THE WO RLD BANK OPERATIO NS EVALUATIO N DEPARTMENT The Consultative Group to Assist the Poor Addressing Challenges of Globalization: An Independent Evaluation of the World Bank's Approach to Global Programs Case Study Edward Bresnyan Director-General, Operations Evaluation: Gregory K. Ingram Director: Ajay Chhibber Task Team Leader: Uma Lele 2004 The World Bank This paper is available upon request from OED. Washington, D.C. ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCE AND INDEPENDENCE IN EVALUATION The Operations Evaluation Department (OED) is an independent unit within the World Bank; it reports directly to the Bank's Board of Executive Directors. OED assesses what works, and what does not; how a borrower plans to run and maintain a project; and the lasting contribution of the Bank to a country's overall development. The goals of evaluation are to learn from experience, to provide an objective basis for assessing the results of the Bank's work, and to provide accountability in the achievement of its objectives. It also improves Bank work by identifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings. OED Working Papers are an informal series to disseminate the findings of work in progress to encourage the exchange of ideas about development effectiveness through evaluation. The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Contact: Operations Evaluation Department Partnerships & Knowledge Programs (OEDPK) e-mail: eline@worldbank.org Telephone: 202-458-4497 Facsimile: 202-522-3125 http:/www.worldbank.org/oed Abbreviations and Acronyms AFCAP Eastern/Southern Africa Capacity ­Building hub AsDB Asian Development Bank AfDB African Development Bank CG Consultative Group/Council of Governors CAPAF Francophone Africa Capacity-Building hub CAS Country Assistance Strategy CGAP Consultative Group to Assist the Poor DGF Development Grant Facility EXCOM Executive Committee of CGAP IDB Inter-American Development Bank MDG Millennium Development Goal MF Microfinance MFI Microfinance institution PAG Policy Advisory Group PRSP Poverty Reduction Strategy Paper SGP Special Grants Program SME Small and medium enterprise USAID United States Agency for International Development i Table of Contents Acknowledgements ..........................................................................................................iii Preface............................................................................................................................... iv Executive Summary........................................................................................................ vii Genesis, Objectives, and Activities....................................................................... vii Design and Implementation.................................................................................. vii OED Findings...................................................................................................... viii Relevance: Are the Program's Objectives Right? ................................... viii Efficacy: Has the Program Achieved Its Stated Objectives?.......................x Efficiency: Has the Program Been Cost-Effective?................................... xi Bank Performance..................................................................................... xii Lessons..................................................................................................................xiv 1. Introduction and Context: Global Challenges in the Sector....................................1 Global Challenges Facing the Sector......................................................................1 International Consensus for Global Collective Action............................................3 2. Program Alignment with Global Challenges and Bank Priorities..........................6 Genesis and Objectives of CGAP ............................................................................6 CGAP Members and Stakeholders...........................................................................7 Organization and Key Activities..............................................................................8 Relevance of Program to Global Challenges and Bank Priorities..........................9 Consistency with the Bank's Strategic Focus for Global Programs and the Subsidiarity Principle...........................................................................10 Consistency with the Bank's Sector Strategy............................................14 3. Outcomes, Impacts, and Their Sustainability.........................................................16 Monitoring and Evaluation....................................................................................16 Evaluation of CGAP ..............................................................................................17 Grants to Selected MFIs to Spur Learning and Innovation .......................18 Best Practice Dissemination ......................................................................20 Mainstreaming Microfinance in Member Donor Agencies.......................23 Fostering National Policy Environments Conducive to Microfinance......24 Donor Peer Reviews ..............................................................................................25 4. Program Governance, Partnerships, Management, and Financing .....................26 ii Governance and Management...............................................................................26 Partnerships and Participation .............................................................................27 Financing of CGAP................................................................................................31 Risks and Risk Management ..................................................................................34 5. Role of the World Bank.............................................................................................35 Comparative Advantage of the Bank .....................................................................36 Linkages to the Bank's Country Operations..........................................................36 The Bank as Donor to CGAP.................................................................................41 Disengagement Strategy ........................................................................................44 6. Findings and Lessons.................................................................................................46 References.........................................................................................................................49 Annex A. Evaluation Framework for Phase 2 Report and Case Studies ...................51 Annex B. Program At a Glance ......................................................................................59 Annex C. Stated Mission, Objectives, Outputs, and Activities....................................63 Annex D. Genesis, Evolution, and Exit Strategies of CGAP........................................64 Annex E. Relationship of CGAP to the Millennium Development Goals...................66 Annex F. Rating of CGAP Activities According to the Bank's Strategic Focus for Global Programs and OED Subcategories..............................................................67 Annex G. Evaluations and Audits of CGAP..................................................................70 Annex H. Governance of CGAP.....................................................................................75 Annex I. Management of CGAP.....................................................................................78 Annex J. Current Membership of the CGAP Council of Governors, Executive Committee, and Investment Committee ..................................................................80 Annex K. CGAP, Sources and Uses of Funds, 2003 .....................................................89 Annex L. World Bank Lending to "Micro & SME Finance"......................................90 Annex M. Overall Assessment in Relation to Development Committee Criteria......95 Annex N. Results from the Partner Survey...................................................................96 iii Acknowledgements This report was prepared by Edward Bresnyan under the leadership of Uma Lele and with inputs from Chris Gerrard. Editorial assistance was capably provided by Lisa Block. The author wishes to thank the CGAP Operational Team based in Washington, D.C., particularly Elizabeth Littlefield, Ousa Sananikone and Tamara Cook, all of whom gave amply of their time as this study took shape, as well as Richard Rosenburg for comments on the initial draft. A note of appreciation is also extended to members of CGAP's Council of Governors (CG) and Executive Committee (EXCOM) ­ including Carlos Cuevas, the Bank's representative on both the CG and EXCOM ­ many of whom graciously agreed to be interviewed for the report. Following the distribution of an initial draft of the study, thoughtful comments were received from several CG and EXCOM representatives: Bernd Balkenhol, Richard Boulter, Brian Branch, Odoardo Como, Nimal Fernando, Carlos Labarthe, Alvaro Ramirez, David J.N. Stanton, Marilou van Goldstein, Koenraad Verhagen and Johan de Waard, as well as from Ulrike Haupt, of the German Ministry for Economic Cooperation and Development. The author would also like to acknowledge the myriad other professionals in the microfinance industry, whose comments and observations also contributed to the findings of this study. iv Preface This report on the Consultative Group to Assist the Poor (CGAP) is one of 26 case studies (see list on page vi) that have been prepared as source material for the second phase of OED's independent evaluation of the Bank's involvement in global programs. The Phase 1 Report, The World Bank's Approach to Global Programs, which was presented to CODE in June 2002, focused on the strategic and programmatic management of the Bank's global portfolio of 70 programs in five Bank Networks (a cluster of closely related sectors). The Phase 2 Report is based on case studies of 26 global programs and it derives additional lessons for the Bank's strategic and programmatic management of global programs as well as lessons for the design and management of individual programs. The first and largest case study ­ for the Consultative Group on International Agricultural Research (CGIAR) ­ was completed in April 2003. OED reports typically contain recommendations only in those reports presented to the Bank's Board or its committees such as the Committee on Development Effectiveness (CODE). While the case studies that underlie OED's Phase 2 Report were not presented to CODE individually, they were distributed in draft to program partners to obtain their feedback, which was taken into account in the final versions of each report before being disclosed to the public. Each case study follows a common outline and addresses four major evaluation issues, which correspond to the four major sections of each report: · The overarching global relevance of the program · Outcomes and impacts of the program and their sustainability · Organization, management, and financing of the program · The World Bank's performance as a partner in the program These four issues correspond roughly to OED's evaluation criteria of relevance, efficacy, efficiency, and Bank performance, appropriately adapted for global programs. Each case study addresses 20 evaluation questions related to these four evaluation issues (Annex A, Table A.1) that have been derived from OED's standard evaluation criteria (Table A.2), the 14 eligibility and approval criteria for global programs that have been endorsed by the Development Committee and established by Bank Management (Table A.3), and the 8 eligibility criteria for grant support from the Bank's Development Grant Facility (Table A.4). Nineteen out of the 26 case study programs and about two-thirds of the Bank's total portfolio of 70 global programs have received DGF grants. Global programs are defined as "partnerships and related initiatives whose benefits are intended to cut across more than one region of the world and in which the partners (1) reach explicit agreements on objectives, (2) agree to establish a new (formal or informal) organization, (3) generate new products or services, and (4) contribute dedicated resources to the program." (OED, The World Bank's Approach to Global Programs: Phase 1 Report, p. 3). Since November 2000, all new global and regional programs have had to be approved at the initial concept stage, based upon the six approval criteria in Table A.3, by the managing director responsible for the Network or Regional Vice Presidential Unit advocating the v Bank's involvement. Such approval authorizes the respective VPU to enter into agreements with partners and to mobilize resources for the program ­ whether from the DGF, trust funds, or the Bank's administrative budget. Both before and after November 2000, the Bank's participation in some high-profile programs ­ such as the Global Environment Facility, the Multilateral Fund for the Implementation of the Montreal Protocol, the Prototype Carbon Fund, the Global Fund to Fight AIDS, TB, and Malaria, as well as CGAP (in March 1995) ­ has been considered and approved by the Bank's Executive Board. Once a program has been approved at the initial concept stage, the Network vice presidencies are responsible for oversight, management, and quality assurance of their respective portfolios of global programs. This includes establishing priorities among programs in their Networks, ensuring their coherence with the Bank's strategy for each sector, sponsoring applications for DGF grants, managing programs that are housed inside the Bank, fostering links to the Bank's country operations, and promoting synergy among programs within the Network, with the rest of the Bank, and externally with partners. Regional vice presidents oversee and manage the portfolio of regional programs and partnerships in their respective Regions. While regional programs are not covered in this OED evaluation, many global programs have strong regional dimensions, which are addressed in the case studies, in addition to their links to the Bank's country-level economic and sector work, policy advice, and lending. vi List of 26 Case Studies in Phase 2 of OED's Evaluation of the Bank's Involvement in Global Programs Acronym/ Operational Size (US$ Short Form Full Name Start Date millions)1 Environnent & Agriculture 1. CGIAR Consultative Group on International Agricultural Research 1972 395.0 2. GEF Global Environment Facility 1991 387.53 3. MLF Multilateral Fund for the Implementation of the Montreal Protocol 1991 158.6 4. ProCarbFund Prototype Carbon Fund 2000 6.5 5. CEPF Critical Ecosystem Partnership Fund 2000 20.19 6. GWP Global Water Partnership 1997 10.25 7. GIF Global Integrated Pest Management Facility 1996 1.3 Health, Nutrition & Population 8. TDR Special Programme for Research and Training in Tropical Diseases Dec 1975 47.5 9. Global Forum Global Forum for Health Research Jan 1997 3.07 10. UNAIDS Joint United Nations Programme on HIV/AIDS Jan 1996 95.0 11. RBM Roll Back Malaria Nov 1998 11.4 12. Stop TB Partnership Stop TB Partnership July 1999 20.8 13. GAVI Global Alliance for Vaccines and Immunization Oct 1999 124.1 Infrastructure & Private Sector Development 14. WSP Water and Sanitation Program March 1978 12.4 15. ESMAP Energy Sector Management Assistance Programme Jan 1982 7.58 16. CGAP Consultative Group to Assist the Poorest August 1995 12.67 17. infoDev The Information for Development Program Sept 1995 6.07 18. PPIAF Public-Private Infrastructure Advisory Facility Dec 1999 15.61 19. CA Cities Alliance Dec 1999 13.25 Social Development & Protection 20. PostConFund Post-Conflict Fund 1998 10.6 21. UCW Understanding Children's Work 2000 0.56 Trade & Finance 22. IF Integrated Framework for Trade-Related Technical Assistance 1997 2.71 23. FSAP Financial Sector Assessment Program May 1999 10.46 24. FIRST Financial Sector Reform & Strengthening Initiative July 2002 4.64 Information & Knowledge 25. GDN Global Development Network Dec 1999 8.67 26. World Links World Links for Development 1998 6.5 /1 FY04/CY03 expenditures. For the following cases updated, audited data was not readily available so the previous fiscal or calendar year expenditures were used: Global Integrated Pest Management Facility, Water & Sanitation Program, Integrated Framework for Trade-related Technical Assistance. vii Executive Summary GENESIS, OBJECTIVES, AND ACTIVITIES Consultative Group to Assist the Poor (CGAP) 1. CGAP traces its roots to a June 1994 Established: 1995 Donor Working Group meeting in Paris. Objectives: To help build financial Subsequently, nine donors came together to systems that work for the poor, providing large numbers establish the Consultative Group to Assist of people with diverse the Poorest in 1995. Funding and financial services through a administration of CGAP were sought from wide range of organizations. the World Bank, which approved a US$30 Major activities: (1) Disseminating microfinance best practice, million grant from the Special Grants (2) grant-making to selected Program in March 1995 for an initial three- MFIs, (3) mainstreaming year period (Phase I). Following a 1998 microfinance among member donors, and (4) fostering mid-course review, the members of the national-level policy on Consultative Group renewed CGAP for an microfinance. additional five-year period through 2003 FY03 US$15.2 million (Phase II). After a second CGAP evaluation expenditures: took place in 2002, the Consultative Group FY03 DGF US$6.7 million authorized Phase III for a further five-year allocation: period beginning July 2003. FY03 other US$5.7 million donor contributions: 2. CGAP's membership now Location: World Bank comprises 28 bilateral and multilateral Governing 28 member donors, development institutions and private partners: comprising 17 bilateral foundations. Its current objective is to agencies, 9 multilateral support the development of financial agencies (incl. World Bank), and 2 private foundations. systems that work for the poor, by Latest program- 2002 improving the capacity of microfinance level evaluation: institutions to deliver flexible, high-quality financial services to the poor on a sustainable basis. Four major activities during Phases I and II have been: (1) disseminating microfinance best practices, (2) grants to selected microfinance institutions (MFIs) to spur learning and innovation, (3) mainstreaming microfinance within member donor agencies, and (4) fostering national policy environments conducive to microfinance. DESIGN AND IMPLEMENTATION 3. At its core, CGAP is a forum ­ now called the Council of Governors (CG) ­ for donor agencies to coordinate support to microfinance, share best practices, and develop consensus on policy guidelines. Comprised of all 28 member donors and chaired by the Infrastructure Vice President of the World Bank, the CG approves CGAP's strategy, sets policy, provides input into the annual work plan and budget, and receives and approves annual reports. CGAP's Executive Committee (EXCOM) ­ comprised of four CG members elected on a constituency basis, one member appointed by the World Bank, four microfinance industry leaders, and the CGAP Executive Director (ex officio) ­ provides strategic direction and support for the CGAP Operational Team and approves the annual work plans and budgets. viii The CGAP Operational Team ­ housed within the World Bank and headed by the Executive Director ­ carries out activities in accordance with the approved work plans and budgets in the pursuit of the mission of CGAP. The Investment Committee (IC) ­ composed of seven World Bank and IFC staff members ­ exercises fiduciary responsibility for CGAP's grant making to external entities. 4. CGAP's products, which are directed at three major client groups ­ MFIs, member donors, and the broader MFI industry ­ have consisted of grant funding for MFIs and MFI networks, technical tools, technical advice and exchange, and training and capacity-building. Grants to MFIs and MFI networks represented 67 percent of CGAP expenditures during Phase I. However, CGAP repositioned itself during Phase II as a generator of knowledge and tools, and as a provider of in-kind technical expertise and capacity building, primarily to MFIs but also increasingly to member donors. Grants to MFIs and MFI networks decreased sharply to 28 percent of total expenditures during Phase II, while expenditures on technical tools, advice and exchange increased from 5 percent to 47 percent of total expenditures between Phases I and II. CGAP's strategic priorities for Phase III represent a further shift in focus. CGAP now sees itself as promoting a diversity of financial institutions that serve the poor ­ not only the NGO/MFI model, but also other financial intermediaries such as credit unions and the commercial banking sector. Overall, nearly 40 percent of CGAP's expenditures during Phases I and II were considered global, with the remaining 60 percent almost equally divided among Africa, Asia, and Latin America. Eastern Europe and the Middle East accounted for less than 2 percent of total expenditures. OED FINDINGS Relevance: Are the Program's Objectives Right? 5. There exists a strong international consensus that MFIs can be both financially sustainable and reach the poor, and that access by the poor to microfinance can help to reduce poverty. First, insofar as microfinance can be mainstreamed into the national formal financial architecture, it increases financial intermediation, which promotes economic growth. Second, from an institutional perspective, MFIs have shown an ability to reach the poor with financial services. Third, empirical evidence, although by no means fully conclusive, does support the claim that microfinance contributes to poverty reduction. Fourth, and perhaps most importantly, evidence supports the claim that MFIs can both achieve financial sustainability and serve the poor at the same time. But there is less consensus on precisely what kind of global collective action can most effectively enhance the role of microfinance in developing countries. Advocacy will be required to promote overall macro prudential and other regulatory regimes within which MFIs can effectively operate as they are increasingly woven into the formal financial economy. Yet even such reforms to achieve greater stability in domestic financial markets will not, by themselves, be sufficient to catalyze finance toward small and medium enterprises ­ let alone micro-enterprises ­ as noted in a recent joint statement by the Latin American Shadow Finance Regulatory Committee (2004). The Shadow Committee further notes that employment in Latin America is substantially dependent on SMEs, and their improved access to finance through the removal of remaining regulatory impediments as well as a greater technical focus on ix outreach by financial institutions or new banking schemes (i.e., second tier banking) will, in large part, determine the extent of job creation in the region. 6. From the Bank's point of view, CGAP's objectives and strategies are broadly consistent with the Bank's most recent Financial Sector Strategy. Upfront, this strategy confirms that the Bank will support MFIs primarily through improving the framework for microfinance, training and building financial institutions, and avoiding the provision of funds for on-lending either at subsidized rates or where local institutional capacity is inadequate. But to what extent are CGAP's current activities consistent with the subsidiarity criterion for grant support from the Bank's Development Grant Facility (DGF) ­ that DGF grants (CGAP's principal source of funding) should not compete or substitute for regular Bank instruments? And what is the value added of the CGAP partnership to the Bank's clients beyond what the Bank could do in its regular country operations acting through partnerships at the country level alone? 7. During Phase I, CGAP saw itself as adding value with its ability to give grants to and otherwise support MFIs directly, as opposed to brokering its assistance via national governments, as for the Bank's regular country operations. CGAP could operate more quickly, flexibly, and agilely at a level below that of most of Bank's lending operations. Such retailing of grants to MFIs is clearly consistent with the DGF subsidiarity criterion and with the newly clarified (December 2003) rules on the allowable use of Bank budget resources, trust funds, and DGF grants for global programs that DGF grants must flow to entities outside the Bank for funding costs of externally managed entities. 8. Now that CGAP has shifted its focus to a mixture of activities ­ some of which are similar to the Bank's country operations (such as supporting financial sector reform) and others of which are similar to the Bank's Networks (such as knowledge creation and dissemination) ­ its compliance with the subsidiarity criterion is less obvious. However, it is clear from the lower than anticipated financial resources mobilized from its member donors, that CGAP does not have a comparative advantage as a major financing facility for MFIs, as was originally envisaged when it was founded in 1995. The Bank initially contributed US$30 million to CGAP's Phase I in the specific expectation that other donors would contribute an additional US$70 million to such a facility. Yet, as things have turned out, the Bank provided US$65.3 million (68 percent) of the US$96.1 million that CGAP received during Phases I and II combined, while other donor members provided US$20.8 million (32 percent). By way of comparison, the World Bank committed an average of US$220 million a year to the micro and SME finance components of 97 investment projects between 1996 and 2003 inclusive, and disbursed an average of US$140 million a year during this same time period. Over roughly the same time period, USAID's annual grant funding to MFIs averaged US$155 million, reaching more than 3.7 million micro-entrepreneurs. CGAP's shift in focus towards technical assistance and knowledge creation and dissemination activities ­ a shift applauded by its constituent members - is also consistent with similar shifts that have occurred in the other infrastructure and private sector development global programs that are housed inside the Bank. x Efficacy: Has the Program Achieved Its Stated Objectives? 9. In each phase, CGAP has sought to expand access to microfinance for the world's poor to financially sound MFIs that can continue to deliver such services over the long run. While these objectives are clear and evaluable ­ since one can measure the financial soundness and sustainability of MFIs, as well as the extent to which they are actually serving poor people (e.g., by means of a poverty assessment tool) ­ there remains an attribution issue. CGAP does not itself provide financial services to the poor. It simply supports MFIs who provide such financial services, both directly through grants to MFIs (primarily under Phase I activities) and indirectly through its financial and technical support to the donor community and to the MFI industry at large (more so under Phases II and III). 10. CGAP's Phase II evaluation found that, while CGAP has been successful in (1) spurring learning and innovation through grants to selected MFIs and (2) promoting best practices in microfinance, its achievements have been much more modest with respect to (3) mainstreaming microfinance in member donor agencies and (4) fostering national policy environments conducive to microfinance. The programmatic partners of CGAP surveyed by OED ­ the 44 members of the CGAP's Council of Governors, its former Policy Advisory Group, and its Investment Committee ­ also rated CGAP the most successful in creating and disseminating knowledge of best practices in microfinance. In many cases, as comments from CG members attest, the donors turn to CGAP to "do the thinking" about microfinance and how best to operationalize it in their existing programs. Fully 96 percent rated CGAP's achievements in this area as high or substantial, compared to the rating of only 46 percent for improving donor coordination and 29 percent for supporting national policy reforms. Yet, while CGAP has indeed produced and disseminated many technical tools, available via the internet, we know little ­ verifiable through rigorous monitoring and evaluation ­ about the actual uptake of these by MFIs and, more importantly, about their impact on the administration, outreach, and sustainability of these MFIs. Anecdotal evidence, albeit favorable in some cases, is insufficient to assess the extent to which CGAP is contributing to increased access of the poor to sustainable financial services. 11. Assessing the impact of CGAP remains a thorny issue and has generated a great deal of discussion among CGAP members in the course of the review of the earlier draft of this study. On the one hand, a considerable body of anecdotal evidence ­ both documented in the body of this paper and supplemented by comments and observations from CG and EXCOM members ­ asserts the positive impact of CGAP tools, advice and expertise on both donor practices and industry standards. Some examples are the MIX, the MicroBanking Bulletin, the Microfinance Gateway, the IDB-CGAP Rating Fund, the annual microfinance training in Boulder, CO, and the Donor Peer Review process. On the other hand, insofar as they reflect inputs toward the objective of microfinance expansion to the poor, the above examples cannot simply be assumed to generate either the expected outputs or the subsequent impact. To assess the latter, monitoring data are essential and an evaluation framework within which to assess and confirm impact of interventions ­ especially in regard to their impact on reducing poverty ­ is critical. CGAP members commenting on the earlier draft of this report expressed a concern about the difficulty of measuring outputs or impacts because of the "degrees of separation" of CGAP members from the institutions that actually provide finance, and the likely high cost of measuring outputs and impacts. A variety of new techniques for measuring impacts now exist. CGAP needs to move forward on both fronts. xi Efficiency: Has the Program Been Cost-Effective? 12. Some 66 percent of the programmatic partners surveyed by OED rated the implementation, monitoring, evaluation, and oversight of CGAP as high or substantial. Nonetheless, CGAP's Phase II evaluation identified some weaknesses in its governance and management. For instance, the evaluation found that the EXCOM ­ established at the onset of Phase II to better facilitate communication between the CG, the Policy Advisory Group (PAG), and the Investment Committee ­ served primarily as a conduit of information, largely through e-mails, and members tended to participate in more of a personal capacity rather than as a true representative of their intended constituencies. In addition, EXCOM members were not expected to poll their constituents before deciding on issues under consideration, nor did EXCOM play an active role in gathering together the views of its constituencies. 13. It is too early to tell whether the merging of the PAG and EXCOM at the onset of Phase III will improve the effectiveness of CGAP's governance. On the one hand, microfinance expertise is now a requirement for those serving on the EXCOM. This move, combined with the inclusion of additional expertise from industry leaders, should aid in deepening the EXCOM's ability to more effectively guide the activities of CGAP. On the other hand, if the constituency-based model of the EXCOM continues to be ineffective, communication flows may actually worsen. The newly configured EXCOM (under Phase III) may be a means of expediting decision-making at the risk of decreasing member donor representation in the process. The internal strength of the CGAP Operational Team might lead to much of CGAP's decision-making as well as its operations effectively residing with the Operational Team. 14. Although CGAP's Phase II made explicit the responsibilities of member donors with regard to the financing of the program, information sharing, and mainstreaming microfinance, their adherence to and compliance with these responsibilities have been weak. The CGAP Phase II evaluation concluded that only 15 of the 28 member donors actively participated in the CG with appropriate personnel and continuity of representation, 8 others did not, and performance was indeterminate for the remaining 6. The frequent turnover of staff within member donor agencies also made for a weak "institutional memory" in the CG. This raises questions concerning the degree of "buy-in" of member donors to CGAP's objectives and activities. In particular, the low compliance with respect to sharing information about member donors' microfinance portfolios, which has consistently been an agreed commitment upon joining CGAP, strikes at the heart of CGAP's role as a clearinghouse for microfinance knowledge sharing. 15. While CGAP has worked with more than 400 "partners" at the activity level, CGAP remains at its highest governance level essentially a "donor club. Neither developing country governments nor MFIs are currently represented on the Council of Governors, yet the presence of MFIs representatives and industry experts in the revised EXCOM under Phase III is a positive step toward their greater participation in CGAP governance. For an international public sector organization that is concerned, among other things, with improving national policies with respect to microfinance, their absence questions the program's legitimacy and relevance to its intended beneficiaries. If the goal of scaling-up microfinance is to be realized, developing countries and their financial institutions will need to be engaged more directly, especially given the observation from the recent Donor Peer Reviews that there is a xii dissonance within some agencies concerning what microfinance is and how it can contribute to agency objectives and the larger development picture, with some increasingly viewing microfinance as a small, insignificant activity, unable to produce systemic results. Nevertheless, donor representatives attending recent CGAP meetings have undertaken strong steps towards harmonization of policies and practices of their agencies. 16. CGAP's shaky financial base is the greatest threat to its ability to maintain operations over the long run. Given member donors' relative unwillingness ­ apart from the World Bank ­ to make cash contributions to CGAP's core budget, CGAP might be better served by adopting a financing strategy that concentrates more on activities in which member donors could participate via co-financing. Additionally, greater consideration of a "fee-for-services" model may be warranted. But the bigger question remains as to why member donors are so unwilling to channel microfinance funds through the CGAP mechanism. The simple answer may be that member donors do not want to finance the program, and remain content to free ride off any positive outcomes that CGAP offers. Would member donors adopt a different strategy if both the legitimacy of CGAP were raised by increasing stakeholder representation in governance and participation, while simultaneously forging greater transparency in decision-making? Bank Performance 17. As a partner in CGAP, the Bank has utilized its comparative advantage more at the global level than at the country level, particularly its convening power and its global mandate and reach. Yet, in retrospect, the Bank's role at the inception of CGAP may have been excessive. Effectively, the Bank established the conceptual basis for CGAP, chaired the initial Donor Working Group, and led the thought process to operationalize CGAP. The Bank's founding contribution of US$30.0 million, along with housing the secretariat, conveyed early on that CGAP was largely a Bank program. Had the Bank held off in making the US$30.0 million available only after other member donors had actually mobilized the expected remaining US$70.0 million, CGAP might have developed a stronger ownership at start-up. 18. A review of 52 Country Assistance Strategies, CAS Progress Reports and CAS Updates (from October 2001 to March 2003) indicates that CGAP's linkages to country programs have been rather minimal, at least as conveyed through these documents. While 29 of the 52 CASs treat microfinance explicitly, more often than not this is in relation to collaboration with IFC in the context of the Bank's overall assistance strategy. A review of 37 Poverty Reduction Strategy Papers (PRSPs) finds that 28 mention some form of strategic approach to microfinance, but none mention CGAP explicitly. The influence of CGAP would appear to be indirect at best, or perhaps non-existent in the exercises. This cursory review of linkages to the Bank's country programs raises the question of CGAP's capacity to advocate for microfinance across the countries represented in the aforementioned CAS and PRSP exercises. CGAP would likely be more effective in leveraging its scarce human resources by targeting its advocacy for microfinance among the country economists and other technical experts who typically comprise the CAS and PRSP teams. 19. The Bank has provided its financial contributions to CGAP, before 1998 under the Special Grants Program (SGP) and after 1998 through the Development Grant Facility xiii (DGF). When the Bank established the DGF as an umbrella for most of the Bank's grant programs and formalized its grant-giving criteria in OP 8.45, the Bank grandfathered CGAP along with the other pre-1998 programs, on the understanding that it would in due course comply with the DGF eligibility criteria. Six years after the establishment of the DGF, CGAP still does not comply with several of these criteria: (1) that the DGF grant should not exceed 15 percent of total funding, (2) an arm's length relationship with the Bank, (3) donor funding of in-house secretariat costs, and potentially (4) the subsidiarity criterion. 20. In their comments on the earlier draft of this paper, some members acknowledged the Bank's dominant role in CGAP ­ housing the secretariat, chairing the governing body, and providing most of the funding ­ and that this has not only hindered compliance with these DGF funding criteria but also reduced the responsibility for shared governance and placed an undue share of intellectual and fiduciary responsibility, accountability and risk on the Bank for program performance. To its credit, CGAP has recently instituted two arrangements that serve to reduce the role of the Bank in general and of the Infrastructure Vice Presidency in particular in the governance and management of CGAP, thereby fostering more of an arm's length relationship with the Bank. First, while the Bank's Infrastructure Vice President remains the chair of the Council of Governors, EXCOM elects a chairperson from among its member donors for a two-year renewable term. Second, the Bank's representative on both the CG and EXCOM is a senior Bank staff member (lead economist) located in the Financial Sector Vice Presidency, and therefore outside the Bank's management chain responsible for CGAP ­ the only such example of the separation of oversight and management among the 12 global programs reviewed by OED that are located inside the Bank. In addition, the FSE Vice Presidency provides an annual budget allocation of US$30,000 for him to exercise this oversight role. To the outside world, and in particular to the MFI industry, CGAP also seems to have developed an identity at least as independent from the Bank, if not more so, than the other five infrastructure global programs reviewed by OED. Yet it is paradoxical that this independent identity has been established without donors taking on more of the burden sharing in the overhead costs of CGAP and with some reducing the priority they accord to microfinance. 