent of deposits lost in the lightly supervisory inspections in 1985 identified three com- supervised nonbank cooperatives. And macroeco- mercial banks whose solvency was threatened by prob- nomic disaster was averted. The economy recovered in lem portfolios (but whose management was reluctant 1987, and stock and property prices and bank balance to acknowledge the full scope of the problem). Addi- sheets recovered with it. Prompt action had made it tionally, twenty-four nonbank deposit-taking coopera- possible to identify and address problems early, while tiveswith over 522,000 depositors and about $1.5 disciplined rescue was still affordable. 68 WORLD DEVELOPMENT REPORT gg I Rules governing asset class/ication. Requiring that banks tors, without having these prerequisites in place. The con- classify the quality and risks of their loan portfolio sequences have often been disastrous. A recent World according to specific criteria, and define and identify Bank study identified over 100 major episodes of bank nonperforming loans, can provide early warning of insolvency in ninety developing and transition economies problems. from the late 1970s to 1994. In twenty-three of the thirty Audit requirements. Minimum auditing standards and countries for which data were available, the direct losses disclosure requirements can make reliable and timely sustained by governments in these episodes exceeded 3 information available to bank depositors, investors, and percent of GDP (Figure 4.1). In absolute terms, losses creditors. were largest in the industrial countries: official estimates put nonperforming loans in Japan in 1995 at about $400 Building a robust system of prudential regulation and billion; the cost of cleaning up the 1980s U.S. savings and supervision is administratively demanding. It means hav- loan debacle came to $180 billion. But in relative terms ing reasonably reliable accounting and auditing informa- the largest losses were in Latin America: Argentina's losses tion on the financial health of a bank's borrowers. And it in the early 1980s amounted to more than half of its means having a sufficient number of supervisors, not only GDP, and Chile's exceeded 40 percent. Later sections skilled enough to do their job but politically independent examine some ways to guard against bank failure that are enough to do it impartially. not so heavily dependent on formal supervision. Many countries have relied exclusively on prudential PRICE CAPS FOR REGULATING UTILITIES. The use of regulation and supervision to undergird their banking price caps in utility regulation illustrates both the scope of Figure 4.1 Bank crises are all too common and carry enormous fiscal cost Direct cost of banking crisis (percentage of GDP) Argentina 1980-82 Chile 1981-83 Uruguay 1981-84 J Israel 1977-83 COte d'lvoire 1988-91 Venezuela 1994-95 Senegal 1988-91 Benin 1988-90 Spain 1977-85 Mexico 1995 Mauritania 1984-93 Bulgaria 1995-96 Tanzania 1987-93 Hungary 1991-93 J Finland 1991-93 Brazil 1994-95 Sweden 1991 Ghana 1982-89 Sri Lanka 1989-93 Colombia 1982-87 Malaysia 1985-88 Norway 1987-89 United States 1984-91 0 10 20 30 40 50 60 Source: Caprio 1996. FOSTERING MARKETS: LIBERALIZATION, REGULATION, AND INDUSTRIAL POLICY 69 authority of an independent regulator and the role of The shortcomings of top-down environmental regula- institutional checks on arbitrary action. Price-cap regula- tion have been even more obvious in developing coun- tion gives the utility an incentive to be efficient and can tries, many of which responded to the surge of interest encourage innovation, but it rests substantial discre- in environmental issues by establishing new regulatory tionary power with the regulator. In the United Kingdom, agencies modeled on this approach. Poland's regulatory which pioneered price-cap regulation, regulators impose agency, for example, although technically competent, an overall ceiling on utility prices, based on the annual found it had limited leverage in negotiations with plant rate of inflation minus an adjustment factor. The regula- managers in communities that were heavily dependent on tors decide the level of the adjustment factor, which they one or a few large enterprises, which funded many com- can change at defined (usually five-year) intervals. munity services. Chile's highly regarded environmental The U.K. regulators are constrained by carefully agency has spent four years trying, and failing, to imple- designed checks and balances: any decisions that the util- ment a system of tradable permits for industrial emissions, ity opposes must be cleared by both the Monopolies and because of difficulties in setting and later measuring base- Mergers Commission and the Secretary of State for Trade line emissions. and Industry. These checks have been strong enough to permit a highly flexible approach to regulation while still The regulatoiy 'fit" when institutions are weak attracting substantial private investment. If countries with Countries with weaker institutions face a much greater weaker checks and balances sought to adopt this type of risk that relying on administrators' skill and discretion regulation, private investors might reasonably expect the will result in a mass of unpredictable and inconsistent reg- adjustment factor to increase dramatically at the first ulation. The challenge for financial and environmental renewal of the price cap. Consequently, investors either regulation in such countries is to prevent costly oppor- would not invest or would demand very high rates of tunism by private actorsbe it banking fraud or pollu- return to ensure a quick payback. tionwhen the regulatory agencies' authority cannot be INSTITUTION-INTENSIVE APPROACHES TO ENVIRON- relied upon. With regard to utilities, the trick will be to MENTAL REGULATION. A central challenge for environ- convince potential investors that regulators will not mental regulation has always been finding ways to com- engage in arbitrary and expensive rule changes. Table 4.2 bine technical expertise with political legitimacy, to avoid summarized some of the regulatory options available in the sense that scientists and technocrats are making deci- such cases, each of which is discussed below. sions without regard for community or broader public FOSTERING INCENTIVES FOR PRUDENT BANKING. The concerns. In industrial countries, strong institutions have incentives and interests of bank owners, managers, and been the key to striking this balance. In France, Ger- depositors can themselves be a vital complement to super- many, and the United Kingdom, for example, elected leg- vision if they are aligned to be compatible with prudent islators delegate the details of policy to environmental banking. The history of banking offers examples of some authorities, who consult with affected parties and respond unusually sophisticated self-enforcing arrangements for to direct political pressure. Decisions by the U.S. Envi- winning credibility. More recently, the World Bank and ronmental Protection Agency are, like many other execu- the European Bank for Reconstruction and Development tive agency decisions, legally binding only if the public is collaborated on a project in Russia designed to influence given advance notice of rule changes and interested parties banks' incentives: banks were chosen to on-lend funds pro- are able to make formal comments. The Dutch govern- vided they agreed to submit to annual audits by interna- ment provides more than half the funding for thirty to tional accounting firms and to adhere to prudential norms. forty environmental NGOs and routinely consults them Using regulation to raise the stakes for bankers is and other affected parties when preparing environmental another institution-light way to protect the health of the legislation. banking system. It is less expensive to monitor the net Viewed through the narrow lens of economic effi- worth of a bank than to monitor each of its transactions. ciency, even these mechanisms have produced imperfect A bank that has adequate net worth will have the right outcomes. Both Germany and the United States, for exam- incentive to behave prudently. The following measures ple, have been strikingly successful in reducing emissions can all help raise net worth, and hence the cost of bank of some important pollutants. Yet partly because of the failure to bankers: need to be seen as responsive to citizen concerns, both countries continue to rely overwhelmingly on command- Very strict capital requirements on banks: not the and-control approaches to environmental regulation, even modest 8 percent of deposits recommended by the where market- and incentive-based regulation could Basel Committee for industrial countries, but 20 per- achieve similar gains at much lower cost. cent or more 70 WORLD DEVELOPMENT REPORT 1997 Tough restrictions on entry, in part to raise the fran- sustained private investment, but only at the cost of lim- chise value of a banking license for incumbents and iting flexibility. Since independence the industry has been thereby strengthen the incentive to stay in business on a regulatory roller coaster, thriving when the country Ceilings on interest rates for deposits, not only to keep was willing to forgo flexibility, but lagging behind when banks in business but also to create powerful incentives the mood shifted in favor of greater discretion. for banks to extend branch networks, so as to boost Unlike Jamaica, the Philippines has until recently been total deposits and accelerate financial deepening. unable to put in place a regulatory commitment mecha- nism capable of convincing private investors that the rules Another option that builds on prudential incentives is of the game would endure beyond the term of the current punitive contingent liability for bank owners, directors, president. Consequently, from the late 1950s until the and managers in the event of bank failure. Before the mid- early 1990s the country's private telecommunications 1930s, U.S. authorities routinely imposed double liabili- utility rode a political investment cycle. Investment was ties on the shareholders of failed banks. Perhaps in part as high immediately following the inauguration of a govern- a consequence, some 4,500 voluntary bank closures ment aligned with the group controlling the utility, but occurred between 1863 and 1928, but only 650 bank liq- tailed off in that government's later years, and stagnated uidations. New Zealand today imposes stringent require- in periods when relations with those in power were more ments on banks for transparent reporting, coupled with distant. In the electric power industry, the government tough sanctions on bank managers who violate them. resolved the problem of commitment by agreeing on rigid COMMITMENT MECHANISMS TO ATTPACT PRIVATE legal "take-or-pay" agreements with private investors, UTILITY INVESTORS. The Jamaican telecommunications sometimes enforceable offshore. Another option is to use industry vividly shows how private investment can affect third-party guaranteessuch as those offered by the the interplay between institutional capability and regula- World Bank Groupto protect private investors and tory roles (Box 4.4). There the government was able to use lenders against noncommercial risks, including the risk of regulatory commitment mechanisms capable of attracting administrative expropriation. Box 4.4 Telecommunications regulation in Jamaica During much of the colonial period and in the years commission decisions. The result was that price con- immediately following independence, the terms under trols became progressively more punitiveto the point which Jamaica's largest telecommunications utility that in 1975 Jamaica's largest private telecommunica- operated were laid out in a legally binding, precisely tions operator was relieved to sell its assets to the gov- specified, forty-year license contract. Then as now, the ernment. In 1987, after a decade of underinvestment, ultimate court of appeal for Jamaica's independent Jamaica reprivatized its telecommunications utility, this judiciary was the Privy Council in the United King- time using a precisely specified, legally binding license dom. This system was adequate to ensure steady growth contract similar to those used prior to 1965. In the next of telecommunications services, and the number of sub- three years, average annual investment was more than scribers tripled between 1950 and 1962. Yet a newly three times what it had been over the previous fifteen. independent Jamaica chafed under the apparent restric- Private investment came at a cost, however. To tiveness of a concession arrangement that afforded vir- maintain long-standing (and politically difficult to tually no opportunity for democratic participation. eliminate) cross-subsidies between local and long- Consequently, in 1966 the country established the distance services, upon privatization Jamaica awarded a Jamaica Public Utility Commission. Modeled on the single telecommunications provider a twenty-five-year U.S. system, the commission held regular public hear- concession to operate the entire system. Revenues from ings and was afforded broad scope to base its regulatory the highly profitable long-distance network were used decisions on inputs from a wide variety of stakeholders. to extend the unprofitable local fixed-link network. However, Jamaica lacked the other institutions Debate continues on whether, even within its political needed to make such a system workable. Whereas the constraints, Jamaica could have retained room for U.S. system has a variety of constraints on regulatory competition in some value added services, thereby pre- discretion (including well-developed rules of adminis- serving at least a modicum of pressure for innovation trative process and constitutional protections on prop- and productivity improvements in an era of rapid erty), Jamaica had virtually no checks and balances on global technological change. FOSTERING MARKETS: LIBERALIZATION, REGULATION, AND INDUSTRIAL POLICY 71 COMMUNITY PRESSURE TO HELP PROTECT THE ENVI- RONMENT. In settings where institutions are weak, public Box 4.5 Environmental activism in information and community pressure can be powerful Yokohama, Japan spurs to ever more credible and efficient environmental regulation. In 1960 local medical associations in Yokohama Experiments with transparent, information-intensive began to petition against oil refinery emissions and initiatives can help moderate industrial pollution even the health damages they caused. Shortly thereafter when enforceable formal rules are lacking. In Indonesia, the municipal government, which had been drag- for example, a largely voluntary Clean Rivers program, ging its feet on environmental issues, was ousted in launched in 1989, had reduced total discharges of the 100 elections by a reformist mayor who pledged to participating plants by more than a third by 1994. A pro- implement pollution prevention policies. A flurry gram announced in mid-1995 to set, and publicize, envi- of activity followed, punctuated by the establish- ronmental ratings for factories also seems to have induced ment of a new pollution control unit within city many poorly rated factories to improve their performance. government (which by the end of 1964 had a staff In both programs the secret to success was the reputation of ten), a residents' environmental organization, effect of making public to business peers, communities, and a joint advisory group composed of community and consumers the extent to which individual firms were representatives, academics, and business experts. good environmental citizens. Although the city had no legal authority to Environmental programs built entirely around public impose controls on pollution, by December 1964 it information have obvious limits. Nearly half the firms had entered into a formal, voluntary agreement with participating in the Clean Rivers program did not reduce a new coal-fired power plant to drastically reduce the intensity of their polluting activities. Information- emissions. This agreement offered a precedent for driven programs do help signal where the most severe subsequent voluntary agreements with other new problems are to be found, but often additional measures and existing large factories, which reduced emis- are necessary to get heavily polluting firms to clean up. sions to just 20 percent of their earlier projected lev- And clearly, as countries develop they will need to move els. Over the next two decades Yokohama progres- toward more institutionalized approaches that integrate sively increased the stringency of these voluntary community pressures with more formalized mechanisms agreementsand consistently maintained higher for enforcing compliance. environmental control standards than did Japan's In a pattern seen throughout the world, initiatives national government (which itself was continually from the bottom up can set the stage for formal action at raising its standards). the national level. In the first two decades after World War II Japan rushed headlong into industrialization, with little concern for the environmental impact. At the national level this period of neglect ended in 1967, with the landmark Basic Law for Environmental Pollution than as an automatic response to problems. Moreover, Control. But well before then, grassroots initiatives in countries' experiences with financial, utility, and environ- many localities had set in motion sustained environmen- mental regulation show how competition, voice, and self- tal reform (Box 4.5). regulation can achieve social objectives once thought to require rule-based solutions. Lessons: C1arz)5ing regulatoly options Does the country have the institutional and political The reality of imperfect markets brings regulation onto underpinnings necessary for formal rules to serve as a basis the development policy agenda. At the same time, how- for credible regulatory commitments? On the political ever, the reality of imperfect government cautions against front, the relevant question is whether the country has the hasty enactment of institution-intensive regulatory sys- political will to follow through on what it enacts. On the tems in settings where institutions are weak. The key to institutional front, a critical issue is whether the country success is to focus the regulatory agenda and adapt the has an independent judiciary, with a reputation for impar- available regulatory tools to fit the country's institutional tiality, whose decisions are enforced. If not, other com- capability. Two questions can help guide countries in the mitment mechanisms (sometimes extraterritorial) may be search for better regulation. needed. In countries where political coalitions capable of Are formal rules necessary to correct the market imper- amending rules are difficult to stitch together, legislation fections? Regulation's mixed record suggests that the use may suffice; in other countries it may be desirable to of formal rules to regulate markets is better viewed as a embed formal rules in binding legal agreements with indi- complement to other measures (or even as a last resort) vidual firms. 72 WORLD DEVELOPMENT REPORT 1997 If formal rules are called for, these must be workable forms of competition, they will have little incentive to use not just in theory but in practice. In an ideal world flexi- resources efficiently or to innovate, productivity will not ble rules are preferable to rigid ones. But what constitutes improve, and industrial expansion will not be sustained. a good regulatory "fit" in the real world may bear little Third, a country's strategy for industrial development relation to ideal conceptions of efficiency. In countries has to be guided by its evolving comparative advantage that lack appropriate checks and balances, flexibility may by its relative abundance of natural resources, skilled and have to be sacrificed in the interests of certainty and pre- unskilled labor, and capital for investment. Some propo- dictability. What appears at first blush to be less than effi- nents of activist measures have favored efforts to nurture a cient may thus turn out to be the single best solution from nascent comparative advantage by encouraging firms to the standpoint of matching the goals of regulation to the risk more on a new market than they might otherwise strengths and weaknesses of existing institutions. have been willing to invest. Very few, however, would support wholesale leapfrogging: low-income countries, Can state activism enhance market development? say, seeking to subsidize investments in highly technol- Where externalities, lack of competition, or other market ogy-intensive activities. And there is broad agreement that imperfections drive a wedge between private and social high levels of protection to promote infant industries, goals, most people accept that states may be able to en- without compensating pressures to encourage efficiency, hance welfare through regulation. Much more controver- can be fatal to a country's chances of achieving sustainable sial is whether states should also try to accelerate market industrial development. development through more activist forms of industrial policy. The theoretical case for industrial policy rests on Industrial policy in practice the proposition that the information and coordination The many and varied approaches to activist industrial pol- problems identified above can be pervasivemore so in icy can be grouped under three broad headings: invest- developing economiesand can go beyond those ad- ment coordination, network thickening, and picking win- dressed by well-functioning institutions to protect prop- ners. In both the first two approaches the government erty rights. In essence the argument centers on the fact attempts to enhance market signals and private activity that, in underdeveloped markets with few participants, although the institutional demands of investment coordi- learning can be extremely expensive. Information, more nation are much greater than those of network thicken- readily available in industrial countries, here becomes a ing. The third approach involves government seeking to zealously guarded secret, impeding coordination and mar- supersede the market altogether. ket development more generally. INVESTMENT COORDINATION INITIATIVES. The classic, In theory, governments in such economies can act as "big push" rationale for government activism was that brokers of information and facilitators of mutual learning investment in an underdeveloped country posed a huge and collaboration, and thereby play a market-enhancing collective action problem. With markets undeveloped, role in support of industrial development. But whether firms could not perceive the demand for more and better governments can play this role in practice will depend, as products that the very act of producing them would cre- ever, on their institutional capability. Even aggressive pro- ate. Thus, it was argued, countries could benefit from ponents recognize that activism can enhance markets only coordinating such investments, which are mutually bene- if three critical background conditions are in place. ficial to firms but which they are unlikely to undertake by First, and perhaps most important, companies and themselves. Postwar Japan's development of its steel, coal, officials need to be working on a basis of mutual trust. machinery, and shipbuilding industries illustrates this Firms need to be confident, not only that additional coor- rationale for intervention, as well as the stringent institu- dination has merit, but that the government and the other tional prerequisites for success (Box 4.6): firms involved will make good on their commitments. The participants also need confidence that a given set of A domestic private sector capable of efficiently manag- arrangements will be flexible enough to adapt to changing ing complex, large-scale projects circumstances. Ordinarily this will mean a credible gov- a A private sector willing to cooperate with government ernment commitment to involve the private sector in in pursuit of the shared goal of competitive industrial implementation. development Second, initiatives to promote industrial development Strong technical capabilities in public agencies for eval- must be kept honest through competitive market pres- uating private analyses of investment options and, on sures. Competition can come from other domestic firms occasion, generating independent industrial analyses or from imports, or take place in export markets. Unless Sufficient mutual credibility to enable each parry to firms are systematically challenged by one or more of these base its investment decisions on the other's commit- FOSTERING MARKETS: LIBERALIZATION, REGULATION, AND INDUSTRIAL POLICY 73 Box 4.6 Japan's postwar big push in metals industries A coordinated restructuring of the machinery, steel, The coal industry committed itself to invest 40 bil- shipbuilding, and coal industries contributed greatly to lion yen to rationalize production from domestic Japan's economic recovery after World War II. mines, provided the steel firms agreed to purchase Machinery companies identified the high cost of steel coal from them afterward at the new prices, which as a major impediment to penetrating export markets. would be 18 percent below prevailing levels. Steel companies, in turn, identified the high cost of The steel and coal industries agreed on an overall tar- coal as a principal reason for high steel prices. High get price that steel firms would pay for coal, to be coal prices were a consequence of continued mining achieved by mixing domestic purchases and imports. from expensive Japanese mines and the high cost of The steel industry committed itself to invest 42 bil- shipping imported coal to Japan. lion yen to upgrade its facilities. With this invest- Building on institutional arrangements nurtured ment, and lower coal prices, it would be able to during wartime, in 1949 Japan's Ministry of Inter- export steel at competitive prices. national Trade and Industry (MITI) put in place a In return for lower steel prices, the machinery and joint public-private deliberative structure, the Coun- shipbuilding industries were in a position to embark cil for Industrial Rationalization. Composed of on large, export-oriented investment programs. representatives of industrial associations, leading en- These commitments provided the domestic market terprises from each industry, and public officials, that the steel industry needed to embark on its the council included twenty-nine sectoral branches investment program, and confidence that the ship- and two central branches. Three of the council's ping cost of imported coal would decline. branchesiron and steel, coal, and coordination worked closely together and agreed on the following Once the Japan Development Bank (after careful commitments: technical analysis, and in consultation with both MITI and the Bank of Japan) agreed to participate in these The steel and coordination branches identified the projects, providing financing at only moderately subsi- price of coal that would make it possible to produce dized interest rates, Japan's largest banks took the lead export steel competitively. in mobilizing the investment funds. ments, and to adapt its actions in response to changing cial turmoil surrounding the fall of President Ferdinand circumstances without undermining the overall com- Marcos' regime intervened. By late 1987 five of the eleven mitment to collaborate. initial projects, accounting for almost $4 billion of the $5 billion, had been shelved as infeasible. A sixth project had Pursuing this style of investment coordination presup- been abandoned because its lack of economic potential poses levels of public and private institutional capability became apparent. A fertilizer plant, completed at a cost of that are beyond the reach of most developing countries. $550 million, was suffering losses that were being shoul- The Philippine experience of the late 1970s and 1980s dered by government. Only four projects, accounting for shows what can happen when the ambitions of policy just $800 million, were operating profitably. do not match up to institutional reality, and efforts to NETWORK-THICKENING INITIATIVES. Activist initia- coordinate investment are pursued where government is tives need not be large in scaleimposing commensurably swayed by powerful private interests. large demands on public and private institutional capabil- Driven in part by the desire to create new business itiesor solely devoted to increasing investment. They can opportunities for domestic allies, in 1979 the Philippine also aim to strengthen the private-to-private networks that government announced a new $5 billion program of flourish in mature market systems. Domestic, regional, "major industrial projects," all in heavy, capital-intensive and international networks create numerous sources of industries. Within a year of the announcement the gov- learning and opportunity for firms: specialized buyers ernment, responding to pressure from critics, agreed to open up new market niches and offer information on subject the projects to another round of economic and product standards, equipment providers transfer techno- financial scrutiny. Soon thereafter the political and finan- logical know-how, input suppliers help with product and 74 WORLD DEVELOPMENTREPORT 1997 process innovations, and competitors are a rich source of tries as diverse as Argentina, India, Jamaica, Mauritius, new ideas. Often, clusters of firms, buyers, equipment sup- Uganda, and Zimbabwe. pliers, input and service providers, industry associations, Using public procurement to foster competitive private sec- design centers, and other specialized cooperative organiza- tor development. In Brazil's state of Ceará an innovative tions come together in the same geographic region. cost- and quality-driven procurement program worked Countries whose markets are underdeveloped may through associations of small producers to transform need some catalyst, public or private, to set this cumula- the economy of the town of São João do Arauru. Before tive process of market thickening and network develop- the program the town had four sawmills with twelve ment in motion. There are three leading examples. employees. Five years later forty-two sawmills em- The first is special support for exports. Participating in ployed about 350 workers; nearly 1,000 of the town's export markets brings firms into contact with interna- 9,000 inhabitants were directly or indirectly employed tional best practice and fosters learning and productivity in the woodworking industry; and 70 percent of output growth. It can also be a useful measure of the effective- was going to the private sector. ness of government efforts at industrial promotion. Many countries have directed credit in favor of exporters and SUPERSEDING MARKETS. Sometimes information and set up export promotion organizations. With few excep- coordination problems are so severemarkets so under- tions, most of them in East Asia, these bodies became developed, and private agents so lacking in resources and expensive white elephants. Other export support measures experiencethat market-enhancing initiatives are un- have also been tried, with mixed results. World Trade likely to yield any response, As a way of kicksrarting Organization rules may well rule out future experiments industrial growth, states have been tempted to supplant along these lines. market judgments with information and judgments gen- A second type of effort focuses on strengthening local erated in the public sector. These efforts rarely work, infrastructure: physical, human, and institutional. The although the success of some ventures by Korea's chaebol history of Korea's once-lagging Cholla region illustrates (interlinked business groups), made at the initiative of the impact local infrastructure can have. In 1983 this government, suggests that the quest to pick winners is not southern region opened its first large-scale industrial inevitably a fool's errand. estate. Its success set in motion a cumulative process of What distinguished Korea's success from others' fail- learning by local authorities about how to plan, finance, ures was that these initiatives were channeled through the build, and operate such estatesthree more followed. It private sector, whereas most such efforts (including some also helped catalyze a transformation of the business envi- in Korea) have been implemented by state enterprises. ronment, from one bogged down by red tape and other When state firms are used as implementing agencies, the bureaucratic obstacles to one of close cooperation and opportunities for venalityor fanciful romanticismare coordination between the local government and the pri- virtually limitless. A number of countries have subsidized vate sector. By 1991 Cholla accounted for 15 percent of money-losing state enterprises, to the severe detriment of industrial land in Korea, up from 9 percent in 1978, and fiscal performance. The generally sorry experience with the rate of growth of regional manufacturing output was investment in state enterprises has convincingly demon- above the national average. strated that the production of tradable products is best left Third, and increasingly popular, are public-private exclusively to private firms. partnerships, with the public partners drawn from either local or regional governments. These can take a variety of Walking the industrial policy tightrope forms, including: These experiences highlight why the debate over indus- trial policy has been unusually heated: industrial policy is Initiatives directed at individualfirms or groups offirms. combustible. Economic theory and evidence suggest that Sometimes these are focused events, such as joint par- the possibility of successful, market-enhancing activism ticipation in a trade fair. Others are aimed at achieving cannot be dismissed out of hand. But institutionaitheory a broader shift in the business culture to favor increased and evidence suggest that, implemented badly, activist cooperation. A promising approach involves giving industrial policy can be a recipe for disaster. How, then, matching grants to firms, typically on a 50-50 cost- might countries proceed? sharing basis, to help penetrate new markets and Taken together, the economic and institutional per- upgrade technologies. Easy to implement, with man- spectives suggest drawing a sharp distinction between ini- agement delegated to private contractors, and demand- tiatives that require only a light touch from government driven, with participating firms paying for half of any (for example, some network-thickening initiatives) and initiative, such programs are now under way in coun- initiatives that require high-intensity government support FOSTERING MARKETS: LIBERALIZATION, REGULATION, AND INDUSTRIAL POLICY 75 (such as coordinating investment or picking winners). to prove ineffective at best, and at worst a recipe for cap- High-intensity initiatives should be approached cau- ture by powerful private interests or predation by power- tiously, or not at all, unless countries have unusually ful and self-interested politicians and bureaucrats. strong institutional capability: strong administrative capa- How, then, should countries proceed if they lack the bility, commitment mechanisms that credibly restrain administrative and institutional wherewithal to make such arbitrary government action, the ability to respond flexibly approaches work? The long-run strategy, explored in Part to surprises, a competitive business environment, and a Three, is to strengthen and build the requisite institu- track record of public-private partnership. tions. In the meantime this chapter has indicated two pos- By contrast, light-touch initiatives (those that are sible pathways toward reform. One is to focus on the inexpensive, and supportive rather than restrictive or com- essentials and take on a lighter agenda for state action. mand-oriented) offer more flexibility. The essential insti- The second, which need not conflict with the first, is to tutional attribute for success is an unambiguous commit- experiment with tools for state action that are better ment by government to public-private partnership. When aligned with the country's capability. Much remains to be this commitment exists, when countries do not overreach learned, but this chapter has highlighted two strategies their institutional capabilities, and when the business envi- that appear to have great potential even where institu- ronment is reasonably supportive of private sector devel- tional capability is weak: opment, the benefits of experimentation with light-touch initiatives can be large, and the cost of failure iow. Specify the content of policy in precise rules, and then lock in those rules using mechanisms that make it Strategic options: Focusing on the workable costly to reverse course: in utility regulation, for exam- In the realm of liberalization and privatization, regulation, ple, these might include take-or-pay contracts with and industrial policyindeed, in the full range of state independent power producers. actions probed in this Reportthere is no one-size-fits-all Work in partnership with firms and citizens, and, where formula. Privatization and liberalization are the appropri- appropriate, shift the burden of implementation entirely ate priorities for countries whose governments have been outside government. In industrial policy this may mean overextended. Every country must also look to build and fostering private-to-private collaboration rather than adapt its institutions, not dismantle them. This chapter building a large industrial bureaucracy. In financial reg- has distinguished between institution-intensive and insti- ulation it means giving bankers an incentive to operate tution-light approaches to regulation and industrial pol- prudently, rather than just building up supervisory capa- icy, stressing how the choice of approaches might appro- bility. And in environmental regulation it means using priately vary with a country's institutional capability. information to encourage citizen initiatives, rather than Successful institution-intensive approaches generally promulgating unenforceable rules from the top down. share two characteristics. They require strong administra- tive capability. And they delegate substantial discretion The policies that rely on these approaches may not be for policy and implementation to a public agency, embed- first-best policies in a textbook sense. But as state capabil- ded in a broader system of checks and balances that pre- ity grows, countries can switch to more flexible tools, capa- vents that discretion from degenerating into arbitrariness. ble of squeezing out further efficiency gains. Throughout, If institutions are strong, these state actions can contribute states must maintain the confidence of firms and citizens to economic well-being. If they are not, the evidence and that flexibility will not be accompanied by arbitrary behav- analysis of this chapter suggest that such actions are likely iorelse the foundation for development crumbles. - iT CON tribution to economic development when its role matches its institutioll bilityc But capability is not destiny. It can and must be improved if governments are to promote further improvements in economic and social welfare. This part of the Report discusses how this can be achieved. State capabil1trefers to the ability of the state to undertake collective actions at least cost to society. This notion of capability encompasses the administrative or technical capacity of state officials, but it is much broader than that. It also includes the deeper, institutional mechanisms that give politicians and civil ser- vants the flexibility, rules, and restraints to enable them to act in the collective interest Three interrelated sets of institutional mechanisms can help create incentives that will strengthen the state's capability. These mechanisms aim to: U Enforce rules and restraints in society as well as within the state Promote competitive pressures from outside and from within the state, and U Facilitate voice and partnerships both outside and within the state Over the long term, sustainable institutions have been built on formal checks and balances, anchored around core state institutions such as an iiidependent 78 WORLD DEVELOPMENT REPORT 1997 judiciary and the separation of powers. These are essential for ensuring that nei- ther state officials nor anyone else in society is above the law. But rule-based government is not enough. State capability will also be improved by institutional arrangements that foster partnerships with, and provide competitive pressures from, actors both outside and within the state. Partnerships with and participation in state activities by external stakeholdersbusinesses and civil societycan build credibility and consensus and supplement low state capa- bility Partnerships within the state can build commitment and loyalty on the part of government workers and reduce the costs of achieving shared goals. The flip side of partnership is competitive pressurefrom markets and civil society and within the state itself. Such pressure can improve incentives for per- formance and check the abuse of the state's monopoly in policymaking and service delivery. Similarly, competitive or merit-based recruitment and promotion are crucial for building a capable bureaucracy. This part of the Report starts with a look at ways to reinvigorate state insti- tutions, by building the foundations of an effective public sector (Chapter 5) and, more broadly, by instituting formal checks and balances and controlling corrup- tion (Chapter 6). Chapter 7 then examines how to sharpen competitive pressure, strengthen voice, and promote partnership by bringing the state closer to the peo- plean important means of which is decentralization, or the shifting of power and resources to lower levels of government. Finally, Chapter 8 discusses partner- ships and competitive pressures that cross the boundaries of nation-states, and how countries can cooperate in providing collective goods. BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR Sire, a vast majority of civil servants are il/paid. . . . The result is that skilled and talented men shun public service. The Government of Your Ivlajesty is then forced to recruit mediocre personnel whose sole aim is to improve their weak pecuniary situation. .. intelligent, hardworking, competent, and motivated individuals should direct Your Empire civil service. . . . It is Your Majesty prerogative to introduce the indispensable principle ofaccountability, without which all progress is retarded and work inevitably destroyed. From The Political Testaments ofAli Pasha, Grand Vizier to Ottoman Sultan Abdulaziz, about 1871 (quoted in Andic and Andic 1996) SAME PROBLEMS THAT PLAGUED SULTAN ABDUL- emphasis here is on the incentive framework guiding THE aziz haunt today's public bureaucracies. The history behavior-what government agencies and officials do and of development in Europe and North America in the how they perform. Few countries consciously set out to nineteenth century, and that of East Asia in this one, have encourage bureaucratic corruption and inefficiency. But shown the economic rewards of building an effective pub- the mere existence of formal rules forbidding bribes, say, lic sector. But examples of this kind of institution build- or the abuse of patronage will rarely be enough to root ing are all too rare. these things out. What matters is whether the actual rules Building the institutions for a capable public sector is and incentive mechanisms embedded in the system can essential to enhancing state effectiveness, but also im- translate the fine words into reality. mensely difficult. Once poor systems are in place, they The foundations of an effective public sector can be very difficult to dislodge. Strong interests develop in maintaining the status quo, however inefficient or Evidence across a range of countries has shown that well- unfair. And those who lose out from present arrangements functioning bureaucracies can promote growth and reduce may be unable to bring effective pressure to bear for poverty (Chapter 2). They can provide sound policy inputs change. Even when the incentives are there to improve and deliver critical public goods and services at least cost. public sector performance, formidable information and During the nineteenth century most industrializing states capacity constraints often thwart the attempt. modernized their public administrations. Early leaders in- How can governments with ineffective public insti- cluded France, Prussia, and the United Kingdom. The suc- tutions begin to put things right? The complex problems cess of the British civil service reforms in the last century involved in building and managing a public bureaucracy provides an early example of the importance of nurturing do not lend themselves to clear, unambiguous solutions. effective, rule-based bureaucracies. As Box 5.1 describes, But this chapter outlines some institutional building these reforms heralded the careful cultivation of a profes- blocks of an effective public sector and discusses promis- sional meritocracy, which helped pave the way for a half- ing options for putting these in place. This focus on insti- century of English dominance over international commerce. tutions is very different from the traditional approach of More recently, some East Asian economies have estab- technical assistance, which emphasizes equipment and lished and nurtured the foundations of capable bureau- skills and administrative or technical capacity. The cracies. Many low-income countries, however, have been SD WORLD DEVELOPMENT REPORT 197 Box 5.1 Laying bureaucratic foundations: The Northcote-Trevelyan reforms in the United Kingdom Until the early nineteenth century the affairs of the ing the government's work into two classesintellec- state in the United Kingdom were administered by tual (policy and administration) and mechanical (cler- public officials who owed their positions to political ical)and creating a career civil service to carry it out. patronage and influence. There was no common sys- Staff capable of performing the intellectual work tem of pay, bribes augmented official salaries, and would be recruited from the newly reformed universi- officeholders, who viewed their positions as property ties; the best talent would be selected through tough that could be sold, often engaged and paid their own competitive examinations supervised by a board of civil staff. Although the system did not rule out advance service commissioners. by individual ability, it was not a basis for sound Opposition was strong. Although a civil service administration. commission was established in 1855, many government As the Victorian era progressed, however, the departments continued recruiting in their accustomed United Kingdom underwent a period of intense way until 1870, when patronage was abolished and the reform driven by social and economic change and the two grades were made compulsory for all departments. demands of an expanding, educated middle class. Uni- The Northcote-Trevelyan reforms were followed by versities, the armed forces, the judiciary, and central reforms in the armed forces, the judiciary, and, later, and local bureaucracies were all reformed. municipal government. There were also extensive The blueprint for civil service reform was the changes in the rules and restraints governing policy Northcote-Trevelyan Report of 1854, which advo- formation and implementation. By the end of the nine- cated the creation of a modern bureaucracy based on a teenth century the United Kingdom had laid the foun- career civil service. Drawing on ideas advanced for the dations of a modern government and formally institu- Indian civil service by Thomas Macaulay, Sir Stafford tionalized the values of honesty, economy, and political Northcote and Sir Charles Trevelyan proposed divid- neutrality. unable to create even the most rudimentary underpin- their expert counselors the flexibility they need to for- nings of a rule-based civil service. Their formal systems mulate policies, but embed their decisionmaking in often resemble those of industrial countries on paper. But processes that allow for inputs and oversight from in practice informality remains the norm. Merit-based stakeholders. personnel rules are circumvented, and staff are recruited Efficient and effective deliveiy systems. Here, too, reform or promoted on the basis of patronage and clientelism; involves setting the right balance between flexibility budgets are unrealistic and often set aside in any case by and accountability. For activities that are contestable ad hoc decisions during implementation. At bottom, all (that is, where there is scope for actual or potential these problems can be traced back to weaknesses in the competition from various suppliers) and easily speci- underlying institutions: poor enforceability of the rule of fied, market mechanisms and contracting out of ser- law both within and beyond the public sector; a lack of vices can often improve delivery dramatically. But for built-in mechanisms for listening to, and forming part- many other services there is often no substitute for nerships with, firms and civil society; and a complete delivery by the core public sector. Here giving citizens absence of competitive pressure in policymaking, the greater voice and allowing client feedback can exert delivery of services, and personnel practices. pressure for better performance, but ultimately perfor- To tackle these problems at their root and lay the foun- mance will depend on the loyalty of civil servants and dations of an effective public sector, countries need to their compliance with established rules. focus on three essential building blocks: Motivated and capable staff These are the lifeblood of the executive. Able and dedicated staff inject energy into Strong central capacity for formulating and coordinating the public sector. Uncommitted staff stifle it. Civil ser- policy. This is the brains of the system. Politicians for- vants can be motivated to perform well through a range mulate visions and set goals, but for these to material- of mechanisms, including merit-based recruitment and ize they must be translated into strategic priorities. This promotion, adequate pay, and a strong esprit de corps. requires mechanisms that lead to well-informed, disci- plined, and accountable decisions. A constant challenge Effective bureaucracies take decades to develop. And in for all countries is to set rules that give politicians and seeking to buildor rebuildthe foundations of such a BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR Si bureaucracy, reformers must, as always, be conscious of capacity in these countries tends to be weak and frag- what they are building on. For example, where countries mented, with few institutionalized mechanisms for input have been unable to establish credible controls over man- or oversight. agers' use of resources, giving them greater flexibility will only encourage arbitrariness and corruption. But certain Policymaking mechanisms in industrial countries reforms can generate early payoffs even in the worst sys- Many OECD countries have built up well-functioning tems. These are discussed further below. policymaking mechanisms over time. At the heart of these systems are mechanisms for properly preparing policy Strengthening institutions for policymaking proposals, estimating the costs of alternative proposals Politicians set goals and broad strategic directions. But within a disciplined overall budget framework, ensuring sound institutional arrangements can determine whether their critical evaluation through consultation and debate, the visions of political leaders get translated into effective and reaching and recording decisions and monitoring policy priorities. They can make transparent the costs and their implementation. A vital complement to these mech- benefits of competing policy proposals. And given that anisms is effective capacity at the center of government adequate information is inevitably in short supply, they the Cabinet Secretariat in France, the Ministry of Finance can make sure that leaders are as well informed as possi- in Japan, the Office of Management and Budget in the ble, through processes that provide input and oversight United Statesto facilitate consultation and coordinate from internal and external stakeholders. All these mecha- proposals among ministries before they are submitted. nisms will help produce better-informed decisions and There is always room for improvement. Australia is a raise the credibility of policymaking in general. good example of an industrial country that has introduced The rules and norms embedded in the policy- reforms explicitly aimed at making the policymaking making process should be designed to curb the kind of process more transparent, competitive, and results- uncoordinated political pressures that can lead to poor oriented. Several features of these reforms are of particular decisionmaking and bad outcomes. If politicians or relevance to other countries: the emphasis on publicizing bureaucrats pursue only their own or their constituents' the medium-term costs of competing policies; the effort immediate interests as they are voiced, the result may be to facilitate debate and consultation on policy priorities, collectively undesirable, even destabilizingthere is no within hard budgets, both in the cabinet and among agen- invisible hand in statecraft, automatically shaping individ- cies; and the attention to results (Box 5.2). ual initiatives toward a common good. The aim must be The United States and some continental European to build mechanisms to discipline and coordinate the pol- countries have instituted other mechanisms for consulta- icy debate and call competing policy proposals into ques- tion and oversight in policymaking. Corresponding to tion. In some countries politicians have delegated macro- the cabinet in a parliamentary system, congressional com- economic and strategic policy coordination to capable, mittees in the U.S. presidential system are the principal relatively autonomous central agencies, whose activities arena for policy debate and consultation. U.S. executive are guided by consultative processes that are transparent to agencies, for their part, are governed by the Adminis- outsiders. In others, politicians themselves collectively trative Procedures Act of 1946, which imposes certain restrain and challenge each other in established forums procedural requirements, enforceable in the courts, such where decisions are made in common. But many countries as public announcement of new policies, while pre- have none of these mechanisms, and their absence reveals serving flexibility in the substance of policymaking. This itself in incoherent strategic policies and macroeconomic procedures-oriented approach to policy formulation instability. allows legislators to shift the substance of policymaking to Although the precise institutional arrangements vary, specialist agencies and other interested parties closer to effective public sectors the world over have generally been the problem. This kind of decentralized mechanism uses characterized by strong central capacity for macro- citizen voice and the judiciary to ensure accountability, economic and strategic policy formulation; by mecha- but with the inevitable side effect of slowing decision- nisms to delegate, discipline, and debate policies among making. government agencies; and by institutionalized links to Many continental European countries rely on admin- stakeholders outside the government, providing trans- istrative law and specialized courts for judicial review of parency and accountability and encouraging feedback. As administrative actions. Citizens can challenge administra- discussed below, systems in many industrial countries and tive decisions on legal grounds or for factual errors. The in much of East Asia exhibit many of these characteristics. European Union has adopted this system for the Euro- Their absence in many developing economies is a major pean Court of Justice, allowing it to oversee decisions obstacle to building a more effective state. Policymaking made by EU institutions. B2 WORLD DEVELOPMENT REPORT 1997 Box 5.2 Australia's mechanisms for transparent, competitive, and results-oriented policymaking One of the main objectives of Australia's public sector addition, the projections made resource flows to the reforms has been to institute a process for disciplining line ministries more predictable, since the projected and coordinating policies and exposing them to vigor- figures were automatically rolled over into the actual ous debate. Some of the challenges the Labor govern- budget if no changes in policy intervened. This helped ment faced when it came to power in the early 1980s improve decisionmaking and the operational efficiency were similar to those confronting many developing of line agencies. countries today: the new government had to manage The reforms also required that line ministries propos- both an immediate fiscal crisis and the unsustainable ing any new policy, or any changes in existing policy long-term fiscal commitments of previous policies. that would increase spending, also propose offsetting To discipline policy formulation and win political savings (see figure at right). This ensured that spending support for a resetting of national strategic priorities, stayed within the resource envelope agreed to in the cab- the new administration decided to publish estimates inet. The cabinet focused on changes in strategic priori- of future spending under existing policies. These pro- tieswhich new policies to adopt and which existing jections painted a bleak picture of unsustainable real ones to cutto stay within macroeconomic constraints. growth in spending requirements, underscoring the Policy proposals were debated vigorously within the need to scale back. Once the government had pub- cabinet, and all affected ministries and agencies were lished these estimates, however, it became incumbent required to submit written comments on the sound- upon the government to continue to do so, to show ness of other agencies' proposals. This helped legitimize the continuing declines in future commitments that it and build consensus on policy priorities. Finally, the had promised. Indeed, succeeding projections fell reforms focused attention on results, through mandated markedly, in what became called the "falling man" periodic evaluation of new and existing policies and pattern (see figure at left). Open financial markets through reporting on performance and outcomes. imposed additional discipline. The results? Australia's deficit of 4 percent of GDP The reforms also required the government to pub- in 1983 became a surplus by the end of the decade. lish a reconciliation table, showing how the projections Accompanying this achievement were significant for existing policies differed from those for the new changes in the composition of public expenditures, policies. These measures helped make apparent the reflecting both broad strategic shifts identified by the changes in the government's strategic priorities, as well cabinet and changes in priorities within ministries, as in the medium-term costs of new commitments. In often identified by the line agencies themselves. Projected growth in total spending changes by category, 1983/84 to 1992/93 Percent (cumulative) Billions of Australian dollars 15 5 March 1984 4- New spending Net change Projection dates 3 10 2 May 1985 November 1985 December 1986 August 1989 Savings from existing policy 0 1982/83 1984/85 1986/87 1988/89 Defense Education Health Social Transpor- Other security tation Fiscal years Source: Adapted from campos and Pradhan 1996; Dixon 1993. BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR 83 Elite central agencies in East Asia public-private deliberation councils, which would expose The successful East Asian economies have adopted a pro- costly programs to rigorous review. Such transparent and cedural approach to policymaking that shares some key institutionalized consultation mechanisms gave those out- aspects of systems in industrial countries. Several East side of government power to restrain or even veto its Asian leaders have formulated long-term visions for their actions, while preserving flexibility for policymakers to countries; examples include, in postwar Japan, the Liberal adapt to changing circumstances. Democratic Party's declared aim to catch up with the Deliberation councils require substantial technical West, and more recently, in Malaysia, Prime Minister capability if they are to use the information supplied by Mahathir Mohamad's Vision 2020. They have then private participants to make coherent policy. There is also worked to create the institutional arrangements needed the problem that such councils exclude broader segments to translate their vision into a highly focused set of strate- of society from their deliberations. If a country's private gic priorities. Powerful elite central agencies have been sector is small, deliberation councils can all too easily delegated the authority to develop policies that will achieve degenerate into well-oiled mechanisms for unproductive the leaders' long-term objectives. Although relatively rent extraction. This probably explains why African ver- autonomous, these agencies' deliberations have always sions of deliberative councilssuch as the francophone been embedded in processessuch as public-private coun- countries' experiments with economic and social coun- cilsthat provide input and oversight from private firms. cilshave generally been ineffective (with the notable These elite agencies often play an immensely important exception of Botswana's). If such councils are to support role in setting their countries' economic course. The Min- sustainable development, they will need to be comple- istry of International Trade and Industry (MITT) in Japan mented, as they were in East Asia, by attempts to win and the Economic Planning Board in the Republic of legitimacy from society more broadly. Korea are considered the prime movers behind their coun- tries' industrial policies and use of administrative guidance. Weak capacity and fragmented policymaking in in Thailand the Ministry of Finance, the budget bureau, developing countries the central bank, and the National Economic and Social Many developing countries, especially in Africa, Central Development Boardthe so-called gang of fouract in America, and the Caribbean, lack the critical mass of unison to cap spending and control inflation. In Indonesia effective capacity and internal coherence to formulate and the Ministry of Finance and the planning agency Bapennas coordinate macroeconomic and strategic policies. Central have been the guardians of the purse and the brains of the capacity is weak, stretched thinly among a handful of civil service. These central agencies are staffed by profes- senior officials who must attend to numerous tasks. These sional and capable employees recruited on the basis of strains are compounded by problems in the bureaucracy: merit, often through highly competitive examinations. low pay at senior levels, rampant political patronage, and The delegation of macroeconomic policy to competent an absence of meritocratic recruitment and promotion. and reputable technocrats has recently been a common One especially costly consequence of weak central feature in several Latin American countries as well, includ- capacity is an inability to make budget forecasts based on ing Argentina, Colombia, Mexico, and Peru. The effects sound and realistic assumptions. This undermines trans- on performance have been noticeable. This delegated poi- parency and predictability in decisionmaking. For exam- icymaking, combined with the kind of hierarchical and ple, in recent years the difference between budgeted and transparent budget procedures described in Chapter 3, actual recurrent expenditures has averaged more than 50 was important in reducing inflation in these countries percent in Tanzania and more than 30 percent in Uganda. during the late 1980s and early 1990s. Chile, in particu- Transparency and coherence are also compromised by the lar, appears to have developed the kind of esprit de corps use of extrabudgetary funds (equivalent to more than half among senior officials that has long fostered partnerships of total federal expenditures in Nigeria, for example) and in the Indonesian technocracy and Japan's MITI, among by long lags in the production of financial accounts and others. Most striking are the parallels between Chile's audits. Often decisionmakers will have little sense of the group of high-level advisersthe Chicago boysand costs or outcomes of policies. Partly as a result, budgeting Indonesia's Berkeley mafia and Thailand's gang of four. focuses almost exclusively on the allocation of inputs As already noted, although the central agencies in East rather than the results they are intended to achieve. Asia had considerable flexibility in policymaking, they In aid-dependent countries donors sometimes allevi- were embedded in a larger network of deliberation coun- ate, but too often worsen, the problem of weak central cils and external think tanks. In Japan, Korea, Malaysia, capacity. To the extent that their policy advice supple- Singapore, and Thailand, information about the costs of ments weak capacity at the center, they help solve the industrial policies was distributed through a range of short-term problem at hand. But such advice does nothing 84 WORLD DEVELOPMENT REPORT 1997 to build long-term capacity if politicians fail to recognize would need to triple over the succeeding four years, the need ultimately to rely on local experts. Donors may implying drastic cuts in other expenditures (left panel of also fragment central capacity for policy formulation, Figure 5.1). Moreover, the recurrent costs of donor-led entering with ministries into bilateral deals on multiple investment projects were shown to be unsustainable (right projects without determining whether their cumulative panel of Figure 5.1). The same inadequacies play out effects are collectively sustainable or mutually consistent. across the developing world, as newly built roads fall into In many countries public investment programs have be- disrepair, and schools find themselves without textbooks come passive repositories of donor-driven projects, whose and health centers without drugs. recurrent costs after completion continue to accumulate, Several initiatives have been launched to address these contributing to an expansionary fiscal bias. Lack of coor- problems, but all are still in their early stages. The Africa dination between the ministry of planning and the min- Capacity Building Initiative seeks to strengthen the capac- istry of finance sometimes further impedes the integration ity of African governments in policy analysis through a of capital and current expenditures. more professional civil service, improved information All these problems seriously erode the capacity to coor- systems, and enhanced external inputs from African uni- dinate, challenge, and discipline decisionmaking. Guinea versities and civil society more broadly. Governments and provides a case in point. Although the government has donors have also launched sectoral investment programs designated primary education, public health, and road to coordinate donor assistance. The Agricultural Sector maintenance as spending priorities, funds often end up Investment Program in Zambia replaces 180 individual being allocated to other areas instead. And no system donor projects. But although they consolidate fragmented exists for costing out policy proposals or subjecting them policies in a shared arena, such efforts can create coordina- to rigorous scrutiny. An exercise to cost out Guinea's poli- tion problems of their own as long as capacity remains cies to meet the government's stated priorities revealed weak. Malawi and Uganda are among those countries that the share of priority programs in total spending moving to the next, crucial level of reform: developing a Figure 5.1 Guinea's policy goals and spending allocations do not add up Government expenditure 1994 Projected costs of investment projects Road Millions of constant Guinea francs maintenance 400 Public health Other 350 15% 85% Required recurrent Shortfall 300 costs Primary education 250 Projected 2000 Actual 200 recurrent costs Road maintenance 150 - Public Other Capital hea spending 56% 100 44% 50 0 Primary 1996 1997 1998 1999 2000 education Source: Adapted from World Bank 1996d. BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR 55 systematic process to set strategic priorities across sectors remains highly centralized. The Apparat of the Cabinet of and within aggregate spending constraints. Colombia is Ministers has retained responsibility for policy formula- instituting cx post evaluation systems to assess whether tion and coordination and directs the activities of central policies and programs already in place are achieving their government departments. The number of central govern- intended outcomes. ment bodies remains large (over 110), their responsibili- ties often overlap, and lines of accountability are unclear Policy coordination in transition economies (Figure 5.2). The cumbersome structure makes coordina- Although the shortage of administrative expertise is not as tion difficult, delays decisionmaking, and reduces trans- much of a problem in the transition economies of Central parency. However, efforts to reform this system are now and Eastern Europe and the former Soviet Union, their under way following the adoption of a new constitution experience shows the equal importance of mechanisms in July 1996. that can channel that expertise into coherent policy. Some Central and Eastern European countries with When the communist regimes in these countries col- similar although perhaps less severe problems have initi- lapsed, so did the centralized decisionmaking apparatus ated promising reforms of their central decisionmaking for coordinating the activities of ministries and depart- mechanisms. Poland and Hungary have both introduced ments. As a result, confused and overlapping responsibili- reforms to streamline multiple and conflicting responsi- ties and multiple rather than collective accountability bilities and speed decisionmaking. In Georgia, streamlin- emergeda sure-fire formula for policy disaster. ing has removed overlapping and conflicting positions, Ukraine exemplifies such problems in the extreme. and the fate of draft laws is now decided in the presence Following independence in 1991, a central machinery of all members of the president's economic council before of government was established which reflects many of submission to the parliament. Such reforms have aided the features of a former Soviet system. Decisionmaking consultation and coordination in central government Figure 5.2 Decisionmaking in Ukraine is bogged down by overlapping responsibilities First Vice Prime Minister President's Administration z Apparat of Cabinet of Ministers Departments and Directorates Line Ministry Departments Departments Departments Departments Source: Adapted from World Bank 1997. 86 WORLD DEVELOPMENT REPORT 1997 decisionmaking. But most countries have a long way to go to build the institutional capability needed to respond effectively to the many demands of transition. Figure 5.3 Most firms rate government services poorly, but some services score Reforming institutions for delivery higher than others The best-designed policy will achieve little if it is badly Percentage of respondents implemented. Poor quality, high cost, waste, fraud, and corruption have marred the delivery of services in many 100 developing countries. More often than not, the problems can be traced back to a belief that government ought to be 80 the dominantif not the soleprovider. The expansion of the state under the state-dominated development strategies outlined in Chapter 1 has pro- 60 vided countless opportunities for politicians to exploit their positions for political gain (Box 5.3). More broadly, 40 governments have simply overextended themselves, with disastrous results. Examples abound of services not being delivered or being delivered badly. Power system losses in 20 low-income countries are more than twice those in other 0 Mail Customs Roads Health care Box 5.3 The mushrooming of Bangladesh's Rating High government of service Moderate efficiency: Low Since independence in 1971 the government of Ban- gladesh has effectively doubled in size. The number of ministries increased from twenty-one to thirty- Source: Private sector survey conducted for this Report. five over twenty years, and between 1990 and 1994 the number of departments and directorates went from 109 to 221. Public sector employment in- countries. In China nearly 1 million hectares of irrigated creased from 450,000 in 1971 to almost 1 million in land has been taken out of production since 1980 because 1992a compound rate of increase of 3.6 percent a of insufficient maintenance. Only 6 percent of domestic year, compared with population growth of 2.5 per- businesspeople surveyed for this Report in fifty-eight de- cent during the same period. Civil service pay has veloping economies rated government service delivery as fallen considerably, especially for those at the top. efficient, while 36 percent rated them very inefficient. The base pay of a permanent secretary (the most Mail delivery fared reasonably well, followed by customs senior civil service position) has declined by 87 per- and roads; health services scored the worst (Figure 5.3). cent in real terms since 1971. These results indicate a broader pattern: the lower-rated New ministries, divisions, and departments were services are delivered by government departments (as created in part to meet emerging needs such as envi- opposed to state enterprises) and produce outputs that are ronmental protection and women's issues. But the harder to measure and monitor. state has also spread its wings into commercial activ- What is needed is better management of the principal- ities. Growth has often been stimulated by political agent relationships inherent in service delivery. In many considerations. The increase in ministries allowed countries delivery suffers because neither the principals new ministerial positions to be handed out to more (the politicians) nor the agents (the bureaucrats) live up to intraparty groupsand, of course, created more their side of the bargain. Politicians interfere in the day- lower-level jobs to be dispensed. Aside from its bud- to-day operations of public agencies; services are delivered getary effects, this expansion has stretched imple- through government departments whose managers have mentation capacity, compounded coordination little operational flexibility and whose resource flows are problems, and made regulation more intrusive. It unpredictable. Even where managers have been able to get has also created vested interests that have blocked around these rules, they have had few incentives to efforts at rationalization and reform. achieve better results. In the many countries where the BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR 87 public sector has assumed a monopoly role in the delivery of many services, external pressures for better performance Figure 5.4 Three strategies for improving are eliminated. And few countries have set credible government service delivery restraints to hold managers accountable for their use of inputs or for the achievement of measurable outcomes. Increase exit Strengthen possibilities voice mechanisms Institutional mechanisms to improve delivery Governments are experimenting with a range of institu- tional mechanisms to improve service delivery. Greater use of markets is creating competitive pressures and more exit optionsalternatives to public provision for users seek- ing better quality or lower cost. Also expanding these options is the practice of contracting out service delivery to private firms or NGOs. Some governments are setting up performance-based agencies in the public sector and enter- ing into formal contracts with these agencies, providing Improve them greater managerial flexibility while holding them ac- compliance countable for specified outputs or outcomes. Others are and loyalty relying on more traditional bureaucratic forms in the core public sector, emphasizing accountability in the use of in- puts, meritocratic recruitment and promotion, and the cul- budget deficit equivalent to 9 percent of GDP into a sur- tivation of an esprit de corps to build loyalty and improve plus during the 1980s and cut the unit cost of delivery by performance. Finally, user participation, client surveys, pub- more than 20 percent in some agencies. lished benchmarks, and other mechanisms for increasing cit- Several developing countries are now emulating these izen voice are providing external pressures for better delivery. reforms. But what is feasible in New Zealand may be These initiatives can be classified into three broad cat- unworkable in many developing countries. It takes con- egories: expanding exit options, strengthening voice, and siderable capability and commitment to write and enforce improving compliance and loyalty. Which of these apply contracts, especially for difficu1t-to-specif,r outputs in the depends on the incentive environment through which the social services. Which mechanism will prove most effec- service can be delivered (Figure 5.4). Markets and con- tive in improving performance depends on both the char- tracts with the private sector offer primarily exit options acteristics of the service and the capability of the state to for better performance. In the broader public sector enforce internal and external contracts (Table 5. 1). including corporatized state enterprises and performance- For instance, for services that are contestablesuch as based agenciesexit options are fewer, but voice begins most commercial products and, more recently, telecom- to become influential. For activities whose outputs are munications and electric power generationmarket difficult to specifr and that are not contestable, the core mechanisms can generate powerful competitive pressures civil service remains the agency of choice, but it provides for improved delivery. For services whose outputs the no realistic exit options and has considerably less flexibil- state can specifj and enforce at low transactions cost, con- ity in financial and personnel management. Here the tracting out to private firms and NGOs is an attractive more relevant instruments switch to citizen voice, the by- option. Countries with strong capability and commit- airy of the civil service, and well-specified rules for them ment are setting up performance-based agencies and for- to follow. mal contracts even for complex activities within the core Recently, the so-called new public management public sector, such as defense, education, and health care. reforms in industrial countries have sought to move deliv- But countries with little capability to enforce complex ery away from the core public sector (the center of the cir- contracts, and weak bureaucratic controls to restrain arbi- cle in Figure 5.4), primarily by using market mechanisms trary behavior under more flexible management regimes, and formal contracting. New Zealand provides the most need to proceed with caution. dramatic example. Beginning in the early 1 980s, com- mercial and other contestable activities were hived off, Using competitive markets to improve delivery corporatized, and often privatized. The remaining large, Overstretched and with budgets increasingly tight, gov- multipurpose ministries were split up into focused busi- ernments are relying more on market mechanisms to ness units headed by managers on fixed-term, output- improve the delivery of contestable services. At the same based contracts with considerable autonomy (including time, a flood of innovation at both the technological and the right to hire and fire). These reforms helped turn a the policy level has been pushing back the boundaries of 88 WORLD DEVELOPMENT REPORT 197 Table 5.1 Mechanisms to improve service delivery Service characteristics and state capability Easy to specify outputs Environment and enforce perfonnance Private sector Contract out to for-profit or nonprofit agencies Create ets, e.g., by issuing vouc Broader public sector Enhance in Set up performance-based competition agencies Set hard budgets and Corporatize state enterprises divest state enterprises and establish enforceable performance contracts Strengthen voice mechanisms Core public sector what can be considered contestable. Chapter 4 showed market: governments can contract with private firms, how deregulation and the unbundling of activities in selected through competitive bids, to provide services. state-dominated industries such as electric power genera- Contracting out is now a widespread practice in many tion have led to significantly lower unit costs and a rapid industrial countries. Victoria State in Australia provides a expansion in services. particularly dramatic example: each local council con- Other recent innovations such as vouchers and capita- tracts out at least half its annual budget through compet- tion grants have even increased the scope for competition itive tender, including complex community care services. in the provision of some social services. Experience re- In developing countries, where both markets and state mains limited, however, and confined largely to educa- capacities are weak, options for contracting out are fewer. tion. Chile's capitation grant system allows students to Still, where outputs are easily specified and direct compe- enroll in any school, public or private; schools receive a tition is impossible, competition managed through such payment from the state based on enrollment (Box 5.4). arrangements as service contracts, management contracts, Private enrollments have increased, but the program's leases, and long-term concessions can yield efficiency effects on school performance are not yet clear. Vouchers gains. In Brazil, for instance, contracting out road main- are promising but carry a risk of increasing social polar- tenance to private contractors led to savings of 25 percent ization if not properly regulated. over the use of government employees. Leases have in- Indeed, greater use of market mechanisms must be creased the technical efficiency of water supply in Guinea accompanied by effective regulatory capacity. As discussed and of the operation of Port Kelang in Malaysia. in Chapter 4, this is not always easy to achieve. The diffi- Governments are also contracting out the delivery of culties of regulation are even more daunting in social ser- social services, especially to NGOs. Even though outputs vices than in, say, infrastructure. For instance, the private here are difficult to specify, governments have taken this delivery of health care services is unregulated in many de- route where NGOs are perceived to be committed to high veloping countries (Brazil is a notable exception) because quality or where, because of their religious or ideological regulating the vast number of small-scale providers is orientation, they can better serve certain groups (for beyond the government's capacity. example, the Netherlands has long contracted with NGOs for education services). In Bolivia an arrangement with a Contracting out to the private sector and NGOs local church organization to manage public schools is pro- In those areas where competition in the market is not fea- ducing promising results (Box 5.5). And in Uganda the sible, it may still be possible to foster competition for the government is entering into partnerships with NGOs to BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR 89 Box 5.4 Vouchers and school choice Vouchers can increase the scope for competition in began providing capitation grantsfixed payments providing education. Students are given vouchers per student enrolledto both public and private funded by public tax dollars but redeemable at any schools. Because the per-student payment was based school, private or public. Letting parents choose the on the average cost of education in the public sector, school their children will attend should induce schools and expenditures per student were 70 percent less in to compete for students. the private sector, private schools eagerly vied for stu- Opponents of vouchers claim that they would lead dents. By 1986 primary enrollment in private schools public school studentsespecially the better onesto had more than doubled, from 14 to 29 percent of total leave in droves, gutting the public system. Such an out- enrollment, and enrollment in private secondary come would be wasteful and might lead to even poorer schools had increased almost fourfold. The effects on education for those who choose to remain. Yet in a school performance are not yet clear. 1993 pilot program in Puerto Rico, the 18 percent of Vouchers also present potential risks. The ones most students who did transfer to private schools were largely commonly cited are increased stratification among ser- offset by the 15 percent who transferred from private to vices and polarization among users. Some analysts have public schoolshardly a mass exodus. Puerto Rico's criticized the unfettered rush toward school choice in experiment was so successful that in its second year the the former Soviet republics, arguing that it will exacer- number of applicants jumped from 1,600 to 15,500. bate social tensions in rapidly polarizing societies. The Public funding for private schools is nothing new. underlying concern is that, in the absence of national In the Netherlands two-thirds of students attend pub- controls, school curriculums will become divisive and licly funded private schools. When Chile reformed its parochial, and an essential role of the statethat of education system in 1980, the Ministry of Education ensuring social cohesionwill be undermined. deliver both preventive and curative health services previ- Imp roving delivery through the broader public sector ously in the public domain. Notwithstanding the growing opportunities for private Like regulatory contracts and vouchers, however, con- participation, the public sector will inevitably continue to tracting out is not a panacea. In general, contracting deliver a large number of services. Here the challenge is to works best where outputs are easy to specify and markets create an enabling environment that provides incentives are strong, so that the effectiveness of alternative suppliers for better performance. can be readily judged. For activities that are complex or INTERNAL COMPETITION. Some industrial countries are nonroutine, contracting out will inevitably incur higher experimenting with ways to increase competition within transactions costs. Contracting is also prone to corruption the public sector, to improve delivery of services for which and mismanagement, much as are contracts within the neither market competition nor contracting out is feasi- public sector. The Inspector General of Uganda, to take ble. In the United Kingdom, for example, an internal but one example, reported several instances of fraud in market has been created within the national health service contracts with the private sector, including payments for by transforming local health authorities and groups of roads that were never built or maintained. general practitioners into purchasers of hospital services Finally, competition in contracting does not necessar- on behalf of their patients. This arrangement has created ily mean that the private sector will outperform the pub- competition among hospitals, acting as a decentralized lic sector. In one of the most extensive U.S. experiments mechanism for reallocating resources. Although internal in public-private competition, the Public Works Depart- competition can enhance efficiency, it is important to ment of Phoenix, Arizona, developed innovative strategies ensure that equity concerns are safeguarded in the process. and outperformed private competitors on several con- Decentralization of deliverymoving resources and tracts between 1984 and 1988. Competition provided the responsibilities to lower levels of governmentis another stimulus for reform in that case, but equally important potentially powerful means of introducing internal com- was the strong partnership forged between management petitive pressure, particularly for the provision of local and labor through quality circles and labor-management public goods with few interjurisdictional spillovers or productivity committees to build on workers' expertise economies of scale. Local governments get the flexibility and jointly identify competitive solutions. to match supply to local preferences or demands, while 90 WORLD DEVELOPMENT REPORT 1997 into executive agencies charged with specific delivery func- tions. These changes have been accompanied by substan- Box 5.5 Contracting with NGOs for better tial devolution of managerial authority and accountability schooling in Bolivia for results. In Australia, Denmark, Ireland, and Sweden, for example, detailed and itemized administrative costs In an experimental program in Bolivia, the govern- have been consolidated into a single budget line item, ment contracted with the church-based Fe y AlegrIa making it possible for managers to reallocate resources in to manage a certain number of mostly secondary accordance with changing priorities and needs. public schools. Before agreeing to do so, Fe y Ale- Among developing countries, Singapore has perhaps grIa demanded (and received) the right to appoint the longest head start in creating focused business units. principals and teachers and to allow teachers to As early as the 1970s the Singaporean civil service was work both the morning and afternoon shifts rather organized around the concept of statutory boards. Else- than the three-and-a-half hours allotted for instruc- where, Jamaica has selected eleven pilot agencies for con- tion in the public schools. version into executive agencies. In all other ways Fe y AlegrIa schools are identi- But countries with inadequate controls over inputs and cal to other Bolivian public schools. Although hand- weak capacity need to proceed with caution. The indus- picked, teachers receive little special training and are trial countries that have now relaxed detailed control over paid the same salary as other public school teachers. inputs did so from a position of strength, having devel- Fe y AlegrIa schools receive no additional money for oped over many years a series of credible restraints on books or supplies, and their curriculums and teach- arbitrary behavior. For the many countries that have not ing methods are the same as those used in other yet succeeded in instituting credible controls over the use public schools. of inputs, greater managerial flexibility will only increase The only comparative advantage Fe y AlegrIa arbitrary and corrupt behavior. Furthermore, writing and schools have is an exceptional esprit de corps among enforcing contracts, particularly for complex outputs, students, parents, and staff. Teachers and students require specialized skills that are often in scarce supply. flock to Fe y AlegrIa schools, with many families Some of these concerns are borne out by a recent study paying extra fees for their children to attend. On of state enterprises, which found the overall record of the rare occasions when innovative teaching meth- performance contracts in developing countries extremely ods have been tried (such as a mathematics course disappointing. Yet while performance contracts have transmitted over a public radio station), they have not succeeded in most developing countries, many have also proved popular, both in the schools and sought to create performance-based agencies for easily throughout the community. This public-private specified and high-priority tasks such as road maintenance partnership between government and a religious or tax collection. These agencies are typically set up as NGO appears to be so successful that the govern- enclaves within the civil service, with greater managerial ment is studying it as a possible model for national flexibility, better pay, and greater accountability for re- education reform. sults. In Sub-Saharan Africa, for instance, performance- based agencies have been created to achieve tax collection targets in Ghana, Uganda, and Zambia. Other countries local accountability and interjurisdictional competition in appear ready to follow suit. supply provide potential restraints. But as discussed in In these instances, enclaving tax collection has been Chapter 7, appropriate institutional preconditions need considered a prerequisite for boosting a government's to be in place if decentralization is to improve efficiency capacity to raise revenues and improving incentives for the and equity. rest of the civil service. The results have been impressive. PERFORMANCE-BASED AGENCIES. The so-called new Ghana was the first country in Sub-Saharan Africa to public management reforms in industrial countries have introduce a performance-based approach to tax and cus- sought to break up the core public sector into a series of toms revenue collection. Total revenues nearly doubled in distinct business groups or special-purpose agencies. In the first five years, from 6.6 percent of GDP in 1984 to general these agencies have greater managerial flexibility in 12.3 percent in 1988, largely thanks to better collection. the allocation of financial and human resources and greater But the scheme was not without its problems. The rest of accountability for results. Sweden and some other Nordic the civil service chafed at the special treatment afforded countries have long separated cabinet ministries from tax collectors, and the Ministry of Finance objected to its agencies with specific purposes. In the United Kingdom loss of authority. The program could not have gone for- nearly two-thirds of the civil service has now been moved ward without strong support from the top. BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR 91 More troublesome has been the establishment of devel- tralia, Colombia, Uganda) are emphasizing outcomes-- - opment project enclaves in several aid-dependent coun- the impact of outputs on beneficiariescombined with ex tries, each with its own system of remuneration and ac- post evaluation. countability. Often, donors have created these enclaves But even with greater clarity of purpose and task, pub- with little if any systematic consideration of the nature of lic sector managers will not perform well if the flow of the services provided or of the optimal sequencing of insti- budgeted resources is highly uncertain. This brings home tutional reforms. And they have often created disparities. the downstream importance of building effective policy- As discussed in Chapter 9, enclaves are usually de- making mechanisms at the center, as discussed earlier in signed as quick fixes. Although they have sometimes this chapter. A credible medium-term expenditure frame- accomplished short-term goals, they can create obstacles work, such as that in Australia (see Box 5.2), provides to deeper institutional reform. Where output is easily greater consistency between policies and resources and specifiedtax revenues collected, for exampleenclaves therefore more predictability in the flow of resources to may be useful as an experimental stage of reform that can units downstream. Malawi and Uganda are beginning to then be progressively extended, and as a demonstration institute such frameworks. that reforms can be effective. But it is important to FINANCIAL AND MANAGEMENT CONTROLS. Because employ systematic criteria in selecting which agencies to public sector outputs are often difficult to measure and hive off. And although they are a useful first step, enclaves monitor, financial control and accountability are needed cannot substitute for the longer-term institutional reforms to keep managers honest, prevent the poor useor needed to create a motivated, capable civil service. abuseof public resources, and improve service delivety. An expenditure tracking exercise in Uganda revealed that Imp roving delivery through the core public sector a significant portion of funds allocated for basic social ser- Contracting out, setting up performance-based agencies, vices never reach the intended health clinics or schools, and ensuring formal accountability for results are not particularly in rural areas. In many countries public finan- viable options for many services in countries with weak cial accounts and audits are outdated and inadequate and capacities. The challenge is particularly acute for street- therefore do not provide credible restraint. level bureaucracies whose operators (police, irrigation To improve the transparency and quality of their finan- patrollers, health workers, teachers, extension workers) cial accounting and auditing systems and reduce lags, interact daily with the people they serve, are geographi- countries are modernizing their financial information sys- cally dispersed, are granted substantial discretion, and tems. Countries as diverse as Bolivia, China, Indonesia, produce outputs that are difficult to monitor and are not and Moldova are writing sound accounting principles into subject to competitive pressure. law, backed by strong professional associations both Experience across a range of countries suggests that a within government and in the private sector. combination of mechanisms can boost incentives to per- Although systematic evidence about the effects of these form in such areas. A study comparing irrigation agencies reforms is not available, some useful policy lessons can in India and Korea revealed that the Indian organization nevertheless be gleaned. Modern, computer-based infor- provided few incentives for conscientious work, whereas mation systems can both improve transparency and the Korean agency was full of such incentives, Korean irri- strengthen aggregate control while reducing the need for gation patrollers had greater clarity of purpose and were controls on specific transactions. Controls on inputs can subject to random monitoring from three separate agen- be broadened sequentially as systems are expanded and cies. Supervision techniques in India sought to discover trust is built up: countries can gradually move away from grounds for punishment; in Korea they sought to solve detailed, ex ante line-item controls to broader budget cat- problems. With staff from all parts of the organization trav- egories, salary scales with broader bands, and greater lati- eling frequently up and down the canals in Korea, there tude in procurement and recruitment. The shift from ex was more external pressure from farmers, and stronger ante to cx post transaction-specific input controls should partnerships could be built for better performance. proceed carefully and only as controls over budgetary PERFORMANCE ORIENTATION AND PREDICTABILITY IN aggregates are improved, as agencies demonstrate that RESOURCE FLOWS. An important starting point in raising they can be trusted with greater autonomy, and as ex post the emphasis on performance within public sector organi- controls build strength through improved accounting and zations is greater clarity of purpose and task, Colombia, auditing capacities. Mexico, and Uganda are introducing performance mea- But experience also suggests that moving from a highly surement to orient managers to achieving desired results. centralized, transaction-specific control regime to a more Whereas some countries (New Zealand is one) have decentralized one can encounter resistance. For instance, stressed outputs as performance measures, others (Aus- in Ecuador a plan to devolve payment controls, although 92 WORLD DEVELOPMENT REPORT 1997 and attracting capable staff, require long-term career rewards, adequate pay, and mechanisms to instill an esprit Figure 5.5 Recruiting and promoting on merit de corps through, for instance, strong partnerships be- improve bureaucratic capability tween management and labor. Index of bureaucratic capability STRENGTHENING MECHANISMS OF CITIZEN VOICE. More Instituting credible bureaucratic controls takes time. capable Meanwhile, where internal monitoring and enforcement A capacity are weak, clients and beneficiaries can exert pow- erful pressures to improve performance, and so help reduce monitoring costs until these capacities are built up. Coproduction and other forms of government partnership with the community in providing services, even if indirect, create incentives to press for better service. Feedback mechanisms such as the client surveys used in India, Nicaragua, and Uganda increase transparency and im- prove accountability by making more people aware of the agency's performance. Citizen charters, in which agencies publicly commit themselves to minimum standards of ser- Less vice provisionas applied in Belgium, Malaysia, Portugal, capable and the United Kingdomcan be a fulcrum for increased 0 2 4 6 8 10 12 14 external and internal pressures to meet performance tar- Index of meritocracy gets, and help focus the attention of clients and staff on service quality. Chapter 7 puts the use of voice and partic- Note: Results are based on a regression using data from thirty- ipation in the broader context of raising state capability. five developing countries during 1970-90, controlling for other variables including income and education. See the Technical Note Fostering motivated, capable staff for details. Source: Evans and Rauch 1996. Whether making policy, delivering services, or adminis- tering contracts, capable and motivated staff are the proposed in 1995, has yet to be implemented, largely lifeblood of an effective state. Efforts to build a competent because of central agencies' fears of fiscal indiscipline. and dedicated civil service usually focus almost exclusively Trust needs to be built up first by strengthening systems on pay. Pay is certainly important, but so are other things, of performance measurement and ex post input controls. such as merit-based recruitment and promotion and esprit As these systems become more credible, resistance to the de corps. Meritocratic recruitment and promotion restrain changeover will fade, and managers can be provided political patronage and attract and retain more capable greater flexibility, with stronger accountability for results. staff. A healthy esprit de corps encourages closer identifi- LoYALTY, MOTIVATION, AND COMPETENCE. Better sys- cation with an organization's goals, reduces the costs of tems of monitoring, accounting, and auditing alone will making people play by the rules, and nurtures internal not suffice to improve the delivery of many services. partnerships and loyalty. Mechanisms for enhancing the loyalty, motivation, and A recent cross-country study shows why these things competence of the civil service are also needed. Loyalty matter. The authors found that an index representing promotes staff identification with an organization's goals meritocratic recruitment and promotion and adequacy and a willingness to take a longer-term view of responsi- of pay was correlated with economic growth as well as bilities. Loyalty is essential in the core public sector, where with investors' perceptions of bureaucratic capability, even activities are not easily specifiable or monitorable, and exit after allowing for income and educational differences has no meaning. (Figure 5.5). Clearly, then, finding the right institutional The civil services of France, Germany, Japan, and recipe for recruiting and motivating capable staff can reap Singapore all seek to ensure that the loyalties of a small sizable rewards. group of professionals are fundamentally aligned with the interests of the state. Job security in the civil service was Merit-based recruitment and promotion meant to foster this alignment of interests. It has worked Making a meritocracy of the civil service helps bring in in some contexts but not in many others. As discussed in high-quality staff, confers prestige on civil service posi- the next section, the experience of successful countries tions, and can do a great deal to motivate good perfor- suggests that building this commitment, and motivating mance. In many countries (Japan and Korea, for example) BUJLDINI3 INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR 93 a national civil service entrance examination uses tough rather than merit. Figure 5.6 shows that political appoint- standards to winnow the more skilled from the less skilled ments run much deeper in the Philippines than in other applicants. In others, academic (often college) perfor- East Asian countries. This, combined with poor pay, has mance is the primary filter. Pressure to perform is sus- resulted in lower bureaucratic capability, which has also tained after recruitment by specifdng clear objectives and worsened over time. The Philippine government is under- criteria for upward mobility and offering rewards for mer- taking reforms to introduce greater meritocracy in its civil itorious long-term service. In Korea, for instance, promo- service. Of course, political appointments can be quite tion is based on a formula that combines seniority with extensive in industrial countries, such as the United merit-based elements. Where instead promotions are per- States, as well. But countries with weak institutions and sonalized or politicized, civil servants worry more about inadequate checks and balances are better off relying on pleasing their superiors or influential politicians, and more transparent and competitive mechanisms. efforts to build prestige through tough recruitment stan- Even countries that have managed to install merit-based dards are undercut. recruitment and limit political appointments can suffer Meritocracy has not yet become established in many from rampant political interference in employee transfers. countries. Instead the state has often become a massive In India, for example, senior civil servants are transferred source of jobs, with recruitment based on connections frequently: the average tenure of field officers in some states Figure 5.6 Lack of meritocracy and poor pay in the Philippines' civil service have lowered capability Depth of political appointments Philippines Other East Asia Leve 3 Index of bureaucratic capability Level 4 1.0 0.8 0.6 Division chief Political appointees 0.4 Ratio of public to private wages (percent) 0.2 100 80 0.0 66 Philippines Other East Asia 60 40 Note: Other East Asia includes Indonesia, Korea, Malaysia, 26 Singapore, Taiwan (China), and Thailand. Pay differential is for 20 senior civil servants. Data are for 1984-88. Source: Adapted from Campos and Root 1996; Commander, Davoodi, and Lee, background paper. Philippines Other East Asia 94 WORLD DEVELOPMENT REPORT 1997 can be as tow as eight months. Partly as a result, the once- more difficult to maintain professional standards and legendary Indian civil service is no longer perceived as a esprit de corps. model of efficiency and effectiveness. Merit-based recruitment and promotion systems are of Adequate compensation two broad but not mutually exclusive types: mandarin sys- As countries build prestige for their civil service through tems and open recruitment systems. Mandarin systems, merit-based recruitment and promotion, government such as those found in France, Germany, and Japan, are employment becomes more attractive. But if civil service closed-entry, hierarchical systems with highly competitive compensation trails far behind that in the private sector, entrance requirements (Box 5.6). Where well-qualified prestige alone will not make up the difference. human resources are in short supply, mandarin systems A rough benchmark for evaluating the adequacy of may well be the preferred basis for a more selective public sector compensation is the gap between public and approach to personnel development. Open recruitment private sector compensation, discounted somewhat to systems, such as those in New Zealand and the United adjust for the fact that civil service employment is gener- States, provide a more flexible, decentralized, and increas- ally more secure. Precise comparisons are tricky, given ingly market-driven system of civil service recruitment. differences in benefits and perquisites, job requirements, Open systems give more flexibility to managers to match and the like. As a rule, however, civil servants nearly every- job requirements with available skills, including hard-to- where are paid less than their private sector counterparts. find technical expertise, although at the cost of making it In the Philippines, for example, public pay averages 25 Box 5.6 Cultivating the best and the brightest: Mandarin versus open systems Most countries have adopted one of two broad ap- driven approach to civil service recruitment. The U.S. proaches to meritocratic recruitment. The first, called system, for example, in stark contrast to the mandarin the mandarin system, is a hierarchical system with model, permits entry at any point in the hierarchy, entry limited to promising candidates at the outset of without age restrictions. Centralized competitive en- their careers. "Mandarin system" traditionally referred trance examinations have been replaced with profes- to an elite group of civil servants in certain East Asian sion-specific exams, and managers have been granted bureaucracies; in its more modern sense it refers to a more autonomy in hiring. The United States, like Aus- "corps-career" system that also includes lower and mid- tralia, complements its horizontal recruitment system dle levels of the civil service. Recruitment is centralized with a Senior Executive Service aimed at building an and highly selective, generally on the basis of a rigorous elite group from within the civil service. entrance examination. The successful candidates are The boldest approach to open recruitment and placed on a fast track into the best jobs in government. career development is found in countries pursuing the For the most part these recruits, who are mostly gener- "new public management" reforms. These countries alists by educational background, are hired into a have significantly devolved recruitment responsibili- career stream or corps rather than for specific jobs. ties. In New Zealand, for example, agency managers France and Japan best exemplify the mandarin sys- can hire staff at market salaries. tem. At France's National School of Administration, Countries with critical shortages of well-qualified future high-level cadres are put through a one-year human resources may find mandarin systems useful for professional internship followed by fifteen months of establishing a selective rather than a comprehensive coursework. Japan's Tokyo University produces that approach to personnel development. Moreover, a pres- country's administrative elite, most of whom have a tigious corps can have spillover effects, motivating other legal or generalist education, possibly supplemented parts of the civil service to perform well. Open recruit- with in-service technical training. Variants on the ment, on the other hand, gives managers greater flexi- system are found in Singapore, where two-year cadet- bility in finding candidates with needed skills, includ- ships rotate promising recruits, and Germany, where a ing hard-to-find specialist expertise. And open systems "practicum" system offers practical, on-the-job intern- discourage insularity in the civil service by bringing in, ships for outstanding candidates. at all levels, staff with fresh perspectives and new ideas. The second system, known as open recruitment, is The downside is that professional standards are less easy a more flexible, decentralized, and increasingly market- to maintain across the service, as is esprit de corps. BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR 95 Figure 5.7 In Africa, as public employment has risen, wages have fallen Average real central government wages relative to GDP per capita High A I S S.. I I I S . I IT ) . I ia , U. .