WORLD DEVELOPMENT REPORT 2002 22825 September 2001 =~~'-a Building Institutions for Markets THE WORLD BANK WORLD DEVELOPMENT REPORT 2002 Bul in.lg nstitutions for Markets Published for the World Bank Oxford University Press Oxford University Press OXFORD NEW YORK ATHENS AUCKLAND BANGKOK BOGOTA BUENOS AIRES CALCUTTA CAPE TOWN CHENNAI DAR ES SALAAM DELHI FLORENCE HONG KONG ISTANBUL KARACHI KUALA LUMPUR MADRID MELBOURNE MEXICO CITY MUMBAI NAIROBI PARIS SAO PAULO SINGAPORE TAIPEI TOKYO TORONTO WARSAW and associated companies in BERLIN IBADAN C) 2002 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. Published by Oxford University Press, Inc. 200 Madison Avenue, New York, N.Y. 10016 Oxford is a registered trademark of Oxford University Press. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Cover design by Debra Naylor, Naylor Design, Inc. Cover photographs: top, © 2001 Stephen Simpson/FPG; middle, C 2001 Stone/Paul Chesley; bottom, Yosef Hadar, World Bank. Inside design and typesetting by Barton Matheson Willse & Worthington, Baltimore. Manufactured in the United States of America First printing September 2001 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 0-19-521607-5 clothbound ISBN 0-19-521606-7 paperback ISSN 0163-5085 Text printed on recycled paper. r o revvo r T his World Development Report is about building We recognize the central importance of institutions in market institutions that promote growth and the development process through the Comprehensive I reduce poverty, addressing how institutions Development Framework, which stresses the interde- support markets, what makes institutions work, and pendence of institutions with the human, physical, and how to build them. macroeconomic sides of development. This theme is a natural continuation of last year's The Report emphasizes the importance of histori- Report, which demonstrated that markets are central cal context: where countries are today affects where to the lives of poor people, that institutions play an im- they can go. It also takes a pragmatic approach to in- portant role in how markets affect people's standards of stitution building, focusing on what can be done prac- living and help protect their rights. This Report identi- tically rather than on what should be done in an ideal fies how institutions can promote inclusive and inte- world. Social and political factors affect the pace of grated markets, and ensure stable growth and thus dra- change, and sweeping reforms are not always possible. matically improve people's incomes and reduce poverty. It is important to work on the areas where opportuni- It is about equal opportunity and empowerment for ties present themselves; each step can take countries people, especially the poor. forward-if correctly designed. And smaller reforms Some countries have successfully harnessed market- can build constituencies for larger ones. oriented reforms to improve the welfare of all their peo- This Report recognizes that one size does not fit all ple. But in other countries, markets have not given peo- in institution building and provides policy guidance on ple as much incentive to engage in wider trade, the how to develop appropriate institutions. Building on ability to use fully their skills and resources, and oppor- the successes of countries, and learning from the fail- tunities to increase their income. ures, the Report provides a deeper understanding of Effective institutions can make the difference in the market-supporting institutions and a better apprecia- success of market reforms. Without land-titling insti- tion of how people may build such institutions. In iden- tutions that ensure property rights, poor people are un- tifying how to promote institutional change, it looks at able to use valuable assets for investment and income the roles of private and public, and national, local, and growth. Without strong judicial institutions that en- international, actors. It draws on a wealth of research force contracts, entrepreneurs find many business ac- and practical experience from inside and outside the tivities too risky. Without effective corporate gover- Bank, as well as on insights from many disciplines, pre- nance institutions that check managers' behavior, firms senting new research and data on institutions. waste the resources of stakeholders. And weak institu- From these diverse sources, the Report distills four tions hurt the poor especially. For example, estimates lessons on building effective institutions: show that corruption can cost the poor three times as much as it does the wealthy. a Design them to complement what exists-in terms Addressing the challenge of building effective institu- of other supporting institutions, human capabilities, tions is critical to the Bank's mission of fighting poverty. and available technologies. The availability and costs III IV WORLD DEVELOPMENT REPORT 2002 of supporting institutions and capacity determine * Promote competition among jurisdictions, firms, the impact of any particular institution. By under- and individuals. Developing country market actors standing how institutions interact, we can identify often face too little competition, and changing this priorities, will significantly improve institutional quality. * Innovate to identify institutions that work-and Greater competition modifies the effectiveness of ex- those that do not. Sometimes this requires experi- isting institutions, creates demand for new ones, and mentation. Even in countries with similar incomes increases choice for consumers. Competition among and capacities, innovation can create stronger insti- jurisdictions highlights successful institutions and tutions because of differences in local conditions, promotes demand for them. Competition among differences that range from social norms to geogra- firms and individuals does the same. phy. Countries can gain from expanding successful public innovations and adopting private innova- These broad lessons, as well as the detailed analysis tions. But they must also have the courage to drop and many examples throughout this Report, will help failing experiments. us and policymakers build institutions that ensure sta- * Connect communities of market players through ble and inclusive growth and thus improve people's in- open information flows and open trade. Exchanging comes and reduce poverty. information changes behavior. It creates demand for institutional change by holding people to account and by supplying ideas for change from outside the community. Linking communities of people in net- works of information and trade is thus a priority for those building market-supporting institutions. James D. Wolfensohn FOREWORD V This Report has been prepared by a team led by Roumeen Islam and comprising Arup Banerji, Robert Cull, Asli Demirgiic-Kunt, Simeon Djankov, Alexander Dyck, Aart Kraay, Caralee McLiesh, Russell Pittman; and Helena Tang, Nazmul Chaudhury, Jeffrey Hammer, Richard Messick, and Tatiana Nenova made additional contributions. The team was assisted by Theodora Galabova, Paramjit Gill, Yifan Hu, Olga loffe, Claudio Montenegro, Stefka Slavova, Mahesh Surendran, and LiHong Wang. Andrei Shleifer and Joseph E. Stiglitz provided valuable suggestions during the writing of the Report. Bruce Ross-Larson is the editor of the overview, chapter 1, and chapter 10. Andrew Balls provided editorial assistance. The work was carried out under the general direction of Nicholas Stern. The team was advised by a panel of experts comprising Carl Anduri, Abhijit Banerjee, Kaushik Basu, Tim- othy Besley, Francois Bourguignon, Antonio Estache, Cheryl Gray, Avner Greif, Nurul Islam, Emmanuel Jimenez, Daniel Kaufmann, Michael Klein, Yingyi Qian, and Kenneth Sokoloff. Many others inside and outside the World Bank provided helpful comments and wrote background papers and other contributions, and their names are listed in the bibliographical note. Much of the background re- search was supported by a generous grant from the Dutch government. Research was also supported by the Swiss Trust Fund. The World Bank Development Data Group contributed to the data appendix and was re- sponsible for the Selected World Development Indicators. The team undertook a wide range of consultations for this Report, from the initial outline to the final draft. During the Report's planning stage in 2000, a February workshop in Berlin and a July workshop in Washington, D.C. provided an exchange of ideas among academics and policymakers from around the world. During the drafting stage in 2001, a consultative meeting on the media was held in April, and a consultative meeting on judicial systems was held in May. In addition, the authors held consultations with a wider com- munity that included nongovernmental organizations, holding meetings in Paris (with representatives from the French Conseil d' Etat, the Organisation for Economic Co-operation and Development, the French De- velopment Agency, and the Competition Council); in London (Department for International Development, Overseas Development Institute, and nongovernmental organizations); and in Amsterdam (Amsterdam Insti- tute for International Development). The team also conducted a series of videoconferences with audiences in Bangladesh, Brazil, Egypt, Japan, Mexico, Morocco, South Africa, and Thailand. A consultation with the In- ternational Confederation of Free Trade Unions was also held. Rebecca Sugui served as executive assistant to the team; Leila Search as program assistant and technical support; and Shannon Hendrickson, Joanna Kata-Blackman, Mei-Ling Lavecchia, and Rudeewan Laohakit- tikul as team assistants. Maria Ameal and later Eva Santo Domingo served as resource management assistant. Book design, editing, and production were coordinated by the Production Services Unit of the World Bank's Office of the Publisher. PART q N7iROdUCTiON I Buildinr lnstitutions: Complement lrtrovzi ' C or nect, and Compete 3 How do institutions support markets?. .. 5 How do institutions support growth and poverty reduction?. . . 9 How do you build effective institutions? ...10 Organization and scope of the Report . . .25 Conclusions. . . 26 PART Xi F I R ! S L Farmers Building more secure and transferable rural land institutions .32 Building effective and accessible rural financial institutions .39 Building effective institutions for agricultural technology and innovation .43 Conclusions .52 3 Governance of Firms 55 What firms around the world look like .57 Private governance institutions for firms .58 Laws and formal intermediaries .63 Conclusions .73 4 Financial Systenmc 75 Should policymakers promote bank-based or market-based financial systems? .77 What form should financial regulation take? .79 Enhancing efficiency in the financial sector: the role of ownership and competition .84 How foreign entry and e-finance can change the nature of financial markets .88 How to enhance access to financial services .91 Conclusions .96 VII VIII WORLD DEVELOPMENT REPORT 2002 PART III GOVERNMENT 5 Political Institutions and Governance 99 Political institutions and policy choices ................................................. 101 Corruption ..................................................................... 105 Politics, institutions, and taxation . .................................................... 110 Conclusions ................................................................... 115 6 The Judicial System 117 Comparison of legal and judicial systems ............................................... 120 New evidence on two aspects of judicial efficiency: speed and cost ............................ 121 Judicial reform efforts .............................................................. 124 Fairness .................................................................... 129 Conclusions ................................................................... 131 7 Competition 133 Domestic competition . ............................................................ 135 International competition ........................................................... 142 Conclusions ................................................................... 149 8 Regulation of Infrastructure 151 Competition in infrastructure sectors .................................................. 154 Structure of the regulatory system ..................................................... 158 Designing infrastructure regulation to deliver services to poor people .......................... 161 Conclusions ................................................................... 166 PART IV SOCIETY 9 Norms and Networks 171 Informal institutions in markets: their utility and shortcomings .......... .................... 172 Building and adapting formal institutions ............................................ 176 Integrating informal and formal institutions ............................................ 178 Conclusions ............................................ 179 10 The Media 181 Independence ............................................ 183 Quality ............................................. 188 Broadening the media's reach ............................................ 190 Institutions to complement the media ............................................. 192 Conclusions ............................................ 192 Bibliographic Note 195 Selected World Development Indicators 229 Boxes 1.1 A poem on the problems of trade .................................................... 5 1.2 What are institutions? ............................................................. 6 1.3 Institutional evolution and economic development: private traders and public rulers .............. 7 1.4 Courts and the expansion of trade ................................................... 9 1.5 Weak institutions hurt poor people .................................................. 9 CONTENTS IX 1.6 Who builds institutions? .11 1.7 Human capital and institutional design .13 1.8 Computerization and land registration in Andhra Pradesh, India .14 1.9 Private innovation supported by formal institutional change .15 1.10 Distributional effects of innovation depend on who innovates: bankruptcy law in two countries .16 1.11 Experimentation and adaptation: bankruptcy institutions in Hungary .17 1.12 Trade and institutional change in Thailand .17 1.13 Institutional evolution of rice markets and standardization in Japan, 1600-1920s .18 1.14 Role of the news media in fighting corruption in Kenya .19 1.15 Competition and the evolution of corporate law .20 1.16 The interplay of social, political, and economic forces in the reform of land institutions in China .. 