38291 Economics and Governance of Nongovernmental Organizations in Bangladesh Bangladesh Development Series Paper No: 11 The World Bank Office, Dhaka April 16, 2006 www.worldbank.org.bd/bds Document of the World Bank Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not 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 any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Copyright Statement: The material in this publication is copyrighted. The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. ISBN: Design: Cover designed by Progressive Printers. Published by Creative Ideas. CONTENTS Executive Summary………………………………………………………………………………………… ..............i Chapter 1: Introduction..............................................................................................................................................1 1.1 Characteristics and evolution of the Bangladesh NGO sector..........................................................................2 1.1.1 The role of government in creating a supportive environment .................................................................4 1.2 Current debates.................................................................................................................................................5 1.3 Assessing service provision for the poor in Bangladesh: An analytical framework.........................................7 1.3.1 The characteristics of services ..................................................................................................................8 1.3.2 Institutional actors and relationships of accountability.............................................................................8 1.3.3 The political context of public and private service provision .................................................................10 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance .........................................13 2.1 Defining and classifying NGOs in Bangladesh ..............................................................................................13 2.2 The range and emphasis of NGO activities ....................................................................................................15 2.3 Whom do NGOs reach and where do they operate?.......................................................................................16 2.4 A closer look at three services ........................................................................................................................18 2.4.1 Microcredit: Facts and issues..................................................................................................................18 2.4.2 The impact of microcredit in Bangladesh ...............................................................................................23 2.4.3 Health: Facts and issues ..........................................................................................................................25 2.4.4 The impact of NGO health programs......................................................................................................27 2.4.5 Education: Facts and issues ....................................................................................................................29 2.4.6 The impact of NGO education programs ................................................................................................32 2.5 NGOs and advocacy issues.............................................................................................................................33 2.6 Cross-cutting themes ......................................................................................................................................38 Chapter 3: Financing NGOs: Trends and Prospects..............................................................................................41 3.1 External financing ..........................................................................................................................................41 3.1.1 Issues with direct donor-NGO financing ................................................................................................43 3.1.2 Issues with government contracting of NGO services ............................................................................44 3.1.3 Issues with private charitable donations .................................................................................................47 3.2 Internally generated funds ..............................................................................................................................49 3.2.1 Microfinance interest income..................................................................................................................49 3.2.2 NGO-managed commercial enterprises ..................................................................................................52 Chapter 4: The Regulatory Framework and Financial Accountability Issues ....................................................57 4.1 An overview of the laws affecting NGOs.......................................................................................................57 4.2 Implementing agencies ...................................................................................................................................61 4.4 Improving the institutional framework for NGO regulation...........................................................................65 