48380 ResearchDigest World Bank VOLUME 1 ✬ NUMBER 1 ✬ FALL 2006 Money for Nothing: The Dire Straits of Medical Practice in Delhi, India ● Jishnu Das and Jeffrey Hammer IN THIS ISSUE A cademic and policy discussions to their “knowledge frontier” than those Money for Nothing: The Dire about health care increasingly paid fixed salaries. Furthermore, effort Straits of Medical Practice in view the quality of care as a cru- and competence are complements, not Delhi, India … page 1 cial determinant of health outcomes. substitutes. More competent doctors Why does the quality of medical care vary Policy levers to address quality dispar- spend more time with patients, ask in India—especially Delhi? ities in health care must cover a vari- more questions, and do more exami- World Bank Research Digest: ety of underlying causes. If geographic nations. Less competent practitioners An Effective Way to Disseminate variations in practice quality reflect deliver worse care because of the direct Research Findings … page 2 differences in provider competence, effects of lower competence and the Research informs all World Bank the solution lies in geographic incen- indirect effects of lower effort. work—and this digest summarizes some tives for doctors, greater movement of The new paper explains that the of the most important recent findings patients, or both. If variations result differences between what doctors from differing provider effort, stronger know and what they do reveal dispari- Bank Supervision and Corruption in Lending … page 3 performance incentives are needed. ties across institutional settings and In a new paper, Das and Hammer geographic locations. Private sector Differences in approaches to bank explore differences in doctor compe- practitioners do more or less what is supervision affect the obstacles that companies face in raising finance tence and incentives to explain varia- expected of them by patients. Poorer tions in the quality of care in India. A patients receive low-quality advice Who Benefits from Residential Water companion paper was published in and spend a fair amount of money for and Electricity Subsidies? … page 4 2005 in the Journal of Development Eco- nothing—on low-value advice and un- Improving the design of water and nomics. Both papers rely on interviews necessary drugs. Wealthier patients electricity subsidies could help extend with 205 doctors using hypothetical get better advice, both because they services to more poor people medical conditions as well as records see more competent providers and Child Labor and Agriculture of what actually happened, and on because their providers put in more ef- Shocks … page 5 measures of effort that reflect what fort. But wealthier patients also spend Child labor could be reduced with policies doctors do when they show up for a fair amount of money for nothing, on that insure farming households against work (abstracting from the number unnecessary drugs. crop losses and other shocks of hours they work). Doctors in the Public sector practitioners spend Market Access for Sale … page 6 public sector are often absent, though less time and effort on patients and Foreign lobbying by exporters from less so in urban than rural areas. do less than they know they should. the Americas has focused on receiving The authors find that provider com- But this public-private disparity masks preferential access to U.S. markets—with petence and effort both play a role in variations in the public sector. Public great success the quality of care. The data indicate providers in smaller clinics and dispen- that what doctors actually do is very saries substantially underperform pub- Public Disclosure: A Tool for different from what they know they lic providers in hospitals; the latter are Controlling Pollution … page 7 should do. This gap reflects the cost of comparable to private practitioners. Long-running experimentation and effort relative to reward and illustrates Unfortunately, poor people do not evaluation have taught the Bank a response to incentives: doctors paid really use public hospitals. Over a important lessons about policy toward for each service they provide are closer (continued on page 6) public disclosure of industrial pollution 2 World Bank ResearchDigest World Bank ations within the Bank. The industrial aware of the findings of research to pollution projection system, featured improve their diagnostics, monitor- Research Digest: in this Research Digest, relies on close ing, and decisionmaking. This aware- An Effective Way collaboration between environmental agencies in developing countries and ness does not require a perfect match between the findings of research and to Disseminate World Bank researchers. It has been used in several Bank projects and is the questions faced by specific prac- titioners. Often, ideas for reforming Research Findings being applied by several non-Bank us- ers around the world. a policy or launching a new develop- ment program come from learning Poverty assessments are now com- about different policies and programs ● François Bourguignon mon practice in many countries. They in other countries. were developed by Bank researchers in This public good aspect of re- response to a need from practitioners search shows the importance of ef- T his new quarterly publication to base poverty reduction policies on ficient dissemination of results. Re- aims to improve dissemination deep knowledge of poor populations. searchers, not practitioners, are