22 1.17 Applying the lessons to the social sectors .............................................. 24 1.18 Crises and institutional change in Malaysia .25 2.1 Agricultural marketing institutions .33 2.2 Early institutions of land ownership in Mesopotamia and Egypt .34 2.3 Quilombos in Brazil: infrastructure, social change, and a new demand for land registration .34 2.4 Examples of policy barriers to the operation of land markets .35 2.5 Informal collateral transactions using land titles in Thailand .37 2.6 Do indigenous land rights constrain agricultural investment and productivity in Africa? .37 2.7 A transparent and accessible institutional framework for granting land rights in Mexico .38 2.8 The intersection of formal and informal lending: marketing agents in the Philippines .41 2.9 Livestock as savings: contrasting evidence from India and Burkina Faso .43 2.10 Creating an information-sharing network for the poor: SRISTI in India .47 2.11 Private sector extension services in Argentina .47 2.12 Increasing information flows between farmers and researchers in Ghana ....................... 51 2.13 International spillovers and the CGIAR .51 3.1 Business groups and restrictions on competition in Kazakhstan .62 3.2 Business associations and trade credit .63 3.3 The need for formal laws: the development of corporate law .65 3.4 Measuring the strength of legal protections for shareholders .65 3.5 Legal and regulatory change in Brazil .66 3.6 Limitations to private governance in accounting .71 3.7 Evolution of international accounting standards .72 4.1 The financial revolution versus the industrial revolution .76 4.2 Financial structure varies across countries: better information and legal systems that protect property rights play a role .78 4.3 Promoting stock markets in developing countries .78 4.4 Designing a bank safety net: the role of deposit insurance .81 4.5 The role of subordinated debt in establishing credibility: the case of Argentina .82 4.6 Institutional design for bank supervisors .83 4.7 The political economy of banking reform .87 4.8 Strengthening bank exit mechanisms: lessons from Latin America .89 4.9 The effects of foreign ownership of banks in Hungary .91 4.10 Technology and provision of financial services .92 4.11 Credit registries .95 5.1 Political institutions, property rights, and fiscal outcomes in 17th century England .100 5.2 Packaging trade reforms ........................................................... 105 X WORLD DEVELOPMENT REPORT 2002 5.3 Political connections and firm value in Indonesia ........................................ 106 5.4 Discretion and truck inspection in Gujarat, India ........................................ 107 5.5 Constitutional reform for anticorruption in Thailand ..................................... 108 5.6 Business taxation in Uganda . ...................................................... 111 5.7 Market discipline versus state discipline: municipal bankruptcy in Hungary .................... 113 5.8 Tax sharing with weak tax administration: the case of Russia ............................... 114 6.1 How mediation resolves disputes . ................................................... 118 6.2 Who benefits from better courts? ..................................................... 119 6.3 Surveys on judicial performance ..................................................... 120 6.4 Comparing judicial efficiency ....................................................... 121 6.5 Index of the complexity of litigation . ................................................ 122 6.6 Debt recovery in Tunisia .......................................................... 123 6.7 The creation of a specialized commercial court in Tanzania ................................ 126 6.8 Alternative dispute resolution in Bangladesh ........................................... 127 7.1 Adam Smith on competition, 1776 . .................................................. 134 7.2 Measuring competition ........................................................... 134 7.3 Labor regulations and rigidities in the labor market: the example of India ...................... 136 7.4 Increased product market competition and increased labor market flexibility in India ..... ........ 137 7.5 Differences between the United States and the EU on competition law and its enforcement .... .... 141 7.6 Open trade and institutional change: product markets in India .............................. 143 7.7 Benefits of liberalization of industrial country markets for agriculture and textiles ................ 144 7.8 Lack of competition in services restricts gains from merchandise trade liberalization ..... ......... 145 7.9 Weak IPR systems promoted access to technology and growth in East Asia ..................... 147 8.1 Private provision: recent evidence from concession arrangements ............................ 153 8.2 Vertical integration and discrimination in the provision of Internet services .................... 157 8.3 Water concession in Tucuman, Argentina .............................................. 163 8.4 Targeting subsidies: Chile's approach . ................................................ 165 9.1 Exclusion in trading in African history ................. ............................... 175 9.2 Tiripur in Tamil Nadu (India): insiders and outsiders in the use of informal institutions ..... ...... 175 9.3 The influence of formal institutions on norms: colonial Uganda ............................. 177 9.4 Islamic banking: informal and formal approaches ........................................ 178 9.5 Education among the Orma in Kenya: adapting well-established norms ....................... 178 10.1 The media's role in reducing corruption in Peru .182 10.2 Improving education through the media in Panama .182 10.3 Measuring media independence through data on media ownership .183 10.4 Media ownership influences content: Ukraine .185 10.5 Controlling news on the Internet .187 10.6 Improving access to information in Thailand .190 10.7 Increasing access to the media: wall newspapers in Nepal .................................. 192 Figures 1.1 The concentration of ownership varies tremendously across countries ......................... 8 1.2 Financial depth generates growth .................................................... 10 1.3 Complexity of procedures in debt collection ............................................ 12 1.4 Cost of business registration across countries ........................................... 13 CONTENTS Xi 1.5 Greater openness and quality of institutions .18 1.6 Diversity of information providers and quality of institutions .19 2.1 Poverty headcounts, urban versus rural, selected countries, 1990s .32 2.2 Evolution of formal institutional borrowing of farm households, selected Asian countries .39 2.3 Financial performance of generalized agricultural insurance programs .44 2.4 Median rates of return on agriculture research and extension by region .45 2.5 Agricultural research intensity, public and private, 1993 .48 3.1 Flows of new investment are insensitive to value added in developing countries .56 3.2 Proportion of assets in publicly traded firms accounted for by group-affiliated firms .61 3.3 Shareholder rights and stock market development .66 3.4 Accounting standards across countries .70 3.5 Capital raised through new depository receipt programs .73 4.1 Financial system development across income groups .77 4.2 State ownership in banking, 1998-99 .85 4.3 Evolution of the Hungarian banking sector .86 4.4 Increase in the market share of majority foreign-owned banks, selected countries, 1994 and 1999... 89 5.1 Variation in the quality of policies around the world ..................................... 101 5.2 Divided governments have difficulty with fiscal adjustments ................................ 103 5.3 Governments with fewer checks and balances than others are less likely to enforce banking regulations during crises ................................................................... 104 5.4 Tax collection around the world ..................................................... 110 6.1 (a) Procedural complexity reduces efficiency ............................................ 122 (b) Rich countries also have complex regulations, but ..................................... 122 (c) They have more efficient systems because of complementary institutions and capacity .... ...... 122 6.2 Excessive written procedures limit access to justice ....................................... 124 6.3 The independence of the judiciary enhances property rights ................................ 129 7.1 The size of the unofficial economy rises with the number of procedures required to start up new business .................................................... 138 7.2 Corruption rises with the number of procedures required to start a new business ..... ........... 139 7.3 Effectiveness of competition law increases with per capita income ............................ 141 7.4 Openness reduces price-cost margins ................................................. 143 7.5 Patent strength rises with per capita income ............................................ 147 10.1 Who owns the media? ..................................................... 184 10.2 State media ownership and low competition are associated with poor outcomes ..... ............ 185 10.3 Media penetration rates by region and by OECD compared with non-OECD countries ........... 191 Tables 2.1 Types of rural lending institutions ............. ...................................... 40 2.2 Where will the private sector invest in agricultural technologies? ............................. 46 3.1 Share of small formal sector firms in selected economies, selected years ........................ 57 3.2 Financial statements do not disclose useful information for resource providers .................. 70 6.1 Inputs into the judicial system for selected countries, 1995 ................................. 120 XII WORLD DEVELOPMENT REPORT 2002 7.1 Benchmarks of product market dominance in competition laws around the world ..... .......... 140 8.1 Investment in infrastructure projects with private participation in developing countries by sector and region, 1990-99 ...................................................... 152 8.2 Strategy for vertical separation or integration ........................................... 156 8.3 Access to electricity, water, sanitation, and telephone services in 22 Latin American countries, 1986-96 ................................................................... 162 9.1 Types of informal sanctions in contract-enforcement mechanisms ............................ 173 Definitions and data notes The countries included in regional and income The following abbreviations are used: groupings in this report are listed in the Classification of Economies table at the end of the Selected World EU European Union Development Indicators. Income classifications are FDI Foreign direct investment based on GNP per capita; thresholds for income clas- GATT General Agreement on Tariffs and Trade sifications in this edition may be found in the Intro- GDP Gross domestic product duction to Selected World Development Indicators. GNP Gross national product Group averages reported in the figures and tables are HIWVAIDS Human immunodeficiency virus/ unweighted averages of the countries in the group un- acquired immune deficiency syndrome less noted to the contrary. IPR Intellectual property rights The use of the word countries to refer to economies . N o n n implies no judgment by the World Bank about the NOE Nrgaoisanion tforoEconomiciCo legal or other status of a territory. The term developing OECD Organisation for Economic Co- countries includes low- and middle-income economies operation and Development and thus may include economies in trasition from cen- PPP Purchasing power parity tral planning, as a matter of convenience. The term de- R&D Research and development velopedor industrial countries may be used as matter of SMEs Small and medium-size enterprises convenience to denote the high-income economies. TRIPS Trade-Related Aspects of Intellectual Dollar figures are current U.S. dollars, unless other- Property Rights wise specified. Billion means 1,000 million; trillion VTO World Trade Organization means 1,000 billion. PART I Intro uction CHAPTER 1 15Ua ilng nstitutLions: f . 