4.5 Improving the laws relating to NGOs.............................................................................................................67 4.6 Improving corporate governance and financial accountability practices within NGOs .................................70 Chapter 5: Scaling Up through a Strategic Compact.............................................................................................73 5.1 Does further NGO growth make sense from a public policy standpoint? ......................................................73 5.2 Is further NGO growth possible?....................................................................................................................78 5.2.1 Management, incentives, and systems ....................................................................................................78 5.3 Managing a larger and more focused NGO program......................................................................................82 Bibliography ..............................................................................................................................................................99 LIST OF BOXES Box 1.1: Eight sizes fit all: Lessons from WDR 2004 ................................................................................. 7 Box 1.2: Market failures and the case of intervention ................................................................................. 9 Box 2.1: World Bank Survey of NGOs in Bangladesh.............................................................................. 14 Box 2.2: Tackling poverty in remote regions: The Integrated Development Foundation.......................... 19 Box 2.3: Combining financial services with social protection: The IGVGD program .............................. 22 Box 2.4: High interest rates but still a popular microlender: BURO, Tangail ........................................... 23 Box 2.5: The Gonoshasthaya Kendra story: Cost-effective health care for the poor ................................. 29 Box 2.6: BRAC’s Non-formal Primary Education Program...................................................................... 32 Box 2.7: A selection of NGOs involved in advocacy in Bangladesh......................................................... 34 Box 2.8: The controversy over Proshika.................................................................................................... 36 Box 2.9: What affects the community perception of NGO services? ........................................................ 39 Box 3.1: Using an autonomous foundation to contract NGOs: Lessons from PKSF................................. 46 Box 3.2: Anjuman Mufidul Islam: A Welfare-oriented “Islamic NGO” ................................................... 48 Box 3.3: Linking NGOs to commercial bank financing ............................................................................51 Box 3.4: Moving out of poverty using microcredit and a mobile phone.................................................... 53 Box 3.5: NGO-linked commercial ventures in other countries.................................................................. 54 Box 3.6: A typology of NGO businesses and their legal status ................................................................. 55 Box 4.1: Suspension and rebirth of a large NGO: The case of Gono Shahajjo Sangstha .......................... 59 Box 4.2: Microfinance Transparency Award............................................................................................. 66 Box 4.3: NGO certification in Pakistan ..................................................................................................... 67 Box 5.1: The transition from NGO to community management................................................................ 76 Box 5.2: Managing NGO health professionals: Are there lessons for the public sector? .......................... 81 LIST OF TABLES Table 2.1: Focus Group Rankings of Community Priorities (May 2003) .................................................... 16 Table 2.2: The Scale of Non-Government Microfinance Activity ............................................................... 18 Table 2.3: Health Spending by NGOs compared to other sources ............................................................... 25 Table 2.4: Expenditures on Health, by type of facility and income quintile ................................................ 26 Table 2.5: Composition of Education Expenditure, By Types of School..................................................... 30 Table 2.6 Literacy Rate by Type of School................................................................................................. 33 Table 2.7 Advocacy and awareness raising: RDRS’s progress in 2004 ...................................................... 38 Table 3.1: Aid Flows to NGOs..................................................................................................................... 41 Table 3.2: Profitability of NGO microfinance institutions, 2001-4.............................................................. 50 Annex Table 2.1: Microfinance as a Share of Total Private Sector Credit........................................................... 88 Annex Table 2.2: NGO Education Programs ....................................................................................................... 88 Annex Table 2.3: Donor Financing of Education Programs of Big NGOs (Million Taka).................................. 