WORLD BANK DISCUSSION PAPER NO. a-- Woik In progce. for public dl_ounC 'ne 1C 8 World Bank HIV/AIDS Interventions Ex-ante and Ex-post Evaluation Backgroundpaperfor L'O'O R O CONFRONTING 4- ~ ~ - v AI .3-Y '~~~~~~~~~~~I 4 Julia Dayton Recent World Bank Discussion Papers No. 321 Clearance and Settlement Systems for Securities: Critical Design Choices in Emerging Market Economies. Jeff Stehm No. 322 Selecting Development Projectsfor the World Bank. Jean Baneth No. 323 Evaluating Public Spending: A Frameworkfor Public Expenditure Reviews. Sanjay Pradhan No. 324 The Bangladesh Rural Advancement Committee's Credit Programs: Performance and Sustainability. Shahidur R. Khand- ker and Baqui Khalily No. 325 Institutional and Entrepreneurial Leadership in the Brazilian Science and Technology Sector:Setting a New Agenda. Edited by Lauritz Holm-Nielsen, Michael Crawford, and Alcyone Saliba No. 326 The East Asian Miracle and Information Technology: Strategic Management of Technological Learning. Nagy Hanna, Sandor Boyson, and Shakuntala Gunaratne No. 327 Agricultural Reform in Russia: A Viewfrom the Farm Level. Karen Brooks, Elmira Krylatykh, Zvi Lerman, Aleksandr Petrikov, and Vasilii Uzun No. 328 Insuring Sovereign Debt Against Default. David E Babbel No. 329 Managing Transboundary Stocks of Small Pelagic Fish: Problems and Options. Max Aguero and Exequiel Gonzalez No. 330 China: Issues and Options in Greenhouse Gas Emissions Control. Edited by Todd M. Johnson, Junfeng Li, Zhongxiao Jiang, and Robert P. Taylor No. 331 Case Studies in War-to-Peace Transition: The Demobilization and Reintegration of Ex-Combatants in Ethiopia, Namibia, and Uganda. Nat J. Colletta, Markus Kostner, Ingo Wiederhofer, with the assistance of Emilio Mondo, Taimi Sitari, and Tadesse A. Woldu No. 333 Participation in Practice: The Experience of the World Bank and Other Stakeholders. Edited by Jennifer Rietbergen-Mc- Cracken No. 334 Managing Price Risk in the Pakistan Wheat Market, Rashid Faruqee and Jonathan R. Coleman No. 335 Policy Options for Reform of Chinese State-Owned Enterprises. Edited by Harry G. Broadman No. 336 Targeted Credit Programs and Rural Poverty in Bangladesh. Shahidur Khandker and Osman H. Chowdhury No. 337 The Role of Family Planning and Targeted Credit Programs in Demographic Change in Bangladesh. Shahidur R. Khand- ker and M. Abdul Latif No. 338 Cost Sharing in the Social Sectors of Sub-Saharan Africa: Impact on the Poor. Arvil Van Adams and Teresa Hartnett No. 339 Public and Private Roles in Health: Theory and Financing Patterns. Philip Musgrove No. 340 Developing the Nonfarm Sector in Bangladesh: Lessonsfrom Other Asian Countries. Shahid Yusuf and Praveen Kumar No. 341 Beyond Privatization: The Second Wave of Telecommunications Reforms in Mexico. Bjorn Wellenius and Gregory Staple No. 342 Economic Integration and Trade Liberalization in Southern Africa: Is There a Rolefor South Africa? Merle Holden No. 343 Financing Private Infrastructure in Developing Countries. David Ferreira and Karman Khatami No. 344 Transport and the Village: Findingsfrom African Village-Level Travel and Transport Surveys and Related Studies. Ian Barwell No. 345 On the Road to EU Accession: Financial Sector Development in Central Europe. Michael S. Borish, Wei Ding, and Michel Noel No. 346 Structural Aspects of Manufacturing in Sub-Saharan Africa: Findingsfrom a Seven Country Enterprise Survey. Tyler Biggs and Pradeep Srivastava No. 347 Health Reform in Africa: Lessons from Sierra Leone. Bruce Siegel, David Peters, and Sheku Kamara No. 348 Did External Barriers Cause the Marginalization of Sub-Saharan Africa in World Trade? Azita Amjadi Ulrich Reincke, and Alexander J. Yeats No. 349 Surveillance of Agricultural Price and Trade Policy in Latin America during Major Policy Reforms. Alberto Valdes No. 350 Who Benefitsfrom Public Education Spending in Malawi: Resultsfrom the Recent Education Reform. Florencia Castro-Leal No. 351 From Universal Food Subsidies to a Self-Targeted Program: A Case Study in Tunisian Reform. Laura Tuck and Kathy Lindert No. 352 China's Urban Transport Development Strategy: Proceedings of a Symposium in Beijing, November 8-10, 1995. Edited by Stephen Stares and Liu Zhi No. 353 Telecommunications Policiesfor Sub-Saharan Africa. Mohammad A. Mustafa, Bruce Laidlaw, and Mark Brand No. 354 Saving across the World: Puzzles and Policies. Klaus Schmidt-Hebbel and Luis Serven (Continued on the inside back cover) WORLD BANK DISCUSSION PAPER NO. 389 World Bank HIV/AIDS Interventions Ex-ante and Ex-post Evaluation Julia Dayton The World Bank W2ashington, D.C Copyright (D 1998 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing June 1998 Discussion Papers present results of country analysis or research that are circulated to encourage discussion and comment within the development community. The typescript of this paper therefore has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to mem- bers of its Board of Executive Directors or the countries they represent. The World Bank does not guaran- tee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this vol- ume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Drive, Danvers, Massachusetts 01923, U.S.A. Cover credits: Confronting AIDS cover design, Edward Towles. Source of photos used in the design are as follows: Mother and child, Curt Carnemark/World Bank; condom box, Population Services International (PSI); Anti-AIDS Club signboard, Warren Parker/PSI; clinic worker and patient, UNAIDS/Yoshi Shimizu. ISSN: 0259-210X Julia Dayton is a Ph.D. candidate at Yale University. Library of Congress Cataloging-in-Publication Data Dayton, Julia M., 1967- World Bank HIV/AIDS interventions: ex-ante and ex-post evaluation / Julia M. Dayton. p. cm. - (World Bank discussion papers ; 389) Includes bibliographical references (p.). ISBN 0-8213-4251-7 1. AIDS (Disease)-Prevention. 2. AIDS (Disease)-Economic aspects. 3. World Bank. I. Title. II. Series. RA644.A25D39 1998 362.1'969792-dc2l 98-23816 CIP Contents Foreword iv Abstract v I. Introduction 1 II. Public economics and priorities for HIV prevention and treatment 2 Prioritizing among HIV/AIDS interventions 4 III. World Bank projects that support HIV/ AIDS interventions 5 World Bank projects focusing primarily on HIV/ AIDS 9 Providing information 10 Promoting safer behavior among those most likely to contract and spread HIV 10 HIV/ AIDS interventions supported by other Bank projects 12 IV. Ex-ante and ex-post project evaluation 13 V. Conclusions 21 Appendix A: Country assistance strategies and HIV/ AIDS 23 Appendix B: Analysis of HIV/AIDS in World Bank economic and sector work 26 Appendix C: World Bank projects with an HIV/ AIDS component 31 Appendix D: The link between economic and sector analysis and project success 33 References 35 Foreword The AIDS epidemic is one of the greatest development challenges of the 20th century. Since 1986 the World Bank has supported member country efforts to fight the epidemic in many ways, including lending, economic and sector analysis, grant support, technical assistance, and policy advice. This paper is the first comprehensive desk study of all World Bank HIV/AIDS- related activities from the perspective of public economics. It was prepared as a background paper for the recent World Bank Policy Research Report, Confronting AIDS: Public Priorities in a Global Epidemic. The paper is intended primarily for the use of World Bank staff, but we hope that it will also provide guidance to governments, other international agencies, and non- governmental agencies in the design and implementation of HIV prevention and mitigation programs. Learning from World Bank experience can help the Bank, together with its member countries, to confront the AIDS epidemic more effectively. iv Abstract This paper evaluates World Bank activities to prevent and mitigate the effects of AIDS in all regions of the developing world during fiscal years 1986 to 1996. It first prioritizes HIV- prevention and treatment interventions using the principles of public economics. Based on this framework, it assesses the appropriateness of Bank lending and grants, Bank Country Assistance Strategies, and country economic and sector work. While the Bank provided extensive assistance for efforts to collect information (surveillance and behavioral studies), support was less extensive for interventions that focus on reducing risky behaviors of those at highest risk for contracting and transmitting HIV. Few of the projects reviewed in the paper relied on strong economic analysis in ex-ante or ex-post evaluation. These findings suggest two primary challenges for the World Bank: (i) to focus its support for HIV prevention interventions that reach groups at highest risk to contract and spread HIV; and (ii) to improve the economic analysis used in preparing Bank HIV-related projects and in evaluating their effectiveness. v I. Introduction Between 1986 and 1996, the World Bank committed over US $ 550 million to HIV/AIDS prevention and mitigation efforts. This paper assesses the appropriateness of Bank interventions from the perspective of public economics and reviews the economic evaluation and implementation of projects. It focuses on 27 countries, selected either because AIDS is a particularly severe problem or because the Bank is especially active in the country (Table 1).1 This study follows three previous reports on HIV/AIDS interventions supported by the World Bank. The first set out an agenda for World Bank action in Africa (World Bank, 1988). The second and third reports also focused on Africa and updated the Bank strategy (Lamboray and Elmendorf 1992 and World Bank 1996). Lamboray and Elmendorf (1992), building on the 1988 strategy, concluded that additional World Bank efforts should focus on promoting behavioral change among core groups, treating other sexually transmitted diseases, and on convincing central financial and planning agencies in African governments to focus on AIDS and its implications for development. The third report, AIDS Prevention and Mitigation in Sub- Saharan Africa: A Strategy for Africa (World Bank 1996), identified five new tasks for the Bank: (i) help to generate political commitment to OAU declarations on AIDS; (ii) work more vigorously to change behavior; (iii) intensify national programs according to a typology of countries based on severity of prevalence levels; (iv) increase the analysis of AIDS and its impact on development goals in economic and sector work; and (v) improve the design and implementation of cost-effective approaches to mitigate the consequences of AIDS. Whereas past reviews have taken primarily a public health perspective, this paper evaluates the Bank's role from the perspective of public economics. It also reviews, for the first time, World Bank AIDS activities in all regions of the world. The paper focuses on World Bank lending, although reviews of HIV/AIDS coverage in Country Assistance Strategies and country economic and sector work are included in Appendices A and B, respectively. Section II discusses the public economics rationale for AIDS interventions. Section III reviews Bank lending for HIV/AIDS prevention and mitigation interventions. Section IV evaluates ex-ante and ex-post project evaluation. The final section draws conclusions. This paper reviews the same 12 African countries reviewed in World Bank (1996). 1 Table 1: Stages of the Epidemic for 27 Developing Countries Nascent Concentrated Generalized Less than S percent of high risk More than 5 percent of high risk 5 percent or more of women in groups infected groups infected, less than 5 urban antenatal clinics infected percent of women in urban antenatal clinics infected Indonesia Brazil Botswana Morocco China Burkina Faso Philippines Honduras Burundi Yemen India Central African Republic Mexico Congo Thailand Cote d'lvoire Vietnam Haiti Kenya Malawi Rwanda Stage of Epidemic Unknown: Tanzania Kyrgyz Republic Uganda Romania Zambia Zimbabwe Source: Statistical appendix, Table 2, Confronting AIDS: Public Priorities in a Global Epidemic II. Public economics and priorities for HIV prevention and treatment2 The principles of public economics provide a framework for making decisions about which activities merit government action and, among these, which should be priorities. From this perspective, there are two main reasons for government intervention: to maximize total social well-being - usually correcting a market failure and thereby enhancing efficiency - and to promote a more equitable distribution of well-being among social groups. Based on these principles, the recent World Bank Policy Research Report Confronting AIDS: Public Priorities in a Global Epidemic identified three activities in which governments have an indispensable role in ensuring the efficiency and equity of HIV prevention programs. First, governments have a role in providing public goods related to prevention, such as the collection and dissemination of information about the epidemic. Information about the HIV/AIDS epidemic and how to avoid infection is a true public good. This means that those who gain access to new information subtract none of its value from others. Although in some case it is possible to restrict access to information by, for example, selling it, the information is often passed on to people who did not purchase it. Because of this, private firms, on their own, 2 This section is based on Confronting AIDS: Public Priorities in a Global Epidemic, World Bank, 1996. 