1etet Innovate, k_..otnlect, anl (0otnpete How do we accountfor the persistence ofpoverty in the ties. And the rules of the organization, although not midst ofplenty? If we knew the sources ofplenty, why written, were self-enforcing. Remaining in the coali- don't poor countries simply adopt policies that make tion of traders best served each member's interests. So- forplenty?... We must create incentivesforpeople to cial ties cemented mutually beneficial business rela- invest in more efficient technology, increase their skills, tionships, and cross-border trade flourished. and organize efficient markets. Such incentives are Today, a millennium later, people everywhere face embodied in institutions, similar problems in striving to improve their well- -Douglass C. North, 2000 being through market activity. African entrepreneurs lack information about potential business partners. Poor farmers in Latin America lacking formal title to - n the 11th century the Maghribi traders of North their land cannot use it as collateral to secure access to Africa wanted to expand business across borders, credit. Budding entrepreneurs in Central Asia, trying all around the Mediterranean. Trade in each center to start new businesses, run into political obstacles was free of formal regulations and restrictions, and from established firms and the state. competitive, with many buyers and sellers negotiating Despite the problems, many people in rich countries prices through brokers, open-bid auctions, and direct and poor are engaged in productive-and rewarding- dealings. Cross-border trade also was generally free of market activity. As World Development Report 2000/ formal regulations and restrictions. But it was fraught 2001 argued, income from participating in the market with uncertainty about selling prices, the quality on is the key to boosting economic growth for nations and arrival, and the possibility of theft. Only if merchants to reducing poverty for individuals. This Report is traveled with their goods to distant markets could they about enhancing opportunities for poor people in mar- ensure the safe arrival and sale of their merchandise. kets, and about empowering them. What makes mar- Such risks and costs naturally limited trade. ket activity rewarding and possible for some, and not So in all major trading centers around the Mediter- others? Why are some market systems inclusive and in- ranean, the Maghribis set up overseas agents to repre- tegrated, allowing benefits to flow to the poor as well sent their interests and exchange information about as the rich, the rural people as well as the urban? And markets. Being from the same community, these agents why are other markets localized and segmented? were seen as trustworthy. And with fewer contractual The Maghribi example illustrates some of the rea- problems, Maghribi merchants no longer needed to sons. Markets allow people to use their skills and re- travel to ensure that they would not be cheated. Infor- sources and to engage in higher-productivity activities mation flowed freely in this network bound by social if there are institutions to support those markets. What 3 4 WORLD DEVELOPMENT REPORT 2002 are these institutions? Rules, enforcement mechanisms, particular countries. Institution building is generally a and organizations supporting market transactions. Ex- cumulative process, with several changes in different tremely diverse across rich and poor communities and areas building up to complement and support each nations, they help transmit information, enforce prop- other. This Report identifies elements of such a strat- erty rights and contracts, and manage competition in egy. Even small changes can build momentum for fu- markets. All market-supporting institutions do one or ture changes. The whole is greater than the parts, and more of these things. And in so doing, they give peo- even moderate progress in the parts can contribute to a ple opportunity and incentives to engage in fruitful better system to promote growth and reduce poverty. market activity. Four main lessons emerge for institution building. This Report is about people building institutions that The first two are about supplying effective market- support the development of markets. The 2000/2001 supporting institutions. But supplying institutions is Report underscores the importance of institutions in af- not enough. People must want to use them too. Thus, fecting poor people's participation in markets. This Re- the second two lessons are also about creating the de- port discusses both institutions that support growth and mand for such institutions, and about the forces for those that directly affect access of people left out of change within countries. many market activities. It considers those institutions To ensure effective institutions: that provide opportunities for people and that empower them. It goes beyond the 2000/2001 Report by analyz- * Design them to complement what exists-in terms of ing what institutions do to promote growth and facili- other supporting institutions, human capabilities, and tate access and by suggesting how to build effective in- available technologies. The reason? The availability and stitutions. And it emphasizes how institutions can help cost of supporting institutions, existing levels of cor- people make better use of the assets they own and how ruption, human capacity and technology determine to accumulate more. In focusing on institution build- the impact of a particular institution. That is why in- ing, it does not devalue the importance of policy. But stitutions that achieve their goals in industrial coun- good policies are not enough. The details of institution tries may not do so in developing ones. Much of the building matter for growth and poverty reduction. important work in building institutions lies in modi- The Report contributes to existing work on institu- fying those that already exist to complement better tions and markets in several novel ways. It provides a other institutions and in recognizing what not to build diagnostic framework for understanding how institu- in a particular context, as much as what to build. "Best tions support market activity. Bridging the gap between practice" in institutional design is a flawed concept. theory and evidence across disciplines, it also builds on * Innovate to design institutions that work-and drop existing evidence on the role of institutions and insti- those initiatives that do not. Even in countries with tutional change. It extends previous empirical work on similar incomes and capacities, innovation can create institutional change to developing countries and pre- stronger institutions because of differences in local sents a framework for institutional change. It confirms conditions-differences ranging from social norms that one size does not fit all in considering institutional to geography. Experimentation, which has some costs design. But it does more than that. It illustrates how to that must be recognized, can nevertheless help iden- proceed in building more effective institutions. It pro- tify new and more effective structures. Countries can vides policy guidance by taking a pragmatic approach. gain from expanding successful public innovations The aim is not to define what should be done in an and adopting private innovations. But they must also ideal world, but what can be done in today's world. have the courage to drop failing experiments. In understanding what drives institutional change, * Connect communities of market players through open the Report emphasizes the importance of history. Many information flows and open trade. Exchanging goods developing countries have been nation-states for a short and services outside existing networks and commu- time compared with industrial countries. The evolution nities creates demand for market-supporting institu- of nations teaches that building institutions takes time tions. Exchanging information through open debate and that the process within each country may stall or creates demand for institutional change by holding reverse because of political conflicts or economic and people to account, by changing behavior, and by sup- social conditions. It offers lessons about the process of plying ideas for change from outside the community. change and the importance of norms and culture in Linking communities of people in networks of infor- BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE 5 mation and trade is thus a priority for policymakers Box 1.1 building market-supporting institutions. A poem on the problems of trade * Promote competition among jurisdictions, firms, and individuals. Greater competition modifies the effec- If I knew you and you knew me tiveness of existing institutions, changes people's in- 'Tis seldom we would disagree; But never having yet clasped hands centives and behavior, and creates demand for new Both often fail to understand institutions. Developing country actors may face too That each intends to do what's right little competition, often because of current institu- And treat each other "honor bright" tional structures. Changing this will improve the qual- How little to complain there'd be ity of other institutions. Competition among jurisdic- If I knew you and you knew me. tions-for example, among different states within a When'er we ship you by mistake, country or between countries-highlights successful Or in your bill some error make institutions and promotes demand for them. Compe- From irritation you'd be free If I knew you and you knew me. tition among firms and individuals does the same. Or when the checks don't come on time And customers send nary a line, This chapter first provides a framework for evaluat- We'd wait without anxiety, ing the role of institutions in supporting market trans- actions, growth, and poverty reduction. It then focuses If knew you and you knew me. on the four main lessons on institution building, fol- Source: Who's Who in the Grain Trade 35 (June 20, lowed by a discussion of the impact of political and so- 1922-23); cited in Bernstein 2001, World Development cial forces on institutional evolution. Report 2002 background paper. How do institutions support markets? Small vendors engage in simple spot-market transac- Yet not all institutions promote inclusive markets. tions, with buyers and sellers dealing face to face in The Maghribis lowered transaction costs among them- fairly standard products whose quality is easy to verify. selves, but in so doing excluded other communities. In- A rural vegetable market in a poor country is such a stitutional designs that evolve through either historical market. Large multinational firms exchange more dif- circumstances or directed action by policymakers are ferentiated products, facing greater difficulties in veri- not necessarily the best institutions for all society-or fying quality and bigger separations in time and space for economic growth and poverty reduction. Moreover, between the quidand the quo. International exchange institutions that once supported market transactions of food products is an example of such a market. Most can outlive their usefulness-for example, privatization economies have both types of markets-the first more agencies and bank restructuring agencies. The challenge common in developing countries, the second in indus- for policymakers is to shape policies and institutional trial economies. development in ways that enhance economic develop- Developed markets, more global, inclusive, and inte- ment. The Maghribis operated under a policy of free grated, offer more opportunity choice. Underdeveloped trade that enhanced their opportunities. It was to take markets, more likely in poor countries, are more likely advantage of these opportunities that they developed to be local and segmented. So, compared with farmers their institutions. in Canada, poor farmers in Bangladesh have fewer op- Clearly there is no unique institutional structure portunities-and far fewer formal institutions (such as guaranteed to lead to economic growth and poverty banks and formal courts) to reduce their risks and in- reduction. Large firms in the United States and the crease their opportunities. United Kingdom are often publicly held, with dis- What limits market opportunities? Transaction costs persed ownership, and are widely traded. But that is not from inadequate information, incomplete definition the case in other high-income countries such as France and enforcement of property rights, and barriers to or Canada, where ownership structures are highly con- entry for new participants.' What increases them? In- centrated (figure 1.1). And to promote competition, stitutions that help manage risks from market ex- policymakers can use quite different guidelines. In East change, increase efficiency, and raise returns (boxes 1. 1, Asia competition authorities consider a market share of 1.2, and 1.3). 50 to 75 percent to be evidence of possible monopoly 6 WORLD DEVELOPMENT REPORT 2002 Box 1.2 What are institutions? Institutions are rules, enforcement mechanisms, and orga- and public and private organizations operating under public nizations. This Report considers those institutions that sup- law. For example, organizations include firms operating port market transactions.2 Distinct from policies, which are under corporate law. Informal institutions, often operating the goals and desired outcomes, institutions are the rules, outside the formal legal system, reflect unwritten codes of including behavioral norms, by which agents interact-and social conduct. Examples include land inheritance norms the organizations that implement rules and codes of conduct and moneylenders using social networks to determine cred- to achieve desired outcomes. Policies affect which institu- itworthiness based on the reputation of the agents involved. tions evolve-but institutions too affect which policies are People in both rich and poor countries rely on informal adopted. Institutional structure affects behavior. But behav- institutions to facilitate transactions, but these institutions ior may also change within existing institutional structures. are relatively more important in poor countries where for- Institution builders can be diverse-such as policymak- mal institutions are less developed. Moreover, poor people ers, businesspeople, or community members. Corporate, in poor countries are often ill served by the limited formal collateral, and bankruptcy laws are public institutions, as are institutions available. In poor countries, and poor regions in the judiciary, tax collection agencies, and regulatory agen- particular, informal institutions substitute for formal institu- cies. Banks, reciprocity between community members, and tions (box figure). Countries and communities can go a long land inheritance norms are private institutions. Many private way toward resolving information and enforcement prob- institutions exist under the aegis of public institutions. Pri- lems without using their formal public legal systems. vate banks, for example, operate within the framework of public law. Social norms exist within (or without) formal laws. The enforcement of rules can be internal, implemented Informal rural credit in selected developing by the parties affected by the rules, or external, imple- countries, 1980s and 1990s mented by a third party. Informal institutions and private for- Percentage of total credit mal mechanisms generally rely on their own members for 100 enforcement. Individual agents organize themselves into in- formal groups, such as business associations (chapter 3) or mutual insurance systems (chapter 9) when the cost of col- 80 - lective action is low and the rules can be easily monitored. In these groups, expulsion from the community is a form of punishment. 