89 Annex Table 3.1: Estimated IDA Disbursed through NGOs (1995-2004)........................................................... 90 Annex Table 3.2: Donor Financing Mechanisms ................................................................................................. 91 Annex Table 3.3: Micro-finance Efficiency Indicators ........................................................................................ 92 Annex Table 3.4: Outreach and Average Loan Size by NGO size....................................................................... 93 Annex Table 3.5: An Illustrative Range of NGO Businesses............................................................................... 94 Annex Table 3.6: Average Profitability of Four NGOs with commercial activities 2001-04 .............................. 95 LIST OF FIGURES Figure 2.1: Main Services Provided by NGOs .............................................................................................. 15 Figure 2.2: Communities with at least one NGO Program, 1995-2000 ......................................................... 17 Figure 2.3: Average Number of NGO Programs per Community by Region, 1995-2000............................. 17 Figure 2.4: Coverage of Poor Households by Microfinance Programs, by district ....................................... 20 Figure 2.5: Distribution of NGO Expenditures by Health Program .............................................................. 27 Figure 2.6: Distribution of Primary Students by Facility and Quintile .......................................................... 30 Annex Figure 1.1 NGO staff perceptions of relations with Government counterparts ........................................ 87 Annex Figure 2.1: How NGOs Assess Community Needs .................................................................................. 87 CURRENCY EQUIVALENTS Taka (Tk) US $1.00 = Tk 71.02 (April 5, 2006) GOVERNMENT'S FISCAL YEAR July 1–June 30 ACRONYMS AND ABBREVIATIONS ADAB Association of Development Agencies in IDA International Development Association (of Bangladesh the World Bank) ASA Association for Social Advancement IDCOL Infrastructure Development Company ASIRP Agricultural Services Innovation and Limited Reform Project IDF Integrated Development Foundation ASK Ain o Shalish Kendra IGVGD Income Generation for Vulnerable Group BLAST Bangladesh Legal Aid and Services Trust Development BNP Bangladesh Nationalist Party INGO international nongovernmental organization CAMPE Campaign for Popular Education JC Jagoroni Chakra CCDB Christian Commission for Development in MFI microfinance institution Bangladesh MRRU Microfinance Research and Reference Unit CDF Credit and Development Forum NFPE Non Formal Primary Education CGAP Consultative Group to Assist the Poor NGO nongovernmental organization CIDA Canadian International Development NGOAB NGO Affairs Bureau Agency NHA National Health Accounts DAM Dhaka Ahsania Mission ORT oral rehydration therapy DFID Department for International Development PBO public benefit organization (UK) PCP Pakistan Centre for Philanthropy DOTS Directly Observed Treatment, Short-Course PDBF Palli Daridro Bimachan Foundation DSW Department of Social Welfare PKSF Palli Karma-Sahayak Foundation EC European Commission PRSP Poverty Reduction Strategy Paper FEMA Fair Election Monitoring Alliance RDRS Rangpur Dinajpur Rural Service FIVDB Friends in Village Development Bangladesh RJSC Registrar of Joint Stock Companies FNB Federation of NGOs in Bangladesh SDC Swiss Development Corporation GDP Gross Domestic Product TB tuberculosis GK Gonoshasthaya Kendra thana subdistrict (also known as upazila) GKT Gano Kallayan Trust TIB Transparency International Bangladesh GSS Gono Shahajjo Sangstha TMSS Thengamara Mohila Sabuj Sangha HEED Health Education and Economic UCEP Underprivileged Children’s Educational Development Program HIV/AIDS human immunodeficiency virus/acquired UPHCP Urban Primary Health Care Project immune deficiency syndrome USAID US Agency for International Development ICAB Institute of Chartered Accountants of WDR World Development Report Bangladesh Vice President : Praful Patel, SARVP Country Director : Christine I. Wallich, SACBD Sector Director : Sadiq Ahmed, SASPR Sector Manager : Ijaz Nabi, SASPR Task Leader : Hassan Zaman, SASPR Acknowledgments This report was prepared by the following team: Hassan Zaman (task manager and lead author), Nagavalli Annamalai, Irajen Appasamy, Stephen Rasmussen, Suraiya Zannath (World Bank); Frank Matsaert (DFID); Dewan Alamgir, Mirza Hasan, Naomi Hossain, Safi Khan (consultants). The team is grateful for the guidance and support provided by Christine Wallich, Sadiq Ahmed and Ijaz Nabi. Professor Wahiddudin Mahmud (Dhaka University) and Jeffrey Hammer (World Bank) were the peer reviewers and the report benefits from their insights. Valuable inputs and suggestions were received from Nilufar Ahmad, Shamsuddin Ahmad, Junaid Ahmad, Anthony Bottrill, Rafael Cortez, Amit Dar, Maitreyi Das, Shantayanan Deverajan, Raihan Elahi, Julia Galef, Varun Gauri, Mohammad Hossain, Swarna Kazi, Bazlul Khondker, Qaiser Khan, Sarwar Lateef, Syed Mahmood, Aneeka Rahman, Nermeen Rouf and Robert Saum. The document was formatted by Joyce Das. The report was prepared in consultation with the Government of Bangladesh, the NGO community, donor representatives and several academics. Extensive discussions were held