21. Concerning the subsidiarity criterion, now that CGAP has significantly scaled back its grant-making activities, the justification for the Bank's continuing participation in CGAP rests largely on its activities characterized by substantial economies of scale and scope, such as knowledge creation and dissemination, capacity building, and improving donor coordination, that add value to the Bank's clients beyond what the Bank's country operations can do acting through partnerships at the country level alone. While evidence is so far sufficiently compelling with respect to CGAP's knowledge creation and dissemination activities, it is less so with respect to donor coordination or impact on Bank operations. CGAP needs to put in place more systematic monitoring and evaluation processes, including a structured set of quantitative or qualitative output, outcome, and impact indicators to demonstrate the continuing value added of its global partnership to the Bank's client countries. 22. It is not the place for this review, which is a background paper for OED's larger review of the Bank's involvement in global programs, to make a recommendation regarding a disengagement strategy for CGAP. At one extreme, CGAP would appear to be a good candidate for mainstreaming into the Bank's regular budget, supported also by donor trust xiv funds, in the way that the Water and Sanitation Program (WSP) and the Energy Sector Management Assistance Program (ESMAP) currently operate. This would also be more consistent with the newly clarified rules on the allowable use of Bank budget resources, trust funds, and DGF grants for global programs. The negative side of this option is the risk that CGAP might lose some of its independent identity from the Bank, which it has fostered and which it enjoys in the MFI industry at large. 23. At the other extreme, CGAP could pursue an independent legal status and exit the Bank. This would cement a distinct persona for CGAP, one which many of the member donors express already exists for the program. It might also spark a greater degree of ownership and burden sharing on the part of the other member donors. Encouraging developing countries to become full-fledged members of a CGAP as its own independent legal entity would also enhance its legitimacy in the eyes of its developing country clients. 24. While it may be possible to find a sustainable course of action between these two extremes, staying on the present course seems unsustainable. CGAP would continue its uneasy relationship with its principal source of funding ­ the DGF. In the face of increasing competition for DGF funds, pressures will mount to enforce DGF's funding criteria equitably among all programs, those that started both before and after 1998. Enforcing the DGF criteria would result in a drastic reduction in Banks' financial support to CGAP, or, in the extreme, a complete withdrawal of DGF support. Under either of these scenarios, and without an alternative plan of action, CGAP would likely cease operations, given the historical importance of DGF support to its livelihood and the as yet unrealized substantive cost sharing from non-Bank member donors. LESSONS 25. In the case of programs which were explicitly intended to benefit the poor as was the case with CGAP, Bank-sponsored global programs need to align more closely ­ and in some cases, realign ­ with the mission of "a world free of poverty." CGAP was conceived as a program to foster and expand access to the poor to microfinance, which it pursued quite actively under its Phase I as a retailer of grants to MFIs under a demand-driven approach. However, a shift in focus under Phase II (continued under the current Phase III) has brought about greater concentration of CGAP activities in standard-setting, technical tools development and capacity building, increasingly directed to its member donors as well as to the microfinance industry at large. Grant-making to MFIs, primarily through its Appraisal and Monitoring service, also became more supply-driven since member donors now select those MFIs they have targeted for financing and other assistance. Whether these shifts have been effective in expanding the access of the poor to microfinance remains unclear. Yet, the modest progress toward mainstreaming microfinance in member donor agencies, coupled with donors' diminished view of microfinance, raises questions about this approach. 26. Adequate monitoring and evaluation must be present to determine progress toward program objectives, realization of intended outputs, and extent of impact from program interventions. CGAP has indeed produced and disseminated many technical tools, available via the internet, in addition to best-practice exchanges and ongoing policy dialogues. But we know little that is verifiable through rigorous monitoring and evaluation about the actual uptake of these technical tools by MFIs and, more importantly, the impact that such adoption may have had xv in the administration, outreach and sustainability of these MFIs. Anecdotal evidence, albeit favorable in some cases, is nonetheless insufficient to conclude that CGAP is generating the needed impact toward increasing access among the poor to financial services. Commentators on this case study have generally accepted three key points: the need for (1) impact monitoring, (2) expanding participation to developing/beneficiary countries and (3) greater financial burden sharing. 27. A broader stakeholder representation, not just the voices of donors alone, could contribute to greater legitimacy and relevance of the objectives and activities of global programs like CGAP. While developing countries and MFIs have benefited from CGAP activities, they are excluded from the strategic direction and oversight processes of the program at the highest governance level and there is little evidence of their influence in CGAP strategy or policy setting. Sustainable microfinance will undoubtedly require a scaling-up of microfinance worldwide, in which both developing country governments and the commercial financial sector must be engaged. Furthermore, donors alone will likely be unable to meet the capital requirements implicit in this scaling-up. The comparative advantage of the private sector, in the guise of commercial financial institutions, can and needs to be brought to bear. 28. The Bank could be more proactive and selective in its financing obligations, eliminating the incentive for other donors to "free-ride." While financial contributions to CGAP's budget have, since the inception of the program, been a condition of membership, some members have never made a financial commitment, while others have done so only sporadically. With the Bank contributing a much larger than proportional share and greatly exceeding the 15 percent criterion of the DGF, other member donors have little incentive to carry their own weight, despite the explicit requirement under the CGAP Charter and the Guidelines set forth by the DGF for in-house secretariats. 29. Tighter linkages to Bank operations are crucial. Among CGAP's objectives is that of dissemination of best practice, yet the true test of this is the adoption of these practices by the member donors, including the Bank. The evidence suggests that CGAP has had minimal input into both the PRSP and CAS exercises, which serve as base documents for Bank lending programs in its client countries. One reason for this may be the program's lean staffing in relation to the myriad PRSP and CAS processes that are ongoing at any one time. While Bank lending to microfinance has been growing since it bottomed out in 1993, this is usually as a component within a larger overall project. This seems to signal that both the Bank and its client countries do not see the sector as meriting increased activity, except as part of larger, multi-sectoral projects. 30. An overwhelmingly dominant role in which the World Bank houses the secretariat, chairs the governing body, and provides most of the funding even while ensuring the ownership of other partners, reduces their responsibility for shared governance, and places an undue share of responsibility, accountability and risk on the Bank for program performance. Proactive establishment of an independent identity for global programs like CGAP would help to determine early on whether sufficient ownership exists among the program partners to sustain its operations. In the absence of the strong and sustained financial support of the Bank, CGAP would likely have ceased activities at the close of its Phase I. 1 1. Introduction and Context: Global Challenges in the Sector GLOBAL CHALLENGES FACING THE SECTOR 1.1 While absolute numbers upon which to base an estimate for demand for microfinance are elusive, there appears to be a strong consensus ­ at least among the donor community ­ that access by the poor to formal financial markets remains extremely scarce in the developing world.1 For millions of people worldwide, limited access to formal sources of financing has constrained asset formation, whether in the form of acquiring land, buying homes, or establishing and building businesses. Yet the evidence clearly underscores the notion that the poor are indeed "bankable," and that institutions that serve them as a market niche can operate on a financially sustainable basis. 1.2 Any discussion of the potential role of microfinance in reducing world poverty must address the larger issue of financial intermediation and its role in economic development. There is a growing consensus as to the role that increased financial depth can play in spurring economic growth, first through mobilizing resources through savings and subsequently by making these resources available via the lending function for consumption and productive activities. 1.3 Yet increasing financial intermediation among the poor poses myriad challenges. First, asymmetric information is likely to exist, in that the borrower knows more than the financial intermediary about his own financial circumstances and ability to repay a loan, a constraint usually addressed by the requirement of collateral from the prospective client. De Soto (2000) finds that by merely formalizing the title to assets already in their possession, many in the developing world would become potential clients for financial services (namely, credit) through the creation of a negotiable asset to be used as collateral. Much of this involves the so-called "informal sector" and how (or if) it can be successfully absorbed into the formal economy. 1.4 A few examples from De Soto (2000) illustrate this point: · In 1995, 280,000 farmers in the former Soviet Union ­ out of a possible 10 million ­ actually owned their land, as evidenced through formal titling. · In the Philippines, about 57 percent of city dwellers and 67 percent of rural people live in housing that is not formally titled and therefore non-negotiable (termed by De Soto as dead capital). For Peru, these figures are 53 percent and 81 percent, respectively. · For the developing world and transition economies, about 85 percent of urban parcels and 40-53 percent of rural parcels are deemed dead capital, the total value of which is estimated at US$ 9.3 trillion, a figure that is 20 times the total of direct foreign investment into these countries in the decade after 1989, 46 times as much as all World Bank loans of the past three decades, and 93 times all development assistance from all advanced countries to the Third World over the same period. 1. Informal sources (such as moneylenders or relatives) are options open to most. 2 However, contributing to "assetization" and hence to financial intermediation in developing economies, while important, does not speak directly to the objective of expanding such access by the poorest to financial intermediation within these economies. 1.5 In 1997, the Microcredit Summit Campaign, drawing on the participation of nearly 3,000 people from 137 countries, launched a nine-year campaign to reach 100 million of the world's poorest families with microcredit by the year 2005. This provides a benchmark for determining the scope of the demand for microfinance. Others (Otero 1999, Robinson 2001) have placed the demand for microfinance as high as 500 million clients. Such financial intermediation has become commonplace in the developed, industrialized world, as can be measured by the number of bank branches per 1,000 as well as the proliferation of consumer banking and financial services in industrialized countries that are targeted to the needs of small businesses and ordinary citizens. 1.6 According to a Microcredit Summit estimate, only 134 million of the world's poorest currently have access to microfinance services.2 This represents only about 5 percent of people living on less than US$ 2 per day (Table 1). The knowledge that, at most, about 5 percent of those living on less than US$ 2 a day have accessed microfinance is another indication of the vast need which remains to be fulfilled. Yet evidence from poverty assessments, conducted by CGAP, on a number of MFIs indicates that while all were reaching the very poor (defined as in the bottom third of the population, when compared to their neighbors), few were actually reaching large numbers of this group. Most MFIs were biased toward less-poor clients, or at best reflected the general population distribution of the neighborhoods being served. Table 1. World Population Living on < US$ 1 and < US$ 2 a day (by region) Less than US$ 1 per day Less than US$ 2 per day Region 1999 Percent 1999 Percent East Asia and Pacific 279.0 15.6 897.0 50.1 Excluding China 57.0 10.6 269.0 50.2 Europe and Central Asia 24.0 5.1 97.0 20.3 Latin America and the Caribbean 57.0 11.1 132.0 26.0 Middle East and North Africa 6.0 2.2 68.0 23.3 South Asia 488.0 36.6 1,128.0 84.8 Sub-Saharan Africa 315.0 49.0 480.0 74.7 Total 1,169.0 23.2 2,802.0 55.6 Total excluding China 945.0 25.0 2,173.0 57.5 Source: World Bank data. 2. In 2002, the Microcredit Summit Campaign reported that, as of December 31, 2001, some 2,186 MFIs had reached 54.9 million clients, of whom 26.8 million (or 48 percent) were among the poorest when they received their first loan (Tables 2 and 3). Assuming, on average, five persons per household, this implies some 134 million individuals have been impacted by microfinance. 3 1.7 Establishing and supporting MFIs that are capable of extending their reach to the poor while gaining operational sustainability will also be required. Here, technological advancements that work to minimize transactions costs can aid in achieving sustainability while active poverty targeting can assist in increasing outreach to the poor and very poor. The path to sustainability may (and historically does) require operational subsidies, which can be justified under the proviso of effective poverty targeting of microfinance. But how to identify actual and potential winners among the gamut of MFIs will be just as important as providing the operational subsidies needed to permit them to successfully navigate the path toward sustainability. Advocacy will also be required to promote prudential and other regulatory regimes within which MFIs can effectively operate as they are increasingly woven into the formal financial economy. Yet, even such reforms to achieve greater stability in domestic financial markets will not be sufficient to catalyze finance toward small and medium enterprises, let alone micro-enterprises, as noted in recent joint statement by the Latin American Shadow Finance Regulatory Committee (2004). The Shadow Committee further notes that employment in Latin America is substantially dependent on SMEs and their improved access to finance through the removal of remaining regulatory impediments. According to the report, a greater technical focus on outreach by financial institutions or new banking schemes (i.e., second tier banking) will, in large part, determine the extent of job creation in the region. INTERNATIONAL CONSENSUS FOR GLOBAL COLLECTIVE ACTION 1.8 Why does microfinance matter to the global community? What can it offer to improve the quality of life for the nearly one-half of the world's population that exists on less than US$ 2 per day? Attention to microfinance can be justified on at least four fronts. First, microfinance, insofar as it can be mainstreamed into the national formal financial architecture, increases financial intermediation, which, in turn, promotes economic growth. In this instance, microfinance is merely one set among several products within the overall financial sector, specifically designed to address the financial needs of, but not limited to the poor. The key here is that financial access is often a precursor to long-term economic growth. 1.9 Second, from an institutional perspective, MFIs have shown an ability to reach the poor with financial services. Nearly one-half of the clients already documented under the Microcredit Summit Campaign (26.8 million households) were determined to be among "the poorest" in their respective countries (Tables 2 and 3).3 While methods for assessing poverty have varied, both across countries and among MFIs, clearly these institutions are demonstrating an ability to target their products toward the poor. Yet, the institutional sustainability of these MFIs, in terms of attaining both operational and administrative self- sufficiency, remains an open question. Several sources in the literature provide ample evidence of strong outreach on the part of MFIs. SHARE in India and CRECER in Bolivia are two examples:4 72.5 percent of SHARE clients and 41 percent of CRECER clients are 3. CGAP states that it only works with MFIs with a client base of at least 3,000 micro-entrepreneurs. This would include, at most, about 27 percent of the MFIs in Table 2. 4. These are cited in Harris, ed. 2002, and Morduch and Haley 2002. SHARE provides financial services to poor women in Andhra Pradesh, India. As of March 2002, SHARE had a client base of 109,000 across 22,000 "banking centers" (consisting of 35-40 members each) with an average client loan size of US$ 85, at 20 percent a.p.r., payable over 50 weeks. CRECER provides integrated financial and educational services to very poor 4 living in absolute poverty, defined as less than US$1 per day. Some 60 percent of SHARE clients and 73 percent of CRECER clients live below their respective national poverty lines, and 58 percent of SHARE clients and 39 percent of CRECER clients are among the poorest one-third of their communities. Finally, both SHARE and CRECER demonstrate a high depth of outreach, since 33 percent of SHARE clients and 20 percent of CRECER clients are among the poorest 20 percent in their own communities. The MFI clientele, at least in regard to capacity to service debt, appear to demonstrate strong sustainability, given the debt loss rates among the best MFIs of between 3-5 percent, rivaling that of many commercial banks. Table 2. Reporting MFIs (as of December 2002) Size of Institution Total Poorest Clients Percent of Total (# poor clients) Number of Institutions (millions) Poorest Clients 1 million or more 8 13.545 32.6% 100,000 to 999,999 25 6.414 15.4% 10,000 to 99,999 222 5.962 14.3% 2,500 to 9,999 410 1.959 4.7% Less than 2,500 1,904 1.003 2.4% Networks 3 12.711 30.6% Total 2,572 41.594 100.0% Source: State of the Microcredit Summit Campaign Report 2003. Table 3. Regional Breakdown of Reporting MFIs (millions of clients) Region # MFIs Percent # Clients # Poorest # Poorest Women of total 2001 2002 2001 2002 2001 2002 Africa 811 32% 4.608 5.762 3.462 4.202 2.362 2.612 Asia 1,377 54% 47.892 59.632 22.34 36.304 18.099 29.423 Latin America & Caribbean 246 10% 1.973 1.942 0.928 0.976 0.644 0.589 Middle East 23 1% 0.068 0.083 0.036 0.038 0.017 0.012 Developing world total 2,457 96% 54.541 67.419 26.766 41.520 21.122 32.636 North America 47 2% 0.263 0.047 0.032 0.022 0.017 0.012 Europe and NIS 68 3% 0.127 0.14 0.081 0.052 0.031 0.028 World total 2,572 100% 54.931 67.606 26.879 41.594 21.170 32.676 Source: State of the Microcredit Summit Campaign Report 2003. 1.10 Third, empirical evidence, although by no means fully conclusive, does support the claim that microfinance contributes to poverty reduction. Out of SHARE clients with three or more years of participation, 76 percent experienced a significant reduction in economic poverty, with one-third no longer being classified as poor. For CRECER, 66 percent reported women and their families in the rural and marginal urban areas of Bolivia. As of December 2001, CRECER had about 31,000 clients in some 1,700 village banks, across five of Bolivia's nine departments. 5 an increase in incomes. In both cases, these increases were attributed to investment of loans in income-generating enterprises. A detailed impact assessment of BRAC in Bangladesh suggested that members who stayed in the program for more than four years increased household expenses by 28 percent and assets by 112 percent. Finally, an early study of Grameen Bank found that its members' incomes were 43 percent higher than incomes of control groups in non-program villages and 28 percent higher than non-members in Grameen villages.5 1.11 Spanning the 13 years from 1990-2003, Khandker studied three MFIs in Bangladesh6 ­ BRAC, Grameen Bank, and RD-12 ­ and found that (1) as much as 5 percent of program- participating households should be able to lift their families out of poverty every year by borrowing from a microfinance program, (2) microfinance helps reduce extreme poverty much more than moderate poverty, and (3) the welfare impact of microfinance is also positive for all households, including non-participants, indicating that microfinance programs are helping the poor to increase their incomes, and are not just redistributing existing income. Programs have spillover effects in local economies, Khandker notes, thereby increasing local village welfare. 1.12 Fourth, and perhaps most importantly, evidence supports the claim that MFIs can both achieve financial sustainability and serve the poor at the same time. Microfinance practitioners have tended to divide philosophically along these two major fronts ­ financial systems and poverty lending. Under the first front (financial systems), the emphasis has been placed on forming and cementing strong institutions capable of serving micro-clients over the long run, under the premise that reaching self-sustainability ­ meaning no subsidies for administrative support or portfolio expansion ­ should be the goal for these institutions. Conversely, the second front (poverty lending) has stressed provision of financial services to the poor, which may require more intensive resource allocation and subsidies, precisely for the same reasons that the poor have historically been "off the radar screen" in terms of access to financial services. To some, there is a perceived tradeoff between sustainable MFIs and targeting the poor with financial services. Yet the evidence provided by the results from both SHARE and CRECER tend to bolster the claim that these two objectives are mutually achievable. Both SHARE and CRECER have achieved 100 percent financial self-sufficiency and are rated investment grade by internationally recognized rating agencies.7 Additionally, the MicroBanking Bulletin reports that only 2.8 percent of loans held by the 124 MFIs in its database were delinquent by more than 90 days. 1.13 The above four points make a strong case for the Bank's involvement in increasing access to microfinance as a means of achieving its expressed mission of "a world free of poverty." In addition, greater access to microfinance, with its demonstrated ability to reach the poor and to reduce poverty can potentially make a substantial contribution toward achieving the Millennium Development Goals. Furthermore, in the 1990 World Development Report, the Bank noted that poverty can be reduced most effectively through a two-fold strategy of (1) promoting the productive use of the poor's most abundant asset, labor, and (2) 5. Daley-Harris, ed. 2002, and Morduch and Haley 2002. 6. Khandker 1998, and Khandker 2003. 7. Daley-Harris, ed. 2002, Chapter 1. 6 providing basic social services to the poor. Microfinance, at least through its more traditional outlet of micro-enterprise lending, makes available financial resources to catalyze this labor into income-generating activities. 1.14 As CGAP affirms in its vision statement, "the vast majority of the world's people are poor and need access to basic financial services. The poor require the means to manage financial resources and increase and diversify income and assets, thereby reducing their vulnerability to shocks and enabling greater access to health services and education."8 Thus, access to microfinance can make an essential contribution to reducing poverty. Yet what are the challenges that must be confronted in order for microfinance to make such a contribution? What kind of global collective action can most effectively enhance the role of microfinance in developing countries? Going to scale with microfinance will require much greater capitalization within the sector than currently exists. To reach the goal of supplying 100 million of the poorest with microfinance, an additional US$7.3 billion will be required (assuming an average outstanding loan balance of US$ 100). Applying the same assumption to the earlier cited upper bound of 500 million clients would require some US$47 billion. Applying a standard capital-adequacy ratio of 8 percent under each case would still require from US$650 million to US$3.8 billion as leverage capital.9 Some of this capitalization may be available from existing or new established MFIs, yet the greater promise of mobilizing and deploying such levels of financial resources lies in the commercial banking sector, where emphasis is currently being placed. 2. Program Alignment with Global Challenges and Bank Priorities GENESIS AND OBJECTIVES OF CGAP 2.1 CGAP traces its roots to a June 1994 Donor Working Group meeting in Paris, where the creation of a Consultative Group mechanism was proposed to provide assistance to the very poor through a microfinance program. Subsequently, nine donors came together to establish the Consultative Group to Assist the Poorest in 1995.10 Funding and administration of CGAP were sought from the World Bank, which approved a US$30 million grant from the Special Grants Program in March 1995 for an initial three-year period (Phase I). Following a 1998 mid-course review,11 the members of the Consultative Group renewed CGAP for an additional five-year period through 2003 (Phase II). After a second CGAP evaluation took place in 2002,12 the Consultative Group authorized Phase III for a further five-year period beginning July 2003. The present case study concentrates on the activities of CGAP under Phases I and II. 8. CGAP Charter, September 2002. 9. Gibbons and Meehan in Daley-Harris, ed. 2002, chapter 5. 10. The founding members were Canada, the Netherlands, the United States, AfDB, AsDB, IFAD, UNCDF, UNDP, and the World Bank. 11. Egger et al. 1997. 12. Fox, Havers, and Maurer 2002. 7 2.2 As a precursor to its Phase III, the member donors of CGAP composed and ratified a Charter to sharpen CGAP's mission in the area of microfinance and to formalize the roles, responsibilities, and accountabilities of the officers and bodies that govern and manage the CGAP partnership. The governance structure, broadly similar to other infrastructure and private sector development programs that are housed in the World Bank, has remained fairly constant over the eight-year period, with two noteworthy changes ­ the creation of an Executive Committee (EXCOM) during Phase II and the merging of the Policy Advisory Group (PAG) into EXCOM at the beginning of Phase III ­ which are discussed further in Chapter 4. CGAP MEMBERS AND STAKEHOLDERS 2.3 Fundamentally, CGAP's authority is derived from the 28 donor organizations that make up the membership of its Council of Governors. Whereas the Bank's Executive Board initially authorized the establishment of CGAP in March 1995, the Council of Governors alone adopted its new Charter for Phase III in 2002.13 CGAP has a broad view of partnerships which extends beyond these 28 programmatic partners. Its 2003 Annual Report notes that, through October 2003, CGAP has partnerships with nearly 400 organizations, including MFI grant recipients, international and regional MFI networks, central bankers, rating agencies, and commercial banks, all of which may be viewed as stakeholders.14 2.4 By comparison, beginning in February 1997, the Microcredit Summit Campaign brought together more than 2,900 people representing 1,500 institutions from 137 countries under the common objective of reaching 100 million of the world's poorest families, especially the women of those families, with credit for self-employment and other financial and business services by 2005. The Campaign continues to bring together microcredit practitioners, advocates, educational institutions, donor agencies, international financial institutions, non-governmental organizations and others involved with microcredit to promote best practice in the field, to learn from each other, and to work toward reaching this goal.15 In fact, CGAP was itself a key player in helping to organize the Campaign. While CGAP and the Microcredit Summit Campaign have distinct agendas and priorities, it is nonetheless instructive to note the strides achieved by the Summit in regard to monitoring and documenting (1) MFI outreach toward the poor and (2) the impact of microfinance, as was noted in the preceding chapter. Furthermore, the Summit, while including aid agencies and donors among their membership, has moved beyond these groups toward practitioners, country representatives and the private sector. By contrast, membership on CGAP's Council of Governors remains restricted to "public bilateral and multilateral donor agencies and private foundations with a significant involvement in the development of financial services for the poor."16 As discussed in Chapter 4, this may raise questions about the legitimacy of 13. Under the Phase III Charter, the Consultative Group (CG) was renamed the Council of Governors. 14. Stakeholders are "parties who are interested in or affected, either positively or negatively, by a program." The programmatic partners of CGAP are a subset of the stakeholders. 15. See the RESULTS International website for greater detail on the Microcredit Summit Campaign: www.microcreditsummit.org. 8 CGAP as an international public sector organization serving clients in developing and transition countries.17 ORGANIZATION AND KEY ACTIVITIES 2.5 With a current membership of 28 donor organizations, annual financial contributions of about US$15.3 million in FY03, and an operational, Washington-based staff of 22, CGAP pursues its strategic themes through activities in about 50 countries worldwide.18 As a global program housed in the Bank, CGAP benefits from substantial core funding from the Bank's Development Grant Facility (DGF),19 works within the Bank's internal bureaucracy, and taps into the substantive convening power of the Bank to pursue its agenda of expanding the reach and sustainability of microfinance worldwide. 2.6 CGAP's financial and human resources are directed at three client groups: (1) MFIs, (2) member donors, and (3) the broader MFI industry. CGAP's products to these clients have consisted of funding for MFIs and MFI networks, technical tools, technical advice and exchange, and training and capacity-building (Table 4). Over the life of CGAP, grants to MFIs and MFI networks have accounted for about one-half of CGAP's expenditures (Table 5).20 Yet these grant-making activities were heaviest under Phase I, and they diminished substantially under Phase II. Grants to MFIs are typically tranched and conditioned on attainment of agreed-upon performance targets for the MFI, which are detailed in a formal partnership agreement signed between CGAP and the MFI. To be eligible for CGAP funding, MFIs must (1) use mechanisms and methodologies to respond to the needs of poor entrepreneurs and be devoted to serving the poor, (2) make at least 50 percent of loans to poor women, (3) serve at least 3,000 micro-entrepreneurs, and (4) have achieved operational self-sufficiency and be on a path to full financial self-sufficiency.21 Overall, nearly 40 percent of total CGAP expenditures are considered global, with the remaining 60 percent almost equally divided among Africa, Asia, and Latin America. Eastern Europe and the Middle East account for less than 2 percent of total expenditures. 16. CGAP Charter, September 2002, paragraph 3.1. While this paragraph states that "other types of institutions involved in microfinance may be considered for membership by the CG at some future point," this has not yet happened. 17. Legitimacy can be defined as "the extent to which the authorizing environment for the program is effectively derived from those with a legitimate interest in the program (including donors, developing and transition countries, clients, and other stakeholders), taking into account their relative importance." 18. CGAP has also recently established a small liaison office in Paris with two professional staff. 19. This accounted for 70 percent of CGAP's funding during its first eight years (1995 to 2003), and 52 percent in fiscal year FY03. See Table 12. 20. CGAP uses the term "investments" rather than "expenditures." In a strict accounting sense, these grants are not investments since they do not represent an asset on CGAP's balance sheet. 21. CGAP: Status Report and Renewal, March 1998, p. 21. Operational self-sufficiency indicates that the MFI is able, through portfolio revenues, to cover all administrative (i.e., operational) expenses, excluding the cost of capital. Financial self-sufficiency includes the cost of capital. These criteria were revised (as reflected here) in 2002. 9 RELEVANCE OF PROGRAM TO GLOBAL CHALLENGES AND BANK PRIORITIES 2.7 The Bank has four major criteria for assessing the relevance of its global programs: · International consensus: The program reflects an emerging international consensus that global action is required (endorsed by the Development Committee on September 25, 2000). · Strategic focus: The program (1) provides global public goods, (2) supports international advocacy for reform agendas that significantly address policy frameworks relevant for developing countries, (3) is a multi-country program that crucially depends on highly coordinated approaches, or (4) mobilizes substantial incremental resources that can be effectively used for development.22 Table 4. CGAP Deliverables Matrix Clients Products MFIs Member Donors Microfinance Industry Technical handbooks Appraisal Format Poverty Microfinance Gateway Tools Assessment Manual Occasional papers Microbanking Bulletin Global Donor Portfolio Focus Notes Training and Regional hubs and training Donor staff training External Audit Capacity- capacity building partners courses Building Program Technical advice Appraisals, monitoring, Portfolio reviews Regulation and supervision and exchange business planning Relationship managers Policy Issues Funding (retail and Funding in MFIs and Products and services for Initiatives to benefit the wholesale) networks member donors entire industry Source: CGAP Annual Reports. Table 5. CGAP Expenditures by Product Type and Phase US$ Millions Percent of Total Product Phase I Phase II Total Phase I Phase II Total Funding for MFI networks 5.500 0.750 6.250 22 % 3 % 12 % Funding for MFIs 11.383 4.263 15.646 45 % 16 % 30 % Funding Networks 3.039 2.325 5.364 12 % 9 % 10 % Technical Advice and Exchange 0.840 2.940 3.780 3 % 11 % 7 % Technical Tools 0.461 9.424 9.884 2 % 36 % 19 % Training and Capacity Building 3.990 6.762 10.752 16 % 26 % 21 % Total 25.213 26.464 51.676 100 % 100 % 100 % Source: CGAP Annual Reports, 1998-2002. 22. A global program has to meet only one of these criteria to be considered eligible for Bank support. As shown in Annex A, Table A.3, "providing global public goods" and "supporting international advocacy" are direct references to the Bank's global public goods and corporate advocacy priorities as enunciated in the Strategic 10 · Subsidiarity: The program does not compete with or substitute for regular Bank instruments (established by the DGF Council on October 28, 1998). · Consistency with sector strategies: The program has a clear strategic rationale consistent with the relevant sector strategy paper. 2.8 The issue of international consensus was discussed in Chapter 1. While there exists a strong international consensus that MFIs can be both financially sustainable and reach the poor, and that access by the poor to microfinance can help to reduce poverty, there is less consensus on precisely what kind of global collective action can most effectively enhance the role of microfinance in developing countries. And while CGAP has worked with more than 400 "partners" at the activity level, CGAP itself represents primarily a consensus among its 28 member donors at the programmatic level. Some 71 percent of the programmatic partners surveyed by OED acknowledge that the involvement of developing and transition countries in the design, governance, implementation, and monitoring of CGAP at the global level has been modest or negligible. Nonetheless, 71 percent also rate the activities of CGAP as relevant to the priorities of developing and transition economies (Box 1). Consistency with the Bank's Strategic Focus for Global Programs and the Subsidiarity Principle 2.9 Among the Bank's ten global public goods and corporate advocacy priorities, CGAP is a corporate advocacy program under the category of "investment climate" and the subcategory of "financial sector reform." It advocates an approach to microfinance that focuses on building pro-poor financial systems and the human, technical, institutional, and information infrastructure necessary to support them. This immediately raises the subsidiarity issue, whether an activity should be carried out by the global program rather than, as the preferred option, implemented through country operations. To what extent do CGAP's current activities compete or substitute for regular Bank instruments? What is the value added of the CGAP partnership to the Bank's clients beyond what the Bank could do in its regular country operations acting through partnerships at the country level alone? 2.10 Initially, in Phase I, CGAP saw itself as adding value through its ability to give grants to and otherwise support MFIs directly, rather than brokering its assistance via national governments, as is the case with the Bank's regular country operations. CGAP could operate more quickly, flexibly, and agilely at a level below that of most of the Bank's lending operations. Accordingly, grants to MFIs and MFI networks constituted 67 percent of CGAP expenditures under Phase I (Table 5). 2.11 In Phase II, however, grants to MFIs and MFI networks decreased sharply to 28 percent of total expenditures, while expenditures on technical tools, advice, and exchange increased from 5 percent of total expenditures in Phase I to 47 percent in Phase II. This shift in focus has been a response to the views among member donors and the MFI industry that CGAP should not be providing direct funding to MFIs and MFI networks, as had been the case under Phase I. Specifically, the Phase II evaluation found that a number of donors Directions Paper for FY02-04, March 28, 2001. By contrast, each global program is supposed to meet, at the concept stage, all six of the approval criteria for global programs established by Bank Management in April 2000. 