24897 December 2001 The World Bank Research A bstracts Program of Current Studies 2001 m Contents Introduction 1 Studies by Subject Area 3 Abstracts of Current Studies 9 Poverty and Social Development Health and Population 40 Education, Labor, and Employment 46 Environment 58 Infrastructure and Urban Development 68 Agriculture and Rural Development 87 Macroeconomics and Growth 100 International Economics 106 Domestic Finance 120 Industry and Private Sector Development 130 Governance and Public Sector Management 139 Bank Research Output 148 iii Introduction The World Bank's research, described annually in this have received intensive analytical training. Those work- publication, is evaluated periodically by external experts ing on the Bank's lending operations or providing advi- and occasionally by policymakers in developing and sory services wrestle constantly with the needs and transition economies. Evaluators generally commend problems of developing and transition economies. the Bank's research program for its focus on important Through their international experience, they often dis- topics and its empirical testing of theoretical concepts in cern patterns in the successes and failures of policies and the light of international experience. Those character- strategies-patterns that would not be evident at the istics reflect a review process established more than 30 national or even regional level. By rigorously analyzing years ago and the training and experience of a unique these patterns, Bank staff have advanced our under- international staff. standing of a wide range of critical development issues. The review process has been developed, revised, and Experience has revealed that a large number of managed by the World Bank's Research Committee, factors-beyond economics and finance-shape suc- which allocates central funds for research within the cessful development. The Bank's mandate has broadened Bank. The committee's mandate is to ensure that Bank over time to accommodate these factors. This breadth has research is relevant to policy and meets high standards come at a cost, however. Responsible for a widening of analytical quality. Its review process is competitive, and range of projects and programs, Bank staff must keep funds are available not only in Bank research offices but abreast of developments across an expanding profes- also to staff througlhout the Bank. Both Bank staff and sional portfolio and make long and frequent trips abroad. external experts review grant proposals, and the com- These demands absorb time that could otherwise be mittee makes final decisions. used to test the validity of the insights suggested by The committee's chairman is Nick Stern, the Bank's experience. As a result, research by the staff closest to chief economist and senior vice president of development government policymakers is declining, which could economics. Its members include regional chief economists reduce the relevance and pragmatism of the Bank's from lending operations offices, sectoral experts and research. advisers to Bank member countries, and managers of the World Bank Institute's learning programs, the Bank's A New Initiative Promotes Research evaluation department, and the International Finance Critical for Policy Corporation. These committee members bring to their reviews both comprehensive knowledge of analytical To address these issues, this year the Bank established methods and familiarity with the most important devel- the Regional and Network Research Initiative to focus opment problems facing policymakers in Bank member research on issues critical for development policy and countries. Bank operations and to help reverse the decline in research by lending operations staff and Bank sectoral World Bank Staff l(eep a Sharp Focus on Policy experts. The program will assist operations and network staff But the results of the committee's review process are only in planning and developing research projects, fund advice as good as the best of the proposals submitted-and and participation by experts in these projects, and increase those are both rigorous and relevant, thanks to the edu- the incentives for conducting research by formally cation and day-to-day work of the Bank's staff. Drawn acknowledging the operations and network teams that from the world's leading academic institutions, the staff make the most valuable contributions in research. The initiative also will encourage research on priority regions, * Infrastructure and urban development. such as Africa, and on priority issues that remain poorly * Agriculture and rural development. understood, such as poverty, inequality, governance, and * Macroeconomics and growth. private sector development in transition economies. * International economics. Research funded under this new initiative will be high- * Domestic finance. lighted in future editions of the Abstracts of Current * Industry and private sector development. Studies. * Governance and public sector management. The appendix lists reports and publications produced What Research Is Now Under Way at the Bank? from Bank research and explains how to obtain them. This volume reports on research projects initiated, under How to Get More Information way, or completed in fiscal 2001 (July 1, 2000, through June 30,2001). The abstracts in the volume describe, for This is an annual compendium; readers interested in each project, the questions addressed, the analytical obtaining more timely information on World Bank methods used, the findings to date, and their policy research and its findings may wish to subscribe, free of implications. Each abstract also identifies the expected charge, to the quarterly Policy and Research Bulletin (see completion date, the research team, and any reports or the subscription request form at the back of the vol- publications produced. To make it easier to obtain infor- ume). Each issue includes information on recent World mation and data, each abstract gives the email address Bank publications and working papers, abstracts of newly for the research project's supervisor. initiated research projects and summaries of recent The abstracts cover 150 research projects from research findings, and a column identifying electronic throughout the Bank, grouped under 11 major head- information available from the Bank. ings: The Bulletin, Abstracts of CurrentStudies, articles from * Poverty and social development. the Bank's two research journals-the World Bank Research * Health and population. Observer and World Bank Economic Review-and many * Education, labor, and employment. research reports and publications are available on the Web * Environment. at http://econ.worldbank.org. 2 Studies by Sub ject Area Poverty and Social Development 11 The Impact of the Financial Crisis on the Farm Sector in Thailand 11 Technology and Poverty Alleviation 11 Does Microcredit Empower Women? A Study of Grameen Bank, BRAC, and the RD-12 Projects in Bangladesh 13 How Do the Poor Cope with Widespread Natural Disaster? A Study of the 1998 Flood of Rural Bangladesh 13 Long-Term Impacts of Microcredit Programs: A Study of the Grameen Bank and Other Programs in Bangladesh 14 Policy Research Report on Gender and Development 15 The Impact of Prices, Taxes, Subsidies, and Stipends on Poverty 17 A New Analytical Framework for Evaluating Social Programs 18 Private Transfers in a Cross-Section of Developing Countries 19 The Impact of Growth on Poverty in Latin America 20 Using Semi-Parametric Methods for the Evaluation of Social Programs and Policies 20 Economic Growth and Household Welfare: Policy Lessons from Vietnam 21 Household Coping with AIDS in Tanzania 22 Informal Safety Nets in Nicaragua 22 Social Exclusion and Poverty 23 The Dynamics of Urban Poverty in Rio de Janeiro and Implications for Public Policy 24 Extreme Poverty and Social Exclusion in Latin America 25 The Methodology of Poverty Assessments 26 Safety Nets in an Emerging Market Economy 27 Will Social Security Reform in Latin America Provide Greater Retirement Security? 28 World Poverty Monitoring 28 Patterns of Inequality and Government Intervention 29 Public Spending and the Poor in Latin America 30 Urban Poverty, Risk Management, and Social Capital . 31 Cultural Change, Community Mobilization, and Participatory Development 31 Innovations in the International Crafts Market and Empowerment of Artisans in India 32 Organizational Cultures and Spaces for Empowerment: Interactions between Poor People's Organizations and World Bank Poverty Programs 33 3 Evaluation of the Impact of Investments in Early Child Development on Nutrition and Cognitive Development 33 Poverty and the International Economy: What Are the Links? 35 Welfare Impacts of Policy on Disadvantaged Groups 36 Social Capital 37 Health and Population 40 The Impact of the AIDS Epidemic on the Health of the Elderly in Northwestern Tanzania 40 Optimal Policies for Controlling Infectious Diseases 40 Africa Nutrition Database Initiative 41 Poverty and Health 42 The Economics of Malaria 43 The Public Economics of Health Reform 44 Education, Labor, and Employment 46 The Impact of Labor Market Policies and Institutions on Economic Performance 46 Income Support for the Unemployed: Mandatory Severance Pay in Peru 47 Assessing the Impact of Early Childhood Malnutrition on Academic Achievement 48 Evaluation of Active Labor Market Programs in China 48 Migration, Remittances, and Poverty in Latin America 49 Improving Primary Education in Kenya: A Randomized Evaluation of Different Policy Options 50 Providing Unemployment Benefits through Unemployment Insurance Savings Accounts 51 The Impact of Deworming Treatment on Primary School Performance in Busia, Kenya 52 Educational Enrollment and Dropout 53 Gender Data and Labor Markets 54 Educating Nomadic Herders out of Poverty 55 Public Sector Retrenchment, Phase 2 56 Environment 58 Economic and Environmental Impacts of Lowland Agricultural Development on Poor Upland Communities in Palawan, Philippines 58 Factors Affecting the Pace of Deforestation in Northern Thailand 58 Economic Instruments for Habitat Conservation 59 Global Overlay Brazil 60 Nature Tourism's Contribution to Economic Development and Conservation Finance 61 Understanding and Improving the Environmental Performance of China's Township-Village Industrial Enterprises 62 4 Studies by Subject Area Making Long-Term Growth and Development More Sustainable 63 Valuing Mortality Risk Reductions 64 Measuring the Economic Value of Environmental Protection Projects: Methodology and Application to Armenia's Lake Sevan 65 Pesticide Use in Brazil 65 Environmental Data Accounting 66 Biodiversity, Habitat, Conservation, and Valuing Natural Resources in the Amazon Estuary 67 Infrastructure and Urban Development 68 The Impact of Regulatory Risk on the Cost of Private Debt for Infrastructure Projects in Emerging Markets 68 Yardstick Competition across Ports: An Illustrated Guide for Regulators 69 Community Comanagement of Urban Environmental Quality: Water, Sanitation, and Water Pollution Control 70 Institutions, Politics, and Contracts: Private Sector Participation in Urban Water Supply 71 Competition and Privatization in Urban Water Supply 72 The Effects of Telecommunications Infrastructure on Investment: An Empirical Analysis 73 Privatization and Basic Infrastructure Services for the Urban Poor 74 The Sustainable Financing of Investments in Municipal Infrastructure: Cost Recovery for Solid Waste Management in the Philippines 75 Comparative Smtdy of Water Institutions and Their Impact on Water Sector Performance in Selected Countries 76 The Impact of Rural Roads 77 The Impact of Urban Spatial Structure on Travel Demand 78 Linking Culture and Poverty Reduction in the Himalayas: A Comparative Study in Bhutan, Nepal, and Ladakh (India) 79 The Macroeconomics of Infrastructure in Latin America 80 Policy Research Report on Regulation 80 Economic and Engineering Evaluation of Alternative Strategies for Managing Sedimentation in Storage Reservoirs 81 Privatization of Telecommunications in Sub-Saharan Africa 82 Connecting Cities with Macroeconomic Concerns: The Missing Link 83 Emergence from Subsistence: Infrastructure, Location, and Development in Nepal 83 Information, Knowledge, and Capacity Building for Effective Urban Strategies: Information-Based Instruments for Urban Management 84 Privatized Utilities 85 Rural Roads: Welfare Impact Evaluation 86 Studies by Sublect Area Agriculture and Rural Development 87 The Impact of Public Spending in Uganda 87 Market Development and Allocative Efficiency: Irrigation Water in the Punjab 87 The Dynamics of Rural Sector Growth 88 Guidelines for Pricing Irrigation Water Based on Efficiency, Implementation, and Equity Considerations 89 Reforming China's Agricultural Research System: A Research Production Function Approach to Measuring Economies of Size and Scope and Efficiency 90 Weather-Based Index Insurance 91 Agriculture in Transition: Land Reform and Farm Restructuring in Formerly Socialist Countries 92 Sending Farmers Back to School: An Econometric Evaluation of the Farmer Field School Extension Approach 94 Welfare Outcomes of Decentralized Land Reform 95 Commodity Risk Management and Export Finance 96 Land Markets, Gender, and Access to Land in Latin America 96 Land Reform 97 Land Tenure in Rural China 98 Might Culture Pay Off? Evaluating the Effects of Farming Innovations and Cultural Empowerment among Lowland Amerindians in Bolivia 99 Macroesconomics and Growth 100 Crisis, Polarization, and Reform 100 The Quest for Growth 100 Structural Adjustment Participatory Review Initiative 101 Capital Subsidies and the Quality of Growth 102 Trade and Fiscal Reform and Investment Booms in Latin America: The Chilean Case and Application to Other Reformers 102 Economic Growth, Social Capability, and Preindustrial Development 103 The Quality of Fiscal Adjustment 104 Geography, Growth, and Comparative Advantage in Sub-Saharan Africa 104 International Economics 106 Aid and Reform in Africa 106 The Antiexport Bias of Duty Drawbacks 107 Comprehensive Database on Trade and Production 107 The Dynamic Impact of Trade Liberalization in Developing Countries 108 Forecasting Capital Flows 109 International Capital Flows 109 Regionalism and Technology Transfers 110 Micro Foundations of International Technology Diffusion 111 6 Studies by Subject Area Trade Liberalization, Industrial Performance, and Export Growth in India 112 The Effects of American Depository Receipt Trading on Local Markets 113 Financial Development and Contagion 113 Trade Policy in Transition Economies 115 African Trade Facilitation and Standards: Bridging the Divide 117 Trade in Services 118 Trade, Standards, and Regulatory Reform 119 Domestic Finance 120 Deposit Insurance: Issues of Principle, Design, and Implementation 120 Policy for Small Financial Systems 121 Financial Liberalization 122 Bank Insolvency 123 Incentives in Banking 123 Corporate Governance, Corporate Finance, and Economic Performance in Developing Countries 125 Twin Crises and Government Policy 125 Institutional Investors 126 Bank Supervision and Regulation 127 Taxation of Financial Intermediation 127 Bank Privatization in Developing Countries 128 Financing of Small and Medium-Size Enterprises 129 Industry and Private Sector Development 130 The Economics of Ethnicity and Entrepreneurship in Africa 130 Corporate Governance in East Asia 131 East Asian Competitiveness Study 132 An Empirical Study of the Surgical Instrument Cluster of Sialkot, Pakistan 133 Firm Analysis and Competitiveness Surveys 134 Privatization and Corporate Governance in Transition Economies 135 Firm Analysis and Competitiveness Research 135 Regional Program on Enterprise Development in Africa 136 Governance and Public Sector Management 139 Agency Independence and Political Institutions 139 Analyzing Problems in Public Hospital Corporatization Using Information Economics 139 Migration, Decentralization, and the Provision of Local Public Goods to the Poor 140 Service Delivery 141 The Causes of Corruption 141 Database on Institutions for Government Decisionmaking 142 The Economics of Political and Criminal Violence 143 Studies by Subject Area The Impact of Institutions 144 Managing Windfalls: Institutional Strategies for Avoiding the Voracity Effect 145 An Analysis of Use Patterns for Latin American Judiciaries 145 Public Policy toward Nongovernmental Organizations in Developing and Transition Economies 147 8 Studies by Subject Area Abstracts of Current Studies Poverty and Social Development The Impact of the Financial Crisis The key policy conclusion is that interventions that on the Farm Sector in Thailand influence farmgate prices or enhance the availability of credit will probably have limited impact on the poor. By This study identified and measured the short- and contrast, well-targeted social programs designed to fill long-term effects of the shocks from the East Asian gaps in the safety net will do more to alleviate rural financial crisis on the well-being and behavior of rural poverty. households in Thailand. The aim was to contribute to the The results were presented at a seminar in Bangkok design of effective interventions to mitigate the adverse on June 22, 2000, attended by senior officials and staff social impacts of the crisis and to a national consensus on of the Ministry of Agriculture and Cooperatives as well short- and medium-term measures for restoring equitable as representatives of the World Bank, the Mekong Envi- and sustainable rural growth. ronment and Resources Institute, Asia-Europe Meeting The study used two data sources: a detailed, (ASEM) affiliates, other donors, local nongovernmental nationwide, farm household survey conducted by organizations, universities, and other elements of civil the Ministry of Agriculture and Cooperatives in 1996, a society. All survey data from the research will be publicly year before the onset of the crisis, and a follow-up released. survey of a large subsample of the same households, Responsibility: Development Research Group, Rural Devel- conducted with the assistance of a ministry research opment-Gershon Feder (gfeder@worldbank.org), Jaime team. These panel data were used to analyze how Quizon, and Hanan Jacoby. With Tongroj Onchan, different indicators of shock are related to changes in Charles Mehl, and Anthony Zola, Mekong Environment household input expenditures, land use, investment, and Resources Institute, Thailand; and Fabrizio Bresciani, output, and consumption. The distribution of the effects University of Maryland. The ASEM Trust Fund of the crisis across poor and nonpoor rural households was contributed funding for the research. also examined. Completion date: August 2000. The main findings of the study are as follows: * There is little sign of massive urban to rural migra- Report tion, although there appears to have been a slowdown in Bresciani, Fabrizio, Gershon Feder, D. Gilligan, Hanan Jacoby, lbn- rural emigration. groj Onchan, and Jaime Quizon. Forthcoming. "Weathering the * Consumption and income fell much less in the Storm: The Impact of the East Asian Crisis on Farm House- central region than in the northeast or the north, which holds in Indonesia and Thailand." Wor/dBank Research Observer was hit particularly hard. * The poor have borne the brunt of the crisis, with their Technology and Poverty Alleviation expenditures and income (farm and off-farm) falling, both in real terms and relative to those of the rich. These This project looked at the effects of post-green revolu- findings are consistent with the fact that the poor derive tion technological change on poverty, assessing the most of their income from the labor market (which per- relative importance of its direct effects (on the income formed badly) rather than from farming (which generally or consumption of adopting households) and indirect performed well), while richer farmers derive most of effects (on food prices, wages, employment, and their income from farming. sectoral linkages). To do so, it estimated computable o No evidence of a credit crunch was found: general equilibrium models for archetypal economies outstanding loans did not decline substantially. representing poor countries. The models characterize sources of income to identify the real income effects of ping contracts to acquire seeds and chemical inputs. The food price declines and farm income improvements result- contracts provide insured credit: in the event of a crop fail- ing from technological advances. ure, the farmers pay nothing for the inputs they received The study found that in a typical African context, from the landlord. But calculations suggest that the price where the agricultural sector is large and most of the farmers pay for inputs through sharecropping is twice poor are smallholders, direct effects are very important. the market price on average. Although sharecropping Thus targeting technological change to poor farmers- provides informal insurance for poor farmers, the high price to their crops, farming systems, market failures, and insti- they pay may offset the benefits conferred by the new tutional gaps-is essential for reducing poverty. In Asia, technologies. One option would be to offer credit programs by contrast, where most of the poor are rural and landless, that include limited liability, thereby enabling farmers to indirect effects captured by the labor market are more purchase inputs in advance. important. So in this case targeting technological change In addition, analysis of data from a test of agricultural toward employment creation is fundamental for reducing knowledge included in the survey questionnaire showed poverty. In Latin America, where poverty is largely urban that poor farmers have significantly less knowledge of the and most of the land is in the hands of large farmers, most use of disease-resistant potato varieties and pesticides than real income gains for the rural poor come through indi- better-off farmers. And controlling for socioeconomic rect effects, mainly from falling food prices. In this case factors, the analysis found that farmers with lower test the main role of technological change may be in the scores were more likely to suffer high crop losses from El fields of large farmers. Nifio. These results underscore the need for agricultural The project also assessed the potential contributions extension programs to accompany the release of new of recent biotechnological advances to reducing poverty technologies. in the future. It identified the main features of agricul- The study's results were presented at a workshop tural biotechnology research that could have direct and organized by the World Bank in Malaysia on May 10-12, indirect effects on poverty, analyzed the patterns of gen- 1999; at the Bank's Summer Research Workshop on eration and ownership of Bacillus thuringiensis (Bt) patents, Poverty in Washington, D.C., on July 6-8, 1999; and at a and developed a simulation model of the determinants conference, The Shape of the Coming Agricultural of adoption of Bt seeds by small landholders. The Biotechnology Transformation: Strategic Investment and project found that agricultural biotechnology has great Policy Approaches from an Economic Perspective, at the promise for reducing poverty, through both direct and University of Rome on June 17-19,1999. The results also indirect effects. But the institutional requirements for served as input to the World Bank's World Development making this promise a reality are demanding. Thus suc- Report 2000/2001: Attacking Poverty (New York: Oxford cess lies less in the ability to progress in biological sciences University Press, 2000). And they are informing ongoing than in the ability to put in place the institutions for efforts by the World Bank's Biotechnology Task Force to generating, transferring, delivering, and adopting biotech- define guiding principles, strategies, and priorities in nological innovations favorable to poverty reduction. support of capacity building in agricultural biotechnology, Finally, the project conducted a household survey in to enhance its impact on poverty. the northern highlands of Peru to assess whether poor Responsibility: Development Research Group, Rural Devel- farmers were benefiting from modern potato varieties opment-Rinku Murgai (rmurgai@worldbank.org) and introduced in the region. The survey results highlight the Gershon Feder. With Oscar Ortiz and Rebecca importance of taking risk into account when assessing the Nelson, International Potato Center, Lima, Peru; Alain potential of genetically improved crops to reduce poverty. de Janvry and Erin McCormick, University of California Although poor farmers have access to credit, they are at Berkeley; and Paul Winters, University of New unwilling to take on the risk of a loan to purchase the seeds England, Armidale, Australia. of modern potato varieties. Instead, they use sharecrop- (omplelion date: September 2000. 12 Poverty and Social Development Reports ness. These findings shed light on how microcredit can deJanvry, Alain, Gregory Graff, Elisabeth Sadoulet, and David Zil- help advance women where they lack decisionmaking berman. 2000. "Agricultural Biotechnology and Poverty: How power in the family and the society. to Make the Promise a Reality." University of California at Responsibility: World Bank Institute, Economic Policy and Berkeley. Poverty Reduction Division, and Development Research McCormick, Erin. 2001. "Reducing Poverty in the Andes with Group, Rural Development-Shahidur R. Khandker Genetically Improved Potatoes: The Importance of Risk and (skhandker@worldbank.org); and World Bank Institute, Knowledge." University of California at Berkeley. Economic Policy and Poverty Reduction Division- Hussain Samad. With Mark Pitt, Brown University; and Does Microcredit Empower Women? Rita Afsar, Bangladesh Institute of Development A Study of Grameen Bank, BRAC, Studies. and the RD-1 2 Projects in Bangladesh Completion dote: December 2000. This research investigated whether microcredit Report programs in rural Bangladesh empower women, the main Pitt, Mark M., Shahidur R. Khandker, and Jennifer Cartwright. 2001. participants in these programs. Indicators were developed "Does Microcredit Empower Women? Evidence from for three concepts of women's empowerment- Bangladesh." World Bank, World Bank Institute, Economic economic, social, and political-through participatory Policy and Poverty Reduction Division; and Development and focus group research involving women, men, village Research Group, Washington, D.C. and urban elite, acaclemics, program organizers, and lead- ers of women's groups. How Do the Poor Cope with Widespread The research used both quantitative and qualitative Natural Disaster? A Study of the 1 998 methods based on household and community survey Flood of Rural Bangladesh data. Questionnaires based on the indicators of empow- erment were administered to women in both program and Large shocks such as floods, cyclones, and earthquakes nonprogram households in program villages and in have dramatic effects on the well-being of rural target and nontarget households in nonprogram villages. households, particularly the poor. Bangladesh experi- The research used participatory and other qualitative enced severe human suffering in 1998, after one of approaches to study similar empowerment issues in the worst floods of the century, when two-thirds of selected households in the same study villages. the country remained under water for more than four Answers to the questionnaires were processed with months. appropriate weights to develop indexes of social, This flood provided a unique opportunity to study how economic, political, and overall empowerment. The the poor respond to catastrophic events and how quantitative analysis controlled for the unobserved governmental and other agencies could improve the empowerment of women before program participation; coping mechanisms of the poor. To do so, this project otherwise, it would be unclear whether participating in collected detailed information at the household and a microcredit program empowers women-or whether village level on the flood damage, the coping mechanisms only empowered women participate. adopted by households, and the speed of the recovery. Key findings suggest that microcredit targeted to This survey-which collected information on assets women has a positive influence on their power to before and after the flood, distress sales of assets, credit purchase household goods; their ability to make decisions market activity, interhousehold transfers, and tempo- about fertility, children's education, and general house- rary and permanent migration-was administered as part hold welfare; their mobility; their access to household of a household survey conducted under the research assets and savings; and their social and political aware- project Long-Term Impacts of Microcredit Programs: Poverty and Social Development 13 A Study of the Grameen Bank and Other Programs in Policy and Poverty Reduction Division; and Development Bangladesh (see the abstract in this volume). Research Group, Washington, D.C. Draft. To assess the vulnerability of the poor, the study used the household data to measure the effect of the flood on Long-Term Impacts of Microcredit Programs: wages, distress sales of assets, and starvation. On aver- A Study of the Grameen Bank and Other age, the male wage fell by 10 percent in the agricultural Programs in Bangladesh sector and 22 percent in the nonagricultural sector. But the labor market rebounded quickly after the flood, with An earlier study (Credit Programs for the Poor: wages surpassing those before the flood. Only 10 percent Household and Intrahousehold Impacts and Program of affected households made distress sales (selling land Sustainability) based on a household and community or other assets, or making advance sales of labor or crops) survey in Bangladesh in 1991-92 found that microcredit during the flood. About a third of the households in programs have a substantial effect in reducing poverty flood-affected villages skipped at least one meal a day and that credit given to women has a substantially larger during the crisis. effect than credit given to men. But how sustainable Interventions by the government and by nongovern- are microcredit programs, and how sustainable are their mental organizations and other informal sources helped benefits? Building on the earlier study, this research the poor cope. The assistance was well targeted; house- project explored these and related issues by resurveying holds with low per capita income and net worth received the households and communities surveyed in 1991-92 most of it. Among those receiving some help, some 6 and constructing a panel data set on borrowing from all percent received help from relatives and another 6 microfinance programs, including Grameen Bank, the percent help from other people, while 78 percent received Bangladesh Rural Advancement Committee (BRAC), help from institutional sources. Among the institutional other nongovernmental organizations, and government sources, nongovernmental organizations accounted for 12 programs. percent of the monetary help received by households in The follow-up survey revisited all 1,798 households, all villages. Preflood assets helped households cope with in 72 program villages and 15 control villages, during the calamity, reducing distress sales and starvation. Micro- F ebruary-September 1999. In addition, it added 17 new credit also helped, increasing household net worth and villages to allow comparison of program impacts between reducing the probability of being poor. old and new program villages. And in each program Preliminary results of the analysis were presented in village a few new participants were interviewed to Bangladesh to local staff of the World Bank field office allow a comparison of impacts between old and new in Dhaka. participants. Responsibility: World Bank Institute, Economic Policy and The study addressed these questions: Do the Poverty Reduction Division, and Development Research estimated effects of microcredit programs remain over Group, Rural Development-Shahidur R. Khandker time? Are there diseconomies of scale in villages as (skhandker@worldbank.org); and World Bank Institute, microcredit programs expand? Do microcredit programs Economic Policy and Poverty Reduction Division- have any spillover effects? Do they have noncredit Hussain Samad. With Mark Pitt, Brown University; and effects? And does the group play an essential part in M. Abdul Latif and Binayak Sen, Bangladesh Institute group-based microcredit? of Development Studies. Preliminary findings show that microcredit programs (ompletion date: December 2000. have significant positive effects on household welfare, as measured by such indicators as net worth and per capita Report consumption. These effects are much larger for female Khandker, Shahidur R., Mark M. Pitt, and M. Abdul Latif. 2001. borrowing than for male borrowing. In contrast, the "How Do the Poor Cope with Natural Disaster? Evidence expected impact of microfinance on contraceptive use and from Bangladesh." World Bank, World Bank Institute, Economic fertility is more pronounced for male borrowing than for 1 4 Poverty and Social Development female borrowing. Microfinance increases the labor icy analysis and design still is not widely understood, nor supply of both men and women, supporting the view that have the lessons for development been fully integrated these programs generate employment for the underem- by donors and national policymakers. And despite con- ployed and unemployed of both genders. siderable advances in gender equality in recent decades, Results also show that the effects of credit change over gender discrimination remains pervasive. time. The marginal return to consumption, for example, This project brought together existing and new is higher for women's past borrowing than for their research from several social science disciplines to exam- current borrowing. This may reflect diminishing returns ine the costs of persistent gender disparities to well- to consumption from microfinance as borrowers being and to countries' development prospects, accumulate assets through higher incomes. investigate how formal and informal institutions shape The study found evidence of village diseconomies, gender roles and relations and how household decisions or market saturation for microfinance borrowers. In and behaviors reproduce those roles, analyze the rela- particular, the effects of male borrowing are smaller in tionship between economic development and gender villages with higher aggregate microfinance borrowing. equality, and consider the role of public policy and civic But this is not the case for female borrowing, suggesting action in promoting gender equality. The findings are that men and women pursue different activities. reported in a World Bank Policy Research Report, Engen- Noncredit inputs play an important role for both male dering Development-Through Gender Equality in Rights, and female borrowers. In fact, they seem to be more Resources, and Voice (New York: Oxford University Press, important than credit in influencing some behavioral 2001). outcomes. The report strengthens the analytical and empirical Preliminary findings were disseminated at. the Asia underpinnings of promoting gender equality and, in Poverty Forum organized by the Asian Development doing so, clarifies the value added of bringing a gender Bank in February 2001 and at a World Bank seminar orga- perspective to the analysis and design of development nized by the Development Research Group in April 2001. policies and projects. Based on the evidence, it argues Responsibility: World Bank Institute, Economic Policy and that development policies that do not take gender roles Poverty Reduction Division-Shahidur R. Khandker and relations into account and that do not address gen- (skhandker@worldbank.org) and Hussain Samad; and der disparities will have limited effectiveness. Among its Development Research Group, Rural Development- key conclusions: Gershon Feder. With Mark Pitt, Brown University; and * Despite progress, gender inequalities are still per- Binayak Sen and M. Abdul Latif, Bangladesh Institute vasive worldwide and exist across many dimensions of of Development Studies. life. In no developing region do women experience Completion date: December 2000. equality with men in legal, social, and economic rights. Gender gaps remain widespread in access to and control Report of resources, in economic participation, in power, and in Khandker, Shahidur R., and Mark M. Pitt. 2001. "The Impact of political voice. These gaps are wider in poorer countries Group-Based Credit on Poor Households: An Analysis of Panel and in the poorest groups within countries. Data from Bangladesh." World Bank, World Bank Institute, 0 While girls and women are most disadvantaged by Economic Policy and Poverty Reduction Division, Washington, gender disparities, these inequalities reduce the well- D.C. Draft. being of all people. Societies that discriminate on the basis of gender pay a significant price-in more poverty, slower Policy Research Report on Gender and Development economic growth, weaker governance, and a lower qual- ity of life. Where gender inequalities impose high human Over the past several decades gender issues have gained costs and constrain countries' development prospects, greater prominence in the debate on development. Yet there is a strong argument for a state role in promoting the importance of bringing a gender perspective to pol- gender equality. Poverty and Social Development is * Promoting gender equality in basic rights and Reports economic development is central to a long-term strategy Cox-Edwards, Alejandra. 2001. "Social Security Reform and for achieving equality. Societies that establish an insti- Women's Pensions." Policy Research Report on Gender and tutional environment supportive of gender equality and Development Working Paper 17. World Bank, Development that promote economic development are likely to be Research Group, Washington, D.C. more effective in reducing gender disparities than those Das Gupta, Monica, Sunhwa Lee, Patricia Uberoi, Danning Wang, that focus on growth-or on rights-alone. Lihong Wang, and Xiaodan Zhang. 2000. "State Policies and * But even an approach of equal rights and economic Women's Autonomy in China, the Republic of Korea, and development may not lead quickly to major gains. Insti- India, 1950-2000: Lessons from Contrasting Experiences." tutional change and economic growth and development Policy Research Report on Gender and Development Work- can be slow and uneven. Active measures are needed to ing Paper 16. World Bank, Development Research Group, redress persistent gender disparities in the short to Washington, D.C. medium term. Dollar, David, and Roberta Gatti. 1999. "Gender Inequality, Income, * The evidence argues for a three-part strategy to and Growth: Are Good Times Good for Women?" Policy Research promote gender equality: reforming legal, economic, Report on Gender and Development Working Paper 1. World and social institutions to establish a foundation of equal Bank, Development Research Group, Washington, D.C. rights and equal opportunities for women and men; Dollar, David, Raymond Fisman, and Roberta Gatti. 1999. "Are implementing policies for sustained economic growth and Women Really the 'Fairer' Sex? Corruption and Women in development; and initiating active policy measures to Government." Policy Research Report on Gender and Devel- reduce gender disparities in the command of resources opment Working Paper 4. World Bank, Development Research and political voice. Group, Washington, D.C. Drafts of the report were broadly reviewed and dis- Filmer, Deon. 1999. "The Structure of Social Disparities in Edu- cussed inside and outside the World Bank. Discussions cation: Gender and Wealth." Policy Research Report on Gen- were held on the Web through the Global Development der and Development Working Paper S. World Bank, Forum and in meetings with staff of governments, uni- Development Research Group, Washington, D.C. versities, donor agencies, and nongovernmental organi- llahi, Nadeem. 2000. "The Intra-Household Allocation of Time zations, including a panel session at the Beijing Plus-5 and Tasks: What Have We Learnt from the Empirical Litera- meetings at the United Nations in New York in June 2000. ture?" Policy Research Report on Gender and Development The report's authors have presented the findings in sev- Working Paper 13. World Bank, Development Research Group, eral countries. The report's executive summary and most Washington, D.C. of the background papers (see below) are available on the Klasen, Stephan. 1999. "Does Gender Inequality Reduce Growth Web at http://www.worldbank.org/gender/prr. and Development? Evidence from Cross-Country Regres- Resporsibility: Development Research Group, Public Services sions." Policy Research Report on Gender and Development for Human Development-Elizabeth M. King Working Paper 7. World Bank, Development Research Group, (eking@worldbank.org); Poverty Reduction and Eco- Washington, D.C. nomic Management Network, Gender Group-Andrew Lampietti, Julian A., and Linda Stalker. 2000. "Consumption D. Mason and Karen Mason; Operations Evaluation Expenditure and Female Poverty: A Review of the Evidence." Department-Ananya Basu; World Development Report Policy Research Report on Gender and Development Work- Office-Claudio Montenegro; and Global Development ing Paper 11. World Bank, Development Research Group, Network-Lyn Squire. With Tai Lui Tan, Lihong Wang, Washington, D.C. Cristina Estrada, Owen Haaga, and Branko Jovanovic. Lampietti, Julian A., Christine Poulos, Maureen L. Cropper, Haile The government of the Netherlands and the Norwegian Mitiku, and Dale Whittington. 1999. "Gender and Preferences Royal Ministry of Foreign Affairs contributed funding for for Malaria Prevention in Tigray, Ethiopia." Policy Research the research. Report on Gender and Development Working Paper 3. World (ompletion dote: January 2001. Bank, Development Research Group, Washington, D.C. 16 Poverty and Sociol Development Lokshin, Michael M. 2000. "Effects of Child Care Prices on ment Working Paper 6. World Bank, Development Research Women's Labor Force Participation in Russia." Policy Research Group, Washington, D.C. Report on Gender and Development Working Paper 10. World World Bank. 2001. EngenderingDevelopment-Through GenderEqual- Bank, Development Research Group, Washington, D.C. ity in Rights, Resources, and Voice. World Bank Policy Research Lokshin, Michael M., Elena Glinskaya, and Marito Garcia. 2000. Report. New York: Oxford University Press. "The Effect of Early Childhood Development Programs on Women's Labor Force Participation and Older Children's The Impact of Prices, Taxes, Subsidies, Schooling in Kenya." Policy Research Report on Gender and and Stipends on Poverty Development Working Paper 15. World Bank, Development Research Group, Washington, D.C. Policies affecting the prices of goods have important Long, Lynellyn D., Le Ngoc Hung, Allison Truitt, Le Thi Phuong effects on the poor. There are many such policies, imple- Mai, and Dang Nguyen Anh. 2000. "Changing Gender Rela- mented through such instruments as import tariffs; sales tions in Vietnam's Post DoiMoi Era." Policy Research Report and other indirect taxes; price subsidies for, say, food, on Gender and Development Working Paper 14. World Bank, energy, or transport; and stipends, for example, for increas- Development Research Group, Washington, D.C. ing retention rates in schools. Newman, Constance. 2001. "Gender, Time Use, and Change: The tools available for evaluating the poverty impact Impacts of Agricultural Export Employment in Ecuador." Pol- of changes in such policies have limits. They rely on icy Research Report on Gender and Development Working particular measures of poverty (such as those of the Paper 18. World Bank, Development Research Group, Wash- Foster-Greer-Thorbecke class), particular poverty lines, ington, D.C. and particular indicators of well-being (such as per capita Paul-Majumder, Pratima, and Anwara Begum.. 2000. ".The Gen- income) without taking into account differences in need der Imbalances in the Export-Oriented Garment Industry in (as measured, for example, through alternative equiva- Bangladesh." Policy Research Report on Gender.and Devel- lence scales). That is, the tools do not provide tests for opment Working Paper 12. World Bank, Development Research the robustness of the analysis-and thus the policy Group, Washington, D.C. conclusions-to alternative value judgments relating to Quisumbing, Agnes R., and John A. Maluccio. 1999. "Intra- the poverty measure, poverty line, and indicator of well- household Allocation and Gender Relations: New Empirical Evi- being. With different assumptions and methodologies, dence." Policy Research Report on Gender and Development the ranking of alternative pricing policies might be Working Paper 2. World Bank, Development Research Group, reversed. Without tests for robustness, an analyst sug- Washington, D.C. gesting policy changes to benefit the poor might end up Rama, Martfn. 2001. "The Gender Implications of Public Sector proposing a regressive policy. Downsizing: The Reform Program of Vietnam." Policy Research The goal of this research project was to provide new Report on Gender and Development Working Paper 19. World analytical tools that can contribute to the design of robust Bank, Development Research Group, Washington, D.C. policies relating to prices, taxes, subsidies, stipends, and Rice, Patricia. 1999. "Gender Earnings Differentials: The Euro- related instruments. These new tools were developed pean Experience." Policy Research Report on Gender and theoretically and applied empirically to household-level Development Working Paper 8. World Bank, Development data for several Latin American countries, including Research Group, Washington, D.C. Guatemala (income and fuel poverty) and Honduras Sen, Samita. 2000. "Towards a Feminist Politics? The Indian (income poverty and housing). Women's Movement in Historical Perspective." Policy Research The first set of tools starts with consumption domi- Report on Gender and Development Working Pap;r 9. World nance curves. These curves can be traced to test whether Bank, Development Research Group, Washington, D.C. a particular value added tax reform involving several van der Meulen Rodgers, Yana. 1999. "Protecting Women and commodities will reduce or increase poverty and whether Promoting Equality in the Labor Market: Theory and the result depends on the choice of poverty line, poverty Evidence." Policy Research Report on Gender and Develop- measure, or both. The research extended the concept of Poverty and Soioal Development 17 the consumption dominance curves to the analysis of cash A New Analytical Framework transfer and other government programs, also using for Evaluating Social Programs graphical techniques. The second set of tools focuses on decomposing the impact of transfer programs on poverty World Bank staff constantly confront the difficult prob- into a targeting component (who gets the transfer) and lem of evaluating social programs and policies. This an allocation component (how much different benefi- research project developed a new analytical framework ciaries get). While the first set deals with stochastic dom- for performing such evaluations. The framework takes inance, the second set deals with sequential stochastic into account not only flexible distributional weights for dominance, in which poverty comparisons and impact translating individual welfare gains into aggregate social analysis are conducted without specifying the equivalence gains but also the targeting performance of programs scale used to compare households of different sizes. and the allocation of benefits among program participants. Papers from the project have been presented at sev- The use of distributional weights is rarely made eral conferences and seminars, including a workshop at explicit in Bank work or in the broader literature, in part Tinbergen Institute, Amsterdam (March 2001), a World because the implications for policy may be disturbing. Bank seminar (May 2001), and conferences organized by Yet the fact that policy reforms are evaluated using Societe Canadienne de Science Economique (May 2001), poverty measures implies that such distributional weights the Canadian Economic Association (June 2001), the are being used. One problem with distributional weights Canadian Public Economics Study Group (June 2001), based on standard poverty measures of the Foster-Greer- the University of York, and GREQAM, Marseille, France. Thorbecke class is that they place no weight at all on the Responsibility: Latin America and the Caribbean Region, welfare of the nonpoor. The framework developed in this Poverty Sector Unit-Quentin Wodon (qwodon project provides an alternative in which the gains to all @worldbank.org). With Paul Makdissi, .University of members of society are taken into account, though with Sherbrooke, Canada; Jean-Yves Duclos, University of different weights. Laval, Canada; Jean-Philippe Tre; and Shlomo Yitzhaki, Starting from a well-known social welfare function, two Hebrew University, Israel. summary parameters are estimated for each program or Completion date: June 2001. policy to assess its impact on social welfare. The parameters are flexible enough to take into account Reports weighting schemes that place varying degrees of empha- Duclos, Jean-Yves, Paul Makdissi, and Quentin Wodon. 2001. sis on poor members of society. The summary parame- "Assessing the Robustness of a Poverty-Reducing Indirect ters consist of a growth term (the mean benefit of the Tax Reform." Draft. program) and a distributional term (who is covered by the .2001. "Socially Efficient Marginal Tax Reforms." Draft. program and in what amount). The distributional term Foster, V, Jean-PhilippcT re, and Quentin Wodon. 2001. "Energy can be decomposed into several components. The first Consumption and Income: An Inverted-U at the Household measures the targeting performance of the program (who Level?" Draft. is participating and who is not). The second measures the .2001. "Energy Prices, Energy Efficiency, and Fuel Poverry." impact of the allocation rules for distributing the bene- Makdissi, Paul, and Quentin Wodon. 2001. "Consumption fits among program participants. The third takes into Dominance Curves: Testing for the Impact of Indirect Tax account the fact that welfare measures for some Reforms on Poverry." nonparticipants are in the range of those for participants, .2001. "Fuel Poverty and Access to Electricity: Compar- which means that if targeting is not perfect, a loss will ing Households When They Differ in Needs." be incurred. .2001. "Income Transfers, Family Size, and Poverty Order- The framework has been applied to social programs ings." Draft. and policies in several Latin American countries using - . 2001. "Migration, Poverty, and Housing: Welfare Com- household-level data. The results show how the perfor- parisons Using Sequential Stochastic Dominance." mance of programs depends on both targeting and 18 Poverty and Sociol Development allocation rules and how the choice of programs can be Wodon, Quentin, and Shlomo Yitzhaki. 2000. "Evaluating the given a welfare interpretation by using distributional Impact of Government Programs on Social Welfare: The weights. In Mexico, for example, the research found Role of Targeting and the Allocation Rules among Program that a large program of cash transfers to rural farmers has Beneficiaries." little impact on inequality. Although the program is well - . 2001. "The Effect of Using Grouped Data on the Gini targeted, the allocation procedure favors beneficiaries with Income Elasticity Estimator." Draft. large amounts of land. Another program in Mexico- - . 2001. "Inequality and Social Welfare." In World Bank, Programa de Educaci6n, Salud y Alimentaci6n, or Poverty Reduction Strategy Sourcebook CD-ROM. Poverty Reduc- Progresa, which provides cash transfers to poor families tion and Economic Management Network, Washington, D.C. as long as members visit health centers and the - . 2001. "The Robustness of Regression Coefficients: A children attend school-was found to be much more Geometrical Exposition." effective. In Chile, where the research looked at a series - . Forthcoming. Evaluating Development Programs and Poli- of programs that target benefits using a common means- cies: A New Framework. testing mechanism, it found that utility subsidies were much less efficient in reducing inequality and improv- Private Transfers in a Cross-Section ing welfare than such programs as family allowances, of Developing Countries noncontributory pensions, and subsidized child care. Poor performance of utility subsidies was.also qbserved Comparisons of private transfers across countries can in Honduras for electricity, where the eligibility rules for provide insights into the relationship between country- lifeline consumption do not prevent errors of inclusion. specific institutions and private transfers (those from The project made another important methodological adult children to their elderly parents, for example, contribution by showing that with little risk of estima- might be less important in countries with more gener- tion bias, the techniques can be applied using grouped ous public pensions). But most studies of private trans- data rather than the unit-level data from household fers examine a single country. This study departs from surveys. Simulation tools for use with grouped data are the single-country case study approach by building a being created to ease the application of the techniques. database on several developing countries in different A survey of the techniques is included in the Poverty regions. Reduction Strategy Sourcebook CD-ROM (World Bank, The study begins with a simple question: How Poverty Reduction and Economic Management Net- similar are patterns in private transfers around the world? work, Washington, D.C., 2001). It then highlights differences in private transfer behav- Findings have been presented at Hebrew Univer- ior, particularly with respect to age patterns. Finally, it sity, in April 2001, and at World Bank and International draws inferences about the relationship between public Monetary Fund seminars. and private transfers across countries. Responsibility: Latin America and the Caribbean Region, The main data sources are Living Standards Mea- Poverty Sector Unit-Quentin Wodon (qwodon surement Study (LSMS) surveys in Albania, Bulgaria, @worldbank.org). With Shlomo Yitzhaki, Hebrew Uni- Jamaica, Kazakhstan, the Kyrgyz Republic, Nepal, versity, Israel. Nicaragua, Panama, Peru, the Russian Federation, and Completion date: June 2001. Vietnam. The LSMS surveys' extensive coverage provides a unique opportunity for a consistent, cross- Reports country econometric analysis of the determinants and Castro-Fernandez, Rodrigo, and Quentin Wodon. 