the of World Bank research in the de- The assessments rely on household main readers of research work. This is velopment community. Knowledge is surveys pioneered by the Living Stan- understandable: research is a cumu- an essential input to World Bank op- dards Measurement Study, launched lative process, and researchers must erations in general and lending in par- by the Bank in the 1980s. somehow incorporate all relevant ticular—to the extent that the World Impact evaluations are the best past research in their work. Practitio- Bank is sometimes called the “Knowl- example of the close in- ners have other obliga- edge Bank.” This strong complemen- tegration between World tions and do not have tarity between knowledge and lending Bank research and op- Development time to regularly browse has sometimes been described as the erations. Based on com- researchers, in the growing number of “lending-learning-knowledge” cycle. parisons between agents academic journals pub- World Bank loans to developing coun- subject to the interven- the World Bank lishing the most salient tries for projects, programs, and policy tions being evaluated and elsewhere, research work. Hence reforms provide a unique opportu- and control groups, im- the importance of de- nity to discover what works and what pact evaluations permit must be willing veloping more direct doesn’t in development depending the attribution of key to listen and dissemination channels on the format of the interventions fi- development outcomes that may inform devel- nanced by the Bank and their context. such as poverty, health observe opment practitioners This knowledge should make World status, school enroll- on remarkable results Bank assistance to developing coun- ment, employment, and business cre- that they should not miss. That is the tries more effective and its financing ation to the interventions, rather than objective of the World Bank Research more attractive. other factors. Essentially demand Digest. Most of the learning part of this driven, the multiplication of these The efficiency of dissemination cycle occurs through research—hence evaluations will help the develop- implies tradeoffs between quantity the importance of research for the ment community evaluate particular and precision. The choice is between Bank. Relying on actual experience, operations and determine what works featuring many research papers with researchers try to understand the and what doesn’t in specific areas. succinct descriptions of their findings relationship between the design of These examples show why ap- and fewer papers with longer accounts projects or policies and their devel- plied development researchers, in of their results and implications. The opment outcomes depending on the the World Bank and elsewhere, must Research Digest takes the second ap- characteristics of the area or country be willing to listen and observe. They proach, and each quarter will offer where the project or policy was imple- can do so only through direct contact six or seven one-page summaries mented. To do so, they rely on ad- with Bank task managers as well as of remarkable World Bank research. vances in knowledge from the research government counterparts, analysts, The papers featured will be chosen community, to which they sometimes and researchers in partner countries. by Bank research managers based contribute. They also develop new If researchers get much of their inspi- on relevance, originality, and impor- analytical tools that can be used by ration from actual development inter- tance. (Readers wishing to access the practitioners when establishing policy ventions that they have been directly Policy Research Working Papers and diagnostics or monitoring aspects of or indirectly involved in, the reverse their abstracts, our monthly Research development. must also be true. E-Newsletter, the Research Highlights of There are many examples of this For the lending-learning-knowledge DEC’s Research Group, and other tight integration of research and oper- cycle to work, practitioners must be (continued on page 8) World Bank Research Digest 3 Bank Supervision and cies than in countries with weaker agencies. Corruption in Lending The finding that powerful supervi- sion agencies are associated with lower integrity of bank lending pro- ● Thorsten Beck, Asli Demirgüç-Kunt, and Ross Levine vides support for the political/regula- tory capture view, which emphasizes that these agencies are prone to cap- B anks provide a large share of ity and incentives of private agents to ture and manipulation by politicians, external finance to businesses overcome information and transaction regulators, or both. But this conclu- around the globe. The Basel costs, so that private investors can sion needs to be tempered. Powerful Committee, International Monetary exert effective governance over banks. supervision is so strongly correlated Fund (IMF), and World Bank promote Beck, Demirgüç-Kunt, and Levine with poor national institutions (inef- the development of powerful supervi- use data on supervision from Rethink- fective government, absence of the sion agencies to monitor and disci- ing Bank Regulation: Till Angels Govern by rule of law, high corruption) that it is pline bank behavior. Yet there are few Barth, Caprio, and Levine and firm- difficult to identify an independent re- studies of international differences in level data from the 1999 World Business lationship between supervision power bank