11 complained (when interviewed) about competition from CGAP in working with the better- performing MFIs. Nonetheless, whether as a strategic turn or in response to criticism from member donors, there is ample evidence that, following the status review of CGAP in 1998, during which the member donors and the Bank considered whether to support a Phase II, CGAP has repositioned itself as primarily a generator of knowledge and tools, and as a provider of in-kind technical expertise and capacity-building, primarily to MFIs but also increasingly to member donors. Box 1. Partner Survey Results In June 2002, OED administered a survey questionnaire to the 44 members of the CGAP's Consultative Group, the Policy Advisory Group, and the Investment Committee. Fifty-seven percent (25 members) responded to the questionnaire, the results of which are summarized below. Relevance · 71 percent agreed that the activities of the program are relevant to the priorities of developing and transition countries. · 43 percent said that it would take CGAP 5-10 years to achieve its current objectives, and 30 percent more than 10 years. Outcomes and Impacts · 73 percent said that the principal benefits of CGAP accrue at the global level. · 96 percent rated the success of CGAP as high or substantial in knowledge creation and dissemination, 84 percent as high or substantial in implementing global rules, standards, and norms, and 79 percent as high or substantial in advocacy. Participation · 83 percent agreed that the objectives of CGAP can be achieved more cost-effectively by the partners of the program working together within the framework of the program than by the individual partners acting alone. · 71 percent rated the involvement of developing and transition countries as modest or negligible in the design, governance, implementation, and monitoring and evaluation of the program at the global level. · 64 percent rated the involvement of beneficiaries as modest or negligible in the design, implementation, and monitoring and evaluation of the activities of the program at the country level. Governance, Management, and Financing · 58 percent agreed that the objectives of the program are realistic compared to the currently available resources. · 83 percent said that the Consultative Group is accountable to donors for its results and impact, and 17 percent to developing countries. · 66 percent rated the implementation, monitoring, evaluation, and oversight of CGAP as high or substantial. World Bank's Role · 88 percent rated the World Bank's global mandate and convening power as high or substantial comparative advantages that the World Bank brought to CGAP relative to other partners in the program. · 79 percent said that the World Bank's involvement was critical for the success of CGAP. · 40 percent said that there are drawbacks to the World Bank's involvement in CGAP. Source: See Annex O for the complete survey results, including the written responses to the open-ended questions. 12 2.12 The program's Phase III strategic priorities (Table 7) reflect further significant changes in the focus of CGAP, when compared to the previous two phases of its operation. CGAP now sees itself as promoting a diversity of financial institutions that serve the poor, not only the NGO/MFI model, but also other financial intermediaries such as credit unions and the commercial banking sector. It sees itself as the standard-setter for the microfinance industry, playing a unique role in facilitating consensus on performance standards for MFIs and on guidelines to donors. Through the Microfinance Gateway, CGAP also sees itself as the resource center and information clearing-house for the microfinance industry. Supporting national level policy and institutional reform also looms large among the Phase III themes, as seen in the emphasis on legal and regulatory frameworks that underpin the financial sectors of developing countries. 2.13 CGAP appears to have taken the criticism to heart (as noted in its Phase II evaluation), that its emphasis on the NGO/MFI model during Phases I and II had made it less relevant for member donors that worked with commercial banks and the formal financial sector. In addressing the connection of microfinance and financial sector deepening, the Phase III strategy notes: It is becoming increasingly apparent that large-scale sustainable microfinance can be achieved only if financial services for the poor are integrated into the overall financial system. This means that microfinance, or financial services for the poor, becomes the lower end of the entire financial sector, opening up access and markets to increasingly poor and geographically remote clients. It means (1) achieving financial sustainability not through high interest rates alone, but by leveraging technology and streamlining business processes to increase cost efficiency in the face of competition; (2) attracting new players and new sources of commercial and quasi-commercial debt and equity capital; (3) promoting transparency that allows the outside world to understand and judge the performance of microfinance vis-à-vis other financial and non-financial services and other development interventions; (4) governmental policies that promote financial sector deepening and expansion; and (5) seeing microfinance as greater than microcredit and microenterprises ­ as financial services for the poor.23 23. CGAP Phase III Strategy, p. 6. 13 2.14 These changes in CGAP's focus from Phase I to III raise two important questions. First, are these changes consistent with CGAP's comparative advantage as a global partnership? Second, now that CGAP has shifted its focus from activities such as grant- making that were "below" that of the Bank's lending operations, to a mixture of activities, some of which are similar to the Bank's country operations (such as supporting financial sector reform) and others of which are similar to the Bank's Networks (such as knowledge creation and dissemination), what is the value added of the CGAP partnership to the Bank's clients beyond the Bank's regular country-level and Network operations? 2.15 With respect to the first question, it is clear that, after nearly ten years of operation and from the lower than anticipated financial resources mobilized from its member donors, CGAP has not demonstrated a comparative advantage as a major financing facility for MFIs, as was originally envisaged when it was founded in 1995. The Bank contributed US$30 million to CGAP's Phase I in the expectation that other donors would contribute an additional US$70 million to the facility. While the Bank commitment to CGAP was fully financial in nature (and explicitly expected to be such), other multilateral and bilateral donors seeking CGAP membership were afforded the option of supporting CGAP either in cash or in kind, through the pledging of their respective microfinance portfolios to be collectively managed by CGAP. The minimum member donor contribution was set at US$2 million, in funds or eligible programming, plus an initial contribution of US$250,000 (either in cash or in kind) toward the cost of the CGAP Secretariat. 2.16 Out of the ten founding donor members, only the Netherlands (in addition to the World Bank) made a cash contribution to CGAP Phase I, with four others making in-kind contributions. Four founding member donors (ADB, AfDB, Canada, UNDP) made no contribution to CGAP Phase I. During the first CG meeting of CGAP, it was agreed that the required US$250,000 contribution to the Secretariat would be waived for the multilateral member donors. Subsequently, the World Bank, realizing that waiting for US$250,000 from the other bilateral member donors would seriously hamper Secretariat operations, decided to Table 7. Evolution of Strategic Themes, CGAP, 1995-present Phase I (1995-1998) Phase II (1999-2002) Phase III (2003-2008) Learning and disseminating best Support institutional development Fostering a diversity of financial practice to governments, donors, in the microfinance sector institutions that serve the poor and MFIs Support changes in CG Member Facilitating the access of the poor Contributing to supportive policies Donor practices to further improve to a wide range of flexible, for microfinance activities and the quality of their microfinance convenient financial services institutions in selected countries operations Improving the availability and Coordinating donor initiatives and Increase understanding on poverty quality of information on the mainstreaming microfinance best outreach of MFIs performance of institutions practice into donor policies and Improve the legal and regulatory Promoting a sound policy and operations framework of the industry legal framework for microfinance Channeling funds to broaden and Facilitate `commercialization' of the deepen the reach of MFIs that industry serve the poor and strive for sustainability Source: CGAP Annual Reports, 1997-2002; CGAP Phase III Charter. 14 reallocate an additional US$3 million of its US$30 million contribution to the Secretariat (in addition to its US$750,000 contribution to the Secretariat also made during Phase I). 2.17 Thus, the Bank provided US$65.3 million (68 percent) of the US$96.1 million that CGAP received during Phases I and II combined, bilateral donor members provided US$28.6 million (30 percent), and other multilaterals and private foundations the remaining US$2.15 million (2 percent).24 By way of comparison, the World Bank has committed an average of US$220 million a year to the micro and SME finance components of 97 investment projects between 1996 and 2003 inclusive, and it disbursed an average of US$140 million a year during this same time period.25 Over roughly the same time period, USAID's annual grant funding to MFIs averaged US$155 million, reaching more than 3.7 million micro- entrepreneurs.26 2.18 With respect to the second question, retailing grants to MFIs is clearly more consistent with the spirit of OP 8.45 on the use of DGF grants ­ that the DGF should provide grants out of the Bank's net income to external partners, either directly to global and regional partnerships located outside the Bank, or indirectly through arm's length secretariats located inside the Bank. It is also more consistent with the newly clarified rules on the allowable use of Bank budget resources, trust funds, and DGF grants for global programs that DGF grants must flow to entities outside the Bank for funding costs of externally managed entities.27 Yet, CGAP's shift in focus towards technical assistance and knowledge creation and dissemination activities is consistent with similar shifts that have occurred in the other infrastructure and private sector development global programs that are housed inside the Bank. The value added of these activities will be addressed in Chapters 3 and 4. Consistency with the Bank's Sector Strategy 2.19 CGAP's objectives and strategies are broadly consistent with the Bank's most recent Financial Sector Strategy. Upfront, this Strategy confirms that the Bank will support MFIs primarily through improving the framework for microfinance, training and building financial institutions, and avoiding the provision of funds for on-lending either at subsidized rates or where local institutional capacity is inadequate.28 2.20 The 2001 Strategy notes that, for many years, countries tried unsuccessfully to provide agriculture, small-scale industry, and the poor with access to credit. With the support of the 24. See Chapter 4, Table 11. 25. See Chapter 5 and Annex L. 26. See www.usaidmicro.org. 27. While this clarification is a welcome development, it still leaves some unanswered questions in relation to CGAP and other in-house global programs. Will CGAP's annual DGF allocation, which, unlike the case for most in-house programs, is transferred into a Bank-administered trust fund, have to comply with the rules for DGF grants or for trust funds? If the former, will secretariat costs have to be financed entirely from the Bank's administrative budget and donor trust funds? Will there be a phase-in period for these new rules? 28. It should also be recognized here that various social funds and Community-Driven Development (CDD) projects financed by the Bank have components which act as ad-hoc microfinance (albeit through small-scale grants) for both income-generation and asset formation (such as housing), clearly at odds with the intent of OP 8.30 discussed below. 15 Bank and other donors, programs and institutions were set up to provide credit to these sectors with little attention to the provision of other financial services, such as safe savings instruments paying reasonable rates or insurance to help manage risks. A major reason for the failure of these programs was the attempt to provide access to credit below the true cost of doing so, under the rationales that (1) the poor "could not pay" high interest rates and (2) it was politically difficult for state-run programs to charge high rates, in spite of the fact that such high rates are paid regularly to informal moneylenders. 2.21 A second reason for the programs' failures was the multiplicity of their goals (e.g., access to credit, improving the welfare of the poor through transfers via credit, and below market rate credit to offset distortions against agriculture and small-scale industry). A third reason was the neglect of the underlying reasons for lack of credit ­ the weak legal framework and the biases in the incentive framework against agriculture and small firms. These programs did not bring lasting benefit to borrowers, improve the growth of agriculture or small-scale industry, or increase employment, and usually did not address demand for non-credit financial services by the poor. 2.22 In the late 1980s, the Bank began to think more strategically about its interventions in the financial sector, especially as regards what were then termed Development Finance Institutions. The Report of the Task Force on Financial Sector Operations (1989) made note of the distortions generated by targeting credit and subsidizing interest rates, citing that these were prone to abuse and failed in achieving their stated objectives. As a result of the Task Force report, Operational Directive 8.30 was redrafted in 1992 and again in 1998 (as OP 8.30). The recasting of OP 8.30 rested on three principles, namely: (1) all financial sector lending interventions should be consistent with the CAS and justified as strategically important, (2) expected overall net benefits in the financial and real sectors resulting from a proposed operation should be positive, with those operations that are primarily within the real sector doing no harm to the financial sector, and (3) the analysis of all economic, financial and institutional factors of the intervention should be of the highest professional standard. 2.23 A reading of the current OP 8.30 establishes these same principles and further clarifies the parameters for pursuing Financial Intermediary Loans (FILs), a primary Bank lending instrument to the financial sector. The objectives of FILs include: (1) supporting reform programs in the financial sector or related real sectors, (2) financing real sector investment needs, (3) promoting private sector development, (4) helping to stabilize, broaden, and increase the efficiency of financial markets and their allocation of resources and services, (5) promoting the development of the participating financial intermediaries, and (6) supporting the country's poverty reduction objectives. Specifically, OP 8.30 states that Bank involvement in the financial sector, via FILs, aims to remove or substantially reduce subsidies, whether provided through interest rates, directed credit, or institution-building grants. 2.24 The 2001 Financial Sector Strategy notes that two models have gradually emerged as alternatives to Development Finance Institutions for improving access to credit sustainably for the poor: (1) the Grameen Bank model for microcredit using collective liability (i.e., peer- group lending) and (2) for larger loans, or where the use of collective liability is not effective, a model based on establishing powerful incentives to select good borrowers and to collect payments. These models use much higher interest rates than the subsidized programs, because they must cover the higher costs of collection and risk of default. Various financial institutions 16 are used to reach the borrowers, including commercial banks, specialized credit institutions, NGOs, grassroots saving groups, cooperatives, and credit unions. The Financial Sector Strategy concludes that, because the approach is seen as profitable, the private sector becomes involved. 2.25 The 2001 Strategy also focuses attention on improving the "enabling" environment ­ stabilizing the macro-economy and reducing the biases in the incentive system against small- scale industry and agriculture, for example by relating agricultural prices more closely to the world market prices. Development of the legal and judicial framework helps, especially protection of property rights and collateral, particularly in agriculture where land titling may be needed. Development of credit bureaus increases the incentives to repay loans. But, in many countries, credit, while essential, is not the only financial service necessary to address the needs of small-to-medium scale enterprises and low-income families. In some cases, for instance, in rural areas in the poorest parts of Africa the total absence of bank branches makes access to means of payment, cash transfers and savings instruments an even more basic need. 2.26 The Bank's 2001 Strategy resonates with the spirit of CGAP's Guiding Principles.29 Indeed, CGAP arose in the mid-1990s as part of the shift in the Bank's strategic thinking towards financial sector lending in general, and microfinance in particular. Yet, CGAP's direct grants to MFIs and MFI networks, while based on performance-based criteria, nonetheless represent subsidies. CGAP does make a strong case for these subsidies ­ that grant making is justified if targeted to MFIs that both serve the poor and offer the hope of expanding access to microfinance services. To keep the grant making of CGAP in context, it is also important to recognize that many of the member donors pursue grant making to MFIs in their own programs. Such grant-making can be effective if it affords MFIs a "breathing space" within which they can align their business model toward maximizing sustainability, through a focus on achieving operational self-sufficiency. 3. Outcomes, Impacts, and Their Sustainability MONITORING AND EVALUATION 3.1 According to CGAP's Charter, the CGAP Operational Team submits Quarterly Reports to EXCOM, as well as an Annual Report for EXCOM approval and further submission to the Council of Governors (CG). In both cases, these reports are expected to detail CGAP's major results, activities, and financial information, including the approvals of new projects and grants. Additionally, an annual work plan is sent to the CG prior to the CGAP annual meetings to solicit feedback on its content. For the past six consecutive years, CGAP has produced annual reports, many of which are available on its web site. These annual reports provide substantive information on the history, objectives, governance structure and activities of CGAP, and should be commended for contributing to the overall transparency of what CGAP is and how it operates. 3.2 Key questions regarding monitoring and evaluation are the extent to which CGAP has: 29. Donor Working Group on Financial Sector Development 1994. 17 · Clear and coherent objectives and strategies that give focus and direction to the program, and provide a basis for evaluating its performance; · A structured set of qualitative or quantitative indicators; · Systematic and regular processes for data collection and management; · Regular, independent program-level evaluations; and · Effective feedback mechanisms to reflect evaluation findings on strategic focus, organization, management, and financing of the program. 3.3 Across each of its three phases, CGAP has sought to expand access to microfinance for the world's poor while simultaneously striving for financially sound MFIs that can continue to deliver such services over the long run. These objectives are clear and evaluable, since one can measure the financial soundness and sustainability of MFIs as well as the extent to which they are actually serving poor people (e.g., by means of a poverty assessment tool). But there remains an attribution issue. CGAP does not itself provide financial services to the poor. It simply supports MFIs who provide such financial services, both directly through grants to MFIs and indirectly through its financial and technical support to the donor community and to the MFI industry at large. This implies two criteria for assessing the effectiveness of CGAP ­ (1) what evidence exists of MFIs that have been successful at both reaching the poor with microfinance as well as attaining financial sustainability in the process; and (2) to what extent has CGAP contributed to this success?30 3.4 Assessing the impact of CGAP remains a thorny issue, which has generated a great deal of discussion among CGAP members in the course of the review of the earlier draft of this study. On the one hand, a considerable body of anecdotal evidence ­ both documented in the body of this paper and supplemented by comments and observations from CG and EXCOM members ­ asserts the positive impact of CGAP tools, advice and expertise on both donor practices and industry standards. Some examples are the MIX, the MicroBanking Bulletin, the Microfinance Gateway, the IDB-CGAP Rating Fund and the annual microfinance training in Boulder, CO, and the Donor Peer Review process. On the other hand, insofar as they reflect inputs toward the objective of microfinance expansion to the poor, the above examples cannot simply be assumed to generate either the expected outputs or the subsequent impact. To assess the latter, monitoring data are essential and an evaluation framework within which to assess and confirm impact of interventions, especially in regard to their impact on reducing poverty, is critical. CGAP members commenting on the earlier draft of this report expressed a concern about the difficulty of measuring outputs or impacts because of the "degrees of separation" of CGAP members from the institutions that actually provide finance, and the likely high cost of measuring outputs and impacts. A variety of new techniques for measuring impacts now exist. CGAP needs to move forward on both fronts. EVALUATION OF CGAP 3.5 CGAP has been the subject of evaluations during both Phase I and Phase II (Annex G). The most recent evaluation, an external assessment commissioned by the member 30. While beyond the scope of this study, data sources such as the MIX, MicroBanking Bulletin and the more extensive database of the Microcredit Summit ­ with nearly 2,600 MFIs ­ could provide the foundation for a more detailed assessment of both MFI effectiveness and outreach and CGAP's role. See also CGAP Occasional Paper No. 8, Financial Institutions with a Double Bottom Line: Implications for the Future of Microfinance. 18 donors,31 reviewed the Phase II achievements in relation to four major activities: (1) grants to selected MFIs to spur learning and innovation, (2) disseminating microfinance best practices, (3) mainstreaming microfinance within member donor agencies, and (4) fostering national- level policy environments conducive to microfinance.32 The evaluation concluded the following: · CGAP's work has been highly relevant, of great importance to the development of the microfinance industry, with its secretariat operating at a high level of technical capability, and highly respected throughout the industry for its vision and leadership. · The effectiveness of CGAP has been less clear. While CGAP has been successful in both direct grants in MFIs and promotion of best practices in microfinance, its achievements have been much more modest on donor mainstreaming and the fostering of favorable national policy environments. · It is too early to address the efficiency of CGAP. Capacity building accounts for about 26 percent of total expenditures across Phases I and II (see Table 5). These investments merit some gestation time before they can be fully assessed in terms of outcomes. Grants to Selected MFIs to Spur Learning and Innovation 3.6 While grants to MFIs and MFI networks decreased markedly under Phase II of CGAP, in line with the shift in thematic priorities under Phases II and III, this activity has nonetheless yielded many lessons for the industry at large, as well as for the member donors. The Partnership Agreements that articulated the funding relationship between MFIs and CGAP were innovative in that they were tied to results-based performance benchmarks, which at the time was not a common practice among donors active in grant making to MFIs. This delegated the decisions regarding application of these grant funds to the beneficiary MFI. The only "conditionality" was the expected progression of the MFI toward administrative and operational self-sufficiency by improving the quality and breadth of its portfolio and building efficiency into financial operations. Many member donors have adopted such performance-based agreements for their MFI programs, precisely because CGAP took the risk to field-test this innovation.33 3.7 Under Phase I, the period during which the bulk of direct grants to MFIs and MFI networks occurred, CGAP tended to make these grants independently, without the participation and consultation of its member donors. The grant making was essentially a demand-driven process, with interested MFIs completing a standardized application, which the CGAP Operational Team then vetted and approved, with guidance from the Investment 31. Fox, et al. 2002. 32. Given that Phase III activities have only recently begun (July 2003), this section focuses on CGAP under Phase II. 33. In its own internal assessment of MFI grants and lessons learned, CGAP has noted that it has not always been consistent in how it has enforced partnership agreements signed with MFIs prior to their receipt of funds. In some cases, thresholds were changed at mid-course (for reasons external or internal to the MFI). MFIs missed thresholds without repercussions, while in other cases partnerships were closed due to poor performance. 19 Committee. But both the level of direct grants and the methodology for approving them underwent a transformation late in Phase II, with the advent of the Appraisal and Monitoring Service in 2001, which serves as an example of how member donors can collaborate while exploiting the core technical services that CGAP offers. This service conducts joint due- diligence of MFIs with interested donors and investors and institutes a consortium approach for funding these MFIs that consolidates the reporting and monitoring requirements. 3.8 This approach changed the grant making to a supply-driven phenomenon, with the member donors recommending to CGAP those MFIs they wished to appraise for potential funding. An advantage of this consortium approach is that it permits CGAP to leverage its scarce resources against those of the member donors in co-financing these selected MFIs. From 2002 to the present, 17 MFIs and MFI networks have received US$19.4 million through the Appraisal and Monitoring Service, about two-thirds of which came from member donors in the form of co-financing (Table 8). Yet one must consider the wisdom of stepping away from a demand-driven approach, which opened the door for MFIs to benefit directly from CGAP. 3.9 CGAP staff members clearly enjoy CGAP's grant-making activities and find these help to keep them grounded in the day-to-day operational aspects of microfinance. On the downside, the absence of country-level CGAP representation makes follow-up on such grants both time-consuming and a potential drain on other aspects of the CGAP work program. One issue to be raised in this regard is how better to mobilize the member donors in both funding and supervision of these CGAP grants to MFIs. The Appraisal and Monitoring Service is an example of how CGAP's limited human resources and counterpart financing of member donors can be leveraged to assist MFIs. 3.10 Three issues arise from CGAP's grant-making activities: (1) grants made to MFIs and their potential for crowding-out of hard money by other donors, (2) the capacity (and merits) of member donors undertaking similar activities in the absence of CGAP, and (3) the potential global payoff for these local-level investments. In response to the crowding-out of member donor investments, the potential for this has been greatly diminished through CGAP's revised approach to MFI grants under the Appraisal and Monitoring Service. CGAP takes its cue from the requests of member donors, thus matching CGAP's flexibility in dealing directly with MFIs with the member donors' desire to use CGAP as a complement for (rather than substitute for) their respective microfinance portfolios. Another element of the success of the Appraisal and Monitoring Service appears to be the co-financing approach, through which the donors contribute independently, but collectively, to supporting the Table 8. Grants Approved Under CGAP Appraisal and Monitoring Service, 2002-03 US$ Millions Grant Category # Recipients CGAP Commitments Member Donor Leveraged Funds MFI 10 2.115 4.100 Network 6 3.853 8.273 Through network 1 1.065 Total 17 7.033 12.373 Source: CGAP 2002 Annual Report, CGAP Web site. 20 appraised MFIs, as opposed to channeling their own funds through CGAP. Finally, the "global payoff" for these local-level investments is simply the lessons learned that can be applied to other MFIs under similar circumstances. Best Practice Dissemination 3.11 Major activities by CGAP in disseminating best practices in microfinance have been (1) technical tools (Table 9), publications and the CGAP web-site and (2) the capacity- building initiative. In fact, under Phase II, technical tools and capacity-building garnered the largest share of CGAP expenditures (Table 5). The development of technical tools by CGAP, and their subsequent use by member donors, MFIs and the overall industry, is considerably more challenging to assess, particularly due to the absence of a solid monitoring and evaluation framework on the part of CGAP. Such monitoring and evaluation was to be spearheaded by the Donor Committee (proposed under the CGAP Phase II proposal), which never became operational. Clearly, CGAP has been adept at generating knowledge products for the microfinance sector; yet the uptake of these products is neither well-understood nor well-documented. Anecdotal evidence from member donors does, however, confirm that these tools have been helpful in their work with MFIs and have provided a convenient "off- the-shelf" product that member donors with limited staffing, frequent turnover and smaller programs can exploit at little or no additional cost. 3.12 The programmatic partners of CGAP surveyed by OED felt that CGAP has been the Table 9. Examples of Technical Tools Generated by CGAP Technical Tool Content Observations MIS Well-received by member donors Planning and financial modeling Low level of familiarity with specific items Technical Guides External audit among MFIs, donors, esp. in Middle East Appraisal and Africa Poverty assessment Complexity and over-design an issue Focus Notes Short overviews of current topics and Multi-language, distributed to >5,000 trends in microfinance institutions in 100 countries In-depth, analytical treatments of Occasional Papers microfinance topics intended for specialists 124 leading MFIs worldwide provide Boosts transparency in the industry; now MicroBanking Bulletin financial data confidentially to CGAP for part of MIX exchange developing performance benchmarks Peer-monitoring among MFIs 1,350 hits in December 1999 CGAP Web site Repository of CGAP objectives, 6,600 hits in December 2001 composition, activities, and publications Average visitor accessed site twice a month Web-based bibliographic database Frequently cited as CGAP's most valuable Microfinance Document exchange, discussion product Gateway groups, overall "broker" for microfinance "Open forum" to share ideas and information exchange resources within the microfinance industry Source: CGAP publications and Web site. 21 most successful, among Figure 1. In your opinion, how successful has the program all its activities, in been in achieving the following objectives? creating and disseminating knowledge of best practices in Knowledge management microfinance ­ more Advocacy successful than, for Global rules & standards example, in improving donor coordination and Research & development in supporting national Capacity building level policy and Donor coordination institutional reform with Mobilizing resources respect to microfinance National policy reform (Figure 1). 0% 20% 40% 60% 80% 100% 3.13 The Guiding High Substantial Modest Negligible Principles on Regulation Significantly different among objectives: P = 0.0% and Supervision of Microfinance, agreed by donor agencies as part of the CGAP 2002 Phase III Charter, define two degrees of self- sufficiency for MFIs and establish indicative `timetables' for the period over which MFIs should be able to achieve them. CGAP has further clarified these definitions. "Operational self-sufficiency" requires MFIs to cover all administrative costs and loan losses from operating income. This is calculated by dividing operating income by operating expenses. The Guiding Principles suggest that, based on international experience, successful MFIs should be able to achieve operational self-sufficiency within three to seven years. "Financial self-sufficiency" requires microfinance programs to cover all administrative costs, loan losses and financing costs from operating income, after adjusting for inflation and subsidies and treating all funding as if it had a commercial cost. It is suggested that successful intermediaries should achieve financial self-sufficiency within 5 to 10 years. These have proven to be useful performance benchmarks for the member donors, yet the degree to which each member donor applies such criteria in its country-level operations remains to be seen. 3.14 Yet this consensus on the regulation and supervision speaks nothing about the donors' degree of consensus on using microfinance to reduce poverty. In fact, there appears to be a broad note of discord on this issue, despite the general view of donors that microfinance can and should reach the poor and that it can have a significant impact on bringing them out of poverty.34 Take, for example, the current debate over the so-called "50-percent solution," proffered by USAID (Box 2). The U.S. government has mandated that methods must be developed to verify that at least 50 percent of its microfinance grants to MFIs are reaching the poorest, defined as either those living on less than US$ 1 per capita per day, or those found in the bottom half of the distribution below the national poverty line. Five of six EXCOM members in a virtual discussion group hosted through the Microfinance Gateway voiced opposition to such policies, generally noting that the legislation would place unnecessary transactions costs on MFIs and stifle their growth. 34. See earlier findings on this account from Khandker 1998. 22 Box 2. Partners in Poverty-Focused CGAP Disagree on Need to Verify Program Impact on the Poor The U.S. Microenterprise for Self-Reliance Act of 2000 made available $310 million over a two-year period for grants to microfinance institutions. The 2002 reauthorization of the act allocates an additional $176 million for FY03 and stipulates that USAID, in consultation with microfinance institutions and other organizations, develop and certify at least two methods for measuring the poverty levels of microfinance clients served by microfinance institutions that receive USAID grants. These methods are meant to ensure that at least 50 percent of USAID micro-enterprise assistance is set aside for the "very poor," defined as those who either live on less than $1 a day or who are in the bottom half below a country's poverty line. Prior to the reauthorization, the Microfinance Gateway (an Internet-based forum for microfinance professionals) hosted a virtual discussion, the objective of which was to "better inform [the microfinance community] on ground-level realities and thus enable [it] to take well-reasoned positions to promote a financial sector that serves the needs of the poor." Six members of the CGAP Executive Committee, representing the partners, participated and offered their views on the subject. Five voiced opposition to the required outreach verification, saying it would "stifle [MFI] freedom and growth," "increase compliance costs, deter investments," and result in "formulaic restrictions (to) choke private sector incentives to serve the poor." One EXCOM member indicated that the new U.S. legislation would advance the poverty focus of microfinance by: (a) encouraging more effort in deepening outreach, and (b) increasing openness and transparency in its assessment. Another discussant noted that this is not regulation, but rather an investment target for subsidies paid for by U.S. taxpayers and added, "Is there a cost to getting to know your clients? Yes. Is investing in that knowledge bad for business? Absolutely not! Any [microfinance institution] who does not believe that has a range of other investors who will offer different constraints and expectations. That's the nature of the market for sourcing funds (both publicly and privately). If you need a subsidy and can provide some informed analysis about the wealth of your clients, go to USAID. If you don't feel knowing the wealth of your clients is worth the effort, go somewhere else." Source: CGAP Web site, Microfinance Gateway. 3.15 To place this move in perspective, USAID, over the three years from 2000 to 2003, made available US$486 million to MFIs as grant support for on-lending to micro-enterprises. This exceeds CGAP grant-making to MFIs over its eight-year history by a factor of twenty. This level of funding is also "nipping at the heels" of the range of required funding to reach the Microcredit Summit Challenge goals. CGAP has strongly promoted its poverty assessment tool, and this would appear to be an excellent opportunity to deploy it for maximum visibility and validity, yet not even the EXCOM concurs that such verification of clientele is necessary. 3.16 As noted in its Phase II evaluation, CGAP is in the process of merging the various regional capacity-building programs (Table 10) into a global initiative under a common strategy, branding, and marketing platform. Assessments for all of these regional initiatives are currently under way and, when completed, will provide the necessary inputs into the formation of this strategy. A major challenge confronting CGAP is how best to expand the base of service providers for capacity-building, especially in South and East Asia, where demand is high. Similarly, coverage among the stronger MFIs appears high, while outreach to the less experienced and smaller MFIs appears to be sparse. This also may speak to a need to more tightly design and adapt current course content to the disparate demands among the MFIs in terms of capacity deficits. Nonetheless, the open access, via the Internet, to training materials and course guides is a step in the right direction and can help to build a broader base of service providers. 