2 are likely to produce less information than is socially optimal and so there is reason for the. public sector to subsidize or provide public goods. This is particularly true of epidemiological surveillance of infection rates, for which the value lies in making the results well-known. But the public role for producing information also includes country-specific information on how to identify and reach people at the highest risk of becoming infected with HIV and spreading it to others. In addition, research that will improve the effectiveness of prevention interventions is also a public good. Second, governments need to support programs that reduce the negative externalities of risky behavior among people most likely to contract and spread HIV. An externality occurs when a transaction between two parties creates an unpriced effect on another party. In the context of HIV in developing countries, the largest externalities arise in the context of consensual sexual relations. Although the individuals involved can weigh their own individual risks, their activities still have consequences for their other (current and future) sexual partners. Ideally, those involved would consider these effects and decide to engage in safer sex. But the problem is that behavior -- safe or risky -- is difficult to prove. This means that individuals are likely to under- invest in activities that reduce the risk of infecting future partners, such as using condoms and treating other STDs. Because of the enormous positive externalities (of lowering HIV transmission) associated with these two activities, there is a clear role for governments to invest in them. Since HIV is transmitted sexually, the positive externalities of using condoms and being free of STDs are greatest for those individuals who have many sexual partners or needle- sharing partners. The more focused prevention interventions are on these groups, the more effective they will be in reducing HIV transmission. Another example of an externality is mother-to-child transmission. Although this is a clear example of an externality in that the infant cannot control his or her exposure to HIV, the role for developing governments in directly preventing mother-to-child transmission may be limited in the context of scarce resources. This is because the externalities created by mother-to- child transmission are not very large as compared with externalities created others (e.g. those with many sexual or needle-sharing partners). Although mother-to-child is a horrible tragedy, it does not result in many secondary infections. Preventing the mother from becoming infected by -IJIV in the first instance may be the best strategy for developing countries, and this can be done 3 most effectively with HIV prevention programs for those who have the largest number of partners. Third, the government has a role in promoting equity by ensuring that the most destitute are not denied access to the means to protect themselves from HIV. The government may want to subsidize clean blood for the poor, who cannot afford to purchase it. Safety net programs that help to reduce the vulnerability of poor households are especially important for those households that experience the death of a prime-age adult from AIDS. When women have low status, they are in a weak position for bargaining safer sexual relations. Thus, governments should also work to improve the status of women by expanding educational and employment opportunities and providing more legal protection. These policies merit support for economic and social reasons beyond the context of the AIDS epidemic, but the presence of the epidemic exacerbates the need for them. Prioritizing among HIV/AIDS interventions Using these principles of public economics and some general intuition about cost- effectiveness, it is possible to prioritize among types of interventions (Box 1). The exact mix of interventions will necessarily vary by country, depending on: (i) the stage and characteristics of the AIDS epidemic, which vary dramatically from country to country; (ii) the cost-effectiveness of particular interventions in that country; (iii) the level of resources available to confront HIV/AIDS, which usually depends on the level of economic development of the country; and (iv) the political support and implementation capacity in the country. 3 Three prevention priorities are particularly efficient and potentially very cost-effective: (i) condom promotion/subsidy for high risk groups; (ii) STD treatment for high risk groups; and (iii) financing operational research and surveillance. There are other important policies that should be supported for reasons other than AIDS. These include improving the status of women, enforcing laws against rape and exploitation of minors, and supporting anti-poverty programs that help reduce the vulnerability of poor household. Finally, broad-based mass-media information campaigns, subsidized HIV testing and financing clean blood are not generally not as efficient or cost-effective. 3Political support and implementation capacity in the country will not be discussed in this paper. 4 Box 1: Prioritizing Government HIV/AIDS Interventions Based on Public Economics PRIORITY INTERVENTIONS: * Finance operational research and disease surveillance * Treatment STDs, particularly for high risk groups * Promote/subsidize condoms for high risk groups . Improve status of women by expanding educational and employment opportunities and more legal protection' * Enforce laws against rape and exploitation of minors' * Support anti-poverty programs that help reduce the vulnerability of poor household to the loss of a prime age adult' OTHER (NON-PRIORITY) INTERVENTIONS * Broad-based mass-media information campaign * Subsidy for HIV testing for individuals * Provision of safe blood There are other social reasons for supporting these policies, but the AIDS epidemic increases their importance. III. World Bank projects that support HIV/AIDS interventions4 Between 1986,when it started lending for HIV/AIDS interventions, and 1996, the World Bank financed 60 projects in 41 countries, for a total commitment of US$ 552 million. This included nine free-standing AIDS projects and 52 projects with an AIDS component. Eight of the projects have been completed and about a dozen projects with an AIDS component were under of preparation as of June 1996. Since then, another freestanding AIDS project, the Argentina AIDS and STD Control Project (US$ 15 million) has been approved and at least two more are planned (for India and Guinea Bissau). All except one of the existing project are in the health or human resources sector and the implementing agency is almost always the Ministry of Health. The exceptions is the Uganda Transport Rehabilitation Project, which supports an AIDS awareness canpaign in Districts that participate in the rural feeder roads campaign. The Bank also provides support for HIV prevention and mitigation through various grant programs (Box 2). 4This review is based on the project descriptions in the Staff Appraisal Reports (SAR). Any additions and/or changes made to the project since then have not yet reflected in this discussion. 5 Box 2: World Bank Grant Support for HIV/AIDS Prevention and Mitigation The Bank finances grants for HIV prevention and mitigation and it also administers grant programs for other organizations. Most grant support is provided through four umbrella programs: the Special Grants Program, the Japanese Program for Human Resources Development, the NIARSH East Africa Initiative and the World Bank Small Grants Program. The Special Grants Program recently funded a $9 million dollar, three-year program to support UNAIDS and several regional HIV prevention initiatives. The main purpose of the grant is to support the global effort to prevent the transmission of HIV/AIDS. UNAIDS, which has been established to inspire, focus and strengthen this effort, is the executor of this 'bundle' grant to support four activities: * UNAIDS. The grant provides US $ I million to support efforts to coordinate HIV/AIDS policies, programs and funding. UNAIDS helps to establish a stronger link between AIDS and broader social, developmental and humans rights issues, in addition to providing policy, strategic and technical guidance. * The Western Africa HI VIAIDS Prevention Project. The project facilitates a coordinated response to the epidemic in three ways: (i) advocacy -- mobilizing political and opinion leaders throughout West Africa to address HIV/AIDS issues; (ii) pilot interventions - testing innovative ideas, particularly those relating to cross- border issues and commercial sex work; (iii) capacity strengthening-- enhancing capacity to identify, design, implement and monitor/evaluate inter-country and multi-sector projects. * The Latin American and Caribbean Regional Initiative for AIDS/STD Control. The overall objective of the grant is to mobilize and unify national and intemational efforts against HIV and other STDs. The project will be implemented over a period of three years and will finance (i) a series of regional studies on the epidemiology of HIV/AIDS, the impact of the epidemic on health service utilization and costs, and cost-effectiveness of alternatives, among other topics; (ii) a series of regional seminars and (iii) one regional conference. * The South-EastAsia HIV/AIDSProject. The project serves Cambodia, Laos, Malaysia, Myanmar, Philippines, Thailand, Viet Nam, and the Yunnan Province in China by assisting them to strengthen national and regional responses to HIV/AIDS. This includes promoting and facilitating analytical work, policy tools and intervention options, consensus building and institutional development. The Japanese Program for Human Resources Development (PHRD) sponsored a range of small research projects, mostly on the impact of AIDS, that help to guide the design of forthcoming Bank projects. The NIARSH Eastern Africa Initiative, financed by a group of donors, provides 1.3 million in funding for a range of prevention activities in Eastern Africa. The World Bank Small Grants Program has made five grants in support of AIDS related activities, for a total of US $ 56,000.' In four of the five grants, the funding helped to support an NGO or a network of NGOs to host a conference or workshop. These have included a workshop sponsored by Education International, an NGO based in Togo, on 'School Health and HIV/AIDS prevention' and another workshop organized by HelpAge Zimbabwe on the impact of AIDS on older people. A grant to an Indonesia NGO Yayansan Kusuma Buana (YKB) supported the publication and dissemination of IEC materials for distribution at STD service outlets, public health facilities, and by other NGOs. 