% S I ?. S Low / Low High Central government employment relative to population ote: Data are panel data for twenty African countries for the period 1972-93. Both axes are scaled logarithmically. See the Technical Note for Jetails. Source: World Bank staff calculations based on data in Kraay and Van Rijckeghem 1995 and World Bank data. percent of private wages; in Somalia the figure is 11 per- The contrast between the size of the wage bill and the cent. (At the other extreme, public sector salaries in Sin- relative level of wages shows through in another major gapore average 114 percent of those in the private sector, problem, which will not be solved by across-the-board and in fact senior Singaporean civil servants are better increases. This is the relatively low level of remuneration paid than their U.S. counterparts.) The gap is widening for senior officials. In many countries the wages of higher- in some countries. In Kenya, for instance, the disparity level civil servants are allowed, often for political reasons, to between public and private wages grew by 3 percent a year erode by more than those of the lower echelons; this wage during 1982-92. Contributing in many countries to the compression makes it even harder to attract and retain relative erosion of public sector wages were fiscal austerity high-quality staff at the vital senior levels. A study of ten measures during the 1980s that tended to lower real wages African countries found that, on average, the ratio of the rather than employment. salaries of the highest- to those of the lowest-grade civil In many African countries, public employment has not servants had declined from thirteen to one to nine to one merely been maintained in the face of rising fiscal pres- during the 1980s. suresit has actually risen, especially at lower skill levels Prompted by a desire to raise public sector salaries, and (Figure 5.7). Governments have become employers of last by the need to correct aggregate fiscal imbalances, some resort and dispensers of political patronage, offering jobs countries have embarked on initiatives to reduce employ- to family, friends, and supporters. As a result, individual ment, decompress the wage structure, and raise average wages have often been iow even though the overall wage pay in the civil service. These efforts have met with mixed bill has been high. Moreover, growth in the wage bill has success. A study of civil service reform in fifteen countries often outpaced expenditures on operations and mainte- in various regions between 1981 and 1991 found that nance, leading to the familiar stories of teachers without wage bill reduction and salary decompression had been textbooks and extension workers without bicycles. achieved in fewer than half the cases. Employment had 96 WORLD DEVELOPMENT REPORT 1997 been reduced in more than half, but reversals were some- members a sense of purpose and belonging and imposes times reported later, and the cuts were rarely adequate to self-discipline that guides members toward achieving the finance substantial salary increases for higher-level staff. In group's objectives. King Arthur's Knights of the Round Peru, for instance, some 250,000 workers were cut from Table, the samurai in Japan, and even the mafiosi of past the civil service over three years, but 163,000 of them were generations all embodied some form of esprit de corps. A subsequently rehired; in addition, poor targeting of the few of today's civil services are also said to do so, includ- cuts resulted in the departure of the most qualified staff. ing those in Chile, France, Germany, Japan, Korea, and This mixed and often disappointing experience with the United Kingdom. Most, however, do not. civil service reform nevertheless provides some lessons for It is not impossible to build an esprit de corps within future efforts. First, strategies have focused exclusively on the bureaucracy from scratch. Singapore's civil service is pay and employment and, within those issues, on reduc- now well known for its coherence and sense of purpose, ing numbers (the wage bill and employment). These are even though these characteristics barely existed in the important, but so are other, complementary elements: early 1960s. Getting there was difficult, but the steps were merit-based recruitment and promotion, performance straightforward enough. Each year prospective recruits are measurement and orientation, mechanisms to improve taken from the top 200 (less than 5 percent) of the grad- accountability, and esprit de corps. uating class at the National University of Singapore (and A more careful sequencing of reforms is called for, start- more recently the Nanyang Technological University) ing with wage decompression. Even within constrained and put through a one-year training program. Their edu- overall wage bills, wages at the top of the scale can be cation and training give the recruits a common under- raised relative to lower levels, to attract more qualified standing of what is expected of them as civil servants and people and concentrate scarce skills in strategic areas. In help build trust among them. The country's meritocratic governments that are considerably overstaffed, reforms promotion system gives officials a stake in achieving their have been too modest to downsize them to sustainable agency's goals. The single-mindedness of Singapore's levels. And they have tended to be one-shot reductions leadership and its continuous efforts to imbue the civil rather than a steady program to redimension government service with its desired values help strengthen the bond over the longer term. among civil servants. Some of Singapore's lessons are now Inevitably, pay and employment reforms will face being learned in Botswana, in twinning arrangements that political obstacles, although fears of political backlash have emphasize the two key ingredients of teamwork and often been exaggerated. Some countries have viewed civil group performance. servants as partners in reform and have consulted exten- Worker dedication and commitment are not confined sively with them to find politically acceptable solutions. to the industrial countries and East Asia. In Brazil's poor For instance, in the province of Santa Fe, Argentina, a northeastern state of Ceará, steps to increase worker com- close dialogue between the governor and the local civil ser- mitment dramatically improved the quality of public vice union helped the parties agree on measures for mod- services delivered (Box 5.7). The state government pro- ernizing provincial public administration, including moted a sense of calling among workers and conferred new expenditure cuts of some 10 percent. In addition, the prestige on their jobs. These feelings were reinforced by experience with civil service reform has helped develop a innovative practices such as worker participation and self- good set of technical toolscivil service censuses, func- managed worker teams, multitasking, and flexibly orga- tional reviews, better-designed severance packagesfor nized or specialized production. These practices involved managing and implementing reforms more effectively. greater worker discretion and flexibility, greater coopera- But civil service reforms will generate losers, who can be tion between labor and management, and greater trust important constituencies of the political leadership and between workers and customers. Such experiences also therefore a force to be reckoned with. The political econ- underscore the importance of nonmonetary rewards omy of reforms is discussed further in Chapter 9. recognition, appreciation, prestige, and awardsin moti- vating staff, over and above the adequacy of pay and mer- Building esprit tie corps itocratic recruitment and promotion. Effective and capable bureaucracies share a commitment to their organization's objectives. This esprit de corps Strategic options: Steps toward an effective includes a common understanding about what is desirable public sector and undesirable behavior, manifested in formal and infor- Some developing countries lack the most basic under- mal norms and grounded in a set of objectives, and a pinnings of a professional, rule-based bureaucracy. Even devotion to upholding the honor of the group, based on reform-minded leaders cannot translate their goals into this common understanding. An esprit de corps gives reality because the machinery linking policy statements to BUILDING INSTITUTIONS FOR A CAPABLE PUBLIC SECTOR 97 B ox 5.7 Building worker dedication: Good government in Brazil's Ceará State In 1987 the state government of Ceará in northeastern larger variety of tasks than usual, often voluntarily. Brazil confronted a crippling fiscal crisis, superimposed Granted greater autonomy and discretion, workers on a legacy of mediocre administrative performance. were able to provide more customized service. This Yet within four years the fiscal crisis had been over- greater discretion did not result, as it often does else- come, and the quality of services had improved where, in greater opportunities for rent seeking because dramatically. Vaccination coverage for measles and of pressures to be accountable. Workers wanted to per- polio more than tripled, from 25 percent to 90 per- form better in order to live up to the new trust placed cent of the child population. The state's public works in them by their clients. This in turn was the result of program employed more than 1 million unemployed the more customized arrangements of their work and farmers during droughts. And its business extension the public messages of respect from the state, At the and public procurement program for small firms was same time, the communities where these public ser- saving more than 30 percent over its previous overall vants worked monitored them more closely. The expenditure. state's publicity campaigns and similar messages had Much of the credit for this success is owed to the armed citizens with new information about their rights civil service itself. The state government contributed in to better government and about how public services an unusual and sometimes inadvertent way to public were supposed to work. Thus government played a workers' newfound dedication. Using rewards for good powerful role in monitoring, but it did so indirectly. performance, public screening methods for new These mechanisms created a virtuous cycle in which recruits, orientation programs, and sheer boasting workers reported feeling more appreciated and recog- through the media about its successes, the state created nized, not necessarily by superiors, but by their clients a strong sense of mission around key programs and and the communities where they worked. This, in their workers. Highly motivated workers carried out a turn, reinforced their dedication to the job. action has ceased to function. As a result, a vast gap has task and by introducing performance measurement on a opened up between what the state says it will do and what selective, sequential basis. 'XThen output measurement is it doesbetween the formal rules of public institutions strengthened and credible controls over inputs are insti- and the real ones. The first step toward building a more tuted, managers can be granted more operational flexibil- effective public sector in such countries must be to close ity in exchange for greater accountability for results. these gaps: to reestablish the credibility of the government's Instituting a professional, rule-based bureaucracy will policies and the rules it claims to live by, making sure they take time. In the meantime some other measures can be operate in practice. This includes setting hard budget lim- implemented more quickly, some of which can generate its, implementing budgets and other policies as approved, early payoffs. Well-functioning policymaking mecha- making the flow of resources predictable, instituting nisms make transparent the costs of competing policies accountability for the use of financial resources, and curb- and encourage debate and consultation among all stake- ing rampant political patronage in personnel decisions. holders. Using the market to deliver contestable ser- Where these preconditions are absent, the new public vicestoo many of which are now the sole domain of management must be introduced cautiously. If informal governmentcan lower costs and improve service qual- norms have long deviated significantly from formal ones ity. Likewise, contracting out easily specified activities (with regard to personnel practices, for example), simply through competitive bids can reduce the burden on over- introducing new formal rules will not change much. stretched capacity and build partnerships with markets Where countries have been unable to establish credible and NGOs to improve efficiency. And reformers need not controls over inputs, giving managers greater flexibility rely only on internal controls: creating more points of will only encourage arbitrary actions and corruption. And access for feedback from firms and the people who use where specialized skills are in short supply, performance public services can do a lot to generate external pressures contracts and other output-based contracts for complex for better performance while internal capacity and en- services may absorb a large share of scarce bureaucratic forcement mechanisms are still developing. capacity to specify and enforce them. Nevertheless, coun- As previous chapters have stressed, getting the state tries can begin by providing greater clarity of purpose and out of the business of providing many of the goods and 98 WORLD DEVELOPMENT REPORT 1997 services it now provides will still leave it plenty to do. tiveness of salaries at the upper end can be raised, to attract Where it is no longer a direct provider, the state must capable staff, even within constrained wage bills, in part by become a partner and facilitator, regulating markets, reducing excess employment among the rank and file. enforcing contracts, and pursuing all the essential roles There is bound to be opposition to these and other outlined in Chapters 3 and 4. That means attracting and reforms from those who stand to lose. But as discussed in keeping capable staff. Rule-based restraint on political Chapter 9, windows of opportunity for reform can open patronage in recruitment and promotion, and more com- and widen. Reform-minded governments should seize petition through meritocracy, will be necessary to build these opportunities to build consensus, address the obsta- this capability. In countries where rapid employment cles to change, and initiate and sustain reforms to build an expansion has taken place in the lower echelons, reforms to effective public sector. The resulting reinvigoration of reduce employment are inevitable. But the relative attrac- public institutions will generate large payoffs. RESTRAINING ARBITRARY STATE ACTION AND CORRUPTION EFFECTIVE STATE CAN CONTRIBUTE POWERFULLY the payoffs from, corruption while raising the price and ANto sustainable development and the reduction of the probability of being caught. poverty. But there is no guarantee that state intervention Formal checks and balances will benefit society. The state's monopoly on coercion, which gives it the power to intervene effectively in eco- In framing a government to be administered by men nomic activity, also gives it the power to intervene arbi- over men, the great difficuliy lies in this: you must first trarily. This power, coupled with access to information enable the government to control the governed; and in not available to the general public, creates opportunities the next place oblige it to control itself for public officials to promote their own interests, or those James Madison, FederalistNo. 51(1788) of friends or allies, at the expense of the general interest. The possibilities for rent seeking and corruption are con- Restraining the potential use and abuse of state power is siderable. Countries must therefore work to establish and a challenge for any country. Harder still is doing it with- nurture mechanisms that give state agencies the flexibility out depriving state agencies of the flexibility they need to and the incentive to act for the common good, while at do their job. The misuse of state power creates serious the same time restraining arbitrary and corrupt behavior problems of credibility, whose effects linger long after in dealings with businesses and citizens. the event. The expropriation of property and the harass- Chapter 5 focused on building the capability of the ment of entrepreneurial Asian minorities in Uganda public sector. Many of the reforms discussed there will under Idi Amin left a legacy of distrust, which initially contribute to reducing arbitrariness and corruption. Par- posed enormous problems when the current administra- ticularly useful toward that end are instilling a rule-based tion tried to attract private investment. But arbitrary and culture in public institutions, and curbing patronage in capricious state action undermines more than credibility. the civil service. This chapter broadens that discussion to It undermines the rule of law itself, by weakening the look at mechanisms to restrain arbitrary state action and force of whatever rules the state has set in place. And it corruption more generally. fosters conditions that encourage state officials to place The chapter examines first the formal checks and bal- themselves above the law and tempt the rest of society to ances that need to be built into the structure of govern- do the same. Development, in these circumstances, hits a ment, including judicial independence and the separation brick wall. of powers. These promote credibility and accountability. But formal instruments of restraint are seldom enough, Instruments of restraint particularly in countries where corruption has become Sustainable development generally calls for formal well entrenched. Therefore this chapter also analyzes the mechanisms of restraint that hold the state and its officials options for these states and others seeking to make a dent accountable for their actions. To be enduring and credible, in corruption, by examining its root causes. One impor- these mechanisms must be anchored in core state institu- tant lesson is that anticorruption efforts must proceed tions; if these are too weak, external mechanisms such as along many fronts, to reduce the opportunities for, and international adjudication may substitute temporarily. The 100 WORLD DEVELOPMENT REPORT 1q97 two principal formal mechanisms of restraint are a strong, bailiffs are not under the control of judges but are em- independent judiciary and the separation of powers. ployed by the Ministry of Justice. Thus, although judges JUDICIAL INDEPENDENCE AND EFFECTIVENESS. To are competent and reasonably efficient, enforcement is prosper, economies need institutional arrangements to slow and often ineffective because the number of bailiffs resolve disputes among firms, citizens, and governments; has not kept pace with the rising caseload. to clarify ambiguities in laws and regulations; and to Developing relationships among the judiciary, legisla- enforce compliance. Societies have devised a broad array ture, and executive that ensure judicial independence and of formal and informal mechanisms to do this, but none reliable enforcement is a gradual process. Studies show more important than the formal judiciary. It alone has that private sector confidence in the rule of law increases access to the coercive authority of the state to enforce with each year a stable regime remains in place. More judgments. And it alone has the formal authority to rule broadly, as Box 6.1 illustrates for Peru, the success of on the legality of actions by the legislative and the execu- third-party mechanisms for enforcement depends in large tive branches. This special relation to the rest of the state part on citizens viewing those mechanisms as legitimate. puts the judiciary in a unique position to support sustain- In countries where judicial institutions are weak, it may be able development, by holding the other two branches at least as important to demonstrate to citizens and firms accountable for their decisions and underpinning the the potential benefits of a well-functioning judiciary, and credibility of the overall business and political environ- to win support for good laws and impartial enforcement, ment. Yet judiciaries can play this role only when three as it is to proceed with wholly technocratic programs of core conditions are met: independence, the power to judicial reform. enforce rulings, and efficient organization. The third component of judicial effectiveness is organi- Independence from the rest of government is the most zational efficiency, which is needed to avoid long delays in important of these. Whatever the precise character of clearing cases. The average case takes 1,500 days to clear in judicial relations with the legislature and the executive, all Brazil and Ecuador, but only 100 days in France. Long de- industrial countriesand many developing countries lays raise the transactions costs for dispute resolution and rely on the judiciary to hold the executive accountable may block access for some potential users; however, the under the law and to interpret and enforce the terms of internal efficiency of the judiciary is less critical than its the constitution. independence and its enforcement authority. As discussed Judicial independence has been repeatedly compro- in Chapter 3, even when saddled with cumbersome and mised in some countries, and in no country has the judi- costly procedures, judicial systems can strengthen credibil- ciary been immune from political efforts to override its ity in countries as long as their decisions are perceived to decisions. Legislatures and executives have used a variety be fair. Any state beginning from a weak institutional base of gambits to rein in their judiciaries: should consider building this aspect of judicial perfor- mance its first priority. Judges of the superior court in Malta were suspended SEPARATION OF POWERS. Judiciaries may be capable of one hour before a case challenging executive actions enforcing rules, but if the public has little faith in those was to be heard. rules remaining stable, the state's credibility can still be A succession of Pakistani governments in the past compromised. The classic constitutional mechanism for appointed temporary judges, whose lack of tenure restraining constant legislative changes is the horizontal made them more vulnerable to political influence. and vertical separation of powers. Although Ukraine's constitution declares that the courts Power can be divided horizontally among the judi- are independent of the executive, judges remain largely ciary, the legislature, and the executive, and vertically be- dependent on local authorities for their housing. Judges tween central and local authorities. The patterns of a who have ruled against city officials appear to be partic- country's political party organizationwhich can range ularly susceptible to long delays in getting housing. from a small number of highly disciplined parties to a large number of parties whose members only loosely abide The effectiveness of the judiciary also depends on its by a party line, and that can govern only by forming multi- decisions being enforced. In practice that means that party coalitionsalso influence the extent to which polit- other branches of the government must consent to pro- ical power is concentrated or diffused. vide the resources needed for enforcement, including per- The broader the separation of powers, the greater will sonnel authorized by law to serve court documents, to be the number of veto points to be navigated to change seize and dispose of property, and to turn the proceeds any rule-based commitments. Thus the separation of over to the winning parry. In many countries this enforce- powers increases confidence in the stability of rules. Mul- ment capability is constrained. In Poland, for instance, tiple veto points can be a double-edged sword, however: RESTRAININC ARBITRARY STATE ACTION AND CORRUPTION 101 Box 6.1 How popular participation improved property rights and dispute resolution in Peru Until 1989 most Peruvians living in marginal urban monopolies that held a stake in the old system, such as settlements and rural areas did not enjoy the security lawyers' associations and notaries, strongly opposed the provided by formal ownership of their real property proposed reforms, community-level support carried the (70 percent of the population in urban areas and 80 day. percent in rural areas). The traditional system of prop- In 1989 a new property registration system was erty registration was run by the Ministry of Justice, and enacted into law. The new system dramatically reduced conflicts were resolved by the judiciary. The system transactions costs and uncertainty by reducing the was perceived by poor urban and rural property own- power of the professional monopolies. Instead the sys- ers as a system for the rich, who could better afford the tem uses community norms, such as neighbors vouch- high transactions costs. ing for a party's ownership claim, to establish property In the early 1980s the Institute of Liberty and rights and resolve conflicts. Subject to administrative Democracy (ILD), an NGO, began a campaign to requirements specified by the law, any lawyer may serve improve the property rights of poor Peruvians. The as a third-party verifier, sign the property titles, and ILD began by holding extensive public hearings to resolve conflicts in the field. Contested or complicated gather complaints, identify reasons why citizens did cases are resolved by the new system's chief registrar, not formally register their property, and publicize the who is appointed by the Ministry of Housing, not the potential benefits and costs of registration and secure judiciary. Only after these mechanisms have failed can property. The group coupled this participatory process the conflict be taken to a judge. with a study of the laws and regulations governing By 1994 the new system had registered nearly property registration and enforcement. Based on the 120,000 entities, and between 1994 and 1996 it regis- resulting diagnosis, the ILD then developed concrete tered an additional 170,000. Spurred by its demand- proposals for reform. These were publicly debated and driven design, the system continues to evolve, and ini- Line-tuned starting in 1986. Although the professional tiatives are under way to expand it nationwide. they make it just as hard to change harmful rules as to anchored by any domestic institution. On the trade front, change the beneficial ones. both the European Union and the North American Free Many developing countries, including some with for- Trade Agreement have been able to play this role, and mal separation of powers, have few effective checks and many countries will find it an important reason to join the balances on the actions of political leaders. In some coun- World Trade Organization. Clearly, sovereign countries tries legislative oversight is weak because of poor capacity can still reverse course on, for example, trade policy by and inadequate information. In others the executive dom- withdrawing from such agreements. But they then have to inates a compliant legislature. But like the development of calculate not just the benefits and costs of the policy rever- a well-functioning judicial system, the formal elaboration sal, but also the broader costs of reneging on an interna- of constitutional checks and balances, or their more effec- tional commitment for which their partners will hold tive institutionalization, is a gradual process. them accountable. The threat of international censure EXTERNAL MECHANISMS. To some extent, extraterritor- makes countries less likely to reverse course. ial and international restraints can substitute for limita- Agreements with multilateral organizations, such as the tions on the ability of national institutions to enforce rules IMF or the World Bank, often include some degree of pol- or to signal credibly that the rules will remain reasonably icy conditionality: in order to borrow funds, for example, stable over time. One option is to use extraterritorial adju- countries undertake certain reforms. This can have a simi- dication to underpin the domestic judicial system. Confi- lar benefit for some countries. These conditionalities can be dence in the Jamaican judicial system is buttressed by the viewed as a sign of national commitment to the policies that fact that the United Kingdom's Privy Council serves as its are included as conditions. Countries with weak domestic appellate court of last resort. Because of the weaknesses of commitment mechanisms can strengthen their credibility the Philippine judicial system, many firms, domestic as by binding themselves to pay a penalty should they violate well as foreign, prefer to adjudicate their contracts offshore. the agreement. One of the intentions behind World Bank As Chapter 3 noted, international agreements are a sec- guarantees is to accelerate the flow of private finance to ond mechanism for strengthening commitments not developing countries by underpinning such commitments. 102 WORLD DEVELOPMENT REPORT 1997 Building in flexibility 1996 a blockbuster movie, Hindustani, expressed an ex- Instruments of restraint are a vital foundation for sustain- treme form of popular outrage over corruption. The able development. But excessive restraint can lead to movie depicts horrific tales of callous politicians and paralysis. Instruments for restraining government need to bureaucrats willing to let hospital patients die and poor be complemented by institutional arrangements that pensioners starve unless they receive their cut. build in flexibility for the executive branch in formulating Fiction thrives on exaggeration. But it is also a mirror and implementing policies and adapting to new informa- that society holds up to itself to reflect entrenched prob- tion and changing circumstances. lems. These talesa century and a hemisphere apart As discussed in Chapter 5, countries have tried a vari- remind us that corruption is nothing new, nor is it con- ety of institutional arrangements that combine flexibility fined to any particular corner of the world. It is a problem with restraint. Some arrangementssuch as deliberation that has deeply affected the lives and stirred the resent- councils in East Asia and the Administrative Procedures ment of citizens and businesspeople the world over. Act in the United Statesdelegate substantial autonomy Today, citizens everywhere demand greater probity of to executive agencies to define the substance and under- government officials, and the new transparency in domes- take the implementation of policy. But they also require tic and global markets brings corruption more quickly to these agencies to follow procedures that open their deci- the public eye. In the past few years allegations of corrup- sions to input and oversight by other arms of the state and tion have contributed to the fall of governments through- by civil society and businesses. Other arrangements rely out the world. Two former presidents of the Republic of on mechanisms within the executive branch to promote Korea have been prosecuted and indicted. A president of flexibility within restraints, such as the devolution of Brazil was impeached on charges of corruption. In Octo- managerial authority to executive agencies within set bud- ber 1996 more than 250,000 people protested the Belgian gets and performance targets. government's handling of a pedophile ring and alleged But even if bureaucracies are embedded in processes corruption in judicial appointments and enforcement. that provide ample opportunity for outside input and And a fierce debate has erupted over the financing of oversight, the risk remains that officials will pursue per- political campaigns and its influence on public policy in sonal rather than organizational goals. Self-seeking behav- the aftermath of the 1996 U.S. presidential election. ior can degenerate into corruption when private interests Corruption has been defined in many ways. This wield their influence in illegal and secret ways, circum- Report defines it as the abuse of public power for private venting the legal and bureaucratic rules designed to keep gain. Although corruption tends to get the most atten- them out. Whether public institutions succumb to these tion, it is a symptom of a more general problem of per- and other sources of corruption will depend on the verse underlying incentives in public service. Corruption strength of their institutional defenses. How these are flourishes where distortions in the policy and regulatory built and maintained is the subject of the next section. regime provide scope for it and where institutions of restraint are weak. The problem of corruption lies at the Controlling corruption intersection of the public and the private sectors. It is a .A A Congressional appropriation costs money. . . two-way street. Private interests, domestic and external, majority of the House Committee, say $10,000 wield their influence through illegal means to take advan- apiece$40, 000; a majority of the Senate Commit- tage of opportunities for corruption and rent seeking, and tee, the same eachsay $40,000; a little extra to one public institutions succumb to these and other sources of or two chairmen of one or two such committees, say corruption in the absence of credible restraints. $10,000 each. Then seven male lobbyists at $3,000 Corruption violates the public trust and corrodes social each; one female lobbyist, $10,000; a high moral Con- capital. A small side payment for a government service gressman or Senator here and therethe high moral may seem a minor offense, but it is not the only costcor- ones cost more. ruption can have far-reaching externalities. Unchecked, A U.S. railroad company owner in Mark Twain the creeping accumulation of seemingly minor infractions and Charles Warner, The Gilded Age: can slowly erode political legitimacy to the point where A Tale of Today (1877) even noncorrupt officials and members of the public see little point in playing by the rules. Mark Twain's damning tale was a thinly veiled caricature Studies have shown a clear negative correlation between of corruption in the U.S. Congress in the 1 870s. Twain's the level of corruption (as perceived by businesspeople) novel followed closely on the heels of the infamous Cre'dit and both investment and economic growth. This is con- Mobilier scandal, in which two prominent businessmen firmed for investment levels by the results of the private brazenly bought their way into the Congress. In India in sector survey conducted for this Report (Figure 6.1). As we RESTRAININ3 ARBITRARY STATE ACTION AND CORRUPTION 103 saw in Chapter 3, the survey identified corruption as one of the major obstacles to doing business in many countries. Figure 6.1 High and unpredictable corruption Yet it is not just a cost of doing business. Other surveys and hurts investment anecdotal evidence suggest that the greatest victims of petty corruption are usually the poor. Gross investment GDP ratio Despite such evidence, many parts of the developing (percent) world retain a certain ambivalence toward corruption. A 30 commonly heard view is that corruption merely greases the wheels of commerce, and that without it there would 25 be no transactions and no growth. Apparent support for this argument comes from the fact that some countries 20 that rank high in surveys of the level of corruption have also excelled in economic growth. The predictability of 15 corruptionboth that of the amount one has to pay and 10 that of receiving the outcome one has paid forprovides some insights into this apparent paradox. For a given level of corruption, countries with more predictable corruption have higher investment rates (Figure 6.1). But even in High these countries corruption has an adverse impact on eco- Low nomic performance. Figure 6.1 also shows that, no matter High Low how high the degree of predictability of corruption in a Level of Predictability country, its rate of investment would be significantly corruption of payments and outcomes higher were there less corruption. Countries that have so far achieved high rates of eco- Note: Each value is an average for a group of countries. Results nomic growth despite serious corruption may find them- are based on a regression using data from thirty-nine industrial and developing countries, controlling for income, education, and selves paying a higher price in the future. Tolerating cor- policy distortion. See the Technical Note for details. Source: ruption that siphons off payments of, say, 10 percent on World Bank staff calculations based on data from the private average may generate pressures to increase the take to 15 sector survey Conducted for this Report. or 20 percent. Corruption feeds on itself, creating a widening spiral of illegal payoffs until ultimately develop- ment is undermined and years of progress are reversed. And the very growth that permitted corruption in the past endemic, these officials may create additional red tape and can produce a shift from productive activities to an unpro- delays to induce even higher payments. Of course, corrup- ductive struggle for the spoils. Over time corruption tion also occurs at the highest levels of government, in the becomes entrenched, so that when governments finally do awarding of major contracts, privatization, the allocation move to contain it, they meet powerful resistance. of import quotas, and the regulation of natural monopo- lies. This helps explain why corruption is more prevalent Causes of corruption in countries with highly distorted policies, as measured by Incentives for corrupt behavior arise whenever public offi- variables such as the black market exchange rate premium cials have wide discretion and little accountability. Politi- (top left panel in Figure 6.2). Any policy that creates an cians, bureaucrats, and judges control access to valuable artificial gap between demand and supply creates a prof- benefits and can impose costs on private citizens and busi- itable opportunity for opportunistic middlemen. nesses. Public officials may be tempted to use their posi- The probability of being caught and punished (for the tions for private gain by accepting bribes; for their part, person paying the bribe and for the official receiving it) private individuals may be willing to make illegal pay- also affects the level of corruption. Economic analysis of ments to get what they want from government. Thus, a the law suggests that individuals weigh the expected ben- necessary condition for corruption is that public officials efits of breaking the law against the expected costs (the have rewards and penalties at their disposal. probability of being caught and punished multiplied by Some corruption stems from opportunities generated the level of punishment). Corruption may be high in a by the policy environment, at the bottom or the top of country where the government system does little to deter the hierarchy. Payoffs are frequent to lower-level officials bribes. Lawbreakers may believe that there is little chance charged with collecting tariffs, providing police protec- of being caught or, if caught, of having to pay the penalty, tion, issuing permits, and the like. When corruption is since they believe that the system of justice itself can be 104 WORLD DEVELOPMENT REPORT 1997 Figure 6.2 Some factors associated with corruption Index of corruption Index of corruption 6 6 5 4 3 2 1 0 0 1 2 3 4 5 6 7 8 2 3 4 5 6 Policy distortion index Predictability of judiciary index Index of corruption Index of corruption 6 6 5 5 4 4 3 3 2 2 1 1 0 0 0.5 1.0 1.5 2.0 25 0.5 1.0 1.5 20 Ratio of civil service wages Merit-based recruitment index to manufacturing wages Note: Data for the top left panel are from thirty-nine industrial and developing countries during 1984-93 (for the policy distortion index) and 1996 (for the corruption index). The top right panel is based on a regression using data from fifty-nine industrial and developing countries during 1996. Data for the bottom left panel are from twenty industrial and developing countries in the late 1980s to early 1990s. The bottom right panel is based on a regression using data from thirty-five developing countries during 1970-90, controlling for income. See the Technical Note to Figure 3. Source: World Bank staff calculations (top two panels); Van Rijckeghem and Weder, background paper (bottom left panel); Evans and Rauch 1996 (bottom right panel). corrupted. Corruption can even persist in countries with a serious threat if more remunerative positions are avail- substantial press freedom and public resentment against able legally in the private sector. Hence corruption is it, if there is little hope of independent judicial resolution often positively associated with the difference between of important cases. This Report's private sector survey public and private salaries, or what may be termed the found a negative correlation between reported levels of "rate of temptation" (bottom left panel in Figure 6.2). But corruption and judicial predictability (top right panel in simply raising civil service salaries may not reduce corrupt Figure 6.2). behavior. Pay reform must be combined with credible Finally, corruption may thrive if the consequences of monitoring and law enforcement. Merit-based recruit- being caught and disciplined are low relative to the bene- ment and promotion mechanisms that restrain political fits. Officials frequently control the allocation of benefits patronage and create a more impartial public service are and costs whose value far exceeds their own salaries, Cor- also associated with lower corruption (bottom right panel ruption becomes especially likely if the wages of public in Figure 6.2), service do not reflect the comparable private wage. 'Where civil service wages are very low, officials may try to eke out Reducing corruption a middle-class standard of living by supplementing their Several countries have managed to reduce endemic cor- pay with illegal payoffs. The risk of being fired from a ruption over time. The struggle of the Progressive move- low-paying civil service job because of corruption is not ment against the power of U.S. urban political machines RESTRAINING ARBITRARY STATE ACTION AND CORRUPTION 105 of the nineteenth century is a case in point (Box 6.2). theme of Chapter 5: creating a rule-based bureaucracy Containing corruption requires an understanding of the with a pay structure that rewards civil servants for honest benefits and costs under public officials' control. Many efforts, a merit-based recruitment and promotion system officials remain honest despite considerable temptation, to shield the civil service from political patronage, and and many ordinary people and businesses refuse to pay credible financial controls to prevent the arbitrary use of bribes despite the promise of short-term gain. But others public resources. Here we focus on the remaining two succumb. It is unwise to deal with the possibility of cor- parts of the strategy. The first is to reduce the opportuni- ruption by assuming that government officials are of ties for officials to act corruptly, by cutting back on their higher moral standing than the rest of the population. discretionary authority. The second aims at enhancing The actual extent of bribery and other corruption accountability by strengthening mechanisms of monitor- depends not just on the potential gains and risks, but also ing and punishmentusing not only criminal law but on the relative bargaining power of the buyer and the also oversight by formal institutions and ordinary citizens. seller of public favors. Reformers must also consider the REDUCING OPPORTUNITIES FOR CORRUPT PPACTICE. fact that anticorruption efforts have marginal costs as well In general, any reform that increases the competitiveness as marginal benefits; the efficient level of bribery will sel- of the economy will reduce incentives for corrupt behav- dom be zero. ior. Thus policies that lower controls on foreign trade, Corruption cannot be effectively attacked in isolation remove entry barriers to private industry, and privatize from other problems. It is a symptom of problems at the state firms in a way that ensures competition will all intersection of the public and the private sectors and support the fight. If the state has no authority to restrict needs to be combated through a multipronged strategy. exports or to license businesses, there will be no opportu- Recent reforms in Uganda illustrate such an approach nities to pay bribes in those areas. If a subsidy program is (Box 6.3). One part of the strategy focuses on a major eliminated, any bribes that accompanied it will disappear Box 6.2 Urban political machines in the United States and their reform In the late nineteenth and early twentieth centuries wealthy property owners and introduced a plan to many U.S. cities were dominated by political machines, assess property at market value. The plan increased the defined by one scholar "as a political party in which a assessed value of real estate, lowered the tax rate, and boss oversees a hierarchy of parry regulars who provide increased revenues. The city's budget was cut by $1.5 private favors to citizens in exchange for votes and million as patronage appointees were removed from who expect government jobs in return." Machine- office. Reform mayors in many other U.S. cities fol- controlled cities also typically made corrupt, collusive lowed similar policies. deals with private businesses seeking contracts, fran- Reform also involved municipal franchises. In chises, or protected markets. The politicians driving Philadelphia, for example, the machine-controlled city these machines operatedand flourishedin nomi- council regularly awarded a gas franchise in return for nally democratic environments. contributions to the Republican Party. In 1905 reform Machines were costly to the communities they dom- mayor John Weaver vetoed the franchise bill, ap- inated. Spending per capita for general administration pointed a supporter to a key position on the city coun- and for police and fire servicesboth areas with lots of cil to ensure that the veto would be upheld, and had patronage jobswas 34 percent and 17 percent higher, machine adherents arrested on charges of corruption. respectively, in machine-controlled than in nonmachine Cities dominated by machines paid a high cost in cities. To take one extreme case, in Boston the number the form of inflated budgets and inequitable tax and of city clerks increased by 75 percent between 1895 and spending systems. Although many people benefited 1907, while the population increased by less than a from the jobs and patronage dispensed by the quarter; meanwhile growth in productivity fell by half. machines, the losers were more numerous still. It was The Progressive movement in the United States had they who eventually organized to elect reform candi- as one of its main goals the reform of machine-domi- dates in many cities. The wave of reform mayors nated cities. Reform frequently meant property tax effected real change, which persisted even when the reform. Seth Low, New York's reform mayor in the machines were returned to power, mainly because the early 1900s, was distressed by the favoritism shown to reforms were popular and hard to reverse. 106 WORLD DEVELOPMENT REPORT 1997 Box 6.3 Fighting corruption in Uganda Uganda, long plagued by systemic corruption, has civil service (including thousands of "ghost workers") launched a multipronged battle against it. The effort and a thin and porous tax base. enjoys support from the country's leaders, who seem Cleaning up the civil service will take years, but committed to the goal of sound governance. Uganda is making some progress. The effort includes in the immediate postcolonial period Uganda was a policy reform and deregulation to remove opportunities kleptocratic state. By 1967 the regime ruled without for rent seeking; civil service reform to streamline the holding elections. These beginnings set the stage for public work force, improve remuneration, provide train- Idi Amin's rise to power in 1971. Under Amin, gov- ing, and instill a code of ethics; revival of the Public ernment became little more than a system of organized Accounts Committee of Parliament; a strengthened crime used to extract rents from the public. Their auditor general's office; and a public relations campaign depredations took many forms, including support for against and prosecution of corruption, under the author- economically irrational projects, exorbitant military ity of an inspector general with powers to investigate and expenditures, kickbacks on state contracts, extortionate prosecute. Much remains to be done before corruption import controls, and expropriation of the properties of can be said to be under control. The inspector general, Asians. Upon emerging from civil war in 1986, the however, has announced prosecutions against common new Ugandan government under President Yoweri examples of rent seeking (such as customs and procure- Museveni inherited a weak, underpaid, and overstaffed ment fraud), which should have a deterrent effect. as well. If price controls are lifted, market prices will can help blunt the incentive to be corrupt. Police offi- reflect scarcity values, not the payment of bribes. cers, for example, must have discretionary authority to Needless to say, reducing official discretion does not make law enforcement decisions on the spot, but pub- mean eliminating regulatory and spending programs with lic complaints will often restrain any abuses. strong justifications. Such programs must be reformed, Introduce market-based schemes that limit the discretion of not eliminated. Abolishing taxes is not a sensible way to regulators. This approach also has the virtue of pro- root out corruption among tax collectors; a corrupt police ducing an economically efficient allocation of re- force cannot simply be closed down. Several measures sources. The sale of water and grazing rights, pollution have proved effective in reducing official discretion in on- rights, and import and export licenses can improve going programs: the efficiency of government operations while limiting corruption. Clar and streamline laws in ways that reduce official Adopt administrative reforms that introduce competitive discretion. Mexico's customs reforms cut the number of pressures into government. Open, competitive bidding steps in the process from twelve to four; the remaining for public procurement contracts can reduce opportu- steps were streamlined to reduce delays. nities for corrupt deals. Creating overlapping, compet- Contract for services with a private company, possibly a itive bureaucratic jurisdictions can greatly diminish the foreign firm with no close ties to the country. When bargaining power of individual officials. If clients can Indonesia contracted with a Swiss firm for customs turn to a second official when the first demands a bribe, preinspection and valuation and for help in collecting no single official has the power to extract a large payoff import duties, corruption declined. Contracting out so long as applicants are eligible for the service. And monitoring functions is pointless, however, unless the if it is the applicants who are seeking something ille- government makes use of the reports it receivesand gal, overlapping enforcement areas can help to check that does not always happen. payoffs as well. For instance, when the state wants to Make rules more transparent. Simpler, nondiscretionary control illegal businesses, police officers can be given tax, spending, and regulatory laws can limit opportuni- overlapping enforcement areas to reduce opportunities ties for corruption. Sometimes a certain risk of corrup- for corruption. tion is tolerated because the benefits of a discretionary approach to program administration exceed the costs of STRENGTHENING MECHANISMS FOR MONITORING AND corruption. But even then transparency and publicity PUNISHMENT. Independent watchdog institutions that are RSTRAINJINJ6 ARBITRARY ST,A,TE ACTIDNI .AMD CDRRUPTIDNI 107 part of the government structure can also curb corruption. units to root out corrupt contractors and propose ways Countries have experimented with various approaches: to reorganize the agency to reduce corruption. Whistleblower statutes protect and reward public Some countries have independent anticorruption com- employees who report the malfeasance of co-workers or missions or inspectors genera! that can investigate alle- government contractors, The United States, for exam- gations and bring cases to trial. The most famous is the ple, has a statute that calls for rewarding workers who Independent Commission against Corruption in Hong report irregularities in government contracts. Such an Kong (China), which reports exclusively to the highest incentive for reporting is often necessary, since people authority and has extensive powers (Box 6.4). Singa- who report co-workers' misdeeds are frequently ostra- pore and Botswana have similar institutions. cized. Such measures are hollow, however, unless pros- Ombudsmen hear citizen complaints and can help ecutors follow up, courts are incorruptible and effi- increase the accountability of government agencies. ciently run, and penalties are severe enough to deter Under the Ombudsman Act of 1991, South Africa has potential offenders. established a public protector to investigate alleged improprieties (malfeasance, corruption, human rights Watchdog organizations should focus not only on abuses) by public officials and to prepare reports, which those who receive bribes, but also on those who pay them. are usually made public. The office cannot initiate legal It takes two to tango, and penalties should be equally actions but will refer cases to offices that can. severe on both sidesusually a multiple of the bribes Some public agencies, such as the School Construction received or pald. Penalties for bribe payers should also Authority in New York City, have established internal include the prospect of being barred from contracting E ox 6.4 Hong Kong's independent commission against corruption Corruption was endemic in Hong Kong (China) dur- at first backed down and granted an amnesty for ing the 1960s. Its entrenched character is suggested by offenses committed before January 1, 1977. This set- expressions popular at the time: people had the choice back was harmful to the commission's prestige, but it of "getting on the bus" (actively participating in cor- was able to recover with a vigorous public education ruption) or "running alongside the bus" (being a campaign. Public surveys carried out between 1977 and bystander who did not interfere with the system). 1994 indicate that public perceptions of corruption "Standing in front of the bus" (reporting or resisting have fallen significantly. Indirect evidence suggests that corruption) was not a viable option. active corruption has declined as well. Spurred to action by a scandal involving a high- Still, the ICAC is not without its problems. The anking police officer, the then-governor general estab- main one is that it reports only to the governor. An lished the Independent Commission Against Corrup- anticorruption commission reporting to an autocratic tion (ICAC) in 1974. The commission reports only to ruler could be used as an instrument of repression the governor and is independent of the police force. against political opponents, and the ICAC has not ICAC officials are paid more than other government been immune to such charges. The ICAC's broad workers and cannot be transferred to other depart- powers could be abused in systems less committed to ments. No one may leave the ICAC to work for senior the rule of law. A series of oversight committees and officers who have been the subject of an investigation. an independent judiciary act as a check on the ICAC, The ICAC has the power to investigate and prosecute but even so the occasional scandal surfaces. As a fur- corruption cases as well as to sponsor public education ther control on its power, such an agency might re- campaigns. The government's commitment to reform port not to the chief executive but to the legisla- was further indicated by the appointment of a person ture, as do Uganda's Inspector General and the U.S. of unquestioned integrity as the first head of the com- General Accounting Office. A tough, independent mission and by a policy of investigating and prosecut- anticorruption agency is a potent tool and represents ing "big tigers" from the outset, a credible long-term commitment, but there should Early efforts to clean up corrupt syndicates within also be checks on its ability to be misused for politi- he police force, however, met with protests. The ICAC cal ends. 108 WORLD DEVELOPMENT REPORT 1997 with the government for a period of years. Industrial their constituents. (However, if corrupt payoffs are used countries with strong monitoring capacity can enforce to buy benefits for individual voters, knowledge of cor- such measures on their multinational companies conduct- ruption may do little to stop it.) ing business overseas. But except for the United States, If courts are independent and citizens can sue to force which adopted the Foreign Corrupt Practices Act in the government to comply with the law, this opens 1977, countries have been reluctant to act unilaterally for another route to control government malfeasance. fear of subjecting their businesses to more stringent stan- Public exposure of corruption through the media is dards than their foreign competitors. another option. Even undemocratic rulers are likely to be In this context, international organizations provide a sensitive to public opinion, if only because they wish to forum for agreeing on common standards and coordinat- avoid being overthrown. A free press can be a vital check ing action. Regional organizations such as the Organization on abuses of power, especially in countries that lack other of American States have sponsored international conven- means of constraining politicians and bureaucrats. tions making bribery, including international bribery, a crime. A recent initiative by the OECD encourages ending Yet even if both the necessary information and the the tax deductibility of bribes and criminalizing the bribing means of punishing corrupt practices are available, indi- of foreign officials. It makes recommendations to its mem- vidual citizens are unlikely to act alone. Laws that make it ber countries on how to deal with bribery in international easy to establish associations and nonprofits can help business transactions. International organizations are also resolve this collective action problem. Such groups might working to coordinate the Light against money laundering not only seek information from government, but also sup- and, in particular, to expand the list of offenses, including ply information to government about citizens' opinions of corruption. The forty recommendations of the Financial the quality of public services. As discussed in Chapter 7, Action Task Force on Money Laundering include nondrug the nonprofit Public Affairs Centre in Bangalore is en- criminal activities. This opens the way for countries to gaged in a promising experiment to publicize the per- make illegal the use, deposit, or transfer of money acquired formance of Indian public agencies. An international through corruption. nonprofit organization, Transparency International, is Citizens' groups can also be an important check on the working to mobilize citizens around the world to fight arbitrary abuse of government powerif people can orga- corruption and to publicize countries' track records. Yet nize, and if they can find out what is happening. Govern- precisely because open information can be so potent in ments should publish budgets, revenue collection data, promoting government reform, many countries limit such statutes and rules, and the proceedings of legislative bod- groups or make it costly for them to organize. ies. Financial data should be audited by an independent authority like the U.S. General Accounting Office. Unau- Strategic options: Balancing flexibility with restraints dited secret funds or extrabudgetary funds available to Pressures for reform are on the rise everywhere. Private en- chief executives are an invitation to corruption. trepreneurs and firms want the credibility of state actions Freedom-of-information acts in the United States and anchored by a well-functioning system of property rights. a number of European countries are an important tool for Citizens are demanding more responsive and effective public oversight. A recent directive of the European delivery of public services and greater probity in the use of Union requires member states to pass freedom-of-infor- public resources. At the same time, globalization is in- mation laws covering environmental information. Such creasing demands for a more agile state, one that can laws enable citizens to obtain government information respond quickly to changing circumstances. These pres- without having to show how their lives are affected by it. sures have magnified the state's dilemma: how to check The availability of information helps citizens discipline arbitrary decisionmaking without building rigidities that public officials at the ballot box and through other inhibit innovation and change. The fundamental chal- avenues of protest, such as legal challenges and direct peti- lenge is to devise institutional arrangements that sustain a tions to decisionmakers. workable balance between flexibility and restraint. Coun- Information is of little value, however, without mech- tries with strong institutions or track records of following anisms for using the knowledge gained to influence gov- through on commitments may have room to respond flex- ernment behavior: ibly (even at the cost of some corruption), but countries with dysfunctional and arbitrary governments may not. In democracies, citizens can vote officials out of office if States in many developing countries have demon- they believe them to be corrupt. This gives politicians an strated a clear imbalance between flexibility and restraint. incentive to stay honest and work for the interests of They have generally not been credible, accountable, RESTRAINING ARBITRARY STATE ACTION AND CORRUPTION 109 sponsive, or agile. In several countries the capricious exer- the civil service (for instance, by raising pay and restrain- cise of state power coupled with rampant and unpre- ing political patronage in recruitment and promotion), dictable corruption has undermined development. States reducing opportunities for officials to act corruptly (for with too much flexibility and not enough restraint will instance, by increasing competition and reducing officials' find that their actions are not viewed as credible, and discretionary authority), and enhancing accountability are investment and growth will suffer. These countries need other essential steps. Strengthening mechanisms for mon- to strengthen the formal instruments of restraintjudi- itoring and punishmentof the people who pay bribes cial independence, effective separation of powersto as well as those who accept themwill require vigorous enhance the credibility and accountability of the state. enforcement of criminal law. But it will also require over- International commitment mechanisms can serve as a sight by formal institutions such as statutory boards and short-term substitute while these institutions are built up. by ordinary citizens (through voice and participation). Yet these actions will not be enough to stop the rot in These efforts can help not only in controlling corruption countries where endemic and entrenched corruption has but also in improving many other functions of the state, undermined key functions of the state. Strengthening such as policymaking and service delivery. The use of formal instruments of restraint is only one element of a voice and participation to reinvigorate public institutions multipronged strategy to control corruption. Reforming is the subject of Chapter 7. BRINGING THE STATE CLOSER TO PEOPLE And, tell me, what use is the ship-of-state fall are not on board? From Tijan M. Sallah, The State (1996) ARE THE MEANS AND THE END OF DEVELOP- ment, business, and civic organizations, can also enhance PEOPLE ment. But they have different amounts of power their supply. But effective citizen involvement does nor and resources, and different interests, all of which the state come easily. One lesson of many such experiments is that must try to represent and respond to if it is to act effec- effective participation requires enlightened government tively. In nearly all societies the needs and preferences of intervention, including improving the institutional envi- the wealthy and powerful are well reflected in official pol- ronment in which social and human capital is created. icy goals and priorities. But this is rarely true of the poor There is another important lesson: bringing govern- and the marginalized, who struggle to get their voices ment closer to some people can risk taking it even fur- heard in the corridors of power. As a result, these and other ther away from others. Not all organizations of civil society less vocal groups tend to be ill served by public policies and are adequately accountable, either to their own members or services, even those that should benefit them most. to the public at large. And although some groups may be A state that ignores the needs of large sections of the quite vocal, the interests they represent may not be widely population in setting and implementing policy is not a shared. In reaching out to groups in civil society, govern- capable state. And even with the best will in the world, ment must be conscious of the interests those groups rep- government is unlikely to meet collective needs efficiently resent, but also of those they do not. Otherwise its inter- if it does not know what many of those needs are. Rein- vention risks creating new disparities between the newly vigorating public institutions must, then, begin by bring- enfranchised and those whose voices remain unheard: ing government closer to the people. That means bringing women and ethnic groups, for example, or people whose popular voice into policymaking: opening up ways for interests have not been adopted by an active NGO. individual users, private sector organizations, and other Some of the same concerns arise with regard to decen- groups in civil society to have their say. In the right set- tralization. Carefully managed, decentralization can do ting, it can also mean greater decentralization of govern- much to improve state capability, creating pressures for ment power and resources. better marching of government services to local prefer- This chapter discusses a wide range of mechanisms to ences, strengthening local accountability, and supporting ensure that policies and programs better reflect the full local economic development. But there are pitfalls: policy- panoply of society's interests. Informing and responding makers in central government may lose control over the to citizens will raise the state's effectiveness by improv- macroeconomy as a result of uncoordinated local deci- ing monitoring of public goods and services and forcing sions, and regional disparities can widen, exacerbating greater transparency in decisionmaking. Encouraging economic and social tensions. Local governments can fall wider participation in the design and delivery of these under the sway of particular interests, leading to abuse goods and services, through partnerships among govern- of state power and even less responsive and accountable SRINGINC THE STATE CLOSER TO PEOPLE 1 1 1 government. The message, here as elsewhere, is that bring- formance of some state institutions is falling short of ex- ing government closer to the people will only be effective pectations (Box 7.1). if it is part of a larger strategy for improving the institu- tional capability of the state. Electoralparticipation In a democratic society elections are the primary manifes- Greater accountability and responsiveness tation of citizen voice. The number of democratically through participation elected governments has increased sharply in recent Throughout history nearly all societies have grappled with decades, giving many citizens new opportunities to voice how to make the state reflect the needs and interests of the their opinions through the ballot. In 1974 only 39 coun- population. But it is an especially relevant question today, triesone in every four worldwidewere democratic. with the spread of education and information and the Today 117 countriesnearly two in threeuse open growing pluralism of nations creating new pressures on elections to choose their national leadership, and two- states to listen and respond to the voices of their citizens. thirds of the adult population in developing countries are A sample of public opinion around the world suggests eligible to participate in national elections (Figure 7.1). that belief in government remains solid, but that the per- The trend is especially striking in Central and Eastern Box 7.1 Public opinion and the state Views on the state vary widely, reflecting perceptions of In Latin America 69 percent of citizens believe that wide-ranging political and economic variables. To assess the state should intervene to reduce income differences these perceptions and concerns, public opinion surveys between rich and poor. Latin American respondents have been carried out in various countries and regions. also regard as critical the government's role in main- For example, a 199 1-92 survey found that 49 percent taining law and order, but 65 percent have little or no of respondents in the United Kingdom, and 44 percent confidence in the judiciary or the police force. Only 24 in the United States, felt excluded from public decision- percent have some confidence in these institutions. making that directly affected their lives. Even so, satis- These results closely mirror the concerns voiced by faction with the way their democracies are developing is the region's entrepreneurs in the survey reported in relatively high in Western Europe and North America, Chapter 3. ranging from 54 to 64 percent of respondents. By con- In India a 1996 survey found that, despite strong trast, in Latin America and the transition economies of support for the democratic system, voter trust in their Eastern Europe only 30 to 40 percent of respondents representatives has fallen since 1971. Trust in some report being satisfied with the way their democracies are public institutions was also iow, particularly the police working. In Latin America this negative perception may (28 percent) and the bureaucracy (37 percent). Still, be related to the fact that 52 percent of respondents in a the poll found strong popular participation and twelve-nation survey felt that the administration of elec- involvement in politics: the number of respondents tions in their country was fraudulent. who said they participated in a social organization or In Europe public support for the state and its services political parry doubled between 1971 and 1996. has been consistently strong since the 1970s. In 1990 In contrast to Europe, Latin America, and to some large majoritiesmore than 70 percentin seven West- extent Asia, where statist traditions remain strong, in ern European nations still believed that the government the United States 80 percent of respondents in a recent should provide health care, services to the elderly, aid to poll said they mistrusted the government because of its the unemployed, and assistance to industry, as well as perceived inefficiency and waste. In addition, public reduce income differences. A 1996 poll across the fifteen support for welfare programs, particularly targeted wel- countries of the European Union found that 51 percent fare spending, has declined. A 1993 poli found that of citizens believe their governments should maintain only one-sixth of Americans thought that the welfare current social benefits and protection levels, compared system was working very well or fairly well, and in with 12 percent who think they should be cut consider- 1995 two-thirds thought that too much was being ably to enhance EU competitiveness in world markets. spent on welfare programs. 1 12 WORLD DEVELOPMENT REPORT 1 97 Figure 7.1 The world has become much more democratic since 1980 Index of democracy (most democratic = 10) 10 OECD 8 Eastern Europe 6 and Central Asia 4 South and East Asia 2 and Pacific Latin America 0 4 6 Sub-Saharan Africa Middle East and North Africa 8 1960 1965 1970 1975 1980 1985 1990 1994 Note: This particular index of democracy is calculated for 177 countries from scores on five component indicators: competitiveness of political participation, regulation of political participation, competitiveness of executive recruitment, openness of executive recruitment, and constraints on the chief executive. The World Bank does not endorse any particular democracy index. See the Technical Note for details. Source: Jaggers and Gurr 1996. Europe and Central Asia, where the fall of communist democracies in the 1 980s shows that women's parliamen- regimes in 1989 and 1991 sparked a series of major polit- tary representation and voter turnout are higher in systems ical changes across the region. These events had repercus- based on proportional representation than in winner-take- sions in other regions as well, particularly in Sub-Saharan all systems. Where women are better represented in par- Africa. In Latin America the gradual movement toward liament, their interests are better protected, through poli- democracy started somewhat earlier. Now all but two cies relating to maternity leave, child care, and flexibility countries in the region have democratically elected gov- in retirement systems. This suggests caution in making ernments, and close to 13,000 units of local government simple generalizations about the representativeness of for- are electing local leaders (such as mayors), compared with mal electoral arrangements: it is important to analyze how fewer than 3,000 in the late 1970s. they play out in practice. The principle of one person, one vote is fundamental to the representative purpose of elections. Without the Diversity and representation necessary safeguards, however, political interference and Concern about the effects of majoritarian rule has some- electoral fraud can seriously affect representativeness and, times led to changes in electoral arrangements to ensure thus, the legitimacy of electoral outcomes. As Box 7.1 adequate representation of minority groups. As the num- noted, for instance, a majority of poll respondents in ber of ethnic conflicts around the world attests, the belief twelve Latin American countries felt that the administra- of certain groups that they are being left behindin terms tion of elections was fraudulent in their country; only 30 of income, assets, or employmentcan be a powerful to 40 percent felt satisfied with the way democracy was source of frustration, This can spill over into direct con- working. Constitutional and institutional arrangements, flict if the lagging groups lack adequate means of repre- compounded by inadequate information, also affect the senting their concerns. Political elites who mirror ethnic ability of minority interests to gain effective representa- differences in competing for power and distributing tion. For example, a study of a number of European patronage will add fuel to the fire. In the contemporary BRINGING THE STATE CLOSER TO PEOPLE 113 world, states play a significant part in shaping ethnic rela- tions through two related channels: Box 7.2 Managing multiethnic societies in Malaysia and Mauritius The expansion of political authority enables states to create a competitive arena for the distribution of state Ethnic divisions in Malaysia have their roots in the resources and access to education, employment, land, colonial period, when large numbers of ethnic Chi- and credit. nese and Indians immigrated to take jobs and com- The extension of political patronage, whether for ad- mercial opportunities that were not being filled by ministrative convenience or for enhanced control, lets the largely agrarian native Malays (bumzputra). The governments favor certain ethnic groups over others. 1957 constitution enshrined consociational princi- ples, achieved through extensive negotiation among Ethnic differences and conflicts are a part of most soci- the major groups. In 1971, however, the new parlia- eties, yet some societies have managed them better than ment passed a constitutional amendment that firmly others. Techniques for alleviating such pressures have established Malay primacy. The amendment made included dispersing the "points of power" (decentraliza- it illegal to "question publicly or even in Parliament tion or devolution), supporting electoral arrangements the status of Malay language, the sovereignty of the that encourage cooperation among ethnic groups (elec- Malay rulers, the special position of Malays or the toral incentives and coalitions), and giving preferences to citizenship rights of the immigrant communities." less privileged ethnic groups in public sector employment This ruling changed the character of electoral cam- and other areas. Box 7.2 describes how constitutional paigns, since parties could no longer gain votes by change and preferential policies were used in Malaysia and arousing ethnic antagonisms. Mauritius. By looking at some of the basic rules of the A second initiative, the New Economic Program game, including elements of the political or institutional (1970-90), boosted the economic position of structure, both countries seem to have found a way to Malays. The program had two main elements. The manage ethnic differences effectively. first was the promotion of full, productive employ- ment of Malays and an expansion of the supply Alternative strategies for voice and participation of skilled Malay labor. Preferential university Voting rules and electoral incentives can be reformed to admissions standards for Malays almost tripled make them more representative, but the simple fact that their enrollment, to three-quarters of the total. The elections and referendums are held relatively infrequently second was the gradual redistribution of asset own- (Switzerland is an exception, with an average of five local ership. The government made it clear that it would referendums each year since 1945) limits their scope for not confiscate Chinese economic wealth, but that it communicating timely information about societal prefer- would promote Malay participation in a growing ences. In most societies, democratic or not, citizens seek economy. Strong growth allowed non-Malays to representation of their interests beyond the ballot: as tax- continue to gain, while the New Economic Pro- payers, as users of public services, and increasingly as gram ensured that the growth would be shared by clients or members of NGOs and voluntary associations. all citizens. Against a backdrop of competing social demands, rising Mauritius has at least three major ethnic groups: expectations, and variable government performance, these Indo-Mauritians, Creoles, and Chinese. The de- expressions of voice and participation are on the rise. signers of the electoral system, anxious to avoid cre- The rapid growth of NGOs illustrates the trend most ating institutions that might exacerbate the coun- vividly. Since the late 1980s the number of NGOs oper- try's ethnic divisions, structured the system to force ating in parts of Africa and Asia has almost doubled. In the main parties to seek support from all communi- Central and Eastern Europe and the CIS the number of ties. Moreover, Mauritian governments have gener- NGOs may have increased three- or fourfold from its very ally chosen broad-based growth and distribution low base of 1989. In some countries of the OECD, oper- policies over ethnic preferences. Formal preferences ating expenditures in the voluntary or NGO sector now in employment and in education have never been account for almost 4 percent of GDP. used. And all governments since independence have NGOs have myriad organizational forms and func- had to form multiethnic coalitions in order to tions, from labor unions to professional associations to assume and maintain power. Growth with redistri- neighborhood groups to philanthropic trusts. Among the bution has tended to blunt the appeal of communal most active NGOs today are those providing services politics. directly to individuals and communities, from health and 114 WORLD DEVELOPMENT REPORT 1997 education to microcredit, vocational training, and profes- have had a direct impact on this isolated community. The sional services. In the OECD countries many NGOs committee has an elected executive board consisting of operate alongside public providers. In Japan and the members from each part of the village; villagers rank it as United Kingdom, for example, a large percentage of non- the most relevant and effective organization in the village. profit organizations are active in education. In the United Similar locally based organizations include rotating credit States they figure prominently in health care. But unlike associations, farmers' associations, worker cooperatives, public providers, most NGOs are not obliged to cater to parent-teacher associations, and even religious congrega- the general needs of the population; this makes it simpler tions. These associations are valuable not only for their for them to provide services of a particular type and qual- ability to meet basic needs, but also for the role they play ity to specific groups. in building trust and a sense of public connectedness In most developing countries NGOs engaged in ser- among those excluded or alienated from the formal politi- vice delivery are small in scale, working in communities cal process. But organizations such as these also face limi- and settings where the reach of government or private tations, including narrowness of membership and repre- providers is weak or nonexistent. In the West Bank and sentation, limited management capacity, and the risk of Gaza, for example, an estimated 1,200 NGOs provide 60 co-option by traditional power holders or political factions. percent of primary health care services, up to half of sec- ondary and tertiary health care, and most agricultural ser- The institutional basis ofparticiation vices, low-cost housing schemes, and microcredit. In The depth and intensity of popular collective activity Cambodia some thirty to forty NGOs provide micro- obviously differ by social and institutional setting. One credit to rural and urban entrepreneurs in the absence of explanation for these differences lies in differing endow- alternative government programs for poverty alleviation. ments of social capital, the informal rules, norms, and The numerical importance of these NGOs reflects their long-term relationships that facilitate coordinated action ability to substitute for weak public sector capacity and to and enable people to undertake cooperative ventures for mobilize funds from a range of different sources, includ- mutual advantage. The presence of rules conducive to ing national and international organizations. social organization can improve the collective efficiency Yet not all NGOs are involved in the delivery of ser- of a community, but the absence of such rules is not nec- vices. Many others are research and civic education essarily a permanent condition. They can be generated groups, advocacy organizations, and professional and busi- by participation itself, and here governments and other ness associations that represent particular interests or seek formal organizations such as labor unions can play a pos- to educate the public about issues in their collective inter- itive role. Government efforts to improve the manage- est. The Socio-Ecological Union in Belarus, for example, ment of irtigation systems in Taiwan (China) and the is actively engaged in public education about industrial Philippines, for example, have yielded numerous collec- pollution and its consequences. The West Africa Enter- tive responses from farmers in managing operations and prise Network is a business-oriented network, with some maintenance and in collecting water user fees. By accept- 300 members from twelve nations, representing the inter- ing co-management of water resources by farmers, public ests of domestic entrepreneurs in dealings with govern- officials achieved a substantial improvement in irrigation ment. In many countries trade unions play an important management. In the state of Ceará in northeastern Brazil role in generating and disseminating information on labor (see Box 5.7 in Chapter 5), community monitoring of an and policy-related issues. The growth of these intermedi- innovative government health program provided the basis ary organizations reflects the larger movement toward not only for a highly successful program but also for more democracy in many regions and, in some countries, the effective cooperation among community members on need to bridge the "missing middle" between citizens and other mutually beneficial courses of action. the state. Unlike the electoral process, however, where all The debate about the contribution of social capital to votes are counted equally, not all these organizations are economic and social development is just beginning, and equally representative, either of their clients' interests or of the early evidence is by no means unambiguous. But some the public interest more broadly. studies are already demonstrating its potential impact on Most intermediary NGOs tend to be one step removed local economic development, on the provision of local from ordinary citizens; by contrast, grassroots organiza- public goods, and on the performance of public agencies tions, community-based groups, and people's organizations (Box 7.3). engage them directly. In Umu-Itodo, a village in Enugu The social mechanisms that constitute social capital and State, Nigeria, for instance, the Community Development the myriad forms of more formal nongovernmental activity Committee formed in 1986 has been responsible for directly reflect the heterogeneity of needs and preferences in numerous development and infrastructure projects that society. At the same time there are no guarantees that these BRINGING THE STATE CLOSER TO PEOPLE 1 15 Box 7.3 Does social capital matter? A study of regional government in Italy during the justed incomes per capita than do households in vil- 1970s and 1980s found that although political and lages with low levels of social capital. When other non- economic failures were widespread, some regional gov- social capital determinants are controlled for, there also ernments, particularly those in the north, had per- appears to be a strong correlation between a village's formed well. Northern regional governments were well-being and its level of social capital. This result notable for developing innovative day care programs points to important spillover effects at the village level and job training centers, promoting investment and arising from individual participation in local associa- economic development, managing the public's busi- tions and groups. Although no general conclusions ness efficiently, and satisfying their constituents. could be drawn about the impact of social capital on Regional governments in the south, by contrast, government performance, the study points to a num- showed much weaker responsiveness and performance. ber of important linkages, including a positive associa- The study attributed the better performance of north- tion between social capital and the quality of local ern governments to the external pressures created by schooling (see figure). The implication is that where dense networks of civic associations and citizen parents can organize to monitor and pressure local gov- involvement in local public affairs. ernment into maintaining local schools, school quality A recent study of villages in rural Tanzania found is enhanced. Even though the direct benefits of school- that households in villages with high levels of social ing accrue mainly to individuals, the benefits of moni- capital (defined in terms of the degree of participation toring local government performance accrue to all as a in village-level social organizations) have higher ad- public good. Social capital, household expenditure, and school quality in Tanzania Household expenditure per capita (thousands of Tanzanian shillings) Index of school quality 250 4.0 200 3.5 . S . 150 3.0 - S S I. S S . 100 . S 50 2.5 . S . S S S 0 2.0 -2 -1 0 1 2 3 -2 -1 0 1 2 3 Index of social capital Index of social capital Note: Data for the social capital index are from a 1995 survey of 1,376 Tanzanian households. The social capital index measures the prevalence of residents' participation in groups and voluntary associations. The school quality index measures households' perceptions of school quality. See the Technical Note for details. Source: Narayan and Pritchett 1997. 116 WORLD DEVELOPMENT REPORT 1997 organizations are adequately addressing citizens' needs or Improving institutional capability that they are genuinely concerned with promoting the pub- Increasing opportunities for voice and participation can lic interest. Most NGOs provide services of high quality, improve state capability in three ways. First, when citizens but some suffer from serious problems, including poor can express their opinions, formally or informally, and quality of service, inadequate community participation, press their demands publicly within the framework of the and weak accountability. Some NGOs are created oppor- law, states acquire some of the credibility they need to tunistically, to advance the interests of narrow and privi- govern we!!. Broad-based discussion of policy goals can leged constituencies, often at the expense of the less vocal also reduce the risk that a powerful minority will monop- and less powerful. And the same social rules and norms that olize the direction of government. States that achieve facilitate collective action among citizens can preserve credibility are then allowed more flexibility in policy im- inequalities and power differences within communities. plementation and have an easier time engaging citizens in Yet given the many obstacles facing ordinary citizens, the pursuit of collective goals. This does not mean that especially the poor, in articulating and pressing their Western-style democracy is the only solution. Experience needs, these associations play a vital role in channeling from parts of East Asia suggests that where there is wide- their voice and in building capacity for participation in spread trust in public institutions, effective ground-level public affairs. And those organizations that genuinely seek deliberation, and respect for the rule of law, the condi- to work in the public interest can be valuable partners in tions for responsive state intervention can be met. economic and social development. Working from this Second, where markets are absent, as is the case for premise, many governments are searching for new institu- most public goods, popular voice can reduce information tional arrangements for providing public goods, which problems and lower transactions costs. Where incentive involve both the private sector and groups in civil society problems and weak state capability lead to inefficient pub- (Figure 7.2). The next section sets these efforts in a lic services, user groups and citizen associations can inform broader context, exploring the range of mechanisms for public officials of the problem and pressure them for im- increasing popular participation in the design and imple- provements. A recent user assessment of the water supply mentation of public policy. system in Baku, Azerbaijan, for example, revealed not only significant problems of water leakage and water-related health problems, but the high costs that the city's unreli- able water supply imposed on low-income consumers. Per- Figure 7.2 Organizations at the interface of haps most interesting, users also revealed themselves will- state, markets, and civil society ing to pay between two and five times more than they were actually paying for a reliable and safe water supply. State (hierarchy and Third, no matter how dedicated, hardworking, or pub- control) lic spirited state officials are, they cannot anticipate all the public goods and services that citizens desire. The emer- gence of private and NGO alternatives to public provision can help meet gaps in the supply of public goods, as well as provide those goods and services that individuals are willing to pay for out of their own pocket. NGOs can be both partners and competitors in the delivery of public services. And when backed by citizen voice, they can exert useful pressure on government to improve the delivery and quality of public services. There is no blueprint for finding the appropriate bal- ance of voice, participation, and bureaucratic control in the provision of public goods. The solution depends on Private sector Civil society the capabilities of the public agencies in question and (profit and (voice and competition) collective action) other providers and on the characteristics of the public good or service being provided. As discussed below, effi- ciency and equity dictate some degree of centralized gov- ernment control and coordination when it comes to goods and services that have jurisdictional spillovers, are subject to economies of scale, or raise distributional concerns. In the technical and often sensitive area of economic man- BRINGING THE STATE CLOSER TO PEOPLE 1 17 agement, for example, some insulation of decisionmaking legislation can also strengthen public accountability and from the pressure of political lobbies is desirable. In set- responsiveness by requiring agencies to announce and ting standards, such as in the provision of basic education, enforce service norms and standards, provide public infor- there is also an important role for centralized decision- mation, and respond to consumer complaints. making. But the process by which broad policy directions Specific techniques and mechanisms for consulting and standards are set should not be insulated from public users and intended beneficiaries can also help improve the discussion. And in the management of common-property quality of specific public services. As the cases in Box 7.4 resources, the production of basic infrastructure, and the illustrate, feedback mechanisms such as client or user sur- delivery of essential services, there is considerable scope veys can provide valuable information about an agency's for involving people directly, both in the formulation of performance and the type and quality of services reaching policy and in its implementation. consumers. Simplified surveys such as the report card used in Bangalore, India, can be particularly useful when insti- Participatory mechanisms tutional capability is limited. MECHANISMS FOR INFORMING AND CONSULTING. It is Besides increasing the flow of information to public generally accepted that some areas of public decisionmak- officials, techniques for citizen and client consultation can ing require insulation from political pressure. In others, introduce more openness and transparency into the sys- however, public and private interests coincide to such an tem. As more people become aware of the performance of extentefforts to raise agricultural production, for exam- specific agencies or officials, they are more likely to exert ple, or reforming the health systemthat some level of collective pressure on the agency to perform better. At the public-private deliberation is not just desirable but in same time public agencies will have less opportunity for fact critical to success. In East Asia, by institutionalizing arbitrary action. public-private deliberation councils comprising represen- But no mechanism for consultation automatically tatives of labor unions, industry, and government, policy- reaches all the appropriate individuals and groups. There makers were able to get broad agreement on economic are costs to acquiring and providing information, and the policy issues and the necessary commitment to intervene low income or subordinate position of some groups in quickly and flexibly. Other nations with very different in- society makes them nearly invisible to public officials. In stitutional settings, such as Botswana, Chile, Mexico, consulting with users or clients, every effort must be made Senegal, Uganda, and the United States, have also sought to identify all relevant social groups and ensure that they to implement deliberative mechanisms on issues ranging are represented. from economic policy to institutional reform. MECHANISMS FOR DESIGN AND IMPLEMENTATION. By embedding the voice of powerful interest groups in Improving government performance does not end with mutually acceptable rules, public-private deliberation improved consultation. There is also compelling evidence councils can reduce transactions costs by reducing the that arrangements that promote participation by stake- scope for opportunistic behavior. But these councils do holders in the design and implementation of public ser- not always succeed. Systematic evaluations of performance vices or programs can improve both the rate of return and are not available, but a number of conditions and char- the sustainability of these activities. acteristics appear to affect success. These include broad The education sector has proved particularly fertile enough representation and public education for the ground for this kind of experimentation. Weak monitor- process to have widespread support, technical support and ing and supervision of local schools is a perennial problem assistance to the councils, and an emphasis on building for governments. But several are finding that these prob- trust and mutual monitoring among both public and pri- lems can be addressed by increasing the involvement of vate sector participants. parents and communities in school management. In New Deliberative mechanisms are unlikely to enable gov- Zealand, elected boards of trustees that manage schools are ernments to be effective in the long run if their policies composed of parents of children at the school. Legislation appear illegitimate or unresponsive to crucial societal de- enacted in Sri Lanka in 1993 established school develop- mands. Efforts to reach out to citizens must reach all the ment boards to promote community participation in way down. At the very least, reforms and programs must school management. Many countries have also found that be made intelligible to the public, for example by encour- communities that participate in school management are aging widespread media coverage of budget debates. In more willing to assist in school financing. Singapore the Division of Public Feedback systematically Yet effective citizen involvement in school manage- gathers citizens' comments on a variety of national poli- ment does not come easy, nor is it a panacea. New cies and invites interest groups to public hearings with Zealand realized after it had embarked on its reform ministers and senior officials. As discussed in Chapter 5, that newly elected trustees required intensive training. 