with a range of stakeholders prior to finalizing the report. The report team is deeply grateful to those who provided written comments on the draft report, to the institutions that hosted consultations and to the many individuals who kindly participated in the discussions. Foreword Bangladesh has made striking progress on a range of social indicators over the last 15 years, an achievement widely credited to the country’s pluralist service provision regime. Nongovernmental organizations (NGOs) have significantly expanded their services during this period and have shown that it is possible to scale up innovative antipoverty experiments into nationwide programs. The unique role of Bangladesh’s NGOs is not confined to the delivery of social services and pro-poor advocacy. NGOs have developed commercial ventures in order to link poor producers with input and output markets, as well as to develop a source of internally generated revenue for the organizations. As we look forward, the 2005 Poverty Reduction Strategy Paper (PRSP) views the NGOs’ role as an integral part of efforts to achieve national poverty reduction targets, particularly by delivering and facilitating pro-poor services. At the same time, the rapid growth and diversification of the NGO sector has also given rise to public policy challenges and questions. These include the viability of a regulatory framework developed when the size and scope of NGOs was far more limited, the appropriate political and commercial spaces for NGO activities, trade-offs between NGO sustainability and pro-poor orientation, the impact and quality of NGO services as they have scaled up, NGO corporate governance, and the implications of different ways of financing NGOs. There has been little systematic review of the public policy implications of the changing character of NGOs in Bangladesh. The present report seeks to augment this effort, by addressing these debates using a public economics framework, and is the product of the teams’ own work and extensive consultations with a wide range of stakeholders on an earlier draft. We hope that it will contribute to a better understanding of the NGO sector in Bangladesh. Sadiq Ahmed Christine Wallich Sector Director Country Director Poverty Reduction and Economic Management World Bank Office, Dhaka Executive Summary Bangladesh has made striking progress on a range of social indicators over the last 15 years, an achievement widely credited to the country’s pluralist service provision regime. Nongovernmental organizations (NGOs) have significantly expanded their services during this period and have shown that it is possible to scale up innovative antipoverty experiments into nationwide programs. Notable innovations that were expanded include delivering credit to the previously “unbankable” poor, developing a non-formal education program to cater to poor children, particularly girls, and using thousands of village-based community health workers to provide doorstep services. The fact that poor women constitute a large proportion of NGO beneficiaries, despite the persistence of strong patriarchal norms, also testifies to institutional innovation. The unique role of Bangladesh’s NGOs is not confined to the delivery of social services and pro-poor advocacy. NGOs have developed commercial ventures in order to link poor producers with input and output markets, as well as to develop a source of internally generated revenue for the organizations. As we look forward, the 2005 Poverty Reduction Strategy Paper (PRSP) views the NGOs’ role as an integral part of efforts to achieve national poverty reduction targets, particularly by delivering and facilitating pro-poor services. However, the rapid growth and diversification of the NGO sector has also given rise to questions and concerns. These include the viability of a regulatory framework developed when the size and scope of NGOs was far more limited, the appropriate political and commercial spaces for NGO activities, trade-offs between NGO sustainability and pro-poor orientation, the impact and quality of NGO services as they have scaled up, NGO corporate governance, and the implications of different government-NGO partnerships. There has been little systematic review of the public policy implications of the changing character of NGOs in Bangladesh. The present report seeks to augment this effort. The first part of the report presents the Bangladesh context, the current debates surrounding NGOs, and an analytical framework within which to examine these issues. We then turn to the questions of what NGOs do, whom they cater to, how their programs and expenditures differ from those of other providers, and what impact their programs have had on individual, household, and community welfare. Although NGOs provide a diverse range of services, we have chosen to limit our analysis to the three key areas of microcredit, education, and health, as well as to the advocacy activities undertaken by some NGOs. The report next takes a close look at issues related to the financing of NGO activities through donor support, government contracts, private contributions, microfinance income, and commercial activities. The basic questions here concern the relative importance of these various sources of income, changing trends in financing, and the implications of these trends for the nature of NGO activity in the medium term. We then assess the status of the legal and regulatory framework, relating