supervision and how they influ- Environment Survey of more than 2,500 and bank corruption when controlling ence the obstacles that corporations small, medium-size, for these institutional face in raising external finance. There and large firms in 37 Bank supervision traits. is also little cross-country evidence on countries. The authors Finally, the authors’ which bank supervision policies facili- estimate the relation- should focus on findings are consistent tate efficient corporate finance and ship between the degree forcing accurate with the private monitor- which reduce bank corruption. of corruption—based on ing view: bank supervi- Beck, Demirgüç-Kunt, and Levine the answer to the ques- information sion strategies that focus empirically assess the relationship tion, “Is the corruption disclosure on forcing accurate infor- between bank supervision policies of bank officials an ob- mation disclosure—and and the degree to which corruption stacle for the operation not distorting the incen- in lending prevents firms from rais- and growth of your business?”—and tives of private creditors to monitor ing external finance. Three theories measures of supervision power (such banks—facilitate efficient corporate of government regulation provide a as the ability to intervene in banks, finance. This view recognizes that framework for their findings. The first, replace managers, force provisioning, private agents face substantive in- known as the supervisory power view, stop dividends and other payments, or formation and enforcement costs holds that strong official supervision acquire information) and of the degree when monitoring banks, but also that of banks can improve their corporate to which regulations require informa- politicians and regulators act in their governance, because private agents tion disclosure by banks and give own interests and not necessarily to often lack the incentives and capabili- private creditors incentives to monitor reduce market friction. Private moni- ties to monitor powerful banks. This banks (such as whether bank directors toring has an especially beneficial theory assumes that governments and officials face criminal prosecution effect on the integrity of bank lending have the expertise and incentives to for failure to disclose information, in countries with sound legal and ad- ameliorate market imperfections and whether banks must disclose consoli- ministrative institutions. improve bank governance. dated accounts, whether international Bank supervision clearly matters. An alternative theory, the accounting firms audit banks, and Policies that redress market failures political/regulatory capture view, whether there is implicit or explicit de- by forcing accurate disclosure of in- argues that politicians and supervi- posit insurance). The authors control formation reduce the obstacles that sors do not maximize social welfare, for a range of firm- and country-specif- firms face in raising external finance, but rather their private welfare. Thus, ic characteristics and use instrumental and active bank supervision can help if bank supervision agencies have variables to control for endogeneity. reduce information costs and improve the power to discipline noncompli- The results strongly refute the view the integrity of bank lending. Powerful ant banks, politicians and supervi- that powerful supervision agencies supervision agencies too often do not sors may use this power to induce with the authority to monitor and dis- act in the best interests of society. banks to divert credit to politically cipline banks facilitate efficient corpo- connected firms. rate finance. Firms tend to face greater Thorsten Beck, Asli Demirgüç-Kunt, and Ross Finally, the private monitoring view obstacles in obtaining bank loans as a Levine. Forthcoming. “Bank Supervision and argues that bank supervision policies result of corrupt bank officials in coun- Corruption in Lending.” Journal of Monetary should focus on enhancing the abil- tries with stronger supervision agen- Economics. 4 World Bank ResearchDigest Who Benefits half as much of the value of the subsi- proxy means-testing and self-selec- dy as they would if the subsidies were tion present relatively progressive from Residential distributed randomly across the entire distributions. Still, these schemes ex- Water and population—while many poor house- holds are excluded altogether because clude a substantial proportion of poor people—due to low utility coverage Electricity they are not connected to the network. Only 2 of 26 quantity-based subsidy for this group. The large difference in access be- Subsidies? programs come close to achieving even a neutral subsidy distribution. tween poor and nonpoor households suggests that connection subsidies are All the rest are regressive. a better way to reach the poor. Assum- ● Kristin Komives, Vivien Foster, This means that all or nearly all ing that all unconnected households Jonathan Halpern, and Quentin residential customers receive a sub- are offered and accept subsidized con- Wodon sidy—and richer consumers receive nections, the distribution of benefits more than poorer ones. In most wa- from connection subsidies is nearly ter programs studied, the poorest 40 always progressive—meaning that U tility subsidies are often seen percent of the population receives lower-income households receive more as a way to expand coverage only 5–20 percent of subsidy benefits. subsidies than higher-income ones. and make services