118 WORLD DEVELOPMENT REPORT 197 Box 7.4 Client surveys to motivate service improvements in India, Uganda, and Nicaragua In several countries client surveys have helped motivate including Mumbai (Bombay), have started using the better public sector performance. By tapping the expe- report card approach. rience of citizens and having them monitor and evalu- Working with NGOs and communities, Uganda's ate services, surveys have helped identify problems and government is also surveying views on service delivery. design and implement innovative solutions. The first survey found that just 11 percent of rural In Bangalore, India, "report cards" ask citizens and households had ever been visited by an agricultural businesses to rate the public agencies they use to solve extension worker. Several districts have incorporated problems or get services. The report cards, adminis- the survey findings into their district plans. One dis- tered by the Public Affairs Centre Bangalore, an trict has instituted further training for extension work- NGO, assess the quality and cost of citizens' interac- ers and is lobbying the central government for permis- tions with public agencies. In the first round of report sion to spend more of its budget on extension workers. cards the Bangalore Development Authority, respon- Nicaragua's surveys, like Uganda's, were initiated sible for housing and other services, scored the lowest by the government. The first survey, in 1995, found in several categories, including staff behavior, quality that 14 percent of bus riders had at some time been of service, and information provided. Only 1 percent assaulted on a bus. It also found that 90 percent of bus of respondents rated the authority's services as satis- drivers did not respect the official fare of 85 cordobas: factory. Rather than viewing the results as a threat, they did not return the 15 cordobas in change from a however, the authority's director took them as an op- 100-cordoba note to riders. Moreover, the survey portunity, launching a citizen-government initia- revealed that people were willing to pay higher bus tive to address delivery problems. Other agencies in fares. Based on these findings, the fare was raised to Bangalore have also taken action inspired by the re- one U.S. dollar. In a follow-up survey in 1996, 90 per- port cards. And groups in five other Indian cities, cent of users said that the official rate is respected. Botswana found it difficult to attract qualified people to protecting and conserving wildlife to local communities. lower secondary school boards of governors, especially in And across Africa, Asia, and Latin America high levels of rural areas. In Uganda, community training for parent- beneficiary participation in the design and management teacher associations and school management committees of rural water supply projects have been shown to be is being provided in two districts by an international highly correlated with project success (Box 7.5). NGO, to ensure that the quality of schooling and of However, the same study of rural water supply also school administration is enhanced. revealed that, among the highly participatory projects, Citizen participation may also be crucial in programs only half adequately involved women. The explanation for the management of natural and common-property was found in factors specific to women's participation, resources such as grazing lands, wildlife, forests, and water including time constraints and cultural barriers. Conse- sources. Exclusive bureaucratic control of such resources quently, innovative participatory mechanisms are required has proved inadequate in many different institutional set- that explicitly seek women's involvement in the design and tings, in some cases leading to confrontation between the implementation of projects that directly affect them. One users of these resources and the public officials seeking to such effort can be found in the Philippines, where a series manage them. Recognizing the importance of participa- of measures over more than two decades has gradually led tory natural resource management, forestry officials, to the integration of gender issues in the government's NGOs, and local communities in India are now under- agenda, partly through greater participation of women in taking a variety of initiatives. The National Forestry Pol- planning and implementing policies, and partly through icy embraces increased participation of local people in specific programs for women. The expected payoff to such managing forests. In India's joint forest management pro- measures lies not just in improvements in the process grams, forest departments and local user groups share of public policymaking but in the economic returns to decisionmaking authority and control over forest lands better-designed and better-targeted public investments. and products and revenue. The result has been reduced conflict and increased productivity of the land. Making particzation succeed will take hard work. . Practices are also changing elsewhere. In Zimbabwe These illustrations suggest that in the provision of certain the CAMPFIRE program seeks to return the benefits of local public goods or shared serviceswhere the people BRINI3ING THE STATE CLOSER TO PEOPLE 1 19 who pay are also the direct and principal beneficiaries using the institutional capacity closest to the client can Box 7.5 Does participation improve project improve the quality and effectiveness of public action. performance? Capable states are therefore likely to be those that strengthen and increase the efficiency of local organiza- Using data from 121 diverse rural water supply proj- tions and associations rather than replace them. But ects in forty-nine countries in Africa, Asia, and reaching out to citizens as co-managers or co-producers Latin America, a recent study tested the relationship does not necessarily reduce the role of the state, nor is it between participation and project performance. costless or quick to implement. To get users or clients to Participation was measured on a continuum ranging become partners, public agencies often must invest con- from simple information sharing through in-depth siderable time and energy in building ties with communi- consultation with beneficiaries to shared decision- ties, in building commitment among their own staff, and making to full control over decisionmaking. The in ensuring that minimum standards of quality and equity authors found a strong correlation between high are maintained. levels of beneficiary participation, especially in deci- In a pioneering case in Recife, Brazil, where the intro- sionmaking, and project success. Of the forty-nine duction of low-cost condominial sewers in low-income projects with low levels of participation, only 8 per- neighborhoods changed the relationship between the state cent were successful. But of the forty-two projects agency and the sewer users, it took two years for public with high levels of beneficiary participation, 64 per- officials, working intensively and in multidisciplinary cent were successes. teams with residents, to figure out how to make the con- Case studies support these conclusions. The first dominial system work. Even after the process was better phase of the Aguthi Rural Water Supply Project in understood, successful implementation took another four Kenya was conducted without community partici- to six months in each neighborhood. An evaluation of the pation. The project, which involved piped water project showed that, by fostering an active and vocal con- systems, was so plagued with problems that it came stituency, the scheme not only generated considerable sav- to a standstill and had to be redesigned. Working ings but also put in motion mechanisms for accountabil- with project staff, local leaders, organized as the ity that were critical for good agency performance. Aguthi Water Committee, mobilized community Greater responsiveness means changing not only the support for the project. Following public confer- way state agencies work with clients, but also the way ences with stakeholders, community members those agencies are organized and reward their workers. began to contribute labor and funds. Phase II of the Effective participation is more likely when opportunities project was completed on schedule and within bud- for internal participation exist within the public agency. get. The communities continue to pay monthly tar- In addition, the overall incentive environment must re- iffs for the new water service, and operation and ward higher-level staff for responsiveness to clients and maintenance are handled successfully in coopera- must provide adequate support to street- or field-level tion with the relevant government agency. workers in their efforts to work with clients. Without such measures, resistance to working with clients can be high, creating an atmosphere that is incompatible with a more ing Committee, which gives local representatives a voice participatory approach. in national policy. Working closely with people also often requires redefining tasks and responsibilities, reallocating staff And a supportive environment resources, and developing new mechanisms for learning Government can also support participation indirectly and experimentation. In Benin the Ministry of Health through its influence on the enabling environment. States gave local health management committees decision- have great power over individuals and organizations making control over resources. Committee members are through the information they make public and through elected democratically; anyone may serve, provided that at the laws they enact and administer. The rule of law that least one member is a woman. The committee is directly protects both persons and personal property is important involved in preparing the health center's annual budget to a healthy, vibrant civil society. Governments can facil- for submission to the ministry. It is responsible for col- itate participation by safeguarding the right of people to lecting and accounting for funds paid to the health cen- organize, to gain access to information, to engage in con- ter for services and drugs. Representatives of local com- tracts, and to own and manage assets. The constitutions of mittees sit on the board of the government's new drug Bolivia, Brazil, Colombia, and the Philippines explicitly procurement agency, which is one way of keeping the encourage the development and participation of NGOs at agency accountable, and on the Health Sector Coordinat- all levels of decisionmaking. In Singapore the government 120 WORLD DEVELOPMENT REPORT 1997 helps NGOs recruit staff, allocates unused government the production and delivery of goods and services should buildings to them at nominal rents, and funds up to half be rendered to the lowest unit capable of capturing the the capital and operating costs of facilities run by NGOs associated costs and benefits. In many countries this will for social welfare purposes. Without a credible legal en- involve scaling back the power of central government, but vironment that requires NGOs and public agencies to reformers must be discriminating. Depending on the act openly and transparently, legitimate organizations are institutional environment, decentralization can improve deprived of an opportunity to develop, or, perhaps worse, state capability by freeing it to focus on its core functions; the door is left open for unhealthy or corrupt activities it can also, however, undermine that capability. that taint the reputation of all NGOs. The aim must be to The demand for formal, political decentralization has strike the right balance between regulations and reporting been driven by at least three major recent developments: requirements that fosters growth of NGOs, while guard- ing against corruption and malpractice. The minimum size of self-sufficient government has The benefits of greater consultation and partnership declined. New technological options and new demands with civil society show up in improvements in the process from citizens, producers, and consumers mean that of public policymaking, in the quality of service delivery, some of the advantages (security, for example) that kept and, in some instances, in improved rates of return. They countries, regions, and provinces working together also manifest themselves in the greater flexibility afforded under a central government have become less impor- to public agencies and officials in the way they intervene. tant. In Europe and North America the pressure from But without effective monitoring this flexibility can give global markets is creating strong demand for local and rise to capricious or arbitrary action. Again, Finding the regional governments that can better provide the in- right balance between participatory mechanisms and en- frastructure and skilled labor force that multinational lightened government control is crucial. The next section businesses need. examines some of these concerns in the context of the Political changes have given voice to local demands. Cen- debate over decentralization and the fostering of greater tralized authority in Czechoslovakia, the Soviet Union, public accountability from below. and Yugoslavia collapsed once the unifying force of the Communist Parry disappeared. Elsewhere, regions and Decentralization: Matching services with subnational governments benefited from the political local preferences vacuum created before and during regime changes, as Decentralizing state power and resources seems a logical in Argentina and Brazil in the late 1980s, and South continuation of the many recent efforts to bring govern- Africa in the 1990s. ment closer to the people. Like the broad range of partici- Countries often turn to local and regional governments patory mechanisms described earlier, decentralization when the central government has persistently failed to pro- offers the chance to match public services more closely with vide essential services. In the second half of the 1980s, local demands and preferences and to build more respon- Colombia embarked on a path of decentralization and sive and accountable government from below. But decen- political reform that reversed a long tradition of central- tralization also has its pitfalls, including the possibility of ism. A new government changed direction, transferring increased disparity across regions, loss of macroeconomic social services delivery to the local level and opening up stability, and institutional capture by local factions, espe- the rigid political appointment system to local electoral cially in highly unequal societies. This section focuses on choice. Similarly, in Venezuela and other countries in some of the factors explaining recent trends in decentral- South America, active local governments have made ization, and on some areas where it has been shown to have local administration more responsive and improved the a positive impact, including bringing citizens into public quality of services provided, often dramatically. affairs and stimulating local economic development. The section concludes by exploring the risks of decentralization Before assessing how governments can act to meet such and the implications for governments of differing capabili- demands, it is worth asking what decentralization really ties starting out on the decentralization path. means. In fact, the term encompasses a wide range of dis- tinct processes. The main ones are administrative decon- The age ofdecentralization? centration, or the transfer of state functions from higher The rising demand for decentralization has come as part to lower levels of government while retaining central con- of the broader process of liberalization, privatization, and trol of budgets and policymaking; fiscal decentralization, other market reforms in many countries. These reforms or the ceding of influence over budgets and financial deci- are distinct from one another, but their underlying ratio- sions from higher to lower levels; and devolution, or the nale is similar to that for decentralization: that power over transfer of resources and political authority to lower-level BRINGING THE STATE CLOSER TO PEOPLE 121 authorities that are largely independent of higher levels of is as fruitless here as in other aspects of state reform. But government. Rarely does decentralization embrace all a number of important analytical principles are available three. This wide diversity of experience makes it difficult to guide reformers. The clearest and most important prin- to compare trends across countries or draw many hard ciple (often referred to as subsidiarity) is that public goods and fast conclusions. and services should be provided by the lowest level of gov- Richer and larger countries tend to be more decentral- ernment that can fully capture the costs and benefits. ized, in terms of the share of subnational governments in Applying this principle is no simple matter, however. total public expenditure and revenue. In the aggregate, Table 7.2 illustrates some of the demand and supply char- however, industrial countries have become slightly more acteristics of goods and services that will be relevant to the centralized since 1974 (Table 7.1). This is especially true decision. As already discussed, for some local goods, such of the United Kingdom, whereas Australia, France, Spain, as those with common-property characteristics, organiza- and the United States are continuing to decentralize cen- tions outside of government such as forest or water user tral government functions. Developing countries, most of associations or NGOs may be the most appropriate insti- which went through a nation-building phase of develop- tutional mechanism for delivery. More generally, where ment in the aftermath of colonialism in the 1950s and preferences or demands differ from one community to the 1960s, have become more decentralized since the 1970s. next, local governments can better match supply to suit Striking examples include Argentina, Brazil, and Colom- local tastes. Decentralized service provision can also bia. In both groups of countries decentralization of expen- enhance efficiency and interj urisdictional competition in diture has gone significantly further than that of corre- supply, providing consumers (at least in theory) with the sponding revenue. option to exit to other jurisdictions. On the other hand, where economies of scale or interjurisdictional spillovers How to think about decentralization are present on the supply sideas in the construction and As the above discussion has made clear, what constitutes maintenance of interurban highwaysor where mini- the best structure of intergovernmental arrangements will mum standards (such as for primary schooling) and other be highly country-specific. The one-size-fits-all approach consumption externalities apply on the demand side, Table 7.1 Changes in subnational finance in selected countries (percentage of expenditures or revenues for all levels of government) Subnational expenditures Subnational revenues Country 1974 1994 Trend 1974 1994 Trend Argentina 25 45 1* 25 37 'V Australia 47 49 20 27 'V Brazil 30 38 Iw 23 25 Canada 61 60 39 44 'V Chile 2 9 'V 2 5 Colombia 25 33 16 18 France 18 19 6 13 'V Germany 44 40 34 30 India 45 49 27 25 Indonesia 11 15 3 3 Iran, Islamic Rep. of 1 5 1 6 Malaysia 18 14 13 8 Romania 16 10 12 6 South Africa 24 41 'V 4 12 'V Spain 10 34 'V 5 12 'V Sweden 44 34 28 32 Thailand 17 8 5 5 United Kingdom 33 28 15 8 United States 45 44 33 36 Zimbabwe 26 25 24 15 Note: Data are for all levels of government other than central government. Data include transfers from central government to subeational governments. Arrows indicate changes of 5 percentage points or more. Where data for 1974 or 1994 were unavailable (indicated by italics), data for the closest available year were used. Data for Germany for 1974 refer to the preunification territory. Source: IMF, various years (a). 122 WORLD DEVELOPMENT REPORT 1q97 lize resources, improve service delivery, and stimulate pri- Table 7.2 Demand and supply characteristics vate sector development. of local and national public goods The benefits of improved local accountability and incentives Level of public Demand-side Supply-side CITIZEN PARTICIPATION. In theory, decentralization can provision factors factors strengthen and complement the measures to broaden pop- Local Variation in local Potential for ular participation described earlier in this chapter. Like taste (Street competition be- them, it can help guard against majoritarian tyranny by lighting, zoning) tween jurisdictions moving government closer to people and by facilitating (police protection, local definition of issues and problems, especially those of Common property road maintenance) minority groups. The contrasting situations of Oaxaca and (urban roads, waste disposal) Chiapas, two of the poorest states in Mexico, provide a telling example of these effects at work. The two states National Spatial consump- Economies of have similar resource endowments and development tion externalities scale (defense) potential, and both have a high percentage of poor and (control of indigenous populations. Yet the outcomes of antipoverty Cross-jurisdictional epidemics) externalities (inter- programs are generally regarded as good in Oaxaca, Equity concerns urban highways) whereas neighboring Chiapas has a bad record. The dif- (minimum stan- ference seems to stem from the degree of popular partici- dards for primary education) pation in policy decisions and implementation. Oaxaca has a long tradition of participatory mechanisms for indigenous populations and the poor. In Chiapas, on the other hand, the denial of such options, coupled with wide- centralized control (whether at the national or the provin- spread official corruption, has led to poor services and ris- cial level) is likely to be preferable. ing tensions, including armed conflict since early 1994. Matching services to local preferences can lead to lower Where public office is contested and people can par- transactions costs (particularly information costs), effi- ticipate in elections and decide on representatives at dif- ciency gains, and incentives to local economic develop- ferent levels of government, the number of political ment. But even where a service might, in principle, seem choices also increases, thus stimulating competition bet- a candidate for local provision, the benefits and costs of ween levels of government. Local participation can also decentralization will vary by setting. And experience sug- mean greater confidence in and acceptance of policy deci- gests that decentralization is unlikely to work without sions by constituents. Decentralization can therefore in- effective institutional arrangements to foster accountabil- crease local options for policymaking while holding local ity at the local level, and fiscal restraint on the part of both officials accountable for what they do and how they do it. local and national governments. This is best explained in Recent evidence from Latin America, particularly Colom- terms of two separate but interrelated sets of relationships bia, suggests that once local policymakers are held ac- facing local governments, both of which need to be con- countable for their actions and made aware that their jobs sidered in gauging the scope for decentralization. depend to a large extent on citizens' assessments of their The first are horizontal relationships between local performance, they tend to be much more concerned with government and citizens, NGOs, and private businesses. the quality of their staff and of the tools they have to run Institutional arrangements, for example local elections or their offices effectively. In Pôrto Alegre, Brazil, an inno- referendums, can create or influence such relationships, vative process of public investment planning and manage- providing incentives for cooperation, accountability, and ment was launched in 1989, to mobilize citizen groups to improved local government performance. The second set take part in formulating the municipal budget. In 1995 of relationships is vertical, between levels of government. some 14,000 people were engaged in the process through Most countries have formal institutional arrangements assemblies and meetings. Indirectly, an estimated 100,000 that define the role and functions of each tier of govern- people were linked to "participatory budgeting" through ment, particularly as they affect intergovernmental fiscal local associations and popular organizations. relations. Both vertical rules and horizontal incentives are LOCAL SERVICE PROVISION. Many governments have essential if local governments are to perform their func- responded to fiscal crises, the availability of new technolo- tions well (Figure 7.3). The next section discusses some of gies, and citizen concerns by transferring resources and the ways in which horizontal relationships can encourage responsibility for service provision, especially in education local governments to enhance their responsiveness, mobi- and health, to local authorities. In many cases this has BRINGING THE STATE ELOSER TO PEOPLE 123 given rise to new and often creative arrangements among LocAl. ECONOMIC DEVELOPMENT. The participation local governments, NGOs, and local businesses. Although of local businesses can also play a crucial role in decentral- relatively little comparative evidence is available with ization, shaping incentives at the local level. Entrepreneurs which to evaluate the relationship between decentralized have for centuries strongly influenced both the pace of government and service quality, some recent examples development and intergovernmental relationships in in- from Latin America are illustrative. In the 1980s the pri- dustrial countries. Property-owning classes with command mary education system in the state of Minas Gerais, in over local resources exerted considerable pressure on pub- southeastern Brazil, faced many of the problems common lic entities. To spur expansion, private actors and public among education systems in developing countries: high officials were encouraged to cooperate. Much of this began repetition rates, low graduation rates, and low achievement in local environments. Members of the business commu- scores. Contributing to these problems were overregulated nity often participated in local legislatures. Provided that and centralized management, inadequate funding, and rent seeking is minimized by effective competition poli- poorly trained teachers. In the 1990s a series of measures, ciesa function for higher levels of governmenta including the introduction of autonomy to elected boards strong local private sector may promote better administra- in each local school (composed of teachers, parents, and tive performance. And local governments that provide and students over 16), together with grants from central gov- maintain credible frameworks for local economic develop- ernment based on enrollment and special needs, have pro- ment end up promoting private investment, which over duced some encouraging early results: achievement scores time increases local government revenues. have risen by 7 percent in science, 20 percent in Por- The world is replete with examples of local govern- tuguese language skills, and 41 percent in mathematics. ments that have stimulated economic development in In Teocelo, a town in Mexico's state of Veracruz, their communities, and of decentralized institutional decentralization has created opportunities to organize arrangements that have contributed to growth. The cities resources for health more efficiently by identifying the of Greenville and Spartanburg, South Carolina, are small, population's needs and designing strategies to foster par- inland, and far from major U.S. population centers. Yet ticipation through community organization and health they have the highest foreign investment per capita of any education. Coverage increased for both preventive and metropolitan areas in the United States. They are host to curative care, the quality of services improved enormously, 215 companies from eighteen countries, seventy-four of and infant mortality rates fell. In addition, users of health which are headquartered there. Visionary decisionmakers facilities reported that the attitudes of health personnel with a strong private sector approach to local development and the quality of services had improved greatly. have established a solid base of innovative small and Figure 7.3 Vertical rules and horizontal incentives shape local government's capability Vertical accountability... and intergovernmental rules Horizontal flexibility and accountability 124 WORLD DEVELOPMENT REPORT 1997 medium-size enterprises, employing a work force whose must at any given time be clear rules speciing the range skiiis are regularly upgraded. of responsibilities for each level of government. On the other side of the world, the city government MACROECONOMIC DIMENSIONS OF INTERGOVERNMEN- of Wuhan, in central China, decided in the early 1990s TAL FINANCE. Macroeconomic control is universally re- to transform the old city and build a new one on a large garded as a function of central government. Centralization, scale. To this end it relaxed controls on foreign in- or at least strong central guidance, in budget and financial vestment in two development zones, opened a third, matters has proved critical in ensuring sound public passed local regulations to supply foreign investors with a finances and a reliable framework for economic develop- legal foundation for business operation, strengthened the ment in most industrial economies. Because decentral- management of real estate and land rentals, and under- ization increases the number of actors and of budgetary took several projects to improve the infrastructure for for- accounts, countries facing serious budgetary and inflation- eign investment. As a result, in 1992 alone the number of ary pressures will be confronted with additional challenges foreign investment projects approved for the city was and risks should they embark on decentralization. more than two-and-a-half times the total for the previous Intergovernmental fiscal relations mainly affect the eight years, producing a threefold increase in total capital macroeconomy through three channels: the assignment invested. Not satisfied, the city government organized a and sharing of tax bases and expenditures, the match of huge investment promotion mission in 1993, which gar- tax and expenditure decisions, and levels of subnational nered agreements worth $5 billion from Singapore and borrowing. Hong Kong (China). Serious macroeconomic imbalances can occur if major Despite such encouraging cases, experience suggests tax bases are inappropriately assigned. In India, for exam- that successful decentralization may be short-lived, or dif- ple, important tax bases have been assigned to subnational ficult to replicate, without effective rules for intergovern- governments. This has left the central government, mental collaboration. Horizontal incentives for improved despite a growing public debt and pension liabilities, with performance tell only part of the story. In education, for a tax base, consisting mainly of income, foreign trade, and example, involvement of higher levels of government may excise taxes, that is too small to cope comfortably with its be needed to prevent fragmentation and to minimize dif- expenditure responsibility. The sharing of major tax bases ferences in the quality of education in different commu- also has the potential to dilute the impact of deficit reduc- nities. And in the health sector the appropriate allocation tion at the central government level. This happened in of responsibilities across levels of government is rarely Argentina in the early 1 990s, when increased tax revenues clear-cut. Immunizations, tuberculosis surveillance, and following a tax reform had to be shared with provincial vaccine storage all need strong, effective management governments. Provincial governments essentially took a from higher levels of government. In addition, localities free ride on the central government's efforts and used the may not provide the right framework for policy formula- extra revenue to expand their work forces. tion and implementation. Consequently, decentralization Expenditures with national benefits and costs should not become a rigid, doctrinaire exercise, pushing national public goodsshould be the responsibility of functions onto communities and municipalities or artifi- central government. These include the costs of economic cially separating levels of government. Instead it should be stabilization and redistribution. But many local expendi- a practical endeavor to find the right balance between the tures also affect income distribution, such as the provision roles of different levels of government, to ensure that of health and housing subsidies in transition economies high-quality services are provided in a timely manner. As and many developing countries. In addition, where the the following sections describe, the quest for that balance benefits of local public expenditure are concentrated must take place within a framework of credible rules. within the jurisdiction doing the spending, but the costs (in the form of general taxes or negative spillovers) are The pitfalls of decentralization spread more widely, subnational governments have an In many cases decentralization is not the result of any incentive to spend beyond their means. The effect on carefully designed sequence of reforms, but has occurred national fiscal policy can be severe. in a politically volatile environment in which the level of SUBNATIONAL BORROWING. Borrowing by local gov- trust is iow and policymakers respond unsystematically to ernments can contribute to macroeconomic instability emerging demands from below. Such weak policy frame- when the central government fails to impose hard budget works can lead to serious economic problems, including constraints and there is no effective mechanism for moni- loss of macroeconomic control, regional disparities in ser- toring debt obligations, particularly when there are vice provision, and misallocation of resources as a result of multiple lenders. Another problem is asymmetric informa- local capture. The lesson, for all governments, is that there tion on the part of borrowers (subnational governments) BRINGING THE STATE CLOSER TO PEOPLE 125 and lenders (central government and international capital bility (Box 7.6). The combined indebtedness of the Brazil- markets). In China, for example, provincial governments ian states exceeds $100 billion, close to the levels of total are not allowed to finance budget deficits through borrow- federal and central bank debt. Unless the growth of this ing. But in the early 1990s all-but-uncontrolled borrowing debt is curtailed, the federal government will have to by state enterprises at the subnational level contributed to reduce its own spending, raise taxes, or resort to inflation- economic overheating and imperiled macroeconomic sta- ary financing to cover the stares' debts. Box 7.6 Pitfalls in intergovernmental relations: The experiences of Brazil and China Democratization and constitutional revisions in the Government revenues as a share of GDP declined 1980s increased the amount of resources under subna- precipitously. By contributing to increased indus- tional control in Brazil and the degree of local auton- trial competition, decentralization helped lower the omy in their use. Local governments now account for profits of industrial state enterprises, previously the half of total public spending. main source of tax revenues. Increasingly autono- Although decentralization shifted resources down- mous local governments reduced revenues even fur- ward, there was no corresponding clarification or ex- ther by granting tax exemptions to improve the pansion of local responsibilities. Subnational govern- after-tax earnings of the state enterprises under their ments were not prepared to assume new tasks, and control. were neither required to perform specific functions nor The growing autonomy of local governments also prohibited from performing functions already per- made it difficult for the central government's invest- formed by other levels of government. As a result, local ment planning system to control the investments of governments used much of their windfall to increase provincial governments and state enterprises under staffing and launch questionable new projects. There is their control. Since tax revenues at the provincial scant evidence that the overall efficiency of public sec- level had declined and were inadequate to cover tor spending improved. Decentralization also increased these investments, local branches of the state banks the fiscal deficit, as large states used their improved were usually prevailed on to lend for these projects. political autonomy to extract federal resources: by the Local branches of the central bank were given discre- mid-1990s nearly a third of the growing federal deficit tionary authority over 30 percent of the central was due to subnational debt. bank's annual lending to the financial system. When Brazil's experience shows that political and fiscal local branches of the state bank needed additional re- decentralization does not guarantee improved public sources to support investments by local governments sector efficiency, and may threaten macroeconomic and state enterprises, they turned to the local branch stability. To achieve its objectives, fiscal decentraliza- of the central bank for an infusion of liquidity. tion must be accompanied by a corresponding de- centralization of expenditure responsibilities; state and The resulting overheating of the Chinese economy municipal governments' institutional capacities should in 1992-93 posed considerable risks to stability. Infla- be improved; and the federal government should im- tion climbed to its highest level in several decades. Real pose hard budgets in its fiscal and financial relation- GDP growth reached an amazing 14.2 percent in 1992 ships with subnational governments. and 13.5 percent in 1993. Eighty percent of this China's experience in the early 1990s demonstrates growth came from growth in investment, most of it by the pitfalls of decentralization that is not accompanied state enterprises under the supervision of provincial by parallel reforms and macroeconomic safeguards. governments. Beginning in 1978, central authority over investment The authorities responded quickly with a combina- and allocation decisions was gradually decentralized to tion of measures. The most important of these were provincial governments, enterprises, financial institu- administrative restrictions on investment by provincial tions, and even households. This was a crucial element governments and state enterprises and a reassertion of of China's economic liberalization and a key factor in authority by the central bank over lending to state the economy's impressive growth over the past two banks. These measures and others helped bring the decades. At the same time, however, three conse- economy to a soft landing. By 1995 inflation had quences of decentralization undermined the central fallen below 7 percent, while GDP growth had been government's control over macroeconomic aggregates: maintained at around 9 percent. 126 WORLD DEVELOPMENT REPORT 1997 All three channels can lead to undesirable macroeco- ernmental rules, local governments may benefit from nomic outcomes. But some of the channels are quantita- sources of income that have not been formally allocated to tively more important in some countries than in others. them under an appropriate fiscal decentralization scheme. Whether they generate macroeconomic instability will In Poland, for example, the introduction of local self- depend on the relative importance of each channel, the government has seen many local authorities begin to act relative capacity of central and local government policy- like pressure groups, eager to extract more benefits from making and implementation, and the central government's Warsaw for local clienteles. As a consequence, inequalities commitment to overall macroeconomic objectives, such have risen among jurisdictions, leading to new forms of as growth and price stability. social conflict. REGIONAL DISPARITIES AND INEQUALITIES IN SERVICE In Pakistan decentralization has been accompanied PROVISION. Rough national equality in living standards by a subtle recentralization of functions at the provin- and in access to public services is an overarching goal cial level and difficult intergovernmental relationships. even a constitutional mandatein many countries. Cen- Provincial governments, which have expanded their role tralization allows the national government more discre- in the provision of education and other local public ser- tion to counter regional income disparities by managing vices since the 1960s, have increasingly adopted an intru- regional differences in levels of public service provision sive, centralist attitude toward municipal governments. and taxation. With decentralization, an equitable out- Instead of being encouraged to assume new tasks and re- come can no longer be guaranteed or, at least, may be sponsibilities, municipal governments are being denied more difficult to achieve. And wealthier local govern- opportunities to succeed. Increased provincial control has ments and regions may benefit disproportionately from not led to any noticeable improvement in service delivety, being given greater taxing power. however. In China, for example, disparities in real income per Industrial countries that have decentralized began the capita between provinces have been growing in recent process with a strong legal framework, which ties subna- years. Income per capita in the richest province, Guang- tional governments to credible rules. Many of these coun- dong, is now four times greater than that in the poorest, tries also have mechanisms in place, such as fiscal transfer Guizhou. Some provinces on the southern coast, such as policies and equalization schemes (Box 7.7), hard budget Fujian, Guangdong, and Hainan, have done better than constraints, and limitations on local borrowing, to coun- the western, interior provinces largely because of their terbalance negative outcomes. Some countries have exper- central location for transport and communications and imented with participatory mechanisms (blue-ribbon com- their proximity to Hong Kong (China). These natural missions, for example), which bring stakeholders together advantages have been reinforced by official policies that and provide options for feasible and manageable policies favor coastal provinces, including tax breaks to foreign within a mutually agreed time frame. investors locating in special economic zones near the coast, large allocations of credit (relative to population) Lessons for successful decentralization through China's government-directed banking system, Ideally, policymakers would embark on decentralization and registration requirements that discourage the poor by gradually phasing in the reassignment of revenue from migrating to the booming coast. authority and of expenditure authority and responsibili- In Russia income inequality across oblasts is high. The ties in ways that are compatible and consistent with pre- ratio of expenditure per capita between the lowest- and viously defined needs and responsibilities. At the same the highest-spending oblast is estimated to have risen to time, they would develop a system of intergovernmental one to seven by 1992, with better-off regions receiving grants to cover gaps between expenditures and revenues at disproportionately high budgetary expenditure allocations the local level and to correct imbalances in efficiency and and rural areas relatively ill served. Subnational taxation effectiveness, preferably with built-in incentives for local or sharing of federal revenues from natural resources on resource mobilization. So much for the ideal. The real-life the basis of their origin could create even greater fiscal dis- business of designing a successful decentralization pro- parities. Both Russia's and China's experiences highlight gram tends to be more complex. the need to design appropriate equalization schemes to ASSIGNING EXPENDITURE AND REVENUE AUTHORITY. deal with rising inequality during periods of accelerated This brings a host of problems. Information on the true growth or macroeconomic stabilization. distribution of benefits and taxes both within and be- RESOURCE MISALLOCATION AS A RESULT OF LOCAL CAP- tween jurisdictions is imperfect. And economies of scale TURE. Economic and financial distortions may also arise in revenue collection and in the production of services from subnational governments' ability to exploit weak- may partly negate the efficiency advantages of a decentral- nesses at the center. In the absence of agreed intergov- ized system. In addition, the costs of alternative options BRINGING THE STATE CLOSER TO PEOPLE 127 Box 77 Calculating fiscal equalization grants Fiscal equalization programs compensate provinces the arithmetic mean of all provinces as a standard, the whose fiscal capacities are below the average. In addi- province's equalization entitlement for a given revenue tion to safeguarding national objectives of providing source is determined by the difference (if positive) minimum levels of public services nationwide, an between the average potential yield at the national equalization program can foster the participation of average rate of taxation in all provinces in the aggre- member provinces in a federation. Thus fiscal equaliza- gate, and the potential yield obtained in the province tion is often viewed as the glue that holds a federation when the national average tax rate is applied to its together. Economists have long recognized that equal- revenue base. In this way the fiscal capacity of below- ization is justified on grounds of horizontal equity, and average provinces is brought up to the median, the in recent years it has become clear that under certain arithmetic mean, or some other norm. Because the conditions it can promote economic efficiency as well. data on tax bases and tax collections required to imple- In Pakistan, for example, a representative tax system ment a representative tax system are published regu- has been proposed to equalize fiscal capacity across larly by various levels of government in many coun- regions. Designing such a system involves, first, calcu- tries, implementing such a system does not impose lating the revenue that could be raised if a provincial new data requirements and could be implemented as a government used all the standard sources of revenues federal fiscal equalization program in lieu of revenue at the nationwide average intensity of use. Then, using sharing by the population. for service production are often unknown. Some possible arbitrary action of all participants, experience suggests tax and expenditure assignments are shown in Table 7.3. that it is difficult to force agreement and that the result DESIGNING INTERGOVERNMENTAL GRANTS. Intergov- may not be sustainable. This is especially true of develop- ernmental grants are important sources of revenue for ing economies. Lessons from a variety of countries indi- many subnational governments. In Brazil between 1970 cate that key policy decisions are more likely to be sus- and 1992, grants from the federal government financed tained when they are based on a broad consensus among 64 percent of local government expenditure. In South stakeholders. Thus, the prospects of successful decentral- Africa grants from the central government to newly ization are greater with institutional settings and processes elected provincial governments account for about 90 per- that allow for articulation of interests and consensual cent of total revenues for the latter. By their nature, inter- policymaking, as spelled out, for example, in the Euro- governmental grants tend to divorce local spending from pean Charter of Local Government of 1985. local resources, and the benefits of providing local public In the absence of agreed-on guiding principles, what services from the costs. The separation of benefits and can governments committed to decentralization do to get costs and the limited ability of local governments to mobi- the process started? Some models already exist. In the lize revenues for themselves can reduce the transparency early 1 990s the government of Uganda established a con- of local budgetmaking and the accountability of local gov- sultative process with different stakeholderscommunity ernments to local citizens, leading to inefficient and groups, agricultural producers, and government represen- inequitable delivery of public services. Clearly, then, any tativesto decide on the best way to proceed with de- system of intergovernmental grants in developing coun- centralization. The decision was made for a staged and tries will need to be designed extremely carefully. gradualist approach. Other countries have opted for com- No blueprint exists for an optimal system of intergov- missions made up of informed leaders from different lev- ernmental grants, but a good system should have certain els of government, academia, and sometimes unions and characteristics. Above all, it should be predictable and business associations, drawing on foreign advice where transparent and embody the relevant principles set forth necessary. South Africa's recent experience is interesting in Table 7.4. in this respect. The new constitution calls for a fiscal com- AGREEING ON THE RIGHT APPROACH. Clear-cut rules mission to deal with the intergovernmental structure of are essential for imposing restraints on actors at each level the country. The president, in charge of setting up the of government. Equally important seems to be the process commission, provided different representatives with a by which the rules are agreed on. Although, in principle, constitutionally guaranteed forum for articulating their rules could be imposed from the top down to restrain the interests. Although it is too early to judge its success, the 128 WORLD DEVELOPMENT REPORT 197 Table 7.3 Possible tax and expenditure assignments by level of government Central government State government Local or provincial government Revenues Value added tax Individual income tax Property taxes Individual income tax Surcharges on national taxes Vehicle taxes Corporate income tax Retail sales taxes User charges Excise taxes Excise taxes Licenses and fees Natural resource taxes Property taxes Customs duties Vehicle taxes Export taxes Expenditures Tertiary health care (control of infectious Secondary health care (hospitals, Primary health care diseases, research) curative care) Primary and secondary education University education University and secondary education Roads and highways (intracity) Roads and highways (intercity) Roads and highways (intercity) Public transportation (intracity) Public transportation (intercity) Public transportation (intercity) Air and water pollution Natural resource management Air and water pollution Solid waste disposal, water, Defense Natural resource management sewerage, fire protection Police protection Land use regulation and zoning Housing Cultural policy Promotion of tourism Police protection process created a widespread sense of expectation and history tells us, paradoxically, is that unless states have generated demands to design appropriate decentralization achieved a certain level of centralization and effective rules measures. for overall macroeconomic control and sound policymak- Commissions may serve short-term interests well. ing, decentralization may be difficult to implement and Over the longer term, however, more-durable solutions may create imbalances. Decentralization, whether it may be needed to allow for formal representation of sub- occurs through gradual evolution or design (or both), can national governments' interests in national policymaking create additional momentum for development, but only if and legislation. Second-chamber institutions, such as the these conditions are met. German Bundesrat, have often been used in this context. Any strategy of decentralization must begin with an Such arrangements offer an institutionalized mechanism assessment of institutional capability at the various levels for articulating interests from below, while providing the of government (Table 7.5). Where there is weak central means to develop widely accepted and credible rules for government capability to manage national fiscal and mon- intergovernmental collaboration, which are essential for etary policy, to enact and enforce credible rules for inter- sustainable decentralization. They also help different tiers governmental affairs, or to provide a framework for bring- of government adjust to emerging needs over time. ing stakeholders together, decentralization will be hard ENSURING THAT THE INSTITUTIONAL GROUND IS to pull off. Strongly polarized relationships between or READY. In principle, bringing policymaking and imple- within tiers of government and extremely weak organiza- mentation closer to the communities they serve, and tional capacity at the subnational level will also be cause involving citizens in shaping policies, lead to greater for concern. In these circumstances policymakers would accountability and improved local checks and balances. do well to postpone decentralization, or eschew ambitious But as the above discussion of the potential for macroeco- strategies in favor of a more carefully staged or sectoral nomic instability shows, government actions at the center approach. Decentralization might begin, for example, can be undermined at the local level if there is no strong with certain priority areas such as education, health, or pattern of fiscal restraint and if there are no enforceable infrastructure. Strong monitoring mechanisms could pro- rules governing intergovernmental relations. In most fed- vide opportunities for learning and for gradually phasing eralized systems effective checks and balances between lev- in new policies. Countries with greater capability at the els of government have evolved over long periods. What central and the local levels have more options to choose BRINGING THE STATE CLOSER TO PEOPLE 129 frombut their preferences will vary. What may be Where appropriate, states should work to ensure broad- important in one country (say, decentralized service deliv- based public discussion and evaluation of key policy ery) may not be desired (or desirable) in others. directions and priorities. At a minimum this means making available information in the public interest and Strategic options: Bridging the gap between state establishing consultative mechanisms such as delibera- and citizen tion councils and citizen committees to gather the The evidence presented in this chapter has shown that views and preferences of affected groups. improving the capability and effectiveness of the state rests They should encourage, where feasible, the direct par- with mechanisms to increase openness and transparency, ticipation of users and beneficiary groups in the design, to strengthen incentives for participation in public affairs, implementation, and monitoring of local public goods and, where appropriate, to bring government closer to the and services. And they should enhance the capacity and people and communities it is meant to serve. Building efficiency of accountable local organizations and insti- capability in this way will take time and careful attention tutions rather than replace them. to the dangers: efforts to open up government to a broader Where decentralized service delivery is considered desir- array of needs and interests will not improve effectiveness able, states should adopt a carefully staged or sectoral or accountability if they shut other groups further out. approach, beginning in priority areas such as health, But the experience of governments the world over sug- education, or infrastructure. They should introduce gests some clear starting points: strong monitoring mechanisms and make sure that Table 7.4 Principles and best practices in grant design Grant objective Principles of grant design Best practices Practices to avoid Bridging fiscal gap Reassignment of Tax abatement in Canada Deficit grants responsibilities between Tax base sharing in Brazil, Tax-by-tax sharing as in levels of government Canada, Pakistan, and India and Pakistan Tax abatement South Africa Tax base sharing Reducing regional fiscal General nonmatching Fiscal equalization General revenue sharing disparities grants programs in Australia, using multiple-factor Fiscal capacity equalization Canada, and Germany formulas transfers Compensating for benefit Open-ended matching Transfers for teaching spi lovers transfers with matching hospitals in South Africa rate consistent with estimated spillover Setting national minimum Conditional nonmatching Roads and primary Conditional transfers with standards block transfers with education grants in conditions on spending conditions on standards of Indonesia alone service and access Education transfers in Ad hoc grants Colombia and Chile Influencing local priorities Open-ended matching Matching transfers for Ad hoc grants in areas of high national transfers with matching social assistance in but low local priority rate varying inversely with Canada local fiscal capacity Stabilization Capital grants, provided Limited use of capital Stabilization grants with no maintenance is possible grants with encouragement future upkeep requirements of private sector participation through guarantees against political and policy risk 130 WORLD DEVELOPMENT REPORT 1997 Table 7.5 Matching decentralization strategy to government capacity Central government capacity Local government capacity Low High Low Cautious decentralization strategy with pilot Deconcentration of some priority services testing Delegation of some functions to NGOs and Delegation of some functions to NGOs and communities communities Massive institutional strengthening at both Targeted strengthening of local entities during levels, particularly in public finances (most Sub- transfer of responsibilities (e.g., Hungary, Saharan African countries) Mexico, Thailand) High Separatist or secessionist tendencies Delegation or devolution of functions according to government priorities and preferences as well Delegation or devolution according to priorities of as articulated needs (most industrial countries) governments (e.g., Santa Cruz Province, Bolivia; parts of former Soviet Union) sound intergovernmental rules are in place to restrain gridlock or of capture by vocal interest groups. And if no arbitrary action at the central and the local levels. clear-cut rules impose restraints on the different tiers of At the local level, states should focus on the processes government, and no incentives encourage local account- and incentives for building accountability and compe- ability, the crisis of governance that now afflicts many tition. Where local governments are weakly account- centralized governments will simply be passed down to able and unresponsive, improving both horizontal ac- lower levels. However, as Part Four of this Report argues, countability (to the public) and vertical accountability the obstacles on the path to reform of the state are not (to the center) will be a vital first step toward achieving insurmountable. The first step toward bringing govern- greater state capability. ment closer to people will be to make the objectives of reform clearly intelligible to citizens and the business Certain dangers are inherent in any strategy aimed at community. Efforts at communication and consensus opening and decentralizing government. Expanded building will have a double benefit, increasing the support opportunities for voice and participation increase the for reform as well as arming the government with a better demands made on the state, which can increase the risk of sense of how to do it right. FACILITATING INTERNATIONAL COLLECTIVE ACTION 7 EXPLAINED HOW NEW PARTNERSHIPS AND they achieve little when countries have signed on without competitive pressures can enhance the state's effec- CFIAPTER fully understanding, or accepting, the likely costs. tiveness at home. But the challenge of reforming the state Funding and provision of international public goods does not stop at the state's borders. In an increasingly interdependent world, one country's actions will often Not so long ago the standard policy advice with regard to have implications for its neighbors and for the rest of the the provision of public goods relied almost exclusively on world. And there is a growing recognition that some state intervention. Depending on the circumstances, the needed public goods and services can only be secured prescription might be to introduce a subsidy, a tax, a new through international cooperation. Thus, building state liability rule, a new regulation, or a new program for direct capability will mean building more-effective partnerships public provision of the good in question. But this approach and institutions internationally as well as at home. usually fell flat when it came to the provision of interna- The need for international cooperation stems from tional public goods. In a world of sovereign nations, volun- global and regional manifestations of the problems de- tary cooperation becomes the only answer. But why would scribed in earlier chapters, such as missing markets and the countries undertake cooperatively actions that they have presence of externalities. World peace, a sustainable global little or no incentive to carry out individually? environment, a single world marketplace for goods and Experience and a better understanding of how eco- services, and basic knowledge are all examples of interna- nomies work have since led us to recognize a richer set of tional public goods. They will be underprovided without motives for collective action and to devise better institu- conscious, concerted, and collective efforts to provide tional arrangements for carrying them out, be they them. Development aid, although not a public good in the national or global. As previous chapters have shown, states strict sense, also justifies international cooperation because are setting aside monopolistic, command-and-control of global equity considerations. approaches to governing in favor of a more participatory This chapter discusses the ways in which governments approach involving civil society, markets, and local might help ensure more effective global provision of inter- authorities. At the global level, the participatory approach national public goods. It begins by examining the volun- goes a step further, since it relies on international cooper- tary mechanisms already established to coordinate inter- ation without the use of coercive power. Today, the key national collective action. Although the evidence is clear mechanisms for the provision of international public that cooperation to achieve global collective goals brings goods are based entirely on voluntary action. global benefits, not every such action will bring benefits In international markets for trade and investment, for all. Hence it will not always be in every country's inter- countries have collaborated to develop common rules est to participate. Some international public goods may and norms of conduct and to institutionalize them simply not be valued as highly by some countries as by through various formal arrangements. These have in- others, and sometimes the domestic costs of complying cluded regional arrangements such as the Asia-Pacific Eco- with an agreement may outweigh the benefits. A major nomic Cooperation (APEC) forum and Mercosur in lesson of experience with voluntary agreements is that South America, as well as multilateral ones such as the 132 WORLD DEVELOPMENT REPORT 197 World Trade Organization (WTO) and its precursor the Collective provision of public goods generally requires General Agreement on Tariffs and Trade (GATT). Al- balancing three principles: openness, diversity, and cohe- though all these arrangements are based entirely on sion. Each has considerable merit, but pursuing any one to voluntary participation, they have attracted growing mem- the extreme risks compromising the others. Openness to berships, contributing hugely to growth in world trade the global economy involves a commitment to transparency and improvements in the participants' welfare. Organizing and clear rules, for example prohibiting discriminatory and operating these arrangements are costly, but members trade practices. Yet openness without diversity could lead have deemed the benefits sufficient to justify the costs. to backsliding by countries that desire more freedom to dif- fer, and thus to greater fragmentation of the world econ- When is cooperation desirable? omy. Likewise, acceptance of diversityfor example in the For any country, the decision to cooperate in important mutual recognition of different national standardsfosters international endeavors will be a complex one, condi- innovation but may be detrimental to cohesion among tioned by the country's social values and by its assessment countries or among communities within a country. Cohe- of the long-term national interest. The balance of the sion, of course, is also generally desirable, but not when it costs and benefits of cooperating depends on the type of involves sacrificing too much openness and diversity. activity, the mechanisms proposed, and the social and These considerations suggest that although there is economic conditions the country faces. much need for collective provision of international public There is greater recognition today not only of the exis- goods, it is not the answer to all problems for all countries. tence and benefits of international public goods, but also Each country has to decide, on a case-by-case basis, of the implications of failing to provide them adequately. whether to participate. An appropriate global framework History has shown what can happen when the commu- for organizing collective action must therefore allow for nity of nations is more fragmented, by war or by trade and multiple arrangements and institutions, all based on vol- investment barriers, than it is today. Without a forum for untary participation. One possible approach to organiza- nation-states to discuss and negotiate orderly changes in tional design is to think in terms of a series of groups, each national policies and standards, small economies may end with a different objective: up having to adopt the practices of the dominant eco- nomic powers, in a process of "imperial harmonization." Functional groups to deal with specific issues such as And failure to provide effective foreign aid or support for macroeconomic policy, environmental protection, basic research to meet the needs of poor countries dimin- labor standards, and international conflicts (for ex- ishes the prospects of those countries ever developing into ample, the International Labour Organisation and the vibrant economies and profitable trading partners. Inter- Bank for International Settlements) national cooperation is critical to making events turn out Regional groups to deal with multiple issues of interest significantly better. to neighboring countries (but preferably open to all But again, not all countries will in all cases wish to that wish to join), including trade and investment participate in the provision of international public goods. (for example, the North American Free Trade Area, or At the very least, cooperation can restrict a country's free- NAFTA, and APEC) dom to act. Often the benefits will exceed the costs of ced- Coordinating groups to link the functional and ing some national autonomy, but not always. In many regional groups and create a broader network for all areasmacroeconomic policy coordination is onethere members (for example, the OECD). is inevitably some uncertainty about what kind of joint action to take. In others, such as environmental protec- This framework provides for a reasonable balance tion and climate change, there is uncertainty about the among openness, diversity, and cohesion and may well be extent to which the key players will participate. Uncer- adequate to prevent fragmentation and imperial harmo- tainties such as these blunt the likely effectiveness and nization. Individual countries' interest and participation dilute the expected benefits of cooperation. in the various arrangements at any given time will vary, Even when the relevant facts are more certain, differ- and groups may be relatively inactive for long periods. But ing perceptions and priorities can still preclude coopera- when the timing is right, with ideas and circumstances tion. Many developing countries, for example, are reluc- converging, groups may witness a surge of interest in their tant to adopt the labor laws and pollution standards of the activities, as occurred, for example, in the concluding richer countries for fear 0f losing their competitive advan- stages of the Uruguay Round negotiations of the GAYT. tage and jeopardizing growth. And some countries may choose not to cooperate in certain activities out of a belief Ensuring more effective cooperation that private research and experimentation will ultimately There is no guarantee that participating states will always produce cheaper solutions. fully comply with external commitments. In the absence FACILITATING INTERNATIONAL COLLECTIVE ACTION 133 of some global authority with universal jurisdiction and implementation reports by member countries. These bod- coercive power, enforcing international agreements and ies, however, lack enforcement power. They also have lim- treaties is up to the states themselves. Hence mechanisms ited capacity to verify implementation unless countries are clearly needed to ensure compliance when commit- cooperate in providing information. More sophisticated ments are not being honored voluntarily. Recent experi- legal agreements, including some recent environmental ence in international law suggests a few such mechanisms. conventions, add the element of supervision. A supervi- Countries fail to comply with international commit- sory body can help by following up on reporting require- ments for different reasons. The incentives to comply may ments and by disseminating information on the impact of be weak, because of changes in political priorities or in the convention at the domestic level. underlying economic conditions. Or the necessary capa- The procedure for noncompliance established under bility, including the technical expertise and organizational the Montreal Protocol on Substances that Deplete the skills needed to ensure timely action, may be lacking. Per- Ozone Layer is a good example of this new approach. Any haps most common among developing countries, the req- party to the agreement that has doubts about the correct uisite financial resources may not be available. application of the protocol by another party can initiate Where its incentives have become incompatible with the procedure, as can the secretariat itself or any party hav- fulfilling its international obligations, a country may need ing trouble meeting its commitments. NGOs and indi- to reconsider its participation. The effectiveness of the viduals also have access to this procedure: they can trans- agreement (or international organization) will be under- mit information about possible noncompliance to the mined if many members simultaneously face strong in- secretariat. The Committee on Implementation may try centives not to abide by its provisions. In practice, how- to bring about an amicable settlement, or recommend ever, states have perceived their self-interest broadly, technical or financial assistance if the failure to apply the recognizing the impact of their actions on the community protocol is due to lack of capacity. It can also suspend the of nations, on their own reputation, and on the possibil- rights and privileges of parties in violation. ity of entering into future reciprocal arrangements. Continuous cooperation among national agencies is Lack of capacity and financial resources is often more the foundation of this process-oriented approach. The manageable. Many international agreements take the building blocks include such facilities as permanent com- capacity and financial constraints of some members into munication networks, periodic reporting or implementa- account. Provisions can be included to ensure that neces- tion, periodic review of legal provisions, and regular meet- sary personnel and financial resources are available to all ings of decisionmakers and staff. All these help maintain members. Where such needs are not fully anticipated, awareness among responsible officials of the goals being mechanisms for communication and supervision can be sought by the agreements and the means for achieving devised to address emerging issues. Allocating responsibil- them, and keep the public informed of emerging issues. ities realistically and providing for necessary resource The Geneva-based International Register of Potentially transfers in advance can improve implementation and re- Toxic Chemicals, which is founded on national regulatory duce violations. decisions rather than international regulatory action, is a Traditional legal mechanisms often fail to address the good example of such a facility. root cause of compliance problems, instead relying on Current provision of international public goods breach of the agreement to trigger action. This approach highlights the violation of commitment and is confronta- This section reviews some of challenges and opportunities tional. Necessary remedial actions may come too late facing states as they seek in five selected areas to balance the damage may have been done. Relations among mem- self-interest and the common interest in an increasingly bers may sour, making future cooperation more difficult. interdependent world. An alternative, more process-oriented approach pro- motes the observance of commitments on a continuous Expanding world markets basis. The goal is not to condemn wrongdoing but to keep The liberalization of trade and investment laws around the states in compliance with their obligations and prevent world has contributed to an enormous increase in the vol- violations. This approach relies to a much greater extent ume of world trade and foreign direct and portfolio invest- on communication, consultation, monitoring, the sharing ment, whose impact on the welfare of participants has been of information, and technical and financial assistance. considerable and for the better, Multilateral and regional Recent conventions, particularly in the environmental agreements have supported market expansion, as greater domain, incorporate mechanisms for monitoring and economic interdependence has made it necessary to main- facilitating compliance. These include conferences among tain and extend an international system of liberal trade and the parties, separate secretariats, and financial assistance investment. Invigorated by buoyant trade, the global econ- arrangements that ensure the submission and review of omy has grown rapidly, and that growth shows little sign 134 WORLD DEVELOPMENT REPORT 1997 Box 8.1 The World Trade Organizationan international mechanism for bringing credibility to national policy The WTO was established in January 1995 to admin- Of these matters, two have completed the entire ister multilateral trade agreements negotiated by its process, and two further panel reports have been member countries. The WTO can be regarded both as issued. Countries can no longer block the establish- an institution embodying a set of rules and principles ment of arbitration panels or ignore their findings, as on the use of policies that affect trade flows, and as a they could under the GATT. And although they may "market" in which members exchange market access appeal, the decision of the appeals body is final. Every concessions and agree on rules of the game. The WTO stage of the process is subject to strict time limits, and requires its members to ensure that their trade policies countries that fail to comply face authorized trade are largely nondiscriminatory and their rules and en- sanctions. forcement procedures transparent. The WTO also The largest trading nations and customs territories provides legal mechanisms for countries to signal the continue to dominate the dispute settlement process, seriousness of their commitments, and improved dis- and the credibility of the system depends on their will- pute settlement procedures for resolving conflicts be- ingness to comply with judgments against them. But tween member states. encouraged by the nature of the WTO system, includ- In the first two years of its existence, the WTO ing the right to redress, developing countries are turn- dispute settlement system received a total of sixty-two ing to the dispute settlement process far more often cases involving more than forty-three distinct matters. than they did under the GATT. of abating. International migration of people in search of tries between the mid-1980s and the mid-1990s. Such dis- work is the laggard in this story. As World Development parity in the speed and extent of integration reflects how 1995 showed, annual migratory flows from developing well different regions have succeeded in raising their vol- countries are no greater now, relative to total population, ume of trade with the rest of the world. 'While East Asia, than in the 1 970s. Most workers in poorer countries are for instance, has consistently expanded its trade over sev- only beginning to experience the benefitsand the costs eral decades, that of Sub-Saharan Africa has actually fallen, of global migration. But the expansion of markets and the In addition to liberalizing trade, more countries are increase in competitive pressure will leave some unpre- also gradually removing restrictions on cross-border pared countries highly vulnerable to unforeseen shocks and movements of capital, either unilaterally or as part of policy mistakes. As Chapter 3 explained, countries will regional initiatives. The number of countries with liberal need to adopt prudent, consistent, and credible policies at or mostly liberal capital regimes has grown from nine to home to prepare for the new global environment. Interna- thirty in the past two decades, while the number of coun- tional collective action can help support these efforts by tries with relatively restrictive rules has dropped sharply, offering ways for countries to make external commitments from seventy-three to fifty-three (Figure 8.1). that will give these policies more credibility. Just as countries differ markedly with respect to The growing global consensus on the benefits of more growth in trade, so there is considerable disparity in coun- liberal trade and international market expansion is tries' ability to attract foreign capital. Although worldwide reflected in the large and growing membership of the private and official capital flows have expanded by about WTO (Box 8.1). The most recent series of multilateral a factor often in the past two decades, developing regions negotiations toward trade liberalization, the Uruguay have fared unequally in attracting these flows. Much of Round, led to significant reductions in both tariff and the expansion has been in private flows, and among devel- nontariff barriers to trade in goods and services, particu- oping regions most of these go to East Asia and Latin larly among developing countries. America. One estimate suggests that more than half the Yet reducing border barriers is only one of the precon- population of the developing world has been little ditions for participating more actively in the global trad- touched by this aspect of globalization. ing system. Countries also need a competitive exchange Of particular concern to developing countries is the rate, good availability of foreign exchange, and a transport composition of these growing private capital flows. infrastructure that can support expanded trade. Thus, de- Whereas many developing countries actively seek foreign spite spreading trade liberalization, the share of trade in direct investment, they regard portfolio investment with GDP fell in forty-four of ninety-three developing coun- ambivalence. Foreign portfolio investors can help develop FACILITATING INTERNATIONAL COLLECTIVE ACTION 135 Figure 8.1 Many countries are loosening restraints on international capital Number of countries 45 40 35 30 25 1975 1985 1994 20 15 10 5 0 Liberal Mostly liberal Partly liberal Restrictive Very restrictive Rules governing foreign capital transactions Note: Data are for 102 industrial and developing countries. Liberal means no restrictions; mostly liberal means a few restrictions by industry; partly liberal means many restrictions on the size and timing of transactions; restrictive means that domestic investment by foreigners or foreign investment by domestic residents requires official approval; very restrictive means that all cross-border transactions require official approval. Source: Adapted from Gwartney, Lawson, and Block 1996. local financial markets by providing liquidity and by flows. A better understanding of' these risks and greater influencing the regulatory framework and corporate gov- confidence in managing them would encourage countries ernance. But they also bring the risk of sudden capital to participate more actively in world markets. More-open flight, whose destabilizing effects were dramatically illus- and better-functioning capital markets in developing trated by Mexico's crisis of 1994-95. countries would improve the use of global resources and Managing the risk of capital flight, and of large capital increase portfolio diversification. flows generally, has been a challenge for most developing What kind of collective actions could help achieve countries. Increasingly the risk is regarded as a welcome these benefits? Closer consultation among central banks source of government discipline, which discourages capri- and financial regulators could help upgrade national reg- cious and irresponsible policies, and many countries have ulatory frameworks and financial practices. And greater relaxed capital controls (see Chapter 3). Still, large flows cooperation among national authorities could help estab- in either direction can accentuate a country's vulnerabili- lish procedures for mutual assistance in crises, such as the ties through large external imbalances, rising inflation or IMF's new facility to help member countries absorb exter- interest rates, or exuberant credit expansion that could nal shocks. compromise the soundness of banks. Another concern is growing regionalism. The past The means at governments' disposal to keep them- two decades have witnessed a sharp increase in the num- selves out of trouble are almost all a matter of domestic ber of regional market-opening agreements, including policy: in particular, prudent fiscal policies, credible mon- NAFTA, Mercosur, and APEC. Regionalism is not sim- etary and exchange rate regimes, a sound and prudent ply about trade. In the case of the European Union, for banking system, and, possibly, measures that reduce the example, it also reflects the desires of neighboring nations public's expectation that the government will bail them for greater political integration in response to common out if investments turn sour. security concerns, for cost sharing for infrastructure and But the international community has important inter- institutions, and for increased bargaining power in inter- ests at stake in addressing the risks associated with capital national negotiations. 136 WORLD DEVELOPMENT REPORT 1q97 Opinion is divided on the merits of regional arrange- production. New knowledge can make a dramatic differ- ments, and the evidence remains inconclusive. Some ence in people's lives, as it did with the eradication of argue that regionalism will divert attention and resources smallpox, the containment of malaria and river blindness, away from the more important multilateral processes and and the significant increases in agricultural productivity undermine progress toward global nondiscriminatory made possible by the green revolution. But these successes trade rules. Others contend that regionalism enables states are few and far between. And new breakthroughs are to undertake innovative market-opening measures that unlikely in the absence of assured continued support for will eventually serve as building blocks for multilateral ini- well-directed efforts. tiatives. Regional partners have indeed pioneered arrange- In developing countries research suffers from several ments later adopted in multilateral agreements; an exam- disadvantages. First, research activity in these countries is ple is the European Union's treatment of trade in services, usually given low priority. In Sub-Saharan Africa, for Some of the concerns about regionalism may be legiti- example, less than 2 percent of health budgets is spent on mate. But regional arrangements can be made more con- health research; the result is shortages of research institu- sistent with more-open and integrated world markets. One tions, facilities, and scientists. Second, the limited funding way is to open membership in such arrangements to any that is devoted to research is usually misallocated. Pneu- trading partners that wish to join, rather than restrict it to monia and diarrheal diseases account for 15 percent of the countries within the region. Another option is to establish, disease burden in developing countries, yet only 0.2 per- through a multilateral mechanism, a time-bound conver- cent of medical research funding in developing countries gence process for cutting differences between internal and is directed toward studying these diseases, This misalloca- external trade barriers to a stipulated minimum. tion often reflects a lack of basic information and of the skilled personnel needed to develop an appropriate re- Supporting basic research and the creation of knowledge search agenda. Third, there is little coordination and ex- Knowledge is an international public good whose benefits change by researchers across developing-country borders; accrue to all. International collective action can direct re- the results are a considerable overlap in research activities search toward the needs of developing countries, where and missed opportunities for cost saving. most research activities that exist are fragmented, poorly The international community can do more to assist funded, and inadequately directed. International assis- developing countries in generating the new knowledge tance can help in assessing needs, developing a cost- that will address their needs. Through foreign aid, donors effective agenda, encouraging international exchange and can help governments develop a research agenda based on collaboration, and providing additional funding where it careful assessment of needs, and help finance a higher but is needed. Successes such as those of the Consultative sustainable level of research spending. The industrial Group on International Agricultural Research (CGIAR) countries can also help alleviate brain drainthe exodus and the World Health Organization suggest that the re- of skilled developing-country researchersby providing turn on investment in research in developing countries competent researchers and scientists to work with in- can be substantial. stitutions in developing countries to develop training Basic research is a classicand globalpublic good. capacity and research programs that encourage the reten- The benefits, although uncertain beforehand, and hard to tion of local personnel. Donors can help establish and measure after, often prove exceptionally high. The trans- fund regional research institutes to encourage cross-fertil- formation of the global economy and of entire societies ization of ideas and limit redundant research activities. has as its basis the knowledge gained from new discover- And international institutions can help disseminate the ies. Yet the incentives to conduct and fund basic research new knowledge gained to promote productivity, more are extremely weak: the benefits of greater knowledge are effective treatment of diseases, and healthier lifestyles. nonexcludable, and few constituencies lobby for more The scope for such efforts has been well illustrated by research. Governments in rich countries often regard re- successful cooperation in many areas of research and dis- search as a luxury. Governments in poor countries seldom semination. The CGIAR, for example, was instrumental pay it much attention. in developing more productive crop varieties and promot- Perhaps the greatest mismatch between potential ing more efficient and environmentally friendly agricul- returns and actual investment in research is to be found in tural methods. A network of sixteen agricultural research developing countries. The scope for building human cap- centers around the world, the CGIAR is supported by ital there is enormous, but the process is complicated by fifty nations in its primary goal of alleviating hunger in childhood malnutrition, debilitating diseases, and degra- developing countries. But the benefits of its activity have dation of the natural resources that support agricultural not been limited to those countries (Box 8.2). FACILITATING INTERNATIONAL COLLECTIVE ACTION 137 International cooperation in health research has also led to important advances. The World Health Organiza- Box 8.2 How international agricultural tion, for example, played a major role in the eradication of research benefits donors as well smallpox, But much scope remains for collective action. Of total worldwide spending on health research in the In 1993 the United States produced about 12 per- early 1990s, an estimated 95 percent was devoted to health cent of the world's wheat crop. The United States problems of concern mainly to industrial countries, and is also a major rice exporter, accounting for nearly only 5 percent to the health needs of developing countries. 18 percent of international trade in the commodity. Several research activities especially warrant more in- Most wheat and rice varieties grown in the United ternational support: States were developed through crop improvement research. Many were developed through the work Improving the understanding of tropical diseases, par- of two research centers in the CGIAR network: the ticularly those affecting children and rural dwellers of International Center for Maize and Wheat Im- Sub-Saharan Africa provement (known by its Spanish abbreviation Controlling the spread of the human immunodefi- CIMMYT) and the International Rice Research ciency virus (HIV), which causes AIDS Institute (IRRI). Both are supported in part by con- Improving treatment and prevention of noncommuni- tributions from the U.S. government. cable diseases, which affect an increasing number of A recent study sought to measure the benefits to people in developing countries the U.S. economy from CGIAR research. It was Finding or developing disease- and pest-resistant vari- estimated that, during 1970-93, gains from the eties of such crops as cotton, cocoa, rice, and yams, use of improved wheat varieties developed by which play a key role in many economies CIMMYT amounted to $3.4 billion to $13.7 bil- Developing mining and farming technology to mini- lion. The benefit-cost ratio for U.S. government mize soil erosion and deforestation. support of CIMMYT was as high as 190 to 1. IRRI research was linked to $20 million to $1 billion in Protecting the global environment gains in rice revenues, yielding a benefit-cost ratio A severe threat to development comes from environmen- tal degradation, at both the global and the local level. Par- of as much as 17 to 1. Thus even though U.S. in- ticularly worrisome global environmental issues include vestments in international agricultural research on climate change (Box 8.3), loss of biodiversity, and protec- wheat and rice were made primarily on humanitar- tion of international waters. At the local level the most ian grounds, they have yielded direct benefits to the pressing problems are urban air and water pollution, U.S. economy that far outweigh the costs of sup- deforestation, and soil and rangeland degradation. Inter- porting the CGIAR. And as the study concludes, national collective action can help mitigate these problems "international agricultural research is an investment through better coordination, increased public awareness, in international stability and economic growth technology transfers, cost sharing, and consultation to overseas, which reaps further rewards for the United help shape national and local policies and practices. States and other donor nations." International cooperation is now recognized as the cornerstone of a sustainable environmental regime. In the past two decades the number of international environ- mental agreements has grown significantly. The wide arid and semiarid areas, forested areas, or areas vulnerable array of interests at stake means that activities must be to forest decay; and countries with areas prone to natural coordinated at the international level to ensure stable and disaster. predictable patterns of behavior and to establish coopera- Institutional and financial support is often needed to tive management systems. Although willingness to partic- enable certain countries to meet their obligations. One of ipate in international collective action implies recognition the outcomes of the 1992 U.N. Conference on Environ- of a common objective, different countries have different ment and Development in Rio de Janeiro, for example, interests at stake, and these, too, must be recognized. was a commitment by the industrial countries to provide Article 4 of the 1992 Climate Change Convention, for financial resources to meet the costs incurred by devel- example, requires that parties to the convention give full oping countries in implementing obligations set out in consideration to the interests of (among others) small the conference's Agenda 21. The Global Environment island countries; countries with low-lying coastal areas, Facility, conceived to finance the incremental costs of 138 WORLD DEVELOPMENT REPORT 1997 Box 8.3 The challenges of global climate change for international cooperation Although some doubts remain about the magnitude of ments, such as tradable carbon emissions entitlements, global climate change and the urgency of dealing with to reduce greenhouse gas emissions. In a tradable per- it, a consensus is emerging that the problem is real and mits system, permits corresponding to a targeted ceiling potentially dangerous, and that reasonable and appro- of greenhouse gas emissions would be issued, and emis- priate measures should not be postponed. The Inter- sions without a permit would be prohibited. Countries governmental Panel on Climate Change has predicted for which the costs of reducing carbon emissions are that over the next 100 years the earth's surface will low would have an incentive to undertake those reduc- warm by an average of 1.5 to 6.3 degrees Fahrenheit, tions and sell their unneeded permits to countries for and sea levels will rise 6 to 38 inches. These changes which the costs of emissions reduction are high. A would bring more frequent and intense droughts, the recent World Bank study estimates that cutting emis- spread of disease, the retreat of mountain glaciers, and sions in the countries of the OECD by 20 percent at storms of greater malevolence. least cost could require global trading of entitlements What challenges does climate change pose for inter- worth $30 billion to $40 billion annually. Allowing national cooperation? Under the auspices of the trading would generate savings equivalent to 65 percent United Nations Framework on Climate Change, of global abatement costs. signed in 1992 and ratified by 159 countries, an inter- The barriers to implementing this global market are national agreement to limit the greenhouse gas emis- largely political. The market's very existence depends sions that contribute to climate change is being nego- on governments' willingness to create and regulate it. tiated and may be adopted by the end of 1997. But (The financial resources for purchasing entitlements climate change, if it brings the dire effects that are pre- are expected to come from the private sector.) A crucial dicted, will take far bolder cooperationpolitical, eco- step in establishing the market will be the initial allo- nomic, and financialto meet needs projected to cation of entitlementsthis will have to be deter- reach $50 billion a year by 2040. mined by a global climate change protocol. Although Recent analysis of climate change provides a strong many formulas have been suggested, this contentious economic rationale for adopting market-based instru- issue has not yet been resolved. projects that have global environmental impact, plays a Preventing and controlling conflict crucial role in implementing these obligations (Box 8.4). For most of the twentieth century the world has lived Many of the most immediately pressing environmental under the specter of major war. The first half of the cen- problems facing developing countries, such as urban water tury witnessed two global conflicts, catastrophic destruc- and air pollution and soil degradation, are mainly local tion of lives and resources, and decades of rehabilitation rather than global. But they have major implications for and reconstruction. During the second half the Cold War productivity, health, and the quality of life within these loomed large, with the threat of even greater destruction countries' borders. Progress in alleviating these problems from nuclear weapons. Global tensions led many coun- has been slow, with lack of capacity and political will, at tries to devote a substantial share of national output to the both the national and the local level, a major stumbling military. Only in the past ten years have these tensions block. begun to subside, providing an opportunity for nations to The lesson of recent experience is that bringing about reduce military spending and reap the dividends of peace both local and global environmental integrity and sus- (Box 8.5). tainability will require a coordinated international effort, The threat of nuclear war has been replaced by a pro- one that blends careful attention to financial incentives, liferation of smaller conflicts, bringing costly problems of market forces, laws, and national interests. Equally refugee relief and rehabilitation. Existing cooperative important, the international community must help raise mechanisms have had limited success in managing these public awareness of the dangers of environmental degra- conflicts, or in helping to avoid them. The problems often dation, so as to change the political incentives facing lead- spill over and engulf neighboring countries, as they did ers, which often work against the goal of environmental in Southeast Asia and much of southern Africa in past integrity. decades and are doing now in Central and West Africa. FACILITATING INTERNATIONAL COLLECTIVE ACTION 139 These conflicts are not confined to the poorest countries, but can also break out in middle-income countries such as Box 8.4 Sharing the burden of environmental the former Yugoslavia and Lebanon. The challenge facing protection the international community is to find new ways to pre- vent such conflicts or to manage them at an early stage The Global Environment Facility (GEF) was set up before they turn into tragedies. in 1991 to help developing countries finance the in- The end of the Cold War brought rising optimism that cremental costs of new environmental investments many of the issues that had contributed to instability and with global benefits in four areas: climate change, conflict throughout the world would be resolved. Instead preservation of biodiversity, protection of the ozone at least thirty major armed conflicts (defined as those layer, and protection of international waters. A joint causing more than 1,000 deaths in a year) have taken undertaking of the United Nations Environment place worldwide in recent years. The fragile peace settle- Programme, the United Nations Development Pro- ments in Cambodia and Mozambique now seem the gramme, and the World Bank, the GEF has led to exception rather than the norm. We have seen: new institutional arrangements for the provision of collective goods. A rising number of refugees and internally displaced The GEF has more than 165 member states and persons, and a disproportionate number of women and is governed by a board of representatives from children who lack access to the basic resources needed thirty-two countries, each representing a con- for repatriation or rehabilitation (Figure 8.2) stituency. There are sixteen constituencies for devel- A rise in the number of humanitarian emergencies, oping countries, fourteen for industrial countries, from an average of five a year in 1985-89 to twenty in and two for Eastern Europe. The countries in each 1990, twenty-six in 1994, and twenty-four in 1995 constituency choose a board member and an alter- The erosion or total collapse of legitimacy and author- nate, and each constituency determines its own ity in many states, including Afghanistan, Liberia, process of consultation and decisionmaking. New Rwanda, Somalia, and the former Yugoslavia, as a re- members join an existing constituency. This innov- sult of extended civil war or genocide. ative arrangement thus combines representativeness with efficiency. The relationship between refugees and the state is inex- tricable. States are major actors in responding to and defining refugee crises. Indeed, international law defines "refugee" in relation to the state. Although NGOs and receiving countries and communities play essential roles but by late 1996 an external refugee population of in providing for displaced persons, the scale of displace- about 1.5 million remained dependent on international ment in recent years has required that statesunilaterally assistance. or within the framework of multilateral organizations States also differ in their ability to avoid or limit mobilize and deliver protection, relief, and assistance, In refugee flows. Stronger states are more effective at denying addition, states acting together, or as members of interna- entry to refugees and asylum seekers. It is often the weaker tional organizations, have initiated the negotiations that states, with the most limited resources, that shoulder the have brought to an end several refugee-generating armed greatest burden in protecting refugees and repatriating conflicts, including those in Cambodia, Mozambique, them when conflicts end. and the former Yugoslavia. Today's international collective response to refugee Nevertheless, the disincentives to cooperate on problems relies heavily on multilateral organizations. The refugee issues are powerful. One is the difficulty in secur- Office of the United Nations High Commissioner on ing commitments in situations where a state may perceive Refugees has seen its budget doubled and its mandate no direct interest of its own. Another is the prospect stretched in the 1 990s. It has provided in-country human- of burden-sharing arrangements that require a state to itarian relief in Bosnia, cross-border operations in Soma- accept refugees into its territory, often at high politi- lia, assistance to internally displaced persons in Sri Lanka, cal and financial cost. The case of Rwanda illustrates and repatriation of refugees in Central America and the high cost of providing relief in large-scale humani- Mozambique. These activities have required highly com- tarian emergencies. Between April and December 1994 plex coordination: in Mozambique, for example, relief the international community allocated about $1.4 bil- operations in 1991 involved twenty-six United Nations lion for relief in Rwanda and neighboring coun- agencies, forty-four bilateral donors, six other multilateral tries. Rehabilitation efforts were gradually introduced, institutions, and 180 NGOs. It is estimated that more 140 WORLD DEVELOPMENT REPORT 1997 than 16,000 NGOs are working to provide relief and Box 8.5 How large the global peace humanitarian assistance worldwide. dividend? Imp roving the effectiveness offo reign aid Global military spending has fallen significantly, Foreign aid is not strictly a public good, but it can be jus- from about 4 percent of GDP in 1990 to 2.7 percent tified by considerations of international equity, particu- in 1994 to 2.4 percent in 1995 (see figure). This larly concerning the future productivity and well-being of drop, in sharp contrast to the rising trend of the pre- people in poor countries. A vital part of improving the vious two decades, resulted from the breakup of the climate for development assistance must be to make for- Soviet Union, a changing global political climate, eign aid more effective from the standpoint of both the increased democratization, and a fall in military aid. borrower and the donor. Recent research suggests that this But has the reduction in military spending im- can be achieved by linking aid more closely to recipients' proved growth and welfare? The relationship be- policies. tween military spending and economic development The success or failure of aid-financed development depends on a variety of factors and circumstances. projects, even in the social sectors, is particularly depen- Empirical results vary, depending on the assump- dent on the quality of a country's macroeconomic poli- tions and methodologies used. Some studies show cies. A project to expand primary education, for example, that global reductions in military spending have is more likely to succeed where macroeconomic policies indeed generated a peace dividend in the form of are sound. If projects are the vehicle for development, faster output growth. Others suggest that the rela- macroeconomic policies can be seen as the fuel and lubri- tionship between military spending and growth is cants that keep the vehicle going. not linear but quadratic: at low levels of military Moreover, only in a good policy environment will for- spending, increased spending contributes to faster eign aid have an impact on growth. In countries that have growth; at higher levels further military spending pursued the key economic policies for growthwhich siows growth. Once the highest-spending nations are empirical research has identified as ensuring fiscal disci- excluded from the sample, the relationship between pline, preventing high inflation, and maintaining a rea- military spending and growth is not significant for sonably open economyforeign aid has significantly in- most developing countries (in peacetime). In such creased economic growth (Figure 8.3). Countries that cases the biggest dividend ultimately may come from have achieved a good policy environment and received a country's perceived security and increased investor significant amounts of aid in recent yearsBolivia, El Sal- confidence, rather than from reductions in military vador, Mali, and Uganda, for examplehave grown faster spending per Se. than would have been predicted by their policies alone. The clear implication is that foreign aid would be more Military expenditure by industrial, developing, and effective if it were either more systematically targeted to former Soviet bloc countries poor countries with good economic reform programs or Percentage of GDP used to promote good policies. At the same time, donors 8 share much of the responsibility for ensuring that foreign Former Soviet Union aid is dispensed responsibly and effectively. 7 and CEE The past decade has seen a trend toward economic lib- 6 eralization in the developing world, indicating an improv- Industrial ing climate for effective assistance. For example, India and 5 countries Vietnampopulous countries that undertook good re- 4 form programs in the early 1 990shave built environ- ments where foreign aid is likely to have a greater impact 3 on growth and poverty reduction. But the track record of 2 targeting aid to poor countries with good policies was gen- - Asia erally weak between 1970 and 1993. Bilateral aid showed 1 Sub-Saharan Africa no tendency to favor good policies, whereas multilateral 0 aid reflected only modest favoritism toward countries 1990 1991 1992 1993 1994 1995 with good policies (as shown in a study that controlled for Note: Data are averages of the countries in each group, income and population). Clearly, a high priority for aid weighted by GDP. Source: Gupta, Schiff, and Clements 1996. agencies is to channel resources more systematically toward poor countries with good policies. FACILITATING INTERNATIONAL COLLECTIVE ACTION 141 Figure 8.2 Refugees have been flooding Africa, Asia, and Europe Millions of persons 12 10 - - 1985 1990 1995 Latin America North America Europe Asia Africa Note: Data include refugees, returnees, and internally displaced persons. Source: Office of the United Nations High commissioner for Refugees 1995. Can aid help poor countries improve their policies and once good policies and good institutional structures are in institutions? This is a difficult but critical question for the place, financial assistance can accelerate the transition to allocation of aid. There has been little systematic research a more rapid growth path. The experience of successful into this question, but the available results are suggestive. economies shows that the need is temporary: as a track Structural adjustment lending to support policy re- record of good policies and performance develops, private form has been more successful where local "ownership" of capital flows increase and gradually eliminate the need for the reform program has been strong. Although adjust- foreign aid. ment lending can provide a useful support to an existing Strategic options: Furthering the provision of reform program, it is not likely to generate reform on its international collective goods own: experience strongly suggests that donors cannot "bribe" governments to introduce policies for which there More effective international cooperation can expand op- is no domestic support. portunities and help nations cope with the new global Where domestic social and political forces have initi- challenges. Each country has to evaluate the merits of each ated programs to reform policies and institutions, foreign proposed cooperative endeavor and decide on a case-by- aid can provide effective support by bringing technical case basis whether or not to participate. But this chapter expertise and lessons from other countries into a receptive has pointed to several areas where cooperation could be of environment. Good examples of such positive interaction great value: are Indonesia, Mauritius, and Uganda. But where there is little domestic movement toward reform, assistance aimed Expansion and preservation of open world markets, at institution building and policy reform has had little including mitigation of the risk associated with volatile impact. capital movements. Many developing countries are con- Thus, in some environments it may be difficult for cerned about more-open capital markets because of the foreign assistance to accomplish anything beyond peace- possibility of sudden outflows that can destabilize eco- keeping and emergency relief. But once domestic social nomic management. and political forces have generated momentum for Basic research directed at the needs ofdeveloping countries. reform, foreign aid can provide important support for The green revolution, made possible by the support both policy reform and institutional development. And of the CGIAR, shows that investing in research and 142 WORLD DEVELOPMENT REPORT 1997 mental problems by improving coordination, increasing Figure 8.3 Poor policies nullify the effect public awareness, transferring technology, and provid- of aid ing incentives for appropriate national environmental policies and enforcement. Marginal impact of aid on GDP growth Peacekeeping and the prevention of armed conflicts. The (percentage points high human and financial cost of warsand of associ- per year) ated relief and rehabilitation effortsis well known, but existing mechanisms have had little success in pre- 0.6 venting conflicts or in resolving them before they 0 become large-scale human tragedies. Improving the effectiveness of foreign aid. Linking aid more closely to recipients' policies can make aid more effective: for any level of foreign aid available to a coun- 03 try, economic performance rises with the quality of pol- icy and governance. Recipients' policies appear to have 0.2 influenced the allocation of multilateral but nor of 01 bilateral aid. Low The appropriate catalyst for greater cooperation will Good Medium vary, both according to the goal and according to the Average Poor High range of countries likely to participate. In several areas, Policy Level of aid new functional or regional groups may be useful in help- ing coordinate and enforce more effective voluntary col- Note: Each value is an average for a group of countries. Results laboration. These groups can seek to develop common are based on a regression using data from fiftysix countries for the period 1970-73 to 1990-93. The level of aid is measured rules and mechanisms for pursuing designated objectives. as a percentage of the recipient countly's GDP. Source: Adapted But the credibility and effectiveness of any such effort will from Burnside and Dollar 1996. rely critically on striking the right balance among the competing values of openness, diversity, and cohesion. They will rely, too, on the political incentivesand the commitmentof participants. Attempts to improve the development can bring rich returnsfor both the effectiveness of international collective actions, like simi- donors and the intended beneficiaries. lar domestic efforts, will bear fruit only if leaders are will- Protection of the environment. International collective ing not merely to promise change, but to take the steps action can help mitigate both global and local environ- required to deliver it. THE CHALLENGE OF INITIATING AND SUSTAINING REFORMS The innovator makes enemies of all those who prospered under the old order, and only lukewarm support isforthcomingfrom those who would prosper under the new. Niccolè Machiavelli, The Prince (1513) CHAPTERS HAVE SHOWN HOW THE STATE temporary. Thus, radical reforms have often been under- PRECEDING can improve its effectiveness by matching its role taken in response to an external threat or economic crisis, more closely to its capability, and working to enhance that or during the "honeymoon" period of a new administra- capability over time. Integral to this approach is a better tion or regime, when incumbents with a strong vested understanding of why some countries do not meet even interest in the old system have been displaced. the most basic requirements for sound economic manage- The second lesson is that, given such an opportunity, ment, and why so few developing countries have managed reformers can make the best use of the time available to create effective state institutions. But understanding by adopting a strategy that understands the likely obsta- the problem and fixing it are two very different things. cles and seeks to mitigate them. Tactically designing and The basic questions remain: why and how is it that some sequencing the reforms can help, as can measures to make countries have been able to throw off this legacy of failure institutions less susceptible to capture by special interests and move forward with reform, while others have not? and gridlock, and, perhaps most important, building a This chapter looks for some answers to these questions consensus in favor of reform. by examining the main obstacles to reform and how they Finally, the message of the many reform successes can be overcome. Three factors turn out to be critical: the and failuresanalyzed in this chapter is that break- distributional characteristics of reform (the likely winners throughs will rarely happen by accident. At any given and losers), the political strength of key groups (particularly time, the forces favoring the status quo are likely to pre- those that will lose out), and the design of existing state vail. Reforms only succeed if they are directed by leaders institutions. Sometimes a reform will be politically unde- with a clear vision of the way things could be, and a con- sirable because the likely losers are part of the political lead- tagious determination to turn that vision into reality. ership's support base. Even when the political will to Obstacles to reform change is present, reformers can find their efforts derailed by constraints embedded in state institutions, which make The obstacles to reform in any country will be many and it easier for opponents to maintain the status quo. varied. The recipe for reform failure is no more suscepti- But the fact that opposition to reform can be rooted ble to easy generalizations than the recipe for success. But deeply in a country's institutions need not be a counsel of chief among the barriers to change will always be the despair. To the contrary, close examination of the imped- powerful interest groups who stand to lose y it. Resis- iments to reform yields three pieces of practical advice for tance will be even stronger when the prospective losers are reformers. The first is that windows of reform opportu- among the political leadership's core constituents. In nity do open; they tend to be those times when the nor- short, the redistributive effects of a reform and the politi- mal rules of the game are in flux for some reason, however cal strength of groups affected by it may simply render THE CHALLENGE OF INITIATING AND SUSTAINING REFORMS 145 some policy changes politically undesirable. Yet even ent groups, whose precise composition will depend on the politically desirable reforms may fail because of con- reform in question. For example, public sector reform, straints embedded in state institutions, which tilt the play- which is central to reinvigorating state institutions, can ing field firmly in the opponents' favor. Thus, policy out- sometimes be thwarted by civil servants who run the risk comes can usually be seen as the combined effect of the of unemployment or finding themselves worse off in pri- characteristics of the reforms themselves, the political vate sector employment. Politicians who use public em- strength of different actors, and the design of existing state ployment as a source of patronage may also see an interest institutions. in blocking certain kinds of reform. Decentralization, for instance, raises the prospect of reallocating resources out- Distributional conflicts, uncertainty, and reform side the political leadership's constituency. In Peru, a de- Some common types of reform and the groups that stand centralization program that would have transferred re- to gain or lose from them are listed in Table 9.1. These sources for financing primary and secondary education to alignments do not apply in all circumstances. But the provincial municipalities was halted in 1993, following truth remains that resistance to reform is often triggered widespread victories by independents and opposition par- by the potential redistribution of resources among differ- ties in the municipal elections. Table 9.1 Alignment of interest groups, political costs, and tactical sequencing of reform by reform type Interest groups Determinants of Tactical Type of reform Against For political cost sequencing Other issues Trade Holders of import Consumers, ex- Redistribution )+) Reduce quantitative liberalization quotas porters, the treasury Efficiency gains (-) restrictions before Protected (if revenues will tariffs. industrialists increase) Pension Trade unions Employers Wealth reduction )+) Allow participants to Young workers may privatization Pensioners Financial institutions Reduced coverage (+) opt Out of public be willing to forgo associations Young workers Older median voter (+) scheme, then phase some of their Bureaucracies (labor Efficiency gains )-) it Out. acquired rights. ministry, social security agency) Decentralization Functional Top officials and Top officials and staff Redistribution )+) Build consensus, Need to mitigate staff in central in local administra- Political contest- phase in pilot fiscal imbalances and administration tions, consumers, ability )+) program, design grant design new schemes citizen beneficiaries, Efficiency gains (-) schemes. for allocation of local businesses grants across jurisdictions Political Top decisionmakers Local decisionmakers in political parties in political parties, associations, and labor unions; NGO5; taxpayers Fiscal Top officials in Department of finance ministry of finance ri local administra- and strategic planning tion, local planning (or public investment) and investment units commission Public sector Employees and Private business, Redundancy (+) Eliminate ghost Incentives: reform managers of public rural elite, central Unemployment (+) workers, encourage severance, buyouts, enterprises, agencies, taxpayers Relative wages )+) voluntary and early capitalization politicians prone to Efficiency gains )-) retirement, ensure scheme, training, patronage retrenchment without private sector revolving door. placement, credit schemes Note: A plus sign indicates a factor that increases, and a minus sign one that lowers, the political cost of reform. 