it to the scope of activities and financing trends discussed earlier. As part of this, the report assesses the state of financial accountability and corporate governance in the NGO sector. The main issues include the extent to which current legislation needs updating, the need for reform of institutional structures overseeing this legislation, and the extent to which NGOs are following best-practice norms in internal governance and accountability. The final Economics and Governance of Nongovernmental Organizations in Bangladesh chapter builds on the analysis to focus on the questions of whether NGOs should expand or diversify their programs, whether they have the capacity to do so, and what will be required from all stakeholders to ensure that any change in direction maximizes benefits to the poor. A framework for analyzing NGO activity The Bangladesh NGO sector and individual organizations within it stand out by virtue of their scale. There are an estimated 2,000 development NGOs in Bangladesh, and a small group of them are among the largest such organizations in the world. These big NGOs—BRAC, ASA, and Proshika—have nationwide programs, with tens of thousands of employees and multimillion-dollar budgets. Most NGOs in Bangladesh are small, however, and have limited managerial and staff capacity. For instance, in a sample of 720 NGOs, 90 percent had programs in less than five (out of 64) districts. An assessment of NGOs requires a framework for analyzing how these organizations can contribute to improving services to the poor. We use a framework built around: (a) the characteristics of services provided by NGOs, (b) the accountability relationship between policy makers, providers, and service users, and (c) the overall national context for pro-poor service delivery by both the public and private sectors. 1 We assess the characteristics of NGO services using public economics tools. To the extent possible, we assess the rationale for NGO provision on equity and effectiveness grounds, using information on expenditure patterns, success in targeting the poor, and indicators of impact. Relations of accountability between the main actors—poor users of services, policy makers, and service providers—ultimately determine the quality of services. In theory, service users can influence public service quality through their political power over policy makers, and NGOs can help by amplifying the voices of poor users through advocacy activities. However, when donors fund NGOs to provide services directly, this may weaken mechanisms of accountability between policy makers and providers. The merits and drawbacks of different donor financing methods are assessed in this light. The increasing prominence of NGO services reflects NGOs’ comparative advantage in bringing services directly to poor users. However, it also highlights the need for a closer look at the conditions for accountability by NGOs to beneficiaries, as we discuss later. The overall national context for pro-poor service provision is mixed. There are many impressive examples of successful service delivery outcomes in Bangladesh. As discussed earlier, they have resulted from the pluralist service provision system comprising government, NGOs, and the private sector. At the same time, there are certain clear areas for improvement in both the private and public sectors. The private sector in services such as education, health, and banking has rudimentary corporate governance arrangements, is overseen by an overstretched regulatory framework, and in many cases is enmeshed with the politics of the country. Moreover, the impact of increasingly progressive public spending allocations in Bangladesh has been diluted by weaknesses in public sector service delivery. 1 These themes are adapted from World Development Report 2004: Making Services Work for Poor People (World Bank 2003d). ii Executive Summary The analysis of NGOs’ activities, their corporate governance, and their financing arrangements needs to be considered in light of this more general framework of service provision in Bangladesh as a relative benchmark of NGO activities. At the same time, there is often too little emphasis on the importance of the space made for NGO activity by successive governments of Bangladesh. In comparative perspective, certainly within the South Asia region, public policy toward NGOs in Bangladesh has been unusually successful in balancing the need for official oversight with the operational autonomy necessary for NGOs to operate and innovate. However, the recent souring of the relationship between government and a few NGOs suggests that the political space for NGO service provision and advocacy activities cannot be taken for granted. Key activities and their impact: An assessment of NGO performance NGOs in Bangladesh provide a strikingly homogeneous set of services, with credit dominating. A survey of 300 NGO branches carried out by the World Bank in 2003 showed that while the total range of NGO interventions is wide, the typical NGO branch provides only three or four services. 2 About 90 percent of all NGO branch offices provide credit services, followed by health (56 percent), sanitation (52 percent), and education (45 percent). 