2001. "Protect- magnitude of private transfers. These surveys overcome ing the Unemployed in Chile: From State Assistance to Indi- one of the main problems in the literature-the difficulty vidual Insurance?" Background paper for Chile: Poavrty in a of defining interhousehold transfers. The questionnaire High-Growth Economy. World Bank, Latin America and the design and data quality controls allow comparable defi- Caribbean Region, Poverty Sector Unit, Washington, D.C. nitions of private transfers and income. Poverty and Social Development 19 Some common threads emerge across the countries in Responsibility: Latin America and the Caribbean Region, the sample: Private transfers appear to flow from high- Poverty Sector Unit-Quentin Wodon (qwodon income to low-income households. They increase the @worldbank.org). With Amos Golan, American Univer- share of income accruing to the poor in the national per sity; and Corinne Siaens. capita distribution. And they tend to be targeted to Completion date: October 2001. female-headed households and to households in which a member suffered from a health shock. Using Semi-Parametric Methods for the Evaluation Profiles of age and private transfer receipts tend to be of Social Programs and Policies U-shaped for all countries, but with distinct patterns according to whether the transfers are targeted to young Econometric work to evaluate the impact of social households (as in Kazakhstan or the Kyrgyz Republic) programs and policies typically assumes that the impact or to older ones (as in Panama or Vietnam). The is the same throughout the distribution of the indicator preliminary results suggest an inverse relationship under review-for example, throughout the distribu- between private and public assistance to the elderly. tion of income. This assumption stems from the fact That inverse relationship holds when a more general that in a traditional regression setting, parametric meth- definition of private help to the elderly is used, one that ods of estimation yield one parameter estimate for the encompasses private interhousehold transfers received impact, whether the program or policy is captured in or co-residence. the data through a continuous or a categorical (for exam- The preliminary results were presented in the Devel- ple, dichotomous) variable. Even when analysts use opment Research Group's seminar series on poverty, interaction effects in their specification, which helps in household economics, and rural development. enriching the analysis, the estimates of the impact are gen- Responsibility: Development Research Group, Poverty erally assumed to be the same for all those with the Team-Emanuela Galasso (egalasso@worldbank.org); given interaction. But in reality this need not be the and South Asia Region, Education Sector Unit- case. Some households may benefit (or suffer) more Emmanuel Jimenez. With Don Cox, Boston College. than others from specific programs or policies. To avoid Completion dote: September 2001. imposing strong assumptions in the estimation of impacts, analysts can rely on semi-parametric methods. The Impact of Growth on Poverty in Latin America This research aims to better equip World Bank staff to use semi-parametric methods by providing the appro- This research is using new econometric methods (based priate statistical programs and showing how the methods on maximum entropy) adapted to the available data to can be applied to a wide range of issues relating to social analyze the relationships between poverty, growth, and programs and policies. The research focuses on devel- inequality in Latin America and in selected countries oping and applying semi-parametric methods for cases within Latin America. The aim is to obtain better esti- in which the program or policy variable is dichotomous. mates of the elasticities of poverty reduction to growth, It builds on earlier work by analysts who applied semi- taking into account the impact of growth on inequality, parametric methods to evaluate the impact of the min- and to analyze why elasticities differ between countries imum wage and unionization on the distribution of or between sectors (such as urban and rural, or agricul- earnings in the United States. ture, manufacturing, and services) within countries and In intuitive terms, the advantage of this semi-para- within the region. metric approach is that it helps in analyzing the effect of The results should be useful both for World Bank staff different factors (such as institutional features, social and for government staff in Bank client countries, allow- programs, and individual attributes) on the full distrib- ing them to be more specific and better prepared when ution of earnings or on the distribution of any other setting poverty reduction targets and evaluating progress variable to which the method is applied. This effect is in reducing poverty. estimated by applying kernel density methods to the 20 Poverty and Social Development appropriately weighted samples. The procedure pro- This research project studies Vietnam's recent success vides a visual representation of where in the density in promoting economic growth, reducing poverty, and of earnings (or other variables) the factors exert the raising living standards with the aim of providing policy greatest impact. lessons for other low-income countries. Specifically, it The method makes it possible to differentiate impacts addresses the following questions: according to where the individuals are located in the * What policies led to Vietnam's high rate of eco- distribution of wages (or another variable). It also allows nomic growth and allowed it to continue that growth the impact of changes in a factor over time to differ for (though at a lower rate) during the recent economic different individuals. The same method can be used to crises in East Asia and elsewhere? compare the impact of policies or factors that differ * Which Vietnamese households experienced growth between two sectors (say, urban and rural areas) rather in their incomes and expenditures, and which compo- than two periods. And it can be used to analyze the nents of their incomes increased most quickly? What pub- impact of factors or policies on nonmonetary indicators lic policies brought about these changes in household rather than wages or household per capita income. incomes? Responsibility: Latin America and the Caribbean Region, * How have increases in income affected other aspects Poverty Sector Unit-Quentin Wodon (qwodon of living standards in Vietnam, such as child nutrition, @worldbank.org). With Diego Angel; Shlomo Yitzhaki, school enrollment, adult health, fertility, child labor, and Hebrew University, Israel; and Rodrigo Castro- infant and child mortality? What has played a more Fernandez. important role in these socioeconomic outcomes-income (ompletion date: October 2001. or the more direct effects of such public policies as social safety nets? Economic Growth and Household Welfare: The research addresses the first question by using Policy Lessons from Vietnam macroeconomic data and cross-country data analysis, and the second and third by using data from the 1992-93 Most economists and other social scientists would agree and 1997-98 Vietnam Living Standards Surveys. that economic growth is essential for improving Preliminary results were presented to donor household welfare and that the form of that growth has,, agencies and Vietnamese researchers and policymakers important implications for poverty reduction. Yet this con- at a workshop in Hanoi in May 2001. sensus still leaves many questions unanswered. Two Responsibility: Development Research Group, Macroeco- stand out: What policies promote broad-based economic nomics and Growth-David Dollar (ddollar@worldbank. growth? And what policies should countries adopt to org), and Public Services for Human Development- ensure that economic growth quickly reduces poverty- Dominique van de Walle; East Asia and Pacific Region, both raising income and consumption and improving Poverty Reduction and Economic Management Sector other indicators of living standards? These questions Unit-Nisha Agrawal and Nguyet Nguyen Nga; South are difficult ones, and more empirical evidence is needed Asia Region, Human Development Sector Unit- to provide recommendations sufficiently precise to guide Emmanuel Jimenez; and Human Development Net- policymakers. work, Health, Nutrition, and Population Team-Adam Examining countries unsuccessful in increasing growth Wagstaff. With Robert Baulch; Paul Glewwe, University and reducing poverty can provide lessons on what not to of Minnesota; Sarah Bales, Dwayne Benjamin, and Loren do, but the most useful information comes from Brandt, University of Toronto; Donald Cox, Boston countries that have been successful in both. A recent College; John Gallup and Dominique Haughton, Bent- example of success is Vietnam. Growth in Vietnam has ley College; Jonathan Haughton, Suffolk University; averaged about 8 percent a year since around 1988, while Nguyen Phong and Pravin Trivedi, Indiana University; the incidence of poverty declined from about 75 percent and Wim Vijverberg, University of Texas at Dallas. in 1988 to 55 percent in 1993 and 37 percent in 1998. Completion date: December 2001. Poverty and Social Development 21 Reports received differs substantially depending on the house- Baulch, Robert, Truong Thi Kim Chuyen, Dominique Haughton, hold's resources and the type of transfer. For households and Jonathan Haughton. "Ethnic Minority Development in poor in financial and human resources (at the 10th Vietnam: A Socioeconomic Perspective." Draft. percentile of the resource index), most assistance comes Benjamin, Dwayne, and Loren Brandt. "Agriculture and Income from public sources or through credit. In contrast, Distribution in Vietnam during the Reform Period." Draft. relatively wealthy households (at the 90th percentile) Dollar, David. "Reform, Growth, and Poverty in Vietnam." Draft. receive substantial private transfers but little public Glewwe, Paul, and Hanan Jacoby. "Economic Growth and the assistance and are not forced to borrow. Demand for Education: Is There a Wealth Effect?" Draft. These findings have two implications for policies Glewwe, Paul, and Nguyen Phong. "Economic Mobility in aimed at mitigating the impact of adult deaths in coun- Vietnam." Draft. tries heavily affected by the AIDS epidemic. First, Glewwe, Paul, Stephanie Koch, and Bui Linh Nguyen. "Child targeting public transfers to the most vulnerable house- Nutrition and Economic Growth in Vietnam in the 1990s." holds-those with the smallest endowments of financial Draft. and human capital-will greatly improve the efficiency Minot, Nicholas, and Robert Baulch. "The Spatial Distribution of of assistance. And second, the reliance of the poorest Poverty in Vietnam and the Potential for Targeting." Draft. households on credit suggests that expanding micro- Trivedi, Pravin. "Patterns of Health Care Utilization in Vietnam: credit programs might help poor as well as less poor Analysis of 1997-98 VLSS Data." Draft. households. van de Walle, Dominique. "The Static and Dynamic Incidence The findings were presented at conferences in of Vietnam's Public Safety Net." Draft. Durban, South Africa, and in York, England. Vijverberg, Wim, and Jonathan Haughton. "Household Enter- Responsibility: Development Research Group, Infrastruc- prises in Vietnam: Survival, Growth, and Living Standards." ture and Environment-Mead Over (meadover Draft. @worldbank.org) and Mattias Lundberg. With Phare Mujinja, University of Dares Salaam; and Kathleen Bee- Household Coping with AIDS in Tanzania gle, Rand Corporation. The Joint United Nations Pro- gramme on HIV/AIDS (UNAIDS) is contributing Evidence from earlier research on the impact of AIDS funding for the research. in Kagera, Tanzania, suggests that less poor rural house- Completion date: December 2001. holds are able to cope economically with the short-term impact of an adult death. Less is known, however, about Report how households cope. To shed light on this question, this Lundberg, MaEtias, Mead Over, and Phare Mujinja. 2000. "Sources research project analyzed household receipts of transfers of Financial Assistance for Households Suffering an Adult after an adult death in Kagera. The analysis shows that Death in Kagera, Tanzania." South African Jour-nalofEconom- households receive significantly more transfers after a ics 68(5): 856-87. (Also issued as Policy Research Working death than before, but the type of transfers received Paper 2508, World Bank, Development Research Group, Wash- varies by type of household. ington, D.C., 2001.) The study first constructed an index of household resources from the first principal component of house- Informal Safety Nets in Nicaragua hold assets, average body mass index, household size, and the sex, age, and education of the household head. It then How did the macroeconomic shocks experienced by estimated the impact of an adult death in the household Nicaraguan households as a result of Hurricane Mitch in on the receipt of each of three types of transfers, inter- 1998 affect informal safety nets? This research is acting the household resource index with whether a examining how private consumption insurance mecha- death occurred in the household and the time since the nisms within communities and across different groups death. The results show that the pattern of transfers (including the extended family) shifted as a result ofthe 22 Poverty and Social Development income shocks induced by Hurricane Mitch. It is also makers and external partners involved in designing social analyzing the extent to which the relative control over protection programs and national poverty reduction strate- resources within the household (for example, by men or gies. The data analysts will make their statistical pro- women) affected human development outcomes and grams available within the World Bank and provide the economic behavior of household members, and how seminars to other data analysts interested in econometric resource control within the household shifted as a result analysis of consumption insurance arrangements and intra- of the economic crisis. household resource allocation in other countries. The research uses both quantitative and qualitative Responsibility: Poverty Reduction and Economic Manage- methodologies. The quantitative research consists of ment Network, Poverty Reduction Strategy Initiative- econometric analysis of a panel data set covering periods Jeni G. Klugman (jklugman@worldbank.org) and Kate before and after Hurricane Mitch in Nicaragua (from Withers; and Latin America and the Caribbean Region, Living Standards Measurement Study household surveys Poverty Sector Unit-Florencia Castro-Leal. With Gabriel in 1998, 1999, and 2001). This analysis is investigating Arturo Basaluzzo, University of Pennsylvania; and Diana the extent and strength of informal consumption insur- I. Kruger, University of Maryland. ance mechanisms and exploring the determinants of Completion date: December 2001. household demand for goods and services, with direct links to poverty and human development outcomes. Social Exclusion and Poverty Qualitative research by national social science researchers explores the anthropological and cultural basis of risk shar- This project examines how social exclusion (including ing practices that influence private transfer arrange- gender exclusion) prevents people from participating in ments among different population subgroups, and social and benefiting from the opportunities provided by human norms relating to resource allocation within the house- development programs and economic growth. It looks at hold and the extended family. These qualitative data how social institutions interact with formal institutions sources inform the econometric specifications and the to shape development outcomes in poor communities. interpretations of statistical results. It investigates how policies can be reshaped to reduce The research will provide concrete results on how social exclusion and increase synergies between informal economic and poverty outcomes are influenced not only and formal institutions. And it looks at how poor people's by shocks and economic institutions, but also by cultural health can be improved by reducing their exposure to customs, beliefs, and institutions that govern resource communicable diseases. allocation across households and within the family. Several studies served as background work for the In particular, the research will provide a better under- World Bank's Wor/dDevelopmentReport2OOO/2OOl:Attack- standing of which groups in Nicaragua face the greatest risk ing Poverty (New York: Oxford University Press, 2000): of being excluded from informal and public transfer a theoretical analysis of social exclusion, a cross-country programs, as well as a better understanding of the study of how land reform and other measures to reduce cultural institutions that determine how shocks affect the agrarian hierarchies interact with changes in local admin- functioning of informal safety nets. It will also aid in under- istrative institutions to generate state-community syn- standing how the control of financial resources within the ergies for development, the ramifications of gender-based household-and cultural norms that influence intrafamily exclusion for poverty and development, and the links resource control-influence the poverty and health out- between social exclusion, poverty, and health. Two comes of vulnerable members. The findings will help to additional studies are under way-an analysis of social guide the design of effective social protection measures for exclusion in the context of urban poverty and a study on Nicaragua as well as labor market, health care, and consumer ways to enhance public health programs for communi- subsidy policies and programs. cable disease control through improved governance and In addition to an academic paper, the research will community participation, focusing on diseases affecting produce a policy paper intended for national policy- poor people in India. Poverty and Social Development 23 The research has drawn on extensive analysis of Washington, D.C. (Also issued as Policy Research Working secondary source material, existing data sets, and freshly Paper 2497, World Bank, Development Research Group, Wash- collected field data. It has been conducted in close ington, D.C., 2001.) collaboration with major universities, government Prasad, Kameshwar, Paolo Belli, and Monica Das Gupta. 1999. agencies, and key research institutions in developing "Social Exclusion, Health, and Poverty." Background paper for countries. Results have been widely disseminated, with World Development Report 2000/2001: Attacking Poverty. World some incorporated in WorldDevelopment Report 2000/2001: Bank, Washington, D.C. Attacking Poverty. Woolcock, Michael, and Hilary Silver. 1999. "Social Exclusion and Responsibility: Development Research Group, Rural Devel- Social Cohesion." Background paper for WorldDevelopmentReport opment-Gershon Feder (gfeder@worldbank.org), 2000/2001: Attacking Poverty. World Bank, Washington, D.C. Poverty Team-Ananya Basu, Vijayendra Rao, and Michael Woolcock, and Public Services for Human The Dynamics of Urban Poverty in Rio de Janeiro Development-Monica Das Gupta. With Kameshwar and Implications for Public Policy Prasad, All-India Institute of Medical Sciences, New Delhi; Paolo Belli, Harvard University; Devendra Gupta, This study takes advantage of high-quality data gathered National Council of Applied Economic Research, New more than 30 years ago to shed light on the intergener- Delhi; Peyvand Khaleghian, Sekhar Bonu, and William ational persistence of urban poverty. The baseline data Reinke, Johns Hopkins University; and V. R. Muraleed- come from interviews conducted with 750 residents of haran, Indian Institute of Technology, Chennai. squatter settlements in Rio de Janeiro in 1969. These Completion date: December 2001. interviews formed the basis of Janice Perlman's award-winning book The Myth of Marginality (Berkeley: Reports University of California Press, 1976). This study has Basu, Ananya, and Monica Das Gupta. Forthcoming. "Family tracked down more than half the original respondents and Systems and the Preferred Sex of Children." In Jan Hoem, ed., administered a specially structured interview tied to the InternationalEncyclopedia oftheSocialandBehavioralSciences. Vol. original. It has also interviewed past and present on Demography. New York: Elsevier Sciences. community leaders. Das Gupta, Monica. 1999. "Social Exclusion and Poverty." Data from the interviews make it possible to con- In Gudrun Kochendorfer-Lucius and Boris Pleskovic, eds., struct life trajectories of the families, tracing their fortunes Inclusion, Justice, and Poverty Reduction. Villa Borsig Workshop and the main events in their lives over the past 30 years. Series. Berlin: German Foundation for International Devel- The study maps these trajectories against the major opment. political and economic transformations in Brazil over - "Bias against Daughters in China, India, and the Repub- the same period, shedding new light on the effects of pub- lic of Korea: What Policies Are Most Likely to Work?" World lic policies, including some (such as state housing pro- Bank, Development Research Group, Washington, D.C. grams) targeted to the poor. The study also explores the Das Gupta, Monica, Helene Grandvoinnet, and Mattia Romani. survival strategies of households and the effectiveness 2000. "State-Community Synergies in Development: Laying of community and nongovernmental organizations as the Basis for Collective Action." Policy Research Working resources and as mediators of the effects of policy and Paper 2439. World Bank, Development Research Group, economic and political change. Washington, D.C. Many government and World Bank policies operate Das Gupta, Monica, Sunhwa Lee, Patricia Uberoi, Danning Wang, on the received wisdom that improving access to Lihong Wang, and Xiaodan Zhang. 2000. "State Policies and education and strengthening tenure will help lift the Women's Autonomy in China, the Republic of Korea, and poor from poverty. But preliminary data suggest that India, 1950-2000: Lessons from Contrasting Experiences." this may not be so. Policy Research Report on Gender and Development Work- The results of the research could inform Bank policy ing Paper 16. World Bank, Development Research Group, and assistance relating to basic sanitation, primary 24 Poverty and Social Developmenl education, primary health care, and upgrading of squat- In other work, based on three case studies, the ter settlements. The findings of the first phase of work project has shown that combining quantitative and were presented at a Bank seminar in November 1999. qualitative methods can provide a better understanding Findings of the main phase will be disseminated through of issues relating to poverty and social exclusion. The working papers and a book. three case studies are on the CAS (benefit) targeting Responsibility: Latin America and the Caribbean Region, system in Chile, the marginalization of the urban poor Brazil Rain Forest Unit-Joseph Leitmann (jleitmann in Uruguay, and reproductive health in rural Argentina. @worldbank.org). With Janice Perlman and Joanna In fiscal 2002 the study will attempt to make three Wheeler, Trinity College. The Tinker Foundation has contributions to the analysis of empowerment, extreme contributed funding for the research. poverty, and inclusion in Latin America and the Complelion date: June 2002. Caribbean. First, it will develop an informed framework for discussing these concepts and the relationships Report between them in the region. Second, it will conduct Perlman, Janice, Carlos Vainer, Pedro Abramo, Rick Huber, and quantitative analysis designed to suggest ways to Tim Campbell. 1999. "The Dynamics of Urban Poverty and measure extreme poverty and social exclusion with Implications for Public Policy: The Case of Rio de Janeiro, survey data, and to analyze the relationships between the 1969-99." World Bank, Private Sector Development and Infra- different dimensions of extreme poverty and social structure Vice Presidency, Infrastructure and Urban Develop- exclusion. Third, it will use case studies to analyze the ment Department, Washington, D.C. institutional processes leading to extreme poverty and social exclusion. It will also look at the practical policy Extreme Poverty and Social Exclusion implications for the Bank's operations in the region. in Latin America The study has led to presentations in Argentina and at a number of Bank seminars in 1999-2001. The more the World Bank promotes participation by Responsibility: Latin America and the Caribbean civil society and the poor in its programs and in the Region, Environmentally and Socially Sustainable poverty strategies developed by governments, the more Development Sector Unit-Estanislao Gacitua-Mario necessary it will become to be specific about the nature (egacituamario@worldbank.org), and Poverty Sector of empowerment and inclusion and the ways in which Unit-Quentin Wodon and Carinne Clert. With Huguette they can be promoted. This study aims to contribute to Redegeld and Bruno Tardieu, International Movement a better understanding of what empowerment and inclu- ATD Fourth World; Charles Courtney, Drew University; sion mean for development work. Jona Rosenfeld, Hebrew University, Israel; and Leandro Work so far has looked at what it means to live in Despouy. extreme poverty, how to reach the very poor through Completion date: June 2002. programs and interventions, and how to make private and public institutions more responsive to their aspirations. Reports It has also analyzed the relationship between extreme Gacitua-Mario, Estanislao, and Quentin Wodon, eds. ForEheom- poverty and human rights. And it has looked at the con- ing. Combining Quantitative and Qualitative Research Methodsfor tribution that the International Movement ATD Fourth the Analysis of Poverty and Social Exclusion. World Bank World-a grassroots nongovernmental organization with Technical Paper. Washington, D.C. (mostly local) staff in 25 countries-has made to the Redegeld, Huguette, Q. Tran, and Quentin Wodon. 2001. "Reach- understanding of the very poor and what is needed to ing the Poorest and Measuring Success: A Qualitative fight extreme poverty. The project has involved collab- Perspective." Draft. oration with this organization that will continue in the Wodon, Quentin, ed. 2001. AttackingExtremePoverty:Learningfrom coming fiscal year, with joint work foreseen in Bolivia, the Experience of the International Movement ATD Fourth World. Guatemala, and Peru. World Bank Technical Paper 502. Washington, D.C. Poverty and Social Development 25 Wodon, Quentin, G. Gonzalez, and Corinne Siaens. 2001. "Does dentis pooror not. Aseries ofcase studies (includingJamaica, Conditionality in Social Programs Exclude the Poorest? Nepal, and the Russian Federation) are examining these Demand-Side Schooling Interventions in Latin America." methods as potential complements to more conventional, World Bank, Latin America and the Caribbean Region, Poverry objective methods of measuring poverty and welfare. Sector Unit, Washington, D.C. The research project is tailored to the problems faced by World Bank staff undertaking poverty assessments and The Methodology of Poverty Assessments includes an active program of training and dissemination. Responsibility: Development Research Group, Poverty Policies to fight poverty rely increasingly on data about Team-Martin Ravallion (mravallion@worldbank.org), the living conditions of the poor, as reflected in a poverty Peter Lanjouw, and Michael Lokshin, and Public Services profile. But constructing a poverty profile that can be for Human Development-Berk Ozler; and East Asia and relied on to guide policy choices is often difficult. And Pacific Region, Poverty Reduction and Economic Man- the data and methods used can matter greatly for the agement Sector Unit-Menno Pradhan. choice of policies. At their worst, poorly devised poverty (ompletion date: June 2002. profiles can misdirect poverty reduction efforts-for example, by channeling scarce resources to cities when Reports poverty is worse in rural areas, or vice versa. Chaudhuri, Shubham, and Martin Ravallion. 1994. "How Well What methods are available for constructing poverty Do Static Indicators Identify the Chronically Poor?" Journalof profiles? What are their strengths and weaknesses? This Public Economics 53(March): 367-94. research project seeks to answer those questions by Jalan, Jyotsna, and Martin Ravallion. 2000. "Is Transient Poverty investigating the theoretical and empirical foundations Different? Evidence for Rural China." Journal of Development of the methods used in constructing poverty profiles. Studies 36(6): 82-99. Typically, too little work goes into assessing the robust- Lanjouw, Jean O., and Peter Lanjouw. 1996. "Aggregation- ness of poverty comparisons as the underlying mea- Consistent Poverty Comparisons: Theory and Illustrations." surement assumptions change. Many of the data routinely World Bank, Policy Research Department, Washington, D.C. used in poverty analysis are full of errors-a situation Lanjouw, Peter, and Martin Ravallion. 1994. "Poverty and House- unlikely to change. There also are unavoidable value hold Size." Economic Journal 105(November). (Also issued as judgments underlying measurement practices. The Policy Research Working Paper 1332, World Bank, Policy quality of the World Bank's policy assessments and Research Department, Washington, D.C., 1994.) prescriptions may or may not depend on these errors and Lokshin, Michael, and Martin Ravallion. 2000. "Welfare Impacts assumptions. An important task is to find out just how of Russia's 1998 Financial Crisis and the Response of the Pub- confident analysts can be in forming poverty comparisons. lic Safety Net." Economics of Transition 8(2): 269-95. The project examines the properties of the measures Pradhan, Menno, and Martin Ravallion. 2000. "Measuring Poverty used for assessing individual welfare, including the Using Qualitative Perceptions of Consumption Adequacy." practices used in comparing the welfare of different Review of Economics and Statistics 82(3): 462-71. demographic groups (such as large and small house- Ravallion, Martin. 1994. "A Better Way to Set Poverty Lines." holds). It also investigates the different methods used for Outreach 15. World Bank, Policy Research Department, setting poverty lines and how much they matter to the Washington, D.C. policy conclusions drawn. And it studies ways of making . 1994. Fundamentals of Pure and Applied Economics. Vol. better use of such non-income indicators of welfare as 56, Poverty Comparisons. Chur, Switzerland: Harwood Academic. health and education indicators. . 1994. "Poverty Rankings Using Noisy Data on Living A new strand of the research is investigating the prop- Standards." Economics Lettera 45: 481-85. erties of subjective welfare indicators, such as self-rated . 1996. "How Well Can Method Substitute for Data? Five assessments of the minimum income needed to make ends Experiments in Poverty Analysis." WorldBank Research Observer meet, of consumption adequacy, and of whether the respon- 11(2): 199-22 1. 26 Poverty and Sociol Development . 1998. Poverty Lines in Theory and Practice. Living Stan- performed a broad qualitative assessment, identifying key dards Measurement Study Working Paper 133. Washington, issues on which more needs to be learned. It reviewed D.C.: World Bank. existing public safety net programs, the principal sources .2000. "On DecomposingChanges in Poverty into Growth of household vulnerability, and what is known about and Redistribution Components." Journal of Quantitative coping strategies, and outlined an agenda for strength- Economics 16(1): 105-18. ening the design and implementation of the main safety . 2000. "Should Poverty Measures Be Anchored to the net programs. National Accounts?" EconomicandPolitical Weekly (August). The recently released 1998 Vietnam Living .2001. "On Assessing the Efficiency ofthe Welfare State." Standards Survey provides previously unavailable data Kyklos 54(1): 115-23. on policy coverage across communes and, in some cases, Ravallion, Martin, and Benu Bidani. 1994. "How Robust Is a households. These data allow quantitative analytical Poverty Profile?" World Bank Economic Review 8(1): work to assess the performance of programs aimed at 75-102. reducing poverty and providing insurance to poor house- Ravallion, Martin, and Michael M. Lokshin. 1999. "Subjective holds. Information on who the poor are can now be com- Economic Welfare." Policy Research Working Paper 2106. bined with information on needs, program participation, World Bank, Development Research Group, Washington, social relief disbursements, and outcomes. In addition, D.C. .. the new survey's coverage of many of the households and .2000. "Who Wants to Redistribute? Russia's Tunnel Effect communities interviewed in the 1992 Living Standards in the 1990s." Journal of Public Economics 76(1): 87-104. Survey creates a panel that can be used to test and . Forthcoming. "Identifying Welfare Effects from Subjec- monitor the performance of interventions. tive Questions." Economica. Drawing on these data, the second part of the project Forthcoming. "Self-Rated Economic Welfare." European examines the performance of decentralized programs Economic Review. and expenditures for poverty reduction in targeting poor Ravallion, Martin, and Shankar Subramanian. 1996. "Welfare communes and poor people. It explores sensitivity to the Measurement with and without Substitution." World Bank, definition of poverty and assumptions about household Policy Research Department, Washington, D.C. behavioral responses to the programs. The study examines whether programs perform a safety net Safety Nets in an Emerging Market Economy function, recognizing that this involves both protection from poverty and promotion from poverty and assessing Vietnam's transition to a market economy has stimu- how well the safety net performs each function. The work lated strong growth. Continued broad-based growth will also examines the role of non-income factors, including help raise living standards for most people. But many whether equally poor communes in different provinces people will continue to be poor, and most will remain are treated equally and, if not, what accounts for the vulnerable to risk. Under the former command economy differences. The research models the determinants of lack of mobility ensured community and family solidar- consumption changes over time as a function of (among ity, and households belonged to local cooperatives that other things) program placement at the household level. provided for the welfare of their members. Developing To assess the performance of the safety net, it will then a reliable and effective system of redistributive transfers compare poverty transition matrices with and without and safety nets to replace these faltering institutions is safety net programs. an important part of the successful transition to a Research results have been disseminated through a market economy. conference in Hanoi in May 2001. This project consists of two parts. The first used Responsibility: Development Research Group, Public Vietnam as a case study with which to assess the strengths Services for Human Development-Dominique van de and weaknesses of an existing safety net program. Walle (dvandewalle@worldbank.org). Because of an earlier lack of good data, the study Completion dole: June 2002. Poverty and Social Development 27 Reports The financial analysis uses data on investment van de Walle, Dominique. 1999. "Protecting the Poor in Viet- regulations, real returns, and new financial products nam's Emerging Market Economy." Vietnam's Socioeconomic provided by the regulators of the new private pension Development:A Social Science Review 19: 59-80. funds and the private insurance markets. The principal .1999. "SafetyNets in an EmergingMarket Economy." In variables include private pension assets as a share of Jennie 1. Litvack and Dennis A. Rondinelli, eds., MarketReforn GDP, the supply of short-term and long-term securities, in Vietnam:Building InstatitonsforDevelopment. Westport, Conn.: real returns from investment, and the supply of new Quorum Books. annuity products. - 2001. "The Static and Dynamic Incidence of Viet Nam's The coverage analysis draws on household surveys in Public Safety Net." World Bank, Development Research 12 countries; a survey on pensions and self-insurance in Group, Washington, D.C. Santiago, Chile, in 2000; country-level data from pension authorities; and the World Bank's SIMA database. The Will Social Security Reform in Latin America principal variables are both country level and household Provide Greater Retirement Security? level. Country-level variables include the share of the labor force contributing to the formal pension system, the In response to changing economic and demographic share in self-employment, the total payroll tax for social trends, Latin American governments embarked on social insurance programs, and the share of the payroll in indi- security reform in the 1980s and 1990s, dismantling vidual retirement accounts. centrally managed single-pillar pension systems and Responsibility: Latin America and the Caribbean Region, replacing them with multipillar systems. This study Human Development Sector Unit-Indermit Gill attempts to answer a difficult question: Will the new (igillCa)worldbank.org) and Truman Packard, and Finance, social security systems in Latin America provide better Private Sector, and Infrastructure Sector Unit-P. S. retirement security than the institutions they replaced? Srinivas; and Human Development Network, Social The study defines better as higher value for contributions, Protection Team-Asta Zviniene. With Juan Yermo, lower risk of loss, and greater coverage of the population- Oxford University and Organisation for Economic Co- with lower poverty and greater equity. The analysis operation and Development; and Salvador Valdes, spans issues in fiscal management, financial sector Catholic University of Santiago, Chile. development, labor market efficiency, and poverty Completion date: June 2002. reduction. The study draws heavily on the macroeconomic and World Poverty Monitoring microeconomic literature on pension reform and theo- retical and empirical research completed or under way Is poverty declining in the developing world? Because in the region, applying a comprehensive insurance frame- little effort has gone into compiling and analyzing work to consolidate the formidable amount of material. distributional data on a reasonably comparable basis, It takes a comparative approach to issues, with reference this question is surprisingly difficult to address. Yet there to the reformed pension systems in Argentina, Bolivia, is a clear need to do so, both to help monitor progress in Chile, Colombia, El Salvador, Mexico, and Uruguay as reducing poverty and as a first step toward understand- well as to social security in countries that have not under- ing the causes and effects of changing distribution. taken reform, such as Brazil. This project monitors progress in reducing aggregate The fiscal analysis is based on PROST simulations poverty using a consistent compilation of distributional using data on contributors and beneficiaries provided by data from household surveys. Estimates of various poverty the social security authorities in the sample countries. The measures are available from numerous studies of principal variables of interest include the implicit individual countries, but their use for monitoring world pension debt and current balance as a share of GDP poverty is questionable because of comparability under different macroeconomic scenarios. problems. Past work at the country level has used poverty 28 Poverty and Sotiol Development lines appropriate to each country, but the real value of Ravallion, Martin. 1994. "Measuring Social Welfare with and with- local poverty lines tends to increase with the average out Poverty Lines." American Economic Review (May). income of a country. This fact clouds attempts to com- . 1997. "Can High-Inequality DevelopingCountries Escape pare and aggregate across countries using the poverty data Absolute Poverty?" Policy Research Working Paper 1775. World available in standard (secondary) sources. The use of Bank, Policy Research Department, Washington, D.C. official exchange rates also biases international poverty .2000. "Growth and Poverty: Making Sense of the Debate." comparisons. World Bank, Development Research Group, Washington, D.C. This project uses primary data sources and reesti- - . Forthcoming. "Growth, Inequality, and Poverty: Looking mates all poverty measures on a consistent basis, beyond Averages." World Devlopment. converting local currencies to constant purchasing power Ravallion, Martin, and Shaohua Chen. 1997. "What Can New parity values. It also tests the robustness of comparisons Survey Data Tell Us about Recent Changes in Poverty and across regions and over time to measurement assump- Distribution?" WorldBank Economic Review 11(2): 357-82. tions. The data set covers 83 countries, with data for - . 2000. "How Did the World's Poorest Fare in the 1990s?" two or more points in time for 50 of them. Policy Research Working Paper 2409. World Bank, Develop- The estimates indicate that by 1998, 1.2 billion ment Research Group, Washington, D.C. people-roughly a quarter of the population covered- Ravallion, Martin, Gaurav Datt, and Dominique van de Walle. 1991. were living on less than a dollar a day at 1993 prices. The "Quantifying Absolute Poverty in the Developing World." incidence of absolute poverty in the developing world Review of Income and Wealth 40(December): 359-76. as a whole fell slightly between 1987 and 1998, while the total number of poor people remained the same. But there Patterns of Inequality and was marked variation among regions and countries, with Government Intervention the number of poor rising in most of Africa, Europe and Central Asia, and Latin America and the Caribbean and The Deininger-Squire database on income distribution falling in East Asia. has become a standard source in cross-country analysis Results of this research are reported in the World of the relationship between inequality and growth. This Bank's annual World Development Indicators (Washing- project aims to update and improve the database, thus ton, D.C.) and in its WorldDevelopmentReport2000/2001: increasing its usefulness. Attacking Povery (New York: Oxford University Press, Taking advantage of the large number of microeco- 2000). Data from this project have also been used in a nomic data sets that have become available since the number of studies of cross-country differences in progress database was first put together, the project will increase in reducing poverty and inequality. both the number of countries and the observations Responsibility: Development Research Group, Poverty covered by the database. It will incorporate data on Team-Martin Ravallion (mravallion@worldbank.org) decile income shares and a rural-urban disaggregation. and Shaohua Chen. And it will add data on access to key social services by Completion date: June 2002. income decile, facilitating more comprehensive analysis of changes in the welfare of the poor, their access to Reports economic opportunities, and the role of government Anand, Sudhir, and Martin Ravallion. 1993. "Human Development programs in expanding these opportunities. in Poor Countries: On the Role of Private Incomes and Responsibility: Development Research Group, Rural Devel- Puiblic Services." Journal of Economic Perspectives 7(1): opment-Klaus Deininger (kdeininger@worldbank.org), 133-50. and Development Economics Senior Vice Presidency, Bidani, Benu, and Martin Ravallion. 1994. "Decomposing Social Global Development Network-Lyn Squire. With Indicators Using Distributional Data." Policy Research Work- Kihoon Lee, University of Maryland; and Changquing ing Paper 1487. World Bank, Policy Research Department, Sun, University of California at Berkeley. Washington, D.C. Completion date: July 2002. Poverty and Social Development 29 Public Spending and the Poor in Latin America Responsibility: Latin America and the Caribbean Region, Poverty Sector Unit-Quentin Wodon (qwodon This research project looks at recent patterns of public @worldbank.org) and Judy Baker. With Mohamed Ihsan expenditures in the social sectors in Latin America, Ajwad; Bernadette Ryan; Corinne Siaens; Jean-Philippe focusing on health, education, social protection, and Tre; and Shlomo Yitzhaki, Hebrew University, Israel. poverty reduction. The project, which has four compo- (ompletion date: September 2002. nents, addresses two main questions: To what extent do public social expenditures benefit the poor? And how Reports could the benefits for the poor be increased? Ajwad, Mohamed Ihsan, and Quentin Wodon. 2001. "Supply-Side The first component is a theoretical framework for Interventions: Spillover Effects from Preschools to Secondary evaluating public spending and social programs, with Schools." World Bank, Latin America and the Caribbean empirical applications. The framework uses a flexible Region, Poverty Sector Unit, Washington, D.C. Draft. social welfare function that takes into account not only Castro-Fernandez, Rodrigo, Carinne Clert, and Quentin Wodon. absolute but also relative deprivation (people assess their 2001. "Government Programs for the Insertion of Youths and welfare in part by comparing themselves with others). Women in Chile's Labor Market: A Discussion." Background Two summary parameters are estimated to assess the paper for Chile: Poverty in a High-Growth Economy. WVorld Bank, impact of public expenditures on welfare. The first para- Larin America and the Caribbean Region, Poverty Sector Unit, meter accounts for the impact of programs and policies Washington, D.C. on growth, and the second accounts for the impact on . 2001. "Protecting the Unemployed in Chile: From State distribution (inequality). Various decompositions are Assistance to Individual Insurance?" Background paper for provided for the second parameter. Chile: Poverty in a High-Growth Economy. World Bank, Latin The second component is a database with social indi- America and the Caribbean Region, Poverty Sector Unit, Wash- cators for about 12 countries, with summary statistics ington, D.C. by area (urban or rural), by gender, and by income or Christiaensen, L., C. Scott, and Quentin Wodon. 2001. "Devel- wealth quintile. The database covers health, education, opment Targets: Feasibility and Costs." In World Bank, Poverty employment, and access to basic infrastructure, and Reduction Strategy Sourcebook CD-ROM. Poverty Reduction includes updated poverty and inequality numbers for and Economic Management Network, Washington, D.C. Latin America through 1998. Wodon, Quentin, and Shlomo Yitzhaki. 2000. "Evaluating the The third component is a study on public spending Impact of Government Programs on Social Welfare: The Role and the poor in Latin America (including new method- of Targeting and the Allocation Rules among Program Benefi- ologies for marginal benefit incidence analysis and assess- ciaries." World Bank, Latin America and the Caribbean Region, ments of program impacts and targeting performance). Poverty Sector Unit, Washington, D.C. The study focuses on health, education, and social pro- . 2001. "The Effect of Using Grouped Data on the Gini tection, but also examines such issues as the impact of Income Elasticity Estimator." World Bank, Latin America and indirect tax reform on poverty and the marginal the Caribbean Region, Poverty Sector Unit, Washington, D.C. efficiency cost of funds. Draft. The project's fourth component is a set of Excel-based . 2001. "Inequality and Social Welfare." In World Bank, simulation tools (SimSIP) that allow users to set targets for Poverty Reduction Strategy Sourcebook CD-ROM. Poverty Reduc- poverty and social indicators and estimate the cost of reach- tion and Economic Management Network, Washington, D.C. ing those targets. The simulation tools, which will be made . 2001. "The Robustness of Regression Coefficients: A available free of charge, are being used for these purposes Geometrical Exposition." World Bank, Latin America and the by a number of World Bank client countries. Caribbean Region, Poverty Sector Unit, Washington, D.C. Results from the project have been presented at sem- Wodon, Quentin, Mohamed Ihsan Ajwad, Bernadette Ryan, inars at Hebrew University, the International Monetary Corinne Siaens, and Jean-Philippe Tre. 2001. "SimSIP: Fund, and the World Bank. Simulations for Social Indicators and Poverty." 30 Poverty and Social Development Wodon, Quentin, Rodrigo Castro-Fernandez, G. Lopez-Acevedo, tute of Economic Growth, Delhi; Soumya Chattopad- Corinne Siaens, C. Sobrado, and Jean-Philippe Tre. Forth- hyaya, University of Maryland at College Park; Lester coming. "Poverty in Latin America: Trends (1986-98) and Coutinho, Johns Hopkins University; and SaumitraJha, Determinants." Cuaderwoste Economia. Stanford University. The Netherlands Trust Fund is contributing funding for the research. Urban Poverty, Risk Management, (ompletion date: October 2002. and Social Capital Reports This project has both a substantive and a methodologi- Jha, Saumitra, Michael Woolcock, and Vijayendra Rao. 2001. cal purpose. Its substantive purpose is to study aspects "Governance in the Gullies: The Determinants of Power of the relationship between social institutions and poverty Networks in Delhi Slums." Paper presented at a conference of in urban India in order to help improve the effectiveness the Society for Behavioral Economics, Washington, D.C., of poverty reduction policies and programs. Its method- June. ological purpose is to help develop an integrated quali- . 2001. "States, Social Institutions, and the Management of tative and quantitative approach that combines Collective Action Problems in Delhi Slums." Paper presented econometrics and ethnographic methods with economic at annual meeting of the Association of American Geographers, and social theory to study how poor urban households use New York, March 1. social resources to manage risk. The project is preparing for a qualitative analysis of networks for survival and Cultural Change, Community Mobilization, mobility, and tests of some hypotheses from the and Participatory Development qualitative work using quantitative data. Work to assess living standards is under way. Alarming infection rates for HIV and other sexually The study draws on primary survey data and qualita- transmitted diseases prompted a group of sex workers to tive information from focus groups and in-depth create the Durbar Committee for Coordination of Women interviews. Fieldwork began in October 1999 and was (DMSC) in West Bengal, India, primarily to promote completed in January 2000. As analysis began, some sexual health and HIV prevention among sex workers. problems with the data were discovered that required But DMSC gradually began to address broader social and additional data collection in the field. Analysis of the data cultural issues, calling for the recognition and protection also revealed several econometric issues better addressed of sex workers' rights. Its activities reflected a transfor- by collecting panel data, which it is hoped can be done mation in the sex workers' capacity, self-image, and in the coming year. perception of their social position-a profound change The results of the data analysis are expected to inform in their cultural attitude. the design of slum infrastructure projects and social Drawing on the analytical insights of sociology, anthro- protection programs for the urban poor. The research has pology, and social history, this research will examine the already contributed to several World Bank operational pro- mechanisms that made this cultural change possible, jects through an improved understanding of the nature including the catalytic role of development interven- of governance and networks among the urban poor. And tions. The research will: the work in Delhi to develop a methodology for under- * Examine the processes of dynamic transformation standing urban poverty is contributing to a review of of cultural norms and reformulation of values and belief urban poverty in East Asia. The main findings of the systems among the poor that enable them to engender study have been presented in several World Bank developmental change. seminars. * Identify forms of cultural change among the poor that Responsibility: Development Research Group, Poverty enable them to redefine their self-identity and recon- Team-Vijayendra Rao (vrao@worldbank.org), Michael ceptualize their potential and capabilities as human Woolcock, and Ana Maria Ibaniez. With Arup Mitra, Insti- subjects and social agents. Poverty and Social Development 31 o Investigate the processes through which cultural Although most producers in India are highly skilled and change leads to a sense of community identity and many are true artisans, the vast majority are poorly edu- cohesion among the poor where none existed before. cated and have low social status. o Identify forms of policy intervention that can cre- The objective of this project is to empower Indian arti- ate an enabling environment for cultural change among sans and reduce poverty among them by: the poor, in which they emerge as the major actors and o Developing expanded market outlets in Europe agents of cultural rethinking. and North America for high-quality crafts from India The research will be a qualitative, cross-sectional and creating more direct links between groups of artisans study based on focus group discussions, unrestricted in India and international market outlets-to improve the interviews, participant observation, and desk-based flow of information to retailers on the craft products research. It will involve DMSC in the research and available from India and to artisans on improvements in dissemination, helping to strengthen the organization's technologies that would make their products more institutional capacity. serviceable. The research is expected to contribute to the o Expanding the incomes going directly to the theoretical and analytical literature on the motivation for artisans. collective action, especially cultural factors, and to o Enabling groups of poor artisans in India to sustain shed light on the relative importance of collective livelihoods they value and improve their well-being. self-representation, expression of community identity, and The project models market accessibility strategies the quest for material improvement as the motor of using an approach that recognizes the role of interme- group activity. In addition, the study will provide a diaries in facilitating artisans' links to the market. The policy framework for creating an enabling environment shops of the Smithsonian's Freer and Sackler Galleries forculturalchangeamongsexworkersandotherpoorand will organize a buyers network of museum shops and marginalized groups. And its findings will contribute to stores in North America and possibly Europe that are the World Bank's initiative on AIDS control in South Asia interested in fair trade in high-quality craft products, to by pointing to the types of health interventions most ensure that a larger share of the profits goes to the arti- likely to succeed. sans. The Mehrangarh Museum Trust in Jodhpur will Responsibility: Social Development Department-Kreszen- help organize a network among artisans, focusing on tia Duer (kduer@worldbank.org) and Sabina Alkire. With groups of poor artisans in Gujarat, Madhya Pradesh, and Nandini Gooptu, Oxford University; and Nandinee Rajasthan. The trust will help the artisans network nego- Bandyopadhyay. The Netherlands Trust Fund is con- tiate with the buyers network, fill orders, and resolve logis- tributing funding for the research. tical problems in exporting, and hold workshops with the Completion date: December 2002. artisans to evaluate the process. The project will conduct a baseline study of the par- Innovations in the International Crafts Market ticipating artisans' groups and a control group, focusing and Empowerment of Artisans in India on their incomes, their values, their security, their aspirations for their lives, their perception of their oppor- Crafts form a significant sector in India, not only because tunities, the physical conditions under which they work, of their intrinsic cultural and aesthetic value but also and the degree of social organization and bonding in because of their promising potential for economic devel- their cooperative. The project will resurvey these factors opment. Recent estimates suggest that handicrafts each year for three years, evaluating changes and employ more than 9 million people in India and con- analyzing conditioning factors. tribute about $1.6 billion to export earnings and $4 bil- The project is expected to contribute to the under- lion to domestic earnings. But those producing the crafts standing of the role of crafts and culture in poverty reap the fewest benefits from the lucrative market, and reduction, economic growth, and development. And by even the most talented often live in abject poverty. establishing a network of stakeholders to coordinate 32 Poverty and Sotial Development negotiations for higher prices for artisans' products, the groups. The research will thus contribute to the Bank's project is expected to improve the understanding of the ability to carry out a poverty reduction agenda need for interventions in the crafts industry. that includes empowerment as a key dimension of Responsibility: Social Development Department-Kreszen- poverty. tia Duer (kduer@worldbank.org) and Sabina Alkire. With Responsibility: Social Development Department-Kreszen- Frank Penna, Policy Sciences Center; Marty Bernstein, tia Duer (kduer@worldbank.org) and Sabina Alkire. With Smithsonian Institution; and Maureen Liebl, Mehran- Anthony Bebbington, University of Colorado; and David garh Museum Trust, Jodhpur. The Netherlands Trust Lewis and Simon Batterbury, London School of Eco- Fund is contributing funding for the research. nomics. The Netherlands Trust Fund is contributing (ompletion date: December 2002. funding for the research. Completion date: December 2002. Organizational Cultures and Spaces for Empowerment: Interactions between Poor People's Evaluation of the Impact of Investments Organizations and World Bank Poverty Programs in Early Child Development on Nutrition and Cognitive Development As part of a broadening poverty agenda and a growing commitment to community-driven development, the Early child development projects account for a rapidly World Bank is increasingly engaging with poor people's growing share of World Bank lending. These programs organizations and designing investment projects that combine nutrition, health care, and cognitive develop- directly finance such organizations. How will this engage- ment to offset the disadvantages of growing up in poverty. ment affect these organizations? In particular, how will Substantial evidence shows that poor health and an it affect their culture? And what are the implications of inadequate early learning environment lead to handicaps this cultural impact for the organizations' effectiveness difficult to reverse later in life. Many children may never in empowering their members and increasing their go to primary school without interventions that reach incomes? them in their early years. If they do attend primary Through case studies in Bangladesh, Burkina Faso, school, they may be at high risk of dropping out early or and Ecuador, this research will study the link between repeating grades. That lowers the return to both public culture and poverty at several levels, addressing these and private schooling. So early child development questions: programs may both increase the efficiency of a range of * How does the culture of poor people's organizations other investments in human capital and promote equity -and the culture of intermediary organizations working in the population they serve. on behalf of the poor-affect the extent to which they Through early coordination with World Bank operations, are able to address poverty? this research is evaluating the impact of investments in * How does the World Bank's culture influence the early child development in three projects: the Bolivian Inte- effectiveness of Bank-supported poverty reduction pro- grated Child Development Project, the Uganda Nutrition jects implemented through poor people's organizations? and Early Child Development Project, and the Philippine And how do the cultural differences between poor Early Childhood Development National Investment Pro- people's organizations and the World Bank affect the gram. All three case studies investigate the effect of early possibility of their engaging with each other in poverty child stimulation and coaching on the age of school enroll- reduction programs? ment and on indicators of cognitive development. They The research is expected to contribute to the World also relate indicators of early cognitive development to early Bank's understanding of its potential role in empower- primary school grade progression and performance. In ing culturally marginalized social groups, particularly by addition, the research assesses the suitability of measures using investment operations to create spaces for the of cognitive development and achievement for evaluat- direct involvement of organizations representing these ing projects. Poverty and Social Development 33 The case studies include both home-based (center- The deworming pilot in Uganda will follow a standard based) and parental education approaches to child stim- experimental methodology in which there will be a ran- ulation. Moreover, because Bolivia is considering the dom assignment of treatment and control communities. sustainability of its day care program, which operates in Both the treatment and the control groups will be the homes of service providers but functions as a center- chosen from communities deemed especially at risk based program, that case study compares the cost- based on results of a survey of parasites undertaken as effectiveness of thatapproach with the cost-effectiveness part of the project in 1998. The treatment began in of new, indirect approaches that will be implemented on October 2000 and will continue for two years or until a a pilot basis. significant impact is measured. Initial analysis will be For Bolivia's early child development program, the conducted jointly with the Imperial College School of initial evaluation was based on a comparison of the Medicine in London and Ugandan counterparts. differences across groups between the baseline and the sec- The Uganda project includes demand-driven ond round of data collection, conducted two years after the components (with communities selecting the programs project was initiated. The results showed an unambigu- they will support). So longitudinal data and community ous improvement in the cognitive development of program fixed effects will be used to address selection bias. T'he participants relative to the control groups. Because the pro- baseline data were collected in early 2000, and prelimi- gram selected children of low height, those who have nary analysis has been undertaken by a staff member of been in the program a short time are smaller than their the Ugandan Institute of Public Health. Since service counterparts. But as time in the program increases, the gap delivery was not initiated until November 2000, the diminishes. The program appears to be particularly effec- second round of household data collection will not be tive in preventing severe malnutrition. undertaken until 2002. While the program is expensive-costing more than Implementation of the Philippine project was also $40 per child a month, including donated food-the delayed, as was the collection of baseline data. Both are analysis found that the benefits exceeded costs by up to now moving in parallel. 70 percent. These benefits include the expected increase Responsibility: Rural Development Department-Harold in earnings directly due to higher cognitive develop- Alderman (halderman@worldbank.org); and Develop- ment and greater physical stature, and the indirect impact ment Research Group, Public Services for Human Devel- on years of schooling mediated by improved health and opment-Elizabeth King. With Jere Behrman, Yingmei learning capacity. Finally, the program appears to be Cheng, and Petra Todd, University of Pennsylvania; well targeted to the poorest households in the selected Patrice Engle, California Polytechnic State University; communities. Donald Bundy, Oxford University; and N. B. Kabater- A third round of data collection in Bolivia is scheduled eine, Vector Control Division, Uganda. for 2001. This data collection will allow a comparison of Completion dale: June 2003. two variants of the intervention and analysis of the project's impact on primary school performance. Reports The Uganda case study will evaluate separately the Alderman, Harold, Jere Behrman, Hans-Peter Kohler, John A. Maluc- impact of deworming on children under the age of six. cio, and Susan Cotts Watkins. 