'The five grants were: US$ 10,000 to HelpAge Zimbabwe for a workshop; US$ 6,000 to Swaziland Network of AIDS Service Organizations for a workshop; US $15,000 to Education International (Togolese NGO) for a workshop; US$ 14,000 to YayasanKusuma Buana (Indonesian NGO) for IEC materials; and US$ 11,000 to the Asia Center in Thailand for an AIDS prevention seminar in Bangkok. Source: Internal Bank grant proposal documents A regional breakdown of this lending shows that almost half of the lending, US$ 274 million was directed towards 42 projects in the Sub-Saharan Africa region (Figure 1). Lending for four projects in Latin America amounted to US$ 173 million, much of which was allocated for the large loan for Brazil. The Bank has supported seven loans, equivalent to US$ 131 million, for HIV/AIDS interventions in South and East Asia, including two free-standing loans 6 for national AIDS prevention programs in India and Indonesia. Bank financing of AIDS prevention and mitigation activities in other regions of the world was more limited. Three projects in each the Eastern and Central Europe and the Middle East and North Africa regions have supported AIDS interventions. Overall, most Bank lending for AIDS interventions took place between 1992 and 1994, when the Bank financed many of its large, free-standing AIDS projects, such as those for India (US$ 84 million) and Brazil (US $ 160 million) (Figure 2). Lending in the 1980s was disbursed in smaller amounts, usually as components of larger loans to countries in Africa. Figure 1: World Bank Lending for HIV/AIDS Interventions, by Region, 1986-96 300 250 o 200 150 100 50 0 _ ~ ~~e --_ , CZ I-~~~~~~~~8 Region Source: Author's calculations 7 Figure 2: World Bank Lending for HIV/AIDS Interventions by Fiscal Year, 1986-96 250 200 150 100 > 50 0 0000000 ON ON O NO NO Fiscal Year Source: Author's calculations One-third of the projects with HIV/AIDS components (21 project) and about the same proportion of lending (US$ 202 million) was for countries with a generalized epidemic (Table 2). Over half of the lending (US$ 291 million) was for the 26 projects in countries with a concentrated epidemic. Total lending for countries with generalized epidemics was less than lending to countries with concentrated epidemics, because the former are mostly in Sub-Saharan Africa where loan sizes are smaller on average. The remaining US$ 59 million was lent for 13 project in countries with a nascent epidemic or for which the stage of the epidemic is unknown. Table 2: Bank Lending for HIV/AIDS Interventions Total Lendingfor HI VIAIDS Components Stage of the Number of Projects Number of Countries (US $ millions) Epidemic Generalized 21 14 202 Concentrated 26 16 291 Nascent 8 6 36 Stage unknown 5 5 23 TOTAL 60 41 552 Source: Appendix C 8 World Bank projects focusing primarily on HIV/AIDS An in-depth review of the ten Bank HIV/AIDS-focused projects (including Argentina) evaluated the extent to which the Bank has supported interventions in three priority areas. These areas, which were identified in Confronting AIDS, include: 1. Information collection (including level and trends of HIV/other STDs, prevalence of high-risk behavior and characteristics of those who practice it, cost of interventions and their impact on HIV incidence) 2. Promoting safer behavior among those most likely to contract HIV (including lowering costs of condom use, lowering costs of safe injecting behavior, treating STDs, working with NGOs, and improving the status of women) 3. Promoting equitable access for the poor to prevention and treatment. Generally speaking, it was not possible to evaluate the exact allocation of funds to these areas, as project documents rarely disaggregated spending along the same categories. A rough approximation was possible for only two projects, Indonesia and Kenya (Box 3). Box 3: Spending Allocations for Priority Interventions for Two Bank Projects What share of Bank project funds support interventions in the three priority areas? This question is answered for the two HIV/AIDS-focused projects for which this infornation was available in the Staff Appraisal Reports. (Other Bank projects did not disaggregate spending along similar lines.) These projects are not necessarily representative of the Bank's entire lending portfolio; it would be helpful to collect this type of informnation on all Bank projects with HIV/AIDS components to indentify patterns in spending. The Indonesia HIV/AIDS and STDs Prevention and Management Project devoted about 40 percent of project base costs for interventions in the three priority areas. This included: 5 percent for targeted IEC, 23 percent for STD services, and 12 percent for surveillance. In addition, 27 percent of project base costs financed other types of IEC, including mass media programming and campaigns in the workplace, schools and universities and among health providers. Another 22 percent of the base cost financed the improvement of laboratory services. In the Kenya Sexually Transmitted Infections Project, 42 percent of total project costs (US$ 42.7 million) have been allocated to interventions in the three priority areas. This includes support for surveillance, operational research, monitoring and evaluation, condom promotion, treatment for STDs, and support for innovative NGO outreach activities to target high-risk groups. The rest of the project finances other activities such as mass media, programs to build national and district capacity, project management, activities to care for HIV/AIDS patients and efforts to improve occupational safety. Source: Staff Appraisal Reports One issue that complicates the evaluation of Bank loans is that they often provide financial support for interventions not already being financed from other sources. In some situations, therefore, the Bank does not support priority interventions because these have already received adequate support from government agencies or other donors. To take this into account, 9 this review also looked at the extent to which the project documentation discussed the role of other donors and organizations and efforts to collaborate with them. Providing information The Bank provided strong support for surveillance of HIV and other STDs. All 10 projects supported activities to improve national surveillance systems. The India project has financed sentinel surveillance among high-risk groups (sex workers, IDUs, truck drivers and TB patients). The Burkina Faso Population and AIDS Control Project is supporting special prevalence studies among miners, prisoners, migrants, TB patients, and pregnant women. The Argentina AIDS and STD Control Project plans to establish HIV and STD surveillance system with about 20 sentinel centers, although it is not clear whether these will focus on high-risk groups. Most of the projects supported behavior studies, but only two identified the specific need to focus on groups most likely to engage in high-risk behavior. Studies are planned on the behavioral risk factors of IDUs in Brazil and on the commercial sex industry in Indonesia. All projects support at least one study of cost-effectiveness or the economic impact of AIDS. Promoting safer behavior among those most likely to contract and spread HIV Seven of the 10 projects (those for Argentina, Burkina Faso, India, Indonesia, Kenya, Uganda, Democratic Republic of Congo) had some discussion of the need to target prevention services to the highest risk groups. Project documentation included limited discussion about how these groups were identified, how they would be reached, or what additional information was needed to reach them effectively. Eight of 10 projects planned to contract NGOs to provide outreach services to high-risk groups. The projects in Burkina Faso, Chad and Kenya have established separate funds to support innovative NGO activities. Four projects discussed efforts to target condom promotion to high-risk groups. In Brazil, the project supported an intervention to promote condoms among IDUs. The project in Burkina Faso subsidized condoms for those at highest risk. The project for Uganda also promoted condoms with an emphasis on expanding access for high risk groups. The project for Democratic Republic of Congo promoted condom distribution at non-traditional distribution 10 points such as bars, nightclubs, and hotels, as well as promoting condom high-risk groups, including sex workers. Students and employees were also identified by the SAR as high-risk groups, although no evidences was provided to support this claim. Only the Brazil project discussed specific efforts to lower costs of safe injecting behavior. This included outreach efforts to promote and distribute products for needle cleansing, experimentation with syringe distribution and the distribution of 2.5 million condoms to IDUs. We might have also expected this type of programmatic support in India or Indonesia. These two projects provided targeted IEC for IDUs but had no support for behavioral interventions to lower risk. All ten projects supported STD treatment services, but only the Kenya project documentation discussed specific efforts to target these services to those most at risk. There was a special emphasis on reaching those at increased risk: youth, men living away from home, truck drivers, and sex workers. One project supported efforts to improve the status of women. The Burkina Faso project provides support for enhancing public understanding of women's rights and problems, including financing seminars and intemational study tours. Promoting equitable access for the poor to prevention and treatment. The Argentina project will establish entities to provide medical care to those without insurance and for care for AIDS orphans and hospice care. Several of the projects support various types of treatment, but do not focus these efforts towards the most disadvantaged. Collaboration with other donors. The degree of collaboration with other donor organizations varied substantially across the projects. The projects for Burkina Faso, Chad, Uganda were jointly financed with others donors and the project documentation discussed a well-coordinated effort. The project for Indonesia, although financed solely by the World Bank, included an extensive discussion of the project would complements efforts of other organizations. In Argentina and Brazil, there was no discussion of the HIV/AIDS programs that already exist in country or of efforts to collaborate with other donors. Other project documents had very limited discussion of the roles of other donors, and it was therefore difficult to know if they supported priority interventions. 11 HIV/AIDS interventions supported by other Bank projects Among the 51 projects with a component for HIV/AIDS-related activities, most projects supported four types of interventions: (i) operational research, surveillance, and monitoring; (ii) information, education and communication (IEC) regarding HIV/AIDS awareness; (iii) treatment of STDs; and (iv) condom promotion (Appendix C). The provision of information (i) is always a priority, although operational research is most useful when it focuses on groups at highest risk for contracting and transmitting HIV. Interventions in the other three categories are appropriate when they reach groups at highest risk. Over half of the 51 projects (57 percent) funded operational research, surveillance, and/or monitoring and evaluation activities. Projects in countries with a concentrated epidemic were more likely to support operational research and surveillance (68 percent) than were projects in other countries. Operational research addressed topics such as measuring the cost effectiveness of various interventions, assessing the economic impact of AIDS and program monitoring and evaluation (see Box 3). Most of these projects also finance epidemiological surveillance of HIV, and some of them also support surveillance of other STDs. Almost half of the operations (43 percent) funded mass-media education for the population at-large. Fewer projects (about 33 percent) supported IEC targeted to high-risk groups, although these types of interventions are likely to be much more effective than mass-media campaigns in reducing HIV transmission. Half of the projects financed assistance for the treatment of STDs, including support for developing protocols for treating STDs, training health care providers on how to diagnose and treat STDs, providing drugs to treat the STDs, and paying the full costs of STD treatment. It is not clear the extent to which these interventions targeted groups at highest risk for infection, although expanding -.eatment for these groups is most important. About a third of the projects also support treatment for HIV/AIDS patients such as treating opportunistic infections (especially TB), providing drugs and counseling, designing protocols for treatment and training medical staff on appropriate treatment. The Uganda Sexually Transmitted Infections Prevention and Control Project, for example, supports all of these as well as financing community and home-based health care and social support for people with AIDS. 12 About half of the projects promote condom use and/or finance the provision of condoms. For example, the project in Guinea Bissau support the social marketing of condoms. The extent to which Bank support was focussed on the highest-risk groups was difficult to establish, but not enough project documents discussed which groups were at highest risk for sexual transmission and how to best reach them. Box 4: The Socioeconomic Impact of AIDS in Madagascar As of 1995, only 20 cases of AIDS had been reported in Madagascar. Yet, a combination of factors could cause the epidemic to explode in the near future. These include: high levels of STDs, extensive prostitution and promiscuity, low use of condoms, high rates of poverty, and an increase in the dislocation of the family and internal migration. Nevertheless, Madagascar has the opportunity to control the spread of HIV. In support of efforts to control the spread of AIDS, the Bank's Economic Management and Social Action Project sponsored operational research to estimate the potential growth of the epidemic and on the socio-economic impact of AIDS in Madagascar. Although only 20 cases of AIDS had been reported, it is estimated that at least an additional 130 cases are unreported. About 5,000 individuals are estimated to be infected with HIV. Two future scenarios were modeled: a controlled epidemic resembling that of Thailand and a rapid growth epidemic like that being experienced by Kenya. In the slow-growth scenario, HIV prevalence would reach 3 percent of the general population in 2015, and in the full-blown scenario, HIV prevalence would reach 15 percent by 2015. The direct costs of caring for AIDS patients in 2015 will be between US$ 10.75 and 52.75 millions (1996 dollars at 4000 FMG/dollar), depending on the severity of the epidemic. These figures are likely to be on the low side as they do not include the costs of treating opportunistic infections. The analysis considers the burden on public health expenditures and concludes that at the present rates of growth of the Ministry of Health budget, the sector will not be able to absorb these costs, particularly the high scenario. A range of indirect costs are also considered but not estimated. By 2015, there will be between 107,000 and 548,000 AIDS orphans. Source: Etude de l'impact socio-economic du VIHISIDA a Madagascar (Juin 1996) IV. Ex-ante and ex-post project evaluation Two recent studies have shown that good economic analysis leads to better projects. Analysis by Belli and Pritchett (1995) found that projects with poor economic analysis were 7 times more likely to perform poorly than were projects with good economic analysis. Belli (1996) repeated the analysis an additional (fourth) year of project performance rating and found that the relationship holds even stronger. In a separate analysis, Robert Schneider (World Bank 1995) measured the intensity of sector work in the three years preceding a project and found that the more sector work done, the higher the probability of success. This section assesses the quality of the economic analysis in project preparation, and Box 6 discusses the available evidence from ex-post evaluation. An analysis of the link between economic and sector analysis and project success is presented in Appendix D. 13 Devarajan, Squire and Suthiwart-Narueput (J 995) identified ways in which the World Bank can improve the economic analysis of its projects. Hammer (1996) applied their approach to the project analysis in the health sector. He proposed four aspects of economic analysis that should be covered in health project evaluation, which are outlined in Box 5. Box 5: Desiderata for Project Analysis in the Health Sector 1. Establish a finn justification for public involvement. This usually involves identifying market failures and discussing how and why public intervention is appropriate. 