3 A parallel community survey conducted as part of the 2003 NGO survey shows that the service delivery priorities identified by communities closely match the services that NGOs provide. Public awareness and advocacy are also common areas of NGO work: 93 percent of NGO branches reported awareness-raising activities, usually relating to sanitation, health, and social issues, while 42 percent reported having lobbied local or national government during the previous year. We now turn to the assessment of the three key services of microcredit, health/sanitation, and education, as well as advocacy activities. Microcredit Distribution of m icro-credit loan portfolio (2005) One of the main reasons for the growing (percent) presence of NGO programs is the expansion in Other NGO MFIs 12% microfinance. Microcredit now reaches as many as Proshika 7% Grameen 43 percent of all Bangladeshi households and about Bank 35% 70 percent of poor households. The sector is ASA dominated by the Grameen Bank, BRAC, ASA, 22% and Proshika, which between them lend to almost BRAC 87 percent of all borrowers from nongovernment 24% microfinance institutions. 4 2 Aside from microcredit, education, and health, NGOs provide many other services, often in partnership with central or local government. These include safety net programs, agricultural extension, social forestry, disaster management, skills training, and so on. 3 These ratios are expressed as percentages of NGO branches compiled by the World Bank in a nationally representative survey in 2003. Since the branches of the large organizations all provide microcredit, over 92 percent of branches surveyed provide microcredit. 4 The Grameen Bank is registered as a specialized bank and is not an NGO. However, its credit delivery model has been widely replicated by the NGO sector in Bangladesh, and for this reason Grameen’s experiences are often included in a discussion of NGOs. iii Economics and Governance of Nongovernmental Organizations in Bangladesh Microcredit generally succeeds in reaching the poor, though there are several geographical pockets of poverty in the country where microcredit coverage is relatively low. Contrary to popular perception, a significant number of the poorest Bangladeshis are microcredit group members, though the scaling up of successful experiments targeting financial services to the extreme poor will lead to even greater participation. The impact of microcredit on smoothing incomes and reducing household vulnerability to seasonal and other shocks is of critical importance to the rural poor. Improvements in key social indicators of well-being are also associated with microcredit borrowing, most notably measures of female empowerment, children’s schooling, and health status. These social gains in part reflect the complementary social mobilization, training, and awareness-raising activities that typically go hand in hand with microcredit. While microcredit has brought benefits to borrowing households, these have not been large enough to have had a significant impact on community- level employment creation and growth. The strong emphasis on financial sustainability, vital to the sector’s success, has led to controversy about purportedly high interest rates. The comparative under regulation of microfinance also poses clear risks. We discuss both these issues in greater detail below. Health /sanitation NGOs use village-based community health workers to provide door-to-door health services, focusing mainly on preventative care and simple curative care for women and children. While a nationwide network of these NGO para-professionals is successfully extending health care to large numbers of poor households, NGO facility-based care is relatively sparse. Hence, while expenditures by NGOs on health have grown significantly since the mid-1990s, they constitute only about one-third of public sector expenditures and less than 10 percent of total expenditures on health (the latter include household spending on private care). NGOs also contribute to health outcomes by providing water and sanitation services, with notable successes in community-based programs promoting behavioral change. Achievements in health include programs on child nutrition and tuberculosis treatment in partnership with government. The impact of NGO interventions on a range of health and nutritional indicators is striking. Cure rates averaged 85 percent in the tuberculosis program. Malnutrition rates dropped by about 20 percent among the poor due to the presence of NGOs in the community, after controlling for other factors. Neonatal mortality has been found to be significantly lower among NGO clients than in a control group of households. Education About 1.5 million children—approximately 8 percent of primary enrollment—are in schools run by NGOs, most in non-formal primary schools for which the NGO sector is best known. The NGO education sector is highly skewed, with one large organization, BRAC, receiving about three-fourths of donor resources and accounting for a similar share of primary enrollment in NGO schools. BRAC also franchises its model by subcontracting 200 small NGOs to deliver non-formal education programs. iv Executive Summary Incidence analysis comparing Distribution of primary students by facility and quintile different providers of primary schooling show that NGO education 120 Fifth quintile % of primary students 1 programs are effectively targeted to the 100 18 28 15 16 Fourth quintile 80 poor, and to poor girls in particular. 