2000. "Attrition in Longitudinal While deworming programs worldwide are targeted to Household Survey Data: Some Tests for Three Developing- school-age children, for whom worm loads are highest, Country Samples." Policy Research Working Paper 2447. World work in progress in Lucknow, India, shows the potential Bank, Development Research Group, Washington, D.C. for dramatic increases in weight for younger children Behrman, Jere, Yingmei Cheng, and Petra Todd. "The Impact of following a mass deworming campaign. As no compara- the Bolivian Integrated 'PIDI' Preschool Program." University ble evaluation has been undertaken for this age group in of Pennsylvania, Philadelphia. other countries or for other indicators, the Uganda exper- Kabatereine, N. B., E. Tukahebwa, S. Brooker, Harold Alderman, iment will be particularly informative. and A. Hall. 2001. "The Epidemiology of Intestinal Helminth 34 Poverty and Sotiol Development Infections among Schoolchildren in 18 Discricts of Southern impacts on budget shares on the expenditure side is Uganda." East African Medical Journal (May). simply not enough. A key problem identified at the conference is the Poverty and the International Economy: lack of information for evaluating poverty impacts at the What Are the Links? national level. Information is needed on where house- holds at different income levels get their incomes. How trade liberalization affects poverty is clearly a Moreover, the data need to be collected in a form that crucial question for the World Bank's work. This project will make it possible to map from changes in factor will produce both research and practical tools on the returns to impacts on households, and thus at a level of subject. Trade policies can be expected to affect the product disaggregation comparable to that in available poor through their impact on the consumer and factor economic models. prices faced by the poor (or on the demand for their In future work the project will prepare a policy services). The impact of these price changes on the real summary drawing out some of the key messages from the income of the poor depends on their ability to adjust to research and identify the research needed to move ahead the changes. Changes in the provision of public goods or with formulating well-founded policy responses. The safety nets that may be required as a result of changes project will also develop a computable general equilib- in tariff revenues, or chosen as part of comprehensive rium model that addresses policymakers' needs in assess- reform, will also affect the welfare of the poor. And these ing the impact of trade liberalization on the poor. The effects involve a dynamic or time dimension, since the model would include diverse households, distinguished ability to adjust changes over time, as do investment by income and sources of factor income, and different and technology decisions affected by trade policy. labor markets, recognizing that the poor are often con- Preliminary findings from work on a number of these centrated in the informal sector. The model should be issues were presented at the conference on Poverty and a useful tool for designing pro-poor policies. the International Economy held in Stockholm on In addition, the project will collect available data on October 20-21, 2000 (the papers from this conference are household incomes and expenditure of the poor and available on the Web at http://wwwl.worldbank.org/ organize the data into coherent patterns allowing infer- wbiep/trade/poverryconf.html). The research program on ences about poverty impacts in countries for which com- which this conference wvas based began with detailed prehensive models and data sources are unavailable. studies of Indonesia and South Africa and has since been And in a backward-looking element the project will broadened to include studies of Mexico and Turkey and examine a country or countries that have undertaken trade a global assessment. Several of these studies used policy reform and investigate the effect of the reform on computable general equilibrium models to trace out the poor. effects on prices and then assessed the effects of the price Responsibility Development Research Group, Trade-David changes on households. One study examined the dynamic Tarr (dtarr@worldbank.org), Will Martin, Alessandro links between trade reform, growth, and poverty. And Nicita, and Isidro Soloaga, and Macroeconomics and two focused on developing practical methodologies Growth-David Dollar and Aart Kraay; Human Devel- forassessingeffectsonpoverty-oneatthecountrylevel opment Network, Office of the Vice President- (relevant for much Bank country policy work) and Shantayanan Devarajan; and Development Prospects one at the global level (needed for the broader policy Group-Dominique van der Mensbrugghe. With Steven debate). Berry, Yale University; Anne Case, Princeton University; An important methodological finding is that the James Levinsohn and Jed Friedman, University of impacts on the poor seem to come more strongly through Michigan; Thomas Rutherford, University of Colorado; factor prices than through consumer prices, implying Glenn Harrison, University of South Carolina; Thomas that researchers need to do the hard work of tracing the Hertel and Paul Preckel, Purdue University; Dani Rodrik, effects of trade reform on factor prices. Focusing on the Harvard University; and L. Alan Winters, University Poverty and Social Development 35 of Sussex. The Swedish International Development Research Working Paper 2142, World Bank, Development Cooperation Agency, the Swedish Ministry of Foreign Research Group, Washington, D.C., 1999.) Affairs, and the Netherlands Ministry of Development Cooperation (through the World Bank-Netherlands Part- Welfare Impacts of Policy nership Program) are contributing funding for the on Disadvantaged Groups research. Completion dale: December 2003. This study examines how policies affect the welfare of the poor and disadvantaged when there is inequality Reports among groups, such as ethnic and gender groups. The Case, Anne. 2000. "Implications of Trade Policy Reform Given project began by examining the sources of inequality Income Distribution and Expenditure Patterns in South Africa." among ethnic groups in Vietnam, where ethnic minori- Paper presented at the conference Poverry and the International ties tend to be concentrated in remote rural areas and Economy, Stockholm, October 20-21. typically have lower standards of living than the ethnic Cranfield, John A. L., Thomas Hertel, and Paul Preckel. 2000. majority. It asked how much of minorities' lower living "Trade and Poverty Reduction in a WTO Round." Paper standards is due to poor economic characteristics and how presented at the conference Poverty and the International much to low returns to characteristics. And it investigated Economy, Stockholm, October 20-21. whether a self-reinforcing culture of poverty exists, Devarajan, Shantayanan, and Dominique van der Mensbrugghe. reflecting patterns of current and past discrimination. The 2000. "Trade Reform in South Africa: Impacts on Households." results show that differences in returns to productive Paper presented at the conference Poverty and the International characteristics are an important explanation of ethnic Economy, Stockholm, October 20-21. inequality, particularly in areas where both minority and Dollar, David, and Aart Kraay. 2000. "Growth, Trade Reform, and majority groups reside. There is evidence of compen- Poverty: A Macroeconomic Approach." Paper presented at sating behavior on the part of minorities. the conference Poverty and the International Economy, The results of this first part of the research suggest that Stockholm, October 20-21. it is not enough to target poor areas to redress ethnic Friedman, Jed. 2000. "Implications of Trade Reform for the Income inequality. Policies need to reach minority households Distribution in Indonesia." Paper presented at the conference within poor areas and to explicitly recognize behavioral Poverty and the International Economy, Stockholm, October patterns that have served the minorities well in the short 20-21. term but intensify ethnic differences in the longer term. Harrison, Glenn, Thomas Rutherford, and David Tarr. 2000. The majority group's model of income generation can "Trade Reform in Turkey and the Impact on the Poor." Paper serve as a guide on how to fight poverty among ethnic presented at the conference Poverty and the International minority groups. Economy, Stockholm, October 20-21. A second part of the research focuses on the delivery lanchovichina, Elena, Alessandro Nicita, and Isidro Soloaga. 2000. of services to disadvantaged groups. In some countries "Implications of Trade Reform for the Distribution of House- certain disadvantaged groups-for example, women or hold Income in Mexico." Paper presented at the conference ethnic minorities-may be socially or economically Poverty and the International Economy, Stockholm, October excluded from society in ways that differ from those for 20-21. the rest of the poor. This difference may be in part the Levinsohn, James, Steven Berry, and Jed Friedman. 2000. "Impacts result of current or past discrimination. How does this of the Indonesian Economic Crisis: Household Evidence." affect how services are delivered? Do services compen- Paper presented at the conference Poverty and the International sate for such inequalities or perpetuate them? Should ser- Economy, Stockholm, October 20-21. vices be delivered in different ways when dealing with Matusz, Steven, and David Tarr. 2000. "Adjusting to Trade Pol- disadvantaged groups, or should all groups be treated the icy Reform." In Anne Krueger, ed., Economic Policy Reform. same? Can better delivery methods help redress such Chicago: University of Chicago Press. (Also issued as Policy inequalities? The project will do some follow-up work 36 Poverty nnd Sotiol Development on Vietnam's ethnic minorities and on India's sched- York: Oxford University Press, 2000) and continues to be uled castes and tribes. drawn on as the Bank incorporates the report's findings The research has been disseminated through seminars into its operations and as it launches new initiatives on at the World 13ank (August 1999); ARQADE, University community-driven development. of Toulouse (October 1999); DELTA-INRA, Paris The research is expected to improve the quality of (December 1999); and Tinbergen Institute, Amsterdam Bank-supported projects concerned with encouraging (March 2000). the participation of the poor (through such initiatives as Responsibilitfy Development Research Group, Public Services community-driven development) and improving insti- for Human Development-Dominique van de Walle tutional quality. Bank operations staff have been asked (dvandewalle@worldbank.org). With Dileni Gunewar- to contribute to future work examining the impact of dena, University of Peradeniya, Sri Lanka. social capital on development projects. Completion date: December 2003. The work has become the foundation for major conceptual and policy analysis in Bank client countries. Report In Colombia social capital has served as the basis for a van de Walle, Dominique, and Dileni Gunewardena. 2001. "Sources new urban renewal initiative designed to reduce crime of Ethnic Inequality in Vietnam." Journal of Development and improve citizens' well-being. In Romania a social Economics 65(1): 177-207. (Also issued as Policy Research Work- capital framework helped in building a dialogue among ing Paper 2297, World Bank, Development Research Group, stakeholders (unions, government officials, media Washington, D.C., 2000.) representatives) that for decades have not communi- cated with one another. In Guatemala a large, integrated Social Capital quantitative and qualitative poverty analysis has been conducted, the first of its kind on this scale. In Indone- The social dimensions of development are increasingly sia new studies are looking at the impact of community- recognized as important determinants of individual based projects on local institutions. And in India the well-being and institutional performance. This project findings of a study have led to changes in the public consists of a number of studies that investigate the social distribution system for food grains. dimensions of development, with the aim of improving the Results from the research project have been theoretical foundations, empirical support, and quality of presented at many conferences and workshops for the policy recommendations emerging from research on academics, politicians, policymakers, civil servants, the the subject. The studies use several analytical approaches, media, business leaders, and civil society representatives. including formal modeling, econometric analysis, and In addition, the work formed the basis for a new course, qualitative analysis. They draw on a variety of sources, rang- Social Institutions and Economic Development, offered ing from literature reviews and secondary sources (such at Harvard University's Kennedy School of Government. as the World Bank Institute) to original fieldwork consisting Responsibility: Development Research Group, Poverty of interviews and household surveys. Team-Vijayendra Rao (vrao@worldbank.org) and The principal findings pertain to how different types Michael Woolcock. of social networks in poor communities are deployed (ompletion date: Ongoing. for different purposes in managing risk and opportunity. The findings have implications for how external agents Reports (governments, aid agencies, nongovernmental organi- Brautigam, Deborah, and Michael Woolcock. Forthcoming. "Small zations) can best intervene in poor communities and States in a Global Economy: The Role of Institutions in Man- how they can leverage-or even build-social ties con- aging Vulnerability and Opportunity in Small Developing necting the poor to markets and to formal institutions. Countries." In Mansoob Murshed, ed., Globalization and the This work directly informed the World Bank's World Obstacles to the SuccessfulIntegration ofSmall Vulnerable Economies. Development Report 2000/2001: Attacking Poverty (New New York: Oxford University Press. Poverty and Social Development 37 Grootaert, Christiaan, Michael Woolcock, and Deepa Narayan. . 2001. "The Place of Social Capital in Understanding Forthcoming. "Integrated Questionnaire for the Measurement Social and Economic Outcomes." In John Helliwell, ed., The of Social Capital." World Bank, Poverty Reduction and Eco- Role of Human Capita/andSocialCapitalin EconomicGrowth and nomic Management Network, Washington, D.C. Well-Being. Paris: Organisation for Economic Co-operation Rao, Vijayendra, and Ana Maria lbahiez. 2001. "The Social Impact and Development and Human Resources Development of Social Funds: Participation and Collective Action in the Canada. Jamaica Social Investment Fund." World Bank, Development . 2001. "Social Capital in Strategies for Global Economic Research Group, Washington, D.C. Development and Development Cooperation." Paper Ritzen, Jo, and Michael Woolcock. Fcrthcoming. "Social Cohesion, presented at international workshop on Social Capital and Public Policy, and Economic Growth: Implications for Coun- Civic Activities, Bundestag, Berlin, Germany, June 25. tries in Transition." In Boris Pleskovic and Nicholas Stern, .2001. SocialCapita/in Theory and Practice: ReducingPoverty eds., Annual World Bank Conference on Development Economics, by Building Partnerships between States, Markets, and Civil Society. Europe. Washington, D.C.: World Bank. Paris: United Nations Educational, Cultural, and Scientific Ritzen, Jo, William Easterly, and Michael Woolcock. 2000. "On Organization. 'Good' Politicians and 'Bad' Policies: Social Cohesion, Insti- . 2001. "Social Theory, Economic Development, and the tutions, and Growth." Policy Research Working Paper 2448. Politics of Ideas." Paper presented at Brown University, World Bank, Development Research Group, Washington, D.C. Watson Institute for International Studies, Providence, R.I., Szreter, Simon, and Michael Woolcock. Forthcoming. "Social March 8. Capital and Public Health: Bringing the Stare and Society . 2001. "Stronger Communities, Better Workers, Higher Back In." Journalof Epidemiology and Public Health. Productivity: The Role of Social Capital in Enhancing Woolcock, Michael. 2000. "How and Why Planners Should Take Competitiveness." Keynote address at Fourth National Social Capital Seriously." Paper presented at annual meeting Symposium on Competitiveness and Productivity, Pereira, of the Associated Collegiate Schools of Planning, Atlanta, Ga., Colombia, February 15. November 3. . 2001. "Structures and Distribution of Social Capital: .2000. "The Role of Social Capital in Shaping Survival and Themes and Issues." Paper presented at Economic and Social Mobility Strategies in Poor Communities." Paper presented at Research Council Seminar Series on Social Capital, University annual meeting of the Poverty Reduction. and Economic of Warwick, England, March 30. Management Network, World Bank, Washington, D.C., . 2001. "What Is Social Capital and What Role Does It November 21. , Play in Economic and Social Development?" Paper presented - 2000. "Social Capital, Economic Development, and the at seminar on social capital, Department of Social, Community, Politics of Ideas." Paper presented at University of Tasmania, and Family Affairs, Dublin, Ireland, March 29. Department of Education, Launceston, Australia, October 2; . 2001. "Why Lawyers (and Everyone Else) Should Care and at Tasmania Department of State Development and about Social Capital." Keynote address at World of Opportu- Premier's Office, Hobart, Australia, October 4. nities Forum, Harvard University Law School, Cambridge, - 2000. "Social Capital: What Does It Mean for Public Mass., March 17. Policy?" Paper presented at University of Canberra, National . 2001. "Why Should We Care about Social Capital?" Institute for Governance, Canberra, Australia, September Canberra Bulletin of PublicAdministration 98: 17-19. 27. . Forthcoming. "Why and How Planners Should Take - 2000. "Theory and Methods in the Study of Social Social Capital Seriously." Journal of the American Planning Capital." Paper presented at Romanian Academic Society's Association. Summer Institute in Public Policy Evaluation, Sinaia, Woolcock, Michael, and Deepa Narayan. 2000. "Social Capital: Romania, September 1. Implications for Development Theory, Research, and Policy." - 2001. "The Place of Social Capital in Understanding World Bank Research Observer 15(2): 225-49. Social and Economic Outcomes." Canadian Journal of Policy . 2000. "Capital Social?" Informativo MensualdelaSociedad Research 2(1): 11-17. Nacional de Minerifa, Petrdleo, y Energia 10(3): 42-43. 38 Poverty and Sotial Development Woolcock, Michael, and Larry Pruzak. 2000. "Social Capital in Woolcock, Michael, Anthony Bebbington, Scott Guggenheim, and Nations and Organizations: A Contrasting Dialogue." Paper Elizabeth Olson. Forthcoming. "Mapping Discourses, Explain- presented at member workshop, IBM Institute for Knowledge ing Action: Social Capital in Theory and Practice at the World Management, Santa Fe, N.M., October 24. Bank." In Anthony Bebbington, Michael Woolcock, and Scott Woolcock, Michael, Sabiria Alkire, and Vijayendra Rao. Guggenheim, eds., Practical Theory, ReflectiveAction: SocialCap- Forthcoming. "Culture and Development Economics: Theory, italandDevelopmentProjectsatlthe World Bank. New York: Oxford Evidence, Implications." I'olitica Externa (Foreign policy review, University Press. Romania). Poverty and Sociol Development 39 Health and Population The Impact of the AIDS Epidemic on the Health the physical well-beingof the elderly-raising incomes, of the Elderly in Northwestern Tanzania improving road infrastructure, and preventing outbreaks of communicable disease. In Sub-Saharan Africa, as a result of the AIDS epidemic, The findings were presented to an African policy rhe elderly often become the caregivers for adult children audience at African Development Forum 2000 in Addis stricken with HIV/AIDS, the guardians of orphaned Ababa, Ethiopia, in December 2000, and at the annual grandchildren, and substitute workers for ill or deceased meetings of the Population Association of America, in adults in the home and on the farm. But until now most Washington, D.C., in March 2001. Results are also avail- of what has been known about the impact of AIDS on able on the Web at http://www.iaen.org. the elderly has been anecdotal, based on case studies of Responsibility: Development Research Group, Public those highly affected and without comparison to a Services for Human Development-Martha Ainsworth control group. Are the impacts identified in these (mainsworth@worldbank.org). With Julia Dayton, Yale studies typical? Who is most seriously affected among the University, Department of Epidemiology and Public elderly? What policies can be most cost-effective in Health. The Joint United Nations Programme on helping those affected? HIV/AIDS (UNAIDS) contributed funding for the research. This study analyzed the impact of mortality among Completion date:June 2001. prime-age adults on the nutritional status of the elderly, as measured by body mass index (BMI), while control- Reports ling for other factors that affect nutritional status. Ainsworth, Martha, and Julia Dayton. 2000. "Is the AIDS Epidemic Drawing on data from a longitudinal household survey Having an Impact on the Coping and Health Status of the in the Kagera region of Tanzania in 1991-94, the study Elderly? Evidence from Northwestern Tanzania." Background analyzed the BMI of 695 people over the age of 50 who paper for African Development Forum 2000, Addis Ababa, were observed at least once and as many as four times. It Ethiopia, December. estimated the determinants of BMI in a single period and - . 2001. "The Impact of the AIDS Epidemic on the Health those of changes in BMI over time (first differences), con- of the Elderly in Northwestern Tanzania." Policy Research trolling for unobserved individual heterogeneity. Working Paper 2649. World Bank, Development Research The analysis showed that the elderly in nonpoor Group, Washington, D.C. households have higher BMI, but their households are more likely to have an adult death and they are more Optimal Policies for Controlling Infectious Diseases likely to suffer a decline in BMI before the death. The elderly in both poor and nonpoor households experi- To characterize the essential features of optimal strate- ence a significant drop in BMI following an adult death, gies for controlling infectious diseases, this research but BMI recovers over time and there is no long-run project merged two distinct literatures: the mathemati- association between BMI levels and recent adult deaths. cal epidemiology literature that models the transmis- The elderly in communities with poor road infrastruc- sion of disease with sets of differential equations and the ture have substantially lower BMI, and those in literature on the optimal control of such sets of equations communities with recent epidemics of communicable widely used in economics. Previous work by the research disease (in the previous six months) had short-term team had shown that obtaining general analytical reductions in BMI. These findings suggest three policies, solutions to the problem is difficult and that further in addition to HIV/AIDS prevention, that could improve progress would require numerical analysis of alternative 40 specifications. Other policy-oriented analyses of these trient deficiencies, food security, demography, and models in the literature are flawed by overly simple poverty. The ANDI Web site (http://www.africa characterizations of policy and behavior. To correct this nutrition.net) allows users to compare African countries flaw, this study analyzed differences in the scale on several indicators and provides access to subnational- economies of control technologies and individual behav- level information. ior in preventing illness and in seeking care once ill. ANDI has developed a novel approach to databases The study selected and solved a few common that has the potential to yield savings in both cost and models of disease transmission, then examined the time in obtaining good-quality information. ANDI allows timing and intensity of the policies involved. The mod- users to go to a single site rather than to the separate sites els of directly transmittable diseases (such as tuberculosis of the participating agencies, and because it uses data that or sexually transmitted diseases) yielded subsidies on are already available and are continuously updated by preventive and curative interventions that were the same these agencies, the maintenance costs are negligible. A (in percentage terms). This outcome was somewhat midterm evaluation in 2001 concluded that ANDI was surprising given the general presumption of a larger answering a clear need and proposed that it continue with public sector role for preventive than for curative its focus on Africa, trying also to create links with oper- measures. This presumption is still justified if public ational information. actors can take certain actions (such as providing true The World Bank served as a catalyst for the initiative, public goods) that individuals cannot take. But the proposing the effort and funding work on the technical externality of actions that both public and private actors aspects of developing the interagency database. The can carry out is just as large for preventive as for curative United Nations Administrative Committee on Coordi- measures. nation, Subcommittee on Nutrition is coordinating the Models describing common vector-borne diseases work of the participating United Nations agencies. A (such as malaria or schistosomiasis) generated a set of memorandum of understanding for database sharing has rich results that yielded complex mixes of preventive been signed by the Food and Agriculture Organization (vector control) and curative activities. Most of the strate- (which has provided parallel financing for this effort), the gies can be characterized as relying more on vector United Nations Children's Fund, and the World Bank, control measures early on, and more on curative care with the United Nations Educational, Scientific, and later. But sensitivity analyses show substantial variation Cultural Organization and the World Health Organiza- in parameter values, and some combinations of para- tion as cooperating partners. meters change this order. ANDI has provided a good foundation for preliminary Responsibility: IDevelopment Research Group, Public Eco- work on models for in-country nutrition information sys- nomics-Jeffrey S. Hammer (jhammer@worldbank.org). tems in three pilot countries: Eritrea, Mali, and Zambia. With Mark Gersovitz, Johns Hopkins University. A nutrition information system is being established in Completion date: June 2001. Mali, and discussions on doing so in Eritrea are under way. These efforts will build local capacity to use the infor- Africa Nutrition Database Initiative mation for better research and more effective planning, monitoring, and evaluation of nutrition programs. Lack of access to good-quality nutrition information for Responsibility: Africa Region Technical Families, Human Africa has hampered the formulation of effective nutri- Development 2-Richard Seifman (rseifman tion policies and strategies and impeded proper planning, @worldbank.org) and Alexandre Abrantes, and Institu- monitoring, and evaluation of nutrition programs. To tional and Social Policy-Antoine Simonpietri. With address this problem, the Africa Nutrition Database Dario Berardi, Data Analysis Support Center, Italy. The Initiative (ANDI) was launched in February 1997. This Italy-Special Studies Consultant Trust Fund contributed initiative was aimed at creating a user-friendly database funding for the initiative. on the Web covering anthropometric indicators, micronu- Completion date: August 2001. Health and Population 41 Poverty and Health across the income distribution. Pakistan and India, for example, have similar rates of under-five mortality, but 'rhis research project takes as its starting point the two Pakistan outperforms India on the index of achieve- principal objectives of the World Bank's Health, Nutri- ment because its under-five mortality varies much less tion, and Population Team: to improve health and nutri- across income groups. tion outcomes for the world's poor and to reduce the The decomposition method was applied in an analy- impoverishing effects of adverse outcomes. Using data sis of the sources of the increase in inequality in from household surveys, the research aims to shed light malnutrition in Vietnam between 1993 and 1998. The on questions raised by these objectives. For example, how results indicate that much of the rise in inequality large are the gaps in health and nutrition outcomes was due to an increase in the impact of income on between the poor and the nonpoor in different countries? malnutrition and to increased income inequality at the Why do the gaps exist, and what explains the differ- community level. Results of an analysis of health care ences in the size of such gaps across countries and over payments in Vietnam indicate that although payments time? What role do inequalities in access to health are broadly proportional to household consumption, they services play, for example? What effects have programs nonetheless have an impoverishing effect, driving already and policies had on the differences in outcomes? To poor households further into poverty. what extent do adverse health and nutrition outcomes Methods developed by the project have begun to be impoverish households, and how do the effects vary used by researchers participating in three international across countries? What role do health insurance and initiatives, sponsored by the European Commission, by prepayment schemes play in cushioning households the Pan-American Health Organization, and by the Rock- from these impoverishing effects? efeller Foundation. In addition, the methods have been In the past year the study has extended methods, used in a number of developing countries, often in work based on concentration indexes, that allow analysts to funded by the U.K. Department for International Devel- specify attitudes toward inequality in health (by income opment. All these initiatives involve local researchers. level) and to compute an index of achievement combining The project's work has been presented in the World information on the mean level of health and the degree Bank Institute's health economics and health financing of inequality in health between the poor and the nonpoor. course and in a Health, Nutrition, and Population Team The study also extended a decomposition method that training course on health and nutrition programs and allows analysts to unpack the causes of changes in inequal- the poor. Results have been presented at a conference ity over time or of differences across countries or across of the Canadian Health Economics Association in Toronto regions within a country. (April 2001); to the World Health Organization Com- The research on the impoverishing effects of adverse mission on Macroeconomics and Health (April 2001); at health and nutrition outcomes has focused on the the U.K. Department for International Development in reduction in living standards associated with out-of- London (July 2001); at an International Health Eco- pocket payments for health services. It has investigated nomics Association congress in York, England (July 2001); different approaches to measuring such impacts, distin- and at seminars at the World Bank (September 2000), guishing among the analysis of fairness or equity in Princeton University (December 2000), the University health care payments, the analysis of the degree to which of Sussex (March 2001), Catholic University of Santiago, payments can be construed as "catastrophic," and the Chile (April 2001), and the Bank's Hanoi Country Office analysis of the impoverishing effects of health care (May 2001). payments. Responsibility: Human Development Network, Health, An application of the index of achievement to data on Nutrition, and Population Team-Adam Wagstaff health and nutrition indicators used in the international (awagstaff@worldbank.org). The Belgian and Japanese development goals showed the importance of looking Trust Funds contributed funding for the research. beyond national averages to the differences in outcomes Completion dale: August 2001. 42 Health and Population Reports 1990s in more than 20 Sub-Saharan African countries to Wagstaff, Adam. 2000. "Caring More about the Poor: Inequality explore the geographic relationship between poverty Aversion, Health Inequalities, and Health Achievement." and the incidence of fever, as well as the household-level World Bank, Human Development Network, Health, Nutrition, relationship between household wealth, the incidence of and Population Team, Washington, D.C. fever, and treatment-seeking behavior. Results show . 2000. "Measuring Equity in Health Care Financing: that poverty and the incidence of fever tend to be only Reflections on and Alternatives to WHO's Fairness of Financ- weakly related, but that a household's economic status ing Index." World Bank, Human Development Network, strongly influences treatment-seeking behavior. The Health, Nutrition, and Population Team, Washington, D.C. analysis also looks at household data from Eritrea, focus- .2001. "Economics, Health, and Development: Some Eth- ing on the links between household income, the reported ical Dilemmas Facing the World Bank and the International incidence of malaria, and its treatment. Community." Journalof MedicalEthics 27(4): 262-67. The second component, an analytical review of social . 2001. "Poverty and Health." Paper presented to the World and epidemiological conditions for malaria, looks at how Health Organization Commission on Macroeconomics and these conditions might affect policy prescriptions. T'his Health. review has led to several findings. First, epidemiologi- Wagstaff, Adam, and Nga Nguyen. 2001. "Poverty and Survival cal and ecological situations do not have a one-to-one Prospects of Vietnamese Children underDoiMoi." World Bank, relationship with economic situations, so public inter- Human Development Network, Health, Nutrition, and Pop- ventions need to incorporate information on both of ulation Team, Washington, D.C. these to promote malaria control. Second, strengthening Wagstaff, Adam, and Eddy van Doorslaer. 2001. "Paying for Health the health sector and improving the government's Care: Quantifying Fairness, Catastrophe, and Impoverishment, regulatory effectiveness are key to improving outcomes. with Applications to Vietnam, 1993-98." World Bank, Human Third, economic analyses should incorporate information Development Network, l-ealth, Nutrition, and Population on the behavior of households and the functioning of the Team, Washington, D.C. health sector, which cost-benefit and cost-effectiveness Wagstaff, Adam, P. Paci, and others. 2001. "Inequalities in Health: analysis typically do not do. And finally, more research Who You Are? Where You Live? Or Who Your Parents Were? is needed to ascertain which determinants of demand for Evidence from a Cohort of British 33-Year-Olds." World Bank, and supply of malaria control have the greatest impact Human Development Network, Health, Nutrition, and on its efficiency. Population Team, Washington, D.C. The third component of the project is a modeling Wagstaff, Adam, Eddy van Doorslaer, and others. 2000. "On exercise that combines epidemiological models with Decomposing the Causes of Health Sector Inequalities, with economic behavioral parameters to assess the sensitiv- an Application to Malnutrition Inequalities in Vietnam." World ity of policy recommendations to assumptions about Bank, Human Development Network, Health, Nutrition, and these parameters. The exercise uses a tractable choice Population Team, Washington, D.C. theoretic approach (relating to prevention and treatment options) to generate the policy implications of different The Economics of Malaria patterns of behavior, as well as the emergence of drug resistance. According to recent estimates, malaria leads to 1.5-2.7 The fourth component has analyzed monthly data million deaths and around 600 million clinical cases a year. on malaria incidence and interventions in 45 malaria This research project is aimed at contributing to the zones over five years in the Solomon Islands. Since the renewed effort to fight the disease. It examines data are nonexperimental, ordinary least squares multi- economic aspects of malaria and malaria control, with the ple regression produced biased estimates of the impact goal being to improve policy recommendations. of the interventions, especially for house spraying with The project has several components. The first uses DDT Applying instrumental variables estimation Demographic and Health Survey data collected in the improved the plausibility of the results, which suggested Heolkh and Populafion 43 that DDT spraying has contributed more than three ear relationships, with the mortality rate falling steeply other common interventions to the observed decline in relative to "wealth" at various points in the wealth dis- the incidence of malaria over the period. tribution in most countries. This effect is not noticeable The fifth component involves working with the with the common nonparametric measures of quantiles, African Economic Research Consortium (AERC) to since those confound the effect of income on health identify and sponsor African researchers to carry out with the distribution of income. analyses of the relationship between malaria and poverty. For a subset of countries for which data are available The AERC has commissioned studies now under way on the distribution of income or consumption in terms in Cameroon, Kenya, Nigeria, and Zambia. of purchasing power parity U.S. dollars, the study has Another activity has focused on the economic impli- scaled the wealth index to that distribution to yield a cations of the emergence of drug resistance. Since the common consumption standard. It then assesses the malaria parasite is increasingly resistant to drugs now used relative contribution of country-specific and common to treat malaria, combination therapy is being patterns. This analysis shows that the more uniform the advocated. This activity is developing a framework for measure of income, across and within countries, the assessing the conditions under which the economic greater the impact of income on health appears to be benefits of combination therapy would make it the among the very poor. preferred policy option. * Efficiency. Private health care is characterized by Responsillity Development Research Group, Public Services two major types of market failure and the interaction for Human Development-Deon Filmer (dfilmer between them: the principal-agent problem associated @worldbank.org) and Varun Gauri, and Infrastructure and with a health care provider's knowledge, and the general Environment-Mead Over and Maureen Cropper. With absence of insurance markets, due primarily to adverse Christie Poulos, University of North Carolina; Lakshmi K. selection and moral hazard. This study attempts to Raut, Yale University; Fadi Balesh, McGill University, develop tools for modeling markets with these charac- Montreal; Bernard Bakotee, government of the Solomon teristics, measuring the welfare loss from the lack of Islands; and Patricia Graves, University of Colorado. insurance and assessing the effect of different payment Completion date: June 2002. systems (capitation, salaried positions, fee for service) on health care. The study also attempts to measure the The Public Economics of Health Reform externalities in the health sector, for example, in the impact of neighbors' education on fertility and This project encompasses research efforts that apply mortality. standard techniques of public economics-and develop * Imp/ementability. In part as a result of the efficiency and apply extensions to those techniques-to the problems, the public sector has always been a major problems of the health sector in developing countries. provider of medical services. For standard reasons in The research formulates the problems in a way consis- public economics-the inherent market failures- tent with a public expenditure review-that is, it public intervention is warranted. But the quality of examines the equity, efficiency, and implementability of public provision has ranged from good to horrendous. This health sector policies and expenditure regimes. study looks at alternative incentive arrangements for * Equity. Based on analyses of about 60 Demographic the delivery of public services, focusing initially on and Health Surveys, this component has introduced the decentralization and contracts with nongovernmental innovation of extracting a measure of wealth from these organizations. A particular problem it addresses is the surveys. It uses this index to examine the relationship allocation of doctors to rural areas. between health measures (infant and child mortality, Responsibility' Development Research Group, Public Services nutritional status, incidence of disease) and family wealth, for Human Development-Jeffrey Hammer (jhammer using a design-adaptive nonparametric regression. Results @worldbank.org), Varun Gauri, and Deon Filmer. With for infant and child mortality show pronounced nonlin- Paolo Belli, Harvard University; William Jack, George- 44 Heolth and Population town University; and Maria Eugenia Bonilla-Chacin, Policy Research Working Paper 2624. World Bank, Develop- Johns Hopkins University. ment Research Group, Washington, D.C. (ompletion date: Ongoing. Hammer, Jeffrey S., and Maria Eugenia Bonilla-Chacin. 2001. "Diminishing Returns of Income on Health." World Bank, Reports Development Research Group, Washington, D.C. Belli, Paolo. 2001. "How Adverse Selection Affects the Health Jack, William G., and Jeffrey S. Hammer. 2001. "The Design of Insurance Market." Policy Research Working Paper 2574. Incentives for Health Care Providers in Developing Coun- World Bank, Development Research Group, Washington, D.C. tries." Policy Research Working Paper 2547. World Bank, Gauri, Varun. 2001. "Are Incentives Everything? Payment Mech- Development Research Group, Washington, D.C. anisms for Health Care Providers in Developing Countries." Health and Population 45 Fducafianp Laho; and EmdIymw The Impact of Labor Market Policies results with lessons drawn from other research on labor mar- and Institutions on Economic Performance ket policies and institutions in developing countries. Responsibilily: Development Research Group, Public Services This study, the second phase of a research project that for Human Development-Martin Rama (mrama assessed the effect of labor market interventions on labor @worldbank.org). With Raquel Artecona; lyabode Fahm; market outcomes and aggregate economic performance, Donna Maclsaac; Alema Siddiky; Jean-Paul Azam, Uni- aimed to improve the understanding of how best to versit6 des Sciences Sociales, Toulouse; Ann Bartel and approach the reform of such interventions. The study Ann Harrison, Columbia University; Kristin Butcher, focused on a set of well-defined departures from the undis- Boston College; Alex Cukierman, Tel Aviv University; torted, partial equilibrium model of the labor market: min- Francesco Daveri, University of Brescia, Italy; Alvaro imum wages, mandated benefits (or nonwage costs), payroll Forteza, Universidad de la Repiublica, Uruguay; Anna taxation (including social security contributions), public sec- Fruttero and Guido Tabellini, Universita Bocconi, Milan; tor employment, job security, and unionization. Catherine Ris, University of Auvergne; Dani Rodrik, The research combined theoretical work, detailed case Harvard University; Cecilia Rouse, Princeton University; studies based on household- and plant-level data from Manisha Singh, University of Maryland; and Jan van countries in which labor market distortions are believed Ours, Tilburg University, Center for Economic Research, to be acute, and cross-country analysis. Several detailed Netherlands. studies were carried out for each labor market interven- Completion date: June 2001. tion. The cross-country analysis was made possible by the construction of a cross-country time-series database of Reports labor indicators, which comprises data collected from Azam, Jean-Paul, and Catherine Ris. 2001. "Rent-Sharing, Hold- cross-country and country-specific sources. The database Up, and Manufacturing Wages in C6te d'lvoire." Policy was used to extend the now standard growth regression Research Working Paper 2600. World Bank, Development analysis so as to take labor market policies and institutions Research Group, Washington, D.C. into account. The theoretical work focused on the polit- Bartel, Ann, and Ann Harrison. 1999. "Ownership versus Envi- ical economy of labor market distortions and its implica- ronment: Why Are Public Sector Firms Inefficient?" NBER tions for the design of economic reforms. Working Paper 7043. National Bureau of Economic Research, The cross-country database includes 44 indicators Cambridge, Mass. relating to labor force participation, employment Bell, Linda. 1995. "The lmpactofMinimum Wages in Mexicoand and unemployment, wages and productivity, conditions Colombia." Journal of Labor Economics 15(3): S102-35. of work and benefits, trade unions and collective Butcher, Kristin, and Cecilia Rouse. 2001. "Wage Effects of Unions bargaining, public sector employment, and labor and Industrial Councils in South Africa." Policy Research standards. It covers 121 countries and 11 five-year Working Paper 2520. WVorld Bank, Development Research periods, starting immediately after World War II. No Group, Washington, D.C. comparable labor market database is available to Currie, Janet, and Ann Harrison. 1997. "Sharing the Costs: The researchers and practitioners. Impact of Trade Reform on Capital and Labor in Morocco." The research findings have been disseminated through Journal of Labor Economics 15(3): S44-71. workshops, training sessions, and seminars for researchers, Daveri, Francesco, and Guido Tabellini. 1997. "Unemployment, government officials, and trade union leaders. A policy- Growth, and Taxation in Industrial Countries." CEPR Work- oriented volume is planned that would combine the study's ing Paper 1681. Centre for Economic Policy Research, London. 46 Forteza, Alvaro, and Martin Rama. 2001. "Labor Market Rigidity .2000. "DIsequilibre salarial dans les pays de la zone CFA: and the Success of Economic Reforms across More than 100 les politiques de l'emploi sont-elles responsables." Revue Countries." Policy Research Working Paper 2521. World Bank, d'Economie du Diveloppement, no. 4, pp. 35-77. Development Research Group, Washington, D.C. . 2000. "Wage Misalignment in CFA Countries: Are Labor Freeman, Richard. 1994. "A Global Labor Market? Differences in Market Policies to Blame?" Journal of African Economics 9(4): Wages among Countries in the 1980s." World Bank, Policy 475-511. Research Department, Washington, D.C. . Forthcoming. "The Consequences of Doubling the Gruber, Jonathan. 1997. "The Incidence of Payroll Taxation: Evi- Minimum Wage: The Case of Indonesia." Industrialand Labor dence from Chile." JournalofLaborEconomics 15(3): S72-101. Relations Review. Maclsaac, Donna, and Martin Rama. 1997. "Determinants of Rama, Martin, and Raquel Artecona. 2000. "A Database of Labor Hourly Earnings in Ecuador: The Role of Labor Market Reg- Market Indicators across Countries." World Bank, Development ulations." Journal of Labor Economics 15(3): S1 36-65. Research Group, Washington, D.C. Rama, Martin. 1994. "Flexibility in Sri Lanka's Labor Markct." Rama, Martin, and Guido Tabellini. 1998. "Lobbying by Capital Policy Research Working Paper 1262. World Bank, Policy and Labor over Trade and Labor Market Policies." European Research Department, Washington, D.C. Economic Review 42(7): 1295-1316. . 1995. "Determination des salaires au paradis: une analyse Revenga, Ana. 1997. "Employment and Wage Effects of Trade du marche du travail P'] le Maurice." Revue d'Economie du Liberalization: The Case of Mexican Manufacturing." Journal Deve1oppement 2: 3-27. of LaborEconomics 15(3): S20-43. 1995. "Do Labor Market Policies and Institutions Matter? Rodrik, Dani. 1997. "What Drives Public Employment?" NBER The Adjustment Experience in Latin America,and the Working Paper 6141. National Bureau of Economic Research, Caribbean." Labour, pp. S243-69. Cambridge, Mass. . 1995. "Unions and Employment Growth: Evidence Weiss,Yoram. 1996. "Growth and LaborMobility." InSolomon W. from Jamaica." World Bank, Policy Research Department, Polachek, ed., Research in LaborEconomics. Greenwich, Conn.: Washington, D.C. JAI Press. . 1997. "Distortions des marches des biens et du travail: Determinants et consequences." In Jaime de Melo and Patrick Income Support for the Unemployed: Guillaumont, eds., Commerce Nord-Sud, migration et delocalisa- Mandatory Severance Pay in Peru tion: cons6quences pour les salaires et l'emploi. Paris: Economica. .1997. "Imperfect Rent Dissipation with Unionized Labor." This research was part of a larger project on income Pub/ic Choice 93: 55-75. support programs for the unemployed in Latin America, .1997. "LaborMarket Institutions and the Seond-BestTar- undertaken jointly by the Latin America and the iff." Scandinavian Journal of Economics 99(2): 299-314. Caribbean Region's Office of the Chief Economist and 1997. "Organized Labor and the Political Economy of Prod- the Development Research Group. The objectives of the uct Market Distortions." WorldBankEconomicReview 11(2): 327-55. larger project were to improve the understanding of who .1997. "Trade Unions and Economic Performance: East Asia the unemployed are and to assess five income support and Latin America." In James McGuire, ed., Rethinking Devel- programs for the unemployed that have been tried in opment: EastAsia andLatin America. Los Angeles: Pacific Coun- middle-income countries: public works, unemployment cil on International Policy. insurance, mandatory severance pay, training for the . 1998. "How Bad Is Tunisian Unemployment? Assessing unemployed, and individual unemployment savings Labor Market Efficiency in a Developing Country." World accounts. Bank Research Observer 13(1): 59-78. This research looked at mandatory severance pay, . 1999. "The Sri Lankan Unemployment Problem Revis- focusing on Peru in the 1990s. As in many other devel- ited." Policy Research Working Paper 2227. World Bank, South oping countries, in Peru employers have a legal obliga- Asia Region, Poverty Reduction and Economic Management tion to compensate workers dismissed through no fault Sector Unit, Washington, D.C. of their own. The research assessed whether this oblig- Educalion, Labor, and Employment 47 ation is an efficient mechanism for providing income Although many cross-sectional studies have examined support to the unemployed. The analysis, which was these relationships, many have documented associations based on individual records from a household survey between preschool malnutrition and subsequent attain- with a panel structure and relied on five indicators of ments, not causal relationships. coverage, found that in the private sector roughly one in This study explored these issues using longitudinal five workers, and one in three wage earners, are legally data collected in rural Zimbabwe between 1982 and entitled to severance pay. Coverage is more common 2000. Using a variety of estimators, the study produced among wealthier workers. Results based on several empir- preliminary results showing that improved anthropo- ical strategies suggest that workers "pay" for their metric status for preschool-age children, as measured entitlement to severance pay through lower wages. by height for age, is associated with earlier enrollment Results also show that consumption among unemployed in school, a lower likelihood of grade repetition, faster workers who receive severance pay is 20-30 percent progress through school, the completion of more grades, greater than among those who do not. In fact, and greater height as a young adult. But the study did consumption among these workers is higher than not find a robust association between preschool anthro- consumption among employed workers, suggesting that pometric status and achievement on Zimbabwe's primary mandatory severance pay is overly generous in Peru. school leaving examination. The findings of the larger project were disseminated The results also show that short-term shocks (such as through a conference in Washington, D.C., in June 2000, conflict and drought) have long-term consequences for and through a report released at that conference (David nutrition and for education, suggesting a need for invest- de Ferranti, Guillermo E. Perry, Indermit$..QiUl,,nd Luis ment in both social protection and nutrition programs. Serven, Securing Our Future in a Global,E4,conomy, Wash- The results were presented at a seminar at the Inter- ington, D.C.: World Bank, 2000). national Food Policy Research Institute in Washington, Responsibility Development Research Group,Public Services D.C., on June 26, 2001, and at a World Bank seminar for Human Development-Martin Rama (mrama attended by research staff and the Zimbabwe country @worldbank.org) and Donna Maclsaac. The Canadian team on June 28, 2001. A seminar is planned for August Trust Fund contributed funding for the research. 2001 in Harare, Zimbabwe. Completion date: June 2001. Responsibility: Rural Development Department and Devel- opment Research Group, Public Services for Human Report Development-Harold Alderman (halderman@worldbank. Maclsaac, Donna, and Martin Rama. 2000. "Mandatory Sever- org). With John Hoddinott, Dalhousie University, Hali- ance Pay: Its Coverage and Effects in Peru." Policy Research fax, Canada; and Bill Kinsey, University of Zimbabwe. Working Paper 2626. World Bank, Development Research Completion date: July 2001. Group, Washington, D.C. Report Assessing the Impact of Early Childhood Alderman, Harold, John Hoddinott, and Bill Kinsey. 2001. "Assess- Malnutrition on Academic Achievement ing the Impact of Early Childhood Malnutrition on Academic Achievement and Adolescent Height in Rural Zimbabwe." Improving the health and nutrition of preschool-age World Bank, Development Research Group, Washington, D.C. children is widely recognized as an important develop- ment objective in its own right. If improving the Evaluation of Active Labor Market nutritional status of preschool-age children increased Programs in China the acquisition of knowledge at school and led to greater height as an adult, it would also have instrumental value As reforms of state enterprises in China have accelerated, where positive associations existed between schooling and addressing labor redundancy in these enterprises has productivity and between height and productivity. gained urgency. But downsizing has been hindered by 48 Education, Labor, and Employment enterprise-based provision of social benefits to workers. ence with active labor market policies, and evaluation A practical response to this problem has been the emer- methodologies. A conference is planned to disseminate gence of a new category of workers, xiagang workers, the study's results to policymakers and practitioners in who are laid off but retain their links to their enterprises China. as well as subsistence wages and access to housing and Responsibility: East Asia and Pacific Region, Poverty Reduc- medical benefits. tion and Economic Management Sector Unit-Tamar The government's main response to the unemploy- Manuelyan Atinc (tmanuelyanatinc@worldbank.org) and ment threat has been the Reemployment Project, Benu Bidani. With Christopher O'Leary, W. E. Upjohn launched as a pilot in 30 cities in 1994 and expanded to Institute for Employment Research; and Institute of 200 cities in 1995-96. The project encompasses a range Labor Studies, Beijing. of active labor market policies, including retraining, job Completion date: October 2001. search assistance and counseling, wage subsidies and tax breaks for enterprises that employ laid-off workers, Migration, Remittances, and Poverty and assistance for self-employment designed to redeploy in Latin America unemployed workers in productive activities. This research is evaluating the cost-effectiveness of Migration is one of several ways through which house- retraining and employment services in two;Chinese holds adapt to such structural shifts as the transformation cities, Shenyang and Wuhan. The study uses a quasi- of a rural agricultural economy into an urban industrial experimental evaluation design that involves collecting one. Households may use migration, which offers the data from individuals who participated in retraining and promise of higher earnings, as a strategy for escaping data from a comparison group drawn randomly from a list poverty. They may use migration to overcome credit of retrenched workers. Using these data, the study com- constraints due to imperfect markets or to overcome pares such outcomes for the groups as probability of other forms of rationing, such as lack of access to edu- reemployment and earnings after reemployment to derive cation and health services. And they may use migration the net impact of the programs. Using data on the costs -permanent or temporary, internal or international of administering the programs, it also compares.the cost- -as a strategy for coping with both macroeconomic and effectiveness of different interventions., idiosyncratic shocks. The random sample of retraining participants was This study aims to provide a balanced view of the gains drawn from program lists kept at the training institutes. from and costs of migration by analyzing its impact on For the comparison group a random sample was drawn both monetary and nonmonetary outcomes. It is both from the lists of redundant workers maintained at the reviewing the literature and undertaking new research municipal labor bureaus. The Institute of Labor Stud- on the geographic determinants of income and other ies completed the data collection for a sample of partic- indicators of well-being, the impact of rural cash trans- ipants and for the comparison group in May-June 2000 fers on migration, and the impact of remittances on and has also collected the data on program costs. inequality, education, and housing. The World Bank already funds projects with active Preliminary results from the research show that remit- labor market components, and such programs are.increas- tances are used to fund public goods-that is, goods that ingly being considered in several East Asian economies benefit both the migrant and the family at home. The affected by the economic crisis. This study will add to marginal impact of remittances on children's enrollment the body of research on thei evaluation of active labor mar- in school and on investments in housing (two goods ket policies, and its results will be useful in designing and likely to enter the utility function of the migrant) exceeds evaluating such interventions in other countries. that of other sources of income for the family at home. A conference on labor market policies was held in The study has found that remittances increase inequal- Beijing in May 1999 to discuss the urban labor. market icy in poor areas but reduce inequality in better-off areas. adjustment in China, lessons of international experi- And while cash transfers to rural farmers reduce perma- Edutation, Labor, and Employment 49 nent migration, they have little or no impact on tempo- done because of their high costs and because of public rary migration. officials' reluctance to vary the level of inputs among Responsibility: Latin America and the Caribbean Region, schools. This research takes advantage of an opportunity Poverty Sector Unit-Quentin Wodon (qwodon in Kenya to perform randomized evaluations of several @worldbank.org) and Corinne Siaens. With Gabriel Gon- different education policy options in collaboration with zalez Konig, American University; and Victoria Malkin, an international nongovernmental organization. New York University. The study began with 14 schools in 1995. From these, 7 (ompletion date: October 2001. were randomly chosen to receive the standard package of assistance. Pretests were given at the beginning of 1995, and Reports posttests in the fall of 1995. Preliminary results show a large Gonzalez Konig, Gabriel, and Quentin Wodon. 