2. Examine the counterfactual. What would happen with and without the project? 3. Determine the fiscal effect of the project. In the health sector, this usually involves discussing the appropriate level of user fees. 4. Acknowledge the fungibility of project resources and examine the incentives facing public servants. Source: Hammer (1996) With a very few exceptions, only the free-standing AIDS-focused projects included economic analysis related to the AIDS interventions. For this reason, this section is limited to the nine completed or ongoing projects that address AIDS as their primary focus. Table 3 discusses the projects in terms of the desiderata identified by Hammer (1996). The first, rationale for public involvement, was almost never explicitly discussed, but rather implicit in the project document. An exception is the Brazil AIDS and STD Control Project, which cited the large negative externalities to the AIDS epidemic and the problems with imperfect information. Two forecasting models were utilized. The WHO epidemiological model developed by Chin and Sato was used in the analyses for the projects in Brazil, India and Democratic Republic of Congo (which only forecasted future AIDS incidence, without estimating the project impact). The InterAgency Working Group-AIDS model developed by the US Census Bureau was used for Indonesia. The assumptions listed in Table 7 indicate wide variation from project to project. The fact that no assumptions were discussed for several of the projects suggests that future project analysis should carefully document the assumptions under which the calculations were made, as they can dramatically affect the results. Only three of the nine projects identified a counterfactual for the project and presented cost-benefit or cost-effectiveness analysis comparing the various options. The project analyses 14 for Brazil, India and Indonesia provided extensive analysis of the alternatives. The Brazil project used the WHO epidemiological model to estimate the future incidence of HIV/AIDS in the presence and absence of the project. The analysis then calculated the direct and indirect costs saved with the project in place. Both the India and Indonesia project analyses included the 'without project' scenario and three possible 'with project' scenarios, which varied according to the effectiveness of the project interventions. For example, the three scenarios in the analysis for India were reducing the incidence of HIV/AIDS by 20, 30 or 40 percent. The 30 percent scenario (considered by the authors to be conservative) was then used to calculate the direct and indirect costs saved by the project. The analysis also calculates the internal rate of return on government direct costs, which was estimated to be 26.7 percent. None of the project analyses estimate the impact separately for each component of the projects. Among the other six projects, direct and indirect benefits were only estimated partially. For the projects in Burkina Faso and Democratic Republic of Congo, direct and indirect costs per AIDS case were calculated, but the number of cases averted as result of the project was not estimated.5 The analysis for Chad, Kenya, Uganda and Zimbabwe did the opposite: they estimated how many people would be reached by the project (although not how many AIDS cases would be averted), but did not estimate the indirect or direct costs saved by the project. With respect to the last two desiderata, only a few of the projects discuss the fiscal impact of the project, mostly raising cost recovery issues and estimating the impact of the project on recurrent costs. Problems related to fungibility were rarely discussed. 5 The project for Burkina Faso estimates how many couples will be using condoms and assumes 90 percent efficacy of condoms use. It then estimates the direct costs saved, assuming that all of these couples would become HIV-positive without wearing condoms. According to epidemiological trends in HIV/AIDS, this last assumption is completely unreasonable. 15 Table 3: Economic Analysis for Nine AIDS-focused Projects Project Rationale Analysis of Costs, Benefits and Counterfactuals Fiscal Fungibility for Public Impact Involvement Assumptions Model Costs Counter- Benefits Considered factual ? Direct Indirect Brazil AIDS and STD Control Imperfect Two sets of WHO IBRD and Yes, 2 300,000 lives saved by Indirect Incremental Four components: information; assumptions: Expanded govemment scenarios 2001, equaling $US 594 plus direct recurrent (i) prevention through IEC and extemalities I.(i) as of 1991, Program of costs million in direct treatment costs cost medical staff training; 425,000 people Immunization considered costs (present value in 1992) estimated at estimated at (ii) treatment services; currently model US$1.2 US$ 150.3 (iii) institutional development infected; (ii) billion million to build capacity to deal with widespread (present HIV/AIDS (mainly training transmission value in and upgrading of laboratories); began in 1980; 1992) (iv) surveillance, research and (iii) annual evaluation. incidence of HIV continues to increase until 1999. 2. (i) and (ii) - same as above, plus (iii) incidence increases only until 1995 and then decrease Burkina Faso Population and Not explicitly 1. a loss per None Bank, other 1992 Ambulatory treatment and Cost per AIDS Control discussed AIDS death of 43 donors, and scenario hospitalization cost $US 416 AIDS case Three components: yrs for men and govemment (pre- (1992) per case. = $US (i) family planning; 49 yrs per costs project) is 7,488 (ii) STD/HIV/AIDS prevention woman calculated the only Costs averted by project not (1992). and treatment; 2. incidence rate one accurately calculated. (iii) grants for private sector. of .173 per considered Project thousand benefits not 3. discount rate calculated. of 3 percent Chad Population and AIDS Not explicitly None given None IDA, other No Project would 'alleviate the None given Very Not Control discussed specified donor, burden' of HIV/AIDS on limited discussed Four components: govemment about 12,000 AIDS patients, incremental (i) reinforce implementation of and 21,000 families, 45,000 effect on natl population policy community orphans and the nation. IEC recurrent (ii) HIV/AIDS/STDs costs are and social marketing are costs prevention analyzed expected to produce (iii) social marketing of behavioral changes among condoms; 415,000 high risk people. (iv) promoting participation of Benefits of project not private sector and NGOs . valued in monetary terms. I Table 3 (contin ed): Economic Analysis for Nine AIDS-focused projects Project Rationale Analysis of Costs, Benefits and Counterfactuals Fiscal Fungibility for Public Impact Involvement Assumptions Model Costs Counter- Benefits Considered factual ?. Direct [Indirect India National AIDS Not explicitly 1. 1991 underlying WHO IDA, other Yes, By 2000, AIDS By 2000, Recurrent Control Project discussed prevalence of 200,000 HIV + Expanded donors and without incidence loss to costs in risk groups and 200,000 Program of government project and would be national implications Five components: people in the general Immunization costs three 'with 500,000/yr income are small but (i) promote public population Model considered project' without the avoided costs of awareness and 2. 8 percent discount rate scenarios - project and US$2.5-3.0 dealing with community support 3. each AIDS case is AIDS 100,000, billion AIDS (ii) improve blood associated with 30 related reduced by 150,000, or patients will safety illnesses, or which 20 will 20 percent, 200,000 in each likely (iii) build surveillance incur medical care at US$ 10 30 percent of the AIDS escalate and clinical per illness (1992 $), or 40 scenarios. management capacity increasing 8 percent/yr. percent as a (iv) Improve STD 4. every patient with AIDS result of the Direct cost clinical services will require 250 days of care, project savings at (v) strengthen of which 25 will require current prices management capacity hospital care at US$30/day for the 30 at national and state percent scenario levels would total US$388.4 million for the govt, and US$ 100.4 in private costs. Intemal rate of return on govt direct costs = ______________ ~ ~ ~ ~ ~ 126.7 percent_______ Indonesia HIV/AIDS Not explicitly 1. demographics based on InterAgency IBRD and Yes, 4 total 5000 fewer Total: Impact on Not discussed and STDs Prevention discussed 1980 and 1990 censuses Working government (baseline AIDS cases by US$140 govemment and Management 2. host/virus contact: Group AIDS costs plus three) 2005; 22,500 million by recurrent (i) HIV 'seed' was 1000 in model considered by 2010. 2005 and costs not 1990; US$551 evaluated (i) two pilot programs (ii) STD and circumcision Present value of million by for IEC and rates assumed direct treatment 2010 strengthening of STD 3. Sexual mixing patterns costs in Jakarta care based on census and US$6.3 for (ii) central activities to community surveys 2005 and improve STD delivery, 4. Mode of transmission: 95 US$19.2 for surveillance, laboratory percent sexual contact 2010 improvements, and IEC Table 3 (continued): Economic Analysis for Nine AIDS-focused projects Project Rationale for Analysis of Costs, Benefits and Counterfactuals Fiscal Fungibility Public Impact Involvement Assumptions Model Costs Counter- Benefits _ _ _ _ _ _ Considered factual _ _ Direct [Indirect Kenya STI Project Not explicitly None given None IDA, other No Evidence on the Total direct and Recurrent Risk that discussed presented donors and govt success in lowering indirect costs of cost broad Three components: costs are transmission of AIDS to the implications government (i) strengthen considered AIDS and cost- country could are estimated support may institutional effectiveness of reach 15 percent to be small not capacity treating STDs in a of GDP by the materialize (ii) prevention of core group in year 2000 (about and govt STIs Nairobi. $US 19,000 per capacity may (iii) address social AIDS case). be too weak to and economic Condoms will be implement consequences, distributed to 5 Indirect costs including drugs for million people. 