146 WORLD DEVELOPMENT REPORT 1997 Ranking policy changes according to their political erations can also affect policy outcomes. Most public pen- costs and benefits can help policymakers design the tacti- sion schemes are unfunded, financed by current payroll cal sequencing of a comprehensive reform program. taxes rather than past contributions. The resulting high Although such an exercise is highly country specific, a marginal tax on labor and the weak link between contri- good starting point is to compare the expected redistribu- butions and benefits create distortions in the labor market. tional effects of planned reforms with their expected effi- In mature systems these distortions are compounded by ciency gains. Some reforms, for example, are difficult to the low implicit return on pensions relative to the market implement because in the short run they seem to be merely return on capital. The distortions could be reduced by, reshuffling opportunities and incomes. Although such among other things, strengthening the link between con- reforms, if they increase efficiency, ultimately increase the tributions and benefits and privatizing and funding pen- size of the pie, their short-run redistributive effects may sion schemes. But such reforms would affect different gen- exceed even the immediate benefits. In that case, other erations of workers differently. For example, privatizing things equal, reform will not occur, because the political and fully funding the U.S. Social Security system could difficulties of reforming outweigh the rewards. This polit- generate net gains, including gains in efficiency, but ical cost-benefit approach can be applied to a broad range younger workers would realize most of those gains, and of reforms, such as trade liberalization (Box 9.1). older workers would lose (Figure 9.1). This dilemma helps Applying the redistributional calculus to the case of explain why reform is so politically sensitive. Similarly, pension reform shows how conflicts of interest across gen- redistribution away from the elderly helps explain why countries in CEE and the CIS are resisting increasing the pension age. In Ukraine, for example, a uniform pension age of 65 would ease the system's actuarial imbalance but Box 9.1 Weighing the political costs and benefits of reform would also reduce the pension wealth of workers (in pres- ent value terms) by about 25 percent of GDP. Efficiency-enhancing reforms are often difficult to The difficulty of reforming universal pension programs implement because they create both winners and thus stems from the anticipated reform's redistributive losers, and there may be no way to compensate the effects and the power of elderly voters. Meanwhile unborn losers. Defusing opposition is even more difficult generations, who would benefit the most from reform, when the efficiency gains are low relative to the have no voice in the decision. The political costs of reform redistributive effects. Applying a rough political in the United States have risen over time, as the gap in cost-benefit ratio to reform measures can show how voter turnout between the young (those between 25 and much redistribution takes place for a given amount 40 years of age) and those over 65 has widened in favor of of efficiency gain. A policy that increases the income the latter, to some 12 percentage points. Clearly, reform- of one group without taking income away from ing public pension and health care financing programs for another, for example, would have a ratio of zero. the elderly, however fiscally unsustainable, is a difficult For trade liberalization the political cost-benefit feat to pull off, but should be a high priority. Even when ratio is inversely related to the tariff rate, the share one-time, comprehensive reform is not feasible, more- of imports in total consumption, and the elasticity gradual changes and grandfathering of existing beneficia- of import demand. In Sub-Saharan Africa the aver- ries, in the recognition that changes are likely to be gener- age import tariff exceeds 30 percent, and the share ational, with long lead times, may reduce opposition to of imports in total consumption is about 40 per- reform. cent. Assuming an import demand elasticity of two In many countries utilities are inefficiently run public (demand for imports rises by 2 percent for every 1 monopolies. Consumers would realize significant gains if percent decline in price), the cost-benefit ratio of these utilities were privatized and effective regulatory trade liberalization is more than four. Thus, for any agencies established to watch over them. Witness the case given amount of efficiency gain, the redistriburive of Argentina, which began privatizing its state-run infra- effects would be four times that amount. When a structure services in 1989. All income groups have bene- reform program combines trade liberalization with fited from the efficiency gains induced by privatization, a stabilization program that increases output, the and these gains (relative to spending on utilities) have ratio declines significantly. A stabilization program been similar across income groups (Table 9.2). In that increases GDP growth by 1 percentage point Uruguay, for example, a 1989 plebiscite rejected privati- would be sufficient to lower the ratio from four to zation legislation. Yet a recent study shows that inefficien- less than one. cies in public utilities add 30 percent to the average Uruguayan's electricity, water, and telephone bills. And as THE cHALLENE OF INTIATINO AND SUSTAININt3 REFORMS 149 can be used to block reform, we focus here on two princi- Figure 9.1 Older workers will lose from pal ones: the electoral and party system and the system of reforming pensions, but the young will gain checks and balances. The point is not that institutions should be redesigned and changed frequently to facilitate Projected net gains from pension privatization in the reform, or that a single design is desirable for all countries United States (thousands of dollars per contributor) and situations. The aim, rather, is to show how elements 10 of the underlying institutional framework can condition 8 both the attempt to reform and the response. ELECTORAL AND PARTY SYSTEMS. As has been empha- 6 sized throughout this Report, institutional choices are sel- 4 dom clear-cut: they will involve a careful tradeoff between allowing state officials flexibility and imposing appropri- 2 ate restraints on them. Experience with proportional elec- 0 toral systems is a case in point. Such systems are associated with coalition governments, which can be desirable to the 2 extent that they bring more voices to the cabinet table and put a high value on consensus. Yet the same characteris- 4 tics can also be a barrier to reform, leading to long delays 6 in policymaking and higher fiscal deficits because of the 5 10 15 20 25 30 35 40 45 50 55 60 need to buy off sectoral or regional interests. Research has Age (years) shown that countries with large and fragmented coalitions tend to have more difficulty adjusting to external shocks, Note: Data are aggregate lifetime gains or losses, at 1995 present value using a 5 percent discount rate. Source: Feldstein such as the 1973-74 oil price hikes. Both Belgium's and and Samwick 1996. Italy's vety high levels of public debt have been partly attributed to two decades of being governed by large and unstable coalitions. Brazil provides another illustration of how electoral we saw in Box 4.2, many of the arguments commonly and party systems can interact with economic policy. The offered against privatization are not valid. social security bill proposed by the administration of Pres- So why is privatization still resisted in some countries? ident Fernando Henrique Cardoso in June 1996 was This reluctance to implement welfare-improving reforms defeated in the lower house of the legislature, despite a is linked to at least three factors: formal majority in favor of the government alliance, be- cause certain interest groups (civil servants and teachers, The perceived uncertainty of the outcome of reform, among others) exploited constitutionally protected privi- which impedes the creation of a strong constituency for leges and a political system that discourages stable voting reform and raises concerns that the immediate response majorities in the congress. The fact that deputies belong- may be social unrest, while the benefits are only realized ing to the alliance voted against the bill reflects the later The fact that private operators typically have to make changes that are detrimental to certain groups in order for the efficiency gains to materialize Table 9.2 Estimated efficiency gains from The fact that different groups may hold conflicting privatizing utilities in Argentina views about the role of the statefor instance, in many Gains per dollar countries that have previously relied entirely on state Efficiency gains of expenditure utilities, many groups continue to resist privatization Income (millions of 1993 on utilities on ideological grounds. quintile dollars) (percent) Poorest 205 30 Second 222 27 Institutional design Middle 342 34 The preceding discussion has identified the losers from Fourth 335 27 reform as potentially powerful obstacles. But whether or Richest 549 31 not these groups prevail in a given instance will often be Total 1,653 30 determined by the design of state institutions. Rather than Source: Chisari, Estache, and Romero 1996. attempt an exhaustive account of how state institutions 148 WORLD DEVELOPMENT REPORT 1997 unusual autonomy of elected officials from political par- larly, as Chapter 7 showed, an ill-designed decentraliza- ties that characterizes Brazil's system of proportional rep- tion that leads to capture of local authority by special resentation. A 1991 study of Brazil's electoral and party interests can prevent the adoption of sound reforms. system found that representatives had belonged to an Although the presence of many veto points can some- average of three political parties, and that in 1987-90 times contribute to delays, some evidence suggests that it one-third of the 559 representatives had switched parties has helped contain the expansion of the welfare state. As since being elected in 1986. A bill to reform some aspects Figure 9.2 shows, the constitutional division of power of party legislation may be voted in the congress in 1997. ranks second only to aging in explaining changes in wel- In Uruguay a number of institutions have speeded fare spending. As incomes per capita increase, demand for some reformsand held back others. One of the pecu- government transfers also rises. Countries with fewer veto liarities of the country's electoral system, prior to the points in their state structure (such as Sweden and Den- recent reform, was that party primaries and general elec- mark) are more receptive to these demands. By contrast, tions were held simultaneously. As a result, the winning the greater number of veto points in Switzerlanda fed- presidential candidate received only a minority of the total eral state with a bicameral legislaturehas blocked many vote and had to build alliances with opponents in the par- initiatives to expand welfare programs. Reforming the liament. In the November 1994 election the winning can- welfare state will therefore require going beyond stream- didate received only 24 percent of the vote, and the three lining the functioning of transfer programs. Indeed, this main parties each received more than 30 percent. Such an was recognized by the 1992 Swedish commission on state electoral system tends to be candidate centered and con- reform, which recommended strengthening the executive ducive to faction. Hence, groups that are able to mobilize by introducing a constructive vote of no confidence, politically reap the most benefits. Another distinctive allowing the government to ask parliament to vote on feature of Uruguay's political system is its heavy reliance measures as a package rather than line by line, extending on direct democracy (through plebiscites) to decide on the time between elections from four to five years, and features of the public pension system. In 1992 voters reducing the size of the parliament by half. reversed major privatization legislation. And a 1989 POLITICAL REGIME. The choice of political regime has plebiscite, initiated by the association of pensioners, guar- justifications that go far beyond economic conditions. Yet antees full wage indexation every three months. These the strong links between state institutions and policy out- institutional designs help explain why pension expendi- tures relative to GDP are about 35 percent higher in Uruguay than in the United States, even though the two Figure 9.2 Multiple veto points help countries countries have roughly the same proportion of elderly in resist pressure to expand welfare the population (16 percent). Uruguay has since recognized that its electoral system Contributors to welfare expenditure is an impediment to a well-functioning state. A new sys- tem was approved by the parliament in October 1996. It Higher Type of unemployment government would end simultaneous voting for primaries and general 13% 4% elections, and it would require a second-round ballot between the two presidential front-runners when neither Z More efficient bureaucracy has managed to secure 50 percent of the vote. These changes are expected to strengthen party discipline and deter factionalism. CHECKS AND BALANCES. Chapter 6 showed how and why inadequate checks and balances can lead to arbitrary Constitutional Aging of Higher GDP government decisions and behavior. The veto points arise structure with population per capita fewer veto points at three levels: the separation between the legislative and 51% 8% 15% the executive branches, the division of the legislature into separate chambers, and the division of power between national and subnational governments. A bias toward the Note: Each slice represents the estimated contribution of status quo develops when the state supports many institu- that factor to observed government expenditure on pensions, unemployment benefits, and family assistance. Results are tional veto points, and groups that oppose change can based on a regression on data from twenty-two OECD countries exert power at one or more levels. In a presidential system, for the period 1965-93. See the Technical Note for details. for example, gridlock can emerge when different parties or Source: World Bank staff calculations. coalitions control the executive and the legislature. Simi- THE CHALLENGE OF INITIATING AND SUSTAINING REFORMS 149 Box 9.2 The predatory state under the Duvalier dynasty in Haiti Haiti became independent in 1804. A predatory state tion. Following a business strike in 1957, the police arose there between 1843 and 1915, a period charac- were authorized to open the shops of striking mer- terized by short-lived rulers, often removed by coups, chants and distribute their merchandise. Significant driven by the creation of personal wealth, Of twenty- resources were devoted to protecting Duvalier himself: two administrations during that period, eleven held 30 percent of total expenditures during the first half of office for less than a year, and only one managed to the 1960s. Agriculture, particularly coffee, was heavily complete its term. taxed. Some sources estimate that Duvalier transferred The United States occupied the country from 1915 more than $7 million a year out of Haiti for personal to 1934, but the logic of a predatory state remained purposes. Large-scale bribes also took place, through unchanged. In 1957 the democratically elected govern- deals with foreign investors on projects that often ment of François (Papa Doc) Duvalier took that logic never materialized. Extortion under the veil of "volun- to a new level, starting with an unprecedented purge of tary" donations was institutionalized under the Mouve- civil society, the inherited army, political opposition, ment de Renovation Nationale. A pseudo-old-age and other branches of government. Within two pension with a 3 percent deduction was created, and months of coming to power, Duvalier had jailed 100 government employees were forced to buy a $15 book political opponents. The Catholic Church was per- containing Duvalier's speeches. An autonomous gov- ceived as a threat, and spiritual leaders were expelled. ernment fund collected taxes and levies, which were The mass media were silenced through the expulsion of excluded from the budget, and no accounting was foreign journalists, and a 1958 code allowed the gov- made of their use. ernment to shoot reporters charged with spreading After nearly thirty years of rule, the Duvalier "false news." Imprisonment of the parents of striking dynasty fell in 1986, when Jean-Claude (Baby Doc) students was made mandatory. After lifting parliamen- Duvalier, who had inherited the presidency from his tary immunity in 1959, Duvalier dissolved both the father, went into exile in France with an estimated $1.6 Senate and the Chamber of Deputies. Modern military billion at his disposal. Haiti's history as a predatory equipment was stored in the basement of the presiden- state goes a long way toward explaining its dismal eco- tial palace, and more than 200 officers were removed in nomic performance. During 1965-90 growth in GNP Duvalier's first eleven years of power. In 1964 Duvalier per capita averaged 0.02 percent, and social indicators declared himself president for life. remain the worst in the Western Hemisphere. Given The economic pillars of Haiti's predatory state were the legacies of the predatory state, Haiti's history expropriation, extortion, the inflation tax, and corrup- remains perhaps the biggest obstacle to change. comes raise the question of whether obstacles to reform The end of the Cold War, combined with pressure may be embedded in the political regime. Some observers from citizens, should lessen the risks of extreme state cap- have argued that nondemocratic regimes, by having fewer ture embodied in a predatory state, with many countries veto points, are more conducive to economic development. now having adopted elements of democratic regimes (such The reality is more complicated: no single type of regime as free and open elections). But researchers have yet to can guarantee economic and social progress. We do know, reach a consensus on the precise relationship between however, that one kind of regimethe so-called predatory growth and democracy: about one-fifth of the studies find statecan be almost guaranteed to produce economic a positive relationship, one-fifth a negative relationship, stagnation. The focus of such a state is on the extraction of and the rest are inconclusive. The analysis of the determi- economic rents from the citizenry by those in power. It nants of growth summarized in Chapter 2 found no statis- does so by specifying property rights in a way that maxi- tically significant correlation between the two. And cer- mizes the revenue of the group in power, regardless of the tainly, economic performance among developing countries impact on the wealth of the society as a whole. Haiti under classified as sustained democracies has varied considerably. the Duvalier regime (Box 9.2) and Romania under Nicolae The experience of countries that have combined polit- Ceausescu ate prime examples. A predatory state is incon- ical transformation with the transition from a planned to sistent with economic development because it discourages a market economy suggests equally mixed conclusions productivity and leads to misallocation of resources, culmi- with regard to the relationship between democracy and nating sometimes in the collapse of the state itself. reform. As noted in Chapter 7, using the ballot box to 150 WORLD DEVELOPMENT REPORT 1997 punish or reward politicians for past performance (retro- spective voting) could be a powerful way to ensure Box 9.3 Reform under external threat: accountability and good policy outcomes. But the road The Meiji restoration in Japan can be rocky at first. Indeed, governments' initial reaction to greater political contestability may add to the difficulty The intrusion of Western powers into national of improving institutions, and transitions to democracy commerce was the impetus for reform in Japan. are sometimes associated with increased budget deficits Starting in the 1840s, Japan was under increasing and inflation. pressure from the West to grant commercial and Sub-Saharan Africa has witnessed about twenty-seven trade privileges. The Tokugawa leadership was elections since 1990, twenty-one of which were in coun- aware that Japan's technological and military tries holding elections for the first time. Young democra- prowess lagged, and in 1854 Japan had little choice cies are not immune to the electoral cycle. Before Ghana's but to accede to Commodore Perry's demand that 1992 elections, the government increased outlays and the some ports be opened to American ships. Similar wage bill at the cost of macroeconomic stability and sub- arrangements with other foreign governments fol- sequent inflation. In Sub-Saharan Africa the numbers of lowed. By 1865 Western powers had restricted ministerial positions and legislative seats have increased Japan's ability to levy import tariffs: the highest rate by 22 percent during the political transition that began was set at 5 percent. In 1868 a coalition of feudal in 1989, The governments of Cameroon, Malawi, and lords overthrew the Tokugawa family, which had Senegal all have more than thirty ministers each. A careful ruled Japan for more than two centuries, and prioritization of policy issues is difficult to achieve in such replaced it with a leadership that would modernize an atmosphere. Bolivia, for instance, has responded by Japan and transform it into a country better able to putting legal restrictions on the size of the cabinet: only face up to foreign threat. This so-called Meiji two additional ministries may be created, and those on a restoration marked the beginning of modern temporary basis. These experiences suggest that states need Japan's economic growth. skill to manage the political transition in such a way that Reforms following the restoration transformed it does not impede the development agenda. the Japanese state and society. The class system was When and why do countries reform? abolished, a new system of local and national gov- ernment was set up, and conscription into the army Insights about the circumstances under which reforms are and the navy was instituted. To establish the basis likely to succeed are as useful as insights about the obsta- for a sound fiscal system, the government under- cles to reform. Indeed, the two are related. If circum- took a land survey, established titles, and imple- stances are conducive to reform, the first step is to alter mented a land tax payable in cash. Education was the dynamics that created the status quo. The following gradually made compulsory, so that by the mid- sections describe how an external threat or an economic 1880s almost half of school-age children were crisisreal or perceivedmay override resistance to attending school. The Bank of Japan (the central change. They may. But they have not always done so. It bank) was created, and bureaucratic reforms were remains a puzzle why some countries reform in such undertaken, laying the ground for the recruitment extreme circumstances and others do not. of state officials based on merit, not patronage. Moreover, the new regime undertook initiatives External threat that today would be called industrial policies: it A growing external military threat has often triggered established and operated factories (silk, brick, glass, reform. Until recently, a country's lagging technological cement, textiles, shipyards), subsidized industries, and economic performance became evident only during imported technicians, and sent students abroad. wartime. In the 1700s and 1800s the leadership of the Ottoman Empire reorganized the military and made broad reforms in education and governance, in response important demonstration effects. Chile's economic success to battlefield losses to European powers. Similarly, the clearly inspired other Latin American countries to under- Meiji restoration in Japan in 1868 was motivated by a take reform in the late 1980s, as did the success of Japan desire to strengthen the state against the encroachments of and, later, that of the Republic of Korea and Taiwan Western powers (Box 9.3). (China) in East and Southeast Asia. Economic reform in Today, military confrontation plays a smaller role in China can be explained by many factors. But among them driving reform. But the perception that a country is lag- was the demonstration effect of its neighbors' economic ging behind its neighbors economically often leads to success and an unwillingness to be left behind. THE CHALLENGE OF INITIATING AND SUSTAINING REFORMS 151 Economic crisis and problems of social mistrust. A balance must be struck Since the early 1980s, economic crisis has been by far the between clear rules that circumscribe the freedom of state most important factor driving the introduction of ambi- officials to act opportunistically, and the need for them to tious reforms. As the failure of prevailing policies becomes act flexibly and responsively. An effective state operates widely recognized, popular demands for reform become with clear and transparent rules, yet is quick to exploit more vocal, and politicians more willing to risk radical opportunities and to reverse course when circumstances change. Economic crisisparticularly hyperinflation and demand it. deep recessionpreceded economic reform in, for exam- The obstacles to building an effective state are not insur- ple, Indonesia in 1961-64 and Peru in 1990. Yet other mountable. Change has a better chance of success when countries in dire economic straits have failed to take cor- policymakers do three things: tactically design and rective measures, while several countries have not needed sequence reforms, compensate losers, and build consensus. a deep crisis to spur reform. Economic crisis cannot be We take up each of these in the following discussion. But credited with stimulating reform in Australia (1983), in some cases institutions must be modified to make the Colombia (1989), or Portugal (1985). state function better: it is no longer a question of tactics but Economic crisis and civil conflict often feed on one of fundamental reform. One lesson is clear: all these another, leading to near disintegration of the state (for changes are a good deal more difficultimpossible, even example, in Liberia and Somalia). Such crises have in the absence of leaders with a clear vision of the future. immense social costs and offer little hope for quick reso- lution, for state capability matters when it comes to turn- Tactical design and sequencing ing a crisis into an opportunity. Yet political leadership lactical design and sequencing of reform can improve the and political entrepreneurship are also important. And chances of success by recognizing the constraints of exist- often political leaders who can convert opportunities for ing state capabilities, diluting resistance to change, and economic gain into reality can capitalize on the benefits of building a constituency in favor of reform. successful reform. MATCHING ROLE TO CAPABILITY. As emphasized Even in time of crisis, incumbents are often more reluc- throughout this Report, a good match between the state's tant to embrace reform than are newcomers. Thus, a role and its capability is the key to effective policies. A change in government in the middle of an economic crisis mismatch between capability and action can compromise (as in Peru and Poland in 1990) may provide the added the sustainability and effectiveness of reforms even in the impetus needed to get reform started. Terrorism, hyper- absence of political obstacles. Regulatory reform inflation, and the poor performance of traditional parties whether of antitrust, environmental, or financial regula- in Peru all gave the new president, Alberto Fujimori, room tionshould reflect institutional capabilities (see Table to maneuver. On the other hand, Colombia's reforms took 4.2). For instance, price-cap regulation, in which the reg- place in 1989, toward the end of the administration of ulator sets the adjustment factor used in determining a Virgilio Barco, when the economy was not in crisis. So monopoly utility's prices, is more suited to countries with honeymoon periods and economic crisis provide an oppor- relatively strong institutions. Similarly, mechanisms to tunity, but not the only opportunity, for embarking on improve the delivery of services have to take into account reform. More important, even in countries where eco- not only the characteristics of the service but also the nomic crisis has triggered reform, the depth of reform has capability of the state (see Table 5.1). Within the core tended to be modest. Lasting improvements in economic public sector (in education and health care, for example), performance have often remained elusive. Economic crisis using performance-based agencies and formal contracts may provide an opportunity to go beyond stabilization, requires an institutional capability that many develop- but whether or not countries seize the opportunity de- ing countries lack. In these cases such institution-heavy pends on the redistributive effects, the state's initial capa- approaches are unworkable. Capability constrainrs should bility, and political leadership. Deep institutional reforms also loom large in the choice of a decentralization strategy are unlikely to result from a reform agenda triggered and (see Table 7.5). Where capability is low in both cen- driven by crisis alone. tral and local government, a cautious decentralization strategy with pilot testing is probably the best course. But, Implementing and sustaining reform as Part Three of this Report showed, capability is not Reform of the state means not only reform of policies but destiny. And the return to improving capability is signifi- also an institutionalization of good rules of behavior for cant indeed. government agencies. Institutions must be created that STRATEGIC SEQUENCING AS A FIRST STEP TO IMPROVING help avoid heavy discounting of the benefits from CAPABILITY. Where administrative cap ability is weak, rather reforms, paralysis due to unfamiliar new circumstances, than attempt comprehensive reforms of all institutions, 152 WORLD DEVELOPMENT REPORT 1997 countries might do better to reform key functions and The strategic enclave approach allows countries to agencies selectively. Besides being consistent with fiscal adopt the first generation of reforms (Table 9.3). These and human resource constraints, this approach has two reformswhich can mostly be enacted through executive advantages. First, it allows reformers to learn from the ordertypically involve stabilization and selected struc- unavoidable mistakes associated with institution building. tural reforms. But an excessively narrow approach to en- Second, by beginning with the most promising agencies, claving can impede the deeper institutional changes reformers can count on demonstration effects for the rest required by the second generation of reforms. And prog- of the public sector. These two advantages increase the ress in the social sectors has generally been modest. For likelihood of achieving a series of success stories that will example, in Ghana, one of the pioneer reformers in Sub- maintain political support of an overall reform program. Saharan Africa, health expenditures have become even Many countries have used this strategic approach, more regressive since reforms began. Deep institutional beginning reform with a few critical enclaves. Candidates reforms take time and are complex, and opposition by for early treatment typically include the ministry of interest groups (for example, the teachers union in finance, the central bank, and the tax collection agency. In Colombia) is often strong. For countries trapped in the both Peru and Ghana, for example, very low tax revenues first generation of reforms, sustainable long-term develop- forced the government to make swift changes in tax col- ment will be elusive. lection. Yet any country following such a strategy must Countries can implement a strategy that enables them ensure that the assignment of policy responsibilities to escape the enclave trap. Such a strategy involves, above among agencies and ministries takes into account where all, agreeing on clear rules for the conditions under which the relevant professional expertise is concentrated, and agencies outside the original enclave will be brought that mandates are broadly consonant with the public within the reform program. These rules provide a bridge interest at large. For instance, a tariff regime that is man- between the first- and second-generation reforms, while aged out of the finance ministry is likely to emphasize rev- mitigating animosities from agencies outside the enclave. enue goals over industry protectionthe priorities would The Bolivian civil service reform has rightly moved in that probably be reversed if the ministry of commerce were in direction. Embedded in the 1990 Financial Administra- charge. Transforming quantitative restrictions into tariffs tion Act and associated agency-specific regulations are a would typically receive more support from the treasury series of rules governing which entities are eligible to (see Table 9.1). The appropriate assignment of policy recruit highly paid civil servants, what conditions these responsibilities can help sustain reform by influencing entities have to meet in order to qualifr, and what is what gets approved and in what order. expected from them once they have implemented the Table 9.3 First- and second-generation reforms First generation Second generation Main objectives Crisis management: reducing inflation and Improving social conditions and competitiveness, restoring growth maintaining macroeconomic stability Instruments Drastic budget cuts, tax reform, price Civil service reform, labor reform, restructuring of liberalization, trade and foreign investment social ministries, judicial reform, modernizing of liberalization, deregulation, social funds, the legislature, upgrading of regulatory capacity, autonomous contracting agencies, some improved tax collection, large-scale privatization, privatization restructuring of central-local government relationships Actors Presidency, economic cabinet, central bank, Presidency and cabinet, legislature, civil service, multilateral financial institutions, private financial judiciary, unions, political parties, media, state groups, foreign portfolio investors and local governments, private sector, multilateral financial institutions Main challenge Macroeconomic management by an insulated Institutional development highly dependent on technocratic elite middle management in the public sector Source: Adapted from Naim 1995. THE CHALLENGE OF INITIATING AND SUSTAINING REFORMS 153 reform. Yet Ecuador's experience with a similar reform reduction in rates, which was supposed to be accompanied strategy illustrates that it requires a firm commitment to by a removal of tax holidays. But the powerful agricultural the program to ensure that these rules are implemented in lobby blocked the lifting of exemptions, and an attempt in practice. Two months after finalizing the rules governing 1993 to introduce a tax on rich farmers was circumvented administrative reform, the secretary in charge declared all by raising the exemption ceiling tenfold. By contrast, even central administration entities "restructured" and granted a revenue-neutral tax rate reduction combined with a salary enhancement, although none had met the eligibil- broadening of the tax base would have cut distortions ity conditions. Rules can underpin a commitment to without creating losers. A back-of-the-envelope estimate of reformbut they cannot substitute for it. the benefit of reduced distortions suggests it would be GRADUAl. PHASING OUT. Matching the state's role to upward of 1.4 percent of GDP (see the Technical Note). its capability and going beyond enclaving sometimes EFFICIENCY VERSUS TACTICAL SEQUENCING. The opti- require replacing a public agency with a private one. This, mal sequencing from an efficiency standpoint may not be in turn, may call for a two-stage strategy to circumvent politically feasible. For example, efficiency considerations resistance. During the first phase, an opting-out mecha- dictate setting up a credible and stable regulatory agency nism might be put in place, to allow people to switch to before privatizing telecommunications. This sequencing private providers if they want to. Wider recognition of the reduces the risk involved in the purchase, hence increasing benefits of better services can then make it easier to carry the selling price of the company. Argentina, however, did out the second phase: getting rid of the public providers. not follow this sequence. Instead the country's telephone Sri Lanka's 1991 Telecommunications Act shows the monopoly was sold a year before a new regulatory agency benefits of such a strategy. This act created the regulatory was set up. This strategy was adopted to speed privatiza- agency and allowed private operators to compete with the tion and prevent opposition to the reform. Regulatory state monopoly Sri Lanka Telecom (SLT) in value added uncertainty may have reduced the selling price, but the telecommunication services. The legal and regulatory political feasibility of the reform was greatly enhanced. framework has contributed to making Sri Lanka one of And as noted above, the efficiency gains induced by the the most liberalized telecommunications markets in Asia. overall privatization program were significant (Table 9.2). By 1995 there were four mobile cellular operators, five Moreover, countries choosing for political reasons to paging companies, three data transmission service reverse the most efficient sequencing might mitigate the providers, and one Internet provider. At the end of 1995, disadvantages of the lower initial selling price by selling 20 percent of all telephone subscribers were connected to shares in stages, as the reform's credibility improves. cellular services. Competition between the cellular opera- COMBINING AND PACKAGING REFORMS. Introducing tors has led to some of the lowest tariffs in the region, and the right mix of reforms could allow key constituencies to these services are increasingly seen as a credible alternative balance their gains and iosses, hence reducing the political to SLT's wired services. To reduce unsatisfied demand cost of the reform (see Box 9.1). This strategy was pursued quickly, in early 1996 the telecommunications regulatory by New Zealand's Labour government in the 1980s. Min- agency licensed two fixed wireless private operators for ister of Finance Roger Douglas persuaded agricultural basic telecommunications services. These impressive groups that losing their subsidies was essential to the total results have created pressure for better performance by the reform package, which benefited farmers by cutting tar- public telecommunications company. Consistent with a iffs, lowering inflation, and addressing the bias against two-stage phasing-out strategy, the government has exports. Similarly, in Bolivia the broad reform package announced the sale of 34 percent of SLT's equity to a introduced in 1985 by the government of Victor Paz strategic investor. Estenssoro, in a context of hyperinflation, managed to cir- Peru's pension reform illustrates how gradual phasing cumvent labor opposition, which had vetoed previous out can be applied to the social sectors as well. When the reform plans. Although support from the two main polit- reform was launched in 1993, workers were allowed to ical parties helped, the speed and comprehensiveness of choose between public and private pension providers. In the reform prevented pressure groups from organizing to 1996 disincentives for joining a private provider were derail it. removed, leading to a de facto phasing out of the public When deep macroeconomic imbalances have to be cor- scheme. During the second stage a strong constituency in rected, packaging some reforms could increase their politi- favor of the reform was formed, comprising workers cal feasibility. For example, trade liberalization is often eas- who had already shifted to a private provider and pension ier to implement in conjunction with an adjustment fund managers. By contrast, the sequencing of Pakistan's program, because the gains from improved macro- direct tax reform seems to have greatly reduced the chances economic policy (in lower inflation and positive growth) of success. The government began the reform with a can offset the distributive effects of liberalization (see Box 154 WORLD DEVELOPMENT REPORT 197 9.1), Broad reforms may also enhance credibility. In 1990 Tangible benefits accrue to citizens early in the process. the Polish government freed 90 percent of prices, elimi- Beginning in May 1997 each person over 65 will get an nated most trade barriers, abolished state trading monopo- annuity from the capitalization program with an esti- lies, and made its currency convertible for current account mated value of $200 to $225. Bolivia's income per transactions. After an initial decline in output in 1990-9 1, capita, for comparison, was $770 in 1994. the Polish economy has recorded vigorous growth. A concern often raised by opponents of privatization, namely, that it provides an opportunity for corruption, Compensation is somewhat mitigated (whether warranted or not) SEVERANCE PACKAGES. Both matching the state's role to because the state does not receive funds. its capability and enhancing its capability require not only tactical design and sequencing but also the compensation The Bolivian and Czech experiences show how a care- of groups adversely affected by reform, to secure their sup- fully designed privatization program can enhance political port. These groups are not always the poorest in society. feasibility and achieve sustainable efficiency gains, but They may, for example, include bureaucrats whose jobs poorly designed versions of these strategies can backfire. are being eliminated, managers of privatized state enter- prises, and businesspeople used to operating behind high Building consensus levels of trade protection. Although compensation might Reforming the state requires cooperation from all major be economically costly in the short run, it will pay off in groups in society. Deep-seated differences and mutual the long run to the extent that it eases opposition to suspicions among groups can delay or kill reform. There reform. A recent study of retrenchment programs found are no quick fixes for reversing age-old enmities, but social that, on average, the associated benefits in terms of pro- pacts can help. In a social pact, business, labor, and agri- ductivity gains and wage bill savings offset the compensa- cultural interests negotiate the terms of a contract with tion cost after only 1.7 years. Three factors show why sev- government leaders, setting clearly defined responsibilities erance payments can be so important to the success of for each group. This approach has proved successful in reform. First, political feasibility may require that countries as diverse as Spain and Benin. In Spain a minor- retrenchment be done on a voluntary basis. Second, even ity government was able to impose wage restraint by where political constraints are not a factor, the law may bringing all political parties to the negotiating table preclude involuntary separation, as in the case of the Cen- and developing a common program, later known as the tral Bank of Ecuador. Third, most developing countries Moncloa Pacts (Box 9.4). Benin's second democratically do not have an unemployment insurance scheme, and sev- elected government organized a consultation with politi- erance pay then becomes a close substitute. cal parties and civil society on taking office in May 1996. Designing severance schemes that take into account This National Economic Conference created a number of broader characteristics of workers can help make politi- sectoral working groups and made specific recommenda- cally feasible retrenchment less costly and better targeted. tions about the role of the state. Whether this initiative One such mechanism is to set ceilings on the number of will translate into a consensus around an economic pro- departures by skill level. In Argentina, for instance, gram remains to be seen. trained professionals from the National Institute of Agri- cultural Technology were made ineligible for voluntary Leadership and vision retirement packages. It is also important when designing There is no universal blueprint for reform. But almost retrenchment programs to establish built-in mechanisms all successful reform episodes in developing economies that prevent rehiring, which would defeat the purpose. have had one common feature: they have been crafted by EQUITY INCENTIVES. In some cases compensation goes dynamic leaders who shepherded changes through com- beyond generous severance payments: labor, manage- plicated political terrain. Such leaders seize opportunities ment, or the public at large may be given a share of priva- as they appear, but they also create them, by identifying tized enterprises. Bolivia's capitalization program is a very and reaching out to potential beneficiaries, reshaping innovative approach. At least three features of the pro- institutions, and articulating a compelling and achiev- gram increased its political acceptance without compro- able vision for the future. Political leadership is particu- mising on efficiency gains: larly important in countries that lack trust and cohesion among different social groups. When businesspeople mis- By distributing shares to employees and pensioners, the trust bureaucrats, workers are wary of managers, and program circumvented the resistance to privatization farmers are suspicious of everyone, sensible reforms can that often occurs when assets are sold entirely to foreign stall. Leaders must instill a sense of common purpose that companies. minimizes polarization. THE CHALLENGE OF INITIATING AND SUSTAINING REFORMS 155 Box 94 The Moncloa Pacts in Spain Adolfo Suárez was named chief of government by agreements, reached in October 1977, have come to be Spain's King Juan Carlos I in July 1976, early in the called the Moncloa Pacts. Agreement among the polit- transition to democracy following Generalissimo Fran- ical parties proved easier to achieve because the party cisco Franco's death in November 1975. Suárez started leaders were more moderate than the interest groups by instituting political reforms: he recognized free they represented, and because the far left (which had trade unions and the right to strike, legalized all polit- no parliamentary representation) could be excluded. ical parties (including the Communist Party), pro- The pacts went far beyond wage restraint and claimed a political amnesty, and passed a new electoral encompassed provisions on monetary and fiscal poli- law. The first free elections since the Spanish civil war cies and structural reforms as well. They promised a were held on June 15, 1977. Suárez won the elections, "new framework for labor relations" with increased but his newly created party, the Center Democratic labor market flexibility, a more progressive tax system, Union, held only 47 percent of the seats in parliament. and the rationalization and decentralization of public The elections took place as the economy was falling enterprises. Importantly, the pacts contained measures into a crisis. Inflation and unemployment were both to compensate workers for some of the costs that on the rise, and the external balance was deteriorating adjustment was expected to impose on them. Increased rapidly. As part of its anti-inflationary drive and its state spending on job creation and unemployment policy of external liberalization, the Suárez government insurance, the progressive extension of unemployment sought to reduce wage growth. But rather than con- insurance to all the unemployed, and some price con- front the labor movement and the trade unions, Suárez trols were among the compensatory measures. Given took a consensual approach to incomes policy. Peak that Spain already had the highest level of worker con- business and labor associations were brought together flict in all of Europe, getting labor to go along with in late summer 1977 to forge a common position. fundamental reforms without these concessions was a These attempts failed, however, because of deep divi- major achievement. sions within business and labor. Labor, for example, Although not all the promises made in the Moncloa was represented by officials from at least four distinct Pacts (especially those in the areas of labor market political leanings. No one wanted to appear to be mak- reform and public enterprises) were fulfilled, the pacts ing concessions. were successful in achieving their primary targets. Price Suárez then changed strategy and sought agreement and wage inflation was sharply reduced after 1977, and among party rather than class leaders. The resulting the current account gap was closed. The purpose of reform is to enhance economic well- contrast, Prime Minister Mahathir Mohamad's policy ini- being. The consequences of reform are often measured tiatives in the 1990s were packaged in his Vision 2020, using quantifiable yardsticks, such as national income, which set the eye-catching target of raising Malaysian living exports, or inflation. But an equally important aspect of standards to industrial-country levels by 2020. reform is whether it succeeds in reshaping the values and Strategic options: Finding the route to reform norms of the state and the state's relationship to the econ- omy. It is this transformation that ultimately legitimates Machiavelli rightly recognized that distributional conflicts reforms in the public eye. Thus political leaders must offer lie at the heart of the difficulties of reforming the state. Yet a compelling vision, beyond the dry realities of economic such conflicts, and the constraints embedded in state insti- efficiency, about where their societies are headed. Such a tutions which can exacerbate them, are not immutable. vision can motivate and rally support for reforms. Change will come when the incentives to throw out the For example, in some transition economiessuch as the old policies and institutional arrangements become Czech Republic, Hungary, and Polandreform was aided stronger than the incentives to keep them. An economic by the prospect of joining the European Union. The same crisis or an external threat may provide the impetus for ini- desire motivated reform in Spain and Portugal in the early tiating reforms. But its precise timing can be prolonged if 1980s. In other instances a clear vision does not present those who control state power stick with outdated policies itself so easily. Venezuela's reforms under Carlos Andrés because it is in their (or their allies') interest that it be so. Perez were a political failure because there was no coherent Sometimes the delays can be painfully long, as in Haiti vision to help sell the reforms (Box 9.5). In Malaysia, by under the Duvaliers or Zaire today. 156 WORLD DEVELOPMENT REPORT 1997 Box 9.5 Venezuela's 1989 reform program and its reversal In the late 1980s the Venezuelan economy was experi- community. Venezuela's bond rating dropped 20 per- encing a deep crisis, with internal and external imbal- centage points below its 1991 level, In 1994 real GDP ances generated by an overextended state and a mis- dropped by 3 percent, and inflation soared to 71 per- managed economy. In 1989 Carlos Andrés Perez, who cent. in the second quarter of 1996 the administration had served as president in the 1970s, was reelected. His began adopting more orthodox economic policies, 1989 stabilization plan included a sharp devaluation of supported by an arrangement with the IMF. These the boilvar and a lifting of price and interest rate con- have been implemented with some success thus far. trols. The plan restored internal and external balance How far-reaching and sustainable this program will be and was accompanied by structural reforms such as remains to be seen. trade liberalization, privatization, and increased central Venezuela's experience highlights some important bank autonomy. points about the sustainability of reform. Economic These reforms were made possible by the broad reform is more susceptible to reversal when it is sup- executive powers of the Venezuelan presidency. But ported by only a few technocrats, without backing the consensus on them did not last long. After a period from political parties or other groups. Reforms linked of political instability, in February 1994 Rafael Caldera to stabilization are easier to implement than are struc- was elected as leader of a coalition government. The tural reforms requiring congressional approval. More- new government started by repudiating some of the over, economic reforms are harder to implement in an reforms, eliminating the value added tax and returning environment prone to political risk. Introducing new to price and interest rate controls. The autonomy of taxes is contentious in an environment where state the central bank was also compromised, prompting its control over natural resources (in this case petroleum) president to resign. Constrained by international gives the appearance of cost-free public services. A cri- agreements, the administration did not reverse trade sis may be sufficient to create the conditions for initi- liberalization, but instead relied on nontariff protec- ating reforms, but sustaining them requires much tion. The program's lack of coherence caused an ero- more. Long-term performance requires vision and sion of confidence in the international investment unity of purpose. This chapter has shown how windows of opportunity beyond reforming a few agencies, and thus escaping the can open, and how important it is to seize these opportu- enclave trap. This involves defining clear rules under nities to bring about changethrough compensation of which agencies outside the original enclave will be brought potential losers, the skillful choice of tactics, and building into the reform program. Rules and tactics, however, are consensus. As emphasized throughout this Report, it is no substitute for commitment and political leadership. important to reckon on state capability when designing Leadership is not everything: even committed and reforms. But reformers cannot afford to stop there: they visionary leaders are not always able to throw off the heavy must also have a strategy for improving capability. A more legacy of years, perhaps decades, of poor performance. capable state can broaden the scope of policy options Choosing reform in these circumstances usually involves a and significantly improve economic performance. Strate- leap of faith, a leap that can be as frightening to those who gic sequencing, and even well-thought-out enclaving, are a will ultimately gain as it is to the losers. In these circum- good first step and are consistent with the budgetary and stances the presence of someone who can convince the other constraints facing many developing countries. Early public that the leap is worth making is a potent reform on, however, countries should design a strategy for going weapon indeed. THE AGENDA FOR CHANGE REPORT HAS SHOWN THAT THE STATE HAS ENOR- their decisions. It takes time for national and provincial THIS mous sway over a country's economic and social legislatures, political appointees, judges, civil servants, development and whether that development is sustain- public-private deliberation councils, independent watch- able. The state's potential to leverage, promote, and medi- dogs, and nongovernmental organizationsarrayed in ate change in pursuit of collective ends is unmatched. unique relations to one another in different societiesto Where this capacity has been used well, economies have learn to respect the limits of one another's authority flourished. But where it has not, development has hit a and to work together. It takes time to lay the foundations brick wall. of a professional, rule-based bureaucracy. Still, it is pos- The process is not irrevocable: examples through his- sible to sequence reforms in a manner that yields some tory, and throughout the world, have shown that countries early payoffs. Such early measures can include strengthen- can change track, reforming policies and institutions to ing the capacity of central government, raising upper- improve the state's effectiveness and forward development. end salary scales to attract capable staff, inviting more But why do so many of these efforts fail to achieve their inputs into policymaking and making deliberations more goals? Even more important, why is it that so many states, open, hiving off contestable and easily specified activi- in desperate circumstances, do not even attempt them? ties for private sector involvement, and seeking more Politics provides much of the answer. But it is not simply feedback from clients. In general, however, there are few a matter of democracy versus authoritarianism. We need to quick fixes. go beyond those broad concepts of political organization Second, the task is difficult because the same institu- to understand the incentives that inspire state organiza- tions that can foster credibility and accountability can also tions to work better. We need to understand better how be constraining. The same rules that prevent abuse of state and when the economic and political interests that favor authority can also lessen the ability to use that authority development can be harnessed to bring about the institu- well. The challenge is to devise institutional arrangements tional changes needed to make development happen. that provide flexibility within appropriate restraints. In the end each country must strike its own balance. Efforts to restart development in countries with in- Countries with strong track records of following through effective states must start with institutional arrange- on commitments, and whose state institutions have deep ments that foster responsiveness, accountability, and roots in society, are likely to have room to experiment and the rule of law. to respond flexibly to unanticipated events, with little cost to credibility. But countries emerging from long periods Enabling the state to do more good for the economy of arbitrary and dysfunctional government might be bet- and society means building confidence; people must have ter off forswearing the short-term benefits of flexibility in trust in the basic rules governing society and in the public favor of the long-term goal of building credible and sus- authority that underpins them. The task is difficult for tainable institutions. two reasons. Where prospects for reform are strong, this Report has First, it requires patience. It takes time for judiciaries suggested a two-part approach to improving the effective- to convince firms and citizens that they are impartial in ness of the state: 158 WORLD DEVELOPMENT REPORT 1997 Matching role with capability, and involved is institutional, and institutional change is path- Reinvigorating state capability by subjecting the state dependent. We examine below some of the key opportu- to more rules and restraint and greater competitive nities and challenges facing states in each developing pressure, and making it more transparent and open. region. But the advice is premised on the assumption that there is, at least, a state there to reform. Increasingly, even Focusing limited state capability on the basics is a this basic precondition for improving development out- badly needed first step in a wide range of coun- comes does not apply. We turn first to these most extreme triesespecially in Africa, the CIS countries, and cases of institutional failure, when, for a period, the state parts ofLatin America, the Middle East, and South itself as a legitimate and functioning order is gone. Asia. When states collapse Matching role with capability must come first. In In recent years a growing number of countries have seen many parts of the world the state is not even performing virtually all of the functions and institutions of' govern- its basic functions: safeguarding law and order, protecting ment collapse, often in the context of civil war. 'X'hen the property rights, managing the macroeconomy, providing state ceases to perform even its most basic functions, the basic social services, and protecting the destitute. associated crisis can be prolonged and severe. Structures In some cases the state has been overregulating the that might normally mitigate the impact of the crisis and economy, even though it lacks the capability to enforce reg- provide a vehicle for eventual recovery are frequently ulations systematically. One consequence has been the destroyed, making the more generic prescriptions for en- spread of corruption, which corrodes capability even fur- hancing state capability somewhat inapplicable. thersometimes rendering the state incapable of deliver- ing even basic services. Reasonably well functioning The causes of state collapse bureaucracies have been weakened over the years and will State collapse is not a new phenomenon. It does, how- now need considerable overhaul. Deregulating these eco- ever, seem to have become more common in the 1980s nomies is vital. and 1990s than in the earlier postwar years, and it is a Reforms supported by the World Bank and other matter of grave concern for other states, individually and international organizations have tried to help countries collectively. Collapsed states such as Afghanistan, Cambo- design reforms to match the state's role to its capability. dia, Liberia, Rwanda, and Somalia have been the settings Many reforms over the last decade or so have tried to trim for some of the worst humanitarian disasters of recent back the state's role in some areas of the economy and years. These often spill over into neighboring countries in refocus the released resources on core functions. These are the form of violence, banditry, and refugee flows. They a start, but more is needed. also send countries into arrears, destroy economic assets However, as this Report has shown, matching role to and infrastructure, claim huge amounts of international capability is not simply a matter of reducing the state's assistanceand, of course, waste countless lives. Three role. It is not just a question of what to do, but also a broad and overlapping pathologies of state collapse can question of how to do it: how to manage the financial sec- be identified: tor, how to regulate economic activity, how to protect the environmentall these choices depend critically on the States that have lost (or failed to establish) legitimacy in state's institutional capability. the eyes of most of the population notionally under Efforts to focus the state's existing capability more their authority, and are therefore unable to exercise that effectively must be complemented by reforms to improve authority capability by reinvigorating public institutions. In many States that have been run into the ground by leaders areas the state will only be able to improve its effectiveness and officials who are corrupt, negligent, incompetent, by forging new partnerships with other organizations of or all three civil society. In other instances it will become more effec- States that have fragmented in civil war, and in which tive only if its decisions and actions can be contestedif no party is capable of reestablishing central authority. people and business have choices and if the state's monop- oly is broken. What all these states have in common is a fundamen- tal loss of institutional capability. As outlined in Chapter Imp roving state capability is not simply a matter of 1 (see Table 1.1), at a minimum the state must perform more technical assistance. It takes the right incentives. the most basic functions of maintaining law and order, providing national defense, and establishing a framework The precise recipe for improved state capability will for managing economic transactions. A collapsed state, vary from country to country, since so much of the reform then, is not one that fails to do the right things, but one THE AGENDA FOR CHANGE 159 that fails to do much of anything effectivelyeven main- omy of armed violence has emerged, based on looting, tain repressive order. A collapsed state is not the same as a protection rackets, drug trafficking, money laundering, state defeated in war, or a state split apart into two or and the extraction of crude resources such as gems, min- more states, where, despite some incapacity, systems con- erals, and tropical timber (Box 10.1). Civil warfare in tinue to function more or less effectively. Nor is a col- these countries has its roots in political or ethnic rivalries, lapsed state the same thing as a "predatory" or repressive but it has gradually shifted character and is now cen- state. Successful repression of a whole population can re- tered around the control of economic assets, which pro- quire considerable organizational capacity. Such acts may vide the source of financing for the war and for private provoke the replacement of one state structure or regime enrichment. with another, as in Cambodia or Rwanda, but do not nec- In such countries factional warfare is the main system essarilyor even probablybring about its collapse. of resource allocation, and violence the source of power. These economies operate independently of any state insti- Factors perpetuating the conflict tutions, and indeed would likely be hindered by them. In most countries where the state has collapsed, there Thus, strong economic forces are at work perpetuating the are forces that have an interest in perpetuating a state fighting. The role of economic factors as a driving cause of anarchy, and whose unbridled pursuit of riches or of the war has been especially pronounced in the case of power would be constrained by a state with the capability Liberia (Box 10.2), but it can also explain the prolonga- to make rules, collect revenue, and enforce the law. The tion of wars in Angola and Sierra Leone. Ordinary people most troubling and intractable of collapsed states are pay a high price for their helpless proximity to these sys- those in which these forces predominate. In Angola, tems. Normal international economic transactions are dis- Liberia, and Somalia, for example, a self-sustaining econ- rupted and, often, corrupted by them. Box 10.1 State collapse and beyond in Somalia Somalia is one of the most ethnically homogeneous some reconstruction from late 1992 to early 1995, but countries in Africa; one of the most common factors in these gains were reversed after U.N. forces withdrew, state collapseinterethnic conflictplayed no part in and anarchy returned. its collapse. Rather, the dynamic of Somalia's collapse Some elements of the private, informal market have was set in motion during a long period (1969-91) of shown an extraordinary, if somewhat perverse, resili- dictatorial rule and egregious economic mismanage- ence to the chaos and violence that surrounds them. ment by Mohamed Siad Barre, followed by a fierce Merchants and entrepreneurs can operate only with contest for power after he was violently deposed. Siad clan protection; those who benefit from this protection, Barre played off the Cold War superpowers against and from their clan's control of economic assets such as each other. His external support, first from the Soviet ports, mineral deposits, and agricultural lands, finance Union and then from the United States, unbalanced clan militias. With no government regulations or con- the fragile social ecology of Somalia's clan politics and trols, the prices of goods and foreign currencies float introduced high levels of lethal weaponry. A north- freely. Transactions costs are high, but markets operate south conflict inherited from the preindependence era, efficiently. The Somali shilling has value despite the an irredentist war to seize Ethiopia's Somali-populated absence of a central bank, but banknotes are physically Ogaden region, and the suspension of economic and wearing out. Remittances from abroad are handled military aid in 1989 all further contributed to the ero- quickly and reliably through clan networks. sion of state capability. Today, Somalia exemplifies a post-state economy. Since 1991, continuous civil war between rival Productive capacity, private investment, and employ- clans and factions has completed the destruction of the ment are extremely low; risks are high, and violence is formal economy and left Somalia with no state institu- the normal means of competition and resolution of tional or public administration. Physical infrastructure economic disputes. If a functioning state with legiti- and economic assets have been destroyed, private busi- mate institutions were able to restore law and order nesses looted, and agricultural and pastoral production without reverting to the heavy-handed economic mis- drastically reduced. Professionals, technocrats, and management of the Siad Barre regime, the private sec- civil servants have fled. International intervention to tor has shown that it could weather the high risk and halt a major war-induced famine brought a respite and respond to the challenges of the market. 160 WORLD DEVELOPMENT REPORT 1997 Box 10.2 The economic underpinnings of conflict: The case of Liberia Initially, the war in Liberia was fought largely for social fighters rarely receive direct payments, instead relying and political motives, with control over the central on looting and pillaging for their livelihood. government as the central objective. Foreign support Because fighting actually runs down the "surplus," helped finance the launching of the war. Gradually as long as the prospects for private enrichment remain control over Liberia's rich natural resources and other high the incentives are strong for factions to main- assets, in addition to being a means of ensuring fund- tain a "no war, no peace" equilibrium. This helps ex- ing for the war, has become an end in itself for the plain the difficulties in reaching a lasting peace agree- fighting factions. ment in Liberia. The factions depend on a steady supply of income Thus, to restore peace, any intervention will have to from the export of rubber, timber, iron ore, gold, and include actions that make the "surplus" accruing to the diamonds. During periods of intense conflict, much of fighting organizations zero or negative, while simulta- the income is used to purchase arms and ammunition. neously eliminating individual fighters' income from When no fighting is taking place, a revenue "surplus" looting, or raising the associated risk. Such actions is created, which provides for the personal enrichment range from economic sanctions on the export of tim- of the leaders and for the payoffs that keep faction ber and rubber exports to international agreements members loyal. This, together with the uncertainty that increase the cost of arms. In addition, combatants' that faction members face regarding their economic incentives to continue the war must be removed prospects after the war, creates strong pressures down through a combination of carefully planned demobi- the ranks to share in the surplus. However, lower-rank lization and enforcement of criminal laws. Although the diminishing returns of conflict and re- embed bureaucratic decisionmaking within appropriate source extraction imply that wars will eventually fade restraints. The integrity and professionalism of the civil away without international intervention, this point is service are an important element in this. reached only after nearly all the assets of a country have been destroyed. This leaves other states, and intergovern- An active civil society and a competent and profes- mental organizations, facing two urgent questions: How sional bureaucracy are twin pillars ofa constructive can state collapse be prevented? And how can a function- relationship between state and society. When com- ing state be rescued from the ashes of a collapsed one? prehensive collapse of the state is a danger, these twin pillars may reduce the risk. Preventing state collapse The enormous toll of collapsed states has naturally turned The challenge of reconstruction attention to prevention as a preferable and potentially less Rapid retrieval of a lost state is even more difficultif not costly course of action. But there are no easy shortcuts. impossible. Most collapsed states are immersed in, or just Preventing state collapse ultimately involves the whole emerging from, bitter internecine conflict. Politics is often gamut of institutions and policies that affect social, polit- highly charged with suspicion of complicity; mistrust is ical, and economic relations within a society. Neverthe- widespread. Cooperation among former rivals is difficult less, two interrelated approaches have emerged as ways of to achieve, and the desire for justice on the part of some reducing the likelihood that political conflict will spiral may look like vengeance seeking to others. There may be down into a comprehensive collapse of the state. little confidence that peace will be sustained, or policies The first strategy emphasizes the reinforcement of civil maintained, and social and human capital has often been society as a way of increasing the resilience of social insti- severely damaged. tutions that may be able to fend off anarchy even if the External actors face unusual difficulties when they state is very weak. A rich associational life may enable decide to become engaged in a country without a func- communities to maintain local law and order, support a tioning state. A basic question arises: how and with whom safety net, and resist official corruption or exploitation. should one work? One strategy involves choosing among The second approach, drawn from the East Asian experi- local factions or leaders and attempting to reinforce the ence, points to the need to build bureaucratic structures position of the chosen partner. This is risky. The chosen that are insulated from political interference, and to partner may be unreliable or may abuse the position of THE AGENDA FOR CHANGE 161 leadership. There is also the risk that another external force ECONOMIC RECOVERY. Although the elements of a will pick an alternative winner, prolonging the conflict, as postconflict stabilization and recovery program may differ occurred in Mozambique and Angola. Alternatively, out- little from programs applied elsewhere, there are grounds siders can choose to work with those local authorities and for caution and for examining the components of the stan- institutions that often retain legitimacy and capability long dard policy package in light of the conditions and after central institutions collapse. Such a strategy can help distortions peculiar to the postconflict environment. Civil rebuild local confidence and strengthen local decisionmak- conflict erodes the social capital that inspires trust and ing capacity. But it can also prolong the period of state facilitates investment. Consequently, fear may be a bigger fragmentation if it is not combined with a strategy for obstacle to the revival of investment than damaged infra- rebuilding the authority of a central government. Somalia structure. Policies to spur the private sector's recovery is a case in point. should therefore avoid overly aggressive increases in taxa- In many instances external actors focus on mitigating tion and should stress the early sequencing of investment- the human cost of state collapse through relief and rehabil- sensitive reforms, including the preservation of low infla- itation programs, rather than intervening directly in civil or tion, the sale or restitution of expropriated housing stock, political reconstruction. Recent experience suggests, how- and restraint in revenue collection. Avoiding aggressive ever, that after some time these programs can lead to revenue collection does not, however, mean offering overly dependency among the population, and even undermine generous tax incentives for private foreign investment. As the rebuilding of state capability by drawing attention away Cambodia's experience now shows, these may be counter- from minimal state functions. By contrast, more-compre- productive, depriving the government of substantial rev- hensive peace approaches, such as adopted in Angola, enue for many years and taking government further from Cambodia, and Mozambique, place far-reaching formal the all-important task of reviving domestic investment. authority in the hands oi international organizations to PROMOTING FUNDAMENTAL SOCIOPOLITICAL RECON- oversee peace agreements, deliver public services, and facil- CILIATION. Economic policy and sectoral projects need to itate the repatriation of refugees. All this is with a view to be tailored to take account of the special problems of trau- handing over power to an indigenous authority after elec- matized populations and societies with heavily damaged tions have been held. The international operations in social capital. The sequencing of standard economic pol- Cambodia and Mozambique are generally considered to icy reform and of governance measures also needs to be have been successful in this regard. But both were enor- tailored to avoid threatening the sustainability of peace mously costly, and external actors cannot and should not agreements and to take account of the typically severe dis- be expected to replace absent state capability indefinitely. tortions in economic conditions. The choice of strategy for external actors will differ BUILDING CAPABILITY. The problem at the heart of according to the particular pathology of state collapse attempts to sustain recovery and reconstruction in post- involved. As the discussion in Chapter 8 suggests, there is conflict situations is lack of capability. Yet even where an increasing awareness of the role that external donors effective macroeconomic and political reforms have been and agencies may need to play in the future to allow for undertaken, improvements in bureaucratic capability have the long process of reconstruction and rehabilitation. often lagged well behind (as in Uganda, for example). Var- Some of the challenges facing external actors are men- ious options have been tried to jump-start the capacity- tioned below. building process, including drawing on a diaspora of pro- REESTABLISHING SECURITY. The end of civil war does fessionals to occupy positions within the new government not lead automatically to the end of insecurity. Fear of (as in Cambodia, Lebanon, and Haiti) and providing for- personal violence or theft may actually increase after the eign experts to work as advisers to their local counterparts. war, and without measures to increase the opportunity As the strategy for enhancing state capability implies, con- cost of warfare for individual combatants, the end of the ventional technical assistance is rarely effective in settings war may signal increased opportunities for criminal activ- where even minimal rules and restraints on arbitrary action ity. Well-designed demobilization programs, like that in are lacking. The absence of such rules is a defining feature Uganda, that assure combatants of their reintegration into of a failed state. Thus, it is vital that imported technical civil society after the war, and so give them a legitimate assistance come with efforts to implement and enforce the exit strategy, are therefore a priority. Also essential will be most basic rules of accountability and restraint within gov- efforts to strengthen local police forces and tighten en- ernment and so begin to rebuild its lost credibility. forcement of criminal laws and other measures, to reduce The collapse of a state is an extreme phenomenon, and the risks of banditry and criminal enterprise. The long- each case is unique. Hence there are no simple generaliza- term response, however, must be to address the socioeco- tions about their causes or effects. Each produces its own nomic root causes of the conflict. challenges for the afflicted nation and for the international 152 WORLD DEVELOPMENT REPORT 1997 system. The consequences are, however, almost uniformly governments have felt a decreasing need to explain and borne by ordinary people, illustrating once again the crit- justify their actions to the domestic population. ical importance of an effective, responsive state to the long-term health and wealth of society. The urgent priority in Africa is to rebuild state capa- bility through an overhaul ofpublic institutions and Some regional steps-and missteps credible checks on the abuse of state power. Inevitably, the elements of any strategy for improving the state's effectiveness will differ hugely from country to The clear need in most African countries is for a thor- country, according to the institutional and political set- ough reform of the machinery of the state, so that it can ting. The key features and challenges facing states in vari- deliver quality public services and facilitate private activity ous regions are summarized below. These are necessarily rather than impede both. As this Report has emphasized, broad generalizations: each region includes several coun- the second part of the strategyrebuilding capability tries whose experiences are entirely different. must also start on an urgent footing. In this it is essential to It is in Sub-Saha ran Africa that the deterioration in the start in some areas of priority. Strengthening the rule of law state's effectiveness has been most severethe result of must be a vital first step. Another is to strengthen the capa- eroding civil service wages, heavy dependence on aid, and bility for legislative oversight of the executive. But strength- patronage politics. There are a few exceptions, such as ening the executiveparticularly the central government's Botswana. But a recent report commissioned by a group capacity to formulate macroeconomic and strategic policy, of African finance ministers, echoing themes raised in the and its incentives to deliver core public goods efficiently preceding chapters of this Report, concluded that the is also a key priority. Disappointingly, with some promis- majority of countries in Sub-Saharan Africa now have ing exceptions (Botswana and more recently Uganda), lower capability (including state capability) than they did improving the effectiveness of the state in Africa has gener- at independence. As a consequence, many are trapped in ally been limited to tinkering around the edges and to a vicious circle of declining state capability and, thus, promulgating reforms on paper. But there are signs that declining credibility in the eyes of their citizens. these issues are now being taken seriously. Recent reform programs pay much greater attention to institutional There is a crisis of statehood in much ofAfricaa strengthening, and macroeconomic management is im- crisis of capability and of legitimacy. proving across a broad swath of African economies. An important constraint on systemic institutional The first part of the two-part strategy recommended reform of the state in Africa is its cost, but resources can be by this Report is perhaps the more urgently needed in found if the priorities are clear. Most countries will need Sub-Saharan Africa. A sharp refocus of the state's priori- to redirect existing, misallocated resources to raise real ties is badly needed. In a few cases this refocus has taken public wages (more than proportionately to any savings place, although only after protracted crisis. Typically, the from further retrenchment), increase spending on social reach and effectiveness of the state have withered away, services, and undertake vast investments in personnel and perforce the state has in effect withdrawn. Unfortu- management, retraining, and accountability. This major nately, since the refocus has not been carefully managed, rebuilding of state capability cannot happen without inter- the state has also weakened or withdrawn from areas that national assistance. But an exceptional degree of coopera- are its legitimate function. An institutional vacuum of sig- tion between governments and external agencies has cul- nificant proportions has emerged in many parts of Sub- minated in a pattern of external dependency. The types of Saharan Africa, leading to increased crime and an absence assistance and the incentives it generates also need to be of security, affecting investment and growth. reexamined, to ensure that assistance supports coherent Achieving a turnaround in the effectiveness of the state and well-contested policies and strengthens the overall will not be easy, since the roots of state failure are many framework of incentives within and outside the state. Pri- and complex. Chief among them has been a continuing oritizing and even strategic enclaving are necessary, to start struggle between traditional forms of governance and the reform process, but must be part of a well-thought-out social organization (often based on tribes, lineages, and plan to broaden and deepen the reforms. language and kinship groups) and modern forms of gov- ernment. High military expenditure and dysfunctional Imp roving the delivery ofpublic and collective ser- behavior of military personnel (in the absence of other vices will require closer partnersh:ps with the private checks and balances) have been other important impedi- sector and civil society. Such partnerships should be ments. These have often reduced the transparency and encouraged, especially when the links between the accountability of public institutions to the extent that state and civil society are underdeveloped. THE AGENDA FOR CHANGE 163 A hopeful recent development is a growing array of Throughout the region substantial economic rents self-help community initiatives, particularly in elementary (flowing from legal monopolies or restrictions, or from in- education, basic health care, and local services such as fluence and corruption) continue to coexist with great pri- waste disposal. These initiatives have often sprung from vate sector dynamism and international engagement. Tol- the state's own failure to provide such services effectively. erance of these rents is probably decliningwitness the Although they can seldom fully substitute for a well- recent anticorruption offensives in China, Korea, and Viet- functioning government administration, they offer a par- nam. And heightened international competition may mean tial escape from the current morass. That said, it is difficult that inefficient, protected sectors can no longer be carried to imagine how reform of the role of the state and im- on the shoulders of dynamic and efficient ones, either in provement in its capability can be realized in most African Japan and Korea or in the newly emerging economies. countries without the stronger incentives that this Report This suggests that deeper, modernizing reforms will be also recommends: more competitive pressure, greater voice needed in much of East Asia to develop robust regulatory and transparency, and rules and restraints including the structures to support competition and, in areas of contin- rule of law. ued government action, mechanisms to ensure greater The situation in Asia is different. In the newly industri- transparency and accountability. Particularly vital for these alizing countries of East Asia, the state is generally viewed countries' long-term development prospects will be more as effective, engaged in a productive partnership with the efficient financing and delivery of infrastructure in part- private sector. With few exceptions, it has matched its role nership with the private sector. Although many high- to its capability very well and thereby enhanced its effec- profile projects are under way, East Asia needs to move tiveness. Whereas ineffective authoritarian states have been faster in developing regulatory frameworks for creating directly responsible for economic decline in Africa, many competition, managing the contracting process, and regu- East Asian countries have experienced remarkable growth lating monopolies. (with some improvement in equity) under authoritarian At the same time, many East Asian societies are likely to regimes. As the previous chapter implied, the link between face new areas where enhanced state action is called for. authoritarianism and economic decline, so evident in These are likely to include formal mechanisms such as Africa, has been inoperative in Asian countries, largely be- social and health insurance schemes, to help manage the cause of their powerful commitment to rapid economic new risks facing individuals and households in increasingly development, strong administrative capability, and institu- urbanized and aging societies, and more vigorous efforts to tionalized links with stakeholders such as private firms, as protect the environment. Public action will also be needed well as their ability to deliver on the economic and social to reach those left behind by rapid development, for exam- fundamentals: sound economic management, basic educa- ple the rural populations in some inland provinces of tion and health care, and infrastructure. China, the relatively poorer populations of eastern Indone- sia, and, in some countries, ethnic minorities. New mech- In East Asia the state 'c ability to change in response anisms to respond to rising inequality may also be needed, to important new challenges will be critical to the for example in China, Malaysia, and Thailand. region 's continued economic success. At the same Finally, many countries in the region face a need to time, steps to modernize public administration and build more-effective core government institutions (follow- achieve effective decentralization will be critical. ing the agenda laid out in Chapter 5) and address the problem of corruption (Chapter 6). Institution building is Yet throughout East Asia there are major new chal- a high priority for those countries where the civil service is lenges to the role and flsnctioning of the state. China and relatively weak (Indonesia, for example) and where tradi- Vietnam, the region's transition economies, are strug- tional state capabilities have eroded somewhat (such as the gling with the need to redefine the state's role in the Philippines and Thailand). Efforts to build bureaucratic enterprise and financial sectors. Both countries have capability at lower levels of government are also urgently combined rapid growth with the maintenance of large needed in countries such as China, Indonesia, the Philip- public enterprise sectors, but these sectors appear in- pines, and Vietnam, where the decentralization of some creasingly fragile financially. There is also a continuing government functions puts power in the hands of lower- debate in these economies about the potential role for an quality staff and raises risks of local government capture activist industrial policy in guiding investment. However, (see Chapter 7). the conditions facing these two economies appear to be very different from those prevailing in Japan and the The situation is different in South Asia, where in Republic of Korea in their early stages of industrialization many countries state inefficiency and corruption (Chapter 4). have coexisted with a relatively competent and 164 WORLD DEVELOPMENT REPORT 1997 efficient civil service, albeit one whose quality has undergoing the transition from central planning. There suffered a noticeable decline. are also broad differences between the CEE countries as a group and the CIS countries. Although both face a com- South Asia, too, suffers from a mismatch between the mon challenge of economic and political transforma- state's role and its capability. Here it is not that the state's tionand hence of a fundamental refocusing of state capability was historically weak, as in Africa. In fact it was activitiesthe CIS countries (except Russia) face the not, but the state pursued an overly activist agenda that gigantic, and historically unique, double challenge of eco- went well beyond its capability. As a result, development nomic transformation and state building: they have had suffered, and over time the state's capability declined as no central government within living memory, nor have well. The agenda now is to refocus the state's role to they constituted for centuries (most never) independent match its current capability, stem its further decline, and political entities within their present boundaries. improve that capability. The challenge of bringing more effective government The main issue in South Asia is overregulation and an to these transition economies is in some ways easier and in overextended stateboth cause and effect of bloated pub- other ways more difficult thanbut certainly quite dif- lic employment, and the surest route to corruption. As ferent fromthat faced in most developing economies. stressed in Chapter 4, regulatory simplification and pub- The first part of the two-part strategy, that of matching lic enterprise reform, and the resulting contraction of the role to capability, is critical, but especially difficult in a sit- role of the state, will be complex and politically difficult. uation where both role and capability are changing But such measures stand to boost economic efficiency, rapidly. The extent of deterioration in capability varies increase competitive pressure, reduce corruption, and pro- enormously, with islands of excellence in a sea of declin- duce substantial fiscal savings. Another imperative is to ing effectiveness and huge problems of basic governance. build stronger partnerships with, and listen more effec- In this context matching role to capability is not a simple tively to, business and civil society, to improve feedback matter of focusing on the fundamentals. It is not that the and supplement the state's capability. fundamentals are not criticalwithout them progress Pervasive political interference is a second major bar- toward a market-based economic system will be rocky. rier to state effectiveness in many countries of the region. But the challenge in these countries is providing these As discussed in Chapter 5, India's bureaucratic capability fundamentals while at the same time maintaining existing suffers not merely from legendary amounts of administra- centers of excellence that have been developed over time, tive red tape, but from the fact that the autonomy of its and providing the population with the mechanisms of highly qualified civil servants is severely circumscribed in social protection to which they are accustomed. Given the practice (despite statutory protection from interference in inherent capabilities of a highly educated, socially con- their individual actions) by frequent, often politically scious population, the task of rebuilding state capability motivated transfers of personnel and other arrangements. the second part of the strategyis here more than else- In Pakistan the machinery of the state has come where one of changing incentives and the culture of increasingly under pressure from powerful business and bureaucracy. How quickly this reorientation is managed feudal interest groups. More recently the resulting erosion will determine how quickly the transition is completed. of state authority and effectiveness has spurred an attempt to reform the very machinery of the state. Here, too, a The job of reorienting the state toward the task of competent civil service, allied with an activist judiciary, "steering, not rowing" isfarfrom complete in Cen- offers grounds for optimism that reform is not only possi- tral and Eastern Europe. But most countries have ble, but will occur. In Bangladesh, recent political devel- made progress in refocusing the state's role and are on opments have also created conditions for beginning to the way to improving capability and accountability. address the decline of the state's authority and effective- ness, using the pillars of strong top-level political com- It is no surprise, then, that improvements in the effec- mitment, popular participation, and decentralization, tiveness of the state have been more noticeable in CEE, buttressed by competent and professional elements of the especially in countries such as the Czech Republic, Hun- civil service. Similar possibilities for improving govern- gary, and Poland. Much of this has come from a very swift ment and streamlining public administration are emerg- refocusing of the state's rolepart one of the two-part ing elsewhere in the region. strategy. However, progress in southeastern Europe has As noted earlier, substantial differences exist among generally been much less significant, both in terms of pol- countries in all regions, and regional generalizations are icy reform and on the public administration front. useful only as a first approximation. This is particularly In particular, several CEE countries have made true of the countries of Central and Eastern Europe (GEE) advances toward setting up competitive systems for service and the Commonwealth of Independent States (CIS), now delivery and toward relying much more on the private sec- THE AGENDA FOR CHANGE 15 tor (including, in part, for education and health care). Outside assistance and encouragementto tighten the Autonomous professional associations are also beginning to accountability of executives to legislatures, to strengthen take root in these countries, with healthy potential im- legislative oversight capacity, to streamline and reinforce plications for institutional capability outside the govern- judicial systems, to clarify property rights and provide for ment (capability that is needed to help implement the accountable management and oversight of public assets, reforms) and for the rebuilding of social capital. And sub- to radically simplify the regulatory framework and move stantial regulatory reform and simplification have occurred, rapidly toward rules and away from discretion, and, not aided by support from the European Union. Again, in least, to bring criminal activities under control and sever southeastern Europe progress in all these areas tends to their links to the public sectorare clearly essential to have been much slower. The difficulty is not so much a improve the functioning of the state and promote eco- lack of consensus on the desirability of these reforms, but a nomic recovery throughout the region. severe lack of capability at the core of government to for- Genuine administrative reform to improve bureau- mulate and guide themcompounded, in some countries, cratic capability has been conspicuous by its absence in by a fluid political situation. most CIS countries. Staffing of central government tends to be low, and although government workers are often Low state capabiliiy in the former Soviet republics highly educated and technically knowledgeable, they lack is a serious and mounting obstacle to further prog- the skills needed for administration in a noncommand ress in most areas of economic and social policy. economy. They are, moreover, deprived of basic supplies and resources and very badly paid. Unsurprisingly, during In the CIS the process of reorienting the state is still at the first stage of the economic transformation, attention an early stage, and a host of severe problems have and financial resources were devoted to the urgent tasks of emerged. Unlike in GEE, in the CIS there is often no clear privatization, to reforming the price and foreign exchange definition of the state's new role. And even where it is system, and to establishing some measure of fiscal and clear in principle, in practice weak central capacity has monetary control. But it is equally unsurprising that the made it extremely difficult to translate those principles next stage of the transformation, that dealing with reforms into reality. In most cases the basic institutional structures that require decisive and clear administrative action, is for an effective state are weak. The first and fundamental now severely hampered by the absence of the mechanisms problem is the weak accountability of the executive to the and of the government personnel and resources necessary legislature in most CIS countries. In the very early 1990s, to implement those reforms. parliaments in most of the new countries were the only The role of the state in Latin America and the repository of whatever political legitimacy existed after the Caribbean has been undergoing profound change over the collapse of the Soviet Union. There were even instances of past decade, fueled by fiscal crisis and rising expectations excessive interference by the legislature in routine execu- born of the return of democracy and the emergence of tive functions (which remains a problem in Ukraine). But civil society. Competitive pressures from globalization the situation in most countries today is one of a dominant have made it increasingly urgent for government to priva- executive and weak legislative oversight. tize or to contract out activities in which it lacks compar- Beyond the issue of increasing executive accountabil- ative advantage, and to improve its efficiency and effec- ity, three further sets of problems face most CIS countries. tiveness in producing public goods. Between them these problems not only produce enor- But while the first-stage reforms have been imple- mous economic costs, but also give rise to pervasive cor- mented successfully in a number of countries in the ruption. This last is rightly decried almost everywhere in region, progress has lagged in the second-generation re- the region, and confirmed in the results of the survey of forms, where institutional change is vital. For example, private business commissioned for this Report. First, weak civil service reform and reforms in the social services are and slow judicial systems severely handicap both reform being held back by political constraints and vested inter- and economic activity. As is well known, weak enforce- ests. Such reforms (recently begun in Argentina and a few ment of the law and weak judicial procedures have per- other countries) cannot be undertaken in isolation, but mitted the unprecedented rise of organized crime. Sec- must be part of a broader decentralization of political ond, fuzzy boundaries between private and public administration and financial management, already under property rights have generated massive rent seeking, with way in many countries. officials often exploiting public assets, including those of state enterprises, for private profit. Third, opaque and still Governments in Latin America are rethinking their hugely complex regulatory frameworks have fueled cor- approach to the alleviation ofpoverty, particularly ruption across the spectrum of interactions between the important in a region with highly skewed income public and the private sectors. distributions. 