2001. "Do Cash increase in attendance, probably due to the provision of Transfers to Rural Farmers Reduce Migration?" World Bank, free school uniforms, which effectively lowered the price of Latin America and the Caribbean Region, Poverty Sector Unit, schooling. However, test scores showed no improvement. Washington, D.C. The project expanded in 1996 to 100 schools, of which . 2001. "Remittances and Income Inequality." World Bank, 25 were randomly chosen to receive a large number of Latin America and the Caribbean Region, Poverty S.ector Unit, textbooks. Pretests were administered in early 1996, and Washington, D.C. posttests near the end of that year. The children were - . 2001. "Remittances, Schooling, and Cbjl!d Labor." World followed for three more years (1997-99) to assess whether Bank, Latin America and the Caribbean. Fegion,. Poverty the initial effects endure for several years. Preliminary Sector Unit, Washington, D.C. results suggest that the impact of textbooks is less strong Makdissi, Paul, and Quentin Wodon. 2001. "Migration,.Poverty, than some earlier studies had indicated. Of particular inter- and Housing: Welfare Comparisons Using Sequential Sto- est is the finding that only the top 20 percent of students, chastic Dominance." World Bank, Latin America and the as identified by the scores on the 1996 pretests, benefited Caribbean Region, Poverty Sector Unit, Washington, D.C. from the provision of textbooks. The textbooks had little Wodon, Quentin, Gabriel Gonzalez Konig, and Corinne Siaens. effect on dropout rates or grade repetition. 2001. "Remittances and Housing." World Bank,. Latin Amer- In 1997 another 25 of the 100 schools were selected ica and the Caribbean Region, Poverty Sector Unit, Washing- to receive block grants that could be spent on several ton, D.C. options, such as textbooks, other school supplies, or Wodon, Quentin, Gabriel Gonzalez Konig, Victoria Malkin, and construction of new classrooms. The purpose of this Corinne Siaens. 2001. "Migration, Remittances, and Poverty." intervention was to see whether funds are used more World Bank, Latin America and the Caribbean Region, Poverty effectively when schools are given a choice on how to Sector Unit, Washington, D.C. spend them. Analysis of the results indicates a small but statistically significant impact on test scores after one year, Improving Primary Education in Kenya: but the impact seems to diminish over time. As with A Randomized Evaluation of Different textbooks, the better students seem to have benefited Policy Options the most. The results show statistically significant effects in reducing dropout rates and increasing repetition rates. Many econometric studies have tried to estimate the In 1998 another randomization was done for all 100 effect of education policies on school performance in schools. Fifty schools participated in a program in which industrial countries. But even the most sophisticated teachers received rewards if their students' performance econometric techniques may yield biased estimates if on standardized tests improved; the other 50 schools schools with different levels of inputs also differ sys- did not participate. Preliminary results show statistically tematically in other, unobserved ways. Randomized significant effects on test scores after two years. experiments overcome many of the statistical problems Findings have been presented at Brown, Cornell, inherent in these econometric studies, but they are rarely Harvard, Hebrew, Michigan, Minnesota, Toronto, and 50 Edutation, Labor, and Employment Yale Universities, at the MacArthur Foundation, and at ard inherent in traditional schemes. And it could con- the International Food FPolicy Research Institute. ceivably provide the same protection to the unemployed. Responsibility: Development Research Group, Public But too little is understood about how the system func- Services for Human Development-Harold Alderman tions to know for which groups of workers, and under (halderman@worldbank.org). With Paul Glewwe, Uni- what conditions, this would hold true. versity of Minnesota; Michael Kremer, Harvard Uni- Moreover, application of the system in developing versity; Sylvie Moulin, Hillary Rodham Clinton Women's countries would raise a host of issues: Empowerment Center, Morocco; Stacy Nemeroff; Nau- * How feasible is the system? Unemployment insur- man Ilias, University of Pennsylvania; and Alexander ance savings accounts eliminate pooling across individ- Wolfson and Eric Zitzewitz, Massachusetts Institute of uals. If a significant share of workers could not generate Technology. The International Christian Humanitarian sufficient savings to draw on during spells of unem- Services, the Netherlands; and the National Science ployment, such a system might not be viable. Foundation, United States, are contributing funding for * What distributional effects would the system have? the research. In particular, what effects would it have on consumption Completion date: December 2001. smoothing, observed income distribution, personal sav- ings, and private transfers? Reports * What redistributional effects would replacing tra- Glewwe, Paul, Michael Kremer, and Sylvie Moulin. "Textbooks ditional unemployment insurance with the savings and Test Scores: Evidence from a Prospective Evaluation in account system have? Research shows that such effects Kenya." Draft. are likely to be small for the United States, but what can Glewwe, Paul, Nauman Ilias, Michael Kremer, and Sylvie Moulin. be expected for developing countries? "Teacher Incentives and Student Outcomes: Evidietce from a * What is the optimal design of a savings account sys- Randomized Evaluation in Kenya." Draft. . tem? For example, what rules should govern the with- Glewwe, Paul, Lakshmi Iyer, Michael Kremer, and Sylvie Moulin. drawal of benefits? "School Funding and Student Performance: Evidence from The research investigates these issues through sim- Kenya." Draft. ulation studies and theoretical modeling. To examine Glewwe, Paul, Michael Kremner, Sylvie Moulin, and Eric Zitzewitz. whether the lack of cross-pooling would make unem- "Do Wall Charts Improve Student Performance in Primary Schools? ployment savings accounts nonviable, and how replac- Evidence from a Randomized Evaluation in Kenya." Draft. ing a traditional system with the savings account system Kremer, Michael, Sylvie Moulin, David Myatt, and Robert would affect income distribution, the study simulates the Namunyu. "The Quantity-Quality Tradeoff in Education: introduction of unemployment accounts in Estonia, Evidence from a Prospective Evaluation in Kenya." Draft. Malaysia, and Slovenia (countries for which rich panel data exist for a representative sample of labor force partici- Providing Unemployment Benefits through pants). The simulation consists of applying the rules of Unemployment Insurance Savings Accounts unemployment insurance savings accounts to a repre- sentative sample of labor force participants over their Traditional unemployment insurance systems are prone entire working life and determining contributions and to moral hazard, creating adverse incentives that reduce withdrawals for each worker. job search efforts and increase unemployment. This Preliminary results for Estonia and Malaysia show research examines an alternative income support scheme, that the savings account system is a viable alternative to based on unemployment insurance savings accounts, in traditional unemployment insurance. For example, which workers would be required to save a fraction of their assuming a 3 percent contribution rate and a 60 percent earnings and draw unemployment compensation from replacement rate, the Estonian study shows that only 9-17 these savings. By internalizing the costs of unemployment percent of workers would end their working life with a benefits, the system is expected to avoid the moral haz- negative cumulative balance in their savings account, and Eduatoion, Labor, and Employment 51 that 30-45 percent would experience a negative balance In the first year of the project (1998) deworming at least once during their working life (by assuming no treatment was provided to 25 of the 75 schools. The changes in hazard rates from unemployment, these results remaining 50 were a control group. During the second underestimate the viability of the savings account and third years (1999 and 2000) treatment was provided system). to the initial 25 schools plus an additional 25. Twenty- Using theoretical modeling, the study aims to five schools thus remained as a control group. (These last formulate an optimal design for an unemployment 25 schools are currently receiving deworming treatment.) insurance savings account system by evaluating This experimental design allows an unbiased estimation alternative specifications of its parameters, including of the impact of deworming treatment on education the rules for contributions and withdrawals, the limit on outcomes-school attendance, grade progression, dropout account balances, and the mix of financial sources. The rates, and academic performance. modeling will incorporate the basic tradeoff between A randomized evaluation suggests that the school- the provision of insurance and the incentives for based mass treatment with deworming drugs reduced reemployment. By deriving optimal consumption paths absenteeism by a quarter, with especially large gains and corresponding reemployment probabilities, the study among the youngest children. There is no evidence that will investigate the extent to which a fine-tuned savings the deworming treatment improved academic test scores, account system could adequately approxin,ate such an however. optimal scheme. The study will also. e,etermine the By reducing disease transmission, the deworming optimal combination of mandatory savings bvworrkers and also created large positive externalities-health and payroll taxes paid by employers. The modeling of finan- school participation benefits-among untreated cial sources will be based on the fact that introducing children in the treatment schools and among children in unemployment insurance savings accounts Improves neighboring schools, suggesting a rationale for public incentives for reemployment while also.imposing costs deworming subsidies. Existing experimental studies, in on workers by distorting intertemporal consumption which deworming treatment is randomized among indi- decisions. viduals in the same school, find small and insignificant Responsibility: Human Development Network, Social effects on education; however, these studies underesti- Protection Team-Milan Vodopivec (mvodopivec mate true treatment effects if deworming creates @worldbank.org). With Hugo Hopenhayn, University positive externalities for the control group. The findings of Rochester and Universidad Torcuato Di Tella, Buenos suggest that deworming is an attractive public policy Aires; Jungyoll Yun and Yoonjung Koh, Ewha University, under a variety of approaches. Seoul; Tomaz Rejec, Jozef Stefan Institute, Ljubljana, Responsibility: Development Research Group, Public Services Slovenia; Ulle Pettai, Statistical Office of Estonia; and for Human Development-Harold Alderman (halderman Zala Kalan, Statistical Office of Slovenia. @worldbank.org) and Sylvie Moulin; and Human Devel- Complelion date: March 2002. opment Network, Education Team-Donald Bundy. With Michael Kremer, Harvard University; Edward The Impact of Deworming Treatment on Primary Miguel, University of California at Berkeley; Simon School Performance in Busia, Kenya Brooker, Imperial College, London; Alfred Luoba, Vec- tor Borne Diseases, Kenya Division; and Internationaal This study is evaluating the effect of mass deworming Christelijk Steunfonds, Nairobi. treatment (treating all students in a school) on education Completion date: April 2002. outcomes in Busia District, Kenya. The deworming treatment was phased in at 75 primary schools with an Report enrollment of about 30,000 students. The treatment was Miguel, Edward, and Michael Kremer. "Worms: Health and Edu- provided to randomly assigned schools, while the schools cation Externalities in Kenya." University of California at not selected for treatment served as a control group. Berkeley, Department of Economics. 52 Education, Labor, and Employment Educational Enrollment and Dropout fall into distinctive regional patterns. In some regions the poor reach nearly universal enrollment in first grade, but Patterns of educational enrollment and attainment vary then drop out in droves, leading to low attainment (typi- greatly across countries and, within countries, across cal of Latin America). In other regions the poor never population groups (particularly wealth groups). For some, enroll in school (typical of South Asia and West Africa). The basic education is practically universal; for others, edu- study found enormous differences across countries in the cational attainment is dismal. This research has docu- "wealth gap," the difference between rich and poor in mented these differences using a unique collection of enrollment and educational attainment. In some countries comparable household data sets. The research first the difference in median years of school completed is addressed a methodological issue-establishing the valid- only I or 2 years; in others it is 9 or 10. The attainment ity of ranking householcls by their wealth status in the profiles can be used as diagnostic tools to examine issues absence of data on household consumption expendi- in the education system, such as where to focus inter- tures. The study then conducted several analyses: inves- ventions or the extent to which low enrollment is due to tigating the relationship between education outcomes and unavailability of schools. wealth status in 35 countries, identifying determinants The research used NFHS data from India to esti- of education outcomes in states of India, and examining mate the determinants of child enrollment (for those the interaction of gender and wealth in generating ages 6-14) and educational attainment of a recent cohort inequalities in educational enrollment and attainment in (ages 15-19). The analysis found enormous gaps in 40 countries. enrollment and attainment between children from Cross-country comparisons are often hinderecl by lack rich and poor households. While 82 percent of children of comparability across data sets. There,eare;data sets from rich households complete grade 8, only 20 percent collected using almost identical methods-the.,Demo- of those from poor households do. The wealth gaps graphic and Health Surveys (DHS) for a large number vary widely across states of India. Moreover, gender of developing countries and the National Family Health differences exacerbate the gaps: while 80 percent of Survey (NFHS) for India. But the data do not,include girls from rich households complete grade 8, only 9.5 consumption expenditures, usually used,to rank house- percent of girls from poor households do. The physical holds by economic status. This research established an presence or absence of schools in rural villages alternative approach to ranking households. As a proxy explains only a small part of differences in enrollment. for long-run household wealth, it constructed a linear Across states there are huge gaps in enrollment index from a set of asset indicators using principal com- rates between observationally equivalent households, ponents analysis to derive the weights, then validated the especially among the poor. For example, enrollment index using data from India, Indonesia, Pakistan, and rates in Kerala are 44 percentage points higher than for Nepal. The study found that the asset index.,has rea- observationally equivalent poor households in Bihar. sonable coherence with current consumption expendi- The study used DHS data to investigate the ways in tures and, most important, works as well as, or better than, which gender and wealth interact in generating within- traditional expenditure-based measures in predicting country inequalities in enrollment and attainment in enrollment status. This index approach has allowed 40 countries. In addition, it carried out a multivariate reanalysis of existing data by researchers in the World analysis in each country to assess the independent effects Bank and elsewhere, focusing on socioeconomic inequal- of the education of adults in the household and the ities in human development outcomes. presence of schools in the community, and how these The study used DHS data from 44 surveys (in 35 interact with gender. The study found a large female countries) to document differences in the enrollment and disadvantage in education in countries in South Asia attainment patterns of children from rich and poor house- and North, West, and Central Africa. But while gender holds, ranked by the asset-based index. It found that the gaps are large in a subset of countries, wealth gaps are enrollment: profiles of the poor differ across countries but large in almost all the countries. The education of adults Education, Labor, and Employment 53 in the household has a significant relationship with the Paper 2268. World Bank, Development Research Group, enrollment of children in all the countries, and the rela- Washington, D.C. tionship with women's education is larger than that with . 2001. "Inequalities in Education: International Experi- men's in some but not all of the countries. The presence ence." In United Nations Educational, Scientific, and of a primary and secondary school in the community has Cultural Organization (UNESCO), Encyclopedia of LifeSupport a significant relationship with enrollment in some Systems. countries (notably in West and Central Africa), and the Filmer, Deon, and Lant Pritchett. 1999. "Educational Enrollment relationship does not appear to differ systematically with and Attainment in India: Household Wealth, Gender, Village, the gender of the child. and State Effects." Journal of Educational Planning and This research has led to two new strands of investi- Administration 13(2): 135-64. gation. First, the study will look at the diffusion of edu- . 1999. "The Effect of Household Wealth on Educational cation and the use of public services. Assessments of the Attainment: Evidence from 35 Countries." Population and impact of education often include the education of only Development Review 25(1). heads of household (or of mothers) in regressions of . 2001. "Estimating Wealth Effects without Expenditure individual and household behaviors and outcomes. Since Data or Tears: An Application to Educational Enrollments in the education of other people can also influence behav- States of India." Demography 38(1): 155-32. iors and outcomes, this research will use comparable (DHS) data from many poor countries to examine whose Gender Data and Labor Markets education matters. Second, methodological work will look at the implica- Labor market outcomes differ by gender in Latin tionsofusingnontraditionaldatatoanalyzeinequalitiesin America, as they do in all regions, but there is scant human development outcomes. This research will compare evidence on why they differ and what implications the consumption-based welfare measures and a variety of differences have for the well-being of men and women. other measures for many countries using data sets with both Moreover, little is known about how these differences types of measures. It will address several questions: How have evolved over the past 10 years of economic reform. closely correlated are the measures? How do different This research project identifies the changes in labor measures affect conclusions about the behaviors and out- market outcomes for men and women in Latin America comes under study? What are the issues in adjusting wel- over the past decade and examines the importance of fare measures to control for household size and composition? gender roles in shaping these outcomes. And are there special purposes forwhich non-consumption- The project has two parts. First, it is constructing a based measures can be or should not be used? database of gender-disaggregated statistics on a variety Results have been presented at many sessions in the of labor market and demographic indicators to permit World Bank and at the annual meetings of the Popula- comparisons across countries and over time. The data- tion Association of America. The findings are available base draws on household and labor market surveys pro- on the Web at http://www.worldbank.org/research/ viding statistics on wages, education, time use, gender projects/edattain/edattain.htm. roles, social capital, household structure, labor force par- Responsithy Development Research Group, Poverty Team- ticipation, and informal sector employment for 15 coun- Deon Filmer (dfilmer@worldbank.org); and Africa Region tries. Statistics on health, violence, and gender expecta- Technical Families, Regional Human Development- tions and attitudes have been compiled from other Jee-Peng Tan. With Lant Pritchett, Harvard University. sources. The data are being formatted into a Web-based Completion date: May 2002. database with a search engine. Second, the project is undertaking an empirical study Reports of the causes of observed differences in wages, unem- Filmer, Deon. 1999. "The Structure of Social Disadvantage in ployment, entrepreneurship, labor force participation, Education: Gender and Wealth." Policy Research Working informal sector employment, and sectoral allocation of 54 Education, Labor, and Employment labor between men and women and within each gender Luz Saavedra, University of South Florida; Diana Prieto group. The study draws on the database and comple- and Diego Angel, Georgetown University; and Maria ments it with econometric analysis of the determinants Lourdes Gallardo, Columbia University. The Danish, of the differences in labor market outcomres. The results Netherlands, and Norwegian Trust Funds have con- are used to build alternative explanations for gender dif- tributed funding for the research. ferences in the labor market by taking into account the (ompletion date: June 2002. opportunities and constraints created by gender roles and how these shape labor market choices and outcomes. Reports The study will include all countries in the database to Arias, Omar. "Are All Men Benefiting from the New Economy? permit cross-country cornparisons. Male Economic Marginalization in Argentina, Brazil, and Costa Three preliminary results have emerged from the Rica." World Bank, Latin America and the Caribbean Region, ongoing work: Washington, D.C. * Women's and men's labor market behavior and Cunningham, Wendy V. 2000. "Sectoral Allocation by Gender in returns are becoming more similar, but that does not the 1990s: Evidence from Argentina, Brazil, and Costa Rica." necessarily mean that women are better off than they World Bank, Latin America and the Caribbean Region, were before. New challenges are arising for both men and Washington, D.C. women in the home and workplace as workloads increase . "Breadwinner versus Caregiver: Labor Force Participation and wages become more equal. and Sectoral Choice over the Mexican Business Cycle." World * The interface between household roles and gender Bank, Latin America and the Caribbean Region, Washington, is increasingly important in explaining observed labor D.C. Draft. market outcomes. Often there are more similarities in . "Mexican Female Small Firm Ownership: Motivations, labor market behavior between those with similar roles Returns, and Gender." World Bank, Latin America and the in the household (primary breadwinner or primary care- Caribbean Region, Washington, D.C. Draft. giver) than between those of the samrre sex. Cunningham, Wendy V., and Raquel Artecona. 2001. "Labor * The most vulnerable groups-women, informal Market Outcomes in Latin America and the Caribbean: Recent sector workers, and the least educated-saw the most Trends and the Role of Gender." World Bank, Latin America rapid gains in wages, employment, and formal sector and the Caribbean Region, Washington, D.C. employment in the 1990s. Women experienced partic- Saavedra, Luz. 2000. "Female Wage Inequality in Latin American ularly rapid gains. Labor Markets." World Bank, Latin America and the Caribbean The study's results challenge the conventional think- Region, Washington, D.C. ing on what causes gender differences in the labor mar- ket and suggest alternative policies that could be more Educating Nomadic Herders out of Poverty effective in narrowing those differences. The database has provided inouts to several World The world's nomadic herders, mainly in Africa, the Bank reports, including the Policy Research Report Middle East, and South, Southwest, and Central Asia, Engendering Development-Through Gender Equality in number in the tens of millions and are among the poor- Rights, Resources, and Voice (New York: Oxford University est and most vulnerable population grcups. Rcaching Press, 2001). The database will be available on the Web them with formal schooling is difficult, and millions of at http://www.worldbank.org/lacgender in December nomadic pastoral children remain outside the education 2002. system. This situation will continue until more effective Responsibility: Latin America and the Caribbean Region, ways are found to bridge the gap between what formal Poverty Reduction and Economic Management Sector education systems teach and what pastoral children need Unit and Gender Sector Unit-Wendy Cunningham to know. (wcunningham@worldbank.org) and Raquel Artecona. This research project will investigate alternative With Omar Arias, Inter-American Development Bank; approaches to education for nomadic pastoralists. It aims Education, Labor, and Employment 55 to improve the understanding of the relationship between of displacement. And it used the principal-agent approach culture, indigenous knowledge, social institutions, and to identify the self-selection mechanisms appropriate poverty in the context of nomadic pastoralism, and to iden- when individual productivity is not observable in the pub- tify the conditions under which more effective schooling lic sector. could be provided for those wishing to remain in the However, the research made it obvious that several pastoral livelihood system. The research will focus on adja- important policy questions had not been addressed. It also cent areas of northern Uganda and northern Kenya. showed that the answers to some of the questions that Building on a literature review on nomadic education were addressed have never been applied in practice, so commissioned by the World Bank's livestock group, the that the potential obstacles to their implementation are research will investigate three questions: unknown. And it left unclear whether the partial imple- * How is indigenous knowledge embedded in social mentation that can be expected in practice would sub- organizations? stantially improve the outcome of downsizing operations. * What are the outcomes of formal education for nomadic This second phase of the research combines further pastoralists, particularly its effect on social capital? analytical and empirical work with experimentation in * What alternative approaches to education could downsizing operations. The experimentation provides an better equip pastoralists to fight poverty? opportunity to assess how public sector units adjust The research is expected to lead to a better under- under different downsizing strategies and whether the standing of how indigenous knowledge is linked to social assistance received by redundant public sector workers structure and economic expectations, how to integrate is sufficient to cope with the welfare loss from job sep- indigenous knowledge and formal education in efforts to aration. The main research advantage of the experi- prepare nomadic pastoralists for a future within the pas- mentation and evaluation component is that it toral economy or outside it, and what policies and inter- circumvents some of the typical identification problems ventions governments, the World Bank, and other in econometrics. The approach also ensures that the development agencies might pursue to do so. research has a direct impact on operations. Because of the Responsibility: Social Development Department-Kreszen- time span of downsizing operations, some of the results tia Duer (kduer@worldbank.org) and Sabina Alkire. With of the research will not be available for several years. Jeremy Swift and Saverio Kratli, University of Sussex. Present efforts focus on setting up information systems, The Netherlands Trust Fund is contributing funding for including baselines and tracer surveys. the research. The analytical and empirical work in this second (ompletion date: December 2002. phase of the research involves ex ante assessment of labor redundancies, across public sector units and across Public Sector Retrenchment, Phase 2 countries. It also involves evaluating the impact of the downsizing of public enterprises on their labor produc- Economic reforms often require shrinking or divesting tivity and on their privatization prices. Other work looks public sector agencies and enterprises, which in turn at the use of labor market information in designing sometimes requires massive layoffs. But experience assistance packages for redundant workers, estimates shows that the process can be badly mishandled. This the economic returns to downsizing operations when research looks at how to deal with redundant labor in the product markets are imperfectly competitive, and public sector. investigates the gender implications of public sector The first phase of the research investigated the opti- downsizing. mal design of downsizing operations, combining three Findings have been disseminated through numerous approaches. It used the public economics perspective to presentations on public sector downsizing for policy- assess the financial and economic returns to downsizing. makers in many countries. More recently, a module on It used the labor economics perspective to predict the public sector downsizing was prepared for a World Bank losses public sector workers may experience as a result Institute course on labor market policies for developing 56 Educafion, Labor, and Employment countries, and a pilot version of the course was delivered Chong, Alberto, and Martin Rama. 2001. "Do Compensation Pack- in Dublin, Ireland, in October 2000 to policymakers ages Need to Be That Generous? Public Sector Workers in from Central Asia and Eastern Europe. The first full Guinea-Bissau." In Shantayanan Devarajan, F. Halsey Rogers, version of the course was offered in Washington, D.C., and Lyn Squire, eds., World Bank Economists' Forum. Vol. 1. in April 2001. In addition, the project has developed Washington, D.C.: World Bank. an Excel-based toolkit, the Downsizing Options Simu- Haltiwanger, John, and Manisha Singh. 1999. "Cross-Country Evi- lation Exercise (DOSE), to predict the effects of down- dence on Public Sector Retrenchment." World Bank Economic sizing strategies combiningearly retirement and severance Review 13(1): 23-66. pay. Country-specific versions of the toolkit have been Jeon, Doh-Shin, and Jean-Jacques Laffont. 1999. "The Efficient prepared for Guinea-Bissau, Guyana, Morocco, and Mechanism for Downsizing the Public Sector." World Bank Vietnam. Economic Review 13(1): 67-88. Resporsibilily: Development Research Group, Public Services . 2001. "Labor Market Information Acquisition and Down- for Human Development-Martin Rama (mrama sizing." Universite des Sciences Sociales, Toulouse, France. @worldbank.org). With Sarah Bales; Patrick Belser; Levy, Anat, and Richard McLean. 1996. "Optimal and Subopti- Alberto Chong; Alice Hong; Elizabeth Ruppert; Ragui mal Retrenchment Schemes: An Analytical Framework." Assaad, University of Minnesota; Kaushik Basu and Gary Rutgers University. Fields, Cornell University; Jose Manuel Campa, New Rama, Martin. 1997. "Efficient Public Sector Downsizing." Finance York University; John Haltiwanger and Manisha Singh, and Development 34(3): 40-43. (Also issued as Policy Research University of Maryland; Doh-Shin Jeon, UJniversitat Working Paper 1840, World Bank, Development Research Pompeu Fabra, Barcelona; Jean-Jacques Laffont, Uni- Group, Washington, D.C., 1997.) versit6 des Sciences Sociales, Toulouse; Anat Levy, Cen- . 1999. "Public Sector Downsizing: An Introduction." World tral Bank of Israel; Florencio L6pez-de-Silanes, Harvard Bank Economic Review 13(1): 1-22. University; and Aysit Tansel, Middle East Technical . 2000. "Downsizing in the Presence of Monopoly Rights: University, Ankara. The Road to Riches." World Bank, Development Research Completion date: June 2003. Group, Washington, D.C. - .2001. "The Gender Implications of Public Sector Down- Reports sizing: The Reform Program of Vietnam." Policy Research Assaad, Ragui. 1999. "Matching Compensation Payments with Working Paper 2573. World Bank, Development Research Worker Losses in the Egyptian Public Sector." World Bank Group, Washington, D.C. Economic Review 13(1): 117-54. Rama, Martin, and Donna Maclsaac. 1999. "Earnings and Welfare Bales, Sarah, and Martdn Rama. 2001. "Are Public Sector Workers after Downsizing: Central Bank Employees in Ecuador." World Underpaid? Appropriate Comparators in a Developing Coun- Bank Economic Review) 13(1): 89-116. try." World Bank, Development Research Group, Washing- Rama, Martin, and Kinnon Scott. 1999. "Labor Earnings in One- ton, D.C. Company Towns: Theory and Evidence from Kazakhstan." Basu, Kaushik, Gary Fields, and Shub Debgupta. 1996. "Retrench- World Bank Economic Review 13(1): 185-209. ment, Labor Laws, and CGovernment Policy: An Analysis with Robbins, Donald, Martfn Gonzalez Rosada, and Alicia Menendez. Special Reference to India." Cornell University, Ithaca, "Public Sector Retrenchment and Efficient Severance Payment N.Y. Schemes: A Case Study of Argentina." Harvard Institute for Belser, Patrick, and Martin Rama. 2001. "State Ownership and International Development, Cambridge, Mass. Labor Redundancy: Estimates Based on Enterprise-Level Ruppert, Elizabeth. 1999. "The Algerian Retrenchment System: Data from Vietnam." Policy Research Working Paper 2599. A Financial and Economic Evaluation." World Bank Economic World Bank, Development Research Group, Washington, D.C. Review 13(1): 155-84. Campa, Jose Manuel. 1996. "Public Sector Retrenchment: Spain Tansel, AysiL "Eamings of Turkish Workers before and after Dismissal in the 1980s." New York University. due to Privatization." Middle East Technical University, Ankara. Educalion, Labor, and Employmenl 57 Economic and Environmental Impacts of Lowland Responsibility: Environment Department-Stefano Pagiola Agricultural Development on Poor Upland (spagiola@worldbank.org). With Jerry Shively, Purdue Communities in Palawan, Philippines University; and Richard Yao, University of the Philippines at Los Banios. Purdue University contributed staff time Through a detailed case study, this research investigated to the research. how intensification of agricultural production is affect- (ompletion date: June 2001. ing patterns of labor demand and pressures on forest areas in the province of Palawan, a frontier area of the Factors Affecting the Pace of Deforestation Philippines. To do so, it administered a survey to 200 in Northern Thailand households to extend a small panel data set previously collected in a low-income farming community. When seeking policies to alter the rate of deforestation, The results of the analysis show that the increased off- it is important to distinguish two competing theories of farm opportunities provided by lowland irrigation were agricultural expansion-the population-driven, or sub- more likely to accrue to larger and poorer households, that sistence, model and the market-driven expansion, or the resulting income gains were likely to lead to fertil- perfect markets, model. According to the subsistence izer purchases, and that concomitant increases in labor model, agricultural decisions are made by households that productivity probably undermined incantives to clear are imperfectly mobile and have limited opportunities forest in subsequent periods. for off-farm work. In the simplest such model each The insights generated by the project should help in household's goal is to achieve a fixed level of consump- identifying potential "win-win" investments and possible tion, either by consuming what it produces or by selling tradeoffs between poverty reduction and environmental its output in the market. Two key predictions of this objectives. The results are most directly applicable in model are that an increase in agricultural population will East Asia, where the combination of extensively cultivated increase the area cultivated and that an increase in agri- uplands and intensively cultivated lowlands, like that cultural productivity will reduce the area cultivated. An found in the study area, is particularly common. immediate implication is that programs aimed at inten- The work was carried out in collaboration with sifying agricultural production, such as subsidies for researchers from the University of the Philippines at inputs or provision of seeds for high-yielding varieties, Los Banos. The results will be disseminated in the will ease the pressure on forests. Philippines through presentations (for example, to the The perfect markets model leads to the opposite Department of Agriculture's Bureau of Agricultural implication. This model assumes that complete Research, to the Philippine Institute of Development labor markets exist, that on- and off-farm labor are per- Studies, and in the academic community) and through fect substitutes, and that agricultural households maxi- articles in academic journals. The results will also be dis- mize profits. The amount of agricultural land cultivated seminated through working papers, World Bank seminars, does not depend directly on population, but instead and World Bank Institute training activities, as well as depends on output and input prices, especially the through presentation at the annual meetings of the Amer- wage rate. So in the perfect markets model an ican Agricultural Economics Association. The panel data exogenous increase in output per hectare increases set generated by the research, including the two previ- the cultivated area, implying that programs to raise ous rounds of data collection, will be made available to farm incomes may have adverse environmental any interested researcher. consequences. 58 This study used data for 361 villages in reserved Responsibility: Development Research Group, Infrastruc- forests in Chiang Mai, a province in northern Thailand, ture and Environment-Maureen Cropper (mcropper to examine factors affecting the rate of agricultural expan- @worldbank.org). With Jyotsna Puri, University of Mary- sion between 1986 and 1996. The key question from a land. Data for the study were obtained from Jariya Boon- policy perspective is whether this agricultural expan- jawat and Chetpong Buttathep, Chulalongkorn sion was population driven-due to increases in the University, Bangkok. number of hill tribe people who practice shifting Completion dote: June 2001. cultivation--or market driven-due to increases in production (and thus acreage) for cash crops. Report In reality, elements of both models are likely to Cropper,Maureen,JyotsnaPuri,and CharlesGriffliths. 2001. "How affect agricultural expansion within villages in Chiang the Location of Roads and Protected Areas Affects Deforesta- Mai. The study therefore attempted to explain variation tion in North Thailand." Policy Research Working Paper 2583. in agricultural land across villages as a function of World Bank, Development Research Group, Washington, village population, the price of agricultural outputs, the D.C. cost of transporting goods to market, and policy vari- ables such as credit availability. Because the impact Economic Instruments for Habitat Conservation of these variables may differ from one crop to another, the study estimated separate models to explain areas As pressures for both agricultural expansion and devoted to paddy rice, short-run crops, and long-run biodiversity protection increase, land must be allocated crops. and managed as efficiently as possible to meet Preliminary findings suggest that paddy rice and short- these two needs. This project is aimed at improving run crops (primarily maize and soybeans) are more the ability to meet those dual needs through three sensitive to increases in population than are long-run crops activities: (primarily upland rice), but the elasticity of acreage * Developing a quantitative, operational definition of planted with respect to population is well below one for biodiversity useful for assessing large-area conservation both rice (0.3) and short-run crops (0.6). All three types policies. of crops are sensitive to changes in the cost of * Developing and demonstrating a methodology for transporting goods to market, but the effects are small. assessing biodiversity and development tradeoffs arising Programs to promote high-yielding varieties of paddy rice from land allocation decisions or policies. appear to have reduced acreage planted, while making * Applying that methodology to assess the potential credit available through the Bank of Agriculture and environmental impact of economic instruments for con- Agricultural Cooperatives has increased acreage devoted servation, with particular attention to incentive pay- to rice and short-run crops. ments for land uses consistent with conservation. Results have been presented at several conferences: The study site for the research is the Atlantic Forest the American Agricultural Economics Association meet- of southern Bahia, in Brazil, an area that exemplifies the ings in Tampa, Florida (August 2000); the conference on issues that arise when economic pressures threaten Global Change and Sustainable Development in South- important biodiversity resources. The Atlantic Forest east Asia organized in Chiang Mai, Thailand, by the has an extremely high level of biodiversity; many Southeast Asian Regional Committee for START conservation biologists rank it among the habitats with (System for Analysis Research and Training) in collabo- highest priority for conservation. The Bahian section of ration with the United Nations Development Programme the forest is now reduced to relatively small fragments, (February 2001); and the Land Cover and Land Use making up perhaps7percentof the forest's original area. Change Conference in Chiang Mai, Thailand, spon- Less than a fifth of this is protected. The remaining sored by the National Research Council of Thailand area is threatened by conversion, often to uses with low (August 2001). economic value and little impact on employment. Environment 59 The project has built a spatially explicit bioeconomic Reporls model. The model's economic component represents Hardner, Jared. 1999. "Land Use Trends and Conservation Oppor- landholder responses to alternative economic instru- runities in the Atlantic Forest of Southern Bahia, Brazil." World ments and assesses the instruments' economic and social Bank, Development Research Group, Washington, D.C. impacts. The biological component assesses the impact Stoms, David, and Frank Davis. 2001. "Economic Instruments for of resulting land use configurations on the ecosystem, Habitat Conservation." taking into account both the representation of diverse species or elements and the viability of plant and animal Global Overlay Brazil populations. The model uses secondary data, including geographic information system (GIS) data and data from This study analyzes emerging policy issues and the census and other sources, and primary data on land proposed economic mechanisms related to protecting values and biodiversity collected by Brazilian partners and establishing forests in Brazil, looking at the impli- under a parallel project. It is informed by focus group cations for biodiversity protection, mitigation of climate interviews with stakeholders on the acceptability of change, and the local economy. The study uses spatial alternative economic instruments. simulation and other methodologies to assess policy Initial results of the bioeconomic model suggest impacts. substantial flexibility in achieving ecological goals, with The study has three components: relatively low opportunity costs for assembling repre- * Transferable development rights and greater flexibility sentative sets of viable habitats. A draft version of the under the legal forest reserve requirement. Brazilian law model was presented to a broad group of regional stake- requires land owners to maintain at least 20 percent of holders at a workshop in June 2001. each property under natural forest cover as a forest The research is expected to contribute directly to reserve. There is active discussion about allowing prop- the design and implementation of the Centra! Corridor erty owners to meet this requirement for one piece of Project, administered by the World Bank for the Pilot Pro- property on another. This component examines the costs gram for the Tropical Brazilian Forests. And it will shed and benefits of alternative proposals for such a transfer- light on the feasibility of new, market-like approaches to able development rights system, focusing on the state of conservation that may be of broad interest to the Bank Minas Gerais. Analysis suggests that restricting enforce- and its clients. ment and trading of legal reserves to the largest proper- Responsibility: Development Research Group, Infrastruc- ties might substantially reduce compliance and ture and Environment-Kenneth Chomitz (kchomitz enforcement costs while satisfying much of the aggregate @worldbank.org) and Timothy Thomas. With Instituto forest cover requirement. de Estudios S6cioambientais do Sul da Bahia, Brazil; * Nativeforests, plantedforests, and coke in Minas Gerais. the Institute for Computational Earth System Science, This component examines a nexus of issues related to University of California at Santa Barbara; Wayt Thomas; the substitutability between charcoal from native wood- and Andre Mauricio de Carvalho, CEPLAC and Indus- lands, charcoal from plantations, and mineral coke in trial Economics. The Rain Forest Trust Administration meeting the energy needs of Minas Gerais's large iron and Unit is contributing funding for the research. Conser- steel industries, assessing policies that would internal- vation International (Brazil and the United States), the ize the environmental benefits of using sustainable Federal University of Minas Gerais, and Instituto de sources of energy. Estudios S6cioambientais do Sul da Bahia are con- * Land use in the Amazon. This component examines tributing data and staff time, and Instituto Brasileiro de the economic and environmental costs and benefits of Geografia e Estatistica is contributing data. The paral- various land use policies. The study has found that most lel project is funded by Programa Estadual para a Con- agricultural land in Amazonia has very low productivity servacao da Biodiversidade (PROBIO). and that under current forms of agriculture the wet West- Completion date: December 2001. ern Amazon is probably even less productive. Ongoing 60 Environment analysis examines the potential effectiveness of defor- Schneider, Robert R., Eug8nio Arima, Adalberto Verissimo, Paulo estation taxes and transferable development rights in Barreto, and Carlos Souza Junior. 2000. "Amaz6nia Susten- encouraging productive agriculture while preserving ravel: limitantes e oportunidades para o desenvolvimento rural." environmental assets. Serie Parcerias. World Bank, Brasilia Country Office. The research project is expected to contribute to policy discussions in Brazil, to the design and implemen- Nature Tourism's Contribution to Economic tation of World Bank projects relating to land use and Development and Conservation Finance sustainable development, and to worldwide discussions of innovativeapproachestoconservation.Preliminaryresults Improperly managed nature tourism can degrade or have been presented at seminars and meetings in Brazil destroy the environmental assets on which it is based. But and abroad and were featured in a Bank policy paper. if managed wisely, it offers potentially valuable oppor- Responsibility: [evelopment Research Group, Infrastruc- tunities for generating revenues, not only for develop- ture and Environment-Kenneth Chomitz (kchomitz ment but also for conservation. @worldbank.org) and Timothy Thomas. With Aline This study investigates the contribution that nature Tristao Bernardes; Antonio Salazar Brandao, Santa Ursula tourism can make to conservation and the economy. It University, Rio de Janeiro; Peter May, Federal Rural also examines the choices that determine the sustain- University of Rio de Janeiro; Instituto do Homem e ability of nature tourism assets and the process by which Meio Ambiente da Amaz6nia; Instituto Brasileiro de these assets are exploited in developing countries. The Geografia e Estatistica; and Funda,co Biodiversitas. The research explores these questions: Global Overlay Program (Danish Trust Fund) is con- * Is it better to raise conservation funds through tributing funding for the research, and Instituto Brasileiro changes in destination pricing or through taxes on the de Geografia e Estatistica is contributing data. tourism trade? Completion date: December 2001. * How should park entrance fees be structured? Should differential pricing be used (for example, for Reporls local and foreign visitors)? Arima, Eug8nio. 2001. "Incentivos fiscais e de cr6dito para pecuaria * At what level should entrance fees be set to maxi- na Amaz6nia Legal." mize the economic returns to both the site and the Bernardes, Aline Tristao. 1999. "Environmental Inspection, tourism sector? Enforcement, and Monitoring System, Minas Gerais, Brazil." * At what level should entrance fees be set to mini- . 1999. "Some Mechanisms for Biodiversity Protection in mize environmental degradation and its associated costs Brazil, with Emphasis on Their Application in the State of and to maximize revenues? Minas Gerais." * What are the main managerial options for improv- Chomitz, Kenneth. Forthcoming. "Transferable Development ing the resource base or minimizing degradation? Rights and Forest Protection: An Exploratory Analysis." Inter- * What are the net local economic benefits from national Regional Science Reviev. nature tourism? That is, what are the leakages and local Chomitz, Kenneth, and Timothy Thomas. Forthcoming. "Geo- multipliers involved? How much do the poor in local com- graphic Patterns of Land Use and Land Intensity in the Brazil- munities adjacent to parks benefit? Are there ways to ian Amazon." Policy Research Working Paper. World Bank, increase those benefits? Development Research Group, Washington, D.C. * Is there potential for local tourism activities by pre- FundasAo Biodiversitas. 2000. "The Use of Charcoal as an Ener- viously disadvantaged groups, and if so, how can they be getic Input for Siderurgy and Its Consequences for Environ- encouraged? mental Conservation in Minas Gerais, Brazil." There has been little applied economic analysis of the Gerwing,JeffreyJ., RodneySalomao, and Christopher Uhl. 2000. supply and demand relationships in nature tourism "Land Cover and Carbon Density Maps for the Brazilian Legal markets or of the relationships among the markets. For Amazon." this reason, this project uses a multimarket model Environment 61 approach. Where possible and appropriate, the: research Natal, in June 2001, to discuss the results of all these stud- attempts to incorporate environmental damage func- ies with key stakeholders and obtain their feedback. tions into the analytical framework, methodology, and case The studies will be revised and pulled together for an studies. Through evaluation of the links between nature integrated multisectoral policy analysis. tourism and the environment, it attempts to highlight Responsibility: Africa Region, Environmental, Rural, and potential threshold effects and critical features of the Social Development Department-Ernst Lutz (elutz relationship. @worldbank.org); and Envhionment Department-John Since the political, institutional, and managerial con- Dixon. With Bruce Aylward; Geert Creemers, Pete Good- texts play a fundamental role in the tradeoffs in nature man, and Stuart Ferrer, KwaZulu-Natal Nature tourism, the project seeks collaboration and cooperation Conservation Service, South Africa; Kreg Lindberg, Grif- with state agencies, the private sector, nongovernmen- fith University, Australia; Barry James, Brousse-James & tal organizations, and civil society to identify local Associates, South Africa; Dawie Mullins, Conningarth limitations. Nonetheless, as an effort in applied Consultants, South Africa; Aki Stavrou; and DRA- economic research, the project questions these limitations. Development, Durban, South Africa. The KwaZulu- The project has developed a general sectoral Natal Nature Conservation Service has made contribu- multimarket analytical framework that clarifies the main tions in kind. direct and indirect links between the economic, Completion dole: December 2001. environmental, social, and policy variables involved in nature tourism. The model attempts to respond to four Understanding and Improving the Environmental key issues: the sustainability of resource use, the relative Performance of China's Township-Village benefits from and interactions among multiple users, Industrial Enterprises the roles of the public and private sectors in nature tourism activities, and social equity. The goal was to In less than a decade China's township-village industrial develop a framework that would allow policy analysts to enterprises (TVIEs) have risen to near parity with the tra- assess these issues in different circumstances and regions. ditional industrial sector. Recent estimates suggest that The framework was applied to the northern area of the these small and medium-size rural enterprises now KwvaZulu-Natal Province in South Africa in partnership account tor about 50 percent of China's industrial out- with the KwaZulu-Natal Nature Conservation Service put and more than 30 percent of rural income. Emissions and the KwaZulu-Natal Tourism Organization. The from these enterprises remain only partially regulated, intent was to illustrate the utility of the model, particu- however, and China's State Environmental Protection larly in shedding light on the relative importance of Administration seeks a more systematic and compre- different policy variables in a particular ecological and hensive approach to pollution control for TVIEs. economic context. The application was also expected to This project will contribute to the discussion of demonstrate how parameters for the model relationships regulatory reform in several ways. First, its results should can be estimated and what data can reliably be found for provide useful insights for decentralization of the national this purpose. regulatory system. Reforms now under discussion may The project has carried out an on-site survey (in two involve changes in the administration of national regu- phases), an ecological survey, a geographic information latory standards, pollution charges, and monitoring and system (G.S) analysis, a producer survey, a focus group enforcement procedures. The research will point to survey involving previously disadvantaged groups, a appropriate local adjustments to respond to the gap household survey, social accounting matrix work, and an between actual and optimal emissions. origin-country survey (in the Netherlands). In addition, Second, the research will focus attention on pollution it is expected that a game sales survey will be und_rtaken. exposure risks for workers in TVIEs. The welfare of A hunter survey had to be dropped because of resource workers in heavily po!luting factories may be severely constraints. A workshop was held in Mkuzi, KwaZulu- affected even when community-level air quality is rela- 62 Environment tively good. The study will contribute to the under- Responsibilily. Development Research Group, Infrastructure standing of the issue by sharpening the overall cost- and Environment-Hua Wang (hwang@worldbank.org), benefit assessment through separate impact estimates for David Wheeler, Susmita Dasgupta, Jun Bi, Lei Liu, and Xun exposed workers and neighboring communities. In addi- Wu. With Fengzhong Cao, China's State Environmental Pro- tion, its econometric analysis will attempt to identify tection Administration; Zhifeng Yang, Beijing Normal Uni- the main determinants ofvwithin-plant pollution. Variables versity; Genfa Lu, Nanjing University; and Guangshou to be tested include sector, output, wages, vintage, Peng, Guizhou Environmental Protection Institute. human resources, environmental management strate- (ompletion dole: December 2001. gies, quality of environmental information, and alterna- tive employment opportunities. Making Long-Term Growth and Development Third, the project will provide insights into the impact More Sustainable of more general policy reforms on TVIE pollution. Research in other Asian countries has suggested that Development strategies generally emphasize sustained policies affecting input prices, worker education, enter- growth as a means of raising average incomes. But failure prise scale, and technology choices can have effects on to account for environmental degradation can erode the pollution similar in size to those of direct regulation. capital base for future development. This research aims Results from this research should promote a broader to improve the understanding of the links between eco- dialogue on the relationship between environmental nomic development and the environment and to identify and economic policy reforms in China. policies that will help maintain long-term growth prospects. Fourth, the project will use local survey data to assess The project, which focuses on case studies of Brazil, the impact of environmental information and the local Chile, and Sri Lanka, adopts a variety of approaches. Each capacity to use it effectively. The results should improve case study includes an action-impact matrix, a cost- the understanding of environmental information as a effective approach that uses available data (primarily source of pressure on TVIEs to control pollution. An from in-country sources, the World Bank, and the Inter- important role for this factor would suggest allocating national Monetary Fund) to show the key environmen- greater regulatory resources to collecting and dissemi- tal impacts of development policies. Other approaches nating appropriate environmental information. range from fairly simple models to quite complex (and Fifth, the project will provide insight into the use of thus costly) ones. The study compares their usefulness environmental information by TVIEs. Recent evidence and cost-effectiveness in modeling the links between from Mexico suggests that internal environmental man- growth and the environment. agement and training strategies can have a significant Initial work focused on organizing strong, in-country impact on factory emissions, because they increase teams of researchers who would be responsible for crit- responsiveness to regulatory incentives for pollution ical elements of the work, consistent with the project's control. The project will assess the importance of such emphasis on capacity building. It also included gather- strategies in TVIEs. Strong results would suggest a need ing data, prioritizing issues, identifying options, and for greater focus on pilot technical assistance and train- refining the analytical models. ing programs for plant-level environmental management The second phase of research is examining how econ- as complements to (but not substitutes for) more tradi- omywide, growth-inducing policies (such as liberaliza- tional regulatory development programs. tion), and the growth that ensues, may exacerbate the Personal interviews of community leaders, plant environmental harm caused by policy distortions (such managers, workers, and heads of household and the as subsidized prices for natural resources). The work is collection of plant-level pollution data have been also identifying conditions under which growth could completed, and the data are being analyzed. A project worsen the adverse environmental effects of market fail- workshop