1.7 saved from opportunistic million people will project not infections and receive care for calculated. support to NGOs. infections. Direct cost of project not ._________ _ calculated. Uganda STI Project Not explicitly None None IDA, other No 6650 TB patients None Recurrent Risk that discussed donors and govt will be treated; cost broad Three components: costs are Condoms will be implications government (i) prevent sexual considered distributed to 4 are estimated support may transmission of million people; 1.9 to be small not AIDS & other STDs million people with materialize (ii) mitigation of HIV will receive and govt personal impact of care for infections; capacity may AIDS 7,000 health be too weak to (iii) institutional workers will receive implement development protective supplies. Direct costs saved _____________ _______________ are not estim ated. __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Table 3 (continued): Economic Analysis for Nine AIDS-focused projects Project Rationale Analysis of Costs, Benefits and Counterfactuals Fiscal Fungibility for Public Impact Involvement Assumptions Model Costs Counter- Benefits Considered factual? . Direct Indirect Democratic Republic of Congo Not explicitly 1. current level WHO IDA, other No, Direct costs of AIDS Healthy life Cost Poor National AIDS Control discussed of infection of 6 model donors, scenario presented for each year 89- years saved per recovery managerial percent in urban govemment without 93 and for 2000 and 2010. case of assessed and capability is Four components: and I percent in and project is The average cost of each infection would project a project risk (i) IEC and condom distribution rural areas beneficiary the only AIDS patient (per year -- on be 6.2 and affordability (ii) integration of AIDS control 2. conversion costs one average patients die within based on compared activities rate of 4 considered considered the year after clinical annual incomes with other (iii) operational research percent/year manifestations ofthe in Democrac health (iv) institutional strengthening 3. disease) is US$ 229 in Republic of ent seroprevalence constant 1986 dollars. Congo, this is interventions stays at 2 per equivalent to 1000 Direct benefits of project not US$5,512 in 4. estimates of calculated. urban areas and the number of US$ 893 in treated cases rural areas. base on The average projected health would be about coverage. US$4,600, which is about 20 times higher that the direct costs of AIDS. Benefits of project not estimated. Zimbabwe STI Prevention and Care Not explicitly None given None IDA, other Only 65,000 TB patients will Indirect costs Cost discussed donor and current receive treatment; condoms are about 20 recovery Five components: govt costs costs will be distributed to times direct discussed (i) condom provision considered considered 700,000 people; 180,000 costs -- as (ii) STD treatment people with HIV will receive much as (iii) drugs for opportunistic infections care; 5,000 health care US$21,300 per (iv) supplies for HIV diagnosis and workers will have protective case. blood screening supplies; and 575,000 people (v) biomedical security supplies will be tested. Benefits of project not Direct benefits of project not estimated. ._________________ _________________ .____________ _. _______________ ______ _calculated. _. _ Source: Project StaffAppraisal Reports. Box 6: Lessons Learned from Completed Projects Eight projects with AIDS components have been completed, including one free-standing AIDS project (Democratic Republic of Congo). The four available completion reports indicate mixed results (see table below). Two of the four project outcomes (Democratic Republic of Congo and Cameroon) were rated as unsatisfactory by both the task manager (ICR in table) and the operations evaluation department (EVM in table). In both cases the project had only modest impact and was not considered to be sustainable in the future. Weak project management and limited government commitment were cited as reasons for poor performance in both projects. Most of the four project completion reports identified similar factors (not limited to the AIDS-related intervention) that contributed to poor project implementation, suggesting priorities for future projects: * A strong government commitment is key: The Niger project credited government commitment as the main reason for project success and the other three projects indicated that lack of commitment to the goals of the project hindered project implementation. * Simple project design is best: 'Simple' is broadly defined as supporting few interventions, that are easily implemented and working with a limited number of implementing agencies. Both the Gambia and Cameroon projects found that the complex design of several of their components hindered project implementation. * Strengthening management capacity is important: All four projects cited poor management capacity, especially financial management, as an impetus to effective project implementation. * Appraisal must be thorough: Weak project appraisal was cited as one of the main reasons for poor project implementation in Cameroon. The Democratic Republic of Congo project did not adequately consider the possibilities of rapid macroeconomics change, which led to severe project implementation problems. Additional project experience on AIDS intervention (World Bank 1996) provided the following additional lessons: . Interventions in the earlier stages of the epidemic have a greater impact and a higher benefit-cost ration that interventions at a later stage. * Large-scale condom promotion and marketing has resulted in large observed changes in sexual behavior and significant increases in condom use. * Efforts to control STDs should be integrated with AIDS prevention efforts. * NGOs can play an important role on reaching high risk groups. Project Ratings for 4 completed projects Project: Outcome Sustainabiity Institutional US$ Percent (rating source) Development Committed Canceled Cameroon SDA(ICR) unsatisfactory uncertain modest 21.5 million 81.7 (EVM) unsatisfactory unlikely negligible Gambia Hlth Dev (ICR) 4.7 million 6.9 Niger I st Health (ICR) satisfactory likely partial 25.1 million 1.3 Democratic Republic of Congo AIDS (ICR) unsatisfactory unlikely negligible 8.1 million 60.9 Notes: ICR= implementation completion report (prepared by task manager); PCR=project completion report (prepared by task manager); EVM=evaluative memorandum (prepared by OED). The possible ratings are: (i) outcome: highly satisfactory, satisfactory, unsatisfactory, highly unsatisfactory; (ii) sustainability: likely, unlikely, uncertain; (iii) institutional development: substantial, partial, and negligible. Source: World Bank project completion reports 20 V. Conclusions Between 1986 and 1996, over US $ 550 million in World Bank loans has supported interventions specifically aimed at preventing HIV/AIDS and mitigating its effects. This paper reviews Bank activities in all regions of the world from a public economics perspective. The paper focuses on the Bank's lending program, although the conclusion presented here also draw on review of the Bank Country Assistance Strategies and country economic and sector work (Appendices A and B, respectively). The recent World Bank Policy Research Report, Confronting AIDS: Public Priorities in a Global Epidemic, identified three priority areas for government support: (i) information collection; (ii) promotion of safer behavior among those most likely to contract and transmit HIV; and (iii) and protection for the poorest groups in society from contracting HIV and support to mitigate the negative consequences of HIV/AIDS among these groups. Bank projects are evaluated based on the extent to which they supported interventions in these areas. The Bank provided extensive assistance for efforts to collect information. All ten of the HIV/AIDS-focused projects financed surveillance of HIV and other STDs. Most of them also supported behavior studies, and all ten had plans for at least one cost-effectiveness study. Only a few, however, indicated a plan to focus these activities on the population groups most at risk in the country. About 60 percent of the 51 projects with an HIV/AIDS component supported operational research, surveillance, evaluation and monitoring activities, although the focus on groups at highest risk was limited. Support for interventions to reduce risky behavior among those most likely to contract and transmit HIV was less extensive. Most of the ten HIV/AIDS-focused projects contracted NGOs to provide outreach services to high risk groups, but there was very limited discussion of how these groups had been identified or what additional information was needed to reach them effectively. Four projects supported condom promotion programs targeted to high-risk groups. Only the Brazil project provided specific support for interventions to support safe injecting behavior. All ten projects financed STD treatment, but only one planned to focus these services on those most at risk. Finally, only one project provided support for efforts to protect the poorest groups in society from contracting HIV or programs to mitigate the negative consequences of HIV/AIDS among these groups. 21 Most of the projects were not based on strong economic analysis. With a very few exceptions, only the free-standing AIDS-focused projects included any ex-ante economic analysis of the proposed AIDS interventions. Only one-third of these projects, however, prepared adequate cost-benefit or cost-effectiveness analysis. As of October, 1996, ex-post evaluations were only available for four of the eight completed projects and their assessment was generally unsatisfactory. Country Assistance Strategies and sector analyses, have only addressed HIV/AIDS issues in limited manner. Consideration of the AIDS epidemic was not central in most of the Country Assistance Strategies, the principle strategic planning documents, although coverage was most extensive in countries with generalized epidemics. The best documents discussed how the development strategy would affect the HIV/AIDS epidemic, or vice-versa, and what the government was doing about this problem. Most of the sector analysis of HIV/AIDS was in health sector reports, and very few reports estimated the economic effects of AIDS. The coverage and quality varied, and few reports provided estimates of the potential public (and/or private) costs of AIDS for the health sector or the cost-effectiveness of the various health interventions. Analysis of the broader consequences of AIDS such as its relationship with poverty, and its effects on other sectors and the macroeconomy was limited. This reviews the extensive support provided by the World Bank for a wide-range of HIV/AIDS interventions. It also identifies a weakness in the Bank's lending program: many of the activities supported by the Bank have not been well-focused on the groups in the population most at risk for HIV infection. Few of the HIV/AIDS Bank projects based support for interventions explicitly on the principles of public economics or have relied on sound economic analysis in ex-ante or ex-post evaluation. Thus, the two primary challenges for the Bank are to provide more support for prevention activities that focus on the groups at highest risk for contracting and spreading HIV and to improve the ex-ante and ex-post economic evaluation of HIV/AIDS projects. 22 Appendix A: Country assistance strategies and HIV/AIDS The Country Assistance Strategy (CAS) is the central planning document guiding World Bank activities in each borrowing country. It is prepared by World Bank staff and consists of a short, concise discussion of the major developmental challenges facing the country, the Government's economic development plan, and the World Bank's strategy to support government efforts. It includes a description of the current and future lending program and the plan for analytical work. How do the CASs incorporate concerns about the AIDS epidemic and what sorts of strategies do they recommend for dealing with the epidemic? (Box Al provides general guidelines for incorporating AIDS issues in CASs.) A review of the CASs for 25 countries6 found that coping with AIDS was not central to any of the country strategies. The AIDS epidemic was not discussed in 11 of the 25 CASs, including those for Haiti, where the incidence of AIDS is relatively high, and in Viet Nam, where the disease is spreading quickly (Table Al). An additional seven CASs briefly mentioned AIDS as a health issue, but did not describe the problem or how to address it in any detail. The remaining seven CASs included a more extensive discussion of AIDS issues. Box Al: When Should a CAS Discuss AIDS? At a minimum all CASs for countries with a mature or developing AIDS epidemic should consider AIDS issues in the CAS. First, the CAS should discuss whether the development strategy proposed by the country and the Bank has either (a) the potential to exacerbate the epidemic or conversely, (b) whether the presence of the epidemic could hamper the achievement of the development objectives. To the extent that there are such effects, the document should say what the government is doing to address them. Second, the Bank's overarching objective is to help countries to reduce poverty. In countries with a severe epidemic, the epidemic is likely to impoverish many and push the already-poor deeper into poverty. When this is the case, the CAS should include some discussion of the implications of the AIDS epidemic on the poverty reduction strategy in the country. How do AIDS survivors fit into the current safety nets? Are they only candidates if otherwise poor (usually the best policy)? If there is no safety net, is the country really too poor to afford one? Is the government doing all it can to promote, rather than hamper, private assistance efforts by NGOs and/or communities? The degree to which AIDS issues were addressed in the CAS varied by stage of the epidemic. Those countries with a high HIV prevalence (over 5 percent of the general adult population, according to WHO estimates) included a more in-depth discussion of the AIDS 6 There is no CAS for Botswana or Central African Republic. 