23 Table 10. Regional Initiatives Funded Under CGAP Region Programs US$ Million Africa AFCAP (East Africa); CAPAF (West Africa); Microsave Africa 5.206 Asia Microfinance Training Center (China); Capacity-building Asia (India, Philippines); BRI Int'l Visitors Program 1.736 E. Europe Microfinance Center (Poland) 0.438 Global Product Manager, new course development 0.500 Other CGAP donor training; staff training UNDP/UNCDF; external audit capacity-building 1.427 Total 9.307 Mainstreaming Microfinance in Member Donor Agencies 3.17 CGAP has pursued a number of initiatives in this area over the years. First, it offered to conduct portfolio reviews for member donors, yet only two member donors ­ the World Bank and UNCDF ­ submitted to such reviews, both in 1997. Again in 2003, both the Bank and UNDP asked for such reviews. Second, in response to member donors, CGAP agreed in 2000 to manage a joint donor portfolio database as a means of consolidating data on microfinance worldwide. Since that time, only 11 member donors (less than one-half of the total CG) have provided information for the database, and often these data are inconsistent with the information independently disseminated by the member donors.35 It now appears that CGAP has abandoned this activity. Third, appraisal and monitoring services were offered by CGAP to member donors in the preparation of microfinance projects under their respective aid portfolios. Through FY03, 17 appraisals had been completed. Fourth, donor working groups ­ chaired by different CG members and including academic researchers, microfinance practitioners, and other member donors ­ were established in four key areas: impact assessment methodologies, poverty yardsticks and measuring tools, agricultural development bank reform, and savings mobilization.36 The Phase II evaluation concluded that these working groups were not well integrated into overall CGAP activities, especially in regard to the CGAP Operational Team. 3.18 Fifth, CGAP has been quite successful in delivering donor training, specifically courses in the basic principles of microfinance and appraisal of MFIs. Sixth, CGAP has completed standardized appraisal formats for MFI assessment, as well as a poverty assessment tool and poverty audit. Drafts of three other tools are also available to member donors (financial statements disclosure guidelines, consensus guidelines on regulation and supervision, and consensus guidelines on supporting savings institutions). Seventh and finally, CGAP has assigned relationship managers (i.e., a CGAP staff member) as a focal point for contacts with each member donor. 35. For example, the database reports IDB disbursed about US$3.0 million annually for microfinance in 1999 and 2000, while the IDB reports some US$295 million in microfinance projects approved from 1997-99, and another US$30.0 million in 2000. 36. The CGAP 2002 Annual Report, however, only cites three working groups to which CGAP provides support: savings mobilization, micro-insurance, and public banks. 24 Box 3. Selected CGAP Initiatives and Programs Microfinance Information eXchange (MIX): The MIX, launched in October 2002, integrates two CGAP transparency initiatives. The first ­ the MIX Market (http://www.mixmarket.org/) ­ is a web-based information service linking MFIs, donors, and investors. The service provides in-depth information on MFI performance, including financial statements, ratings, outreach and impact data, together with terms and conditions of donor and investor funding. To date, 153 MFIs, 36 investors and 48 markets facilitators (raters and networks) have posted their information on the MIX Market site. A second initiative ­ the Microbanking Bulletin ­ is billed as the premier benchmarking source for the microfinance industry. Its benchmarks are widely used by investors, donors and other service providers to facilitate greater standardization and a better understanding of the development of the microfinance sector. IDB-CGAP Rating Fund: A joint IDB-CGAP initiative, the Rating Fund has approved partial financing for 66 ratings and evaluations of MFIs. Most of these MFIs are publicly disclosing these ratings for the first time. Ample funding remains for more MFIs interested in sharing the cost of a professional rating and/or evaluation. Microfinance Regulation Database: CGAP has signed an agreement with the IRIS (Institutional Reform and the Informal Sector) Center at the University of Maryland to develop a database on country laws and regulations relevant to microfinance. When the database is functional, it will be available through the MIX. Rural Pro-Poor Innovation Challenge (PPIC): Launched in 2000 jointly with IFAD and now administered by them independently, the Rural PPIC supports pro-poor innovations in rural microfinance. From 2000-02, 24 MFIs received an aggregate US$ 1.1 million. In FY03, 10 recipients were chosen from an applicant pool of 500. Agricultural Microfinance Research: In response to the increasing need and interest of CGAP's member donors in guidelines and models for rural and agricultural finance, CGAP has been assessing nearly 80 promising agricultural microfinance providers with a combination of desk-based research, communication, and brief consultant field visits. CGAP hopes to uncover and profile models and approaches that offer potentially sustainable financial services for poor agriculture-dependent households. Source: CGAP Quarterly Report, September 2002. 3.19 On the whole, as stated in its own Phase II Evaluation, CGAP efforts toward donor mainstreaming have not yielded much success. That CGAP should have little success in this area raises fundamental questions as to the overall "buy-in" of member donors to its objectives and as to the constraints all donors face in improving their effectiveness in areas such as private sector development and financial sector development that require high technical inputs and relatively lower financial sums. Information sharing of member donor's microfinance portfolios was an agreed commitment upon their joining CGAP under both Phase I and II, and it remains a condition of membership under the CGAP Phase III Charter. The lack of compliance with this commitment strikes at the heart of CGAP's purpose and would appear to be another instance of a disconnect between the positive rhetoric about CGAP and the member donors' actual participation in it. Fostering National Policy Environments Conducive to Microfinance 3.20 On the whole, the CGAP Phase II evaluation notes that policy work has received less emphasis than other CGAP activities. The major activities in pursuit of this objective have been a model for supervising credit unions in Guatemala, a workshop for West African MFIs on developing a legal framework for microfinance in the region, and widely circulated publications such as "The Rush to Regulate: Legal Frameworks for Microfinance" and "CGAP Consensus Guidelines on Microfinance Regulation and Supervision." There has also been considerable work on issues of MFI transparency, as evidenced by efforts on the MicroBanking Bulletin, standardized MFI appraisal formats, and overall MFI ratings. According to the CGAP Operational Team, these are key inputs toward the longer-run achievement of a favorable policy environment for microfinance. Responding to specific 25 country requests, CGAP has provided analysis of proposed laws or regulations in about two dozen countries in the last three years. Finally, CGAP has recently developed a database on microfinance-relevant laws and regulations in 25 countries (soon to be on Microfinance Gateway) with another 25 countries in process. More recently and on a global front, the EXCOM suggested principles for the UN year of Microfinance, which were presented to and adopted by the G8 Group in Spring 2004. DONOR PEER REVIEWS 3.21 Subsequent to the CGAP Phase II evaluation, and perhaps in an effort to bolster the effectiveness of its mainstreaming activities with member donors, CGAP undertook a series of Donor Peer Reviews. At present, 17 member donors have submitted to peer reviews since they were initiated in 2002. As of February 2004, the World Bank had not yet taken part in this exercise. Rather than concentrate on constraints at the country level (such as governance, corruption, and macroeconomic instability), the peer reviews focus on what donor agencies can most directly influence: their own procedures, practices, processes and systems. The reviews identify success factors and constraints to good practices in microfinance and provide concrete recommendations for each agency.37 Several donor agencies that took part in these reviews indicate that policy changes are underway in response to these findings. CGAP has circulated a series of documents that analyze the emerging lessons from the reviews ­ among them a summary of the major findings from a policy perspective, aimed at policy-makers and top-level management, and a synthesis report of the first six peer reviews.38 3.22 Key findings emerging from the peer reviews were the following: · Lack of a clear vision by the member donors reviewed on the role of the financial sector ­ and microfinance as an element within it ­ in development; · Increased focus on moving "upstream," concentrating on policy and sector-level work, while de-emphasizing retail-level work with MFIs; · Little accountability for quality; · Inability to work directly with the private sector, a key element in expanding financial access among the poor; · Thin technical capacity among donor staff; · Interest (on the part of the donors) in defining their comparative advantage and in collaboration across donors; and · Recognition of common challenges that need to be tackled by working together. 3.23 Donors themselves are uncertain as to their role vis-à-vis increased access to financial services and its contribution to the MDGs, which may help to explain the perception of decreased interest in microfinance among the agencies reviewed. Microfinance, at least by these member donors, is increasingly viewed as a small, insignificant activity, unable to 37. CGAP 2003c. 38. One MFI practitioner notes that communication from donors, in the post peer review period, is more consistent and that better quality guidelines for microfinance are being formed. 26 produce systemic results.39 In sum, a disconnect between support to microfinance expansion and how it can contribute to financial sector deepening and overall improvement in financial intermediation is often present among the very member donors who have banded together under CGAP. That a review covering 60 percent of CGAP member donors would generate such a finding also implies a disconnect with the increasing evidence that microfinance can indeed produce lasting results in terms of poverty reduction, as cited earlier in this study. 4. Program Governance, Partnerships, Management, and Financing GOVERNANCE AND MANAGEMENT 4.1 At its core, CGAP is a forum for donor agencies to coordinate support to microfinance, share best practices, and develop consensus on policy guidelines.40 This forum ­ the Council of Governors (CG) ­ is the membership body of CGAP. Comprised of all 28 member donors and chaired by the Infrastructure Vice President of the World Bank, the CG approves CGAP's strategy, sets policy, provides input into the annual work plan and budget, and receives and approves annual reports (Annex H). CGAP's Executive Committee (EXCOM), comprised of four CG members elected on a constituency basis - one member appointed by the World Bank, four microfinance industry leaders, and the CGAP Executive Director (ex officio) - provides strategic direction and support for the CGAP Operational Team and approves the annual work plans and budgets. The CGAP Operational Team, housed within the World Bank and headed by the Executive Director, carries out activities in accordance with the approved work plans and budgets in the pursuit of CGAP's mission. 4.2 The Investment Committee (IC), composed of seven World Bank and IFC staff, exercises fiduciary responsibility for CGAP's grant making to external entities. As custodian of the CGAP Investment Fund, the IC ensures adherence to the overall strategy for funding MFIs as established by the CG. The IC evaluates and approves the CGAP Operational Team's recommendations on grants to be made from the Investment Fund. 4.3 CGAP first established its EXCOM at the onset of Phase II to better facilitate the communication between the CG, its Policy Advisory Group (PAG), the Investment Committee, and the CGAP Operational Team.41 Originally, this six-member constituency- based committee was expected to meet once or twice annually, with membership rotating among the member donors every 1-2 years. Under both Phase I and II, the PAG, composed of recognized, national-level microfinance practitioners and experts worldwide, provided technical advice and guidance to the CG. With the ratification of the CGAP Phase III Charter, the PAG has now been merged with the EXCOM. 4.4 Some 66 percent of the programmatic partners surveyed by OED rated the implementation, monitoring, evaluation, and oversight of CGAP as high or substantial 39. Ibid, page 2. 40. CGAP Charter, p. 2 (para. 3.2) 41. See CGAP Annual Report 2000, among others. 27 (Figure 2). Nonetheless, the Figure 2. From your perspective, to what extent do you Phase II evaluation found that believe that the program has effective? the interaction of CGAP's Operational Team was stronger with EXCOM than with the Implementation CG. For most of the constituencies, the respective Monitoring EXCOM member served primarily as a conduit of Independent information, largely through e- Evaluation mails, and tended to participate in more of a personal capacity Donor Oversight rather than as a true representative of their intended 0% 20% 40% 60% 80% 100% constituency. In addition, the High Substantial Modest Negligible Phase II evaluation found that Not significantly different among activities: P = 40.6% the EXCOM was not expected to poll its constituents before deciding on issues under consideration, nor did it take an active role in gathering together the views of its constituencies. 4.5 The next evaluation of Phase III of CGAP needs to address the question as to whether the merging of the PAG and EXCOM has brought about an effective improvement in the governance structure of CGAP. On the one hand, microfinance expertise became a requirement in Phase III for those wishing to serve on the EXCOM. This move, combined with the inclusion of additional expertise in the guise of industry leaders, should aid in deepening the EXCOM's ability to more effectively guide the activities of CGAP. On the other hand, if the constituency-based model of the EXCOM continues to be ineffective (as the Phase II evaluation concluded), communication flows may actually worsen. The newly configured EXCOM may be a means of expediting decision-making at the risk of decreasing member donor representation in the process. The internal strength of the CGAP Operational Team might lead to much of CGAP's effective decision-making (as well as its operations) truly residing with the Operational Team. PARTNERSHIPS AND PARTICIPATION 4.6 CGAP membership has historically been open to both multilateral and bilateral donor agencies, and more recently to private foundations. While there was ambiguity regarding the financial responsibilities of member donors during Phase I of CGAP, the Phase II renewal brought greater clarity to the member donors' role by explicitly defining their responsibilities. The Phase III Charter enumerates four explicit conditions with which prospective and current members of the Council of Governors are expected to comply (Box 28 Box 4. Conditions for Membership on the Council of Governors · Contribute funding in cash, through flexible mechanisms, to carry out CGAP's operations. · Adopt and actively promote implementation of standards and basic principles of donor support to MF as outlined in CGAP consensus documents including, "Micro and Small Enterprise Finance: Guiding Principles for Selecting and Supporting Intermediaries" or successor documents to be agreed upon. · Participate actively in the CGAP Annual Meetings with representatives who are appropriately placed within the organization and have the necessary knowledge to support mainstreaming of MF best practice within their respective agency. · Agree to open the institution to sharing experiences with other donors, including providing aggregate information on its portfolio of MF activities, reporting on how it is applying the consensus documents, and providing contact information for all staff working in microfinance. This information should be submitted to CGAP on an annual basis prior to the CGAP Annual Meeting. Source: CGAP Charter, September 2002. 4). But the Phase II evaluation found that, on most of these criteria, past adherence to the agreed responsibilities had been weak.42 4.7 Contribution (in cash or in kind) to CGAP. Drawing on CGAP Annual Reports for 2002 and 2003, all 16 bilateral members had contributed in cash to the program, providing, on average, US$242,000 annually, primarily under Phase II. Among the ten multilateral donors (aside from the World Bank), the AsDB began contributing in FY02 with US$250,000 and the IDB made a one-time contribution of US$50,000 in 1996 under Phase I. Those who have not yet contributed are AfDB, EBRD, EC, ILO, and UNCDF/UNDP. While many of the multilateral member donors are precluded from making cash contributions under their respective charters and legal frameworks (primarily since CGAP funds are administered by the World Bank), one must question their wisdom in agreeing to participate in an alliance with other donors that they are prohibited from supporting financially. 4.8 Official adoption of Donor Committee Guidelines and willingness to apply them. The Phase II evaluation found that the CGAP Secretariat did not have a system for determining whether adoption had taken place, or whether these Guidelines were being applied in programming within the donor agency. Note also that the Phase III conditions of membership stipulate that the consensus documents, including, Micro and Small Enterprise Finance: Guiding Principles for Selecting and Supporting Intermediaries (1995), would be adopted and applied in member donor agencies. CGAP's web site currently lists five consensus documents: · Key Principles of Microfinance; · Guiding Principles on Regulation and Supervision of Microfinance Institutions; · Definitions of Selected Financial Terms, Ratios, and Adjustments for Microfinance; · Developing Deposit Services for the Poor: Preliminary Guidance for Donors; and · Disclosure Guidelines for Financial Reporting by Microfinance Institutions. 42. The responsibilities of Council of Governor members (Phase III), as given in Box 3, are the same as those of the Consultative Group members under Phase II. Hence, in this section, the findings of membership compliance under Phase II are reported. 29 4.9 Active participation in the CG, with appropriate personnel and continuity of representation. The Phase II evaluation concluded that 15 member donors met this criterion, 8 did not, and the performance was indeterminate for 6 others.43 Even where the representative was sufficiently grounded in the subject of microfinance, the frequent turnover of staff within member donor agencies made for a weak "institutional memory" in the CG. 4.10 Willingness to mainstream microfinance. While there appeared to be progress made in consolidating microfinance within donor programming, the Phase II evaluation found that CGAP did not have a mechanism to track such compliance. To their defense, many member donors had established internal microfinance policies and issued operational guidelines. Those with Microfinance Sector Policies in place during Phase I of CGAP included USAID and BMZ/Germany (1994), IDB (1997), and SDC/Switzerland (1998). Under Phase II, the list extends to CIDA/Canada and UNCDF (1998), AfDB, IFAD, and AsDB (2000). The IDB also produced a Rural Finance Strategy in 2000. Similarly, in 1999 and in preparation for the Microcredit Summit Meeting of Councils, the World Bank issued its Microfinance Institutional Action Plan, which it has updated annually ever since. 4.11 The Phase II evaluation concludes that donors have moved away from direct funding of MFIs ­ precisely when scaling-up of microfinance is paramount ­ and toward capacity- building (as is the case for CGAP). This finding is also consistent with the previously mentioned Peer Review exercise, in that donors are repositioning toward the "high end" or "upstream" activities of policy and sector work, and away from retail microfinance activity. The donors now almost universally accept the commitment to sustainable MFI operations, combined with effective targeting of the poor, at least in rhetoric. While a review of the above mentioned microfinance strategies for member donors indicates that, at least as conveyed in these documents, CGAP does not play a particularly active role in support of operationalizing the strategies, subsequent comments from CG members on an earlier draft of this study indicate otherwise, for example, in regard to the recent drafting and approval of the Key Principles of Microfinance, as well as their endorsement by the G8 summit in June 2004. 4.12 Provision of information to CGAP on donor microfinance activities. While some member donors had demonstrated a willingness to share experiences through dissemination of studies and research (e.g., USAID, Germany, IDB), the sharing of donor microfinance portfolio information, especially in the context of the desired CGAP Donor Portfolio database, was characterized as poor. 4.13 At this juncture, an obvious question arises as to the enforceability of the responsibilities outlined in the CGAP Charter. Specifically, who is tasked with enforcing the responsibilities laid out for the CG, EXCOM, Investment Committee, and CGAP Operational Team? Are there remedies or penalties for non-compliance? The Charter is decidedly silent on all these issues. 4.14 On the issue of membership and compliance with the conditions thereof, only one member donor, UNCTAD, has withdrawn from CGAP membership since its inception. Hence, we are left with the dilemma not only as to why some participate in spite of their 43. These numbers include UNCTAD, which was still part of the CG at the time of the Phase II evaluation. 30 weak adherence to the Figure 3. To whom is the governing body accountable for membership criteria, but also its results and impact? why other potential members of the CG are excluded from the process in an apparently ad Donors hoc fashion when current member donors are not penalized for non-compliance Developing with their conditions of countries membership. Scientists/ 4.15 At the highest professionals governance level, CGAP remains essentially a "donor 0% 20% 40% 60% 80% 100% club." The programmatic Most important Intermediate Least important partners surveyed by OED confirm this reality (Figure 3). Significantly different among stakeholders: P = 0.0% As an international public sector organization that is concerned, among other things, with improving national level policies with respect to microfinance, there are gaps in the representation that might be expected within the CG. The most obvious gap is the lack of formal representation from the Bank's client countries. This lack of representation by developing country governments is common to all six infrastructure and private sector development global programs reviewed by OED, with the recent exception of the Cities Alliance, where the Government of Brazil became the developing country member in 2003. While sponsorship of a program by international organizations like the World Bank and the United Nations (sometimes enshrined in formal co-sponsorship agreements) may enhance the relevance of a global program in the Bank's client countries, this is not a sufficient condition for developing countries' ownership of the program or for ensuring its development effectiveness. Relevance and ownership by the Bank's client countries is more assured if they demand the program or if it removes some critical constraint in their countries, such as lack of access to international markets, lack of infrastructure, lack of human skills, or simply a lack of computers and staff to more effectively perform the tasks that global programs are advocating. 4.16 Global programs that are housed in U.N. organizations (such as Roll Back Malaria and Stop TB in WHO) as well as global programs that are independent legal entities (such as the Multilateral Fund for the Implementation of the Montreal Protocol, the Global Water Partnership, the Global Forum for Health Research, and UNAIDS) tend to have direct representation of developing country governments on their governing bodies. Such representation has generally had a positive impact on the relevance and ownership of global program activities from the point of view of developing countries, although there is concern among some CGAP members that including developing country government representatives would result in political responses.44 44. Bilateral donors also represent their respective governments and are no less immune to political incentives, which would likely impact the role of developing country representatives within the CG. 31 4.17 CGAP's Council of Governors also lacks representation from MFIs or umbrella organizations of financial institutions such as the World Council of Credit Unions, although four individual industry leaders are members of EXCOM.45 With ready portfolios of microfinance clients, track records of financial and administrative performance, and knowledge of the challenges that they and other practitioners face in expanding the market of microfinance, MFI representation could add first-level pragmatic insight that currently is not available on the CG.46 MFIs could also provide a counterpoint to the donor-beneficiary relationship that prevails in much of CGAP's interaction with them.47 4.18 While some may argue that the PAG historically provided such insights (and, with its absorption into the EXCOM under Phase III, perhaps even more effectively than in the past), the Phase II CGAP evaluation notes that the PAG's technical and advisory role had diminished significantly in recent years. The same evaluation also found that most current PAG members, as directors of growing MFIs, had both limited time to devote to CGAP and were relatively removed from the broad industry-wide issues that CGAP attempts to address. Thus far under CGAP Phase III, according to several CG and EXCOM representatives who commented on an earlier draft of this study, the revised composition of the EXCOM has substantially improved the level of engagement on the part of MFIs, including some from developing countries. This is a positive development. 4.19 Yet, the primary governing body, the CG, lacks representation in two key elements of current and future microfinance delivery ­ developing country governments and MFIs themselves. Without their active participation, it is questionable whether CGAP's stated Phase III strategy to widen the access of financial services for the poor can be achieved. FINANCING OF CGAP 4.20 While member donors, over most of CGAP's existence, have been afforded the option of in cash or in-kind contributions, the program nonetheless requires cash resources to defray its ongoing activities. Since FY95, member donor financial contributions have totaled US$96.1 million, with the Bank providing US$65.3 million, bilateral member donors an additional US$28.6 million and other multilaterals and private foundations the remaining US$2.15 million (Table 11). Through both the Special Grants Program (SGP) and later the DGF, the Bank has provided 68 percent of CGAP's core funding. Under Phase I, in two of the four calendar years, virtually 100 percent of CGAP core funds were resourced from the Bank's SGP. Donors were slow to make financial commitments to CGAP, despite the aforementioned member donor responsibilities. Seventeen donor organizations were admitted as CGAP members, yet made no financial contributions at the time of their admission. 45. In fact, while the President of the World Council of Credit Unions does serve on the EXCOM, he does so in his own capacity as a recognized leader in the microfinance industry, not because the World Council has a permanent seat on EXCOM. 46. However, it is arguable whether MFIs would be able to meet the financial obligations of CG membership as set forth in the Phase III Charter. Other global programs have different contributing requirements for membership for low-income countries or those representing the poor. 47. The Cities Alliance provides an example of a global program housed inside the Bank that includes representation from umbrella organizations of clients ­ in its case, associations of urban municipalities ­ on its highest level governing body. 32 4.21 Phase II brought a decrease in CGAP's dependency on Bank funding, with other member donors ramping up their collective contributions. Financial commitments from bilateral member donors have steadily increased, with over one-half of all these commitments coming from five donors ­ Netherlands, Norway, United States, Switzerland, and Canada. In FY03, the Bank still provided 54 percent of CGAP's core funding, and is projected to provide 50.4 percent in FY04 (Figure 4). 4.22 This being said, several CG representatives, in comments on the earlier draft of this study, expressed a concern that the analysis of CGAP financing alone does not adequately reflect the level of donor buy-in and commitment to CGAP, particularly with respect to the more recent Phase III activities in which they indicate there is substantially greater donor ownership. Others have challenged the notion as to whether the financial contribution either to the secretariat (as one of several criteria outlined in the Phase III Charter) should be a defining element of CGAP membership, particularly regarding any eventual inclusion of either MFI representatives or developing country governments in the CG. Still others have offered the suggestion of charging user fees for CGAP's products and services, but whether and by how much this would dampen the uptake of tools and materials by MFIs remains an open question. A "fee-for-services" model could serve to better inform CGAP as to the applicability of the products it generates, thereby fine-tuning Figure 4. The World Bank still provides more than half of supply with demand. Expanding CGAP's core funding. CGAP membership, the 14 financing obligations for the program, and the degree to 12 Total Contributions which these remain linked in the 10 future will need further exploration. 8 64 % 62 % 56 % 54 % Millions 64 % 50.4 % 4.23 While member donors 6 World Bank Contributions have been slow in making cash US$ 4 contributions to CGAP, they have nonetheless participated 2 with CGAP in co-financing certain activities directly from 0 1999 2000 2001 2002 2003 2004 their own funds.48 Overall, member donors have matched CGAP expenditures with a collective US$48.1 million, or about 72 percent of the level of total CGAP expenditures (Table 12). The significance of these matching funds is that they were committed by the individual member donors, rather than pooled under the CGAP umbrella. More non-Bank funds have been mobilized through this co-financing mechanism than through direct cash contributions to CGAP. Member donor financial contributions (other than World Bank) under this co-financing regime over Phases I and II total US$27.4 million, which is just 56 percent of total co-financing over the same period. This leads to the question of why member donors are more comfortable in their role as co-financiers than in complying with their membership responsibilities under CGAP. 48. For example, one EXCOM member notes that approximately 15 percent of their staff time is currently allocated to CGAP-related activities. 33 4.24 To what extent has the Bank's presence been critical in coalescing support from other donors? From an inaugural membership base of 9 donors, CGAP has tripled in size to include 26 multilateral and bilateral donors and 2 private foundations. Something has attracted them to CGAP, and it is likely that it has been the strong technical expertise that CGAP brings to microfinance and the simple fact that the donors active in microfinance have a convenient forum, under the Bank's auspices, through which they can interact, exchange lessons learned, and otherwise attempt to bring a degree of homogeneity to the field. 4.25 Is there a compelling reason for a larger financial resource base for CGAP? In other words, is CGAP in need of larger budget in order to achieve its Phase III objectives? Some 58 percent of the programmatic partners surveyed by OED agreed that the objectives of CGAP are realistic compared to its currently available resources (Box 1). As CGAP has now successfully (and, according to many of the member donors, correctly) moved from primarily grant-making to MFIs toward a much stronger program of technical assistance, knowledge sharing, standard-setting and information exchange, budget growth may not be an issue. However, the scale effects of the knowledge being generated by CGAP, in the form of MFI assessment tools and guidelines should be of value to the member donor community, and this value should be recognized through greater, collective financial burden sharing of CGAP's annual budget envelope. 4.26 CGAP's shaky financial base is the greatest threat to its ability to maintain operations over the long run. That member donors, aside from the World Bank, have not been strong in providing cash contributions to CGAP must be recognized as a chief obstacle. CGAP might be better served by adopting a financing strategy that concentrates more on activities in which member donors could participate via co-financing, as distinct from requiring contributions to CGAP's core budget. Additionally, greater consideration of the "fee-for- services" model discussed earlier may be warranted. Perhaps the experience under the Appraisal and Monitoring Service can serve as a lesson for how best to engage the donors in activities for which they are also willing to bring money to the table. But the bigger question is why member donors are not willing to channel microfinance funds through the CGAP mechanism. The simple answer may be the best: the member donors simply don't want to finance the program, and are more than content to free-ride off any positive outcomes that CGAP offers. Would member donors adopt a different strategy if the legitimacy of CGAP were raised by increasing stakeholder representation in governance and participation, while forging greater transparency in decision-making? 4.27 Furthermore, how should CGAP position itself in light of the future capital requirements in microfinance? With a range of US$650 million to US$3.8 billion needed to meet the expected demand for microfinance worldwide, should CGAP be mobilized to aid in resourcing such capital? Or, as evidence from Phase II indicates and comments from CG members support, should CGAP continue to primarily make intellectual contributions toward the expansion of microfinance, through tools, best-practice dissemination and efforts to promote greater transparency in MFIs? Over its eight-year history, CGAP's own funding base has totaled some US$93 million, of which some US$27.0 million has been expensed as additional capital funding to MFIs. There is no doubt therefore that a "capital gap" must be closed in order to bring the benefits of microfinance to a greater portion of the poor. But the figures above would suggest that the capital being mobilized for microfinance will have to be sourced outside the auspices of CGAP. While this is at odds with CGAP's original objective 34 of mobilizing a US$100 million fund for expanding access to microfinance for the poor, it is nonetheless clear that the private sector, and specifically commercial banking, will be the key player in closing this financial gap. Table 11. Member Donor Contributions to Core CGAP Budget, FY95-FY03 US$ Millions Percent of Total Donor Member Group Phase I Phase II Phase I Phase II (FY95-98) (FY99-03) TOTAL (FY95-98) (FY99-03) TOTAL World Bank 30.750 34.593 65.343 80.6% 59.7% 68.0% Bilaterals 6.755 21.862 28.617 17.7% 37.7% 29.8% Other Multilaterals 0.650 0.900 1.550 1.7% 1.6% 1.6% Private Foundations 0.000 0.600 0.600 0.0% 1.0% 0.6% Total 38.155 57.955 96.110 100.0% 100.0% 100.0% Source: CGAP Annual Report 2002, Annex 3, and CGAP Annual Report, Table 1. Note: These include the estimated dollar value of in-kind contributions (primarily staff secondments) provided by France, IFAD, Netherlands, U.K., and U.S.A. in various years. Table 12. CGAP and Member Donor Co-financing, by Expenditure Category, Phase I and II (in US$ millions) Phase I Phase II Total Expenditure Member Member Member Category CGAP Donors CGAP Donors CGAP Donors Capacity 2.829 2.175 9.608 5.438 12.437 7.613 MFI 11.349 8.300 4.339 5.107 15.688 13.407 Network 2.788 0.000 6.128 14.373 8.916 14.373 Other 1.758 0.085 13.187 3.464 14.945 3.549 TA 0.189 0.037 2.269 0.000 2.459 0.037 Through network 5.750 2.500 1.815 0.000 7.565 2.500 Donor 0.000 0.000 3.904 6.700 3.904 6.700 Total 24.663 13.097 41.251 35.082 65.914 48.179 Source: CGAP. Note: Excludes Banco do Nordeste (US$51.6 million) under Phase I and ABL (US$50.0 million) under Phase II RISKS AND RISK MANAGEMENT 4.28 The risks to the World Bank of its involvement with CGAP result primarily from the significant financing burden that the Bank bears in the program, and the close, in-house relationship that the CGAP Operational Team enjoys in the use of various Bank services. 4.29 Financial risk. Grants approved through CGAP's Investment Committee are formally executed by means of legal agreements signed between CGAP and the recipient institution. These agreements are drafted by the Bank's legal department, and the Executive 35 Director of CGAP signs them as a representative of the World Bank. Given the financing burden already carried by the Bank, as majority funder of CGAP, any reduction in its support without a commensurate increase in burden sharing by the remaining member donors would require cuts in CGAP core programming. Yet, this is precisely the logic behind the DGF's 15 percent criterion ­ to effectively spread the financial risk among the partners to the program, such that others can "pick up the slack" if there are changes in the funding dynamics of any individual member. While member donors (other than the World Bank) collectively increased their financial contributions under Phase II, the Bank nevertheless remained the dominant financier, contributing 61 percent of total member donor contributions to date (see Table 11). 4.30 Institutional risk. This would appear to be substantial, given the level of "embeddedness" of Bank staff and Bank processes within CGAP. The Chair of the CG is the Infrastructure Vice President, the staff of the Operational Team are Bank employees, and the CGAP Investment Committee is composed of senior-level World Bank staff considered to be experts in the field of microfinance. While the risk is indeed substantial, the mere fact that the Bank is in control of so many of the issues that could potentially generate the downside risk plays to its favor. Yet again, the sheer weight of Bank involvement calls into question whose voices are being heard (and listened to) within the CG, EXCOM and the Investment Committee, as well as why other member donors are not actively seeking a greater role in such leadership. 4.31 Also, CGAP's ability to deal directly with MFIs opens the door for the potential of a disconnect between on-going Bank projects and CGAP activities in the respective countries. Minimizing this risk, therefore, becomes a function of the coordination of CGAP's activities in the context of the CAS and PRSP exercises. 