60 * External enforcement mechanisms, such as judicial sys- tems or third-party arbitration, are critical mechanisms for the development of integrated markets. They allow access 40 - to market opportunities for a broader group of market par- ticipants. For external enforcement mechanisms to be ef- 20 fective, the legitimacy of the enforcer is vital. When the 20 state acts as an agent that shares the objectives and beliefs of its citizens-and implements rules consistent with 0 them-it is more likely to build effective formal institutions Nigeria India Nepal Sri Lanka Pakistan Thailand to support market development.3 11980s) (1981) (1995-96) (1981) (1991) (1980sa Effective institutions are those that are incentive-compati- Source: Kochar 1997; Besley and others 2001; Ijere 1986, cited in ble. Institutions with internal enforcement mechanisms are ef- Adegbite 1997; Mansuri 1998; Desai and Mellor 1993. fective because there is a mutually recognized system of re- wards and penalties. An important issue in the design of public institutions is ensuring that the incentives that are created ac- Networks, such as those of the Maghribis, that are tually lead to desired behavior. Take the example of deposit in- based on common ethnic, religious, and other common ties, surance, which is designed to protect depositors from the risks are closed groups; that is, entry into the group is limited. In inherent in financial institutions (chapter 3). Experience has such groups, costs of information processing and definition shown that deposit insurance can weaken the incentives of fi- and enforcement of property rights are lowered by mutual nancial managers to lend depositors' funds prudently and can ties or trust. Although these transaction costs are lower in lead to excessive risk-taking. In circumstances like this, com- closed groups, the informal and norm-based institutions that plementary regulations are required to realign incentives, such such groups rely on tend to support a less diverse set of ac- as regulations to ensure that bank managers have a significant tivities than do formal legal institutions. As countries de- financial stake in bank performance. velop, the number and range of partners that market partic- ipants deal with increases and market transactions become Informal andformal institutions more complicated, demanding more formal institutions. Formal institutions include rules written into the law by gov- Conversely, public or private agents may build formal insti- ernment, rules codified and adopted by private institutions, tutions to enable a more diverse set of activities. BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE 7 Box 1.2 (coritinueci] Legislators may purposefully base formal law and judi- state power, overtaxation, corruption, short time horizons, cial practice on social norms. In some cases this may con- cronyism, and the inability to uphold public order. For exam- sist of simply codifying and modifying existing practices and ple, governments may establish restrictive trading rules in re- writing them into law (Bernstein 1999). But this is not sim- sponse to lobbying by business monopolies intent on safe- ple, particularly in heterogeneous societies. Choosing how guarding their monopoly interests. The balance between to weigh each group's norms and standards is critical in de- markets and state power, and between business and social termining not just the efficiency impacts, but also legitimacy interests, is a delicate one in the course of institutional devel- and distributional implications.4 For example, in multiethnic opment. Historically, the government's role in the protection Uganda, English was adopted as a neutral common lan- of property rights and the provision of other public goods has guage for the formal functions of the state. Such concerns been closely linked to its role in ensuring peace or law and extend to standards or rules in international markets as well. order. Conflicts over property between private agents, and Ideally, informal and formal institutions should com- between the state and private agents, are some of the most plement each other. Together, they can reduce transaction important issues that governments have had to deal with, be- costs more than either can alone. Formal courts, for example, cause they often lead to a breakdown of law and order. deter litigation and facilitate informal settlement simply by Market development and private business flourish when providing the threat of enforcement (chapter 6). Far more dis- the behavior of those who govern is not arbitrary (see box putes arise in business transactions than go through a formal 1.3 ). For example, detailed analyses of the evolution of cor- dispute resolution process (Bernstein 1999). porate law in several countries show that in the early stages of development, private business was typically subject to Putblic versus pritwte roles the arbitrary whims of those in power. The state, with pri- Governments have an important role in providing public mary control rights, granted the permission to incorporate goods, such as laws that delineate property rights and the ju- case by case (Pistor and others 2000). At later stages, the dicial institutions that enforce these rights and establish the right to incorporate was no longer a personal favor but was rule of law. But governments have been known to impede granted to any entrepreneur that met a set of predeter- the development of markets through arbitrary exercise of mined conditions. Box 1I3 Institutional evolution and economic development: private traders and public rulers In medieval Europe, the political power of local rulers was ex- among traders, and increased problems of information and en- tensive. Local rulers could confiscate the property of individual forcement. Growth meant trading with members from other traders from other regions without incurring penalties. In re- social and ethnic backgrounds, which meant that social con- sponse, private mercantile guilds evolved to promote trade and nections could not easily be used as a basis for information or to guard against the arbitrary action of local rulers. These guilds enforcement. established agreements with merchants in foreign cities and Members no longer wanted to be collectively responsible with local authorities themselves. Arbitrary confiscation was for individual breaches of contract. So leaders pushed for an punished by the withdrawal of large amounts of business by enforcement and sanctioning system based on individual re- the guild, and so local rulers were forced to respect the rights sponsibility rather than community responsibility. To the ex- of its members. This change in the balance of power helped tent that community growth implied more intracommunity so- to promote the security of foreign traders. cial and economic diversity, it also reduced the political viability In the 12th century, traders in Europe established commu- of the community. But the extent to which communities could nity-based mechanisms to facilitate the exchange of credit and abolish community-based mechanisms depended on a reliable trade across borders. These mechanisms were based on the third party to enforce contracts. In England, the monarch per- community accepting responsibility for the performance of its formed this role, and in 1275 King Edward I issued a statute members vis-a-vis other communities. For example, when a outlawing community responsibility for debts. Genoese merchant defaulted on a loan from a merchant in The example illustrates a general principle: as economies London, community leaders in Genoa were responsible for en- grow and develop, different types of institutions are needed forcing the contract by imposing sanctions on the defaulter. to facilitate transactions. Many different actors can push for Community origin was easily established, meaning that repu- new institutions. But the role that the state plays depends tation within the community was important, and agents could on its capacity and political viability: a strong state that re- be trusted not to renege on their contracts. spects the law itself and refrains from arbitrary action is a criti- As cities grew in size and number, so did the communities cal factor. of merchants and traders, making collective action more diffi- cult. Unrestricted entry into trading led to more competition Source: Greif 1997a. 8 WORLD DEVELOPMENT REPORT 2002 Figure 1.1 rights to assets and income and being able to protect The concentration of ownership varies those rights are critical for market development. tremendously across countries These include the rights of the private sector in rela- tion to the state. Institutions can reduce the poten- Ownership concentration tial for disputes and help enforce contracts. Examples 0.58 - , include a country's constitution, its judicial system, 0.58 - * ++ + , and the full array of social networks. * * + + + * They increase competition in markets-or decrease it. 0.48 - Competition gives people incentives to do better and * ' + promotes equal opportunity. In competitive markets 0.38 -0 resources are more likely to follow the merits of a project than the social or political connections of an 0.28 - entrepreneur. The degree of competition also affects *.18 ,1 I I t I# innovation and economic growth (chapters 2 and 7). 0.18 u l , , + I I a But while some institutions facilitate competition, 5.5 6.5 7.5 8.5 9.5 10.5 Income others impede It. For example, by overregulatmg the (logarithm of per capita income} entry of new business, governments can constrain Note: Ownership concentration is measured by the combined stakes competition. And by organizing market activities of the three largest shareholders in the 10 largest privately controlled around a closed group of participants-recall the firms. Source: La Porta, Lopez-de-Silanes, and Shleifer 1999. Maghribis-outsiders will find it harder to compete _____________ _ even while opportunities for those in the group may increase (chapters 3 and 9). power, whereas in Africa the range is 20 to 45 percent. Within South Asia some farmers rely on cooperatives The transaction costs of acquiring information, en- to market their goods; others use informal contracts forcing property rights, and restraining competition with private traders. can prevent the emergence of inclusive markets. But ef- This Report provides a framework that applies across fective institutions can reduce those costs. Consider the the range of market-supporting institutions. It cuts following example. If the quality and value of the grain through the complexity and diversity of institutional that traders buy from a farmer cannot be easily deter- structures by focusing on what institutions do. Under- mined, and if traders have little information about a standing what they do is the first step in building effec- farmer, they have to inspect each bag of grain to assess tive institutions. Institutions do three main things:5 quality. Traders also provide credit to farmers. But if traders have little information on the ability of farmers * They channel information about market conditions, to repay the debt-or if farmers cannot use the assets goods, andparticipants. Good information flows help they own as security-providing credit is risky. These businesses identify partners and high-return activi- problems are magnified for smaller and poorer farmers. ties-and assess their creditworthiness. Information The trader may impose higher interest rates on poorer about businesses helps governments regulate effec- farmers, and the farmers may be more likely to default tively. Institutions can affect the production, collec- than if they were exposed to competition.7 tion, analysis, verification, and dissemination-or the Through these three functions, all institutional struc- withholding-of information and knowledge. They tures affect the distribution of assets, incomes, and costs do this for participants in, and between, communi- as well as the incentives of market participants and the ties and markets. Examples include accounting firms efficiency of market transactions. By distributing rights and credit registries, which facilitate information pro- to the most efficient agent, institutions can enhance pro- cessing, or government regulations on the media, ductivity and growth. By affecting the incentives to which restrict the dissemination of information.6 invest-for example, through strengthening property * They define and enforce property rights and contracts, rights-they can affect investment levels and adoption determining who gets what and when. Knowing one's of new technology. By delineating market rights, such BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE 9 Box 1.4 Box 1.5 Courts and the expansion of trade Weak institutions hurt poor people Studies of manufacturing firms in eight African countries Mounting evidence shows that the poor bear the greatest demonstrate the supporting role institutions play in mar- burden of institutional failure. Take corruption, a highly re- ket development. These country studies show that the ab- gressive tax. Demands for bribes and unofficial fees for sence of effective public dispute resolution mechanisms services hit poor people hardest. In far too many cases in cases of breach of contract has limited the expansion legal systems and the judiciary fail to serve poor people. of trade and market development. Courts tend to be slow Their illiteracy and inability to pay for legal representation and inefficient. The absence of formal contract enforce- put formal legal institutions beyond reach. The failure of ment mechanisms has limited the growth of firms and the the state to protect property also hurts the poorest dispro- development of financial institutions. The small scale of portionately, because they cannot afford to protect them- the formal productive sector has, in turn, prevented the selves from crime. And badly designed regulatory institu- development of complementary institutions. tions reduce the provision of infrastructure to the poorest Another study analyzing six countries in Africa (Bu- in society. rundi, Cameroon, C6te d'lvoire, Kenya, Zambia, and Zim- World Development Report 2000/2001 stressed that babwe) shows that among these