23 epidemic. They discussed the incidence or prevalence of the disease, in some cases indicating changes over time and the main modes of transmission. For example, the CAS for Malawi asserted that AIDS is spreading in epidemic proportions, with HIV infection in the 15-19 age group estimated at about 20 percent in urban and 8 percent in rural areas. Life expectancy in Malawi has declined from 46 years in 1980 to 44 in 1993 as a result of AIDS mortality. Two of the CASs, those for Tanzania and Uganda, also considered problems with the spread of other STDs. Table Al: Selected HIV/AIDS Issues covered in 25 Country Assistance Strategies Describes Incidencel Recommends Recommends Recommends Targeted Prevalence or Means Analysis of STD/AIDS Programs for Vulnerable of Transmission of HI V/A IDS Prevention (9) Groups (e.g. aidfor widows Country HI VAIDS or Care (+) & orphans) Generalized Epidemic Burkina Faso V / / Burundi V/ Congo, / Democratic Republic of Cote d'Ivoire Haiti Kenya Malawi V , Rwanda V Tanzania V Uganda V/ / + Zambia V Zimbabwe Concentrated Epidemic Brazil China Honduras V India Thailand V Vietnam Nascent Epidemic (or *for stage not known) Indonesia Kyrgyz Republic* Mexico Morocco Philippines Romania* Yemen Total: 25 7 5 5 1 Note: There is no CASs for Botswana or Central African Republic. Source: Author's review 24 Few of the documents communicated a country strategy for dealing with HIV/AIDS. Only ten CASs discussed any actions taken by the Government or the Bank to fight the HIV/AIDS epidemic, and all of these were for countries with either high or medium prevalence rates. These CASs included five countries that proposed STD and/or AIDS prevention or care, five that reported plans for analysis of HIV/AIDS and six that mentioned other HIV/AIDS interventions in a very general way. The Burundi CAS, after identifying the 'looming threat of AIDS' as a key constraint to economic development, mentioned that the Government is supporting education campaigns and the promotion of condoms to help slow the spread of STDs and HIV/AIDS. It did not, however, detail the Bank's support for these efforts except to mention that an AIDS rapid assessment survey was on-going. Research from the Kegara region of Tanzania indicated that the poor and other vulnerable groups are the most severely affected by a death from AIDS of an adult in the household. These findings suggested that targeted programs for the poor are likely to be effective in mitigating the affects of AIDS (Over et al, forthcoming). This suggests the need for strong links between targeted poverty programs and AIDS mitigation efforts. The CASs for Burundi and Uganda were the only ones to make this link when they identify plans for specific targeted programs. The Uganda CAS, which indicated the need for specific targeted programs for vulnerable groups such as AIDS patients and AIDS widows and orphans, describes how two Bank projects will support the areas in Uganda most devastated by the AIDS epidemic. Educating girls and other interventions to raise the status of women are additional ways in which governments can help to slow the spread of AIDS. About half of the 25 CASs discuss the poor status of women and/or the low educational enrollment rates of girls. Almost all of these CASs also outline measures supported by the Bank to improve the status of women or expand education for girls. Yet none of them made the link between support for these measures and potentially lowering the transmission of HIV/AIDS. There are several other issues related to HIV/AIDS that one might have expected to see discussed in the CASs. The financial impact of the epidemic on the health sector and recommended policies to mitigate it has traditionally been one of the first policy considerations. Only two of the CASs (C6te d'Ivoire and Tanzania) alluded to this issue generally, but neither provided actual estimates of the cost of the impact. The effects on labor force productivity or on macroeconomic indicators could have been covered in some of the countries with high HIV 25 prevalence, but they were not. The relationship between the AIDS epidemic and family planning activities was not discussed in any of the CASs. Appendix B: Analysis of HIV/AIDS in World Bank economic and sector work This section discusses the coverage of AIDS in World Bank economic and sector work from 1990 to the present for the 27 selected countries. A previous review (World Bank 1996) of the sector work completed since 1991 for 12 African countries found that consideration of AIDS was greatest in poverty assessments, policy framework papers, and country assistance strategies, and least in public expenditure reviews and country economic memorandums. As a follow-up to that study, this paper reviews the following for each country (when available): country economic memorandum (CEM), poverty assessment, public expenditure review (PER), health sector reviews and gender analysis. When the analysis for each country is considered as a whole, the economic and sector work for countries at an advanced stage of the AIDS epidemic included more extensive analysis of AIDS issues than did the sector work for countries where AIDS is less widespread (Table B 1). The sector work for four (Malawi, Tanzania, Uganda, and Zambia) of the countries with a mature AIDS epidemic included economic analysis of the epidemic, in addition to a summary of trends in incidence and/or prevalence. In particular, the CEMs and poverty assessments for these countries considered the impact of AIDS on the economy at large, or on efforts to reduce poverty. Consideration of the effects of the epidemic on the macroeconomy has been most comprehensive for Tanzania, Zambia, and Uganda, and the main findings of the analysis are presented in Box B 1. Overall, however, the inclusion of economic analysis of the AIDS epidemic was very limited in the reports reviewed and, with the above exceptions, AIDS was almost always considered in the more narrow context of a health sector issue. Among the report types, the health sector analysis and joint CEM/Poverty assessment documents contained the most extensive economic analysis of AIDS. The most comprehensive analysis in the joint CEM/poverty assessments was highlighted above. Three-fourths of the health/social sector analyses reviewed included substantial coverage of AIDS issues. The most in-depth analysis is presented in the Tanzania AIDS Assessment and Planning Study, which includes analysis of incidence and prevalence of the epidemic, the likely demographic impact, 26 and economic analysis of the costs of the epidemic and possible sectoral and macroeconomic effects.7 The Malawi Population Sector Study contains extensive modeling of the potential demographic impacts of HIV/AIDS, considering several scenarios depending on the severity of the AIDS epidemic. Discussion of AIDS was generally absent among the stand-alone CEMs, with the exception of the reports for C6te d'Ivoire, Indonesia and Zambia. The Cote d'Ivoire CEM draws on analysis done for other countries to predict how the epidemic is likely to affect the Ivoirian economy. Based on this analysis, the report lays out priority activities that should be part of a comprehensive government strategy. The Indonesia CEM highlighted the fact that STDs are becoming a more serious health problem and considers the per capita costs of several basic health care interventions, including those for TB and STD care and AIDS prevention. The Zambia CEM considered various ways in which the AIDS epidemic threatens to undermine prospects for economic growth, as discussed in Box B 1. Among the poverty assessments, analysis of AIDS issues varied. The reports for Cote d'Ivoire, Kenya and Malawi focused on how AIDS increased poverty by imposing additional costs on an already impoverished society. The report for Zimbabwe asserted that the AIDS crisis is exacerbated by the poverty situation. Poor men often go to urban areas to seek work, leaving the wife and family at home, and are at higher risk of sexual encounters outside of marriage and of infecting their family. In addition, the burden of caring for AIDS patients is often left to women in rural areas. The report suggests that a systematic and vigorous IEC campaign, along with affordable and easy access to condoms are needed to reduce transmission of AIDS. 7 Eight more AIDS assessments are at various stages of preparation. 27 Table B1: AIDS Analysis in World Bank Economic and Sector Work, 1990-96 Poverty Health Sector Country CEM Assessment PER Work Generalized Epidemic Botswana Burkina Faso Burundi ++R* Central African Republic ++R Congo, Democratic Republic C6te d'lvoire ++R ++R - Haiti Kenya ++ Malawi ++ - ++R Rwanda ++R* Tanzania ++R* +R +++R Uganda ħħħR* + ħħ+R Zambia ... ++ħR R Zimbabwe ++R* + Concentrated Epidemic Brazil China Honduras + India - -++R Mexico Thailand Vietnam + Nascent Epidemic or Unknown Stage Indonesia ++- +R + Kyrgyz Republic Morocco Philippines - +R Romania Yemen * Joint PER/Poverty Assessment. Legend: A blank indicates that there was no recent report to review - no discussion of HIV/AIDS + a brief mention of HIV/AIDS, but no in-depth discussion ++ incidence and/or prevalence rates discussed +++ incidence and/or prevalence rates discussed , plus economic analysis, including cost analysis and/or macro/sectoral analysis R policy recommendations given Source: Author's review 28 This review confirms the finding in the World Bank report (1996) that PERs generally included very little analysis of the AIDS epidemic and its potential impact on the public budget or civil service. A few of the PERs did, however, provide a discussion of the public economics rationale for government intervention in the health sector, often finding a role for the government in fighting AIDS. For example, the Indonesia PER outlines three economic rationales for public involvement in the health sector: public good, equity and market failures. It cites communicable disease control and epidemiological surveillance as two types of public goods that should be publicly supported. Treating STDs, including AIDS, should be publicly supported because of their equity implications, as the report asserts (without evidence) that the poor are more likely to get these diseases. The PER for the Philippines asks the question: What can national government do that will not be done by the people themselves or by the local government? The main candidates are infectious disease control and health education campaigns of various types (smoking, HIV prevention, and nutrition), where many of the benefits of the activity will accrue to people outside the jurisdiction in which the health expenditures are borne. Local government may underprovide these policies, either because effects directly transcend political boundaries or because migration spreads the effects of interventions acrossjurisdictions. None of the PERs reviewed, however, took the next logical step of analyzing costs of the alternative public interventions to fight AIDS. 29 Box Bi: Analysis of the Impact of AIDS on Growth and Macroeconomic Projections Tanzania. The rising prevalence of AIDS can be expected to affect the macroeconomy through a number of channels, which can be grouped into two broad categories. First, declines in 'healthiness' have four effects: reduced labor productivity; increased health care expenditures; reduced savings; and reduced human capital investments. Second, rising mortality rates, particularly among children and sexually active adults, reduce the population growth rate and change the age structure. A modeling exercise that took into account these effects, estimated that the presence of AIDS reduced the average real GDP growth rate in the 1985-2010 period by between 15 and 28 percent, from 4.0 percent per annum to 2.9-3.4 percent per year (depending on the productivity and savings parameters chosen). Over a 25 year period this decreases potential output by between (1980) Tsh 15 billion to Tsh 25 billion. The impact on growth of potential per capita GDP is more moderate, decreasing it by only 12 percent in the worst case scenario. Under the simulation, per capita GDP growth is forecast to grow an average annual rate of 0.7 percent in the hypothetical situation without AIDS, while with AIDS growth rates range between 0.3 and 0.8 percent per annum. (Tanzania AIDS Assessment and Planning Study. World Bank, 1992) Zambia. It is not easy to separate the impact of AIDS from the impact of other factors on such important variables as fertility rates and birth control practices, but recent research suggests that the impact of AIDS on population growth rates will be larger than earlier estimated. From an aggregate level of well over 3 percent in the 1980s, the current population growth rate is estimated at 2.8 percent. Estimates from various sources for the rate of population growth in 2005 vary from 1.7 to 2.3 percent. It is even more difficult to assess the