5. Role of the World Bank 5.1 This chapter assesses the Bank's performance as a partner in CGAP in accordance with four criteria: · Comparative advantage: Whether the Bank is employing its comparative advantages in relation to other partners in the programs (endorsed by the Development Committee September 2000).49 · Global-country linkages: Whether the global program has effective operational linkages to the Bank's country operational work, where appropriate (one of the six approval criteria established by Bank Management in April 2000). · Oversight: Whether the Bank is exercising effective and independent oversight of its involvement in the program, as appropriate, for in-house and externally managed programs. · Exit strategy: Whether the Bank is facilitating effective, flexible, and transparent disengagement strategies, as appropriate (established by the DGF Council in October 1998). 49. This is also one of the six criteria for approving a global program at the initial concept stage established by Bank Management in April 2000 and one of the eight eligibility criteria for grant support established by the DGF Council in September 1998. 36 COMPARATIVE ADVANTAGE OF THE BANK 5.2 According to the programmatic partners surveyed by OED, the Bank has utilized its comparative advantage as a partner in CGAP more at the global level than at the country level (Figure 5), particularly in regards to the Bank's global mandate and reach and its convening power relative to its expertise in country and sector-level analysis, its multi-sector capacity, and its country-level knowledge. With the Bank as its backdrop and institutional anchor, CGAP gains significant legitimacy by accessing the Bank's substantial goodwill among both donors and developing countries. The Bank's convening power is self-evident, with its own membership base of 184 countries, and CGAP has made extensive use of such power in bringing its member donors together and building a consensus around the basic parameters of how best to build strong financial institutions for the poor. Almost in unison, donors are voicing their collective belief that microfinance serves as one among several alternatives through which the poor worldwide can effectively emerge from poverty. 5.3 But in retrospect, the Bank's role at the inception of CGAP may have been excessive. Effectively, the Bank established the conceptual basis for CGAP, chaired the initial Donor Working Group, and led the thought process to operationalize CGAP. The Bank's founding contribution of US$30.0 million, along with housing the secretariat at the Bank's headquarters in Washington, D.C., conveyed early on that CGAP was largely a Bank program. Had the Bank held off in making the US$30.0 million available only after other member donors had actually mobilized the expected remaining US$70.0 million, CGAP may have developed a stronger ownership at start-up. LINKAGES TO THE BANK'S Figure 5. The Bank is employing it comparative advantage COUNTRY OPERATIONS better at the global level than at the country level. 5.4 In its engagement Global with the World Bank, the mandate shift in CGAP's focus Convening power toward more policy-level Mobilizing financial work at the country level resources will make possible a Analytical expertise stronger and more consistent Multi-sector capacity linkage to country operations than has been Country-level knowledge possible up to the present, 0% 20% 40% 60% 80% 100% but it will also require a reallocation of scarce staff High Substantial Modest Negligible time from other CGAP products and services. The Significantly different among comparative advantages: P = 0.0% stronger focus on financial sector policy plays both to CGAP's strengths in terms of global knowledge of best practices and the Bank's capacity to open doors to senior level policy makers to more effectively place microfinance on the table in ongoing consultations with member countries in the process of preparing both the CAS and/or the PRSP. Yet, if this policy dialogue is to be effective, 37 stronger ties with core country teams, especially during the CAS preparation stage, will be an important element. 5.5 But the broader membership of CGAP should also be able to avail itself of much stronger linkages to their respective microfinance operations. This brings to light two issues for the future of CGAP ­ its human and financial resources, and its own institutionality. On the resource side, effective country-level engagement will require either substantially greater staffing and budget levels than CGAP currently possesses or equipping in-country member donor staff with the right microfinance messages, information, and technical tools that they can share and disseminate to national-level policymakers and other stakeholders. Maintaining CGAP under the auspices of the World Bank headquarters will also perpetuate the sense of entitlement among CGAP staff (which remain Bank staff in actuality). This may be precisely why the member donors have been hesitant to take up a greater burden sharing of the financing of CGAP. 5.6 Perhaps the best indicator of the Bank's role as a development partner with CGAP would be the extent to which microfinance has been "mainstreamed" or internalized within the Bank's country-level operations. Clearly, the Bank has been a strong partner with CGAP in terms of the financial and institutional support that it has provided (e.g., staffing, access to the Bank's legal and other business services, management of trust funds). Yet, the expected outcome of this participation should have been a deeper appreciation for the role that microfinance can play in creating assets for the poor by providing a potential pathway out of poverty. 5.7 A review of Bank lending indicates that the Bank's Board approved some 97 new projects with "micro and SME finance" components from FY96 to FY03 inclusive, corresponding to CGAP's Phases I and II (Table 13). Europe and Central Asia, Sub-Saharan Africa, and Latin America and the Caribbean have been the dominant regions in the micro and SME portfolio in terms of the number of lending operations, while Latin America and South Asia have received the largest volume of commitments. More than three-quarters of these operations had micro and SME finance components that constituted less than 50 percent of the total project costs, suggesting that a general strategy for microfinance lending at the Bank entails "bundling" microfinance within a multi-component, multi-sectoral context, as opposed to a separate focus on microfinance. 38 Table 13. Bank-Financed Operations, Micro and SME Finance, FY96-03 Region # Operations Percent of Total US$ Millions Percent of Total Africa 24 25% 154.5 9% East Asia and Pacific 12 12% 246.9 14% Europe and Central Asia 27 28% 265.7 15% Latin America and Caribbean 20 21% 715.8 40% Middle East and North Africa 9 9% 100.4 6% South Asia 5 5% 294.0 17% Total 97 100% 1,777.3 100% Percent of Micro and SME Finance Loans # Operations Percent of Total US$ Millions Percent of Total 1-10% 26 27% 146.1 8% 11-25% 30 31% 212.6 12% 26-50% 19 20% 345.6 19% 51-75% 5 5% 71.8 4% 76-99% 15 15% 921.9 52% 100% 2 2% 79.3 4% Total 97 100% 1,777.3 100% Source: Annex O. 5.8 Figure 6 shows the impact on Bank lending of the Report on the Task Force on Financial Sector Operations (1989), the recasting of the Bank's strategic thinking towards interventions in the financial sector, and the redrafting of Operational Directive 8.30 in 1992 (discussed in Chapter 2). New commitments to micro and SME finance plummeted to a mere US$20.1 million in 1993. Since that time, however, there appears to be a discernable upward trend, albeit fluctuating considerably from one year to the next (as Bank commitments tend to do in every sector). Plotting a log-linear trend line to the data, new commitments have increased at an average annual rate of about 15 percent a year between 1993 and 2003. While actual disbursements have been relatively flat, the rising level of commitments appears finally to be having an impact on disbursements in 2003 and 2004. 5.9 It may be tempting to attribute these rising trends to the presence of CGAP. In fact, CGAP's view has been that the Bank spends too much money on "bad" microfinance, precisely in multi-sector projects. As CGAP sees it, a large proportion of CGAP-Bank interactions have involved trying to dissuade project staff from doing "bad" projects. A review of 52 Country Assistance Strategies (CAS), CAS Progress Reports and CAS Updates (reported and approved by the Board over the period October 2001 to March 2003) indicates that CGAP's linkages to country programs, at least as conveyed through these documents, has been rather minimal. Explicit mention of CGAP is made in only one CAS (Mexico). While 29 of the 52 CASs treat microfinance explicitly within their strategy, more often than not it is in relation to collaboration with IFC in the context of the Bank's overall assistance strategy. Given that these are consensus documents developed over several months of in- country consultation, it is somewhat striking that, despite the rather notable collective emphasis given to microfinance across these documents, CGAP is not more frequently cited. 39 Figure 6. Bank Lending to Micro and SME Finance, New Commitments and Annual Table 14. Sectoral distribution of lines of credit (LOC) Disbursements, 1990 to 2004 operations, by number of operations, FY93-03 600 Sector Percent all Percent Percent LOC regular micro 500 Rural 31.4 26.4 42.5 Commitments Finance 16.9 20.2 9.6 Urban 14.0 19.0 2.7 400 PSD 9.7 11.7 5.5 Social protection 9.7 3.1 24.7 Millions 300 Energy 4.7 4.3 5.5 Environment 3.8 4.3 2.7 US$200 Water and sanitation 3.0 3.7 1.4 Other /1 6.8 7.4 5.5 100 Total 100.0 100.0 100.0 /1Public sector, transport, economic policy, education, HNP Disbursements 0 1990 1992 1994 1996 1998 2000 2002 2004 Source: Annex O. Volumes correspond only to the micro and SME finance components of Bank lending activities. Date for FY04 is only through March 31, 2004. 5.10 One might also ask whether CGAP's interactions with Bank staff have resulted in significant quality improvements in the Bank's microfinance operations. It would appear that they have not, as evidenced by a recent OED study of Bank Lines of Credit (LOC) operations. For the period FY93-03, a total of US$13.4 billion was committed for LOC operations (representing 8.4 percent of total Bank commitments during the period), of which US$11.7 billion was for regular LOC and US$1.7 billion was for microfinance LOC. Over this time period, the share of microfinance LOC has been more or less steady in terms of overall commitments and it increased slightly in terms of number of operations. The rural sector accounted for the largest share (31 percent) of all LOC operations, while the financial sector accounted for only 17 percent (Table 14). The pre-dominance of LOC in sectors other than finance persists across time and for all Regions. 5.11 As a result, microfinance lending has become a substantial part of both the number and amounts of Bank lending operations for LOC. This may be partly the result of the increased focus on poverty alleviation and the perception that microfinance is an effective and efficient tool for fighting poverty. 5.12 An analysis, conducted by CGAP in conjunction with the OED LOC report, reveals an overall inferior quality among the microfinance LOCs. A scoring exercise was performed to determine the quality of microfinance LOC. The projects were scored on an index from 1 (exceptionally bad) to 5 (exceptionally good), with 3 being a weak but acceptable quality and 2 unacceptable. The average score for 64 microfinance LOC was 2.77, or on the low end of weak. Only 11 of the 64 were rated "good" or better. Thus, CGAP has been instrumental in identifying areas of needed improvement in the Bank's microfinance LOC portfolio, yet the 40 time frame of analysis (FY93-03) ­ encompassing seven of CGAP's eight-year history ­ gives pause to reflect on CGAP's role (if any) in the decision to initially pursue these LOCs. 5.13 Another country-level exercise, the Poverty Reduction Strategy Paper (PRSP), provides the basis for Bank assistance as well as debt relief under the HIPC initiative. PRSPs are designed to be country-driven, comprehensive in scope, partnership-oriented, and participatory.50 A country prepares a PRSP every three years. To date, some 44 PRSPs have been prepared worldwide. As stated in its FY02 Annual Report, CGAP staff contributed to the World Bank PRSP Sourcebook, numerous specific country PRSPs, and project appraisals. Other technical advice included the review of Bank-funded microfinance and rural finance projects in 15 countries.51 In a review of 37 PRSPs, while 28 mention some form of strategic approach to the microfinance sector, none mention CGAP by name. This would seem to imply that, while the microfinance sector is gaining more weight in discussions concerning effective methods to achieve poverty reduction, the influence of CGAP is indirect at best in this process. 5.14 This cursory review of CASs and PRSPs leads to the bigger question of the capacity of CGAP to effectively advocate microfinance issues across the 89 countries represented in these CAS and PRSP exercises. Raising awareness among those embedded in the CAS process on the potential for poverty reduction through microfinance interventions requires frequent and consistent dialogue. Given that the CAS process is typically a three-year cycle, missing the window of opportunity to influence the policy discussion could effectively stifle both the policy engagement on microfinance and the lending pipeline that could help to facilitate it. CGAP, at present, has a staff of 22, only a subset of whom would have the requisite expertise to actively participate in such consultations. Rather than focusing on direct participation, CGAP would likely be more effective by targeting its advocacy for microfinance among the country economists and other technical experts who typically comprise the CAS and PRSP teams. This would allow CGAP to better leverage its scarce human resources in a wholesale manner, perhaps in the form of periodic workshops and scheduled seminars, in which other member donors could also take part. 50. This contrasts with country assistance strategies (CASs), which are Bank-driven, albeit with greater country-level consultation of late. For low-income, HIPC-eligible countries, however, PRSPs are supposed to feed into and influence the CASs. 51. Afghanistan, Bulgaria, El Salvador, Eritrea, East Timor, Ghana, Kazakhstan, Mali, Malawi, Mexico, Nicaragua, Pakistan, the Philippines, Sri Lanka, and Tajikistan. 41 5.15 Not only the World Bank, but also the International Finance Corporation (IFC) supports the development of microfinance in developing countries. In 2000, IFC independently launched its Small and Medium Enterprise Capacity Building Facility (CBF) to finance, through one-time grants, promising pilot projects with replication and scaling-up potential. Partnerships are also sponsored under the CBF, which are characterized as broader, enduring strategic alliances with on-the-ground institutions that can stimulate other new future projects. CBF funding is drawn from IFC net income, similar to that of DGF funding for CGAP. With annual allocations of approximately US$7.0 million, CBF has funded eight partnerships and over 30 pilot projects, totaling US$9.7 million under FY01 and FY02. Whether partnership or pilot project, the CBF funding, on average, constitutes slightly more than 20 percent of total project cost for any given investment. Interestingly, about one-half of the CBF commitments over this period have been made to MFIs or Microenterprise capacity- building institutions, some of which (e.g., Acción, Women's World Banking, FINCA) have also benefited from CGAP assistance. The IFC, through the CBF, appears to be following a similar path to that pursued by CGAP during its Phase I, since grant-making to MFIs and related institutions constitutes a primary activity. The cost-sharing approach under the CBF is also comparable in design to the Appraisal and Monitoring service instituted by CGAP in Phase II. However, the IFC generally finances MFIs at a later and more developed stage in comparison with CGAP. Few of the MFIs in which CGAP has invested would have qualified at the time for IFC investments. THE BANK AS DONOR TO CGAP 5.16 The Bank's financial support to CGAP has always been through grants, first under the Special Grants Program (SGP) during Phase I, and later under the DGF in Phase II. In moving from the SGP to the DGF, the Bank recognized the importance of grants as a complement to the Bank's core business of operational lending. When criteria were Table 15: IFC SME Capacity-Building Facility, Commitments FY01 and FY02 US$ Millions Percent of Total FY01 FY02 TOTAL FY01 FY02 TOTAL Partnerships 2.570 2.200 4.770 10 % 9 % 10 % Pilots 1.855 3.075 4.930 7 % 13 % 10 % Leveraged Capital 20.875 19.129 40.004 83 % 78 % 80 % TOTAL 25.300 24.404 49.704 100 % 100 % 100 % established for DGF funding in 1998, CGAP, like other programs that had received grants from the SGP, was grandfathered into the DGF, with the proviso that it would in due course comply with the DGF eligibility criteria (Annex A, Table A.4). 5.17 Six years after the establishment of the DGF, CGAP still does not comply with several criteria for DGF funding. First, Bank support constituted 68 percent of CGAP's core funding from FY95 to FY03. While Bank support declined to 60 percent of total support during Phase II (the period during which DGF support began), and to 54 percent of core funding in FY03, this remains well in excess of the DGF criterion of 15 percent. 42 5.18 To its credit, CGAP has made strides to correct this funding imbalance. Its 2000 Annual Report clearly confirms CGAP's own awareness of the need to diversify its funding base and reduce Bank participation. In fact, the EXCOM and the Operational Team launched an initiative that same year to boost commitments from existing multilateral and bilateral members, as well as recruit new members to CGAP, particularly private donors.52 CGAP cited three reasons for the need to broaden its funding base: (1) to reflect its multi-donor nature, (2) to comply with the 15 percent DGF guideline and (3) to demonstrate a disengagement strategy from Bank funding. The Bank reinforced this initiative in 2001, when Bank President James Wolfensohn, in a speech to the Spring Meetings, made the case for "equal partnership in funding such important partnerships [as CGAP]."53 Despite these efforts, as well as a noticeable increase in bilateral commitments during the Phase II period, goals for decreased Bank share in financing CGAP have clearly not been achieved. 5.19 Second, the arm's-length relationship of CGAP with the Bank is questionable. The CGAP Phase II evaluation asks the question, "Who owns CGAP?" and responds by finding that ownership lies squarely with the World Bank: CGAP [Operational Team] staff are employees of the World Bank, the CG chair is a World Bank employee, oversight of CGAP investments is provided by a group of World Bank employees (on the Investment Committee), more than 60 percent of the CGAP budget comes from the World Bank (through the DGF), and fiduciary responsibility for CGAP financial management is exercised by the World Bank."54 5.20 The Charter reinforces this notion in its declaration that CGAP is not an independent entity, but rather a group of trust funds managed by the World Bank. As the CGAP Charter states, "Bilateral trust fund agreements between CGAP member donors and the World Bank govern CGAP, and these same agreements delegate to the Bank the role of execution and oversight of CGAP."55 This recognition of the Bank's role in CGAP, while simultaneously defining a governance role for the CG, sets up an internal dissonance within CGAP. Technically, and perhaps legally, the Bank is CGAP and CGAP is the Bank, while philosophically, rhetorically and operationally, CGAP is termed a consortium of its member donors. Here again, the voices of several CG members, in providing inputs on an earlier draft of this study, convey a solid sense of ownership, as least in regard to Phase III of CGAP. Yet, one might expect to see this ownership more substantially reflected in the CGAP Charter, for example in a reduced role of the Bank in the execution and oversight of CGAP and a commensurate increase in such responsibilities by the remaining CG members. Another sign of greater ownership would be the CG members' greater input in developing strategies and assessing their impacts in light of their own operational experience on the ground rather than "leaving much of the thinking to CGAP secretariat," as some acknowledged may be the case currently. 52. CGAP Annual Report 2000, pg. 15. 53. CGAP Annual Report 2001, pg. 23. 54. CGAP External Evaluation, 2002, p. 30. 55. CGAP Charter, p. 2. 43 5.21 Again to its credit, CGAP has instituted two arrangements to reduce the dominant role of the Bank, in general, and of the Infrastructure Vice Presidency, in particular, in the governance and management of CGAP, and thereby foster more of an arm's length relationship with the Bank. First, while the Bank's Vice President for Infrastructure remains the chair of the Council of Governors, EXCOM elects a chairperson from among its member donors for a two-year renewable term.56 Second, the Bank's representative on both the CG and EXCOM is a senior Bank staff member (lead economist) located in the Financial Sector Vice Presidency, and therefore outside the Bank's management chain responsible for CGAP ­ the only such example of the separation of oversight and management among the 12 global programs reviewed by OED that are located inside the Bank. In addition, the FSE Vice Presidency provides an annual budget allocation of US$30,000 for him to exercise this oversight role.57 To the outside world, and in particular to the MFI industry, CGAP also seems to have developed an identity from the Bank that is at least as independent, if not more so, than the other five infrastructure global programs. 5.22 Nonetheless, CGAP has not yet complied with the DGF guidelines that were approved in June 2000 in relation to the funding of in-house secretariat costs. Under "promoting and reinforcing partnerships" these guidelines provide the following: Except during the first year of operation where majority funding from the Bank could be necessary as partnership support crystallizes, the Bank should not fund more than fifty percent of such secretariat costs to avoid a program's over-reliance on the Bank. After no more than three years, a decision should be made to either move the secretariat out of the Bank, keep it in the Bank with strong donor support, or discontinue the effort due to lack of donor interest or other reasons. In exceptional cases where there is strong donor interest in maintaining an in-house secretariat in the Bank after three years, then this secretariat should be financed 100% by partners.58 5.23 There is also the subsidiarity issue that was first introduced in Chapter 2. Now that CGAP has significantly scaled back its grant-making activities, the justification for the Bank's continuing participation in CGAP rests largely on its activities characterized by substantial economies of scale and scope ­ such as knowledge creation and dissemination, capacity building, and improving donor coordination ­ that add value to the Bank's clients beyond what the Bank's country operations can do acting through partnerships at the country level alone.59 The evidence presented in Chapter 3 is so far sufficiently compelling with respect to CGAP's knowledge creation and dissemination activities, but less so with respect to donor coordination. CGAP needs to put in place more systematic monitoring and evaluation processes, including a structured set of quantitative or qualitative output, outcome, 56. CGAP Charter, p. 4. The representative of the Department for International Development (U.K.) is currently the chair of EXCOM. 57. Presumably, this means that any conflicts that arose between the INF and FSE Vice Presidencies in terms of the strategic direction or performance of CGAP would be resolved at the Vice Presidential or Managing Director levels. 58. Guidelines for DGF Funding for Programs with In-House Secretariats (approved June 2000), http://wbln0023/rmc/rmc.nsf/DOCs/DGF+Guidelines+on+In-House+Secretariat?OpenDocument . 59. This corresponds to the Bank's third strategic focus for global programs, "multi-country programs that crucially depend on highly coordinated approaches." See Annex A, Table A.3. 44 and impact indicators to demonstrate the continuing value added of its global partnership to the Bank's client countries. 5.24 CGAP has clearly been a positive beneficiary of the differential application of DGF funding rules between those global programs that started before 1998 and those that started afterwards. But this lack of compliance with so many DGF funding rules produces a double- edged dilemma for the Bank that is the subject of the next section. DISENGAGEMENT STRATEGY 5.25 OED's review of the Bank's involvement in global programs distinguishes exit strategies from three perspectives: (1) the program declares "mission accomplished" and closes; (2) the program continues, but the Bank withdraws from all aspects of its participation; and (3) the program continues and the Bank remains engaged, though the degree of engagement in some or all aspects declines over time. 5.26 Regarding (1), the present mandate of CGAP extends through FY08. The Council of Governors will conduct a review in FY06 in order to determine post FY08 options for CGAP such as disbanding, expanding or transforming.60 Regarding (2), the two sponsoring vice presidencies, FSE and INF, have no strategy at the present time for completely disengaging the Bank from CGAP. Regarding (3), CGAP was designated a Window 1 program when the DGF introduced its two-window approach commencing with the FY02 DGF funding allocation. This means that the DGF regards CGAP either (a) as an established program with long-term development objectives, a track record of success, and strong partnership support, or (b) as a priority program with a time horizon for DGF funding that goes beyond three years.61 CGAP's current strategy for disengaging from DGF financial support, which was instituted in FY02, is to reduce its financial dependence on DGF funding by US$400,000 each year. 5.27 It is not the place for this review, which is a background paper for OED's larger review of the Bank's involvement in global programs, to make a recommendation regarding a disengagement strategy for CGAP. In any event, what the Bank, the DGF, and CGAP's partners decide to do in this regard will clearly be influenced by the Bank's Board and Management response to OED's recommendations in its larger review. Rather, this review simply offers some insights into CGAP's current situation. 5.28 At one extreme, CGAP would appear to be a good candidate for mainstreaming into the Bank's regular budget, supported also by donor trust funds, in the way that the Water and Sanitation Program (WSP) and the Energy Sector Management Assistance Program (ESMAP) currently operate. Indeed, in shifting the focus of its activities from grant-making to in-kind technical assistance and knowledge creation and dissemination activities, CGAP has been transforming itself into a program that is structurally very similar to WSP. Mainstreaming CGAP into the Bank's regular budget would also be more consistent with the 60. See Annex D. 61. DGF Guidelines on the Two-Window Approach, http://wbln0023/rmc/rmc.nsf/DOCs/DGF+Two+Window+Approach?OpenDocument. 45 newly clarified rules on the allowable use of Bank budget resources, trust funds, and DGF grants for global programs that "DGF grants must flow to entities outside the Bank for funding costs of externally managed entities." While the DGF might choose to grandfather CGAP once again, or CGAP might escape the application of these rules on a technicality,62 using DGF grants to fund the staff costs of a growing secretariat to provide in-kind technical assistance is clearly inconsistent with the spirit of these rules. The negative side of this option is the risk that CGAP might lose some of its independent identity from the Bank, which it has fostered and which it enjoys in the MFI industry at large. Pursuing this option would also require the Bank to replace CGAP's DGF grant with the Bank's regular budgetary resources, and for CGAP and the Bank to make a more concerted effort to raise trust fund resources from its member donors. 5.29 At the other extreme, CGAP could pursue an independent legal status and exit the Bank, like the Global Water Partnership (GWP), the Global Development Network (GDN), and World Links have done. This would cement a distinct persona for CGAP, one which many member donors express already exists for the program, and might also spark a greater degree of ownership and burden sharing on the part of the other member donors. While the Bank could continue to provide DGF grants to CGAP without being in violation its own funding rules, the Bank would scale back its role in CGAP to a level more commensurate with that of a member donor, while other donors would "step up to the plate" and begin to comply with the conditions of their own membership in CGAP. Greater proportionality in terms of financial burden sharing, given the varying capital bases among the member donors, and perhaps making this explicit within the CGAP Charter, would be steps in the right direction. Encouraging developing countries to become full-fledged members of a CGAP that is its own independent legal entity would also enhance its legitimacy in the eyes of its developing country clients. 5.30 It may be possible to find a sustainable course of action between these two extremes. But staying on its present course seems unsustainable. CGAP would continue its uneasy relationship with its principal source of funding ­ the DGF. In the face of increasing competition for DGF funds, pressures will mount to enforce DGF's funding criteria equitably among all programs ­ those that started both before and after 1998. Enforcing the DGF criteria would result in a drastic reduction in Banks' financial support to CGAP in order to comply with the 15 percent criterion, or, in the extreme, a complete withdrawal of DGF support. Under either of these scenarios, and without an alternative plan of action, CGAP would likely cease operations, given the historical importance of DGF support to its livelihood and the as yet unrealized substantive cost-sharing from non-Bank member donors. Even if the CGAP partnership is objectively judged as adding value to the Bank's clients beyond what the Bank can do through its country operations alone, why should the Bank unilaterally continue to support its activities and disregard its own policies regarding grant- making? Equally perplexing, why do the other donors remain so reticent about financing CGAP? 5.31 There is little doubt that increasing the access of the poor to sustainable microfinance will remain a long-term development challenge for the immediate future, and that this is 62. Unlike the case for most in-house programs, CGAP's annual DGF allocation is transferred into a Bank- administered trust fund, and trust funds can be used to pay for in-house secretariat staff costs. 46 based on a solid international consensus that microfinance institutions can be both financially sustainable and reach the poor, and that access by the poor to microfinance can help to reduce poverty. The Microcredit Summit Campaign once again provides a snapshot of this challenge in expanding access to microfinance and achieving its stated goal of 100 million clients by 2005. It is also clear that CGAP has contributed both to increasing the reach and sustainability of selected MFIs and to increasing the world's knowledge about best practices in microfinance. It is less clear what CGAP's contribution to the broader microfinance industry has been in the past, and what kind of global collective action can most effectively enhance the role of microfinance in developing countries in the future. While the donor- driven model of expanding access to microfinance will likely remain in place for some time to come, this is something that individual donors, including the Bank, can undertake by themselves. One can also envisage a progression away from donor interventions and toward an ever-increasing presence of commercial financial institutions engaged in both retail or wholesale lending for microfinance. In fact, this is precisely what must occur if the desired mainstreaming of microfinance is ever to take place. It may be necessary for CGAP's partners to answer the question of the most effective role of global collective action in this process before CGAP will be able to reconfigure itself on a more sustainable course of action. 6. Findings and Lessons 6.1 In the case of programs which were explicitly intended to benefit the poor as was the case with CGAP, Bank-sponsored global programs need to align more closely ­ and in some cases, realign ­ with the mission of "a world free of poverty." CGAP was conceived as a program to foster and expand financial access to the poor, which it pursued quite actively under its Phase I as a retailer of grants to MFIs under a demand-driven approach. However, a shift in focus under Phase II (and continued under the current Phase III) has brought about greater concentration of CGAP activities in standard-setting, technical tools development and capacity building, increasingly directed to its member donors as well as to the microfinance industry at large. Grant-making to MFIs ­ primarily through its Appraisal and Monitoring service ­ became supply-driven, in that member donors now select those MFIs they have targeted for financing and other assistance. Whether this shift in focus away from MFIs directly and toward member donors has been effective in expanding financial access to the poor is still unclear, yet the modest progress toward mainstreaming microfinance, coupled with donors' diminished view of microfinance (as given in the Peer Review exercise) suggest a weakened effectiveness of this approach. 6.2 Adequate monitoring and evaluation must be present to determine progress toward program objectives, realization of intended outputs, and extent of impact from program interventions. CGAP has indeed produced and disseminated many technical tools, available via the internet, in addition to its grant-making to MFIs and MFI networks, best- practice exchanges and ongoing policy dialogues. But we know little ­ verifiable through rigorous monitoring and evaluation ­ about the actual uptake of these by MFIs and, more importantly, their impact on the administration, outreach, and sustainability of these MFIs. Anecdotal evidence, albeit favorable in some cases, is insufficient to conclude that CGAP is increasing the desired access of the poor to sustainable financial services. A lesson from the actions of one of its member donors ­ USAID, through its so-called "50-percent solution" ­ can and should be applied to CGAP in terms of tighter monitoring and evaluation processes, 47 and the construction of objectives and activities that are indeed evaluable and can be assessed over time. Future Bank support to CGAP should be contingent on the adoption of such procedures. Commentators on this case study have generally accepted three key points: the need for (1) impact monitoring, (2) expanding participation to developing/beneficiary countries, and (3) greater financial burden sharing. 6.3 A broader stakeholder representation, not just the voices of donors alone, could contribute to greater legitimacy and relevance of the program objectives and activities of global programs like CGAP. While developing countries and MFIs have benefited from CGAP activities, they are excluded from the strategic direction and oversight processes of the program at the highest governance level (the Council of Governors) and there is little evidence of their influence in CGAP strategy or policy setting. Sustainable microfinance will undoubtedly require a scaling-up of microfinance worldwide, in which both developing country governments and the commercial financial sector must be engaged. Furthermore, donors alone will likely be unable to meet the capital requirements implicit in this scaling-up. Clearly, CGAP has not demonstrated a capacity to "go to scale" with microfinance, and the donors ­ as of Phase II ­ have signaled a greatly reduced role for CGAP in the actual capital financing of MFIs. The comparative advantage of the private sector ­ in the guise of commercial financial institutions ­ can and should be brought to bear. 6.4 The Bank could be more proactive and selective in its financing obligations, eliminating the incentive for other donors to "free-ride." Financial contributions to CGAP's budget have, since the inception of the program, been a condition of membership, yet some members have never made a financial commitment, while others have done so, albeit sporadically. With the Bank continuously maintaining a much larger than proportional financial burden of CGAP ­ and greatly exceeding the 15 percent criterion of the DGF ­ other member donors have little incentive to carry their own weight, despite the explicit requirement under the CGAP Charter and the Guidelines set forth by the DGF for in-house secretariats. 6.5 Tighter linkages to Bank operations are crucial. Among CGAP's objectives is that of dissemination of best practice, yet the true test of this is the adoption of these practices by the member donors, including the Bank. The evidence suggests that CGAP has had minimal input into both the PRSP and CAS exercises, which serve as base documents for Bank lending programs in its client countries. One reason for this may be the program's lean staffing in relation to the myriad PRSP and CAS processes that are ongoing at any one time. While Bank lending to microfinance has been growing since it bottomed out in 1993, this is usually as a component within a larger overall project. This seems to signal that both the Bank and its client countries do not see the sector as meriting increased activity, except as part of larger, multi-sectoral projects. 6.6 An overwhelmingly dominant role in which the World Bank houses the secretariat, chairs the governing body, and provides most of the funding, even though ensuring the ownership of other partners, reduces the responsibility for shared governance, and places an undue share of responsibility, accountability and risk on the Bank for program performance. The CGAP Phase III Charter clearly places fiduciary responsibility for the program with the Bank. The weight of the Bank is also felt in both the Council of Governors (CG) and the Investment Committee ­ which is comprised in its 48 entirety of Bank and IFC staff with CG and EXCOM representatives as observers. Add to these the financial weight carried by the Bank in the context of CGAP's operational budget, and the line dividing the Bank and CGAP is further blurred. Proactive establishment of an independent identity for global programs like CGAP would help to determine early on whether sufficient ownership exists among the program partners to sustain its operations. In the absence of the strong and sustained financial support of the Bank, CGAP would likely have ceased activities at the close of its Phase I. 49 References CGAP. 1998. "CGAP: Status Report and Renewal, March 1998." ______. Occasional Paper No. 8, Financial Institutions with a Double Bottom Line: Implications for the Future of Microfinance. ______. 1999. Annual Report. ______. 2000. Annual Report. ______. 2001. Annual Report. ______. 2002a. Annual Report. ______. 2002b. "Tackling Aid Effectiveness from the Top: Microfinance as a Test Case, Donor Peer Reviews." ______. 2003a. Annual Report. ______. 2003b. "Phase III Strategy: 2003-2008." ______. 2004. "Aid Effectiveness Initiative: Microfinance Donor Peer Reviews". Mimeo. Daley-Harris, Sam, ed. 2002. Pathways out of Poverty. Bloomfield, CT: Kumarian Press. Daley-Harris, Sam. 2003. State of the Microcredit Summit Campaign Report 2003. de Soto, Hernando. 