countries, the presence poor people are often more vulnerable than others to of a more developed legal system encouraged firms to macroeconomic crises and natural disasters. Market insti- undertake riskier activities because well-functioning legal tutions that support growth of overall incomes can reduce systems helped to adjudicate and settle disputes that their vulnerabilities to shocks and help them insure against arose from such market activities, bad times. Some of the institutions discussed in this Report have an important and direct role in this. For example, finan- Source: Bigsten and others 2000; Collier and Gunning cial institutions help mitigate their risks, allowing individuals Source: Bigsten and others 2000; Collier and Gunning to diversify their savings and risks and allowing them to smooth their consumption over good times and bad. Source: World Bank 2000d. as through competition law, they limit producer rents and protect consumers from high prices. And by clarify- overlapping sets of institutions. For example, the success ing rights for the disadvantaged in markets, institutions of the state in providing laws and the performance of the can directly affect the lives of poor people. For example, judiciary and police reflect whether citizens and in- giving formal titles to poor people whose occupancy vestors perceive the state as respecting property rights. rights were not recognized by lenders allows them to Access to financial services and the sophistication of fi- borrow and invest. nancial markets reflect how successfully institutions pro- tect the property rights of borrowers and lenders. High How do Institutions support growth Hnw dovinrtitrductions supportgrowtlevels of public corruption reflect how the behavior of public agents in state institutions responds to the types Institutions that support market transactions can thus of incentives that exist for politicians and civil servants affect poor farmers in Latin America as much as they af- to pursue the public good over their self-interest. fect wealthy businessmen in Canada. Country case stud- Positive relationships between economic develop- ies, as well as cross-country empirical work, provide im- ment and these indicators of institutional success have portant insights into institutional development and been widely documented. But most studies do not es- market development (box 1.4). They confirm how mar- tablish links between specific institutions and specific ket-supporting institutions affect people's lives by influ- outcomes. Instead, they highlight the wide variety of encing growth, determining people's access to markets, institutions that support markets. For example, income and enabling poor and rich people to make the best use and the rule of law-encompassing the collective im- of their assets. Moreover, weak market-supporting insti- portance of property rights, respect for legal institu- tutions can hurt the poor disproportionately (box 1.5). tions, and the judiciary-are highly correlated. For an- A growing body of research links institutional success other example, the development of financial institutions (and failure) to economic growth and market develop- predicts growth (figure 1.2). ment over time and across countries. A wide range of in- On institutional development and economic growth, dicators captures the performance of different, often important differences have been found between coun- 10 WORLD DEVELOPMENT REPORT 2002 Figure 1.2 institutions for conflict management.9 In divided soci- Financial depth generates growth eties, such as those with ethnic fragmentation or high inequality, low-quality institutions for managing con- Ratio of liquid liabilities to GDP, 1960 flict-including low-quality government institutions Deep (greater and inadequate social safety nets-magnify external than 0.5) shocks, triggering distributional conflicts and delaying policy responses. Prolonged uncertainty in the eco- 0.25toO.5 nomic environment and delayed policy adjustments curtail subsequent economic growth. How do you build effective institutions? 0.15 to 0.25 Recalling the framework of information, enforcement, and competition, policymakers building institutions first need to assess what is inhibiting market develop- Shallow (less ment or leading to certain market outcomes (box 1.6). than 0.15) Rather than focusing first on specific structures, they l l l l need to focus on the functions that are missing and de- 0 1 2 3 termine why. Policymakers need to ask: Average GDP growth 1960-95 Note: Figure based on partial scatter from the instrumented cross- * Who needs information on what? For example, do sectional regressions in Beck, Levine and Loayza 2000. bankers lack information on the creditworthiness of potential borrowers? * Are everybody's property rights and contracts clearly de- tries that once were colonies and are now industrialized fined and enforced? For example, do farmers have en- and former colonies that are still developing. Both groups forceable rights to land they use? trace key features of their institutions to former settlers. * Is there too little competition-or too much? For exam- A big part of the difference in later institutional develop- ple, is an infrastructure monopoly inhibiting entry ment-and its impact on growth-is the effort of set- or are firms not undertaking high-return research be- tlers in establishing well-functioning legal institutions.8 cause they lack safeguards on intellectual property? In the United States and New Zealand, colonizers settled in large numbers and transplanted institutions Once the institutional gap is identified, the next step common to, and understood by, the general populace, is to design the appropriate institution. Both supply and mostly new immigrants. In such countries the trans- demand factors are important. Moreover, as countries planted legal institutions were widely used, adapted to change and develop, so will the appropriate institution. local circumstances, and changed with economic de- To be effective, such an institution must be designed so velopment. Developing countries on every continent that the incentives of market actors are aligned to also received formal legal systems, transplanted by col- achieve the desired outcome. Four key approaches to- onizers. But their indigenous populations had little ac- ward institution building hold across all sectors and cess to or understanding of these legal systems. So the countries: complement what exists, innovate to identify institutions were not adapted to local circumstances. institutions that work, connect communities through Cross-country evidence suggests that the quality of in- information flows and trade, and promote competition. stitutions that support growth and poverty reduction through market development is lower in these coun- Completnent what exists tries than in the former group and has therefore not Developed market economies have institutional struc- supported economic growth and poverty reduction to tures that depend heavily on a capable state-a pro- the same extent. vider of public goods, a regulator, and an adjudicator. Institutions also affect how countries deal with con- But the involvement of the state in markets must be flict. A recent study found that growth and poverty out- consistent with its capacity. World Development Report comes in Asia, Latin America, and Sub-Saharan Africa 1997 emphasized matching the capability of the state since the mid-1970s have depended on the quality of with the tasks that government organizations take on. BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE II Box 1.6 Who builds institutions? Institutional reform is not just the preserve of national govern- mestic reform in Mexico. Some countries in Eastern Europe ments. Individuals and communities, local entrepreneurs, multi- are implementing wide-ranging institutional reforms as they national companies, and multilateral organizations can build strive to become members of the European Union. institutions, often in partnership with each other. National gov- Recent developments surrounding the AIDS crisis illustrate ernments may initiate reform or may simply respond to pres- how different groups may affect the process of institutional sures from the private sector or from external actors. change. In some cases of systemic institution building, govern- ments have been effective in successfully transplanting laws, Many agenits of ehange: healtlh crises andpatents organizations, and agencies. In other cases systemic reforms More than 95 percent of HIV/AIDS cases are in developing did not have the desired outcomes. The contrast between countries. But the average cost of the antiretroviral treatments, Poland and Russia is instructive in this regard. Poland had a which have reduced AIDS mortality by 70 percent in industrial more recent history of a market system, and Polish policymak- countries, is still more than $10,000 a year, far beyond the ers and business people had a better understanding of the req- reach of most people in poor countries. uisite institutional framework. Polish reforms focused on clari- Some developing countries-Brazil, India, South Africa, and fying property rights between the state and private actors-for Thailand-have taken steps to reduce the cost of AIDS treat- example, by imposing hard budget constraints on public firms. ment through the design and application of their intellectual Russia did not have a recent history of market development, property rights laws-an international institution-to allow com- and reforms did not initially have the desired effects, partly be- pulsory licenses permitting the production of generic drugs and cause there was no clear delineation between private and pub- the import of cheaper generic drugs. In Thailand generic drugs lic institutions. Firms were not immediately exposed to hard became available at just 10 percent of the price of the patented budget constraints, as shown by widespread arrears in taxes product. and other payments (Recanatini and Ryterman 2000). These measures led to threats of trade sanctions and law Institution building at the sectoral level has also met with suits from the drug manufacturers. But collective action, initi- varying success. In Tanzania and Zambia the public sector in- ated by international agencies and NGOs, helped increase tervened in agricultural marketing with the stated aim of stabi- access to AIDS drugs by enforcing existing public health safe- lizing farmer incomes. In most cases these reforms failed- guards, permitted under the Trade-Related Aspects of Intellec- leading to lower marketable output and often corruption. tual Property Rights agreement but not previously imple- Worse, the experiences affected perceptions of the overall in- mented. The news media were instrumental in publicizing the tegrity of public institutions. Successes include the reform of disparities in the availability of AIDS treatment and promoting business registration in Bulgaria, now conducted online and public debate on the issue. As a result, the U.S. government taking around two days, not three weeks as in the past. retracted its trade sanction threats. And pharmaceutical com- Local business interests, the foreign business community, panies agreed to reduce prices-and more recently to drop a nonprofit organizations, the media, and international organiza- lawsuit on intellectual property rights against the South African tions have all been involved in direct institution-building efforts government. in developing countries. For example, membership in the North American Free Trade Association has hastened the pace of do- Source: Perez-Casas and others 2000. This Report builds on that analysis by examining how a Costs, relative to per capita income, of establishing existing information, enforcement costs, and the cost and maintaining institutions of building and maintaining institutions affect the way * Administrative capacity, including human capabilities governments support private transactions in markets. a Technology. It also examines how market development is affected by the extent to which government actors themselves Both existing and newly transplanted institutions respond to the institutions they build. As countries de- can be more effective in poor countries if they are sys- velop, the types of institutions they need and demand tematically modified to take these differences into ac- also change. count."0 This may sometimes mean changing priorities One of this Report's messages is that institutions in terms of which types of institutions to build first, that work in industrial countries may not produce sim- and whether to build at all at a given time. ilar results in poorer countries because of differences in: Complementary institutions. Government interven- tions can reduce many market failures, but governments * Complementary institutions, such as those promot- may also fail in trying to support market transactions. ing transparency and the enforcement of laws For example, governments may impose regulations to * Existing levels and perceptions of corruption try to compensate for market failures or as a way of re- 12 WORLD DEVELOPMENT REPORT 2002 stricting private activity. Choosing between market fail- Figure 1.3 ures and potential government failures is not easy, but Complexity of procedures in debt collection measures can be taken to limit both. However, the lim- ited capacity of developing country governments to im- Economy plement regulations means that many activities in Philippines poorer countries are overregulated. Peru For regulatory systems in developing countries to - have a realistic chance of success, they need to be sim- Senegal pler, often less information-intensive, and less burden- Mozambique some on the courts. Many developing countries, how- Bulgaria ever, despite their weaker judicial systems, tend to have Ecuador very complex debt collection procedures (figure 1.3). Venezuela, Regulations in industrial countries can also be very R.B. de complex, but they do not impose as many additional Argentina costs as they do in poorer countries-for several reasons. El Salvador Enforcement capacity in richer countries is stronger, and Ghana judges may face other incentives that affect their perfor- Kazakhstan