affordable Yet for poor households that do ben- But this finding assumes that, when to poor people. Yet a new study by efit, utility subsidies represent 3–4 introducing connection subsidies, un- Komives, Foster, Halpern, and Wodon percent of household income. connected households at each income shows that such subsidies tend to Why do quantity-based utility sub- level will connect at the same rate. benefit the middle class and well off. sidies perform so poorly? That assumption is unlikely to hold, The study examines 32 subsidy • Poor households are less likely because utilities often face constraints programs from 13 water utilities and than others to be connected to in expanding their networks into poor 27 electricity utilities in developing networks. Households that lack areas. And even where networks are countries. Most of the programs in- access to services cannot benefit already present, many poor house- volve quantity-based subsidies (which from consumption subsidies. holds face nonfinancial obstacles to are equally common for water and • Differences in consumption connecting—such as not having legal electricity)—that is, basing service between poor and better-off title to the property they occupy. prices on consumption. The study households are less than is often The study notes that despite large decomposes determinants of target- assumed, so using quantity as a consumption subsidies, only 20–30 per- ing performance to examine varying basis for targeting is ineffective. cent of poor households in Africa con- access, consumption, and targeting • Even when poor households do nected to utility networks even when performance across rich and poor consume less than better-off they were available. Unless those rates populations. Explicit utility subsidies ones, tariff structures often fall are dramatically improved, connection come in two forms: those that cut the heaviest on those who consume subsidies will continue to dispropor- cost of consumption and those that the least. Households that con- tionately benefit nonpoor households, reduce connection charges. sume small quantities may face particularly in Sub-Saharan Africa. Consumer subsidies are a fea- much higher unit prices than Because most water and sanita- ture of water and electricity services do larger consumers. Moreover, tion subsidies are captured by the around the world. Most water utilities failure to apply full cost recovery nonpoor, most households would see charge tariffs far below levels required tariffs to households consuming higher bills if subsidies were elimi- to recover costs. Nearly 40 percent do at higher levels means that even nated or restructured. But the bulk of not even cover operation and main- large consumers receive substan- any price increase would be paid by tenance costs. Average water tariffs tial benefits from subsidies. nonpoor households. in low-income countries are about a Quantity-based subsidies are not The study concludes that there is tenth of those in high-income coun- the only type. Many utilities offer con- potential for improving utility sub- tries. For electricity, about 15 percent nection subsidies, targeted by geo- sidies, whether by switching from of utilities have average tariffs that are graphic or proxy means-test criteria. quantity-based to more sophisticated too low to even cover operation and In addition, self-selection is used approaches or by making more use of maintenance costs. to target subsidies for lower-quality connection subsidies. The study’s results are sober- services such as public standpipes. ing. The quantity-based subsidies These schemes perform somewhat Kristin Komives, Vivien Foster, Jonathan Halpern, practiced by four-fifths of water and better than quantity-based ones. The and Quentin Wodon. 2006. “Water, Electricity, electricity utilities surveyed are starkly geographic schemes examined present and the Poor: Who Benefits from Utility Subsi- regressive. Poor households capture an almost neutral distribution, while dies?” World Bank, Washington, D.C. World Bank Research Digest 5 Child Labor and Agriculture Shocks a shock and when they hold durable assets. Moreover, the probability of taking a loan in response to a shock ● Kathleen Beegle, Rajeev Dehejia, and Roberta Gatti is higher for households with durable assets. Crop shocks significantly increase But the authors Using four rounds of Child labor has acknowledge that there child labor and decrease school household panel data are different possible in- enrollment—but households with from the Kagera region traditionally terpretations of their re- assets can almost entirely offset of Tanzania, the authors been viewed as sults. The effect of shocks these effects find that income shocks on households could be increase child labor by a consequence explained by myopia or W hat are the links between 50 percent. Households of poverty, but by an extremely high dis- crop shocks, household as- in Kagera use almost count rate relative to the sets, and child labor? In no purchased inputs some recent interest rate. Still, the a new article, Beegle, Dehejia, and and rudimentary tech- country studies authors believe that they Gatti analyze the extent to which in- nology, and wage labor offer the most plausible come shocks increase child labor and is limited. When hit by have called that interpretation of their decrease enrollment, and whether a shock, households into question. findings. household assets mitigate the effects tend to increase