impact on GDP projections. In some areas. such as cotton. it is not likely to have much direct effect, but for domestic food crops, such as maize. one would expect a proportionate reduction. Looking at the economy as a whole, the impact of AIDS on total output depends on the extent to which the impact is disproportionately on the more highly skilled workers. the degree of incapacity to work of AIDS patients, and the substitutability of unemployed workers for AIDS victims. One estimate using cross sectional production functions suggests that with prevalence rates of 20 percent. a twvo to one ratio of skilled incidence to unskilled incidence, and assuming that 50 percent of the direct AIDS costs come out of direct savings, the impact on per capita output is a negative 0.2 percent per year. In other words. if using these assumptions AIDS decreases annual population growth rates by one percentage point. AIDS would decrease output growth rates by 1.2 percentage points. (Zambia Countrv Economic MemoraniduLmt. World Bank, 1996) Ugartda: Prospects for accelerated growth are likely to be adversely affected by the AIDS epidemic. WHO projects that the number of HIV-infected people could increase to more that 1.9 percenit bx 199S. In 1993-98, 565,000 adults and 250,000 children will die from AIDS. HIV prevalence is found to be hialier in urban areas and among the more educated occupational groups in Uganda. Instead of the labor force growing from 400,000 to 1. I million by 2010, it is estimiated to reach only 740.000 because of deaths anion- workers between the ages of 35 and 50. Studies on the economic impact of AIDS are less conclusive. Nonetlheless, private savings and investment, labor supply, and henlce growth are likely' to be adversely affected by the epidemic. In the hatd- hit areas, less land is under cultivation and farmers have shifted to less-labor intensive crops (n1o specific evidence is cited to support these assertions). Interest in long-term planninig and investmllenlts has decreased. Health expenditures per capita are likely to go Up, but as riesources are limited, the time required to care for the sick may further reduce labor input and henice growth. (Uganda: The Challenoe of Grow th and Pov ert! Reduction, World Bank. 1996) -: Appendix C: World Bank Projects with an HIV/AIDS Component PRIORITYINTERVENTIONS OTHER INTERVENTIONS (when targeted to high-risk groups) Country Project Condom STD Targeted Operational Mass- Treatment Blood Safety HIV Testing Total Amount Promotion Treatment IEC Research, media of AIDS (equipment and Bank for Surveillance, IEC and supplies Counseling HI VIAIDS Evaluation, Operational and training) and Monitoring Infections (US S millions) Generalized Epidemic Benin Health Services 1 18.60 0.60 Benin Health and Population 1 I 1 I ? 27.80 1.20 Burundi Population and Health I I I 1 14.00 1.20 C6te d'lvoire Integrated Health Services I I 1 I 1 40.00 3.00 Guinea-Bissau Social and Infrastructure I 1 1 5.00 0.05 Guinea-Bissau Social Sector 1 1 8.80 1.00 Haiti First Health I 1 28.20 1.70 Kenya Third Population 1 12.20 7.00 Kenya Fourth Population 1 35.00 5.00 Kenya.. Health Rehabilitation I I 1 31.00 1.00 Lesotho Second PHN I I 1 1 1 1 12.10 0.90 Malawi PHN 1 1 1 55.50 0.23 Rwanda First Population I ? 19.60 2.40 Tanzania Health and Nutrition 1 47.60 5.00 Uganda First Health I I 1 1 1 52.50 1.66 Uganda Transport 1 75.00 0.50 Zambia Health Sector I 1 1 1 56.00 tbd Sub-total 3 8 8 9 8 7 4 3 538.90 32.44 Share 18 % 47 % 47 % 53 % 47 % 41% 24 % 18% Concentrated epidenmc Angola Health I 1 19.90 0.75 Brazil NE Endemic Disease I I I 1 109.00 6.60 Cameroon SDA I I I I 1 21.50 040 Cameroon Health, Fertility, and Nutrition I 1 1 43.00 0.40 Chad Health and Safe Motherhood 1 1 1 18.50 0.10 China Infectious and Endemic Disease I 1 90.00 0.48 China Comprehensive Maternal and Child I 100.00 0.40 China Disease Prevention I I 1 1 1 129.60 0.48 Gambia National Health Development I I 1 5.60 0.10 Guinea Health Services I 1 19.70 1.00 Guinea Health and Nutrition I I 1 24.60 tbd Honduras Nutrition and Health I I I I I 1 25.00 0.67 Malaysia Health Development I 1 50.00 16.00 Mali 2nd Pop and Rural Water 1 26.60 1.40 Niger flealth I I I ? ? 27.80 0.35 Niger Population I I 1 17.60 0.60 Niger Health 11 I 1 1 1 1 1 40.00 1.70 Nigeria National Population 1 1 1 78.50 tbd Nigeria Imo Health and Population I I I 1 27.60 1.00 Nigeria Health System I 1 70.00 tbd Senegal Human Resources Development I I 1 35.00 0.60 Togo Population and Health I I I I I 1 14.20 1.00 Sub-total 14 12 6 15 11 6 5 3 993.70 34.03 Share 64 % 55 % 27 % 68 % 50% 27% 23% 14% Nascent Epidemic Madagascar Economic Management and 1 22.10 0.43 Social Action Madagascar Health Sector Improvement 1 1 1 1 31.00 1.60 Mauritania Health and Population I 1 15.70 0.20 Morocco Social Priorities-Basic Health I I I 1 68.00 tbd Morocco Second Health 104.00 8.00 t'.. Papua New Pop and Family Planning 1 I I 1? 6.90 0.66 Guinea Yemen Family Health 1? 1? 1? 1? 36.60 0.25 Sub-total 3 3 2 3 2 1 0 0 284.30 11.14 Share 43% 43% 29% 43% 29% 14% 0% 0% Stage of epidenic unknown Comoros Population and Human Resources I I 1 13.00 0.29 Equatorial Guinea Health Improvement I 1 5.50 0.23 Kyrgyz Republic Health Sector Reform I 1 18.50 0.50 Macedonia, FYR Health Sector I? 1 16.90 0.77 Romania Health Services I 1 150.00 21.50 Sub-total 1 2 1 2 1 1 1 2 203.90 23.29 Share 20% 40% 20% 40% 20% 20% 20% 40% Grand Total 21 25 17 29 22 15 10 8 2020.80 100.90 Share 41% 49% 33% 57% 43% 29% 20% 16% Notes: A question mark (?) indicates the SAR was ambiguous regarding this intervention; tbd = to be determined. Source: Author's review Appendix D: The link between economic and sector analysis and project success The relationship between the amount of sector work prepared prior to the project and project success was investigated following a method similar to the one used in World Bank (1995). 'Project success' was measured using the average project implementation rating from project supervision reports (as reported in the MIS), evaluated on a four-scale criteria: high satisfactory (1), satisfactory (2), unsatisfactory (3), and unsatisfactory (4). The average project implementation rating was 'satisfactory ' (2) and there was very little variation by region (Table D-1). The amount of high quality sector work was measured by the number of sector reports completed in the three years preceding the project, as documented in the World Bank Reports database.8 Another measure of the amount of sector work is the number of staff weeks dedicated to sector analysis during the three years preceding project approval, although this measure does not take into account the quality of the analysis.9 Projects in Sub-Saharan Africa had the fewest average number of completed health sector reports per project (0.5 per project). In contrast, projects in South Asia were most likely to have more completed sector work - an average of almost 1 completed report per project-although the countries in this region seemed to have spent more staff weeks preparing sector work. Neither of these variables was significant in predicting project success, nor were they significantly related to the measure of project success.10 This may be due to the small sample size (53 projects were include in the analysis) or the limited variation in project implementation ratings (almost all project received a '2'). 8 The assumption is that reports must be of a sufficiently high quality to be recorded in this database. The source for this information is the MIS. 10 Regressing the project implementation rating on the number of reports completed and, separately, the number of staff weeks both yielded statistically insignificant results (N=53). The Pearson's correlation between project success and number of reports completed was 0.07 (, p=-.75) and between project success and staff weeks of sector work was 0.08, (p=0.56). The correlation between staff weeks spent health sector analysis and reports completed was also insignificant (Pearson's correlation 0.19, p=. 15), suggesting that these two indicators are not measuring the same thing and are not good proxies for each other. 33 Table D-1: Implementation Ratings and Previous Health Sector Work for Projects with AIDS components, by Region Number of Completed Staff Weeks Spent on Average Health Sector Reports Health Sector Work in Region Implementation Rating' in Previous 3 Fiscal Previous 3 Fiscal Years Years Sub-Saharan Africa 2.0 0.5 44.4 South and East Asia 2.1 0.9 212.5 Latin America 2.0 0.8 33.0 Other regions 1.8 0.6 30.9 Average of all projects 2.0 0.6 45.8 ' The average of all supervision implementation ratings as of 10/31/96. Scale: I=highly satisfactory; 2=satisfactory; 3=unsatisfactory; 4=highly unsatisfactory. Source: World Bank MIS reports The nine projects that focused primarily on AIDS prevention and mitigation tend to perform the same as the overall average (1.9 compared with 2.0) (Table D-2). A higher average number of sector reports was completed (0.9 per project) and the average staff weeks spent on health sector work during the previous three years was over twice as high as the average for all projects - 111.6 compared with 45.8. Based on previous analysis (World Bank 1995), we would expect these projects to have a higher probability of success than the other AIDS related projects. Table D-2: Implementation Ratings and Sector Work Preceding Project for the Nine Free-standing AIDS Projects Number of Staff Weeks Average Completed Spent on Health Implementation Health Sector Sector Work in Project Rating' Reports in Previous 3 Fiscal Previous 3 Fiscal Years Years Brazil AIDS and STD Control 2 2 54.3 Burkina Faso Population and AIDS Control 2 0 10.9 Chad Population and AIDS Control I 1 45.4 India National AIDS Control 2 2 212.5 Indonesia HIV/AIDS and STD 2 1 Kenya STI 2 0 195.8 Uganda STI 2 0 205.3 Democratic Republic of Congo National AIDS 2 0 76.7 Control Zimbabwe STI Prevention and Care 2 2 91.6 Average 1.9 0.9 111.6 Source: World Bank MIS tables 34 References Devarajan, S., L. Squire, and S. Suthiwart-Narueput. 1995. "Reviving Project Appraisal at the World Bank." Policy Research Working Paper 1496. World Bank. Washington, D.C. Futures Group International. 1996. "Etude de l'Impact Socio-economique du VIH/SIDA a Madagascar." Projet Banque mondiale Credit CR1697-MAG. Washington, DC. Hammer, J. 1996. Economic Analysis for Health Sector Projects. Policy Research Department World Bank. Washington, D.C. Lamboray, J-L. and A.E. Elmendorf. 1992. Combating AIDS and Other Sexually Transmitted Diseases in Africa: A Review of the World Bank's Agenda for Action. World Bank Discussion Paper, Africa Technical Department Series Number 181. Washington, D.C. World Bank. 1995. Economic and Sector Work and Results on the Ground. Operations Policy Department. Washington, D.C. World Bank. 1996. "AIDS Prevention and Mitigation in Sub-Saharan Africa: A Strategy for Africa" Human Resources and Poverty Division, Technical Department, Africa Region. Report number 15569. Washington, D.C. World Bank. 1997. Confronting AIDS. Public Priorities in a Global Epidemic. Policy Review Report. New York: Oxford University Press. 