is scheduled for September 21-22, 2001, in ures. The three country case studies are in the final Beijing. stage of preparation. Environment 63 Several in-country workshops that included senior a stated amount. This question is preceded by exercises decisionmakers have been held, and more are planned to familiarize respondents with the concept of risk of for reviewing and disseminating the results of the study. dying and with their own baseline risk of dying over the If resources are available, a concluding conference next 10 years. Also preceding it is a section discussing will be organized to communicate the study's main measures that people ages 40-75 commonly take to findings. prolong their lives (such as screening for cancer and drug Responsibility: Environment Department, Office of the therapy for high cholesterol or high blood pressure) and Director-Mohan Munasinghe (mmunasinghe the quantitative risk reductions provided by such @worldbank.org); and World Bank Institute, Economic measures. Policy and Poverty Reduction Division-Jorge Araujo. This survey will be administered in six countries to With Noreen Beg; Peter Meier, IDEA Inc., United King- compare and contrast estimates of the value of a statis- dom; Chitru Fernando, ESI Inc., Sri Lanka, and Tulane tical life for people over the age of 40. Thus far the sur- University; Ronaldo Seroa da Motta and Claudio Ferraz, vey has been administered to 1,200 people in the United Instituto de Pesquisa Econ6mica Aplicada (IPEA), Brazil; States, 930 people in Canada, and 600 people in Japan. Carlos Young, Federal University of Rio de Janeiro; Current plans call for administering the survey in France, Osvaldo Sunkel and Carlos de Miguel, Center for the Italy, and the United Kingdom. Analysis of Public Policies, Chile; Shavi Fernando and Results from the Canadian and U.S. surveys are D. C. Wijeratne, Ceylon Electricity Board; Nimal Siri- remarkably similar, implying values of a statistical life of pala, Ministry of Finance, Sri Lanka; U. Sapukotane, between $950,000 and $3 million (U.S. dollars), lower than Ministry of Environment, Sri Lanka; and D. Chan- the values currently used by either Health Canada or the drasekere, Ceylon Petroleum Corporation. U.S. Environmental Protection Agency. The reason for (ompletion date: January 2002. these lower estimates is not, however, the age of the respondents. The study has found that the willingness Report to pay to avoid the risk of death eventually falls with age, Munasinghe, Mohan, Osvaldo Sunkel, and Carlos de Miguel. but this effect is modest: at age 70 it is 65 percent of the 2001. The Sustainability of Long-Term Growth: Socioeconomic and peak willingness to pay, which occurs at age 54. The lower Ecological Perspectives. London: Edward Elgar. willingness-to-pay value is consistent with the results of other stated preference surveys, which tend to produce Valuing Mortality Risk Reductions values for a statistical life lower than those in revealed preference studies. In most industrial countries the mortality benefits of Interestingly, having a chronic heart or lung condition environmental programs accrue primarily to older has no impact on willingness to pay. But a respondent's people. In the case of air pollution controls the age mental health score (based on a standardized health distribution of statistical lives saved parallels the age questionnaire) is a significant correlate of willingness to distribution of deaths, implying that 75 percent of pay: other things equal, people with higher mental health people saved are over 65 years old. Yet the most common scores are willing to pay more to increase their life method of valuing these risk reductions is to use expectancy. compensating wage differentials from the labor Results of the research have been presented at the market. American Economic Association meetings in Boston To remedy this problem, this study has developed a (January 2000); the Wharton School, University of Penn- survey that asks people ages 40-75 what they would pay sylvania (March 2000); North Carolina State University, to reduce their risk of dying. Specifically, it asks respon- Raleigh (May 2000); Florida State University, Tallahas- dents what they would pay for a drug (not covered by see (March 2001); the John F. Kennedy School of Gov- health insurance) that, if taken for the next 10 years, ernment, Harvard University, Cambridge, Massachusetts would reduce their chances of dying over this period by (May 2001); and the Summer Institute of the National 64 Environment Bureau of Economic Research, Cambridge, Massachu- person interviews), and different payment vehicles (util- setts (July 2001). ity bills, a trust fund, tourist fees). Responsibility: Development Research Group, Infrastruc- Surveys in Armenia have been completed, and the ture and Environment-Maureen Cropper (mcropper resulting data are being analyzed. Surveys in France and @worldbanlk.org). With Anna Alberini, University of the United States are expected to be finished soon. Maryland; Ronald Goeree and Bernard O'Brien, McMas- The findings from the research will serve as input to ter University, Ottawa; Alan Krupnick, Resources for government policy decisions relating to restoration of the Future, Washington, D.C.; and Nathalie Simon, U.S. Lake Sevan. In addition, the results from testing survey Environmental Protection Agency. The National methods will be useful for managers of projects dealing Science Foundation (United States), the U.S. Environ- with large nonuse or nonmarket environmental benefits mental Protection Agency, and Resources for the Future and for practitioners of environmental valuation, partic- have contributed funding for the research. Health Canada ularly in developing countries. provided funding for administering the Canadian Responsibility: Development Research Group, Infrastruc- survey. ture and Environment-Hua Wang (hwang@worldbank. Completion dote: April 2002. org), Maureen Cropper, Benoit Laplante, Craig Meisner, Xun Wu, Wenhua Di, and Yanghong Jin. With Vic Reporl Adamowicz and Dale Whittington. Krupnick, Alan, Maureen Cropper, Anna Alberini, Nathalie Completion dole: June 2002. Simon, Bernard O'Brien, Ronald Goeree, and Martin Heintzel- man. Forthcoming. "Age, Health, and the. Willingness to Pesticide Use in Brazil Pay for Mortality Risk Reductions: A Contingent Valuation Survey of Ontario Residents." Journal of. Risk and Growing use of chemical pesticides and fertilizers has Uncertainly. caused serious damage to human health and the envi- ronment in many developing countries. According to Measuring the Economic Value of Environmental estimates by the World Health Organization and United Protection Projects: Methodology and Application Nations Environment Programme, pesticide poisoning to Armenia's Lake Sevan injures 1-5 million agricultural workers a year. At least 20,000 workers die from exposure to pesticides every year, Lake Sevari, a large, high-altitude lake in Armenia with most of them in developing countries. Chemically pol- much symbolic, cultural, and historical importance to luted runoff from fields has contaminated surface and Armenians, has been significantly depleted over the past ground waters, damaged fisheries, destroyed freshwater 50 years. Withdrawals have lowered the water level and ecosystems, and created growing "dead zones" in the shrunk the surface area, destroying fisheries and reduc- ocean. ing tourism. The Armenian government is considering Despite growing awareness of the human health haz- actions to stop or partially reverse the damage. The costs ards and environmental degradation from pesticide use, of these actions are estimated to be high. The benefits policymakers in developing countries have shown little have not been estimated. response. The reason, in large part, is that most devel- This study is using contingent valuation methods to oping countries have little or no reliable information estimate the benefits related to the lake's recreational and about their pesticide application and the associated nonuse values. Through surveys, it is assessing the damage-and even less information about what deter- willingness of Armenians-in the country and in the mines pesticide use and how it would respond to regu- diaspora-and tourists to pay to prevent further degra- latory instruments. Until recently, lack of data has dation of the lake. It is also testing different formats for prevented systematic research in this area. eliciting value (open-ended, close-ended, likelihood), Drawing on a range of new data from Brazil, this different survey modes (mailed questionnaires, in- research attempts to help close that gap by quantifying Environment 65 the negative externalities of pesticide poisoning and Environmental Data Accounting ecosystem damage throughout the country and by draw- ing a correlation between these results and the spatial dis- This project investigates how the depletion of natural tribution of intensive pesticide use at the county resources and degradation of the environment by air (muniaipio) level. This work will give policymakers a and water pollution could be incorporated into a system more reliable measure of the problem and help to iden- of conventional national accounts to measure economic tify areas where the problem is most intense and where performance and sustainability. Through participation in targeted monitoring and intervention could yield the the London Group on Environmental Accounting, the maximum social benefits. Recognizing the wide varia- project supports international initiatives to revise the tion in use throughout Brazil, the analysis will attempt United Nations' Recommended System of Integrated to examine regional differences and offer alternative Environmental and Economic Accounts (SEEA). In con- strategies for a responsive agricultural (pesticide) policy. junction with related research activities by the United The research, a collaborative effort with Funda,ao Nations, Eurostat, the Organisation for Economic Co- Instituto Brasileiro de Geografia e Estatfstica (IBGE), operation and Development, and leading industrial coun- involves compiling data on pesticide use from secondary tries, the project is working to develop a general measure sources in Brazil, mapping comprehensive profiles of of sustainability. It is also reviewing different approaches pesticide application at the county level, modeling to creating a coherent set of environmental accounts regional differences in pesticide use in Brazil, and sys- and assessing how well each fits with internationally tematically tracking cases of pesticide poisoning and established economic accounting methods that link asset ecosystem damage. stocks with flows. The study has found that Brazil's agricultural growth All these activities are closely tied to efforts within the in the 1990s was associated with a more than .doubling World Bank to expand the range of environmental indi- of pesticide use, leading to serious damage to human cators, strengthen their relevance, and produce and refine health and the environment. The main message emerg- associated statistical series. This initiative thus forms ing from the findings is nevertheless hopeful. The results an important part of the Bank's commitment to provid- suggest that pesticide use in Brazil is heavily skewed ing suitable data for monitoring newly agreed on inter- toward a few cash crops for export, suggesting that tar- national development goals for the environment in the geting interventions to these few crops offers promise for 21st century. offsetting the rapid increase in pesticide use and conse- Some of the data compiled to better understand quent harm. the links between the economy and the environment Responsibility: Development Research Group, Infrastruc- appear regularly in the World Bank's annual World Devel- ture and Environment-Susmita Dasgupta (sdasgupta opment Indicators. In particular, "genuine savings," an @worldbank.org). With Craig Meisner; Nlandu Mamingi, indicator of sustainability, has been estimated for some University of West Indies; Guilherme Soria Bastos Filho; 120 countries and published in the World Development and SCIENCE, Escola Nacional de Ciencias Estatisti- Indicators and the Bank's Little Green Data Book. The cas, Brazil. project has also produced working papers on the Completion date: June 2002. revised SEEA for the London Group on Environmen- tal Accounting. Report Responsibility: Development Data Group-M. Saeed Dasgupta, Susmita, Nlandu Mamingi, and Craig Meisner. Ordoubadi (mordoubadi@worldbank.org); and Environ- Forthcoming. "Pesticide Use in Brazil in the Era of ment Department, Office of the Director-Kirk Agroindustrialization and Globalization." Environment and Hamilton. DevelopmentEconomics (special issue on agroindustrialization and Completion date: December 2002. globalization). 66 Environmeni Biodiversity, Habitat, Conservation, and Valuing ject is developing a land use model that provides the eco- Natural Resources in the Amazon Estuary nomic framework for evaluating the ecological effects of land conversion. This leads to the estimation of three The Amazon estuary is relatively pristine. But ranching models: the demand for fish, the demand for cattle prod- activities are encroaching on its wetland habitats, putting ucts, and the effect of land on ranch profits. at risk fisheries with important commercial and subsis- Data are being collected for the ecological assess- tence value. This project investigates whether unre- ment, biodiversity analysis, and evaluation of the estu- stricted expansion of cattle ranching is a good land use ary's main economic activities. These include data on for the Amazon estuary and will attempt to establish structural components (especially vegetation types and whether joint management of cattle ranching and fish- biodiversity) of estuarine wetland habitats, the spatial dis- eries would optimize economic returns and conservation tribution of estuarine habitats, major impacts (especially benefits for the region. 'I'he project's principal objectives deforestation) on the habitats, estuarine fisheries and are to: inland fisheries (migratory species from the Amazon * Identify, analyze, and evaluate major wetland estuary), estuarine livestock, estuarine timber and non- resource uses in the Amazon estuary. timber resources, and market prices for marketed species. * Call attention to the need to incorporate the over- The land use analysis is expected to provide useful whelming influence of urban-driven exploitation of wet- insights into optimal patterns of land use in this and land resources into conservation policy for the Amazon similar regions. Estuarine biodiversity and habitat data- wetlands. bases will be developed and made available on diskette * Spatially define wetland habitats and patterns of and through Web sites to local research and government resource use in the Amazon estuary in order to develop institutions in Brazil. In addition, the work will support a realistic geographic basis for conservation planning. ongoing efforts by the World Bank and the Global Envi- * Analyze the estuary's ecological and economic rela- ronment Facility to promote stakeholder dialogue and the tionship to inland wetlands, using cattle ranching and dissemination of information within the estuary and migratory catfish as examples. across the Amazon Basin. * Produce a report that can be used by policymakers Responsibility: World Bank Institute, Environment and Nat- to launch sound conservation measures in the Amazon ural Resources Division-Gayatri Acharya (gacharya estuary. @worldbank.org). With Nigel Smith, University of The project's analytical framework integrates an eco- Florida; Michael Goulding, Amazon Conservation Team; logical assessment and biodiversity analysis of the estu- Ronaldo Barthem, Museo Goeldi; and Kenneth ary's principal habitats with an economic valuation of the McConnell, University of Maryland. most important wetland resources exploited. The pro- Completion date: June 2003. Environmeni 67 The Impact of Regulatory Risk on the Cost identify variables that drive the perception of gover- of Private Debt for Infrastructure Projects nance risk in a given country. in Emerging Markets o It complemented the analysis based on cross- sectional data by constructing a database of time-series During the 1990s capital markets became the main yield data on infrastructure bonds traded in the sec- source of funding for infrastructure projects, once financed ondary market and examining factors that affect bond risk overwhelmingly by governments. In developing and in actual market conditions. transition economies infrastructure bonds were the most o It investigated the effect of industry structure common method of financingoil, gas, transport, electricity, and regulation on infrastructure bonds in the power and telecommunications projects. Since investments in sector. infrastructure are particularly susceptible to the risks of o It compiled and analyzed the rating history of an government interference, countries raising private finance entire universe of infrastructure bonds, then used the for infrastructure projects need to ensure that their polit- history of upgrade and downgrade events to investigate ical and regulatory frameworks allay investors' concerns the effects of governance over the lifetime of the bonds. about the risk of default. o Finally, the study analyzed the information about Through analyses of rich cross-sectional and time- key governance risks to get at the implications for series data on the fixed rate infrastructure bonds issued policy. and traded during the 1990s, this research examined the The relationship between regulatory policy and the effect of political and regulatory frameworks on investors' cost of private debt for infrastructure in developing perceptions of risk and thus on the cost of debt. The countries has received little attention, yet has impor- research considered such aspects of polity and governance tant implications for the World Bank's policy advice on as the tradition of law and order, the presence of cor- building and strengthening regulatory regimes to promote ruption, the quality of the bureaucracy, the strength of private sector development. This research contributed the contract enforcement system, the extent of regula- to that advice as well as to the Bank's methodology tory discretion and the political constraints on this, and for pricing risk insurance to facilitate private project the likelihood of expropriation and other political risks. finance. The research also produced a database of It also identified the political and regulatory risks that con- regulatory risk indicators in industrial and developing cern investors most. The research compared infrastruc- countries, which will be made available to researchers and ture bond markets of developed and emerging economies practitioners. to see how the factors driving infrastructure finance in Responsibility: Europe and Central Asia Region, Energy the two country groups differ and to draw policy con- Sector Unit-Laszlo Lovei (llovei@worldbank.org). With clusions from these differences. Nina B. Bubnova and John M. Quigley, University of The study included several methodological innovations: California at Berkeley; and Ilya Lipkovich, Virginia o It enhanced the explanatory capacity of credit risk Polytechnic Institute. models by combining firm-specific and macroeconomic (ompletion dote: October 2000. factors and adding the full spectrum of risks associated with the quality of governance and regulation in the Report country and sector issuing a given bond. Bubnova, Nina. 2000. Governance Impact on Private Investment: o It applied hierarchical cluster analysis to examine an Evidencefrom the InternationalPatterns of infrastructure BondRisk array of indicators of regulatory and political risk and to Pricing. World Bank Technical Paper 488. Washington, D.C. 68 Yardstick Competition across Ports: for the water sector in Africa and the electricity sector in An Illustrated Guide for Regulators Latin America. To motivate similar research in Africa and Latin America, the project involved reviewers from these This research showed how to use relatively standard regions and offered seminars to develop the interest of methodologies in measuring the efficiency gains from local researchers. reforms in the organization of port infrastructure-and Seminars and presentations have included a regula- then how to use these measures to promote competition tory accounting workshop at the Mexican port authority between ports and introduce regulation driven by built- for participants involved in tariff revision (June 2000); a in incentives. Measuring efficiency in the context of lecture on efficiency measures in regulated industries at incentive-based regimes and yardstick competition is a regulatory economics course for utility regulators from among the key policy issues for regulated infrastructure seven Latin American countries (January 2001); a three- activities, and the study showed that despite the well- day course on competition in regulated industries at the known difficulties of getting good data on the infra- Chinese Academy of Social Sciences in Beijing for reg- structure sector, it is possible to collect enough ulators and academics from all sectors (February 2001); comparable data on production to establish cost-cutting a workshop at a regulation and intermodal integration incentives and to promote ex post competition in the course in Barcelona, Spain, for port regulators from Spain, sector. Portugal, and eight Latin American countries (May 2001); Based on performance and productivity data assem- and a course on measuring efficiency for African infra- bled in collaboration with the Mexican port regulator, the structure regulators in Dakar, Senegal (June 2001). A study estimated a stochastic production frontier. It then presentation is also planned for a conference of Latin applied this frontier in an analysis of the effects of Mex- American regulators and academics in Santiago, Chile, ico's 1993 port reform (luring 1996-99, the first effi- in January 2002. ciency analysis of port restructuring in .a deyeloping Responsibility: World Bank Institute, Governance, Regula- country. The results showed that Mexico has achieved tion, and Finance Division-Antonio Estache (aestache annual efficiency gains of 6-S8 percent in the use of port @worldbank.org). With Antonio Alvarez, University of infrastructure since assigning the management of the Oviedo, Spain; Tim Coelli, University of New England, infrastructure to independent, decentralized operators. Armidale, Australia; Marianela Gonzalez and Lourdes The size of the gains is roughly the magnitude by which Trujillo, Universidad de Las Palmas de Gran Canaria, the price cap should be cut in the next tariff revision, in Spain; Eugene Kouassi, University of Cocody, C6te order to penalize operators performing below average and d'Ivoire; Sergio Perelman, University of Liege, Belgium; reward those performing above average. The relative Martin Sgut; G. de la Merced; S. Lamarroy; and J. Pas- performance ratings over time are revealing: they iden- trana. The government of Mexico contributed staff time tify consistent sets of leaders and laggards, including for the research. some that would not have been identified with the par- Complelion date: November 2000. tial productivity indicators commonly used in the sector. Through continuing collaboration with the Mexican Reporls port authority, work is being done to update the results Coelli, Tim, Antonio Estache, Sergio Perelman, and Lourdes Tru- with final data for 2000. In addition, Mexico is using jillo. Forthcoming. A Primeron Efficiency Measurementfor Utili- the results in its tariff revision. ties and Transport Regulators. World Bank Institute Development The project also established initial contacts with coun- Studies Series. Washington, D.C.: World Bank. tries where the project team hopes to continue working Estache, Antonio, Marianela Gonzalez, and Lourdes Trujillo. 2001. with local researchers (Argentina, Brazil, Chile, China, "Technical Efficiency Gains from Porc Reform: The Potential Colombia, Cote d'lvoire, Peru, and Senegal). Several of for Yardstick Competition in Mexico." Policy Research Work- these countries are considering a similar analysis of their ing Paper 2637. World Bank, World Bank Institute, Gover- port sectors. In addition, studies are under preparation nance, Regulation, and Finance Division, Washington, D.C. Infrastructure and Urban Development 69 Community Comanagement of Urban sible lessons and reduced the universe of cases to which Environmental Quality: Water, Sanitation, they might be extrapolated. Moreover, failure of gov- and Water Pollution Control ernment agencies to participate prevented exploration of methodologies to support negotiation of burden shar- The main goal of this research was to answer this ques- ing and resource allocation by service providers (gov- tion: What is the best decisionmaking procedure for ernment and private) and consumers (the community). coordinating the actions and resources of the municipal The project applied the planning methodology in government, the community, and the private sector in three low-income communities in Soyapango (a suburb designing water and sanitation, solid waste management, of San Salvador): Las Margaritas (with a population of and water pollution control systems that are sustainable, 275), Reparto San Luis (235), and Nueva Esperanza equitable, effective, and efficient? (385). Fifty-six alternative service systems were deter- From this main objective came three specific objec- mined to be feasible for Las Margaritas and Reparto San tives. The first was to develop a planning methodology Luis, and 16 for Nueva Esperanza. Through intensive consisting of a set of decisionmaking procedures and a consultation in each community, in group forums and indi- set of guidelines for selecting the procedure most appro- vidual discussions, hierarchies of objectives for the sys- priate for a given situation. The decisionmaking proce- tems were developed, focusing on costs, benefits, and dures needed to be designed to determine the best opportunities. In each community a workshop was then systems for serving a given urban community, the conducted at which members of the neighborhood board resources required for those systems, and the contribu- would design several alternative plans, evaluate them, and tions to be provided by the municipality, the community, select the preferred one. The work to design alternative and the private sector. The second objective.,was to plans was based on a multiobjective mixed-integer opti- apply the procedures in an urban environment in Latin mization model in which the main decision variables America. And the third was to evaluate, the procedures were the number of households to be served by a given on the basis of this application. system. The project developed a multiobjective participatory Using operational versions of this model that combined planning methodology for helping service users and a weighting method and compromise programming, the providers collaborate in determining the best technology project generated 6-10 solutions for consideration by mix and the best way to share the burdens of finance, the board. After some were eliminated as of no interest labor, and materials. The planning methodology focuses and the remaining ones were ranked, a variant of the on the household as a fundamental unit of decision- weighted Tchebycheff method was used to generate making and resource allocation; links three levels three or four more alternatives. Board members then of analysis-the household, the community, and the chose the three most attractive plans for a detailed eval- municipality-in an operational procedure for allocating uation based in part on the hierarchies of objectives. resources; recognizes the link between environmental Although no group completed the full assessment dur- management and economic development; and,offers an ing the workshop, the Las Margaritas and San Luis operational framework for linking sustainability, effi- groups made sufficient progress to select their preferred ciency, effectiveness, and equity. plan. The initial intention was to apply the methodology in The application of the methodology led to several three varied case studies. But as a result of insurmount- lessons: able difficulties, case studies at two of the three proposed * Plan implementation. Omission of a project compo- sites failed to get off the ground, while those conducted nent to implement the plan chosen was a sticking point at the third site, in Soyapango, El Salvador, failed to that hampered the investigation. Given the time and elicit the collaboration of government service providers, energy that communities are expected to give to stud- despite earlier commitments to participate. The similarity ies such as this one and the problems that can occur ofthecasesinSoyapangodiminishedtherichnessofpos- when implementation is not part of the research, it 70 Infrastructure and Urban Development appears that studies should guarantee funds for imple- development of the urban water sector or how institu- mentation. But if a community or agency knows that tional and political factors affect options for reform. money will be available for whatever technologies it Using a case study approach, this research attempted selects, it will have no incentive to consider anything but to fill that gap by systematically analyzing the experience the most convenient choices or to consider cost-hardly with reform and private participation in the provision of a realistic approach. water in six cities in Africa and Latin America: Buenos * Beneficiary initiative. Lack of interest by some par- Aires, Argentina; Santiago, Chile; Abidjan, C8te d'lvoire; ties during the case studies required the team to "mar- Conakry, Guinea; Mexico City, Mexico; and Lima, Peru. ket" the project, underscoring the importance of ensuring This sample represents the main contracting options that the expected beneficiaries are the driving force and used in water supply, provides significant variation in the that they have something at stake. institutional setting, and allows comparison with coun- * Strengthened ties. Participation in the project strength- tries that have not used private participation. ened the ties between community leaders and the peo- The study analyzed the effects of political and con- ple they represent and improved their ability to discuss tracting institutions on the design and performance of con- and tolerate different viewpoints. tracts for reform of urban water supply, comparing * Community realism. 'IThe communities came to under- different forms of private participation and reforms under stand that real constraints-legal, technical, financial- public ownership. It examined the design and perfor- were at play, precluding some solutions and demanding mance of reform (including the contractor selection tradeoffs among objectives. For example, despite frequent process, contract design, and regulatory framework), assertions that they would not accept latrines,.San Luis looking at how the different types of reform addressed and Las Margaritas ultimately selected systems utilizing (or failed to address) possible problems of incentives, latrines because of the cost of alternatives. information asymmetries, and credibility. Finally, the Don Bosco University, a private Catholic institution research assessed the results of the reform, looking at dif- in the heart of Soyapango, collaborated in the research ferent indicators of performance and measuring the wel- and served as principal liaison with the communities. fare effects of reform compared with a counterfactual Responsibility: Latin America and the Caribbean Region, (using the methodology developed by Ahmed Galal and Environmentally and Socially Sustainable Development others, We/fareConsequencesofSellingPublic Enterprises:An Sector Unit-Richard M. Huber. With Mark Ridgley EmpiricalAnalysis, New York: Oxford University Press, and Aimee del Aguila, Terra Group; Maria Cristina de 1994). Barahona, Francisco Rivas, Nelly Castillo, and Saul Ben- The research produced several important findings: itez, Don Bosco University, El Salvador; and Arnold vd * In institutionally weak environments it is difficult Klundert, Maria Muller, Jaap Rijnsburger, and Lex Heme- to provide institutionalized protection against expropri- laar, WASTE, the Netherlands. The Netherlands Part- ation of quasi rents and regulatory capture. nership Trust Fund contributed funding for the research. * Constraints such as the political importance of water, Completion date: December 2000. the cost and scarcity of water resources, and the extent of unmet demand profoundly affect the design and dif- Institutions, Politics, and Contracts: Private Sector ficulty of reform. Participation in Urban Water Supply * The political motivation for, feasibility of, and com- mitment to reform affect the choice of design for reform Since the late 1980s many developing countries have and its success. reformed urban water utilities to improve performance, * Even in difficult institutional environments and involving the private sector through such approaches as with serious failings in design or implementation, private leases, concessions, and service contracts. Despite the participation produced gains over reasonable counter- wide variation in approaches, no rigorous analysis had crit- factuals. Most notably, despite the very difficult institu- ically examined how the design of reform affects the tional environment in Conakry, consumers, the Infrastructure and Urban Development 71 government, and the foreign firm that entered the sec- ing Paper 2311. World Bank, Development Research Group, tor all benefited from privatization. Washington, D.C. * The large health and environmental externalities Alcazar, Lorena, Colin Xu, and Ana Maria Zuluaga. 2000. "Reform- associated with water provision make it fundamentally ing Urban Water Supply: The Case of Lima, Peru." World different from other infrastructure sectors. Bank, Development Research Group, Washington, D.C. Draft. * In the African case studies one of the largest prob- Clarke, George R. G., Claude Menard, and Ana Maria Zuluaga. lems was nonpayment by the government. Although the 2000. "The Welfare Effects of Private Sector Participation in private operator in Conakry could cut off nonpaying Guinea's Urban Water Supply." Policy Research Working Paper agencies, it chose not to, fearing retaliation. It is impor- 2361. World Bank, Development Research Group, Washington, tant that governments recognize the cost of nonpay- D.C. ment, particularly to the poor, who cannot afford to Haggarty, Luke, Penelope Brook Cowen, and Ana Maria Zuluaga. subsidize government consumption. 1999. "Institutions, Politics, and Contracts: Private Sector Par- * The gains from having a politically independent reg- ticipation in Urban Water Supply Systems-The Case of Mex- ulator are significant. Performance improved in Santiago ico City Water Sector Service Contracts." World Bank, without privatization, apparently as a result of improved Development Research Group, Washington, D.C. regulation. Santiago's regulator was powerful, indepen- Menard, Claude, and George R. G. Clarke. 1999. "Reforming dent, politically insulated, and guided by detailed laws Urban WaterSupply: The Case of Abidjan, C6te d'lvoire." Pol- that left little room for discretion. Moreover, it paid icy Research Working Paper 2377. World Bank, Development salaries above the civil service norm, and its staff was Research Group, Washington, D.C. regarded as honest, professional, and competent. . 2000. "A Transitory Regime: Water Supply in Conakry, The case studies will be published in a forthcoming Guinea." Policy Research Working Paper 2362. World Bank, book and will also be used by the World Bank Institute Development Research Group, Washington, D.C. in training programs on the water sector and on regula- Noll, Roger, Mary Shirley, and Simon Cowan. 1999. "Reforming tion. Results have been presented at a World Bank Insti- Urban Water Systems: Theory and Evidence from Developing tute clinic in India and at the Regional Conference on Countries." World Bank, Development Research Group, Wash- Reform of the Water Supply and Sanitation Sector in ington, D.C. Draft, Africa, in Kampala, Uganda, on February 26-28, 2001. Shirley, Mary M., ed. Forthcoming. ThirstingforEfficiency:Experi- Responsibi.y Development Research Group, Regulation and ences in Reforming Urban Water Systems. Oxford: Elsevier. Competition Policy-Mary Shirley (mshirley@worldbank. Shirley, Mary, and Claude Menard. 1999. "Cities Awash: Reform- org), Luke Haggarty, Colin Xu, Ana Maria Zuluaga, and ing Urban Water Systems in Developing Countries." World George R. G. Clarke; and Private Sector Development Bank, Development Research Group, Washington, D.C. Draft. Department, Private Participation in Infrastructure-Pene- Shirley, Mary, L. Colin Xu, and Ana Maria Zuluaga. 2000. "Reform- lope Brook. With Claude Menard, Sorbonne University; ing Urban Water Supply: The Case of Santiago, Chile." Policy Simon Cowan, Worcester College, Oxford; Manuel Abdala, Research Working Paper 2294. World Bank, Development Navigant Consulting and Universiry of San Andr6s, Buenos Research Group, Washington, D.C. Aires; Lorena Alcazar, University of the Pacific; Douglass North, Washington University at St. Louis; Scott Masten and Competition and Privatization Keith Crocker, University of Michigan; Roger Noll, Stanford in Urban Water Supply University; Dale Whittington, Northeastern University; and Matthew McCubbins, University of Califomia at San Diego. Compared with other infrastructure sectors, urban water Completion date: December 2000. supply has been privatized more slowly while involving the private sector less. One plausible reason for this is that Reports introducing product market competition is much harder Alcazar, Lorena, Manuel A. Abdala, and Mary M. Shirley. 2000. in the water sector. Consequently, competition must be "The Buenos Aires Water Concession." Policy Research Work- introduced indirectly-in the form of competition for the 72 Infrastructure and Urban Development market (competitive bidding for a lease or concession), yard- tizing the water authorities was originally proposed in stick competition (benchmarking), and competition in the 1985, the regulator was not fully operational until after capital market. Rather than relying on competition for the the prices were set for the first five years following pri- market, the government of the United Kingdom decided vatization. Failure to have an independent regulator in to rely on yardstick and capital market competition, com- place not only increases the probability of errors, but also bined with a new regulatory regime to protect consumers. opens the government up to charges of politicizing the This project evaluated the performance of this approach privatization process. in the United Kingdom. Using data from company balance Preliminary results were presented at a World Bank sheets and income statements, the research used a partial seminar in April 1998, and more recent results at the equilibrium cost-benefit analysis (developed by Ahmed Regional Conference on Reform of the Water Supply and Galal and others, We/fare Consequences ofSelling Public Enter- Sanitation Sector in Africa, in Kampala, Uganda, on Feb- prises:AnEmpirica/Analysis, New York: Oxford University ruary 26-28, 2001. Press, 1994) to assess the effect of privatization on welfare Responsibility: Development Research Group, Competition and identify who gained and who lost. and Regulation Policy-George R. G. Clarke (gclarke The research found that privatization appears to have @worldbank.org) and Ana Maria Zuluaga. With Simon resulted in net societal gains while failing to benefit Cowan, Oxford University. either consumers or the government-but allowing big Completion dote: June 2001. gains to stockholders. The main reason is that the gov- ernment allowed the water companies to increase prices Reports significantly. Thus shareholders earned large windfall Clarke, George R. C., Simon Cowan, and Ana Maria Zuluaga. gains-despite the establishment of a well-funded, inde- 2000. "Competition and Regulation in Urban Water Supply: The pendent, and highly competent regulator and the Privatization of Thames Water." World Bank, Development attempts to introduce cornpetition. Research Group, Washington, D.C. Several broad lessons can be drawn from this experience: . 2000. "A Cost-Benefit Analysis of the Privatization of * When the cost of meeting future obligations is Thames Water." World Bank, Development Research Group, highly uncertain, as it was at the time of privatization in Washington, D.C. the United Kingdom as a result of a tightening of envi- ronmental laws, setting prices for long periods in advance The Effects of Telecommunications Infrastructure can be costly. The large price increases proposed to cover on Investment: An Empirical Analysis the cost of the new requirements resulted in large losses to consumers and large gains to shareholders when it This research studied the effcct of telceommunications turned out that the water authorities could meet the infrastructure on the allocation of foreign direct invest- requirements at far lower cost than originally thought. ment and on the outcome of World Bank projects through * This miscalculation occurred despite a relatively apo- cross-country econometric analysis. The research involved litical, highly competent, and noncorrupt regulator and creating and analyzing a data set and carrying out sup- civil service. In many developing countries setting appro- porting case studies. priate price caps for long periods under great uncer- The research found a positive relationship between tainty is likely to be even more difficult. teledensity (telephone lines per capita) and inflows of for- * Price caps that are generous to shareholders (and less eign direct investment. It also found that privatization so to consurners) are more likely when soft price caps of telecommunications was significantly and positively serve the government's political purposes, as they did in related to growth in foreign direct investment inflows. the United Kingdom. However, only by using some methodologically irregu- * It is important to have a strong, independent regu- lar techniques could the research find a relationship lator in place well before the main parameters that affect between teledensity and World Bank project perfor- privatization outcomes are established. Although priva- mance that was supported by the data. Infrastructure and Urban Development 73 Material from the econometric research was com- been absent from the reform process in Latin America, bined with case study material on the use of information they have often been secondary to fiscal considerations. and communications technologies in developing coun- As a result, it is difficult to generalize about the overall tries as they might relate to foreign direct investment and impact of reform on poor households. World Bank project performance. The resulting working This study examined the impact of infrastructure pri- paper has been used in several courses on information vatization on the poor in Latin America. It reviewed the and communications technologies and development. transmission mechanisms through which infrastructure The research results, data set, and case studies have also reform may affect the poor, focusing on microeconomic provided input to many Bank reports and papers. The issues related to the poor's access to infrastructure ser- data set is being updated and will be used in future vices and the affordability of those services. Using house- studies, and the case studies are being incorporated into hold survey data from 12 countries, the study reviewed articles. the trends in access to infrastructure services, examin- Responsibility: World Bank and International Finance Cor- ing whether the poor are benefiting from increases in con- poration, Global Information and Communications Tech- nections. It also looked at the policy options for improving nology Department, Policy Division-Charles Kenny access and ensuring affordability of service and assessed (ckenny@worldbank.org) and Christine Zhen-wei Qiang. how priorities should be set. With Taylor Reynolds and Jia Liu, American University; The study found that in many cases the poor have ben- and Jeremy Grace, State University of New,York at efited from utility reform, but there is little doubt that Geneseo. more could have been achieved if reform had focused Completion dote: June 2001. more explicitly on poverty from the outset. In two areas in particular reforming countries consistently failed to take Reports the interests of the poor into account: the affordability Grace, Jeremy, Charles Kenny, and Christine Zhen-wei Qiang. of service and that of connections. 2001. "ICTs and Broad-Based Development." World Bank Throughout Latin America cross-subsidies have been and International Finance Corporation, Global Information the preferred method for safeguarding the interests of and Communications Technology Department, Washington, poor households. But there is evidence that many cross- D.C. subsidy mechanisms are poorly designed, directing as Liu, Jia, Charles Kenny, and Christine Zhen-wei Qiang. 2001. much as 60-80 percent of cross-subsidies to households "T'elecommunications Infrastructure and Success of World well above the poverty threshold and failing to benefit Bank Projects: An Empirical Analysis." World Bank and as many as 80 percent of poor households. The alterna- International Finance Corporation, Global Information and tives to cross-subsidies have often been overlooked by Communications Technology Department, Washington, policymakers. One alternative is to provide poor house- D.C. holds with the option of receiving a lower quality of ser- Reynolds, Taylor, Charles Kenny, and Christine Zhen-wei Qiang. vice in return for more affordable tariffs. Another is to 2001. "Telecommunications Infrastructure and Foreign Direct provide customers with better ways of budgeting for Investment: An Empirical Analysis." World Bank and Inter- utility costs. national Finance Corporation, Global Information and Com- Another common failing in the reform has been the munications Technology Department, Washington, D.C. introduction of prohibitively high connection charges that fail to take into account the difficulty poor house- Privatization and Basic Infrastructure holds have in making one-time capital payments. For poor Services for the Urban Poor households affordability can be much more of a barrier to access than to service use. There are several strategies Latin America has been at the forefront of the worldwide for circumventing this problem. The utility could pro- movement toward reform of utility services (water, energy, vide customers with a credit line, allowing them to pay telecommunications). While social concerns have not the connection charges in several installments. The cost 74 Infrostructure and Urban Development of connection could be reduced by using innovative low- sented at the conference Infrastructure for Development: Pri- cost technologies or allowing poor households to con- vate Solutions and the Poor, Public-Private Infrastructure Advi- tribute labor. And the cross-subsidy approach could be sory Facility, U.K. Department for International Development, applied to financing connections rather than reducing and World Bank, London, May 31-June 2, 2000.) service tariffs. Foster, Vivien, and Jean-Philippe Tre. 2000. "Measuring the The findings of the study have been widely dissem- Impact of Energy Interventions on the Poor: An Illustration from inated. They are being used in two World Bank Institute Guatemala." Paper presented at the conference Infrastructure courses on private participation in infrastructure and its for Development: Private Solutions and the Poor, Public- implications for the poor and have served as input to a Private Infrastructure Advisory Facility, U.K. Department for World Bank energy strategy and to poverty diagnostic International Development, and World Bank, London, May reports for several Latin Arnerican countries. In addition, 31-June 2. findings have been presented at seminars and confer- Foster, Vivien, Jean-Philippe Tre, and Quentin Wodon. 2001. ences, including at the Caniadian Economic Association, "Energy Consumption and Income: An Inverted-U at the the Canadian Public Economics Study Group, Household Level?" GREQAM (Nlarseille, France), the Planning Ministry of . 2001. "Energy Prices, Energy Efficiency, and Fuel Poverty." Uruguay, the University of York (England), Soci6te Cana- Gomez-Lobo, Andres, and D. Contreras. 2001. "Subsidy Policies dienne de Science Economique, and the World Bank. for the Utility Industries: A Comparison of the Chilean and Responsibility: World Bank Institute, Governance, Regula- Colombian Water Subsidy Schemes." Draft. tion, and Finance Division-Antonio Estache (aestache Makdissi, Paul, and Quentin Wodon. 2001. "Consumption Dom- @worldbank.org); and Latin America and the Caribbean inance Curves: Testing for the Impact of Indirect Tax Reforms Region, Poverty Sector Unit-Quentin Wodon and Vivien on Poverty." Paper presented at the workshop Progress in the Foster. With Mohamed Ihsan Ajwad; Corinne Siaens; and Analysis of Poverty and Income Inequality, Tinbergen Institute, Jean-Philippe Tre. Amsterdam, March. Completion date: June 2001. Siaens, Corinne, and Quentin Wodon. 2001. "Impact of Access to Basic Infrastructure Services on Poverty." Draft. Reports Wodon, Quentin. 2000. "Low-Income Energy Assistance and Dis- Ajwad, Mohamed Ihsan, and Quentin Wodon. 2001. "Do Local connection in France." Applied Economics Letters 7(12): 775-79. Governments Maximize Access Rates to Public Services across . 2000. "Public Utilities and Low-Income Customers: A Areas? A Test Based on Marginal Benefit Incidence Analy- Marketing Approach." International Journal of Public Sector sis." Paper presented at the World Bank Economists' Forum, Management. Washington, D.C., May. Wodon, Quentin, and Mohamed Ihsan Ajwad. 2001. "Marginal Ben- .2001. "Estimating the Welfare Impact of Privatization: Elec- efit Incidence Analysis Using a Single Cross-Section of Data: tricity in Bolivia." Draft. An Alternative Approach." . 2001. "Infrastructure Services and the Poor: Providing Wodon, Quentin, Mohamed Ihsan Ajwad, and Corinne Siaens. Connection or Consumption Subsidies?" Draft. 2001. "Targeting Electricity Subsidies: Lifeline or Means- Duclos, J.-Y., Paul Makdissi, and Quentin Wodon. 2001. "Socially Testing?" Draft. Efficient Marginal Tax Reforms." Estache, Antonio, Vivien Foster, and Quentin Wodon, with con- The Sustainable Financing of Investments tributions from Mohamed Ihsan Ajwad, Corinne Siaens, Jean- in Municipal Infrastructure: Cost Recovery PhilippeTre,and A.Wellenstein.2001.InfrastructReformand for Solid Waste Management in the Philippines the Poor: Learningfrom Latin America's Experience. Washington, D.C.: World Bank Institute. This study uses contingent valuation surveys to measure Estache, Antonio, Andres Gomez-Lobo, and Danny Leipziger. the willingness to pay for garbage collection and dis- Forthcoming. "Utilities Privatization and the Poor: Lessons and posal. Results of a 1998 survey of 500 representative Evidence from Latin America." World Development. (Also pre- households, 300 representative commercial establish- Infrastructure and Urban Development 75 ments, and 150 itinerant vendors in Iloilo, a medium-size Study Tour on Waste Management in Scandinavia, on city in the Philippines, reveal that on average households June 7-11, 1999. are willing to pay 24.45 pesos and commercial enter- Responsibility: Latin America and the Caribbean Region, prises 89.15 pesos a month for solid waste collection. (The Finance, Private Sector, and Infrastructure Sector Unit- number of usable responses from the itinerant vendor Sheoli Pargal (spargal@worldbank.org); and Develop- sample was too low to obtain robust results.) In contrast, ment Research Group, Infrastructure and Environment willingness to pay for solid waste disposal was close to -Maureen Cropper. With Nathalie Simon, U.S. Envi- zero for both households and commercial entities. Sim- ronmental Protection Agency; and the University of the ilar results were obtained in a second medium-size city, Philippines. The Japanese government contributed fund- Naga. ing for the research. Household income, the respondent's level of school- (ompletion date: July 2001. ing, male gender, hiring help to dispose of garbage, and recycling waste in the household were all positively (omparative Study of Water Institutions related to the willingness to pay for garbage collection. and Their Impact on Water Sector Performance Age and burning garbage were negatively related to will- in Selected Countries ingness to pay. These results clearly underline the impor- tance of education and information campaigns in In response to increasing water scarcity, water institutions increasing awareness of the importance of solid waste -formal and informal policy, legal, and administrative management-and thus the willingness to pay for the entities-are undergoing far-reaching changes world- service. wide. What has been the cross-country experience with The data are also being used to study the feasibility the evolution of water institutions? Do the changes and applicability of the benefit transfer methodology, in advance or impede the water sector's financial viability which the benefits associated with an improvement pro- and the sustainability of water resources? gram are assessed by using the results from.one city to This project addressed these questions through a predict the response in another. The household survey comparative study of water institutions and their impact data from Naga and Iloilo on willingness to pay for a on the sector's performance in selected industrial and garbage collection program were examined using survival developing countries. The interdisciplinary study com- time models and a Weibull distribution under the pared the water institutions of the 35 sample countries same model specification for each city. Respondent within a common framework defined by a set of insti- characteristics-age, gender, income, home ownership, tutional (legal, policy, and administrative) features, focus- household size, level of schooling, and level of garbage ing on economic and operational performance. It then collection service-were used as explanatory variables. compared water sector performance, based on qualitative The results were mixed. Although overall mean and and quantitative variables. By linking the comparative median willingness-to-pay values for the service in the analysis of institutions with the analysis of sector two cities were similar, explanatory variables entered performance, the research identified both common the equation with differing levels of statistical significance. and unique features in best-practice and worst-case Additional work in this area is looking at the effects of institutions. the choice of distribution on model fit and the resulting The analysis drew primarily on data from personal implications for benefit transfer. interviews on key water sector challenges and recent Preliminary analytical results were presented to may- institutional responses and from a detailed question- ors and city officials in Iloilo and Naga, the Philippine naire administered to water sector experts to obtain fac- Department of Environment and Natural Resources, tual and subjective information on water institutions local nongovernmental organizations, and academics in and sector performance. Manila in February and March 1999. Results were also Despite variations in resources and political econ- presented to World Bank operational staff during the omy situations, common trends and patterns emerged. 7 6 Infrastructure and Urban Development The key issues are no longer resource development and University of Illinois at Urbana, and at a conference of water quantity but resource allocation and water quality. the European Association of Environmental and Resource The notion of water provision as a public good and a wel- Economists. A final report on the results is being fare activity is being replaced by the concept of water as prepared. an economic good and an input in economic activity. Responsibility: Rural Development Department-Ariel Dinar And the old development paradigm centered on cen- (adinar@worldbank.org). With R. Maria Saleth, Institute tralized decisionmaking, administrative regulation, and for Social and Economic Change, Bangalore, India. The bureaucratic allocation is fast giving way to a new para- Netherlands Trust Fund contributed funding for the digm rooted in decentralized allocation, economic instru- research. ments, and stakeholder participation. Completion date: October 2001. For the World Bank and other international develop- ment agencies, the findings suggest a critical need to con- Reports centrate efforts in countries, areas, and subsectors that Saleth, R. Maria, and Ariel Dinar. 1999. Evaluating Water lnstitu- already have a critical mass of institutional capacity, iionsand WaterSectorPerformance. World BankTechnical Paper where lower transaction costs increase the probability of 447. Washington, D.C. success. The institutional changes occurring across coun- - . 1999. "Water Challenge and Institutional Response: A tries suggest that the potential net gains from such Cross-Country Perspective." Policy Research Working Paper changes are beginning to exceed the corresponding trans- 2045. World Bank, Rural Development Department, Wash- action costs in most contexts. ington, D.C. The multidimensional links among key sector prob- . 2000. "Institutional Change in the Global Water Sector: lems suggest that an integrated approach to reform will Trends, Patterns, and Implications." WaterPoliq- 2(3): 175-99. have the greatest impact. At the heart of such an approach . 