their Regardless of how of these shocks. use of child labor—typ- The article the empirical results are Child labor has traditionally been ically by having children examines the interpreted, the fact re- viewed as a consequence of poverty, substitute for adults mains that households but some recent country studies have in household activi- role that child increase child labor in called that assumption into question. ties such as gathering labor plays as a response to crop losses. The article examines the role that firewood and water. Child labor is a major child labor plays as a buffer against The data also show buffer against policy problem—not only crop losses—and suggests that insur- that shocks cause a 20 crop losses and for moral reasons but ance or access to credit might reduce percentage point drop also because it slows the its extent. in school enrollment suggests that accumulation of human The authors also explore how as- (compared with an insurance or capital and is inimical to sets affect household responses to average enrollment development. The find- income shocks such as crop losses. rate of 70 percent). access to credit ings of this article imply As such, the article relates to recent Households with as- might reduce that child labor could development research examining sets can offset income be reduced with policies how credit constraints affect child shocks to a substantial its extent that insure agricultural labor. degree: at the mean households against crop Child labor entails a tradeoff for level of asset holdings, losses and other shocks. households between immediate ben- households can offset more than 80 Increasing household access to credit efits and, to the extent it interferes percent of the effect of income shocks. in response to crop shocks would also with the development of a child’s The article finds that household as- reduce child labor and raise house- human capital, potential long-run sets decrease in response to shocks hold welfare. costs. When faced with a transitory but that wealthier households draw shock, households would ideally use down assets to a lesser extent—sug- Kathleen Beegle, Rajeev Dehejia, and Roberta asset holdings—either as a buffer gesting that they may be borrowing Gatti. Forthcoming. “Child Labor and Agri- or as collateral to obtain credit—to in response to shocks. The results culture Shocks.” Journal of Development offset it. also suggest that poorer households Economics. The article also relates to the use assets as a buffer stock, drawing broader development literature on the them down in times of need, whereas permanent income hypothesis and wealthier households’ behavior is con- consumption smoothing. If house- sistent with access to credit. holds succeed in smoothing their Beegle, Dehejia, and Gatti present consumption but lack buffer stocks a range of evidence to corroborate this or are credit constrained, they must interpretation of their results. They use other mechanisms to cope with show that households are more likely income shocks. to take loans when they experience 6 World Bank ResearchDigest (continued from page 1) Market Access for Sale two-year period about 55 percent of medical visits by members of low- and ● Hiau Looi Kee, Marcelo Olarreaga, and Peri Silva middle-income households were to private doctors. Just 31 percent were to clinics and dispensaries—and only Has foreign lobbying by exporters ucts and countries at a statistically 13 percent to public hospitals. The from the Western Hemisphere significant level. When setting prefer- public sector spends more than 80 improved their access to U.S. ences, the U.S. government puts five percent of the government’s health markets? times more weight on foreign lobby- budget on salaries for doctors and ing contributions than on forgone tar- heavy subsidies to educate them. This F or many countries foreign lobby- iff revenues. These estimates suggest is another form of money for nothing: ing is likely to be associated with that market access is for sale—and substantial public spending on cur- preferential rather than nondis- foreign firms are buying it. sory, poorly delivered health care. criminatory access to markets—for These striking results contrast with The concentration of more com- two reasons. First, there are stronger evidence on domestic lobbying. In a petent providers in richer neighbor- incentives to lobby for preferential ac- 2004 review of empirical approaches hoods, combined with the low use of cess. Exporters expect to grab market to the political economy of trade public hospitals, implies that poor share away from exporters from other policy, Gawande and Krishna conclude urban residents are particularly under- countries, not just from domestic pro- that the weight granted to domestic served—for several reasons: ducers in the importing country—as producers and their lobbying in the • The private practitioners they visit under a most favored nation (MFN) U.S. government’s objective function have low competence. tariff reduction. Second, developing is close to the weight granted to social • They receive worse medical care countries have played little role in welfare. both due to the direct effects of multilateral trade negotiations but Kee, Olarreaga, and Silva offer two lower competence and the indirect have benefited from several bilateral explanations for the difference between effects of lower effort. and unilateral preference schemes their estimates and those in the earlier • Weaker incentives in the public when exporting to developed markets. literature. Their article assumes that sector offset the protective effects The recent proliferation of prefer- contribution functions are not differen- of higher competence. ential agreements between the United tiable, so the Nash equilibrium solu- Thus the poor receive low-quality States and its trade partners in the tion to the lobbying game differs from care from the private sector because Americas indicates that their foreign the one in the Grossman and Helpman doctors do not know much—and low- lobbying is likely to focus on prefer- article. In this setup the government quality care from the public sector ential access. In the early 1990s only is left on its participation constraint, because doctors do not do much. Poor about 15 percent of Latin American whereas in Grossman and Helpman’s households are better off visiting less- exports entered the United States work the government grabs part of the qualified private practitioners than under preferential schemes. By the lobbying rent. This additional restric- more qualified public doctors. early 2000s that share was 50 percent. tion implies that smaller amounts of The findings of Das and Hammer Kee, Olarreaga, and Silva explore foreign lobbying contributions will be suggest that weak medical services, at the role of foreign lobbying in obtain- sufficient to move the government away least for poor people, are more a func- ing tariff preferences. They first develop from welfare maximization, which may tion of a lack of incentives for public a simple framework, based on a model explain larger tariff preferences. providers than a lack of competence developed by Grossman and Helpman, The authors also assume that the among private providers. This would to explain the extent to which lobbying government grants different weights tend to indicate that solutions are by foreign firms affects their preferen- to tariff revenues than to other com- more likely to come from improving tial access to U.S. markets. They then ponents of social welfare in its objec- information for consumers and reduc- estimate an equilibrium contribution tive function. If the weight put on tariff ing the demand for extensive inappro- schedule for 34 countries in the West- revenues is lower than the weight put priate treatment. The authors suggest ern Hemisphere using data on foreign on other components of social wel- changing the incentives of public pro- lobbying expenditures from the U.S. fare, that could explain the difference viders rather than increasing provider Department of Justice and on tariff between the authors’ estimates and competence through training. preferences at the industry level from those in the rest of the literature. the International Trade Commission. Jishnu Das and Jeffrey Hammer. Forthcoming. The authors find that foreign lob- Hiau Looi Kee, Marcelo Olarreaga, and Peri “Money for Nothing: The Dire Straits of Medical bying contributions explain variations Silva. Forthcoming. “Market Access for Sale,” Practice in Delhi, India.” Journal of Develop- in U.S. tariff preferences across prod- Journal of Development Economics. ment Economics. World Bank Research Digest 7 Public Disclosure institutions value the Bank’s contribu- The environmental performance —A Tool for tions—particularly its strong policy orientation, extensive field experience, rating programs adopted by Indonesia, the Philippines, China, Vietnam, and Controlling knowledge of related initiatives in other countries, analytical skill, and, India include the following elements: • They focus on conventionally regu- Pollution perhaps most important, capacity to sustain an innovative vision while lated pollutants rather than toxic chemicals. engaged in the complex institutional, • They use locally recognized bench- ● Susmita Dasgupta, Hua Wang, and technical, and political activities that marks to grade performance. David Wheeler always surround new policy initiatives. • They compare audited factory Second, a successful transition emissions reports with benchmark from pilot experimentation to large- standards and grade the results in Though the research in this article scale implementation depends on sev- a few overall categories. is not new, it has had a major impact eral factors: credible demonstration of • They assign locally understood on policy in several Asian countries impact; widespread dissemination of color codes or symbols to perfor- in recent years—offering lessons information about the pilot program, mance grades. from a policy experience with a long its rationale, and results; and sus- • Performance rating agencies main- time horizon tained technical assistance delivered tain close communications with to collaborating institutions, both dur- audited facilities. ing the pilot phase and in scaling up. Table 1 summarizes evidence on M ost pollution regulation is Third, long time horizons are the impacts of these collaborative still based on the quantity critical for this kind of work. The full disclosure programs in Indonesia, the of emissions. Market instru- development cycle, from catalytic re- Philippines, China, and Vietnam. In all ments are rarely used, but public search to national scale-up, may take four countries the programs strength- disclosure of pollution has shot up 5–10 years. ened conventional regulation by sig- since the 1980s. Public disclosure Finally, flexibility is essential. Suc- nificantly increasing compliance. has particular appeal in developing cessful collaboration to promote pol- countries, where corruption and weak icy reform does not happen smoothly, Susmita Dasgupta, Hua Wang, and David enforcement often make it difficult predictably, and on a fixed schedule. Wheeler. 