35 Distributors of ~ COLOMBIA GERMANY Tel: (52 5) 624-2800 POLAND Faa: (94 1) 432154 Dntoenlace Lda. UNO-VeRtag ISRAEL Fax. (525)624-2822 Intemational Publishing Service E-mail: LHL@srilanka net W orld Bank Carrera 6 No. St-21 PoppolsdoderAllee55 Yozmo Literature Ltd. E-mail: intotec@rnnet.mx Ul. Piekna31/37 Apartado Aereo 34270 53115 Bonn PO. Box 56055 D0O077 Warzawa SWEDEN Publications Santatd de Begot6, D.C. Tel: (49 228) 949020 3 Yohanan Hasandlar Street Mundi-Prenxa Mexico S.A. de C.V. Tel: (482(8628-6089 Wnnnergren-Williams AB Pricbslirad creditaterms oars' fr Tel (57 1)285-2798 Fax (49 228) 217492 Tel Aviv 61560 c/Rio Panuco, 141-ColoniaCuauhtemoc Fax: (48 2)621-7255 W n0. Baa 1305 count rnd carndf terms varoyi"t irou Fax (571( 285-2798 E-mail: unovertag@aol.com Tel: (972 3) 5285-397 06500 Mexico, D.F E-mail: books%iFs@ikp atm com.d S-171 25 Solna countryto cortny. Consultyour ' Fax: (9723)5285-397 Tel: (52 5)533-5658 Tel: (468)705-97-50 local distributor before placing an COTE D'IVOIRE GHANA Fax: (52 5) 514-6799 PORTUGAL Faxn (46 8) 27-00-71 order Center d'Edition et de D11usion Africaixes Ep SBooks Services RO8Y xntemational Livradaha Pa ugal E-mail: mailOwwi.se (CEDA) P0 Box 44 PO Ban 13056 NEPAL Apartodo 2681, Raa Do Cairns 70-74 ARGENTINA 04 . P 541 TUC Tel Asia 61130 Everest Media Intemational Services (P) Ltd. 1200 Lisbon SWITZERLAND Oficina deltLibro Intemacional AbsidjanO04 Accra Tel: (972 3( 5461423 GPO Box 5443 Tel: (1( 347-4982 Librairie Payot Service lnstitationnel As. Cordoba 1877 Tel: (225) 24 6519:24 6511 Faa: (972 3(5461442 Kathmandu Faa: (1( 347-0264 COtes-de-Mantbeenon 30 1120 Buenos Aires Fa-(2(557GREECE E-mail: royilBnervision.net.il Tel. (977 1( 472 152 1902 Lassasne Tel: (541)815-8354 Fax (225)250567 Papasotiriou SA. Fax: (977 1)224431 ROMANIA Tel: (41 21) 341-3229 Fan: (54 1) 815-8156 CYRS35, Sloumara Stri Palestinian Authsrity/Middle East CompaniOne Librarii Bacaresti S.A. Faa: (41 21) 341-3235 E-mail: nlilibn@satlinlr.camCeoPRUS pledRsac 106 82 Athena Index Isformation Services NETHERLANDS Sir Lipscani rno. 26, sector 3 E-mailolilibrosatinkcom Center fC r AppliedgResearch Tel: (30 1(364-1826 PFOB 19502 Jeruslem Dn LindeboomlsnOr-Publikaties Bucharese ADECO Van Diermen EdiionsTechniques AUSTRALA, FIJ, APUNEWGUIEA, 6,Diogens StrlegetEgm Fan (30 1( 364-8254 Tel: (972 2( 6271219 PO. Box 202, 7480 AE Haaskbergen Tel: (40 1(613 9645 Ch. de Lacuez 41 AUSTRALIA, FIJI, PAPUANEWGUINEA, 6 Diogene Street, Engom290 Fax. (972 2)6271634 Tel: (31 53) 574-0004 Fax: (40 1(312 4000 CHI L807Blonay SOLOMON ISLANDS, VANUATU, AND PO. Box 2006 HAITI Fax: (31 53) 572-9296 Tel: (41 21) 943 2673 SAMOA Nicosia Services Cuture Diffusion ITALY E-mail: lindeboo@woddonline nl RUSSIAN FEDERATION Faa: (4121) 943 3605 648 Atrheaorse road Tel: (3572) 44-1730 5, Rue Capois Licosa Commissionaria Sansoni SPA 9a,KolpachniytPereuok THAILAND 648 WhRehorse Road Faa: (357 2(46-2051 C.P 257 Via Duca Di Calabria, 1/1 NEW ZEALAND Ma ospchni Pere31 nkntHaIlANoksDisbto Mtcham 3132 Fog-as-Prince Casella Postale 552 EBSCO NZ Ltd. Moscow7 1095 17831 9 0 Cetrlo Booksdinrb Victoria CZECH REPUBLIC Tel. (509) 23 9260 50125 Firenze Private Mail Bag 99914 Tel: (7095) 917749 30Ban kok 0500 Tel: (61( 3 9210 7777 USIS, NIS Prode)na Fan: (509( 234858 Tel: (55) 645-41 5 New Market Fa 79(195 ogo 00 Fax: (61( 39210 7788 Haselkova 22 Fan (55) 641-257 Auckland SNAOE AWN HN;Tl 6 )2550 E-mail: service@dadirect.com.au 13000 Prague3 HONG KONG, CHINA; MACAO E-mail: licxas5 tlb-c.i Tel: (64 9)524-8119 SINGAPORE; TAIWAN, CHINA; Tel: (66 2)235-5400 Tel: (420 2(2423 1406 Asia 29000 Ltd Fax: (64 9(524-6067 MANMgaR; PbliUhnEg Faa:(662)ic237-8321 AUSTRIA Fax: (4202)24231114 Sales & Circulation Deparment JAMAICA - A lgate Publishing Asia PaclicPIe d TRINIDAD & TOBAGO Gerold and Co. ~~~~~~~~~~~Seabird Hause, unit 1101-02 Ion Randle Pablishers Ltd. NIGERIA 41oKldseg Padluiding Roda-sAND THE CARRIBBEAN Weihburggasse 26 DENMARK 22-28 Wyndham Street, Central 206OdiHope Rod Kingston 6 LNt G Prs L Goldnq Pudd Boiding A-lOll Wino SarvtsndsLitleratar Hong Kong Singapora 349316 ldSystRoa,ematics6 SiveslyPrssodiiedeingpoes3931 St AuusineShopiLtd. te AT10 ) Wien Rsendc A1u Te (852) 2530-1409 Tel: 876-927-2085 Three Crowns Building Jericho Tel- (65)741-5166 SiAastm SingRoad,Stcom Ce ine Tel: (431) 512-47-310 Rosenoems AIIO 1 1Tel: (852) 2530-1409Fax: 876-977-0243 Pris Ine Mail Bag 5095 Faa: (65) 742-9356 TrsinmdMainRobado, Wst AIgasines Faa: (43 1( 512-47-31-29 DK-1 970 Frederiksberg C Foe: (852) 2526-1 107 E-al lclscmlaa -alahaeainonccmTrisidad68 Tobgo,Wea4 Inia Tel: (45 31( 351942 E-mail ssles@asia2000.cormhk E-mail:i234@22is.41 Ih3d6 BANGLADESH Fax:(4531)357822 HUNGARY JAPAN Fax: (234 22) 41-2056 SLOVENIA Fax (868) 645-9467 Micro Industries Development HUNADOR East!m SenvicepBH 3 13Hongok Servichme,Bunkyo-kuNORWA Gospodarski Vestnik Publishing Group E-mail: tobe trinidad.nel Assistance Sociely (MIDAS) ECUADOR Ears Into Service 3an13 Hanko ServcheBnykuNRA Gunaspkark Vensni PabishAgNras Hoase 5, Road 16 Libi Mundi Margitszgeti Europa Hoz 3-ko 13Hng2-hm,Bkyk NIInoRWAYS10 Dxn bljaka caUAD Dhbamondi RArea Libreria lntemscional H-1 138 Budapest To 1(81 818 8 ok Depa men,Postboks6512Eerstad 03 1)30 GustroLd Dhaka 1209 PD. Ban 17-01-3029 Tel: (36 1) 350 80 24, 350 80 25 Te.(P3O8806 asBoamn,Psbk6s2f ld Tl 361(384:313 atox Ltd. Mdhai uldn Tel. (8802(326427 Joan Less Main 851 ~~~~~~~~Fas: (36 1) 350 90 32 Fax: (81 3) 3818-0864 N-0606 Oslo Faa: (380 61) 13398030 PD Boot14 9997, Msdh. aaidn Fax: (880 2)(81186 JuansLeonMa651 E-mail: euroinlomail.matav.hu E-mail: orders@svt-ebs.co.jp Tel: (47 22) 974500 E-mail: repansekj@gvestnik l Pt1614Jia Rd. Fax: (80 2) 811188 Qufto ~~~~~~~~~~~~~~~~~~~~~~~~Faa: (47 22) 97-4545Kapl Tel (593 2) 521-606; (593 2) 544-185 KENYA SOUTH AFRICA, BOTSWANA Tel: (256 41) 251 467 BELGIUM Fax: (593 2) 504-209 Allied Publishers Ltd Atica Book Service (E.A.) Ltd. PAKISTAN For singte titles: Fax: (25041(a251 46 Jean De Lannoy E-mail librimul @librimundi.com.ec 751 Mount Road Quaran Housea Mlangano Street Mirza Book Agency Oxford UnwiersLy Press Southem Africa E-mail: gus@nsganda.com 1090 Bdsu l CoiEU Madras -6000002 PO. Sax 45245 65, Shsahrah-e-Quaid-e-Azam Vasco Boalesard, Goodwoad 1060 Brussels CODEU 44) 852 Nairobi Lahore 54090 PO. Sox 12119,1 City 7463 UNITED KINGDOM Tel: (91 44) 852-3938 ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Microinfa LId. Tel: (322)538-5169 Ruiz de Castilla 763. Edd Expocolor Fax (91 44) 852-0649 Tel (2542) 223 641 Tel: (92 42) 735 3601 Cape Town AltoUd. Fax: (322)53-0841 Primer piso, O #2 Fax. (254 2)330272 Fax (92 42) 576 3714 Tel: (27 21) 595 4400 PO. Sx3 AlIas HampshireGU342PG Quo INDONESIA Fan: (27 21) 595 44350 England BRAZIL TeVFax: (593 2) 507-383: 253-091 Pt Indira Limied KOREA, REPUBLIC OF Oxford Univershy Press E-mail: oxford@oup.co.za Tl (441420 846 Publicacoes Tecnicas lntemacionais LUda. E-mail codeu@impsat.net.ec Jalan Borobudur 20 Dae.Oo Trading Co. Ltd. B Bangalore Town Fax: (44 1420) 89869 Ruas Peixoto Gomide, 209 P0. Ban tUl PD. Sax 34, Youida, 706 Snoun Bldg Sharae Faisal For subscription orders. E-mail: wbank@ukrninoa.demon.co.uk 01409 Sao Paulo, SP EGYPT, ARAB REPUBLIC OF Jakada 10320 44-6 Youido-Dong, Yeongchengpo-Ku PO Box 13033 Intemational Subscription Service Tel: (5511) 259-6644 Al Ahram Distnbution Agency Tel: (62 21)390-4290 Seoul Karachi-75350 PO. Box 41095 The StationeryOffice Fax: (5511)258-6990 Al Galaa Street Fax: (6221)390-4289 Tel: (62 2) 785-1631/4 Tel: (92 21) 446307 Crighall 51 Nine Elms Lane E-mail: postmaster@pti.uol br Cairo Fax (82 2)784-0315 Fax. (92 21) 4547640 Johannesburg 2024 London SW8 5DR Tel.(202)578-6083 IRAN E-mail: ouppak@TheOftice.net Tel: (27 11) 880-1448 Tel. (44 171)873-8400 CANADA Fax: (202)578-6833 Ketab Sara Co Publishers LEBANON Fax: (27 11) 880-6248 Fax (44 171)873-8242 Renous Publishing Co. Ltd. Khaled Eslamboli Ave., 6th Street Librairie du Liban Pak Book Corporation E-mail: iss@is.co.za VENEZUELA 5369 Canotek Road The Middle East ObseNer Delafrooz Alley No. 8 PO. Bax 11-9232 Aziz Chambers 21, Queen's Road Onawa, Onlado K1J 9J3 41 Sherd Street PO. Box 15745-733 Beirut Lashore SPAIN Tecni-Ciencia Libros, S.A Tel: (613) 745-2665 Cairo Tehran 15117 Tel: (9619)217944 Tel: (92 42) 636 3222; 636 0885 Mundi-PrensaLibros, S.A. CentroCuidadComercial Tamanco Fax: (613) 745-7660 Tel:(202)393-9732 Tel: (98 21) 8717819: 8716104 Fax: (961 9)217 434 Fax (9242)6362328 Castello 37 NivelC2, Caracas E-mail: order.dept@senoutbooks.com Fax: (202)393-9732 Fax: (98 21) 8712479 E-mail: pbc@brain.nel.pk 28001 Madrid Tel: (582)9595547,5035;0016 E-mail ketab-sara@neda.net.ir MALAYSIA Tel: (341)431-3399 Fax: (582)9595636 CHINA FINLAND ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Universey at Malaya Cnoperative PERU Fax. (34 1) 575-3999 China Financial 6 Economic Akateeminen Kirjakauppa Kowkab Publishers Bookshop, Limited Edoorial Desarrollo SA E-mail libreria@mUndryrensb en ZAMBIA Publishing House PDO Box 128 PO. Box 19575-S11 PO. Box 1127 Apaaado 3824, Limal UniesiryBookohop,UsiseroityatZambia 8BeDsiFa SiiDn nggJin FIN-00101 Helsinki Tehran Jalan Pantai Baru Ts (51 14) 285380 Mundi-Prensa Barcelona Oreat East Road Campas Benijing Tel: (358 0) 121 4418 Tel: (99 21) 258-3723 59700 Koala Lurmpar Faa: (51 14) 286629 80 C a rseloeCnt, 391usaBok337 Tel. (86 10) 6333-8257 Fax: (3580)1214435 Fax: (9821)258-3723 Tel: (60 3) 756-5000 Tel. (34B3) 488-342 Teln (260L1)25257 Faa: (86 10(6401-7365 E-mail- skahlaus@slackmane Ii ~~~~~~~~~~~~~Faa: (60 3) 755A424 PHILIPPINES Tel. (343)488-34592 Tel: (260 1(2525762 Fax: (86 10 64017335 EmaI akatilaus@stockmannf tl GoveIRELAND E-mail: umkoopRtm net my nematiana Boobsourca Center Incl Faa ( a Fax: (2601) 253 952 China Sank Imnport CentreoRAC Gasetoment Supplies Agency 1127-A Anhipolo SI, Barangay, Vesezuela E-mail: baucelona@mundiprenna as PD. Boa 2825 Wodd~~~FANC BakFbiOin ifig an ISol6lhair MEXICO Makati City ZIMBABWE Beijing 4-5 Harcourt Road INFOTEC Tel. (63 2) 896 6501, 6505, 6507 SL T Acadasic and Basb Books (Pvt Ltd 751 Paris ,0-6 Dblia 2 As. San Femrdo Na 37 Faa (632)00 16 D41 Lake House bo skop 4 Conald Road, Granituside Tel: (331) 40-69-30-06/57 Tel (353 1)601-31 11 Cal. Tsriells Coasts 100. Sir Chillamyalam Dardiverittmpalam Ga Maner thaw PDa.PBuno567 Faa: (2331) 40-69-30-5605 Fax: (353 1) 475-2670 14050 Mexico OF Clmo2Hrr Fax: (33 1) 40-69-30-68 ~~~~~~~~~~~~~~~~~~~~~~~~~~Tel: )94 1) 32105 Tel: 263 4755033 Fan 7634 701913 Recent World Bank Discussion Papers (continued) No. 355 Agriculture and German Reunification. Ulrich E. Koester and Karen M. Brooks No. 356 Evaluating Health Projects: Lessonsfrom the Literature. Susan Stout, Alison Evans, Janet Nassim, and Laura Raney, with substantial contributions from Rudolpho Bulatao, Varun Gauri, and Timothy Johnston No. 357 Innovations and Risk Taking: The Engine of Reform in Local Government in Latin America and the Caribbean. Tim Campbell No. 358 China's Non-Bank Financial Institutions:Trust and Investment Companies. Anjali Kumar, Nicholas Lardy, William Albrecht, Terry Chuppe, Susan Selwyn, Paula Perttunen, and Tao Zhang No. 359 The Demandfor Oil Products in Developing Countries. Dermot Gately and Shane S. Streifel No. 360 Preventing Banking Sector Distress and Crises in Latin America: Proceedings of a Conference held in Washington, D.C., April 15-16, 1996. Edited by Suman K. Bery and Valeriano F. Garcia No. 361 China: Power Sector Regulation in a Socialist Market Economy. Edited by Shao Shiwei, Lu Zhengyong, Norreddine Berrah, Bernard Tenenbaum, and Zhao Jianping No. 362 The Regulation of Non-Bank Financial Institutions: The United States, the European Union, and Other Countries. Edited by Anjali Kumar with contributions by Terry Chuppe and Paula Perttunen No. 363 Fostering Sustainable Development: The Sector Investment Program. Nwanze Okidegbe No. 364 Intensified Systems of Farming in the Tropics and Subtropics. J.A. Nicholas Wallis No. 365 Innovations in Health Care Financing: Proceedings of a World Bank Conference, March 10-11, 1997. Edited by George J. Schieber No. 366 Poverty Reduction and Human Development in the Caribbean: A Cross-Country Study. Judy L. Baker No. 367 Easing Barriers to Movement of Plant Varieties for Agricultural Development. Edited by David Gisselquist and Jitendra Srivastava No. 368 Sri Lanka's Tea Industry: Succeeding in the Global Market. Ridwan Ali, Yusuf A. Choudhry, and Douglas W. Lister No. 369 A Commercial Bank's Microfinance Program: The Case of Hatton National Bank in Sri Lanka. Joselito S. Gallardo, Bikki K. Randhawa, and Orlando J. Sacay No. 370 Sri Lanka's Rubber Industry: Succeeding in the Global Market. Ridwan Ali, Yusuf A. Choudhry, and Douglas W. Lister No. 371 Land Reform in Ukraine: The First Five Years. Csaba Csaki and Zvi Lerman No. 373 A Poverty Profile of Cambodia. Nicholas Prescott and Menno Pradhan No. 374 Macroeconomic Reform in China: Laying the Foundation for a Socialist Economy. Jiwei Lou No. 375 Design of Social Funds: Participation, Demand Orientation, and Local Organizational Capacity. Deepa Narayan and Katrinka Ebbe No. 376 Poverty, Social Services, and Safety Nets in Vietnam. Nicholas Prescott No. 377 Mobilizing Domestic Capital Marketsfor Infrastructure Financing: International Experience and Lessonsfor China. Anjali Kumar, R. David Gray, Mangesh Hoskote, Stephan von Klaudy, and Jeff Ruster No. 378 Trends in Financing Regional Expenditures in Transition Economies: The Case of Ukraine. Nina Bubnova and Lucan Way No. 379 Empowering Small Enterprises in Zimbabwe. Kapil Kapoor, Doris Mugwara, and Isaac Chidavaenzi No. 380 India's Public Distribution System: A National and International Perspective. R. Radhakrishna and K. Subbarao, with S. Indrakant and C. Ravi No. 382 Public Expenditure Reform under Adjustment Lending: Lessons from World Bank Experiences. Jeff Huther, Sandra Roberts, and Anwar Shah No. 383 Competitiveness and Employment: A Frameworkfor Rural Development in Poland. Garry Christensen and Richard Lacroix No. 384 Integrating Social Concerns into Private Sector Decisionmaking: A Review of Corporate Practices in the Mining, Oil, and Gas Sectors. Kathryn McPhail and Aidan Davy No. 385 Case-by-Case Privatization in the Russian Federation: Lessonsfrom International Experience. Harry G. Broadman, editor No. 386 Strategic Managementfor Government Agencies: An Institutional Approachfor Developing and Transition Economies. Navin Girishankar and Migara De Silva No. 387 The Agrarian Economies of Central and Eastern Europe and the Commonwealth of Independent States: Sistuation and Perspectives, 1997. Csaba Csaki and John Nash No. 388 China: A Strategyfor International Assistance to Accelerate Renewable Energy Development. Robert P. Taylor and V. Susan Bogach THE WORLD BANK 1818 H Street, N.W Washington, D.C. 20433 USA Telephone: 202-477-1234 Facsimile: 202-477-6391 Telex: MCI 64145 WORLDBANK MCI 248423 WORLDBANK World Wide Web: http://www.worldhank.org/ E-mail: books@worldbank.org I111111 g11 1 4 251 9 780821 342510 ISBN 0-8213-4251-7