2000. The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else. New York: Basic Books. Development Grant Facility. "Guidelines on In-House Secretariats," updated July 23, 2003. Donor Working Group on Financial Sector Development. 1994. Micro and Small Enterprise Finance: Guiding Principles for Selecting and Supporting Intermediaries. Egger, Ruth, Klaus Maurer, and Cristina Ortiz. 1997. "CGAP Mid-Course Review." Fox, James W., Mark Havers, Klaus Maurer. 2002. "Evaluation and Strategic Review of the Consultative Group to Assist the Poorest (CGAP)." Khandker, Shahid. 1998. Fighting Poverty with Microcredit: Experience in Bangladesh. Oxford University Press. New York. _______. 2003. "Does Microfinance Really Benefit the Poor? Evidence from Bangladesh." Paper delivered at Asia and Pacific Forum on Poverty: Reforming Policies and Institutions for Poverty Reduction. Asian Development Bank, Manila. Latin American Shadow Financial Regulatory Committee. 2004. "Small and Medium Size Enterprises Finance in Latin America: Developing Markets, Institutions and Instruments." Statement #11. Mimeo. 50 Morduch, Jonathan and Barbara Haley. 2002. "Analysis of the Effects of Microfinance on Poverty Reduction." New York University Wagner Working Paper Series no. 1014. Operations Evaluation Department. 1998. Financial Sector Reform. A Review of World Bank Assistance. Washington, D.C.: The World Bank. Otero, Maria. 1999. "Bringing Development Back into Microfinance." Paper delivered at Conference for New Development Finance, Goethe University, Frankfurt. Robinson, Marguerite S.. 2001. The Microfinance Revolution: Sustainable Finance for the Poor. World Bank/Open Society Institute. United States General Accounting Office. 2003. Microenterprise Development: USAID's Program Has Met Some Goals; Annual Reporting Has its Limitations, Report number GAO 04-171, November 2003. CGAP 51 Annex A Annex A. Evaluation Framework for Phase 2 Report and Case Studies The Phase 2 Report and each case study follows a common outline and addresses 20 evaluation questions (Table A.1) that have been derived from OED's standard evaluation criteria (Table A.2), the 14 eligibility and approval criteria for global programs (Table A.3), and the 8 eligibility criteria for grant support from the Development Grant Facility (Table A.4). The sheer number of these criteria, some of which overlap, can be daunting even to an evaluator. Hence the OED evaluation team has reorganized these criteria into four major evaluation issues, which correspond to the four major sections of each report (Table A.1): -The overarching global relevance of the program -Outcomes and impacts of the program and their sustainability -Governance, management, and financing of the program -The World Bank's performance as a partner in the program These four issues correspond roughly to OED's evaluation criteria of relevance, efficacy, efficiency, and Bank performance, appropriately interpreted and expanded for the case of global programs. In the case of global programs, relevance must be measured not only against individual borrowing countries' priorities and Bank priorities, but also in terms of the interplay between global challenges and concerns on the one hand and country needs and priorities on the other. The former are typically articulated by the "global community" by a variety of different stakeholders and are reflected in a variety of ways such as formal international conventions to which developing countries are signatories; less formal international agreements reached at major international meetings and conferences; formal and informal international standards and protocols promoted by international organizations, NGOs, etc.; the Millennium Development Goals; and the Bank's and the Development Committee' eligibility criteria for global programs. While sponsorship of a program by significant international organizations may enhance "legitimacy" of a global program in the Bank's client countries, it is by no means a sufficient condition for developing country ownership, nor for ensuring its development effectiveness. "Relevance" and ownership by the Bank's client countries is more assured if the program is demanded by them. On other hand some "supply-led" programs may also acquire ownership over time by demonstrating substantial impacts, as in the case of the internet. Assessing relevance is by far the most challenging task in global programs since global and country resources, comparative advantages, benefit, costs, and priorities do not always coincide. Indeed the divergence of benefits and costs between the global level and the country level is often a fundamental reason for the provision of global public goods. Evaluating the relevance of global action to the Bank's client countries is however important because the global development agenda is becoming highly crowded and resources to finance it have remained relatively stagnant, therefore highlighting issues of selectivity. For the global programs that have been operating for some time, efficacy can be assessed not only in terms of program outcomes but more crucially in terms of impacts on the ground in developing countries. Outcomes and impacts in turn depend on the clarity and evaluability of each program's objectives, the quality of the monitoring and evaluation of results and, where appropriate, the effectiveness of the links of global program activities to the country level. Since global programs are partnerships, efficiency must include an assessment of the extent to which the benefit-cost calculus in collective organizational, management and financing arrangements is superior to achieving the same results by the individual partners acting alone. The institutional CGAP 52 Annex A development impact and the sustainability of the program itself (as opposed to that of the outcomes and impacts of the program's activities) are also addressed in this section of each report. Finally, this being an OED evaluation, it focuses primarily on the Bank's strategic role and performance in playing up to its comparative advantage relative to other partners in each program. The Bank plays varied roles in global programs as a convener, trustee, donor to global programs, and lender to developing countries. The Bank's financial support to global programs ­ including oversight and liaison activities and linkages to the Bank's regional operations ­ comes from a combination of the Bank's net income (for DGF grants), the Bank's administrative budget, and Bank-administered trust funds. In the case of the Global Environmental Facility (GEF) the Bank is a trustee and in the case of the Global Fund to Fight HIV/AIDS, Tuberculosis, and Malaria (GFATM), a "limited" trustee. In the case of GEF and MLF the Bank is also an implementing agency. Thus, the assessment of Bank performance includes the use of the Bank's convening power, the Bank's trusteeship, Bank financing and implementation of global programs, and, where appropriate and necessary, linkages to the Bank's country operations. Bank oversight of this entire set of activities is an important aspect of the Bank's strategic and programmatic management of its portfolio of global programs. The first column in Table A.1 indicates how the four sections and 20 evaluation questions addressed in the Phase 2 Report and case studies relates to the eight evaluation issues that were raised by the Bank's Executive Board in the various Board discussions of global programs during the design phase of OED's global evaluation and identified in the OED's Evaluation Strategy paper:1 -Selectivity -Monitoring and evaluation -Governance and management -Partnerships and participation -Financing -Risks and risk management -Linkages to country operations The third column in Table A.1 indicates how the four sections and 20 evaluation questions relate to OED's standard evaluation criteria for investment projects (Table A.2), the 14 criteria endorsed by the Development Committee and established by Bank management for approving the Bank's involvement in global programs (Table A.3), and the 8 criteria for grant support from the Development Grant Facility (Table A.4). The 14 eligibility and approval criteria for the Bank's involvement in global programs have evolved since April 2000 when Bank management first proposed a strategy to the Bank's Executive Board for the Bank's involvement in global programs and include the four overarching criteria endorsed by the Development Committee, and the four eligibility criteria and six approval criteria presented by Bank management to the Bank's Executive Board. Each global program must meet at least one of the four relatively more substantive eligibility criteria and all six of the relatively more process-oriented approval criteria. The first two eligibility criteria relate directly to the Bank's global public goods and corporate advocacy priorities (Table A.3). Although the six approval criteria resemble the topics covered in a project concept or appraisal document for Bank lending operations, unlike for Bank lending operations, there is currently only a one-step approval process for new global 1OED, The World Bank and Global Public Policies and Programs: An Evaluation Strategy, July 16, 2001, page 21. "Partnerships and participation" were originally listed as two separate evaluation issues in the evaluation strategy document. "Monitoring and evaluation" is now interpreted more broadly to include not only an assessment of the monitoring and evaluation procedures of each program but also the findings of previous evaluations with respect to the outcomes and impacts of each program, and their sustainability. CGAP 53 Annex A programs ­ at the concept stage and not at the appraisal stage. And new global programs only have to be approved by the Bank managing director responsible for the Network proposing a new program, not by the Bank's Executive Board. While the approval of new global programs is logically separate from and prior to their financing (whether from the DGF, trust funds, or other sources), the eight DGF eligibility criteria for grant support from the DGF (Table A.4) were actually established in 1998. Twenty out of the 26 case study programs and about two-thirds of the Bank's total portfolio of 70 global programs have received DGF grants. Table A.1. Key Evaluation Issues and Questions Evaluation Evaluation Questions Reference Issues Section I. Overarching Global Relevance of the Program Relevance. To what extent are the programs: A modification of OED's - Addressing global challenges and concerns in relevance criterion (Table the sector A.2) for the purpose of - Consistent with client countries' current global programs. development priorities The third bullet also relates - Consistent with the Bank's mission, corporate to managing director (MD) priorities, and sectoral and country assistance approval criterion #1 strategies? regarding a "clear linkage to the Bank's core institutional objectives" (Table A.3). International consensus. To what extent did the Development Committee programs arise out of an international consensus, (DC) criterion #4 (Table Selectivity formal or informal: A.3). - Concerning the main global challenges and concerns in the sector - That global collective action is required to address these challenges and concerns? Strategic focus. To what extent are the programs: The four bullets - Providing global and regional public goods correspond to the four MD - Supporting international advocacy to improve policies eligibility criteria (Table at the national level A.3). - Producing and delivering cross-country lessons of relevance to client countries - Mobilizing substantial incremental resources? Subsidiarity. To what extent do the activities of the DGF eligibility criterion #1 programs complement, substitute for, or compete (Table A.4). with regular Bank instruments? Section II. Outcomes, Impacts, and their Sustainability Efficacy. To what extent have the programs achieved, OED's efficacy criterion or are expected to achieve, their stated objectives, (Table A.2). taking into account their relative importance? Value added. To what extent are the programs adding The first bullet corresponds value to: to DC criterion #1 (Table Monitoring and - What the Bank is doing in the sector to achieve its A.3). evaluation core mission of poverty alleviation and sustainable development - What developing and transition countries are doing in the sector in accordance with their own priorities? CGAP 54 Annex A Evaluation Evaluation Questions Reference Issues Monitoring and evaluation. To what extent do the MD approval criterion #6 programs have effective monitoring and evaluation: (Table A.3), since effective - Clear program and component objectives verifiable by communications with key indicators stakeholders, including the - A structured set of quantitative or qualitative indicators Bank's Executive - Systematic and regular processes for data collection Directors, requires good and management monitoring and evaluation - Independence of program-level evaluations practices. - Effective feedback from monitoring and evaluation to program objectives, governance, management , and financing? Sustainability of outcomes and impacts. To what OED's sustainability extent are the outcomes and impacts of the criterion (Table A.2). programs resilient to risk over time? Section III. Organization, Management, and Financing of the Program Efficiency. To what extent have the programs A modification of OED's achieved, or are expected to achieve: efficacy criterion for the - Benefits more cost-effectively than providing the same purpose of global service on a country-by-country basis programs (Table A.2). - Benefits more cost-effectively than if the individual The first bullet also relates contributors to the program acted alone? to MD eligibility criterion #3 (Table A.3) and DGF eligibility criterion #3 (Table A.4). Legitimacy. To what extent is the authorizing A modification of OED's Governance environment for the programs effectively derived evaluation criteria (Table and from those with a legitimate interest in the program A.2) for the purpose of management (including donors, developing and transition global programs. countries, clients, and other stakeholders), taking into account their relative importance. Governance and management. To what extent are the MD approval criterion #5 governance and management of the programs: (Tables B.3) and DGF - Transparent in providing information about the eligibility criterion #5 programs (Table A.4). - Clear with respect to roles & responsibilities - Fair to immediate clients - Accountable to donors, developing and transition countries, scientists/professionals, and other stakeholders? Partnerships and participation. To what extent do DGF eligibility criterion #8 developing and transition country partners, clients, (Table A.4). Partnerships and beneficiaries participate and exercise effective and voice in the various aspects of the programs: - Design participation - Governance - Implementation - Monitoring and evaluation? Financing. To what extent are the sources of funding MD approval criterion #4. for the programs affecting, positively or negatively: (Table A.3). Financing - The strategic focus of the program The third bullet also relates - The governance and management of the program to OED's sustainability - The sustainability of the program? criterion (Table A.2). CGAP 55 Annex A Evaluation Evaluation Questions Reference Issues Bank action to catalyze. To what extent has the DC criterion #2 (Table A.3) Bank's presence as a partner in the programs and DGF eligibility criterion catalyzed, or is catalyzing non-Bank resources for #4 (Table A.4). the programs? Institutional development impact. To what extent has A modification of OED's the program established effective institutional institutional development arrangements to make efficient, equitable, and impact criterion (Table A.2) sustainable use of the collective financial, human, for the purpose of global and other resources contributed to the program. programs. Risks and risk Risks and risk management. To what extent have the MD approval criterion #3 management risks associated with the programs been identified (Table A.3). and are being effectively managed? Section IV. World Bank's Performance Comparative advantage. To what extent is the Bank DC criterion #3 (Table playing up to its comparative advantages in relation A.3), MD approval criterion to other partners in the programs: #2 (Table A.3), and DGF - At the global level (global mandate and reach, eligibility criterion #2 convening power, mobilizing resources) (Table A.4). - At the country level (multi-sector capacity, analytical expertise, country-level knowledge)? Linkages to country operations. To what extent are MD approval criterion #1 there effective and complementary linkages, where (Table A.3) regarding needed, between global program activities and the "linkages to the Bank's Linkages to Bank's country operations, to the mutual benefit of country operational work." country each? operations Oversight. To what extent is the Bank exercising This relates to DGF effective and independent oversight of its eligibility criterion #6 on involvement in the programs, as appropriate, for in- "arm's length relationship" house and externally managed programs, (Table A.4). respectively. Both questions 17 and 18 together relate to OED's Bank performance criterion (Table A.2). Disengagement strategy. To what extent is the Bank DGF eligibility criterion #7 facilitating effective, flexible, and transparent (Table A.4). disengagement strategies, as appropriate? CGAP 56 Annex A Table A.2. Standard OED Evaluation Criteria Criterion Standard Definitions for Lending Operations Possible Ratings The extent to which the project's objectives are consistent (1) with High, substantial, modest, the country's current development priorities and (2) with current negligible. Relevance Bank country and sectoral assistance strategies and corporate goals (expressed in Poverty Reduction Strategy Papers, Country Assistance Strategies, Sector Strategy Papers, Operational Policies). The extent to which the project's objectives were achieved, or High, substantial, modest, Efficacy expected to be achieved, taking into account their relative negligible. importance. The extent to which the project achieved, or is expected to High, substantial, modest, Efficiency achieve, a return higher than the opportunity cost of capital and negligible. benefits at least cost compared to alternatives. The extent to which the authority exercised by the program is High, substantial, modest, effectively derived from those with a legitimate interest in the negligible. Legitimacy /1 program (including donors, developing and transition countries, clients, and other stakeholders), taking into account their relative importance. The extent to which a project improves the ability of a country or High, substantial, region to make more efficient, equitable and sustainable use of its negligible, modest. Institutional human, financial, and natural resources through: (a) better development definition, stability, transparency, enforceability, and predictability impact of institutional arrangements and/or (b) better alignment of the mission and capacity of an organization with its mandate, which derives from these institutional arrangements. IDI includes both intended and unintended effects of a project. Sustainability The resilience to risk of net benefits flows over time. Highly likely, likely, unlikely, highly unlikely. Highly satisfactory, satisfactory, moderately Outcome The extent to which the project's major relevant objectives were satisfactory, moderately achieved, or are expected to be achieved, efficiently. unsatisfactory, unsatisfactory, highly unsatisfactory The extent to which services provided by the Bank ensured quality Highly satisfactory, Bank at entry and supported implementation through appropriate satisfactory, unsatisfactory, performance supervision (including ensuring adequate transition arrangements highly unsatisfactory. for regular operation of the project). The extent to which the borrower assumed ownership and Highly satisfactory, Borrower responsibility to ensure quality of preparation and implementation, satisfactory, unsatisfactory, performance and complied with covenants and agreements, toward the highly unsatisfactory. achievement of development objectives and sustainability. /1 This represents an addition to OED's standard evaluation criteria in the case of global programs, since effective governance of global programs is concerned with legitimacy in the exercise of authority in addition to efficiency in the use of resources. CGAP 57 Annex A Table A.3. Selectivity and Oversight of Global Programs Selectivity Criteria for Bank Involvement in Global Public Goods: Endorsed by Development Committee (September 2000) /1 An emerging international consensus that global action is required A clear value added to the Bank's development objectives The need for Bank action to catalyze other resources and partnerships A significant comparative advantage for the Bank. Approval Criteria for Bank Involvement in Partnership Initiatives Beyond the Country Level: Established by Bank Management (November 2000) /2 A clear linkage to the Bank's core institutional objectives and, above all, to the Bank's country operational work A strong case for Bank participation based on comparative advantage A clear assessment of the financial and reputational risks to the Bank and how these will be managed A thorough analysis of the expected level of Bank resources required, both money and time, as well as the contribution of other partners A clear delineation of how the new commitment will be implemented, managed, and assessed A clear plan for communicating with and involving key stakeholders, and for informing and consulting the Executive Directors. Global Public Goods Priorities /3 Corporate Advocacy Priorities Communicable diseases Strategic Focus for Oversight Empowerment, security, and social HIV/AIDS, tuberculosis, malaria and of Global Programs: inclusion childhood communicable diseases, Established by Bank Gender mainstreaming including the relevant link to education Management (March 2003) /4 Civic engagement and participation Vaccines and drug development for major Social risk management (including communicable diseases in developing disaster mitigation) countries Provide global public Investment climate Environmental commons goods Support to both urban and rural Climate change development Water Infrastructure services to support private Forests sector development Biodiversity, ozone depletion and land Support international advocacy Regulatory reform and competition policy degradation for reform agendas which in Financial sector reform Promoting agricultural research a significant way Public sector governance Information and knowledge address policy framework Rule of law (including anti-corruption) Redressing the Digital Divide and conditions relevant for Public administration and civil service equipping countries with the capacity developing countries reform (incl. public expenditure to access knowledge accountability) Understanding development and poverty Access to and administration of justice reduction Are multi-country programs (judicial reform) Trade and integration which crucially depend on Education Market access highly coordinated Education for all, with emphasis on girls' Intellectual property rights and standards approaches education International financial architecture Building human capacity for the Development of international standards knowledge economy Financial stability (incl. sound public debt Mobilize substantial incremental Health management) resources that can be Access to potable water, clean air and International accounting and legal effectively used for sanitation framework development. Maternal and child health /1 From the Development Committee Communiqué issued on September 25, 2000. Both the Development Committee and Bank Management envisaged global programs as being the principal instrument for Bank involvement in providing global public goods. /2 The Initiating Concept Memorandum in the Partnership Approval and Tracking System (PATS) was initially organized according to these six criteria. /3 These are the five corporate advocacy priorities and the five global public goods priorities (and bulleted sub-categories) from the Strategic Directions Paper for FY02-04, March 28, 2001. Within the Partnership Approval and Tracking System (PATS), global programs are expected to identify, for tracking purposes, their alignment with at least one of these ten corporate priorities. CGAP 58 Annex A Table A.4. Eligibility Criteria for Grant Support from the Development Grant Facility Subsidiarity The program contributes to furthering the Bank's development and resource mobilization objectives in fields basic to its operations, but it does not compete with or substitute for regular Bank instruments. Grants should address new or critical development problems, and should be clearly distinguishable from the Bank's regular programs. Comparative The Bank has a distinct comparative advantage in being associated with the program; it does not advantage replicate the role of other donors. The relevant operational strengths of the Bank are in economic, policy, sector and project analysis, and management of development activities. In administering grants, the Bank has expertise in donor coordination, fund raising, and fund management. Multi-country The program encompasses multi-country benefits or activities which it would not be efficient, practical benefits or appropriate to undertake at the country level. For example, informational economies of scale are important for research and technology work, and operations to control diseases or address environmental concerns (such as protect fragile ecosystems) might require a regional or global scope to be effective. In the case of grants directed to a single country, the program will encompass capacity-building activities where this is a significant part of the Country Assistance Strategy and cannot be supported by other Bank instruments or by other donors. This will include, in particular, programs funded under the Institutional Development Fund, and programs related to initial post- conflict reconstruction efforts (e.g., in countries or territories emerging from internal strife or instability). Leverage The Bank's presence provides significant leverage for generating financial support from other donors. Bank involvement should provide assurance to other donors of program effectiveness, as well as sound financial management and administration. Grants should generally not exceed 15 percent of expected funding over the life of Bank funding to a given program, or over the rolling 3-year plan period, whichever is shorter. Where grant programs belong to new areas of activities (involving, e.g., innovations, pilot projects, or seed-capital) some flexibility is allowed for the Bank's financial leverage to build over time, and the target for the Bank grant not to exceed 15 percent of total expected funding will be pursued after allowing for an initial start-up phase (maximum 3 years). Managerial The grant is normally given to an institution with a record of achievement in the program area and competence financial probity. A new institution may have to be created where no suitable institution exists. The quality of the activities implemented by the recipient institution (existing or new) and the competence of its management are important considerations. Arm's length The management of the recipient institution is independent of the Bank Group. While quality an arm's relationship length relationship with the Bank's regular programs is essential, the Bank may have a role in the governance of the institution through membership in its governing board or oversight committee. In cases of highly innovative or experimental programs, Bank involvement in supporting the recipient to execute the program will be allowed. This will provide the Bank with an opportunity to benefit from the learning experience, and to build operational links to increase its capacity to deliver more efficient services to client countries. Disengage- Programs are expected to have an explicit disengagement strategy. In the proposal, monitorable action ment strategy steps should be outlined indicating milestones and targets for disengagement. The Bank's withdrawal should cause minimal disruption to an ongoing program or activity. Promoting Programs and activities should promote and reinforce partnerships with key players in the development partnerships arena, e.g., multilateral development banks, UN agencies, foundations, bilateral donors, professional associations, research institutions, private sector corporations, NGOs, and civil society organizations. Source: World Bank, Development Grant Facility documentation. CGAP 59 Annex B Annex B. Program At a Glance Table B.1. General Information Operational start date August 1995. Organizational status Formal. Independent legal entity No ­ housed in World Bank. Initial authorizing environment Bank's Executive Board, March 21, 1995. Current authorizing environment Self-authorizing Council of Governors (CG). Current cosponsors, if any None. Terminology not used. Written charter Yes. Approved September 2002 by CG members. Present location World Bank HQ (Washington, D.C.), with one regional office in Paris. Internet address www.cgap.org Program manager Elizabeth Littlefield, Director, INF/CGP. Bank task manager /1 Carlos Cuevas, Lead Financial Economist, FSE/OPD. Sponsoring Bank unit(s) INF and FSE. Reviewing sector board PSD ­ Private Sector and FSE ­ Financial Sector Sector Finance Sub-sector Micro- and SME finance Theme Financial & private sector development Sub-theme Small & medium enterprise support Bank sector strategy paper FSE ­ Strategy for the Financial Sector, March 2001 PSD ­ Private Sector Development Strategy, April 2002 GPG Priority None CA Priority Investment climate ­ financial sector reform FY02 program expenditures $9.17 million DGF status Window 1 DGF grant (FY03) $6.73 million Bank-administered trust funds Yes TF contributions (FY02) $5.90 million (excluding DGF contribution to the trust fund) /1 Person who is immediately responsible for oversight of the program. Sources: The information in this and subsequent annexes and tables has been assembled from the CGAP Charter, the CGAP web-site, the Bank's Partnership Approval and Tracking System (PATS), and other program documents. CGAP 60 Annex B Table B.2. Partners and Participants Global/Program Level Programmatic Country/Activity Institutional and Governing Body Executive Body Level Other (CG) (Excom) International/ AfDB, AsDB, EBRD, World Bank plus one Numerous regional regional IDB, IFAD, ILO, other representative partners such as the organizations UNCDF, UNDP, (AsDB) China Microfinance World Bank Training Center, EDA Rural Systems (India), and the Microfinance Center (Poland). Industrialized Australia, Belgium, Two representatives Co-funding of countries Canada, Denmark, (currently UK and activities with EU, Finland, France, Norway) member donors. Germany, Italy, Japan, Luxembourg, Netherlands, Norway, Sweden, Switzerland, UK, USA Developing Regulatory and countries supervisory authorities (such as BCEAO in West Africa) Foundations Argidius Foundation, Ford Foundation Co-funding of Ford Foundation activities with Argidius and Ford. Commercial Financial institutions. private sector Private sector service providers such as auditors and MIS specialists. Industrialized Microfinance country CSOs networks (including umbrella organizations) Developing Local-level MFIs and country CSOs microfinance networks. Individuals Four microfinance Practitioners industry leaders General view of partnerships: At the global/program level, CGAP is a consortium of donors working together to build sustainable financial services for the poor. CGAP functions as a global convening platform and resource center for the microfinance industry, providing different services to a broad range of audiences and clients. At the country/activity level, CGAP does almost everything with partners. CGAP has established and defined a typology of partnerships in order for both parties to clearly understand both the nature of each partnership and their respective roles, mutual responsibilities, and expectations: · Investment grant partnership ­ between CGAP and a grant recipient · Joint investment partnership ­ between CGAP and another donor or service provider as co-investor CGAP 61 Annex B · Collaborative partnership ­ between CGAP and service provider or donor · Contractual partnership ­ between CGAP and service provider Perspective on commercial private sector partnerships: At the program level, membership of CGAP is only open to public bilateral and multilateral donor agencies and private foundations with significant involvement in the development of financial services for the poor. Other types of institutions may be considered for membership by the Council of Governors (governing body) at some future point. At the activity level, CGAP works in partnership with a broad range of private sector service providers. These include community-level microfinance institutions (MFI NGOs), transformed NGOs that have become regulated commercial banks (such From your perspective, to what degree have developing as Compartamos in Mexico), and transition countries been involved in the design, traditional commercial banks that governance, implementation, and monitoring and are providing financial services to evaluation of the program at the global level? the poor (such as Banco do Nordeste in Brazil), and private service providers such as IT companies (Hewlett-Packard, MIS Design firms), local and international consulting firms, auditors, and credit rating agencies. Governance Programmatic vs. institutional partners at the global level: Implementation Legally, CGAP is a group of trust funds managed by the World Monitoring & Bank on behalf of other donors. Evaluation CGAP members are expected to contribute cash, through flexible 0% 20% 40% 60% 80% 100% mechanisms, to carry out CGAP's High Substantial Modest Negligible operations. Not significantly different among activities: P = 78.1% Voice of developing countries: At the program level, this was previously provided by the Policy Advisory Group, all eight From your perspective, to what degree have members of which were leaders of beneficiaries been involved in the design, microfinance institutions or implementation, and monitoring and evaluation of the microfinance networks from activities of the program at the country level? developing and transition economies. Under the new governance structure of CGAP III, the Policy Advisory Group has been incorporated into the Design Executive Committee of the Governing Body (the CG). The new Excom now comprises four Implementation microfinance industry leaders, as well as representatives of member donors. That is, the Excom has been restructured under CGAP's Monitoring & new governance structure to Evaluation include non-donor stakeholders as opposed to involving only donor 0% 20% 40% 60% 80% 100% members. High Substantial Modest Negligible Not significantly different among activities: P = 79.6% CGAP 62 Annex B Table B.3. World Bank's Roles Founder Yes Chair of governing body Yes If yes, who? Nemat Shafik, INFVP Member of the governing body Yes If yes, who? Carlos Cuevas, Lead Financial Economist, FSE/OPD In-house secretariat Yes Funding /1 DGF TF trustee Yes TF manager /2 Yes Lender to the sector Yes Convener in the sector /3 Yes /1 Financial contributions to the program itself, not including BB resources spent on oversight and liaison activities. /2 Involves responsibility for oversight and management of how the trust fund resources are utilized. /3 The World Bank takes the initiative to organize meetings and conferences in the sector on issues related to but outside the scope of the program in order to advocate change, reach consensus, and/or mobilize resources with respect to emerging issues in the sector. Global task manager perspective on how the Bank's role has changed during the life of the program The Bank's role vis-à-vis CGAP has evolved in significant ways since the creation of CGAP in 1995. In the first phase of CGAP, microfinance expertise and coordination in the Bank were extremely limited, and a major priority of the Secretariat was placed on mainstreaming microfinance best practices in the Bank, more so than for any other member donor. As the Bank progressively developed a small core of microfinance expertise and with the Financial Sector Vice Presidency assuming the coordination of microfinance policy and knowledge dissemination within the Bank, CGAP's support to the Bank has become more consultative and informative rather than one of direct operational support. Indeed, in 1998, CGAP's mandate was changed from "mainstreaming best practices in the World Bank" to "improving microfinance best practices in all member donors." CGAP's support to the Bank is now focused on high-level policy dialogue and upstream inputs into selected CASs and country- level projects. The role of the Bank has also changed in terms of funding. When CGAP was established, it was with the full expectation that the Bank would remain the majority funder because of CGAP's global mandate. Since CGAP's integration in the DGF, non-Bank member donors have contributed financially and some have increased their funding commitments in order to meet the DGF's 15 percent funding criteria. This has resulted in an increase in ownership among other members and a more balanced treatment of all donors, even though the Bank remains CGAP's largest donor.2 2Response from the TM questionnaire, question 7, which was completed by the Bank's task manager or program manager for each global program in the fall of 2001. Revised August 30, 2003. CGAP 63 Annex C Annex C. Stated Mission, Objectives, Outputs, and Activities Mission/Goal To help build financial systems that work for the poor, providing large numbers of people with diverse financial services through a wide range of organizations. Development To support the development of financial systems that work for the poor, by improving the Objective capacity of microfinance institutions to deliver flexible, high-quality financial services to the poor on a sustainable basis. Outputs/ Five strategic priorities for CGAP's third phase of operations (2003-2008): Strategy · Fostering a diversity of financial institutions that serve the poor · Promoting a broader range of financial services available to the poor · Improving the availability and the quality of information on the performance of microfinance providers · Promoting a sound policy, legal and regulatory framework for microfinance · Improving aid effectiveness in microfinance. Five core strategic themes for CGAP's second phase of operations (1998-2003) · Supporting the development of microfinance institutions. · Supporting changes in the practices of member donors to improve their microfinance operations. · Increasing the poverty outreach of microfinance institutions. · Improving the legal and regulatory framework for microfinance institutions. · Facilitating the commercialization of the industry. Activities/ CGAP works in each of these strategic areas by: Products · Providing technical advice · Developing and setting standards, facilitating consensus building · Conducting research and development · Advancing knowledge and information sharing · Offering training and capacity building services with other actors · Funding innovative projects or models Major Clients To achieve its goals and objectives, CGAP works with three major groups of clients: · Local-level microfinance institutions (MFIs) and other financial institutions and practitioners · Member donor organizations · The microfinance industry as a whole (including government policymakers, regulatory and supervisory authorities, and private sector service providers such as auditors, credit rating agencies, MIS specialists, and consulting firms.) MFI Services · Technical Guides · Skills for Microfinance Managers (training) · Information Systems Services · Rating Fund · Funding Donor · Technical Tools & Standards Services · Staff Training · Donor Peer Reviews and follow up technical assistance · Donor-specific resources: publications (highlights of specific topics, best practice case studies of donors' role in microfinance), presentation modules, staff training CGAP 64 Annex D Annex D. Genesis, Evolution, and Exit Strategies of CGAP GENESIS AND EVOLUTION Jointly recognizing the powerful role of microfinance as a development tool, nine leading donors and practitioners formed CGAP in 1995 to develop and share best practices, set standards, and develop tools and models for microfinance. CGAP is currently in its third phase. External program-level evaluations were carried out in 1997 and 2002 (towards the ends of Phase I and II), and a third evaluation will be carried out towards the end of Phase III to recommend whether and how CGAP should continue after June 2008. · Phase I (1995-1998): This initial phase focused on the development of a common language for the microfinance industry, catalyzing the movement towards best practice performance standards, and building a consensus among its many and varied stakeholders primarily on the importance of financial sustainability. · Phase II (1998-2003): A more explicit focus on capacity building at the retail level through the development of technical tools and the provision of training, advisory services, and other forms of assistance to microfinance institutions. This phase also focused on better understanding the impact of microfinance on the poor, increasing commercialization of microfinance, improving the policy and regulatory environment, and mainstreaming good practices in donor agencies. · Phase III (2003-2008): A focus on scaling up outreach by (1) promoting greater diversity of institutions providing microfinance; (2) encouraging a broader range of financial services for the poor (beyond microcredit); (3) fostering a good policy framework; and (4) improving aid effectiveness. In addition, CGAP plans to leverage its resources by decreasing one-on-one grant-making to microfinance institutions and working through wholesale-level intermediaries such as national, regional, or international level networks; and co-funding with its member donors. CGAP also plans to become much more of an operational and learning institution, engaging in more action research, prospecting of innovations and promising institutions, disseminating information, and sharing of lessons learned. EXIT STRATEGIES Regarding the program The present mandate of CGAP extends through FY08. The Council of Governors will conduct a review in FY06 in order to determine post FY08 options for CGAP (disband, expand or transform). Regarding the Bank's involvement in the program The two sponsoring vice presidencies, FSE and INF, have no strategy at the present time for completely disengaging the Bank from CGAP. However, they are supporting CGAP's current strategy of reducing its financial dependence on the Bank (amount requested from DGF is reduced by US$400,000 every year). Regarding DGF funding CGAP is a Window 1 program. When CGAP was established, it was with the full expectation that the Bank would remain the majority funder because of CCAP's global mandate. Since CGAP's integration into the DGF, non-Bank member donors have contributed financially and some have increased their funding commitments in CGAP 65 Annex D order to meet the DGF's 15 percent funding criterion. This has resulted in an increase in ownership among other members and a more balanced treatment of all donors, even though the Bank remains CGAP's largest donor. CGAP's present strategy is to gradually reduce the share of DGF funding in its total budget by US$400,000 annually, by seeking increased contributions from other member donors as well as new member donors. Accordingly, the DGF allocation declined from US$7.5 million in FY01 to US$7.13 million in FY02, to US$6.73 million in FY03, and to US$6.33 million in FY04. CGAP 66 Annex E Annex E. Relationship of CGAP to the Millennium Development Goals CGAP's activities are aimed at reducing poverty by facilitating the poor's access to financial services and enhancing their economic well-being. Access to financial services allows the poor to improve their lives in different ways, enabling them, in effect, to achieve most of the MDGs, but on their own terms and in a self-sustainable way. Extensive evidence shows that microfinance directly contributes to many of the MDG goals and targets. Goals Targets Comments 1. Eradicate 1. Halve, between 1990 and Access to financial services helps reduce extreme poverty extreme 2015, the proportion of and hunger by increasing and diversifying incomes, building hunger and people whose income is less assets, and reducing vulnerability. poverty than one dollar a day. 2. Achieve 3. Ensure that, by 2015, Households that have access to microfinance spend more on universal children everywhere, boys their children's education. This enables families to send primary and girls alike, will be able to several children to school at one time and reduces drop-out education complete a full course of rates in higher primary grades. primary schooling. 