2005. “Disclosure Strategies for Pollu- for regulators to control pollution by It requires constant attention to re- tion Control.” In T. Tietenberg and H. Folmer, more conventional means. lationships, operational details, and eds., The International Yearbook of En- Although weaknesses in conven- emerging opportunities, as well as vironmental and Resource Economics tional regulation open the door for consistent application of high techni- 2005/2006: A Survey of Current Issues. community pressure through public cal standards and analytical skills. Cheltenham, U.K.: Edward Elgar. disclosure, public interpretations of emissions data can be hampered by low education levels and a scarcity of Table 1: Performance Rating Programs: Changes in Polluters’ Compliance Status technically informed nongovernmen- tal organizations (NGOs) to help with Number Share of factories interpretation. The World Bank has of factories (percent) Increase in compliance Country, program Noncompliant Compliant Noncompliant Compliant (percentage points) worked on this issue with environ- Indonesia, mental agencies from several develop- PROPER ing countries, with the goal of devel- 1995 92 54 63 37 oping and evaluating environmental 1997 57 89 39 61 24 Philippines, performance ratings. Indonesia began EcoWatch its disclosure program in 1994, fol- 1997 48 4 92 8 lowed by the Philippines’s EcoWatch 1998 19 26 42 58 50 in 1997, China’s GreenWatch in 2000, China, GreenWatch Zhenjiang Vietnam’s Environmental Information 1999 23 68 25 75 Disclosure System for Hanoi in 2002, 2000 14 77 15 85 10 and two programs in Uttar Pradesh, China, Greenwatch Hohhot India, in 2002 and 2004. 1999 43 13 77 23 After 10 years of collaborative work, 2000 21 35 38 62 39 the Bank has learned several lessons Vietnam, Hanoi about this kind of policy experimenta- 2001 45 5 90 10 tion and evaluation. First, counterpart 2002 38 12 76 24 14 8 World Bank ResearchDigest (continued from page 2) Recent Policy Research Working Papers World Bank research publications WPS3961 Global Redistribution of Income François Bourguignon Victoria Levin should visit http://econ.worldbank.org) David Rosenblatt Development research is a global WPS3962 Untangling the Maze of European Union Funds to Bulgaria Gallina Andronova public good, and the audience of this Vincelette publication—if it achieves its goal—is Iglika Vassileva potentially the whole development WPS3963 Financial Development in Latin America: Big Emerging Issues, Augusto de la Torre community. Given the mission of the Limited Policy Answers Juan Carlos Gozzi World Bank—to fight poverty and im- Sergio L. Schmukler prove the living standards of people WPS3964 Income Diversification in Zimbabwe:Welfare Implications from Lire Ersado in the developing world—its research Urban and Rural Areas activity covers practically every aspect WPS3965 Household Financial Assets in the Process of Development Patrick Honohan of development in its partner coun- WPS3966 On the Financial Sustainability of Earnings-Related Pension David A. Robalino tries. It also bears on issues that go Schemes with “Pay-As-You-Go” Financing and the Role of András Bodor beyond national borders, such as in- Government Indexed Bonds ternational trade, migration, HIV/AIDS, WPS3967 Can Information Campaigns Spark Local Participation and Abhijit Banerjee and climate change. Improve Outcomes? A Study of Primary Education in Uttar Rukmini Banerji Over time, we intend to reflect all Pradesh, India Esther Duflo this diversity of themes and geograph- Rachel Glennerster ic focus in the Research Digest. We hope Stuti Khemani that all those interested in the results WPS3968 Measuring Corruption in Eastern Europe and Central Asia: Stephen Knack of development research will find this A Critique of the Cross-Country Indicators new publication both easily accessible WPS3969 Deposit Insurance Design and Implementation: Policy Lessons Asli Demirgüç-Kunt and helpful. from Research and Practice Edward J. Kane Luc Laeven François Bourguignon is Senior Vice President and Papers can be downloaded at http://econ.worldbank.org Chief Economist of the World Bank. To download the World Bank Research E-Newsletter, go to Data & Research at http://www.worldbank.org The World Bank Research Digest is a quarterly publica- The Research Digest is financed by the Bank’s The Editorial Committee is Asli Demirgüç-Kunt, Jean- tion aimed at disseminating findings of World Bank Research Committee and managed by DECRS, the Jacques Dethier (managing editor), and Alan Gelb. Thi research. The views and interpretations in the articles research support unit of the Development Economics Trang Linh Phu provides research assistance and Evelyn are those of the authors and do not necessarily repre- Vice Presidency (DEC). The Research Digest is not Alfaro-Bloch produces the Digest. For information or sent the views of the World Bank, its Executive Direc- copyrighted and may be reproduced with appropriate free subscriptions, email researchdigest@worldbank. tors, or the countries they represent. source attribution. org or visit http://econ.worldbank.org/research_digest The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A Printed on Recycled Paper