2000. "Institution-Performance Interaction within the lie institutional changes to strengthen and modernize the Water Sector: Empirical Results with Cross-Country Data." legal, policy, and administrative arrangements governing Paper presented at a conference of the European Association the water sector. of Environmental and Resource Economists, Crete, June The second phase of the study developed an analyt- 30-July 2. ical framework to identify layers of institutional links and links between institutions and performance. It then eval- The Impact of Rural Roads uated these layers of links using an evaluation method- ology employing perception-based cross-country data. Rural roads can boost agricultural growth and invest- These analytical and empirical efforts were used to iden- ment, household food security, and investment in human tify key inputs for a generic strategy of water institution capital. And they can improve access to markets for rural reform. products and reduce transaction costs. But the returns to The results show the relative strength, direction, and rural road investments are not known, in part because of significance of the effects of institutions on performance. methodological problems. Even if the returns are cal- They suggest that the interaction of institutions and culated using the internal rate of return approach, the esti- performance can be determined by the general socio- mates are generally so low that the investments do not economic, political, and resource-related environment in appear viable. Moreover, the effects of rural roads are long which it occurs. And they strongly favor a sequential term and cannot be captured through cross-sectional strategy for institutional reform. survey data. The third phase of the study analyzed the entire data This research project has designed a long-term impact set, which includes 127 observations from 43 countries. study of a World Bank-financed rural roads project in The results of the study have been presented at sem- Bangladesh. It is providing technical help to the inars at the World Bank, the University of Indiana at Bangladesh Institute of Development Studies (BIDS) in Bloomington, and the Water Resources Center of the carrying out baseline and follow-up surveys and support Infrastructure and Urban Development 77 in analyzing the descriptive data. Based on the panel data of jobs and housing on annual household vehicle miles on households and communities, the study is calculat- traveled? How does improved access to public transit ing the economic and social returns of rural roads. affect annual household vehicle miles traveled and the The BIDS conducted the baseline survey in program transport mode used for commuting to work? How does and nonprogram areas in 1997. The follow-up survey, the density of the road network and its shape affect planned for 1999, had to be postponed because of the dev- annual miles traveled? astating flood of 1998, which delayed completion of The empirical strategy involves two main steps. First, some of the project roads. The first follow-up survey was in contrast with previous studies, this project constructs carried out between November-December 2000 and a variety of citywide measures that capture different March-April 2001. The data are now being processed by aspects of the spatial configuration of cities and are exoge- the BIDS; once data processing is completed, analysis will nous to individual households. For example, it measures follow. the share of the population living within a certain num- Responsibility: World Bank Institute, Economic Policy and ber of miles from the central business district and uses the Poverty Reduction Division, and Development Research area under this distribution to measure the centrality of Group, Rural Development-Shahidur R. Khandker the population in a city. It constructs a similar centrality (skhandker@worldbank.org); World Bank Institute, Eco- measure for employment. To measure the compactness of nomic Policy and Poverty Reduction Division-Hus- the city, the study computes the share of the population sain Samad; and South Asia Region, Infrastructure Sector living within a certain percentage of the area around the Unit-Thampil Pankaj. With M. A. Latif, Bangladesh central business district and uses this to construct a spa- Institute of Development Studies. tial Gini coefficient of residential sprawl. It also constructs Completion date: December 2001. a measure of the balance between jobs and housing. Second, the study merges its measures of urban spa- The Impact of Urban Spatial Structure tial structure, computed for the 130 largest metropolitan on Travel Demand areas in the United States in 1990, with the 1990 National Personal Transportation Survey. It uses the resulting This research is quantifying the effects of urban sprawl data set of about 8,000 households to estimate equa- and the availability of public transportation on the travel tions for annual household vehicle miles traveled and demand (annual miles driven and choices of tranport choice of transport mode for commuting. This yields mode for commuting) of U.S. households. Decentral- estimates of the elasticity of travel demand with respect ization of jobs and housing has been presumed to greatly to population and employment sprawl and characteris- increase travel demand, but attempts to measure this tics of the transport system, controlling for such house- effect have been hampered by the use of data on city- hold characteristics as income, education, and household level travel, which is determined simultaneously with size and composition. urban spatial structure. This study overcomes this prob- Preliminary results suggest that urban form has a non- lem by matching city-level measures of sprawl and tran- negligible effect on travel demand. For example, if Wash- sit supply, which are exogenous to individual households, ington, D.C., were to become as sprawling as Atlanta (in with household-level data on annual miles driven and the terms of the population centrality measure), households principal mode of commuting to work. that currently drive would drive 8 percent more on aver- By estimating the effects of the spatial distribution of age. If Washington, D.C., were to become as unbalanced jobs and housing, the characteristics of the road net- as Dallas in the distribution of jobs and housing, house- work, and the characteristics of public transit on the holds would drive about 4 percent more. And if Wash- travel demand of urban households in the United States, ington, D.C., had a rail system similar to that in Dallas, the research can answer a range of questions important households would drive 7 percent more. for transport economists, environmental economists, and Although the analysis uses U.S. data, its methods can urban planners: What is the effect of the decentralization also be applied to developing country cities. The research 78 Infrostruclure and Urban Development was presented at the National Bureau of Economic was studied in three places that share a common cultural Research Summer Institute in Cambridge, Massachusetts, and historical identity but have responded quite in July 2001. differently to change-Leh, Kathmandu Valley, and Responsibility: Development Research Group, Infrastruc- selected Bhutanese towns. ture and Environment-Maureen L. Cropper (mcropper With the broadening of the focus from traditional @worldbank.org). With Antonio Bento, University of architecture to built form (both buildings and urban California at Santa Barbara; and Ahmed Mushfiq Mobarak spaces), the study adapted its methodology to give greater and Katja Vinha, University of Maryland. emphasis to participatory research. It has organized work- Completion dole: December 2001. shops of practitioners and artisans and civic consultations to envision future town development. It has also used par- Linking Culture and Poverty Reduction ticipatory exercises to define an urban vision that would in the Himalayas: A Comparative Study serve as a basis for developing guidelines within which in Bhutan, Nepal, and Ladakh (India) innovation and tradition can interact and redefine the built environment. This research set out to explore the feasibility of using A workshop conducted in Bhutan in June-July 2001, traditional technologies and architecture in modern con- with the Thimphu Municipal Corporation, focused on struction as a means of conserving the cultural identity the question of how to conserve the traditional built of towns and cities. But in the course of field surveys it form under the pressures of globalization. The research became evident that the issue was linked with funda- team worked with local planners and officials to identify mental questions about urban livability and the way and document "innovations" in different types of build- societies define and manage their built environment. ings. And the group discussed public and private concerns The study therefore shifted its focus to examining the about the impact of Western influences on the tradi- process of change in the built environment and its impact tional built form. on that environment, with the aim being to identify the Early findings were incorporated in an urban man- conditions that allow social and economic transformations agement course organized by the World Bank Institute to take place. in Bhutan on June 17-29, 2001, which focused on urban Innovation in the built environment occurs at two livability and the importance of culture. levels: in the management and organization of urban The research supports the Bhutan Urban Develop- life, particularly with regard to the relationship between ment Project. In addition, the research findings are communities and urban space, and in the construction expected to influence the World Bank's emerging agenda process, through its impact on the labor and materials for livable cities. markets. Thus the impact of innovation on the urban poor Responsibility: Transportation and Urban Development is mediated through several processes: the creation of liv- Department, Urban Unit-Evangeline Kim Cuenco able spaces for the urban poor, the promotion and pro- (ecuenco@worldbank.org). With Solomon Benjamin; tection of supporting economic systems for poverty Biresh Shah; Karma Jimba; Dasho Karma Gelay; Luke reduction, and the redefinition of the knowledge Young; Kirk Nyland; Reinhard Goethert, Dennis French- economy to benefit the poor. man, Jan Wampler, and Mike Priore, Massachusetts The research uses the built form as an entry point for Institute of Technology; Daniel Pini, University of Fer- examining a range of issues that affect the way in which rara; Franco Frizon; Andre Alexandre, Tibet Heritage livable cities are created, including urban land, local Trust; Christian Hlade; E. Seckler, Harvard University; governance and planning, local economic development, Carl Puscha, College of Fine Arts, Vienna; and Meghraj and employment. The study aims to define the Adhikari, Dawa Zangmo, Tashi Wangm, and Manjusha parameters for intervention, going beyond traditional Rai, Department of Urban Development and Housing, concepts of historical preservation to broader issues of Bhutan. urban planning and development. The process of change Completion date: December 2001. Infrostructure and Urban Development 79 The Macroeconomics of Infrastructure of new firms, or allowing entry of foreign firms? What kind in Latin America of regulatory framework has yielded the biggest payoff in attracting private initiative? The progress in stabilization and reform that many Latin To address these questions, the research combines American countries have achieved in the past decade has microeconomic and macroeconomic data from Latin been based largely on strong fiscal adjustment efforts. America and elsewhere. It will produce a data set on infra- There is a wide perception that public infrastructure structure stocks and flows for the major economies in investment (and, in many cases, infrastructure mainte- Latin America. Results will be disseminated at a con- nance) has been a casualty of this fiscal retrenchment. Evi- ference in the fall of 2001. dence shows that across Latin America fiscal corrections Responsibility: Latin America and the Caribbean Region, have been associated with a slowdown in the growth of Office of the Chief Economist-Luis Serv6n (Iserven infrastructure stocks, such as road networks and power @worldbank.org), and Finance, Private Sector, and Infra- generation capacity. Deficiencies in infrastructure could structure Unit-Antonio Estache, Sheoli Pargal, and hamper private sector activity and growth in the medium Marianne Fay. With Cesar Calder6n, University of term, eventually constraining the expansion of tax bases Rochester. and making the fiscal effort self-defeating. Complelion dale: December 2001. In part in response to this risk, but also as part of the regionwide shift away from state-led development, most Policy Research Report on Regulation Latin American countries have allowed increased private sector involvement in infrastructure, such as ports, roads, Since the late 1980s many developing and transition railways, power, telecommunications, and water and san- economies have implemented far-reaching structural itation. Indeed, among developing regions, Latin Amer- reform and privatization programs in key infrastructure ica has advanced furthest in privatizing and liberalizing sectors. The policy outcomes clearly reveal the critical basic infrastructure sectors. importance of the organization and architecture of post- This study asks a fundamental question: How has privatization markets and governance structures to the this redefinition of the private and public sector roles continuing success of the reform. Indeed, unless the affected the volume and quality of infrastructure services reform is accompanied by appropriate policies for pro- and the economy's investment and growth prospects, moting competition and regulating monopoly, improve- and with what fiscal costs or benefits? More specifi- ments in performance will be limited and are unlikely cally, it addresses the following policy issues: to be sustainable in the long run. Moreover, the poten- o What are the short- and long-run fiscal costs and ben- tial benefits could be substantially undermined, and efits of the new public-private partnership-and of the perhaps transformed into harm, if the regulatory archi- reduced public sector involvement in infrastructure? tecture and rules that evolved in industrial countries are How should they be reflected in the priority ranking of uncritically applied in developing and transition public expenditures at a time that fiscal restraint is a dom- economies with entirely different economic characteristics inant concern? and institutional endowments. o How have the opening of infrastructure industries This research will look at competition and regula- to the private sector and the new forms of public-private tion in the infrastructure sectors of developing and tran- partnership affected the volume and quality of infra- sition economies, giving equal emphasis to the structure stocks and services? Where have crowding-in institutional issues of regulation and the substantive reg- effects between public and private investment been ulatory policies. This balance is consistent with the reinforced, and where have they been hampered? emerging conventional wisdom that the procedural o What has been more effective in attracting the aspects of regulation are critical in determining how well private sector to "fill the gap" left by public sector the substantive goals of regulatory policy are achieved. retrenchment-privatizing existing firms, allowing entry Specifically, the research will: 80 Infrastructure and Urban Development * Review the economic rationale for regulation (eco- Responsibility: Development Research Group, Regulation and nomic, social, and administrative) in infrastructure and Competition Policy-loannis N. Kessides (ikessides the worldwide trend toward regulatory reform. @worldbank.org), Mary Shirley, George R. G. Clarke, * Assess the role of regulatory policy as an integral part Luke Haggarty, and L. Colin Xu; Latin America and the of national economic policy and its contribution to improv- Caribbean Region, Finance, Private Sector, and Infra- ing resource allocation, productive efficiency, economic structure Sector Unit-Jose L. Guasch; and Infrastruc- growth, and prosperity. ture and Urban Development Department, Transport * Identify the important elements of the framework Division-Kenneth M. Gwilliam. With Paul Joskow, that must be established by government and the com- Massachusetts Institute of Technology; Jean-Jacques mercial culture for competition to work effectively- Laffont, University of Toulouse; Roger Noll, Stanford laws that establish rights to tangible and intellectual University; and David Newbery, Cambridge University. property and create the ability to undertake transactions Completion date: June 2002. at reasonable cost, contract law and contract dispute res- olution that facilitate flexible and reliable commercial Economic and Engineering Evaluation agreements, and central banking services and orderly of Alternative Strategies for Managing structures for capital markets that foster the efficient Sedimentation in Storage Reservoirs provision of currency transactions and credit. * Evaluate the efficacy of the new regulatory princi- The traditionally accepted practice in designing and ples reflecting a preference for competition and market- operating reservoirs is to allow them to fill slowly with like solutions, and assess their applicability to developing sediment, leaving the consequences of the sedimenta- and transition economies. tion and abandonment of reservoirs for future generations. * Analyze the transitional problems of deregulation, That stage has already arrived for many reservoirs, and especially those relating to labor entrenchment, tariff suitable sites for new reservoirs are limited. What is rebalancing, business liquidation, and universal service. needed is sustainable management of reservoirs. But * Identify the substantive regulatory issues likely to not all reservoirs can be managed sustainably, and eco- arise in countries implementing competitive restructur- nomic guidelines are needed to identify those that can ing and privatization programs, and suggest strategies for be. addressing them. This research is developing a mathematical opti- * Review the institutional architecture for enforcement mization model that can be used to determine the eco- of regulation and competition and the procedural safe- nomic feasibility of sustainable management of storage guards needed to ensure the integrity, transparency, and reservoirs, taking into account economic, social, envi- accountability of regulatory and antitrust intervention. ronmental, and engineering factors. The model is being * Identify the fundamental principles that must be designed to provide policymakers with a general frame- publicly articulated by regulatory agencies as the basis work for assessing whether a particular site can support for their policy analysis and decisions. a sustainable reservoir and whether that mode of oper- * Identify the important elements and goals of regula- ation makes economic sense. The model has been val- tory dynamics-how regulatory mandates and rules should idated through five case studies (Baira and Ichari, India; adapt to new problems, changed circumstances, and new Sefid-Rud, Islamic Republic of Iran; Tarbela, Pakistan; information on the workings of the regulated sectors. and Gebidim, Switzerland) as well as in nine additional The research will draw on reports issued by the World reservoirs in Morocco, and sensitivity analysis has been Bank and other multilateral institutions, on databases of carried out. external consulting firms, and on information from the The study was launched with a World Bank workshop sectoral ministries, regulatory agencies, and regulated firms on December 8-9, 1999, attended by people from both of Bank client countries. The results will be published in developing and industrial countries (Austria, Brazil, Italy, a forthcoming World Bank Policy Research Report. Japan, Malaysia, the Philippines, South Africa, Turkey, Infrostrurture and Urban Development 81 the United Kingdom, and the United States) and by in additional capacity? How can telecommunications Bank staff. The workshop was intended to help outline reform be structured so that it gains the support of key the first phase of the research. In addition, the research stakeholders and fits well with the country's institu- has been discussed at several other events: the Interna- tional capacity? tional Water Resources Association Workshop on Dams, The research analyzes these questions though its Development, and the Environment on February 14-16, broad data set and through case studies of countries that 2000, in Sao Paulo, Brazil; the conference on Reservoir have tried a mix of reform approaches, including no Sedimentation onJune 20,2000, in Wallingford, England; reform. (The set of countries includes C6te d'Ivoire, the annual meeting of the U.S. Commission on Large Ghana, Malawi, Senegal, Tanzania, and Uganda.) The Dams on July 10-13, 2000, in Seattle, Washington; and work began in 1999 with a pilot case study (Ghana), fol- the Third Workshop on Development and Management lowed by team meetings to ensure consistency in the con- of Reservoirs in the Rio de la Plata Basin, Posadas, ceptual framework and the quality of the work. Four case Argentina, on March 9-17, 2001. studies have been discussed with project advisers, revised, The results of the research may influence the way pol- and finalized, and are expected to be disseminated icymakers and engineers approach the design of dams. through a regional conference in Africa in late 2001. And the research is expected to initiate a new generation Additional developments will be tracked and analyzed, of operations in the Bank, focusing not only on dam resulting in a set of updated cases and papers for a con- safety but also on reservoir conservation. ference in late 2002. Responsibility: Rural Development Department-Alessandro The rigorous empirical analysis not only will assist Palmieri (apalmieri@worldbank.org) and Ariel Dinar. With donors but also will help the proponents of reform make Farhed Shah, University of Connecticut; and George their case. The project is also strengthening policy and Annandale, Hydrosystems Engineering, Denver, Colorado. regulatory skills in the region by joining African research Completion date: October 2002. institutes in close partnership with senior researchers from academia and the World Bank, all using a single Privatization of Telecommunications methodological and conceptual framework. Over the in Sub-Saharan Africa project's four years the participating African research institutes-the Ivoirian Center for Economic and Social Much of the policy advice on privatization and regula- Research in C6te d'Ivoire and the Economic and Social tion of telecommunications is based on the experience Research Foundation in Tanzania-are taking on more of high- and middle-income countries. Yet a growing and more lead responsibility for the research, with the number of low-income countries in Sub-Saharan Africa goal being to create an institutionalized regional capac- have privatized their telecommunications firms or allowed ity to advise on telecommunications reform and on infra- private cellular operators to enter their domestic markets. structure regulation more broadly. A better understanding of this experience would help Findings have been discussed with World Bank donors develop policy advice tailored to the region's operational teams working on telecommunications reg- institutional environments and market conditions, avoid- ulation in Benin, Ghana, Malawi, Niger, Tanzania, Togo, ing "one size fits all" approaches to reform. and Uganda. The research has also been discussed with This research is analyzing in depth the experiences members of the World Bank and International Finance of Sub-Saharan African countries with telecommunica- Corporation's joint Global Information and Communi- tions reform and tracking changes over the period cations Technology Department, who have applied the 1999-2002. The research explores three key questions results in a number of countries. And the research has con- faced by policymakers in the region: How can greater tributed to policy discussions on regulating entry, com- competition be encouraged by facilitating efficient entry? petition, and interconnection in telecommunications. How can the incumbent telecommunications provider be Responsibility: Development Research Group, Regulation and motivated to use its existing assets better and to invest Competition Policy-Mary M. Shirley (mshirley 82 Infrastrutture and Urban Development @worldbank.org), Luke Haggarty, George R. G. Clarke, growth and employment associated with inadequate pro- and Frew Gebreab. With David Sappington and Mark vision of local services. The research is expected to con- Jamison, University of Florida; Jean-Jacques Laffont, tribute to World Bank operations by sharpening the Jean-Paul Azam, and Mlagueye Dia, University of analytical tools needed to understand the factors ham- Toulouse; Tchetche N'CGuessan and Mathieu Meleu, pering the developmentof small and medium-size enter- Ivoirian Center for Economic and Social Research, C6te prises at the local level and by highlighting the public d'Ivoire; Samuel Wangwe and Haji Semboja, Economic services that most need improvement. and Social Research Foundation, Tanzania; Francis The study is being conducted in partnership with Tusubira, Uganda Communication Commission and four local institutions. The results will be incorporated Makere University, Kampala; Mavis Ampah and Larry in an urban management course in Central America. Attipoe, Spectrum International, Accra, Ghana; and Pau- Responsibility: Latin America and the Caribbean Region, los Nyirenda, Sustainable Development Network Pro- Finance, Urban Cluster-Mila Freire (mfreire gramme, Malawi. The U.S. Agency for International @worldbank.org). With Fundacao Joao Pinheiro, Belo Development is contributing funding for the research. Horizonte, Brazil; Programa de Desarrollo Urbano Completion dale: October 2002. Sostenible (Produs), University of Costa Rica; Mario Lungo; Institut National de Recherche Scientifique, Connecting Cities with Macroeconomic Concerns: Montreal; and Benemerita Universidad Auton6ma de The Missing Link Puebla, Mexico. Completion dote: December 2002. Income and employment creation (and thus poverty reduction) requires local public goods-such as secu- Emergence from Subsistence: Infrastructure, rity and urban transport-that make urban locations Location, and Development in Nepal more productive (beyond the "pure" gains from agglom- eration) or, more precisely, that lower firms' unit pro- Intuitively, the economic effect of such projects as roads, duction costs. Thus an understanding is needed of the irrigation projects, fertilizer distribution programs, and links between local production costs and job creation. small business assistance schemes would seem to depend This study will measure and document these links and critically on where they are located. But research has shed determine how the quality of local public goods can little light on the effect of location on economic outcomes. affect them. The analysis will draw on data from surveys The purpose of this research is to begin to fill this gap of firms in Brazil, Canada, Costa Rica, El Salvador, and by studying the relationships among infrastructure, geo- Mexico (about 100 firms in each country). graphic location, and economic development in Nepal, Production costs and markets may not appear to be whose extreme diversity in accessibility and proximity "social" issues. But in market economies it is private firms to urban centers makes it a particularly suitable place to that must generate employment if poverty is to decline, study spatial specialization. and the study identifies private firms as the primary To study how proximity to towns and cities affects actors in urban economies, an essential link between households' participation in labor and output markets and microeconoinic and macroeconomic concerns. This focus their allocation of land, the research uses the von Thunen differs from the usual one in efforts to reduce urban model of specialization. It modifies the simple von poverty, where the dominant approach has been to Thunen model to account for several facts: that the size directly target poor people or poor neighborhoods through of the market (measured by city population) affects the local projects, such as slum upgrading, housing con- width of the circle, that the presence of a network of cities struction, and service delivery. The perspective of this enables villagers to trade different goods in different study is metropolitan rather than neighborhood-focused. markets, and that workers in villages surrounding cities The research will shed light on the economic cost of commute to urban centers while workers farther away urban failures, quantifying the lost opportunities in migrate permanently. In the econometric estimation the Infrastructure and Urban Development 83 research takes a nonparametric approach that allows for of households is partly endogenous. If, for example, a vil- a flexible relationship between household decisions and lage is good for vegetable production, it may attract peo- proximity to cities of different sizes. ple good at producing vegetables. Over the years roads The first stage of the research combined household are likely to have been built to serve more promising or data from the 1995-96 Nepal Living Standards Survey productive areas. (NLSS) with geographic information system data on To account for individual-specific fixed effects and the travel time to major cities, using the 1991 population cen- endogeneity of road placement, the second stage of the sus for data on urban populations. Estimation based on project will resurvey the NLSS households in 2001-02, a cross-section of 3,300 households reveals a strong spa- after the population census is completed. 'rhe new sur- tial division of labor. Nonfarm employment is heavily con- vey will provide data on how urbanization and the con- centrated in and around cities (up to four hours of travel struction and upgrading of roads since 1995 have affected time), while agricultural wage employment dominates vil- market participation and geographic patterns of spe- lages located farther away (three to eight hours). Isolated cialization. The research will also investigate how liber- villages (more than eight hours from the nearest city) are alization of trade with India has affected border trade and essentially self-subsistent in both farm and nonfarm the division of labor across space within Nepal. products. Vegetable and cereal production for sale takes The results of the research should help in identify- place near urban centers, while oilseed and other ing strategies for placing infrastructure and develop- commercial crops are more important at intermediate ment projects. And the updated, representative data set distances. produced by the study can be used for further poverty These findings are consistent with the von Thunen assessment in Nepal. model of concentric specialization, and they also show Responsibility: Development Research Group, Rural Devel- the importance of city size. The research also finds that opment-Forhad Shilpi (fshilpi@worldbank.org). With border effects are significant: proximity to Indian towns Marcel Fafchamps, Oxford University; Isuzu Otsuka, does not have the same effect on local specialization as Johns Hopkins University; and Jyotsna Puri, University proximity to Nepalese towns. of Maryland. The Japanese, Netherlands, and Swedish The research used data from the 1999 Labor Force Consultant Trust Funds have contributed funding for the Survey (covering 14,355 households in 719 villages across research. 73 districts) to study the pattern of individual and village- (omplelion date: February 2003. level specialization. The econometric estimation, based on a simple theoretical model of specialization in the pres- Reports ence of increasing returns and agglomeration effects, Fafchamps, Marcel, and Forhad Shilpi. 2000. "CiEies and Spe- reveals that villages in and near cities have more diver- cialization: Evidence from South Asia." World Bank, Devel- sified and more market-oriented activities. This finding opment Research Group, Washington, D.C. implies externalities harnessed through markets. In man- - . 2000. "The Spatial Division of Labor in Nepal." World ufacturing, proximity to cities is associated with larger firm Bank, Development Research Group, Washington, D.C. size and more diversified employment structures. The research finds that urban women specialize more Information, Knowledge, and Capacity Building than rural women. While urban women either work in for Effective Urban Strategies: Information-Based paid jobs or perform household chores, rural women Instruments for Urban Management tend to do both. This finding suggests that with better infrastructure, not only will the pattern of specialization Decisionmakers in urban areas increasingly are becom- change dramatically but so will women's role in the work ing responsible for designing policies and developing pro- force. grams to improve the quality of life of urban residents. Although the results are promising, they are based on To help improve the efficacy of such policies and pro- cross-sectional analysis, in which the geographic location grams, this project has initiated research and analytical 84 Infrastructure and Urban Development work to develop detailed information systems within Somik Lall, Uwe Deichmann, and David Wheeler. With cities for identifying, evaluating, and prioritizing issues Hoon Chang; Vernon Henderson; Marcela Gonzales; related to urban management, urban productivity, and Sudeshna Ghosh; Pulin Nayak; Atul Sarma; and the urban poverty. Focusing initially on Bangladesh and Society for Development Studies, India. The U.K. India, the project is initiating collaborative efforts between Department for International Development is con- the World Bank and local governments, research insti- tributing funding for the research. tutions, the private sector, and nongovernmental orga- Completion date: March 2003. nizations to encourage the collection and use of spatially detailed data and related analytical methods for urban Privatized Utilities planning and policy. Large-sample, geo-referenced house- hold surveys are being administered to capture hetero- Lack of data about the effects of infrastructure privati- geneity within cities in living standards, service delivery, zation has severely constrained the World Bank's ability and resource mobilization. to systematically assess the policy alternatives for such The underlying premise of the work is that public efforts and how these relate to a country's institutional disclosure of credible information offsets costs created environment. This project is developing an extensive by weak institutions. By emphasizing the collection, database of variables crucial to understanding the expe- validation, and public dissemination of data-and thus rience in infrastructure privatization, to enable policy- encouraging transparency, accountability, and participa- makers and Bank staff to learn from the successes and tion in decisionmaking--it is possible to improve the failures in infrastructure reform. The rapid increase in effectiveness of public programs and enhance the countries attempting infrastructure privatization has quality of urban management. made this a difficult undertaking. Early efforts to collect Government officials and researchers have been information from Bank sources identified substantial involved from the outset in designing and implement- gaps that must be filled through country visits and fur- ing the project. In India a national steering committee ther work. of senior government officials and eminent researchers The database currently covers telecommunications in has been formed to provide feedback and ensure that the 40 countries and includes financial information and per- research findings are useful for policy analysis and deci- formance indicators before and after sale, the terms and sionmaking. The project also involves coordination with conditions of the privatization transaction, and details on World Bank operational staff to enhance the utility of the the regulatory framework, as well as a number of insti- work for Bank lending and policy dialogue. tutional variables, although not all series are complete. The project will produce analytical and empirical It also provides thin coverage of electricity, with an ini- publications on the quality of urban life, service deliv- tial series for about 15 countries. ery, and public finance. Findings also will be disseminated When complete, the database will fill important gaps through national and regional workshops. Information on in the Bank's institutional memory and allow staff prepar- the project is available on the Web at http://www. ing and supervising infrastructure projects to quickly info-urban.org. compare performance measures and regulatory frame- Once the data are collected, validated, and analyzed, works with those for successful and unsuccessful priva- efforts will be undertaken to train local government coun- tized firms. terparts in using the data in economic and spatial analy- Analysis using the data will develop detailed guide- sis, build capacity for using and sharing spatial data for lines based on what worked and what failed in privatiz- urban management, and provide training in conducting ing infrastructure. In addition, the data will be an surveys to monitor service delivery and quality of life. important input for the forthcoming World Bank Policy Responsibility: World Development Report Office-Zmarak Research Report on regulation. Shalizi (zshalizi@worldbank.org); and Development Responsib:lity Development Research Group, Regulation and Research Group, Infrastructure and Environment- Competition Policy-Mary Shirley (mshirley@worldbank. Infrastructure and Urban Development 85 org), Luke Haggarty, Colin Xu, and George R. G. Clarke. commune-level data in context, and an extensive With Roger Noll, Stanford University. Stanford Univer- province-level database was created to help understand sity is contributing staff time for the research. the selection of provinces for the project. Because the Completion dote: June 2003. impact of road projects varies with the size of the result- ing changes and the method of project implementation, Rural Roads: Welfare Impact Evaluation a project-level database for each project area surveyed is also being constructed. Rural roads are often seen as key to raising living stan- The baseline data will be used to model the selection dards in poor rural areas. Yet despite the consensus on of project sites, with a focus on the underlying economic, their importance-and much anecdotal evidence-there social, and political economy processes. Later rounds will is surprisingly little hard evidence on the size and nature be used to understand gains measurable at the commune of their benefits. level, conditional on selection. The general approach This study aims to assess the impact of rural roads on will be in the tradition of double differencing with match- poverty and contribute to policy discussions on how best ing methods. Matching methods can be used to select ideal to allocate scarce public resources. The empirical inves- controls from among the 100 sampled nonproject com- tigation is being conducted in Vietnam, where the World munes, and outcomes in project communes can be com- Bank is financing and helping to implement a large- pared with those in the control communes before and after scale rural roads project for poverty reduction. the road projects. Outcome indicators to be examined The study focuses on how the determinants of living include commune-level agricultural yields, income diver- standards change over time in communes that have road sification, employment opportunities, availability of goods, projects compared with ones that do not. It addresses sev- land use and distribution, services and facilities, and asset eral key questions: What is the relationship between wealth and distribution. spending on rural roads and poverty reduction? What Responsibility: Development Research Group, Public impacts do rural roads have on living standards, broadly Services for Human Development-Dominique van de defined to include health, education, and security, and Walle (dvandewalle@worldbank.org). With Vu Tuan Anh, what factors affect those impacts? Do roads really reduce Economics Institute, Hanoi; Dorothy Jean Cratty, Uni- poverty? How do the benefits and their distribution versity of Maryland at College Park; and Jyotsna Jalan, change over time? And how much do the benefits depend Indian Statistical Institute, New Delhi. The Canadian on other investments, such as those in human capital? International Development Agency Trust Fund and the A survey collected baseline data for a random sample U.K. Department for International Development Viet- of 100 project communes and 100 nonproject communes nam Poverty Trust Fund are contributing funding for the in the spring of 1997. A second round of surveying was research. conducted in June 1999, although floods delayed the sur- (ompletion date: December 2005. vey in a few provinces. The data from these two rounds should allow the assessment of initial impacts. To get at Reports longer-term impacts, a third round of data collection is van de Walle, Dominique. 1999. "Assessing thc Poverty Impact of planned for June 2001, and a fourth for two years later. Rural Road Projects." World Bank, Development Research Four other surveys were also conducted in each round. Group, Washington, D.C. In each sampled commune a questionnaire was admin- . 2000. "Choosing Rural Road Investments to Help Reduce istered to 15 randomly sampled households. A short Poverty." Policy Research Working Paper 2458. World Bank, district-level survey was conducted to help put the Dcvelopment Research Group, Washington, D.C. 86 Infrostrucdure and Urban Development Agriculture and Rural Development The Impact of Public Spending in Uganda . 2001. "Uganda's Rural Economy, 1992 to 2000: Accom- plishments and Challenges." World Bank, Development This research used data fiom a series of household sur- Research Group, Washington, D.C. veys (spanning 1992-99) in Uganda to investigate a set Deininger, Klaus, and Donald Larson. 2001. "Crop Markets and of questions about social and economic development in Household Participation in Uganda." In Rirva Reinikka and Paul the country's rural areas. The study produced two main Collier, eds., Ugandas Recovery: The Role of Farms, Firms, and findings. Government. Washington, D.C.: World Bank. First, while education, access to roads, and access to Deininger, Klaus, and John Okidi. 2001. "Market Participation, extension have a significant positive impact on agricul- Agricultural Productivity, and Nonfarm Activities." In Ritva tural production, rural producers overuse nontraded Reinikka and Paul Collier, eds., Uganda's Recovery: The Role of inputs (homegrown seed) and underuse purchased inputs Farms, Firms, and Government. Washington, D.C.: World Bank. (fertilizer). Credit constraints significantly affect demand for inputs. Lack of access to financial services (as mea- Market Development and Allocative Efficiency: sured by distance to the closest bank) negatively affects Irrigation Water in the Punjab the start-up of nonagricultural enterprises and integra- tion into markets for agricultural produce, supporting the With the rapid development of private tubewells, ground- government's emphasis in its rural development strategy water markets have flourished across South Asia in the past on improving access to financial services. few decades. Barriers to entry and extreme spatial frag- Second, knowledge about the causes and treatment mentation in these markets have raised concerns about of diseases, in addition to access to health services and local monopoly. This study used data from Pakistan's their quality, has a significant impact in reducing indi- Punjab, collected by the International Water Manage- viduals' propensity to fall sick. Even under conservative ment Institute, to examine monopoly power in the ground- assumptions, the monetary benefits (derived by valuing water market, with the aim of clarifying a role for policy. the number of days lost at the local wage rates) can be The study used two complementary approaches to test very high. for monopoly power. First, since groundwater markets and Responsibility: Development Research Group, Rural Devel- tenancy contracts are linked in the Punjab, a monopo- opment-Klaus Deininger (kdeininger@worldbank.org). listic tubewell owner would be expected to charge his own With John Okidi, Makerere University, Kampala, Uganda. tenants a lower price than other buyers. Using data on (ompletion date: December 2000. groundwater transactions over an 18-month period, the analysis compared prices charged by a tubewell owner Reports to different customers, including his own tenants. Sec- Deininger, Klaus. 2000. "Does Cost of Schooling Affect Enrollment ond, since the shadow price of groundwater to the tube- by the Poor? Universal Primary Education in Uganda." World well owner (as cultivator) is the marginal extraction cost, 13ank, Development Research Group, Washington, D.C. he would be expected to use more of it per acre than a . 2001. "Causes and Consequence of Civil Strife: Micro- groundwater buyer facing a monopoly price. The study Level Evidence from UJganda." World Bank, Development thus compared groundwater use across buyers, tubewell Research Group, Washington, D.C. owners, and their tenants, at the plot level, as another test .2001. "Long-Term Welfare and Investment Impacts of Sud- for monopoly distortion. den Changes in Family Composition: Evidence from Uganda." The study also analyzed weekly panel data on canal World Bank, Development Research Group, Washington, D.C. water endowments and use over three agricultural sea- 87 sons to test whether informal exchange of canal water fos- region's policies, institutions, and mechanisms of rural ters allocative efficiency, given monopoly pricing of development. groundwater. This research examines the mechanisms of rural The results strongly support the existence of monop- development in Indonesia, the Philippines, and Thailand olistic price discrimination and corresponding water mis- in a historical context. The research has several parts. allocation in one watercourse in Pakistan. But evidence First, it reviews the development of policies that have from canal water transactions indicates that the effect of had direct or indirect effects on the rural sector, partic- this misallocation on crop yields may be somewhat ularly on rural income growth, over the past three decades. blunted by reallocations within the season. Using both It uses quantitative analysis to evaluate the impact of dif- the price and the quantity data, a welfare analysis showed ferent policies on rural income and examine the politi- that monopoly power in the groundwater market has cal, social, and economic background of policy choices. only limited effects on efficiency and equity. In the long Second, a cross-country comparative study looks at invest- run a policy aimed at eliminating monopoly pricing ment and productivity and at trade and domestic mar- would do little to help the poorest farmers. kets. Applying quantitative and comparative institutional The study was done in collaboration with the Inter- economic analysis, the study examines the dynamics of national Water Management Institute field office in rural development and the impact of policies on it. Lahore, Pakistan. Its findings have been presented at a Finally, the research synthesizes the mechanisms of rural World Bank seminar, sponsored jointly by the Devel- development, the impact of policies on the rural sector, opment Research Group and the Water Resources The- and the political, social, and economic background of matic Group, in December 2000, and at seminars at these policies. several universities and research centers during the first Responsibility: Development Research Group, Rural Devel- half of 2001 (Yale University, University of Toulouse, Uni- opment-Takamasa Akiyama (takiyama@worldbank.org), versity of Maryland, Johns Hopkins University, and Maria Amelina, and Donald Larson. With 'Toshihiko INE/DELTA, Paris). Kawagoe and Masa Honma, Seikei University, Tokyo; Jun Responsibility: Development Research Group, Rural Devel- Teranishi, Hitotsubashi University, Tokyo; Y. Hayami, opment-Hanan G. Jacoby (hjacoby@worldbank.org) Aoyama Gakuin University, Tokyo; Nobuhiko Fuwa and and Rinku Murgai. With Saced Ur Rehman, International Professor Kikuchi, Chiba University, Tokyo; N. Okura, Water Management Institute, Lahore, Pakistan. Senshu University, Tokyo; Yair Mundlak, University of Completion dole: June 2001. Chicago; Peter Timmer, Harvard University; Shin Shige- tomi, Institute of Developing Economies, Japan; Professor Report Manarungsan, Chulalongkorn University, Bangkok; Pro- Jacoby, Hanan G., Rinku Murgai, and Saeed Ur Rehman. 2001. fessor Balisacan, University of the Philippines; and Dr. "Monopoly Power and Distribution in Fragmented Markers: The Dillon, Center for Agricultural Policy Studies, Indone- Case of Groundwater." Policy Research Working Paper 2628. sia. The Japanese Consulting Trust Fund for Large- World Bank, Development Research Group, Washington, D.C. Scale Studies has contributed funding for the research. Completion date: October 2001. The Dynamics of Rural Sector Growth Reports Although many studies have analyzed the spectacular eco- Akiyama, Takamasa. "Agricultural Pricing Policies in Three South- nomic growth of Asian countries, they have given little east Asian Countries: lime-Series Analysis over Four Decades." attention to the rural sector, the mechanisms of its growth, Draft. or the impact of government policies on the sector. Asian . "Agricultural Sector Growth: Any Peculiarities with South- countries have achieved impressive agricultural and rural east Asia?" Draft. development, and the World Bank and its client coun- Amelina, Maria, and E Plessman. "Do Different Regimes Distort tries would benefit from a comprehensive study of the Differently?" Draft. 88 Agriculture and Rural Development Butzer, Rita, Yair Mundlak, and Donald Larson. "Intersectoral Turkey-with different physical, social, economic, and Migration in Thailand, Indonesia, and the Philippines." institutional conditions. It developed a cost-benefit Draft. framework to evaluate the performance of different water Fuwa, Nobuhiko, and others. "The Political Economy of Rural pricing methods. And it is carrying out an economywide Development in the Philippines since the 1960s." Draft. analysis (applying a computable general equilibrium Hayami, Y. Forthcoming. "Ecology, History, and Development: A framework) in Morocco. Perspective from Rural Southeast Asia." World Bank Research Country reports and databases for China, Morocco, Observer. South Africa, and Turkey have been completed, and Honma, Masa. "A Comparative Study on Agricultural Exports of reports and a database for Mexico are being finalized. Three Southeast Asian Countries." Draft. Mathematical programming models that capture various Kawagoe, Toshihiko. "Political Economy of Rural Development aspects of water pricing at the regional level have been in Indonesia." Draft. completed for several perimeters in Morocco and for one Larson, Donald, and F. Plessman. "Do Farmers Choose to Be irrigation scheme in China, in South Africa, and in Turkey. Inefficient? Evidence from Bicol." Draft. Analysis in five irrigation districts in four of the coun- Mundlak, Yair, Rita Butzer, and Donald Larson. "Determinants of tries (China, Morocco, South Africa, and Turkey) has Agricultural Production in Thailand, Indonesia, and the Philip- focused on estimating the (derived) demand for irriga- pines." Draft. tion water under the conditions in each location. Fol- Okura, N., and Jun Teranishi. "Impact of Politics on Rural Resource lowing are some of the important findings: Allocation." Draft. * Although volumetric pricing is both efficient and Shigetomi, Shin. "Development with Poor Farmers: Four Decades equitable, the costs of implementing it can be prohibi- of Thai Agriculture and Related Policies." Draft. tive. A combination of per-area charging and water quo- tas can achieve efficiency results similar to those of Guidelines for Pricing Irrigation Water volumetric pricing. A combination of tiered pricing (tak- Based on Efficiency, Implementation, ing care of variable costs) and per-area charges (taking care and Equity Considerations of fixed costs) can achieve similar efficiency and equity results-and better cost recovery. World Bank-supported irrigation projects increasingly * Similar pricing policies can have very different emphasize pricing as a primary means of regulation. effects under different conditions, as reflected in the Conditions for the disbursement of a loan for construct- shape (elasticity) of the derived demand curves. Farms ing an irrigation project often require "appropriate" pric- with steep curves (inelastic demand) will be less respon- ing of the irrigation water that will be generated. Yet it sive to price increases than farms with more horizontal is unclear what "appropriate" water prices are or how they curves (elastic demand). should be applied. Disagreements among competing * Farmers respond to water prices. The nature of that groups of water users are common, particularly if they are response (derived demand) reflects both endogenous in different economic sectors. Surprisingly, economists conditions (big or small farm, rich or poor farm, crop also disagree on these issues. mix) and exogenous conditions (water supply reliability, This project clarifies the basic concept of water pric- water institutions, prices of other inputs and outputs, ing and lays out a set of guidelines on how to price irri- extension and availability of appropriate technologies, pro- gation water under different circumstances. The primary duction quotas, access to markets and credit). To be measure of performance is efficiency, broadly defined to effective, policy interventions therefore need to address include implementation costs. Income distribution, avail- these conditions. able water institutions, and political constraints are con- Because agriculture remains an important sector in sidered in case studies. most low- and middle-income countries, and because the The project conducted detailed studies in five agricultural production process is affected by many poli- countrics--China, Mexico, Morocco, South Africa, and cies, water pricing policies must be viewed in an econ- Agriculture and Rural Development 89 omywide context. The economywide analysis in Morocco Findings were presented at a workshop for World has produced such findings as these: Bank staff and participants from the U.S. Department of o Trade liberalization increases the shadow price Agriculture, the U.S. Environmental Protection Agency, (opportunity cost) of water for fruits and vegetables, and the International Food Policy Research Institute compared with the values before reform, and reduces the on February 27, 2001. shadow cost of water for the crops that were protected Responsibility: Rural Development Department-Ariel Dinar before reform. (adinar@worldbank.org). With Yakov Tsur and Vladimir o In the post-trade reform environment a water allo- Lubinsky, Hebrew University, Rehovot, Israel; Terry cation mechanism allows Moroccan water authorities to Roe and Robert Johansson, University of Minnesota; manage the sector exactly as they have historically (for Rachid Doukkali, Institute of Veterinary Science and example, by allocating water quotas to farmers and crops), Agriculture, Rabat, Morocco; Michael Schur, Ministry of except that they now grant farmers water rights to their Finance, South Africa; Enrique Aguilar; and Zhou Yaozhou quotas and permit them to rent water out, rent water in, and Wei Bingcai, Ministry of Water Resources, China. or simply proceed as in the past. This mechanism allows Completion date: October 2001. water markets to clear at a relatively uniform shadow price throughout a perimeter, allows farmers to internalize Reports the rents earned (and thus to recoup much of the loss from Dinar, Ariel. 2001. "Insights from the Study 'Pricing Irrigation trade reform), and allows expanded production of the Water: Efficiency, Equity, and Implementation Costs."' crops in which the country has a comparative advantage Dissemination Note. World Bank, Rural Development Depart- in world markets. ment, Washington, D.C. o Thus trade reform (and, by extrapolation, related Johansson, Robert C. 2000. "Pricing Irrigation Water: A Literature reforms) may create an opportunity to introduce water Survey." Policy Research Working Paper 2449. World Bank, pricing reforms with less political controversy, because Rural Development Department, Washington, D.C. it allows more flexibility in adjusting to the pricing reform Johansson, Robert C., Yacov Tsur, 'Ferry L. Roe, Rachid M. and to the internalization of water rents. Doukkali, and Ariel Dinar. 2001. "Pricing and Allocation of o In the context of economic growth these new incen- Irrigation Water: A Review of Theory and Practice." World tives cause capital deepening in the comparative advan- Bank, Rural Development Department, Washington, D.C. tage crops. Moreover, since the economy is more efficient and savings are now a larger share of a larger GDP, cap- Reforming China's Agricultural Research System: ital accumulation occurs in the other agricultural sub- A Research Production Function Approach sectors. As a result, for many of the crops that the country to Measuring Economies of Size and Scope cannot produce as cheaply as the rest of the world (such and Efficiency as soft wheat), production often approaches pre-reform levels after about seven years (reassuring from a food secu- China's agricultural research system, the world's largest, rity perspective). has a strong record of past accomplishments, but today o The growth in agricultural GDP encouraged by it can be characterized as bloated, sluggish, and unre- these policies also increases the social returns to invest- sponsive to client needs. To overcome the most serious ments in canals, dams, and other fixed structures. In an inefficiencies, China's government plans radical changes environment where the marginal cost of water supply is to the system beginning in 2001. Planned reforms involve minuscule relative to these fixed costs, however, the cutting up to 80 percent of the agricultural scientists marginal cost pricing that leads to the efficient allocation and increasing salaries and research support for the of water does not lead to total cost recovery. The econ- remaining 20 percent-the best and the brightest-by omywide framework helps to assess the least resource- 200-300 percent. distorting mechanisms that might be used to recover To assist policymakers and research administrators the remaining fixed costs. in defining efficient research policy and to provide guid- 90 Agriculture and Rural Development ance for the reform, this research project is developing a Weather-Based Index Insurance methodological framework for studying the process of creating new agricultural technology, particularly new cul- This study is assessing the feasibility of weather-based tivars. The research focuses on understanding the nature index insurance in four countries-Ethiopia, Morocco, of the economies of size and scope that exist (or do not Nicaragua, and Tunisia. Weather insurance is triggered exist) in China's agricultural research system, addressing by certain weather events (such as rainfall, temperature, such questions as these: What is the most efficient size for and wind speed). Because it does not require inspecting research institutes? Are several large institutes better than individual farms, it reduces administrative costs and low- many small ones? What types of research institutes have ers the risk of adverse selection and moral hazard asso- been most productive-those with only breeders or those ciated with traditional insurance. with a broad range of agricultural scientists? In other In the four study countries droughts have severely words, how strong are the economies of scope? The affected agricultural production, GDP, and farmers' abil- research also seeks to identify other institutional factors ity to repay loans. Linking drought insurance to credit that may affect agricultural research productivity in China. could significantly improve loan recovery. Moreover, The project will construct a production (or cost) func- weather-based insurance could lessen the pressure on gov- tion for research on crop improvement, in which the ernment budgets from ad hoc drought relief. output is the number of varieties produced. Inputs The feasibility analysis for Morocco and Tunisia was include research expenditures and number of researchers based on rainfall and yield data for critical crops in spe- in breeding programs, research expenditures and num- cific areas of each country. The analysis computed a ber of scientists in other programs at the same institutes, value at risk defined as the revenue from the crops, and institutional factors. The project will test for which is equal to area planted times yield times price (area economies of size and scope based on the estimation planted and price are fixed-that is, only yield varies). results of the research production function. The project Using information from agronomic analysis on the crit- is funding a survey of about 100 wheat and maize breed- ical periods of rainfall that affect the crops in the selected ing programs throughout China to provide the basic data regions, the analysis established the correlation between for estimating the research production function. Data will crop revenue and rainfall variations. The results indicate be collected on costs and outputs over a 10-year period. a wide variation in the correlation coefficients among The research is being implemented by the Chinese regions in each country. Center for Agricultural Policy and carried out by Chinese For Morocco the correlation between crop revenue and scientists and economists, with inputs from other team rainfall was found to be sufficiently strong in three cli- members. matic zones that account for 71 percent of the estimated The project will provide key information for design- revenue for the crops. The correlation coefficients for ing a World 13ank loan to support agricultural research in 1979-99 ranged between 60 percent and 85 percent. China. Findings will be widely discussed with Chinese For Tunisia the correlations between crop revenue and policymakers in a series of seminars and workshops. rainfall were somewhat weaker, ranging between 50 per- Responsibility: Rural Development Department-Derek cent and 70 percent in several regions. Byerlee (dbyerlee@worldbank.org); and East Asia and Next, the analysis simulated the performance of a Pacific Region, Beijing Country Office-Jiurgen Vogele. proportional rainfall contract with 5 percent and 10 per- With Linxiu Zhang, Jikun Huang, and Ruifa Hu, Chi- cent pure premium cost. For Morocco insurance would nese Center for Agricultural Policy, Beijing; and Songqing lower crop revenue risk by up to 30 percent. For Tunisia Jin, Scott Rozelle, and Julian Alston, University of Cal- it would lower the revenue risk by about 20 percent. ifornia at Davis. The Chinese Center for Agricultural For Nicaragua data limitations prevented the study Policy and the University of California at Davis are from establishing a correlation between yields and rain- contributing staff time for the research. fall. But the project continues to investigate the impact (ompletion date: October 200(1. of rainfall on agricultural production. It also has simulated Agriculture and Rural Development 91 the performance of a proportional weather contract with establishing the Global Weather Risk Facility to help a 5 percent and 10 percent premium cost. Historical weather-exposed businesses and farmers in emerging simulations showed that this contract would have made markets to manage their risks and improve their credit- some indemnity payments about every three to four worthiness. years, payments that producers might value if they pre- Responsibility: Development Research Group, Rural Devel- fer more frequent payments and are affected by less cat- opment-Panos Varangis (pvarangis@worldbank.org) astrophic but more frequent droughts. Analysis of farm and Donald Larson; Financial Sector Development budgets indicates that a drop in yields below 20 percent Department-Vijay Kalavakonda, Rodney Lester, and could significantly affect many farmers' profitability and Don Mclsaac; Middle East and North Africa Region, their ability to repay loans. Yield losses of 40-50 percent Private Sector Development and Finance Group- or more would probably cause crucial problems in loan Stephanie Gober; and Human Development Network, repayments. Social Protection Team-Paul Siegel. With Peter Hazell, In Ethiopia the study has found that weather-related International Food Policy Research Institute; Pasquale risks are a fundamental obstacle to development. Weather Scandizzo, University of Rome and Sichelgaita, Salerno, station data confirm that weather patterns vary through- Italy; Roberto Pasca, University of Rome; Andrea Stoppa, out Ethiopia. The cost of weather insurance would there- Procom, Italy; Sonia Carbone and Marco Zupi, Sichel- forevary-sometimesdramatically-amongcommunities. gaita, Salerno, Italy; Jerry Skees, University of Ken- But the pure risk premiums for some stations would not tucky; and Mario Miranda, Ohio State University. The preclude insurance markets. The institutions charged Italian Trust Fund and the Development Marketplace with gathering, analyzing, and disseminating the basic are contributing funding for the research. data needed to support weather insurance in Ethiopia are (ompletion dote: November 2001. adequate. But the private markets needed to distribute weather insurance instruments are weak and are further Report hampered by regulations precluding foreign partner- Skees, Jerry, Stephanie Gober, Panos Varangis, Rodney Lester, and ships. And oversight of the insurance and microfinance Vijay Kalavakonda. 