166 WORLD DEVELOPMENT REPORT 1997 A number of Latin American countries, having services renderedas axiomatic. But four decades of cen- launched major economic and structural reforms, have tralized systems in Latin America and the Caribbean have now created social investment funds for the poorer seg- broken this unspoken bargain of trust, and poor cost ments of society. At a broader level, governments in the recovery and low tax revenues are just two of the conse- region have been faced with the bankruptcy of highly in- quences. The outstanding feature of the new style of gov- equitable, state-run, pay-as-you-go pension systems. Chile erning, and one bright prospect for the future, is the privatized its system in 198 1, reducing the role of govern- restoration of this critical fiscal link at the local level. ment to that of regulator; other countries (Argentina and Governments are also moving to create more open Colombia, for example) have pursued dual systems. A administrations, reflecting the emergence of a more forceful third group of countries are just now starting to contem- civil society (media, private think tanks, nongovernmental plate how pension reform should proceed. But more organizations) demanding greater access to decisionmaking, aggressive action will be needed on the social fundamen- particularly in the Commonwealth Caribbean through cit- tals, to complement reform on the economic fundamen- izens' charters (in Jamaica, for example). Efforts are under tals, if Latin America is to compete with East Asia. way in countries such as Colombia to develop performance The priority in Latin America is to reinvigorate the indicators (of economy, efficiency, effectiveness, service institutional capability of the state, and here some major quality, and financial performance) that will provide ac- initiatives are under way. Both unitary and federal states countability for public sector managers and politicians. have turned to decentralization, particularly to improve Court systems in most Latin American countries have service delivery. Authority and responsibility for both rev- suffered from major inefficiencies, widespread corruption, enue and spending have been transferred, although often and political interference. For the new legislative and these have been mismatched at first. Strengthening insti- institutional structures to work effectively, the judicial sys- tutional capability at the provincial and the local level is a tem must work well and fairly. Needed reforms, already major challenge in countries with federal systems, such as under way in a number of countries (including Bolivia, Argentina and Brazil, which have a legacy of overly in- Ecuador, Trinidad and Tobago, and Venezuela) include debted local governments and weak state or provincial legal education, better court administration and case flow banks. Nevertheless, examples abound of successful local management, and procedural reform, including the devel- government reforms (for example, through public-private opment of alternative dispute resolution systems outside partnerships) emerging in time of crisis, where dynamic the courts. Efforts are also needed to strengthen the crim- local leaders have been able to foster reforms on a man- inal justice system, to address the growing problem of ageable scale. crime and violence, which is partly related to drug traf- ficking in the region. Decentralization ofpower and ofspending, coupled with democratization, has dramatically trans- In the Middle East and North Africa, unemploy- formed the local political landccape in Latin Amer- ment is by far the single greatest economic and social ica, in what some have called a "quiet revolution." problem and makes government downsizing espe- A new model ofgoverning is emerging in the region. cially dftIcult. The first stages of local reform in Latin America, cov- Countries in the Middle East and North Africa assigned ering roughly the period 1983-90, created an institu- vast economic responsibilities to the state in the 1960s tional environment that nurtured a fresh generation of and 1970s. More recently, governments have trimmed office seekers, more professional and more reformist than back their roles to some extent, but much remains to be their predecessors. Leaders in scores of cities were drawn done, within government and in the state enterprise sec- by the ineluctable logic of popular participation as a nat- tor. There has also been no proportionate slimming of the ural means of sounding out the wishes of voters. This large civil services built up in those earlier years. With a same logic led to a strengthening of the fiscal link between few exceptions (Tunisia is one), regulation is excessive, as government and the governed, making more obvious the is state involvement in economic activity, and delivery of connection between the public works and service im- public services is inefficient. provements that neighborhood residents say they want, Skilled civil servants are often in short supply, and the and the payment burdens that authorities say residents delivery of public services is often inadequate. Overregula- must bear to achieve cost recovery. For instance, many tion has produced a bloated work force, pervasive interfer- local governments tie improvements to user fees or better- ence in private economic activity, and widespread corrup- ment levies, or rely on referendums. Traditional models of tion. Little effort has been devoted to streamlining agencies government take this exchangepayment for goods and and improving state efficiency. And the system's capacity THE AGENDA FOR CHANGE 167 to change has siowiy atrophied over time. Regional conflict fix capacity-building solutions have been attempted over has also distracted attention from state reform, but the end the past twenty years, with limited impact. These efforts of the Cold War, the conclusion of the Gulf War, and the have focused largely on training, on building skills, and on change in attitudes brought about by the peace process importing technical systems. They have not focused on have led governments in the region to begin to give the incentives, which come from competitive pressures, part- issue more attention. nership, and transparency, and from rule-based systems. The first steps in most countries in this region must be Reforms pushed forward too rapidly can produce new to prevent any further growth in central government risks as well: they may be blocked by those likely to lose employment, and to liberalize the economy. These steps from change, and the danger is always present that the have begun in a few cases. Because the political and social reforms will lead to fragmentation and the creation of an difficulties of reform are considerable, although not insur- institutional vacuum. mountable, one approach might be to decentralize selected International cooperation and decentralization are a services to improve responsiveness and accountability, and potentially positive trend, since they create the opportunity focus on selective state enterprise reforms while preparing to improve international collective action and the delivery the ground for more comprehensive state reform. of local public goods. The two trends, in fact, reinforce In the countries of the Organization for Economic Coop- each other. But the gains will be realized only if the pitfalls eration and Development (OECD), where the overall capa- are avoided. International cooperation will yield gains only bility of the state is high, efforts are ongoing to improve its if countries believe that they will benefit from international effectiveness. Driven in part by globalization and technol- integration. This can happen only if the risks and uncer- ogy, changes are under way that will open greater oppor- tainties of globalizationfor households, workers, the tunities in many OECD countries for government to enter poor, and the vulnerableare skillfully handled. Decen- into partnerships with the private sector and civil society, tralization will also need careful management to ensure that especially in the area of social insurance and assistance. the costs, such as loss of macroeconomic control and rising Some of the most comprehensive reforms of the state regional inequalities, do not overwhelm the benefits. worldwide have occurred in OECD countries, driven pri- In the past, it seemed clear that reform delayed was marily by the rising cost of government and citizens' growth merely deferred, that the cost of missing an oppor- demands for more value for money. It is too soon to ren- tunity to reform would be, at worst, continued stagnation. der a definitive judgment on most of these efforts. But the But a deeper understanding of the way in which declines depth of dissatisfaction with the effectiveness of the state in state effectiveness and credibilityand resistance to in these countries suggests that the search for more effec- reformcan magnify over time, coupled with examples tive government, for better public services at lower overall of collapsed states at the end of these downward spirals, cost to society, will and must continue. suggests the costs are far greater. If governments cannot grasp the nettle of improving their effectiveness and rein- Even small steps can make a big difference to the vigorating public institutions, the prospects for significant state's effectiveness, leading to better standards of improvement in economic and social welfare, in some living and opening opportunities for more reforms. countries, may be bleak indeed. The challenge is to take those small steps that can The approach of the twenty-first century brings enor- open the way to virtuous cycles. mous promise of change and reason for hope. In a world of dizzying changes in markets, civil societies, and global The two-part strategy laid out in this Report is only a forces, the state is under pressure to become more effec- broad guide for the many different agendas of reform being tive. But it is not yet changing rapidly enough to keep pursued throughout the world. Likewise, international pace. There is no unique model for change, and reforms assistance for state reform must go beyond a one-size-fits-all will often come slowly, because they involve fundamental approach, or one based on donors' preferences rather than alterations in the roles of institutions and in interactions recipients' needs. It must be based on a clearer diagnosis of between citizens and government. This Report has shown the case at hand and of how reforms would fit into broader that reform of state institutions is a long, difficult, and political and social changes going on in each country. politically sensitive task. But if we now have a better sense An important component of the reforms discussed in of the dimensions of the challenge, we are also now much this Report is institutional, and so will take time. Quick- more aware of the costs of leaving things as they are. TECHNICAL NOTE Overview education, and policy distortion in Table TN 1). The Data in Figure 1 are at current national prices. Data for heights of the other two bars are the values of the depen- the OECD countries come from Tanzi and Schuknecht dent variable obtained by evaluating the estimated regres- 1995, with updates from OECD, various years. The sion at the sample mean of all the right-hand-side vari- countries included are Austria, Belgium, France, Ger- ables except the variable of interest (credibility in Table many, Italy, Japan, the Netherlands, Norway, Spain, Swe- TN I), which is evaluated at one standard deviation above den, Switzerland, the United Kingdom, and the United the sample mean for the high category and one standard States. Data for the developing countries come from IMF, deviation below the sample mean for the low category. various years (b). Developing countries include low- and In figures where a bar diagram is based on a simple cor- middle-income countries as defined by the World Bank; relation (for example, the top left panel of Figure 3 in the the countries included vary across the period. Overview), the following methodology is used unless oth- The regional indexes in the left panel in Box 2 are erwise stated. Countries in the sample are ranked accord- derived from averages of credibility indexes for the coun- ing to their values for the variable on the horizontal axis. tries in the region. On construction of the credibility The low, medium, and high categories are then defined as index see "The private sector survey" below and the follows. The medium category includes countries within Brunetti, Kisunko, and Weder background papers. one standard deviation of the sample mean, and the The other two panels in Box 2 are based on regressions height of the bar is the average for those countries. The reported in Table TN 1. The GDP growth regression fol- remaining countries, in the upper and lower tails of the lows the instrumental variable (IV) methodology, using distribution, then make up the high and low categories, the Freedom House Index of Political Rights (Freedom respectively, and again the bar height is the average for the House, various issues) as an instrument for the level of countries in the category. credibility. The regression for the share of investment in The top left panel in Figure 3 is based on a simple cor- GDP reports ordinary least-squares (OLS) results. The relation between a policy distortion index (averages for credibility index is taken from the Brunetti, Kisunko, and 1984-93) and a corruption index (1996 data). The corre- Weder background paper (a). The index is normalized lation coefficient is 0.53 with a t-statistic of 3.79. The such that the level for the high-income OECD countries policy distortion index is taken from the Commander, equals one. The source for the remaining variables is the Davoodi, and Lee background paper. Higher values of the Commander, Davoodi, and Lee background paper. index mean more distorted policies. The corruption index The methodology used to generate the two right-hand is taken from the private sector survey conducted for this panels in Box 2 is routinely used in multiple regression Report (see below). See the technical note for Figure 5 for analysis and is used throughout the Report unless stated the construction of the policy distortion index. otherwise. The height of the vertical bar associated with The top right panel in Figure 3 is based on a regres- the medium category is the value of the dependent vari- sion, reported in Table TN2, of an index of corruption on able obtained by evaluating the estimated regression at the an index of predictability of the judiciary (higher values sample mean of all the right-hand-side terms (for exam- mean greater predictability), controlling for initial GDP ple, the constant term, credibility, initial income, initial per capita and education (data from the World Bank data TECHNICAL NOTE 19 Table TN1 Regressions of growth and Table TN2 Regression of level of corruption investment on level of credibility and other on predictability of the judiciary and crime variables and theft Regression coefficient Independent variable Regression coefficient Growth in GDP Investment- Constant Independent per capita GDP ratio (0.703) variable (IV method) lOIS method) Predictability of judiciary _955Q** _36.841*** (0.10) Constant Log of initial GDP per capita in (414) (9.03) 1990 (0.162) Level of credibility 13.44** 17.54** Log of 1990 secondary school 0.39 (0 worst, 1 best) (6.34) (7.53) enrollment (0.185)** Log of initial GDP 0.048 5.025** per capita (0.77) (1.43) Adjusted R2 0.603 Log of initial 0.255 1.109 No. of observations 59 mean years (0.85) (1.85) of schooling *4<* Significant at the 1 percent level. Policy distortion 0.256 0.628 Note: Numbers in parentheses are standard errors. Estimation is by the OLS method. (0.64) (1.51) Adjusted P2 0.264 0.674 No. of observations 33 33 ***Sjgnifjcant at the 1 percent level. The correlation coefficient for the data in Figure 4 is **significant at the 5 percent level. 0.35, with a t-statistic of 3.65. Note: Numbers in parentheses are standard errors. Figure 5 is from the Commander, Davoodi, and Lee background paper and is based on the IV regression, with the interaction term, reported in TN3. The data are from base). Data on level of corruption and predictability of the both World Bank sources and the Summers-Heston judiciary were taken from the private sector survey (see world tables (National Bureau of Economic Research below). The statistical methodology for generating the bars 1997). The state variables are the logarithm of initial is described in the technical note to Box 2. GDP per capita at 1985 international prices, educational The bottom left panel in Figure 3 is derived from a attainment (as measured by the logarithm of mean years regression reported by Evans and Rauch 1996. The de- of schooling among the working-age population), the pendent variable is corruption, data for which are taken population growth rate, and the logarithm of the share of from various issues of the International Country Risk investment in GDP at 1985 international prices. The Guide, a publication of Political Risk Services, as com- control variables include the logarithm of the share of gov- piled by the IRIS Center, University of Maryland (see also ernment consumption in GDP (government size) at 1985 Keefer and Knack 1995), and rescaled so that higher val- international prices, a policy distortion index, a measure ues indicate more corruption. The regression includes a of the quality of bureaucracy or institutional capability, constant term and, as independent variables, real GDP the average percentage change in the terms of trade, and per capita and a merit-based recruitment index. Higher decade and regional dummies. The regression with the values of this index indicate that a greater proportion of interaction term estimates the combined effect of govern- higher-level officials in the core economic agencies en- ment size and bureaucracy on growth of GDP per capita. tered the civil service through a formal examination sys- The institutional capability variable is a composite tem, and that a higher proportion of those who did not index of measures of the quality of government and draws enter through examinations have university or postgradu- on Knack and Keefer 1995 and Mauro 1995, among oth- ate degrees. The statistical methodology for generating the ers. This evaluation is put together from a set of responses bars is described in the technical note to Box 2. by foreign investors that focus on the extent of red tape The bottom right panel in Figure 3 shows the relation involved in any transaction, the regulatory environment, between, on the one hand, civil service wages relative to and the degree of autonomy from political pressure. manufacturing wages and, on the other, the index of cor- Although responses by foreign investors are likely to be ruption from the International Country Risk Guide. The biased, these series are the only currently available large- overall correlation coefficient was 0.65 with a t-statistic of scale, cross-country evaluations of the way in which 3.61. Data are taken from the Van Rijckeghem and government bureaucracies function. All responses have Weder background paper. been resealed to range from zero to one, with higher scores 170 WORLD DEVELOPMENT REPORT 1997 Table TN3 Ordinary least-squares and instrumental variable regressions with growth of GDP per capita as the dependent variable OLS method IV method Without With Without With Independent variable interaction term interaction term interaction term interaction term Constant 0. 171* * * 0.161*** 0.167*5* 0.136*** (0.022) (0.024) (0.027) (0.038) Dummy variable for 1974-83 _0.015* * * -0.015' * * _0.015*** -0.014 (0.003) (0.003) (0.004) (0.004) Dummy variable for 1984-93 -0.0 17 * * * _0.016*** -0.0 17 * * * _0.016*** (0.004) (0.004) (0.004) (0.004) Initial GDP per capita -0.019*** _0.021*** _0.021*** (0.003) (0.003) (0.003) (0.003) Initial schooling 0.003 0.003 0.003 0.003 (0.003) (0.003) (0.003) (0.003) Population growth -0.184 -0.209 -0.2 60 -0.304 (0.192) (0.192) (0.204) (0.203) lnvestment-GDP ratio 0.009*** 0.009*5* 0.008*5 0.0075* (0.003) (0.003) (0.003) (0.003) Government size _0.016*** _0.022*** _0.023*** _0.038*** (0.004) (0.008) (0.008) (0.015) Institutional capability 0.017* 0.041 0.027 0.085 (0 worst, 1 best) (0.009) (0.027) (0.010) (0.044) Government size x Institutional capability 0.014 0.033 (0.014) (0.024) Policy distortion _0.006*** _0.006* * * _0.005* * * _Q.005*** (0.002) (0.002) (0.002) (0.002) Terms-of-trade changes 0.034 0.034 0.042 0.044 (0.040) (0.040) (0.042) (0.042) Latin America dummy variable -0.0 17 * * * -0.0 17 * * * -0.015*** -0.015 * * (0.004) (0.004) (0.004) (0.004) Sub-Saharan Africa dummy variable _0.030*** -0.030 _0.028*** -0.028 * * (0.006) (0.006) (0.006) (0.006) Socialist dummy variable -0.008 -0.008 -0.013 _0.013** (0.006) (0.006) (0.005) (0.005) No. of observations 271 271 258 258 R2 0.5196 0.5213 *** Significant at the 1 percent level. ** Significant at the 5 percent level. * Significant at the 10 percent level. Note: Growth in GDP per capita is at 1985 international prices. Standard errors, corrected for heteroskedasticity, are in parentheses. a. The 92 is not an appropriate measure of goodness of fit with instrumental variable regressions. indicating better bureaucracies. The policy distortion IV estimates are very close; the discussion concentrates on index is obtained by principal component analysis of three the latter. The instruments for the policy distortion index key indicators: the degree to which an economy is open and the investment share of GDP are their own lagged (as measured by the share of trade in GDP), the degree to values from five years previously. The instrument for the which a country's currency has been overvalued (as mea- government size variable is the prediction from the IV sured by the black market premium on the exchange rate), regression for government size reported in the Comman- and the degree to which local prices have departed from der, Davoodi, and Lee background paper. All other vari- international prices. Higher values of the index indicate ables in the regression were treated as exogenous. greater policy distortion. The state variables all have the predicted signs. The co- The data for Figure 5 were pooled so as to exploit the efficient on the initial income term indicates a conditional information in ten-year averages covering the periods convergence rate of 2.1 percent per year, which is close to 1964-73, 1974-83, and 1984-93. Both OLS and IV the 2.6 percent reported by Barro and Sala-i-Martin 1995. regressions are reported in Table TN3. The OLS and the Human capital formation, as given by years of schooling, TECHNICAL NOTE 171 affects growth positively, but the effect is not statistically ernment credibility. The terms-of-trade and policy distor- significant. The investment rate, on the other hand, affects tion variables (see the note to Figure 5 above) are from the growth quite significantly. Population growth exerts a Commander, Davoodi, and Lee background paper. Data negative effect on GDP growth per capita, whereas the on level of credibility are from the private sector survey terms-of-trade variable has a positive effect. Both, how- conducted for this Report (see "The private sector survey" ever, are insignificant. There is an unambiguously nega- below). The regression also included a constant term. tive, and statistically significant, effect from government Chapter 3 consumption spending. A one-standard-deviation increase in government consumption is associated with a decline of The data in Figure 3.8 are simple averages of pensions, 0.65 percentage point per year, which is close to the 0.7 unemployment benefits, and family assistance expendi- percentage point reported by Barro and Sala-i-Martin ture shares in GDP across twenty-two OECD countries. 1995. There is also an unambiguously negative growth The series for the years 1965-79 is constructed from data effect of policy distortion, which is significant at the 1 per- in ILO, various years. The 1980-93 data are from OECD cent level. This indicates that policy distortions, as mea- 1996. sured by the index used here, will have a predictably neg- Chapter 5 ative effect on growth. However, the size of that effect, as given by the coefficient on the policy term (and control- In Figure 5.5 the dependent variable, an index of bureau- ling for other variables) is not that large, at least relative to cratic capability, is the bureaucratic quality rating variable the government size variable (0.5 percentage point per reported in the International Country Risk Guide, various year). By contrast, the institutional capability variable issues, resealed so that higher values signify higher bureau- exerts a sizeable, positive, and significant effect on growth. cratic quality. The independent variables are initial GDP Similarly, interacting the government consumption term per capita and an index of meritocracy. The regression with the institutional capability variablean attempt to also includes a constant term. coax out the implications of high values of the two vari- The correlation coefficient for the data in Figure 5.7 is ables occurring simultaneouslyyields a positive coeffi- 0.37 with a t-statistic of 10.14. The countries repre- cient. When evaluating the consequences of government sented are Botswana, Burkina Faso, Cameroon, Chad, for growth, it is not simply size that is relevant. Bad poli- Egypt, Gabon, the Gambia, Ghana, Kenya, Madagascar, cies, as represented by overvalued currencies and pervasive Malawi, Mauritania, Mauritius, Morocco, Nigeria, trade restrictions, hold down a country's growth, while the Rwanda, Senegal, Somalia, Sudan, and Togo. The sample quality of government can exert a positive effect on per- covers variation over time and across countries in pay and formance. And it is clear that countries and regions that employment. The wage variable is the central government have fared least well tend to do worst on all three indica- real average wage relative to real GDP per capita, both tors. It is the combination of government size and the measured in local currency; the employment variable quality of policy and institutions that seems to matter. is central government employment relative to the total Chapter 2 In Figure 2.2 the IV estimates from the growth regression Tabte TN4 Regression of rates of return on reported in the Commander, Davoodi, and Lee back- World Bank-financed projects on level of ground paper are used to decompose the sources of credibility and other variables growth in Sub-Saharan Africa and East Asia over the period 1964-93. These sources (explained and unex- Independent variable Regression coefficient plained) are added sequentially to GDP per capita in Sub- Constant 7.080 (12.87) Saharan Africa in 1964. 35.55** Level of credibility See the technical note to Box 2 for details regarding the (16.18) (0 worst, 1 best) top two panels in Figure 2.4. The bottom panel in Figure Change in terms of trade, 1984-93 8.078 2.4 is derived from the regression reported in Table TN4. (31.07) The dependent variable is the country-average rates of Policy distortion, 1984-93 2.481 return for a sample of 312 development projects financed (2.76) by the World Bank. Project evaluation has been com- Adjusted R2 0.088 pleted for the projects chosen. The data are from the No. of observations 30 World Bank's Operation and Evaluation (OED) data ** Statistically significant at the 5 percent level. base. The independent variables are the change in the Note: Numbers in parentheses are standard errors. Estimation terms of trade over the period 1984-93, the policy distor- is by the OLS method. tion index for 1984-93, and an index of the level of gUy- 172 WORLD DEVELOPMENT REPORT 1997 population. Employment and nominal wages are taken from Kraay and Van Rijckeghem 1995. The real average Table TN5 Regression of share of gross wage is constructed by deflating total nominal wages by investment in GDP on level and predictability the product of the consumer price index and employ- of corruption and other variables ment, Population, the consumer price index, and real Independent variable Regression coefficient GDP per capita are taken from the World Bank data base. Constant 19.523 Chapter 6 (13.49) Level of corruption _5.814** Figure 61 is based on cross-country regression results pre- (2.23) sented in Table TN5. For the countries included and def- Predictability of corruption 6.309** (2.62) initions of the corruption variables see "The private sector Log of secondary school 1.987 survey" below. The dependent variable, the investment- enrollment in 1990 (2.18) GDP ratio, is a simple average of the share of gross invest- Log of initial GDP per capita 1.149 ment in GDP over the period 1990-94. The predictabil- in 1990 (1.87) ity of corruption is a combination of predictability of Policy distortion 1.959 (1.46) outcomes and of the size of the additional payment to be made, The regression controls for initial level of education Adjusted P2 0.24 (measured as the logarithm of secondary school enroll- No. of observations 39 ment in 1990), initial income (logarithm of 1990 GDP ** Significant at the 5 percent level. per capita, measured in PPP terms), and existing policy Note: Numbers in parentheses are standard errors. Estimation is by the OLS method. distortion. The policy distortion variable is taken from the Commander, Davoodi, and Lee background paper (see the technical note to Figure 5). Data on investment, edu- cation, and initial income were obtained from World coefficients of 0.92 and 0.87, respectively, for the period Bank sources, The statistical methodology for construct- 1973-94. ing the figure is the same as for Figure 5. The level and The figures in Box 7.3 are based on simple correlations predictability of corruption are important factors in deter- and do not control for other characteristics of the villages. mining the share of investment and are significant at the However, the relationships still hold when these charac- 5 percent level. teristics are controlled for. The data on social capital are from the Social Capital and Poverty Survey (SCPS), Chapter 7 which was carried out in rural Tanzania in April and May The index of democracy in Figure 7.1 is based on Polity 1995 as part of a larger World Bank participatory poverty III data developed by Jaggers and Gurr 1996; regional assessment exercise. Expenditure data are from the same groupings follow standard World Bank classifications. source, although that part of the survey covered house- The index is derived from "institutionalized democracy" holds in only fifty-three of the eighty-seven clusters. The and "institutionalized autocracy" indicators for each year village-level social capital index is the average number of from 1800 to 1994, for all independent countries with groups (for example, churches, women's groups, farmers' populations greater than 500,000 in the early 1990s (177 groups) to which respondents belonged, multiplied by an countries in all). The index is calculated by subtracting index of the average characteristics of those groups along the latter indicator from the former. Each indicator con- three dimensions: kin heterogeneity, income heteroge- sists of five components: competitiveness of political par- neity, and group functioning. The data on school quality ticipation, regulation of political participation, competi- are from the Human Resource Development Survey tiveness of executive recruitment, openness of executive (HRDS), carried out in Tanzania in 1993, which can be recruitment, and constraints on the chief executive. Each matched at the cluster level with the eighty-seven clusters of the components is scored according to Jaggers and from the SCPS. The school quality index is a cluster-level Gurr 1995 (p. 472). Both indicators are additive on an average derived from respondents' answers to questions eleven-point scale from 0 to 10. The advantage of using on the relative importance of five attributes of schools, the difference between the two indicators is that it shows and their assessment of the quality of their local school a regime type along a political continuum of which along those same five dimensions. The HRDS was a democracy (+ 10) and autocracy (-10) represent the two nationally representative survey of 5,000 households in ends. The index of democracy is highly correlated with Tanzania. The survey was a joint effort undertaken by the the Freedom House index of political rights and civil Department of Economics of the University of Dar es rights (Freedom House, various issues), with correlation Salaam, the Government of Tanzania, and the World TECHNICAL NOTE 173 Bank, and was funded by the World Bank, the Govern- of constitutional structure, type of government, and ideo- ment of Japan, and the British Overseas Development logical orientation of the legislature (data for the last two Agency. See Ferreira and Griffen 1995. are from Alesina and Perotti 1995). Constitutional struc- ture is measured as the veto power given under the con- Chapter 9 stitution to minorities and interest groups to block social Figure 9.2 is based on the generalized least squares (GLS) legislation. Higher values of the index imply less obstruc- regression reported in Table TN6. The regression captures tive capacity. A positive coefficient implies that countries the effects of initial conditions and institutional, demo-. with low veto power are more likely to indulge in welfare graphic, and economic variables on welfare spending. expansion. Government is classified into six types ranging The dependent variable is government spending on from single-party to multiparty minority. Ideological ori- pensions, unemployment, and family assistance as a share entation is classified as right-wing, right-center, balanced, of GDP (see the technical note to Figure 3.8). Initial con- left-center, or left-wing. A positive coefficient for govern- ditions are captured by the initial Gini index for income ment type implies that single-party governments are less distribution (Deininger and Squire 1996) and the time prone toward expansionary policies; for ideological orien- elapsed since the start of the social security program in tation it implies that left-wing governments are more each country (U.S. Department of Health and Human prone to such policies. The percentage of population Services 1994). The institutional variables include an above 65 years of age (aging), number of children 14 years indicator of bureaucratic efficiency (institutional capabil- and under, unemployment rate, real GDP per capita, and ity, from the Commander, Davoodi, and Lee background inflation are taken from the World Bank's Social Indica- paper) and three indicators of the political environment. tors data base; OECD, various years; ILO 1986 and 1994; Higher scores imply greater efficiency. Although efficient IMF, various years (b); and the World Bank Economic bureaucracies are more likely to curb fraud and abuse, and Social Data Base. A positive coefficient is expected for they are also more likely to grant benefits to all eligible the aging and unemployment rate variables. Finally, the beneficiaries. Thus, the sign of the coefficient on the more affluent the country, the bigger its welfare bill. This bureaucratic efficiency variable depends on the relative effect is picked up by a positive coefficient for real GDP importance of these two effects. A positive coefficient per capita. implies that more efficient governments provide greater A new variable, "residual average welfare spending," is welfare coverage. Political environment indicators consist defined by subtracting, from the sample average of welfare spending, the constant term of the regression and the fol- lowing term: (coefficient x sample average of independent variables not significant at the 10 percent level or less). Table TN6 Regression of welfare spending on The contribution of each variable to welfare spending is constitutional structure and other variables obtained by multiplying the estimated coefficient on each Regression Standard variable by the sample average of that variable and divid- Independent variable coefficient error ing the result by residual average welfare spending. The Constant 6.37 4.229 reported decomposition of welfare spending is robust to Constitutional structure 4.40 2.189' retaining only those variables in the regression that were Aging 0.583 0.082*** significant at the 10 percent level or better. Unemployment rate 0.361 0.029*** The back-of-the-envelope calculation for the reduction Type of government 0.227 0.063*** 0.521*** in distortions resulting from tax reform in Pakistan is Efficiency of bureaucracy 1.76 GDP per capita x l0 0.94 0.495* based on the so-called Harberger triangle. The deadweight Ideology of government 0.04 0.045 loss (DWL) equation is given by: Inflation rate 0.01 0.017 No. of children x 108 2.37 4.09 DWL = (0.5) x (t2) x (1 - t)1 x n7x TI Starting year of program 0.006 0.039 Gini index 0.004 0.109 where: R2 0.63 t = tax rate No. of observations 365 = compensated demand elasticity Significant at the 1 percent level. TI = taxable income. ** Significant at the 5 percent level. Significant at the 10 percent level. Note: Estimation is by the generalized least-squares method. The prereform tax rate used is 60 percent, and the post- The R2 does not have all the properties of the OLS P2. reform rate 30 percent. The compensated demand elastic- ity used is 0.5, the same as in Feldstein 1995. The relevant 174 WORLD DEVELOPMENT REPORT 1997 ratio of taxable income to GDP is 7.22 percent. Based on of senior management's time is spent dealing with legal these figures, the deadweight loss reduction (prereform requirements. minus postreform) associated with a revenue-neutral Uncertainty created by lack ofgovernment efficiency in pro- reform is estimated at 1.4 percent of GDP. viding services. These questions concentrate on whether and how efficiently the government delivers certain basic The private sector survey services such as mail, health care, telephone service, and Why this survey? roads. A vast number of anecdotal reports document the harm done to private sector development by uncertainty about Implementation and results laws, policies, and regulations. Examples are Dc Soto 1989 The questionnaire was originally written in English. on the problems of informal firms in Peru, the description However, worldwide distribution required its translation by Klitgaard 1990 of the uncertainties of doing business in into several major languages: French, German, Portu- Equatorial Guinea, and the analysis of institutional uncer- guese, Russian, and Spanish. Wherever possible the ques- tainty in Nicaragua by Borner, Brunetti, and Weder 1995. tionnaires were administered in one of these languages or Adequate data for such an analysis have been hard to in English. At times, however, it was vital to provide come by, however. The private sector survey conducted translations for a single country. This was done in the for this Report sought to fill that gap by creating an inter- cases of Albania, Bulgaria, the Czech Republic, Hungary, nationally comparable data set, for a broad cross section of Italy, Poland, the Slovak Republic, and Turkey. countries, on different aspects of institutional uncertainty The survey was implemented between August 1996 as perceived by private entrepreneurs. and January 1997. At its conclusion sixty-nine countries had participated. These include the following: industrial The survey questionnaire countries, Austria, Canada, France, Germany, Ireland, The questionnaire first asks about some general character- Italy, Portugal, Spain, Switzerland, the United Kingdom, istics of the responding firm. Five different dimensions are and the United States; South and Southeast Asia, Fiji, considered: size (fewer than 50 employees, between 50 India, and Malaysia; Middle East and North Africa, Jor- and 200 employees, or more than 200 employees); the dan, Morocco, and the West Bank and Gaza; Central and firm's line of business (manufacturing, services, or agri- Eastern Europe, Albania, Bulgaria, the Czech Republic, culture); its headquarters location (capital city, other large Estonia, Hungary, Latvia, Lithuania, Macedonia, Poland, city, or small city or countryside); the presence or absence the Slovak Republic, and Turkey; Latin America and the of foreign capital participation; and whether the firm Caribbean, Bolivia, Colombia, Costa Rica, Ecuador, exports its products. Jamaica, Mexico, Paraguay, Peru, and Venezuela; Sub- The main part of the questionnaire consists of twenty- Saharan Africa, Benin, Cameroon, Chad, the Congo, five multiple-choice questions, which are grouped in five Côte d'Ivoire, Ghana, Guinea, Guinea-Bissau, Kenya, sections, each with its own focus. The questionnaire is Madagascar, Malawi, Mali, Mauritius, Mozambique, divided into the following five sections: Nigeria, Senegal, South Africa, Tanzania, Togo, Uganda, Zambia, and Zimbabwe; Commonwealth of Independent Predictability of laws and policies. These questions seek States, Armenia, Azerbaijan, Belarus, Georgia, Kazakstan, to assess the uncertainties created by the lawmaking the Kyrgyz Republic, Moldova, Russia, Ukraine, and process. Uzbekistan. Political instability and security ofproperty. These ques- The companies selected ran the gamut of firm size, tions ask about the uncertainties that arise from both geographic location within the country, and type of busi- regular and irregular transfers of government power. ness, and companies both with and without foreign capi- Government-business interface. The questionnaire lists tal participation were well represented. Questionnaires fifteen areas where the firm is confronted with govern- were sent by mail where possible, but were delivered by ment action and asks it to evaluate the degree to which hand in some countries where mail delivery was unreli- each of these areas creates obstacles to doing business. able. Table TN7 provides details on response rates. Con- Law enforcement and bureaucratic red tape. These ques- sidering other experience with mailed surveys, the high tions focus on the degree of corruption and whether overall response rate for the mailed survey in developing it is a predictable transactions cost or a source of un- countries (30 percent) is remarkable. The high rate can be certainty. A problem in analyzing the responses is, of attributed to two factors: the fact that the survey raised course, firms' reluctance to openly admit that they pay questions of great concern for local businesspeople, and bribes. In addition, the questionnaire asks directly the fact that the survey was sponsored by an international whether uncertainties in dealing with the state have sti- organization with considerable name recognition in devel- fled planned investment projects, and what percentage oping countries. TECHNICAL NOTE 175 rable TN7 Responses to the private sector survey by region No. of questionnaires returned No. of countries No. of firms Region or group surveyed surveyed Average Median Minimum Maximum ndustrial countries 11 254 23 20 14 56 south and Southeast Asia 3 139 46 45 41 53 AiddIe East and North Africa 3 109 36 42 15 52 EE 11 771 70 70 46 114 _atin America and Caribbean 9 474 53 47 17 87 Sub-Saharan Africa 22 1,288 59 48 13 124 lS 10 650 65 62 31 91 All developing countries 58 3,431 59 51 13 124 All countries 69 3,685 53 50 13 124 Because of budget and time constraints, in some coun- tries' investment climate, which concentrated entirely on tries the responding firms did not represent a random the perceptions of multinational firms. sample of those to which questionnaires had been distrib- uted. In other countries, political and economic condi- Construction of the credibility indicator tions allowed only limited geographical coverage. On The credibility indicator was designed as a broad measure average, however, the survey achieved its goal of fifty of the reliability of the institutional framework as per- responses per country. ceived by private entrepreneurs. It encompasses several dif- Almost half of the firms were small (fewer than fifty ferent sources of uncertainty in the interaction of govern- employees); the rest were divided more or less equally ment and the private sector and summarizes these into one between the two larger firm sizes. The survey intended global indicator. The credibility index is constructed as the that the companies in the sample represent a variety of simple mean of the average answers to five subindicators, geographic locations within each country. Firms located which is then normalized such that the index for the high- in the capital city constituted about half of those respond- income OECD countries equals one: ing. It is encouraging, however, that management of almost one-quarter of the firms was located in a small city Predictability of rulemaking, or the extent to which or in the countryside. entrepreneurs have to cope with unexpected changes in The aggregate results, however, hide considerable vari- rules and policies, whether they expect the government ation within countries. The share of firms located in the to stick to announced major policies, the degree to capital city varies between 100 percent and 0 percent. which entrepreneurs are usually informed about impor- Such variation can be explained by the distribution of pri- tant rule changes, and whether they have an opportu- vate businesses over country territory. In some former nity to voice concerns when planned changes affect republics of the Soviet Union more than half of registered their business businesses are in the capital. In other countries the socio- Subjective perception of political instability, or whether economic and political situation limited distribution of changes in government (constitutional or unconstitu- the questionnaire to the more remote parts. In some tional) are perceived to be accompanied by far-reaching countries the unreliability of the mails made it infeasible policy surprises that could have serious effects on the to distribute questionnaires to remote places and have private sector them returned in a timely fashion. Security of persons and property, or whether entrepre- Services and manufacturing were represented about neurs feel confident that the authorities will protect equally among respondents, but responses from agricul- them and their property from criminal actions, and tural enterprises were relatively few. This bias can be whether theft and other forms of crime represent seri- explained by geography: more than three-quarters of the ous problems for business surveyed firms were headquartered in the capital city or Predictability ofjudicial enforcement, or the degree of another large city, where few agricultural firms are located. uncertainty arising from arbitrary enforcement of rules Firms were evenly distributed with respect to foreign by the judiciary, and whether such unpredictability capital participation and access to foreign markets. Two- presents a problem for doing business thirds of the surveyed companies reported no foreign par- Corruption, or whether it is common for private entre- ticipation. The results therefore contrast with those of preneurs to have to make some irregular additional pay- other earlier attempts at subjective measurement of coun- ments to get things done. BIBLIOGRAPHICAL NOTE REPORT HAS DRAWN ON A WIDE RANGE OF tance at various stages. Helpful contributions on individ- THIS World Bank documents and on numerous outside ual chapters are listed under the chapter acknowledgments. sources. World Bank sources include ongoing research as The survey of private sector firms conducted specially well as country economic, sector, and project work. These for this Report was made possible through the assistance and other sources are listed alphabetically by author or and cooperation of staff in the operational complex of the organization in two groups: background papers commis- World Bank. Particular thanks go to the coordinators of sioned for this Report and a selected bibliography. The the survey in the regional vice presidencies as well as the background papers, some of which will be made available resident representatives and the staff of rhe resident mis- through the Policy Research Working Paper series, and sions in the participating countries. the rest through the World Development Reportoffice, syn- A wide range of consultations was undertaken for the thesize relevant literature and Bank work. The views they Report. We particularly wish to thank the following orga- express are not necessarily those of the World Bank or of nizations for arranging consultation meetings: the Re- this Report. search Institute for Development Assistance, Overseas In addition to the principal sources listed, many per- Economic Cooperation Fund, Tokyo; the German Min- sons, both inside and outside the World Bank, provided istry of Cooperation; the Norwegian Ministry of Foreign valuable advice and guidance. Special thanks is owed to Affairs; the North-South Institute, Ottawa; the National Gregory Ingram, Arturo Israel, Ravi Kanbur, and Michael Council for Applied Economic Research, New Delhi; the Walton. Caroline Anstey and Hans Jurgen Gruss helped African Economic Research Consortium, Nairobi; the facilitate inputs and consultations from many sources. Overseas Development Agency, London; the Global Valuable comments and contributions were also provided Coalition for Africa, Addis Ababa; and InterAction, by Sri-Ram Aiyer, Mark Baird, Shahid Javed Burki, Un Washington. Dadush, Partha Dasgupta, Gloria Davis, Shanta Devara- We wish to thank in Tokyo: Yuan Gangming, Kaoru jan, Mamadou Dia, Jessica Einhorn, Gunnar Eskeland, Hayashi, Mr. Hisatake, Naoko Ishii, Shigeru Ishikawa, Francisco Ferreira, Cesar Gaviria, Roger Grawe, Jeffrey Shinichi Jin, Yutaka Kosai, Isao Kubota, Toru Nakanishi, Hammer, Ricardo Hausmann, Enrique Iglesias, Edmundo Nobutake Odano, Tetsuji Okazaki, Yoshio Okubo, Toru Jarquin, Robert Klitgaard, Geoff Lamb, Moises NaIm, Shinotsuka, Masaki Shiratori, Akira Suehiro, Shigeki Te- Gobind Nankani, John Nellis, Richard Newfarmer, jima, Prof. Juro Teranishi, and Yoshio Wada; in Brusse& Guillermo Perry, Guy Pfeffermann, Robert Picciotto, Dominique Be, Brunet Bernard, Noel Coghlan, P. De- Boris Pleskovic, Stephen Pursey, Sarath Rajapatirana, Mal- fraigne, M. de Lange, L. de Richemont, Daniel Guyader, colm Rowat, Salvatore Schiavo-Campo, Nemat Shafik, Maurice Guyader, Ditte Juul Jorgensen, Maral LeRoy, Ibrahim Shihata, Mary Shirley, I. J. Singh, Andrew Steer, Francoise Moreau, L. R. Pench, Regine Roy, G. Tebbe, A. Nicholas Stern, Maurice Strong, Roger Sullivan, Vinod Tincani, J. Vignon, and Rutger Wissels; in the United Thomas, Jacques van der Gaag, Paulo Vieira da Cunha, Kingdom:Mandeep Bains, Bill Baker, Richard Bailey, Kate Steve Webb, iMan Winters, and John Williamson. Bruce Bayliss, Sarah Bernard, Graham P. Chapman, Anne Coles, Ross-Larson, Meta de Coquereaumont, Paul Holtz, and Paul Collier, Sean Collins, Rosalind Eyben, Mick Foster, Alison Strong provided valuable editorial advice and assis- Peter Grant, Mr. Greif, P. Holden, Tony Killick, Robert BIBLIOGRAPHICAL NOTE 177 Lasiett, Andrew Leslie, Deborah McGurk, Dino Merotto, med El Sayed Scum, Mohamed Aboul Enein, Samiha Mick Moore, Peter MountfIeld, Rachel Phillipson, Trevor Fawzy, Ahmed Gweily, Heba Handoussa, Taher Helmy, Robinson, Sally Taylor, Sandra Waliman, Jon Wilms- Mohamed Mahmoud El Imam, Mohamed Lofty Man- burst, and Geoffrey Wood; in Stockholm: Stefan Fölster, sour, Omar Mohanna, Mohamed Ozaip, Ghada Ragab, Jorgen Holmqvist, Erik Johnsson, Assar Lindbeck, Eva Yasser Sobhi, Arvind Subramanian, Fouad Sultan, and El Lindstrom, Per Molander, and Joakim Palme; in Ferney- Sayed Yasseen; in Oslo: Ole Winkler Andersen, Christian Voltaire, France Hans Engelberts, Elie Jouen, and Mike Friis Bach, Ingrid Braenden, Adne Cappelen, Arne Disch, Waghorne; in Bonn: Friedrich W. Bolay, Hans-Gert Thorvald Grung Moe, Tor Halvorsen, Trond Folke Lind- Braun, Hartmut Elsenhans, Ingrid Hoven, Ernst-J. Ker- berg, Desmond McNeill, Lars Mjoset, Frode Neergaard, busch, Elmar Kleiner, Gudrun Kochendörfer-Lucius, Poul Engberg Pedersen, Erik Reinert, Reiulf Steen, Astri RolfJ. Langhammer, Hildegard Lingnau, Peter Molt, Mr. Suhrke, and Lars Udsholt; in the Netherlands M. Bienefelt, Preuss, Dirk Reinermann, Hans-Bernd Schafer, Mr. K. Blekxtoon, J. de Groot, L. de Maat, K. Doornhof, J. Schröder, Christian Sigrist, Klaus Simon, Albrecht Enneking, J. Faber, H. Gobes, A. C. M. Hamer, J. P. Stockmayer, Franz Thediek, Josef Thesing, Dr. Tittel- Ramaker, F. Roos, C. Storm, R. J. Tjeerdsma, A. van't Gronefeld, Kiemens van de Sand, and Peter Wolf; in Parir Veer, F. Ph. M. van der Kraaji, G. van Dijk, G. J. J. M. Sophie Bismut, Catherine Bourtembourg, Christian Cha- van Empel, F. D. van Loon, L. van Maare, A. van Raver- vagneux, Jean Coussy, Maximin Emagna, Bénédicte Etien, stein, and M. van Wier; from Belgium: Guido Dumon, Ulrich Hiemenz, Etienne Le Roy, Turkia Ould-Daddah, Dany Ghekiere, Luc Hubloue, Thomas Lievens, and Guy Michel Pipelier, and Jean Pisani Ferry; at a meeting of Schorochoff; at the World Health Organization: D. Bett- NGO representatives in London: Graham Bray, Joji Carino, cher, A. Moncayo, S. Sapirie, J. Tulloch, J. Visschedijk, Marcus Coichester, Harriet Goodman, Andrew Gray, Car- and Derek Yach; at the Office of the United Nations High oline Harper, Rob Lake, Christine Lippai, Brendon Mar- Commissioner for Refugee Jamal Benomar and Eric Mor- tin, Arthur Neame, Henry Northover, Helen O'Connell, ris; inAddisAbaba: Members of the Economic Committee Robin Poulton, Mohammed Sulliman, Sabjit Tohal, Kitty of the Global Coalition for Africa; in the West Bank and Warnock, Alex Wiiks, Christian Wisskirchen, and Jessica Gaza: Samir Abdallah, Hatem Halawani, Nabil Kassis, Au Woodroffe; at the European Commission: Roderick Abbott, Mahmoud Khadr, Mohammad Zuhdi Nashashibi, Yousif Christoph Bail, Chris Boyd, Gunther Burghardt, Carlos Nasser, and Mohammad Shtayyeh; ata meeting in Cancün, Camino, Jim Cloos, Robert Coleman, Carlos Costa, Pierre Mexico: José Afonso, Pedro Aguayo, Kenny Anthony, Defraigne, Xavier de Larnaudie-Eiffel, Joly Dixon, Nicoiás Ardito-Barletta, Edgardo Boeninger, Juan Bour, Michael Green, Alexander Italianer, Horst Krenzler, Ed Hernán Buchi, Ruben Caries, Alejandro Carrillo, Tarsicio Kronenburg, Francois Lamoureux, Rene Leray, Jean- Castañeda, Pelegrin Castillo, José Dagnino Pastore, Francois Marchipont, Stefano Micossi, Agne Pantelouri, Andrés Dauhajre, Diego de Figueiredo Moreira Neto, Bernard Petit, Juan Prat, Giovanni Ravasio, Alexander Aiberto DIaz Cayeros, Haydee Garcia, Rudolf Hommes, Schaub, Steffen Smidt, Michel van den Abeele, Robert Tasso Jereissati, Arnoldo Jiménez, Eduardo Lizano, Verrue, Jerome Vignon, Heinrich von Moltke, and Jorg Thereza Lobo, RoIf Luders, Gabriel Martinez, Néstor Wenzel; at Georgetown University, Washington: Daniel Martinez, Helen McBain, Ambier Moss, Marthe Muse, Brumberg, Marsha Darling, Bruce Douglas, Steven King, Arturo Nüñez del Prado, Tomás Pastoriza, Ramón Carol Lancaster, Marilyn McMorrow, Dennis McNamara, Pinango, Fernando Romero, Luis Rubio, Ricardo Sama- Gwendolyn Mikell, Howard Schaeffer, and Dan Unger; at niego, Cezley Sampson, Antonio Sancho, Enrique Ves- the Washington College of Law, American University Clau- covi, and Eduardo Wiesner Durán; in India: Swaminathan dia Martin and Rochas Pronk; at a World Bank meeting Aiyar, Yoginder Alagh, Surjit Bhalla, Onkar Goswani, R. with NGO representatives: Peter Bachrach, Deborah N. Maihotra, Rakesh Mohan, and Pai Panandikar; in Brautigam, Jim Cox, George Devendorf, Jack Downey, Bern: Franz Blankart, Thomas Greminger, Beat Kappeler, Justin Forsyth, Jo Marie Griesgraber, Wendy Grzywacz, Luzius Mader, and Mathias Meyer; at the African Economic Kari Hamerschlag, Carola Kaps, Meg Kinghorn, Michael Research Consortium: Ibrahim Elbadawi and Benno Ndulu; Kronthal, Carolyn Long, Claudia Martin, Carmen Mon- in Berlin: Heinz Buhler, Alexander Friedrich, Götz Link, ico, Joe Muwonge, Gabriel Negatu, Michal Nehrbass, Theo Sommer, and Carl-Dieter Spranger; at the Interna- Carolyn Reynolds, Mildred Robbins Leet, Bruce Robin- tional Labour Organisation: Katherine Hagen and Stanley son, Berta Romero, James Rosen, Frances Seymour, Taylor; at the International Confederation of Free Trade Gmakahn Sherman, Carla Simon, Andrea Soccobo, Julia Unions: Gemma Adaba; in Canada: Isabella Bakker, Man- Taft, Nicolas van de Walle, Nick Vanedwaild, and Chuck fred Bienefeld, Jim Carruthers, G. Shabbir Cheema, Roy Woolery; in Cairo: Ismail Sabry Abdallah, Mamdouh El Culpeper, Nasir Islam, Devesh Kapur, Bahman Kia, Peter Beltagy, Ahmed Galal, Abdel Fattah El Gebalyl, Moha- Larson, Caroline Pestieau, and Alison Van Rooy; in Addis 178 WORLD DEVELOPMENT REPORT 1997 Ababa. Addis Anteneh, Tedenekialesh Assfaw, Asrat i-Martin 1995, Bosworth, Collins, and Chen 1995, Deva- Bekele, Befekadu Degefe, Getachew Demeke, Tekalign rajan, Swaroop, and Zou 1996, Kormendi and Meguire Gedamu, Murtaza Jaffer, Rehenia Jingo-Kakonge, Te- 1985, Landau 1986, Ram 1986, and Siemrod 1995. shome G. Mariam, Berhane Mewa, Gabriel Negatu, Flor- Knack and Keefer 1995 look explicitly at the role of insti- ence Nekyon, Tom L. Torome, and Kiffle Wodajo. tutions. Mauro 1995 looks at the consequences of cor- For this consultation process we wish to thank in par- ruption for growth. The discussion of credibility draws on ticular Patricia Dufour, Tomoko Hirai, Mika Iwasaki, Ali Borner, Brunetti, and Weder 1995, and the results re- Khadr, Geoff Lamb, S. Miyamura, Fayez Omar, Sudar- ported are derived from the credibility survey conducted shan Gooptu, Claudia Von Monbart, and Spiros Voy- for this Report, which is described in detail in the Bru- adzis and several Executive Directors of the World Bank netti, Kisunko, and Weder background papers. The gen- and their staff based in Washington. eral discussion on institutions draws on North 1990, Olson 1996, and Dia 1996. The discussion of evolution Chapter 1 in economists' thinking about growth draws on Barro and A bold and panoramic look at the evolution of the state Sala-i-Martin 1995, Solow 1956, Dasgupta 1995, and over the past thousand years is provided in Tilly 1990. Drèze and Sen 1989. Classic texts that deal with the state and its management include Machiavelli's The Prince and Kautiliya's Arthashas- Chapter 3 tra (see Kangle 1965). Helm 1989 analyzes the shifting This chapter benefited from valuable contributions and boundaries of the state after 1945. DIaz Alejandro 1988 suggestions from Richard Ball, Jeanine Braithwaite, Lionel considers shifting fashions in development in the Latin Demery, Jeffrey Hammer, Estelle James, Emmanuel American context. Tanzi and Schuknecht 1995 provide a J imenez, Maureen Lewis, Geoffrey Shepherd, Canlos Silva, recent evaluation of the effectiveness of states. World Bank Kalanidhi Subbarao, and Dominique van de Waite. 199 lb provides an overview of competing paradigms of The discussion on shared growth is based on Aoki, development and the differing roles accorded to the state. Murdoch, and Okuno-Fujiwara 1995 and Ishikawa 1990. Teranishi and Kosai 1993 provide a comprehensive review Box 3.2 is drawn from Guerrero 1996. The discussion of of Japanese economic policies. IMF 1996 provides a mining in Nevada is from Libecap 1996. The material on review of the fiscal policy issues facing developing coun- land titling was provided by Klaus Deininger, with addi- tries. A comprehensive review of the arguments for state tional information from Feder and Nishio 1996. Berry intervention is contained in Stiglitz 1994. See Mueller and Levy 1994 describe patterns of export marketing by 1989 for an alternative approach. Box 1.1 is adapted from Indonesian firms. Box 3.3 builds on Stone, Levy, and Sills 1968, Gould and KoIb 1964, and Kuper and Kuper Paredes 1996. 1996. Box 1.2 is summarized from Stiglitz 1996. Box 1.3 The section on the policy environment draws on draws from a background note by Swaminathan Aiyar. World Bank 1994a for its discussion of price distortions in The historical section benefited from contributions by African agriculture. The discussion of changes in the way Emma Rothschild and draws as well on the background developing countries raise tax revenue is based on infor- paper byAron, Elbadawi, and Ndulu. Box 1.4 is summa- mation provided by Vinaya Swaroop. The discussion of rized from Stigiitz 1986. Table 1.1 benefited from valu- mechanisms for establishing fiscal and monetary credibil- able inputs from Jeffrey Hammer. ity draws on the background papers by Alesina and Bait. The discussion on patterns of public expenditure Chapter 2 draws on Hammer 1997, Pradhan 1996, and World Bank Peter Knight contributed material for Box 2.1. Figure 2.1 1994c. The discussion of benefit incidence of public draws on Alston 1996. On measurement of the state see spending draws on contributions by Lionel Demery and Gemmell 1993 and Lindauer 1988. The standard refer- on van de Walle and Nead 1995. The discussion of the ence on purchasing power parity prices is Summers and historical role of the private sector draws on Psacharopou- Heston 1991. The empirical work in this chapter is los and Nguyen 1997 and van der Gaag 1995. Box 3.6 is based on the Commander, Davoodi, and Lee background based on van der Gaag 1995. paper. There is an extensive literature on the factors deter- Box 3.7 is based on Coloma 1996. Box 3.8 was pro- mining government size: see, among others, Borcherding vided by Nisha Agrawal. 1985, Buchanan 1977, Courakis, Moura-Roque, and Tn- dimas 1993, Lybeck 1986, Meltzer and Richard 1981, Chapter 4 Oxley 1994, Peacock and Wiseman 1961, Ram 1987, and Jcan Adcn, Gerard Caprio, Cheryl Gray, Luis Guasch, Rodrik 1996. On the effects of government size on Robert Hahn, Gordon Hughes, Pablo Spiller, and growth and other indicators of well-being see Alesina and Andrew Stone made valuable contributions to this chap- Perotti 1995, Anand and Ravallion 1993, Barro and Sala- ter. Box 4.1 was prepared by Andrew Stone. BIBLIOGRAPHICAL NOTE 179 The discussion of privatization and liberalization draws tributed Box 5.1. Box 5.2 is from Campos and Pradhan heavily on Galal and others 1994, World Bank 1995c, 1996. The discussion on policymaking as the brains of and the Guasch and Hahn background paper. Lau and government draws on Israel 1990. The discussion of pol- Song 1992 delineate the evolution of public-private own- icymaking in Poland and Hungary draws on Nunberg ership in the Republic of Korea and Taiwan (China). forthcoming. The frameworks for the discussion of financial, utili- The section on service delivery draws on information ties, and environmental regulation draw heavily from, provided by Nicholas Manning and on Israel 1997. Box 5.3 respectively, Caprio 1996, Levy and Spiller 1994, and is from World Bank 1996b. Box 5.4 draws from van der Afsah, Laplante, and Wheeler 1996. The discussion of Gaag 1995 and Heyneman forthcoming. Box 5.5 is from banking supervision draws heavily on Polizatto 1992. Box van der Gaag 1995. The study on performance contracts 4.3 is drawn from Sheng 1992. The World Bank study on with public enterprises is World Bank 1 995c. The compar- episodes of bank insolvency is Caprio and Klingebiel ison of the Indian and Korean irrigation systems draws 1996. The description of price-cap regulation in the upon Wade 1994. The section on financial and manage- United Kingdom is based on Spiller and Vogelsang 1996. ment controls draws on information from Gary Reid, The information on environmental regulation in indus- Chetana Neerchal, George Russell, and Jim Wesberry. trial countries is from Lovei and Weiss 1996 and Rose- The section on building capable, motivated staff bene- Ackerman 1995. The shortcomings of top-down environ- fited from extensive contributions by Barbara Nunberg. mental regulation are examined in Margulis 1996. The section on meritocratic recruitment and promotion Stiglitz and Uy 1996 probe some East Asian ap- draws from Evans 1995 and Campos and Root 1996. Box proaches to maintaining bank solvency, and Saunders and 5.6 is from Nunberg 1995. The discussion on pay and Wilson 1995 synthesize some historical experiences with employment, including the study on wage decompression contingent liability in banking in the Western world. The in Africa, draws on Lindauer and Nunberg 1994; the anal- discussions of telecommunications regulation in Jamaica ysis of the decline of public sector wages is derived from and the Philippines are from Spiller and Sampson 1996 Haque and Sahay 1996. The study of civil service reform and Esfahani 1996. The discussion of Indonesia's experi- efforts over 1981-91 is reported in World Bank 1991a. ence with environmental regulation is adapted from Box 5.7 is from Tendler 1997. Afsah, Laplante, and Makarim 1996. Box 4.5 is from Metropolitan Environment Improvement Program 1996. Chapter 6 The section on industrial policy builds on Aoki, Mur- This chapter has benefited from valuable inputs, sugges- doch, and Okuno-Fujiwara 1995, Ohno 1996, World tions, and comments by Ladipo Adamolekun, Robert Bank 1993, Levy and others 1994, and Humphrey and Bates, Ed Campos, Maria Dakolias, Matthew McCub- Schmitz 1995. Box 4.6 is adapted from Okazaki 1997. bins, Elena Panaritis, Andres Rigo Sureda, Susan Rose- The Philippine experience with capital-intensive projects Ackerman, Kenneth Shepsle, Mike Stevens, Andrew is summarized in World Bank 1987. Lee's background Stone, and Douglas Webb. paper lays out the experience with infrastructure in the The section on the judiciary is based on the Webb Cholla region of the Republic of Korea. Tendler 1997 background paper. The study on property rights and the examines public procurement in Ceará State, Brazil. stability of the regime is Clague and others 1996. Box 6.1 was provided by Elena Panaritis. The section on corrup- Chapter 5 tion draws on the Rose-Ackerman background paper. The This chapter has benefited from valuable contributions, analysis of the predictability of corruption has benefited suggestions, and comments from Ladipo Adamolekun, Ed from discussions with Ed Campos. Boxes 6.2, 6.3, and 6.4 Campos, Migara da Silva, Giulio de Tommaso, Roger were provided by Susan Rose-Ackerman, drawing on Brett Grawe, Jeffrey Hammer, Malcolm Holmes, Arturo Israel, 1993 and Ruzindana 1995 (Box 6.3) and Manion 1996 Klaus Konig, Alexander Kotchegura, Patricia Langan, and Quah 1993 (Box 6.4). There is a vast literature on Nicholas Manning, Ernesto May, Julie McLaughlin, corruption, which includes Klitgaard 1988, Mauro 1995, Amitabha Mukherjee, Vikram Nehru, Chetana Neerchal, Rose-Ackerman 1978, and Shleifer and Vishny 1993. Barbara Nunberg, Gary Reid, Susan Rose-Ackerman, George Russell, Claude Salem, Salvatore Schiavo-Campo, Chapter 7 Mary Shirley, Mike Stevens, Roger Sullivan, Jim Wes- This chapter has benefited from valuable contributions berty, and David Wood. and comments from Junaid K, Ahmad, Dan Aronson, The chapter draws on the Campos and Pradhan back- Katherine Bain, Ela Bhatt, Richard Bird, Tim Campbell, ground paper, the Schiavo-Campo, de Tommaso, and John Clark, Peter Evans, Marianne Fay, Deon Filmer, Mukherjee background paper, and background notes by AshrafGhani, Jim Hicks, Michael Layer, Deepa Narayan, Nicholas Manning and Gary Reid. Mike Stevens con- Vikram Nehru, Samuel Paul, Lant Pritchett, Lester Sala- 180 WORLD DEVELOPMENT REPORT 1997 mon, David Sewell, Anwar Shah, Jerry Silverman, section on basic research makes use of findings reported in Albrecht Stockmayer, and David Wildasin. Ad Hoc Committee on Health Research Relating to The section on voice and participation draws heavily Future Intervention Options 1996. Box 8.2 is based on on ideas developed in Evans 1996a and 1996b, Pardey and others 1996. The discussion of international Hirschman 1970, and Montgomery 1988, Box 7.1 is environmental agreements is based on material supplied based on Linz, Lipset, and Pool n.d.,Weaver and Dickens by Laurence Boisson des Chassournes; examples of global 1995, and polls conducted by Europinion, Brussels, and environmental issues are drawn from Flavin 1996. Box 8.3 India Today 1996. The reference to electoral arrange- was prepared by Carter Brandon and Charles Feinstein. ments under parliamentary systems and women's political Box 8.5 draws on work by Landau 1993 and Knight, representation comes from Lijphart 1995. The section on Loayza, and Villanueva 1995. The discussion of refugees diversity and representation is based on the Brautigam and the role of the state is based on the background paper background paper and a background note by Jalali. The by Suhrke and Newland. Evidence on the effectiveness of distinction among service delivery NGOs, intermediary aid is drawn from Burnside and Dollar 1996. organizations, and primary associations draws on Fisher 1993 and Carroll, Schmidt, and Bebbington 1996. The Chapter 9 reference to Umu-Itodo, Nigeria, comes from Francis and others 1996. Box 7.3 is based on Putnam, Leonardi, and This chapter benefited from written contributions by Nanetti 1993 and Narayan and Pritchett 1997. The dis- Dani Rodrik and Gary Reid and suggestions and com- cussion of the role of social capital in enhancing the qual- ments from Barry Ames, Juan Cariaga, Antonio Estache, ity of public action draws on Evans 1996a and 1996b. Sue Goldmark, Jorge Gorrio, Ravi Kanbur, Octavio The discussion of participatory mechanisms is inspired Amorim Neto, Graham Scott, Mary Shirley, and Zafiris by Campos and Root 1996, Paul 1994, Picciotto 1995, and Tzannatos. Holmes and Krishna 1996. The reference to the user assess- The information on the decentralization program in ment of water supply in Baku, Azerbaijan, comes from Peru is from Graham and Kane 1996. The study on inef- World Bank 1995a. Box 7.4 is based on material supplied ficiency in Uruguay's water utilities is Estache, Rodriguez- by Patricia Langan. The evidence in Box 7.5 is based on Pardina, and Smith 1996. The study on political coali- Narayan 1995 and subsequent empirical work by Isham, tions in Brazil is Alesina and Rosenthal 1995. On pension Narayan, and Pritchett 1995. The example from Recife, expenditure in Uruguay see Kane 1995. Box 9.2 is based Brazil, comes from Orstrom 1996, and the discussion of on Lundahl 1992. Box 9.3 is based on Lewis 1961. The the implications for public agencies and the enabling envi- discussion of telecommunications in Sri Lanka is based ronment draws on World Bank 1996c and 1996e. on World Bank l996g. Box 9.4 is based on Bermeo and The section on decentralization draws on the work of GarcIa-Durán 1994. Box 9.5 is based on Navarro 1996. Bennett 1990, Campbell and Fuhr forthcoming, Oates 1972, Scharpf 1994, Shah 1994, Stiglitz 1977 and 1996, Chapter 10 Tanzi 1 995a, Wallich 1994, and World Bank research led The discussion on state collapse is based on the Suhrke by Hans Binswanger and Anwar Shah. Table 7.2 is based and Newland background paper. It also draws on discus- on contributions by Jeffrey Hammer. Box 7.6 was pre- sions with Mamadou Dia and Steven Holtzman and on pared by Bill Dillinger and Vikram Nehru, and Box 7.7 by Tallroth 1997. Box 10.1 is based on Mubarak 1996. Box Anwar Shah. Table 7.4 was prepared by Anwar Shah. 10.2 is based on Tallroth 1997. The discussion on the Examples of decentralization initiatives in various coun- regional agenda was prepared with the help of Malcolm tries are based on Barzelay 1991, Kanter 1995, Villadsen Rowat, Salvatore Schiavo-Campo, and Michael Walton. and Lubanga 1996, and information provided by Tim Shahrokh Fardoust, Alan Geib, Costas Michalopoulos, Campbell, Florence Eid, Armin Fidler, Vikram Nehru, Alcyone Saliba, Klaus Simon, and Markus Steinich. Marcelo Selowsky, Shekhar Shah, Roger Sullivan, and John Williamson also provided valuable comments. Chapter 8 Background papers This chapter draws extensively on the framework set out Aiyar, Swaminathan. "Evolution of the Role of the State in by Stiglitz 1995. The principles of voluntary cooperation India." are drawn from Lawrence, Bressant, and Ito 1996. The Alesina, Alberto. "Politics, Procedures, and Budget Deficits." section on ensuring more effective cooperation draws on Aron, Janine, Ibrahim Elbadawi, and Benno Ndulu. "The State Shihata 1996. Hoekman 1995 provides useful back- and Development in Sub-Saharan Africa," ground materials on opening world markets. Box 8.1 is Ball, Richard. "The Institutional Foundations of Monetary based on Hoekman 1995 and Financial Times 1996. The Commitment: A Comparative Analysis." BIBLIOGRAPHICAL NOTE 181 Braathen, Einar, and Harald Ekker. "The State and National Adamolekun, Ladipo. 1991. "Promoting African Decentraliza- Reconstruction: Interdependency Between Central and tion." Public Administ ration and Development 11(3): 285-91. Local Level." Aden, Jean. 1996. "Industrial Pollution Abatement in the Brautigam, Deborah. "The State and Ethnic Pluralism: Manag- Newly Industrializing Countries: Korea." Asia Technical ing Conflict in Multiethnic Societies." Department, World Bank, Washington, D.C. Brunetti, Aymo, Gregory Kisunko, and Beatrice Weder. "Cred- Ad Hoc Committee on Health Research Relating to Future ibility of Rules and Economic Growth: Evidence from a Intervention Options. 1996. Investing in Health Research and Worldwide Survey of the Private Sector." (a) Development: Report of the Ad Hoc Committee on Health "Institutional Obstacles for Doing Business: Region- Research Relating to Future Intervention Options. Geneva: by-Region Results from a Worldwide Survey of the Private World Health Organization. Sector." (b) Afsah, Shakeb, Benoit Laplante, and Nabiel Makarim. 1996. Campos, Ed, and Sanjay Pradhan. "Building Institutions for a "Program-based Pollution Control Management: The More Effective Public Sector." Indonesian PROKASIH Program." World Bank Policy Commander, Simon, Hamid Davoodi, and Une J. Lee. "The Research Working Paper No. 1602. Policy Research Depart- Causes and Consequences of Government for Growth and ment, Environment, Infrastructure and Agriculture Divi- Well-Being." sion, World Bank, Washington, D.C. Dc Silva, Migara. "War, Tax Revenue and the Rise of the Mod- Afsah, Shakeb, Benoit Laplante, and David Wheeler. 1996. ern Public Administration in Western Europe." "Controlling Industrial Pollution: A New Paradigm." World Disch, Arne. "The Scandinavian 'Model': Successes and Limita- Bank Policy Research Working Paper No. 1672. Policy tions of the Activist State." Research Department, Environment, Infrastructure and Guasch, J. Luis, and Robert W. Hahn. "The Costs and Benefits Agriculture Division, World Bank, Washington, D.C. of Regulation: Some Implications for Developing Countries." Alesina, Alberto, R. Hausmann, R. Hommes, and E. Stein. Jalali, Rita. "State and Ethnicity." 1996. "Budget Institutions and Fiscal Performance in Latin Lee, Kyu Sik. "Cholla Region Catches Up with Korea: The Role America." NBER Working Paper No. 5556. National of Local Governments." Bureau of Economic Research, Cambridge, Mass. Molander, Per. "Public Sector Spending Control: Swedish Alesina, Alberto, and Roberto Perotti, 1995. "Fiscal Expansions Experiences." and Adjustments in OECD Countries." Economic Policy: A Ohno, Kenichi. "Creating the Market Economy: The Japanese European Forum 21 (October): 205-48. View on Economic Development and Systemic Transition." 1996. "Income Distribution, Political Instability, and Rose-Ackerman, Susan. "When Is Corruption Harmful?" Investment." European Economic Review 40: 1203-28. Schiavo-Campo, Rino. "Civil Service and Economic Develop- Alesina, Alberto, and Howard Rosenthal. 1995. Partisan Poli- mentA Selective Synthesis of International Facts and tics, Divided Government, and the Economy. Cambridge, Experience." England: Cambridge University Press. Schiavo-Campo, Salvatore, Giulio de Tommaso, and Amitabha Aiston, Lee. 1996. "Empirical Work in Institutional Econom- Mukherjee. "An International Statistical Survey of Govern- ics: An Overview." In Lee Alston, T. Eggertsson, and Doug- ment Employment and Wages." lass North, eds., Empirical Studies in Institutional Change. Suhrke, Astri, and Kathleen Newland. "States and Refugees: Cambridge, England: Cambridge University Press. International Cooperation on Issues of Displacement." Amsden, Alice. 1989. Asia's Next Giant: South Korea and Late Van Rijckeghem, Caroline, and Beatrice Weder. "Corruption Industrialization. New York, N.Y.: Oxford University Press. and Rate of Temptation: Do Low Wages in the Civil Service Anand, Sudhir, and Martin Ravallion. 1993. "Human Devel- Cause Corruption?" opment in Poor Countries: On the Role of Private Incomes Webb, Douglas. "The Judiciary: The Arbiter of Rules and and Public Services." Journal of Economic Perspectives 7(1): Resolver of Disputes." 133-50. 'WHO (World Health Organization). "Essential Public Health Andic, Fuat, and Suphan Andic. 1996. The Last of the Ottoman Functions: A New Initiative in Support of Health for All." Grandees: The Lf and Political Testament ofAli Pasha. Istan- "Redefining the Scope of Public Health Beyond the bul: Istanbul's Press. Year 2000." Aoki, Masahiko, and Ronald Dore, eds. 1994. The Japanese "The Role of Government in Public Health Through Firm: Sources of Competitive Strength. Oxford, England: the Ages." Clarendon Press. "The Role of State Action in Disease Eradication and Aoki, Masahiko, Hyung-Ki Kim, and Masahiro Okuno-Fuji- Control." wara. 1997. The Role of the Government in East Asian Eco- nomic Development: Comparative Institutional Analysis. Ox- Selected bibliography ford, England: Oxford University Press. Ahdallah, A. E. A. 1990. "Ethnic Conflict in Sudan." In M. L. Aoki, Masahiko, Kevin Murdoch, and Masahiro Okuno-Fuji- Michael Wyzan, ed., The Political Economy of Ethnic Dis- wara. 1995. "Beyond the East Asian Miracle: Introducing crimination and Affi rmative Action. New York, N.Y.: Praeger. the Market-Enhancing View." Stanford University Center 182 WORLD DEVELOPMENT REPORT 1Y97 for Economic Policy Research Discussion Paper No. 442. Borner, Silvio, Aymo Brunetti, and Beatrice Weder. 1995. Polit- Stanford, Calif. ical Credibility and Economic Development. New York, N.Y.: Arisawa, Hiromi, and Takahide Nakamura, eds. 1990. Data: St. Martin's Press. Design of Postwar Economic Policies, Vol. 1. Tokyo: Tokyo Boston, Jonathan, John Martin, June Pallot, and Pat Walsh. University Press (in Japanese). 1996. Public Management: The New Zealand ModeL New Bahi, Roy. 1994. "Revenues and Revenue Assignment: Inter- York, N.Y.: Oxford University Press. governmental Fiscal Relations in the Russian Federation." In Bosworth, Barry, Susan Collins, and Yu-chin Chen. 1995. Christine I. Wallich, ed., Russia and the Challenge of Fiscal "Accounting for Differences in Economic Growth." Eco- Federalism. World Bank Regional and Sectoral Studies. nomic Studies Program, Brookings Institution, Washington, Washington, D.C.: World Bank. D.C. Bardhan, Pranab. 1996. "Efficiency, Equity and Poverty Allevi- Brautigam, Deborah. 1996. "State Capacity and Effective Gov- ation: Policy Issues in Less Developed Countries." Economic ernance." In Benno Ndulu and Nicholas van de Walle, eds., Journal 106 (September): 1344-56. Agenda for Africa's Economic Renewal, pp. 81-108. Wash- ington, D.C.: Overseas Development Council. Barro, Robert J. 1996. "Determinants of Democracy." Depart- Brass, P. R. 1985. Ethnic Groups and the State. Totowa, N.J.: ment of Economics, Harvard University, Cambridge, Mass. Barnes and Noble. Barro, Robert J., and Xavier Sala-i-Martin. 1995. Economic Bread for the World Institute. 1997. What Governments Can Growth. New York, N.Y.: McGraw-Hill. Do: Seventh Annual Report on the State of World Hunger. Sil- Barzelay, Michael. 1991. "Managing Local Development: ver Spring, Md.: Bread for the World Institute. Lessons from Spain." Policy Sciences 24: 271-90. Brett, E. A. 1993. "Theorizing Crisis and Reform: Institutional Batley, Richard. 1996. "Public-Private Relationships and Perfor- Theories and Social Change in Uganda." Institute for Devel- mance in Service Provision." Urban Studies 33(4-5): 723-5 1. opment Studies, Sussex University, Brighton, England. Bennett, Robert J., ed. 1990. Decentralization, Local Govern- Buchanan, J. M. 1977. "Why Does Government Grow?" In ments, and Markets: Toward a Post-Welfare Agenda. Oxford, Thomas Borcherding, ed., Budgets and Bureaucrats: The England: Oxford University Press. Sources of Government Growth. Durham, NC,: Duke Uni- Bermeo, Nancy, and José Garcia-Durán. 1994. "Spain: Dual versity Press. Transition Implemented by Two Parties." In Stephan Hag- Burki, Shahid J., and Sebastian Edwards. 1996. Dismantling the gard and Steven B. Webb, Voting for Reform: Democracy, Populist State. The Unfinished Revolution in Latin America and Political Liberalization, and Economic Adjustment. New the Caribbean. World Bank Latin American and Caribbean York, N.Y.: Oxford University Press. Studies: Viewpoints. Washington, D.C.: World Bank. Berry, Albert, and Brian Levy. 1994. "Indonesia's Small and Burnside, Craig, and David Dollar. 1996. "Aid, Policies and Medium-Size Exporrers and Their Support Systems." Policy Growth." Policy Research Department, Macroeconomics Research Working Paper No. 1402. Policy Research and Growth Division, World Bank, Washington, D.C. Department, Finance and Private Sector Development Divi- Buscaglia, Edgardo, and Maria Dakolias. 1996. Judicial Reform sion, World Bank, Washington, D.C. in Latin American Courts: The Experience in Argentina and Bhatt, Ela. n.d. "Moving Towards a People-Centered Econ- Ecuador. World Bank Technical Paper No. 350. World omy." The Self-Employed Women's Association, Ahmeda- Bank, Washington, D.C. bad, India. Campbell, Tim, and Harald Fuhr, eds. Forthcoming. Does Bird, Richard M. 1995. "Decentralizing Infrastructure: For Good Decentralization Work? Case Studies on Innovative Local Gov- or for Ill?" In Antonio Estache, ed., Decentralizing Infrastruc- ernment in Latin America. Washington, D.C.: World Bank. ture: Advantages and Limitations, pp. 22-51. World Bank Dis- Campbell, Tim, George Peterson, and José Brakarz. 1991. "Decentralization to Local Government in LAC: National cussion Paper No. 290. Washington, D.C.: World Bank. Strategies and Local Response in Planning, Spending and Bird, Richard M., Robert D. Ebel, and Christine I. Wallich, Management." Report No. 5. Latin America and the eds. 1995. Decentralization of the Socialist State: Intergovern- Caribbean Technical Department, Regional Studies Pro- mental Finance in Transition Economies. World Bank gram, World Bank, Washington, D.C. Regional and Sectoral Studies. Washington, D.C.: World Campos, Ed, and Sanjay Pradhan. 1996. "Budgetary Institu- Bank. tions and Expenditure Outcomes: Binding Governments to Boadway, Robin W., Sandra Roberts, and Anwar Shah. 1994. Fiscal Performance." Policy Research Working Paper No. "The Reform of Fiscal Systems in Developing and Emerging 1646. Policy Research Department, World Bank, Washing- Market Economies: A Federalism Perspective." Policy ton, D.C. Research Working Paper No. 1259. Policy Research Depart- Campos, Ed, and Hilton L. Root. 1996. The Key to the Asian ment, Public Economics Division, World Bank, Washing- Miracle: Making Shared Growth Credible. Washington, ton, D.C. D.C.: Brookings Institution. Borcherding, T. E. 1985. "The Causes of Government Expen- Caprio, Gerard, Jr. 1996. "Bank Regulation: The Case of the diture Growth: A Survey of the U.S. Evidence." Journal of Missing Model." World Bank Policy Research Working Public Economics 28 (December): 359-82. Paper No. 1574. Policy Research Department, Finance and BIBLIOGRAPHICAL NOTE 183 Private Sector Development Division, World Bank, Wash- Deininger, Klaus, and Lyn Squire. 1996. "A New Data Set on ington, D.C. Measuring Income Inequality." World Bank Economic Caprio, Gerard, Jr., and Daniela Klingebiel. 1996. "Bank Insol- Review 10(3): 565-92. vency: Bad Luck, Bad Policy, or Bad Banking?" Paper pre- De Soto, Hernando. 1989. The Other Path. New York, N.Y.: sented at the Annual World Bank Conference on Develop- Harper & Row. ment Economics, World Bank, Washington, D.C., April 25. Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ). Carroll, Tom, Mary Schmidt, and Tony Bebbington. 1996. nd. "Indo-German Watershed Development Programme, "Participation Through Intermediary NGOs." Environment Maharashtra, India." GTZ, Eschborn, Germany. Department Papers No. 031. Environment Department, Deutsche Stiftung für internationale Entwicklung. 1996. World Bank, Washington, D.C. Zweites Deutsches Weltbank-Forum: Verantwortungsbewufte CEPAL/GTZ (Comisión Económica para America Latina/ o'intlich-private Partnerschafi. Berlin: Deutsche Stiftung für Gesellschaft für Technische Zusammenarbeit). 1996. internationale Entwicklung. Descentralización Fiscal en America Latina: Balance y Princi- Devarajan, Shantayanan, Vinaya Swaroop, and Heng-fu Zou. pales Desafios. Santiago, Chile: CEPAL/GTZ. 1996. "The Composition of Public Expenditure and Eco- Chellaraj, Gnanaraj, Olusoji Adeyi, Alexander S. Preker, and nomic Growth." Journal ofMonetary Economics 37: 31 3-44. Ellen Goldstein. 1996. Trends in Health Status, Services, and Dia, Mamadou. 1996. Africa's Management in the 1990s and Finance: The Transition in Central and Eastern Europe. Vol- Beyond: Reconciling Indigenous and Transplanted Institutions. ume IL StatisticalAnnex. Washington, D.C.: World Bank. Directions in Development Series. Washington, D.C.: Chhibber, Ajay, Mansoor Dailami, and Nemat Shafik, eds. World Bank. 1992. Reviving Private Investment in Developing Countries: DIaz Alejandro, Carlos. 1988. Trade, Development and the World Empirical Studies and Policy Lessons. Elsevier Science Pub- Economy: Selected Essays. Oxford, England: Basil Blackwell. lishers, North Holland. Dilulio, John J., Jr., ed. 1994. Deregulating the Public Service: Chisari, Omar, Antonio Estache, and Carlos Romero. 1996. Can Government Be Improved? Washington, D.C.: The "Winners and Losers from Utilities Privatizations: Lessons Brookings Institution. from a General Equilibrium Model of Argentina." Universi- Dillinger, Bill. 1995. "Decentralization, Politics and Public Ser- dad Argentina de la Empresa, Buenos Aires, and World vice." In Antonio Estache, ed., Decentralizing Infrastructure: Bank, Washington, D.C. Advantages and Limitations, pp. 5-21. World Bank Discus- Clague, Christopher, Philip Keefer, Stephen Knack, and Man- sion Papers No. 290. Washington, D.C.: World Bank. cur Olson. 1996. "Property and Contract Rights under Dixon, Geoffrey. 1993. "Managing Budget Outlays 1983-84 to Democracy and Dictatorship." Journal of Economic Growth 1992-93." In Brian Galligan, ed., Federalism and the Econ- 1(2): 243-76. omy: International, National and State Issues. Canberra: Fed- Clark, John. 1995. "The State, Popular Participation and the eralism Research Centre, Australian National University. Voluntary Sector." World Development23(4): 593-601. Drèze, Jean, and A. K. Sen. 1989. Hunger and Public Action. Coase, R. H. 1960. "The Problem of Social Cost." Journal of Oxford, England: Oxford University Press. Law and Economics 3 (October): 1-44. Easterly, William, and Ross Levine. 1996. "Africa's Growth 1988. The Firm, the Market, and the Law. Chicago, Ill.: Tragedy: A Retrospective, 1960-89." World Bank Policy University of Chicago Press. Research Paper No. 1503. Macroeconomics and Growth Coloma, Fernando C. 1996. "Seguro de Desempleo: Teoria, Division, Policy Research Department, World Bank, Wash- Evidencia, y una Propuesta." Cuadernos de Economia 33(99): ington, D.C. 29 5-3 20. Easterly, William, and Sergio Rebelo. 1993. "Fiscal Policy and Conyers, Diana. 1985. "Decentralization: A Framework for Economic Growth." Journal ofMonetary Economics (Nether- Discussion." In Hasnat Abdul Hye, ed., Decentralization, lands) 32(3): 417-58. Local Government Institutions and Resource Mobilization, Economic and Social Research Council. n.d. ESRC Whitehall pp. 22-42. Comilla, Bangladesh: Bangladesh Academy for Programme: The Changing Nature of Central Government in Rural Development. Britain. London: Economic and Social Research Council. Courakis, Anthony, Fatima Moura-Roque, and George Tridi- Edwards, Michael, and David Hulme, eds. 1992. Making a mas. 1993. "Public Expenditure Growth in Greece and Por- ference: NGOs and Development in a Changing World. Lon- tugal: Wagner's Law and Beyond." Applied Economics 25: don: Earthscan Publications. 12 5-34. Elsenhans, Hartmut. 1996. State, Class and Development. New Cox, Gary W., and Mathew D. McCubbins. 1996. "Structure Delhi: Radiant Publishers. and Policy: The Institutional Determinants of Policy Out- Epsing-Andersen, Gusta. 1994. "After the Golden Age: The comes." Department of Political Science, University of Cal- Future of the Welfare State in the New Global Order." ifornia, San Diego, Calif. Occasional Paper No. 7. World Summit for Social Develop- Dasgupta, Partha. 1995. An Inquiry into Well-Being and Desti- ment, Geneva. tution. New York, N.Y.: Oxford University Press. Esfahani, Hadi Salehi. 1996. "The Political Economy of the 1997. "Social Capital and Economic Performance." Telecommunications Sector in the Philippines." In Brian University of Cambridge, Cambridge, England. Levy and Pablo T. Spiller, eds., Regulations, Institutions and 184 WORLD DEVELOPMENT REPORT 1997 Commitment. Comparative Studies of Telecommunications, Ferreira, M. Luisa, and Charles C. Griffen. 1995. "Tanzania pp. 145-201. Cambridge, England: Cambridge University Human Development Survey: Final Report." Population Press. and Human Resources, Eastern Africa Department, World Esman, Milton. 1994. Ethnic Politics. Ithaca, N.Y.: Cornell Bank, Washington, D.C. University Press. Financial Times. 1996. "Antagonists Queue for WTO Judg- Estache, Antonio, ed. 1995. Decentralizing Infrastructure: ment." (August 8). Advantages and Limitations. World Bank Discussion Paper Fischer, Stanley. 1995. "Central-Bank Independence Revis- No. 290. Washington, D.C.: World Bank. ited." AEA Papers and Proceedings (May): 20 1-06. Estache, Antonio, M. Rodriguez-Pardina, and W. Smith. 1996. Fisher, Julie. 1993. The Road from Rio: Sustainable Development "Towards a New Role for the State in Uruguay's Utilities." and the Nongovernmental Movement in the Third World. New Latin America and the Caribbean Country Department I, York, N.Y.: Praeger. Infrastructure and Urban Development Division, World Fiske, Edward B. 1996. Decentralization of Education: Politics Bank, Washington, D.C. and Consensus. Directions in Development Series. Washing- European Commission. 1996. "Towards a More Coherent ton, D.C.: World Bank. Global Economic Order." Discussion Paper. Forward Stud- Flavin, Christopher. 1996. "Facing Up to the Risks of Climate ies Unit and Directorate-General for Economic and Finan- Change." In Lester R. Brown, Christopher Flavin, and Linda cial Affairs, European Commission, Brussels. Starke, eds., State of the World 1996 A Worldwatch Institute Evans, Peter B. 1995. EmbeddedAutonomy: States and Industrial Report on Progress Toward a Sustainable Society. New York, Transformation. Princeton, N.J.: Princeton University Press. N.Y.: Norton. 1996a. "Government Action, Social Capital and Folster, Stephan. 1996. "Social Insurance Based on Personal Development: Reviewing the Evidence on Synergy." World Savings Accounts: A Possible Reform Strategy for Overbur- Development 24(6): 1119-32. dened Welfare States." Industrial Institute for Economic and 1996b. "Social Capital and the Functioning of Bureau- Social Research, Stockholm. cracies in Developing Countries." Paper presented at the Forster, Michael. 1994. The Effects of Net Transfers on Low International Conference on Governance Innovations: Build- Income Among Non-Elderly Families. OECD Economic ing a Government-Citizen-Business Partnership, Manila, Studies No. 22. Paris: OECD. October 20-23. Francis, Paul, J. A. Akinwumi, P. Ngwu, S. A. Nkom, J. Odihi, Evans, Peter B., and James Rauch. 1996. "Bureaucratic Struc- J. A. Olomajeyc, F. Okunmadewa, and D. J. Shehu. 1996. ture and Economic Growth: Some Preliminary Analysis of State, Community and Local Development in Nigeria. World Data on 35 Developing Countries." University of Califor- Bank Technical Paper No. 336. Africa Region Series. Wash- nia, Berkeley, Calif. ington, D.C.: World Bank. Fageberg, Jan, Bart Verspagen, and Nick von Tunzelmann. Francks, Penelope. 1992. Japanese Economic Development: The- 1994. The Dynamics of Technology, Trade and Growth. my and Practice. London and New York, N.Y.: Routledge. Aldershot, Hants, England, and Brookfield, Vt.: Edward Elgar. Freedom House. Various issues. Freedom in the World: The Farrington, John, and David Lewis, eds., with S. Satish and Annual Survey of Political Rights and Civil Liberties. New York, N.Y.: Freedom House. Avrea Miclat-Teves. 1993. Nongovernmental Organizations and the State in Asia: Rethinking Roles in Sustainable Agricul- Freeman, Richard B., Birgitta Swedenborg, and Robert Topel. tural Development. London and New York, N.Y.: Routledge. 1995. "Economic Troubles in Sweden's Welfare State: Intro- Feder, Gershon, and Akihiko Nishio. 1996. "The Benefits of duction, Summary, and Conclusions." NBERISNS Project Land Registration and Titling: Economic and Social Per- Reforming the Welfare State Occasional Paper No. 69. spectives." Paper presented at the International Conference National Bureau of Economic Research, Cambridge, Mass. of Land Tenure and Administration, Orlando, Fla., Novem- Frey, Bruno S., and Reiner Eichenberger. 1994. "The Political ber U. Economy of Stabilization Programmes in Developing Coun- Federalist Papers. 1987. Edited by Isaac Kramnick. Har- tries." European Journal of Political Economy 10: 169-90. mondsworth, Middlesex, England: Penguin. Fuhr, Harald, Klaus Simon, and Albrecht Stockmayer, eds. Feldstein, Martin. 1995. "Tax Avoidance and the Deadweight 1993. Subsidiaritdt in der Entwicklungszusammenarbeit: Loss of the Income Tax." NBER Working Paper No. 5055. Dezentralisierung und Verwaltungsreformen zwischen Struktu- National Bureau of Economic Research, Cambridge, Mass. ranpassung und Selbsrhilfe. Baden-Baden, Germany: Nomos Feldstein, Martin, and Andrew Samwick. 1996. "The Transi- Verlagsgesellschaft. tion Path in Privatizing Social Security." Paper presented at Fukuyama, Francis. 1995. Trust: The Social Virtues and the Cre- a National Bureau of Economic Research conference on ation of Prosperity. New York, N.Y.: Simon & Schuster. Privatizing Social Security, Cambridge, Mass., August 2. Galal, Ahmed. 1996. "Chile: Regulatory Specificity, Credibility Fernandez, Raquel, and Dani Rodrik. 1991. "Resistance to of Commitment, and Distributional Demands." In Brian Reform: Status Quo Bias in the Presence of Individual- Levy and Pablo Spiller, eds., Regulations, Institutions and Specific Uncertainty." American Economic Review 81(5): Commitment: Comparative Studies of Telecommunications. 1146-5 5. Cambridge, England: Cambridge University Press. BIBLIOGRAPHICAL NOTE 185 Galal, Ahmed, Leroy Jones, Pankaj Tandon, and Ingo Vogelsang. Hammer, Jeffrey S. 1997. "Economic Analysis for Health Pro- 1994. Welfare Consequences of Selling Public Enterprises: An jects." World Bank Research Observer 12(1): 47-71. EmpiricalAnalysis. New York, N.Y.: Oxford University Press. Haque, Nadeem UI, and Ratna Sahay. 1996. "Do Government Gemmell, Norman, ed. 1993. The Growth of the Public Sector. Wage Cuts Close Budget Deficits? Costs of Corruption." London: Edward Elgar. IMF Staff Papers 43(4): 754-78. Gould, Julius, and William L. KoIb. 1964. A Dictionary of the Helm, Dieter, ed. 1989. The Economic Borders of the State. Social Sciences. New York, N.Y.: Free Press of Glencoe. Oxford, England: Oxford University Press. Graham, Carol, and Cheikh Kane. 1996. "Opportunistic Gov- Hesse, Jems-Joachim. 1993. "From Transformation to Mod- ernment or Sustaining Reform? Electoral Trends and Public ernization: Administrative Change in Central and Eastern Expenditure Patterns in Peru, 1990-1995." Poverty and Europe." Public Administration 71 (Spring-Summer): Social Policy Department Discussion Paper Series 89. 2 19-57. Human Capital Development and Operations Policy Heyneman, Stephen P. Forthcoming. "Education Choice in Department, World Bank, Washington D.C. Eastern Europe and the Former Soviet Union: A Review Graham, Carol, and Moises NaIm. 1997. "The Political Econ- Essay." Education Economics. omy of Institutional Reform in Latin America." Paper pre- Hirsch, Joachim. 1995. Der nationale Wettbewerbstaat: Staat, sented to the MacArthur Foundation/IDB Conference on Demokratie, und Politik, in globalen Kapitalismus. Berlin: Inequality Reducing Growth in Latin America's Market Edition ID-Archiv. Economies, January 28-29. Hirschman, Albert 0. 1970. Exit, Voice, and Loyalty: Responses Gray, Cheryl. 1996. "In Search of Owners: Privatization and to Decline in Firms, Organizations and States. Cambridge, Corporate Governance in Transition Economies." World Mass.: Harvard University Press. Bank Research Observer 11(2): 179-97. Ho, Luu Bich. 1997. "The Government's Role in Market- Greenwood, Roystone, C. R. Hinnings, and Stewart Ranson. oriented Economic Renovation Process in Viet Nam." Paper 1975. "Contingency Theory and the Organization of Local presented at the International Symposium on Government's Authorities. Part 1: Differentiation and Integration." Public Role in the Market Economy, China Institute for Reform Administration 53: 1-23. and Development, Haikou, China, January 7-8. Gruber, Jonathan. 1994. "The Consumption Smoothing Bene- Hoekman, Bernard M. 1995. Trade Laws and Institutions: Good fits of Unemployment Insurance." NBER Working Paper Practices and the World Trade Organization. World Bank Series No. 4750. National Bureau of Economic Research, Discussion Paper No. 282. Washington, D.C.: World Bank. Cambridge, Mass. Holmes, Malcolm, and Anirudh Krishna. 1996. "Public Sector Guasch, J. Luis, and Pablo T. Spiller. 1997. Managing the Reg- Management and Participation: Institutional Support for Sus- ulatory Process: Concepts, Issues and the Latin America and tainable Development." In Jennifer Rietbergen-McCracken, Caribbean Story Book. Directions in Development Series. ed., Participation in Practice: The Experience ofthe World Bank Washington, D.C.: World Bank and Johns Hopkins Uni- and Other Stakeholders, pp. 29-35. World Bank Discussion versity Press. Paper No. 333. Washington, D.C.: World Bank Guerrero, Rodrigo. 1996. "Epidemiology of Violence: The Case Holsti, K. J. 1995. "War, Peace, and the State of the State." of Cali, Colombia." Paper presented at the Second Annual International Political Science Review 16(4): 319-39. World Bank Conference on Development in Latin America Holtzman, Steven. 1995. "Post Conflict Reconstruction." Envi- and the Caribbean, Bogota, Colombia, July. ronment Department, Social Policy and Resettlement Divi- Guhan, S., and Samuel Paul, eds. 1997. Corruption in India: sion, World Bank, Washington, D.C. Agenda for Action. New Delhi: Vision Books. Hommes, Rudolf. 1995. "Conflicts, and Dilemmas of Decen- Gupta, Sanjeev, Jerald Schiff, and Benedict Clements. 1996. tralization." In Michael Bruno and Boris Pleskovic, eds., "Drop in World Military Spending Yields Large Dividend." Annual World Bank Conference on Development Economics, IMF Survey (June 3). pp. 33 1-50. Washington, D.C.: World Bank. Gurr, Ted Robert, Keith Jaggers, and Will H. Moore. 1990. Huber, Evelyne, Charles Ragin, and John D. Stephens. 1993. "The Transformation of the Western State: The Growth of "Social Democracy, Christian Democracy, Constitutional Democracy, Autocracy, and State Power Since 1800." Stud- Structure, and the Welfare State." American Journal of Soci- ies in Comparative International Development 25(1): 73-108. ology99(3): 71 1-49. Gustafsson, Bo. 1995. "Foundations of the Swedish Model." Hulten, Charles R. 1996. "Infrastructure and Economic Devel- Nordic Journal of Political Economy 22: 5-26. opment: Once More into the Breach." Department of Eco- Gwartney, James D., Robert Lawson, and Walter Block. 1996. nomics, University of Maryland, College Park, Md. Economic Freedom of the World, 1975-1995. Vancouver: Humphrey, John, and Hubert Schmitz. 1995. "Principles for Fraser Institute. Promoting Clusters and Networks of Small and Medium Haggard, Stephan, and Steven B. Webb, eds. 1994. Voting for Enterprises." Discussion Paper No. 1. Small and Medium Reform: Democracy, Political Liberalization, and Economic Enterprise Branch, United Nations Industrial Development Adjustment. New York, N.Y.: Oxford University Press. Organization, New York, N.Y. Hahn, Robert W., ed. 1996. Risks, Costs and Lives Saved: Get- Huntington, Samuel P. 1991. The Third Wave: Democratization ting Better Results from Regulation. New York, N.Y.: Oxford in the Late Twentieth Century. Norman: University of Okla- University Press. homa Press. 186 WORLD DEVELOPMENT REPORT 1997 ILO (International Labour Office). Various years. The Cost of Jhabvala, Renana, and Ela Bhatt. n.d. "The World of Work in Social Security. Geneva: ILO. the People's Sector: And Its Inherent Strength. The SEWA ILO (International Labour Organisation). 1986. Yearbook of Experience." The Self-Employed Women's Association, Labor Statistics. Geneva: ILO. Ahmedabad, India. 1994. Yearbook of Labor Statistics. Geneva: ILO. Jun, Jong S., and Deil S. Wright, eds. 1996. Globalization and IMF (International Monetary Fund). 1986. Manual on Govern- Decentralization: Institutional Contexts, Policy Issues and ment Finance Statistics. Washington, D.C.: IMF. Intergovernmental Relations in Japan and the United States. 1996. World Economic Outlook. Washington, D.C.: Washington, D.C.: Georgetown University Press. IMF. Ka, Samba, and Nicolas van de Walle. 1994. "Senegal: Stalled Various years-a. Government Statistics Yearbook. Reform in a Dominant Party System." In Stephan Haggard Washington, D.C.: IMF. and Stephen B. Webb, Voting for Reform: Democracy, Politi- Various years-b. International Financial Statistics. cal Liberalization, and Economic Adjustment, pp. 290-359. Washington, D.C.: IMF. New York, N.Y.: Oxford University Press. India Today. 1996. "The Maturity of Democracy." (August 31). Kabeer, Naila. 1994. Reversed Realities: Gender Hierarchies in International Countiy Risk Guide. Various issues. New York, Development Thought. London: Verso. N.Y.: International Reports. Kane, Cheikh T. 1995. "Uruguay: Options for Pension International Political Science Review. 1996. New Trends in Reform." Education and Social Policy Paper Series No. 68. Federalism. Special edition of International Political Science Human Resources Development and Operations Policy Department, World Bank, Washington, D.C. Review 17(4). Kane, Cheikh, and Robert Palacios. 1996. "The Implicit Pen- Inter-Parliamentary Union. 1996. "Women in Parliament as at sion Debt." Finance and Development 33 (June): 36-38. 30 June 1995." Inter-Parliamentary Union, Geneva. Kangle, K. P. 1965. The Kautiliya Arthashastra: An English Isham, Jonathan, Daniel Kaufmann, and Lant Pritchert. 1995. Translation with Critical and Explanatoiy Notes. Delhi: "Governance and the Returns to Investment: An Empirical Motilal Banarsidass Publishers. Investigation." World Bank Policy Research Working Paper Kanter, Rosabeth Moss. 1995. "Thriving Locally in the Global No. 1550. Policy Research Department, Poverty and Human Economy." Harvard Business Review (September-October): Resources Division, World Bank, Washington, D.C. 15 1-60. Isham, Jonathan, Deepa Narayan, and Lant Pritchett. 1995. Kaufmann, Daniel. 1996. "Listening to Stakeholders on Devel- "Does Participation Improve Performance? Establishing opment Challenges and World Bank Instruments in their Causality with Subjective Data." World Bank Economic Countries: Myths Meet Some Evidence on Corruption, Eco- Review 9(2): 175-200. nomic Reforms, and Bank Programs." Harvard Institute for Ishikawa, Shigeru. 1990. Basic Issues of Development Economics. International Development, Harvard University, Cam- Tokyo: Iwanami Shoten (in Japanese). bridge, Mass., and World Bank, Washington, D.C. 1996. "From Development Economics to Develop- Keefer, Philip, and Stephen Knack. 1995. "The Effects of Insti- ment Aid Policy." In S. Ishikawa, ed., Theoretical Studies on tutions on Public Investment." Policy Research Department, Development Aid Policy. Tokyo: Institute of Development World Bank, Washington, D.C.; IRIS/University of Mary- Economics (in Japanese). land, College Park, Md.; and School of Public Affairs, Amer- Israel, Arturo. 1990. "The Changing Role of the State: Institu- ican University, Washington, D.C. tional Dimensions." PRE Working Paper WPS 495. Coun- Khaldaun, Ibn. nd. The Muqaddimah. Translated with an try Economics Department, World Bank, Washington, D.C. Introduction by Franz Rosenthal. Princeton, N.J.: Princeton 1997. "A Guide for the Perplexed: Institutional University Press. Aspects of Social Programs." SOC 96-105. Inter-American Kim, Hyung-Ki, Michio Muramatsu, T. J. Pempel, and Kozo Development Bank, Washington, D.C. Yamamura, eds. 1995. The Japanese Civil Service and Eco- Itoh, Motoshige, Kazuharu Kiyono, Masahiro Okuno, and nomic Development: Catalysts of Change. Oxford, England: Clarendon Press. Kotaro Suzumura. 1988. Economic Analysis of Industrial Pol- Klingemann, Hans-Dieter, and Dieter Fuchs, eds. 1995. Beliefs icy. Tokyo: Tokyo University Press (in Japanese). in Government. Volume 1: Citizens and the State. New York, Jaggers, Keith, and Ted Robert Gurr. 1995. "Tracking Democ- N.Y.: Oxford University Press. racy's Third Wave with the Polity III Data." Journal ofPeace Klitgaard, Robert. 1988. Controlling Corruption. Berkeley, Research 32(4): 469-82. Calif.: University oi California Press. 1996. "Polity III: Regime Type and Political Author- Tropical Gangsters. New York, N.Y.: Basic Books. ity, 1800-1994." Inter-University Consortium for Political Knack, Stephen, and Philip Keefer. 1995. "Institutions and and Social Research, Ann Arbor, Mich. Economic Performance: Cross-Country Tests Using Alter- Jalali, Rita, and Seymour M. Lipset. 1992-93. "Racial and Eth- native Institutional Measures." Economics and Politics 7(3): nic Conflicts: A Global Perspective." Political Science Quar- 207-27. terly 107(4). Knight, Malcolm, Norman Loayza, and Delano Villanueva, Jalan, Bimal. 1992. The Indian Economy: Problems and Prospects. 1995. "The Peace Dividend: Military Spending Cuts and New Delhi: Penguin Books India. Economic Growth." IMF Working Paper No. WP/95/53. 8!BLIOGRAPHtCAL NOTE 187 Middle Eastern Department, International Monetary Fund, Levy, Brian, and Pablo Spiller. 1994. "The Institutional Foun- Washington, D.C. dations of Regulatory Commitment: A Comparative Analy- Kohli, Atul. 1994. "Where Do High Growth Political sis of Telecommunications and Regulation." Journal ofLaw, Economics Come From? The Japanese Lineage of Korea's Economics and Organization 10(2): 20 1-46. 'Development State.'" World Development 22: 1269-93. Levy, Brian, and Pablo Spiller, eds. 1996. Regulations, Institu- Konig, Klaus. 1996. "Policy Planning and Management Dia- tions and Commitment: Comparative Studies of Telecommuni- logue with Countries in Transition." Public Administration cations. Cambridge, England: Cambridge University Press. and Development 16: 417-29. Levy, Brian, Albert Berry, Motoshige Itoh, Linsu Kim, Jeffrey 1997. "Drei Welten der Verwaltungsmodernisierung." Nugent, and Shujiro Urata. 1994. "Technical and Market- In Klaus Luder, ed., Staat und Verwaltung: Funfzig Jahre ing Support Systems for Successfi.sl Small and Medium-Size Hochschule für Verwaltungswissenschafren, pp. 399-424. Enterprises in Four Countries." World Bank Policy Research Berlin: Duncker & Humblot. Working Paper No. 1400. Policy Research Department, Kormendi, Roger, and Philip Meguire. 1985. "Macroeconomic Finance and Private Sector Division, World Bank, Wash- Determinants of Growth: Cross-Country Evidence." Journal ington, D.C. ofMonetary Economics 16(2): 141-63. Lewis, Bernard. 1961. The Emergence of Modern Turkey. Lon- Kraay, Aart, and Caroline Van Rijckeghem. 1995. "Employment don and New York, N.Y.: Oxford University Press. and Wages in the Public Sector: A Cross-Country Study." Libecap, Gary D. 1996. "Economic Variables and the Develop- IMF Working Paper WP195I70. Fiscal Affairs Department, ment of Law: The Case of Western Mineral Rights." In Lee International Monetary Fund, Washington, D.C. J. Alston, Thrainn Eggertsson, and Douglass C. North, eds., Kubota, Isao. 1996. "Roles That Governments Play in Devel- Empirical Studies in Institutional Change. New York, N.Y.: oping, Developed States." Japan Times, June 17, p. 15. Kuper, Adam, and Jessica Kuper, eds. 1996. The Social Science Cambridge University Press. Encyclopedia, 2nd ed. London and New York, N.Y.: Rout- Lijphart, Arendt. 1969. "Consociational Democracy." World ledge. Politics 21. Laking, R. G. 1996. "Good Practice in Public Sector Manage- 1995. "The Virtues of Parliamentarism: But 'Vhich ment: Issues for the World Bank." Poverty and Social Policy Kind of Parliamentarism?" In H. E. Chenabi and Alfred Department, World Bank, Washington, D.C. Stepan, eds., Politics, Society, and Democracy: Comparative Lalor, R. Peter, and Hernan Garcia. 1996. "Reshaping Power Studies. Boulder, Cob.: Wesrview Press. Markets in South America." Electricity Journal 9(2): 63-71. Lijphart, Arendt, and Carlos H. Waisman, eds. 1996. Institu- Landau, D. 1986. "Government and Economic Growth in Less tional Design in New Democracies: Eastern Europe and Latin Developed Countries: An Empirical Study for 1960-1980." America. Boulder, Cob.: Westview Press. Economic Development and Cultural Change 35(1): 35-75. Lim, Linda Y., and Peter Gosling, eds. 1983. The Chinese in 1993. "The Economic Impact of Military Expendi- Southeast Asia, vol. 1. Singapore: Naruyen Asia. tures." World Bank Policy Research Working Paper No. Lin, Justin Yifu. 1996. "Comparative Advantage, Development WPS 1138. Policy Research Department, World Bank, Policy, and the East Asian Miracles." Peking University, Washington, D.C. Hong Kong University of Science and Technology, and Aus- Latinobarometro. 1996. "Press Release: Latinobarometro Sur- tralian National University. vey 1996." Latinobarometro, Santiago, Chile. Lin, Justin Yifu, Fang Cai, and Zhou Li. 1996. The China Mir- Lau, Lawrence, and D. H. Song. 1992. "Growth versus Privati- acle: Development Strategy and Economic Reform. A Friedman zation-An Alternative Strategy to Reduce the Public Enter- Lecture Fund Monograph. Hong Kong: The Chinese Uni- prise Sector: The Experiences of Taiwan and South Korea." versity Press. Working Paper, Department of Economics, Stanford Uni- Lindauer, David. 1988. "The Size and Growth of Government versity, Stanford, Calif Spending." Policy Working Paper WPS 44. World Bank, Lawrence, Robert Z., Albert Bressand, and Takatoshi Ito. 1996. Washington, D.C. A Vision for the World Economy. Washington, D.C.: Brook- Lindauer, David, and Barbara Nunberg, eds. 1994. Rehabilitat- ings Institution. ing Government: Pay and Employment Reform in Africa. Leftwich, Adrian, ed. 1996. Democracy and Development: Theocy Washington, D.C.: World Bank. and Practice. Cambridge, England, and Cambridge, Mass: Lindbeck, Assar. 1995. "Hazardous Welfare State Dynamics." Polity Press in association with Blackwell Publishers. Leonard, David K., and Dale Rogers Marshall. 1982. "Institu- Reprint Series No. 538. Institute for International Eco- tions of the Rural Development for the Poor: Decentraliza- nomic Studies, Stockholm University, Stockholm. tion and Organizational Linkages." Institute of International Lindbeck, Assar, Per Molander, Torsten Persson, Obof Peters- Studies, University of California, Berkeley, Calif. son, Agnar Sandmo, Birgitta Swedenborg, and Niels Thyge- Letowski, Janusz. 1993. "Polish Public Administration Between sen. 1994. Turning Sweden Around. Cambridge, Mass.: MIT Crisis and Renewal." Public Administration 71 (Spring-Sum- Press. mer): 1-11. Linz, Juan J., Seymour Martin Lipset, and Amy Bunger Pool. Levine, Ross, and David Renelt. 1992. "A Sensitivity Analysis of n.d. "Social Conditions for Democracy in Latin America: Cross-Country Growth Regressions." American Economic Latin American Barometer Survey Analysis." Yale Univer- Review 82(4): 942-63. sity, New Haven, Conn. 188 WORLD DEVELOPMENT REPORT 1997 Linz, Juan ]., and Alfred Stepan, eds, 1996. Problems of Democ- Milgrom, Paul R., Douglass C. North, and Barry R. Weingast. ratic Transition and Consolidation: Southern Europe, South 1990. "The Role of Institutions in the Revival of Trade: The America, and Post Communist Europe. Baltimore, Md.: Johns Law Merchant, Private Judges, and the Champagne Fairs." Hopkins University Press. Economics and I'olitics 2(1): 1-23. Lipset, Seymour Martin. 1979. The First New Nation: The Montgomery, John D. 1988. Bureaucrats and People: Grassroots United States in Historical & Comparative Perspective. New Participation in Third World Development. Baltimore, Md: York, N.Y.: W. W. Norton & Company. Johns Hopkins University Press. 1996. American Exceptionalism: A Double Edged Sword. Montinola, Gabriella, Yingyi Qian, and Barry R. Weingast. New York, N.Y.: W. W. Norton. 1995. "Federalism, Chinese Style: The Political Basis for Lovei, Magda, and Charles Weiss, Jr. 1996, "Environmental Economic Success in China." World Politics 48: 50-81. Management and Institutions in OECD Countries: Experi- Mosley, Paul, Jane Harrigan, and John Toye. 1995. Aid and ence and Lessons Learned." Environment Department, Power.' The World Bank and Policy-Based Lending. Vol. 1, World Bank, Washington, D.C. Analysis and Policy Proposals, 2nd ed. London and New York, Lundahl, Mats. 1992. Politics or Ivlarkets?: Essays on Haitian N.Y.: Routicdge. Underdevelopment. London and New York, N.Y.: Routledge. Moser, Caroline. 1996. Confronting Crisis: A Comparative Study Lybeck, J. A. 1986. The Growth of Government in Developed of Household Responses to Poverty and Vulnerability in Four Economies. Aldershot, Hants, England, and Brookfield, Vt.: Urban Communities. Environmentally Sustainable Develop- Gower. ment Studies and Monographs Series No. 8. Washington, Ma, Jun. 1996. Intergovernmental Relations and Economic Man- D.C.: World Bank. agement in China. Houndsmill, Basingstoke, and London: Mubarak, Jamil Abdalla. 1996. From Bad Policy to Chaos in Macmillan. Somalia: How an Economy FellApart. London: Praeger. Machiavelli, Niccolô. 1513. The Prince. Translated with an Mueller, D. C. 1989. Public Choice II. Cambridge, England: Introduction by George Bull. London: Penguin Books, 1981. Cambridge University Press. Mainwaring, Scott. 1991. "Politicians, Parties, and Electoral Mugoya, Ndungu. 1996. "Tribalism and the Politics of Patron- Systems: Brazil in Comparative Perspective." Comparative age in Kenya." Finance 31 (May): 14-18. Politics (October): 21-43. Murphy, Ricardo Lopez, ed. 1995. Fiscal Decentralization in Manion, Melanie. 