3. Promote 4. Eliminate gender disparity Microfinance clients are overwhelmingly female. The ability to gender in primary and secondary borrow, save, and earn income enhances women's equality and education, preferably by confidence, giving them more economic freedom and power empower 2005, and to all levels of within their households and communities. women education no later than 2015. 4. Reduce 5. Reduce by two-thirds, Access to financial services contributes to improved nutrition, child mortality between 1990 and 2015, the housing, and health, especially among women clients. It under-five mortality rate. allows poor clients to seek health care services when 5. Improve 6. Reduce by three-quarters, needed, rather than wait until an illness has reached crisis maternal between 1990 and 2015, the proportions. health maternal mortality rate. 8. Develop a 12. Develop further an open, CGAP is a global partnership of public and private sector global rule-based, predictable, non- institutions. partnership discriminatory trading and for financial system development 18. In cooperation with the private sector, make available the benefits of new technologies, especially information and communications. CGAP 67 Annex F Annex F. Rating of CGAP Activities According to the Bank's Strategic Focus for Global Programs and OED Subcategories To what extent is the program involved in the following activities? Modest/ High Sub- stantial Negli- Comments gible 1. Providing global public goods CGAP is the standard setter for the microfinance industry. It plays a unique role in facilitating Implementing consensus on performance standards and guidelines to donors, MFIs, regulators and other players conventions, rules, X such as rating agencies. Recent examples include the Donor Guidelines for Supporting Savings- standards and norms based MFIs, Guidelines for Regulation and Supervision, the Disclosure Guidelines for MFI Financial Statements, the Loan Portfolio Tool, and the development of a set of simple indicators to determine the poverty level of microfinance clients. Financing research & CGAP undertakes research on key issues and new challenges facing the microfinance industry. In development for new X 2002, CGAP carried out research (often in conjunction with other organizations) on agricultural products and technologies microfinance, product costing, business process mapping, and the impact of microfinance on the poor. Financing country-level investments to deliver X global public goods 2. Supporting international advocacy for country-level policy reform agendas CGAP advocates an approach to microfinance focusing on building pro-poor financial systems and the human, technical, institutional, and information infrastructure necessary to support them. Its approach is focused on institutional and financial sustainability and large-scale outreach. CGAP does Advocacy X this by (1) providing strategic advice, technical assistance and training to microfinance institutions; (2) working with donors to improve the quality of their microfinance programming; (3) working with government policymakers, donors, and microfinance practitioners to improve the policy and regulatory framework for pro-poor financial services; and (4) developing and disseminating technical tools and best practice information to the microfinance industry as a whole. CGAP plans to expand its activities in this area in Phase III. As a multi-donor consortium, CGAP has Supporting national-level a comparative advantage in working with multilateral development agencies and other influential policy & institutional X policymakers to improve the legal and regulatory environment for microfinance. CGAP works closely reform with the Bank, the IMF, the BCEAO and other regional and national-level regulatory bodies on policy and regulatory issues affecting microfinance, through consultations in high-level policy forums, providing inputs to selected CASs and PRSPs, and training of central bankers and bank supervisors. CGAP 68 Annex F Modest/ High Sub- stantial Negli- Comments gible Financing country-level investments to deliver X national public goods 3. Multi-country programs which crucially depend upon highly coordinated approaches Through the Microfinance Gateway, CGAP is the resource center and information clearing-house for Knowledge creation and the industry. CGAP plans to expand its role In Phase III as a learning institution by engaging in more dissemination X action research, focusing on incubating ideas, testing innovative models and spinning them off, and sharing lessons learned. CGAP's capacity building programs provide financial and operational management training to MFI staff through local and regional training institutes. The programs seek to build local markets for microfinance by providing course materials, training of trainers, and initial marketing support to local partner organizations that deliver the courses on a commercial basis. CGAP's capacity building initiatives encompass six regional training hubs with local private sector partners serving 40 countries and offering 10 courses in six different languages. CGAP also develops and disseminates technical tools for use by microfinance practitioners. Its series of technical tools includes the Handbook for Management Information Systems (MIS) for MFIs, the Business Planning and Financial Modeling Handbook, and the Product Costing Tool. Capacity building X In addition to training, CGAP provides advisory services to country-level MFIs and commercial banks through institutional appraisals, business planning exercises and specific technical and strategic management advice. CGAP also provides training and strategic and technical services to donors, including portfolio reviews. The bulk of CGAP's activities are not individualized services but rather meant to benefit multiple players in different countries, as in its role in setting standards for donors, regulators and financial service providers. In cases where CGAP provides individualized services, the lessons learned are disseminated and can be applied globally. CGAP is decreasing its direct investments in MFIs in Phase III. CGAP's approach to improving donor coordination is to build consensus among donors regarding Improving donor what works in microfinance so that different donors are not working at cross-purposes from each coordination X other, and to help individual donors improve the effectiveness, efficiency, and accountability of their operations in microfinance. CGAP plans to expand its services to member donors in Phase III in the areas of standards building, information dissemination, training, and technical assistance. CGAP 69 Annex F Modest/ High Sub- stantial Negli- Comments gible 4. Mobilizing substantial incremental resources While mobilizing financial resources is an important priority objective, CGAP is very reluctant to Mobilizing financial accept restricted or tied (non-core) funding from donors. All member donors' funds go into the core resources X fund, from which is financed all of CGAP's work program, which is approved by the Executive Committee of donor members. In your opinion, at what level do the principal benefits of the program accrue? Primarily global Primarily regional Primarily individual Primarily national 0% 20% 40% 60% 80% 100% 1 = most significant 2 3 4 = least significant Significantly different among levels: P = 0.1% CGAP 70 Annex G Annex G. Evaluations and Audits of CGAP Table I.1. Recent Evaluations and Audits of CGAP, and OED Sector Studies Relating to CGAP Type of Date Commissioned Reported to Conducted by Title evaluation/ completed/ by audit expected External November "Renewal Renewal Ruth Egger, Klaus Maurer, CGAP Mid-Course Review Evaluation 1997 Committee" of Committee and and Cristina Ortiz the CG the CG External April 4, 2002 Excom Excom and CG James W. Fox, Mark Evaluation and Strategic Review of the Consultative Evaluation Havers, and Klaus Maurer Group to Assist the Poorest (CGAP) Internal audit April 24, IAD at request of Arun Banerjee and Faten Audit of the Use of Development Grant Facility Funds by 2002 RMC/DGF Hatab the Consultative Group to Assist the Poorest Sector study 1993 OED A Review of Bank Lending for Agricultural Credit and Rural Finance (1948-1992) Sector study 1996 OED A Review of Bank Lending for Agricultural Credit and Rural Finance (1948-1992): A Follow-Up Sector study 1998 OED Nicolas Mathieu Financial Sector Reform Sector study 2005 OED Laurie Effron Financial Sector Reform PROCESS, SCOPE, AND METHODOLOGIES OF THE MOST RECENT PROGRAM-LEVEL EVALUATION Process and Approach The 2002 evaluation was conducted by a team of evaluators engaged by the Consultative Group and it was overseen by the EXCOM (see table above for individual names). Those contracted to perform the evaluation were external to CGAP and selected on the basis of Terms of Reference approved by the EXCOM. Given the involvement of both the CG and the EXCOM in the design of the evaluation, one could not consider the this to have been an independent evaluation. Furthermore, initial findings were vetted with EXCOM for their advice and comment prior to the finalization of the evaluation. CGAP 71 Annex G Scope Table I.2. To what extent did the evaluation address the following issues? A great Some- Not at Comments deal what all Objectives and activities · Strategic focus x Recommends five strategic themes for the CGAP Secretariat under Phase III: (1) institutional diversity, (2) MFI regulatory environment, (3) financial systems development, (4) savings, and (5) multi-sector programs with microfinance components. · Outcomes and impacts x Outputs resulting from CGAP expenditures are well documented, but the impact of these outputs on MFI performance (e.g., extended outreach, self-financial sufficiency) is not provided. Organization, management, and financing · Partnerships and participation x Participation is assessed through the lens of Phase II member donors five obligations: (1) contribution (cash or kind) to CGAP; (2) adoption of Donor Committee Guidelines; (3) active participation in the CG (renamed to Council of Governors); (4) willingness to mainstream microfinance; and (5) information-sharing with CGAP. · Governance x · Management x · Financing x There are descriptive statements, which detail the Bank's contribution vis-à-vis other member donors, but no analytical treatment of this larger-than-proportional financing and the possible reasons for it. · Monitoring & evaluation X Not addressed · Role of the World Bank x Dealt with insofar as the Bank is one of the larger donors. No specific discussion or analysis regarding CGAP and its value-added to the Bank, or vice-versa. · Role of other global partners x Not addressed · Risks and risk management x Not addressed CGAP 72 Annex G Methodologies Table I.3. To what extent did the evaluation utilize the following methodologies? A great Some- Not at Comments deal what all Desk review x See comments for Literature Review below Literature review x No bibliography provided, but, as stated in the evaluation (pg. 9), "voluminous materials" on microfinance written by experts in NGOs, the academic community and the MFI community, were reviewed. Consultations x Interviews with more than 150 people from the MFI and donor communities. All CGAP member donors were contacted, and most were interviewed, either by phone or via email. All EXCOM members were interviewed, as were most PAG members. From the MFI community-at-large, most interviews were in conjunction with other conferences. Surveys x None were conducted for the evaluation; the evaluation does make use of the results from a confidential client survey conducted in 2000 for CGAP, specific to the usefulness of the technical tools developed by the Secretariat (Listening to Clients: Improving CGAP's Effectiveness, by Carla Henry). Site visits x In-country meetings with MFIs and donors were limited to India and Bosnia and Herzegovina. Impact studies X None conducted and none were available for review during the 2002 evaluation. OUTCOMES AND IMPACTS OF THE PROGRAM See comments above. Impact assessment was not a specific objective of the 2002 evaluation. CGAP 73 Annex G EFFECTS OF PREVIOUS EVALUATIONS ON THE PROGRAM Table I.4. Recommendations from 2002 Phase II Evaluation and their Status Phase II Evaluation Recommendations CGAP Phase III Actions Authorize Phase III for CGAP Phase III commenced in July 2003 Restate goal to read: To rapidly increase the sustainable provision of Phase III Charter states, "The mission of CGAP III is to expand and to financial services to poor people, including the very poor." accelerate [the poor's] access to a broad range of convenient and sustainable financial services." Recommended CGAP objectives: (1) to develop, synthesize and Phase III priority themes are (1) fostering a diversity of financial disseminate knowledge and information on good practices and institutions that serve the poor; (2) facilitating the poor's access to a innovations in microfinance; (2) to develop and establish a consensus wide range of flexible, convenient financial services; (3) improving the on principles and standards in microfinance; (3) to improve quality of availability and quality of information on the performance of donor programming in microfinance by mainstreaming principles and microfinance institutions and, (4) promoting a sound policy and legal standards within member donor agencies and by coordinating policy framework for microfinance. and activities among donors. Recommends a two-tier membership to the CG: (1) full membership for No change in membership criteria to the CG. those institutions making financial contributions to CGAP and complying with CGAP norms and (2) associate membership for others not contributing financially (these would also not participate in governance of CGAP). Establish a European office to better serve the majority of its members Paris CGAP office established in 2002. who are headquartered there. CGAP 74 Annex G OVERALL ASSESSMENT OF MONITORING AND EVALUATION Table I.5. To what extent does monitoring and evaluation comply with OED standards for best practice? Sub- Not High stan- Mod- Neg- Don't appli- Comments tial est ligible know cable 1. Clear and coherent program Program objectives and strategies under Phase II were specified, but no objectives and strategies x verifiable indicators were developed to assess achievement toward these objectives. 2. A structured set of quantitative See comments (1) above. or qualitative indicators x 3. Systematic and regular See comments (1) above. processes for data collection x and management 4. Independence of program- The 2002 Evaluation was commissioned and overseen by EXCOM, but level evaluations x managed by the Secretariat. 5. Effective feedback of CGAP is now focusing more on standard-setting and knowledge creation and evaluations on the strategic x dissemination than on capacity-building grants to MFIs. focus of the program 6. Effective feedback of Charter was developed and approved for Phase III. Modifications to evaluations on organization, x composition of EXCOM and the elimination of PAG. management, and financing Note: The criteria are based upon OED's standards of best practice as identified in OED's Report, Monitoring and Evaluation Plans in Staff Appraisal Reports Issued in Fiscal Year 1995, December 29, 1995, Report No. 15222, pp. 23-24. CGAP 75 Annex H Annex H. Governance of CGAP Legal status World Bank is legal entity. Governance and Council of Governors (CG) management bodies Executive Committee (Excom) Operational Team Investment Committee (IC) Location of program World Bank management unit. Written charter and Yes. Approved September 2002. date Approved by CG. Other constitutional- TF agreements between member donors and the World Bank. level documents Governing Body Name Council of Governors (CG) Current size 29 Current membership World Bank, 9 other international/regional organizations, 17 bilateral donors, and 2 private foundations Membership criteria Public bilateral and multilateral donor agencies and private foundations with a significant involvement in the development of financial services for the poor. Other types of institutions involved in microfinance may be considered for membership at some future point. Membership Contribute funding in cash, through flexible mechanisms, to carry out CGAP's responsibilities operations. Adopt and actively promote implementation of standards and basic principles of donor support to microfinance as outlined in CGAP consensus documents including "Micro and Small Enterprise Finance: Guiding Principles for Selecting and Supporting Intermediaries," or successor documents to be agreed upon. Participate actively in the CGAP Annual Meetings with representatives who are appropriately placed with the organization and have the necessary knowledge to support mainstreaming of microfinance within their respective agency. Agree to open the institution to sharing experiences with other donors, including providing aggregate information on its portfolio of microfinance activities, reporting on how it is applying the consensus documents, and providing contact information for all staff working in microfinance. This information should be submitted to CGAP on an annual basis prior to the CGAP Annual Meeting. Minimum annual None financial contribution Functions of governing The Council of Governors has the authority to: body · Set policy and approve strategy. · Elect Excom members on a constituency basis. · Approve Excom members from the microfinance industry leaders. · Confirm the appointment of the Executive Director of the CGAP Operational Team. · Adopt consensus documents. · Amend the CGAP Charter. · Approve extension or disbandment of CGAP. · Provide input to the annual work plan and budget. · Receive and approve annual reports. · Request an external audit in compliance with the bilateral agreements, if needed. CGAP 76 Annex H Meetings Annual Decisions Whenever possible by consensus. When necessary, on the basis of a simple majority of those present. Each contributing member has one vote, but may be represented on the CG by two representatives. Chair of governing World Bank INFVP body Functions of the Chair· To call and host the CG Annual Meetings and other CG meetings. of the governing body · To represent the World Bank senior management at these meetings. · To maintain oversight of the Executive Director Executive Body Name Executive Committee (Excom) Current size 9 members. All serve for two years, renewable for one more year. Five members constitute a quorum. · Four elected by the CG constituencies · One appointed by the World Bank · Four microfinance industry leaders with different backgrounds. Nominations are solicited from the CG, Excom, and the Operational Team. After consultation with the Excom and CG Chair, the Executive Director proposes a shortlist for CG approval. · Executive Director (ex-officio). Excom elects a Chairperson from among its donor members for a two-year term, renewable. Reports to CG Functions. Provide strategic direction and support for CGAP Operational Team, both externally and internally. Engage in improving aid effectiveness in member donor agencies' microfinance operations. Approve annual work plans and budgets. Propose amendments to the Charter, when necessary, for approval by the CG. Advisory Body Name(s) From 1995 to 2003, CGAP had an advisory body called the Policy Advisory Group that was separate from the other governance bodies and was composed of distinguished microfinance practitioners and experts. This was disbanded as a separate body in 2003 and merged into Excom. Proportion from developing countries Tenure Members appointed/ removed by Reports to Functions of advisory body Chair of advisory body Functions of Chair of advisory body CGAP 77 Annex H Other Committees Name, size, and Investment Committee. functions 7 members, including its Chair, designated by the World Bank (given its governance and oversight role delegated by bilateral TF agreements). The majority of members must be senior World Bank staff, one of whom must be from the World Bank Legal Vice-Presidency. Responsibilities: · Fiduciary responsibility for CGAP's grants to external entities · Ensure adherence of grants to the criteria and overall strategy established by the CG. · Approve external grants greater than US$250,000, and approve on a non- objection basis, external grants between US$100,000 and US$250,000. · Reports on investment decisions to Excom. Management Comprises Operational Team headed by Executive Director Program Manager INF VP (= Chair of CG) appointed/ removed by Reports to INF VP (= Chair of CG) Executive Director submits quarterly reports to Excom, and the Annual Report to Excom for approval and further submission to the CG. The Executive Director consults closely with the Excom on all key issues of CGAP's strategy and work program. Functions Responsibilities of the Executive Director include: · Strategic direction · Forward planning. The annual workplan is sent to the Excom and the CG prior to the CGAP Annual Meeting to solicit feedback. · Personnel decisions · Liaising with Excom · Advancing CGAP's vision among broad range of stakeholders in the microfinance sector, in development agencies and donor governments, developing country policymakers, civil society (NGOs, community groups, foundations, etc.) commercial banks, the media, and the general public at large · Approvals of grants or projects up to $100,000 Staffing 22 Location of staff Headquarters in Washington, and a small representative office in Europe to better serve the concentration of CGAP donors and other actors in Europe. Seconded staff 0 Evaluation and Audit Evaluation Previous program-level evaluations were carried out in 1997 and 2002, towards the ends of Phases I and II. A third evaluation to review CGAP's performance will be carried out with a view to recommending whether and how CGAP should continue after June 2008. Audit The World Bank provides for fiduciary oversight and internal audits of CGAP. An internal audit was carried out in FY02 by IAD for the DGF. CGAP 78 Annex I Annex I. Management of CGAP CGAP's Operational Team consists of 22 full-time staff headed by the Executive Director and housed administratively in the World Bank's INF Vice-Presidency. Most staff are located in CGAP's headquarters in Washington, D.C.. CGAP recently established a small representative office in Europe to better serve the concentration of CGAP donors and other actors in Europe. CGAP serves three major clients ­ donors, microfinance institutions and other microfinance providers, and the microfinance industry as a whole. CGAP staff provide services and products directly to clients ­ providing strategic and technical advice, developing new models of financial services provision to the poor, developing technical tools, managing capacity-building programs, delivering training, conducting action research, and disseminating information to the industry at large. Selectivity is critical given CGAP's limited resources and the need to capitalize on the wealth of experience and expertise in the microfinance industry. The Operational Team applies the following criteria in determining which high-priority activities to undertake. · Is there clear demand from our clients for this service? · Does it leverage the CGAP team's human and financial resources, and does it benefit many parties? · Is CGAP uniquely positioned to provide this service, or can someone else do it better? In addition, while CGAP does not principally operate as a donor, it has a small investment fund at its disposal. The Executive Director can approve external grants and projects up to US$100,000, while the Investment Committee approves grants over US$250,000 and approves on a no-objection basis grants between US$100,000 and US$250,000. The criteria for grants and projects are the following: · Activities should reinforce CGAP's strategic priorities determined by the CG and Excom and meet one or more of the following characteristics: provide services to the industry as a whole; promote innovation or learning value for the industry as a whole; and play a catalytic role (i.e. allows CGAP to bring together various stakeholders towards a common goal) · Activities should provide significant leverage for CGAP. These include investments that benefit the microfinance industry as a whole, affecting a large number of institutions and/or having the potential to benefit large numbers of poor clients · Microfinance providers are eligible for grants only if the institution has achieved significant scale (> 3,000 current poor and very poor customers), has achieved operational self- sufficiency and is moving on a clear and credible path to full financial self-sufficiency. CGAP staff manage and monitor all activities but collaborate closely with external partners. Performance standards for each of these activities are regularly reviewed, analyzed and discussed with the relevant partners. CGAP does not contract with Bank staff outside of CGAP to supervise CGAP activities or grants. The Operational Team reports to and interacts with the CG and the Excom constantly about new ideas, ongoing activities, and other developments through email, telephone calls, and meetings. The Executive Director submits quarterly reports to Excom and the Annual Report to Excom for approval and further submission to the CG. These reports include CGAP's major results, activities and financial information, including the approval of new projects and grants. The Executive Director submits the annual workplan to the CG prior to the CGAP Annual Meeting to solicit feedback. CGAP 79 Annex I Overall Assessment of Governance and Management To what extent does governance and management apply the principles of a well- functioning corporate governance system, enunciated by the OECD Business Sector Advisory Group? Sub- Not High stan- Mod- Neg- Don't appli- Comments tial est ligible know cable 1. Transparency /1 X Published Annual Reports open to the public since 1998 and available on web-site. 2. Clarity of roles & New charter was composed and ratified in responsibilities /2 X FY03. 3. Fairness to CGAP has shifted its grant-making from a immediate X demand-driven to a supply-driven approach clients /3 that is less transparent than before. 4. Accountability to CGAP is a "donor club." Membership is donors /4 X restricted to multilateral and bilateral donor agencies and private foundations. 5. Accountability to Developing country governments have no developing direct role in the strategic direction or countries X oversight of CGAP ­ only indirectly through membership on the Bank's Board and as occasional partners in CGAP's country-level activities. 6. Accountability to Technical specialists (such as supervisors, scientists/ auditors, and regulators) in the broader MFI professionals X industry play a role mainly at the activity level, not the program level. 7. Accountability to CGAP sees MFIs as one of its principal client other X groups. Four industry representatives serve stakeholders on EXCOM. /1 Transparency ­ the program provides both shareholders and stakeholders with the information they need in an open and transparent manner (such as accounting, audit, and non-financial but material issues) /2 Clarity of roles and responsibilities ­ of the various officers and bodies that govern and manage the program as well as clear mechanisms to modify and amend the governance and management of the program in a dynamic context /3 Fairness ­ the program does not favor some immediate clients over others (such as Bank staff, central governments and their agencies, municipal agencies, local authorities, private service providers, NGOs, and community organizations) /4 Accountability ­ of the program for the exercise of power over resources to each of the four groups of stakeholders listed here, "other stakeholders" being those not otherwise mentioned with a legitimate interest in the activities of the program (such as international NGOs) CGAP 80 Annex J Annex J. Current Membership of the CGAP Council of Governors, Executive Committee, and Investment Committee3 COUNCIL OF GOVERNORS Bilateral Member Donors (16) Australia Australian Agency for International Development (AusAID) Mr. Bernadette Whitelum Corporate Policy, Policy and Multilateral Branch Belgium Direction Générale de la Coopération au Développement (DGCD) Mr. Philippe Gerard Service Banque de Développement D41 Mr. Charles Bois d'Enghien Service Banque de Développement D41 Mr. Charles Tollenaere EQS Canada Canadian International Development Agency (CIDA) Mr. Jonathan Rothschild Senior Economist Economic Policies (YDA), Policy Branch Ms. Doris L. Wong Enterprise Specialist Denmark Royal Danish Ministry of Foreign Affairs Mr. Morten Elkjaer Chief Adviser Technical Advisory Services 3Updated May 5, 2004. CGAP 81 Annex J Finland Ministry for Foreign Affairs Department for International Development Cooperation Mr. Tuukka Castren Economic Adviser Unit for Sectoral Policy Advice (KYO-12) France Agence Française de Développement (AFD) Ms. Anne Clerc Ms. Martha Stein-Sochas Chef de Division Ingénierie Financière French Ministry of Foreign Affairs Mr. Jean Bruschi Chargé de mission microfinance et qualité Germany Kreditanstalt fur Wiederaufbau (KfW) Mr. Roland Siller Senior Financial Expert Financial Sector Competency Center Mr. Martin Hagen Senior Sector Economist Financial Sector Competency Center Federal Ministry for Economic Cooperation and Development (BMZ) Ms. Ulrike Haupt Leiterin Referat Zusammenarbeit und Entwicklung Mr. Uwe Schmidt Deputy Head of Division Society for Technical Cooperation (GTZ) Mr. Dirk Steinwand Head of Section Financial Systems Development Mr. Roland Gross CGAP 82 Annex J Italy Ministry of Foreign Affairs Directorate General for Development Cooperation - DGCS Mr. Sebastiano Salvatori Mr. Daniele Di Ceglie Japan The World Bank Office of the Executive Director for Japan Ms. Kiyo Oi Advisor to the Executive Director Japan Bank for International Cooperation (JBIC) Mr. Teruhisa Oi Mr. Sonoko Takahashi Ministry of Foreign Affairs Mr. Kazunori Hosoya Ms. Wakana Kanikawa Ministry of Finance Ms. Miwa Miyata Development Institution Division Luxembourg Ministry of Finance Mr. Georges Heinen Head, Multilateral Development Financing Mr. Arsène Jacoby Desk Officer for Microfinance Activities Mr. Miguel Marques Attaché de Gouvernement 1er en Rang CGAP 83 Annex J Ministère des Affaires Etrangères Direction de la Coopération au Développement Mr. Thierry Lippert Desk officer for Microfinance Activities The Netherlands Ministry of Foreign Affairs Mr. Johan de Waard Policy Adviser N(O)VIB Oxfam Netherlands Mr. Stijn C.J. Albregts Coordinator Dutch Microfinance Platform Norway Norwegian Agency for Development Cooperation (NORAD) Ms. Hege Gulli Advisor Technical Department, Unit for Private Sector Development Ms. Brita Naess Advisor Norway Royal Ministry of Foreign Affairs Mr. Berit Fladby Senior Advisor Mr. Torgeir Fyhri Sweden Swedish International Development Cooperation Agency (SIDA) Ms. Eva Bursvik Adviser, FinSys Team Division for Trade, Private Sector Development and Financial Systems Department for Infrastructure and Economic Cooperation Ms. Karin Dahlstrom Microfinance Advisor CGAP 84 Annex J Mr. Jan Grafstrom Coordinator Financial Sector Development Switzerland Swiss Agency for Development and Cooperation Mr. Hansruedi Pfeiffer Senior Programme Officer Responsible for SDC's Financial Sector Operations United Kingdom Department for International Development (DFID) Mr. David J.N. Stanton Chief Enterprise Development Adviser Office of the Chief Adviser Mr. Richard Boulter Team Leader Financial Sector Team Policy Division (9E7) United States USAID Ms. Kate McKee Director Office of Microenterprise Development Multilateral Member Donors (9) African Development Bank Mr. Ross Croulet Coordinator, Central Microfinance Unit (OCMU) Mr. Philibert Afrika Director Operations Policies and Review Department Asian Development Bank Mr. Jan Van Heeswijk Director General Agriculture and Social Sectors Department (West) CGAP 85 Annex J Mr. Nimal Fernando Lead Rural Finance Specialist European Commission EuropeAid Mr. Odoardo Como Mr. Andreas Schwarz Administrator Development Directorate-General European Bank for Reconstruction and Development (EBRD) Ms. Elizabeth Wallace Director Financial Institutions Team Inter-American Development Bank (IDB) Mr. Alvaro Ramirez Chief, Micro, Small and Medium Enterprise Division Mr. Dieter Wittkowski Senior Operations Specialist Micro, Small and Medium Enterprise Division International Fund for Agricultural Development (IFAD) Mr. Henri Dommel Technical Advisor Rural Finance Technical Advisory Division International Labour Organization (ILO) Mr. Bernd Balkenhol Program Manager Social Finance Programme Employment Sector Mr. Craig Churchill Senior Microfinance Expert Social Finance Programme Employment Sector United Nations Development Programme (UNDP) Mr. Peter Kooi Director Special Unit for Microfinance CGAP 86 Annex J United Nations Capital Development Fund/United Nations Development Programme (UNCDF/UNDP) Mr. John Tucker Deputy Director Special Unit for Microfinance United Nations Capital Development Fund/United Nations Development Programme (UNCDF/UNDP) World Bank Ms. Nemat Talaat Shafik, CG Chair Vice-President Infrastructure Mr. William F. Steel Lead Specialist Mr. Carlos Cuevas Principal Financial Specialist Foundation Member Donors (2) The Ford Foundation Mr. Frank DeGiovanni Director of Economic Development Argidius Foundation Mr. Jim Shelter Mr. Koenraad Verhagen Consultant Co-operative Economy and Microcredit CGAP 87 Annex J EXECUTIVE COMMITTEE MEMBERS Mr. Fazle Hasan Abed Founder & Executive Director BRAC Mr. Brian Branch Vice President World Council of Credit Unions Mr. Carlos Cuevas Principal Financial Specialist The World Bank Mr. Frank DeGiovanni Director of Economic Development The Ford Foundation Mr. Nimal Fernando Lead Rural Finance Specialist Asian Development Bank Ms. Hege Gulli Advisor Technical Department, Unit for Private Sector Development Norwegian Agency for Development Cooperation (NORAD) Mr. Carlos Labarthe Director General Financiera Compartamos David Stanton Chief Enterprise Development Adviser Enterprise Development Group Department for International