mance and judicial efficiency (chapter 6). Regulators are more accountable, and complementary institutions (such Sample average as those affecting judges' wages or careers, or those which Japan promote transparency) provide checks and balances to Canada protect market participants. In developing countries, Sweden where there are fewer supporting institutions (for exam- ple, where courts are weak or lack credibility), one solu- United States tion is to write simple rules and have fewer of them. Switzerland Where informal institutions operate effectively, and Germany when formal institutions require supporting institu- Australia tions, building new formal institutions may not be a Hong Kong, priority for policymakers. China Finland * Studies of land titling in various countries show that New Zealand formal titles may not have the desired effects when United Kingdom input, output, and credit markets and institutions are underdeveloped and the demand for agricultural Increasing complexity of litigation goods is low (chapter 2). In such cases traditional community-based mechanisms are more effective in Note: For the definition of complexity see chapter 6. The sample average is based on 96 countries. delineating property rights. Source: Survey done for World Development Report 2002 in * Corporate governance is difficult in poorer countries conjunction with Lex Mundi, an international association of law firms. because of weak legal systems and the lack of private information intermediaries. In this situation concen- trated ownership structures-and business groups countries, the financial cost of complying with regula- and associations-may provide more effective cor- tions for registering a business is very high relative to porate oversight than dispersed ownership structures. per capita gross national product (GNP) (figure 1.4a) and higher than industrial country averages."t Surpris- Costs, capacity, and corruption. The cost of govern- ingly, developing countries that have less administra- ment regulation, whether in financial or other terms, tive capacity also require more procedures to register a needs to be consistent with a country's per capita in- business (figure 1.4b). The high cost, whether in com- come to be effective. For example, a recent study cov- plexity or resources, deters entry into the formal sector, ering 85 countries found that in many developing potentially reducing competition and incurring ad- BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE 13 Figure 1.4a Box 1.7 Cost of business registration (as percentage of Human capital and institutional design GNP per capita) is higher for lower-income countries Human capital affects the quality of the rules that govern market transactions and the enforcement of these rules. Cost of registration (NO Literacy levels and technical skills vary greatly across and 140 - within countries. The poorest economies of the former So- 120 - viet Union have income levels lower than many countries in Asia and Africa but nearly universal primary education. 100 _ So literacy is less of a barrier for Armenians using formal in- 80 - stitutions than it may be for some Angolans-and it is less 60 - of a problem for today's Malaysians than it was for those of a generation ago. The rules and organizations that gov- 40 - ern markets have to allow relevant market actors to use 20 - them easily. This argument holds within countries as well- for example, across poorer rural and richer urban areas. ° ' S 2nd 3rd 4th l The usefulness of institutions also depends on the ca- pability of their administrators. Judges untrained in corpo- Income quartiles rate law and accountancy, for instance, may not be the best arbiters of bankruptcy cases. Successful institution Figure 1.4b builders have had either to tailor institutions to prevailing Lower-income countries have more procedures administrative capacity (using, for example, simpler bank- ruptcy rules) or to complement institution building with a Average per capita GDP (thousands of U.S. dollars) strong focus on concurrently developing technical exper- 25 - tise for administrators (from accountancy skills to regula- tory economics). 20 - 15 - ditional costs in the form of increasing corruption 10 - (figure 1.4c).2 A World Bank study also finds that in 5 - many African countries, restrictive regulations and _I I_ - practices are often aimed at generating rents for offi- o 0 _J I _ S _ | - cials and favored private agents or groups, constraining 1-5 6-10 11 15 16 and more business activity in both agriculture and industry."3 Number of procedures Since building institutions is costly, requiring a min- Figure 1.4c imum threshold demand before they can operate effi- More procedures are associated with higher ciently, small countries can face problems. Small coun- corruption tries and those countries wishing to expedite access to Index of corruption (higher = worse} institutions may wish to rely on foreign institutions- 90 - such as foreign banks or foreign stock market listings- 80 - rather than build supervisory and regulatory capacity 60 - at home (chapter 4). Hungary and Estonia, for exam- 50 - ple, encouraged the entry of foreign banks, supervised 40 - and regulated in their country of domicile. 30 - Human capability. More human capital may be 20 - needed to use some market institutions-such as for- o u- * | | | * S *mal judicial methods to resolve disputes-and to ad- 1-5 6-10 11-15 16 and more minister regulations or develop standards (box 1.7). For Number of procedures example, competition authorities need people who un- derstand the complex details of competition cases. As Note: Costs are defined as official fees as a percentage of 1999 GNP countries build human capabilities, they need to con- per capita. sider where to focus their attention. Human capital and Source: Dlankov and others 2001, World Development Report 2002 background paper. the array of market institutions In an economy have a dynamic relationship. Agents need human capital to 14 WORLD DEVELOPMENT REPORT 2002 benefit from certain institutions. And over time, as Box 1.8 agents learn, institutions need to be adapted. As can be Computerization and land registration in seen from the experience of East Asia, actively promot- Andhra Pradesh, India ing literacy and primary education can have a big pay- off in the eventual quality and success of formal insti- Buying property in Andhra Pradesh used to be complex and take a long time. After the purchase the buyer visited tutions, as both users and administrators are more able the local office of the Sub-Registrar of Assurances in per- to work with market institutions. son, had the property valued and stamp duty calculated, Technology. Infrastructure regulation shows that purchased stamp paper, and had a writer draft the deed in technical standards used in industrial countries may be the requisite legal language. The purchaser also had to pro- vide additional documents related to income and other inappropriate for developing countries (chapter 8). In properties owned. All these documents were then scruti- poor countries service providers using low-cost tech- nized by the registrar, and recorded, before an exact copy nology often operate in the informal sector for parts of of the final deed was copied by hand and certified. society not reached by formal operators. Regulators are In Andhra Pradesh, 387 subregistrar offices registered about 1.2 million documents a year, 60 percent of them typicalyhostletonforma providers.Busomedevelfor agricultural land. A yearly manual update of property in- oping countries recognize the benefits of allowing those formation was carried out, since hundreds of thousands providers to operate. In Paraguay about 400 private of property files were updated with the new sales from water suppliers operate their own wells and provide the year. Land registration offices throughout the state are now piped water to households unserved by the public sec-eqipdwtcouerzdonesudrthCi- tor. Imposng strictstandardson providrsuigim equipped with computerized counters under the Com- ror. Imposing strict standards on providers using sim- puter-aided Administration of Registration Department pie technology would immediately drive these private (CARDi project, initiated and financed by the state govern- suppliers out of business. A more gradual evolution in ment to improve efficiency and increase duty collections. regulation is needed.14 Starting with a pilot project in 214 locations over 15 months, Countries do not have to go through a long learning- the entire database was transferred to computers, the Countres do not have to go through a long learnng- copying and filing system was replaced with imaging, and by-doing process in all aspects of institutional develop- all back-office functions were automated. Standardization ment. They can transplant and modify some insti- and greater transparency in property valuation procedures tu tional forms from other countries and shorten the boosted stamp duty revenues. Registration processing time was cut from 10 days to 1 hour. development process by learning from other countries. They can also use Internet technology to reduce insti- Source: Case study by Dr. Subhash Chandra Bhatnagar, tutional constraints and improve the effectiveness of in- University of Delhi. World Bank 2000, as part of the stitutions. In many developing countries the Internet E-Government Focus Group, available at http://wwwl. is already providing the means for accelerated learning, worldbank.org/publicsector/egov. improved information flows, reduced enforcement costs, and enhanced competition in markets (box 1.8). But to leapfrog stages in development through technol- distributional effects, can discriminate systematically ogy, policymakers need to increase access to technology against poor countries. For example, the Trade-Related for market agents. Market rules affect access. Aspects of Intellectual Property Rights Agreement International rules and standards. Standardizing laws (TRIPS) can impose significant costs on poor countries, and regulations generally reduces the information and because strong patent protection is not as appropriate enforcement costs of transactions across borders and can for them as it is for rich countries. Many industrial enhance trade efficiency. International standards also countries themselves only recently adopted laws safe- have the potential to provide benefits much larger than guarding intellectual property, and the nature of these those under bilaterally agreed standards between coun- laws has evolved over time in response to changing do- tries for both poor and rich countries. But depending mestic economic and political factors. Developing coun- on which standards are chosen, international standards tries also lack the supporting institutions to implement can also be costly for poor countries and can have sig- TRIPS effectively-these will take time and resources nificant distributional consequences between countries. to build. International trading rules and principles, enshrined Another example is the adoption of international ac- in the World Trade Organization (WTO), promote trade counting rules by companies in many developing coun- (chapters 5 and 7). But some standards, through their tries. This has enhanced their access to credit in inter- BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE I5 national markets. Voluntary adoption of standards by Sicw- E %. firms wanting to obtain credit in international markets Private innovation supported by formal is likely to be beneficial. But these standards are not ap- -nstitutioal change_ propriate for smaller firms (chapter 3). And forcing In Bangladesh an economics professor had an idea-to small firms in developing countries to adopt them help poor people help themselves by giving them small would raise their costs and possibly push them into the loans to start businesses despite their lack of collateral or informal sector. credit histories. He started the Grameen Bank in 1976 For international standards to truly benefit all coun- using his social connections in government to manage a tries by facilitating trade-and to avoid systematic bi- village branch of a government bank. The success of this endeavor, followed by expansion to other bank branches, ases against developing countries-the standards need led the government to eventually change the laws govern- to reflect realities in developing countries. These include ing the Grameen Bank. It was established first as an inde- the costs of adhering to standards as well as the benefits, pendent entity with government control, then as an effec- and particularly important are the costs imposed on the tively private bank run by a public official, and finally as an effectively private bank run by a private individual and an poor. Important questions are: W hose standards should independent board of directors. Today, Grameen Bank has be adopted and why, and what is the process under branches in more than half the villages in Bangladesh and which these standards are negotiated? The process of re- more than 2 million borrowers. forming international rules needs to be transparent, and In Peru another innovative individual began with an ex- be active participants to periment. He found that in Lima it took 728 bureaucratic developing countries need to be acuve partlclpants to steps for a person with an informal right to housing to get influence outcomes in their favor. But human capital legal title. He followed up with a 10-year public informa- constraints may prevent developing countries from rep- tion campaign, proving to politicians that there was a "hid- resenting their interests. In such circumstances interna- den consensus for reform" for simplifying the procedures tional donors could help enhance their representation, for formalization. Faced with overwhelming public support for simplification, the Peruvian congress unanimously or developing countries could pool their scarce technical passed legislation to formalize titles. Today, a simple legal skills and have common representation at international procedure for establishing land titles for poorer people negotiations or hire private specialists to represent them. works in parallel with