1996. "Policy Instruments and Political Con- Latin America. Washington, D.C.: Inter-American Develop- text: Transforming a Culture of Corruption in Hong Kong." ment Bank. Paper presented at the Annual Meeting of the Association Musgrave, R. A. 1976. "Adam Smith on Public Finance and for Asian Studies, Honolulu, Ha., April 11-14. Distribution." In Thomas Wilson and Andrew Skinner, Manor, James. 1996. "The Political Economy of Decentraliza- eds., The Market and the State.' Essays in Honour of Adam tion." Agriculture and Natural Resources Department, Smith. Oxford, England: Oxford University Press. World Bank, Washington, D.C. NaIm, Moises. 1995. "Latin America's Journey to the Market: Margulis, Sergio. 1996. "Environmental Regulation: Instru- From Macroeconomic Shocks to Institutional Therapy." ments and Actual Implementation." Environment Depart- International Center for Economic Growth Occasional Paper ment, World Bank, Washington, D.C. Margulis, Sergio, and Paulo Pereira de Gusmao. 1996. "Prob- No. 62. San Francisco, Calif.: Institute for Contemporary Studies Press. lems of Environmental Management in the Real World: The Rio de Janeiro Experience." Environment Department, Narayan, Deepa. 1995. "The Contribution of People's Partici- World Bank, Washington, D.C. pation: Evidence from 121 Rural Water Supply Projects." Martin, Brendan. 1993. In The Public Interest?: Privatization Environmentally Sustainable Development Occasional and Public Sector Reform London: Zed Books. Paper Series No. 1. World Bank, Washington, D.C. Mauro, Paolo. 1995. "Corruption and Growth." Quarterly Jour- Narayan, Deepa, and Lant Pritchett. 1997. "Cents and Socia- nal ofEconomics 110: 681-712. bility: Household Income and Social Capital in Rural Tan- McLure, Charles E., Jr. 1994. "The Sharing of Taxes on Nat- zania." Environment Department and Policy Research ural Resources and the Future of the Russian Federation." In Department, World Bank, Washington, D.C. Christine I. Wallich, ed., Russia and the Challenge of Fiscal National Bureau of Economic Research. 1997. Penn World Federalism, pp. 181-217. World Bank Regional and Sectoral Tables, Mark 5.6 http://nher.harvard.edu/pwt56.html. Studies. Washington, D.C.: World Bank. Navarro, Juan Carlos. 1996. "Reversal of Fortune: The 1995. "Comment on Prud'homme." World Bank Re- Ephemeral Success of Adjustment in Venezuela, 1989-93." search Observer 10(2): 22 1-26. In Leila Frischtak and Izak Atiyas, eds., Governance, Leader- Meltzer, Alan, and S. F. Richard. 1981. "A Rational Theory of the shp, Communication: Building Constituencies for Economic Size of Government." Journal ofPolitical Economy 89: 914-27. Reform. Essays on Venezuela, Malaysia, Tanzania, Bolivia, Metropolitan Environment Improvement Program. 1996. Egypt, Ghana, Pakistan, Turkey, Brazil. Washington, D.C.: "Japan's Experience in Urban Environmental Manage- World Bank Private Sector Development Department. ment-Yokohama: A Case Study." United Nations Devel- Ndulu, Benno, and Nicholas van de Walle, eds. 1996. Agenda opment Programme, New York, N.Y., and World Bank, for Africa's Economic Renewal. New Brunswick, N.J., and Washington, D.C. Oxford, England: Transaction Publishers. BIBLIOGRAPHICAL NOTE 189 Nellis, John R., and Dennis A. Rondinelli. 1986, "Assessing Osborne, David, and Ted Gaebler. 1993. Reinventing Govern- Decentralization Policies in Developing Countries: The ment: How the Entrepreneurial Spirit Is Transforming the Case for Cautious Optimism." Development Policy Review Public Sector. New York, N.Y.: Penguin. 4(1): 3-23. Ottaway, Marina. 1994. Democratization and Ethnic National- Noonan, John T., Jr. 1994. Bribes. Berkeley, Calif.: University ism: African and Eastern European Experiences. Policy Essay of California Press. No. 14. Washington, D.C.: Overseas Development Council. North, Douglass. 1990. Institutions, Institutional Change and Oxley, Les. 1994. "Cointegration, Causality and Wagner's Law: Economic Performance. Cambridge, England: Cambridge A Test for Britain, 1870-1913." Scottish Journal of Political University Press. Economy 41(3): 286-98. 1993. Paper prepared for the Nobel Prize Lecture in Pardey, Philip G., and others. 1996. Hidden Harvest: US. Ben- Economic Science, Stockholm. efits from International Research Aid. Washington, D.C. Nunberg, Barbara. 1995. Managing the Civil Service: Reform International Food Policy Research Institute. Lessons from Advanced Industrialized Countries. World Bank Pasha, Hafiz. 1996. "Governance and Fiscal Reform: A Study of Discussion Paper No. 204. Washington, D.C.: World Bank. Pakistan," In Leila Frischtak and Izak Atiyas, eds., Gover- Forthcoming. The State After Communism: Administra- nance, Leadership, Communication: Building Constituencies for tive Transitions in Central and Eastern Europe. Public Man- Economic Reform. Essays on Venezuela, Malaysia, Tanzania, agement Division, ECA-MENA Technical Department, Bolivia, Egypt, Ghana, Pakistan, Turkey, Brazil. Washington, World Bank, Washington, D.C. D.C.: World Bank Private Sector Development Department. Oates, Wallace E. 1972. Fiscal Federalism. New York, N.Y.: Paul, Samuel. 1994. "Does Voice Matter? For Public Account- Harcourt Brace Jovanovich. ability, Yes." Policy Research Working Paper No. 1388. Pol- 1985. "Searching for the Leviathan: An Empirical icy Research Department, Finance and Private Sector Devel- Study." American Economic Review 74(4): 748-57. OECD (Organization for Economic Cooperation and Develop- opment Division, World Bank, Washington, D.C. ment). 1993. Managing with Market- Type Mechanisms. Peacock, A. T., and J. Wiseman. 1961. The Growth of Public Paris: OECDIPUMA. Expenditure in the United Kingdom. NBER General Series 1994. DAC Orientations on Participatory Development No. 72. Princeton, N.J.: Princeton University Press. and Good Governance, Vol. 2, No. 2. OECD Working Peters, B. Guy. 1996. The Future of Governing: Four Emerging Papers. Paris: OECD. Models. Lawrence, Kan.: University Press of Kansas. 1996. "Social Expenditure Statistics of OECD Mem- Peterson, George E. 1997. Decentralization in Latin America: ber Countries (Provisional Version)." Labor Market and Learning Through Experience. World Bank Latin American Social Policy Occasional Papers No. 17. Paris: OECD. and Caribbean Studies. Washington, D.C.: World Bank. Various years. OECD Economic Outlook. Paris: OECD. Peterson, Paul E. 1995. The Price of Federalism. A Twentieth Office of the United Nations High Commissioner for Refugees. Century Fund Book. Washington, D.C.: Brookings Institu- 1995. The State of the World's Refugees: in Search of Solutions. tion. New York, N.Y.: Office of the United Nations High Com- Picciotto, Robert. 1995. Putting International Economics to missioner for Refugees. Work: From Participation to Governance, World Bank Dis- Ohno, Kenichi. 1996. "Replicability of Selective Industrial Pol- cussion Paper No. 304. Washington, D.C.: World Bank. icy: The Case of Five Capital-Intensive Industries in Viet- 1996. "What Is Education Worth? From Production nam." Saitama University and Tsukuba University, Japan. Function to Institutional Capital." Human Capital Devel- Okazaki, Tetsuji. 1997. "The Government-Firm Relationship opment Working Papers No. 75. World Bank, Washington, in Postwar Japanese Economic Recovery: Co-ordinating the D.C. Co-ordination Failure in Industrial Rationalization." In Picks Currency Yearbook. Various years. New York, N.Y.: Pick Masahiko Aoki, Hyung-Ki Kim, and Masahiro Okuno- Publishing. Fujiwara, eds., The Role of Government in East Asian Eco- Pierson, Paul. 1994. Dismantling the Welfare State. New York, nomic Development: Comparative Institutional Analysis. N.Y.: Cambridge University Press. Oxford, England: Oxford University Press. Pitschas, Rainer, and Rolf SOlzer, eds. 1995. New Institutional- Olson, Mancur, Jr. 1971. The Logic of Collective Action. Cam- ism in Development Policy: Perspectives and General Condi- bridge, Mass.: Harvard University Press. tions of Public Administration Development in the South and 1996. "Big Bills Left on the Sidewalks: Why Some Nations are Rich, and Others Poor." Journal of Economic East. Berlin: Duncker and Humblot. Perspectives 10(2): 3-24. Platteau, Jean Philippe. 1991. "Traditional Systems of Social Orstrom, Elinor. 1996. "Crossing the Divide: Co-production, Security." In E. Ahmad and others, eds., Social Security in Synergy and Development." World Development 24(6): Developing Countries. Oxford, England: Oxford University 1073-87. Press. Orstrom, Elinor, Larry Schroeder, and Susan Wynne. 1993. Polizatto, Vincent P. 1992. "Prudential Regulation and Bank- Institutional Incentives and Sustainable Development: Infra- ing Supervision." In Dimitri Vittas, ed., Financial Regula- structure Policies in Perspective. Boulder, Cob., San Fran- tion.' Changing the Rules of the Game. EDT Development cisco, Calif., and Oxford, England: Westview Press. Studies. Washington, D.C.: World Bank. 190 WORLD DEVELOPMENT REPORT 1997 Poterba, James M. 1994. "State Responses to Fiscal Crises: The 1995. Controlling Environmental Policy: The Limits of Effect of Budgetary Institutions and Politics." Journal of Public Law in Germany and the United States. New Haven, Political Economy 102(4): 799. Conn.: Yale University Press. Ptadhan, Sanjay. 1996. Evaluating Public Spending: A Frame- Forthcoming. "The Political Economy of Corruption." work for Public Expenditure Review. World Bank Discussion In Kimberly Ann Elliott, ed., Corruption in the World Economy. Paper No. 323. Washington, D.C.: World Bank. Washington, D.C.: Institute for International Economics. Pritchett, Lant, and L. H. Summers. 1996. "Wealthier Is Roy, Jayanta, ed. 1995. Macroeconomic Management and Fiscal Healthier." Journal ofHuman Resources 1(4): 84 1-68. Decentralization. EDI Seminar Series. Washington, D.C.: Prud'homme, Remy. 1995. "The Dangers of Decentralization." World Bank. World Bank Research Observer 10(2): 20 1-20. Ruzindana, Augustine. 1995. "Combating Corruption in Psacharopoulos, George. 1995. Building Human Capitalfor Bet- Uganda." In Petter Langseth, J. Katorobo, E. Brett, and J. ter Lives. Washington, D.C.: World Bank. Munene, eds,, Uganda: Landmarks in Rebuilding a Nation. Psacharopoulos, George, and Nguyen Xuan Nguyen. 1997. The Kampala: Fountain Publishers. Role of Government and the Private Sector in Fighting Poverty. Salamon, Lester M., and Helmut K. Anheier. 1994. The Emerg- World Bank Technical Paper No. 346. Washington, D.C.: ing Sector-An Overview. Baltimore, Md.: The Johns Hop- World Bank. kins University Institute for Policy Studies. Psacharopoulos, George, and P. Zafiris Tzannatos. 1992. 1996. "The Nonprofit Sector: A New Global Force." Women c Employment and Pay in Latin America: Overview Working Paper No. 21. The Johns Hopkins Comparative and Methodology. World Bank Regional and Sectoral Stud- Nonprofit Sector Project, The Johns Hopkins University ies. Washington, D.C.: World Bank. Institute for Policy Studies, Baltimore, Md. Putnam, Robert, with Robert Leonardi and Rafaella Y. Nanetti. 1993. Making Democracy Work: Civic Traditions in Modern Saunders, Anthony, and Berry Wilson. 1995. "Contingent Lia- Italy. Princeton, N.J.: Princeton University Press. bility in Banking: Useful Policy for Developing Countries?" Quah, Jon. 1993. "Controlling Corruption in City-States: A World Bank Policy Research Working Paper No. 1538. Pol- Comparative Study of Hong Kong and Singapore." Paper icy Research Department, Finance and Private Sector Devel- presented at a conference on "The East Asian Miracle: Eco- opment Division, World Bank, Washington, D.C. nomic Growth and Public Policy," Stanford University, Palo Scharpf, Fritz W. 1994. Optionen des Foderalismus in Deutsch- Alto, Calif., October 25-26. land und Europa. Frankfurt, Germany, and New York, N.Y.: Ram, Ran. 1986. "Government Size and Economic Growth: A Campus Verlag. New Framework and Some Evidence from Cross-Section Schwart, Joseph. 1994. "Democratic Solidarity and the Crisis of and Time-Series Data." American Economic Review 76(1): the Welfare State." In Lyman Legters, John Burke, and 191-203. Arthur Diquatro, eds., Critical Perspectives on Democracy. 1987. "Wagner's Hypothesis in Time Series and Cross- Lanham, Md.: Rowman & Littlefield. Section Perspectives: Evidence from 'Real' Data for 115 Scott, Graham. 1995. "Improving Fiscal Responsibility." Countries." Review ofEconomics and Statistics 69: 194-204. Extract from Agenda: A Journal of Policy Analysis and Reform Ramey, Garey, and Valerie Ramey. 1995. "Cross-Country Evi- 2(1): 3-16. Heidelberg, Victoria, Australia. dence on the Link Between Volatility and Growth." Ameri- 1996. "The Use of Contracting in the Public Sector." can Economic Review 85(5): 1138-51. Australian Journal ofPublic Administration 55(3). Reinert, Erik S. 1996. "The Role of the State in Economic Sen, Amartya. 1987. The Standard of Living. Cambridge, Eng- Growth." Paper prepared for a conference on The Rise and land: Cambridge University Press. Fall of Public Enterprises in Western Countries in Milan, Senghaas, Dieter. 1985. The European Experience: A Historical October 10-12. Critique of Development Theories. Oxford, England: Berg Rietbergen-McCracken, Jennifer, ed. 1996. Participation in Publishers. Practice: The Experience of the World Bank and Other Stake- Serven, Luis, and Andres Solimano, eds. 1994. Striving for holders. World Bank Discussion Paper No. 333. Washing- ton, D.C.: World Bank. Growth after Adjustment: The Role of Capital Formation. Rodrik, Dani. 1996. "Why Do More Open Economies Have Washington, D.C.: World Bank. Larger Governments?" John F. Kennedy School of Govern- Shah, Anwar. 1994. The Reform ofIntergovernmental Fiscal Rela- ment, Harvard University, Cambridge, Mass. tions in Developing and Emerging Market Economies. Wash- Romer, Paul M. 1994. "The Origins of Endogenous Growth." ington, D.C.: World Bank. Journal ofEconomic Perspectives 8(1): 3-22. 1997. "Fostering Responsible and Accountable Perfor- Rose, Richard, and Christian Haerpfer. 1990. "New Democra- mance: Lessons from Decentralization Experience." Paper cies Barometer III: Learning from What is Happening," presented an the OED Annual Conference, World Bank, Studies in Public Policy. Washington, D.C., April 1-2. Rose-Ackerman, Susan. 1978. Corruption: A Study in Political Sheng, Andrew. 1992. "Bank Restructuring in Malaysia." In Economy. New York, N.Y.: Academic Press. Dimitri Vittas, ed. Financial Regulation: Changing the Rules 1992. Rethinking the Progressive Agenda: The Reform of of the Game. ED! Development Studies. Washington, D.C.: the American Regulatory State. New York, N.Y.: Free Press. World Bank. BIBLIOGRAPHICAL NOTE 191 Shihata, Ibrahim F. I. 1996. "Implementation, Enforcement sented at the United Nations Third Meeting of the High and Compliance With International Environmental Agree- Level Group on Development Strategy and Management of ments-Views from the World Bank." Legal Department, the Market Economy, Helsinki, July 8-9. World Bank, Washington, D.C. 1996. "Keynote Address: The Role of Government in Shiratori, Masaki, and Yoshio Wada. 1996. "World Bank- Economic Development." In Michael Bruno and Boris OECF Periodic Conference: General Meeting on 'Role of Pleskovic, eds., Annual World Bank Conference on Develop- Government.' "World Bank, Washington, D.C. ment Economics, pp. 11-23. Washington, D.C.: World Bank. Shleifer, Andrei. 1996. "Government in Transition." Discus- Stiglitz, Joseph, and Marilou Uy. 1996. "Financial Markets, sion Paper No. 1783. Harvard Institute of Economic Public Policy and the East Asian Miracle." World Bank Research, Harvard University, Cambridge, Mass. Research Observer 11(2): 249-76. Shleifer, Andrei, and Robert W. Vishny. 1993. "Corruption." Stone, Andrew, Brian Levy, and Ricardo Paredes. 1996. "Public Quarterly Journal of Economics 108: 599-617. Institutions and Private Transactions: A Comparative Analy- Shui Yan Tang. 1992. Institutions and Collective Action. Self sis of the Legal and Regulatory Environment for Business Governance in Irrigation. San Francisco, Calif.: Institute for Transactions in Brazil and Chile." In Lee J. Alston, Thrainn Contemporary Studies. Eggertsson, and Douglass C. North, eds., Empirical Studies Sills, David L., ed. 1968. International Encyclopedia of the Social in Institutional Change. New York, N.Y.: Cambridge Uni- Sciences. New York, N.Y.: Macmillan and the Free Press, and versity Press. London: Collier-Macmillan. Strong, Maurice F. 1996. "The CGIAR at Twenty-Five: Look- Silverman, Jerry M. 1992. Public Sector Decentralization: Eco- ing Back and Looking Forward." Sir John Crawford Memo- nomic Policy and Sector Investment Programs. World Bank rial Lecture, International Centers Week, Washington, Technical Paper Number 188, Africa Technical Department D.C., October 28. Series. Washington, D.C.: World Bank. Stuart, Charles E. 1981. "Swedish Tax Rates, Labor Supply, and Slemrod, Joel. 1995. "What Do Cross-Country Studies Teach Tax Revenues." Journal of Political Economy 89(5 1): about Government Involvement, Prosperity and Economic 1020-5 1. Growth?" Brookings Papers on Economic Activity 2: 373-431. Subbarao, Kalanidhi, Aniruddha Bonnerjee, Jeanine Braith- Snider, Lewis. 1996. Growth, Debt and Politics: Economic waite, Soniya Carvalho, Kene Ezemenari, Carol Graham, Adjustment and the Political Performance ofDeveloping Coun- and Alan Thompson. 1997. Safety Net Programs and Poverty tries. Boulder, Cob.: Westview Press. Reduction.' Lessons from Cross Country Experience. Directions Solow, Robert. 1956. "A Contribution to the Theory of Eco- in Development Series. Washington, D.C.: World Bank. nomic Growth." Quarterly Journal of Economics 70: 65-94. Summers, Robert, and Alan Heston. 1991. "The Penn World Soyinka, Wole. 1996. The Open Sore of a Continent: A Personal Table (Mark 5): An Expanded Set of International Compar- Narrative of the Nigerian Crisis. New York, N.Y., and isons, 1950-1988." Quarterly Journal of Economics 106 Oxford, England: Oxford University Press. (May): 327-68. Spiller, Pablo T., and Cezley Sampson. 1996. "Telecommuni- Sweden. Ministry of Health and Social Affairs. 1994. "Pension cations Regulation in Jamaica." In Brian Levy and Pablo Reform in Sweden: A Short Summary." Stockholm: Min- Spiller, eds., Regulations, Institutions and Commitment: Com- istry of Health and Social Affairs. parative Studies of Telecommunications. Cambridge, England: Sweden. Styrelsen for internationell Urveckling. 1994. State, Cambridge University Press. Market and Aid: Redefined Roles. Stockholm: Swedish Inter- Spiller, Pablo, and Ingo Vogelsang. 1996. "The United King- national Development Authority. dom: A Pacesetter in Regulatory Incentives. In Brian Levy Tallroth, Nils Borje, III. 1997. "The Political Economy of and Pablo Spiller, eds., Regulations, Institutions and Commit- Modern Wars in Africa." Paper presented at the Carter Cen- ment: Comparative Studies of Telecommunications. Cam- bridge, England: Cambridge University Press. ter Conference on the Transition from War to Peace, Stein, Freiherr vom. 1807. "Uber die zweckmaige Bildung der Atlanta, February 19-21. obersten und der Provinzial-, Finanz- und Polizeibchorden Tanzi, Vito. l995a. "Fiscal Federalism and Decentralization: A in der preuischen Monarchie (Nassauer Denkschrift)." Review of Some Efficiency and Macroeconomic Aspects." In (Reprinted in Walther Hubatsch, ed., 1959. Freiherr Michael Bruno and Boris Pleskovic, eds,, Annual World Stein. Briefe und amtliche Schrifien, pp. 380-98. Stuttgart, Bank Conference on Development Economics, pp. 295-3 16. Germany: Kohlhammer.) Washington, D.C.: World Bank. Stiglitz, Joseph E. 1977. "The Theory of Local Public Goods." 1995b. Taxation in an Integrating World. Washington, In Martin S. Feldstein and Robert P. Inman, eds., The Eco- D.C.: Brookings Institution. nomics ofPublic Services, pp. 274-333. London: Macmillan. Tanzi, Vito, and Ludger Schuknecht. 1995. "The Growth of 1986. Economics of the Public Sector. New York, N.Y., Government and the Reform of the State in Industrial and London: W. W. Norton. Countries." IMF Working Papers WP/95/130. Interna- 1994. Whither Socialism? Cambridge, Mass.: MIT tional Monetary Fund, Washington, D.C. Press. Tavares, Jose, and Romain Wacziarg. 1996. "How Democracy 1995. "The Theory of International Public Goods and Fosters Growth?" Department of Economics, Harvard Uni- the Architecture of International Organizations." Paper pre- versity, Cambridge, Mass. 192 WORLD DEVELOPMENT REPORT 1997 Tendler, 5udith. 1997. Good Government in the Tropics. Balti- Vittas, Dimitri, ed. 1992. Financial Regulation: Changing the more, Md.: Johns Hopkins University Press. Rules of the Game. EDI Development Studies. Washington, Tengs, Tammy 0., and John D. Graham. 1996. "The Oppor- D.C.: World Bank. tunity Costs of Haphazard Social Investments in Life- von Hagen, J., and I. Harden. 1994. "National Budget Processes Saving." In Robert W. Hahn, ed., Risks, Costs and Lives and Fiscal Performance." European Economy, Reports and Saved: Getting Better Results om Regulation. New York, Studies3: 311-418. N.Y.: Oxford University Press. Wade, Robert. 1994. "The Governance of Infrastructure: Orga- Teranishi, Juro, and Yukata Kosai, eds. 1993. The Japanese nizational Issues in the Operation and Maintenance of Irri- Experience of Economic Reforms. New York, N.Y.: St. Mar- gation Canals." World Bank, Washington, D.C. tin's Press. Wallich, Christine I. 1994. "Intergovernmental Fiscal Relations: Thomas, Vinod, and Jisoon Lee. 1997. "The Payoffs from Eco- Setting the Stage." In Christine I. Wallich, ed., Russia and nomic Reforms." In Nancy Birdsall and Frederick Jaspersen, the Challenge of Fiscal Federalism, pp. 19-63. World Bank eds., Pathways to Growth. Washington, D.C.: Inter-Ameri- Regional and Sectoral Studies. Washington, D.C.: World Bank. can Development Bank. Weal, Albert. 1990. "Equality, Social Welfare and the Welfare Tilly, Charles. 1990. Coercion, Capital, and Europ ean States, AD State." Ethics 100(3): 475. 990-1990. Cambridge, Mass.: Basil Blackwell. Weaver, R. Kent, and William T. Dickens, eds. 1995. Looking Tlaiye, Laura, and Dan Biller. 1994. "Successful Environmental Before We Leap: Social Science and Welfare Reform. Washing- Institutions: Iessons from Colombia and Curitiba, Brazil." ton, D.C.: Brookings Institution. LATEN Dissemination Note No. 12. Latin America Tech- Wiesner Durán, Eduardo. 1992. "Colombia: Descentralización nical Department, Environment Unit, World Bank, Wash- y Federalismo Fiscal: Informe Final de la Misión para Ia ington, D.C. Descentralización." Departarnento Nacional de Planeación, Toye, J. 1992. "Interest Group Politics and the Implementation Presidencia de la Repüblica, Bogota, Colombia. of Adjustment Policies in Sub-Saharan Africa." Journal of Wildasin, David E. 1996. "Introduction: Fiscal Aspects of International Development Policy, Economic and International Evolving Federations." International Tax and Public Finance Relations 4(2), March-April: 183-98 (special issue on The (Netherlands) 3: 21-131. Economic Analysis of Aid Policy). Williamson, John, ed. 1994. The Political Economy of Policy Trebilcock, Michael J. 1996. "What Makes Poor Countries Reform. Washington, D.C.: Institute for International Eco- Poor? The Role of Institutional Capital in Economic Devel- nomics. opment." Faculty of Law, University of Toronto, Toronto. Willman, John, Stephen Pollard, Bernard Jenkin, Madsen Pine, Twain, Mark, Charles Warner, and Ward Just. 1877. The Eamonn Butler, and José Piflera. 1996. "The Need for Wel- GilcledAge: A Tale of Today. Reprinted by New York, N.Y.: fare Reform." In Over to You: The Transition to Funded For- Oxford University Press, 1996. tune Accounts. London: Adam Smith Institute. United Nations Development Programme. 1997. Reconceptual- Williamson, Oliver E. 1996. The Mechanisms of Governance. ising Governance. Discussion Paper 2. New York, N.Y.: Nw York, N.Y.: Oxford University Press. Management Development and Governance Division, Winston, Clifford. 1993. "Economic Deregulation: Days of Bureau for Policy and Programme Support, United Nations Reckoning for Microeconomists." Journal ofEconomic Liter- Development Programme. ature 3 1(3): 1263-89. U.S. Department of Health and Human Services. 1994. Social Wolf, Edward. 1985. "Social Security, Pensions and the Wealth Security Programs Throughout the World. SSA Publication Holdings of the Poor." Institute for Research on Poverty No. 13-11805, Research Report No. 63. Office of Research Discussion Paper No. 799-85. University of Wisconsin- and Statistics, Social Security Administration, Washington, Madison, Madison, Wise. D.C. Wong, John. 1997. "Government Participation in Economic Development: Singapore Experiences." Paper presented at Valdeavilla, Ermelita V. 1995. "Breakthroughs and Challenges the International Symposium on Government's Role in the of Making Philippine Government Work for Gender Equal- Market Economy, China Institute for Reform and Develop- ity." IDS Bulletin 26(3): 94-101. Brighton, England: Insti- ment, Haikou, China, January 7-8. ture for Development Studies, University of Sussex. World Bank. 1983. World Development Report 1983. New York, van der Gaag, Jacques. 1995. Private and Public Initiatives: N.Y.: Oxford University Press. Working Together for Health and Education. Washington, 1987. "The Philippines: Issues and Policies in the D.C.: World Bank. Industrial Sector." Report No. 6706-PH. Asia Region, van de Waite, Dominique, and Kimberley Nead, eds. 1995. Country Department II, World Bank, Washington, D.C. Public Spending and the Poor: Theoiy and Evidence. Balti- l991a. The Reform of Public Sector Management: more, Md.: Johns Hopkins University Press. Lessons ofExperience. Policy Research Series 18. World Bank, Villadsen, Søren, and Francis Lubanga, eds. 1996. Democratic Washington, D.C. Decentralisation in Uganda: A New Approach to Local Gover- 1991 b. World Development Report 1991: The Challenge nance. Kampala, Uganda: Fountain. of Development. New York, N.Y.: Oxford University Press. BIBLIOORAPHICAL NOTE 193 1993. The East Asian Miracle: Economic Growth anti and Finance Division, Country Department I, South Asia Public Policy. A World Bank Policy Research Report. New Region, World Bank, Washington, D.C. York, N.Y.: Oxford University Press. 1 996c. Handbook on Good Practices for Laws Relating to 1 994a. Adjustment in Africa: Reforms, Results and the Nongovernmental Organizations. Prepared for the World Road Ahead. A World Bank Policy Research Report. New Bank by the International Center for Non-Profit Law. York, N.Y.: Oxford University Press. 1 996d. "Republic of Guinea: Public Expenditure 1 994b. Averting the OldAge Crisis: Policies to Protect the Review." Western Africa Department, World Bank, Wash- Old and Promote Growth. A World Bank Policy Research ington, D.C. Report. New York, N.Y.: Oxford University Press. 1 996e. Social Assistance and Poverty- Targeted Programs: 1 994c. World Development Report 1994: Infrastructure A Sourcebook Prepared by the SocialAssistance Program Team. for Development. New York, N.Y.: Oxford University Press. Washington, D.C.: Poverty and Social Policy Department, 1995a. "Azerbaijan: Baku Water Supply Rehabilitation World Bank. Project." Environment Department Papers Series. Assess- 1 996f. Social Indicators of Development 1996 Wash- ment Series Paper No. 017. World Bank, Washington, D.C. ington, D.C.: World Bank. 1995b. Better Urban Services: Finding the Right Incen- 1 99Cg. "Technical Annex: Democratic Republic of tives. Washington, D.C.: World Bank. Sri Lanka: Telecommunications Regulation and Public Enterprise Reform Technical Assistance Project." Report 1995c. Bureaucrats in Business: The Economics and Pol- No. T-6730 CE. Country Department I, Energy and Project itics of Government Ownership. A World Bank Policy Finance Division, South Asia Region, World Bank, Wash- Research Report. New York, N.Y.: Oxford University Press. ington, D.C. 1 995d. Development in Practice: Priorities and Strategies 1996h. The World Bank Participation Sourcebook. for EducationA World Bank Review. Washington, D.C.: Washington, D.C.: World Bank. World Bank. 1996i. World Development Report 1996: From Plan To 1995e. Global Economic Prospects and the Developing Market. New York, N.Y.: Oxford University Press. Countries. Washington, D.C.: World Bank. 1997. "Ukraine Public Sector Reform Loan, Prepara- I 995E "Vietnam Poverty Assessment and Strategy." tion Mission Report, Public Administration." World Bank, Report No. 13442-VN. Country Department I, East Asia Washington, D.C. and Pacific Region, World Bank, Washington, D.C. Forthcoming. The Road to Financial Integration: Pri- l995g. World Development Report 1995.' Workers in an vate Capital Flows to Developing Countries. Washington, Integrating World. New York, N.Y.: Oxford University Press. D.C.: World Bank. 1 996a. "Argentina: Reforming Provincial Utilities: Wyzan, M. L. Michael, ed. 1990. The Political Economy ofEth- Issues, Challenges and Best Practice." Report No. 1 5063-AR. nic Discrimination and Affirmative Action.' A Comparative Infrastructure Division, Country Department I, Latin America Perspective. New York, N.Y.: Praeger. and the Caribbean Region, World Bank, Washington, D.C. Zartman, William I., ed. 1995. Collapsed States: The Disintegra- 1996b. "Bangladesh: Government That Works: tion and Restoration of Legitimate Authority. Boulder, Cob., Reforming the Public Sector." Private Sector Development and London: Lynne Rienner. SELECTED INDICATORS ON PUBLIC FINANCE ON CENTRAL GOVERNMENT REVENUE AND Ex- Goods and services expenditure includes all government DATA penditure are from IMF, various years (a) and (b), payments in exchange for goods and services, whether in and IMF data files. The accounts of each country are the form of wages and salaries to employees or other pur- reported using the system of common definitions and chases. Wages and salaries consists of all cash payments to classifications in IMF 1986. See these sources for com- employees in return for services rendered, before taxes and plete and authoritative explanations of concepts, defini- pension contributions. Interest payments are payments for tions, and data sources. the use of borrowed money to domestic sectors and non- residents. Subsidies and other current transfers includes all Table A.1 Central government revenue unrequited, nonrepayable transfers on current account to Current revenue includes tax receipts and nonrepayable private and public enterprises, and the cost of covering the receipts (other than grants) from the sale of land, intangi- cash operating deficits of departmental enterprise sales to ble assets, government stocks, or fixed capital assets, or the public. Capital expenditure is expenditure to acquire from capital transfers from nongovernmental sources. fixed capital assets, land, intangible assets, government Income, profit, and capital gains taxes are levied on the stocks, and nonmilitary, nonfinancial assets. Also in- actual or presumptive net income of individuals, on enter- cluded are capital grants. prise profits, and on capital gains. Social security contribu- Table A.3 Central government expenditure by function tions includes employers' and employees' contributions and those of self-employed and unemployed people. Central government expenditure data by function are often Domestic goods and services taxes include general sales and incomplete, and coverage varies from country to country turnover or value added taxes, selective excises on goods, because functional responsibilities stretch across levels of selective taxes on services, taxes on the use of goods or government for which no data are available. Consequently, property, and profits of fiscal monopolies. International the data presented, especially those for education and trade taxes include import and export duties, profits of health, may not always be comparable across countries. export or import monopolies, exchange profits, and Health covers public expenditure on hospitals, mater- exchange taxes. Other taxes includes employers' payroll or nity and dental centers, and some clinics; on national labor taxes, taxes on property, and taxes not allocable else- health insurance schemes; and on family planning and pre- where. Nontax revenue includes requited nonrepayable ventive care. Education comprises expenditure on prepri- receipts for public purposes, such as fines, administrative mary, primary, and secondary schools; universities and fees, or entrepreneurial income from government owner- colleges; and vocational, technical, and other training insti- ship of property, and voluntary, unrequited nonrepayable tutions. Social security and welfare covers compensation for receipts other than from government sources. loss of income to the sick and temporarily disabled; pay- ments to the elderly, the permanently disabled, and the Table A.2 Central government expenditure by unemployed; family, maternity, and child allowances; and economic type the cost of welfare services, such as care of the aged, the dis- Government expenditure includes all nonrepayable pay- abled, and children. Defense comprises all expenditures, ments, whether current or capital, requited or unrequited. whether by defense or other departments, to maintain mil- Expenditure can be measured either by economic type as itary forces. Other expenditures include general public ser- shown here or by function as in Table A.3. vices, interest payments, and items not included elsewhere. 196 WORLD DEVELOPMENT REPORT 1997 Table A.1 Central government revenue Percentage of total current revenue Tax revenue Income, profit, Social security Domestic goods International and capital gains contributions and services trade Other taxes Nontax revenue 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 Low-income economies I Mozambique . . .. .. .. .. .. .. .. .. S 2 Ethiopia 27.5 26.5 22.2 25.7 20.2 18.5 0.7 1.7 27.5 23.4 3 Tanzania 28.7 ,. .. 55.0 .. 9.6 1.3 .. 4.6 4 Burundi 22.4 .. 2.9 28.7 .. 24.0 2.3 .. 10.8 5 Malawi 35.0 . . 32.2 . . 19.0 . . 0.5 . . 13.3 6 Chad 19.0 22.6 .. .. 35.3 33.7 27.5 15.3 66 5.2 9.5 21.8 7 Rwanda 17.1 14.7 64 3.5 33.6 35.2 261 31.4 3.3 3.8 13.0 11.2 8 Sierra Leone 26.0 22.8 22.7 35.4 41.2 37.7 1.2 0.3 8.8 3.8 9 Nepal 8.6 8.8 38.7 39.1 29.2 28.0 0.2 17.2 19.3 10 Niger . . .. . . . . 11 Burkina Faso 17.8 21.5 8.2 17.6 . . 35.8 . . 5.1 2.0 13.7 25.0 12 Madagascar 11.4 16.0 .. 21.0 24.6 45.9 46.7 0.7 0.4 19.6 11.1 13 Bangladesh 9.6 .. 25.3 32.2 1.5 29.5 14 Uganda 8.2 .. . . 31.1 58.6 . . 0.1 3.1 15 Vietnam . . , . . . . . . . . . . 16 Guinea-Bissau 9,9 .. 2.1 . 15.8 .. 33.5 8.8 .. 36.6 17 Haiti 14.9 .. 0.3 . 30.7 .. 23.8 62 .. 11.5 18 Mali 12.3 . . 4.7 31.8 . . 21.5 . . 11.0 , , 11.2 19 Nigeria 50.5 ,. .. 69 .. 12.1 .. 0.2 .. 50.6 20 Yemen, Rep. 25.8 22.8 10.2 10.8 168 19.1 4.5 2.7 42.6 42.9 21 Cambodia .. .. .. .. .. .. .. .. 22 Kenya 28.6 28.5 . . 39.9 47.5 20.1 12.4 0.9 1.0 10.6 10.6 23 Mongolia . . 34.8 .. 11.6 . . 24.3 .. 13.1 .. 0.4 .. 16.0 24 Togo 33.5 . . 6.4 11.6 . . 31.2 . . 0.2 . . 13.2 25 Gambia, The 15.8 13.3 . . . . 10.5 38.0 65.4 42.4 0.8 0.4 7.4 5.9 26 Central African Republic 161 .. 64 .. 20.8 . . 39.8 .. 7.4 . . 9.1 27 India 16.0 19.7 .. . . 37.8 32.9 26.2 23.3 0.1 0.2 19.5 23.8 28 LaoPDR 29 Benin .. .. .. .. .. .. .. .. .. 30 Nicaragua 13.5 11.2 9.7 11.8 43.8 45.1 13.2 19.4 2.1 5.3 9.5 6.6 31 Ghana 22.5 17.7 . . 28.2 37,1 37.2 31.2 0.1 . . 12.1 14.0 32 Zambia 33.2 33.5 . . 41.2 42.5 17.8 18.5 1.2 0.1 6.4 5.1 33 Angola , , .. .. .. , , 34 Georgia .. . . . , , , .. .. .. . 35 Pakistan 12.3 14.2 . . , . 33.0 32.0 31.7 26.5 0.0 . , 22.8 26.8 36 Mauritania , , . . . . . , , , . 37 Azerbaijan , , , , . . , , . . . . . , , , . 38 Zimbabwe 44.8 44.4 ., 29.0 263 14.7 19.0 0.8 0.8 10.4 9.3 39 Guinea 8.6 10.4 166 25.1 47.0 461 0.4 0.8 27.4 17.6 40 Honduras 24.2 .. ,. 25.9 42.4 1,0 5.7 41 Senegal 20.7 .. 3.5 260 .. 35.6 .. 2.1 .. 7.1 42 China 31.2 33.5 .. . . 17.6 29.2 13.8 17.7 .. 0.8 37.4 19.4 43 Cameroon 39.0 18.1 5.7 5.6 15.1 18.4 19.7 16.9 2.9 3.6 15.0 33.4 44 Côte d'lvoire . . . , , . . , . . . . . 45 Albania 8.4 15.2 .. 39.6 . 14.0 0.4 21.5 46 Congo 47 Kyrgyz Republic . . .. .. .. .. .. .. .. 48 Sri Lanka 13.4 13.2 38.5 49.6 32.6 22.1 . . . , 12.7 11.1 49 Armenia .. .. .. .. , , .. , . .. Middle-income economies Lower-middle-income 50 Lesotho 11.3 15.4 . . . . 16.8 15.9 59.7 55.1 0.4 0.1 11.9 13.4 51 Egypt, Arab Rep. 16.5 20.6 13.1 10.5 11.4 12,8 15.7 10.4 6.4 9.0 35.9 36.0 52 Bolivia 3.7 4.3 8.7 7.5 37.5 38.5 12.5 6.6 3.8 0.5 29.2 33.9 53 Macedonia, FYR 54 Moldova 55 Uzbekistan .. .. .. .. .. .. .. .. .. 56 Indonesia 61.9 52.2 . 2.0 16.9 28.0 5.2 5.4 0.5 0.5 14.1 11.5 57 Philippines 23.6 29.2 . . . . 36.3 26.7 24.1 28.1 2.1 3.6 13.1 12.2 58 Morocco 19.0 22.6 4.7 3.8 39.4 39.9 17.2 18.0 3.9 1,4 13.5 12.0 59 Syrian Arab Republic 26.6 27.4 20.6 34.0 7.9 11.5 6.1 6.0 36.7 19.9 60 Papua New Guinea 46.0 45.0 . . , . 12.8 10.8 23.9 23.8 1.8 1.9 15.5 18.4 61 Bulgaria 32.4 20.7 20.9 25.4 20.0 21.1 1.8 6.9 1.2 4.3 23.7 21.6 62 Kazaksran . . .. .. . . .. .. .5 , 63 Guatemala 15.7 19.4 . . 39.4 42,4 21.2 20.1 7.5 3.5 14.8 13.7 64 Ecuador 53.9 56.0 -. . . 20.8 24.9 20.2 11.6 1.1 1.0 2.5 5.5 65 Dominican Republic 19.4 17.1 4.1 4.1 28.4 25.1 33.7 44.2 1.4 0.3 12.3 8.6 66 Romania 18.9 32.4 15.4 29.6 32.6 22.7 0.5 4.0 11.1 2.5 68.3 8.8 67 Jamaica 34.5 .. 4.6 38.0 .. 7.7 .. 8.7 .. 4,8 68 Jordan 12.0 11.1 13.8 20.6 34.0 31.6 5.6 5.3 30.0 26.9 69 Algeria .. .. .. .. .. .. S. 70 El Salvador 20.2 22.2 39.9 49.8 25.8 17.0 0.8 0.2 8.3 7.1 71 Ukraine .. -. .. .. .. .. .. .. .. 72 Paraguay 12.7 9.8 13.1 .. 27.7 22.6 14.6 12.7 14.8 30.0 11.9 16.7 Note Components may not add to 100 percent because of adjustments to current revenue totals by the reporting countries. Figures in italics are for years other than those specified. APPENDIX 197 Percentage of total current revenue Tax revenue Income, profit, Social security Domestic goods International and capital gains contributions and services trade Other taxes Nontax revenue 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 73 Tunisia 13.7 13.6 9.0 12.3 21.0 23.2 27.4 28.4 2.6 2.9 24.4 17.8 74 Lithuania . . 20.7 . . 29.8 . . 38.2 . 3.7 . . 0.3 . . 6.5 75 Colombia 24.6 39.3 10.3 28.2 37.2 17.0 9.7 6.2 0.7 14.4 13.2 76 Namibia 34.9 26.2 .. 21.9 26.6 29.3 33.6 0.3 0.4 12.9 12.7 77 Bclarus 11.9 32.2 .. 39.9 .. 4.5 8.7 .. 2.0 78 Russian Federation . 14.2 33.6 . 32.7 . 12.1 1.4 .. 64 79 Latvia .. 11.5 .. 34.9 .. 40.0 .. 3.6 .. 0.4 .. 9.8 80 Peru 14.6 12.5 7.4 12.3 49.8 49.3 20.8 9.8 8.0 7.3 9.1 8.6 81 Costa Rica 12.3 9.8 26.2 27.7 26.0 32.2 25.5 16.1 1.2 0.7 10.6 13.1 82 Lebanon .. 10.3 .. .. .. 68 .. 35.5 .. 67 .. 30.5 83 Thailand 20.3 28.4 0.2 1.1 46.5 40.7 21.3 17.6 0.7 0.8 9.3 9.1 84 Panama 20.7 18.0 22.4 20.5 15.5 16.8 9.9 10.5 1.7 1.7 25.6 31.1 85 Turkey 44.4 37.0 . . . . 28.0 35.4 6.9 4.1 3.7 2.3 15.6 20.1 86 Poland 25.8 27.6 24.0 24.7 30.0 28.0 66 8.1 64 1.1 5.3 10.4 87 Estonia . . 20.5 33.6 38.8 1.8 . . 0.5 4.6 88 Slovak Republic . . . . . . . . . . . . . . . . . 89 Botswana 35.6 29.7 . . . . 1.3 3.0 20.4 18.6 . . . . 42.5 48.6 90 Venezuela 60.0 50.6 3.9 5.5 5.5 10.4 12.6 9.1 1.1 1.3 18.3 22.3 Upper-middle-income 91 South Africa 51.4 50.9 .5 1.8 31.0 34.9 4.4 2.9 1.4 1.3 8.7 6.5 92 Croatia 10.7 . . 36.9 . . 38.1 8.3 . . 0.5 . . 5.1 93 Mexico 27.9 34.2 12.9 17.9 55.3 50.2 10.1 7.1 1.0 2.4 10.7 9.0 94 Mauritius 12.8 12.5 4.4 5.1 19.2 23.9 49.2 41.0 0.4 0.7 10.9 11.1 95 Gabon 40.2 276 0.9 0.8 10.8 23.7 17.4 17.4 1.8 0.9 29.2 29.3 96 Brazil 17.4 16.3 24.8 30.1 21.0 19.2 2.7 1.9 4.5 5.2 29.5 27.2 97 Trinidad and Tobago 62.3 . . . . . . 11.0 8.0 .. 0.6 . . 17.7 98 Czech Republic 16.6 . . 38.0 . . 31.9 3.9 . . 1.7 . . 7.5 99 Malaysia 34.8 34.3 0.6 1.0 17.8 22.0 20.5 14.1 1.8 3.5 24.1 24.6 100 Hungary 16.9 . . 22.8 .. 34.7 6.2 . . 3.5 . . 13.2 101 Chile 13.3 18.1 8.4 6.7 41.8 45.2 8.7 9.6 6.3 3.7 21.1 16.7 102 Oman 24.3 19.9 . . . . 0.7 1.0 2.5 3.2 0.5 0.7 71.9 74.9 103 Uruguay 7.4 7.3 26.1 30.1 42.2 32.4 11.4 5.6 4.9 8.4 6.4 5.9 04 Saudi Arabia 105 Argentina 4.9 216 26.4 45.9 33.1 266 1318 7.9 6.0 4.6 11.9 Al 106 Slovenia 107 Greece 18.5 29.6 31.7 1.1 37.8 66.5 o1 o1 4.7 3.9 9.7 6.4 Highincome economies 108 Korea, Rep. 27.1 31.6 2.2 6.9 40.9 33.6 14.2 6.9 2.7 6.0 11.4 12.1 109 Portugal 21.6 24.7 25.0 25.0 35.1 35.2 3.4 0.2 5.5 2.8 8.3 11.8 110 Spain 25.6 31.2 41.5 38.2 19.9 21.5 3.3 0.6 0.3 . . 7.9 8.1 111 NewZealand 59.7 58.2 .. . . 21.5 27.4 3.4 2.1 1.1 1.4 13.4 10.1 112 Ireland 33.9 38.6 13.6 14.7 31.2 30.9 8.2 5.9 1.4 1.5 10.0 6.4 113 tlsrael 38.2 37.4 8.4 7.7 28.8 35.9 3.9 1.2 3.5 3.1 16.7 13.7 114 tKuwait 1.7 . . . . . . 0.6 . . 1.6 . . 96.0 115 tUnited Arab Emirates . . . . 2.5 2.1 24.1 26.0 . . . . . . . . 77.3 71.9 116 United Kingdom 39.2 35.2 16.9 16.9 29.7 32.3 0.1 0.1 1.0 0.1 10.7 8.3 117 Australia 61.5 63.9 . . . . 22.8 20.4 4.8 3.3 0.5 1.5 10.1 10.9 118 Italy 36.3 36.6 33.7 29.2 25.6 28.8 0.0 0.0 1.4 1.2 4.0 2.9 119 Canada 50.3 49.3 13.6 18.4 19.2 18.2 4.4 2.6 0.0 . . 12.5 11.4 120 Finland 30.5 29.0 9.5 11.4 46.7 44.2 1.1 0.9 0.3 0.3 8.4 11.6 121 tHongKong .. . . .. .. . . .. .. .. .. . 122 Sweden 16.9 8.9 31.5 36.0 28.6 31.6 0.6 0.8 5.1 3.0 14.5 15.8 123 Netherlands 26.0 28.7 39.0 38.5 20.6 21.6 . . . . 0.4 0.9 12.0 8.0 124 Belgium 37.2 33.6 33.4 35.4 23.5 24.7 0.0 0.0 0.1 . . 3.7 3.8 125 France 17.6 17.7 42.1 44.1 29.4 27.5 0.0 0.0 1.2 2.0 7.7 6.8 126 tSingapore 27.2 28.0 . . . . 15.3 18.3 1.7 3.7 8.9 5.4 40.1 36.5 127 Austria 19.6 19.9 36.3 37.5 26.2 24.6 1.4 1.4 6.5 8.3 6.5 8.6 128 United States 51.4 50.9 32.4 34.5 4.4 3.9 1.5 1.5 0.0 0.0 9.2 8.1 129 Germany 17.3 15.5 53.9 47.7 22.5 25.2 0.0 . . .. . 6.2 . 6.2 130 Denmark 36.7 38.3 3.9 3.9 42.6 39.6 0.1 0.1 0.7 1.4 13.2 14.8 131 Norway 21.4 16.0 22.7 24.0 37.9 36.7 0.5 0.6 0.1 0.0 16.4 21.9 132 Japan 67.8 38.7 . . 25.6 17.4 13.9 1.7 1.2 3.6 1.6 5.2 15.4 133 Switzerland 14.9 14.7 48.7 52.6 19.2 16.9 8.4 6.4 .. 6.0 6.6 t Economies classified by the United Nations or otherwise regarded by their authorities as developing. As ofJuly 1, 1997, Hong Kong is a part of China. 198 WORLD DEVELOPMENT REPORT 1997 Table A.2 Central government expenditure by economic type Percentage of total expenditure Subsidies and other Goods and services Wages and salanes Interest payments current transfers Capital expenditure 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 Low-income economies 1 Ivlozarnbique Ethiopia .. .. . .. .. 0 2 78.0 74.6 35.0 43.9 4.9 72 7.2 14.9 20.5 13.1 3 Tanzania 52.4 19.0 .. 8.9 13.1 28.6 4 Burundi 39.4 25.5 .. 1.9 68 . 45.5 5 Malawi 47.9 18.2 17.0 6.5 . 28.6 6 Chad 34.1 40.9 24.3 23.9 1.2 2.8 2.5 2.7 61.7 60.3 7 Rwanda 47.9 60.2 29.3 27.3 5.3 8.9 14.7 7.1 32.9 31.6 8 Sierra Leone 54.1 36.4 27.0 18.8 15.6 23.7 68 19.4 24.3 30.0 9 Nepal .. . . . . 10 Niger 11 Burkina Faso 63.8 460 53.0 37.8 5.8 8.3 12.8 11.6 15.3 34.0 12 Madagascar 37.5 32.1 260 20.9 10.8 21.9 8.1 7.1 39.9 35.4 13 Bangladesh .. . 14 Uganda 16.9 15 Vietnam . . . . . 16 Guinea-Bissau 22.9 13.9 61 .. 4.0 .. 62.1 17 Haiti 70.5 .. .. 3.4 .. 15.2 .. 10.8 18 Mali 37.1 25.8 . . 2.4 6.5 . . 3.3 19 Nigeria 162 .. 9.5 .. 31.8 .. 123 .. 38.5 20 Yemen, Rep. 64.2 68.1 55.0 58.4 8.4 10.0 65 8.2 33.4 13.6 21 Cambodia .. .. .. .. .. .. .. .. 22 Kenya 53.2 49.8 31.4 31.1 15.7 26.3 14.8 8.4 16.3 15.4 23 Mongolia .. 32.1 .. 8.2 .. 2.5 .. 47.7 .. 17.7 24 Togo 48.1 26.2 12.7 13.0 .. 28.5 25 Gambia, The 49.9 25.3 8.0 .. 10.4 .. 30.9 30.7 26 CentralAfrican Republic 67.0 .. 53.6 .. 1.2 .. 160 .. 60 27 India 27.2 22.8 12.3 10.0 17.1 25.6 42.6 40.1 13.1 11.5 28 Lao PDR 29 Benin .. .. .. .. .. .. .. 30 Nicaragua 59.9 47.9 14.6 27.9 4.0 10.7 14.6 19.2 17.2 22.1 31 Ghana 55.5 47.4 32.1 31.1 12.7 13.8 14.8 20.7 16.2 18.1 32 Zambia 47.3 37.6 262 19.6 9.9 18.4 19.0 16.9 19.7 26.0 33 Angola . . .. 34 Georgia .. .. .. .. .. .. .. 35 Pakistan 50.0 45.7 16.9 23.7 18.9 13.7 14.2 16.8 36 Mauritania . . . 37 Azerbaijan .. .. .. . . . . .. .. .. 38 Zimbabwe 48.6 55.1 30.4 38.0 12.5 15.5 30.6 18.2 8.2 11.1 39 Guinea 363 39.4 17.2 22.0 5.8 72 5.8 4.4 52.1 48.9 40 Honduras .. .. 41 Senegal 52.4 .. 363 7.5 169 169 42 China .. .. .. .. .. .. .. 43 Cameroon 47.3 59.0 30.0 47.8 3.0 8.1 11.9 14.4 37.9 16.0 44 Côte d'lvoire . . . . . . . 45 Albania 263 11.8 7.6 .. 48.4 .. 17.8 46 Congo .. .. .. 45.6 47 Kyrgyz Republic .. .. .. .. .. .. .. 48 Sri Lanka 26.3 34.4 14.2 18.4 16.7 21.8 19.6 21.6 37.4 22.2 49 Armenia . . Middle-income economies Lower-middle-income 50 Lesotho 43.8 51.8 23.5 29.3 9.0 7.1 4.9 60 42.3 33.2 51 Egypt, Arab Rep. 42.4 33.6 20.3 17.8 9.2 18.9 33.0 25.2 16.1 22.3 52 Bolivia 62.6 58.0 41.6 32.8 66 8.3 18.9 13.5 11.8 20.2 53 Macedonia, FYR . . 54 Moldova :. .. 55 Uzbekistan . . . . . . .. . . . . . . . . . 56 Indonesia 23.5 25.4 14.3 16.0 9.9 11.3 19.6 14.1 47.0 48.6 57 Philippines 52.0 42.5 29.0 28.0 23.1 29.8 6.8 9.9 18.1 17.8 58 Morocco 47.3 50.6 34.1 37.3 14.6 17.9 12.5 9.1 25.7 22.5 59 Syrian Arab Republic . . . . . . . . . . . . . . . . 30.1 31.9 60 Papua New Guinea 61.9 57.0 35.2 28.7 9.5 9.2 19.1 22.5 9.6 11.3 61 Bulgaria 34.0 26.6 2.8 5.7 69 24.6 54.2 46.0 4.9 2.9 62 Kazakstan .. -. .. , . . . . . .. . 63 Guatemala 52.8 50.2 37.3 34.4 9.7 11.4 14.0 15.5 25.6 25.6 64 Ecuador 32.0 46.1 28.9 42.2 17.8 22.7 265 11.2 19.2 16.7 65 Dominican Republic 47.1 33.9 34.8 22.9 4.1 6.1 13.3 10.2 34.2 48.7 66 Romania 25.9 33.4 5.9 15.8 1.8 1.8 34.2 53.9 39.0 10.8 67 Jamaica 44.1 .. 18.4 .. 30.7 .. 10.5 .. 14.7 68 Jordan 53.9 57.9 48.9 43.4 9.0 12.4 12.4 10.5 26.0 19.6 69 Algeria .. . . . . . , . . . . . . , . 70 El Salvador 61.2 51.3 51.0 41.9 8.3 13.2 14.0 17.4 10.6 16.8 71 Ukraine .. .. .. .. .. .. .. .. 72 Paraguay 55.9 53.7 35.3 42.4 6.2 7.0 20.9 22.5 16.9 16.8 Note: Components may not add to 100 percent because of adjustments to total expenditure by the reporting countries. Figures in italics are for years other than those specified. APPENDIX 199 Percentage of total expenditure Subsidies and other Goods and services Wages and salailesa Interest payments current transfers Capital expenditure 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 73 Tunisia 35.3 35.9 26.9 29.2 7.5 10.3 30.9 32.9 27.0 20.9 74 Lithuania . . 31.5 . . 11.8 . 0.7 . . 55.8 . . 12.2 75 Colombia 28.1 25.2 19.3 16.2 7.3 10.0 45.7 42.7 20.7 22.1 76 Namibia 51.5 72.5 21.8 .. 5.8 1.1 25.7 10.9 169 15.5 77 Belarus .. 364 2.1 .. 1.8 463 15.5 78 Russian Federation .. 39.6 13.7 .. 9.7 49.4 .. 4.6 79 Latvia .. 368 .. 18.6 .. 2.4 .. 566 .. 4.2 80 Peru 46.0 29.7 18.5 15.6 21.6 19.3 15.8 35.7 16.6 15.3 81 Costa Rica 49.4 50.5 38.9 38.0 9.2 15.2 29.2 23.9 15.3 10.5 82 Lebanon .. 52.9 .. 37.0 .. 266 .. .. .. 20.5 83 Thailand 59.9 58.4 30.8 32.1 13.8 5.5 8.0 7.5 18.4 28.6 84 Panama 57.3 53.0 37.6 40.1 15.8 9.7 18.9 28.9 7.9 8.3 85 Turkey 39.9 46.3 25.3 34.5 11.8 14.4 28.1 27.9 20.2 11.4 86 Poland 18.2 261 14.1 0.5 10.3 7415 60.1 69 3.5 87 Estonia 25.1 .. 7.8 .. 0.2 72.8 7.5 88 Slovak Republic . . . . . . . . . . , . . . . . 89 Borswana 48.8 53.0 25.7 27.9 3.3 2.1 26.0 27.0 21.8 17.9 90 Venezuela 37.4 30.9 29.5 24.4 12.4 17.4 29.8 34.7 19.2 15.4 Upper middle mcome 91 South Africa 49.6 47.4 21.5 31.9 11.8 15.2 27.8 29.0 10.7 8.4 92 Croatia . , 57.5 .. 21.3 , . 3.1 . . 33.2 . . 6.2 93 Mexico 23.7 32.6 17.9 23.4 40.6 22.7 19.2 29.9 16.7 15.1 94 Mauritius 43.1 46.1 34.1 35.1 17.7 11.6 24.6 23.5 14.6 18.7 95 Gabon 35.8 .. 23.0 27,1 5.5 .. 62 3.8 45.3 45.6 96 Brazil 16.5 15.1 9.4 8.7 39.7 44.5 46.7 47.3 5.5 3.2 97 Trinidad and Tobago 40.6 . , 34.8 .. 6.2 .. 30.2 .. 23.0 98 Czech Republic . . 19.6 . . 8.6 . . 3.7 .. 66.1 . . 10.6 99 Malaysia 45.1 47.9 31.1 30.7 19.8 16.5 14.6 16.6 20.9 19.6 100 Hungaty 19.6 . . 7.0 . . 3.6 . . 67.9 . . 8.9 101 Chile 30.0 28.4 19.3 18.6 5.9 6.3 54.0 50.9 10.1 14.4 102 Oman 71.4 72.1 16.5 22.4 3.9 5.2 4.7 6.5 20.0 16.2 103 Uruguay 39.9 30.3 24.9 15.8 6.5 5.9 46.4 57.5 7.2 6.3 104 Saudi Arabia . . , . . . . . . . . . . , . . . 105 Argentina 28.4 29.3 19.8 22.2 11.6 10.2 50.6 577 9.5 2.8 106 Slovenia .. .. .. .. .. .. .. 107 Greece 39.2 31.4 24.8 22.1 11.1 25.8 38.6 32.3 11.6 10.4 High-income economies 108 Korea, Rep. 39.9 31.7 14.2 13.1 6.3 3.4 39.0 49.5 14.8 15.5 109 Portugal 31.7 38.2 22.0 28.5 16.6 15.8 40.0 33.4 10.5 12.6 110 Spain 29.0 20.5 21.6 14.9 6.6 10.2 53.5 61.7 10.7 7.7 111 NewZealand 25.2 43.6 16.1 10.5 15.2 13.6 54.5 39.4 5.1 3.4 112 Ireland 18.1 18.5 12.8 13.2 17.9 16.0 56.9 58.1 7.1 7.4 113 5lsrael 38.4 34.3 11.6 13.5 22.6 15.2 35.0 39.6 3.9 11.1 114 tKuwait 46.9 41.9 24.8 19.7 24.1 47.8 29.0 10.3 115 1'United Arab Emirates 88.6 88.1 30.6 34.0 . . 8.8 8.2 4.4 3.8 116 United Kingdom 30.6 29.8 12.8 10.4 10.0 7.2 54.1 56.4 5.3 6.6 117 Australia 22.2 23.5 11.1 10.0 8.3 5.5 63.1 65.7 6.4 5.2 118 Italy 160 15.3 11.7 11.7 17.1 21.8 560 56.5 9.5 6.0 119 Canada 20.6 18.9 9.8 9.2 17.2 17.3 60.4 60.9 1.8 1.4 120 Finland 20.3 17.9 10.4 7.8 3.6 6.3 67.0 71.1 9.0 4.7 121 tHongKong .. .. .. .. . . .. .. . 122 Sweden 14.4 14.7 6.3 5.8 13.7 68.611.7 72.9 3.2 2.5 123 Netherlands 14.9 14.5 9.1 8.7 7.5 69.9 9.1 71.9 7.8 4.5 124 Belgium 19.9 18.0 14.2 14.2 17.1 56.620.1 57.4 6.4 4.6 125 France 27.4 24.9 17.8 16.3 4.2 6.0 63.8 64.3 4.6 4.9 126 l'Singapore 49.2 54.1 26.3 28.2 14.7 10.2 7.4 13.0 28.6 22.7 127 Austria 25.9 24.9 10.2 9.7 7.3 9.5 58.0 57.7 8.8 7.9 128 United States 30.1 25.3 10.5 9.3 14.3 14.6 50.6 55.7 5.1 4.4 129 Germany 33.7 30.1 8.6 7.6 4.9 6.3 56.0 58.4 5.3 5.2 130 Denmark 20.4 19.8 12.6 11.4 15.1 13.6 60.3 63.0 4.2 3.5 131 Norway 18.8 19.7 9.0 8.0 7.0 5.6 70.8 69.4 3.4 5.2 132 Japan 13.4 . . . . . . 18.7 . 53.0 . . . 14.9 . 133 Switzerland 29.3 30.4 65 5.0 2.5 2.8 62.1 62.0 62 4.8 a. Included in goods and services. tEconomies classified by the United Nations or otherwise regarded by their authorities as developing. As of July 1, 1997, Hong Kong is a part of China. 200 WORLD DEVELOPMENT REPORT 197 Table A.3 Central government expenditure by function Percentage of total expenditure Social security Health Education and welfare Defense Other 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 Low-income economies 1 Mozambique .. .. .. .. .. .. .. 2 Ethiopia 3.5 4.2 10.7 12.4 5.2 5.8 32.8 30.2 47.7 47.5 3 Tanzania 5.5 11.8 . 0.5 13.3 . . 68.8 4 Burundi ,. .. . . -. 5 Malawi 6.5 11.9 . . 0.8 . . 6.4 . . 74.3 6 Chad .. .. .. 7 Rwanda .. .. .. .. .. 8 Sierra Leone 7.8 .. 14.8 .. 1.9 61 .. 69.4 9 Nepal 4.6 3.8 10.8 12.5 0.6 5.9 5.7 78.4 78.0 10 Niger . . . . . . . . . . . . . . . . . 11 BurkinaFaso 6.1 69 17.3 17.3 5.5 .. 19.1 14.0 53.2 61.8 12 Madagascar 5.6 5.3 14.3 13.5 2.3 1.6 7.2 5.9 73.0 73.7 13 Bangladesh 5.1 9.1 3.4 . . 10.8 . . 71.6 14 Uganda 3.7 12.6 1.5 20.1 . . 62.0 15 Vietnam . . . . . . . . , 16 Guinea-Bissau 6.1 8.0 . . 4.0 . 7.2 . 75.4 17 Haiti ,. .. .. .. .. 18 Mali 2.6 10.0 .. 4.2 .. 8.3 .. 74.8 19 Nigeria 1.7 .. 61 .. 0.3 7.5 .. 84.5 20 Yemen, Rep. 4.1 4.4 17.4 19.7 . . 29.5 30.3 49.0 45.6 21 Cambodia .. .. .. . . . . .. .. . 22 Kenya 6.4 5.6 20.6 20.3 0.1 0.1 10.7 7.6 62.1 66.4 23 Mongolia .. 2.4 .. 3.6 . . 22.2 .. 11.3 . . 60.5 24 Togo 5.1 18.5 7.8 8.0 60.6 25 Gambia, The 7.4 .. 14.8 2.5 4.3 74.0 26 Central African Republic 5.1 .. 176 .. 62 9.7 . . 61.4 27 India 2.0 1.7 2.2 2.0 . . 19.2 15.1 76.7 81.3 28 Lao PDR 29 Benin .. .. .. .. .. .. .. 30 Nicaragua 13.3 13.4 13.7 15.4 8.3 16.3 28.5 7.6 362 47.3 31 Ghana 8.5 7.8 21.9 23.0 6.0 7.0 5.4 4.7 58.2 57.6 32 Zambia 6.5 9.9 11.6 12.8 1.8 2.9 . , 80.1 74.4 33 Angola . 34 Georgia .. .. .. .. 35 Pakistan .. . - .. .. .. 36 Mauritania . . . . . . . .. . . . 37 Azerbaijan .. . .. .. .. .. .. . 38 Zimbabwe 6.8 21.5 . . 4.5 . . 17.1 . . 50.1 39 Guinea .. .. . . .. . . . 40 Honduras 41 Senegal 4.3 .. 18.0 . . 5.5 .. 10.0 .. 62.3 42 China 0.4 0.4 2.1 2.4 0.1 0.1 15.4 16.3 82.0 80.8 43 Catrieroon 3.8 4.8 11.8 18.0 4.6 1.0 7.5 9.4 72.3 668 44 Côte d'Ivoire .. .. . . . . . . 45 Albania .. 5.6 .. 2.3 .. 21.7 7.1 .. 63.3 46 Congo .. 10.3 4.2 -. .. .. 88.9 47 Kyrgyz Republic .. .. .. . . . . .. .. .. 48 Sri Lanka 4.7 5.5 8.6 10.2 10.8 16.9 5.4 11.8 70.5 55.6 49 Armenia . Middle-income economies Lower-middle-income 50 Lesotho 8.9 12.1 15.7 21.3 1.7 1.3 74 61 663 59.2 51 Egypt,ArabRep. 2.6 2.4 11.2 12.0 11.9 10.4 15.4 9.4 58.9 65.8 52 Bolivia 5.5 6.3 20.4 18.0 17.4 15.3 13.6 9.5 43.2 50.8 53 Macedonia, FYR .. .. .. , 54 Moldova .. 55 Uzbekistan . . . . . . . . . . . . . . . . 56 Indonesia 2.2 2.8 9.2 9.8 . . 5.3 9.7 6.6 78.9 79.5 57 Philippines 5.0 3.8 17.6 15.7 1.6 2.7 12.7 10.5 63.0 67.3 58 Morocco 2.9 3.0 17.4 17.8 5.8 61 14.6 14.3 59.3 58.8 59 Syrian Arab Republic 1.4 2.3 8.8 9.0 4.2 2.0 36.3 35.2 49.3 51.5 60 Papua New Guinea 9.3 8.3 17.4 16.2 0.5 0.8 4.6 3.9 68.1 70.8 61 Bulgaria 1.7 3.2 2.8 3.7 21.8 30.7 8.8 7.0 65.0 55.4 62 Kazakstan .. .. .. .. .. .. .. .. 63 Guatemala 7.7 10.1 14.1 16.8 3.5 4.4 14.1 13.9 60.9 58.2 64 Ecuador 9.0 . . 24.9 . . 1.4 . . 12.2 . . 52.6 65 Dominican Republic 10.1 11.0 12.4 10.1 6.4 3.7 7.2 4.9 63.8 70.4 66 Romania 3.1 8.1 3.4 9.7 21.4 27.0 7.1 8.1 65.0 47.1 67 Jamaica 68 Jordan 4.4 6.5 12.4 14.8 1114 14.6 25.8 2216 46.1 41.5 69 Algeria 70 El Salvador 7.5 8.0 16.7 13.4 3.3 4.5 21.7 14.0 50.8 60.1 71 Ukraine .. .. .. .. .. .. .. .. .. 72 Paraguay 4.3 6.3 11.9 17.6 24.6 15.6 11.9 12.1 47.3 48.5 Note: Components may not add to 100 percent because of rounding. Figures in italics are for years other than those specified. APPENDIX 201 Percentage of total expenditure Social security Health Education and welfare Defense Other 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 1981-90 1991-95 73 Tunisia 6.3 65 15.1 17.5 10.7 14.1 7.7 5.5 60.2 564 74 Lithuania . 6.0 .. 6.9 . . 35.8 . 2.4 . 48.9 75 Colombia 4.6 5.4 21.1 18.3 16.7 8.3 8.0 8.0 49.5 60.1 76 Narnibia . . . . . . . . . 77 Belarus .. 2.5 17.6 .. 365 4.1 .. 39.3 78 Russian Federation .. 1.6 2.7 .. 28.1 14.3 . 53.3 79 Latvia .. 64 . 13,8 .. 39.0 .. 3.0 .. 37.9 80 Peru 5.7 . . 16.9 . . . , . . 20.1 . . 57.2 81 CostaRica 25.0 26.7 19.6 21.0 13.2 14.4 2.5 . . 39.9 38.0 82 Lebanon .. .. .. .. .. 83 Thailand 5.7 7.6 19.7 21.3 3.2 3.6 19.1 15.6 52.4 51.8 84 Panama 17.1 20.5 16.0 18.4 14.5 22.4 72 5.2 49.4 33.5 85 Turkey 2.3 3.0 13.6 16.1 1.2 3.1 12.2 10.0 70.8 67.8 86 Poland .. .. 87 Estonia 13.1 8.2 34.9 2.6 . . 41.9 88 Slovak Republic . . . . .. . . . . . . ,. .. . 89 Botswana 5.5 5.1 19.1 21.3 2.3 2.1 9.3 12.1 63.9 59.4 90 Venezuela 8.6 18.3 .. 6.7 6.3 . . 60.2 Upper-middle-income 91 South Africa , , .. . . , . . . . . . 92 Croatia . . 15.3 . . 6.2 . . 32.2 . . 18.7 . . 27.6 93 Mexico 1.4 3.0 11.9 23.7 10.0 21.0 2.2 3.8 74.5 48.6 94 Mauritius 7.9 8.8 14.5 15.5 15.8 15.9 1.1 1.5 60.7 58.3 95 Gabon .. ., .. .. .. .. .. 96 Brazil 7.1 5.9 3.9 3.5 26.8 30.6 3.9 2.8 58.3 57.3 97 Trinidad and Tobago .. .. .. .. 98 Czech Republic . . 17.2 .. 11.2 . . 27.8 . . 6.1 . . 37.7 99 Malaysia 4.9 5.6 18.7 20.4 4.0 5.9 10.7 11.7 61.8 56.3 100 Hungary 3.4 .. 2.0 .. 24.3 .. 4.5 .. 65.8 101 Chile 7.5 11.4 13.2 13.5 37.7 33.5 11.4 9.2 30.3 32.3 102 Oman 4.2 6.0 8.9 12.1 3.0 3.6 44.4 35.4 41.6 42.9 103 Uruguay 4.1 5.4 7.2 6.7 50.6 58.4 10.8 6.5 27.4 23.0 104 Saudi Arabia , , . . . . . , . , . . . . . , , 105 Argentina 2.0 2.2 7.9 70 35.7 47.5 7.9 68 46.8 364 106 Slovenia . , . . . . . . , . . . .. , . . 107 Greece 10.5 7.7 9.6 8.2 30.6 14.2 10.8 8.6 38.4 61.3 High-income economies 108 Korea, Rep. 1.7 1.1 18.9 18.8 6.9 10.0 29.0 20.0 43.5 50.1 109 Portugal 8.7 .. 9.9 . . 24.2 . . 6.0 . . 51.3 110 Spain 8.2 6.2 5.9 4.4 47.0 39.0 4.9 3.7 34.0 46.7 111 NewZealand 12.8 13.5 11.8 14.4 29.4 37.7 4.9 3.8 41.1 30.7 112 Ireland 13.0 14.1 11.7 12.8 25.4 27.5 3.1 3.1 46.9 42.6 113 tlsrael 3.7 5.4 8.5 12.0 17.1 23.5 27.8 20.3 43.0 38.8 114 tKuwait 6.7 4.5 11.9 8.8 10.6 13.7 14.5 36.8 56.3 36.1 115 tUnited Arab Emirates 6.8 7.1 11.4 16.2 3.1 3.4 43.3 37.5 35.5 35.8 116 United Kingdom 13.6 14.0 2.7 4.2 30.0 30.5 13.0 9.2 40.7 42.2 117 Australia 9.8 12.9 7.4 7.4 27.7 32.5 9.1 7.7 46.0 39.5 118 Italy 11.0 . . 8.2 . . 32.8 . . 3.4 . . 44.6 119 Canada 5.8 4.9 3.3 2.8 34.7 40.6 7.8 6.5 48.5 45.1 120 Finland 10.7 2.9 14.1 12.2 32.2 45.3 5.1 4.2 37.9 35.4 121 tHong Kong . . .. .. .. .. .. .. .. 122 Sweden 1.3 0.5 9.2 7.3 46.1 50.1 6.8 5.6 36.6 36.4 123 Netherlands 11.2 13.9 11.0 10.5 36.6 37.4 5.2 4.3 35.9 33.9 124 Belgium 1.9 . . 12.9 . . 40.3 . . 5.0 . . 39.9 125 France 16.1 17.8 7.5 7.0 43.0 42.9 6.6 5.7 26.8 26.6 126 tSingapore 5.5 6.8 19.1 21.0 1.5 3.3 20.6 26.5 53.2 42.3 127 Austria 12.4 13.3 9.6 9.5 45.5 45.4 2.9 2.3 29.6 29.6 128 United States 11.7 16.9 1.9 1.8 29.9 28.5 19.3 24.2 32.3 33.6 129 Germany 18.6 168 0.7 0.8 48.9 45.3 64 8.8 23.1 30.7 130 Denmark 1.2 1.0 9.4 9.8 38.7 41.1 4.5 5.4 45.2 43.6 131 Norway 9.0 3.5 7.9 6.3 35.2 36.7 7.0 8.2 39.7 46.5 132 Japan . . 1.5 . 6.2 . . . 37.5 . . 4.4 . . 50.3 133 Switzerland 13.0 20.7 3.2 2.7 49.1 46.0 10.4 7.1 24.3 23.4 t Economies classified by the United Nations or otherwise regarded by their authorities as developing. As of July 1, 1997, Hong Kong is a part of China, CONTENTS Introduction to Selected World Development Indicators 206 Key and Primary Data Documentation 210 Tables Summary of socioeconomic development indicators Table 1 Basic indicators 214 Table 2 Macroeconomic indicators 216 Table 3 External economic indicators 218 Human resources Table 4 Population and labor force 220 Table 5 Distribution of income or consumption 222 Table 6 Health 224 Table 7 Education 226 Environmental sustainability Table 8 Commercial energy use 228 Table 9 Land use and urbanization 230 Table 10 Forest and water resources 232 Economic performance Table 11 Growth of the economy 234 Table 12 Structure of the economy: production 236 Table 13 Structure of the economy: demand 238 Table 14 Central government budget 240 Table 15 Exports and imports of merchandise 242 Table 16 Balance of payments 244 Table 17 External debt 246 Table I a. Basic indicators for other economies 248 Technical notes 249 Data sources 263 Classification of economies 264 INTRODUCTION TO SELECTED WORLD DEVELOPMENT INDICATORS SELECTED WORLD DEVELOPMENT INDICATORS tion? What kind of external economic environment do THESE provide a core set of data covering three develop- countries face? ment themes: people, the environment, and the economy. Tables 4 to 7, Human resources, show the rate of The layout of the seventeen tables retains the tradition of progress in social development during the past decade. past editions of World Development Report of presenting Data on population growth, labor force participation, and comparative socioeconomic data covering more than 130 income distribution are included. Measures of well-being economies for the most recent year or period for which such as malnutrition and access to health care, school data are available and one earlier year or period. An addi- enrollment ratios, and gender differences with respect to tional table presents basic indicators for seventy-six adult illiteracy are also presented. economies with sparse data or with populations less than Tables 8 to 10, Environmental sustainability, include 1 million. measures of human impacts on the environmentdefor- Most of the indicators presented here have been estation, changing land use patterns, freshwater with- selected from more than 500 indicators covered in the drawals, and emissions of carbon dioxideand of some of new, freestanding World Development Indicators 1997. the activities that cause these impactsenergy use and Published annually, World Development Indicators is the urbanization. Also included is information on the extent World Bank's flagship statistical publication. It features of protected areas that preserve natural habitat and, hence, a broader, more integrated approach to the presentation biodiversity. of development statistics. In its five main sections it rec- Tables 11 to 17, Economic peformance, present infor- ognizes the interplay of a wide range of issues: human cap- mation on economic structure and growth, as well as on ital development, environmental sustainability, macro- foreign investment, external debt, and degree of integra- economic performance, private sector development, and tion into the global economy. the global links that influence the external environment Because the World Bank's primary business is provid- for development. It also features, for the first time, exten- ing lending and policy advice to low- and middle-income sive documentation of the data to highlight potential pit- member countries, the issues covered focus mainly on falls in intercountry and intertemporal comparisons. these economies. Where available, information on the World Development Indicators is complemented by a new high-income economies is also provided for comparison. CD-ROM data base of over 1,000 data tables and 500 Readers may wish to refer to national statistical publica- time-series indicators for 209 economies. tions or publications of the Organization for Economic Cooperation and Development and the European Union More about the Selected World Development for more information on the high-income economies. Indicators Tables 1 to 3, Summmy ofsocioeconomic development indi- Classification of economies cators, offer an overview of key development issues: How As in the Report itself, the main criterion used to classif,r rich or poor are the people? What is the life expectancy of economies and broadly distinguish stages of economic newborns? What percentage of adults are illiterate? How development is GNP per capita. Countries are classified has the economy performed in terms of growth and infla- into three categories according to income. The GNP per INTRODUCTION TO SELECTED WORLD DEVELOPMENT INDICATORS 207 capita cutoff levels in this edition of Selected World national currency figures are described in the Technical Development Indicators are as follows: low-income, $765 Notes. or less in 1995 (forty-nine economies); middle-income, Summary measures $766 to $9,385 (fifty-eight economies); and high-income, $9,386 or more (twenty-six economies). A further divi- Summary measures, presented in the colored bands on sion, at GNP per capita $3,035, is made between lower- each table, are either totals (indicated by t), weighted aver- middle-income and upper-middle-income. Economies are ages (w), or median values (m) calculated for groups of further classified by region. For a list of all economies in economies. Countries for which data are not shown in the each income group and region, including those with pop- main tables have been implicitly included in the summary ulations of fewer than 1 million, see the Classification of measures on the assumption that they followed the trend Economies table at the end of the Selected World Devel- of reporting economies during the period. The countries opment Indicators. excluded from the main tables (those presented in Table la, Basic indicators for other economies) have been Data sources and methodology included in the summary measures when data are available Socioeconomic data presented here are drawn from sev- or, when not, by assuming that they follow the trend of eral sources: primary collection by the World Bank, reporting countries. This gives a more consistent aggre- member-country statistical publications, nongovernmen- gate measure by standardizing country coverage for each tal organizations such as the World Resources Institute, period shown. Where missing information accounts for a and other international organizations such as the United third or more of the overall estimate, however, the group Nations and its specialized agencies, the International measure is reported as not available. The method used for Monetary Fund, and the Organization for Economic computing the summary measures in each table is stated Cooperation and Development. (See Data Sources at the in the technical note for the table. end of the Technical Notes for a complete listing of Terminology and country coverage sources.) Although international standards of coverage, definition, and classification apply to most statistics The term "country" is not meant to imply political inde- reported by countries and international agencies, there pendence but may refer to any territory for which author- are inevitably differences in coverage, currentness, and ities report separate social or economic statistics. the capabilities and resources devoted to basic data col- Data are shown for countries or economies as they lection and compilation. In some cases, competing were constituted in 1995, and historical data are revised to sources of data require review by World Bank staff to reflect current political arrangements. Throughout the ensure that the most reliable data available are presented. tables, exceptions are noted. Where available data are deemed too weak to provide Data for China do not include data for Taiwan, China, reliable measures of levels and trends, or do not ade- unless otherwise noted. As of July 1, 1997, Hong Kong is quately adhere to international standards, the data are a part of China. not shown. Data are shown separately whenever possible for the Data presented in these tables are consistent with those Czech Republic and the Slovak Republic, the countries in World Development Indicators 1997. Differences be- formed from the former Czechoslovakia. tween data in each annual edition reflect not only newly Data are shown separately for Eritrea whenever possi- received information, but also revisions to historical series ble; in most cases prior to 1992, however, they are and changes in methodology. Thus data of different vin- included in the data for Ethiopia. tages may be published in different editions. Readers are Data for Germany refer to the unified Germany unless advised not to compare data series between publications. otherwise noted. Consistent time-series data are available on the World Data for Jordan refer to the East Bank only unless oth- Development Indicators 1997 CD-ROM erwise noted. Considerable effort has been made to standardize the In 1991 the Union of Soviet Socialist Republics was data, but full comparability cannot be ensured, and care formally dissolved into fifteen countries: Armenia, Azer- must be taken in interpreting the indicators. For example, baijan, Belarus, Estonia, Georgia, Kazakstan, Kyrgyz the indicators in Table 5, Distribution of income or con- Republic, Latvia, Lithuania, Moldova, Russian Federa- sumption, are not strictly comparable across countries, tion, Tajikistan, Turkmenistan, Ukraine, and Uzbekistan. because the underlying household surveys differ in Whenever possible, data are shown for the individual method and in the type of data collected. countries. All dollar figures are in current U.S. dollars unless oth- Data for the Republic of Yemen refer to that country erwise stated. The methods used for converting from from 1990 onward; data for previous years refer to the 208 WORLD DEVELOPMENT REPORT 1997 former People's Democratic Republic of Yemen and the Technical Notes, the Key and Primary Data Documenta- former Yemen Arab Republic, unless otherwise noted. tion table, the Classification of Economies table, and the Whenever possible, data are shown for the individual footnotes to the tables. These describe the methods, con- countries formed from the former Yugoslavia: Bosnia and cepts, definitions, and data sources used in compiling Herzegovina, Croatia, the former Yugoslav Republic of the tables. For more extensive documentation see World Macedonia, Slovenia, and the Federal Republic of Develop ment Indicators 1997. The Data Sources section at Yugoslavia (Serbia and Montenegro). the end of the Technical Notes lists sources that contain more comprehensive definitions and descriptions of the Table layout concepts used. The table format of this edition generally follows thar of For more information about the Selected World previous editions of World Development Report. Eco- Development Indicators and the World Bank's other sta- nomies are listed in ascending order of GNP per capita in tistical publications, please contact: all tables except Table la. High-income economies marked by the symbol are those classified by the United Information Center, Development Data Group Nations, or otherwise regarded by their authorities, as The World Bank, 1818 H Street, N.W. developing. Economies with populations of fewer than 1 Washington, D.C. 20433 million and those with sparse data are not shown sepa- Hotline: (800) 590-1906 or (202) 473-7824 rately in the main tables but are included in the aggre- Fax: (202) 522-1498 gates. Basic indicators for these economies may be found E-mail: info@worldbank.org in Table la. The alphabetical list in the Key and Primary World Wide Web: http:// www.worldbank.org or Data Documentation table shows the reference number for each economy. http://www.worldbank.org/wdi. Technical Notes To order World Bank publications, e-mail your re- Because data quality and intercountry comparisons are quest to books@worldbank.org, or write to World Bank ofren problematic, readers are encouraged to consult the Publications at the address above, or call (703) 661-1580. KEY AND PRIMARY DATA DOCUMENTATION KEY AND PRIMARY DATA DOCUMENTATION TABLE In each statistical table of the Selected World Devel- THE below provides an index to the countries included opment Indicators, economies are listed in ascending in the Selected World Development Indicators and addi- order of GNP per capita. The ranking below by GNP tional information on the sources, treatment, and cur- per capita therefore indicates a country's place in the rentness of the principal demographic, economic, and statistical tables. environmental indicators for the 133 countries included Figures in the colored bands in the tables are sum- in the main statistical tables. mary measures for groups of economies. The letter w The World Bank is not a primary data collection means weighted average; m, median; and t, total. agency for most areas other than living standards sur- Except where noted in the Technical Notes, growth veys and debt. As a major user of socioeconomic data, rates for economic data are in real terms. however, the World Bank places particular emphasis The data cutoff date was February 1, 1997. on data documentation to inform users of data in eco- The symbol . . means not available. nomic analysis and policymaking. Differences in the A blank space means not applicable. methods and conventions used by the primary data The figures 0 and 0.0 mean zero or less than half the collectorsusually national statistical agencies, central unit shown. banks, and customs servicesmay give rise to signifi- Figures in italics indicate data that are for years or cant discrepancies over time both among and within periods other than those specified. countries. See the 1997 World Development Indicators The symbol t indicates high-income economies for a more complete treatment of primary data docu- classified by the United Nations, or regarded by their mentation. own authorities, as developing. Govern- GNP per Vital Latest ment National accounts capitaLatest Latest household regis- water Fiscal finance ranking population or demographic tratzon withdrawal year base price accounting Economy in tables census survey complete data end year valuation concept Albania 45 1989 I 1970 Dec.31 1993 VAP Algeria 69 1987 PAPCHILD, 1992 1990 Dec. 31 1980 VAB Angola 33 1970 1987 Dec. 31 1970 VAP Argentina 105 1991 1976 Dec.31 1986 VAP C Armenia 49 1989 I 1989 Dec.31 1993 VAB Australia 117 1991 I 1985 Jun.30 1989 VAP C Austria 127 1991 I 1991 Dec.31 1990 VAP C Azerbaijan 37 1989 I 1989 Dec. 31 1987 VAP Bangladesh 13 1991 DHS, 1994 1987 Jun. 30 1985 VAP Belarus 77 1989 I 1989 Dec.31 1990 VAB C Groups of economies Low-income economies For this map, economies are classified by income group, as they are for the tables that follow. Middle-income economies Low-income economies are those with a GNP per capita of $765 or less in 1995; middle-income, $766$9,385; high-income, $9,386 or more. Five middle-income economiesAmerican Samoa (US), High-income economies Fiji, Kiribati, Tonga, and Western Samoa, one high-income economyFrench Polynesia (Fr), and Tuvalu, for which income data are not available, are not shown on the map because of space constraints. Data not available tlanwa nlan Nethertard Rusaaan Federation Canada ale OtMan lIMO; a-nm. Lrbaano retard POsed Balarus CtraasneeenBerga 1 UkraIne Kaaakstae LaaaernrbOst I MoIova Mana)aIia Lirtastenat Prance Ru ma Usbekietan Kytqy505ttalaSa men United Stateu Porlual, Sn Tartary Tuelartenrnlan Talilalslaa Gatiraoan (taR- Susan stevE Silica Tars ReP. Iraq 1513mm Rep MBghanzutae of lean Jordan Palastan Bh9tae Bahrahe ForeVer Saudr Veobra Laa Peo(tks Spaden Islands (lila) United stab Boeqlados) Deter. Rep Sans cc Emirates Myaoman' MaunHania Cape Verde Ft MOrtars Belrae Jantarca Snare laleleda luSt '1mIhaiIand Gavtant ala Philippreeu The Gambia Bartdea staeas (GO) El Salvador . Nlcaragaa Oainna-BiSaaa mSlama Fase BehOl Faderalad Slates oyttecrorreoa \lsoaods Costa Aic'\ Nigeeli rNnearaeta Dlaya S L eoeg errs SaliarrO (Fr) Caemetoan ColoenSti F or i.---' -, Equatonat Guinea Saber Cn Paua Scionron Broad stS Uislarta Baa Ballad Paraguay Nyanyare (Fr) Calearoaraa IFS Polard MalvgUa and Barlanda Czech anDublac Ukraine Germany Sloaak Republic Uaza101eaye (Pet awl News Aabttra S Dominica COle Sagantina Hungary Mantanaquo (Fr) Slovenia Ramania New CSL LucIa Fed BaNana Zealan St VIncent and a S Barbadite Rep St the Grenadrnns. Marine &anada cedonia - Tenidad Jsnd Tobago - R Seneca M?,uICII.I SELECTED WORLD DEVELOPMENT INDICATORS 21 1 Govern- GNP per Vital Latest ment National accounts capita Latest Latest household regis- water Fiscal finance ranking population or demographic tration withdrawal year base price accounting Economy in tables census survey complete data end year valuation concept Belgium 124 1991 1980 Dec. 31 1985 VAP C Benin 29 1992 WFS, 1981 1994 Dec. 31 1985 VAP Bolivia 52 1992 DHS, 1994 1987 Dec. 31 1980 VAP C Botswana 89 1991 DHS, 1988 1992 Mar. 31 1986 VAP B Brazil 96 1991 DHS, 1991 1990 Dec. 31 1980 VAB C Bulgaria 61 1992 1988 Dec. 31 1990 VAP C Burkina Faso 11 1985 SDA, 1995 1992 Dec. 31 1985 VAB C Burundi 4 1990 1987 Dec. 31 1980 VAB Cambodia 21 1962 1987 Dec. 31 1960 VAP Cameroon 43 1987 DHS, 1991 1987 Jun. 30 1980 VAP C Canada 119 1991 1991 Mar.31 1986 VAB C Central African Rep. 26 1988 DHS, 1994-95 1987 Dcc. 31 1987 VAB Chad 6 1993 1987 Dec. 31 1977 VAB C Chile 101 1992 1975 Dec. 31 1986 VAP C China 42 1990 Population, 1995 1980 Dec. 31 1990 VAP B Colombia 75 1993 DHS, 1995 1987 Dcc. 31 1975 VAP C Congo 46 1984 1987 Dec. 31 1978 VA!' Costa Rica 81 1984 CDC, 1993 1970 Dec. 31 1987 VAP C Côte d'Ivoire 44 1988 DHS, 1994 1986 Dec. 31 1986 VAB C Croatia 92 1991 Dec. 31 1994 VAB C Czech Republic 98 1991 CDC, 1993 1991 Dec. 31 1984 VAP C Denmark 130 1991 1990 Dcc. 31 1980 VAB C Dominican Republic 65 1993 DHS, 1991 1987 Dcc. 31 1970 VAP C Ecuador 64 1990 DHS, 1994 1987 Dec. 31 1975 VAP B Egypt, Arab Rep. 51 1986 DHS, 1995 1992 Jun. 30 1987 VAB C El Salvador 70 1992 CDC, 1994 1975 Dec. 31 1962 VAP B Estonia 87 1989 1989 Dec. 31 1993 VAB C Ethiopia 2 1994 Fam. & fert., 1990 1987 Jul. 7 1981 VAB B Finland 120 1990 1991 Dec. 31 1990 VAB C France 125 1990 Income, 1989 1990 Dec. 31 1980 VAI C Gabon 95 1993 Dec. 31 1989 VAP B Gambia, The 25 1993 1982 Jun. 30 1976 VAB B Georgia 34 1989 1989 Dec. 31 1987 VAB Germany 129 1991 Dec. 31 1990 VAP C Ghana 31 1984 DHS, 1993 1970 Dec. 31 1975 VAP C Greece 107 1991 1980 Dcc. 31 1970 VAB C Guatemala 63 1994 DHS, 1995 1970 Dec. 31 1958 VAP B Guinea 39 1991 SDA, 1991 1991 Dec. 31 1986 VAP Guinea-Bissau 16 1983 SDA, 1994-95 1987 Dec. 31 1989 VAP C Haiti 17 1982 DHS, 1994-95 1987 Sep. 30 1976 VAP Honduras 40 1988 DHS, 1994 1992 Dec. 31 1978 VAB t HongKong 121 1991 Dec.31 1990 VAB Hungary 100 1990 Income, 1995 / 1991 Dec. 31 1991 VAB C India 27 1991 Nat. fam. hlth., 1975 Mar. 31 1980 VAB C 1992-93 Indonesia 56 1990 DHS, 1994 1987 Mar. 31 1993 VAP C Ireland 112 1996 1980 Dec.31 1985 VAB C t Israel 113 1983 1989 Dec.31 1990 VAB C Italy 118 1991 1990 Dec.31 1985 VAP C Jamaica 67 1991 LSMS, 1994 1975 Dec. 31 1986 VAP Japan 132 1990 1990 Mar. 31 1985 VAP C Jordan 68 1994 DHS, 1990 1975 Dec. 31 1990 VAB B 212 WORLD DEVELOPMENT REPORT 1997 Govern- GNP per Vital Latest ment National accounts capita Latest Latest household regis- water Fiscal finance ranking population or demographic tration withdrawal year base price accounting Economy in tables census survey complete data end year valuation concept Kazakstan 62 1989 V 1989 Dec. 31 1994 VAB Kenya 22 1989 DHS, 1993 1990 Jun. 30 1982 VAB B Korea, Rep. 108 1995 1992 Dec.31 1990 VAP C t Kuwait 114 1995 V 1974 Jun. 30 1984 VAP C Kyrgyz Republic 47 1989 LSMS, 1994 V 1989 Dec. 31 1993 VAB Lao PDR 28 1985 1987 Dec. 31 1990 VA1 Latvia 79 1989 V 1989 Dec.31 1993 VAB C Lebanon 82 1970 1975 Dec.31 1990 VAB Lesotho 50 1986 DHS, 1991 1987 Mar. 31 1980 VAB C Lithuania 74 1989 V 1989 Dec.31 1993 VAB C Macedonia, FYR 53 1994 V Dec. 31 1990 VAP Madagascar 12 1993 SDA, 1993 1984 Dec. 31 1984 VAB C Malawi 5 1987 DHS, 1992 1994 Mar. 31 1978 VAB B Malaysia 99 1991 V 1975 Dec.31 1978 VAP C Mali 18 1987 DHS, 1987 1987 Dec. 31 1987 VAB Mauritania 36 1988 PAPCHILD, 1990 1985 Dec. 31 1985 VAB Mauritius 94 1990 CDC, 1991 1974 Jun.30 1992 VAB C Mexico 93 1990 DHS, 1987 1991 Dec. 31 1980 VAP C Moldova 54 1989 V 1989 Dec. 31 1993 VAB Mongolia 23 1989 1987 Dec. 31 1986 VAB C Morocco 58 1994 DHS, 1995 1992 Dec. 31 1980 VAP C Mozambique 1 1980 1992 Dec. 31 1987 VAB Namibia 76 1991 DHS, 1992 1991 Mar.31 1990 VAB C Nepal 9 1991 1987 Jul. 14 1985 VAB C Netherlands 123 1971 1991 Dec.31 1990 VAP C New Zealand 111 1991 V 1991 Jun. 30 1982 VAP B Nicaragua 30 1995 LSMS, 1993 1975 Dec.31 1980 VAP C Niger 10 1988 Hsld. bgt. & cons., Dec. 31 1987 VAP 1993 Nigeria 19 1991 Cons. cxpenditure, 1987 Dec.31 1987 VAB 1992 Norway 131 1990 V 1985 Dec.31 1990 VAP C Oman 102 1993 Child health, 1989 1975 Dec.31 1978 VAP B Pakistan 35 1981 LSMS, 1991 1975 Jun.30 1981 VAB C Panama 84 1990 1975 Dec. 31 1992 VAB C Papua New Guinea 60 1990 1987 Dec. 31 1983 VAP B Paraguay 72 1992 CDC, 1992 1987 Dec.31 1982 VAP C Peru 80 1993 LSMS, 1994 1987 Dec.31 1979 VAP C Philippines 57 1990 DHS, 1993 1975 Dec.31 1985 VAP B Poland 86 1988 V 1991 Dec.31 1990 VAP C Portugal 109 1991 V 1990 Dec.31 1985 VAP C Romania 66 1992 LSMS, 1995 V 1994 Dec. 31 1993 VAB C Russian Federation 78 1989 LSMS, 1994 V 1991 Dcc. 31 1993 VAB C Rwanda 7 1991 DHS, 1992 1993 Dec. 31 1985 VAB C Saudi Arabia 104 1992 Mat'!. & child hlth., V 1975 Hijriyear 1970 VAP 1993 Senegal 41 1988 DHS, 1992-93 1987 Dec.31 1987 VAP Sierra leone 8 1985 SHEHEA, 1989-90 1987 Jun.30 1985 VAB B t Singapore 126 1990 V 1975 Mar. 31 1985 VAP C Slovak Republic 88 1991 V 1991 Dec.31 1993 VAP Slovenia 106 1991 V Dec. 31 1992 VAB South Africa 91 1991 LSMS, 1993 1990 Mar.31 1990 VAB C SELECTED WORLD DEVELOPMENT INDICATORS 213 Govern- GNPper Vital Latest ment National accounts capita Latest Latest household regis- water Fiscal finance ranking population or demographic Eration withdrawal year base price accounting Economy in tables census survey complete data end year valuation concept Spain 110 1991 1991 Dec. 31 1996 VAP C Sri Lanka 48 1981 DHS, 1993 I 1970 Dec. 31 1982 VAB C Sweden 122 1990 I 1991 Jun. 30 1990 VAB C Switzerland 133 1990 I 1991 Dec. 31 1990 VAP C Syrian Arab Republic 59 1994 1976 Dec. 31 1985 VAP C Tanzania 3 1988 LSMS, 1993 1994 Jun. 30 1992 VAB Thailand 83 1990 DHS, 1987 1987 Sep. 30 1988 VAP C Togo 24 1981 DHS, 1988 1987 Dec. 31 1978 VAP Trinidad and Tobago 97 1990 DHS, 1987 I 1975 Dec. 31 1985 VAB Tunisia 73 1994 1990 Dec. 31 1990 VAP C Turkey 85 1990 Pop. & health, 1983 1991 Dec. 31 1994 VAB C Uganda 14 1991 DHS, 1995 1970 Jun. 30 1991 VAB Ukraine 71 1989 I 1989 Dec.31 1990 VAB I. United Arab Emirates 115 1980 1980 Dec. 31 1985 VAB B United Kingdom 116 1991 I 1991 Dcc. 31 1990 VAB C United States 128 1990 Current pop., 1994 V 1990 Sep. 30 1985 VAP C Uruguay 103 1985 1965 Dec. 31 1983 VAP C Uzbekistan 55 1989 I 1989 Dec. 31 1987 VAB Venezuela 90 1990 LSMS, 1993 1970 Dec. 31 1984 VAP C Vietnam 15 1989 Intercensal demo., 1992 Dec. 31 1989 VAP 1995 Yemen, Rep. 20 1994 DHS, 1991-92 1987 Dec. 31 1990 VAB C Zambia 32 1990 SDA, 1993 1994 Dec. 31 1977 VAP C Zimbabwe 38 1992 DHS, 1994 1987 Jun. 30 1980 VAB C Notes: Latest population census shows the most recent year in which a census was conducted. Latest household or demographic survey gives information on the surveys used in compiling household and demographic data. PAPCHILD is the Pan Arab Project for Child Development; DHS is Demographic and Health Survey; WFS is World Fertility Study; LSMS is Living Standards Measurement Study; SDA is Social Dimensions of Adjustment; CDC is Centers for Disease Con- trol and Prevention; and SHEHEA is Survey of Household Expenditure and Household Economic Activities. Vital registration complete identifies countries judged to have complete registries of vital statistics (v') by the United Nations Depart- ment of Economic and Social Information and Policy Analysis, Statistical Division, and reported in Population and Vital Statistics Reports. Countries with complete vital statistics registries may have more accurate and more timely demographic indicators. Latest water withdrawal survey refers to the most recent year for which data have been compiled from a variety of sources. Fiscal year end is the date of the end of the fiscal year for the central government. Fiscal years for other levels of government and the reporting years for statistical surveys may differ, but if a country is designated as a fiscal year reporter in the following column, the date shown is the end of its national accounts reporting period. National accounts base year is the year used as the base period for constant price calculations in the country's national accounts. Price indexes derived from national accounts aggregates, such as the GDP deflator, express the price level relative to prices in the base year. Constant price data reported in the World Bank are partially rebased to a common 1987 base year. National accounts price valuation shows whether value added in the national accounts is reported at basic or producers' prices (VAB) or at purchasers' prices (yAP). Purchasers' prices include the value of taxes levied on value added and collected from consumers and thus tend to overstate the actual value added in production. Government finance accounting concept describes the accounting basis for reporting central government financial data. For most coun- tries government finance data have been consolidated (C) into one set of accounts capturing all fiscal activities of the central govern- ment. Budgetary central government accounts (B) exclude central government units. 214 WORLD DEVELOPMENT REPORT 1997 Table 1. Basic indicators PPP estimates of GNP Poverty Life GNP per capitaa per capitab % of people expectancy Population Surface area Avg. ann. living on less than at birth Adult (millions) (thousands Dollars growth 1%) Current int'l $ $1 a day (PPP) (years) illiteracy 1%) mid-1995 of sq. hm) 1995 1985-95 1987 1995 1995 1981-95 1995 1995 Low-income economies 3,179.9 t 40,606 t 430w 3.8w 63w 34w Excluding China and India 1,050.3 27,758 t 290w -1.4w 56w 46w Mozambique 16.2 802 80 3.6 2.5 3.0 810' . . 47 60 2 Ethiopia 56.4 1,097 100 -0.3 2.0 1.7 450 33.8 49 65 3 Tanzania1 29.6 945 120 1.0 2.6 2.4 640 16.4 51 32 4 Burundi 6.3 28 160 -1.3 3.2 2.3 630' . . 49 65 5 Malawi 9.8 118 170 -0.7 3.1 2.8 750 . . 43 44 6 Chad 6.4 1,284 180 0.6 2.5 2.6 700' 48 52 7 Rwanda 6.4 26 180 -5.4 3.8 2.0 540 45.7 46 40 8 Sierra Leone 4.2 72 180 -3.6 3.2 2.2 580 40 9 Nepal 21.5 141 200 2.4 4.0 4.3 1,170' 53.1 55 73 10 Niger 9.0 1,267 220 3.6 2.8 750' 61.5 47 86 11 Burkina Faso 10.4 274 230 -0.2 3.3 2.9 780' 49 81 12 Madagascar 13.7 587 230 -2.2 3.1 2.4 640 72.3 52 13 Bangladesh 119.8 144 240 2.1 4.8 5.1 1,380 58 62 14 Uganda 19.2 236 240 2.7 4.7 5.5 1,470' 42 38 15 Vietnam 73.5 332 240 68 6 16 Guinea-Bissau 1.1 36 250 2.0 2.8 2.9 790' 87.0 38 45 17 Haiti 7.2 28 250 -5.2 5.8 3.4 910' 57 55 18 Mali 9.8 1,240 250 0.8 2.3 2.0 550 50 69 19 Nigeria 111.3 924 260 1.2 4.4 4.5 1,220 28.9 53 43 20 Yemen, Rep. 15.3 528 260 53 21 Cambodia 10.0 181 270 53 35 22 Kenya 26.7 580 280 0.1 5.7 5.1 1,380 50.2 58 22 23 Mongolia 2.5 1,567 310 -3.8 10.6 7.2 1,950 65 24 Togo 4.1 57 310 -2.7 5.5 4.2 1,130' 56 48 25 Gambia, The 1.1 11 320 4.5 3.5 930' 46 61 26 Central African Republic 3.3 623 340 -2.4 5.0 4.0 1,070' 48 40 27 India 929.4 3,288 340 3.2' 4.4 5.2 1,400 52.5 62 48 28 Lao PDR 4.9 237 350 2.7 52 43 29 Benin 5.5 113 370 -0.3 6.9 6.5 1,760 50 63 30 Nicaragua 4.4 130 380 -5.4 11.8 7.4 2,000' 43.8 68 34 31 Ghana 17.1 239 390 1.4 7.4 7.4 1,990' 59 32 Zambia 9.0 753 400 -0.8 4.2 3.5 930 84.6 46 22 33 Angola 10.8 1,247 410 -6.1 8.9 4.9 1,310 47 34 Georgiat 5.4 70 440 -17.0 28.1 5.5 1,470 73 35 Pakistan 129.9 796 460 1.2 8.4 8.3 2,230 11.6 60 62 36 Mauritania 2.3 1,026 460 0.5 6.0 5.7 1,540' 31.4 51 37 Azerbaija& 7.5 87 480 -16.3 21.8 5.4 1,460 70 38 Zimbabwe 11.0 391 540 -0.6 8.6 7.5 2,030 41.0 57 '5 39 Guinea 6.6 246 550 1.4 26.3 44 40 Hondutas 5.9 112 600 0.1 7.9 7.0 1,91 46.5 67 27 41 Senegal 8.5 197 600 7.3 6.6 1,780 54.0 50 67 42 China 1,200.2 9,561 620 8.3 6.3 10.8 2,920 29.4 69 '9 43 Cameroon 13.3 475 650 -6.6 15.1 7.8 2,110 57 37 44 Core d'Ivoire 14.0 322 660 8.2 5.9 1,580 17.7 55 60 45 Albania 3.3 29 670 73 46 Congo 2.6 342 680 -3.2 11.5 7.6 2,050 51 25 47 Kyrgyz Republict 4.5 199 700 -6.9 13.6 6.7 1,800 18.9 68 48 Sri Lanka 18.1 66 700 2.6 10.6 12.1 3,250 4.0 72 10 49 Armenia1 3.8 30 730 -15.1 25.4 8.4 2,260 7' Middle-income economies ,9o9 60,838 2,390w -0.7w 68w 18w Lower-middle-income [,i26 40,323 l,670w -1.3w 67w 50 Lesotho 2.0 30 770 1.2 6.1 6.6 1,780' 50.4 61 29 51 Egypt, Arab Rep. 57.8 1,001 790 1.1 14.3 14.2 3,820 7.6 63 49 52 Bolivia 7.4 1,099 800 1.8 9.1 9.4 2,540 7.1 60 17 53 Macedonia, FYR 2.1 26 860 73 54 Moldova1 4.3 34 920 . . 6.8 69 55 Uzbekistant 22.8 447 970 -3.9 12.6 8.8 2,370 70 56 Indonesia 193.3 1,905 980 6.0 9.8 14.1 3,800 14.5 64 16 57 Philippines 68.6 300 1,050 1.5 10.3 10.6 2,850 27.5 66 5 58 Morocco 26.6 447 1,110 0.9 13.2 12.4 3,340 1.1 65 56 59 Syrian Arab Republic 14.1 185 1,120 0.9 18.5 19.7 5,320 68 60 Papua New Guinea 4.3 463 1,160 2.3 8.5 9.0 2,420' 57 28 61 Bulgaria 8.4 111 1,330 -2.6 23.4 16.6 4,480 2.6 71 62 Kazakstant 16.6 2,717 1,330 -8.6 24.2 11.2 3,010 69 63 Guatemala 10.6 109 1,340 0.3 13.2 12.4 3,340 53.3 66 44 64 Ecuador 11.5 284 1,390 0.8 15.8 15.6 4,220 30.4 69 10 65 Dominican Republic 7.8 49 1,460 2.1 13.7 14.3 3,870 19.9 71 18 66 Romania 22.7 238 1,480 -3.8 22.2 16.2 4,360 17.7 70 67 Jamaica 2.5 II 1,510 3.6 11.3 13.1 3,540 4.7 74 15 68 Jordan 4.2 89 1,510 -4.5 23.8 15.1 4,060' 2.5 70 '3 69 Algeria 28.0 2,382 1,600 -2.4 26.5 19.6 5,300 1.6 70 38 70 El Salvador 5.6 21 1,610 2.8 8.2 9.7 2,610 67 29 71 Ukraine1 51.6 604 1,630 -9.2 20.7 8.9 2,400 69 72 Paraguay 4.8 407 1,690 1.2 13.3 13.5 3,650 68 8 Note: For other economies, see Table Ia. For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 215 PPP estimates of GNP Poverty Life GNP per capita per capita" % of people expectancy Population Surface area Avg. ann. living on less than at birth Adult (millionsl (thousands Dollars growth l%l US100 Current int'l $ $1 a day (PPP) lyearsl illiteracy (%l mid-1995 of sq. kml 1995 1985-95 1987 1995 1995 1981-95 1995 1995 73 Tunisia 9.0 164 1,820 1.9 18.3 18.5 5,000 3.9 69 33 74 Lithuania 3.7 65 1,900 -11.7 25.2 15.3 4,120 2.1 69 75 Colombia 36.8 1,139 1,910 2.6 20.7 22.7 6,130 7.4 70 9 76 Namibia 1.5 824 2,000 2.9 15.8 15.4 4,150' 59 77 Belarus 10.3 208 2,070 -5.2 26.3 15.6 4,220 70 . 78 Russian Federationf 148.2 17,075 2,240 -5.1 30.9 16.6 4,480 1.1 65 79 Latvia1 2.5 65 2,270 -6.6 24.5 12.5 3,370 .. 69 80 Peru 23.8 1,285 2,310 -1.6 17.9 14.0 3,770 49.4 66 II 81 Costa Rica 3.4 51 2,610 2.8 19.8 21.7 5,850 18.9 77 5 82 Lebanon 4.0 10 2,660 . . S 68 8 83 Thailand 58.2 513 2,740 8.4 16.2 28.0 7,540 0.1 69 6 84 Panama 2.6 76 2,750 -0.4 26.1 22.2 5,980 25.6 73 9 85 Turkey 61.1 779 2,780 2.2 20.4 20.7 5,580 . . 67 18 86 Poland 38.6 313 2,790 1.2 21.5 20.0 5,400 6.8 70 87 Estoniat 1.5 45 2,860 -4.3 25.5 15,6 4,220 6.0 70 88 Slovak Republic 5.4 49 2,950 -2.8 17.6 13.4 3,610 12.8 72 89 Botswana 1.5 582 3,020 6.1 15.3 20.7 5,580 34.7 68 30 90 Venezuela 21.7 912 3,020 0.5 33.0 29.3 7,900 11.8 71 9 Upper middle income 4383 t 20 514 t 4260w 0 2w 69 w 14w 91 SoutliAfrica 41.5 1,221 3,160 -1.1 22.4 18.6 5,030' 23.7 64 18 92 Croatia 4.8 57 3,250 . . . . 74 93 Mexico 91.8 1,958 3,320 0.1 27.8 23.7 6,400 14.9 72 10 94 Mauritius 1.1 2 3,380 5.4 39.0 49.0 13,210 71 17 95 Gabon 1.1 268 3,490 -8.2 . . . . 55 37 96 Brazil 159.2 8,512 3,640 -0.8 24.2 20.0 5,400 28.7 67 17 97 Trinidad and Tobago 1.3 5 3,770 -1.7 38.1 31.9 8,610' . . 72 2 98 Czech Republic 10.3 79 3,870 -1.8 44.9 36.2 9,770 3.1 73 99 Malaysia 20.1 330 3,890 5.7 22.9 33.4 9,020 5.6 71 17 100 Hungary 10.2 93 4,120 -1.0 28.9 23.8 6,410 0.7 70 101 Chile 14.2 757 4,160 6.1 24.6 35.3 9,520 15.0 72 5 102 Oman 2.2 212 4,820 0.3 33.2 30.2 8,140' 70 103 Uruguay 3.2 177 5,170 3.1 23.6 24.6 6,630 73 3 104 Saudi Arabia 19.0 2,150 7,040 -1.9 43.0 . . . . . . 70 37 105 Argentina 34.7 2,767 8,030 1.8 31.6 30.8 8,310 73 4 106 Slovenia 2.0 20 8,200 . . . 74 107 Greece 10.5 132 8,210 1.3 44.2 43.4 11,710 .. 78 Low- and middle-income 4,770.8 101,444 1,090w 0.4w 65w 30w Sub-Saharan Africa 583.3 24,271 490 w -1.1w 52w 43w East Asia and Pacific 1,706.4 16,249 800 w 7.2 w 68w 17w South Asia 1,243.0 5,133 350w 2.9w 6lw 51w Europe and Central Asia 487.6 24,355 2,220w -3.5w 68w Middle East and N. Africa 272.4 11,021 1,780w -0.3 w 66w 39w Latin America and Caribbean 477.9 20,414 3,320w 0.3 w 69w 13w High-income economies 902.2 32,039 24,930 w 1.9w 77w 108 Korea, Rep. 44.9 99 9,700 7.7 27.3 42.4 11,450 .. 72 h 109 Portugal 9.9 92 9,740 3.6 41.6 47.0 12,670 . 75 110 Spain 39.2 505 13,580 2.6 50.5 53.8 14,520 77 111 New Zealand 3.6 271 14,340 0.8 63.3 60.6 16,360 . 76 h 112 Ireland 3.6 70 14,710 5.2 44.2 58.1 15,680 . . 77 h 113 1' Israel 5.5 21 15,920 2.5 56.3 61.1 16,490 77 114 tKuwait 1.7 18 17,390 1.1 86.3 88.2 23,790' . 76 21 115 t United Arab Emirates 2.5 84 17,400 -2.8 84.4 61.1 16,470 .. 75 21 116 United Kingdom 58.5 245 18,700 1.4 72.0 71.4 19,260 .. 77 h 117 Australia 18.1 7,713 18,720 1.4 70.1 70.2 18,940 77 h 118 Italy 57.2 301 19,020 1.8 72.5 73.7 19,870 78 h 119 Canada 29.6 9,976 19,380 0.4 84.6 78.3 21,130 . . 78 h 120 Finland 5.1 338 20,580 -0.2 72.9 65.8 17,760 76 h 121 tHong Kong 6.2 1 22,990t 4.8 70.7 85.1 22,9509 79 8 122 Sweden 8.8 450 23,750 -0.1 77.7 68.7 18,540 79 h 123 Netherlands 15.5 37 24,000 1.9 70.5 73.9 19,950 . . 78 h 124 Belgium 10.1 31 24,710 2.2 76.3 80.3 21,660 .. 77 h 125 France 58.1 552 24,990 1.5 77.6 78.0 21,030 78 h 126 1 Singapore 3.0 1 26,730 6.2 56.1 84.4 22,770' 76 9 127 Austria 8.1 84 26,890 1.9 75.0 78.8 21,250 .. 77 h 128 United States 263.1 9,364 26,980 1.3 100.0 100.0 26,980 . . 77 h 129 Germany 81.9 357 27,510 . . . . 74.4 20,070 . . 76 h 130 Denmark 5.2 43 29,890 1.5 78.7 78.7 21,230 75 h 131 Norway 4.4 324 31,250 1.7 78.6 81.3 21,940 78 h 132 Japan 125.2 378 39,640 2.9 75.3 82.0 22,110 80 h 133 Switzerland 7.0 41 40,630 0.2 105.4 95.9 25,860 78 h World 56730t 133 483t 4880w 08w tEconornies classified by the United Nations or otherwise regarded by their authorities as developing, a. Atlas method; see the technical notes. Is. Purcha.stng power parity; see the technical notes. c. The estimate is based on regression; others are extrapolated from the latest International Comparison Programme benchmark estimates, d. In all tables, GDP and GNP cover mainland Tanzania. e. GDP growth rates were revised after the statistics for this publication were finalteed. f. Estimates Lot economies of the former Soviet Union are preliminary; their classification will be kept under review. g. Data refer to GDP. h. According to UNESCO, illiteracy ts less than 5 percent. 215 WORLD DEVELOPMENT REPORT 1997 Table 2. Macroeconomic indicators Money and quasi money Nominal interest rates f Average Current Central Gross Net present (average annual international gov't. curr. Avg. ann. Average annual account value of deflcitIsurplus nominal outstanding Deposit Lending inflation 1%) balance reserves (months external debt 1% of GOP) growth 1%) as a % of GDP rate rate IGDP deflator) 1% of GOP) of import coy.) 1% of GNP) 1980 1995 1985-95 1980 1995 1980 1995 1980 1995 1985-95 1980 1995 1980 1995 1995 Low-income economies Excluding China and India 1 Mozambique . . 52.2 -18.1 . . S 333 2 Ethiopia _j,7b 21.6' 42.1 11.5 . 15.1 -4.1" 19 40h 7.0 61 3 Tanzania -1.3 40.7 30.7 4.0 24.6 11.5 42.8 32.3 -7.1 -17.5 0.2 1.5 148 4 Burundi 1.9 13.5 19.9 2.5 12.0 15.3 6.1 .. -0.6 . . 8.7 50 5 Malawi 1.1 . 23.6 18.0 15.2 7.9 37.3 16.7 47.3 22.1 -21.0 -35.0 1.4 1.5 65 6 Chad . -0.6 20.0 13.7 5.5 5.5 11.0 16.0 3.1 1.2 -4.1 1.7 4.2 38 7 Rwanda 3.5 ., 7.3 13.6 15.5 6.3 5.0 13.5 15.0 10.8 -4.2 -65 6.7 4.7 40 8 SierraLeone -4.4 -3.3 51.4 19.4 9.8 9.2 7.0 11.0 28.8 61.6 -14.2 -10.1 0.7 2.7 137 9 Nepal . . . 21.9 33.7 4.0 . 14.0 . 11.6 -4.8 -8.9 8.9 4.9 26 10 Niger 5.0 . 1.1 13.3 14.2 6.2 . 14.5 . 1.3 -10.9 -8.2 1.6 2.6 53 11 Burkina Faso 2.0 . 9.3 13.8 22.3 6.2 . 14.5 . 2.6 -2.9 0.8 1.5 7.1 27 12 Madagascar . -2.8 24.2 18.2 17.9 . . 18.4 -13.8 -8.6 0.1 1.1 98 13 Bangladesh . . 14.1 18.4 35.6 8.3 6.0 11.3 14.0 6.4 -6.5 -3.5 1.5 4.2 32 14 Uganda -2.2 . 69.0 12.7 10.1 6.8 7.6 10.8 20.2 65.7 -6.6 -7.6 0.1 3.8 43 15 Vietnam . . . 88.3 . . -9.9 0.0 138 16 Guinea-Bjssau 13.8 26.5 . 32.9 62.5 -48.0 -16.1 . . 2.5 235 17 Haiti -3.2 16.9 24.0 42.9 10.0 14.7 -6.9 -3.3 0.6 1.6 25 18 Mali -1.4 6.5 17.9 20.2 6.2 14.5 . 4.6 -8.0 -8.9 0.6 4.0 73 19 Nigeria . . 23.8 25.1 5.3 13.5 8.4 20.2 33.0 5.6 -1.9 5.8 1.5 132 20 Yemen, Rep. -13.4 47.0 .. 3.0 .. 2.5 128 21 Cambodia . 7.7 8.7 18.7 .. -6.7 .. 0.8 52 22 Kenya 2.5 -0.3 21.3 29.8 37.6 5.8 13.6 10.6 28.8 13.0 -12.1 -4.4 2.1 1.2 72 23 Mongolia . 8.6 25.7 60.1 114.9 51.6 -34.7 4.5 .. 3.4 39 24 Togo 1.8 . 0.5 29.0 28.9 6.2 14.5 -8.4 -5.8 1.4 3.6 75 25 Gambia, The 6.7 7.4 14.4 21.1 23.1 5.0 12.5 15.0 25.0 9.0 -37.2 -2.1 0.4 5.3 59 26 Central African Republic -2.1 . 6.1 18.9 20.9 5.5 5.5 10.5 16.0 3.8 -5.4 -2.8 2.2 9.0 52 27 India 0.0 -1.6 16.8 34.7 46.0 . 16.5 163 9.8 -1.7 -1.8 8.0 5.2 23 28 LaoPDR 12.7 14.0 25.7 22.6 .. 12.7 . . 1.8 43 29 Benin 10.1 17.1 25.1 6.2 . 14.5 2.5 2.4 0.4 3.2 46 30 Nicaragua -1.5 3.1 836.2 24.0 30.0 . 11.1 . 19.9 961.6 -19.2 -36.9 0.9 1.2 520 31 Ghana -2.9 -0.9 38.4 16.2 15.4 11.5 28.7 19.0 . 28.6 0.7 -6.5 3.1 4.3 61 32 Zambia -8.1 3.4 75.1 28.4 12.6 7.0 . 9.5 113.3 91.5 -13.3 . 1.2 . . 139 33 Angola 169.5 .. -18.1 . . 260 34 Georgia 310.0 .. .. 44 35 Pakistan 1.8 0.3 15.6 38.7 40.9 . . 9.2 -4.8 -3.8 3.1 2.1 38 36 Mauritania . 7.5 20.5 19.3 5.5 . 12.0 . 6.9 -18.8 -2.6 3.6 1.7 166 37 Azerbaijan . . 9.5 . . . 162.5 . . . -10.9 . . 8 38 Zimbabwe -9.0 22.1 29.5 26.0 3.5 25.9 17.5 34.7 20.9 -2.8 -73 2.7 . . 65 39 Guinea . . 8.8 . 17.5 . 21.5 . .. _5.3 . . 1.0 59 40 Honduras . 18.6 21.1 25.2 10.6 12.0 16.5 27.0 14.3 -12.3 -5.1 1.5 1.5 101 41 Senegal 1.8 . 3.9 26.6 20.0 6.2 . 14.5 . 3.7 -12.8 0.1 0.2 1.9 54 42 China . 28.3 33.2 92.4 5.4 11.0 5.0 12.1 9.3 2.8 0.2 10.5 6.3 16 43 Cameroon 5.8 0.2 -2.9 20.6 16.2 7.5 5.5 13.0 16.0 2.0 -10.1 -2.2 1.1 0.1 97 44 Côted'Ivoire 3.8 . 2.2 26.7 26.2 6.2 . 14.5 . 4.0 -18.0 -2.7 0.1 1.5 185 45 Albania . -'4.7 . . 47.5 . 15.3 . 19.7 27.3 . -0.5 . . . 3.7 32 46 Congo 1.1 14.8 14.7 6.5 5.5 11.0 16.0 2.2 -9.8 -26.4 0.9 0.4 325 47 Kyrgyz Republic .. -9.5 .. . . 15 48 Sri Lanka -4.5 -1.5 17.0 28.4 31.8 14.5 16.1 19.0 14.7 11.8 -16.3 -4.7 1.5 4.2 44 49 Armenia . 183.1 . -13.6 . . . 14 Middle income economies Lower middle income 50 Lesotho 1.3 18.0 13.9 39.7 28.9 9.6 13.3 11.0 16.4 13.4 15.3 12.2 1.3 5.4 26 51 Egypt,Arab Rep 9.0 65 19.0 52.2 96.8 8.3 10.9 13.3 16.5 15.7 -1.9 -2.0 3.1 11.8 56 52 Bolivia . -1.2 43.1 16.2 44.9 18.0 18.9 28.0 51.0 18.4 -0.2 -4.0 6.0 6.7 67 53 Macedonia, FYR . S S . . 1.7 57 54 Moldova . 11,5 . . . . . -2.7 . . 2.9 16 55 Uzbekistan . . S .. . 0.0 . . . . . 7 56 Indonesia 9.6 9.9 13.2 6.0 .. 17.1 8.8 -0.6 -3.5 2.9 3.0 54 57 Philippines 4.1 2.0 20.7 22.0 45.4 12.3 8.4 14.0 14.7 9.8 -5.9 -2.7 4.6 2.6 49 58 Morocco 0.5 . 13.1 38.5 65.2 4.9 7.0 10.0 4.8 -7.5 -4.7 1.7 3.6 62 59 Syrian Arab Republic -3.5 7.1 40.9 63.2 5.0 . 16.0 1.9 2.6 2.2 . . 118 60 Papua New Guinea -6.2 -4.1 . 32.9 30.2 6.9 5.1 11.2 9.2 4.5 -11.3 13.8 3.5 1.3 45 61 Bulgaria . -4.1 45.9 4.8 2.7 .. . . 87 62 Kazakstan S .. .. -2.4 ., 2.8 22 63 Guatemala 2.1 0.8 22.7 20.5 23.5 9.0 7.9 11.0 21.2 18.6 -2.1 -3.9 4.3 2.4 19 64 Ecuador 0.9 3.3 55.3 21.2 26.6 43.3 9.0 55.7 45.5 -5.5 -4.6 4.1 3.4 76 65 Dominican Republic 2.9 8.3 31.9 17.8 24.4 26.4 -10.9 -1.1 1.5 0.7 33 66 Romania 15.5 2.2 51.3 33.4 19.9 . 68.7 . -3.8 . 2.1 2.9 18 67 Jamaica 33.4 32.8 44.0 9.5 23.2 15.6 43.6 28.3 -5.1 -5.6 0.8 2.0 123 68 Jordan . 2.5 10.2 . 104.5 . 3.3 9.0 . . . . -9.0 6.3 5.3 108 69 Algeria . . 14.8 53.3 38.8 . 22.9 0.6 -5.6 5.8 5.0 64 70 El Salvador -0.4 0.8 21.2 28.0 36,1 14.4 19.1 14.9 0.9 -0.7 3.6 3.2 22 71 Ukraine . 0.0 70.3 122.7 .. .. -1.4 .. 0.7 10 72 Paraguay 3.3 3.0 36.5 19.8 26.4 . 21.2 . . 31.0 24.9 -13.5 -19.0 6.7 27 Note For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELDPMENTINDICATORS 217 Money and quasi money Nominal interest rates of banks Gross Net present Central Average Current gov't. curr. (average annual value of Avg. ann. Average annual account international deficitJsurplus nominal outstanding Deposit Lending inflation 1%) balance reserves (months external debt 1% of GDPI growth 1%) as a % of GDP rate rate IGDP deflator) (% of GDP) of import coy.) 1% of GNP) 1980 1995 1985-95 1980 1995 1980 1995 1980 1995 1985-95 1980 1995 1980 1995 1995 73 Tunisia 9.2 9.8 37.6 44.3 2.5 . 7.3 6.0 -4.0 -4.1 2.0 2.1 52 74 Lithuania 0.7 . . . . 22.6 . . 8.4 . 27.1 . . . . -8.7 . 2.5 9 75 Colombia 1.5 2.8 31.6 17.1 19.5 . . 32.3 . 42.7 25.2 -0.6 -5.4 12.5 5.0 27 76 Namibia . 0.7 . . . . 38.7 . . 10.8 . . 18.5 10.4 . . 1.6 . . 1.3 77 Belarus . . . . . 10.5 . . 100.8 . 175.0 . . . --1.2 . 0.9 6 78 Russian Federation . -5.6 . . . . 11.6 102.0 . . 319.5 148.9 . . 2.8 . . 2.5 35 79 Latvia . -2.5 . . . 25.0 14.8 . . 34.6 72.5 . . -0.4 . 3.2 7 80 Peru 2.0 0.5 388.5 16.5 17.2 . . 16.0 .36.6 398.5 -0.5 -7.4 6.6 8.6 52 81 Costa Rica -3.5 -0.4 23.6 38.8 32.2 18.3 23.9 25.0 36.7 18.4 -13.7 -1.5 1.2 3.0 40 82 Lebanon . -11.2 63.4 . . 117.6 12.9 16.3 16.8 24.7 . . . . -45.7 . . 14.6 25 83 Thailand -0.1 8.1 19.7 34.5 73.8 12.0 11.6 18.0 14.4 5.0 -6.4 -8.1 3.3 5.0 35 84 Panania 0.4 2,8 11.3 32.8 67.9 . . 7.2 . . 11.1 1.7 -9.2 -2.2 0.2 0.9 98 85 Turkey 2.6 --2.4 73.9 14.2 24.8 8.0 76.1 64.6 -5.0 -1.4 4.3 3.7 43 86 Poland . -0.8 87.2 57.0 31.8 , . 26.8 8.0 33.5 91.8 -6.0 -3.6 0.3 4.9 31 87 Estonia . . . . . 22.5 . . 8.7 . . 16.0 77.2 . . -4.6 2.2 6 88 Slovak Republic . . . . . . 62.9 . . 9.0 . . 15.6 10.6 . . 3.7 . . 4,4 31 89 Botswana 10.8 21.1 19.2 28.2 25.9 5.0 10.0 8.5 14.2 11.5 -15.6 7.9 4.3 22.5 13 90 Venezuela 7.4 2.6 36.3 28.9 23.1 . . 24.7 . . 32.2 37.6 6.8 3.0 9.4 6.3 47 Upper nuddle income 91 SouthAfrica 4.4 -4.2 14.5 50.1 51.7 5.5 13.5 9.5 17.9 13.9 4.5 -2.6 3.6 1.4 92 Croatia . 2.1 . . . . 22.1 . . 5.5 . . 20.2 . . . . -9.5 , , 2.5 18 93 Mexico 3.8 2.4 46.1 25.2 30.7 20.6 39.2 28.1 . 36.7 -5.3 -0.3 1.4 2.1 67 94 Mauritius -1.9 1.9 20.3 40.0 73.2 9,3 12.2 12.2 20.8 8.8 -10.3 -0.6 1.9 4.2 95 Gabon 1.2 15.3 14.6 7.5 5.5 12.5 16.0 5.0 9.0 8.1 0.7 0.8 89 96 Brazil 4.1 -13.3 995.5 11.1 26.1 115.0 52.2 875.3 --5.5 -2.6 2.3 7.9 23 97 Trinidad and Tobago 24.1 . 4.3 27.1 40.1 66 69 10.0 15.2 6.8 5.7 5.5 11.4 1.8 52 98 Czech Republic . 4.0 . . . . 81.0 . . 7.0 . . 12.8 12.2 . . -3.1 . . 6.5 36 99 Malaysia 7.1 7.3 15.5 46.1 85.0 6.2 5.9 7.8 7.6 3.3 -1.1 -5.9 4.6 3.2 39 100 Hungary 4.7 . , . . 43.0 3.0 26.1 . . 32.6 19.9 -2.3 -5.8 . . 6.7 72 101 Chile 6.7 5.3 25.5 21.0 33.9 37.7 13.7 47.1 18.2 17.9 -7.1 0.2 5.9 8.8 41 102 Oman 7.9 -4.3 5.7 13.8 31.3 . . 6.5 . . 9.4 -0.2 15.8 -8.1 3.2 2.6 28 103 Uruguay 2.1 0.6 71.3 31.2 33.4 50.3 38.2 66.6 99.1 70.7 -7.0 -2.0 12.5 5.3 31 104 SaudiArabia . . 5.1 13.8 50.4 . . . 2.8 26.5 -6.5 5.0 2.7 105 Argentina -2.6 257.9 19.0 18.8 79.6 11.9 . . 17.8 255.6 -6.3 -1.4 7.0 6.2 31 106 Slovenia , . . . . 32.5 . . 15.3 . . 24.8 . . . . -0.2 . , 2.1 18 107 Greece -0.4 -14.4 . , 50.5 53.0 14.5 15.8 21.3 23.1 15.4 -5.5 -3.2 3.7 7.0 Low- and middle-income Sub-Saharan Africa East Asia and Pacific South Asia Europe and Central Asia Middle East and N. Africa Latin America and Caribbean High-income economies 108 Korea, Rep. 2.8 6.0 18.5 29.0 40.9 19.5 8.8 18.0 9.0 6.7 -8.3 -1.8 1.3 2.5 109 Portugal -2.7 -3.4 15.5 70.1 78.1 19.0 8.4 18.8 13.8 11.2 --3.7 -0.2 15.2 6.2 110 Spain 0.4 -4.9 11,2 75.4 78.6 13.1 7.7 16.9 10.0 6.3 -2.6 0.2 6.0 3.2 111 New Zealand -1.6 3.8 19.2 26.4 77.6 11.0 8.5 12.6 12.2 3.9 -4.3 -6.6 0.6 2.4 112 Ireland -5.7 -1.8 11.4 43.8 50.1 12.0 0.4 16.0 6.6 2.5 --10.6 2.3 2.7 2.0 113 tlsrael -16.9 -2.4 22.1 14.8 67.3 .. 14.1 176.9 20.2 17.1 -3.9 -6.0 3.6 2.5 14 5 Kuwait 70.5 . . 3.2 33.1 77.7 9.2 6.5 9.2 8.4 -0.5 53.4 15.8 6.2 4.1 115 lUnited Arab Emirates -11.0 -8.8 4.6 19.0 54.0 9.5 . 12.1 . . . , . , . , . , 116 United Kingdom -1.2 -3.4 . 14.1 4.1 16,2 6.7 5.1 1.3 -0.4 2.0 1.3 117 Australia 0.6 -1.9 11.9 36.4 61.3 8.6 , 10.6 . 3.7 -2.8 -5.5 2.5 1.9 118 Italy -6.4 -8.4 . 70.9 62.5 12.7 6.4 19.0 12.5 6.0 -2.3 2.4 6.4 2.4 119 Canada -2.4 8.9 45.1 59.3 12.9 7.1 14.3 8.6 2.9 -0.6 -1.5 2.3 0.8 120 Finland 2.0 -9.0 6.6 39.8 56.7 9.0 3.2 9.8 7.7 3.8 -2.7 4.5 1.6 2.9 121 tHong Kong .. .. 60.7 .. 8.7 -4.4 .. 122 Sweden -2.5 -5.9 . . 11.3 6.2 15.2 11.1 5.5 -3.4 2.0 2.0 3.1 123 Netherlands 1.2 -2.3 5.6 67.3 82.0 6.0 4.4 13.5 7.2 1.7 -0.5 4.1 4.4 2.5 124 Belgium -2.9 -2.6 13.0 45.0 80.2 7.7 4.0 18.0 8.4 3.2 . . . . . , 125 France 2.2 -3.8 3.7 71.6 64.4 7.3 4.5 12.5 8.1 2.8 -0.6 1.1 5.2 1.5 126 tSingapore 9.8 15.0 14.3 57.7 82.6 9.4 3.5 11.7 6.4 3.9 -13.3 18.0 3.0 5.7 127 Austria 0.6 -1.5 7.0 72.6 89.5 5.0 2.2 3.2 -5.0 -2.2 6.4 2.5 128 United States -0.5 -1.4 3.9 60.4 59.4 13.1 5.9 15.3 8.8 3.2 0.1 -2.1 6.2 2.0 129 Germany . . -0.2 8,1 . 62.0 7.9 3.9 12.0 10.9 . . . . -0.9 5.3 2.1 1,30 Denmark -1.2 -1.2 4.4 42.6 57.8 10.8 3.9 17.2 10.3 2.8 -3.3 0,8 1.5 1.5 131 Norway 4.0 1.0 5.9 47.1 55.6 5.0 5.0 12.6 7.8 3.0 1.7 3.0 3.0 5.5 132 Japan --3.2 - . 5.9 83.4 112.7 5.5 0.7 8.3 3.4 1,4 -1.0 2,2 2.8 4.1 133 Switzerland 0.8 s4 4.6 107.4 126.3 8.8 1.3 5.6 5.5 3.4 -0.2 7.2 13.3 77 World i Rckrs to current budget balance excluding grants. ft Includes Eritrea. c. Certificate of' deposh ran 218 WORLD DEVELOPMENT REPORT 1997 Table 3. External economic indicators Net barter Aggregate net Net private terms of trade Trade resource flows capital flows Aid (1987= 100) (% of GDP) (% of GNP) (millions $1 (% of GNP) 1985 1995 1980 1995 1980 1995 1980 1995 1980 1994 Low-income economies 111 m 91 m 3.4w 5.9w 7,368 t 53,446 t 1.9w 4.3w Excluding China and India 112 m 91 m 6.3w 7.3w 4,769 t 5,517 t 4.1w 12.6w 1 Mozambique 113 124 61 102 3.9 76.8 0 67 8.4 101.0 2 Ethiopia 119 74 27 39 8.3a 12.0 26 -42 4.7 22.7 3 Tanzania 126 83 37 96 164 18.1 100 137 12.4 29.9 4 Burundi 133 52 32 43 8.1 24.5 -3 1 12.8 31.6 5 Malawi 99 87 64 69 15.7 22.4 30 -14 12.6 38.0 6 Chad 99 103 65 46 3.3 16.2 0 7 4.9 23.9 7 Rwanda 136 75 41 32 9.3 53.8 14 1 13.4 95.9 8 Sierra Leone 109 89 62 40 5.2 22.7 -7 -28 8.0 36.0 9 Nepal 98 85 30 60 6.5 6.4 0 -2 8.3 10.9 10 Niger 91 101 63 30 12.9 8.6 199 -23 6.8 25.0 11 BurkinaFaso 103 103 43 45 8.4 13.9 4 0 12.5 23.7 12 Madagascar 124 82 43 54 8.7 8.4 131 4 5.8 10.2 13 Bangladesh 126 94 24 37 13.0 3.5 11 10 9.9 6.8 14 Uganda 149 58 45 33 8.9 10.1 44 112 9.0 19.2 15 Vietnam .. 83 .. 4.4 0 1,487 5.9 16 Guinea-Bissau 91 92 52 48 108.0 26.4 18 1 56.6 74.3 17 Haiti 89 52 52 17 5.2 28.0 20 2 7.3 37.3 18 Mali 100 103 51 38 12.1 13.9 10 1 16.6 24.5 19 Nigeria 167 86 48 81 0.9 0.8 694 453 0.0 0.6 20 Yemen, Rep. 131 84 88 3.6 97 -2 .. 4.6 21 22 Cambodia Kenya 124 . 98 .. .. 67 36 72 .. 8.8 15.1 5.6 301 0 164 -42 .. 5.6 14.3 9.7 23 Mongolia . . .. 58 .. 0.0 16.0 0 -4 0.0 27.6 24 Togo 139 90 107 65 12.7 11.3 83 0 8.3 13.8 25 Gambia, The 137 111 119 103 35.0 9.0 21 10 24.4 19.8 26 Central African Republic 109 91 69 46 10.7 12.5 4 3 13.9 19.4 27 India 92 150 17 27 1.4 1.0 868 3,592 1.3 0.8 28 Lao PDR .. .. .. 53 . . 16.1 0 88 .. 14.2 29 Benin 111 110 66 64 7.2 10.7 4 1 6.4 17.4 30 Nicaragua 111 95 68 76 13.8 33.3 -26 -7 10.9 46.1 31 Ghana 93 64 18 59 4.1 17.4 -26 525 4.3 8.5 32 Zambia 89 85 87 71 14.6 12.5 175 30 8.9 20.7 33 Angola 153 86 132 21.3 38 523 .. 11.0 34 Georgia .. .. .. 46 .. 66.2 0 0 . . 7.5 35 Pakistan 112 114 37 36 5.4 4.1 230 1,443 5.1 3.1 36 Mauritania 110 106 104 104 29.4 17.8 27 3 26.2 27.7 37 Azerbaijan . . . . .. 66 . . 7.9 0 110 . . 3.8 38 Zimbabwe 100 84 64 74 4.2 7.2 23 99 3.1 10.2 39 Guinea 120 91 .. 46 .. 9.1 80 20 .. 11.0 40 Honduras 118 77 80 80 11.7 6.5 137 65 4.2 9.5 41 Senegal 107 107 72 69 9.0 8.4 18 -24 9.0 17.2 42 China 109 105 13 40 1.0 7.9 1,732 44,339 0.0 0.6 43 Cameroon 113 79 54 46 10.9 2.8 409 49 4.4 10.0 44 Côte d'Ivoire 109 81 76 76 11.8 7.9 936 36 2.2 24.8 45 Albania .. 52 9.2 0 70 .. 9.1 46 Congo 150 93 120 128 35.5 1.1 440 -49 6.0 24.9 47 Kyrgyz Republic . . . . . . 58 . . 6.1 0 15 . . 5.5 48 Sri Lanka 106 88 87 83 10.6 6.0 129 140 9.8 5.1 49 Armenia .. 85 .. 7.2 0 8 .. 9.8 Middle-income economies 94m 2.8w 4.1 w 44,334 t 130,742 t 0.6w 1.0w Lower-middle-income .. .. .. .. .. 1.3w 1.4w 50 Lesotho . . . . 143 138 10.5 8.4 7 32 14.9 8.9 51 Egypr,ArabRep. 147 95 73 54 14.2 1.9 1,133 294 6.5 6.4 52 Bolivia 130 67 38 47 14.1 12.9 203 191 5.9 10.9 53 Macedonia, FYR .. 86 3.0 0 0 . 54 Moldova 78 4.9 0 79 1.5 55 Uzbekisran . . .. . . 125 . . 2.3 0 235 . . 0.1 56 Indonesia 145 79 53 53 2.5 6.8 987 11,648 1.3 1.0 57 Philippines 99 114 52 80 3.9 5.2 840 4,605 0.9 1.6 58 Morocco 99 90 45 62 8.6 2.9 731 572 4.9 2.2 59 Syrian Arab Republic 138 78 54 . . 19.7 1.9 42 43 13.0 5.3 60 Papua New Guinea 94 90 97 106 16.8 10.8 106 578 13.1 6.4 61 Bulgaria 95 106 66 94 1.7 3.7 339 489 0.0 1.6 62 Kazakstan .. .. .. 69 .. 4.7 0 500 . . 0.2 63 Guatemala 114 93 47 47 2.8 1.7 91 85 0.9 1.7 64 Ecuador 143 71 51 56 7.4 4.8 594 561 0.4 1.4 65 Dominican Republic 115 123 48 55 7.1 1.5 150 237 1.9 0.7 66 Romania 66 lii 75 60 .. 4.4 1,360 687 . . 0.5 67 Jamaica 89 105 102 145 12.3 5.5 9 188 5.1 2.9 68 Jordan 127 128 . . 121 35.0 8.4 28 -143 . . 6.5 69 Algeria 173 83 65 57 3.1 2.6 897 129 0.4 1.0 70 El Salvador 122 89 67 55 3.2 1.6 -17 8 2.8 3.9 71 Ukraine .. . . .. .. .. 0.9 0 247 .. 0.3 72 Paraguay 110 101 44 82 3.6 4.3 121 174 0.7 1.3 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 219 Net barter Aggregate net Net private terms of trade Trade resource flows capital flows Aid (1987r 100) (% of GOP) (% of GNP) (millions $1 1% of GNPI 1985 1995 1980 1995 1980 1995 1980 1995 1980 1994 73 Tunisia 123 91 86 93 7.2 5.7 337 751 2.7 0.7 74 Lithuania . .. .. 108 .. 3.6 0 194 . 1.1 75 Colombia 124 80 32 35 2.9 2.7 688 3,741 0.3 0.2 76 Namibia .. 142 110 .. .. .. 0.0 4.7 77 Belarus . . . . 1.4 0 103 . . 0.5 78 Russian Federation .. .. 44 .. 0.5 2,817 1,116 .. 0.6 79 Larvia . . .. .. 91 .. 2.2 0 224 .. 0.9 80 Peru 111 83 42 30 1.8 6.4 -67 3,532 1.0 0.8 81 CostaRica 111 92 63 81 9.2 1.6 248 384 1.4 0.9 82 Lebanon 105 95 . . 70 11.3 70 1,153 . . 2.5 83 Thailand 103 100 54 90 6.5 6.1 1,465 9,143 1.3 0.4 84 Panama 104 86 . . 79 4.1 10.1 65 228 1.3 0.6 85 Turkey 82 109 17 45 3.0 0.8 660 2,000 1.4 0.1 86 Poland 95 109 59 53 5.5 5.6 2,265 5,058 0.0 2.0 87 Estonia 160 6.5 0 207 1.1 88 Slovak Republic 98 86 .. 124 .. 4.3 0 653 .. 0.6 89 Borswana 97 152 116 101 20.3 2.3 115 64 11.8 2.2 90 Venezuela 166 82 51 49 2.6 0.0 1,825 848 0.0 0.1 Upper-middle-income 110 m 95m . .. .. 0.1w 0.1w 91 SouthAfrica 101 III 64 44 . . . . . . . . 0.2 92 Croatia . . . . . . 93 . . 1.9 0 346 . 93 Mexico 145 92 24 48 4.8 8.7 8,181 13,068 0.0 0.1 94 Mauririus 77 103 113 120 8.4 7.9 48 304 3.0 0.4 95 Gabon 154 90 96 101 -1.9 9.6 -93 -125 1.4 5.6 96 Brazil 101 101 20 15 2.9 2.6 5,656 19,097 0.0 0.1 97 Trinidad and Tobago 138 86 89 68 6.3 6.4 258 271 0.1 0.5 98 Czech Republic 98 86 . . 108 0.0 9.2 0 5,596 0.0 0.4 99 Malaysia 114 92 113 194 8.7 14.7 1,913 11,924 0.6 0.1 100 Hungary 103 97 80 67 3.3 17.4 596 7,841 0.0 0.5 101 Chile 91 94 50 54 8.7 4.7 2,447 4,230 0.0 0.3 102 Oman 182 77 100 89 3.3 1.3 33 126 3.1 1.0 103 Uruguay 91 112 36 41 5.3 1.8 479 217 0.1 0.5 104 SaudiArabia 175 92 101 70 . . . . .. . . 0.0 0.0 105 Argentina 123 120 12 16 4.6 4.1 3,475 7,204 0.0 0.1 106 Slovenia . . . . . . 113 . 4.0 0 838 . 107 Greece 96 111 47 57 . . . . 0.1 0.0 Low- and middle-income 111 m 93 m 2.9w 4.5w 51,702 t 184,188 t 1.0w 3.3w Sub-Saharan Africa 110 m 91 m 5.5w 8.3w 7,906 t 9,128 t 3.4w 16.3w East Asia and Pacific .. .. 2.9w 7.8w 7,135 r 84,137 r 0.7w 1.1 w South Asia 109 m 94m 3.0w 1.6w 1,238 t 5,191 r 2.4w 1.9w Europe and Central Asia .. .. 1.7w 3.8w 26,164 r 30,059 t 0.4w 0.9w Middle East and N. Africa 147 m 92 m 1.9w 0.4w -1,040 t 1,414 r 1.1w Latin America and Caribbean 111 m 94m 4.2w 4.3w 24,590 t 54,261 t 0.3w 1.7w High-income economies 96 m 97m 108 Korea, Rep. 94 102 74 67 0.2 0.0 109 Portugal 117 92 61 66 110 Spain 82 114 34 47 111 NewZealand 90 108 62 62 112 Ireland 96 90 108 136 . 113 lIsrael 99 109 91 69 4.1 1,6 114 tKuwait 165 88 113 104 0.0 0.0 115 t United Arab Emirates 181 93 112 139 0.0 0.0 116 United Kingdom 104 102 52 57 117 Australia 110 101 34 40 118 Italy 84 107 47 49 . 119 Canada 99 100 55 71 120 Finland 88 95 67 68 121 tHongKong 118 87 181 297 0.0 0.0 122 Sweden 92 102 61 77 123 Netherlands 101 103 103 99 124 Belgium 96b 101b 128 143 125 France 89 106 44 43 . . 126 tSingapore 108 89 423 .. 0.1 0.0 127 Austria 92 87 76 77 128 United States 101 102 21 24 129 Germany 84 96 .. 46 130 Denmark 91 100 66 64 .. 131 Norway 142 95 81 71 . . 132 Japan 73 127 28 17 . . 133 Switzerland 85 60 77 68 World a. Includes Eritrea. b. Includes Luxembourg. c. Data refer to the Federal Republic of Germaisy before unification. 220 WORLD DEVELOPMENT REPORT 1997 Table 4. Population and labor force Population Labor force Total Avg. annual Aged 15-64 TotaP Avg. annual (miJilons) growth rate 1%) Imilhions) (millions) growth rate 1%) Female 1%) Agriculture 1%) Industry 1%) 1980 1995 1980-90 1990-95 1980 1995 1980 1995 1980-90 1990-95 1980 1995 1980 1990 1980 1990 Low-income economies 2,378 t 3,180 t 2.0w 1.7w 1,351 t 1,934 t 1,156t 1,575 t 2.2w 1.7w 40w 41 w 73w 69w 13w 15w Excluding China and India 709 t 1,050 t 2.7w 2.4w 371 t 563 t 317 46? 