2001. "Developing Rainfall-Based Index industries is weak because of limited capacity. Public and Insurance in Morocco." Policy Research Working Paper 2577. donor-supported institutions that currently deliver credit World Bank, Development Research Group, Washington, D.C. and input packages could be used to deliver weather insurance. Agriculture in Transition: Land Reform and Farm In future work the study will evaluate the potential Restructuring in Formerly Socialist Countries demand for rainfall insurance in two of the countries and assess the institutional feasibility of supplying such In the formerly socialist countries of Europe and Cen- insurance. tral Asia land reform and farm restructuring arc among The results of the study have been presented at sem- the main elements of the transition to the market in inars and conferences, including a Weather Risk Man- agriculture. The World Bank has been monitoring agement Association meeting in Bermuda on June 5-7, progress in these aspects of transition since 1992 for all 2001; the United Nations University, World Institute 23 countries of the former Soviet Union and Central for Development Economics Research meeting on Insur- and Eastern Europe through country sector reviews sup- ance against Poverty in Helsinki, Finland, on June 15-16, plemented by extensive farm-level surveys. This mon- 2001; and a conference, Agricultural Insurance: New itoring work shows that despite the countries' common Global Trends, organized by the Latin American Asso- heritage of socialist command economy with pervasive ciation for the Development of Agricultural Insurance in collectivization of agriculture, they are not following the Buenos Aires, Argentina, on November 14-17, 2000. same path in market reforms. Based on the results of the study, the International This research is examining the different paths of land Finance Corporation and several private partners are reform and farm restructuring in Europe and Central 92 Agriculture and Rural Development Asia to produce a coherent picture of the agrarian trans- ment Research Group, Rural Development-Gershon formation in the region. It aims to analyze the causes of Feder; and Africa Region Technical Families, Rural the observed patterns of transformation; establish rela- Development 1-Karen Brooks. With Zvi Lerman, tionships between a country's progress in agricultural Hebrew University, Israel. reforms and its social, cultural, political, and economic (ompletion dote: December 2001. profile; and formulate policy conclusions based on an inte- grated overview of the processes and experiences Reports throughout the region. Lerman, Zvi. 1999. "Comparison of Restructured and Nonre- The research combines several methodological structured Farm Enterprises in Ukraine." Paper presented at approaches. The first involves generalizing and synthe- the Fourth Annual Convention of the Association for the Study sizing information on agricultural transformation in the of Nationalities, Columbia University, New York, April. region to evaluate progress in land reform and farm .1999. "Farm Restructuring Experience: Comparison of the restructuring, through a detailed comparison of farming Former Soviet Union and Eastern and Central Europc." Paper structures in the formerly socialist countries with those presented at the Golitsino Workshop on Farm Profitability, in market economies. The second develops a multi- Sustainability, and Restructuring, U.S. Agency for Interna- variate typology of agricultural reform in different coun- tional Development and Institute for Economy in Transition, tries through cluster analysis, supplementing and Moscow, October. validating existing subjective rankings. The third . 1999. "From Commonality to Divergence: How ECE and approach applies statistical and econometric tools to CIS Agricultures Are Drifting Apart." Paper presented at the detect the impact of reform by analyzing differences in conference Land Ownership, Land Markets, and Their performance, income, and satisfaction between different Influence on the Efficiency of Agricultural Production in Cen- groups of stakeholders, between regions with different tral and Eastern Europe, Institute of Agricultural Develop- levels of reform, and between countries with different ment in Central and Eastern Europe and Food and Agriculture political orientations. It applies data envelopment analy- Organization, Halle, Germany, May. sis to estimate differences in efficiency between collec- . 1999. "Land Reform and Farm Restructuring: What Has tives and family farms. Been Accomplished to Date?" American Economic Review 89(2): The research will produce a book summarizing and 271-75. generalizing the experiences of the first decade of agri- . 1999. "Record of Land Reform in Transitional Economies." cultural transition. The book will give policymakers in Paper presented at the Organisation for Economic the region a cross-country perspective and thus enable Co-operation and Development Forum on Agricultural Policies them to judge on the basis of empirical information what in Nonmember Countries, Paris, April. transition strategies work better and why. And it will . 1999. "Status of Land Reform and Farm Restructuring in give international donors and others outside the region the CEE Countries: A Regional Overview." Paper presented a competent overview of the first decade of reforms, at the Second European Union Accession Workshop in the contributing to the development of assistance strategies Rural Sector, World Bank and Food and Agriculture Organi- for the next decade. zation, Warsaw, June. As part of ongoing dissemination efforts, partial results . 2000. "From Common Heritage to Divergence: Why the of the research have been presented to international audi- Transition Countries Are Drifting Apart by Measures of ences, including scholars from formerly socialist coun- Agricultural Performance." American Journal of Agricultural tries in Europe and Central Asia. Upon completion of the Economics 82(5): 1140-48. research, a special workshop will be organized for schol- . 2001. "A Decade of Land Reform and Farm Restructur- ars, policymakers, and government officials in the region. ing: What Russia Can Learn from the World Experience." Responsibility: Europe and Central Asia Region, Environ- QuarterlyJournalof InternationalAgriculture 40(1): 5-28. mentally and Socially Sustainable Development Sector . 2001. "Institutions and Technologies for Subsistence Unit-Csaba Csaki (ccsaki@worldbank.org); Develop- Agriculture: How to Increase Commercialization." Paper Agricullure and Rural Development 93 presented at rhe conference Subsistence Agriculture in For Indonesia a 1990-91 survey with a with-and- Central and Eastern Europe: How to Break the Vicious Circle, without design serves as the baseline. The same house- Institute of Agricultural Development in Central and Eastern holds (about 450) were resurveyed, and the household Europe, Halle, Germany, May. data from the two panels were mapped and are being ana- lyzed. In Peru a baseline survey covered about 500 Sending Farmers Back to School: households. The Peru study analyzed the baseline data An Econometric Evaluation of the but dropped the full before-and-after analysis because Farmer Field School Extension Approach of contamination of the control group by interventions similar to farmer field schools. In the Philippines a sur- Studies of the economic and other effects of agricultural vey of some 300 households (including some of the 45 extension programs, particularly traditional training and farmers tracked by a separate study since 1995) has been visit programs, have yielded a mixed record. The many completed and used to explore the spread effects of problems associated with administering large and often farmer field schools and the retention of knowledge. expanding extension bureaucracies; the growing disen- Evidence from Indonesia and the Philippines chantment with packaged technologies and the top- shows that fiscal unsustainability is a risk that cannot be down, centralized approach to disseminating knowledge; ignored if the farmer field school approach is applied on and the general alienation of farmers from the creation a large scale. Because of the high costs per trained farmer, and dissemination of knowledge have led many to ques- the amount of funding for extension in the Philippines tion staff-intensive public extension systems. And in cannot provide for significant farmer outreach. Moreover, the light of growing fiscal deficits in many countries, a the Philippine case study shows little informal farmer- particular concern is economic sustainability. to-farmer diffusion of knowledge. Program coverage In recent years many development agencies, includ- would have to be extensive to have significant ing the World Bank, have promoted farmer field schools impact. as a more promising approach to extending science- Some view farmer-led field schools as a way out of the based knowledge and practices to farmers. The approach fiscal dilemma if part of the cost is shifted to the com- relies on participatory training methods to convey knowl- munity. But survey data from Indonesia show that farm- edge to field school participants and encourages all grad- ers have taken over little formal training responsibility. uates to share their experiences with other farmers. The And even these farmer-led schools are not funded mainly spread of knowledge from farmer to farmer is expected by community resources. The results suggest a need for to make the approach cost-effective and sustainable. great selectivity and caution in initiating pilot field school This research evaluates the performance of farmer field programs, with a focus on fiscal sustainability if the schools at the farm and program level in Indonesia, Peru, intention is to scale up the program. and the Philippines. At the farm level farm profits are used A case study of rice farmers in Iloilo, Philippines, as a key indicator of success. Farm-level impacts are found that field school graduates generally score higher measured econometrically using both with-and-without than their counterparts on tests of knowledge typically and before-and-after comparisons with panel data, to acquired in a field school, and that there is little diffu- deal with estimation problems arising from individual sion of the knowledge from graduates to other commu- selection and program placement biases. The empirical nity members. There appear to be no significant strategy also seeks to isolate secondary effects of farmer- differences in scores betwveen old and new graduates, sug- to-farmer diffusion of knowledge from direct program gesting that graduates retain the knowledge acquired. effects on participants and to measure the depth and A case study in Peru pointed to several factors explain- breadth of this knowledge diffusion through case stud- ing farmers' participation in farmer field schools: whether ies and participant observation. At the program level or not the schools are based on preexisting local groups the study aggregates the farm effects to compare the with preexisting rules for participation that exclude some program's overall economic benefits with its overall costs. community members; farmers' perceptions of the ben- 94 Agriculture and Rural Development efits of participation, which are influenced by experience Welfare Outcomes of Decentralized Land Reform with other extension organizations and, often, the receipt of short-term economic benefits; the gender division of Major policy reforms in China and Vietnam effectively labor in potato cultivation (which resulted in male- privatized most farmland, one of the most important dominated field schools); and competition from farming assets for the security of poor people. Institutional reform activities. Farmers perceived the cultivation of new on this scale raises many questions: Were central dictates potato varieties (introduced through the farmer field on how the land should be allocated followed locally? school) and the knowledge of which fungicides to use as What were the implicit objectives of local decisionmak- the main benefits of participation. ers? How did they weigh equity against efficiency? Did The research has involved close collaboration with staff they forgo aggregate income gains for their preferred of local organizations in Indonesia, Peru, and the distribution, and how much did they forgo? Did the Philippines. The results of the Philippine study were tradeoff between equity and efficiency improve over presented at seminars in the Philippines and at the World time, through feedback effects of land distribution on Bank in 2001. Preliminary results from the Indonesia growth? How much did the fact that the land realloca- study were presented at seminars in Nairobi, Kenya, tion was limited to existing members of each commune and Bogor, Indonesia, in November 2000. (with no scope for redistribution between communes or Responsibilily: Development Research Group, Rural Devel- for mobility) constrain its overall distributional impact? opment-Gershon Feder (gfeder@worldbank.org), Jaime This project addresses these questions, studying the Quizon, and Rinku Murgai. With Tahlim Sudaryanto, socioeconomic determinants of the land allocations and Sjaiful Bahri, and Hania Rahma, Center for Agro- the premium local decisionmakers appear to have Socioeconomic Research, Indonesia; Agnes Rola, attached to equity. The work draws in part on discussions University of the Philippines at Los Banios; Rebecca with policymakers and others, but much of it involves an Nelson, Paul Winters, and Oscar Ortiz, International econometric analysis of household- and community- Potato Center, Lima, Peru; and Hermann Waibel, level socioeconomic data. University of Hannover, Germany. Household panel data allowing an examination of the Completion date: [)ecember 2001. issues are available for both Vietnam and China. For Vietnam the quantitative data are from the Vietnam Liv- Reports ing Standards Surveys of 1992-93 and 1998, covering Quizon, Jaime, Gershon Feder, and Rinku Murgai. Forthcoming. about 4,000 rural (3,457 panel) households in 120 com- "Fiscal Susrainabiliry of Agricultural Extension: The Case of munes throughout the country. For China similar data are the Farmer Field School Approach." Journal of InternationalAgri- available from the Rural Household Surveys, for four con- cultural and Extension Education. tiguous southern provinces (Guangdong, Guangxi, Rola, Agnes, Jaime Quizon, and S. Jamias. 2001. "Do Farmer Guizhou, and Yunnan). The panel spans 1985-90 and cov- Field School Graduates Recain and Share What They Learn? ers 5,000 households. The panel data feature of both An Investigation in Iloilo, Philippines." World Bank, Devel- surveys permits estimation of dynamic models of con- opment Research Group, Washington, D.C. sumption growth and land allocation and also allows for Rola, Agnes, Jaime Quizon, S. Jamias, M. Paunlagui, and Z. Provido. the possibility of land reallocation. 2000. "Spread of IPM-FFS Knowledge: Lessons Learned from The research will provide a deeper understanding of Rice Farming Communities in Iloilo, Philippines." Institute of the ways in which central policy initiatives affect welfare Strategic Planning and Policy Studies Working Paper 00-07. Uni- when key aspects of policy implementation are decen- versiry of the Philippines at Los Bafnos. tralized. Decentralizing policy implementation is a com- Vasquez-Caicedo, G., J. Portocarrero, Oscar Ortiz, and C. Fon- mon practice. Indeed, the World Bank often advises seca. 2000. "Case Study on Farmers'Perceptions about Farmer central governments to rely more on better informed Field School Implementation in San Miguel, Peru." Interna- local agents. But without a proper understanding of local tional Potato Center, Lima, Peru. political economy, is such advice well founded and will Agritulture and Rural Development 95 the expected gains materialize? This question has rele- rely much more on market solutions than on adminis- vance for a wide range of development initiatives. trative arrangements. Most aim at easing access to risk Responsibility: Development Research Group, Public Services management instruments, such as commodity futures, for Human Development-Dominique van de Walle options, and swaps, and rely on households and firms to (dvandewalle@worldbank.org) and Poverty Team- make appropriate choices. Martin Ravallion. With Tomomi Tanaka, University of The evidence suggests that while farmers' direct Hawaii, East-West Center. access to market-based risk management instruments may Completion dote: April 2002. be limited, they can gain indirect access to these instru- ments through producer associations, rural credit institu- Report tions, and traders and processors. Thus identifying local Ravallion, Martin, and Dominique van de Walle. 2001. "Breaking institutions that could provide price risk management up the Collective Farm: Welfare Outcomes of Vietnam's Mas- instruments to farmers is key. These institutions can then sive Land Privatization." World Bank, Development Research hedge their own exposure in international markets. Group, Washington, D.C. Responsibility: Development Research Group, Rural Devel- opment-Panos Varangis (pvarangis@worldbank.org), Commodity Risk Management and Export Finance Donald Larson, and Takamasa Akiyama. With Christo- pher Gilbert, Free University of Amsterdam; and Alexan- Policymakers have long debated how best to manage der Sarris, University of Athens. volatility in commodity prices. The solutions offered Completion date: June 2002. by policymakers and the resulting institutions- international commodity agreements, marketing boards, Report compensatory financing, stabilization funds-were based Akiyama, Takamasa, Donald Larson, and Panos Varangis, eds. on collective responses largely coordinated by govern- 2001. "Agricultural Risk Management." World Bank, Devel- ment. The dominant concerns of macroeconomic growth opment Research Group, Washington, D.C. Draft. and stability also influenced the approaches that gov- ernments chose. Land Markets, Gender, and Access During the 1990s the focus of the debate shifted from to Land in Latin America governments and sectors to households and firms. More- over, a series of events-including the collapse of the This study investigates how recent property rights reforms Soviet Union-initiated policy and institutional changes aimed at improving the functioning of rural factor mar- that profoundly altered how most commodity markets kets in Honduras, Mexico, and Nicaragua have affected worked. Marketing boards were eliminated, and thus agricultural productivity and the lives of the rural poor. so were trade interventions and publicly financed min- It also examines how the reforms have affected the prop- imum price guarantees. Producing firms and households erty rights of women in rural areas and therefore the were forced to become self-reliant. For many agricultural well-being of rural households. The study is part of a producers who had been implicitly taxed under earlier broader research agenda ultimately aimed at identifying rules, the institutional changes brought higher incomes, policies that can help to improve the functioning of land even as income volatility increased. For other producers markets and increase equity and productivity. It relies on the changes have come at a cost. microeconometric analysis of recently collected panel data This study synthesizes the results and experiences pre- and data collected in the course of the project in Mex- sented in papers that examined institutional aspects of ico and Nicaragua and in Honduras, where data collec- dealing with commodity price fluctuations in develop- tion has only recently been completed. ing countries, focusing on the change in approaches to Results for Mexico indicate that legal and institu- risk management preferred by policymakers. New tional changes, together with a participatory and trans- approaches to reducing uncertainty in commodity prices parent program of tenure regularization (the Program 96 Agriculture and Rural Developmenl for Certification of Ejidal Rights, or Procede), led to Economic Impacts." World Bank, Development Research both economic and noneconomic benefits. The reform Group, Washington, D.C. reduced conflicts, significantly increased access to land . "Mexico's Ejido Reforms: Their Impact on the Func- for formerly marginalized groups, improved governance tioning of Factor Markets and Land Access." World Bank, and accountability at the local level, and had a clearly pos- Development Research Group, Washington, D.C. itive impact on equity. Econometric evidence suggests Deininger, Klaus, and Juan Sebastian Chamorro. "Investment and that Procede, but not the legal reforms, improved the Income Effects of Land Regularization: The Case of Nicaragua." functioning of rental markets and access to common World Bank, Development Research Group, Washington, D.C. property resources, though it had no significant impact . "Land Markets, Land Access, and the Impact of Policy: on land sales or access to credit. In addition, the reform The Case of Nicaragua." World Bank, Development Research significantly reduced the transaction costs associated Group, Washington, D.C. with land rentals. But because of credit market imper- Deininger, Klaus, and Isabel Lavadenz. "Determinants and Impacts fections, the poor were often unable to take advantage of Rural Land Market Activity: Evidence from Nicaragua." of the increased market opportunities. The economic World Bank, Development Research Group, Washington, D.C. benefits were generated mainly through increased Deininger, Klaus, Isabel Lavadenz, Fabrizio Bresciani, and Manuel incorporation of ejidatarios (smallholders) into the Diaz. "Mexico's Second Agrarian Reform: Implementation nonfarm sector. and Impact." World Bank, Development Research Group, In Nicaragua liberalization has improved the func- Washington, D.C. tioning of rental markets, and many landless farmers have been able to make the transition from rental to ownership. Land Reform At the same time, the rental market remains very small. Lack of properly registered titles, a problem experienced New opportunities and lingering problems have renewed disproportionately by the poor, hampers the supply of the interest of developingcountry governments in agrar- land to the rental market. The poor express a clear demand ian reform: for increased tenure security but lack the resources to * The inability to address rural violence, deep-rooted register their titles. Receipt of registered titles increases poverty, and inequality of opportunity through conven- land values by 30 percent and increases the propensity to tional means, and the expectation that a redistribution invest, bringing investment closer to the optimum. These of productive assets would improve the access of the poor findings suggest not only that the legal validity of the to economic opportunities. certificates issued is of overriding importance but also * The potential for improving productivity and equity that titling can have a positive distributional effect. given the large tracts of unutilized or underutilized land Responsibility: Development Research Group, Rural Devel- on large farms, the inverse relationship between farm size opment-Klaus Deininger (kdeininger@worldbank.org). and productivity, and the desire of large landlords to liq- With Michael Carter and Bradford Barham, University uidate their landholdings. of Wisconsin at Madison; and Alain de Janvry and * The failure of old-style land reform, the continuing Elisabeth Sadoulet, University of California at Berkeley. political pressure for land reform, and the drop in land The European Union is funding a major part of the prices following the elimination of many distortions Honduras component, and the U.S. Agency for Inter- favoring agriculture. national Development contributed funding for the This research project aims to quantify the scope for surveys in Honduras and Nicaragua. and economic potential of land reform, help in the design Completion dole: June 2002. of programs that would realize this potential, and mount monitoring and evaluation systems that would allow in- Reports depth evaluation of the impact of such programs on both Deininger, Klaus, and Fabrizio Bresciani. "Implementing Property productivity and poverty reduction. The project combines Rights Reforms in Mexico's Ejidos: Political Preconditions and ex ante analysis of land reform based on farm budget Agriculture and Rural Development 97 analysis with econometric analysis of survey data for Deininger, Klaus, and Hans Binswanger. 1999. "The Evolution of Brazil, Colombia, Guatemala, Malawi, the Philippines, the World Bank's Land Policy: Principles, Experience, and South Africa, and Zimbabwe. Country-specific results and Future Challenges." World Bank Research Observer 14(2): 247-76. international experience provide feedback on ongoing Deininger, Klaus, and Julian May. 2000. "Can There Be Growth efforts. with Equity? An Initial Assessmcnt of Land Reform in South In addition to helping to establish baseline surveys, Africa." Policy Research Working Paper 2451. World Bank, the project has undertaken several case study evaluations. Development Research Group, Washington, D.C. In general, these demonstrate the scope for increasing Deininger, Klaus, M. Maertens, Pedro Olinto, and F. Lara. 2000. productivity through land reform and the advantages of "Redistribution, Investment, and Human Capital Accumula- a demand-driven approach. But they also highlight the tion: The Case of Agrarian Reform in the Philippines." World importance of designing mechanisms for implementing Bank, Development Research Group, Washington, D.C. incentive-compatible reforms to ensure replicability and economic viability. Land Tenure in Rural China The findings support the World Bank's stance in favor of land reform, and the efforts in developing countries This research aims to quantify the costs and benefits of to address long-standing issues of maldistribution of improving the definition of land use rights in China and assets. They also emphasize the importance of placing to determine the distribution of gains and losses accru- such efforts in a broader policy context. ing to different types of farmers. It is based on a resur- The research has contributed to che design of Bank vey of a panel of about 1,000 farm households in three operations by pointing to elements critical to success provinces for which comprehensive data are available (such as focusing on integrated productive projects, from four rounds of the State Statistical Bureau's annual involving financial intermediaries, and using a decen- household survey (1995-99). tralized approach) and by elaborating a framework for The study tests the following hypotheses: monitoring and evaluating projects that will eventually * Even where formal credit markets are absent, secure allow an in-depth assessment of their impact. The tenure is very important forincreasinghousehold invest- research has also helped to inform the policy debate and ment and wealth. clarify the potential and limits of land reform in countries * The greater tenure security resulting from property where it remains controversial. Results have been pre- rights reform increases the propensity to participate in sented at conferences and workshops for policymakers factor markets (for land and, through out-migration and in Brazil, the Philippines, South Africa, and Zimbabwe. off-farm employment, for labor), thus generating gains Responsibility: Development Research Group, Rural Devel- in allocative efficiency that tend to increase household opment-Klaus Deininger (kdeininger@worldbank.org) income and improve rural well-being. and Pedro Olinto. The Swiss Trust Fund and the Brazil- * If initial land endowments are distributed equitably ian government are contributing funding for the research. and education and off-farm employment are widely acces- Complelion date: June 2002. sible, less costly means of insurance than periodic redis- tribution of land are likely to be available, implying that Reports property rights reform will not adversely affect the poor. Deininger, Klaus. 1999. "Making Negotiated Land Reform Work: * The net economic benefits from property rights Initial Evidence from Colombia, Brazil, and South Africa." reform are large, and failure to implement such reforms World Development 27(4): 651-72. carries a high opportunity cost. - 2001. "Negotiated Land Reform as One Way of Land The study goes beyond the current literature in sev- Access: Experience from Colombia, Brazil, and South Africa." eral ways. First, most studies on the subject have been In Alain de Janvry, Elisabeth Sadoulet, and Jean-Philippe Plat- limited to examining the impact of formal title rather than teau, eds., LandReform Revisited: Access to Land, Rural Pove, informal means of increasing tenure security. This, and PublicAction. New York: Oxford University Press. together with the fact that most of these studies find that 98 Agritulture and Rurol Development the main impact of title comes through credit markets, information and bargaining skills to compete successfully implies that little is known about the effects of tenure in a modern market economy. security in areas where credit markets are underdevel- This study will investigate whether introducing farm- oped and most poor people live. This study promises to ing innovations that increase yields and maintain or shed light on rural poverty reduction more generally. improve soil fertility while building on traditional Second, several studies have found the impact of land farming practices would improve the welfare of the tenure on short-term investments to be small, leading to lowland Amerindians in the tropical rain forest of Bolivia. a belief that even the long-term benefits from increased Using a randomized experimental design, the research tenure security would be limited. This study will deter- will estimate the effects of farming innovations and mine whether there is empirical support for such a view. training-and workshops on cultural empowerment, Third, the earlier finding that the nearly universal defense of land rights, and skills for dealing with the access to land in China performs an important safety modern market economy-on a number of outcomes: net function helps focus the study on the distributional wealth, income, consumption, crop yields and prices, impact of land tenure reform. Because attitude surveys health and nutrition, the share of farm output sold and in China have found broad support for periodic land the share consumed, reciprocity within and between redistribution, many believe that tenure reform would villages, defense of territorial rights, and feelings of be antipoor, an issue that the research can directly address. cultural self-worth. Specifically, the study addresses the Finally, the research explicitly considers the following questions: relationship between land tenure and the functioning of * Can farming innovations that build on traditional other factor markets. This is not only important because technologies yield greater dividends when accompanied other markets will affect households' options for by workshops on cultural empowerment, defense of land self-insurance. It is also important for the transferability rights, and marketing skills than when introduced alone of the findings to other Asian countries where legal or with agricultural extension only? restrictions on the functioning of land rental markets * Will the use of an experimental design with random continue to be widespread. assignment of villages to treatment and control groups Responsibility: Development Research Group, Rural Devel- make it possible to obtain unbiased estimates of the opment-Klaus Deininger (kdeininger@worldbank.org); effect of interventions; to estimate the return of invest- and East Asia and Pacific Region, China Country Unit- ments in cultural empowerment; to test whether invest- Jurgen Vogele, and Rural Development and Natural ments in agricultural technology, culture, and skills for Resources Sector Unit--Li Guo. With Yang Yao and dealing with the modern market economy reduce poverty; Yaohui Zhao, Beijing University; Michael Carter, and to learn about the ethical, logistical, and substantive University of Wisconsin at Madison; and Songqing Jin, problems that might arise in using a randomized design University of California at Davis. to evaluate World Bank investment projects? Completion date: December 2002. To address these questions, the study will use several analytical approaches, including qualitative and statisti- Might Culture Pay Off? Evaluating the Effects cal techniques, and draw on sources ranging from inter- of Farming Innovations and Cultural Empowerment views to a literature review. among Lowland Amerindians in Bolivia Responsibility: Social Development Department-Kreszen- tia Duer (kduer@worldbank.org) and Sabina Alkire. With In the past few decades loggers, ranchers, colonists, and Ricardo Godoy, Brandeis University; Tomas Huanca; commercial farmers have encroached on the territory of Victoria Reyes-Garcia and Vincent Vadez, University of lowland Amerindians in Latin America, leading to Florida; and Josh Angrist, Massachusetts Institute of insecurity of land tenure and the depletion of farmland. Technology. The Netherlands Trust Fund is contribut- To get by, lowland Amerindians have been forced to ing funding for the research. tighten their link with the market, but they lack the Completion dote: December 2002. Agriculture ond Rural Development 99 Crisis, Polarization, and Reform . Forthcoming. "Can Institutions Resolve Ethnic Con- flict?" Economic Development and Cultural Change. This research project looked at countries' policy decisions Easterly, William, and Stanley Fischcr. 2001. "Inflation and the as determined by two opposing forces-crisis and Poor." Journal of Money, Credit, and Banking 33(2, 1): 159- polarization-using regressions on data over time for all 78. countries for which data were available. Crisis tends to Easterly, William, and Ross Levine. 1997. "Africa's Growth Tragedy: spur reform, while polarization of society between oppos- Policies and Ethnic Divisions." Quarterly Journalof Economics ing interest groups tends to delay reform. 112(4): 1203-50. The project produced several papers exploring aspects of these issues. One paper examined the ability of The Quest for Growth policymakers to postpone true reform by engaging in illusory reforms, and found that they often do so. Another This extended research project has studied the contri- examined who minds inflation more, the poor or the bution of national policies, institutions, factor accumu- rich, and found that the poor do. Yet another tested the lation, and productivity to the growth of economies, hypothesis that crisis (as reflected by a number of using regressions on data over time for all countries for different indicators) leads to reform, and found that which data were available. The results show that policies while crises characterized by high inflation or a high and institutions affect growth mainly through productivity black market premium do lead to reform, other macro- rather than through investment or education. economic crises do not. And still other papers looked at Since the end of World War II economists have tried the role of ethnic tensions in determining policies, pub- to figure out how poor countries in the tropics could lic service delivery, and government employment-and attain standards of living approaching those of countries found that ethnic divisions inhibit the supply of public in Europe and North America. Attempted remedies services. have included providing foreign aid, investing in Responsibility: Development Research Group, Macroeco- machines, fostering education, controlling population nomics and Growth-William Easterly (weasterly growth, and making aid loans as well as forgiving those @worldbank.org). With Alberto Alesina, Harvard Uni- loans on condition of reforms. None of these solutions versity; Reza Baqir, University of California at Berkeley; has delivered as promised. The problem is not the fail- Allan Drazen, University of Maryland; and Stanley Fis- ure of economics, but the failure to apply economic prin- cher, International Monetary Fund. ciples to practical policy work. (ompletion date: June 2001. A book based on the research, The Elusive Quest for Growth, shows how these solutions all violate the basic Reports principle of economics, that people-private individu- Alesina, Alberto, Reza Baqir, and William Easterly. 1999. "Public als and businesses, government officials, even aid Goods and Ethnic Divisions." Quarterly Journal of Economics donors-respond to incentives. The book analyzes the 114(4): 1243-84. development solutions that have failed and suggests .2000. "Redistributive Government Employment." Jour- alternative approaches. nal of Urban Economics 48(2): 219-41. Responsibility: Development Research Group, Macroeco- Drazen, Allan, and William Easterly. 2001. "Do Crises Induce nomics and Growth-William Easterly (weasterly Reform? Simple Empirical Tests of Conventional Wisdom." @worldbank.org), David Dollar, and Norman Loayza; Economics and Politics 13(2): 129-58. Human Development Network, Office of the Vice Pres- 100 ident-Shantayanan Devarajan; and Office of the Senior adjustment through a broad-based consultative process Vice President, Development Economics-Roumeen including both public forums and field investigations. The Islam. With Peter Montiel; Ross Levine; Michael Bruno; project aimed to improve the understanding not only of and Howard Pack, University of Pennsylvania. the effects of adjustment policies but also of the ways Completion date: July 2001. broad participation of local civil society can improve policymaking. The initiative attempted to identify Reports practical changes in economic policies for governments Bruno, Michael, and William Easterly. 1998. "Inflation Crises and and the World Bank. Long-Run Growth." Journal of Monetary Economics 41(Febru- In each of the countries participating in the project- ary): 3-26. Bangladesh, Ecuador, Ghana, Hungary, Mali, and Devarajan, Shantayanan, William Easterly, and Howard Pack. Uganda-nongovernmental and civil society organizations Forthcoming. "The Productivity of Investment in Africa: Macro set up local networks and representative committees to and Micro Evidence." JournalofAfrican Economies. work with representatives from the Bank and from Dollar, David, and William Easterly. 1999. "The Search for the Key: different parts of the government in designing the work. Aid, Investment, and Policies in Africa." Journal of African The tripartite steering committee in each country was Economies 8(4): 546-77. responsible for planning an opening public forum, a field Easterly, William. 1999. "The Ghost of Financing Gap: Testing the investigation into the impact of selected policies, and a Growth Model of the International Financial Institutions." closing forum at which the results of the analysis were Journal of )evelopment Economics 60(2): 423-38. presented. 1999. "Life during Growth." Journal of Economic Growth The project was launched at a global forum in Wash- 4(3): 239-75. ington, D.C., in July 1997 and ended in July 2001 with . 2001. The Elusive Questfor Growth: Economists' Adventures a second global forum, where the discussion centered on andMisadventures in the Tropics. Cambridge, Mass.: MIT Press. the six country studies and their policy findings. The - .2001. "The Lost Decades: Explaining Developing Coun- policy findings are being summarized in a book. tries' Stagnation in Spite of Policy Reform, 1980-1998." Jour- Responsibiliy. Office of the Senior Vice President, Develop- nal of Economic Growth 6(2): 135-57. ment Economics-Jozef M. Ritzen (jritzen@worldbank. - . Forthcoming. 'The Niddle Class Consensus and Economic org) and John Randa; South Asia Region, Bangladesh Development." Journal of Economic Growth. Country Office-Fred Temple, Syed Nizammuddin, Easterly, William, and Ross Levine. 2001. "It's Not Factor Accu- and Zaidi Sattar; Latin America and the Caribbean mulation: Stylized Facts and Growth Models." World Bank Region, Ecuador Country Office-Marcelo Romero; Economic Review 15(2):177-219. Africa Region, Ghana Country Office-Peter Harrold and Easterly, William, Roumeen Islam, and Joseph Stiglitz. 2000. Kofi Marrah, Mali Country Office-Grace Yabrudy and "Shaken and Stirred: Explaining Growth Volatility." Paper Youssouf Thiam, and Uganda Country Office-James presented at the Annual Bank Conference on Development Adams, Robert Blake, and Mary Bitekerezo; and Europe Economics, World Bank, Washington, D.C., April 18-20. and Central Asia Region, Hungary Country Office- Easterly, William, Norman Loayza, and Peter Montiel. 1997. "Has Roger Grawe and Mihaly Kopanyi. The Belgian, Nether- Latin America's Post-Reform Growth Been Disappointing?" lands, Norwegian, and Swedish Trust Funds-contributed Journal of International Economics 43: 287-311. funding for the research. Completion date: July 2001. Structural Adjustment Participatory Review Initiative Report World Bank. 2001. "Adjustment from Within: Lessons from the This joint project involving six governments and an Structural Adjustment Participatory Review Initiative." Office international network of nongovernmental and civil of the Senior Vice President, Development Economics, society organizations examined the impact of structural Washington, D.C. Macrostonomits and Growth 101 Capital Subsidies and the Quality of Growth a small effect on output and and sometimes even a neg- ative effect on productivity. This research project examines how capital subsidies Partner institutes have been invited to conduct case and public expenditure affect the quality of growth, studies as part of the research. Once the research is com- especially through their impact on investment in human pleted, the results will be incorporated in World Bank capital and exploitation of the environment. Work has Institute courses, seminars, and conferences. focused on building a theoretical model, collecting evi- Responsibility: World Bank Institute, Office of the Vice dence on capital subsidies, and reviewing the literature President-Vinod Thomas (vthomas@worldbank.org), on their impact. The project is also conducting four Economic Policy and Poverty Reduction Division-Yan country studies-on Brazil, Chile, China, and India-and Wang, and Environment and Natural Resources Division collecting evidence on the size of capital subsidies and -Nalin M. Kishor. With Ashok M. Dhareshwar; Ramon their effect on the sustainability of growth. Finally, the E. Lopez, Sumeet Gulati, Gustavo Anriquez, and Andres study will provide evidence on how policies and public Ulloa, University of Maryland; Yudong Yao; Viju Ipe, expenditure allocation have affected the qualiry of growth Illinois State Geological Survey; Ashok Guha, Yale Uni- and what the possible tradeoffs are between capital sub- versity; Claudia Ocana; and Arati Belle. sidies and the quality of growth. The framework and Completion date: September 2001. initial evidence will serve as the basis of a proposal for a larger research project. Trade and Fiscal Reform and Investment Booms The evidence indicates that government subsidies to in Latin America: The Chilean Case and Application industry, agriculture, and infrastructure worldwide are to Other Reformers large. Although the data are fragmented and partial, and the estimates gross rather than net (not accounting for Many developing countries have liberalized foreign trade taxes), initial evidence nonetheless shows that these and investment policies over the past two decades. Some subsidies introduce elements of nontransparency, have achieved excellent-even spectacular-results, discrimination among activities, and pressure on scarce while others have seen little change in trade, invest- resources. ment, and growth. How do policies of openness interact The four country studies show that subsidies to withotherelementsoftheinvestmentclimatetoproduce foreign investors are often significant, with preferential good or bad economic outcomes? This research project tax treatment for foreign firms sometimes costing the addresses that question in order to provide policymak- government in forgone tax revenues. Competition for ers with guidance on what policies are needed to foreign investment is sometimes the reason for these complement openness as countries attempt to integrate subsidies. In Chile, for example, foreign firms account with the global market. for about 60 percent of copper production. Extensive The project has two parts. The first is an in-depth case private investment led to a rapid expansion of produc- study of Chile with new econometric work based on the tion and exports, which worsened the terms of trade. In Chilean census of manufacturing. The study focuses on addition, while private firms paid less than $100 million the reduction in the tax on retained profits in the mid- a year in taxes in 1995-2000, the publicly owned Codelco 1980s. Taxing retained earnings is particularly harmful paid about $600 million. Thus as a result of subsidies for in an economy with poorly developed financial markets foreign investment in the copper industry, the country but favorable macroeconomic policies as well as policies has been losing more than $100 million a year. of openness, such as the Chilean economy in the mid- Recent studies examining the effect of corporate sub- 1980s. When some firms are credit constrained, taxing sidies suggest that as an instrument for increasing pro- retained profits is more distortionary than taxing divi- ductivity and promoting growth, many are counter- dends, and it reduces investment and slows growth. The productive. Detailed firm-level panel data over several study hypothesized that by reducing the tax rate on years for Israel and Sweden show that subsidies have only retained earnings, the government increased the inter- 102 Macroeconomics and Growth nal funds of credit-constrained firms, leading to an invest- If sound institutions are needed to get strong bene- ment boom and subsequent growth. fits from openness, should countries wait until they have To test this theory and evaluate its importance, such institutions to open up? Not necessarily. One the study analyzed annual plant-level data covering reason that liberalizing services trade is so important is all Chilean manufacturing firms with more than 10 that developing countries can use this market to improve employees. It divided the firms into two groups-those the investment climate: allowing foreign firms to provide more likely to face financing constraints and those less power, telecommunications, and financial services can be likely to-and compared their investment behavior. It a good strategy for strengthening the investment found that plants with high correlations between cash flow climate. and investment before the reform increased investment The survey also looks at targeted efforts to attract significantly more during and, to some extent, after the foreign investment. Countries often try to attract invest- reform than did similar firrns with low correlations. Plants ment through subsidies and tax holidays, but the evidence that previously had low short-term reserves also increased suggests that these instruments are not particularly investment rnore during and, to some extent, after the successful; they also discriminate against domestic reforms. The study found no evidence that smaller plants firms. The really successful cases-Taiwan (China), for or plants paying rent benefited disproportionately from example-have created an environment of good gover- the reform (though the size distinction is less likely to nance and good infrastructure in which both foreign and reflect the financial constraints facing a plant in Chile than domestic firms can be competitive. in the United States). Responsibility: Development Research Group, Macroeco- Earlier studies have often pointed to other market- nomics and Growth-David Dollar (ddollar@Cworldbank. oriented reforms in Chile-liberalization of trade, org), Mary Hallward-Driemeier, and Tilahun Temes- liberalization and deepening of financial markets, gen, and Office of the Director-Paul Collier. With privatization of the public pension system-as the under- Chang-Tai Hsieh, Princeton University. lying cause of Chile's boom. But the study's findings do Completion date: September 2001. not suggest that these reforms are irrelevant for growth. Instead, they emphasize that the policies of openness Reports combined with the improvement in tax policy led to an Hallward-Driemeier, Mary. "Openness, Firms, and Competition." investment boom and rapid growth in Chile. World Bank, Development Research Group, Washington, D.C. The second part of the project, a survey of firm-level Hsieh, Chang-Tai, and Jonathan Parker. 2001. "Taxes and Growth studies in open economies, looks more generally at the in a Financially Underdeveloped Country: Explaining the policies that usefully complement openness to foreign Chilean Investment Boom." trade and investment. Individual ca§es and firm-level studies reveal that developing country firms can be Economic Growth, Social Capability, competitive. But they are often hampered by a poor and Preindustrial Development investment climate-corruption, inefficient regulation, infrastructure weaknesses, and poor financial services. A Recent studies have conjectured that the preindustrial recent study of India based on firm surveys highlights the development experiences of countries in Asia, Africa, and critical importance of the investment climate. With the Latin America left these countries differently positioned same (national) trade and macroeconomic policies, Indian for achieving modern economic growth following World states are experiencing widely differing results from lib- War II. Preindustrial development can be conceptualized eralization. Not surprisingly, "good climate" states- as running along a spectrum from primitive, foraging, those with more efficient regulation and better band-type societies, to extensive agricultural and pastoral infrastructure-are experiencing more foreign and domes- small-village societies, to intensive state-level agrarian tic investment, more income growth, and more poverty societies resembling those of Europe and Japan on the reduction, while "poor climate" states lag behind. eve of industrialization. In cross-country growth regres- Macroetonomics ond Growth 103 sions preindustrial development-proxied by popula- Responsibility: Africa Region Technical Families, Macroeco- tion density, farmers per hectare, the irrigated share of nomics 2-Miria Pigato (mpigato@worldbank.org). With farmland, or the first principal component of the three- Louis Putterman and Areendam Chanda, Brown University. has been shown to be a good predictor of per capita Completion date: October 2001. income growth in developing countries in 1960-75, 1975-90, and 1960-90. Report In further explorations using the Barro-Lee education Chanda, Areendam, and Louis Putterman. 2000. "Economic data and the Adelman-Morris social development index, Growth, Social Capability, and Preindustrial Development." the preindustrial development measures usually show a World Bank, Africa RegionTechnical Families, Macroeconomics robust, statistically significant relationship with the eco- 2, Washington, D.C. nomic growth rate, in the direction predicted: when con- ventional explanatory factors are controlled for, countries The Quality of Fiscal Adjustment more densely populated and intensively cultivated at the beginning of the period achieved faster per capita This research project analyzed the quality of fiscal adjust- income growth during postwar periods for which data are ment undertaken by policymakers in response to macro- available. economic pressures. The analysis combined cross-country This research project further investigates the rela- data for Latin America, data for all developing coun- tionship between premodern development and recent tries, and selected case studies. The results show that pol- growth to see whether the original hypotheses can be ver- icymakers sometimes postponed true reform by engaging ified using larger samples, longer time series, earlier data in illusory reforms. And they sometimes sacrificed future for premodern development, and more sophisticated growth for current fiscal adjustment, which worsened their methods. It also explores the connections between pre- long-run budget constraint. modern development, growth, and the demographic Responsibility: Development Research Group, Macroeco- changes of recent decades. And it seeks to improve the nomics and Growth-William Easterly (weasterly understanding of the relationship between premodern @worldbank.org); and Latin America and the Caribbean development, as proxied by population density and sim- Region, Office of the Chief Economist-Luis Serven. ilar variables, and "social capability." With Cesar Calder6n, Rochester University. The research has found that initial population density, Completion date: December 2001. farmers per acre, and irrigation are good predictors of eco- nomic growth rates when other factors are controlled for Reports in cross-country growth regressions of standard form. Calder6n, C6sar, and William Easterly. 2001. "How Did Latin Two other indicators, social development and ethnolin- America's Infrastructure Fare in the Era of Macroeconomic guistic heterogeneity, bear a less robust relationship to Crises?" World Bank, Development Research Group, Wash- growth rates. But experience with state-level structures, ington, D.C. another measure of social development suggested by Easterly, William. 1999. "When Is Fiscal Adjustment an Illusion?" the social evolutionary literature, is a strong predictor of Economic Policy (April): 57-86. growth in such regressions. . 