the formal system. Variation within countriers. Some variation in insti- These two stories show how the state can work with Variationsmaybe w sitn coutri Somhefvaiaincy in ist.i- private actors to promote institutional innovation by directly tutions may be desirable for both efficiency and distri- supporting experiments-or at least by allowing them to butional reasons, even between regions within coun- proceed and be tested and then, if they are successful, tries. Even industrial countries do not standardize all by encouraging their growth. The stories also show the im- laws and regulations within the country. For example, portance of other factors in promoting innovation. Social connections and networks can reduce barriers to experi- Australia and Canada have different laws in different mentation. Openness in information sharing provides the states for secured transactions. Different states in the impetus to adopt and expand successful experiments. United States have different corporate laws. The differ- ences exist because of variations in economic and social source: De Soto 2000; Yunus 1997. structures-variations that can be particularly instruc- tive for large countries such as Brazil, China, India, and the Russian Federation. Of course the costs of standard- however, and these need to be balanced against poten- ization versus diversity will vary depending on the in- tial benefits. stitution and the relative distribution of gains and Policymakers can replicate successful local innova- losses. Where spillover effects across jurisdictions are tions. But they also need to be flexible enough to drop large and not sustainable at the macroeconomic level, unsuccessful experiments. Because innovation can come variation may be less desirable. from many sources, collaboration by the different ac- tors in society is vital, as shown by the development of Innovate to identify5 itnst.ittion7s that work microfinance institutions in Bangladesh, where the Even at similar levels of development, countries differ government adapted its formal legal structure to ac- in many ways-in their norms, geography, and endow- commodate private innovation, and the process of land ments. Innovation, often through experimentation, can titling followed in Peru (box 1.9). help accommodate those differences and produce more In some cases, greater local autonomy and participa- effective institutions. Experimentation also has costs, tion may foster institutional experiments that lead to in- i6 WORLD DEVELOPMENT REPORT 2002 novation. For example, Aguas Argentina, a privatized Box 1.10 monopoly that provides water and sanitation services Distributional effects of innovation depend on in Buenos Aires, worked under a novel institutional who innovates: bankruptcy law in two countries arrangement to design new ways to organize service de- livery. The monopoly worked with local government, a The United Kingdom created its bankruptcy regime livery. T mn lwthrough explicit legislation that recognized the importance low-income community, and a nongovernmental orga- of decentralized contracting: legislation stated that corpo- nization (NGO) to create a new organizational form to rations were free to make the rules under which they improve service delivery. The community was experi- would transact. Lenders and borrowers in the United King- menting with two systems: a low-cost sewerage system dom had the power to innovate through contracts, and aenting a . , over time commercial practice was incorporated into law. and a double water system (with one connection to the In the United States judges and legislators held that power. network for small volumes of potable water and another The U.K. system, designed by private agents engaged drawing on groundwater sources too salty for drinking in borrowing and lending, is today characterized by a great but good for washing and bathing). The double water concentration of rights in favor of the principal lender. The principal claimant appoints a receiver who uses his powers system was dropped at the experimental stage because it for the sole purpose of repaying this principal's debt. The was too expensive to develop, while the sewerage system court's role is much less significant than it is in the United was maintained. To expand its water network, Aguas States, and the judgment is not subject to court review. Argentinas took over those systems built at lower cost In the United States, Chapter 11 bankruptcy law is the community, giving customers a discount on the characterized by a partial dispersion of rights away from by tne communlty, glvmg customers a alscount on tne secured claims (priority lenders). U.S. legislation was price in exchange. In effect, it had contracted out con- amended several times at moments of economic crisis at struction to the community.15 the instigation of the judiciary. At these times preservation Innovation through experimentation can happen at of companies rather than their dissolution was uppermost different,levels..Experimentation and innovation occur in the minds of legislators and judges-leading to a debtor- friendly bankruptcy law. Upon default a company in the on at least three levels: national public policymaking, United States may seek protection from its creditors, usu- private commercial practices, and local action by com- ally retaining control over the business. munities and civil society leaders (see box 1.16). Local experimentation has the advantage of allowing many innovations to be tried simultaneously-with the suc- cessful ones replicated and the failures contained. But not all innovations can be left to local or decentralized novation.16 The tension between experimenting and communities-since local actions may have conse- standardizing public institutions within countries will quences across communities and too much experimen- be settled in favor of the latter when effective institu- tation can lead to each community having different tional forms are found. Policymakers have to ensure that rules. Local innovation can also open institutions to successful local innovations can be scaled up. They must capture by local elites, inviting corruption. When ef- also be willing to drop outdated institutional forms. fective innovations are identified, policymakers can help Hungary, in the early years of its transition, for exam- expand such institutions by replicating them in other ple, experimented with a particular form of bankruptcy areas (for example, through adopting a law) or by shar- law, which was later dropped when conditions changed ing information on the innovation. and a more effective alternative emerged (box 1.11). Who innovates determines institutional evolution. De- pending on who innovates, institutions can evolve in Connect communities through information flows quite different ways (and with quite different distribu- and trade tional consequences), as shown by the evolution of Open information exchange and open trade promote bankruptcy law in the United Kingdom and the United institution building by creating demand for market- States (box 1. 10). As history shows, during the devel- supporting institutions. opment process the institutions adopted favor those Open trade. Going beyond allocative efficiency, open who control the process. trade does more. Debates among people who formulate policy, those who implement it, and those outside government can * It exposes market participants to a larger, more di- help in disseminating information on institutional in- verse, group of trading partners, increasing the de- BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE 17 Box 1.11 Box 1.12 Experimentation and adaptation: Trade and institutional change in Thailand bankruptcy institutions in Hungary In the early 19th century, with labor scarce and land abun- In 1992 the Hungarian government adopted a bankruptcy dant, land had little value in Thailand. Slaves rather than code giving creditors very strong rights to file for bank- land were taken as collateral in financial markets. Corre- ruptcy. The intention was to impose a hard budget con- spondingly, land markets were underdeveloped. There straint on firms, particularly on large enterprises. The law was little demand or need for the development of formal therefore stated that the creditor could file for bankruptcy institutions. But there was a well-developed legal system if a company was three months or more overdue on any to govern transactions in labor commitments. In theory, all debt (known as an automatic trigger). Since accounting land belonged to the king. In practice, individuals could use systems were underdeveloped, information on the true and sell the land, as long as they paid taxes and did not let performance of firms was not readily available, and the it lie fallow for more than three consecutive years. available information was not always reliable. The solvency In the latter part of the century, international trade or insolvency of a firm was therefore hard to measure. opened up, and transport costs declined. A rice export The short time frame and the establishment of such a boom led to a rapid expansion of production and use of strong trigger for bankruptcy proceedings led more than land. Land became more valuable, land disputes more 5,000 firms to file for bankruptcy. The government had not common. The demand for formal institutions, such as reg- expected such a large number of bankruptcies, particularly istries, to convey information and enforce property rights of small firms. The automatic trigger allowed the govern- increased. ment to assess quickly the true condition of firms. But be- The government responded by implementing a series cause the courts dealt with so many cases, they quickly of procedural and administrative changes, beginning in developed experience in handling bankruptcies. The re- 1892. The first initiative, to document land rights, was sult: the authorities abolished the trigger in 1998. Not only modified and improved several times; the final legislation were courts better able to adjudicate bankruptcies, but was passed in 1954. The current legislation is a compro- better information systems had developed to allow credi- mise between traditional practice, which allowed citizens tors to monitor companies. Market dynamics and support- to bring unoccupied forestland under cultivation as private ing institutions had evolved enough so that the law was property, and the more formal requirement of land titling no longer needed. based on detailed land surveys. Such institutional evolution is not unique to Thailand, Source: Gray and others 1996. for industrial countries have also shown that trade, by changing the terms of trade, gives rise to the demand for clear property rights and a need for the state to define them. mand for formal institutions to provide information and enforce contracts.17 Source: Siamwalla and others 1993; Stifel 1976. * It helps firms learn about technology and about or- ganizational and managerial forms. * It exposes markets to greater competition and land titles in many countries (chapter 2) developed changes in relative returns, which induce institu- once markets for goods produced on the land were ac- tional change (see below). cessible or when new members entered the community. * It exposes countries to a different set of risks, possi- Empirical work spanning over 110 countries shows bly supporting the creation of additional institutions that measures of institutional effectiveness (such as the to manage the new risks. quality of institutions for public service delivery, or per- * It brings new market participants from other coun- ceptions of the rule of law) are significantly related to tries or regions who also demand more effective in- openness in international trade. This is so even after ac- stitutions to support market transactions. counting for differences in size, per capita income, legal heritage, years the country has been independent, and The case of Thailand illustrates how liberalization of other factors (figure 1.5).i9 trading rules led to a shift in agricultural returns-and Greater openness in trade and capital markets has to institutional change in the market for land (box been associated with the development of financial sys- 1.12).i8 Similar patterns are observed in other coun- tems, as historical and cross-country analyses clearly tries and sectors. The development of standards for rice show. Large incumbent firms that have access to fi- within Japan was spurred after markets within Japan nance-through either retained earnings or established were connected (box 1.13). The demand for formal links with financial institutions-do not always have an 18 WORLD DEVELOPMENT REPORT 2002 Box 1.13 Figure 1.5 Institutional evolution of rice markets and Greater openness and quality of institutions standardization in Japan, 1600-1920s Rule of law index In Japan's Tokugawa period (1600-1868), local private 0.6 - traders collected and marketed the rice shares of both the (a3 daimyo (feudal lord) and the peasant. The traders had to be big, since poor inland transport meant that rice was 0.4 - shipped in large sailing vessels and later steamships-a costly and risky venture. When the network of railroads was extended to local areas, locally segmented markets 0.2 - began to form a nationwide market. And with the econo- mies of scale in transport and related risks, small traders _ I could market their rice, using small shipments from many local centers. ° - The competition among small traders from different 1st 2nd 3rd rice-producing regions increased the pressure to standard- Openness (terciles) ize rice grades. Better and more stable quality and stan- dards ensured higher prices in urban markets. Groups of Government effectiveness index farmers and traders began taking the initiative by labeling the quality of rice in various regions. By 1900 these volun- (b) tary efforts were transformed into official regulations by 0.4 - local government agencies, which began to set standards for the packaging of rice shipped to other regions. By 1910 there were 33 rice-grading warehouses (beiken soko), managed by private companies or cooperatives, serving 0.2 - several purposes-inspecting, grading, repackaging, and storing. Innovations in finance