2.6w 2.5w 40w 41w 72w 67w lOw 12w 1 Mozambique 12 16 1.6 2.6 6 9 7 8 1.2 2.4 49 48 84 83 7 8 2 Ethiopia 38 56 3.1 1.9 19 28 17 25 2.9 2.3 42 41 86 80 2 2 3 Tanzania 19 30 3.2 3.0 9 15 10 15 3.2 2.9 50 49 86 84 4 5 4 Burundi 4 6 2.8 2.6 2 3 2 3 2.6 2.7 50 49 93 92 2 3 5 Malawi 6 10 3.3 2.7 3 5 3 5 3.0 2.5 51 49 88 95 5 5 6 Chad 4 6 2.4 2.5 2 3 2 3 2.1 2.5 43 44 88 81 3 4 7 Rwanda 5 6 3.0 -1.7 3 4 3 4 3.2 2.0 49 49 93 92 3 3 8 Sierra Leone 3 4 2.1 1.0 2 2 1 2 1.8 2.0 36 36 70 67 14 15 9 Nepal 15 21 2.6 2.5 8 12 7 10 2.4 2.4 39 40 95 95 1 0 10 Niger 6 9 3.3 3.3 3 4 3 4 3.0 2.9 45 44 93 91 3 4 II Burkina Faso 7 10 2.6 2.8 4 5 4 5 2.0 2.1 48 47 92 92 3 2 12 Madagascar 9 14 2.9 3.1 5 7 4 6 2.5 3.1 45 45 85 84 6 7 13 Bangladesh 87 120 2.4 1.6 44 64 41 60 2.8 2.1 42 42 74 64 9 16 14 Uganda 13 19 2.4 3.2 6 9 7 9 2.2 2.7 48 48 89 93 4 5 15 Vietnam 54 73 2.1 2.1 28 43 26 37 2.7 1.9 48 49 73 72 13 14 16 Guinea-Bissau 1 1 1.8 2.1 0 1 0 1 1.3 1.8 40 40 86 85 2 2 17 Haiti 5 7 1.9 2.0 3 4 3 3 1.3 1.7 45 43 71 68 8 9 18 Mali 7 10 2.5 2.9 3 5 3 5 2.3 2.7 47 46 93 93 2 2 19 Nigeria 71 111 3.0 2.9 38 58 30 44 2.6 2.8 36 36 55 43 8 7 20 Yemen, Rep. 9 15 3.3 5.0 4 8 2 5 3.7 4.9 33 29 70 58 13 16 21 Cambodia 6 10 2.9 2.8 3 5 3 5 2.8 2.5 56 53 76 74 7 8 22 Kenya 17 27 3.4 2.7 8 14 8 13 3.6 2.7 46 46 83 80 6 7 23 Mongolia 2 2 2.9 2.1 1 1 1 1 3.1 2.9 46 46 40 32 21 22 24 Togo 3 4 3.0 3.0 1 2 1 2 2.6 2.8 39 40 69 66 10 10 25 Gambia, The 1 1 3.6 3.7 0 1 0 1 3.4 3.2 45 45 84 82 7 8 26 Central African Republic 2 3 2.4 2.2 . . 2 1 2 1.7 1.8 48 47 85 80 3 3 27 India 687 929 2.1 1.8 394 562 300 398 1.9 2.0 34 32 70 64 13 16 28 Lao PDR 3 5 2.7 3.0 2 3 2 2 2.3 2.7 45 47 80 78 6 6 29 Benin 3 5 3.1 2.9 2 3 2 2 2.7 2.5 47 48 67 62 7 8 30 Nicaragua 3 4 2.9 3.1 1 2 1 2 2.9 4.0 28 36 39 28 24 26 31 Ghana 11 17 3.3 2.8 6 9 5 8 3.1 2.7 51 51 61 60 13 13 32 Zambia 6 9 3.0 2.9 3 5 2 4 3.1 2.8 45 45 76 75 8 9 33 Angola 7 11 2.7 3.1 4 5 3 5 2.1 2.8 47 46 76 75 8 8 34 Georgia 5 5 0.7 -0.2 3 4 3 3 0.4 -0.1 49 46 32 26 27 31 35 Pakistan 83 130 3.1 2.9 44 70 29 46 2.9 3.3 23 26 62 56 15 20 36 Mauritania 2 2 2.6 2.5 I 1 1 1 2.0 2.7 45 44 72 55 7 10 37 A.zeebaijan 6 8 1.5 1.0 4 5 3 3 1.0 1.7 47 44 35 31 28 29 38 Zimbabwe 7 11 3.3 2.4 3 6 3 5 3.6 2.2 44 44 74 69 12 8 39 Guinea 4 7 2.5 2.7 2 3 2 3 2.1 2.4 47 47 91 87 1 2 40 Honduras 4 6 3.3 3.0 2 3 1 2 3.6 3.8 25 30 56 40 14 19 41 Senegal 6 8 2.9 2.7 3 4 3 4 2.6 2.7 42 42 81 76 6 7 42 China 981 1,200 1.5 1.1 586 811 539 709 2.2 1.1 43 45 76 74 14 15 43 Cameroon 9 13 2.8 2.9 5 7 4 5 2.4 3.1 37 38 73 70 8 9 44 Côte d'lvoire 8 14 3.8 3.1 4 7 3 5 3.1 2.3 32 33 65 60 8 10 45 Albania 3 3 2.1 -0.1 2 2 1 2 2.6 0.8 39 41 57 55 23 23 46 Congo 2 3 3.1 2.9 1 1 1 1 3.1 2.6 43 43 58 48 13 14 47 Kyrgyz Republic 4 5 1.9 0.5 2 3 2 2 1.6 1.2 48 47 34 32 29 26 48 Sri Lanka 15 18 1.4 1.3 9 12 5 8 2.3 2.0 27 35 52 49 18 21 49 Armenia 3 4 1.4 1.2 2 2 1 2 1.6 1.1 48 48 21 17 43 41 Middle-income economies 591t 1.8w 14w 717t 7f'513t t 21w 1.8w 36w 38w 38w 32w 28w 27w Lower-middle-income 153t 1.7w 14w 527t 71t 387t 05't 18w 1.7w 38w 40w 41w 36w 27w 27w 50 Lesorho 2 2.7 2.1 1 1 1 1 2.3 2.3 38 37 41 41 33 28 51 Egypt, Arab Rep. 41 58 2.5 2.0 23 34 14 21 2.5 2.7 26 29 61 43 17 23 52 Bolivia 7 2.0 2.4 3 4 2 3 2.6 2.6 33 37 53 47 18 18 53 Macedonia, FYR 2 2 0.7 0.9 1 1 1 1 1.2 1.3 36 41 34 22 31 41 54 Moldova 4 4 0.9 -0.1 3 3 2 2 0.2 0.2 50 49 43 33 26 30 55 Uzbekistan 16 23 2.5 2.1 9 13 6 9 2.2 2.8 48 46 38 34 25 25 56 Indonesia 148 193 1.8 1.6 83 120 59 89 2.9 2.5 35 40 59 57 12 14 57 Philippines 48 69 2.4 2,2 27 40 19 28 2.7 2.7 35 37 52 45 15 15 58 Morocco 19 27 2.2 2.0 10 16 7 10 2.6 2.6 34 35 56 45 20 25 59 Syrian Arab Republic 9 14 3.3 3.0 4 7 2 4 3.0 3.5 23 26 39 34 28 24 60 Papua New Guinea 3 4 2.2 2.3 2 2 2 2 2.1 2.3 42 42 83 79 6 7 61 Bulgaria 9 8 -0.2 -0.7 6 6 5 4 -0.4 -0.6 45 48 20 14 45 50 2 Kazaksran 15 17 1.2 -0.2 9 10 7 8 1.1 0.5 48 47 24 22 32 31 63 Guatemala 7 11 2.8 2.9 4 6 2 4 2.9 3.5 22 26 54 52 19 17 64 Ecuador 8 11 2.5 2.2 4 7 3 4 3.5 3.2 20 26 40 33 20 19 65 Dominican Republic 6 8 2.2 1.9 3 5 2 3 3.1 2.6 25 29 32 25 24 29 66 Romania 22 23 0.4 -0.4 14 15 11 11 -0.2 0.1 46 44 35 24 41 47 67 Jamaica 2 3 1.2 1.0 1 2 1 1 2.1 1.8 46 46 31 24 16 23 68 Jordan 2 4 3.7 5.7 1 2 1 1 4.9 5.3 15 21 24 21 32 32 69 Algeria 19 28 2.9 2.2 9 16 5 9 3.7 4.1 21 24 36 26 27 31 70 El Salvador 5 6 1.0 2.2 2 3 2 2 1.7 3.4 27 34 43 36 19 21 71 Ukraine 50 52 0.4-0.1 33 34 26 26 -0.1 -0.2 50 49 25 20 39 40 72 Paraguay 3 5 3.0 2.7 2 3 1 2 2.9 2.9 27 29 45 39 20 23 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELDFMENTINDICATORS 221 Population Labor force Total Avg. annual Aged 15-64 Total Avg. annual (millions) growth rate 1%) (millions) (millions) growth rate (%) Female 1%) Agriculture (%) Industry (%) 1980 1995 1980-90 1990-95 1980 1995 1980 1995 1980-90 1990-95 1980 1995 1980 1990 1980 1990 73 Tunisia 6 9 2.5 1.9 3 5 2 3 2.7 3.0 29 30 39 28 30 32 74 Lithuania 3 4 0.9 0.0 2 2 2 2 0.7 -0.2 50 48 28 18 38 40 75 Colombia 28 37 1.9 1.8 16 23 9 16 3.9 2.7 26 37 39 25 20 22 76 Namihia 1 2 2.7 2.7 1 1 0 1 2.3 2.5 40 41 56 49 15 15 77 Belarus 10 10 0.6 0.2 6 7 5 5 0.5 0.2 50 49 26 20 38 40 78 Russian Federation 139 148 0.6 0.0 95 99 76 77 0.2 0.0 49 49 16 14 44 42 79 Latvia 3 3 1).5 -1.2 2 2 1 1 0.2 -1.1 51 50 16 16 42 40 80 Peru 17 24 2.2 2.0 9 14 5 9 3.1 3.1 24 29 40 36 18 18 81 Costa Rica 2 3 2.8 2.3 1 2 1 1 3.8 2.5 21 30 35 26 23 27 82 Lebanon 3 4 2.5 1.9 2 2 1 1 3.5 2.9 23 28 13 5 26 22 83 Thailand 47 58 1.7 0.9 26 39 24 34 2.6 1.3 47 46 71 64 10 14 84 Panama 2 3 2.1 1.7 1 2 1 1 3.1 2.4 30 34 29 26 19 16 85 Turkey 44 61 2.3 1.7 25 38 19 28 2.9 2.1 35 35 60 53 16 18 86 Poland 36 39 0.7 0.3 19 26 19 19 0.1 0.6 45 46 30 27 38 36 87 Estonia 1 1 0.6 -1.1 1 1 1 1 0.4 -0.8 51 49 15 14 43 41 88 Slovak Republic 5 5 0.6 0.3 3 4 2 3 0.9 0.7 45 48 14 12 36 32 89 Botswana 1 1 3.5 2.5 0 1 0 1 3.4 2.5 50 46 64 46 10 20 90 Venezuela 15 22 2.6 2.3 8 13 5 8 3.5 3.0 27 33 15 12 28 28 Upper-middle-income 331 17w 191t 269r 2.0w 29w 34w 31w 21w 29 27w 91 South Africa 29 41 2.4 2.2 16 24 11 16 2.7 2.4 35 37 17 14 35 32 92 Croatia 5 0.4 0.0 3 3 2 2 0.3 0.1 40 43 24 15 32 32 93 Mexico 67 92 2.3 1.9 35 54 22 36 3.5 2.8 27 31 37 28 29 24 94 Mauritius 1 0.9 1.3 1 1 0 0 2.3 1.8 26 32 27 17 28 43 95 Gabon 1 3.0 2.8 0 1 0 1 2.1 1.9 45 44 76 61 14 19 96 Brazil 121 159 2.0 1.5 71 101 48 71 3.2 1.6 28 35 37 23 24 23 97 Trinidad and Tobago 1 1.3 0.8 1 1 0 1 1.2 1.8 32 36 11 11 39 31 98 Czech Republic 10 10 0.1 -0.1 6 7 5 6 0.2 0.4 47 47 13 11 56 45 99 Malaysia 14 20 2.6 2.4 8 12 5 8 2.8 2.7 34 37 41 27 19 23 100 Hungary 11 10 -0.3 -0.3 7 7 5 5 -0.8 0.1 43 44 18 15 43 38 101 Chile 11 14 1.7 1.5 7 9 4 6 2.7 2.1 26 32 21 19 25 25 102 Oman 2 3.9 6.0 1 1 0 1 3.4 5.1 7 14 50 48 22 26 103 Uruguay 3 0.6 0.6 2 2 1 1 1.6 1.0 31 40 17 14 28 27 04 Saudi Arabia 19 5.2 3.7 5 10 3 6 6.5 3.2 8 13 45 20 16 20 105 Argentina 28 35 1.5 1.3 17 21 11 14 1.3 2.0 28 31 13 12 34 32 106 Slovenia 2 0.5 -0.1 1 1 1 1 0.3 0.1 46 46 15 42 44 107 Grzccz 10 10 0.5 0.6 6 7 4 4 1.2 0.9 28 36 31 23 29 28 Low- and middle-income 3,614 t 4,771 t 2.0w 1.6w 2,069 t 2,916 t 1,669 t 2,263 t 2.2w 1.7w 38w 40w 63w 58w 17w 18w Sub-Sahitrius Africa 381 r 583 t 3.0w 2.6w 196 t 305 t 173 t 257 t 2.7w 2.6w 42w 42w 72w 68w 9w 9w East Asia and Pacific 1,360 t 1,706 r 1.6w 1,3w 796 t 1,119 r 704 t 951 t 2.3w 1.3w 43w 45w 73w 70w 14w 15w South Asia 903 t 1,243 t 2.2w 1.9w 508 t 732 t 389 t 532 t 2.1w 2.1w 34w 33w 70w 64w 13w 16w Europe and Central Asia 437 t 488 t 0.9w 0.3w 277 r 317 t 219 t 238 t 0.6w 0.5w 46w 46w 27w 23w 37w 36w Middle East and N. Africa 175 t 272 t 3.1w 2.7w 91 t 151 t 54 t 88 t 3.2w 3.3w 24w 26w 48w 36w 21w 24w Latin America and Caribbean 358 t 478 t 2.0w 1.7w 201 t 293 t 130 t 197 t 3.0w 2.3w 27w 33w 34w 25 w 25 w 24w High-income economies 816 t 902 t 0.7w 0.7w 522 e 605 t 368 r 432 t 1.2w 0.9w 39w 42w 9w 5w 35 w 31 w 108 Korea, Rep. 38 45 1.2 0.9 24 32 16 22 2.3 1.9 39 40 37 18 27 35 109 Portugal 10 10 0.1 0.1 6 7 5 5 0.4 0.5 39 43 26 18 36 34 110 Spain 37 39 0.4 0.2 23 27 14 17 1.3 1.0 28 36 19 12 37 33 111 New Zealand 3 4 0.8 1.4 2 2 1 2 2.0 1.5 34 44 11 10 33 25 112 Ireland 3 4 0.3 0.5 2 2 1 1 0.4 1.7 28 33 19 14 34 29 113 3 Israel 4 6 1.8 3.5 2 3 1 2 2.3 3.5 34 40 6 4 32 29 114 3Kuwair 1 2 4.4 -4.9 1 1 0 1 5.8 -1.6 13 27 2 1 32 25 115 3 United Arab Emirates 1 2 5.7 5.8 1 1 1 1 5.1 3.9 5 13 4 7 37 24 116 United Kingdom 56 59 0.2 0.3 36 38 27 29 0.6 0.3 39 43 3 2 38 29 117 Australia 15 18 1.5 1.1 10 12 7 9 2.3 1.4 37 43 6 5 32 26 118 Italy 56 57 0.1 0.2 36 39 23 25 0.8 0.4 33 38 13 9 38 32 119 Canada 25 30 1.2 1.3 17 20 12 15 1.9 1.0 40 45 7 3 33 25 120 Finland 5 5 0.4 0.5 3 3 2 3 0.6 0.1 46 48 12 8 35 31 121 I Hung Kong 5 6 1.2 1.6 3 4 2 3 1.6 1.3 34 37 1 1 50 37 122 Sweden 8 9 0.3 0.6 5 6 4 5 1.0 0.3 44 48 . 123 Netherlands 14 15 0.6 0.7 9 11 6 7 2.0 0.6 31 40 6 5 31 26 124 Belgium 10 10 0.1 0.4 6 7 4 4 0.2 0.5 34 40 3 3 35 28 125 France 54 58 0.5 0.5 34 38 24 26 0.4 0.8 40 44 8 5 35 29 126 3 Singapore 2 3 1.7 2.0 2 2 1 1 2.3 1.7 35 38 2 0 44 36 127 Austria 8 8 0.2 0.8 5 5 3 4 0.5 0.5 40 41 10 8 41 37 128 United States 228 263 0.9 1.0 151 172 110 133 1.4 1.1 42 46 3 3 31 28 129 Getmany 78 82 0.1 0.6 52 56 37 40 0.6 0.3 40 42 7 4 45 38 130 Denmark 5 5 0.0 0.3 3 4 3 3 0.7 0.1 44 46 7 6 31 28 131 Norway 4 4 0.4 0.5 3 3 2 2 0.9 0.7 40 46 8 6 29 25 132 Japan 117 125 0.6 0.3 79 87 57 66 1.1 0.6 38 41 11 7 35 34 133 Switzerland 6 7 0.6 1.0 4 5 3 4 1.5 0.8 37 40 6 6 39 35 World 4,429 t 5,673; 1,7w 1.5w 2,590 t 3,521 t 2,037 t 2,695 t 2.0w 1.6w 38w 40w 53w 49w 20w 20w a. Participation rates from ILO a e applied to population estimates to derive labor force estimates. 222 WORLD DEVELOPMENT REPORT 1997 Table 5. Distribution of income or consumption Percentage share of income or consumption Survey Cmi Lowest Lowest Second Third Fourth Highest Highest year index 10% 20% quintile quintile quintile 20% 10% Low-income economies Excluding China and India I Mozainbique .. . . 2 Ethiopia .. . . . . . . . . . . 0 3 Tanzania 1993' 38.1 2.9 6.9 10.9 15.3 21.5 45.4 30.2 4 Burundi . . 5 Malawi 6 Sierra Leone 7 Rwanda 1983-85' 28.9 4.2 9.7 13.2 16.5 21.6 39.1 24.2 8 Chad .. .. .. . .. 9 Nepal 199596a,b 36.7 3.2 7.6 11.5 15.1 21.0 44.8 29.8 10 Niger l992 36.1 3.0 7.5 11.8 15.5 21.1 44.1 29.3 11 Madagascar 1993s.b 43.4 2.3 5.8 9.9 14.0 20.3 50.0 34.9 12 Burkina Faso .. . . .. .. .. .. 13 Vietnam l993' 35.7 3.5 7.8 11.4 15.4 21.4 44.0 29.0 14 Bangladesh l992 28.3 4.1 9.4 13.5 17.2 22.0 37.9 23.7 15 Uganda l992-93' 40.8 3.0 6.8 10.3 14.4 20.4 48.1 33.4 16 Mali .. .. .. .. .. 17 Guinea-Bissau l99l' 56.2 0.5 2.1 6.5 12.0 20.6 58.9 42.4 18 Haiti . . . . . . . . . . . . . 19 Nigeria 1992-93 45.0 1.3 4.0 8.9 14.4 23.4 49.3 31.3 20 Yemen, Rep. 21 Cambodia .. . . . . .. .. . . 22 Kenya 57.5 1.2 3.4 6.7 10.7 17.0 62.1 47.7 23 Togo 24 Mongolia 25 Gambia, The 26 India l992' 33.8 3.7 8.5 12.1 15.8 21.1 42.6 28.4 27 Central African Republic . . . . . . . . . . . . . 28 Lao PDR l992'' 30.4 4.2 9.6 12.9 16.3 21.0 40.2 26.4 29 Benin . . . . . . . . . . . . . 30 Nicaragua 1993'" 50.3 1.6 4.2 8.0 12.6 20.0 55.2 39.8 31 Ghana 1992'"' 33.9 3.4 7.9 12.0 16.1 21.8 42.2 27.3 32 Zambia 1993'"' 46.2 1.5 3.9 8.0 13.8 23.8 50.4 31.3 33 Angola . 34 Georgia .. .. .. .. .. .. 35 Pakistan 199P'' 31.2 3.4 8.4 12.9 16.9 22.2 39.7 25.2 36 Mauritania 1988' 42.4 0.7 3.6 10.6 16.2 23.0 46.5 30.4 37 Azerbaijan . . . . . . . . . . . . . 38 Zimbabwe l990' 56.8 1.8 4.0 6.3 10.0 17.4 62.3 46.9 39 Guinea 1991a,b 46.8 0.9 3.0 8.3 14.6 23.9 50.2 31.7 40 Honduras 1992 52.7 1.5 3.8 7.4 12.0 19.4 57.4 41.9 41 Senegal 1991,b 54.1 1.4 3.5 7.0 11.6 19.3 58.6 42.8 42 China l995' 41.5 2.2 5.5 9.8 14.9 22.3 47.5 30.9 43 Cameroon . . . . . . . . . . . . . 44 Core d'Ivoire 1988'" 36.9 2.8 6.8 11.2 15.8 22.2 44.1 28.5 45 Albania 46 Congo 47 Sri Lanka 30.1 3.8 8.9 13.1 16.9 21.7 39.3 25.2 48 Kvrgvz Republic 49 Armenia Middle-income economies Lower-middle-income 50 Lesorho l986-87' 56.0 0.9 2.8 6.5 11.2 19.4 60.1 43.4 51 Egypt, Arab Rep. 1991a,b 32.0 3.9 8.7 12.5 16.3 21.4 41.1 26.7 52 Bolivia l990' 42.0 2.3 5.6 9.7 14.5 22.0 48.2 31.7 53 Macedonia, FYR 54 Moldova l992 34.4 2.7 6.9 11 16.7 23.1 41.5 25.8 55 Uzbekistan 56 Indonesia 1993 31.7 39 8.7 12.3 16.3 22.1 40.7 25.6 57 Philippines l988' 40.7 2.8 6.5 10.1 14.4 21.2 47.8 32.1 58 Morocco 1990-91'" 39.2 2.8 6.6 10.5 15.0 21.7 46.3 30.5 59 Syrian Arab Republic 60 Papua New Guinea .. 61 Bulgaria 1992a,d 30.8 3.3 8.3 13.0 17.0 22.3 39.3 24.7 62 Kazakstan l993c 32.7 3.1 7.5 12.3 16.9 22.9 40.4 24.9 63 Guatemala 1989c,d 59.6 0.6 2.1 5.8 10.5 18.6 63.0 46.6 64 Ecuador 1994'" 46.6 2.3 5.4 8.9 13.2 19.9 52.6 37.6 65 Dominican Republic l989'" 50.5 1.6 4.2 7.9 12.5 19.7 55.7 39.6 66 Romania 1992" 25.5 3.8 9.2 14.4 18.4 23.2 34.8 20.2 67 Jordan 1991'" 43.4 2.4 5.9 9.8 13.9 20.3 50.1 34.7 68 Jamaica 1991"' 41.1 2.4 5.8 10.2 14.9 21.6 47.5 31.9 69 Algeria 1988'"' 38.7 2.8 6,9 11.0 15.1 20.9 46.1 31.5 70 El Salvador . 0 0 . . . 0 0 0 0 71 Ukraine 1992" 25.7 4.1 9.5 14.1 18.1 22.9 35.4 20.8 72 Paraguay Note: For data comparability and coverage, see the technical notes. SELECTED WORLD DEVELDFMENT INDICATORS 223 Percentage share of income or consumption Survey Gini Lowest Lowest Second Third Fourth Highest Highest year index 10% 20% quintile quintile quintile 20% 10% 73 Tunisia 1990" 40.2 2.3 5.9 10.4 15.3 22.1 46.3 30.7 74 Lithuania l993" 33.6 3.4 8.1 12.3 16.2 21.3 42.1 28.0 75 Colombia 1991c,d 51.3 1.3 3.6 7.6 12.6 20.4 55.8 39.5 76 Namibia .. .. .. .. .. .. 77 Belarus l993 21.6 4.9 11.1 15.3 18.5 22.2 32.9 19.4 78 Russian Federation l993" 49.6 1.2 3.7 8.5 13.5 20.4 53.8 38.7 79 Latvia l993' 27.0 4.3 9.6 13.6 17.5 22.6 36.7 22.1 80 Peru l994' 44.9 1.9 4.9 9.2 14.1 21.4 50.4 34.3 81 Costa Rica 46.1 1.2 4.0 9.1 14.3 21.9 50.7 34.1 82 Lebanon . . 83 Thailand l992' 46.2 2.5 5.6 8.7 13.0 20.0 52.7 37.1 84 Panama l989" 56.6 0.5 2.0 6.3 11.6 20.3 59.8 42.2 85 Turkey .. .. .. .. .. 86 Poland 1992a 27.2 4.0 9.3 13.8 17.7 22.6 36.6 22.1 87 Estonia 1993c,d 39.5 2.4 6.6 10.7 15.1 21.4 46.3 31.3 88 Slovak Republic 1992c 19.5 5.1 11.9 15.8 18.8 22.2 31.4 18.2 89 Venezuela 1990d 53.8 1.4 3.6 7.1 11.7 19.3 58.4 42.7 90 Borswana Upper-middle-income 91 South Africa 1993 58.4 1.4 3.3 5.8 9.8 17.7 63.3 47.3 92 Croatia 93 Mexico l992' 50.3 1.6 4.1 7.8 12.5 20.2 55.3 39.2 94 Mauritius 95 Gabon 96 Brazil l989" 63.4 0.7 2.1 4.9 8.9 16.8 67.5 51.3 97 Trinidad and Tobago .. .. .. .. .. 98 Czech Republic l993' 26.6 4.6 10.5 13.9 16.9 21.3 37.4 23.5 99 Malaysia l989c, 48.4 1.9 4.6 8.3 13.0 20.4 53.7 37.9 100 Hungary l993' 27.0 4.0 9.5 14.0 17.6 22.3 36.6 22.6 101 Chile 1994c,d 56.5 1.4 3.5 6.6 10.9 18.1 61.0 46.1 102 Oman 103 Uruguay 104 Saudi Arabia .. . . . . . 105 Argentina . . . . . 106 Slovenia 1993c,d 28.2 4.1 9.5 13.5 17.1 21.9 37.9 23.8 107 Greece Low- and middle-income Sub-Sabaran Africa East Asia and Pacific South Asia Europe and Central Asia Middle East and N. Africa Latin America and Caribbean High-income economies 108 Korea, Rep. 109 Portugal 110 Spain 1988f 8.3 13.7 18.1 23.4 36.6 21.8 111 NewZealand 5.1 10.8 16.2 23.2 44.7 28.7 112 Ireland 113 tlsrael 1979 6.0 12.1 17.8 24.5 39.6 23,5 114 tKuwait 115 t United Arab Emirates 116 United Kingdom l988 4.6 10.0 16.8 24.3 44.3 27.8 117 Australia l985e1 4.4 11.1 17.5 24.8 42.2 25.8 118 Italy l986F 6.8 12.0 16.7 23.5 41.0 25.3 119 Canada 5.7 11.8 17.7 24.6 40.2 24.1 120 Finland l98l 6.3 12.1 18.4 25.5 37.6 21.7 121 tHongKong l980 5.4 10.8 15.2 21.6 47.0 31.3 122 Sweden 1981 8.0 13.2 17.4 24.5 36.9 20.8 123 Netherlands 1988 8.2 13.1 18.1 23.7 36.9 21.9 124 Belgium 1 978-79 7.9 13.7 18.6 23.8 36.0 21.5 125 France 1989f 5.6 11.8 17.2 23.5 41.9 26.1 126 tSingapore 5.1 9.9 14.6 21.4 48.9 33.5 127 Austria 128 United States 4.7 11.0 17.4 25.0 41.9 25.0 129 Germany 1988f 7.0 11.8 17.1 23.9 40.3 24.4 130 Denmark l98lf 5.4 12.0 18.4 25.6 38.6 22.3 131 Norway I979f 6.2 12.8 18.9 25.3 36.7 21.2 132 Japan l979" 8.7 13.2 17.5 23.1 37.5 22.4 133 Switzerland 1982 5.2 11.7 16.4 22.1 44.6 29.8 World a. Refers to expenditure shares by percentiles of persons. b. Ranked by per capita expenditure. c. Refers to income shares by percentiles of persons. d. Ranked by per capita income. e. Refers to income shares by percentiles of households. f. Ranked by household income. 224 WORLD DEVELOPMENT REPORT 1997 Table 6. Health Maternal Infant mortality Prevalence of Contraceptive mortality ratio Percentage of total population with access to rate (per 1,000 malnutrition prevalence (per 100,000 Health care Safe water Sanitation live births) Total fertility rate live births) 1% under 5) rate 1%) 1980 1993 1980 1994-95 1980 1994-95 1980 1995 1989-95 1989-95 1980 1995 1989-95 Low-income economies 98w 69w 4.3w 3.2w Excluding China and India 116w 89w 6.3w 5.0w 1 Mozambique 9 28 10 23 145 113 . . . . 6.5 6.2 1,512 2 Ethiopia 55 4 27 10 155 112 47 4 6.6 7.0 1,528a 3 Tanzania 72 93 49 86 104 82 28 10 6.7 5.8 748 4 Burundi 80 58 48 121 98 .. .. 6.8 6.5 1,327a 5 Malawi 40 . 54 63 169 133 27 13 7.6 6.6 620' 6 Chad 26 29 32 147 117 .. .. 5.9 5.9 l,594 7 Rwanda . . . .. . . 128 133 28 21 8.3 6.2 1,512 8 Sierra Leone 26 . . 13 . . 190 179 23 .. 6.5 6.5 9 Nepal 10 . 11 48 0 6 132 91 70 . . 6.4 5.3 515C .. 74 593b 10 Niger 30 57 15 150 119 4 4 11 Burkina Faso . 35 . . 5 14 121 99 . . 8 7.5 6.7 939 12 Madagascar . 32 17 138 89 32 17 6.5 5.8 13 Bangladesh 80 74 83 30 132 79 84 40 6.1 3.5 887 14 Uganda . 42 60 116 98 23 15 7.2 6.7 506 15 Vietnam 75 38 21 57 41 45 . . 5.0 3.1 16 Guinea-Bissau 30 . 24 27 20 168 136 . . . . 6.0 6.0 4.4 6OO 17 Haiti . . 28 . . 24 123 72 27 18 5.9 18 Mali 20 . 44 44 184 123 . . . . 7.1 6.8 1,249a 19 Nigeria 40 67 43 63 99 80 43 6 6.9 5.5 20 Yemen, Rep. 16 . 52 .. 51 141 100 30 .. 7.9 7.4 1,47P' 21 Cambodia . . 13 . . 201 108 . . . . 4.7 4.7 22 Kenya 49 43 72 58 23 27 7.8 4.7 23 Mongolia 90 . .. .. 82 55 10 .. 5.4 3.4 24 Togo . 67 .. 20 110 88 6.6 6.4 626a 25 Gambia, The 90 . 42 61 .. 34 159 126 6.5 5.3 26 Central African Republic . . 16 . . . . . . 117 98 . . . . 5.8 5.1 649 27 India 50 . 63 29 116 68 63 43 5.0 3.2 437' 28 Lao PDR . 41 . . 30 127 90 40 6.7 6.5 29 Benin 42 70 22 122 95 36 .. 6.5 6.0 30 Nicaragua . 57 .. . . 90 46 12 44 6.2 4.1 31 Ghana 25 56 29 100 73 27 20 6.5 5.1 742 32 Zambia . . 47 . . 42 90 109 27 15 7.0 5.7 33 Angola 70 24 32 16 153 124 20 6.9 6.9 34 Georgia . . . . . . . 25 18 . . . . 2.3 2.2 35 Pakistan 65 85 38 60 16 30 124 90 40 14 7.0 5.2 36 Mauritania . . 41 64 120 96 6.3 5.2 37 Azerbaijan . . . . .. . . 30 25 . . . . 3.2 2.3 38 Zimbabwe 55 74 5 58 82 55 16 6.8 3.8 39 Guinea 45 49 12 6 161 128 18 .. 6.1 6.5 880' 40 Honduras . 70 .. 68 70 45 19 47 6.5 4.6 41 Senegal 40 . . . . 91 62 20 7 6.7 5.7 42 China . 83 . . 42 34 17 83 2.5 1.9 115 43 Cameroon 20 41 .. 40 94 56 14 16 6.5 5.7 44 Côte d'Ivoire . . 20 82 17 54 108 86 11 7.4 5.3 887 .. 23d 45 Albania 100 92 .. 100 47 30 3.6 2.6 46 Congo 60 9 89 90 .. 6.2 6.0 822 43 30 .. .. 4.1 3.3 80d 47 Kyrgyz Republic 75 53 3Ql 48 Sri Lanka 90 .. 57 66 34 16 38 3.5 2.3 49 Armenia 26 16 . . 2.3 1.8 35' Middle-income economies 65w 39w 3.8w 3.0w Lower-middle-income 68w 41w 3.7w 3Gw 50 Lesotho . . 18 57 12 35 108 76 21 23 5.6 4.6 598 51 Egypt, Arab Rep 100 99 90 84 70 .. 120 56 9 48 5.1 3.4 52 Bolivia . . 60 . . 44 118 69 13 45 5.5 4 53 Macedonia, FYR . . . . 54 23 . . 2.5 2.2 12' 54 Moldova . . . . 50 35 22 . . 2.4 2.0 34 55 Uzbekisran . . . 18 47 30 . . . . 4.8 3.7 43 56 Indonesia . 63 . . 55 90 51 39 55 4.3 2.7 390 52 40 4.8 3.7 208h 57 Philippines 84 . . 75 39 30 58 Morocco 62 32 59 50 63 99 55 9 50 5.4 3.4 372c 45 32 7.4 4.8 179d 59 Syrian Arab Republic 99 71 87 78 56 60 Papua New Guinea 31 .. 26 67 64 .. 5.7 4.8 61 Bulgaria . . 96 99 20 15 2.1 1.2 20" 62 Kazakstan . . . . . . 33 27 . . 2.9 2.3 53" 63 Guatemala . . 64 71 75 44 .. 32 6.2 4.7 464' 64 Ecuador 70 64 67 36 45 57 5.0 3.2 65 Dominican Republic . 79 .. 85 76 37 10 56 4.2 2.9 66 Romania 77 50 49 29 23 .. 57 2.4 1.4 48d 67 Jamaica . 70 .. 74 21 13 10 55 3.7 2.4 68 Jordan 90 89 89 76 30 41 3! 17 .. 6.8 4.8 132 69 Algeria . 77 . . . . 98 34 9 51 6.7 3.5 140" 70 El Salvador 62 .. 73 81 36 22 53 5.3 3.7 71 Ukraine . 97 50 49 17 15 . . . . 2.0 1.5 33' 72 Paraguay . . 30 50 41 4 48 4.8 4.0 180" Note: Fot data comparability and coverage, see the technical notes Figures in italics are for yeats other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS ZZS Maternal Infant mortalit3' Percentage of total population with access to Prevalence of Contraceptive modality ratio rate lper 1,000 malnutrition prevalence (per 100,000 Health care Safe water Sanitation live birthsl Total fertility rate 1% under 5) rate (%l live births) 1980 1993 1980 1994-95 1980 1994-95 1980 1995 1989-95 1989-95 1980 1995 1989-95 73 Tunisia 95 90 72 86 46 72 71 39 .. 5.2 2.9 138 74 Lithuania . . . 20 14 . . . 2.0 1.5 75 Colombia 88 96 70 45 26 10 72 3.8 2.8 107a 76 Namibia 57 .. 36 90 62 29 5.9 5.0 518 77 Belarus . . 50 100 16 13 . . 2.0 1.4 25d 78 Russian Federation . . . . . 22 18 . . . . 1.9 1.4 79 Latvia . . . . . 20 16 . . .. 2.0 1.3 80 Pent . 60 47 81 47 16 55 4.5 3.1 81 Costa Rica . 100 . . 99 20 13 2 3.7 2.8 82 Lebanon . 92 . 59 .. 48 32 .. 4.0 2.8 83 Thailand 30 59 81 87 49 35 13 3.5 1.8 84 Panama . 82 87 32 23 7 3.7 2.7 85 Turkey . 67 92 .. 94 109 48 .. 4.3 2.7 183c 86 Poland 100 67 50 100 21 14 2.3 1.6 lO' 87 Estonia . 17 14 2.0 1.3 88 Slovak Republic . . 43 51 21 11 . . 2.3 1.5 89 Botswana 70 55 69 56 . . 6.7 4.4 220a 90 Venezuela . 88 . . 55 36 23 6 . . 4.1 3.1 200' Upper middle mcome 57w 35 w 39w 2 9w 91 South Africa . . 46 67 50 . . 4.9 3.9 404 92 Croatia . 96 68 21 16 . . 1.9 1.5 10' 93 Mexico 51 87 70 51 33 . . . . 4.5 3.0 94 Mauritius 100 99 . 100 100 32 16 .. 75 2.7 2.2 112' 95 Gabon 67 76 116 89 . . 4.5 5.2 483 96 Brazil . 92 73 70 44 18 . . 3.9 2.4 200' 97 Trinidad and Tobago . 82 .. 56 35 13 .. 3.3 2.1 98 Czech Republic . . . . . 16 8 . . 69 2.1 1.3 l2' 99 Malaysia 88 90 75 94 30 12 23 4.2 3.4 34F 100 Hungary .. 94 23 11 .. .. 1.9 1.6 l0' 101 Chile . 96 . . 71 32 12 1 .. 2.8 2.3 102 Oman 75 89 15 56 .. 72 41 18 . . 9 9.9 7.0 103 Uruguay 83 .. 82 37 18 . . .. 2.7 2.2 104 Saudi Arabia 85 98 91 93 76 86 65 21 ,. 7.3 6.2 l8' 105 Argentina 64 89 35 22 . . 3.3 2.7 140' 106 Slovenia . .. 90 15 7 . . 2.1 1.3 107 Greece . . . 96 18 8 . . 2.2 1.4 Low- and middle-income 87w 60w 4.1w 3.1w Sub-Saharan Africa 114w 92w 6.7w 5.7w East Asia and Pacific 56w 40w 3.1 w 2.2w South Asia 120w 75w 5.3w 3.5w Europe and Central Asia 40w 26w 2.5w 2.0w Middle East and N. Africa 97w 54w 6.1 w 4.2w Latin America and Carribbean 60w 37w 4.1 w 2.8 w High-income economies 13 w 7 1.9w 1.7w 108 Korea, Rep . 100 . 89 . . 100 26 10 . . 2.6 1.8 30d 109 Portugal . . 57 . . . 100 24 7 2.2 1.4 21 110 Spain . 98 99 95 97 12 7 .. 2.2 1.2 111 NewZealand 100 87 .. .. 13 7 . . .. 2.1 2.1 112 Ireland . . . . 100 11 6 . . 60 3.2 1.9 113 IsraeI . 99 . . 70 15 8 . . 3.2 2.4 114 tKuwait 100 . 100 . 100 . . 27 11 . . 5.3 3.0 l8' 115 t United Arab Emirates 96 90 100 98 75 95 55 16 .. 5.4 3.6 20 116 United Kingdom 100 .. 96 12 6 .. 1.9 1.7 117 Australia 99 99 95 99 90 11 6 .. 1.9 1.9 118 Italy . 99 . 99 100 15 7 . . 1.6 1.2 119 Canada . 97 100 60 85 10 6 .. 1.7 1.7 120 Finland 100 100 100 8 5 1.6 1.8 121 tHongKong .. .. 11 5 .. 2.0 1.2 122 Sweden 35 100 7 4 .. 1.7 1.7 123 Netherlands 100 100 100 100 9 6 .. 1.6 1.6 124 Belgium 99 100 12 8 . . 1.7 1.6 125 France . 100 85 96 10 6 . . 1.9 1.7 126 tSingapore . 100 100 .. 100 12 4 14 1.7 1.7 127 Austria 100 85 100 14 6 . . 1.6 1.5 128 United States 90 98 85 13 8 1.8 2.1 129 Germany .. 100 12 6 .. 1.6 1.2 130 Denmark . 100 100 100 100 8 6 .. 1.5 1.8 131 Norway . 100 100 100 8 5 .. 1.7 1.9 132 Japan 100 95 .. 85 8 4 3 1.8 1.5 61 133 Switzerland 100 85 100 9 6 . . . . 1.6 1.5 World 80w 55 w 3.7w 2.9w a. UNICEF/WHO estimate based on statistical modeling. b. Indirect estintatr based on sample survey. c. Based on sample survey. d. Official estimate. e. Based ona survey covering thirty provinces. f. Based on civil registration. 226 WORLD DEVELOPMENT REPORT 1997 Table 7. Education Percentage School enrollment as a % of age group of cohort reaching Prima,y Secondary grade 4 Adult illiteracy (%) Female Male Female Male Tertiary Female Male Female Male 1980 1993 1980 1993 1980 8993 1980 1993 1980 1993 1980 1990 1980 1990 1995 1995 Low-income economies 81 w 98w 104w 112w 26w 41 w 42w 3w 45w 24w Excluding China and India 64w 85w 14w 25 w 3w 55w 37w I Mozambique 84 51 114 69 3 6 8 9 0 0 . 54 60 77 42 2 Ethiopiafl 23 19 44 27 6 11 11 12 0 1 48 56 42 56 75 55 3 Tanzania 86 69 99 71 2 5 4 6 . . 89 90 90 89 43 21 ........ 4 Burundi 21 63 32 76 2 5 4 9 1 1 83 79 83 79 78 51 5 Malawi 48 77 72 84 2 3 5 6 1 1 55 68 62 73 58 28 6 Sierra Leone 43 61 . 8 20 . 1 82 55 7 Rwanda 60 76 66 78 3 9 4 11 0 74 76 73 73 48 30 8 Chad 30 . 62 . . . . . 1 65 74 65 38 9 Nepal 49 87 117 129 9 23 33 46 3 6 86 59 10 Niger 18 21 33 35 3 4 7 9 0 1 79 . 82 . 93 79 11 Madagascar 133 72 139 75 14 . . 14 3 4 . 72 . 68 . 12 BurkinaFaso 14 30 23 47 2 6 4 11 0 . 79 90 79 86 91 71 13 Vietnam 106 111 40 44 . . 2 2 67 . 71 . 9 4 14 Bangladesh 46 105 76 128 9 12 26 26 3 . 30 46 29 44 74 51 15 Uganda 43 83 56 99 3 10 7 17 1 1 . . 50 26 16 Mali 19 24 34 38 5 6 12 12 1 . 77 84 73 87 77 61 17 Guinea-Bissau 43 . 94 . 2 2 10 10 . . . 47 . 63 . 58 32 18 Haiti 70 . 82 . 13 . 14 . . I . 64 60 63 60 58 52 ........ 19 Nigeria 104 82 135 105 14 27 27 32 2 76 74 53 33 20 Yemen, Rep . . . .. 21 Cambodia 46 48 .. .. .. 47 20 22 Kenya 110 91 120 92 16 23 23 28 1 85 78 84 76 30 14 23 Togo 91 81 146 122 16 12 51 34 2 3 84 82 90 87 63 33 24 Mongolia 107 . 107 97 87 85 .. .. 25 Gambia, The 35 61 67 84 7 13 16 25 75 47 26 India 67 91 98 113 20 39 . . 5 . 52 . 57 62 35 27 Central African Republic 51 . 92 92 7 . 21 . . 1 2 48 32 28 LaoPDR 104 92 123 123 16 19 25 31 0 2 31 31 56 31 29 Benin 41 44 87 88 9 7 24 17 2 . 73 58 77 58 74 51 30 Nicaragua 102 105 96 101 45 44 39 39 13 9 55 62 51 55 33 35 31 Ghana 71 70 89 83 31 28 51 44 2 82 . 87 . 47 24 32 Zambia 83 99 97 109 11 22 . . 2 29 14 33 Angola . . .. 0 1 37 49 34 Georgia . .. . . 30 . . 35 Pakistan 27 49 51 80 8 . 20 . . . 41 45 53 55 76 50 36 Mauritania 26 62 47 76 4 11 17 19 . . 4 86 83 96 82 74 50 37 Azerbaijan . 87 91 88 .. 89 25 26 38 Zimbabwe 114 123 40 .. 51 1 6 20 10 39 Guinea 25 30 48 61 10 6 24 17 5 . 78 81 78 50 40 Honduras 99 112 98 111 31 37 29 29 8 9 27 27 41 Senegal 37 50 56 67 7 11 15 21 3 3 90 93 77 57 42 China 103 116 121 120 37 51 54 60 1 4 27 10 43 Cameroon 89 . 107 . 13 . 24 . . 2 2 81 . 81 . 48 25 44 Côted'lvoire 63 58 95 80 12 17 27 33 3 . 91 82 94 85 70 50 45 Albania 111 97 116 95 63 70 8 10 96 . 97 . 46 Congo . .. .. .. 91 . 91 33 17 47 Sri Lanka 100 105 105 106 57 78 52 71 3 6 96 . 95 13 7 48 Kyrgyz Republic .. . . 28 21 49 Armenia . 93 . 87 . 90 . . 80 30 49 . . . .. Middle income economies 99w 101 w 106w 105w 48w 62w 53w 64w 21 w 20w 23w 14w Lowermiddle income 97w 101w 106w 106w 50w 61w 56w 65w 24w 22w 50 Lesotho 120 105 85 90 21 31 14 21 2 2 77 85 1 75 38 19 51 Egypt,Arab Rep 61 89 84 105 39 69 61 81 16 17 83 . 75 61 36 52 Bolivia 81 . 92 . 32 . 42 . . 16 23 50 . 52 24 10 53 Macedonia, FYR 87 88 55 53 28 16 54 Moldova 77 78 72 67 29 35 55 Uzbekisran . 79 . 80 . 92 . . 96 30 33 . . 56 Indonesia 100 112 115 116 23 39 35 48 . . 10 . . 22 10 57 Philippines 112 . 113 . 69 . 61 . . 24 26 . 82 . 78 6 5 58 Morocco 63 60 102 85 20 29 32 40 6 10 89 81 90 83 69 43 59 SyrianArab Republic 88 99 111 111 35 42 57 52 17 18 91 96 94 97 44 14 60 61 62 63 64 PapuaNewGuinea Bulgaria Kazakstan Guatemala Ecuador 51 98 . 65 116 67 84 86 78 122 66 98 77 119 80 87 86 89 124 84 . 8 17 53 10 70 91 23 56 15 85 . . 20 53 15 66 89 25 54 2 16 34 35 8 23 42 . ........ . 95 56 76 70 91 . . 98 66 78 72 93 . 37 . 51 12 19 38 8 65 Dominican Republic 99 95 43 .. 30 .. 18 18 66 Romania 101 86 102 87 69 82 73 83 12 12 94 93 67 Jordan 102 95 105 94 73 54 79 52 27 19 95 97 95 99 21 7 68 Jamaica 104 108 103 109 71 70 63 62 7 6 100 98 11 19 69 Algeria 81 96 108 111 26 55 40 66 6 11 91 95 92 96 51 26 70 El Salvador 75 80 75 79 23 30 26 27 4 15 55 52 30 27 71 Ukraine 87 87 95 65 42 46 . 72 Paraguay 101 110 107 114 38 36 10 8 78 79 9 7 Note: For data comparability and coverage, see the technical notes Figures in italics are for years other than those specified. SELECTED WORLD DEVELOFMENT INDICATORS 227 Percentage School enrollment as a % of age group of cohort reaching Primary Secondary grade 4 Adult illiteracy (%l Female Male Female Male Tertiary Female Male Female Male 1980 1993 1980 1993 1980 1993 1980 1993 1980 1993 1980 1990 1980 1990 1995 1995 73 Tunisia 88 113 118 123 20 49 34 55 5 11 90 93 94 95 45 21 74 Lithuania .. 90 . 95 .. 79 .. 76 49 39 75 Colombia 126 120 123 118 41 68 40 57 9 10 46 74 42 72 9 9 76 Namibia .. 138 .. 134 61 49 .. 3 64 65 77 Belarus .. 95 96 96 89 39 44 78 Russian Federation 102 107 102 107 97 91 95 84 46 45 . . 79 Larvia .. 82 .. 83 .. 90 .. 84 45 39 .. 80 Peru 111 . . 117 . . 54 . . 63 . . 17 40 83 85 . 17 6 Costa Rica 104 106 30 ........8 81 105 106 51 49 44 45 21 84 91 80 90 5 5 82 Lebanon .. 114 .. 117 .. 78 .. 73 30 29 10 5 83 Thailand 97 97 100 98 28 37 30 38 13 19 4 84 Panama 105 .. 108 .. 65 65 58 .. 21 23 88 88 87 85 10 9 85 Turkey 90 98 102 107 24 48 44 74 5 16 98 98 28 8 86 Poland 99 97 100 98 80 87 75 82 18 26 87 Estonia .. 83 84 .. 96 87 43 38 88 89 90 Slovak Republic Venezuela Botswana . . 100 101 97 120 . 83 . 101 95 113 .. 25 20 90 41 55 18 17 . . 87 29 49 . 21 . 1 17 29 3 ........10 98 94 91 . 92 . 40 20 8 Upper-middle-income 103w .. 106w 43w 43w .. 13w 16w 14w 12w 91 SouthAfrica . . 110 . . 111 . . 84 . . 71 . . 13 18 18 92 Croatia . . . . . . . . .. 86 . . 80 . . 27 . . 93 Mexico 121 110 122 114 46 58 51 57 14 14 63 85 . 13 8 94 Mauritius 91 106 94 107 49 60 5! 58 4 97 98 97 99 21 13 ........17 1 95 Gabon .. 136 . . 132 .. .. . . 3 79 80 82 82 47 26 96 Brazil 97 . 101 . . 36 . . 31 . . 11 12 17 97 Trinidad and Tobago 100 94 98 94 78 74 4 8 89 97 83 96 3 1 98 Czech Republic .. 100 .. 99 .. 88 .. 85 18 16 .. .. 99 Malaysia 92 93 93 93 46 61 50 56 4 . . . . 99 . . 98 22 11 100 Hungaty 97 95 96 95 67 82 72 79 14 17 96 97 96 97 101 Chile 108 98 110 99 56 67 49 65 12 27 81 95 78 95 5 5 102 Oman 36 82 69 87 6 57 19 64 .. 5 103 Uruguay 107 108 107 109 62 .. 61 .. 17 30 99 98 93 98 2 3 104 Saudi Arabia 49 73 74 78 23 43 36 54 7 14 90 .. 81 50 29 105 Argentina 106 107 106 108 .. 75 .. 70 22 41 76 73 4 4 106 Slovenia .. 97 .. 97 ., 90 .. 88 .. 28 .. 107 Greece 103 103 . . 77 . . 85 17 98 98 . . . 87w 99w 105w 110 w 33w 49w 45w .. 8w .. 39w 21 w Low- and middle-income Sub-Saharan Africa EastAsiaandPacifit 68w 65w 90w 78w 102w 116w 118w 120w 10w 22w 36w 51w 20w 51w 27w 60w 1w 3w .5w 54w 35w 24w 9w SouthAsia 61w 87w 91w 110w 18w 35w 36w .. 5w 64w 37w Europe and Central Asia . 97w .. 97w .. 90w .. 81 w 31 w 32w .. -. Middle East and N. Africa 74w 91w 98w 103w 32w 51 w 52w 65w 11 w 14w 50w 28w Latin America and Caribbean 105w . . 108w . . 41 w . . 40w . . 14w 15w 14w 12w High-income economies 103w 103w 103w 103w .. 98w .. 97w 35w 56w 108 Korea, Rep. 111 102 109 100 74 92 82 93 15 48 96 100 96 100 b h 109 Portugal 123 118 124 122 40 .. 34 .. 11 23 110 Spain 109 lOS 110 104 89 120 85 107 23 41 94 95 92 94 . Ill NewZealajid 111 101 111 102 84 104 82 103 27 58 . 99 . 98 b b 112 Ireland 100 103 100 103 95 110 85 101 18 34 100 99 97 98 b b 113 tlsrael . . 96 . . 95 . . 91 . . 84 29 35 97 . 98 . 114 115 116 117 118 tKuwajt tUniredArab Emirates LjnitedKingdom Australia Italy 100 88 103 110 100 108 113 107 99 . . 105 90 103 112 100 112 112 108 98 . . 76 49 85 72 70 . . 94 94 86 82 84 55 82 70 73 . 84 91 83 81 . 11 19 25 27 3 . 11 37 42 37 ........b ........b 97 93 100 94 94 98 25 20 b 18 21 h b b 119 Canada 99 104 99 106 89 103 87 104 52 103 97 98 94 95 b b 120 Finland 96 100 97 100 105 130 94 110 32 63 99 98 99 98 b b 121 tHongKong 106 . . 107 . . 65 . . 63 . . 10 21 . . 12 4 122 Sweden 97 100 96 100 93 100 83 99 31 38 100 99 b b 123 124 125 Netherlands Belgium France 101 103 110 99 100 105 99 104 112 96 99 107 90 92 92 120 104 107 95 90 77 126 103 104 29 26 25 45 50 ........b 100 81 97 78 b b b b b ........b 99 126 tSingapore 106 . 109 . . 59 . . 56 . . 8 . 100 100 100 14 4 127 Austria 98 103 99 103 87 104 98 09 22 43 97 99 92 97 b b 128 United States 100 106 101 107 . . 97 . . 98 56 81 b 129 Germany . . 98 .. 97 .. 100 .. 101 .. 36 98 99 96 97 b b 130 Denmark 95 98 96 97 104 115 105 112 28 41 . b b 131 Norway 100 99 100 99 96 114 92 118 26 54 . b b 132 Japan 101 102 101 102 94 97 92 95 31 30 100 100 100 100 b b 133 Switzerland . . 102 . . 100 . . 89 . . 93 18 31 94 . . 92 . b b World 90w 99w 104w 109w 39w 57w 50w .. 13w a. Data for 1980 include Erirrea. b. According to UNESCO, illiteracy is less than 5 percent. c. Data before 1990 refer to the Federal Rcpuhlic of Germany before unification. 228 WORLD DEVELOPMENT REPORT 1997 Table 8. Commercial energy use Net Energy use (oil equivalent) Carbon dioxide emissionsa energy imports Total Per capita Avg. annual GDP per kg. as % of energy Total Per capita (thous. metuic tons) (kg) growth rate 1%) (1987 $1 consumption (mill, metric tons) (metric tons) 1980 1994 1980 1994 1980-94 1980 1994 1980 1994 1980 1992 1980 1992 Low-income economies 587,124 t 1,154,712 t 248w 369w 4.4w 0.9w 1.1 w 2,063 t 3,880t 0.9w 1.3w Exduding China and India 80,087 t 137,034 t 114w 134w 2.3w .. 2.6w 223 t 443 t 0.3w 0.5w 1 Mozambique 1,123 619 93 40 -2.5 1.4 3.3 -15 74 3 1 0.3 0.1 2 Ethiopia 624 1,193 17 22 5.2 .. 6.9 91 87 2 3 0.0 0.1 3 Tanzania 1,023 975 55 34 0.7 . . 4.5 92 83 2 2 0.1 0.1 4 Burundi 58 143 14 23 6.8 13.9 8.3 98 97 0 0 0.0 0.0 5 Malawi 334 370 54 39 1.5 3.2 3.4 70 59 1 1 0.1 0.1 6 Chad 93 100 21 16 0.6 6.2 10.9 100 100 0 0 0.0 0.0 7 Rwanda 190 209 37 34 -0.4 9.3 4.9 85 78 0 0 0.0 0.1 8 Sierra Leone 310 323 96 77 0.5 2.3 2.4 100 100 1 0 0.2 0.1 9 Nepal 174 582 12 28 8.7 12.5 7.3 91 88 1 1 0.0 0.1 10 Niger 210 327 38 37 2.1 12.1 7.3 93 83 1 1 0.1 0.1 11 BurkinaFaso 144 160 21 16 1.1 11.2 16.0 100 100 0 1 0.1 0.1 12 Madagascar 391 479 45 36 1.7 6.7 5.6 90 83 2 1 0.2 0.1 13 Bangladesh 2,809 7,566 32 64 7.7 4.5 3.1 60 28 8 17 0.1 0.2 14 Uganda 320 425 25 23 1.3 .. 22.6 52 58 1 1 0.1 0.1 15 Vietnam 4,024 7,267 75 101 3.1 . . 7.5 32 -55 17 22 0.3 0.3 16 Guinea-Bissau 31 39 38 37 2.1 3.8 5.8 100 100 0 0 0.2 0.2 17 Haiti 240 200 45 29 1.2 9.5 7.9 77 93 1 1 0.1 0.1 18 Mali 164 205 25 22 1.8 11.2 11.5 87 80 0 0 0.1 0.0 19 Nigeria 9,879 17,503 139 162 3.8 3.1 2.2 -968 -484 68 97 1.0 0.9 20 Yemen, Rep. 1,364 3,044 160 206 4.1 . . .. 100 463 3 10 0.4 0.7 21 Cambodia 393 512 60 52 2.2 .. 2.4 97 96 0 0 0.0 0.1 22 Kenya 1,991 2,872 120 110 3.1 3.1 3.3 95 83 6 5 0.4 0.2 23 Mongolia 1,943 2,550 1,168 1,058 2.0 1.2 1.2 38 15 7 9 4.0 4.0 24 logo 195 183 75 46 1.8 6.3 69 99 100 1 1 0.2 0.2 25 Gambia, The 53 60 83 56 0.9 3.5 4.9 100 100 0 0 0.2 0.2 26 Central African Republic 59 93 26 29 2.7 16.2 12.1 71 76 0 0 0.0 0.1 27 India 93,907 226,638 137 248 6.6 1.9 1.6 21 21 350 769 0.5 0.9 28 Lao PDR 107 182 33 38 2.6 . . 9.1 -121 -18 0 0 0.1 0.1 29 Benin 149 107 43 20 -3.5 8.3 18.0 93 -194 0 1 0.1 0.1 30 Nicaragua 756 1,273 270 300 3.3 5.1 2.7 83 63 2 2 0.7 0.6 31 Ghana 1,303 1,542 121 93 2.5 3.6 4.4 57 66 2 4 0.2 0.2 32 Zambia 1,685 1,296 294 149 -2.6 1.3 1.8 32 31 4 2 0.6 0.3 33 Angola 937 931 133 89 0.3 . . 7.0 -722 -2,576 5 5 0.8 0.5 34 Georgia . . 3,325 .. 614 . . .. 0.7 .. 85 . . 14 .. 2.5 35 Pakistan 11,698 32,133 142 254 7.4 1.8 1.5 38 40 32 72 0.4 0.6 36 Mauritania 214 229 138 103 0.5 3.8 4.8 100 100 1 3 0,4 1.4 37 Azerbaijan 15,001 16,274 2,433 2,182 -2.7 . . 0.2 1 1 . . 64 . . 8.7 38 Zimbabwe 2,797 4,722 399 438 3.9 1.5 1.4 28 24 10 19 1.4 1.8 39 Guinea 356 418 80 65 1.3 . . 6.1 89 86 1 1 0.2 0.2 40 Honduras 843 1,173 230 204 2.1 4.2 4.4 76 82 2 3 0.6 0.6 41 Senegal 875 803 158 97 -0.1 4.2 6.3 100 100 3 3 0.5 0.4 42 China 413,130 791,040 421 664 5.0 0.3 0.7 -4 -I 1,489 2,668 1.5 2.3 43 Cameroon 774 1,335 89 103 2.6 10.0 6.9 -269 -333 4 2 0.4 0.2 44 COted'Ivoire 1,435 1,406 175 103 1.4 6.8 6.8 87 70 5 6 0.6 0.5 45 Albania 3,058 1,093 1,145 341 -4.8 0.6 2.4 0 3 7 4 2.8 1.2 46 Congo 262 847 157 331 3.7 5.7 2.8 -1,193 -1,013 0 4 0.2 1.6 47 Kyrgyz Republic .. 2,755 . . 616 . . . . 0.9 .. 47 . . 15 . . 3.4 48 Sri Lanka 1,411 1,728 96 97 1.9 3.4 5.1 91 80 3 5 0.2 0.3 49 Armenia 1,071 1,441 346 384 1.6 4.3 1.4 79 . . 4 . . 1.1 Middle-income economies 1,873,142 t 2,313,337t 1,537w 1,475w -1.7w 2,831 t 7,221 t 2.9w 4.8w Lower-middle-income 1,448,776 t 1,647,009 t 1,632w 1,449w -3.0w .. 1,664 t 5,565 t 2,6w 5.1w 50 Lesotho . . .. . . . . . . . . . . . . . . . 0 . . 0.0 51 Egypt, Arab Rep. 15,176 34,071 371 600 5.7 1.6 1.2 -120 -79 45 84 1.1 1.5 52 Bolivia 1,713 2,698 320 373 1.6 2.7 2.1 -107 -61 5 7 0.8 1.0 53 Macedonia, FYR 2,686 .. 1,279 .. . . 44 . . 4 . . 2.0 54 Moldova 4,763 1,095 .. .. .. 99 0 14 0.0 3.3 55 Uzbekistan .. 41,825 . . 1,869 . . . . 0.3 . . 0 . . 123 . . 5.7 56 Indonesia 25,028 69,740 169 366 8.3 2.1 1.8 -275 -120 95 185 0.6 1.0 57 Philippines 13,406 21,199 277 316 3.5 2.4 1.9 79 71 37 50 0.8 0.8 58 Morocco 4,927 8,509 254 327 4.1 3.1 2.9 87 95 16 27 0.8 1.1 59 Syrian Arab Republic 5,343 13,675 614 997 6.1 1.9 1.2 -78 -130 19 42 2.2 3.3 60 Papua New Guinea 705 990 228 236 2.4 3.9 4.8 89 -150 2 2 0.6 0.6 61 Bulgaria 28,476 20,568 3,213 2,438 -2.7 0.7 1.0 74 56 75 54 8.4 6.4 62 Kazakstan 76,799 56,664 5,153 3,371 -2.3 . . 0.3 0 -25 . . 298 . . 17.6 63 Guatemala 1,443 2,165 209 210 3.2 5.0 4.3 84 74 4 6 0.6 0.6 64 Ecuador 4,209 6,345 529 565 2.7 2.3 2.2 -156 -231 13 19 1.7 1.8 65 Dominican Republic 2,083 2,591 366 337 1.4 2.0 2.5 93 94 6 10 1.1 1.4 66 Romania 63,846 39,387 2,876 1,733 -3.1 0.5 0.7 19 27 191 122 8.6 5.4 67 Jamaica 2,169 2,703 1,017 1,083 2.3 1.3 1.5 99 100 8 8 4.0 3.3 68 Jordan 1,710 4,306 784 1,067 5.2 . . 1.5 100 96 5 11 2.2 3.0 69 Algeria 12,078 24,834 647 906 4.9 4.1 2.6 -452 -318 66 79 3.5 3.0 70 El Salvador 1,000 2,032 220 370 4.0 4.5 2.6 63 70 2 4 0.5 0.7 71 Ukraine 108,290 165,132 2,164 3,180 -1.4 . . 0.4 -I 48 . . 611 . . 11.7 72 Paraguay 550 1,402 175 299 6.9 6.0 3.5 88 -123 1 3 0.5 0.6 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 229 Net Energy use (oil equivalent) Carbon dioxide emissions energy imports Total Per capita Avg. annual GDP per kg. as % of energy Total Per capita (thous. metric tons) (kg) growth rate 1%) (1987 $1 consumption Imill. metric tons) (metric tons) 1980 1994 1980 1994 1980-94 1980 1994 1980 1994 1980 1992 1980 1992 73 Tunisia .. 7,555 .. 595 . .. 0.8 . . 70 .. 22 .. 5.9 74 Lithuania 13,972 22,470 501 2,030 3.5 2.1 2.1 7 -99 39 61 1.4 1.8 75 Colombia . . . . . . 622 .. . . . . .. . . 0 . . 0.0 76 Namibia 2,385 24,772 247 . . 13.4 . . 0.8 -8 88 102 9.9 77 Belarus 750,240 595,440 5,397 2,392 -2.6 0.6 0.5 0 -53 . . 2,103 . . 14.1 78 Russian Federation 3,997 . . 4,014 . 1.2 . . 90 - - 15 - - 5.6 79 Latvia 8,139 8,555 471 1,569 -0.2 2.5 2.7 -36 0 24 22 1.4 1.0 80 Peru 1,292 1,843 566 367 3.5 3.1 3.4 86 67 2 4 1.1 1.2 81 Costa Rica 2,376 3,790 840 558 2.3 97 98 6 II 2.2 2.9 82 Lebanon 12,093 44,395 259 964 13.4 2.8 2.2 96 61 40 112 0.9 2,0 83 Thailand 1,376 1,597 703 769 1.2 3.2 3.9 97 87 4 4 1.9 1.7 84 Panama 31,314 57,580 705 618 4.4 1.9 1.8 45 53 76 145 1.7 2.5 85 Turkey 124,500 92,537 3,499 957 -2.0 0.5 0.7 3 -2 460 342 12.9 8.9 86 Poland 5,560 . . 2,401 0.7 39 0 21 0.3 13.5 87 Estonia 17,343 . . 3,709 0.9 .. 72 - . 37 . . 7.0 88 Slovak Republic 35,011 46,300 2,354 3,243 2.3 1,3 1.2 -280 -269 90 116 6.0 5.7 89 Borswana 424,366 666,328 1,282 387 2.6 2.3 1.7 - . . - 1,167 1,656 3.7 4.0 90 Venezuela 384 549 426 2,186 2.6 2.1 4.7 32 55 1 2 1,1 1.6 Upper-middle-income 424,366t 666,328 t 1,282w 1,544w 2.6w 2 1,167t 1,656 3.7w 4,0w 91 South Africa 60,511 86,995 2,074 2,146 2.1 1.2 1.0 -14 -35 213 290 7.3 7.5 92 Croatia . - 6,667 - - 1,395 .. .. 43 . . 16 - - 3.4 93 Mexico 97,434 140,840 1,464 1,561 2.6 1.3 1.2 -49 -48 260 333 3.9 3.8 94 Mauritius 339 431 351 387 2.8 3.7 6.3 94 92 1 1 0.6 1.3 95 Gabon 759 692 1,098 652 -0.4 5.0 5.5 -1,106 -2,212 5 6 6.9 5.5 96 Brazil 72,141 112,795 595 718 3.9 3.4 2.8 65 39 184 217 1.5 1.4 97 Trinidad and Tobago 3,863 6,935 3,570 5,436 2.9 1.5 0.7 -240 -87 17 21 15.4 16.5 98 Czech Republic 29,394 39,982 2,873 3,868 45.2 . . 0.8 -29 7 .. 136 . . 13.1 99 Malaysia 9,522 33,410 692 1,699 10.0 2,4 1.7 -58 -71 28 70 2.0 3.8 100 Hungary 28,322 24,450 2,645 2,383 -0.9 0.8 1,0 49 47 82 60 7.7 5.8 101 Chile 7,743 14,155 695 1,012 5.0 2.3 2.3 50 68 27 35 2,4 2.6 102 Oman 1,346 5,018 1,223 2,392 9.1 2.9 2.4 -1,024 -787 6 10 5.3 5.3 103 Uruguay 2,208 1,971 758 622 3.2 3.4 4.6 89 67 6 5 2.0 1.6 104 SaudiArabia 35,496 83,772 3,787 4,566 5.5 2.7 1.1 -1,361 -463 131 221 14.0 13.1 105 Argentina 39,669 51,405 1,411 1,504 1.6 2.8 2.7 8 -18 107 117 3.8 3.5 106 Slovenia . - 5,195 - - 2,612 Si . . 6 . . 2.8 107 Greece 15,973 23,560 1,656 2,260 2,8 2.2 77 62 51 74 5.3 7.2 Low- and middle-income 2,460,266 t 3,,#,049t 686w 739w -0.1w 1.4w l.lw 4,893 t 11,101 t 1.5w 2.4w Sub-SaharanAfrica 94,721 t 133,471t 249w 237w 1.2w 2.2w 2.0w 353t 472t 0.9w 0.9w East Asia and Pacific 514,066t 1,000,586 t 378w 593w 4.8w 0.7w 0.9w 1,846 t 3,378t 1.4w 2.1w South Asia 110,906 t 271,293 t 123w 222w 6.4w 2.0w 1.7w 395 t 866 t 0.4w 0.7w Europe and Central Asia 1,279,071 r 1,288,624 t 3,105w 2,647w -415w .. 0.6w 944 t 4,506 t .. 9.3w MiddleEastandN.Africa 143,540t 323,064t 825w 1,220w 4.8w 3.2w 1.7w 500t 849t 2.9w 3.4w LatinAinericaandCaribbean 317,962t 451,011t 888w 960w 2.9w 2.3w 2.0w 855t 1,029t 2.4w 2.3w High-income economies 3,789,479 t 4,543,482 t 4,644w 5,066w 1.9w 2.9w 3.4w 9,877 t l0,246t 12.4w 11.9w 108 Korea, Rep. 41,426 132,538 1,087 2,982 9.5 1.8 1.8 77 86 126 290 3.3 6.6 109 Portugal 10,291 18,090 1,054 1,827 4.7 3.5 2.8 86 88 27 47 2.8 4.8 Ill) Spain 68,692 96,200 1,837 2,458 2.8 3.6 3.6 77 69 200 223 5.4 5.7 111 NewZealand 9,202 15,070 2,956 4,245 4.1 3.4 2.8 39 15 18 26 5.7 7.6 112 Ireland 8,485 11,200 2,495 3,137 2.2 3.1 3.9 78 68 25 31 7.4 8.7 113 flstael 8,616 14,624 2,222 2,717 4.9 3.4 3.7 98 96 21 42 5.4 8.1 114 tKuwait 9,500 13,968 6,909 8,622 0.3 2.7 2.0 739 693 25 16 18.0 11.2 115 5 United Arab Emirates 8,558 25,137 8,205 10,531 6.3 3.6 - . -996 -454 36 71 34.8 33.9 116 United Kingdom 201,200 220,270 3,572 3,772 0.8 2.8 3.5 2 -10 588 566 10.4 9.8 117 Australia 70,399 95,280 4,792 5,341 2.3 2.4 2.6 -22 -83 203 268 13.8 15.3 118 Italy 139,190 154,600 2,466 2,707 1.4 4.8 5.5 86 81 372 408 6.6 7.2 119 Canada 193,170 229,730 7,854 7,854 1.5 1.7 2.0 -7 -47 430 410 17.5 14.4 120 Finland 24,998 30,520 5,230 5,997 1.7 2.9 3.0 72 58 55 41 11.5 8.2 121 tHong Kong 5,628 13,243 1,117 2,185 6.4 5.3 5.3 100 100 16 29 3.3 5.0 122 Sweden 40,992 50,250 4,933 5,723 1.3 3.4 3.3 61 38 71 57 8.6 6.6 123 Netherlands 65,106 70,440 4,601 4,580 1.3 3.0 3.7 10 7 153 139 10.8 9.2 124 Belgium 46,122 51,790 4,684 5,120 1.5 2.8 3.2 83 78 128 102 13.0 10.1 125 France 190,660 234,160 3,539 4,042 2.0 4.1 4.4 75 95 484 362 9.0 6.3 126 tSirrgapore 6,049 23,743 2,651 8,103 9.9 2.2 1.6 100 100 30 50 13.2 17.7 127 Austria 23,449 26,500 3,105 3,301 1.6 4.6 5.4 67 66 52 57 6.9 7.2 128 United States 1,801,000 2,037,980 7,908 7,819 1.6 2.1 2.6 14 19 4,623 4,881 20.3 19.1 129 Germany 359,170 336,490 4,587 4,128 -0.1 . . , . 49 58 1,068 878 13.6 10.9 130 Denmark 19,488 20,700 3,804 3,977 0.8 4.4 5.5 97 28 63 54 12.3 10.4 131 Norway 18,865 23,060 4,611 5,318 1.6 3.9 4.6 -195 -638 40 60 9.8 14.1 132 Japan 347,120 481,850 2,972 3,856 2.8 5.5 6.2 88 81 934 1,093 8.0 8.8 133 Switzerland 20,840 25,380 3.298 3,629 1.7 7.3 7.4 66 57 41 44 6.5 6.4 World 6,249,71St 8.O/1.531t 1,419w 1,433w 1.0w 2.3w 2.4w 14,770t 21,347t 3.6w 4.Ow a. From industrial processes. 230 WORLD DEVELOPMENT REPORT 1q97 Table 9. Land use and urbanization Land use (% of total land areal Urban population Population in urban agglomerations Permanent As % of total Avg. annual of 1 million or more as % of Cropland pasture Other population growth rate 1%) Urban Total 1980 1994 1980 1994 1980 1994 1980 1995 1980-95 1980 1995 1980 1995 Low-income economies 12w 12w 31 w 32w 57w 55w 21w 29w 4.0w 32w 34w 7w lOw Excluding China and India 8w 8w 32w 32w 60w 60w 21 w 28w 4.6w 28w 31 w 6w 9w 1 Mozambique 4 4 56 56 40 40 13 38 8.5 48 36 6 14 2 Ethiopia . . 11 .. 20 . . 69 11 13 4.5 30 29 3 4 3 Tanzania 3 4 40 40 57 56 15 24 6.7 30 24 5 6 4 Burundi 46 46 39 39 15 15 4 8 6.8 0 0 0 0 5 Malawi 14 18 20 20 66 62 9 13 6.0 0 0 0 0 6 Sierra Leone 7 8 31 31 62 62 25 39 4.9 0 0 0 0 7 Rwanda 41 47 28 28 30 24 5 8 4.7 0 0 0 0 8 Chad 3 3 36 36 62 62 19 21 3.4 0 0 0 0 9 Nepal 17 17 14 15 69 68 7 14 7.8 0 0 0 0 10 Niger 3 3 8 8 90 89 13 23 7.2 0 0 0 0 11 Madagascar 5 5 41 41 54 53 18 27 5.7 0 0 0 0 12 Burkina Faso 10 13 22 22 68 65 9 27 11.3 0 0 0 0 13 Vietnam 20 21 1 1 79 78 19 21 2.7 27 31 5 7 14 Bangladesh 70 74 5 5 25 21 11 18 5.6 46 47 5 9 15 Uganda 28 34 9 9 63 57 9 12 5.2 0 0 0 0 16 Mali 2 2 25 25 74 73 19 27 5.3 0 0 0 0 17 Guinea-Bissau 10 12 38 38 51 50 17 22 3.7 0 0 0 0 18 Haiti 32 33 18 18 49 49 24 32 3.9 55 56 13 18 19 Nigeria 33 36 44 44 23 20 27 39 5.6 23 27 6 11 20 Yemen, Rep. 3 3 30 30 67 67 20 34 7.6 0 0 0 0 21 Cambodia 12 22 3 8 85 70 12 21 6.7 . . . . 22 Kenya 8 8 37 37 55 55 16 28 7.0 32 28 5 8 23 Togo 43 45 4 4 53 52 23 31 5.1 0 0 0 0 24 Mongolia 1 1 79 75 20 24 52 60 3.6 0 0 0 0 25 Gambia, The 16 17 19 19 65 64 18 26 6.3 0 0 0 0 26 India 57 57 4 4 39 39 23 27 3.1 25 35 6 10 27 Central African Republic 3 3 5 5 92 92 35 39 3.1 0 0 0 0 28 Lao PDR 3 4 3 3 94 93 13 22 6.3 0 0 0 0 29 Benin 16 17 4 4 80 79 32 42 5.1 0 0 0 0 30 Nicaragua 10 10 40 45 50 44 53 62 3.9 42 44 23 27 31 Ghana 15 19 37 37 48 44 31 36 4.3 30 27 9 10 32 Zambia 7 7 40 40 53 53 40 45 4.0 23 33 9 15 33 Angola 3 3 43 43 54 54 21 32 5.9 63 64 13 20 34 Georgia 17 16 39 24 44 60 52 58 1.3 42 43 22 25 35 Pakistan 26 28 6 6 67 66 28 35 4.6 39 53 11 18 36 Mauritania 0 0 38 38 62 62 29 54 6.8 0 0 0 0 37 Azerbaijan 22 23 27 25 51 52 53 56 1.7 48 44 26 25 38 Zimbabwe 7 7 44 44 49 48 22 32 5.7 0 0 0 0 39 Guinea 3 3 44 44 54 53 19 30 5.8 65 77 12 23 40 Honduras 16 18 13 14 71 68 36 48 5.2 0 0 0 0 41 Senegal 12 12 30 30 58 58 36 42 4.0 49 55 18 23 42 China 11 10 36 43 53 47 19 30 4.2 41 35 8 11 43 Cameroon 15 15 4 4 81 81 31 45 5.3 19 22 6 10 44 Côte d'Ivoire 10 12 41 41 49 47 35 44 5.2 44 46 15 20 45 Albania 26 26 15 15 59 59 34 37 2.1 0 0 0 0 46 Congo 0 0 29 29 70 70 41 59 5.6 0 0 0 0 47 Sri Lanka 29 29 7 7 64 64 22 22 1.6 0 0 0 0 48 Kyrgyz Republic 8 7 47 44 45 48 38 39 1.6 0 0 0 0 49 Armenia . S .. . . . 66 69 1.6 51 51 34 35 Middle-income economies 9w lOw 28w 23w 62w 67w 52w 60w 2.8w 31 w 33w l6w 20w Lower-middle-income lOw 11 w .. 18w .. 71 w 48w 56w 2.8w 28w 30w 13w 17w 50 Lesotho 10 11 66 66 24 24 13 23 6.5 0 0 0 0 51 Egypt,Arab Rep. 2 4 0 0 98 96 44 45 2.5 52 51 23 23 52 Bolivia 2 2 25 24 73 73 46 58 3.9 30 29 14 17 53 Macedonia, FYR . . 26 25 . . 49 53 60 1.5 0 0 0 0 54 Moldova 67 66 11 13 23 21 40 52 2.4 0 0 0 0 55 Uzbekistan 10 11 57 50 33 39 41 42 2.5 28 24 11 10 56 Indonesia 14 17 7 7 79 77 22 34 4.8 33 39 7 13 57 Philippines 29 31 3 4 67 65 38 53 4.9 33 25 12 14 58 Morocco 18 21 47 47 35 32 41 49 3.3 26 37 11 18 59 Syrian Arab Republic 31 30 46 45 24 25 47 53 4.1 60 52 28 28 60 Papua New Guinea 1 1 0 0 99 99 13 16 3.6 0 0 0 0 61 Bulgaria 38 38 18 16 44 46 61 71 0.6 20 23 12 16 62 Kazakstan 13 13 70 70 16 17 54 60 1.6 12 13 6 8 63 Guatemala 16 18 12 24 72 58 37 42 3.6 0 0 0 0 64 Ecuador 9 11 15 18 77 71 47 58 3.9 29 44 14 26 65 Dominican Republic 29 31 43 43 27 26 51 65 3.8 49 51 25 33 66 Romania 46 43 19 21 35 36 49 55 1.0 18 17 9 9 67 Jordan 4 5 9 9 87 87 60 72 5.8 49 39 29 28 68 Jamaica 22 20 24 24 54 56 47 55 2.2 0 0 0 0 69 Algeria 3 3 15 13 82 83 43 56 4.5 25 24 11 13 70 El Salvador 35 35 29 29 36 35 42 45 2.0 0 0 0 0 71 Ukraine 61 59 12 13 27 28 62 70 1.0 22 22 14 15 72 Paraguay 4 6 40 55 56 40 42 54 4.7 0 0 0 0 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 231 Land use 1% of total land area) Urban population Population in urban agglomerations Permanent As % of total Avg. annual of 1 million or more as % of Cropland pasture Other population growth rate 1%) Urban Total 1980 1994 1980 1994 1980 1994 1980 1995 1980-95 1980 1995 1980 1995 73 Tunisia 30 32 22 20 48 48 51 57 3.0 34 40 17 23 74 Lithuania 49 47 8 7 43 46 61 72 1.8 0 0 0 0 75 Colombia 5 5 37 39 58 56 64 73 2.7 34 38 22 28 76 Namibia 1 1 46 46 53 53 23 38 6.2 0 0 0 0 77 Belarus 31 31 16 14 53 55 56 71 2.0 24 24 14 17 78 Russian Federation 8 8 . . 5 . . 87 70 73 0.8 23 26 16 19 79 Latvia 28 28 12 13 60 59 68 73 0.5 0 0 0 0 80 Peru 3 3 21 21 76 76 65 72 2.9 40 44 26 31 81 Costa Rica 10 10 39 46 51 44 43 50 3.7 0 0 0 0 82 Lebanon 30 30 1 1 69 69 73 87 3.7 83 Thailand 36 41 1 2 63 58 17 20 2.6 59 56 10 11 84 Panama 7 9 17 20 75 71 50 56 2.8 0 0 0 0 85 Turkey 37 36 13 16 50 48 44 70 5.4 39 35 17 24 86 Poland 49 48 13 13 38 39 58 65 1.2 31 28 18 18 87 Estonia 24 27 8 7 68 66 70 73 0.4 0 0 0 0 88 Slovak Republic 41 34 13 17 45 49 52 59 1.3 0 0 0 0 89 Venezuela 4 4 20 20 76 75 83 93 3.3 20 29 16 27 90 Botswana 1 1 45 45 54 54 15 31 8.4 0 0 0 0 Upper-middle-income 7w 7w 30w 32w 63w 60w 64w 73 w 2.8 w 38w 38w 24w 28w 91 SouthAfrica 11 11 67 67 22 23 48 51 2.7 23 38 11 19 92 Croatia 29 22 28 20 42 59 50 64 2.0 0 0 0 0 93 Mexico 13 13 39 39 48 48 66 75 3.1 41 37 27 28 94 Mauritius 53 52 3 3 44 44 42 41 0.6 0 0 0 0 95 Gabon 2 2 18 18 80 80 36 50 5.4 0 0 0 0 96 Brazil 6 6 20 22 74 72 66 78 3.0 42 42 27 33 97 Trinidad and Tobago 23 24 2 2 75 74 63 68 1.7 0 0 0 0 98 Czech Republic 41 44 13 12 45 45 64 65 0.2 18 18 12 12 99 Malaysia 15 23 1 1 85 76 42 54 4.3 16 11 7 6 100 Hungary 58 54 14 12 28 34 57 65 0.5 34 30 19 20 101 Chile 6 6 17 18 77 76 81 86 2.0 41 41 33 36 102 Oman 0 0 5 5 95 95 8 13 8.6 0 0 0 0 103 Uruguay 8 7 78 77 14 15 85 90 1.0 49 46 42 42 104 SaudiArabia 1 2 40 56 60 42 67 79 6.0 28 27 19 21 105 Argentina 10 10 52 52 38 38 83 88 1.8 42 44 35 39 106 Slovenia . . 14 . . 25 . . 61 48 64 2.2 0 0 0 0 107 Gteece 30 27 41 41 29 32 58 65 1.4 54 54 31 35 Low-andmiddle-income 10w 11w 29w 27w 60w 63w 32w 39w 3.3w 32w 33w lOw 13w Sub-Saharan Africa 6w 7w 34w 34w 60w 59w 23w 31 w 5.0w 23w 26w 5w 8w East Asia and Pacific 11 w 12w 30w 34w 59w 54w 21 w 31 w 4.2w 37w 34w 8w 11 w SouthAsia 44w 45w 11w lOw 45w 45w 22w 26w 3.4w 27w 38w 6w lOw EuropeandCentralAsia 13w 13w .. 16w .. 71w 58w 65w 1.6w 24w 25w 14w 16w Middle East and N. Africa 5w 6w 21 w 24w 74w 70w 48w 56w 4.2w 36w 36w 17w 20w Latin America and Caribbean 7w 7w 28w 29w 65 w 63 w 65 w 74w 2.8w 37w 38w 24w 28 w High-income economies 12w 12w 25w 24w 62w 63w 75w 75w 0.7w 41 w 43w 31 w 33w 108 Korea, Rep. 22 21 1 1 77 78 57 81 3.5 65 64 37 52 109 Portugal 34 32 9 11 57 58 29 36 1.3 46 53 13 19 110 Spain 41 40 22 21 37 38 73 77 0.6 27 23 20 18 Ill NewZealand 13 14 53 50 34 35 83 84 1.0 0 0 0 0 112 Ireland 16 19 67 45 17 36 55 58 0.5 0 0 0 0 113 tlsrael 20 21 6 7 74 72 89 . . . . 41 . . 37 35 114 5 Kuwait 0 0 8 8 92 92 90 97 0.9 67 68 60 66 115 t United Arab Emirates 0 0 2 2 97 97 72 84 5.8 0 0 0 0 116 United Kingdom 29 25 47 46 24 29 89 90 0.3 28 26 25 23 117 Australia 6 6 57 54 37 40 86 85 1.3 55 68 47 58 118 Italy 42 38 17 15 40 47 67 66 0.1 39 30 26 20 119 Canada 5 5 3 3 92 92 76 77 1.4 38 47 29 36 120 Finland 8 9 1 0 91 91 60 63 0.8 0 0 0 0 121 tHong Kong 7 7 1 1 92 92 92 95 1.5 100 95 91 90 122 Sweden 7 7 2 1 91 92 83 83 0.4 20 21 17 17 123 Netherlands 24 28 35 31 41 41 88 89 0.6 8 16 7 14 124 Belgium . . 24 . . 21 . . 55 95 97 0.3 13 II 12 11 125 France 34 35 23 19 42 45 73 73 0.5 29 28 21 21 126 tSingapore 13 2 0 0 87 98 100 100 1.8 106 95 106 95 127 Austria 20 18 25 24 56 57 55 56 0.5 49 46 27 26 128 United States 21 21 26 26 53 53 74 76 1.2 49 51 36 39 129 Germany 36 34 17 15 47 51 83 87 0.6 46 47 38 41 130 Denmark 63 56 6 7 31 37 84 85 0.2 32 30 27 25 131 Norway 3 3 0 0 97 97 71 73 0.7 0 0 0 0 132 Japan 13 12 2 2 85 87 76 78 0.6 44 48 34 37 133 Switzerland 10 II 40 29 49 60 57 61 1.2 0 0 0 0 World 11w 11w 28w 26w 60w 63w 40w 45w 2.5w 35w 36w 14w 16w 232 WORLD DEVELOPMENT REPORT 1997 Table 10. Forest and water resources Forest areas Annual freshwater withdrawal, 1995b Ann. deforestation, 1980-90 Nationally protected areas, l994 Total area As % of (thousand sq. kml Thousand As % of Total total water Per capita (Cu. ml Thousand % 1990 sq. km change sq. km Number land area lcu. kml resources Domestic Other Low-income economies 7,916 t 65.5 t 0.8w 2,001.1 t 1.842 t .2 w Excluding China and India 6,152 r 53.3 t 0.8w 1,276.9 t 795 t 4.9w 1Mozambique 173 1.4 0.8 0.0 1 0.0 0.6 0.4° 13 42 2 Ethiopia 142 0.4 0.3 60.2 23 6.0 2.2 2.0 6 45 3 Tanzania 336 4.4 1.2 139.4 31 15.8 1.2 1.3° 7 28 4 Burundi 2 0.0 0.6 0.9 3 3.5 0.1 2.8 7 13 5 Malawi 35 0.5 1.4 10.6 9 11.3 0.9 5.0° 7 13 6 Chad 114 0.9 0.7 114.9 9 9.1 0.2 0.4° 6 29 7 Rwanda 2 0.0 0.2 3.3 2 13.3 0.8 12.2 6 18 8 Sierra Leone 19 0.1 0.6 0.8 2 1.1 0.4 0.2 7 92 9 Nepal 50 0.5 1.0 11.1 12 8.1 2.7 1.6 6 144 10 Niger 24 0.1 0.4 84.2 5 6.6 0.5 0.9° 9 33 11 Burkina Faso 44 0.3 0.7 26.6 12 9.7 0.4 1.4 5 13 12 Madagascar 158 1.3 0.8 11.2 36 1.9 16.3 4.8° 16 1,568 13 Bangladesh 8 0.4 4.1 1.0 8 0.7 22.5 1.0° 7 213 14 Uganda 63 0.6 1.0 19.1 31 9.6 0.2 0.3° 7 14 15 Vietnam 83 1.4 1.5 13.3 52 4.1 28.9 7.7 54 361 16 Guinea-Bissau 20 0.2 0.8 . . . . . . 0.0 0.0° 3 8 17 Haiti 0 0.0 5.1 0.1 3 0.4 0.0 0.4 2 5 18 Mali 121 1.1 0.8 40.1 11 3.3 1.4 1.4° 3 159 19 Nigeria 156 1.2 0.7 29.7 19 3.3 3.6 1.3 13 28 20 Yemen, Rep. 41 0.0 0.0 0.0 0 0.0 3.4 136.0 17 318 21 Cambodia 122 1.3 1.0 30.0 20 17.0 0.5 0.1 3 61 22 Kenya 12 0.1 0.6 35.0 36 6.2 2.1 7.0° 14 37 23 Mongolia 139 1.3 0.9 61.7 15 3.9 0.6 2.2 30 243 24 Togo 14 0.2 1.5 6.5 11 11.9 0.1 0.8° 17 11 25 Gambia, The 1 0.0 0.8 0.2 5 2.3 0.0 0.3° 2 27 26 Central African Republic 306 1.3 0.4 61.1 13 9.8 0.1 0.0 5 20 27 India 517 3.4 0.6 143.4 339 4.8 380.0 18.2° 18 594 28 Lao PDR 132 1.3 0.9 . . . . . . 1.0 0.4 21 239 29 Benin 49 0.7 1.3 7.8 2 7.0 0.1 0.4° 7 19 30 Nicaragua 60 1.2 1.9 9.0 59 7.4 0.9 0.5 92 275 31 Ghana 96 1.4 1.4 11.0 9 4.9 0.3 0.6° 12 23 32 Zambia 323 3.6 1.1 63.6 21 8.6 1.7 1.5° 54 32 33 Angola 231 1.7 0.7 26.4 5 2.1 0.5 0.3 8 49 34 Georgia 28 0.2 0.7 1.9 15 2.7 4.0 6.5 156 586 35 Pakistan 19 0.8 3.5 37.2 55 4.8 153.4 32.8° 21 2,032 36 Mauritania 6 0.0 0.0 17.5 4 1.7 1.6 14.0° 59 436 37 Azerbaijan 10 0.1 1.3 1.9 12 2.2 15.8 56.4° 90 2,158 38 Zimbabwe 89 0.6 0.7 30.7 25 7.9 1.2 6.1° 19 117 39 Guinea 67 0.9 1.2 1.6 3 0.7 0.7 0.3 14 126 40 Honduras 46 1.1 2.2 8.6 43 7.7 1.5 2.1° 12 282 41 Senegal 75 0.5 0.7 21.8 9 11.3 1.4 3.5° 10 191 42 China 1,247 8.8 0.7 580.8 463 6.2 460.0 16.4 28 433 43 Cameroon 204 1.2 0.6 20.5 14 4.4 0.4 0.1 17 20 44 COre d'lvoire 109 1.2 1.0 19.9 12 6.3 0.7 0.9 15 52 45 Albania 14 0.0 0.0 0.3 11 1.2 0.2 0.9° 6 88 46 Congo 199 0.3 0.2 11.8 10 3.4 0.0 0.0° 12 7 47 Kyrgyz Republic 7 0.1 1.2 2.8 5 1.5 11.7 24.0 82 2,647 48 Sri Lanka 17 0.3 1.4 8.0 56 12.3 6.3 14.6 10 493 49 Armenia 3 0.2 3.9 2.1 4 7.6 3.8 45.8° 149 996 Middle-income economies 20,913 t 114.4 t 0.5w 2,994.3 t 2,662 t 5.0w Lower-middle-income 13,525 t 65.6 t 0.5w 2,199.7 t 1,664 t 5.6w 50 Lesotho 0 0.0 0.0 0.1 1 0.2 0.1 1.0 7 24 51 Egypt, Atab Rep. 0 0.0 0.0 7.9 12 0.8 56.4 97.1° 67 889 52 Bolivia 493 6.3 1.2 92.3 25 8.5 1.2 0.4 20 181 53 Macedonia, FYR 9 0.0 0.1 2.2 16 8.5 .. . . . 54 Moldova 4 -0.2 -6.7 0.1 3 0.4 3.7 29.1° 60 793 55 Uzbekistan 14 1.0 5.5 2.4 10 0.6 82.2 76.4° 165 3,956 56 Indonesia 1,095 12.1 1.1 185.6 168 10.2 16.6 0.7 12 83 57 Philippines 78 3.2 3.4 6.1 27 2.0 29.5 9.1 123 562 58 Morocco 90 -1.2 -1.4 3.7 11 0.8 10.9 36.2 23 404 59 Syrian Arab Republic 7 -0.3 -4.3 0.0 0 0.0 3.3 9.4 30 405 60 Papua New Guinea 360 1.1 0.3 0.8 5 0.2 0.1 0.0 8 20 61 Bulgaria 37 -0.1 -0.2 3.7 46 3.3 13.9 6.8° 43 1,501 62 Kazakstan 0 0.0 0.0 9.9 20 0.4 37.9 30.2 92 2,202 63 Guatemala 42 0.8 1.8 13.3 18 12.3 0.7 0.6 13 127 64 Ecuador 120 2.4 1.8 111.1 15 40.1 5.6 1.8 41 541 65 Dominican Republic 11 0.4 2.9 10.5 17 21.7 3.0 14.9 22 423 66 Rumania 63 0.0 0.0 10.7 39 4.7 26.0 12.5° 91 1,044 67 Jamaica 2 0.3 7.8 0.0 1 0.2 0.3 3.9 11 148 68 Jordan 1 0.0 -1.0 2.9 10 .3.3 0.5 32.1 50 123 69 Algeria 41 0.3 0.8 119.2 19 5.0 4.5 30.4° 35 125 70 El Salvador 1 0.0 2.3 0.1 2 0.2 1.0 5.3 17 228 71 Ukraine 92 -0.2 -0.3 4.9 19 0.8 34.7 40.0° 108 565 72 Paraguay 129 4.0 2.8 15.0 20 3.8 0.4 0.1° 16 93 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 233 Forest areas Annual freshwater withdrawal, 1995b Total area Ann. deforestation, 1980-90 Nationally protected areas, 1994 total water Per capita (Cu. ml (thousand sq. kml Thousand % Thousand As % of Total 1990 sq. km change sq. km Number land area Icu. kml resources Domestic Other 73 Tunisia 7 -0.1 -1.9 0.4 6 0.3 3.1 79.5' 41 276 74 Lithuania 20 0.0 0.0 6.3 76 9.6 4.4 19.0' 83 1,107 75 Colombia 541 3.7 0.7 93.8 80 9.0 5.3 0.5 71 103 76 Namibia 126 0.4 0.3 102.2 12 12.4 0.2 0.3' 7 103 77 Belarus 63 -0.3 -0.4 2.7 11 1.3 3.0 5.4' 94 200 78 Russian Federation 7,681 15.5 0.2 705.4 209 4.2 117.0 2.7' 134 656 79 Latvia 28 -0.1 -0.2 7.8 45 12.5 0.7 2.2' 110 152 80 Peru 679 2.7 0.4 41.8 22 3.3 6.1 15.3 57 243 81 Costa Rica 14 0.5 3.0 6.5 28 12.7 1.4 1.4 31 749 82 Lebanon 1 0.0 0.6 0.0 1 0.4 0.8 15.6 30 241 83 Thailand 127 5.2 3.5 70.2 111 13.7 31.9 17.8' 24 578 84 Panama 31 0.6 1.9 13.3 14 17.8 1.3 0.9 91 664 85 Turkey 202 0.0 0.0 10.7 49 1.4 33.5 17.3' 140 445 86 Poland 87 -0.1 -0.1 30.7 111 10.1 12.3 21.9' 42 279 87 Estonia 19 -0.2 -1.2 4.1 38 9.7 3.3 21.2 105 1,992 88 Slovak Republic 18 0.0 0.1 10.2 40 21.1 1.8 5.8 . 89 Botswana 143 0.8 0.5 106.6 9 18.8 0.1 0.6' 5 94 90 Venezuela 457 6.0 1.2 263.2 100 29.8 4.1 03' 164 218 Upper-middle-income 7,387 t 48.8 t 0.6w 794.6 t 998 t 3.9w 91 SouthAfrica 45 -0.4 -0.8 69.7 238 5.7 13.3 26.6' 47 348 92 Croatia 20 0.0 0.1 3.9 30 7.0 . . . . . 93 Mexico 486 6.8 1.3 98.5 68 5.2 77.6 21.7 54 845 94 Mauritius 1 0.0 0.2 0.0 1 2.0 0.4 16.4 66 344 95 Gabon 182 1.2 0.6 10.5 6 4.1 0.1 0.0 41 16 96 Brazil 5,611 36.7 0.6 321.9 272 3.8 36.5 0.5' 54 191 97 Trinidad and Tobago 2 0.0 -2.1 0.2 5 3.1 0.2 2.9 40 108 98 Czech Republic 26 0.0 0.0 10.7 34 13.8 2.7 4.7 109 157 99 Malaysia 176 4.0 2.1 14.8 51 4.5 9.4 2.1 177 592 100 Hungary 17 -0.1 -0.5 5.7 53 6.2 6.8 5.7' 59 601 101 Chile 88 -0.1 -0.1 137.3 66 18.3 16.8 3.6 98 1,528 102 Oman 41 0.0 0.0 9.9 28 4.6 0.5 24.0 17 547 103 Uruguay 7 0.0 -0.6 0.3 8 0.2 0.7 0.5' 14 227 104 Saudi Arabia 12 0.0 0.0 62.0 10 2.9 3.6 163.6 224 273 105 Argentina 592 0.9 0.1 43.7 84 1.6 27.6 2.8' 94 949 106 Slovenia 10 0.0 0.0 1.1 10 5.4 . . . . . 107 Greece 60 0.0 0.0 2.2 21 1.7 5.0 8.6' 42 481 Low- and middle-income 28,828 179.8 0.6w 4,995.4 4,504 5.1w Sub-Saharan Africa 5,322 40.7 0.7w 1,362.5 673 t 5.8w East Asia and Pacific 3,986 43.5 1.0w 966.3 1,172 6.2 w South Asia 658 t 5.5 t 0.8 w 212.4 485 t 4.4w Europe and Central Asia 8,630 16.6 0.2w 860.0 964 r 3.6 w Middle East and N. Africa 446 t -1.4 -0.3 w 290.8 170 e 3.0w Latin America and Caribbean 9,786 74.8 0.7w 1,303.4 1,040 6.5 w Hieh-income economies 10,766 -46.4 -0.5 w 3,607.9 5,506 11.9w 108 Korea, Rep. 65 0.1 0.1 6.9 27 7.0 27.6 41.8 117 515 109 Portugal 31 -0.1 -0.5 5.8 24 6.3 7.3 10.5c 111 628 110 Spain 256 0.0 0.0 42.5 214 8.5 30.8 27.6' 94 687 Ill New Zealand 75 0.0 0.0 60.7 182 22.6 2.0 0.6 271 318 112 Ireland 4 0.0 -1.2 0.5 11 0.7 0.8 1.6' 37 196 113 tlsrael 1 0.0 -0.3 3.1 15 14.9 1.9 84.1' 65 343 114 IKuwait 0 0.0 0.0 0.3 2 1.5 0.5 . . 336 189 115 United Arab Emirates 0 0.0 0.0 0.0 0 0.0 0.9 300.0 97 787 116 United Kingdom 24 -0.2 -1.1 51.1 168 21.1 11.8 16.6 41 164 117 Australia 1,456 0.0 0.0 940.8 889 12.3 14.6 4.3 607 327 118 Italy 86 0.0 0.0 22.8 171 7.7 56.2 33.7' 138 848 119 Canada 4,533 -47.1 -1.1 823.6 627 8.9 45.1 1.6 288 1,314 120 Finland 234 -0.1 0.0 27.4 81 9.0 2.2 l.9c 53 387 121 tHong Kong 0 0.0 -0.5 .. .. .. .. .. 122 Sweden 280 -0.1 0.0 29.8 197 7.2 2.9 1.6' 123 218 123 Netherlands 3 0.0 -0.3 4.3 85 12.6 7.8 8.7' 26 492 124 Belgium 6 0.0 -0.3 0.8 3 2.3 9.0 72.2' 101 816 125 France 135 -0.1 -0.1 56.0 102 10.2 37.7 19.1' 106 559 126 S Singapore 0 0.0 2.3 0.0 1 4.9 0.2 31.7 38 46 127 Austria 39 -0.1 -0.4 20.8 170 25.2 2.4 2.6' 101 203 128 United States 2,960 3.2 0.1 1,302.1 1,585 14.2 467.3 18.9' 244 1,626 129 Germany 107 -0.5 -0.4 91.9 497 26.3 46.3 27.1' 64 518 130 Denmark 5 0.0 0.0 13.9 114 32.8 1.2 9.2' 70 163 131 Norway 96 -1.2 -1.4 55.4 113 18.0 2.0 0.5' 98 390 132 Japan 238 0.0 0.0 27.6 80 7.3 90.8 16.6 125 610 133 Switzerland 12 -0.1 -0.6 7.3 109 18.5 1.2 2.4' 40 133 World 39595t 1334t 03w 86032t lOOlOt 67w a Data may refer to earlier years and are the must recent reported by the World Conservation Monitoring Centre b. Refers to any year from 1970 to 1995. c. Total water resources include river flows from other countries. 234 WORLD DEVELOPMENT REPORT 1997 Table 11. Growth of the economy Average annual growth rate 1%) Exports of goods Gross domestic GDP GDP deflator Agriculture Industry Services and services investment 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 Low-income economies 6.0w 6.8w 13.4w 62.0w 3.6w 3.1 w 7.7w 11.6w 6.9w 6.4w 6.1 w 11.0w 6.2w 10.5w Excluding China and India 2.7w 1.8w 28.4w 170.8w 2.6w 1.9w 2.9w .. 2.8 w .. 2.5 w 4.4w -1.3w 3.6w 1 Mozambique -0.2 7.1 38.4 48.6 1.6 2.4 -9.8 -2.4 -0.1 15.0 -5.0 7.2 -2.5 8.6 2 Erhiopia' 2.3 .. 3.4 .. 1.4 1.8 .. 3.1 ... . 0.3 . 3.5 3 Tanzania 3.8 3.2 35.7 22.4 4.9 4.1 3.4 8.4 1.6 1.7 .. .. 4 Burundi 4.4 -2.3 4.4 8.5 3.1 -4.1 4.5 -5.0 5.4 -1.5 4.5 -2.6 4.5 -5.0 5 Malawi 2.3 0.7 15.0 30.4 2.0 1.7 2.9 0.4 3.4 -1.0 2.5 2.3 -2.8 -11.2 6 Chad 6.3 1.9 1.1 8.9 2.7 6.9 8.0 -9.9 9.9 1.2 7.7 -15.8 19.0 -2.9 7 Rwanda 2.3 -12.8 3.9 18.2 0.7 -10.8 1.8 -17.0 5.4 -123 3.5 17.8 3.7 -6.3 8 Sierra Leone 1.6 -4.2 62.2 39.6 4.4 -2.8 5.7 -2.8 -1.1 -5.9 2.8 -15.2 -6.5 -20.0 9 Nepal 4.6 5.1 11.1 11.0 4.0 1.5 6.0 9.3 4.8 7.2 1.1 25.8 1.8 6.3 10 Niger -1.1 0.5 2.9 6.5 1.8 . . -3.3 - - -5.2 . . -4.6 -5.2 -5.9 0.3 11 BurkinaFaso 3.7 2.6 3.1 6.2 3.1 4.6 3.7 1.4 4.7 1.7 -0.6 -1.2 8.6 -5.8 12 Madagascar 1.3 0.1 16.9 23.5 2.5 1.6 0.9 0.5 0.8 -0.6 -2.0 4.9 4.9 -4.5 13 Bangladesh 4.3 4.1 9.5 4.6 2.7 1.1 4.9 7.3 5.7 5.4 7.7 14.2 1.4 8.2 14 Uganda 3.1 6.6 125.6 23.7 2.3 3.8 6.0 11.0 3.0 8.2 2.3 11.7 9.6 7.9 15 Vietnam 8.3 210.7 26.3 . . 5.2 . . . . 16 Guinea-Bissau 4.5 3.5 56.1 49.5 6.7 4.8 0.4 1.9 3.3 2.2 -1.7 11.3 5.8 1.2 17 Hairi -0.2 -6.5 7.5 22.4 . . . . . - .. . . . . 1.2 -19.0 -0.6 -45.7 18 Mali 1.8 2.5 5.3 10.1 4.3 3.1 2.7 5.3 -1.7 1.2 5.2 5.3 5.4 6.1 19 Nigeria 1.6 1.6 16.6 47.1 3.3 2.3 -1.0 -1.2 3.2 4.5 -0.3 1.1 -8.6 1.2 20 Yemen, Rep. .. .. .. .. . . 21 Cambodia .. 6.4 .. 56.2 .. 2.1 .. 11.3 .. 8.3 - . . 22 Kenya 4.2 1.4 9.0 18.5 3.3 -0.4 3.9 1.5 4.9 3.1 4.3 3.0 0.8 0.0 23 Mongolia 5.5 -3.3 -1.2 126.7 2.9 . . 4.6 .. .. . - 3.1 12.8 1.7 24 Togo 1.8 -3.4 4.7 4.9 5.6 3.3 1.1 -60 -0.3 -8.6 0.6 -10.6 2.9 -16.4 25 Gambia, The 3.4 1.6 18.7 5.0 0.4 2.6 6.0 0.4 3.9 2.5 0.6 -0.8 0.8 3.0 26 Central African Republic 1.7 1.0 5.6 8.5 2.7 1.5 3.1 -4.6 -0.1 -1.6 -3.7 4.4 4.8 -8.7 27 Indjar 5.8 4.6 8.0 10.1 3.1 3.1 7.1 5.1 6.7 6.1 5.9 12.5 6.5 5.3 28 Lao PDR .. 6.5 37.8 10.1 .. .. . . .. .. .. .. 5.9 29 Benin 2.6 4.1 1.6 7.9 5.1 4.9 2.1 3.5 1.2 3.5 -2.7 6.3 -6.2 12.1 30 Nicaragua -2.0 1.1 422.6 98.3 -2.2 0.3 -1.7 -4.4 -2.0 2.2 -3.8 7.2 -4.7 4.1 31 Ghana 3.0 4.3 42.4 23.8 1.0 2.4 3.3 4.4 6.4 6.5 2.5 6.7 4.5 0.9 32 Zambia 0.8 -0.2 42.4 107.8 3.6 -0.5 1.0 -1.2 0.1 0.7 -3.3 10.8 -2.7 -10.2 33 Angola 3.7 -4.1 5.9 774.5 0.5 -1.8 6.4 0.9 2.2 -10.8 13.3 2.9 -6.8 0.1 34 Georgia 0.5 -26.9 1.9 2,280.2 0.7 -31.4 1.8 -34.1 -1.4 -22.3 - . .. 0.3 -21.2 35 Pakistan 6.3 4.6 6.7 11.2 4.3 3.4 7.3 5.7 6.8 5.0 8.1 7.7 5.9 4.0 36 Mauritania 1.7 4.0 8.6 6.8 1.7 4.9 4.9 3.9 0.4 3.2 3.4 -1.0 -4.1 -1.3 37 Azerbaijan .. -20.2 .. 747.6 .. .. . . .. .. .. .. 38 Zimbabwe 3.5 1.0 11.5 27.6 2.4 1.6 3.6 -3.6 2.9 1.7 5.4 5.5 1.3 1.5 39 Guinea . . 3.8 . . 10.1 .. 4.5 . - 2.3 . . 4.5 - - 1.3 - - 0.6 40 Honduras 2.7 3.5 5.7 19.2 2.7 2.9 3.3 4.9 2.5 1.3 1.1 0.2 2.9 10.0 41 Senegal 3.1 1.9 6.5 7.6 2.8 1.3 3.7 2.0 3.0 2.0 3.9 -0.9 3.9 4.7 42 China 10.2 12.8 5.8 12.4 5.9 4.3 11.1 18.1 13.6 10.0 11.5 15.6 11.0 15.5 43 Cameroon 3.1 -1.8 5.9 5.1 2.2 2.2 5.9 -6.8 2.1 -.1.4 5.9 2.2 -2.7 -4.1 44 Core d'Ivoire 0.1 0.7 3.4 10.4 -0.5 0.3 4.4 1.7 -1.3 0.2 1.9 -0.9 -28.8 138.3 45 Albania 3.0 1.4 -1.9 76.4 2.4 7.6 3.2 -15.6 3.2 5.9 .. .. -0.3 38.4 46 Congo 3.6 -0.6 0.3 7.8 3.4 -0.9 5.2 1.2 2.5 -2.1 4.8 4.0 -11.9 -7.9 47 Kyrgyz Republic - - -14.7 -. 337.3 .. -7.6 -. .. .. .. .. 48 Sri Lanka 4.2 4.8 10.8 10.4 2.2 2.4 4.6 6.5 4.7 6.3 6.8 11.0 0.6 6.8 49 Armenia 3.3 -21.2 0.3 896.6 -3.9 -0.6 5.1 -28.7 4.6 -19.7 - . . - 6.2 -17.7 Middle-income economies 1.9 ss 0.1 w 64.8w 298.8w .. 0.9w .. .. 2.6w 3.9w Lower-middle-income 2.3 w -1.5 w 17.5w 286.7w .. 0.5w 50 Lesotho 4.3 7.5 13.6 11.0 2.6 -3.4 7.2 12.3 5.2 6.1 4.1 11.4 6.9 12.1 51 Egypt, Arab Rep. 5.0 1.3 11.7 13.3 1.5 2.1 2.6 0.4 8.4 1.5 5.2 4.2 2.7 -1.5 52 Bolivia 0.0 3.8 316.7 10.5 2.0 -2.9 - - -0.1 - - 3.5 6.7 -9.9 4.2 53 Macedonia, FYR 54 Moldova 55 Uzbekistan .. -4.4 .. 628.4 .. -0.9 .. -6.7 . - -6.6 .. .. .. -9.2 56 Indonesia 6.1 7.6 8.5 7.6 3.4 2.9 6.9 10.1 7.0 7.4 2.9 10.8 7.0 16.3 57 Philippines 1.0 2.3 14.9 9.2 1.0 1.6 -0.9 2.2 2.8 2.7 3.5 9.4 -2.1 3.2 58 Morocco 4.2 1.2 7.2 3.9 6.7 -5.9 3.0 1.7 4.2 2.8 6.8 3.1 2.5 -2.5 59 Syrian Arab Republic 1.5 7.4 15.3 8.5 -0.6 . . 6.6 .. . . . . 3.6 . . -7.0 60 Papua New Guinea 1.9 9.3 5.3 5.7 1.8 4.7 1.9 17.8 2.0 4.8 3.3 13.3 -0.9 0.4 61 Bulgaria 4.0 -4.3 1.2 81.2 -2.1 -1.9 5.2 -7.5 7.2 -20.7 -3.5 -0.7 2.4 -7.1 62 Kazakstan . . -11.9 . . 805.5 - . -18.0 - - 19.2 .. 6.1 . . . . . . -167 63 Guaremala 0.8 4.0 14.6 14.2 2.3 2.5 2.1 4.2 2.1 4.9 -2.1 4.8 -1.8 10.7 64 Ecuador 2.0 3.4 36.4 37.2 4.4 2.5 1.2 4.9 1.8 2.7 5.4 7.4 -3.8 5.3 65 Dominican Republic 2.7 3.9 21.5 11.7 0.4 2.5 2.2 3.3 3.7 4.5 2.8 4.6 3.7 4.9 66 Romania 0.5 -1.4 2.5 158.4 .. -0.4 .. -2.1 .. -2.8 . . - - - - -10.0 67 Jamaica 2.0 2.9 18.6 38.5 0.6 8.3 2.4 -0.5 1.9 6.0 5.4 -1.0 -0.1 5.8 68 Jordan -1.5 8.2 7.0 4.7 13.2 10.2 -1.3 7.9 -8.2 6.2 5.9 8.2 7.3 6.5 69 Algeria 2.8 0.1 8.0 25.8 4.6 1.3 2.3 -1.1 3.8 1.3 4.1 0.2 -2.3 -4.7 70 El Salvador 0.2 6.3 16.4 11.2 -1.1 1.2 0.1 2.9 0.7 9.3 -3.4 13.6 2.2 14.7 71 Ukraine . - -14.3 -- 1,040.5 - - -9.7 -. -21.6 . - . - . . .. 72 Paraguay 2.5 3.1 24.4 3.6 18.0 1.4 1.9 -0.3 3.4 4.1 11.5 13.8 -0.8 2.6 Nore: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 235 Average annual growth rate 1%) Exports of goods Gross domestic GOP GOP deflator Agriculture lndustiy Services and services investment 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 1980-90 1990-95 73 Tunisia 3.3 3.9 7.4 5.4 2.8 -2.1 3.1 4.0 3.6 5.6 5.6 5.6 -1.8 1.4 74 Lithuania .. -9.7 .. 241.4 .. .. . .. ... .. .. .. 75 Colombia 3.7 4.6 24.6 23.3 2.9 1.4 5.0 3.0 3.1 6.4 7.5 7.2 0.5 19.0 76 Namibia 1.1 3.8 13.6 9.3 1.8 6.8 -1.1 2.9 2,7 4.6 1.5 6.9 11.9 -2.8 77 Belarus -9.3 . . 878.8 . -11.2 . . -10.9 . -6.9 -17.0 . . 78 Russian Federation 1.9 -9.8 3.2 517.0 .. . . . . . . . . . . . 79 Latvia 3.4 -13.7 . . 149.1 2.3 -16.4 4.3 -25.1 3.1 -2.1 . . . . 3.4 -37.1 80 Peru -0.2 5.3 229.6 62.4 . . . . . . . . . . . 1.7 8.3 -4.2 7.4 81 Costa Rica 3.0 5.1 23.5 19.1 3.1 3.6 2.8 5.2 3.1 5.6 6.1 9.5 5.3 6.6 82 Lebanon .. .. .. .. .. .. .. .. .. 83 Thailand 7.6 8.4 3.9 4.6 4.0 3.1 9.9 10.8 7.3 7.8 14.0 14.2 9.4 10.2 84 Panama 0.3 6.3 2.4 1.8 . . 4.4 . . 14.9 . . 5.5 4.3 . . 15.3 85 Turkey 5.3 3.2 45.3 75.6 1.3 0.9 7.8 4.2 4.4 3.3 16.9 9.4 5.3 2.0 86 Poland 1.9 2.4 53.7 34.9 -0.1 -2.0 -0.9 3.7 5.1 2.4 4.5 9.4 0.9 1.1 87 Estonia 2.1 -9.2 2.4 151.4 .. -8.9 . . -14.9 .. -3.8 -13.4 88 Slovak Republic 2.0 -2.8 1.8 16.0 1.6 1.0 2.0 -10.4 0.8 6.2 17.8 1.1 -7.7 89 Botswana 10.3 4.2 13.1 9.2 2.2 0.7 11.4 1.4 11.0 7.7 90 Venezuela 1.1 2.4 19.3 38.4 3.0 1.9 1.6 3.4 0.5 1.7 2.8 4.9 -5.3 3.8 U. er-middle-income 1.3 138.0w 320.5w 2.4w 1.8w 0.7w 2.6w 91 South Africa 1.3 0.6 14.8 11.5 3.0 -0.3 1.1 0.1 3.1 0.9 1.9 2.4 4.8 4.7 92 Croatia .. .. .. .. .. .. .. .. .. .. .. 93 Mexico 1.0 1.1 70.4 15.5 0.6 0.4 1.0 0.5 1.1 1.5 6.6 6.8 -3.1 -1.2 94 Mauritius 6.2 4.9 9.4 6.7 2.9 -1.4 10.3 5.6 5.4 6.4 10.4 4.8 10.2 1.7 95 Gabon 0.5 -2.5 1.9 13.0 1.7 -0.2 1.0 2.7 -0.3 -10.0 2.8 4.1 -4.6 -0.5 96 Brazil 2.7 2.7 284.5 965.3 2.8 3.7 2.0 1.7 3.5 3.6 7.5 7.4 0.2 3.5 97 TrinidadandTobago -2.5 1.0 4.1 7.2 -5.8 1.3 -5.5 0.2 -3.3 -0.1 8.9 12.5 10.1 1.0 98 Czech Republic 1.7 -2.6 1.5 18.3 . . .. . . . . . . . . .. . . 2.3 0.9 99 Malaysia 5.2 8.7 1.7 3.9 3.8 2.6 7.2 11.0 4.2 8.6 10.9 14.4 2.6 16.0 100 Hungary 1.6 -1.0 8.6 22.3 0.6 -7.0 -2.6 0.5 3.6 -4.6 4.0 -1.5 0.4 6.6 101 Chile 4.1 7.3 20.9 14.7 5.6 5.2 3.7 6.1 4.2 8.4 7.0 9.2 9.6 11.9 102 Oman 8.3 6.0 -3.6 -2.9 7.9 . . 10.3 . . 6.0 .. . . . . . 103 Uruguay 0.4 4.0 61.3 55.6 0.0 4.5 -0.2 0.1 0.9 6.2 4.3 4.4 -7.8 12.9 104 Saudi Arabia -1.2 1.7 -3.7 1.0 13.4 . . -2.3 . . -1.2 .. . . . . . 105 Argentina -0.3 5.7 389.0 20.5 0.9 0.5 -0.9 5.9 0.0 6.4 3.7 6.9 -4.7 16.0 106 Slovenia .. .. .. .. .. .. .. .. .. .. .. 107 Greece 1.4 1.1 18.3 13.1 -0.1 3.1 1.3 -0.8 4.9 0.6 7.1 4.5 -0.9 1.9 Low- and middle-income 2.8w 2.1 w 50.flw 235.8w 3.1 w 2.0w 3.9w 4.9w 3.6w 4.5w .. .. 1.8w 6.5w Sub-Saharan Africa 1.7w 1.4w 19.0w 47.4w 1.9w 1.5w 0.6w 0.2w 2.5w 1.5w 1.9w 2.5w -4.0w 3.4w East Asia and Pacific 7.6w 10.3w 11.5w 4.8w 10.1 w 3.9w 8.9w 15.0w 9.0w 8.4w 8.8w 13.9w 8.5w 14.4w SouthAsia 5.7w 4.6w 8.0w 9.9w 3.2w 3.0w 6.9w 5.3w 6.6w 6.0w 6.4w 11.9w 6.1w 5.3w Europe and Central Aria 2.3w -6.5w 12.2w 461.5w .. .. .. .. Middle East and N. Africa 0.2w 2.3w 8.2w 19.4w 4.5w 3.3w 1.1 w .. 1.2w . . . . . . Latin America and Caribbean 1.7w 3.2w 179.4w 380.9w 2.0 w 2.3w 1,4w 2.5 w 1.9w 3.8 w 5.4w 7.0w -1.5 w 5.7w High-income economies 3.2w 2.0w 4.8 w 2.4w 2.3w 0.6w 3.2w 0.7w 3.4w 2.3w 5.2w 6.4w 4.1 w -0.2 w 108 Korea, Rep. 9.4 7.2 5.9 6.2 2.8 1.3 13.1 7.3 8.2 7.9 12.0 13.4 11.9 7.2 109 Portugal 2.9 0.8 18.1 8.7 8.7 3.3 . 110 Spain 3.2 1.1 9.3 5.2 -1.7 . . . . . . . . 5.7 10.1 5.7 -2.6 111 New Zealand 1.8 3.6 10.8 0.6 4.4 0.9 1.3 3.8 1.7 3.5 4.1 5.2 1.7 12.4 112 Ireland 3.1 4.7 6.6 2.2 8.9 10.7 . . -3.8 113 lIsrael 3.5 6.4 101.5 12.2 5.5 9.5 2.2 11.5 114 f Kuwait 0.9 12.2 -2.4 -2.0 14.7 .. 1.0 .. 0.9 .. -2.3 .. -4.5 115 f United Arab Emirates -2.0 . . 0.7 9.6 9.3 -4.2 -1.8 3.4 4.9 0.0 .. -8.7 116 UniredKingdom 3.2 1.4 5.7 3.6 3.9 4.3 6.4 117 Australia 3.4 3.5 7.3 1.3 3.3 -2.4 2.8 3.3 3.7 3.7 7.0 6.8 2.7 5.8 118 Italy 2.4 1.0 9.9 4.7 0.6 1.6 . . . . . . . . 4.1 8.1 2.1 -3.2 119 Canada 3.4 1.8 4.4 1.5 1.5 0.3 2.9 1.2 3.7 1.8 6.0 9.5 5.2 2.3 120 Finland 3.3 -0.5 6.8 1.8 -0.2 0.0 3.3 -1.2 5.3 -2.7 2.2 9.8 3.0 -8.3 121 tHongKong 6.9 5.6 7.7 8.1 14.4 13.5 4.0 11.7 122 Sweden 2.3 -0.1 7.4 3.2 1.5 -1.9 2.8 -0.7 2.5 -0.1 4.3 6.7 4.3 -7.2 123 Netherlands 2.3 1.8 1.6 2.2 3.4 3.0 1.6 0.4 2.6 2.1 4.5 3.7 3.1 -0.3 124 Belgium 1.9 1.1 4.4 3.1 1.8 4.0 2.2 . . 1.8 . . 4.6 4.6 3.2 -0.9 125 France 2.4 1.0 6.0 2.1 2.0 -1.1 1.1 -1.0 3.0 1.5 3.7 4.0 2.8 -2.8 126 t Singapore 6.4 8.7 2.0 3.7 -6.2 0.5 5.4 9.2 7.2 8.4 10.0 .. 3.7 6.0 127 Austria 2.1 1.9 3.7 3.5 1.1 -1.8 1.9 1.7 2.3 2.2 4.6 2.5 2.5 3.6 128 United States 3.0 2.6 4.1 2.4 4.0 3.6 2.8 1.2 3.1 2.1 5.2 7.3 3.4 4.1 129 Germany" 2.2 .. 2.6 1.7 .. 1.2 . . 2.9 . . 4.4 . . 2.0 130 Denmark 2.4 2.0 5.5 3.1 0.3 2.9 1.6 2.3 1.3 4.4 2.8 4.0 -1.1 131 Norway 2.9 3.5 5.5 1.3 0.9 .. 3.5 . . 2.6 .. 5.0 5.1 0.6 132 Japan 4.0 1.0 1.7 0.9 1.3 -2.2 4.2 0.0 3.9 2.3 4.5 3.4 5.3 -0.8 133 Switzerland 2.2 0.1 3.7 2.3 3.4 1.5 4.9 0.0 World 3.1w 2.0w w 56.6w 2.8w 1.3w 3.3w 1.4w 3.4w 2.6w 5.3w 6.8w 3.7w 0.8w a. Services inclade unallocated items. b Data poor to 1992 include Lntrca c. GDP growth rates were revised after the statistics for this publication were finalized. d. l)ata prior to 1990 refer to the Federal Repuhltc of Germany before unification 236 WORLD DEVELOPMENT REPORT 1 97 Table 12. Structure of the economy: production Distiibution of gross domestic product (%1 Agriculture Industry (Manufactuiing) $ervice&' GOP Imillion $1 value added value added value added value added 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 Low-income economies 739,236 t 1,352,256 t 34w 25 w 32w 38w 21 w 27w 32w 35w Exduding China and India 390,472t 316,889t 33w 25w 13w 41w 1 Mozambique 2,028 1,469 37 33 31 12 .. 32 55 2 Ethiopia 5,179 5,287 56 57 12 10 6 3 31 33 3 Tanzania 5,702 3,602 46 58 18 17 11 8 37 24 4 Burundi 920 1,062 62 56 13 18 7 12 25 26 5 Malawi 1,238 1,465 37 42 19 27 12 18 44 31 6 Chad 727 1,138 54 44 12 22 16 34 35 7 Rwanda 1,163 1,128 50 37 23 17 16 3 27 46 8 Sierra Leone 1,166 824 33 42 21 27 5 6 47 31 9 Nepal 1,946 4,232 62 42 12 22 4 10 26 36 10 Niger 2,538 1,860 43 39 23 18 4 35 44 11 Burkina Faso 1,709 2,325 33 34 22 27 16 21 45 39 12 Madagascar 4,042 3,198 30 34 16 13 . . 13 54 53 13 Bangladesh 12,950 29,110 50 31 16 18 11 10 34 52 14 Uganda 1,267 5,655 72 50 4 14 4 6 23 36 15 Vietnam . . 20,351 28 30 22 42 16 Guinea-Bissau 105 257 44 46 20 24 7 36 30 17 Haiti 1,462 2,043 .. 44 .. 12 .. 9 . 44 18 Mali 1,629 2,431 61 46 10 17 4 6 29 37 19 Nigeria 93,082 26,817 27 28 40 53 8 5 32 18 20 Yemen, Rep. .. 4,790 22 27 14 .. 51 21 Cambodia . . 2,771 .. 51 .. 14 .. 6 . 34 22 Kenya 7,265 9,095 33 29 21 17 13 11 47 54 23 Mongolia 2,328 861 .. .. .. .. .. 24 Togo 1,136 981 27 38 25 21 8 9 48 41 25 Gambia, The 233 384 30 28 16 15 7 7 53 58 26 Central African Republic 797 1,128 40 44 20 13 7 .. 40 43 27 India 172,321 324,082 38 29 26 29 18 19 36 41 28 Lao PDR .. 1,760 .. 52 .. 18 .. 14 .. 30 29 Benin 1,405 1,522 35 34 12 12 8 7 52 53 30 Nicaragua 2,144 1,911 23 33 31 20 26 16 45 46 31 Ghana 4,445 6,315 58 46 12 16 8 6 30 38 32 Zambia 3,884 4,073 14 22 41 40 18 30 44 37 33 Angola 3,722 12 . 59 3 .. 28 34 Georgia 2,325 .. 67 . . 22 .. 18 .. 11 35 Pakistan 23,690 60,649 30 26 25 24 16 17 46 50 36 Mauritania 709 1,068 30 27 26 30 13 44 43 37 Azerbaijan . 3,475 .. 27 . . 32 .. .. .. 41 38 Zimbabwe 5,355 6,522 14 15 34 36 25 30 52 48 39 Guinea .. 3,686 .. 24 .. 31 .. 5 .. 45 40 Honduras 2,566 3,937 24 21 24 33 15 18 52 46 41 Senegal 3,016 4,867 19 20 25 18 15 12 57 62 42 China 201,688 697,647 30 21 49 48 41 38 21 31 43 Cameroon 6,741 7,931 29 39 23 23 9 10 48 38 44 Core d'Ivoire 10,175 10,069 27 31 20 20 13 18 53 50 45 Albania . . 2,192 56 21 . . 23 46 Congo 1,706 2,163 12 10 47 38 7 6 42 51 47 Kyrgyz Republic .. 3,028 .. 44 . . 24 .. . S 32 48 Sri Lanka 4,024 12,915 28 23 30 25 18 16 43 52 49 Armenia . 2,058 44 . . 35 . 25 20 Middle-income economies 2,461,307t 4,033,376t 11w .. 35w 18w 52w Lower-middle-income .. 2,025,853 t .. 13w .. 36w .. .. .. 49w 50 Lesotho 368 1,029 24 10 29 56 7 18 47 34 51 Egypt, Arab Rep. 22,913 47,349 18 20 37 21 12 15 45 59 52 Bolivia 3,074 6,131 18 . 35 . . 15 47 53 Macedonia, FYR 1,975 . . . . . . 54 Moldova 3,518 . . 50 28 . . 26 22 55 Uzbekistan . . 21,590 .. 33 .. 34 .. 18 .. 34 56 Indonesia 78,013 198,079 24 17 42 42 13 24 34 41 57 Philippines 32,500 74,180 25 22 39 32 26 23 36 46 58 Morocco 18,821 32,412 18 14 31 33 17 19 51 53 59 Syrian Arab Republic 13,062 16,783 20 .. 23 . . .. . . 56 60 Papua New Guinea 2,548 4,901 33 26 27 38 10 8 40 34 61 Bulgaria 20,040 12,366 14 13 54 34 . 32 53 62 Kazakstan .. 21,413 .. 12 . 30 6 . . 57 63 Guatemala 7,879 14,489 .. 25 .. 19 .. .. . . 56 64 Ecuador 11,733 17,939 12 12 38 36 18 21 50 52 65 Dominican Republic 6,631 11,277 20 15 28 22 15 15 52 64 66 Romania . 35,533 .. 21 .. 40 .. .. . 39 67 Jamaica 2,679 4,406 8 9 38 38 17 18 54 53 68 Jordan .. 6,105 .. 8 .. 27 .. 14 .. 65 69 Algeria 42,345 41,435 10 13 54 47 9 9 36 41 70 El Salvador 3,574 9,471 38 14 22 22 16 . . 40 65 71 Ukraine . . 80,127 .. 18 . . 42 . 37 . 41 72 Paraguay 4,579 7,743 29 24 27 22 16 16 44 54 Notes For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 237 Distribution of gross domestic product 1%) Agriculture Industry (Manufacturingi Services" GDP (million $) value added value added value added value added 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 73 Tunisia 8,743 18,035 14 12 31 29 12 19 55 59 74 Lithuania 7,089 11 36 30 53 75 Colombia 33,399 76,112 19 14 32 32 23 18 49 54 76 Namibia 2,190 3,033 12 14 53 29 5 9 35 56 77 Belarus 20,561 13 35 22 52 78 Russian Federation 344,711 7 38 .. 31 55 79 Larvia 6,034 9 31 18 60 80 Peru 20,661 57,424 10 7 42 38 20 24 48 55 81 Costa Rica 4,831 9,233 18 17 27 24 19 19 55 58 82 Lebanon 11,143 .. 7 .. 24 .. 10 .. 69 83 Thailand 32,354 167,056 23 11 29 40 22 29 48 49 84 Panama 3,592 7,413 11 15 74 85 Turkey 68,790 164,789 26 16 22 31 14 21 51 53 86 Poland 57,068 117,663 6 39 26 54 87 Estonia 4,007 8 28 17 .. 64 88 Slovak Republic 17,414 .. 6 .. 33 .. .. .. 61 89 Botswana 971 4,318 13 5 44 46 4 4 43 48 90 Venezuela 69,377 75,016 5 5 46 38 16 17 49 56 Upper middle income 91 South Africa 989 317 t 78,744 1 981 511 a 136,035 55Ø,a6 , 7 5 w 47w 50 37 w 31 20 w 23 18w 24 4 43 r$!"* 64 92 Croatia 18,081 .. 12 25 20 62 93 Mexico 194,914 250,038 8 8 33 26 22 19 59 67 94 Mauritius 1,132 3,919 12 9 26 33 15 23 62 58 95 Gabon 4,285 4,691 7 .. 60 .. 5 .. 33 96 Brazil 235,025 688,085 11 14 44 37 33 24 45 49 97 Trinidad and Tobago 6,236 5,327 2 3 60 42 9 9 38 54 98 Czech Republic 29,123 44,772 7 6 63 39 .. .. 30 55 99 Malaysia 24,488 85,311 22 13 38 43 21 33 40 44 100 Hungary 22,163 43,712 .. 8 .. 33 .. 24 59 101 Chile 27,572 67,297 7 37 21 55 102 Oman 5,982 12,102 3 . 69 .. 1 . . 28 103 Uruguay 10,132 17,847 14 9 34 26 26 18 53 65 104 SaudiArabia 156,487 125,501 1 . 81 . 5 . 18 105 Argentina 76,962 281,060 6 6 41 31 29 20 52 63 106 Slovenia . . 18,550 . . 5 . . 39 . 1 . . 57 107 Greece 40,147 90,550 27 21 48 36 30 21 24 43 Low- and middle-income 3,192,729 t 5,393,142 t .. 14w .. 36w . 20w .. 48w Sub-Saharan Africa 292,557 t 296,748 t 24w 20w 36w 30w 12w 15w 38w 48w East Asia and Pacific 464,719 t 1,341,265 t 27w 18w 39w 44w 27w 32w 32w 38w SouthAsia 219,283t 439,203t 39w 30w 24w 27w 15w 17w 35w 41w Europe and Central Asia .. 1,103,330 t .. .. Middle East and N. Africa 463,031 t .. 9w .. 57w .. 7w .. 32w Latin America and Caribbean 758,569 t 1,688,195 t lOw 10w 37w 33w 25w 21 w 51 w 55w High-incomeeconomies 7,758,074t 22,485,548t 3w 2w 37w 32w 24w 21w 58w 66w 108 Korea, Rep. 63,661 455,476 15 7 40 43 29 27 45 50 109 Portugal 28,526 102,337 . . . . 110 Spain 211,543 558,617 . . 3 . . . . . . 111 New Zealand 22,469 57,070 11 31 . 22 . . 58 112 Ireland 20,080 60,780 . . . S 113 tlsrael 22,579 91,965 . . . . . . . . . . . . 114 5 Kuwait 28,639 26,650 0 0 75 53 6 11 25 46 115 tUnitedArab Emirates 29,625 39,107 1 2 77 57 4 8 22 40 116 United Kingdom 537,382 1,105,822 2 2 43 32 27 21 54 66 117 Australia 160,109 348,782 5 3 36 28 19 15 58 70 118 Italy 452,648 1,086,932 6 3 39 31 28 21 55 66 119 Canada 263,193 568,928 5 . . 40 . . 22 . . 55 120 Finland 51,306 125,432 12 6 49 37 35 28 39 57 121 tHongKong 28,495 143,669 1 0 32 17 24 9 67 83 122 Sweden 125,557 228,679 4 2 37 32 25 23 59 66 123 Netherlands 171,861 395,900 3 3 32 27 18 18 64 70 124 Belgium 118,022 269,081 2 2 34 . . 24 . . 64 125 France 664,597 1,536,089 4 2 34 27 24 19 62 71 126 tSingapore 11,718 83,695 1 0 38 36 29 27 61 64 127 Austria 76,882 233,427 4 2 40 34 28 24 56 63 128 United States 2,708,150 6,952,020 3 2 34 26 22 18 64 72 129 Germany .. 2,415,764 . .. .. S - 130 Denmark 66,322 172,220 6 4 33 29 22 21 61 67 131 Norway 63,283 145,954 4 . . 36 . . 15 .. 60 132 Japan 1,059,253 5,108,540 4 2 42 38 29 24 54 60 133 Switzerland 101,646 300,508 . . . . S S World 10768090t 27846,241t 7w 5w '°'S lt w 23w 21w 53w 63w a. Because manufacturing is generally the most dynamic part of the industrial sector, its share is shown separately. b. Services include unallocated items. c. Data prior to 1992 include Erirrca. 238 WORLD DEVELOPMENT REPORT 1997 Table 13. Structure of the economy: demand Distribution of gross domestic product 1%) General govt. Private Gross domestic Gross domestic Exports of goods Resource consumption consumption investment saving and services balance 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 Low-income economies 12w 12w 66w 59w 24 32w 22w 30w 13w 19w 2w 1w Excluding China and India 13w 80w .. 20w .. lOw 24w 8w 6w Mozambique 21 20 78 75 22 60 5 20 23 22 6 11 1 1 2 Ethiopia ]4a 12 83 81 17 3 7 iF' 15 3 Tanzania 12 10 69 97 29 31 19 7 14 30 10 38 4 Burundi 9 12 92 95 14 11 7 9 12 15 18 5 Malawi 19 20 70 76 25 15 11 4 25 29 14 11 6 Chad 8 17 99 93 4 9 6 10 24 13 17 19 7 Rwanda 12 14 83 93 16 13 7 15 6 11 20 8 Sierra Leone 8 11 91 98 18 6 2 9 23 13 16 15 9 Nepal 7 8 82 79 18 23 11 12 12 24 7 12 10 Niger 10 17 67 82 37 6 23 1 24 13 14 4 11 Burkina Faso 10 16 95 78 17 22 6 6 10 14 23 16 12 Madagascar 12 7 89 91 15 11 1 3 13 23 16 13 Bangladesh 6 14 92 78 15 17 2 8 6 14 13 8 14 Uganda 11 10 89 83 6 16 0 7 19 12 6 9 15 Vietnam 7 77 27 16 36 11 16 Guinea-Bissau 29 77 98 30 16 6 13 36 22 17 18 Haiti Mali 10 10 11 8 6 82 91 101 79 17 17 26 2 2 8 710 8 22 16 4 22 9 19 9 16 19 Nigeria 12 10 56 69 22 18 32 20 29 25 10 4 20 Yemen, Rep. 29 61 12 10 43 2 21 Cambodia 11 82 19 6 11 13 22 Kenya 20 15 62 72 29 19 18 13 28 33 11 6 24 Togo Mongolia 22 11 53 74 80 30 46 14 25 9 51 31 5 36 4 23 b 27 19 2 25 Gambia, The 20 19 79 76 26 21 5 47 53 26 15 26 Central African Republic 15 13 94 80 7 15 10 6 26 18 17 9 27 India 10 10 73 68 21 25 17 22 7 12 4 2 28 Lao PDR 29 30 Benin Nicaragua 9 20 14 9 96 83 82 95 15 17 20 18 5 2 99 23 24 27 24 20 19 6 27 31 Ghana 11 12 84 77 6 19 5 10 8 25 1 8 32 Zambia 26 9 55 88 23 12 19 3 41 31 4 7 33 Angola 47 9 27 43 74 33 34 Georgia 13 7 56 103 29 3 9 17 35 Pakistan 10 12 83 73 18 19 7 16 12 16 - 3 36 Mauritania 25 9 68 80 36 15 7 11 37 50 29 3 Azerbaijan 16 37 38 39 Zimbabwe Guinea 20 b 19 8 64 96 64 81 19 16 22 15 16 . 4 17 11 30 27 34 21 3 6 4 40 Honduras 13 14 70 73 25 23 17 14 36 36 8 10 41 Senegal 22 11 78 79 15 16 o io 28 32 16 5 42 China 15 12 51 46 35 40 35 42 6 21 0 2 43 Cameroon 10 9 70 71 21 15 20 21 27 26 1 6 44 Côte dIvoire 17 12 63 67 27 13 20 20 35 41 6 7 45 Albania 9 15 56 93 35 16 . 8 14 24 46 Congo 18 12 47 64 36 27 36 23 60 62 0 4 47 Kyrgyz Republic 23 67 16 10 26 18 48 Sri Lanka 9 12 80 74 34 25 11 14 32 36 23 11 49 Armenia 16 13 47 116 29 9 29 24 53 Middle-income economies 14w 59w .. 25w 25w 24 1w Lower-middle-income 50 Lesotho 36 23 124 85 42 87 60 9 20 21 102 96 51 Egypt, Arab Rep. 16 13 69 81 28 17 15 6 31 21 12 3 52 53 Bolivia Macedonia, FYR 14 13 14 67 79 82 15 15 19 8 4 21 20 37 4 7 11 54 Moldova 20 81 15 7 . 35 7 55 56 Uzbekistan Indonesia 11 25 8 .. 52 59 56 . 24 23 38 37 .. 24 36 33 63 25 13 0 57 Philippines 9 11 67 74 29 23 24 15 24 36 5 8 58 Morocco 18 15 68 71 24 21 14 13 17 27 10 8 59 Syrian Arab Republic 23 . . 67 . . 28 . . 10 . 18 17 60 Papua New Guinea 24 12 61 48 25 24 15 39 43 61 10 15 61 Bulgaria 6 15 55 61 34 21 39 25 36 49 5 2 62 Kaaakstan 15 . 65 22 . . 19 34 3 6 . 63 Guatemala 8 6 79 86 16 17 13 8 22 19 64 Ecuador 15 13 60 67 26 19 26 21 25 29 0 2 65 Dominican Republic 8 4 77 80 25 20 15 16 19 26 10 0 66 Rumania 5 12 60 66 40 26 35 21 35 28 5 67 Jamaica 64 7 68 69 Jordan Algeria 20 14 22 9 16 .. 43 80 75 56 16 39 17 26 32 . 16 43 10 3 29 51 34 69 49 27 4 0 24 3 70 El Salvador 14 8 72 86 13 19 14 6 34 21 1 16 71 Ukraine 3 72 Paraguay 79 32 23 18 14 15 36 13 19 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 239 Distribution of gross domestic product 1%) General govt. Private Gross domestic Gross domestic Exports of goods Resource consumption consumption investment saving and services balance 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 73 Tunisia 14 16 62 63 29 24 24 20 40 45 5 4 74 Lithuania 20 63 19 16 58 3 75 Colombia 10 9 70 75 19 20 20 16 16 15 1 3 76 Namibia 17 31 44 52 29 20 39 17 76 53 10 7 77 Belarus 22 58 25 20 43 5 78 Russian Federation 15 16 62 58 22 25 26 22 3 79 Latvia 8 20 60 65 26 21 .. 16 .. 43 .. 5 80 Peru 11 6 57 83 29 17 32 11 22 12 3 5 81 Costa Rica 18 17 66 60 27 25 16 24 26 41 10 1 82 Lebanon 12 110 29 22 10 62 83 Thailand 12 10 65 54 29 43 23 36 24 42 6 7 84 Panama .. 15 . 64 .. 24 .. 22 . 39 2 85 Turkey 10 10 78 70 18 25 11 20 5 20 7 5 86 Poland 9 18 67 63 26 17 23 19 28 28 3 2 87 Estonia 23 58 27 18 75 9 88 Slovak Republic 20 50 - 28 30 63 2 89 Botswana 19 32 53 45 38 25 28 23 53 49 10 2 90 Venezuela 12 6 55 73 26 16 33 21 29 27 7 5 Upper middle mcome 12w lSw 56w 61w 25 21w 32w 23w 27w 22w 3F$t, 91 SouthAfrica 13 21 50 61 28 18 36 18 36 22 8 0 92 Croatia 33 66 .. 14 1 - 40 13 93 Mexico 10 10 65 71 27 15 25 19 11 25 2 3 94 Mauritius 14 12 75 65 21 25 10 22 51 58 10 3 95 Gabon 13 10 26 42 28 26 61 48 65 61 33 22 96 Brazil 9 17 70 62 23 22 21 21 9 7 2 1 97 TrinidadandTobago 12 13 46 62 31 14 42 25 50 39 11 11 98 Czech Republic 20 60 25 20 - 52 99 Malaysia 17 12 51 51 30 41 33 37 58 96 3 3 100 Hungary 10 11 61 68 31 23 29 21 39 35 2 2 101 Chile 12 9 67 62 25 27 20 29 23 29 4 2 102 Oman 25 31 28 42 22 17 47 27 63 49 25 10 103 Uruguay 12 13 76 74 17 14 12 13 15 19 6 1 104 Saudi Arabia 16 27 22 43 22 20 62 30 71 40 41 10 105 Argentina b b 76 82 25 18 24 18 5 9 1 0 106 Slovenia . . 21 -- 58 .. 22 - . 21 .. 56 - - 1 107 Greece 16 19 60 74 29 19 23 7 21 22 5 12 Low- and middle-income 14w 14w 57w 63w 26 27w 30w 22w 23w 22w 2w 1w Sub-Saharan Africa 14w 17w 60w 67w 23 19w 27w 16w 31w 28w 3w 3w East Asia and Pacific 12w 11w 58w 51w 28 39w 28w 38w 16w 29w 1w 1w South Asia 9w 11w 75w 69w 20 23w 15w 20w 8w 14w 6w 3w Europe and Central Asia Middle East and N. Africa l6w 39w 26 45w .. 47w .. 16w Latin America and Caribbean 11w 12w 67w 67w 25 20w 23w 19w 16w 17w 2w 2w High-income economies 17w 15w 60w 63w 23 21w 23w 21w 22w 22w Ow 2w 108 Korea, Rep. 12 10 64 54 32 37 25 36 34 33 7 1 109 Portugal 14 17 65 65 34 28' 21 18 24 28 13 9 110 Spain 13 16 66 62 23 21 21 22 16 24 2 0 111 NewZealand 18 15 62 60 21 24 20 26 30 32 1 2 112 Ireland 19 15 67 57 27 13 14 27 48 75 13 15 113 lIsrael 39 29 50 58 22 24 11 13 40 29 11 10 114 IKuwait 11 33 31 49 14 12 58 18 78 55 44 6 115 tUnitedArabEmirates 11 18 17 54 28 27 72 27 78 70 43 1 116 UnitedKingdom 22 21 59 64 17 16' 19 15 27 28 2 1 117 Australia 18 17 59 60 25 23 24 22 16 20 2 1 118 Italy 15 16 61 62 27 18 24 22 22 26 3 3 119 Canada 19 19 55 60 24 19 25 21 28 37 2 2 120 Finland 18 21 54 54 29 16 28 24 33 38 1 8 121 tHong Kong 6 9 60 59 35 35 34 33 90 147 2 122 Sweden 29 26 51 55 21 14 19 19 29 41 2 4 123 Netherlands 17 14 61 57 22 22 22 29 51 53 0 7 124 Belgium 18 15 63 62 22 18 19 24 63 74 3 6 125 France 18 20 59 60 24 18 23 20 22 23 2 126 tSingapore 10 9 53 40 46 33 38 . . 207 9 127 Austria 18 19 56 55 28 27 26 26 37 38 2 1 128 UnitedStates 18 16 63 68 20 16 19 15 10 11 1 2 129 Germany - 20 . 58 . - 21 . . 23 . - 23 - - 1 130 Denmark 27 25 56 54 19 16 17 21 33 35 -1 6 131 Norway 18 21 51 50 25 23' 31 29 43 38 6 6 132 Japan 10 10 59 60 32 29 31 31 14 9 1 2 133 Switzerland 13 14 67 59 24 23 20 27 37 36 4 4 World 16w 15w 59w 63w 24 23w 25w 21w 22w 22w 1w 1 a. Includcs Eritrea. Is. General government consumption figures are not available separately; they are included in private consumption. c. lnclude.c statistical discrepancy. 240 WORLD DEVELOPMENT REPORT 1997 Table 14. Central government budget Percentage of GDP Total revenuea Total expenditure Percentage of total expenditureb deficit/surplus0 Tax Nontax Current Capital Defense Social servicesd 1% of GDP) 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 Low-income economies Excluding China and India 1 Mozambique 0 .. . . 2 Ethiopia 12.8 11.9 370 18.00 330 3550 1960 3,]0 -8.5 3 Tanzania 17.1 .. 9.4 19.2 10.4 .. 9.2 35.0 -7.0 4 Burundi 13.2 3.5 .. 11.5 11.0 -3.9 5 Malawi 16.6 5.9 18.0 . 16.6 . 12.8 30.7 -15.9 6 Chad . .. 7 Rwanda 11.0 .. 2.5 .. 9.3 .. 5.0 -1.7 -69 8 Sierra Leone 14.0 12.5 2.5 4.5 19.6 13.3 5.0 65 -12.1 -5.0 9 Nepal 6.6 9.1 2.9 4.3 .. .. .. 6.7 -3.0 10 Niger 12.2 2.6 9.4 9.0 . . 3.8 50.5 -4.7 11 Burkina Paso 10.4 . . 1.9 . . 9.8 . 2.3 . . 17.0 36.0 0.2 12 Madagascar 12.9 8.2 5.2 2.2 . . 11.3 7.5 34.6 -4.8 13 Bangladesh 7.7 2.9 . . . . . . 9.4 2.5 14 Uganda 3.0 1.3 . . 5.3 0.8 .. 25.2 24: -3.1 15 Vietnam . . . . . 16 Guinea-Bissau . . . . .. . . . . . . . . . . . 17 Haiti 9.3 1.6 . . 13.9 . . 3.5 .. . . . . . . -4.7 18 Mali 9.5 4.0 12.3 . . 1.9 . . 11.0 37.8 -4.6 19 Nigrria .. . . .. . . . . .. . . . . .. . . . 20 Yemen, Rep. .. 13.0 2.6 .. 34.7 3.8 .. 30.3 .. 30.8 .. -17.3 21 Cambodia .. .. .. .. .. .. .. .. .. .. .. 22 Kenya 19.1 19.6 8.5 10.0 19.4 22.1 5.9 5.3 16.4 62 36.0 271 -4.5 -3.2 23 Mongolia .. 20.3 .. 5.0 . . 18.1 .. 3.3 .. 11.5 .. 38.8 .. -1.9 24 logo 27.0 .. 4.6 .. 23.7 .. 8.9 .. 71 .. 562 .. -2.0 25 Gambia, The 20.0 21.8 0.8 7.5 16.7 15.8 15.5 4.6 .. 38.2 -4.5 3.5 26 Central African Republic 15.0 .. 3.4 .. 18.5 .. 1.3 .. 9.7 .. 34.5 .. -3.5 27 India 9.8 9.6 5.0 4.1 11.7 14.2 1.6 1.8 19.8 14.5 8.8 11.9 -6.5 -5.4 28 Lao PDR 29 Benin .. .. .. .. .. .. .. ., .. .. .. 30 Nicaragua 20.3 23.6 8.7 10.7 24.8 22.0 4.7 8.4 7.7 0.0 41.3 63.5 -7.2 -4.3 31 Ghana 6.4 12.9 1.9 5.6 9.8 17.6 1.1 3.0 3.7 4.9 38.3 42.3 -4.2 -2.5 32 Zambia 23.1 13.4 10.8 7.0 33.0 10.7 4.0 6.1 0.0 23.4 59.0 -18.5 -2.9 33 Angola .. 34 Georgia .. .. .. .. .. .. .. .. .. .. 35 Pakistan 13.3 15.3 5.5 7.2 14.5 19.1 3.1 4.1 .. . . . . -5.7 -4.8 36 Mauritania . . . 37 Azerbaijan .. .. .. . . . , . . .. .. . . 38 Zimbabwe 19.2 6.7 .. 33.0 . . 1.8 . . 25.0 . . 30.2 . . -10.9 39 Guinea . . . . . . 40 Honduras 13,6 . . 3.5 . . . . . . . . . . -0.2 41 Senegal 20.7 . . 6.3 . . 22.3 1.9 .. 16.8 . . 38.5 0.9 0.0 42 China .. 5.7 .. 4.6 .. .. .. .. .. 12.4 .. .. .. -1.9 43 Cameroon 15.0 9.5 2.9 2.8 10.5 14.0 5.2 1.5 9.1 9.4 38.2 29.0 0.5 1.7 44 Côte d'Ivoire 21.1 .. 5.7 .. 19.1 ,, 9.0 .. .. .. .. .. -10.8 45 Albania . . 18.3 9.2 . . 28.0 . . 6.1 . 7.1 . . 40.2 . . -9.9 46 Congo 27.0 .. 2.7 .. 21.8 .. 17.7 .. 16.1 12.6 .. -5.2 -0.1 47 Kyrgyz Republic . . .. .. .. .. .. .. .. .. .. . . . 48 Sri Lanka 19.1 18.0 5.4 10.8 24.7 22.1 16.6 6.5 1.7 2.6 40.5 46.2 -18.3 -0.1 49 Armenia . Middle-income economies Lower-middle-income 50 Lesotho 29.5 44.4 3.5 7.5 32.9 33.3 .. .. .. .. 31.4 51 EgyptArab Rep. 28.9 26.3 4.0 5.6 366 34.8 9.0 8.0 13.5 8.7 32.1 39.1 -63 2.0 52 Bolivia 11.8 7.1 .. 18.7 5.4 .. 8.2 .. 52.3 .. -3.6 53 Macedonia, FYR 54 Moldova 55 Uzbekistan ,. .. .. .. .. .. .. .. .. .. .. 56 Indonesia 20.2 16.4 1.8 62 11.7 8.4 10.4 7.8 13.5 62 23.7 70.4 2.3 0.6 57 Philippines 12.5 16.0 5.9 4.9 9.9 15.4 3.5 3.0 15.7 10.6 25.4 263 -1.4 -1.5 58 Morocco 20.4 . . 8.1 . . 22.8 . . 10.3 . . 17.9 . . 39.5 . . -9.7 59 Syrian Arab Republic 10.5 17.8 1.4 8.2 30.3 15.5 17.9 11.1 35.8 28.2 28.0 268 -9.7 3.8 60 Papua New Guinea 20.5 18.9 2.8 2.3 29,2 261 5.2 3.3 4.4 3.3 29.8 30.5 -1.9 -4.1 61 Bulgaria . . 29.0 . . 10.4 . . 41.4 1.6 6.3 . . 35.0 . . -5.5 62 Kazakstan .. .. .. .. .. .. .. .. .. .. .. 63 Guatemala 8.7 6.8 2.9 3.7 7.3 69 5.1 2.3 10,6 15.2 47.5 38.4 -3.4 -1.2 64 Ecuador 12.2 13.9 2.2 4.1 11.9 12.4 2.3 3.3 12.5 . . 52.5 . . -1.4 0.0 65 Dominican Republic 11.1 14.9 3.1 5.5 11.4 8.2 5.2 8.7 7.8 4.7 53.0 871 -2.6 0.0 66 Romania 10.1 26.3 0.0 69 29.8 27.7 15.0 4.3 3.8 6.2 28.3 54.8 0.5 0.0 67 Jamaica 27.9 ,. 14.3 ., ,, .. .. .. .. .. .. -15.5 68 Jordan .. 20.4 65 24.7 .. 6.1 25.3 20.7 38.5 468 .. 1.1 69 Algeria . 0 0 0 0 0 0 0 0 70 El Salvador 11.1 12.1 3.4 6.4 11.7 11.2 2.8 3.7 8.8 5.2 47.0 37.2 -5.7 0.0 71 Ukraine .. ,. .. ., .. .. .. .. .. .. .. 72 Paraguay 9.8 9.1 1.9 5.0 7.5 11.1 2.4 1.9 12.4 10.7 47.0 54.1 0.3 1.2 Note: For data comparabiliry and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 241 Percentage of GDP Total revenues Total expenditure Percentage of total expenditure1' deficit/Surpluse Tax Nontax Current Capital Defense Social servicesd 1% of GOPI 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 73 Tunisia 23.9 . . 7.5 . . 22.1 . . 9.4 . . 12.2 . . 53.7 . . -2.8 74 Lithuania .. 24.4 .. 12.6 .. 24.7 .. 2.7 .. 1.9 .. 53.2 75 Colombia 10.3 14.0 2.7 68 10.4 11.9 4.1 2.5 67 8.7 58.5 40.8 -1.8 -0.5 76 Namibia .. 31.4 10.1 .. 34.6 .. 61 .. .. -4.8 77 Belarus .. ,, .. .. .. .. 78 Russian Federation .. 16.1 6.3 .. 25.8 .. 1.3 -. 16.4 .. 34.6 .. -10.5 79 Larvia . . 23.1 . . 10.9 . . 29.2 . . 1.2 . . 2.6 . . 63.7 . . -4.2 80 Peru 15.8 14.4 6.4 7.7 15.0 15.8 4.4 3.1 21.0 . . 27.4 0.0 -2.4 0.0 81 Costa Rica 16.8 22.0 5.4 8.4 21.3 26.0 5.2 2.4 2.6 0.0 73.9 63.0 -7.4 -2.9 82 Lebanon .. 10,8 .. 1.1 .. 25.8 .. ,. .. .. 83 Thailand 13.2 17.1 6.6 7.4 14.4 10.5 4.4 . . 21.7 . . 37.8 57.7 -4.9 1.8 84 Panama 19.7 20.1 4.5 4.8 26.5 25.4 5.9 2.9 .. 5.4 48.5 69.6 -5.5 4.3 85 Turkey 14.3 14.3 3.6 7.3 15.5 20.9 5.9 6.1 15.2 15.8 33.0 21.6 -3.1 0.0 86 Poland .. 36.7 11.6 .. 41.9 1.5 .. .. .. .. -2.3 87 Estonia .. 33.2 .. 13.6 .. .. 3.1 .. .. 1.4 88 Slovak Republic 89 Botswana 24.9 28.1 0.3 1.7 23.1 38.3 10.8 . . 9.8 41.5 356 -0.2 90 Venezuela 18.9 14.8 0.9 4.0 14.9 163 4.0 2.5 5.8 48.0 0.0 -4.1 Upper-middle-income 91 South Africa 20.5 25.2 5.6 10.0 19.1 31.0 3.0 2.2 -2.3 -62 92 Croatia . . 43.0 . . 18.7 . . 42.8 . . 3.7 21.1 . . 60.9 . . -0.9 93 Mexico 14.3 14.8 4.4 8.0 11.3 14.3 5.5 . . 2.3 57.6 65.7 -3.0 94 Mauritius 18.4 18.2 3.6 5.5 22.7 19.1 4.6 4.2 0.8 1.5 55.9 60.0 -10.3 -1.4 95 Gabon 23.6 . . 1.7 6.1 96 Brazil 17.8 18.6 7.3 4.9 18.6 39.0 2.2 .. .. 43.5 34.5 -2.2 97 Trinidad and Tobago 35.7 .. 1.6 .. 18.4 .. 12.0 . . .. .. .. 7.2 98 Czech Republic . . 37.5 . . 13.0 . . 36.9 . . 5.1 . . 5.7 . . 65.7 . . 0.5 99 Malaysia 23.4 20.6 4.4 6.6 19.2 18.1 9.9 5.1 14.8 12.7 45.3 48.0 -6.0 0.8 100 Hungary 44.9 .. 20.5 .. 48.7 .. 7.5 .. 4.4 .. 31.0 .. -2.8 101 Chile 25.6 17.8 11.4 9,7 25.3 16.2 2.7 3.3 12.4 8.8 65.3 77.9 5.4 1.6 102 Oman 10.7 8.5 0.2 0.4 30.3 36.0 8.2 69 51.2 365 9.7 26.3 0.4 -11.2 103 Uruguay 21.0 27.6 9.6 9.6 20.1 29.2 1.7 2.6 13.4 7.3 67.6 79.8 0.0 -2.8 104 Saudi Arabia .. .. .. .. .. .. .. .. .. 105 Argentina 10.4 . . 2.6 18.2 . . 2.7 . . 14.3 28.6 . . -2.6 0.0 106 Slovenia . . 107 Greece 27.4 26.0 9.7 17.8 31.1 38.8 5.5 4,412.6 8.9 58.8 34.1 -5.0 -15.7 Low- and middle-income Sub-Saharan Africa East Asia and Pacific South Asia Europe and Central Asia Middle East and N. Africa Latin America and Caribbean Hish-income economies 108 Korea, Rep. 15.3 17.7 8.0 6.5 14.6 14.2 2.4 3.6 34.3 18.1 30.0 42.1 2.2 -0.2 109 Portugal 24.3 30.9 8.8 12.8 28.9 37.7 4.4 . . 7.4 . . 55.9 . . -8.5 110 Spain 22.2 28.7 3.1 65 23.8 368 3.0 2.6 4.3 6.8 77.4 54.0 -4.2 0.0 111 New Zealand 30.6 34.4 6.1 10.8 35.7 35.0 2.4 1.2 5.1 3.6 63.8 70.1 -6.7 0.1 112 Ireland 30.9 35.1 10.4 11.4 40.4 39.3 4.6 3.2 3.4 3.0 57.7 61.6 -12.5 -0.2 113 lIsrael 43.3 33.4 12.4 13.0 67.2 40.9 2.8 4.3 39.8 19.4 29.0 57.5 -15.6 -2.9 114 tKuwait 2.7 1.2 0.2 0.0 18.9 44.1 8.9 7.3 12.2 25.5 39.1 44.2 58.7 115 tUnited Arab Emirates 0.0 0.6 0.0 0.6 11.2 11.3 0.9 0.5 47.5 37.1 24.0 29.2 2.1 0.2 116 United Kingdom 30.6 33.5 9.8 11.8 36.4 39.6 1.8 4.4 13.8 . . 48.2 54.5 -4.6 0.1 117 Australia 19.6 22.3 5.1 5.2 21.1 26.6 1.5 2.5 9.4 16.8 50.1 58.0 -1.5 -0.1 118 Italy 29.1 38.4 7.7 11.2 37.6 48.0 2.2 1.9 3.4 .. 55.3 . . -10.7 -10.5 119 Canada 16.2 . . 3.1 .. 21.0 . . 0.3 . . 7.7 10.8 46.1 . . -3.5 120 Finland 25.1 29.3 13.3 14.2 25.2 42.0 3.0 1.7 5.6 3.9 59.7 63.5 -2.2 -13.4 121 tHongKong . . . . . . . . . . . . . . . . . . . . . . . . . 122 Sweden 30.1 32.8 10.2 11.2 37.5 43.6 1.8 1.4 7.7 5.6 67.0 64.6 -8.1 -6.9 123 Netherlands 44.2 42.9 10.3 10.5 48.2 48.3 4.6 2.5 5.6 3.9 70.5 68.7 -4.6 -4.9 124 Belgium 41.7 43.7 10.5 11.4 46.6 47.9 4.3 5.5 5.7 7.0 67.0 7.2 -8.2 -0.5 125 France 36.7 38.1 12.2 11.5 37.4 44.4 2.1 2.4 7.4 .. 74.1 72.5 -0.1 -5.5 126 tSingapore 17.5 17.2 4.0 4.6 15.6 10.7 4.5 4.2 25.2 37.4 37.5 48.5 2.1 0.0 127 Austria 32.0 32.9 8.8 9.1 34.0 37.5 3.4 2.9 3.0 3.7 78.8 77.8 -3.4 -0.1 128 United States 18.5 19.0 0.9 0.8 20.7 22.1 1.4 0.8 21.2 18.1 54.2 55.0 -2.8 -2.3 129 Germany . . 30.0 . . 7.3 . . 32.2 . . 1.7 9.1 . . 74.9 . . . . -2.5 130 Denmark 31.3 35.4 16.7 16.5 36.7 42.0 2.7 1.5 6.5 4.0 61.2 56.9 2.7 2.0 131 Norway 33.9 31.6 14.8 15.4 33.4 39.1 1.2 . . 7.7 . . 559 51.1 -1.7 132 Japan 11.0 17.6 2.4 3.0 14.8 . . 3.6 . . . . 4.1 . . . . -7.0 0.0 133 Switzerland 18.3 21.5 3.8 3.4 18.9 25.9 1.4 1.2 10.2 15.2 69.0 75.2 -0.2 0.1 World a. Refers to current revenue. h. 1cludes IrIsdiug minus repaynsents. c. Includes grants. d. Refers tcs eds,cation, health, social security, welfare, housing, and comnsuniry amenities. e. Includes Erirrea. 242 WORLD DEVELOPMENT REPORT 1997 Table 15. Exports and imports of merchandise Exports Imports Total Manufactures Total Food Fuel Average annual growth rate (%l (million SI (% of total) (million 5) (% of total) (% of total) Export volume Import volume 1980 1995 1980 1993 1980 1995 1980 1993 1980 1993 1980-90 1990-95 1980-90 1990-95 Low-income economies 84,204 t 245,456 t 97,748 t 251,806 5.3w 8.3w 1.6w 13.0w 2.7w -4.2w Excluding China and India 2 1 Mozambique Ethiopiaa 58,817 t 281 425 64,769 t 169 423 0 2 20 4 65,465 t 800 717 86,058 t 784 1,033 8 ........ -10.5 6 25 11 1.4w 1.2 -0.3 -1.0 5.0w 2.9 -9.4 3.3 -3.3 3 Tanzania 511 639 14 . 1,250 1,619 13 . 21 -1.8 10.0 -3.3 12.7 4 Burundi 65 106 3 30 168 234 7.4 -4.8 1.4 -14.6 5 Malawi 295 325 7 6 439 491 8 15 0.1 -1.8 1.3 -1.6 6 Chad 71 156 8 74 220 5.4 -10.0 10.5 -12.1 7 Rwanda 72 45 0 243 235 12 13 5.6 -19.6 1.3 -1.9 8 Sierra Leone 224 42 40 27 427 135 24 2 -2.1 -4.3 -9.9 -1.1 9 Nepal 80 348 84 342 1,374 4 7.8 68 ........ 31 18 22.1 4.9 10 Niger 566 225 2 594 309 14 .. 26 -6.4 -2.0 -4.5 2.5 11 Burkina Faso 161 274 359 549 5.4 1.3 2.1 8.3 12 Madagascar 401 364 6 20 600 499 9 11 15 12 -0.1 -68 -4.6 -5.6 13 Bangladesh 793 3,173 69 81 2,600 6,496 24 15 10 14 7.5 12.7 1.8 5.3 14 Uganda 345 461 3 1 293 1,058 8 30 . -1.4 3.9 -0.6 28.7 15 Vietnam 339 5,026 1,310 7,272 . . . 16 Guinea-Bissau 11 23 55 70 20 6 -5.1 -18.3 1.3 -5.4 17 Haiti 226 110 . 375 653 24 13 -2.9 -11.2 -4.4 -6.8 18 Mali 205 326 9 439 529 19 35 2.6 -3.7 1.2 -3.4 19 Nigeria 26,000 11,670 3 16,700 7,900 -2.4 -1.9 -17.5 7.6 ........ 0 17 2 20 Yemen, Rep. 1,937 51 1,962 28 7 1.5 7.2 -5.9 11.1 21 Cambodia .. 855 .. 1,213 . . . 22 23 24 Kenya Mongolia Togo 1,250 338 .. 1,878 324 209 12 11 18 6 2,120 551 .. 223 ........ 2,949 386 17 8 23 8 34 23 33 10 2.6 4.9 .. 166 9.0 1.1 1.1 -5.6 -11.2 25 Gambia, The 31 16 9 37 165 140 23 . 11 2.3 269 1.0 9.0 26 Central African Republic ........ 116 187 29 . 81 174 21 . 2 2.5 3.5 6.0 -3.3 27 India 8,590 30,764 59 75 14,900 34,522 9 4 45 30 6.3 7.0 4.5 2.7 28 Lao PDR 31 348 8 29 587 . . 29 Benin 63 163 8 331 493 26 8 7.7 -0.3 -6.3 29.4 30 Nicaragua 451 520 14 7 887 962 15 23 20 15 -4.4 -8.7 -4.1 7.3 31 Ghana 1,260 1,227 1 23 1,130 1,580 10 27 3.9 9.1 1.6 12.8 269 -62 ........ 32 Zambia 1,300 781 1,340 1,258 5 22 -3.5 -5.0 33 Angola 1,880 3,508 13 . 1,330 1,748 24 1 11.3 4.2 -3.4 -4.1 34 Georgia .. 347 .. 687 . . 35 Pakistan 2,620 ........ 7,992 49 85 5,350 11,461 13 14 27 17 9.5 8.8 2.1 10.3 36 Mauritania 194 404 2 1 286 700 30 14 7.8 3.5 1.1 4.4 ........ 37 Azerbaijan . . 612 . . 955 . . . . . 38 Zimbabwe 1,415 1,885 38 37 1,448 2,241 3 18 1 12 2.2 -66 -2.2 -5.1 39 Guinea 401 583 270 690 -3.6 -8.6 -2.9 -2.8 40 Honduras 830 1,061 13 13 1,010 1,219 10 11 16 13 1.3 10.7 -1.0 7.0 41 Senegal 477 340 15 21 1,050 704 25 29 25 11 2.6 3.6 1.0 61 42 Chioa* 18,100 148,797 48 81 19,900 129,113 . 3 . 6 11.4 14.3 10.0 24.8 43 Cameroon 1,380 2,331 4 13 1,600 1,241 9 16 12 3 4.5 1.7 1.4 11.2 44 Côte d'lvoire 3,130 3,939 . 17 2,970 2,808 13 . 16 3.3 _7.5 -4.0 5.4 45 Albania 367 205 354 679 . . 46 47 48 Congo Kyrgyz Republic Sri Lanka 911 1,070 .. 952 409 3,798 16 7 73 580 .. 2,040 610 ........ 670 5,185 19 20 16 14 24 9 5.5 6.3 .. 9.7 .. 17.0 -2.0 2.0 2.5 15.0 49 Armenia .. 271 .. .. 674 .. . . Middle-income economies 586,567 t 893,331 t 455,925 t 987,309 t 2.6w 69w -0.2w 11.0w Lower-middle-income 50 51 Lesotho Egypt,ArabRep. 58 3,050 143 3,435 11 33 464 4,860 821 ......... 11,739 32 24 1 2 -0.2 . . -0.1 -0.7 -2.9 52 Bolivia 942 1,101 3 19 665 1,424 19 9 1 5 1.7 -5.4 2.8 18.9 53 Macedonia, FYR 1,244 .. 1,420 54 55 56 57 Moldova Uzbekistan Indonesia Philippines 21,900 5,740 . . 746 3,805 45,417 17,502 37 2 .. 53 76 . 10,800 8,300 . 841 3,598 40,918 28,337 ........ 13 7 16 28 8 . 5.3 2.9 . . 21.3 . . 1.2 2.4 9.1 8 8 12 10.2 15.2 58 Morocco 2,490 4,802 24 57 4,160 8,563 20 17 24 14 4.2 0.8 2.9 1.7 59 Syrian Arab Republic 2,110 3,970 6 9 4,120 4,616 14 19 26 4 6.4 -3.2 -9.3 22.3 60 Papua New Guinea 1,030 2,644 12 ........ 3 1,180 1,451 21 . 15 . 4.5 19.3 -0.2 2.1 61 Bulgaria 10,400 5,100 9,650 5,015 . 8 36 62 Kazakstan .. 5,197 .. 5,692 . . 63 Guatemala 1,520 2,156 24 30 1,600 3,293 8 11 24 14 -1.3 8.2 -0.6 19.3 64 Ecuador 2,480 4,307 3 7 2,250 4,193 8 5 1 2 3.0 8.9 -3.9 10.0 65 Dominican Republic 962 765 24 52 1,640 2,976 17 25 -1.0 -10.2 2.6 8.9 66 Rumania 11,200 7,548 77 12,800 9,424 14 26 -6.8 -4.7 -0.9 5.3 67 Jamaica 963 1,414 63 65 1,100 2,757 20 14 38 19 1.2 1.3 3.1 7.0 68 Jordan 574 1,769 34 51 2,400 3,698 18 20 17 13 7.4 7.1 -3.1 13.0 69 Algeria 13,900 8,594 29 ........ -0.8 0 3 10,600 9,570 21 3 1 2.5 -5.1 -5.7 70 El Salvador 967 998 35 48 966 2,853 18 15 18 14 -2.8 13.0 1.3 162 71 Ukraine . . 13,647 . . 15,945 . . . . . . . 72 Paraguay 310 817 12 17 615 2,370 11 . . 12 9.9 -1.9 3.2 7.3 * Data for Taiwan, China 19,800 111,585 88 93 19,700 103,698 8 6 25 8 11.6 5.9 12.8 14.1 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 243 Exports Imports Total Manufactures Total Food Fuel Average annual growth rate (%) (million SI (% of total) (million 5) 1% of totall 1% of total) Export volume Import volume 1980 1995 1980 1993 1980 1995 1980 1993 1980 1993 1980-90 1990-95 1980-90 1990-95 73 Tunisia 2,200 5,475 36 75 3,540 7,903 14 8 21 8 6.2 7.7 1.3 64 74 Lithuania .. 2,707 64 3,083 11 45 .. . 75 Colombia 3,920 9,764 20 40 4,740 13,853 12 8 12 4 9.7 4.8 -1.9 22.3 76 Namibia 1,353 1,196 77 Belarus 4,621 5,149 .. ........ . . 78 Russian Federation 81,500 58,900 . . . . 79 Latvia . . 1,305 . 1,818 . 80 Peru 3,900 5,575 18 17 2,500 9,224 20 20 2 8 -1.9 11.0 -1.0 12.1 81 CosraRica 1,000 2,611 34 33 1,540 3,253 9 8 15 9 4.9 10.1 2.8 15.1 82 Lebanon 868 982 65 3,650 6,721 16 15 -1.2 -7.8 -7.4 23.5 83 Thailand 6,510 56,459 28 73 9,210 70,776 5 5 30 8 14.3 21.6 12.1 12.7 84 Panama 358 625 9 16 1,450 2,511 10 10 31 13 2.6 23.3 -4.1 14.3 85 Turkey 2,910 21,600 27 72 7,910 35,710 4 6 48 14 12.0 8.8 11.3 11.2 86 Poland 14,200 22,892 71 60 16,700 29,050 14 12 18 17 4.8 3.9 1.5 264 87 Estonia 1,847 2,539 ........ . . . 88 Slovak Republic . . 8,585 . . 9,070 . . . . . 89 Botswana 502 2,130 692 1,907 11.4 -0.8 7.7 -5.6 90 Venezuela 19,221 18,457 2 14 11,827 11,968 15 11 2 1 1.6 -0.1 -6.1 19.3 Upper-middle-income 246,329 t 372,898 t 161,848 t 379,450 1.7w 7.3w -0.6w 12.6w 91 South Africa 25,500 27,860 39 74 19,600 30,555 3 6 0 1 0.9 2.8 0.8 5.3 92 Croaria . - 4,633 . - 71 .. 7,582 . - 9 . 10 .. . . . 93 Mexico 15,600 79,543 12 75 19,500 72,500 16 8 2 2 12.2 14.7 5.7 18.7 94 Mauritius 431 1,537 27 67 609 1,959 26 13 14 9 8.6 2.0 11.0 2.5 95 Gabon 2,170 2,713 4 674 882 19 1 0.6 5.7 -2.0 2.0 96 Brazil 20,100 46,506 60 25,000 66 -1.5 ........ 39 53,783 10 10 43 16 6.1 8.5 97 TrinidadandTohago 3,960 2,455 4 34 3,160 1,714 11 15 38 16 4.3 4.9 12.1 8.1 98 Czech Republic . . 21,654 . . 26,523 . . . . . 99 Malaysia 13,000 74,037 19 65 10,800 77,751 12 7 15 4 11.5 17.8 6.0 15.7 100 Hungary 8,670 12,540 66 68 9,220 15,073 8 6 16 13 3.0 -1.8 0.7 7.9 101 Chile 4,710 16,039 10 18 5,800 15,914 15 6 18 10 5.7 10.5 1.4 14.5 102 Oman 2,390 6,065 3 70 1,730 4,248 15 19 11 3 13.1 9.8 -1.6 18.5 103 Uruguay 1,060 2,106 38 43 1,680 2,867 8 8 29 9 2.9 -3.1 -2.0 21.7 104 Saudi Arabia 109,000 46,624 1 9 30,200 27,458 14 .. 1 .. -8.2 4.0 -8.4 5.9 105 Argenrina 8,020 20,967 23 32 10,500 20,122 6 5 10 2 3.1 -1.0 -8.6 45.8 106 Slovenia . . 8,286 . . 86 . . 9,452 . . 8 . . 11 . . . . . - 107 Greece 5,150 9,384 47 48 10,500 21,466 9 6 23 25 5.1 11.9 5.8 12.8 Low- and middle-income 660,833 t 1,152,249 t 547,4171 1,233,7491 3.0w 7.2w 0.2w 11.4w Sub-Saharan Africa 77,237t 72,8471 66,593 t 77,574t 0.9w 0.9w -3.8w 1.9w East Asia and Pacific 69,6231 359,1021 65,1391 368,6831 9.3w 17.8w 7.1w /7.0w South Asia 13,848 t 46,4551 25,8631 60,512 t 6.6w 8.6w 3.5w 5.3w Europe and Central Asia .. . . . . . . . . . . Middle East and N. Africa 203,379 t 106,441 t 103,8501 110,841 t -2.0w i.lw -5.8w 5.9w Latin America and Caribbean 98,589 t 221,210t 107,971 t 237,5761 5.2w 66w -O.5w 15.1w High-income economies 1,393,926 t 3,997,2881 1,503,7431 4,037,671 t 5.2w 5.4w 6.2w 4.6w 108 Korea, Rep. 17,500 125,058 90 93 22,300 135,119 10 6 30 18 13.7 7.4 11.2 7.7 109 Portugal 4,640 22,621 72 78 9,310 32,339 14 19 24 24 12.2 0.5 9.8 2.4 110 Spain 20,700 91,716 72 78 34,100 115,019 13 14 39 11 6.9 11.2 10.1 5.3 111 NewZealand 5,420 13,738 20 27 5,470 13,958 6 8 23 7 3.6 5.4 4.6 5.5 112 Ireland 8,400 44,191 58 75 11,200 32,568 12 10 15 5 9.3 11.4 4.7 5.6 113 tlsrael 5,540 19,046 82 91 9,780 29,579 11 7 27 7 5.9 10.0 4.6 12.3 114 tKuwait 19,700 12,977 10 88 6,530 7,784 15 13 1 1 -2.0 42.3 -6.3 23.0 115 t United Arab Emirates 20,700 25,650 3 . 8,750 21,024 11 . 11 6.1 63 -1.3 21.0 116 United Kingdom 110,000 242,042 74 82 116,000 263,719 13 11 14 5 4.4 1.8 6.3 0.9 117 Australia 21,900 52,692 20 35 22,400 61,280 5 5 14 6 5.8 8.1 4.9 5.1 118 Italy 78,100 231,336 85 89 101,000 204,062 13 13 28 10 4.3 60 5.3 1.7 119 Canada 67,700 192,198 49 66 62,500 168,426 8 6 12 4 5.7 8.4 6.2 63 120 Finland 14,200 39,573 70 83 15,600 28,114 7 7 29 13 2.3 8.7 4.4 -1.9 121 tHong Kong 19,800 173,754 92 95 22,400 192,774 12 6 6 2 15.4 15.3 11.0 15.8 122 Sweden 30,900 79,908 79 85 33,400 64,438 7 8 24 9 4.6 7.4 4.9 5.0 123 Netherlands 74,000 195,912 51 63 76,600 176,420 15 15 24 9 4.5 5.8 4.6 4.3 124 Belgiumb France 64,500 116,000 136,864 71,900 135,000 125,297 ........ 27 9 4.4 4.1 4.2 2.3 4.0 5.0 0.3 0.8 125 286,738 74 78 275,275 10 11 126 tSingapore 19,400 118,268 50 80 24,000 124,507 9 6 29 11 12.2 162 8.6 12.1 127 Austria 17,500 45,200 83 89 24,400 55,300 6 5 16 5 6.4 3.9 5.8 1.9 128 United States 226,000 584,743 68 82 257,000 770,852 8 5 33 10 3.6 5.6 7.2 7.4 129 Germanyc 193,000 523,743 86 90 188,000 464,220 12 10 23 8 4.6 2.2 4.9 2.9 130 Denmark 16,700 49,036 56 66 19,300 43,223 12 13 22 6 4.4 5.4 3.6 3.4 131 Norway 18,600 41,746 32 31 16,900 32,702 8 7 17 3 6.8 65 4.2 0.7 132 Japan 130,000 443,116 96 97 141,000 335,882 12 18 50 21 5.0 0.4 6.5 4.0 133 Switzerland 29,600 77,649 91 94 36,300 76,985 8 7 11 4 6.0 3.3 4.9 -67 World 2,003,797 t 5,144,7701 2,027,0781 5,246,3261 4.7w 6.0w 4.9w 5.8 w a. Data prior to 1992 include Eritrea. b. Includes Luxembourg. c. Data prior to 1990 refer to the Federal Republic of Germany before unification. 244 WORLD DEVELOPMENT REPORT 1997 Table 16. Balance of payments Exports of Imports of Current transfers goods, services, goods, services, Net workers' Other net Current account Gross international and income and income remittances transfers balance reserves (million $1 (millionS) (million $1 (million $1 (million $1 (million 5) 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 Low-income economies Excluding China and India 1 Mozambique 452 490 875 1,368 0 56 .. -367 .. 2 Ethiopiaa 590 828 797 1,400 80 532 -126 -93 262 815 3 4 5 Tanzania Burundi Malawi 762 315 1,253 139 419 1,412 638 . 2,236 297 937 .. 0 0 0 0 0 128 63 354 151 124 -522 -260 -629 -6 -450 20 105 76 270 216 115 6 Chad 71 274 83 540 -4 -15 25 206 9 -38 12 147 7 Rwanda 182 91 335 338 -14 -7 118 196 -48 -129 187 126 8 Sierra Leone 276 137 494 374 -2 0 54 47 -165 -89 31 52 9 Nepal 239 1,110 368 1,592 . . 36 108 -93 -375 272 646 10 Niger 644 291 1,016 496 -47 -41 143 108 -277 -126 132 99 11 Burkina Faso 225 356 596 652 100 29 223 226 -49 15 75 352 12 Madagascar 518 756 1,121 1,161 -30 -2 77 131 -556 -276 9 109 13 Bangladesh 976 4,292 2,622 6,747 . 802 1,426 -844 -1,029 331 2,376 14 Uganda 331 642 450 1,440 0 36 370 -83 -428 3 459 15 Vietnam . . 7,368 . 9,865 . . . . . . 477 -792 -2,021 0 3 16 Guinea-Bissau 17 24 83 95 -14 -1 . 31 -80 -41 . . 20 17 Haiti 309 209 498 780 52 0 36 505 -101 -67 27 106 18 Mali 263 533 537 967 40 69 104 162 -130 -164 26 330 19 Nigeria 27,759 9,879 22,005 50,427 -410 2,567 -166 -1,894 5,178 -510 10,640 1,709 20 Yemen, Rep. . 2,154 3,075 .. .. 1,067 . 146 638 21 Cambodia . 979 . . 1,442 . . 10 . . 267 . -186 . . 192 22 Kenya 2,061 2,974 3,095 3,874 0 -4 156 503 -878 -400 539 384 23 Mongolia 476 511 1,283 550 0 0 0 77 -808 39 .. 158 24 Togo 570 520 752 655 1 5 85 25 -95 -57 85 135 25 Gambia, The 66 181 181 241 0 0 28 52 -87 -8 6 106 26 Central African Republic 205 234 329 319 -19 -27 100 91 -43 -25 62 238 27 India 12,348 40,995 18,130 54,303 . . 2,860 7,478 -2,922 -5,830 12,010 22,865 28 Lao PDR .. 453 .. 673 .. 0 -3 .. -224 . . 99 29 Benin 241 489 428 769 75 65 76 84 -36 36 15 202 30 Nicaragua 514 655 1,049 1,436 0 75 124 0 -411 -706 75 142 31 Ghana 1,213 1,586 1,264 2,264 -4 12 84 264 30 -414 330 804 32 Zambia 1,625 1,500 1,987 1,691 -61 . . -93 . . -516 . . 206 192 33 Angola .. 3,655 4,701 . -83 .. 249 . . -769 34 Georgia .. .. .. .. .. .. .. .. 35 Pakistan 3,011 8,403 6,042 12,758 .. .. 1,895 2,390 -1,137 1,965 1,568 2,528 36 Mauritania 270 533 493 636 -27 -20 117 76 -134 27 146 90 37 Azerbaijan . . . .. 1,273 . . . . .. 110 . . -379 . . 84 38 Zimbabwe 1,714 2,372 1,895 2,836 8 -2 23 41 -149 -425 419 888 39 Guinea . . 714 .. 1,090 . . -10 .. 189 . . -197 . . 87 40 Honduras 967 1,667 1,306 2,110 0 120 22 123 -317 -201 159 270 41 Senegal 830 1,501 1,337 1,898 -15 13 135 375 -387 3 25 283 42 China * 24,729 152,431 19,541 152,248 538 350 -52 1,085 5,674 1,618 10,091 80,288 43 Cameroon 1,813 2,070 2,478 2,250 . . . . -17 9 -682 -171 206 15 44 Côte d'Ivoire 3,640 4,527 4,761 4,502 -716 -449 10 155 -1,826 -269 46 546 45 Albania 386 376 375 865 0 385 6 93 16 -12 . . 265 46 Congo 1,029 1,252 1,195 1,825 -38 -27 37 29 -167 -570 93 64 47 Kyrgyz Republic .. 340 .. 490 .. . . .. 43 . . -288 48 Sri Lanka 1,340 4,843 2,269 6,041 152 715 121 75 -657 -546 283 2,088 49 Armenia .. 301 741 .. 12 .. 148 -279 Middle-income economies Lower-middle-income 50 Lesotho 363 663 482 1,021 0 0 175 471 56 108 50 457 51 Egypt,ArabRep. 6,516 11,337 9,745 17,353 2,696 3,417 95 5,060 -438 956 2,480 17,122 52 Bolivia 1,046 1,283 1,112 1,794 0 -1 60 226 -6 -218 553 1,005 53 Macedonia, FYR . . 1,321 2,184 . . . . . . . . . . . 275 54 Moldova . . 865 999 40 -95 240 55 Uzbekistan . . 3,746 . . 3,253 . . . . . . . . . . -8 56 Indonesia 24,878 52,505 25,694 60,367 0 629 250 210 -566 -7,023 6,803 14,908 57 Philippines 7,997 32,862 10,348 35,722 202 296 232 584 -1,917 -1,980 3,978 7,757 58 Morocco 3,270 9,118 5,807 12,900 989 1,890 141 371 -1,407 -1,521 814 3,874 59 Syrian Arab Republic 2,568 5,929 4,610 6,406 774 385 1,520 532 251 440 828 . 60 Papua New Guinea 1,089 3,014 1,561 2,415 0 0 184 75 -289 674 458 267 61 Bulgaria 9,443 6,680 8,547 6,478 0 0 58 132 954 334 62 Kazakstan . . 5,296 .. 5,874 . . . . . 59 . 519 . 1,660 63 Guatemala 1,834 2,868 2,107 3,933 0 350 110 144 -163 572 753 783 64 Ecuador 2,975 5,298 3,647 6,351 0 0 30 231 -642 -822 1,257 1,788 65 Dominican Republic 1,313 5,106 2,237 6,100 183 795 21 -266 -720 -125 279 373 66 Romania 12,160 9,094 14,580 10,799 0 3 0 360 -2,420 -1,342 2,511 2,624 67 Jamaica 1,422 3,327 1,678 4,107 51 414 70 121 -136 -245 105 681 68 Jordan 1,782 3,606 3,318 5,200 . . . 594 1,118 -942 -476 1,745 2,279 69 Algeria 14,500 10,954 14,552 12,512 241 . . 60 168 249 -2,310 7,064 4,164 70 El Salvador 1,271 2,103 1,289 3,562 11 1,061 41 328 34 -70 382 940 71 Ukraine . 17,337 . . 18,961 . . . . 472 . . -1,152 . . 1,069 72 Paraguay 781 1,399 . . 4,173 2 . -2 . 42 -618 -1,473 783 1,040 * Data for Taiwan, China 22,627 134,484 23,445 126,626 -95 -2,202 -913 5,656 4,055 95,559 Note: For data comparability and coverage, see the technical notes. Figures irs italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 245 Exports of Imports of Current transfers goods, services, goods, services, Net workers' Other net Current account Gross international and income and income remittances transfers balance reserves (million SI (million $1 (million $1 (million $1 (million $1 (million $1 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 1980 1995 73 Tunisia 3,356 8,098 4,119 9,646 304 659 106 152 -353 -737 700 1,689 74 Lithuania 3,242 .. 3,966 .. 1 .. 109 .. -614 .. 829 75 Colombia 5,860 14,794 6,231 19,588 68 172 97 506 -206 -4,116 6,474 8,205 76 Namibia 1,899 .. 2,082 4 230 . . 50 . . 