Development (DFID) Ms. Marilou van Golstein Brouwers Senior Fund Manager Triodos Bank Ms. Elizabeth Littlefield, ex-officio Director and CEO CGAP CGAP 88 Annex J INVESTMENT COMMITTEE MEMBERS Jean-Francois Rischard Vice President The World Bank Carlos Cuevas Lead Financial Economist The World Bank Claudia Morgenstern Senior Adviser The World Bank Elizabeth O. Adu Deputy General Counsel, Operations The World Bank Assaad Jabre Vice President, Operations The World Bank Lynn Bennett Adviser The World Bank Lynne D. Sherburne-Benz Lead Financial Economist The World Bank Antony Thompson Lead Financial Sector Specialist The World Bank Observers Kate McKee USAID Frank De Giovanni Ford Foundation CGAP 89 Annex L Annex K. CGAP, Sources and Uses of Funds, 2003 Sources of funds Uses of funds Revenues 13,577 Expenses 15,247 Interest Income 536 By program activities Financial contributions 13,041 Microfinance institutions By restrictions Member donors Core 12,451 Microfinance industry Designated 590 Publications Other Overhead Core contributions by type of donor By execution Multilateral 7,375 Bank-executed Bilateral 4,776 Recipient-executed Private foundations 300 Third-party executed Core contributions by donor By region World Bank 6,725 Sub-Saharan Africa United States 800 East Asia & Pacific Denmark 424 South Asia Norway 402 Europe & Central Asia Netherlands 401 Middle East & North Africa Switzerland 400 Latin America & Caribbean United Kingdom 400 Global/multi-regional Sweden 381 Canada 318 By cost object IFAD 300 Grants/initiatives committed 9,809 Japan 300 Staff salaries and benefits 2,616 Belgium 295 Office and occupancy costs 884 Finland 251 Travel 334 AsDB 250 Publications, etc. 1,397 Germany 242 CG and EXCOM Meetings 208 Ford Foundation 200 France 162 AfDB 100 Argidius Foundation 100 Australia Italy IDB Core contributions by source BB 0 DGF 6.725 TFs 5,726 Total sources of funds 13,577 Co-financing of CGAP- supported activities 90 Annex L Annex L. World Bank Lending to "Micro & SME Finance" Table L.1. New Commitments and Annual Disbursements, 1990-2004 Number of New Projects with Micro & SME finance Components New Commitments (US$ Millions) Annual Disbursements Micro & SME Micro & SME (US$ Millions) Entire projects finance Entire projects finance /1 components only components only 1990 10 5.4 761.9 377.0 68.3 1991 11 6.8 859.8 503.0 235.3 1992 5 1.6 451.0 116.0 156.3 1993 3 0.9 50.7 21.0 222.6 1994 10 1.9 537.0 115.9 209.2 1995 9 2.3 265.6 66.4 141.6 1996 12 6.6 485.3 295.8 162.2 1997 10 4.6 400.8 205.5 125.0 1998 13 4.9 806.0 161.4 176.1 1999 17 5.1 779.5 176.8 153.5 2000 14 4.3 439.6 128.9 121.8 2001 11 3.0 1,184.1 235.2 116.5 2002 11 2.9 807.9 111.0 114.0 2003 9 2.0 927.6 462.7 166.3 2004 /2 6 1.6 686.0 276.8 264.8 Total 151 53.8 9,442.7 3,253.4 2,433.7 Source: World Bank Business Warehouse (data downloaded on April 1, 2004) /1 Total disbursements by each project in each year are weighted by the proportion of micro & SME finance in each project. /2 Through March 31, 2004 91 Annex L Table L.2. New Commitments by Region, 1990-2004 (No. of Projects with Micro & SME Finance Components) Africa East Asia Europe & Latin Middle East South Total & Pacific Central Asia America & & North Asia Caribbean Africa 1990 5 2 1 2 10 1991 1 4 2 1 1 2 11 1992 1 2 2 5 1993 2 1 3 1994 3 1 2 1 3 10 1995 5 1 1 1 1 9 1996 2 1 6 2 1 12 1997 1 4 3 1 1 10 1998 4 4 2 2 1 13 1999 2 4 6 1 3 1 17 2000 7 1 3 2 1 14 2001 2 5 3 1 11 2002 3 2 1 3 1 1 11 2003 3 4 1 1 9 2004 /1 2 1 1 1 1 6 Total 43 18 35 28 17 10 151 Percent 28% 12% 23% 19% 11% 7% 100% /1 Through March 31, 2004 Table L.3. New Commitments by Region, 1990-2004 (Total of Micro & SME Finance Components Only) Africa East Asia Europe & Latin Middle East South Total & Pacific Central Asia America & & North Asia Caribbean Africa 1990 1.7 2.0 0.9 0.8 5.4 1991 1.0 3.9 0.3 0.0 0.2 1.5 6.8 1992 0.3 1.2 0.2 1.6 1993 0.7 0.2 0.9 1994 0.4 0.5 0.5 0.1 0.5 1.9 1995 1.4 0.5 0.0 0.0 0.3 2.3 1996 0.6 1.0 3.6 0.9 0.5 6.6 1997 0.1 3.0 0.4 0.1 1.0 4.6 1998 1.0 1.0 1.9 0.5 0.4 4.9 1999 1.1 0.6 2.2 0.0 0.7 0.5 5.1 2000 2.4 0.0 0.7 1.1 0.1 4.3 2001 0.2 1.4 0.5 1.0 3.0 2002 0.3 0.4 0.1 1.5 0.2 0.4 2.9 2003 0.3 1.6 0.2 0.0 2.0 2004 /1 0.4 0.3 0.2 0.1 0.7 1.6 Total 11.8 7.8 15.1 9.2 4.2 5.8 53.8 Percent 22% 15% 28% 17% 8% 11% 100% /1 Through March 31, 2004 92 Annex L Table L.4. New Commitments by Region, 1990-2004 (US$ Millions, Entire Project) Africa East Asia Europe & Latin Middle East South Total & Pacific Central Asia America & & North Asia Caribbean Africa 1990 175.8 66.1 170.0 350.0 761.9 1991 11.3 404.3 200.0 20.7 140.0 83.5 859.8 1992 35.0 160.0 256.0 451.0 1993 28.7 22.0 50.7 1994 25.6 96.0 23.4 50.0 342.0 537.0 1995 149.8 0.0 40.0 10.8 65.0 265.6 1996 110.0 122.0 79.8 53.5 120.0 485.3 1997 30.0 84.3 151.5 30.0 105.0 400.8 1998 54.7 271.3 10.0 455.0 15.0 806.0 1999 21.3 405.0 92.5 60.7 110.0 90.0 779.5 2000 117.0 27.5 167.0 53.1 75.0 439.6 2001 30.0 927.8 75.3 151.0 1,184.1 2002 133.8 218.7 13.8 431.5 5.0 5.0 807.9 2003 246.0 593.7 12.4 75.6 927.6 2004 /1 56.8 303.1 28.1 60.0 238.0 686.0 Total 1,225.8 1,544.8 2,101.7 2,305.9 1,166.4 1,098.1 9,442.7 Percent 13.0% 16.4% 22.3% 24.4% 12.4% 11.6% 100.0% /1 Through March 31, 2004 Table L.5. New Commitments by Region, 1990-2004 (US$ Millions, Micro & SME Finance Components Only) Africa East Asia Europe & Latin Middle East South Total & Pacific Central Asia America & & North Asia Caribbean Africa 1990 29.9 66.1 153.0 128.0 377.0 1991 11.3 387.9 25.0 0.6 25.2 53.0 503.0 1992 9.1 76.4 30.5 116.0 1993 17.4 3.5 21.0 1994 3.7 43.2 4.5 6.0 58.5 115.9 1995 45.5 0.0 0.4 0.4 20.2 66.4 1996 33.9 115.9 52.6 33.4 60.0 295.8 1997 3.3 73.1 27.5 1.8 99.8 205.5 1998 8.2 35.2 9.3 102.6 6.2 161.4 1999 16.4 60.5 30.1 0.6 24.3 45.0 176.8 2000 52.2 0.3 21.3 49.9 5.3 128.9 2001 5.2 77.7 7.3 145.0 235.2 2002 13.4 35.1 1.5 57.9 1.1 2.0 111.0 2003 21.9 436.7 1.9 2.3 462.7 2004 /1 10.6 90.9 5.6 3.0 166.6 276.8 Total 282.1 678.0 463.0 825.1 363.7 641.6 3,253.4 Percent 8.7% 20.8% 14.2% 25.4% 11.2% 19.7% 100.0% /1 Through March 31, 2004 93 Annex L Table L.6. New Commitments by Share of the Microfinance Component, 1990-2004 (No. of Projects with Micro & SME Finance Components) 1-10% 11-25% 26-50% 51-75% 76-99% 100% Total 1990 1 1 4 0 1 3 10 1991 2 2 1 0 2 4 11 1992 1 2 1 0 1 0 5 1993 1 1 0 1 0 0 3 1994 3 4 3 0 0 0 10 1995 3 2 3 1 0 0 9 1996 1 4 3 0 3 1 12 1997 2 4 0 0 3 1 10 1998 3 2 5 1 2 0 13 1999 4 5 6 0 2 0 17 2000 3 6 2 2 1 0 14 2001 4 4 0 1 2 0 11 2002 4 3 2 1 1 0 11 2003 5 2 1 0 1 0 9 2004 /1 1 3 1 1 0 0 6 Total 38 45 32 8 19 9 151 Percent 25% 30% 21% 5% 13% 6% 100% /1 Through March 31, 2004 Table L.7. New Commitments by Share of the Microfinance Component, 1990-2004 ($US Millions, Micro & SME Finance Components Only) 1-10% 11-25% 26-50% 51-75% 76-99% 100% Total 1990 1.1 6.2 148.1 0.0 153.0 68.6 377.0 1991 3.6 47.2 27.0 0.0 222.9 202.3 503.0 1992 7.1 40.4 9.1 0.0 59.4 0.0 116.0 1993 0.2 3.5 0.0 17.3 0.0 0.0 21.0 1994 16.2 12.1 87.5 0.0 0.0 0.0 115.9 1995 2.0 3.8 38.2 22.5 0.0 0.0 66.4 1996 2.2 16.4 94.1 0.0 153.2 30.0 295.8 1997 2.1 33.2 0.0 0.0 121.0 49.3 205.5 1998 15.5 43.1 90.4 3.1 9.3 0.0 161.4 1999 6.0 38.7 112.9 0.0 19.2 0.0 176.8 2000 7.6 23.7 12.2 36.0 49.5 0.0 128.9 2001 33.5 20.8 0.0 19.2 161.8 0.0 235.2 2002 53.8 32.6 7.1 13.5 4.0 0.0 111.0 2003 25.4 4.3 29.1 0.0 404.0 0.0 462.7 2004 /1 3.0 16.2 90.9 166.6 0.0 0.0 276.8 Total 179.4 342.0 746.5 278.1 1,357.2 350.2 3,253.4 Percent 6% 11% 23% 9% 42% 11% 100% /1 Through March 31, 2004 94 Annex L Table L.8. Annual Disbursements by Region, 1990-2004 (US$ Millions, Micro & SME Finance Components Only) Africa East Asia Europe & Latin Middle East South Total & Pacific Central Asia America & & North Asia Caribbean Africa 1990 16.0 0.0 20.7 3.0 17.0 11.5 68.3 1991 8.1 31.1 64.6 11.6 108.2 11.8 235.3 1992 20.8 41.1 33.9 10.3 21.8 28.5 156.3 1993 32.0 124.9 10.6 13.1 18.6 23.5 222.6 1994 35.6 132.7 9.0 13.9 7.5 10.5 209.2 1995 34.3 52.6 15.6 19.0 7.3 12.8 141.6 1996 25.9 20.5 56.3 24.2 21.9 13.4 162.2 1997 13.0 9.4 22.1 33.9 20.2 26.5 125.0 1998 22.2 15.2 19.9 23.3 48.5 47.0 176.1 1999 18.7 52.8 18.9 20.0 13.4 29.7 153.5 2000 12.4 41.0 16.8 10.6 14.4 26.7 121.8 2001 15.9 26.0 28.9 16.2 7.4 22.1 116.5 2002 15.1 12.9 9.0 18.3 4.8 54.0 114.0 2003 17.4 11.8 18.7 60.8 5.0 52.5 166.3 2004 /1 20.6 10.4 11.2 191.2 5.9 25.4 264.8 Total 307.8 582.3 356.4 469.4 322.0 395.7 2,433.7 Percent 12.6% 23.9% 14.6% 19.3% 13.2% 16.3% 100.0% /1 Through March 31, 2004 95 Annex M Annex M. Overall Assessment in Relation to Development Committee Criteria Current Level of Consistency with the Development Committee Criteria for the Bank's Involvement in Global Programs Sub- High stan- Mod- Neg- Comments tial est ligible 1. An international consensus CGAP represents a consensus among currently exists that global donors that access to sustainable micro- collective action is required finance institutions can contribute to economic growth and poverty reduction. The X MicroCredit Campaign represents such consensus. However, there remain differences of opinion regarding approaches ­ "financial systems" vs. "poverty lending" ­ and on what kind of global collective action is needed. CGAP's Phase II evaluation concluded that, 2. The program is currently while CGAP has been successful in known to be adding value to promoting best practices and direct achieving the Bank's investments in MFIs, its achievements with development objectives of X respect to donor mainstreaming and fostering poverty alleviation and favorable national policy environments have sustainable development. been much more modest. CGAP members disagree on the need to monitor program impacts on the poor. 3. The Bank's presence is The Bank contributed 70 percent of CGAP's currently catalyzing other non- funding during its first eight years (1995- Bank resources for the X 2003), and still 52 percent in its most recent program. fiscal year (FY03). 4. The Bank is currently playing The Bank's role in CGAP has been too up to its comparative dominant, so that other donors evidence advantages at the global level X weak ownership of CGAP. The Bank has not /1 sought to extend ownership to its developing country members. 5. The Bank is currently playing Linkages between CGAP and the Bank's up to its comparative country operations have been very weak. The advantages at the country Bank has not helped CGAP to engage in level /2 X dialogue with national policy makers to improve national policy environments for microfinance. /1 Global mandate and reach, convening power, and mobilizing financial resources /2 Multi-sectoral capacity, expertise in country and sector level analysis, in-depth country-level knowledge. 96 Annex N Annex N. Results from the Partner Survey Note: This survey was administered to partners involved in the governance of the six infrastructure and private sector development programs, whether serving on the governing body, the professional/technical advisory committee, or, in the case of CGAP, the Investment Committee. The response rate for CGAP was 25 respondents out of 44 (57 percent). Each bullet in the written open-ended responses represents one respondent. Question 1. The objectives of the program can be achieved more cost-effectively by the partners of the program working together within the framework of the program than by the individual partners acting alone. (Results in rank order by program) ESMAP CGAP CA WSP PPIAF infoDev All Agree 90% 83% 79% 86% 80% 50% 79% 18 20 15 6 8 6 73 Partially Agree 10% 17% 21% 0% 10% 50% 18% 2 4 4 0 1 6 17 Disagree 0% 0% 0% 14% 10% 0% 2% 0 0 0 1 1 0 2 No Answer 2 1 1 0 2 3 9 Total 20 24 19 7 10 12 92 The responses are significantly different among programs (P = 3.1 percent). Written Comments · Lack of decentralization. Process competence. · Clear objectives will be achieved the same with or without CGAP. · But individual partners have not "mainstreamed" CGAP well within their organizations and/or field offices. · There is a tendency for CGAP (secretariat and donor members) to dominate too much the microfinance finance industry development agenda. Practitioners are not that well organized. · This way each can leverage resources better. · Except that CGAP has so far taken an unnecessarily narrow view of its mandate, both in content and geography. · Couldn't agree more! · CGAP has been a guide, a standard setter in the industry and this has been very important for the whole field. · We have already seen CGAP play an important part in consolidating many of the advances and standards in the industry. This level of consolidation of thought and standards would not have been achieved by individual partners working on their own. · There is a great need to harmonize standards of best practice in microfinance and agree upon common formats and elements for reporting data from microfinance institutional partners and donor agencies. 97 Annex N Question 2. In your opinion, at what level do the principal benefits of CGAP accrue? (Results in rank order by level) Primarily Primarily Primarily Primarily global regional individual national 1 = most significant 73% 25% 11% 16% 16 5 2 3 2 5% 40% 33% 21% 1 8 6 4 3 14% 25% 28% 32% 3 5 5 6 4 = least significant 9% 10% 28% 32% 2 2 5 6 No Answer 3 5 7 6 Total 22 20 18 19 The responses are significantly different among levels (P = 0.1 percent). Question 3. In your opinion, the objectives of the program are realistic compared to the currently available resources? (Results in rank order by program) PPIAF WSP CGAP ESMAP CA infoDev All Agree 70% 57% 58% 50% 33% 17% 47% 7 4 14 10 6 2 43 Partially Agree 30% 43% 38% 45% 50% 58% 44% 3 3 9 9 9 7 40 Disagree 0% 0% 4% 5% 17% 25% 9% 0 0 1 1 3 3 8 Not Sure 0 0 0 1 1 0 2 No Answer 2 0 1 1 1 3 8 Total 10 7 24 20 18 12 91 The responses are not significantly different among programs (P = 18.9 percent). Written Comments · Realistic but not "dramatic/drastic" enough, having the partners that compose it. · Resources need to be augmented to achieve more effective results. · The funds available for institutional support to scale-up well-managed MFIs are woefully inadequate. · Realistic thanks to a highly effective Secretariat. · As long as the funds contributed by the Bank are leveraged by other donor funds. 98 Annex N · What CGAP achieves is significant, very significant, but even this is probably not up to what CGAP's objectives really call for. For example, there has probably not been a significant increase in services to the very poor (as opposed to the less poor). · If CGAP concentrates on key strategic issues, then I agree. We have already seen CGAP play an important part in consolidating many of the advances and standards in the industry. This level of consolidation of thought and standards would not have been achieved by individual partners working on their own. Question 4. The activities of the program are relevant to the priorities of developing and transition countries. (Results in rank order by program) WSP CA PPIAF CGAP ESMAP infoDev All Agree 100% 89% 80% 71% 70% 50% 75% 7 16 8 17 14 6 68 Partially Agree 0% 11% 20% 25% 25% 42% 24% 0 2 2 6 5 5 20 Disagree 0% 0% 0% 4% 5% 8% 4% 0 0 0 1 1 1 3 No Answer 0 2 2 1 1 3 9 Total 7 18 10 24 20 12 91 The responses are not significantly different among programs (P = 50.0 percent). Written Comments · More needs to be done to encourage or facilitate delivery of services to the very poor. This group makes up at least one-third of the populations of these countries but, as noted above, not much progress has been made in ensuring that this group is reached. · Mainly in Latin America. · The main mission of CGAP is to work in the developing world and I think it has been very successful in doing so. · Microfinance as such is very relevant for poverty reduction and pro-poor private sector development. The focus on donor coordination and mainstreaming of standards is also very valuable to decrease developing countries' transaction costs in connecting with international development cooperation. 99 Annex N Question 5. In your opinion, how successful has CGAP been in achieving the following objectives? (Results in rank order by objectives) Knowledge Global Research Donor National manage- Advocacy rules & & develop- Capacity coor- Mobilizing policy ment standards ment building dination resources reform High 46% 29% 17% 8% 17% 25% 8% 0% 11 7 4 2 4 6 2 0 Substantial 50% 50% 67% 71% 50% 21% 13% 29% 12 12 16 17 12 5 3 7 Modest 4% 13% 17% 21% 33% 50% 63% 50% 1 3 4 5 8 12 15 12 Negligible 0% 8% 0% 0% 0% 4% 13% 17% 0 2 0 0 0 1 3 4 Not 0% 0% 0% 0% 0% 0% 4% 4% Applicable 0 0 0 0 0 0 1 1 No Answer 1 1 1 1 1 1 1 1 Total 24 24 24 24 24 24 24 24 The responses are significantly different among objectives (P = 0.0 percent). Question 6a. From your perspective, to what degree have developing and transition countries been involved in the design, governance, implementation, and monitoring and evaluation of CGAP at the global level? (Results in rank order by aspects of the program) Design Implementation Governance Monitoring & Evaluation Total High 5% 0% 0% 5% 2% 1 0 0 1 2 Substantial 33% 29% 29% 15% 27% 7 6 6 3 22 Modest 24% 38% 38% 50% 37% 5 8 8 10 31 Negligible 38% 33% 33% 30% 34% 8 7 7 6 28 Don't Know 3 3 3 3 12 No Answer 1 1 1 2 5 Total 21 21 21 20 83 The responses are not significantly different among activities (P = 78.1 percent). 100 Annex N Written Comments · For the same reason as question 4. PAG members, though many from these countries, have no official "representation" status of their countries or regions. · Most work is done within the Secretariat, which has capable people, but they don't know much about poverty and microfinance in Asia. · If you mean governments, they have not been much involved (nor is it a big role for them in CGAP). · I think this is one of the main functions of the PAG. From the practitioners' point of view and from the developing world point of view, to influence CGAP so that its objectives, programs, etc. are demand-driven and useful for the final clients. Question 6b. From your perspective, to what degree have beneficiaries been involved in the design, implementation, and monitoring and evaluation of the activities of CGAP at the country level? (Results in rank order aspects of the program) Design Implementation Monitoring & Evaluation Total High 9% 14% 14% 12% 2 3 3 8 Substantial 32% 18% 23% 24% 7 4 5 16 Modest 55% 64% 50% 56% 12 14 11 37 Negligible 5% 5% 14% 8% 1 1 3 5 Don't Know 2 2 2 6 No Answer 1 1 1 3 Total 22 22 22 66 The responses are not significantly different among activities (P = 79.6 percent). Written Comments · Again, most is done in the Secretariat. · From my experience, I can say that local partners are given a good opportunity to play an important role in improving access to microfinance through CGAP programs. · Muhammad Yunus of Grameen Bank in Bangladesh was involved in the early design stage. The "Policy Advisory Group" consisting of practitioners is consulted regarding all four issues. However, the role of PAG has been unclear, and a new governance structure is currently being discussed. 101 Annex N Question 7. To whom is the CGAP Consultative Group accountable for its results and impacts? (Results in rank order by program) Donors Developing Scientists/ countries professionals Most important 83% 17% 0% 20 4 0 Intermediate 17% 43% 39% 4 10 9 Least important 0% 39% 61% 0 9 14 No Answer 1 2 2 Total 24 23 23 The responses are significantly different among stakeholders (P = 0.0 percent). Question 8. From your perspective, to what extent do you believe that the program has effective design, implementation, monitoring, and independent evaluation? (Results in rank order by aspect of the program) Implementation Monitoring Independent Donor Evaluation Oversight Total High 33% 13% 8% 17% 18% 8 3 2 4 17 Substantial 46% 48% 58% 39% 48% 11 11 14 9 45 Modest 21% 39% 29% 39% 32% 5 9 7 9 30 Negligible 0% 0% 4% 4% 2% 0 0 1 1 2 No Answer 0 0 0 2 5 Total 0 2 1 23 94 The responses are not significantly different among activities (P = 40.6 percent). Written Comments · Partners may not be appointing high-level enough staff to really affect changes and decisions. CGAP performs basically on its own because of the high quality of its professional staff & Executive Director. · The Secretariat does pretty much as it wishes. · CGAP produces excellent monthly newsletters, quarterly reports and annual reports. 102 Annex N Question 9. In your opinion, how long from today would the program take to achieve its current objectives? (Results in rank order by program) PPIAF infoDev CGAP WSP CA ESMAP All 1-2 Years 10% 0% 4% 0% 0% 5% 3% 1 0 1 0 0 1 3 3-5 Years 40% 33% 22% 17% 11% 5% 19% 4 4 5 1 2 1 17 5-10 Years 30% 42% 43% 50% 47% 33% 41% 3 5 10 3 9 7 37 Greater than 20% 25% 30% 33% 42% 57% 37% 10 years 2 3 7 2 8 12 34 No Answer 2 3 2 1 1 1 10 Total 10 12 23 6 19 21 91 The responses are not significantly different among programs (P = 49.0 percent). Question 10. In your opinion, what are the particular strengths and weaknesses of CGAP? Strengths: · Technical competence. · The major strength of CGAP is its staff ­ great professionals really committed to CGAP's objectives. Another strength is its global scope and the number of donors it has been able to mobilize. · If CGAP didn't exist, someone would have to invent it to serve the hunger for a global community of interest for microfinance. Independence from a national or institutional agenda. Credibility (perception of independence and high quality of work) and widespread ownership position CGAP to challenge received wisdom without being viewed as heretical. · Information pooling. · Leadership. Increasingly responsive to members. Websites. Technical expertise. · Its location and the partners it has. · Professionalism. Donors' collective effort. World Bank's involvement (image build-up). · Academic excellence. Can easily establish contacts with important donors and financiers. · Leader in microfinance thinking, concepts, training, handbooks, pilot operations, support of role models, catalytic operations, influencing reforms, influencing and worldwide MF practitioners alike. · Can play the role of a global center of excellence in the industry. Can work with so many donors and build knowledge quickly and effectively. Can do things that individual donors cannot do. 103 Annex N · Quality of staff, power to convene and "spotlighting" issues, standard setter/benchmark for the industry. · Complementary to the activities of others, especially the IBRD. Given its size, highly flexible in establishing and reaching targets. Close ties between donors and recipients. Expertise in microfinance. · Current, as opposed to former, leadership. Professional staff. · Highly qualified, effective, and service-minded Secretariat. Good fora for donor coordination, the ongoing "donor peer review" within CGAP is excellent. · (1) The high level of commitment to the industry from its bodies and mainly the Secretariat. (2) The high level of competence in microfinance that the Secretariat has. (3) The influence that CGAP has over the donor community as a standard setter for the work in microfinance. · Improving coordination and consistency of donor programs. Providing clear performance-based objectives and the tools for implementing them. Capacity building ­ both directly in the donor agencies and establishing programs for capacity building of MFIs. · Focus on practitioners, best practice, and tools for monitoring and evaluation. · Powerful sector. Brilliant staff. · (1) Very much respected globally for the professionalism of its work. (2) Seen as fairly neutral and thus respected as a voice for the industry as a whole. · Collective consensus platform, resource center, potential voice in policy and regulatory change · Information pooling. · Knowledge, expertise, global reach, credibility with donors, respect of the microfinance community. · Very professional and focused on the ever-expanding agenda and issues related to microfinance. Its seminars and publications on emerging issues and knowledge are first class, and should be promulgated accordingly. · Good quality Secretariat. High level of commitment and support from a significant proportion of the donor community and from microfinance practitioners. Strategic location in respect to the World Bank, but with increasing recognition that CGAP is separate from the Bank. Weaknesses: · Process competence. Organizational and canceling competence. Misunderstanding Third World country reality. · CGAP is not well known. · The ownership is not clear. · Too technocratic. Washington-based. · Partners seem a bit passive in face of increased questioning of economic model, the state of the world, etc. · Low involvement of practitioners. · Too much linked to the way of thinking inside the "Washington beltway." Not sufficient input from private donors and social investors. 104 Annex N · Hasn't been able to still down some of the quackery and amateurism (generally in Northern circles) that has afflicted the microfinance scene worldwide. CGAP isn't as known as it should be. · On-the-ground involvement at service-provider operational level is limited by nature. · Must coordinate and enforce standards by consensus, difficult to "discipline" its own shareholders. · CGAP is best acting indirectly ­ as catalyst for experiments in partnership with others, as synthesizer and distributor of practical knowledge for both MFIs and donors. Yet some expectations remain that CGAP should be an implementer as well. These expectations have been managed better in recent years, but it's important that the assessment criteria for OED's current exercise not implicitly assume that CGAP's mission is to mobilize financing for, and ensure implementation of specific MFI projects. · Concentration on Latin America, which although declining is still too strong. Lack of knowledge about and attention to poverty and microfinance in Asia. Failure to mobilize equity (donated equity) for well-managed MFIs to scale-up their outreach to the poor. · Difficult to influence in-house donor practices. However, the current "microfinance donor peer review" is the first of a serious of initiatives with the aim of improving donor effectiveness. · (1) The limited human and economic resources that CGAP has. (2) The actual governance structure is very complex. (This is about to change.) · Public perception of CGAP as a big funding agency. Limited ability to participate directly in designing innovative components in projects of donor members. · Too centralized. · Donors' bureaucracy and politics constraints. · (1) Insufficient influence on member donors. (2) Too much influence from Latin American experience as opposed to Asian or African experience. · Too narrow focus on microfinance · The Secretariat often acts in isolation. · Lack of policy framework for working with private sector partners (as opposed to governments/NGOs). Working of governance bodies could be strengthened further (balance of power is too tilted toward management!) · The acceptance of best-practice microfinance, such as in our institution, is weak. CGAP should be more of an active advocate, supporter and component, especially mobilizing political support from the highest levels, to compel implementing partners such as ourselves in mainstreaming microfinance into our activities. · Relatively small secretariat staff located in Washington, D.C. (with a small outpost in Paris) means that CGAP has limited presence in the field. There is a perception, probably well founded, that CGAP caters better to the higher end of the microfinance sector than it does to the middle to lower levels (for example, in the technical tools that it has produced). To a significant extent, CGAP is only as strong as its member donors, and its donors inevitably differ in the degree to which they put into practice the policies that are agreed upon by CGAP. While CGAP has brought a range of 105 Annex N reputable practitioners into the governance of the organization, there has been in my view a tendency not to give sufficient attention to views that don't accord with the mainstream of CGAP (secretariat and donor) opinion. Question 11. In your opinion, what are the most important lessons of CGAP to date­ in one or more of the following areas: establishing objectives, selecting activities, partnerships, participation, governance, implementation, financing, and monitoring and evaluation? · CGAP is not self-critical enough of its strengths and weaknesses. This is particularly reflected in the external evaluation 2001/02. · I think CGAP has been very good at establishing (and maintaining) partnerships (either private or public). · Selecting activities. Monitoring and evaluation. · CGAP is best acting indirectly ­ as catalyst for experiments in partnership with others, as synthesizer and distributor of practical knowledge for both MFIs and donors. Yet some expectations remain that CGAP should be an implementer as well. These expectations have been managed better in recent years, but it's important that the assessment criteria for OED's current exercise not implicitly assume that CGAP's mission is to mobilize financing for and ensure implementation of specific MFI projects. · Governance: Unclear and not in line with a donor consortium and insufficient accountability of Secretariat to member donors. Role of World Bank too important. · Small multilaterals focused on single issues can be very effective. · (1) Placing microfinance as a serious issue on the world development agenda. (2) Documents produced by CGAP are being used in English-speaking countries and state of the art material is disseminated. (3) It may be trying to do too much in too many areas, without insisting on the most important. · CGAP's most important contribution has been in establishing objectives, selecting activities, and monitoring and evaluation for MFIs and the industry as a whole. · One big lesson: The Bank is at its best at creating global initiatives such as CGAP. Don't stifle this with excessive layers of M&E, soul-searching, frameworks, etc. Another: The Bank is great at creating and animating such donors' clubs. A final one: Exacting ex ante "exit strategies" etc. from such initiatives makes little sense beyond the political correctness thereof. (It's about as wise as asking two people getting married to spell out the desired horizon for their marriage and to detail their exit strategies.) · Working with donors can produce substantial positive results. Partnerships are a much better way to achieve results. · CGAP's strategy for leveraging its impact (through the CG, services to donors, standard setting, capacity building, etc.) is a model for all development assistance. 106 Annex N · Its strategy and resources have been far too modest for its vision and objectives. Its geographical outreach has been skewed to Latin America, and to a lesser extent Africa. · The most important lesson is that we have realized the internal agency-specific constraints to implementing good donor practices in microfinance, and that we are currently acting upon them (donor peer review process). Furthermore, the link of this process to overall aid effectiveness will provide valuable input into the donor community as such. · (1) Governance structure must help the day-to-day work with appropriate approval and commitment. (2) The activities that CGAP as a whole body must do are demand-driven and never supply-driven. All the programs must be useful to the CGAP clients. (3) Participation of the whole body (CG, PAG, members) is very important so that the objectives can be achieved. (4) The commitment that the donor members must have is very important because CGAP by its own is too limited to have direct actions into the field. One of the main issues is this. (5) Member donors must be active economically and practically so that CGAP can accomplish its mission. (6) Partnerships have been very important. Some new services and products have been made with this kind of arrangement. · Grant assistance to develop the industry should be provided on a performance- based approach with monitorable indicators of progress toward sustainability and outreach objectives. · Giving visibility to the sector and developing tools. · (1) CGAP has demonstrated a much higher level of professionalism for funding programs. This provides important lessons for donors in the field of microfinance. (2) CGAP has led the microfinance field in terms of agreeing on standards, promoting transparency of performance and creating tools for this transparency. This is important in advancing operational practice and in making it possible for private investors and social investors to consider investing in microfinance programs. (3) The recent CGAP work of donor peer reviews should yield very important improvements in donor activities and effectiveness. This may be a first in terms of donor practice in any field in which donors are active and thus may be particularly important towards improving donor practice in more fields than just microfinance. · It pays to (1) have the right partners, (2) be in a high impact sector, (3) add value to clients and ensure buy-in, and last but not least, (4) have skilled management. · Persistent and consistent effort and application. · The importance of establishing a secretariat of knowledgeable and committed people working on a subject on a full-time basis as a way of building credibility among a wide group of stakeholders. The importance of working on a number of fronts (donors, broad industry issues, technical support for practitioners), but of clearly differentiating between the different stakeholder groups in the interests of clarity of purpose and ease of monitoring and evaluation. The importance of continuity in donor representation as a means of enhancing the extent to which 107 Annex N member donors can hold the secretariat accountable and increase their sense of ownership. Question 12. In your opinion, what comparative advantage does the World Bank bring to CGAP relative to the other partners in the program? (Results in rank order by comparative advantage) Global Convening Mobilizing Analytical Multi-sector Country-level mandate power financial resources expertise capacity knowledge High 64% 52% 24% 12% 8% 4% 16 13 6 3 2 1 Substantial 24% 36% 48% 48% 48% 40% 6 9 12 12 12 10 Modest 12% 12% 24% 32% 36% 48% 3 3 6 8 9 12 Negligible 0% 0% 4% 8% 8% 8% 0 0 1 2 2 2 No Answer 0 0 0 0 0 0 Total 25 25 25 25 25 25 The responses are significantly different among comparative advantages (P = 0.01 percent). Question 13. Is the World Bank's involvement critical for the success of the program? (Results in rank order by program) ESMAP PPIAF infoDev CA CGAP WSP All Yes 100% 100% 100% 83% 79% 57% 88% 20 10 12 15 19 4 80 No 0% 0% 0% 17% 21% 43% 12% 0 0 0 3 5 3 11 No Answer 2 2 3 2 1 0 10 Total 20 10 12 18 24 7 91 The responses are significantly different among programs (P = 1.50 percent). Written Comments · I think the World Bank gives CGAP an additional credibility and eases the whole process of mobilizing funds for microfinance purposes. The Bank also gives a valuable logistics support. · It should be an independent "Consultative Group" and the World Bank should have the same rights as the other members. 108 Annex N · CGAP is too small to have fully independent institutional standing. If not housed within the World Bank, it would require larger and more elaborate institutional structure. The Bank's periodic renewal of medium-term financial support ensures CGAP of a continued level of resources, making CGAP less vulnerable to year-to- year changes in the budgets of donor agencies. · Financial only, not technical at all! · To start CGAP yes, but World Bank presence heavily conditions how the rest will act. · So far yes, because of financial support. Hopefully not so in the future. · No Bank, no CGAP. · Legitimacy. · Because the World Bank has the capacity to mobilize resources and to catalyze. · It is unlikely that any other organization could lead such a broad coalition. · Besides the financial resources the IBRD devotes to CGAP, the overarching multi- sectoral expertise and the global outreach seem to me crucial for CGAP's success. · Not so far. As a member of PAG, I get the feeling that normally the two organizations pay little attention to each other, but I may be wrong. · Not really, CGAP knows more about microfinance than the Bank. However, CGAP is critical for the Bank's success in microfinance. · I think it is very important because no other donor has the capacity to be heard and to influence as the World Bank does. So this has been very powerful for the whole body of CGAP. · Convening power, ability to influence policy environment for microfinance. · Because of global reach. · CGAP is perceived by the other members as the Bank's baby. · Agenda setter. · Financing the operations of CGAP plus giving it its organizational caché and prestige. · The Bank has played a crucial role, not least in providing a major part of CGAP's funding since its inception. This financial role has declined over time but remains important. The participation of the Bank also lends weight to CGAP, given the authority that the Bank exercises in development circles, especially among bilateral agencies. The Bank's participation also means that CGAP has a direct means of influencing the programming of the major actor in development finance. CGAP could now survive without the Bank's involvement, although it would be diminished if the Bank were to withdraw. 109 Annex N Question 14. Are there any drawbacks to the Bank's involvement in the program? (Results in rank order by program) WSP infoDev CA ESMAP PPIAF CGAP All Yes 86% 82% 61% 53% 44% 40% 56% 6 9 11 9 4 10 49 No 14% 18% 39% 47% 56% 60% 44% 1 2 7 8 5 15 38 No Answer 0 4 2 5 3 0 14 Total 7 11 18 17 9 25 87 The responses are not significantly different among programs (P = 11.6 percent). Written Comments · CGAP started as a World Bank-driven initiative which allowed the Bank to catch up on microfinance. Now that his has happened, there is no need for a special place and host function any longer. · CGAP is not sensitive enough to realities outside the U.S. and perhaps Latin America. · Maybe the World Bank takes too much "place" in CGAP. The idea one ends up getting is that the Bank is the most important donor in CGAP and that therefore, CGAP is the "World Bank's creature." · You could link any decision to the World Bank interest and not to the donors' needs. · Day-to-day independence from Bank operations is critical. Otherwise CGAP staff could spend full-time "cooperating" and "collaborating" with the large number of Bank sectoral and regional operations that have a "finance for the poor" dimension. Similarly CGAP should not be expected to draw substantive support for its operations from Bank staff ­ beware of false "efficiencies" that assume CGAP is more effective if it can use the Bank's comparative advantages listed in question 12 above. · Yes and no. It depends on how deep the involvement is. · It is perceived to be a part of the World Bank, rather than a 29-member organization. · CGAP has become one more instrument of World Bank, IMF, etc. Their economic policies are being seriously questioned by civil society everywhere (more importantly in the member countries themselves, finally!) and CGAP acts as if this was not happening at all! · The image of CGAP in developing countries is too much World Bank-linked, dominated, and motivated. · The Bank is conflicted in its own microfinance operations ­ which are sometimes less than world class. · It is really important that the Bank continues to be member of the CG. In the eyes of some, CGAP is seen as a child of the Bank, with no capacity for independent action. For those who are critical of the Bank (e.g. some among the NGO community), this has encouraged a wariness in respect of CGAP and a reluctance to listen carefully to the messages that CGAP is putting out and to assess them on their own terms. However, I would not rank this drawback as particularly significant.