followed. As farmers and traders I - brought ungraded rice to the warehouse, it issued a "rice 0 - exchange note." The precursor to today's inventory credit, 1st 2nd 3rd these notes were also used as collateral for loans from Openness (terciles) banks and pawnshops, easing capital constraints for farm- Note: The figures show the partial relationship (after accounting for the ers and traders. effect of differences in the legal systems, ethnic diversity, GNP per More trade among different communities led to the de- capita, years that the country in question has been independent, velopments of standards, first adopted by private traders country size, and inequality of income) between an indicator of rule of and later by government. These early institutional changes law/ government effectiveness and openness for over 100 countries promoted new institutions to support market exchange. in 1997-98. The countries have been divided into three groups of equal size. Source: Islam and Montenegro forthcoming, World Development Source: Kawagoe 1998. Report 2002 background paper. incentive to promote financial systems that would fa- with underdeveloped supervisory and regulatory sys- cilitate new entry into their markets. Opening the econ- tems. To get around weak judicial systems, poor coun- omy to trade and financial flows can automatically re- tries can export the enforcement of contracts. For infra- duce rents that incumbents receive from preferential structure deals in which private investors from rich access to financial institutions. And over time the lower countries invest in poor countries, for example, interna- rents can reduce opposition to financial sector reform.20 tional arbitration clauses can be used in cases of dispute. Rather than improve their own systems, policymak- Open information flows. Open information exchange, ers in open economies can import whole aspects of the a driver of institutional development, can both improve institutional system: laws, regulations, and enforcement the quality of other existing institutions and create a de- systems. Because of the political problems and costs of mand for new ones. Better information makes monitor- importing foreign agencies, including foreign human ing peoples' behavior easier. This ability to monitor be- capital, there are not many examples. Many countries havior changes behavior and institutional quality even have allowed foreign banks to operate in the domestic when institutional structure does not change. Better in- financial sector, helping financial services grow even formation can also change social norms and so change BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE 19 people's incentives to participate in different institu- ing-both in time spent and in bribes paid-led the tions. And it can inform policymakers and other mar- government to undertake reforms that reduced licens- ket participants about the benefits of institutional re- ing time from years to days.22 With poor information form and about the constraints on institutional reform. flow in an economy, regulatory rules and policies are Information from the media and low-cost informa- unclear. So regulated firms and customers do not know, tion on the Internet can enhance the functioning of or cannot find out, what regulations apply to them or public institutions. Evidence indicates that corruption, how to comply with them.23 for example, is lower in countries with a free press (box 1.14). There is also evidence that free media, by pro- Promlote cuiiipetition--amAxaongj,,m?isdictiotas,firins, viding a check on political actions, can raise policymak- anid individuals ers' awareness of the social effects of policies, improv- Competition among jurisdictions, among firms in ing the provision of social services. A study in India product markets, and among individuals does much found that the media affected how the government re- for institutional change.24 Often, current institutional sponded to floods and famines: the distribution of re- lief was greater in states with higher newspaper circula- Figure 1.6 tions. The more information the local media provided, Diversity of information providers and quality of the more effectively citizens could develop a collective institutions voice and put pressure on the government.21 Recent research for this Report shows that competi- Press freedom index tion in the provision of information can significantly 0.8 - increase the impact of the media on the quality of in- 0-7 - stitutions. For example, where the state does not con- 0.6 - trol information through monopoly or concentrated 0.5 - ownership of the media industry, the media can do _ much in checking corruption (figure 1.6). The effect of 0.4 private monopolies on information flow can be ex- 0.3 - pected to be similar. 0.2 - Information about the potential benefits and costs 0.1 - of particular institutional arrangements can change the 0 L I I I incentives for those who engage in market transactions 0-25 25-50 50-75 75-100 and the demand for institutions. In Nepal the publica- State ownership of press tion of simple facts about the costs of business licens- (percent) Effectiveness of government Box 1.14 0.6 - Role of the news media in fighting corruption 0.4 - in Kenya 0.2 - 0- In 1996 investigative lournalists from a privately owned newspaper uncovered evidence of corruption in Kenya's -0.2 - ministry of health. A purchase of unapproved malaria -0.4 - chemicals was planned through a foreign firm at a substan- 0.6 - tially higher price than local firms were charging. It was____ also reported that the health minister paid the foreign firm -0.8 - I 400 million Kenya shillings, even though no goods were re- -1 L ceived in return for this payment. The press revealed these 0-25 25-50 50-75 75-100 findings and made daily reports on the scandal. Eventually, State ownership of press under unrelenting pressure from the media, the minister (percent) was dismissed. Source: D)ankov and olheTs 2001, Wordd Deve$opment Reporl 2002 Source: Githongo 1997; Stapenhurst 2000. background paper 20 WORLD DEVELOPMENT REPORT 2002 structures may inhibit competition. Competition makes Box 1.15 institutions more or less effective by affecting relative Competition and the evolution of corporate law returns and changing the incentives of agents. For ex- ample, as competition in markets increases, traditional A study investigating legal change in 10 jurisdictions, in- norm-based institutions may become inadequae or .cluding both industrial and developing countries over more nombse ntiuinsmy e in othan 100 years, found competition among firms operating obsolete.'5 Competition can reduce the effectiveness within countries and across borders to be important in pro- of closed groups, such as guilds or business networks, moting changes in the corporate law. The changes were whose existence and effectiveness depend on superior often enacted in response to crises, owing to competitive access to such inputs as information. This can create the pressures, or as a conscious effort to standardize corpo- accessto suc mputsas mtomatlon Thls an crete therate law across countries. demand for new institutions or improve the quality of Studies of Europe in the late 19th century highlight ju- existing institutions by changing behavior. In places as risdictional competition in the development of corporate varied as Thailand and Uganda, greater competition for law. There was a shift from concession systems, in which rulers granted the right to incorporate case by case and land increased land disputes and created the demand for often as a special favor, to a system of registration in more formal procedures for recording transactions. which any company meeting certain minimum require- Competition in product markets has led to institutional ments could incorporate. For example, in France in 1867, change in labor markets (chapter 7). And there is some the shift was induced by the expansion of English compa- nies on the continent. Once France allowed companies in- evidence that competition between firms can be a par- corporated in England to operate as a corporation in France, tial substitute for strong shareholder rights in inducing without special approval by parliament, it faced pressure managers to act in the interest of owners. from domestic companies to drop the concession require- Firms competing in product markets, forced to in- ment at home. Israel in 1999, Japan in the 1990s, Chile in 1981, and crease efficiency, have the incentive to lobby policymak- Delaware in the United States (where there have been ers to implement institutional changes that lower their continuous changes) provide examples of jurisdictions that costs. Competition also affects the distribution of gains changed their corporate laws in response to competitive among market players, and so increases the demand for pressures in the 20th century. institutional change among those who want to maintain Source: Pistor and others 2000, World Development Re- their gains in the light of changing economic factors. port 2002 background paper. But sometimes institutions, such as rules governing in- tellectual property, may be needed to limit the degree of competition in markets and to foster innovation. and corporate laws vary across the states. Education sys- For firms in international capital markets, competi- tems vary across districts. tion can produce demand for better institutions, such Markets with more competition may require fewer as accounting standards (chapter 5). In turn, domestic formal institutions, since competition can substitute banks, to compete with foreign banks outside their for regulation. Take infrastructure: greater competition, home markets, may pressure their regulators to improve possible with technological changes, has allowed regu- prudential regulations. This happened in Mexico after lators to lower the frequency of price reviews (chapter it signed the North American Free Trade Agreement. A 8). Sectors previously considered natural monopolies World Bank study looking at institutional performance became potentially competitive, so governments now cites competition as a key factor affecting institutional rely more on competition to deliver desired outcomes, performance, since it changes the incentives for indi- such as affordable prices for consumers. viduals to succeed.26 But the competition from new infrastructure pro- Jurisdictional competition also fosters institutional viders can also complicate regulation. Before the priva- evolution. A study of corporate law evolution shows tization of state monopolies in many countries, state that competition between countries-and between for- infrastructure monopolies cross-subsidized some cus- eign firms operating in a country-has created pressure tomers-in many cases lowering costs for poorer house- for change in corporate laws (box 1. 1 5). In the United holds by charging higher prices to business users. After States competition between states to attract business privatization, governments aiming to protect poorer has led to institutional evolution of different forms in customers have found it difficult to regulate the priva- the various states. For example, personal bankruptcy tized firms in a way that provided adequate profits for BUILDING INSTITUTIONS: COMPLEMENT, INNOVATE, CONNECT, AND COMPETE 2I firms while providing adequate services to the poor since they could keep any collections above a certain (chapter 8). level. Local governments were also able to raise addi- There are times when institutions restricting com- tional taxes, not shared with the national government. petition are desirable. Some market rents may need to But all reforms are not equally difficult politically. be tolerated to fund the adoption of new technology, Some ineffective institutions may exist in part because and institutions restricting competition may be needed there are no interest groups pressing for change-not to promote market development. And regulating the because some interest groups oppose change. Or it may degree of competition among banks can enhance finan- be that those who would oppose change do not have cial stability by reducing the incentives for risk taking. much political sway. Whatever the reason, reforms in these areas could be accelerated. And as these reforms How1 do poditicaiftrwces. s;ia1lpr pssllres, breed new constituencies and forces, they can lead to a H7mdl shocks affect thc pace of change?J demand for greater change. The key is to find the op- Political forces and social pressures can either acceler- portunities and to work in these areas. ate or retard the development of new institutions. Shift- Although indigenous institutional development re- ing social, political, and economic balances are in turn sponds to changing economic and social conditions, a affected by a government's institutional reform efforts. central issue for transplants is managing distributional In industrial markets, however, the state is constrained conflicts. Institutional change creates winners and losers. from arbitrarily changing rules and laws, and there tend For example, bankruptcy law designates the rights to in- to be more checks and balances on various actors, pub- come and assets for creditors. Corporate law distributes lic and private. rights among owners, managers, and the government. Politicalforces. An institution exists in part because Regulation covering service provision to the poor trans- some constituencies gain from its existence and so have fers economic gains from producers to poor consumers the incentives and influence to support it. This distri- and between levels of government (chapter 8). butional aspect is particularly important when institu- The distribution of power among different levels of tions benefit a small group or minority in society for government largely determines the type of regulatory whom the costs of collective action are low and bene- structure likely to be effective. A study of the evolution fits are large. Checks and balances on political power, of regulation in infrastructure conducted for this Re- from firms and interest groups, can support the inter- port argues that the allocation of regulatory authority ests of the majority. But minority interests may in som