225 77 Belarus 2,773 3,209 . . . . . . 182 -254 . . 377 78 Russian Federation 95,100 . . 85,800 . . . . . . 304 9,604 . . 18,024 79 Latvia 2,151 .. 2,246 .. . . .. 68 -27 .. 602 80 Peru 4,832 7,382 5,080 12,097 0 334 147 157 -101 -4,223 2,804 8,653 81 Costa Rica 1,219 3,945 1,897 4,241 0 0 15 154 -664 -143 197 1,060 82 Lebanon 1,512 . . 6,953 . . . . . 350 . . -5,092 7,025 8,100 83 Thailand 8,575 74,093 10,861 88,134 0 0 210 487 -2,076 -13,554 3,026 36,939 84 Panama 7,853 9,542 8,225 9,584 -36 -7 76 210 -331 -141 117 782 85 Turkey 3,672 38,069 9,251 44,904 2,071 3,327 100 1,169 -3,408 -2,339 3,298 13,891 86 Poland 16,200 33,169 20,338 36,929 0 35 721 -520 -3,417 -4,245 574 14,957 87 Estonia 2,801 3,112 -1 127 -184 583 88 Slovak Republic 11,185 .. 10,629 . . 0 . . 93 . . 648 . . 3,863 89 Botswana 748 2,908 954 2,539 -17 -157 72 129 -151 342 344 4,764 90 Venezuela 22,232 22,406 17,065 20,262 -418 -173 -21 284 4,728 2,255 13,360 10,715 Upper-middle-income 91 South Africa 29,258 33,471 25,989 36,994 0 0 239 23 3,508 -3,500 7,888 4,464 92 Croatia 7,375 .. 9,733 .. .. . . 646 .. -1,712 . . 2,036 93 Mexico 23,987 93,529 35,243 98,145 687 3,672 147 290 -10,422 -654 4,175 17,046 94 Mauritius 579 2,402 718 2,525 0 0 22 101 -117 -22 113 887 95 Gabon 2,434 2,793 1,926 2,415 -143 -152 19 0 384 378 115 153 96 Brazil 23,275 56,098 36,250 77,855 -80 2,773 224 848 -12,831 -18,136 6,875 51,477 97 Trinidad and Tobago 3,371 2,875 2,972 2,577 1 30 -43 -35 357 294 2,813 379 98 Czech Republic 29,399 . . 31,345 . . 0 . . 572 . . -1,374 . . 14,613 99 Malaysia 14,836 84,212 15,100 92,440 0 0 -2 163 -266 -4,147 5,755 24,699 100 Hungary 9,780 17,933 10,374 21,528 0 -14 63 1,073 -531 -2,535 . . 12,095 101 Chile 6,276 20,014 8,360 20,214 0 0 113 357 -1,971 157 4,128 14,860 102 Oman 3,852 6,403 2,650 5,671 -362 -1,740 102 29 942 -979 704 1,251 103 Uruguay 1,594 3,679 2,312 4,069 0 0 9 32 -709 -358 2,401 1,813 104 Saudi Arabia 114,208 55,091 62,710 45,583 -4,094 -16,616 -5,901 -1,000 41,503 -8,108 26,129 10,399 105 Argentina 11,202 28,052 15,999 30,874 0 0 23 432 -4,774 -2,390 9,297 15,979 106 Slovenia 10,731 . . 10,812 . . 53 . . -8 . . -37 . . 1,821 107 Greece 8,374 16,835 11,670 27,707 1,066 2,982 21 5,026 -2,209 -2,864 3,607 16,119 Low- and middle-income Sub-Saharan Africa East Asia and Pacific South Asia Europe and Central Asia Middle East and N. Africa Latin America and Caribbean High-income economies 108 Korea, Rep. 22,477 151,826 28,342 160,490 96 486 496 -73 -5,273 -8,251 3,101 32,804 109 Portugal 6,846 35,666 10,916 43,026 2,928 3,348 78 3,783 -1,064 -229 13,863 22,063 110 Spain 33,863 146,042 41,089 149,863 1,647 2,119 -1 2,983 -5,580 1,280 20,474 40,531 111 NewZealand 6,561 18,572 7,630 22,428 143 174 -47 -96 -973 -3,778 365 4,410 112 Ireland 10,418 53,126 13,754 53,530 0 . . 1,204 1,782 -2,132 1,379 3,071 8,770 113 lIsrael 9,858 28,659 13,458 39,750 0 0 2,729 5,600 -871 -5,491 4,055 8,123 114 tKuwait 27,344 19,276 10,463 13,232 -692 -1,347 -888 -499 15,302 4,198 5,425 4,543 115 t United Arab Emirates 2,355 7,778 116 United Kingdom 201,137 458,728 189,683 452,359 0 -4,592 -11,001 . . 6,862 -4,632 31,755 49,144 117 Australia 26,668 74,417 30,702 93,535 .. ..-416 -67 -4,774 -19,184 6,366 14,952 118 Italy 104,979 330,286 116,668 299,954 1,609 98 -507 -4,724 -10,587 25,706 62,428 60,690 119 Canada 77,980 224,135 79,845 232,458 . . . . 173 -370 -1,691 -8,693 15,462 16,369 120 Finland 17,332 50,798 18,620 44,813 0 . . -114 -343 -1,403 5,642 2,451 10,657 121 tHongKong 24,190 219,346 25,448 219,500 . . . . . . . . -1,258 . . . 122 Sweden 39,388 109,063 42,495 101,439 0 106 -1,224 -3,098 -4,331 4,633 6,996 25,909 123 Netherlands 103,143 250,990 102,850 228,460 -320 -423 -828 -5,916 -855 16,191 37,549 47,162 124 Belgiumb 88,925 305,010 92,625 286,809 -270 -393 -961 -2,848 -4,931 14,960 27,974 24,120 125 France 174,118 498,203 174,156 475,234 -2,591 -1,364 -1,578 -5,162 -4,208 16,443 75,592 58,510 126 t Singapore 25,239 159,437 26,695 143,456 0 0 -106 -888 -1,563 15,093 6,567 68,695 127 Austria 29,152 106,474 32,951 110,085 -67 28 -1,531 1 -3,865 -5,113 17,725 23,369 128 United States 344,470 969,220 333,820 1,082,260 -810 -12,230 -7,690 -22,960 2,150 -148,230 171,413 175,996 129 Germanyc 238,177 706,502 238,524 686,512 -4,437 -5,305 -8,422 -35,661 -13,205 -20,976 104,702 121,816 130 Denmark 23,176 92,772 24,891 90,398 0 0 -161 -961 -1,875 1,413 4,347 11,652 131 Norway 28,252 50,837 26,658 45,573 -23 -236 -493 -1,384 1,079 3,645 6,746 22,976 132 Japan 158,230 687,136 167,450 568,143 0 . -1,530 . -7,747 -10,750 111,246 38,919 192,620 133 Switzerland 59,462 154,840 58,524 129,113 -603 -2,519 -537 -1,586 -201 21,622 64,748 68,620 World a. Data prior to 1992 include Eritrea. b. Includes Luxembourg. c. Data prior to 1990 refer to the Federal Republic of Germany before unification. 246 WDPLD DEVELOPMENT REPOPT 1997 Table 17. External debt External debt as percentage of Debt service Ratio of present Multilateral debt as Total external debt Exports of goods as % of exports of value to nominal % of total external (million SI GNP and services goods and services value of debt (%( debt 1980 1995 1980 1995 1980 1995 1980 1995 1995 1980 1995 Low-income economies 106,209 t 534,794 t 16.3w 38.7w 96.8w 183.9w 9.6w 15.4w 17.2w 25.5 v Excluding China and India I Mozambique .. 5,781 .. 443.6 .. 1,192.5 .. 35.3 76.6 . 22.7 2 Ethiopia 824 5,221 .. 99.9 134.5 458.2 7.3 13.6 65.8 41.2 45.3 3 Tanzania 2,460 7,333 .. 207.4 323.0 585.2 21.1 17.4 73.5 23.0 39.1 4 Burundi 166 1,157 18.2 110.1 . . 829.3 .. 27.7 45.2 35.7 80.1 5 Malawi 821 2,140 72.1 166.8 260.8 499.6 27.7 25.9 47.6 26.7 78.8 6 Chad 285 908 39.5 81.4 399.6 339.0 8.4 5.9 48.7 26.1 73.0 7 Rwanda 190 1,008 16.3 89.1 103.5 657.3 4.1 .. 47.6 47.8 80.4 8 Sierra Leone 435 1,226 38.3 159.7 157.7 1,163.5 23.2 60.3 62.6 14.2 34.3 9 Nepal 205 2,398 10.4 53.3 85.4 198.0 3.3 7.8 49.2 62.0 81.3 10 Niger 863 1,633 34.5 91.2 132.8 571.7 21.7 19.8 62.2 16.5 53.2 11 Burkina Faso 330 1,267 19.5 55.0 88.0 346.1 5.9 11.1 51.0 42.8 77.6 12 Madagascar 1,241 4,302 31.1 141.7 239.3 562.2 20.3 9.2 74.1 14.7 39.2 13 Bangladesh 4,230 16,370 32.6 56.3 360.4 298.2 23.7 13.3 55.9 30.2 59.7 14 Uganda 689 3,564 54.6 63.7 208.1 555.1 17.3 21.3 52.4 11.5 61.8 15 Vietnam . . 26,495 . . 130.2 396.0 5.2 88.0 . . 1.2 16 Guinea-Bissau 145 894 137.8 353.7 .. 1,874.3 .. 66.9 65.1 20.1 56.1 17 Haiti 302 807 20.9 39.8 72.8 386.8 6.2 45.2 49.1 43.8 75.7 18 Mali 732 3,066 45.4 131.9 227.3 467.1 5.1 12.6 58.9 23.7 45.2 19 Nigeria 8,921 35,005 10.1 140.5 32.1 274.5 4.1 12.3 94.1 6.4 14.1 20 Yemen, Rep. 1,684 6,212 .. 155.2 .. 192.1 3.2 82.2 14.9 20.6 21 Cambodia . . 2,031 .. 73.5 .. 205.4 .. 0.6 70.7 .. 5.8 22 Kenya 3,383 7,381 48.1 97.7 164.1 248.2 21.0 25.7 74.1 18.6 39.5 23 Mongolia . . 512 .. 61.5 . . 100.2 .. 9.1 64.2 . . 33.2 24 Togo 1,052 1,486 95.9 121.2 181.3 464.5 9.0 5.7 61.7 11.3 48.4 25 Gambia, The 137 426 61.5 .. 206.6 235.1 6.2 14.0 50.3 29.9 76.0 26 India 20,581 93,766 11.9 28.2 136.0 201.2 9.3 27.9 80.0 29.5 32.0 27 Central African Republic 195 944 24.4 .. 94.8 403.9 4.9 6.8 53.4 27.4 67.2 28 Lao PDR 350 2,165 .. 124.9 .. 478.3 .. 5.8 34.3 5.9 28.7 29 Benin 424 1,646 30.2 81.8 133.1 285.6 6.3 8.4 55.9 24.5 52.3 30 Nicaragua 2,192 9,287 108.5 589.7 426.5 1,272.7 22.3 38.7 88.2 19.2 16.0 31 Ghana 1,398 5,874 31.6 95.1 115.2 366.5 13.1 23.1 64.3 19.9 50.8 32 Zambia 3,261 6,853 90.7 191.3 200.7 528.7 25.3 174.4 72.4 12.2 31.9 33 Angola 11,482 .. 274.9 314.3 12.5 94.6 1.7 34 Georgia . . 1,189 . . 51.6 . . . . . . .. 85.9 . . 19.7 35 Pakistan 9,930 30,152 42.4 49.5 208.7 257.9 18.3 35.3 77.6 15.4 40.5 36 Mauritania 843 2,467 125.5 243.3 306.1 458.5 17.3 21.5 68.3 14.8 36.8 37 Azerbaijan . . 321 . . 9.2 . . . . . . .. 86.3 . . 30.8 38 Zimbabwe 786 4,885 14.9 78.9 45.6 . . 3.8 25.6 82.2 0.4 33.1 39 Guinea 1,134 3,242 .. 91.2 .. 453.4 .. 25.3 64.9 11.5 45.2 40 Honduras 1,473 4,567 60.6 124.6 152.2 255.5 21.4 31.0 81.2 31.1 47.3 41 Senegal 1,473 3,845 50.5 82.3 162.7 224.3 28.7 18.7 65.3 17.8 48.4 42 China 4,504 118,090 2.2 17.2 .. 77.3 .. 9.9 91.4 0.0 13.8 43 Cameroon 2,588 9,350 37.9 124.4 140.7 338.3 15.2 20.1 80.0 16.7 17.9 44 Côted'Ivoire 7,462 18,952 77.1 251.7 205.0 418.6 38.7 23.1 87.6 7.0 20.6 45 Albania . . 709 . . 31.6 . . 93.2 . . 1.0 101.6 . . 15.6 46 Congo 1,526 6,032 99.0 365.8 148.2 481.8 10.6 14.4 88.8 7.7 11.7 47 Kyrgyz Republic .. 610 .. 20.2 .. .. 4.8 73.9 .. 29.9 48 Sri Lanka 1,841 8,230 46.1 64.4 123.4 140.3 12.0 7.3 67.7 11.7 34.7 49 Armenia .. 374 . . 17.6 .. 119.1 .. 2.9 77.9 .. 55.7 Middle-income economies 509,503 t 1,530,883 t 22.4w 39.9w 84.6w 142.6w 13.6w 17.4w 5.6w 10.4w Lower-middle-income .. .. .. .. .. .. .. 50 Lesorho 72 659 11.4 44.6 19.8 108.8 1.5 6.0 58.4 56.0 69.6 51 Egypt, Arab Rep. 19,131 34,116 89.2 73.3 207.7 208.1 13.4 14.6 75.8 13.7 12.4 52 Bolivia 2,702 5,266 93.4 90.6 258.4 410.1 35.0 28.9 74.0 16.5 48.9 53 Macedonia, FYR 1,213 65.8 79.9 11.8 86.1 24.0 54 Moldova 691 .. 17.8 . . 79.9 .. 8.0 90.0 31.3 55 Uzbekistan . . 1,630 .. 7.5 . . 35.3 6.0 91.7 . . 15.1 56 Indonesia 20,938 107,831 28.0 56.9 . . 202.9 . . 30.9 95.7 8.8 18.6 57 Philippines 17,417 39,445 53.7 51.5 212.4 121.7 26.6 16.0 95.8 7.5 21.5 58 Morocco 9,247 22,147 50.7 71.0 213.9 200.9 33.4 32.1 90.0 7.8 30.8 59 Syrian Arab Republic 3,552 21,318 27.2 134.8 106.3 336.8 11.4 4.6 87.7 8.8 4.8 60 Papua New Guinea 719 2,431 28.9 53.3 66.0 80.6 13.8 20.8 84.6 21.2 38.3 61 Bulgaria 10,887 92.3 163.0 18.8 94.6 . . 16.8 62 Kazaksran .. 3,712 .. 23.5 .. 60.8 .. 4.6 92.6 . . 10.6 63 Guaremala 1,166 3,275 14.9 22.3 63.6 101.5 7.9 10.6 85.4 30.0 28.8 64 Ecuador 5,997 13,957 53.8 84.1 201.6 263.4 33.9 26.7 90.1 5.4 21.4 65 Dominican Republic 2,002 4,259 31.2 36.5 133.8 128.5 25.3 7.8 90.5 10.2 24.0 66 Rumania 9,762 6,653 .. 19.5 80.3 73.1 12.6 10.6 94.0 8.3 25.5 67 Jamaica 1,913 4,270 78.0 134.9 129.9 113.2 19.0 17.9 91.2 14.9 28.5 68 Jordan 1,971 7,944 .. 126.2 79.0 163.8 8.4 12.6 85.9 8.0 14.9 69 Algeria 19,365 32,610 47.1 83.1 129.9 264.2 27.4 38.7 76.9 1.5 11.6 70 El Salvador 911 2,583 26.1 27.0 71.1 81.6 7.5 8.9 81.4 28.3 52.4 71 Ukraine .. 8,434 .. 10.7 .. 48.6 .. 5.3 93.4 .. 7.3 72 Paraguay 955 2,288 20.7 29.4 122.2 53.8 18.6 92.7 20.2 34.0 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. SELECTED WORLD DEVELOPMENT INDICATORS 247 External debt as percentage of Debt service Ratio of present Multilateral debt as Total external debt Exports of goods as % of exports of value to Nominal % of total external (million $1 GNP and services goods and services value of debt (%) debt 1980 1995 1980 1995 1980 1995 1980 1995 1995 1980 1995 73 Tunisia 3,527 9,938 41.6 57.3 96.0 113.2 14.8 17.0 89.8 12,3 37.2 74 Lithuania 802 ., 10.1 . 24.7 .. 1.4 87.9 .. 20.7 75 Colombia 6,941 20,760 20.9 28.2 117.1 138.7 16.0 25.2 95.8 19.5 25.6 76 Namibia . . .. .. .. 77 Belarus 1,648 7.9 33.3 . . . . 80.7 . 11.4 78 Russian Federation .. 120,461 37.6 . 126.7 .. 6.6 92.7 .. 1.7 79 Latvia . 462 . . 7.6 . . 21.5 . . 1.6 93.1 . . 30.3 80 Peru 9,386 30,831 47.6 54.1 194.2 399.5 44.5 15.3 96.5 5.5 12.1 81 Costa Rica 2,744 3,800 59.7 42.5 225.2 96.3 29.1 16.4 93.0 16.4 35.5 82 Lebanon 510 2,966 . . 25.5 . . 152.7 . . 13.1 97.2 15.2 6.7 83 Thailand 8,297 56,789 25.9 34.9 96.8 76.6 18.9 10.2 101.2 12.0 5.6 84 Panama 2,975 7,180 81.8 101.4 37.5 74.7 6.2 3.9 97.2 11.0 8.5 85 Turkey 19,131 73,592 27.4 44.1 333.1 177.8 28.0 27.7 91.3 11.2 12.2 86 Poland 42,291 36.1 127.3 12.2 84.5 4.9 87 Estonia 309 6.7 11.0 0.8 92.6 42.2 88 Slovak Republic . . 5,827 .. 33.5 . . 52.1 . . 9.7 93.2 . . 16.3 89 Borswana 147 699 16.3 16.3 19.6 24.0 2.1 3.2 80.0 57.4 68.0 90 Venezuela 29,344 35,842 42.1 49.0 132.0 160.0 27.2 21.7 95.5 0.7 9.2 Upper middle income 91 South Africa . . . . . . . . . . .. . 92 Croatia . . 3,662 . . 20.3 . . 49.7 . . 5.7 88.9 . . 14.4 93 Mexico 57,378 165,743 30.5 69.9 232.4 170.5 44.4 24.2 96.1 5.6 11.2 94 Mauritius 467 1,801 41.6 45.9 80.8 75.0 9.0 9.0 90.6 16.6 15.0 95 Gabon 1,514 4,492 39.2 121.6 62.2 160.3 17.7 15.8 79.8 2.7 14.8 96 Brazil 71,520 159,130 31.2 24.0 306.5 269.8 63.3 37.9 95.4 4.3 5.9 97 Trinidad and Tobago 829 2,556 14.0 53.6 24.6 87.9 6.8 14.8 97.4 8.6 20.7 98 Czech Republic .. 16,576 .. 37.0 .. 67.4 . . 8.7 97.2 . . 6.1 99 Malaysia 6,611 34,352 28.0 42.6 44.6 40.8 6.3 7.8 90.5 11.3 4.8 100 Hungary 9,764 31,248 44.8 72.8 . . 174.2 . . 39.1 99.4 0.0 10.5 101 Chile 12,081 25,562 45.5 43.3 192.5 127.7 43.1 25.7 95.5 2.9 11.2 102 Oman 599 3,107 11.2 29.5 15.4 48.2 6.4 7.5 94.9 5.8 5.7 103 Uruguay 1,660 5,307 17.0 32.4 104.1 144.3 18.8 23.5 96.4 11.0 23.7 104 Saudi Arabia . . . . . . . . . . . . . . .. . . . 105 Argentina 27,157 89,747 35.6 33.1 242.4 320.2 37.3 34.7 92.5 4.0 10.5 106 Slovenia . . 3,489 18.7 33.3 6.7 96.5 15.4 107 Greece . . . . . Low- and middle-income 615,711 t 2,065,676 t 21.0w 39.6w 86.5w 151.4w 13.0w 17.0w 7.6 w 14.3 w Sub-Saharan Africa 84,119 t 226,483 t 30.6w 81.3w 91.7w 241.7w 9.8w 14.5w 9.0w 24.3w East Asia and Pacific 64,600 t 404,458 t 17.3w 32.9w 81.8w 98.3w 11.5w 12.8w 8.4w 13.3w South Asia 38,014 t 156,778 t 17.4w 30.5w 160.5w 218.7w 11.7w 24.6w 24.6w 36.4w Europe and Central Asia 87,919 425,319 t 9.9w 39.9w 47.1 w 130.7w 7.4w 13.8w 5.4w 7.9w Middle East and N. Africa 83,793 t 216,046 t 18.3w 37.3w 41.1w 133.4w 5.7w 14.9w 6.7 w 10.6w Latin America and Caribbean 257,266 t 636,594 t 36.Ow 41.0w 201.8w 212.0w 36.3 w 26.1 w 5.5w 11.4w High-income economies 108 Korea, Rep. 109 Portugal 110 Spain 111 New Zealand 112 Ireland 113 tlsrael 114 I Kuwait 115 5 United Arab Emirates 116 United Kingdom 117 Australia 118 Italy 119 Canada 120 Finland 121 tHongKong 122 Sweden 123 Netherlands 124 Belgium 125 France 126 5 Singapore 127 Austria 128 United States 129 Germany 130 Denmark 131 Norway 132 Japan 133 Switzerland World a. Includes Eritrea. 248 WORLD DEVELOPMENT REPORT 1997 Table la. Basic indicators for other economies PPP estimates of GNP Poverty GNP per capitaa per capita" % of people Life Population Surface area Avg. ann. living on less than expectancy at Adult Dollars US=100 Ithousands) (thousands growth 1%) Current int'l $ $1 a day IPPP) birth (years) illiteracy (%) mid-1995 of sq. km) 1995 1985-95 1987 1995 1995 1981-95 1995 1995 I Afghanistan 23,481 652.09 c ' . . . 44 69 2 American Samoa 57 0.20 d 3 Andorra . . 0.45 e 4 Antigua and Barbuda 65 0.44 d .. .. . . 75 5 Aruba . 0.19 C . . . . . . 6 Bahamas, The 276 13.88 11,940 -0.8 68.8 54.5 14,710 . . 73 2 7 Bahrain 577 0.68 7,840 0.2 54.2 49.7 13,400 . . 72 15 8 Barbados 266 0.43 6,560 0.8 45.1 39.4 10,620 .. 3 9 Belize 216 22.96 2,630 3.9 17.0 20.0 5,4OO 70 10 Bermuda 63 0.50 e .. .. .. II Bhutan 695 47.00 420 4.9 4.5 4.7 1,2601 58 12 Bosnia and Herzegovina 4,383 51.13 c . 13 Brunei 285 5.77 e . . . . . . . . 75 12 14 Cape Verde 380 4.03 960 6.6 6.9 1,870 65 28 15 Cayman Islands 0.26 e 16 Channel Islands 142 0.20 e .. .. .. .. 78 17 Comoros 499 2.23 470 -1.4 6.5 4.9 1,320 . . 56 43 18 Cuba 11,011 110.86 g . . 76 4 19 Cyprus 734 9.25 e 44.5 .. 78 20 Djibouti 634 23.20 g . . 50 54 21 Dominica 73 0.75 2,990 4.1 .. . . 73 22 Equatorial Guinea 400 28.05 380 . . .. . . 49 23 Eritrea 3,574 124.80 c . . . . .. . . . . 48 24 Facroc Islands . . 1.40 e . . . . .. . . . . . 25 Fiji 775 18.27 2,440 2.0 19.1 21.4 5,78O . . 72 8 26 French Guiana .. 90.00 e . . 27 French Polynesia 225 4.00 e . . . . 70 28 Greenland .. 341.70 e .. 68 29 Grenada 91 0.34 2,980 .. . 30 Guadeloupe 424 1.71 d . . .. . . 75 31 Guam 149 0.55 c . . . . .. . . 73 32 Guyana 835 214.97 590 0.6 8.2 9.0 2,420 66 2 33 Iceland 268 103.00 24,950 1.0 88.1 75.8 20,460 79 34 Iran, Islamic Rep. 64,120 1,648.00 g -1.5 21.6 20.3 . . 68 28 35 Iraq 20,097 438.32 g . . . . 66 42 36 Isle of Man .. 0.59 d . . . . 37 Kiribati 79 0.73 920 -0.6 .. .. .. 38 Korea, Dem. Rep. 23,867 120.54 g .. .. 70 39 Liberia 2,733 97.75 c 7.0 . . . . 54 40 Libya 5,407 1,759.54 d 43.9 .. .. 65 41 Liechtenstein . . 0.16 e . . . . . . . . . . . 42 Luxembourg 410 2.59 41,210 0.9 154.1 140.6 37,930 .. 76 h 43 Macao 450 0.02 e . . . . . . . . . 44 Maldives 253 0.30 990 5.9 7.5 11.4 3,080 63 7 45 Malta 372 0.32 d 38.2 77 46 Marshall Islands . . 0.18 g . . 47 Martinique 380 1.10 e .. 77 48 Mayotte . . 0.37 d 49 Micronesia, Fed. Sts. 107 0.70 g 50 Monaco 0.20 e .. 51 Myanmar 45,106 676.58 . . . . . . 59 17 52 Netherlands Antilles 200 0.80 . . . . 77 53 New Caledonia 185 18.58 r . . . . . . . 54 Northern Mariana Islands .. 0.48 e .. .. .. 55 Puerto Rico 3,717 8.90 d .. 41.8 .. .. .. 75 56 Qatar 642 11.00 11,600 -4.2 85.4 65.6 l7,690 . . 72 21 57 Reunion 653 2.51 e . . . . .. 74 58 SaoToméandPrincipe 129 0.96 350 -2.1 . . .. . . 69 59 Seychelles 74 0.45 6,620 . . . . .. .. . . 72 21 60 Solomon Islands 375 28.90 910 3.2 7.9 8.1 2,l90 . . 63 61 Somalia 9,491 637.66 c .. 2.3 .. .. 49 62 St. Kitts and Nevis 41 0.36 5,170 4.8 28.2 34.9 9,410 69 63 St. Lucia 158 0.62 3,370 3.9 71 64 St. Vincent and the Grenadines 111 0.39 2,280 3.8 . . . . 72 65 Sudan 26,707 2,505.81 c . . 8.1 . . . . 54 54 66 Suriname 410 163.27 880 3.5 11.9 8.3 2,250 70 7 67 Swaziland 900 17.36 1,170 -1.4 12.5 10.7 2,880 . . 58 23 68 Tajikistan' 5,836 143.10 340 .. 12.1 3.4 920 67 69 Tonga 104 0.75 1,630 0.5 .. .. 69 70 Turkmenistan' 4,508 488.10 920 . . .. 4.9 71 Vanuatu 169 12.19 1,200 -1.1 9.4 8.5 2,290 .. 72 Virgin Islands (U.S.) 99 0.34 e 0.0 . . . . 76 73 WestBankandGaza .. 6.10 g .. .. .. 74 Western Samoa 165 2.84 1,120 0.2 8.9 7.5 2,O30 . . 68 75 Yugoslavia, Fed. Rep. (Serb/Mont.) 10,518 102.17 g . . . . . . . . 76 Zaire . . . . 120 . 490 . a. Atlas method; see the technical notes. b. Purchasing power parity; see the technical notes. c. Estimated to be low income ($765 or less). d. Upper middle income ($3,036 to $9,385). e. Estimated to be high income ($9,386 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 ($766 to $3,035). h. According to UNESCO, illiteracy is less than 5 percent. i. Estimates for economies of the former Soviet Union are preliminary; their classification will be kept under review. TECHNICAL NOTES TECHNICAL NOTES DISCUSS THE SOURCES AND in the Key and Primary Data Documentation table and in THESE methods used to compile the 124 indicators the notes below. included in the 1997 Selected World Development Indi- cators. The notes are organized by table and, within each Data consistency and reliability table, by indicator in order of appearance in the table. Considerable effort has been made to standardize the data, The 133 economies covered in the main tables are but full comparability cannot be assured, and care must be listed in ascending order of gross national product (GNP) taken in interpreting the indicators. Many factors affect per capita. A separate table (Table la) shows basic indica- availability, comparability, and reliability: statistical sys- tors for an additional 76 economies that have sparse data tems in many developing economies are still weak; statis- or populations of fewer than 1 million. tical methods, coverage, practices, and definitions differ widely; and cross-country and intertemporal comparisons Sources involve complex technical and conceptual problems that The data in the Selected World Development Indicators cannot be unequivocally resolved. For these reasons, al- are taken from World Development Indicators 1997. Ex- though the data are drawn from the sources thought to be cept for a few corrections made to the data base after that most authoritative, they should be construed only as indi- volume went to press, they are identical in source and vin- cating trends and characterizing major differences among tage to the data published there. Although some countries economies rather than offering precise quantitative mea- have produced revised statistical series since the publica- sures of those differences. Also, national statistical agencies tion of World Development Indicators 1997, those revi- tend to revise their historical data, particularly for recent sions are not included here. They will appear in the next years. Thus, data of different vintages may be published in edition of the World Development Indicators. different editions of World Bank publications. Readers are The World Bank draws on a variety of sources for the advised not to compare such data from different editions. indicators published in the World Development Indicators. Consistent time series are available on the World Develop- Data on external debt are reported directly to the World ment Indicators 1997 CD-ROM In addition, data issues Bank by developing member countries, through the have yet to be resolved for the Fifteen economies of the for- Debtor Reporting System. Other data are drawn mainly mer Soviet Union: coverage is sparse, and the data are sub- from the United Nations, its specialized agencies, the ject to more than the normal range of uncertainty. International Monetary Fund (IMF), and country reports Data in italics are for years or periods other than those to the World Bank. Bank staff estimates are also used to specified: up to two years before or after the date shown improve currentness or consistency. For most countries, for economic indicators, and up to three years for social national accounts estimates are obtained from member indicators, because the latter tend to be collected less reg- governments through World Bank economic missions. In ularly and change less dramatically over short periods. some instances these are adjusted by Bank staff to ensure Ratios and growth rates conformity with international definitions and concepts. Most social data from national sources are drawn from For ease of reference, data are often presented as ratios or regular administrative files, special surveys, or periodic rates of growth. The underlying absolute data are available census inquiries. Citations of specific sources are included on the World Development Indicators 1997 CD-ROM 250 WORLD DEVELOPMENT REPORT 1997 Unless otherwise noted, period-average growth rates are weighting process may result in discrepancies between computed using the least-squares regression method (see subgroup and overall totals. Statistical methods below). Because this method takes Table 1. Basic indicators into account all available observations in a period, the resulting growth rates reflect general trends and are not Basic indicators for economies with sparse data or with unduly influenced by exceptional values. To exclude the populations of fewer than 1 million are shown in Table la. effects of inflation, constant-price economic indicators are Population estimates for mid-1995 are based on the de used in calculating growth rates. facto definition of population, which counts all residents regardless of legal status or citizenship. Refugees not per- Constant-price series manently settled in the country of asylum are generally To facilitate international comparisons and capture the considered to be part of the population of their country of effects of changes in intersectoral relative prices for the origin. national accounts aggregates, constant-price data for most Population estimates are derived from national cen- economies are first partially rebased to three sequential suses. Precensus and postcensus estimates are often based base years and then "chain-linked" together and expressed on interpolations or projections. The international com- in prices of a common base year, 1987. The year 1970 is parability of population indicators is limited by differ- the base year for the period from 1960 to 1975, 1980 for ences in the concepts, definitions, data collection proce- 1976 to 1982, and 1987 for 1983 and beyond. dures, and estimation methods used by national statistical During the chain-linking procedure, components of agencies and other organizations that collect the data, In GDP by industrial origin are individually rescaled and addition, the frequency and quality of coverage of popu- summed to provide the rescaled GDP. In this process a lation censuses vary by country and region. For more rescaling deviation may occur between constant-price information on the compilation of population data, see GDP as measured by industrial origin and constant-price the notes to Table 4. GDP as measured by expenditure. Such rescaling devia- Surface area is measured in square kilometers and com- tions are absorbed in private consumption expenditures, prises land area and inland waters. Data on surface area on the assumption that GDP by industrial origin is the come from the Food and Agriculture Organization (FAQ) more reliable estimate. Independent of the rescaling, value and are published in the FAO Production Yearbook. added in the services sector also includes a statistical dis- Gross national product (GNP) per capita is the sum of crepancy as reported by the original source. gross value added by all resident producers, plus any taxes (less subsidies) that are not included in the valuation of Summary measures output, plus net receipts of primary income (employee The summary measures across countries for regions and compensation and property income) from nonresident income groups, presented in the blue bands in the tables, sources, divided by the midyear population and con- are calculated by simple addition when they are expressed verted to U.S. dollars using the World Bank's Atlas as levels. Growth rates and ratios are usually combined by method. This involves using a three-year average of ex- a base-year value-weighting scheme. The summary mea- change rates to smooth the effects of transitory exchange sures for social indicators are weighted by population or rate fluctuations. For further discussion of the Atlas subgroups of population, except for infant mortality, method see Statistical methods below. The growth rate of which is weighted by the number of births. See the notes GNP per capita is computed from GNP measured in on specific indicators for more information. constant 1987 prices using the least-squares growth rate For summary measures that cover many years, calcula- method. tions are based on a uniform group of economies so that GNP per capita is estimated by World Bank staff based changes in the composition of the aggregate do not pro- on national accounts data collected by World Bank staff duce spurious changes in the indicator. Group measures during economic missions or reported by national statisti- are compiled only if data are available for a given year cal offices to other international organizations such as the for at least two-thirds of the full group, as defined by the Organization for Economic Cooperation and Develop- 1987 benchmarks. As long as that criterion is met, coun- ment (OECD). For high-income OECD economies the tries with missing data are assumed to behave like those data come from the OECD. GNP per capita in U.S. dol- that provided estimates. Readers should keep in mind that lars is used by the World Bank to classify countries for the summary measures are estimates of representative analytical purposes and to determine eligibility for bor- aggregates for each topic, and that nothing meaningful rowing. For definitions of the income groups used in this can be deduced about behavior at the country level by book, see the table on Classification of Economies by working back from group indicators. In addition, the Income and Region. SELECTED WORLD DEVELOPMENT INDICATORS 251 PPP estimates of GNPper capita are calculated by con- are available, the average level of income has been adjusted verting GNP to U.s. dollars using purchasing power par- to accord with either a survey-based estimate of mean con- ities (PPP) instead of exchange rates as conversion factors. sumption (when available) or an estimate based on con- The resulting estimates are expressed in international dol- sumption data from national accounts. lars, a unit of account that has the same purchasing power Poverty measures are prepared by the Poverty and over total GNP as the U.S. dollar in a given year. The Human Resources Division of the World Bank's Policy denominator is the midyear population estimate for the Research Department. International poverty lines are year shown. based on primary household survey data obtained from Relative prices of goods and services not traded on government statistical agencies and World Bank country international markets tend to vary substantially from one departments. The poverty measures are based on the most country to another, leading to large differences in the rel- recent PPP estimates, from the latest version of the Penn ative purchasing power of currencies and thus in welfare World Tables (Mark 5.6a). as measured by GNP per capita. The use of PPP conver- Lift expectancy at birth is the number of years a new- sion factors corrects for these differences and may there- born infant would live if prevailing patterns of mortality fore provide a better comparison of average income or at the time of its birth were to stay the same throughout consumption between economies. However, caution its life. Estimates of life expectancy are derived from vital should be used in interpreting PPP-based indicators. PPP registration systems or, in their absence, from demo- estimates employ price comparisons of comparable items, graphic and household surveys using models to obtain but not all items can be matched perfectly in quality age-specific mortality rates. across countries and over time. Services are particularly Adult illiteracy is the proportion of adults age 15 and difficult to compare, in part because of differences in pro- above who cannot, with understanding, read and write a ductivity. Many servicesfor example, government ser- short, simple statement on their everyday life. Literacy vicesare not sold in markets in all countries, so they are and illiteracy are difficult both to define and to measure. compared using input prices (mostly wages). Because this The definition here is based on the concept of "func- approach ignores productivity differences, it may inflate tional" literacy. To measure literacy using such a defini- estimates of real quantities in lower-income countries. tion requires census or sample survey measurements The source of PPP data is the International Compari- under controlled conditions. In practice, many countries son Programme (ICP), coordinated by the U.N. Statisti- estimate the number of illiterate adults from self-reported cal Division. The World Bank collects detailed ICP data or from estimates of school completion. Because of benchmark data from regional sources, establishes global these problems, comparisons across countriesand even consistency across the regional data sets, and computes over time for one countryshould be made with caution. regression-based estimates for nonbenchmark countries. Data on illiteracy rates are supplied by UNESCO (United For detailed information on the regional sources and com- Nations Educational, Scientific, and Cultural Organiza- pilation of benchmark data see World Bank 1993. For in- tion) and published in its Statistical Yearbook. formation on how regression-based PPP estimates are derived see Ahmad 1992. Table 2. Macroeconomic indicators The percentage of people living on less than $1 a day Central go vernment current deficit/surplus is defined as cur- (PPP) at 1985 international (purchasing power parity) rent revenue of the central government less current expen- prices is a widely used measure of poverty. A person is said diture. Grants are not included in revenue. This is a use- to be poor if he or she lives in a household whose total ful measure of the government's own fiscal capacity. The income or consumption per person is less than the poverty overall deficit or surplus, including grants and the capital line. Although it is impossible to create an indicator of account, is shown in Table 14. The data come from the poverty that is strictly comparable across countries, the use IMF's Government Finance Statistics Yearbook. of a standard, international poverty line helps to reduce Money and quasi money comprise most liabilities of a comparability problems in several ways. In estimating liv- country's monetary institutions to residents other than ing standards, nationally representative surveys have been the central government. This definition of the money used, conducted either by national statistical offices or by supply is sometimes referred to as M2. Money comprises private agencies under government or international agency currency held outside banks and demand deposits other supervision. Whenever possible, consumption has been than those of the central government. Quasi money com- used as the welfare indicator for deciding who is poor. The prises time and savings deposits and similar bank accounts measure of consumption is generally comprehensive, that the issuer can exchange for money with little, if any, including that from own production as well as all food and delay or penalty, and foreign currency deposits of resi- nonfood goods purchased. When only household incomes dents other than those of the central government. Where 252 WORLD DEVELOPMENT REPORT 1997 nonmonetary financial institutions are important issuers Net present value of external debt is the value of short- of quasi-monetary liabilities, their liabilities may be term debt plus the discounted sum of all debt service pay- included in quasi money. ments due over the life of existing loans, at current prices. The source of data on the money supply is the IMF's The debt figures are converted into U.S. dollars from cur- International Financial Statistics. Money and quasi money rencies of repayment at end-of-year official exchange rates. are the sum of IFS lines 34 and 35. To calculate the ratio of debt to GNP, GNP is converted The average annual nominal growth rate of the money at official exchange rates or, in exceptional cases, by an supply is calculated from year-end figures using the least- alternative, single-year conversion factor determined by squares method. The average of the year-end figures for World Bank staff. (See also the notes to Tables 12 and 17.) the specified year and the previous year is used to calculate the average outstanding as a percentage of GDP. Table 3. External economic indicators Nominal interest rates of banks show the deposit rate Net barter terms of trade measures the relative movement paid by commercial or similar banks for demand, time, or of export prices against that of import prices. Calculated savings deposits and the lending rate charged by these as the ratio of a country's average export price index to its banks on loans to prime customers. The data are of lim- average import price index, this indicator shows changes ited international comparability, partly because coverage relative to a base year (1987). The data come from the and definitions vary. Interest rates are expressed in nomi- U.N. Conference on Trade and Development (UNC- nal terms; therefore much of the variation among coun- TAD) data base, the IMF's International Financial Statis- tries stems from differences in inflation. The data come tics, and World Bank staff estimates. (See also Table 15.) from International Financial Statistics, lines 601 and 60p. Trade is measured as the ratio of the sum of exports Average annual inflation is measured by the rate of and imports of goods and services to the current value of change in the GDP implicit deflator. The implicit defla- GDP. The trade-GDP ratio is a commonly used measure tor is calculated by dividing annual GDP at current prices of the openness of an economy or its integration with the by the corresponding value of GDP at constant prices, global economy. The data come from the World Bank's both in national currency. The least-squares method is national accounts data files. used to calculate the growth rate of the GDP deflator for Aggregate net resource flows is the sum of net flows the period. of long-term debt (excluding use of IMF credits), official The GDP implicit deflator is the broadest-based mea- grants (excluding technical assistance), net foreign direct sure of inflation, showing price movements for all goods investment, and net portfolio equity flows. Total net flows and services produced in the economy, but like all price of long-term debt are disbursements less repayments of indexes it is subject to conceptual and practical limita- principal on public, publicly guaranteed, and private non- tions. Deflators for developing economies are estimated guaranteed long-term debt. Official grants are transfers from national accounts data collected by the World Bank. made by an official agency in cash or in kind, in respect of Data for high-income economies are derived from data which no legal debt is incurred by the recipient. The data provided by the OECD. are taken from the World Bank's Debtor Reporting Sys- Current account balance is the sum of net exports of tem and from the IMF's International Financial Statistics. goods, services, and income and net current transfers. Cap- Net private capital flows consists of private debt and ital transfers are excluded. (See also Table 16.) The data nondebt flows and bank and trade-related lending. Private come from the IMF's International Financial Statistics and debt flows include commercial bank lending, bonds, and from estimates provided by World Bank country teams. other private credits. Nondebt private flows are made up Gross international reserves comprises holdings of mon- of net foreign direct investment and portfolio investment. etary gold, special drawing rights (SDRs), the reserve posi- Foreign direct investment is investment made to acquire a tion of members with the IMF, and holdings of foreign lasting management interest in an enterprise operating in exchange under the control of monetary authorities. Gross an economy other than that of the investor. It is the sum international reserves in U.S. dollars are shown in Table of net flows of equity capital, reinvested earnings, other 16. Reserve holdings as months of import coverage are long-term capital, and short-term capital as shown in the calculated as the ratio of gross international reserves to the balance of payments. Portfolio investment flows include current U.S. dollar value of imports of goods and services, net non-debt-creating portfolio equity flows (the sum of multiplied by 12, country funds, depository receipts, and direct purchases of The summary measures in this table are computed as shares by foreign investors) and net portfolio debt flows the ratio of group aggregates for gross international re- (bond issues purchased by foreign investors). serves and total imports of goods and services in current The principal source of data on private capital flows is dollars. the World Bank's Debtor Reporting System. Additional SELECTED WORLD DEVELOPMENT INDICATORS 253 data come from International Financial Statistics and time workers, in general the labor force includes the World Bank data files. armed forces, the unemployed, and first-time job-seekers, Aid comprises financial assistance classified as official but excludes homemakers and other unpaid caregivers development assistance (ODA) or official aid (OA) by the and workers in the informal sector. Development Assistance Committee (DAC) of the Average annual growth rate of the labor force is com- OECD. ODA comprises loans and grants made on con- puted using the exponential end-point method. See the cessional financial terms by all bilateral official agencies section on Statistical methods for more information. and multilateral sources to promote economic develop- Females as a percentage of the labor force shows the ment and welfare. Net disbursements equal gross dis- extent to which women are active in the labor force. Labor bursements less payments to the originators of aid for force estimates are derived by applying participation rates amortization of past aid receipts. To qualify as ODA, a from the International Labour Organization to World transaction must meet the following tests: it is adminis- Bank population estimates. tered with the promotion of the economic development Agricultural labor force includes people engaged in and welfare of developing countries as its main objective, farming, forestry, hunting, and fishing. and it is concessional in character and conveys a grant ele- Industrial labor force includes people working in the ment of at least 25 percent. OA comprises assistance pro- mining, manufacturing, construction, and electricity, vided on ODA-like terms to the countries of Eastern water, and gas industries. Europe, the former Soviet Union, and other economies on Activity rates or labor force participation rates of the the DAC's "part II" list. The data on aid are provided by economically active population are compiled by the Inter- the DAC and published in its annual report, Development national Labour Organization from the latest national Co-operation. Data for GNP are World Bank estimates. censuses or surveys and are published in its Yearbook of Summary measures for aid as a percentage of GNP are Labour Statistics. Labor force numbers in some develop- computed from the ratio of group totals for aid and for ing countries reflect a significant underestimation of GNP in current U.S. dollars. female participation rates. Estimates of the rural labor force may also fail to capture the extent of family and sea- Table 4. Population and labor force sonal labor. Population estimates for mid-1995 come from a variety of All summary measures are country data weighted by sources, including the U.N. Population Division, national population or population subgroup. statistical offices, and World Bank country departments. The World Bank uses the de facto definition of a country's Table 5. Distribution of income or consumption population, which counts all residents regardless of legal Survey year is the year in which the underlying data were status or citizenship. However, refugees not permanently collected. settled in the country of asylum are generally considered to The Gini index measures the extent to which the dis- be part of the population of their country of origin. tribution of income (or, in some cases, consumption The notes to Table 1 provide additional information expenditures) among individuals or households within an about population estimates. The Key and Primary Data economy deviates from a perfectly equal distribution. A Documentation table lists the date of the most recent cen- Lorenz curve plots the cumulative percentages of total sus or demographic survey. income received against the cumulative number of recipi- Average annual growth rate ofpopulation is computed ents, starting with the poorest individual or household. using the exponential end-point method. See the section The Gini index measures the area between the Lorenz on Statistical methods for more information. curve and a hypothetical line of absolute equality, and is Population aged 15-64 is the age group generally con- expressed as a percentage of the maximum area under the sidered to be the most economically active. In many line. Thus a Gini index of zero represents perfect equality, developing economies, however, many children under 15 and an index of 100 percent perfect inequality. work full or part time. And in some high-income eco- Percentage share of income or consumption is the share nomies many workers postpone retirement past age 65. that accrues to subgroups of population indicated by Total labor force comprises those people who meet the deciles or quintiles. Percentage shares by quintiles may International Labour Organization's definition of the eco- not sum to 100 because of rounding. nomically active population: all people who supply labor Inequality in the distribution of income is reflected in for the production of goods and services during a specified the percentage share of income or consumption accruing period. It includes both the employed and the unem- to segments of the population ranked by income or con- ployed. Although national practices vary in the treatment sumption levels. The segments ranked lowest by personal of such groups as the armed forces and seasonal or part- or family income typically receive the smallest share of 254 WORLD DEVELOPMENT REPORT 1997 total income. The Gini index provides a convenient sum- has been used rather than income. Households have been mary measure of the degree of inequality. ranked by consumption or income per capita in forming Data on personal or household income or consump- the percentiles, and the percentiles are based on popula- tion come from nationally representative household sur- tion, not households. The comparability of the data for veys. The data sets refer to different years between 1985 high-income economies is more limited, because the unit and 1994. Footnotes to the survey year indicate whether of observation is usually a household unadjusted for size, the rankings are based on income or consumption per and households are ranked according to total household capita or, in the case of high-income economies, house- income rather than income per household member. These hold income. Where the original data from the household data are presented pending the publication of improved survey were available, they have been used to directly cal- data from the Luxembourg Income Study, which ranks culate the income (or consumption) shares by quintile. households by the average disposable income per adult Otherwise, shares have been estimated from the best avail- equivalent. The estimates in the table should therefore be able grouped data. treated with considerable caution. The distribution indicators for low- and middle- Data on distribution for low- and middle-income eco- income economies have been adjusted for household size, nomies are compiled by the Poverty and Human providing a more consistent measure of income or con- Resources Division of the World Bank's Policy Research sumption per capita. No adjustment has been made for Department, using primary household survey data geographic differences in the cost of living within coun- obtained from government statistical agencies and World tries, because the data needed for such calculations are Bank country departments. Data for high-income eco- generally unavailable. For further details on the estimation nomies are from national sources, supplemented by the method for low- and middle-income economies, see Luxembourg Income Study 1990 data base, the Eurostat Ravallion and Chen 1996. Statistical Yearbook, and the United Nations' National Because the underlying household surveys differ in Accounts Statistics: Compendium of Income Distribution method and in the type of data collected, the distribution Statistics (1985). indicators are not strictly comparable across countries. These problems are diminishing as survey methods im- Table 6. Health prove and become more standardized, but strict compara- Access to health care is measured by the share of the popu- bility is still impossible. lation for whom treatment of common diseases and in- The following sources of noncomparability should be juries, including essential drugs on the national list, is noted. First, the surveys differ as to whether they use available within one hour's walk or travel. Facilities tend income or consumption expenditure as the living stan- to be concentrated in urban areas. dard indicator. For thirty-seven of the sixty-six low- and Access to safe water shows the percentage of the popula- middle-income economies for which data are available, tion with reasonable access to adequate amounts of safe the data refer to consumption expenditure. Income is water (including treated surface waters or untreated but typically more unequally distributed than consumption. uncontaminated water from sources such as springs, sani- In addition, the definitions of income used in the surveys tary wells, and protected boreholes). In an urban area such are usually very different from the economic definition of a source may be a public fountain or standpost located not income (the maximum level of consumption consistent more than 200 meters away. In rural areas access implies with keeping productive capacity unchanged). For these that members of the household do not have to spend a reasons, consumption is usually a much better measure. disproportionate part of the day fetching water. The defi- Second, the surveys differ as to whether they use the nition of safe water has changed over time. household or the individual as their unit of observation. Access to sanitation refers to the percentage of the pop- Furthermore, household units differ in size and in the ulation with at least adequate excreta disposal facilities extent to which income is shared among members. Indi- that can effectively prevent human, animal, and insect viduals differ in age and consumption needs. Where contact with excreta. households are used as the observation unit, the deciles or The infant mortality rate is the number of deaths of quintiles refer to the percentage of households rather than infants under 1 year of age per thousand live births in a of population. Third, the surveys differ according to given year. The data are a combination of observed values whether they rank the units of observation by household and interpolated and projected estimates. or per capita income (or consumption). Prevalence of malnutrition is the percentage of children World Bank staff have made an effort to ensure that under age 5 whose weight for age is more than 2 standard the data for low- and middle-income economies are as deviations below the mean of the reference population. comparable as possible. Whenever possible, consumption Weight for age is a composite indicator of weight for SELECTED WORLD DEVELOPMENT INDICATORS 255 height (wasting) and height for age (stunting). Although many countries consider primary school age to be 6 to 11 this indicator does not distinguish wasting from stunting, years, others use different age groups. Gross enrollment it is useful for comparisons with earlier surveys, as weight ratios may exceed 100 percent because some pupils are for age was the First anthropomerric measure in general younger or older than the country's standard primary use. The reference population, adopted by the World school age. Health Otganization (WHO) in 1983, consists of chil- Secondary school enrollment data are calculated in the dren in the United States who are assumed to be well same manner, and again the definition of secondary nourished. For some countries, if weight for age cannot be school age differs among countries. It is most commonly estimated, the prevalence assessment for the country was considered to be 12 to 17 years. Late entry of students as estimated from survey data by the WHO. This approach well as repetition and the phenomenon of "bunching" in has minor effects on the estimated rates, which the WHO final grades can influence these ratios. considers generally comparable across countries. Tertiary enrollment data are calculated by dividing the Contraceptive prevalence rate is the proportion of number of pupils enrolled in all postsecondary schools women who are practicing, or whose husbands are prac- and universities by the population age 20 to 24, although ticing, any form of contraception. Contraceptive usage is people above and below this age group may be registered generally measured for married women age 15 to 49. A in tertiary institutions. few countries use measures relating to other age groups, Percentage of cohort reaching grade 4 is the proportion especially 15 to 44. Data are mainly derived from demo- of children starting primary school in 1980 and 1988 who graphic, health, and contraceptive prevalence surveys. continued to the fourth grade by 1983 and 1991, respec- The total fertility rate represents the number of chil- tively. Figures in italics represent earlier or later cohorts. dren that would be born to a woman were she to live to Data on enrollment flows are compiled by UNESCO the end of her childbearing years and bear children at each from reports by national authorities. age in accordance with prevailing age-specific fertility Adult illiteracy is defined as the proportion of the pop- rates. The data are a combination of observed, interpo- ulation age 15 years and older who cannot, with under- lated, and projected estimates. standing, read and write a short, simple statement on their The maternal mortality ratio is the number of female everyday life. This is only one of three widely accepted deaths that occur during pregnancy and childbirth per definitions, and its application is subject to qualifiers in a 100,000 live births. Because deaths during childbirth are number of countries. The data are from the illiteracy esti- defined more widely in some countries, to include com- mates and projections prepared by UNESCO. plications of pregnancy or the period after childbirth or of The summary enrollment measures in this table are abortion, and because many pregnant women die from weighted by population. lack of suitable health care, maternal mortality is difficult Table 8. Commercial energy use to measure consistently and reliably across countries. Clearly, many maternal deaths go unrecorded, particularly Total energy use refers to domestic primary energy use in countries with remote rural populations. This may before transformation to other end-use fuels (such as elec- account for some of the low estimates shown in the table, tricity and refined petroleum products) and is calculated especially for several African countries. The data are drawn as indigenous production plus imports and stock changes, from diverse national sources. Where national administra- minus exports and international marine bunkers. Energy tive systems are weak, estimates are derived from demo- consumption also includes products for nonenergy uses, graphic and health surveys using indirect estimation tech- mainly derived from petroleum. The use of firewood, niques or from other national sample surveys. For a dried animal excrement, and other traditional fuels, number of developing countries, maternal mortality esti- although substantial in some developing countries, is not mates are derived by the WHO and the United Nations taken into account, because reliable and comprehensive Children's Fund (UNICEF) using statistical modeling data are not available. techniques. Energy use per capita is based on total population esti- All summary measures, except for infant mortality, are mates in the years shown. weighted by population or by subgroups of the popula- GDPper kilogram of commercial energy use is the U.S. tion. Infant mortality is weighted by the number of births. dollar estimate of GDP produced (at constant 1987 prices) per kilogram of oil equivalent. Table 7. Education To calculate net energy imports as a percentage of energy Primary school enrollment data are estimates of the ratio of consumption, both imports and consumption are mea- children of all ages enrolled in primary school to the coun- sured in oil equivalents. A negative sign indicates that the try's population of primary school-age children. Although country is a net exporter. 256 WORLD DEVELOPMENT REPORT 1997 Data on commercial energy use come primarily from Other land includes forest and woodland and the land the International Energy Agency and the U.N. Energy under natural or planted stands of trees, as well as logged- Statistics Yearbook. They refer to commercial forms of over areas that will be forested in the near future. It also primary energypetroleum (crude oil, natural gas liq- includes uncultivated land, grassland not used for pasture, uids, and oil from unconventional sources), natural gas, wetlands, wastelands, and built-up areas. Built-up areas solid fuels (coal, lignite, and other derived fuels), and pri- are residential, recreational, and industrial lands and areas mary electricity (nuclear, hydroelectric, geothermal, and covered by roads and other fabricated infrastructure. other)all converted into oil equivalents. For converting Data on land use are from the Food and Agriculture nuclear electricity into oil equivalents, a notional thermal Organization (FAO), which gathers these data from efficiency of 33 percent is assumed; hydroelectric power is national agencies through annual questionnaires and represented at 100 percent efficiency. national agricultural censuses. However, countries some- Carbon dioxide emissions measures industrial contribu- times use different definitions of land use. The FAO often tions to the carbon dioxide flux from solid fuels, liquid adjusts the definitions of land use categories and some- fuels, gas fuels, gas flaring, and cement manufacture. The times substantially revises earlier data. Because data on data are based on several sources as reported by the World land use reflect changes in data reporting procedures as Resources Institute. The main source is the Carbon Diox- well as actual land use changes, apparent trends should be ide Information Analysis Center (CDIAC), Environmen- interpreted with caution. Most land use data are from tal Science Division, Oak Ridge National Laboratory. 1994. CDIAC annually calculates emissions of carbon diox- Urban population is the midyear population of areas ide from the burning of fossil fuels and the manufacture defined as urban in each country. The definition varies of cement for most countries of the world. These calcula- slightly from country to country. tions are based on data on the net apparent consumption Population in urban agglomerations of] million or more of fossil fuels from the World Energy Data Set maintained is expressed as the percentage of a country's population by the U.N. Statistical Division, and from data on world living in metropolitan areas that in 1990 had a population cement manufacture based on the Cement Manufacturing of 1 million or more people. Data Set maintained by the U.S. Bureau of Mines. Emis- Estimates of the urban population come from the sions are calculated using global average fuel chemistry United Nations' World Urbanization Prospects: The ]994 and usage. Estimates do not include bunker fuels used in Revision. To compute the growth rate of the urban popu- international transport because of the difficulty of appor- lation, the United Nations' ratio of urban to total popu- tioning these fuels among the countries benefiting from lation is first applied to the World Bank's estimates of that transport. Although the estimates of world emissions total population (Table 4). The resulting series of urban are probably within 10 percent of actual emissions, esti- population estimates are also used to compute the popu- mates for individual countries may have larger error lation in urban agglomerations as a percentage of the bounds. urban population. Because the estimates in this table are Summary measures for total energy use and carbon based on different national definitions of what is urban, dioxide emissions are simple totals. The summary growth cross-country comparisons should be made with caution. rates are computed from the group totals using the least- The summary measures for urban population as a per- squares method. For energy consumption per capita and centage of total population are calculated from country carbon dioxide emissions per capita, population weights percentages weighted by each country's share in the aggre- are used to compute group averages. gate population. The other summary measures are weighted in the same fashion, using the urban population. Table 9. Land use and urbanization Cropland includes land used to cultivate temporary and Table 10. Forest and water resources permanent crops, temporary meadows, market and Forest areas refers to natural stands of woody vegetation in kitchen gardens, and land that is temporarily fallow. Per- which trees predominate. manent crops are those that do not need to be replanted Annual deforestation refers to the permanent conver- after each harvest, excluding land used to grow trees for sion of forestland to other uses, including shifting cultiva- wood or timber. tion, permanent agriculture, ranching, settlements, or in- Permanent pasture is land used for five or more years frastructure development. Deforested areas do not include for forage, including natural crops and cultivated crops. areas logged but intended for regeneration or areas de- Only a few countries regularly report data on permanent graded by fuelwood gathering, acid precipitation, or forest pasture, as this category is difficult to assess because it fires. The extent and percentage of total area shown refer includes wild land used for pasture. to the average annual deforestation of natural forest area. SELECTED WORLD DEVELOPMENT INDICATORS 257 Estimates of forest area are derived from country sta- mercial establishments, and homes. Other withdrawals are tistics assembled by the FAQ and the United Nations those for direct industrial use, including withdrawals for Economic Commission for Europe (UNECE). In 1993 cooling thermoelectric plants and for agriculture (irriga- new assessments were published for tropical countries by tion and livestock production). the FAO and for temperate zones jointly by the UNECE Data on annual freshwater withdrawal are subject to and the FAQbut with different definitions. The FAQ variation in collection and estimation methods but are defines natural forest in tropical countries either as closed indicative of the magnitude of water use in both total and forest, where trees cover a large portion of the ground per capita terms. These data, however, also hide what can with no continuous grass cover, or as open forest, a mix of be significant variations in total renewable water forest and grassland with at least 10 percent tree cover and resources from one year to another. They also fail to dis- a continuous grass layer on the forest floor. The UNECE- tinguish the seasonal and geographic variations in water FAQ assessment defines a forest as land where tree crowns availability within a country. Because freshwater cover more than 20 percent of the area. Also included are resources are based on long-term averages, their estima- open forest formations; forest roads and firebreaks; small, tion explicitly excludes decade-long cycles of wet and dry. temporarily cleared areas; young stands expected to The data for water indicators were compiled by the achieve at least 20 percent crown cover on maturity; and World Resources Institute from various sources and pub- windbreaks and shelter belts. lished in World Resources 1996-97. The Département Nationally protected areas refers to areas of at least Hydrogeologie in Orleans, France, compiles water 1,000 hectares that fall into one of five management cate- resource and withdrawal data from published documents, gories: scientific reserves and strict nature reserves; na- including national, U.N., and professional literature. The tional parks of national or international significance (not Institute of Geography at the National Academy of Sci- materially affected by human activity); natural monu- ences in Moscow also compiles global water data on the ments and natural landscapes with some unique aspects; basis of published work and, where necessary, estimates managed nature reserves and wildlife sanctuaries; and pro- water resources and consumption from models that use tected landscapes and seascapes (which may include cul- other data, such as area under irrigation, livestock popu- tural landscapes). This table does not include sites pro- lations, and precipitation. Data for small countries and tected under local or provincial law or areas where countries in arid and semiarid zones are less reliable than consumptive uses of wildlife are allowed. The data are those for larger countries and countries with more subject to variations in definition and in reporting to the rainfall. organizations, such as the World Conservation Monitor- Table 11. Growth of the economy ing Centre, that compile and disseminate them. Annual freshwater withdrawal refers to total water Gross domestic product at purchasers' prices is the sum of withdrawal, not counting evaporation losses from storage the gross value added by all resident and nonresident pro- basins. Withdrawals also include water from desalination ducers in the economy plus any taxes and minus any sub- plants in countries where that source is a significant part sidies not included in the value of the products. It is cal- of all water withdrawal. Withdrawal data are for single culated without making deductions for depreciation of years between 1970 and 1995. Withdrawals can exceed fabricated assets or for depletion and degradation of nat- 100 percent of renewable supplies when extractions from ural resources. nonrenewable aquifers or desalination plants are consider- The GDP deflator is calculated implicitly as the ratio of able or if there is significant water reuse. Data are ex- current-price GDP to constant-price GDP. The GDP pressed as totals and as a percentage of total freshwater deflator is the most broadly based measure of changes in resources, which include both internal renewable the overall price level. (See also the note to Table 2.) resources and, where noted in the table, river flows from Agriculture comprises value added from forestry, hunt- other countries. Internal renewable water resources ing, and fishing as well as cultivation of crops and live- include flows of rivers and groundwater from rainfall in stock production. In developing countries with high lev- the country. els of subsistence farming, much agricultural production Freshwater withdrawal per capita is calculated by divid- is either not exchanged or not exchanged for money. This ing a country's total withdrawal by its population in the increases the difficulty of measuring the contribution of year for which withdrawal estimates are available. For agriculture to GDP and reduces the reliability and com- most countries, data on sectoral withdrawal per capita are parability of such numbers. calculated using sectoral withdrawal percentages estimated Industy comprises value added in mining, manufac- for 1987 to 1995. Domestic use includes drinking water, turing (also reported as a separate subgroup), construc- municipal use or supply, and use for public services, corn- tion, electricity, water, and gas. 258 WORLD DEVELOPMENT REPORT 1997 Manufacturing refers to industries belonging to divi- For information on the OECD national accounts series sions 15-37 in the International Standard Industrial Clas- see OECD, National Accounts, 1960-1994, volumes 1 sification, Revision 2. and 2. The complete set of national accounts time series is Services includes value added in all other branches of available on the World Development Indicators CD-ROM economic activity, such as wholesale and retail trade World Bank staff review the quality of national (including hotels and restaurants), transport, and govern- accounts data and, in some instances, adjust national ment, financial, professional, and personal services such as series. Because of the sometimes limited capabilities of education, health care, and real estate services. Also in- statistical offices and basic data problems, strict interna- cluded are imputed bank service charges, import duties, tional comparability cannot be achieved, especially in eco- and any statistical discrepancies noted by national com- nomic activities that are difficult to meastire, such as par- pilers as well as discrepancies arising from rescaling. allel market transactions, the informal sector, and Exports of goods and services represents the value of all subsistence agriculture. goods and other market services provided to the world. The figures for GDP are U.S. dollar values converted Included is the value of merchandise, freight, insurance, from domestic currencies using single-year official ex- travel, and other nonfactor services. Factor and property change rates. For a few countries where the official income (formerly called factor services), such as invest- exchange rate does not reflect the rate effectively applied ment income, interest, and labor income, is excluded. to actual foreign exchange transactions, an alternative Transfer payments are excluded from the calculation conversion factor is used. Note that the table does not use of GDP. the three-year averaging ("Atlas") technique applied to Growth rates of GDP and its components are calcu- GNP per capita in Table 1. lated using constant-price data in the local currency. Summary measures are computed from group aggre- Regional and income group growth rates are calculated gates of sectoral GDP in current U.S. dollars. after converting local currencies to U.S. dollars using the World Bank's International Economics Department Table 13. Structure of the economy: demand (lEG) conversion factor. Growth rates are estimated by General government consumption includes all current fitting a linear trend line to the logarithmic annual values expenditures for purchases of goods and services by all lev- of the given variable using the least-squares growth rate els of government, but excluding most government enter- method. This produces an average growth rate that corre- prises. Most capital expenditures on national defense and sponds to a model of periodic compound growth. The security are regarded as a general government consump- least-squares growth rate method and the lEG conversion tion expenditure. factor are described in the section on Statistical methods. Private consumption is the market value of all goods In calculating the summary measures, constant 1987 and services, including durable products (such as cars, U.S. dollar values for each indicator are calculated for washing machines, and home computers), purchased or each year of the periods covered, and the values are aggre- received as income in kind by households and nonprofit gated across countries for each year. The least-squares pro- institutions. It excludes purchases of dwellings but in- cedure is used to compute the aggregate growth rates. cludes imputed rent for owner-occupied dwellings. In Table 12. Structure of the economy: production practice, it may include any statistical discrepancy in the use of resources. The definitions of GDP and its components are those of Gross domestic investment consists of outlays on addi- the U.N. System of National Accounts (SNA), Series F, tions to the fixed assets of the economy plus net changes No. 2, Version 3. Version 4 of the SNA was completed in the level of inventories. only in 1993, and it is likely that many countries will con- Gross domestic saving is calculated by deducting total tinue to use the recommendations of Version 3 for the consumption from GDP. next few years. Estimates are obtained from national Exports of goods and services represents the value of all sources, sometimes reaching the World Bank through goods and other market services provided to the world. other international agencies but more often collected by Included is the value of merchandise, freight, insurance, World Bank staff. For definitions of specific components, travel, and other nonfactor services. Factor and property see the technical note to Table 11. income (formerly called factor services), such as investment National accounts data for developing countries are income, interest, and labor income, is excluded. Transfer collected from national statistical organizations and cen- payments are excluded from the calculation of GDP. tral banks by visiting and resident World Bank missions. Resource balance is the difference between exports of Data for industrial countries come from OECD data files. goods and services and imports of goods and services. SELECTED WORLD DEVELOPMENT INDICATORS 259 Summary measures in this table are computed from portrayal of the allocation of resources for various pur- group aggregates of sectoral GDP in current U.S. dollars. poses, especially in countries where lower levels of gov- ernment have considetable autonomy and are responsible Table 14. Central government budget for many economic and social services. In addition, "cen- Total revenue is derived from tax and nontax sources. Tax tral government" can mean either of two accounting con- revenue comprises compulsory, unrequited, nonrepayable cepts: consolidated or budgetary. For most countries, cen- receipts for public purposes. It includes interest collected tral government finance data have been consolidated into on tax arrears and penalties collected on nonpayment or one overall account, but for others only the budgetary late payment of taxes, and is shown net of refunds and central government accounts are available. Because bud- other corrective transactions. Nontax revenue comprises getary accounts do not always include all central govern- receipts that are not compulsory, nontepayable payments ment units, the overall picture of central government for public purposes, such as fines, administrative fees, or activities is usually incomplete. The concept employed by entrepreneurial income from government ownership of the reporting country is noted in the Key and Primary property. Proceeds of grants and borrowing, funds arising Data Documentation table. from the repayment of previous lending by governments, In general, the data presented, especially those for incurrence of liabilities, and proceeds from the sale of cap- social services, are not comparable across countries. In ital assets are not included. many economies, private health and education services are Total expenditure comprises expenditures by all gov- substantial; in others, public services represent the major ernment offices, departments, establishments, and other component of total expenditure but may be financed by bodies that are agencies or instruments of the central lower levels of government. Caution should therefore be authotity of a country. It includes both current and capi- exercised in using the data for cross-country comparisons. tal (development) expenditures. Data on central government revenues and expendi- Defense comprises all expenditures, whether by defense tures are from the IMF's Government Finance Statistics or other departments, on the maintenance of military Yearbook (1995) and IMF data files. The accounts of each forces, including the purchase of military supplies and country are reported using the system of common defini- equipment, construction, recruiting, and training. Also in tions and classifications found in the IMF's A Manual on this category are closely related items such as military aid Government Finance Statistics (1986). For complete and programs. Defense does not include expenditure on pub- authoritative explanations of concepts, definitions, and lic order and safety, which is classified separately. data sources, see these IMF sources. Social services comprises expenditures on health, educa- Table 15. Exports and imports of merchandise tion, housing, welfare, social security, and community amenities. It also covers compensation for loss of income Merchandise exports and imports, with some exceptions, to the sick and temporarily disabled; payments to the covers international movements of goods across customs' elderly, the permanently disabled, and the unemployed; borders; trade in services is not included. Exports are val- family, maternity, and child allowances; and the cost of ued f.o.b. (free on board) and imports c.i.f. (cost plus welfare services, such as care of the aged, the disabled, and insurance and freight) unless otherwise specified in the children. Many expenditures relevant to environmental foregoing sources. These values are in current U.S. dollars. defense, such as pollution abatement, water supply, sani- The categorization of exports and imports follows the tary affairs, and refuse collection, are included indistin- Standard International Trade Classification (SITC), Series guishably in this category. M, No. 34, Revision 1. Manufactures are commodities Overall deficit/surplus is defined as current and capital classified in Sections 5 through 9, excluding Division 68 revenue and official grants received, less total expenditure (nonferrous metals). Food commodities ate those in SITC and lending minus repayments. This is a broader concept Sections 0, 1, and 4 and Division 22 (food and live ani- than the current government deficit or surplus shown in mals, beverages and tobacco, animal and vegetable oils and Table 2. fats, oilseeds, oil nuts, and oil kernels). Fuels are the com- Because of differences in coverage of available data, the modities in SITC Section 3 (mineral fuels, lubricants, and individual components of central government expendi- related materials). For some countries, data for certain ture and revenue shown may not be strictly comparable commodity categories ate unavailable. across all economies. Average annual growth rates of exports and imports are Inadequate statistical coverage of state, provincial, and calculated from values at constant prices, which are de- local governments requires the use of central government rived from current values deflated by the relevant price data; this may seriously understate or distort the statistical index. The World Bank uses the price indexes produced 260 WORLD DEVELOPMENT REPORT 1997 by UNClAD for low- and middle-income economies, the definition of international reserves, in the valuation of and those presented in the IMF's International Financial gold, and in reserve management practices, the levels of Statistics for high-income economies. These growth rates reserve holdings published in national sources may not be can differ from those derived from national sources be- strictly comparable. The reserve levels for 1980 and 1995 cause national price indexes may use different base years refer to the end of the year indicated and are in current and weighting procedures from those used by UNCTAD U.S. dollars at prevailing exchange rates. See Table 2 for or the IMF. reserve holdings expressed as months of import coverage. The main source of current trade values is the UNC- The data for this table are based upon IMF data files. TAD trade data base, supplemented by the data from the World Bank staff also make estimates and, in rare IMP's International Financial Statistics, the U.N.'s Com- instances, adjust coverage or classification to enhance modity Trade (COMTRADE) data base, and World comparability between the national accounts and the bal- Bank estimates. The shares in these tables are derived ance of payments. Definitions and concepts are based on from trade values in current dollars reported in the the IMF's Balance of Payments Manual, Fifth Edition UNCTAD trade data system, supplemented by data from (1993). Values are in U.S. dollars converted at official the U.N. COMTRADE system. exchange rates. The summary measures for the growth rates are calcu- The summary measures are computed from group lated by aggregating the 1987 constant U.S. dollar price aggregates for gross international reserves. series for each year and then applying the least-squares Table 17. External debt growth rate procedure for the periods shown. Total external debt is the sum of public, publicly guaran- Table 16. Balance of payments teed, and private nonguaranteed long-term debt, use of Exports and imports ofgoods, services, and income comprise IMF credit, and short-term debt. Long-term debt has all transactions involving a change of ownership of goods three components: public, publicly guaranteed, and pri- and services between residents of a country and the rest of vate nonguaranteed loans. Public loans are external obli- the world, including merchandise, services, and income. gations of public debtors, including the national govern- Receipts of compensation to employees by, and invest- ment, its agencies, and autonomous public bodies. ment income from, nonresident entities are treated as ex- Publicly guaranteed loans are external obligations of pri- ports; payments to residents by nonresidents are treated as vate debtors that are guaranteed for repayment by a pub- imports. lic entity. Private nonguaranteed loans are external obliga- Net workers' remittances covers payments and receipts tions of private debtors that are not guaranteed for of income by migrants who are employed or expect to be repayment by a public entity. Use of IMF credit denotes employed for more than a year in their new economy, repurchase obligations to the IMF for all uses of IMF where they are considered residents. These remittances are resources, excluding those resulting from drawings in the classified as private unrequited transfers, whereas those reserve tranche. It comprises purchases outstanding under derived from shorter-term stays are included in services as the credit tranches, including enlarged access resources, labor income. The distinction accords with internation- and all special facilities (the buffer stock, compensatory ally agreed guidelines, but some developing countries clas- financing, extended fund, and oil facilities), trust fund sif workers' remittances as a factor income receipt (hence loans, and operations under the enhanced structural a component of GNP). The World Bank adheres to inter- adjustment facilities. Use of IMF credit outstanding at national guidelines in defining GNP, and therefore its year-end (a stock) is converted to U.S. dollars at the definitions may differ from national practices. dollar-SDR exchange rate then in effect. Short-term debt Other net transfers comprises net unrequited transfers is debt with an original maturity of one year or less. It other than workers' remittances. includes interest arrears on long-term debt outstanding The current account balance is the sum of net exports of and disbursed that are due but not paid on a cumulative goods and services and net transfers. basis. Available data permit no distinction between public Gross international reserves comprises holdings of mon- and private nonguaranteed short-term debt. etary gold, special drawing rights (SDRs), the reserve posi- External debt as a percentage of GNP and of exports of tion of members in the IMF, and holdings of foreign goods and services are calculated in U.S. dollars. Workers' exchange under the control of monetary authorities. The remittances are included in exports of goods and services. gold component of these reserves is valued at year-end Debt service as a percentage ofexports ofgoods and services (Deceinber 31) London prices ($589.50 an ounce in 1980 is the sum of principal repayments and interest payments and $386.75 an ounce in 1995). Because of differences in on total external debt. It is one of several conventional SELECTED WORLD DEVELOPMENT INDICATORS 261 measures used to assess a country's ability to service debt. Statistical methods Workers' remittances are included in exports of goods and This section describes the calculation of the least-squares services. growth rate, the exponential (end-point) growth rate, the The ratio ofpresent value to nominal value ofdebt is the Gini index, and the World Bank's Atlas methodology for discounted value of future debt service payments divided estimating the conversion factor used to estimate GNP by the face value of total external debt. The present value and GNP per capita in U.S. dollars. of external debt is the discounted sum of all debt service payments due over the life of existing loans. The present Least-squares growth rate value can be higher or lower than the nominal value of The least-squares growth rate, r, is estimated by fitting a debt. The determining factors for the present value being least-squares linear regression trend line to the logarithmic above or below par are the interest rates on loans and the annual values of the variable in the relevant period. The discount rate used in the present value calculation. A loan regression equation takes the form: with an interest rate higher than the discount rate yields a present value that is larger than the nominal value of debt; log X = a + bt, the opposite holds for loans with an interest rate lower than the discount rate. which is equivalent to the logarithmic transformation of The discount rates used to calculate the present value the geometric growth rate equation: are interest rates charged by OECD countries for officially supported export credits. The rates are specified for the XX(1 + r)t. Group of Seven (G-7) currencies: British pounds, Cana- dian dollars, French francs, German marks, Italian lire, In these equations, Xis the variable, tis time, and a = log Japanese yen, and U.S. dollars. International Bank for Re- X0 and b log (1 + r) are the parameters to be estimated. construction and Development (IBRD) loans and Inter- If b' is the least-squares estimate of 6, then the average national Development Association (IDA) credits are dis- annual growth rate, r, is obtained as [antilog (6*) - 1] and counted at the most recent IBRD lending rate, and IMF is multiplied by 100 to express it as a percentage. loans are discounted at the SDR lending rate. For debt The calculated growth rate is an average rate that is rep- denominated in other currencies, discount rates are the resentative of the available observations over the period. It average of interest rates on export credits charged by other does not necessarily match the actual growth rate between OECD countries. For variable rate loans, for which future any two periods. Assuming that geometric growth is the debt service payments cannot be precisely determined, appropriate way of modeling the data, the least-squares debt service is calculated using the end-1994 rates for the estimate of the growth rate is consistent and efficient. base period specified for the loan. Ivlultilateral debt as a percentage of total external debt Exponential end-point growth rate conveys information about the borrower's receipt of The growth rate between two points in time for certain aid from the World Bank, regional development banks, demographic data, notably labor force and population, is and other multilateral and intergovernmental agencies. calculated from the equation: Excluded are loans from funds administered by an in- ternational organization on behalf of a single donor r=ln(p 1p1)In government. The data on debt in this table come from the World where p and p1 are the last and first observations in the Bank Debtor Reporting System, supplemented by World period, respectively, n is the number of years in the Bank estimates. The system is concerned solely with de- period, and in is the natural logarithm operator. veloping economies and does not collect data on external This growth rate is based on a model of continuous, debt for other groups of borrowers or for economies that exponential growth. To obtain a growth rate for discrete are not members of the World Bank. Debt is stated in periods comparable to the least-squares growth rate, one U.S. dollars converted at official exchange rates. The data takes the antilog of the calculated growth rate and sub- on debt include private nonguaranteed debt reported by tracts 1. thirty developing countries and complete or partial esti- mates for an additional twenty that do not report but for The Gini index which this type of debt is known to be significant. The Gini index measures the extent to which the distrib- The summary measures are taken from the World ution of income (or, in some cases, consumption expen- Bank's Global Development Finance 1997. diture) among individuals or households within an econ- 262 WORLD DEVELOFMENTREFORT 1997 omy deviates from a perfectly equal distribution. A Lorenz / 5$ / ss' curve plots the cumulative percentages of total income received against the cumulative percentage of recipients, e t3 Pt 1Pt 5$ Pt-2 Pt_2) + e_1 s$ \Pt_1 Pt_I I +e starting with the poorest individual or household. The Gini index measures the area between the Lorenz curve and a hypothetical line of absolute equality, expressed as a and for calculating GNP per capita in U.S. dollars for percentage of the maximum area under the line, Thus a year t Gini index of zero presents perfect equality, whereas an index of 100 percent implies maximum inequality. $ =( /N)Ie The World Bank employs a numerical analysis pro- where: gram, POVCAL, to estimate values of the Gini index; see = current GNP (in local currency) for year t; Chen, Datt, and Ravallion 1992. Pt = GNP deflator for year t; et = average annual exchange rate (units of national World Bank Atlas method currency per U.S. dollar) for year t; The Atlas conversion factor for any year is the average of IV = midyear population for year t; a country's exchange rate (or alternative conversion factor) = SDR deflator in U.S. dollar terms for year t. for that year and its exchange rates for the two preceding years, after adjustment for differences in rates of inflation Alternative conversion factors between the country in question and the G-5 countries The World Bank systematically assesses the appropriate- (France, Germany, Japan, the United Kingdom, and the ness of official exchange rates as conversion factors. An United States). The inflation rate for the G-5 countries is alternative conversion factor is used when the official represented by changes in the SDR deflators. This three- exchange rate is judged to diverge by an exceptionally year averaging smoothes annual fluctuations in prices and large margin from the rate effectively applied to domestic exchange rates for each country. The Atlas conversion fac- transactions of foreign currencies and traded products; tor is applied to the country's GNP. The resulting GNP this is the case for only a small number of countries (see in U.S. dollars is divided by the midyear population for the Key and Primary Data Documentation table). Alter- the latest of the three years to derive GNP per capita. native conversion factors are used in the Atlas method and The following formulas describe the procedures for elsewhere in the World Development Indicators as single- computing the conversion factor for year t: year conversion factors. DATA SOURCES Ahmad, Sultan. 1992. "Regression Estimates of Per Capita GDP Based on UNCTAD (United Nations Conference on Trade and Development). Purchasing Power Parities." Policy Research Working Paper 956. Various years. Handbook ofInternational Trade and Development Sta- World Bank, International Economics Department, Washington, D.C. tistics. Geneva. Bos, Eduard, My T. Vu, Ernest Massiah, and Rodolfo A. Bulatao. UNESCO (United Nations Educational, Scientific, and Cultural Orga- 1994. World Population Projections, 1994-95 Edition. Baltimore, nization). Various years. Statistical Yearbook. Paris. Md.: Johns Hopkins University Press. UNICEF (United Nations Children's Fund). 1997. The State of the Council of Europe. 1995. Recent Demographic Developments in Europe World's Children 1997. Oxford: Oxford University Press. and North America. Council of Europe Press. UNIDO (United Nations Industrial Development Organization). Eurostat (Statistical Office of the European Communities). Various 1996. International Yearbook of Industrial Statistics 1996. years. Statistical Yearbook. Luxembourg. Vienna. FAO (Food and Agriculture Organization). Various years. Production United Nations. 1968. A System ofNationalAttounts: Studies and Meth- Yearbook. FAO Statistics Series. Rome. ods. Series F. No. 2. Rev. 3. New York. lEA (International Energy Agency). 1996. Energy Statistics and Bala nces 1985. NationalAccounts Statistics: Compendium ofIncome Dis- ofNon-OECD Countries 1993-94. Paris. tribution Statistics. New York. 1996. Energy Statistics of OECD Countries 1993-94. Paris. 1994. World Urbanization Prospects, 1994 Revision. New York. ILO. (International Labour Organisation). 1995. Year Book of Labour 1996. World Population Prospects: The l996Edition. New York. Statistics. Geneva. Various years. Energy Statistics Yearbook. New York. 1995. Labour Force Estimates and Projections, 1950-2010. Various years. Levels and Trends of Contraceptive Use. New York. Geneva. Various issues. Monthly Bulletin of Statistics. New York. 1995. Estimates oft/se Economically Active Population by Sex Various years. Population and Vital Statistics Report. New York. andAge Group and by Main Sectors ofEconomic Activity. Geneva. Various years. Statistical Yearbook. New York. IMF. (International Monetary Fund). 1986. A Manual on Government Various years. Update on the Nutrition Situation. Administra- Finance Statistics. Washington, D.C. tive Committee on Co-ordination, Subcommittee on Nutrition. 1993. Balance ofPayments Manual. 5th ed. Washington, D.C. Geneva. Various years. Government Finance Statistics Yearbook. Wash- Various years. Yearbook of International Trade Statistics. New ington, D.C. York. Various years. Inttrnacional Financial Statistics. Washington, WHO. (World Health Organization). 1991. Maternal Mortality: A D.C. Global Factbook. Geneva. OECD. (Organization for Economic Cooperation and Development). Various years. World Health Statistics. Geneva. 1988. Geographical Distribution of Financial Flaws to Developing Various years. World Health Statistics Report. Geneva. Countries. Paris. World Bank. 1993. Purchasing Power of Cu rrencies: ComparingNational 1996. Development Co-operation: 19.95 Report. Paris. Incomes Using ICP Data. Washington, D.C. 1996. NacionalAccounts 1960-1994. Vol. 1, Main Aggregates. 1997. Global Development Finance 1997. Washington, Paris. D.C. 1996. National Accounts 1960-1994. Vol. 2, Detailed Tables. World Resources Institute in collaboration with UNEP (United Paris. Nations Environment Programme) and UNDP (United Nations Various years. Development Co-operation. Paris. Development Programme). 1994. World Resources 1994-95: A Ravalbon, Martin, and Shaohua, Chen. 1996. "What Can New Survey Guide to the Global Environment. New York: Oxford University Data Tell Us About Recent Changes in Living Standards in Devel- Press. oping and Transitional Economies?" World Bank, Policy Research World Resources Institute, UNEP (United Nations Environment Pro- Department, Washington, D.C. gramme), UNDP (United Nations Development Programme), and U. S. Bureau of the Census. 1996. World Population Profile. Washing- World Bank. 1996. World Resources 1996-97: A Guide to the Global ton, D.C.: U.S. Government Printing Office. Environment. New York: Oxford University Press. 264 WORLD DEVELOPMENT REPORT 1997 Table 1. Classification of economies by income and region, 1997 Sub-Saha ran Africa Europe and Central Asia East and Asia Eastern Middle East and North Africa Income Southern East Asia Europe and Rest of Middle North group Subgroup Africa West Africa and Pacific South Asia CentralAssa Europe East Africa Americas Angola Benin Cambodia Afghanistan Albania Yemen, Rep. Guyana Burundi Burkina Paso China Bangladesh Armenia Haiti Comoros Cameroon Lao PDR Bhutan Azerbaijan Honduras Eritrea Central African Mongolia India Bosnia and Nicaragua Ethiopia Republic Myanmar Nepal Herzegovina Kenya Chad Vietnam Pakistan Georgia Madagascar Congo Sri Lanka Kyrgyz Malawi Core d'Ivoire Republic Mozarnbique Equatorial Tajikistan Rwanda Guinea Somalia Gambia, The Low- Sudan Ghana income Tanzania Guinea Uganda Guinea-Bissau Zaire Liberia Zambia Mali Zimbabwe Mauritania Niger Nigeria São Tome and Principe Senegal Sierra Leone Togo Botswana Cape Verde Fiji Maldives Belarus Turkey Iran, Islamic Algeria Belize Djibouti Indonesia Bulgaria Rep. Egypr, Arab Bolivia Lesotho Kiribati Estonia Iraq Rep. Colombia Namibia Korea, Dem. Kazakstan Jordan Morocco Costa Rica Swaziland Rep. Lacvia Lebanon Tunisia Cuba Marshall Lithuania Syrian Arab Dominica Islands Macedonia, Republic Dominican Micronesia, FYRa West Bank Republic Fed. Sts. Moldova and Gaza Ecuador Papua New Poland El Salvador Lower Guinea Romania Grenada Philippines Russian Guatemala Solomon Federation Jamaica Islands Slovak Panama Thailand Republic Paraguay Tonga Turkmenistan Peru Vanuaru Ukraine St. Vincent MieIdl.e- Western Uzbekistan and the income Samoa Yugoslavia, Grenadines Fed. Rep.b Suriname Venezuela Mauritius Gabon American Croatia Greece Bahrain Libya Antigua and Mayotte Samoa Czech Isle of Man Oman Barbuda Seychelles Malaysia Republic Malta Saudi Arabia Argentina South Africa Hungary Barbados Slovenia Brazil Chile Guadeloupe Mexico Puerto Rico Upper Sr. Kitts and Nevis St. Lucia Trinidad and Tobago Uruguay Subtotal: 158 26 23 21 8 27 4 10 5 34 SELECTED WORLD DEVELOFMENT INDICATORS 265 Table 1. (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 Pacf1c South Asia CentralAsia Europe East Africa Americas Australia Austria Canada Japan Belgium United States New Zealand Denmark Korea, Rep. Finland France Germany Iceland Ireland OECD Italy countries Luxembourg Netherlands High- Norway income Portugal Spain Sweden Switzerland United Kingdom Reunion Btunei Andorra Israel Aruba French Channel Kuwait Bahamas, The Polynesia Islands Qatar Bermuda Nsn-OECD Guam Cyprus United Arab Cayman Hong Kong Faeroe Islands Emirates Islands countries Macao Greenland French New Liechtenstein Guiana Caledonia Monaco Martinique N. Mariana Netherlands Islands Antilles Singapore Virgin Islands OAE' (U.S.) Total: 210 27 23 34 8 27 28 14 5 44 Former Yugoslav Republic of Macedonia. Federal Republic of Yugoslavia (Serbia/Montenegro). Other Asian economiesTaiwan, China. For operational and analytical purposes, the World Definitions of groups Bank's main criterion for classifying economies is gross These tables classify all World Bank member countries national product (GNP) per capita. Every economy is and all other economies with populations of more than classified as low-income, middle-income (subdivided into 30,000. lower-middle and upper-middle), or high-income. Other analytical groups, based on geographic regions, exports, Income group: Economies are divided according to 1995 and levels of external debt, are also used. GNP per capita, calculated using the World Bank Atlas method. The groups are: low-income, $765 or less; lower- Low-income and middle-income economies are some- middle-income, $766$3 ,03 5; upper-middle-income, times referred to as developing economies. The use of the $3,036$9,385; and high-income, $9,386 or more. term is convenient; it is not intended to imply that all economies in the group are experiencing similar develop- The estimates for the republics of the former Soviet ment or that other economies have reached a preferred or Union are preliminary and their classification will be kept final stage of development. Classification by income does under review. not necessarily reflect development status. Distributors of CHINA China Financial & Economic GERMANY ISRAEL UNO-Verlag Yozmot Literature Ltd. World Bank Publishing House Poppelsdorfer Al lee 55 P0. Box 56055 Publications 8, Da Fo Si Dong Jie 53115 Bonn 3 Yohanan Hasandlar Street Prices and credit terms vary Beijing Tel: (49 228) 212940 Tel Aviv 61560 from country to country. Tel: (86 10) 6333-8257 Fax: (49 228) 217492 Tel: (9723) 5285-397 Consult your local distribu- Fax: (86 10) 6401-7365 Fax: (972 3) 5285-397 tor before placing an order. GREECE COLOMBIA Papasotiriou S.A. ROY. International Intoenlace Ltda. 35, Stournara Str. P0 Box 13056 ARGENTINA Carrera 6 No.51-21 10682 Athens Tel Aviv 61130 Oficina del Libro Intemacional Apartado Aereo 34270 Tel: (30 1)364-1826 Tel: (972 3) 5461423 Av. Cordoba 1877 Santafé de Bogota, D.C. Fax: (30 1)364-8254 Fax: (972 3) 5461442 1120 Buenos Aires Tel: (57 1)285-2798 E-mail: royil@netvision.net.il Tel:(541)815-8354 Fax: (57 1) 285-2798 HAITI Fax:(541)815-8156 Culture Diffusion Palestinian Authority/Middle East COlE D'IVOIRE 5, Rue Capois Index Information Services AUSTRALIA, FIJI, PAPUA NEW Center d'Edition et de Diffusion CR257 P0.B. 19502 Jerusalem GUINEA, SOLOMON ISLANDS, Africaines (CEDA) Port-au-Prince Tel: (9722)6271219 VAN UATU, AND WESTERN 04 BR 541 Tel: (509 1) 39260 Fax: (9722) 6271634 SAMOA Abidjan 04 D.A. Information Services Tel: (225)24 6510;24 6511 HONG KONG, MACAD ITALY 648 Whitehorse Road Fax: (225) 250567 Asia 2000 Ltd. Licosa Commissionaria Sansoni Mitcham 3132 Sales & Circulation Department SPA Victoria CYPRUS Seabird House, unit 1101-02 Via Duca Di Calabria, 1/1 Tel: (61)39210 7777 Center for Applied Research 22-28 Wyndham Street, Central Casella Postale 552 Fax: (61)392107788 Cyprus College Hong Kong 50125 Firenze E-mail: service@dadirect.com.au 6, Diogenes Street, Engomi Tel: (852) 2530-1409 Tel: (55) 645-415 URL: http://www .dadirect.com.au P0. Box 2006 Fax: (852) 2526-1107 Fax: (55) 641-257 Nicosia E-mail: sales@asia2000.com.hk E-mail: licosa@ftbcc.it AUSTRIA Tel: (3572)44-1730 URL: http:llwww .asia2000.com.hk Url: http://www .ftbcc.itllicosa Gerold and Co. Fax: (3572)46-2051 Weihburggasse 26 INDIA JAMAICA A-1011 Wien CZECH REPUBLIC Allied Publishers Ltd. Ian Randle Publishers Ltd. Tel: (43 1)512-47-31-0 National Information Center 751 Mount Road 206 Old Hope Road Fax: (43 1)512-47-31-29 prodejna, Konviktska 5 Madras - 600 002 Kingston 6 URL: http:/Iw'wi .gerold.co/at.online CS-113 57 Prague 1 Tel: (91 44) 852-3938 Tel: 809-927-2085 Tel: (422) 2422-9433 Fax: (91 44) 852-0649 Fax: 809-977-0243 BANGLADESH Fax: (422)2422-1484 E-mail: irpl@colis.com Micro Industries Development URL: http://www .nis.cz/ INDONESIA Assistance Society (MIDAS) Pt. Indira Limited JAPAN House 5, Road 16 DENMARK Jalan Borobudur 20 Eastern Book Service Dhanmondi R/Area SamfundsLitteratur P0. Box 181 3-13 Hongo 3-chome, Bunkyo-ku Dhaka 1209 Rosenoerns AIlé 11 Jakarta 10320 Tokyo 113 Tel: (8802) 326427 DK-1970 Frederiksberg C Tel: (62 21) 390-4290 Tel: (81 3) 3818-0861 Fax: (8802)811188 Tel: (45 31) 351942 Fax: (62 21) 421-4289 Fax: (81 3)3818-0864 Fax: (45 31) 357822 E-mail: orders@svt-ebs.co.jp BELGIUM IRAN URL: http://www .bekkoame.or.jp/ Jean De Lannoy EGYPT, ARAB REPUBLIC OF Ketab Sara Co. Publishers -svt-ebs Av.duRoi2O2 Al Ahram Distribution Agency Khaled Eslamboli Ave., 1060 Brussels Al Galaa Street 6th Street KENYA Tel: (322) 538-5169 Cairo Kusheh Delafrooz No. 8 Africa Book Service (E.A.) Ltd. Fax: (322) 538-0841 Tel: (202) 578-6083 P0. Box 15745-733 Quaran House, Mtangario Street Fax: (202) 578-6833 Tehran P0. Box 45245 BRAZIL Tel: (98 21) 8717819; 8716104 Nairobi PublicacOes Tecnicas Internacionais The Middle East Observer Fax: (9821) 8712479 Tel: (2542) 223 641 Ltda. 41, Sherit Street E-mail: ketab-sara@ neda.net.ir Fax: (254 2) 330 272 Rua Peixoto Gomide, 209 Cairo 01409 Sao Paulo, SR Tel: (202) 393-9732 Kowkab Publishers KOREA, REPUBLIC OF Tel: (5511)259-6644 Fax: (202) 393-9732 P0. Box 19575-511 Daejon Trading Co. Ltd. Fax: (5511)258-6990 Tehran P0. Box 34, Youida E-mail: postmaster@pti.uol.br FINLAND Tel: (98 21) 258-3723 706 Seoun Bldg URL: http://www .uol.br Akateeminen Kirjakauppa Fax: (98 21) 258-3723 44-6 Youido-Dong, Yeongchengpo- P0. Box 128 Ku CANADA FIN-00101 Helsinki IRELAND Seoul Renouf Publishing Co. Ltd. Tel: (3580)121-4418 Government Supplies Agency Tel: (822) 785-1631/4 5369 Canotek Road Fax: (3580) 121-4435 Oitig an tSoláthair Fax: (822) 784-0315 Oft awa, Ontario K1J 9J3 E-mail: akatilaus@stockmann.fi 4-5 Harcourt Road Tel: (613) 745-2665 URL: http://www.akateeminen.com/ Dublin 2 MALAYSIA Fax: (613) 745-7660 Tel: (353 1) 661-3111 University of Malaya Cooperative E-mail:order.dept@renoufbooks.com FRANCE Fax: (3531)475-2670 Bookshop, Limited URL: http// .renoufbooks.coni World Bank Publications P0. Box 1127 66, avenue d'léna Jalan Pantai Baru 75116 Paris 59700 Kuala Lumpur Tel: (33 1) 40-69-30-56/57 Tel: (603) 756-5000 Fax: (33 1)40-69-30-68 Fax: (603) 755-4424 MEXICO PERU SOUTH AFRICA, BOTSWANA THAILAND INFOTEC Editorial Desarrollo SA For single titles: Central Books Distribution Av. San Fernando No. 37 Apartado 3824 Oxford University Press 306 Silom Road Col. Toriello Guerra Lima 1 Southern Africa Bangkok 10500 14050 Mexico, D.E Tel: (51 14) 285380 P0. Box 1141 Tel: (662)235-5400 Tel: (52 5) 624-2800 Fax: (51 14) 286628 Cape Town 8000 Fax: (662) 237-8321 Fax: (525) 624-2822 Tel: (27 21) 45-7266 E-mail: infotec@rtn.net.mx PHILIPPINES Fax: (27 21) 45-7265 TRINIDAD & TOBAGO, URL: http:/Irtn.net.mx International Booksource Center AND THE CARAIBBEAN Inc. For subscription orders: Systematics Studies Unit NEPAL 1127-A Antipolo St. International Subscription Service 9 Watts Street Everest Media International Barangay, Venezuela PD. Box 41095 Curepe Services (P) Ltd. Makati City Craighall Trinidad, West Indies GPO Box 5443 Tel: (632)8966501; 6505; 6507 Johannesburg 2024 Tel: (809) 662-5654 Kathmandu Fax: (632)8961741 Tel: (27 11) 880-1448 Fax: (809) 662-5654 Tel: (977 1)472 152 Fax: (27 11) 880-6248 E-mail: tobe@trinidad.net Fax: (977 1) 224 431 POLAND E-mail: iss@is.co.za International Publishing Service UGANDA NETHERLANDS UI. Piekna 3 1/37 SPAIN Gustro Ltd. De Lindeboom/InOr-Publikaties 00-677 Warzawa Mundi-Prensa Libros, S.A. P0 Box 9997 P0. Box 202 Tel: (482) 628-6089 Castello 37 Madhvani Building 7480 AE Haaksbergen Fax:(482)621-7255 28001 Madrid Plot 16/4Jinia Rd. Tel: (31 53) 574-0004 E-mail: books%ips@ikp.atm.com.pI Tel:(341)431-3399 Kampala Fax: (31 53) 572-9296 URL: http:// Fax: (34 1) 575-3998 Tel: (256 41) 254 763 E-mail: lindeboo@worldonline.nl www.ipscg.waw.pI/ips/export/ E-mail: Iibreria@mundiprensa.es Fax: (256 41) 251 468 URL: http:II URL: http://www .mundiprensa.es/ www.worldonline.nl/-Iindeboo PORTUGAL UNITED KINGDOM Livraria Portugal Mundi-Prensa Barcelona Microinfo Ltd. NEW ZEALAND Apartado 2681 Consell de Cent, 391 P0. Box 3 EBSCO NZ Ltd. Rua Do Carmo 70-74 08009 Barcelona Alton, Hampshire GU34 2PG Private Mail Bag 99914 1200 Lisbon Tel: (343) 488-3492 England New Market Tel: (1) 347-4982 Fax: (343)487-7659 Tel: (44 1420) 86848 Auckland Fax: (1) 347-0264 E-mail: barcelona@mundiprensa.es Fax: (44 1420) 89889 Tel: (649) 524-8119 E-mail: wtank@ukminfo.demon.co.uk Fax: (649) 524-8067 ROMANIA SRI LANKA, THE MALDIVES URL: http://www .microinto.co.uk Compani De Librarii Bucuresti S.A. Lake House Bookshop NIGERIA Str. Lipscani no. 26, sector 3 100, Sir Chittampalam Gardiner VENEZUELA University Press Limited Bucharest Mawatha Tecni-Ciencia Libros, S.A. Three Crowns Building Jericho Tel: (401)6139645 Colombo 2 Centro Cuidad Comercial Tamanco Private Mail Bag 5095 Fax: (40 1) 312 4000 Tel: (941) 32105 Nivel C2 Ibadan Fax: (941)432104 Caracas Tel: (234 22) 41-1 356 E-mail: LHL@sri.lanka.net Tel: (582) 959 5547; 5035; 0016 Fax: (234 22) 41 -2056 RUSSIAN FEDERATION Fax: (58 2) 959 5636 lsdatelstvo SWEDEN NORWAY 9a, Lolpachniy Pereulok Wennergren-Williams AB ZAMBIA NICInfoNS Moscow 101831 P0. Box 1305 University Bookshop Book Depariment Tel: (7095)9178749 S-171 25 Solna University of Zambia Postboks 6512 Etterstad Fax: (7095)9179259 Tel: (468) 705-97-50 Great East Road Campus N-0606 Oslo Fax: (468) 27-00-71 P0. Box 32379 Tel: (47 22) 97-4500 SINGAPORE, TAIWAN, E-mail: mail@wwi.se Lusaka Fax: (47 22) 97-4545 MYANMAR, BRUNEI Tel: (2601)252576 Asahgate Publishing Asia SWITZERLAND Fax: (260 1)253952 PAKISTAN Pacific PIe. Ltd. Librairie Payot Mirza Book Agency 41 Kallang Pudding Road #04-03 Service Institutionnel ZIMBABWE 65, Shahrah-e-Quaid-e-Azam Golden Wheel Building Côtes-de-Montbenon 30 Longman Zimbabwe (Pvt.)Ltd. Lahore 54000 Singapore 349316 1002 Lausanne Tourle Road, Ardbennie Tel: (92 42) 735 3601 Tel: (65) 741-5166 Tel: (41 21) 341-3229 P0. Box ST125 Fax: (92 42) 758 5283 Fax: (65) 742-9356 Fax: (41 21) 341-3235 Southerton E-mail: ashgate@asianconnect.com Harare Oxford University Press ADECO Van Diermen Editions Tel: (2634)6216617 5 Bangalore Town SLOVENIA Techniques Fax: (2634) 621670 Sharae Faisal Gospodarski Vestnik Publishing Ch. de Lacuez 41 P0 Box 13033 Group CH1807 Blonay Karachi-75350 Dunajska cesta 5 Tel: (41 21) 943 2673 Tel: (92 21) 446307 1000 Ljubl(ana Fax: (41 21) 943 3605 Fax: (92 21) 4547640 Tel: (386 61) 1338347; 1321230 E-mail: oup@oup.khi.erum.com. pk Fax:(38661) 1338030 TANZANIA 04,2a/g7 E-mail: belicd@gvestnik.si Oxford University Press Pak Book Corporation Maktaba Street Aziz Chambers 21 P0 Box 5299 Queen's Road Dar es Salaam Lahore Tel: (255 51) 29209 Tel: (92 42) 636 3222; 636 0885 Fax: (255 51) 46822 Fax: (92 42) 636 2328 E-mail: pbc@brain.net.pk World Development Indicators 1997 Formerly the statistical appendix to the World Development Report, these comprehensive data have been enlarged to include more than 75 tables with some 600 indicators. This major new publication provides an expanded view of the world economy for nearly 150 countrieswith chapters focusing on people, economy, environment, states and markets, and global links. Concise, insightful commentary tells the story of how people live and work, and how countries are expanding and changing. 376 Pages Stock no. 13701 (ISBN 0-8213-3701-7) $60.00 World Development Indicators on CD-ROM This comprehensive database, which replaces World Data, contains most of the underlying time-series data for the World Development Indicators and World Bank A/las. We've added powerful new featuresnow you can generate maps and charts, and download your results to other software programs. Requires Windows 3.1 Individual Version: Stock no. 13703 (ISBN 0-8213-3703-3) $275.00 Network Version: Stock no. 13702 (ISBN 0-8213-3702-5) $550.00 One of the Bank's most popular offerings, the Atlas has been redesigned as a companion to the WorldDevelopment Indicators. Tables, charts, and 21 colorful maps address the development themes of people, economy, environment, and states and markets. This easy-to-use, inexpensive book is an international standard in statistical compilations and an ideal reference for office or classroom. Text, maps, and references appear in English, French, and Spanish. 48 Pages Stock no. 13576 (ISBN 0-8213-3576-6) $15.00 World Bank Publications In the USA, contact The Wortd Bank (order form below), P.O. Box 7247-8619, Philadetphia, PA t9t70-5619 or phone: (703) 661-1580, Fax: (703) 661-t5Ot. Shipping and handling: US$5.00. For airmail deliveryoutside the USA, add US$13.00 for one item plus US$6.00 for each addi- tionat item. Payment by US$ check to the Wortd Bank or by VISA, MasteeCard. or American Express. Customers outside the USA, ptease contact your Wortd Bank distributor. To find the distributor in your country, please contact the World Bank Office of the Publisher by fax: 202-522-2631 or check the World Bank website www.worldbank.org/. Click on Publications. Quantity Title Stock # Price Total Price IJ Bank check U VISA U MasterCard U American Express Subtotal cost US$ Shipping and handting US$ credit card account number Total US$ Expiration Date Signature (required to vatidate all orders) PLEASE PRINT CLEARLY Name Address City State Postal Code Country Telephone NK 1818 LI Streci, !Wj hington 1) ( 20 I 't TELEPHONE: 202-477-1234 TELEX: MCI 64145 W4)RLUBANK MCI 248423 WORLI)IJAiK CABLE ADDRESS: tNTBAFRAI) WASH INGTOINDC WORLD WIDE 'dyE B: htip://WWW.ROri(Ibitflk.Ot'g/ E-M Al t..: books@world I)a1Ik.ort COVER DESIGN BY THE MAGAZINE GROUP