2001. "Growth Implosions and Debt Explosions: Do The findings suggest that history matters, that the human Growth Slowdowns Explain Public Debt Crises?" Contribu- factor is crucial to growth, that a society's human capacities tions to Macroeconomics 1(1) (Berkeley Electronic Press). are formed by a wider range of influences than those asso- ciated with formal schooling, and that fostering the devel- Geography, Growth, and Comparative opment of human technological and organizational Advantage in Sub-Saharan Africa capabilities is indeed as important as many have contended. The research has contributed to the debate on growth Sub-Saharan Africa has experienced a sustained eco- prospects in Africa. nomic slowdown for the past 20 years or more and, 104 Mactroeconomics and Growth despite notable success stories, the trend is likely to phy and climate in Africa, and regional cooperation in continue for the region as a whole well into the 21st developing infrastructure and communications to reduce century. Reversing this economic decline is the most economic isolation and expand markets. serious challenge the development community faces. A preliminary analysis of the climate ecology, soil Efforts to identify the root causes of the deep, persistent characteristics, and topography of Sub-Saharan Africa decline have pointed to disadvantages in geography and, suggests that, given current agricultural technology, the more recently, in health and population. If these factors region suffers from adverse endowments. Thus these pre- are found to be critical for Africa's growth and interna- liminary findings appear to support the project's key tional competitiveness, what would the implications be assumption that health and agriculture in Africa require for development policy in the region? a new scientific agenda. This project aims to contribute to the understanding This research effort will contribute to a larger, col- of these issues by analyzing Africa's growth performance laborative research project, Explaining Economic Growth and its potential comparative advantage in labor- Performance in Sub-Saharan Africa, that involves many intensive manufactured exports. The research draws on researchers and policymakers from the region. An inter- global panel data sets on geographic, demographic, and agency initiative, this collaborative project is led by the institutional and policy indicators for the period 1965-97. African Economic Research Consortium (AERC) and The growth analysis is based on a dynamic endogenous includes the United Nations Economic Commission for growth model, and the comparative advantage analysis Africa, the World Bank's Development Research Group, on an extended Heckscher-Ohlin framework. Both analy- Oxford University, and Harvard University. ses emphasize rigorous econometric modeling to ensure Results will be disseminated at an AERC meeting in that the estimated models permit policy simulations. December 2001 or May 2002, and through presenta- If the results should show that Sub-Saharan Africa's tions at the World Bank in February 2002. geography is not destiny, yet does influence the region's Responsibility: Development Research Group, Macroeco- growth and international competitiveness, they would nomics and Growth-Ibrahim A. Elbadawi (ielbadawi suggest a balanced approach to development that would @worldbank.org), and Office of the Director-Paul also require strategic measures in technology and infra- Collier. With Anke Hoeffler, Oxford University; structure. In particular, the approach would need to and Siham Mohamedahmed, University of Arizona, emphasize the development of technology (especially in Tucson. agriculture) to deal with the unique features of geogra- Completion date: June 2002. Macroeconomics and Growth 105 International Economics Aid and Reform in Africa workshops and seminars in Frankfurt, Germany (June 1999); Washington, D.C. (September 1999); Dar es Recent cross-country studies have found that foreign Salaam, Tanzania (November 1999); Nairobi, Kenya aid has a strong, positive effect on a country's economic (September 2000); and Clermont-Ferrand, France performance if the country has undertaken certain (September 2000). Participants included host country policy and structural reforms. But the evidence also officials, policymakers from case study countries, and shows that less aid goes to countries that have undertaken representatives of bilateral donor agencies, international these reforms than to those that have not. Moreover, organizations, and research institutions. there is evidence that "aid cannot buy reform." General project information and the report produced This research project went beyond the cross- by the research project are available on the Web at country regressions to arrive at a better understanding of http://www.worldbank.org/research/aid/. the causes of reforms and of the link between foreign aid Responsibility: Development Research Group, Macroeco- and reform. It focused on the real causes of reform and nomics and Growth-David Dollar (ddollar@worldbank. whether and how aid has encouraged, generated, influ- org), Torgny Holmgren, and Waly Wane, Public Eco- enced, supported, or retarded reforms. Accordingly, it nomics-Shantayanan Devarajan, and Office of the analyzed the processes of reform rather than the results. Director-Paul Collier. With Elliot Berg, Patrick Case studies of 10 African countries examined the Guillaumont, and Sylviane Guillaumont, Auvergne nature of external assistance, investigated the causes University; Jacques Pegatienan, University of Abidjan; and paths of policy reforms, and traced the relationship, Jerome Chevallier; Gilbert Kiakwama; Berhanu Abegaz, if any, between aid and reform. The studies were based College of William and Mary; Samuel Wangwe, Yvonne largely on interviews and on reviews of available data, lit- Tsikata, ard Deogratias Mutalemwa, Economic and erature, and documentation. The data came from sources Social Research Foundation, Dar es Salaam, Tanzania; at the International Monetary Fund, Organisation for Stephen O'Brien; Cheikh Sidibe, Ministry of Finance, Economic Co-operation and Development, and World Mali; Jeffrey Herbst, Princeton University; Charles Bank. Soludo, University of Nigeria; Arne Bigsten, University A major finding is that reforms are generated largely of Goteborg, Sweden; Louis Kasekende, Michael by causes not directly related to aid, such as crises, polit- Atingi-Ego, and Daniel Ddamulira, Bank of Uganda; ical leadership, committed local technocrats, country Terry Ryan, University of Nairobi; Lise Rakner, role models, and consensus among social groups. But in Christian Michelsen Institute, Bergen, Norway; Nicholas certain circumstances and phases, foreign assistance can van de Walle, Michigan State University; Dominic help trigger and sustain reforms through policy dialogue, Mulaisho; Dennis Tao; and Winifred Thompson. The advisory services, and financial aid. Conditions attached governments of France, Germany, the Netherlands, to assistance can help reform-minded technocrats lock Norway, Sweden, and Switzerland contributed funding in reforms, but probably cannot generate sustainable for the research. reforms. These findings could influence how aid is used, (ompletion date: May 2001. and what types and amounts are used, during different phases of a reform program and how donors allocate Report their resources among countries and sectors. Devarajan, Shantayanan, David Dollar, and Torgny Holmgren, Preliminary case study findings and potential eds. 2001. Aid and Reform in Africa: Lessons from Ten Case conclusions on cross-cutting issues were discussed at Studies. Washington, D.C.: World Bank. 106 The Antiexport Bias of Duty Drawbacks Jean-Marie Grether, University of Neuchatel; and Jaime de Melo, University of Geneva. Among the most commonly used instruments for (ompletion date: June 2001. correcting the antiexport bias of protectionist trade regimes are duty drawbacks, which provide duty- Reports free entry of imported inpuits used in producing exports. Cadot, Olivier, Jaime de Melo, and Marcelo Olarreaga. 2000. "Can But by reducing exporters' incentives to participate Duty Drawbacks Have a Protectionist Bias? Evidence from in lobbying against protection, the introduction of duty Mercosur." Policy Research Working Paper 2523. World Bank, drawbacks may lead to greater trade protection- Development Research Group, Washington, D.C. and thus increase the antiexport bias of the trade .2001. "Lobbying and the Structure of Tariff Protection in regime. Poor and Rich Countries." World Bank, Development Research This research project explored the conditions under Group, Washington, D.C. which duty drawbacks correct the antiexport bias of Cadot, Olivier, Jean-Marie Grether, and Marcelo Olarreaga. 2001. trade regimes when trade policy is subject to lobbying "Lobbying and Protection: Who Does It and Who Gets It? The by interest groups. Its empirical investigations focused Indian Case." World Bank, Development Research Group, on India and on Argentina and Brazil, for an analysis of Washington, D.C. the impact of eliminating duty drawbacks on intraregional trade in the context of bilateral trade agreements Comprehensive Database on Trade and Production (Mercosur). The analysis was based on recent models developed in the political economy literature on trade This project constructed a database on trade, production, policy, extended to include counterlobbying by users of and tariffs in 67 industrial and developing countries over intermediate goods. the period 1976-99. The data were filtered into a com- The analysis showed that within a Grossman- mon classification and disaggregated by the 27 industries Helpman model of endogenous protection, equilibrium of the three-digit International Standard Industrial Clas- tariffs escalate with the degree of processing. This result sification (ISIC) and, where available, the 81 industries was empirically verified both for India and for Argentina of the four-digit ISIC. and Brazil. The production data are from the United Nations The research also showed that the presence of duty Industrial Development Organization and include value drawbacks for exporters reduces the incentives to added, total output, average wages, capital formation, counterlobby against higher tariffs on imported inputs, number of employees, number of female employees, a finding confirmed by evidence for India and the and number of firms. The trade data are from the United Mercosur members. In Mercosur eliminating duty draw- Nations' Commodity Trade (Comtrade) database and backs for intraregional exports led to a reduction in the include imports, exports, and mirrored exports. Trade average common external tariff of 3.5 percentage points. flows were aggregated by region and income group, and The findings are being presented at seminars and a separate database was constructed on bilateral trade conferences, including a Latin American and Caribbean flows. The tariff data are based on most-favored-nation Economic Association conference in Montevideo, tariffs from the Trade Analysis and Information System Uruguay; Econometric Society meetings in Buenos Aires, (Trains) maintained by the United Nations Conference Argentina; and seminars at the World Bank, the World on Trade and Development and the World Trade Trade Organization, and several universities. A survey of Organization's Trade Policy Reviews. Input-output the literature on duty drawbacks and the research pro- tables were obtained from the Global Trade Analysis ject's findings is under preparation. Project. Responsibility: Development Research Group, Trade- The database could be used to analyze a wide range Marcelo Olarreaga (molarreaga@worldbank.org). With of issues. The Development Research Group's Trade Olivier Cadot, INSEAD and University of Lausanne; Team is using it to study the impact of trade liberaliza- International Economics 107 tion on the mobility of female workers across sectors and adjustment costs associated with forgone consumption. on the acquisition of foreign knowledge by producers. It assessed the claim that the dynamic welfare gains The database will be made available on demand, in from trade liberalization are considerably larger than the ASCII files and Microsoft Excel worksheets on comparative static estimates. Using computable general CD-ROM. It will also be available in a handbook on the equilibrium models, it also investigated the importance new round of World Trade Organization negotiations. of having a variety of imported inputs available in domes- A descriptive paper accompanies the database. tic production and of opening service sectors to foreign Responsibility: Development Research Group, Trade- competition. The analysis drew on the Global Trade Marcelo Olarreaga (molarreaga@worldbank.org) and Analysis Project database, the Trade Team's database on Alessandro Nicita. tariffs, and input-output tables as appropriate. Completion dote: June 2001. The project produced three papers. The first extends a comparative static analysis of Chile's trade policy The Dynamic Impact of Trade Liberalization options to a Ramsey-type dynamic model of Chile with in Developing Countries constant returns to scale and perfect competition. It shows that simply adding a dynamic element to the International trade economists and World Bank policy analysis does not increase the welfare gains from trade advice have typically argued that an open trade regime is liberalization much. very important for economic growth and development. The second paper develops a stylized, somewhat This view has been based in part on neoclassical trade the- aggregate computable general equilibrium model of a ory, which generally finds that trade liberalization improves small open economy with endogenous growth entering a country's welfare; in part on casual empirical observation through a productivity multiplier of the Ethier-Dixit- that countries that remain highly protected for long peri- Stiglitz variety. Trade liberalization in this model dra- ods appear to suffer significantly and perhaps cumula- matically increases welfare because it results in a tively; and in part on empirical work that also finds trade significant increase in the number of varieties (tech- liberalization beneficial to growth and welfare. nologies) available in the economy. The paper shows that Yet numerical estimates of the impact of trade liber- when the impact of trade liberalization on technology dif- alization have generally shown that it increases the wel- fusion is taken into account, the estimated welfare gains fare of a country by only about 1 percent of GDP. These are very large and consistent with the large econometri- estimates have been based on comparative static mod- cally estimated effects on growth. els, however, and researchers typically claim that the The third paper examines the impact of liberalizing estimates would be much larger if they incorporated the foreign direct investment in service sectors. It shows dynamic gains from trade liberalization. that not only does liberalizing foreign investment have The development of endogenous growth theory has a large welfare impact, but opening business services to provided a clear theoretical link from trade liberalization to foreign competitors is likely to significantly benefit economic growth. Because of the complexity of the mod- domestic skilled labor. Even the pattern of what the els, the theoretical literature has necessarily been based on country exports and imports can change. rather aggregate models, and it has focused on the steady- The study's results support the strong version of the state growth path, making it difficult to gain insight into the link between trade liberalization and economic growth, dynamic growth path of the key variables. Moreover, since as well as the importance of opening service sectors to the theoretical literature does not evaluate the adjustment foreign direct investment. They should buttress the costs, it cannot clearly indicate that welfare is significantly intellectual case in the development community for the increased even if the long-run growth rate increases. importance of openness for growth. This study quantified the dynamic gains from trade Liberalizing services is particularly important in coun- liberalization in a small developing country in an applied tries acceding to the World Trade Organization, and the general equilibrium model, taking into account the project's innovation in allowing a practical assessment of 108 International Economics the impact of service liberalization should prove Reports extremely useful in the policy dialogue with such coun- Mody, Ashoka, and Mark Taylor. "Modeling Capital Crunches." tries. The stucly's results have been incorporated in the Draft. World Bank Institute's course on trade policy. Mody, Ashoka, Mark Taylor, and J. Y. Kim. 2001. "Modeling Fun- Resportsibility: Development Research Group, Trade-David damentals for Forecasting Capital Flows to Emerging Mar- Tarr (dtarr@worldbank.org). With Thomas Rutherford and kets." InternationalJournal of FinanceandEconomics 6(3): 201-16. James Markusen, University of Colorado. - . Forthcoming. "Forecasting Capital Flows to Emerging (ompletion date: June 2001. Markets: A Kalman Filtering Approach." Applied Financial Economics. Reports Markusen, James, Thomas Rutherford, and David Tarr. 2000. International Capital Flows "Foreign Direct Investmen-t in Services and the Domestic Market for Expertise." NBE1R Working Paper 7700. National This research project compiled new measures of finan- Bureau of Economic Research, Cambridge, Mass. cial liberalization and capital controls, gathering data on Rutherford, Thomas, and David Tarr. 1998. "Regional Trading 28 industrial countries and emerging markets (in East Asia Arrangements for Chile: Do the Results Differ with a Dynamic and Latin America) since 1973. The project collected Model?" Paper presented at the conference Using Dynamic information on regulations in the domestic financial sec- Computable General Equilibrium Models for Policy Analysis, tor, including controls on interest rates (lending and Assens, Denmark, June 14-17. deposit), on quantities (credit controls and reserve . Forthcoming. "Trade Liberalization and Endogenous requirements), and on other aspects (foreign currency Growth in a Small Open Economy: An Illustrative Model." deposits). Journal of International Economics. (For another version of this The project collected data on international capital paper, "Trade Liberalization, Product Variety, and Growth in flows for a large cross-section of countries since the a Small Open Economy: A Quantitative Analysis," contact 1960s. The flows were disaggregated into assets and lia- mkasilag@worldbank.org.) bilities (debt instruments, foreign direct investment, and portfolio equity), and a data series was constructed Forecasting Capital Flows for the share of foreign direct investment represented by mergers and acquisitions. The research then identified This study examined three issues relating to developing the basic patterns of international flows-their compo- country borrowing. First, the study looked at what deter- sition, their countries of origin and destination, their mines the maturity of developing country bonds. The trends over time, and their relationship with domestic results of this analysis shed light on what factors might macroeconomic and governance conditions. These styl- be manipulated to move bonds away from short maturi- ized facts will be used to construct a model of capital ties, which have been shown to be a risk factor in recent flows, focusing mostly on foreign direct investment. financial crises. Second, the study examined the effects The project also carried out research on the integra- of a recent sovereign debt restructuring on the country's tion of developing countries with the international finan- access to debt markets and its borrowing costs, an analy- cial system. This research analyzed the benefits and sis that contributes to the debate on bailout and moral risks of financial globalization, the flows of mutual fund hazard. Finally, the study looked at the relationship investments to developing countries, the impact of finan- between exclhange rate regime (hard peg, soft peg, or rel- cial globalization on firms' financing choices, and the atively free float) and the cost of external finance. transmission of international interest rates and credit Responsibility: Development Prospects Group-Ashoka rating upgrades and downgrades to stock market prices Mody (amody@worldbank.org). With Mark Taylor and and country premiums. J. Y. Kim, Warwick University. The research found that developing countries have Completion date: June 2001. gradually lifted restrictions on capital account transactions Internalional Etonomics 109 to take greater advantage of financial globalization- ington University; Richard Lyons and Changqing Sun, though there have also been periods in which restrictions University of California at Berkeley; Andrea Bubula, were reimposed. The most substantial reversals occurred Columbia University; Eduardo Fajnzylber, University of in Latin America in the aftermath of the 1982 debt cri- California at Los Angeles; Federico Guerrero, Francisco sis and in the mid-1990s, and in Asia in the aftermath of Vazquez, and Jose Pineda, University of Maryland; Jon that region's financial crisis. Tong; Nong Thaicharoen, Massachusetts Institute of As a result of the easing of restrictions and of tech- Technology; Sergio Kurlat; Leonor Coutinho-Gouveia; nological advances, net capital flows to developing Chris van Klaveren; Yanchun Zhang; Yaye Sakho; Tatiana economies have increased sharply since the 1970s, Brandao;andAkidoTerada.TheNationalScience Foun- one of the benefits of financial globalization. Capital dation, United States, contributed funding for the flows rose from less than $28 billion in the 1970s to research. about $306 billion (in real terms) in 1997, when they Completion date: June 2001. peaked. The composition of capital flows to developing countries changed significantly during this period. Reports The share of official flows declined by more than half, Frankel, Jeffrey, Luis Servdn, and Sergio Schmukler. "Global while private flows became the major source of capital Transmission of Interest Rates: Monetary Independence and for many emerging economies. The composition of pri- Currency Regime." vate capital flows also changed markedly, with foreign Kaminsky, Graciela, and Sergio Schmukler. 2001. "Short- and direct investment growing continuously throughout the Long-Run Integration: Do Capital Controls Matter?" Brookings 1990s. Trade Forum 2000, pp. 125-78. While financial liberalization brings benefits, it also Schmukler, Sergio, and Pablo Zoido-Lobat6n. "Financial Globaliza- involves clear risks, reflected in the devastating crises in tion: Opportunities and Challenges for Developing Countries." recent years. The research shows that the process of opening leads to a more extreme cycle in financial Regionalism and Technology Transfers markets. In the typical stock market cycle of an open developing country, stock prices increase nearly 100 This research project examines the determinants of tech- percent in the 18 months before the cycle peaks, nology transfers from industrial to developing countries, then fall 20 percent over the first six months of the and the impact of these transfers on total factor produc- downturn. But in the first cycle within three years of tivity. In particular, it investigates whether economic financial liberalization, on average stock prices triple, openness and education in the host country contribute then drop by 50 percent over the first six months of the to the absorption of foreign technological knowledge downturn. Thus a key question for developing countries and thus to growth. The research is based on econo- is whether they have the robust financial institutions metric analysis using an approach for determining the rela- needed to manage this temporary volatility. If not, a seri- tionship between total factor productivity and foreign ous crisis can ensue. research and development. The research also looked at the effect of capital con- The study uses an industry-related database for devel- trols. It found that these controls can work in the short oping countries, which has not been done before in the run, directing flows toward longer maturities, but that literature. The data are drawn from several sources, their effectiveness declines over time. including the World Integrated Trade Solution infor- Responsibility: Development Research Group, Macroeco- mation system, the Barro-Lee cross-country database nomics and Growth-David Dollar (ddollar@worldbank. on educational attainment and the quality of education, org), Sergio Schmukler, Norman Loayza, and Hairong Yu; the Analytical Business Enterprise Research and Devel- and Latin America and the Caribbean Region, Office of opment Database maintained by the Organisation for the Chief Economist-Luis Serven. With Jeffrey Frankel, Economic Co-operation and Development (OECD), Harvard University; Graciela Kaminsky, George Wash- and databases of the World Bank, the United Nations 110 Internalional Economics Industrial Development Organization, and the series data, which are difficult to interpret; or case stud- United Nations Educational, Scientific, and Cultural ies, which are difficult to generalize. The premise of Organization. this research project is that analysis of firm-level panel Preliminary findings show that openness contributes data is most likely to shed light on the issues. to the absorption of foreign technological knowledge The project is documenting cross-country and sectoral and to total factor productivity and that education patterns in the incidence of the activities associated with contributes additionally to this process. These findings technology transfer and, where possible, relating the should strengthen the arguments on the importance of patterns to country policies. The research uses firm- economic openness and good-quality education. The level panel data sets for 10 developing and transition findings also have implications for regional agreements- economies: Bulgaria, Colombia, the Czech Republic, whether countries should enter or form one and, if so, with Ghana, Indonesia, the Republic of Korea, Morocco, which partners. Poland, Romania, and Taiwan (China). Econometric The findings have been discussed with economists and techniques are used to investigate the causal relationship policymakers in Mercosur, at the European Commis- between firm activities and firm performance and the size sion in Brussels, and at the OECD Development Cen- of the effects. tre in 13aris. They have also been presented to World Bank Results to date suggest that: staff and to academic audiences at Yale University, Uni- * In high- and medium-technology sectors foreign versity of Paris 1, University of Clermont-Ferrand, and investors that are technology or marketing leaders in Universidad Catolica, Chile. their industries are more likely to engage in wholly Responsibility: Development Research Group, Trade- owned projects than to share ownership (Poland). Maurice Schiff (mschiff@worldbank.org). With Joseph * Total factor productivity growth tends to be higher Fran,ois; Yanling Wang, Georgetown University; in wholly owned firms than in firms with joint ventures Wolfgang Keller, University of Texas at Austin; Giorgio and firms without foreign partnerships (the Czech Repub- Barba Navaretti, Universita degli Studi di Ancona and lic, Indonesia). Centro Studi Luca d'Agliano (Milan); and Nisha * As foreign owners relinquish their equity stake, Malhotra, University of Maryland at College Park. productivity falls (Indonesia). Completion date: August 2001. * Productivity growth in manufacturing is positively correlated with the complexity of machines imported in Micro Foundations of International a given industry (Morocco, Poland). Technology Diffusion * Firms that neither import intermediate goods nor export tend to produce lower-quality products than Developing country governments often seek to spur counterpart firms engaged in international activities adoption of foreign technologies through policy inter- (Colombia). ventions designed to encourage domestic firms to enter * Differences in technology inflows related to the into joint ventures with foreign partners, import sophis- pattern of imports explain about 20 percent of the vari- ticated capital goods, license technology, and export to ation in productivity growth between countries. industrial country markets. Which of these policy inter- The findings of the project have been presented at ventions (if any) is appropriate depends crucially on the World Bank Trade Seminar Series, the Annual Bank which leads to effects that improve firm performance, how Conference on Development Economics, and other large the effects are, and whether the effects are inter- conferences. nal or external to the firnm. Responsibility: Development Research Group, Trade- Much of the research on these questions for devel- Bernard Hoekman (bhoekman@worldbank.org), Isidro oping countries relies on cross-sectional data, which Soloaga, and Beata Smarzynska, Macroeconomics and make it difficult to identify the direction of causation Growth-Aart Kraay, and Finance-Simon Evenett. between activities and performance; aggregate time- With Gary Anderson, University of Maryland; Bee-Yan International Economics 111 Aw, Mark Roberts, and James Tybout, Pennsylvania Saggi, Kamal. 1999. "Trade, Foreign Direct Investment, and Inter- State University; Marc Bacchetta and Felix Eschenbach, national Technology Transfer: A Survey." Southern Methodist World Trade Organization; Howard Pack, University of University, Dallas. Pennsylvania; Kamal Saggi, Southern Methodist Uni- Smarzynska, Beata. 2000. "Composition of Foreign Direct Invest- versity, Dallas; Francis Teal, Oxford University; Alex ment and Protection of Intellectual Property Rights in Transi- Voicu, Rutgers University; and Giorgio Barba Navaretti, tion Economies." World Bank, Development Research Group, Universita degli Studi di Ancona. Washington, D.C. (ompletion date: December 2001. - . 2000. "Technological Leadership and the Choice of En"ty Mode by Foreign Investors." World Bank, Development Reports Research Group, Washington, D.C. Anderson, Gary. 2000. "Multinational Corporations and Tacit Teal, Francis. 2000. "Micro Foundations of Technological Diffu- Knowledge: Determination of Entry Mode and Impact of sion: An African Data Set." Oxford University, Centre for the Entry." University of Maryland. Study of African Economies. Deardorff, Alan, and Simeon Djankov. Forthcoming. "Knowledge Transfer under Subcontracting: Evidence from Czech Firms." Trade Liberalization, Industrial Performance, WorldDevelopment. and Export Growth in India Djankov, Simeon, and Bernard Hoekman. 1998. "Conditions of Competition and Multilateral Surveillance." World Bank, Over the past decade India has deregulated industries, Development Research Group, Washington, D.C.; and Centre liberalized trade, opened its economy to foreign invest- for Economic Policy Research, London. ment, and reformed its financial sector. How have these - 2000. "Market Discipline and Corporate Efficiency: Evi- reforms affected the economy? To find out, this study dence from Bulgaria." Canadian Journal of Economics 33(1): examines the links between trade liberalization, indus- 190-202. trial performance, and export growth in India during the - Forthcoming. "Foreign Investment and Productivity 1990s. Growth in Czech Enterprises." World Bank Economic Review. The study is carrying out an analysis of manufacturing Evenett, Simon, and Alex Voicu. 2001. "Picking Winners in Tran- at the aggregate level and at the two-digit level of disag- sition: Revisiting the Benefits of FDI in the Czech Republic." gregation. To supplement this, it is also conducting an World Bank, Development Research Group, Washington, D.C. analysis of the auto components and pharmaceutical indus- Kraay, Aart, Isidro Soloaga, and James Tybout. 2001. "Product tries, based on firm-level data, to explore the microeco- Quality, Productive Efficiency, and International Technology nomic responses to the new policy regime. The analyses Diffusion: Evidence from Plant-Level Panel Data." World are based on ordinary least squares and panel regressions Bank, Development Research Group, Washington, D.C. and draw on the World Bank's World Development Indi- Navaretti, Giorgio Barba, and Isidro Soloaga. 2001. "Weightless cators database, data from national sources, and some pri- Maehines and Costless Knowledge: An Empirical Analysis of mary data collected through direct questionnaires. The Trade and Technology Diffusion." Universita degli Studi di work is being carried out by the Indian Council for Ancona, Italy. Research on International Economic Relations (ICRIER). Navaretti, Giorgio Barba, and David G. Tarr. 2000. "International The study's findings may serve as a useful input for Knowledge Flows and Economic Performance: A Review of the the Indian government in determining its stance in the Evidence." WorldBankEconomic Review 14(1): 1-15. World Trade Organization on opening the pharmaceuti- Navaretti,GiorgioBarba,IsidroSoloaga,andWendyTakacs.2000. cal and automobile sectors. The findings will be dis- "Vintage Technologies and Skill Constraints: Evidence from cussed at interim seminars and at a larger seminar once U.S. Exports of New and Used Machines." WorldBank Economic the study is completed. The data collected will be made Review 14(1): 91-109. publicly available. Pack, Howard. 1999. "Modes of Technology Transfer at the Firm Responsibility: South Asia Region, Poverty Reduction and Level." University of Pennsylvania, Philadelphia. Economic Management Sector Unit-Deepak Mishra 112 Internotionol Economits (dmishra@worldbank.org) and Shahnaz Rana. With Isher To conduct these analyses, the project has been Judge Ahluwalia and Basudev Guha, Indian Council for collecting data for a large group of developing countries. Research on International Economic Relations. It has also been collecting data on stock prices in each Complelion date: December 2001. domestic market and on world exchanges, on trading volumes of stocks, and on stock returns in major world The Effects of American Depository Receipt markets. These data will make it possible to compute Trading on Local Markets time-series measures of liquidity, volatility, and integra- tion across the developing country markets. Large companies in developing countries are migrating In addition, the project has assembled the dates of to international equity markets, particularly in London major cross-listings of equities and has been construct- and New York, through the trading of depository receipts ing trading volume series in the local market and in and the cross-listing of their stocks. As the number of com- international markets for companies with and without panies listed domestically shrinks, several local markets cross-listings and American depository receipt programs. are becoming illiquid, making it difficult for the com- The database shows when each company issues equity panies remaining at home to raise equity. Foreign direct in different stock markets around the world. So, for each investment might be exacerbating the liquidity problem: country, data can be obtained for several companies, for when local companies are acquired, they withdraw their several cross-listing events, and for a continuous series listings from local markets (delist). And foreign compa- for each company. nies tend to move their equity abroad. The data come from several sources, including This process could have important policy implica- Bloomberg, DataStream, Euromoney, the Bank of New tions if equity financing becomes more difficult for a large York, and the International Finance Corporation. number of companies. In response, policymakers have Responsibility: Development Research Group, Macroeco- been proposing new local or regional equity markets. To nomics and Growth-Sergio Schmukler (sschmukler contribute to the debate on these issues, this project is @worldbank.org); and Financial Sector Strategy and Pol- evaluating the importance of the liquidity problem and icy Department-Stijn Claessens, Daniela Klingebiel, its determinants. and Ying Lin. With Ross Levine, University of Min- The project will analyze recent trends in domestic nesota; Tatiana Brandao, Catholic University, Rio de equity markets, using country-level data for a large set Janeiro; Yaye Sakho, University of Pennsylvania; and of countries to test whether the introduction of deposi- Francisco Vazquez, University of Maryland. tory receipts and the internationalization of equity Completion date: June 2002. markets are affecting the liquidity and development of local stock markets. It will measure the extent of the Financial Development and Contagion. migration of trading from domestic to international markets at the aggregate level, controlling for This project consists of several studies aimed at improv- macroeconomic and institutional factors affecting the ing the understanding of the functioning of financial development of local stock markets. markets, the benefits of financial integration, and the The project will also analyze the effect of migration effects of financial and banking crises. related to foreign direct investment on the companies First, the project examined the interaction between remaining in the domestic market. Using firm-level data deposit insurance and market discipline in the banking on value traded in the local market and international sector, and the impact of banking crises on market markets, the project will analyze how migration and discipline. This study used bank-level data for Argentina, delisting affect the trading activity of companies Chile, and Mexico, all of which experienced banking without American depository receipt programs. These crises during the sample period (1980s and 1990s). The data will also shed light on how much trading is migrat- results show that in these countries depositors-large or ing to international markets for firms with such programs. small, with local or foreign currency deposits-do exert Intemnational Economics 113 market discipline, punishing banks for risky behavior by alization with prudential regulation to prevent a mis- withdrawing their deposits and by requiring higher inter- match in maturity between assets and liabilities. est rates. Some authors have argued that bank-based and Deposit insurance need not reduce market discipline. market-based financial systems provide different types In Chile uninsured depositors seem to respond more of financing, with firms able to obtain more debt financ- aggressively than insured depositors to bank risks, while ing in bank-based systems and more equity financing in in Argentina no significant differences were observed in market-based systems. The study showed that the finan- the responses of the two groups. The results suggest cial structure of firms in different systems is not notably that depositors, even if covered by deposit insurance, different. might exercise market discipline when there is uncer- Third, the project examined stock market cycles in tainty about the future availability of their deposits-for 28 countries-in the G-7, Europe, East Asia, and Latin example, if the government has reneged on its promises America-characterizing the amplitude and duration of in the past, if the deposit insurance scheme is under- cycles over time and across regions. It also examined the capitalized, or if depositors are concerned about the cost claim that financial cycles are more protracted after of repayment (typically in the form of delays) through the domestic and external financial liberalization. deposit insurance fund. The findings show that financial liberalization does Crises seem to be wake-up calls for depositors. With- not necessarily lead to financial excesses. Financial cycles drawals become more frequent immediately after bank- become more severe only in the aftermath of the open- ing crises, but this kind of market discipline is much more ing of the economy to international capital flows. Over limited before and during crises. Interest rates were time, liberalized capital markets become more stable. responsive to bank risk taking throughout the sample Why? Markets with fewer capital controls are more period. These results suggest that depositors become exposed to shocks from abroad and thus more prone to more aware of the risk of losing deposits, and start to shift contagion. But eliminating capital account restrictions them, after they observe bank failures. favors the development of capital markets. As financial Second, the project studied the effects of financial inte- markets deepen and investors become more diversified, gration on firms' financing choices, using data on a large markets become less prone to wild gyrations. panel of nonfinancial companies in East Asia and Latin The findings also show that the behavior of stock America. This study focused on seven emerging market prices in developing countries differs from a economies that have experienced financial liberaliza- random walk, so that it is meaningful to analyze stock tion and crises-Argentina, Brazil, Indonesia, the Repub- market booms and busts. Monte Carlo experiments lic of Korea, Malaysia, Mexico, and Thailand. The data showed that the characteristics of cycles derived from cover the 1980s and 1990s, allowing comparison of pre- processes in which parameters are obtained from coun- and post-liberalization periods. try data differ from those derived from random processes The main results show that integration of financial in which returns are unpredicted. markets does affect firms' financing choices and that To continue this research, the project is creating a data the effects seem to be uneven. Firms that participate set on financial liberalization for the same 28 countries, in international markets obtain better financing from the early 1970s through the 1990s, with information opportunities-gaining the ability to extend their debt on restrictions on the domestic financial sector and inter- maturity structure, for example. Debt maturity tends to national capital flows. The data set, for which 11 indicators shorten, however, when countries undertake financial lib- of financial liberalization have been collected, will be valu- eralization. This implies that firms that do not participate able for measuring financial integration. in international markets are probably increasing their Fourth, the project examined whether local managers short-term liabilities. The shift in the maturity struc- of firms have an informational advantage in predicting ture toward the short term after financial liberalization currency fluctuations and crises. This study was based suggests that it could be important to accompany liber- on data from the Global Competitiveness Survey, which 114 Inlernationol Economics gathers the perspectives of managers around the world Reports on their country's economic, political, and institutional Kaminsky, Graciela, and Sergio Schmukler. 1999. "On Booms and situation. The data used are from surveys at the end of Crashes: Stock Market Cycles and Financial Liberalization." 1995, 1996, and 1997 and thus precede the crises in Asia, World Bank, Development Research Group, Washington, the Russian Federation, and Brazil. D.C. The results suggest that local managers were able to - . 2001. "Short- and Long-Run Integration: Do Capital predict the crises in the Republic of Korea and Thailand, Controls Matter?" Brookings Trade Forum 2000, pp. 125-78. but not those in Indonesia or Malaysia. The evidence also (Also issued as Policy Research Working Paper 2660, World suggests that there were information asymmetries: local Bank, Development Research Group, Washington, D.C., residents were shifting funds out of the country before 2001.) foreign investors were. Foreign market participants (such Kaufmann, Daniel, Gil Mehrez, and Sergio Schmukler. 1999. "Pre- as international mutual funds) and market analysts largely dicting Currency Fluctuations and Crises: Do Resident Firms did not expect the Asian crisis. Have an Informational Advantage?" Policy Research Working The results also show that local managers' private Paper 2259. World Bank, World Bank Institute, Washington, information can help predict exchange rate fluctuations. D.C. Their information seems superior to that revealed by Martinez Peria, Maria Soledad, and Sergio Schmukler. 2001. "Do macroeconomic and financial data and by markets. Depositors Punish Banks for Bad Behavior? Market Discipline Findings have been presented at the Central Bank of in Argentina, Chile, and Mexico." Journal of Finance 56(3): Chile, the Federal Reserve Board, the Federal Reserve 1029-56. (Also forthcoming in Leonardo Hernandez and Klaus Bank of New York, American Economic Association Schmidt-Hebbel,eds.,Banking,Financial/ntegration,andlnter- meetings (Boston and New York), European Econo- national Crises, Santiago: Central Bank of Chile.) metric Society meetings, Latin American and Caribbean Schmukler, Sergio, and Esteban Vesperoni. 2000. "Globalization Economic Association (LACEA) meetings (Buenos Aires, and Firms' Financing Choices: Evidence from Emerging Argentina; and Rio de Janeiro, Brazil), LACEA Winter Economies." Policy Research Working Paper 2323. World Bank, Camps on International Finance (Cartagena, Colombia; Development Research Group, Washington, D.C. and Santiago, Chile), Society for Economic Dynamics - . Forthcoming. "Firms' Financing Choices in Bank-Based meetings (P'hiladelphia), and World Bank conferences. and Market-Based Economies." In Ashl Demirgus-Kunt and The papers from the project have been posted on the Ross Levine, eds., FinancialStructure andEconomicDeve/opment. Web at http://www.worldbank.org/contagion/. Cambridge, Mass.: MIT Press. Responsibility: Development Research Group, Macroeco- nomics and Growth-Sergio Schmukler (sschmukler Trade Policy in Transition Economies @worldbank.org), and Finance-Maria Soledad Martinez Peria; and World Bank Institute, Governance, Regulation, Integration with the international economy is an essen- and Finance Division-Daniel Kaufmann. With Jon tial part of the transition from central planning to a mar- Tong; Sergio Kurlat; Graciela Kaminsky and Akiko Ter- ket system. Price distortions, so common under central ada, George Washington University; Gil Mehrez and planning, can be maintained only through formidable Esteban Vesperoni, International Monetary Fund; Arun trade and foreign exchange controls that divorce the Sharma, Federico Guerrero, Francisco Vazquez, Jose domestic from the international market. Dismantling Pineda, and Kevin Wang, University of Maryland; Car- these barriers promotes efficient domestic resource los Arteta, University of California at Berkeley; Cecilia allocation. International prices pose a competitive chal- Harun, Columbia University; Marco Sorge, Stanford Uni- lenge to domestic producers and signal needed structural versity; Matias Zvetelman, University of Buenos Aires; changes. Trade policy reform allows the link between Matteo Ciccarelli, Universitat Pompeu Fabra, Barcelona; domestic and international prices and markets and is and Miana Plesca, McGill University, Montreal. thus a key determinant of the pace and scope of the Completion date: June 2002. structural change necessitated by the transition. Infernalional Economics 115 This project has looked at trade policy reform issues tatives of transition economies, and presented at in the countries of the former Soviet Union. A study conferences. summarizing trade performance and the experience with Responsibility: Development Research Group, Trade-David trade policy reform in these countries recommended Tarr (dtarr@worldbank.org) and Bartlomiej Kaminski; strategies for increasing their integration with the inter- and Europe and Central Asia Region, Brussels Office- national economy-strategies that entail actions by the Constantine Michalopoulos. countries as well as by their main trading partners, the Completion date: June 2002. OECD countries. A study of customs unions showed that although Reports preferential trade areas in the Commonwealth of Inde- Cavalcanti, Carlos, Bartlomiej Kaminski, and Beata Smarzynska. pendent States (CIS) may have served a useful purpose 2000. TradeandForeign Investmentin Poland: Wi//Expores Rcover? in the past as a transitional device, the time for customs World Bank Country Study. Washington, D.C. [http://www. unions and free trade areas in the CIS is over. Integrat- worldbank.org/eca/eu-enlargement/index.html]. ing with the world economy should be the highest Dolgov, Sergei, Valery Ponomarev, and David Tarr, eds. 2000. priority now, and customs unions will retard that Russian Trade Policy and WTOAccession(in Russian). Washing- integration. Work on the issue of accession to the World ton, D.C.: World Bank [http://wwwl.worldbank.org/wbiep/ Trade Organization by the transition economies empha- trade/l. sized that accession provides a unique opportunity for the Kaminski, Bartlomiej. 1999. "The EU Factor in the Trade Policies acceding country to "lock in" a trade policy that is of Central European Countries." Policy Research Working beneficial to that country, as long as it adopts a less than Paper 2239. World Bank, Development Research Group, Wash- minimalist approach in its accession offer. ington, D.C. The project has also examined issues relating to tar- . 2000. "The 'EU Factor' in Transition: Credibility of Com- iff policy and tariff uniformity for transition economies, mitment, Institutional Change, and Integration." In S. Antohi focusing on the Russian Federation. And it has assessed and V. Tismaneanu, eds., Between Past and Future: The Revola- the progress of the transition economies in integrating tions of 1989 and the Strugglefor Democracy in Central and East- into the world trading system and acceding to the World ern Europe. Budapest: Central European University Press. Trade Organization. This research concluded that for . 2000. "Industrial Restructuring as Revealed in Hungary's many countries of the former Soviet Union, integrating Pattern of Integration into EU Markets." Europe-Asia Studies effectively will require considerable reform and adjust- 52(3). ment. But the United States and the European Union also . 2000. "Political Economy of the New Europe: EU East- may need to make some changes, especially in their ern Enlargement." In Sabina A.-M. Crisen, ed., NATO and designation of some of these countries as "nonmarket" Europe in the 21st Century: New Rolesfor a Changing Partnership. economies. Washington, D.C.: Woodrow Wilson International Center for Research has also looked at the impact of the Euro- Scholars. pean Union on the trade policies of Central Europe. It - 2001. "HowAccession to the European Union Has Affected found that, except for certain "sensitive products," the External Trade and Foreign Direct Investment in Central link with the European Union has offset protectionist European Economies." Policy Research Working Paper 2578. pressures in Central Europe. And an examination of the World Bank, Development Research Group, Washington, D.C. effect of overvalued exchange rates on growth and on . Forthcoming. "Integration into Global Production and trade policies, initially focusing on Kazakhstan, found that Distribution Networks through FDI: The Case of Poland." the effect on growth is strongly negative, just as in other Communist Economies and Economic Transformation. parts of the world. Kaminski, Bartdomiej, and Michelle Riboud. 2000. Foreign Invest- The study's findings and recommendations have been ment and Restructuring: The Evidencefrom Hungary. World Bank communicated to the governments in policy dialogue Technical Paper453, Europe and Central Asia Poverty Reduc- on international trade reform, discussed with represen- tion and Economic Management Series. Washington, D.C. 116 International Economits Kaminski, Bartlomiej, and Beata Smarzynska. 2001. "Rownanie w Saharan Africa as these countries attempt to strengthen gore: handel zagraniczny w 2001 " (Catching up: Poland's trade industrial performance, increase agricultural productiv- in 2001). Rzeczpospolita (Warsaw; January). ity and competitiveness, and engage in international Kaminski, Bartlomiej, Z. Wang, and L. Alan Winters. 2000. "Zwiazki trade. Expanding access to information on international miedzy stabilizacja makroekonomiczna, liberalizacja a reori- standards and strengthening the implementation of trade entacja eksportu" (Links between macroeconomic stabilization, obligations are important development objectives-goals liberalization, and export reorientation). In T Kowalik and J. that require both stronger empirical understanding and Hausner, ecls., Polscy ekonomisci w swiecie (Polish economists creative programs in capacity building. abroad). Warsaw: Scientific: Publishers. This project is aimed at contributing to those objec- Michalopoulos, Constantine. 1997. "The Economics of Customs tives and thus helping to strengthen Sub-Saharan Africa's Unions in the Commonwealth of Independent States." Post- integration into the world trading system and its capac- SovietGeography and Economics 38(3): 125-43. (Also issued as Pol- ity to exercise rights and meet obligations under World icy Research Working Paper 1786, World Bank, Development Trade Organization agreements on trade standards. The Research Group, Washington, D.C., 1997.) project involves the following activities: - . 1998. "WTO Accession for Countries in Transition." * Through workshops and capacity building efforts, Policy Research Working Paper 1934. World Bank, Develop- assessing and building awareness of the range and impor- ment Research Group, Washington, D.C. tance of international standards and technical regula- - 1999. "Integration of the Transition Economies into the tions affecting Africa's current and prospective trade in World Trading System." Policy Research Working Paper 2182. food, agricultural, and light manufactured products. World Bank, Developmenc Research Group, Washington, D.C. * Analyzing the status of African supply systems and Michalopoulos, Constantine, and David Tarr. 1996. Trade Perfor- regulatory and certification arrangements and the gaps mance and Policy in the Net Independent States. Directions in between these systems and supply chains linked to Development Series. Washington, D.C.: World Bank. (Also OECD markets. The analysis will also cover the strate- published in Russian.) gies that different industries use to comply with the Shatz, Howard, and David Tarr. Forthcoming. "Exchange Rate evolving international and external standards and the bot- Overvaluation and Trade Protection: Lessons from Experi- tlenecks and costs that they face. ence." In World Bank, Trade Policyfor Developing Countries in a * Highlighting best practice models-in Africa and Global Economy: A Sourcebook. Washington, D.C. (Also forth- elsewhere-for implementing reforms and capacity build- coming, in a longer version, in Zdenek Drabek, ed., Globalisa- ing in the area of international standards, including pub- tion under Threat: The Stability of Trade Policy and Multilateral lic, private, and joint public-private initiatives. Agreements, London: Edward Elgar.) * Developing five action plans-for Kenya, Mozam- Tarr, David. 1999. "Design of Tariff Policy for Russia." In Harry bique, Nigeria, South Africa, and Uganda-for expand- Broadman, ed., Russian Trade Policy Reformfor WTO Accession. ing access to and use of information on international World Bank Discussion Paper 401. Washington, D.C. standards. This work will include identifying infra- structure and capacity building needs in the public and African Trade Facilitation and Standards: private sectors. The study expects to extend the work to Bridging the Divide other countries over time, including those in West Africa. * Developing databases and a design for a Standards Trade facilitation and standards are at the center of trade Access Africa network to deliver information on standards policy discussions. Standards and technical regulations and regulations critical to development and trade oblig- may either raise or lower economic efficiency, promote ations. This network will leverage the World Bank's or block competition, facilitate or constrain international Global Distance Learning network. trade, and enable or exclude the participation of the The project is being carried out in partnership with poor in economic activities. These issues are of particu- the African Economic Research Consortium (AERC) lar importance for the least developed countries in Sub- and the Trade and Investment Policy Secretariat of South International Etonomics 117 Africa, among other groups. Fieldwork began in July tion from Asia Pacific Economic Cooperation, the Organ- 2001, and a planning forum hosted by the World Bank isation for Economic Co-operation and Development and the AERC was held that month in Nairobi, Kenya. (OECD), the World Trade Organization, and other National workshops will take place in the first quarter of sources to identify impediments to trade in the sector. 2002, and country action plans will be prepared by Sep- The telecommunications database draws together infor- tember 2002. A regional conference will be held in 2002. mation on more than 100 countries from the International Information on standards and trade can be found on Telecommunication Union, the OECD, the World Bank, the Web at http://www.worldbank.org/research/trade/. and other sources. A new Web site focusing on these issues in Africa is A Web site has been established (http://www. being developed and will be linked to that site. worldbank.org/trade/services.html) as a vehicle for dis- Responsibility. Development Research Group, Trade-John S. seminating research on trade in services conducted within Wilson (jswilson@worldbank.org) and Tsunehiro Otsuki; and outside the Bank. The Web site now includes links and Africa Region Technical Families, Private Sector- to information on trade in services, including measures, Steve Jaffee. The U.S. Agency for International Devel- trade flows, and other economic variables, and will soon opment has contributed funding for the research through be expanded. a trust fund (Africa Trade and Investment Policy Project). The program is being undertaken by World Bank staff Completion date: September 2002. in collaboration with other international organizations-such as the OECD, Andean Pact, World Trade Organization, Trade in Services International Telecommunication Union, Organization of American States, and United Nations Conference on Trade In response to a widely felt need for more research on and Development-and with research networks and trade in services, especially on issues of interest to devel- institutions-notably the Latin American Trade Network, oping countries, this program is generating and drawing the National Council of Applied Economic Research together data and research on services trade. Its aim is (India), the Coordinated African Program of Assistance to improve the formulation of services trade policy and on Services, the Trade Policy Forum of the Pacific Eco- facilitate effective participation by developing countries nomic Cooperation Council, and the Economic Research in the World Trade Organization negotiations on ser- Forum for the Arab Countries, Iran, and Turkey. vices. The program supplements sector-specific work Responsibility: Development Research Group, Trade- in the World Bank on finance, transport, telecommuni- Aaditya Mattoo (amattoo@worldbank.org) and Carsten cations, and other service sectors. Fink. With Randeep Rathindran, University of Maryland; The program has three components: collecting infor- Ileana Cristina Neagu; and Robert Simms. The U.K. mation on trade in services; conducting research pro- Department for International Development is con- jects (on such topics as liberalizing trade in maritime tributing funding for the program. transport services, the determinants of foreign investment Completion date: October 2002. in service industries, the economic performance of the telecommunications sector, and the impact of service Reports sector policies on growth); and mainstreaming trade in Fink, Carsten, Aaditya Mattoo, and llcana Cristina Neagu. 2001. services at three levels-national policies (through direct "Trade in International Maritime Services: How Much Does advice and capacity building), World Bank country oper- Policy Matter?" Policy Research Working Paper 2522. World ations, and the international policy environment, includ- Bank, Development Research Group, Washington, D.C. ing support in negotiations. Fink, Carsten, Aaditya Mattoo, and Randeep Rathindran. 2001. Preliminary databases have been created on maritime "Liberalizing Basic Telecommunications: The Asian transport and basic telecommunications services. The Experience." Paper presented at the Conference on Trade, maritime transport services database covers 58 countries Investment, and Competition Policies, Hamburg Institute of (37 developing economies), drawing together informa- International Economics, Hamburg, Germany. 118 International Etonomits Mattoo, Aaditya. Forthcoming. "Shaping Future Rules for Trade elsewhere on the effects of international standards. in Services: Lessons from the GATS." In Takatoshi Ito and Anne Research has also been completed on the impact of envi- Krueger, eds., Trade in Services. Cambridge, Mass.: National ronmental standards on trade in pollution-intensive Bureau of Economic Research. exports. The case studies show that developing countries Mattoo, Aaditya, and Ludger Schuknecht. 2000. "Trade Policies face major challenges in gaining access to information on for Electronic Commerce." Policy Research Working Paper international standards and that many have difficulty 2380. World Bank, Development Research Group, Washington, implementing World Trade Organization (WTO) oblig- D.C. ations on product standards. The findings suggest a role Mattoo,Aadirya,MarceloOlarreaga, and Kamal Saggi. 2001. "Mode for development agencies in assisting in areas of clear of Foreign Entry, Technology Transfer, and FDI Policy." Paper public good. presented at the World Bank Trade Research Seminar, Wash- The project has organized two seminars on standards ington, D.C. -one in Panama City for representatives of seven Cen- tral American governments and ano