60 25 25 14 27 49 Third quintile 27 NGO schools have a positive impact on 40 23 20 23 19 Second 20 14 12 quintile school enrollment, particularly of girls, 20 14 17 10 44 19 10 10 First quintile 0 and record higher attendance and Madrassa Government subsidized operated Private and NGO- nment- Others Gover completion rates than formal schools. Educational achievements tell a more mixed story: NGO school students perform considerably better than their counterparts in government schools on reading and writing skills, but only slightly better on other basic competencies. A key concern is that the coordination between government and NGO education programs is weak. There is little official recognition of NGO education programs, which continue to depend largely on external grants. One obstacle in increasing NGO-government collaboration in education is that contracting arrangements have had a long history of being problematic, as discussed below. The recent diversification into pre-primary education by the market leader, BRAC, could lead other NGOs to follow suit. This diversification is being actively encouraged by the government and presents an opportunity for improved government- NGO collaboration in future. Advocacy The advocacy activities of NGOs have become somewhat controversial. Most NGO advocacy focuses on issues affecting the poor (e.g., violence against women, dowry, land rights, access to justice, housing, education) and is seen as fully legitimate. However, in 2000–1 the government accused a few NGOs of stretching their advocacy work into partisan political activity and electioneering, and funding for their service delivery programs was sequestered as a result. As government funding for NGO services grows in importance, NGOs have become more reluctant to antagonize the government. Hence, the more prominent advocacy-oriented NGOs tend to be involved only slightly in direct service delivery activities, while large NGOs with many activities and services tend to avoid issues that could seriously antagonize the authorities. Nevertheless, even without taking on contentious issues such as human rights and electoral reform, multi-activity NGOs engage in a range of low-key advocacy activities that are of significant benefit to the poor (for example, campaigns to reduce violence against women and to promote poor people’s access to resources). Advocacy fulfils an essential function in a democracy such as Bangladesh, and must be given space. Therefore, NGOs’ scope to do advocacy should be entirely unconstrained, except for activities that promote one political party, or election candidate, over another. Returning to the analytical framework, we find that the role of NGOs in strengthening relations of accountability between policy makers and poor service users has been mixed, as NGO advocacy efforts are to an extent constrained by their dependence on government. NGOs have had more v Economics and Governance of Nongovernmental Organizations in Bangladesh success in bridging the gap between service users and providers in Bangladesh through their own services and by facilitating government, community, and private sector provision. Financing of NGOs: Trends and prospects The share of aid going to Aid to NGOs NGOs has risen sharply. We 2,000 30.0 estimate that aid to NGOs rose Total aid to Government from an annual average of $233 25.0 1,500 million (0.7 percent of GDP) in Share of aid 20.0 US$ million 1990–95 to $343 million (0.7 to NGOs % 1,000 15.0 percent of GDP) in 1996–2005. Meanwhile, total aid to 500 Aid as share of Total aid to NGOs 10.0 GDP Bangladesh fell from an annual 5.0 average of $1.62 billion (4.9 0 0.0 percent of GDP) to $1.39 billion 0 20 2 3 19 1 19 92 19 3 19 94 19 5 19 96 19 97 19 8 99 99 20 1 00 -0 -0 -9 -9 -9 -0 -9 - - - - - (2.9 percent of GDP) during those 01 02 90 91 92 93 97 94 95 96 98 00 -2 20 19 19 periods. As a result, the share of aid to NGOs in total aid to Bangladesh has risen from 14.4 percent in the first half of the 1990s to 24.6 percent since then. These figures include our estimates of aid to NGOs from multilateral development agency loans, typically through Annual Development Programme projects that contract NGOs. Our estimates suggest that on average 20 percent of the aid that is provided to NGOs comes through government from lending agencies, an amount averaging about $82 million a year. There are a number of ways in which donors provide direct financing to NGOs in Bangladesh. The most common is funding for specific projects. Where financing needs are large, donor funds may be pooled and a donor-liaison function introduced to coordinate support and reduce transaction costs for the NGO. In some instances donors have financed the whole range of NGO activities, and in a few of these cases the institution develops into a different legal entity by the end of the funding period. In recent years, the recognition that partnerships with smaller NGOs carry high transaction costs has increased reliance on wholesale vehicles, in which an agency manages a pool of money for which smaller NGOs compete. Financing of NGOs through the government budget improves the comprehensiveness of the fiscal framework and the quality of public policy choices. However, the manner by which this service delivery partnership is managed matters a lot, as government-NGO contracting has not escaped the governance problems associated with public procurement in general: absence of transparency, bribes associated with contract awards, delays and rent seeking, and emergence of fake, fly-by-night NGOs. The successful contracting practices of the microcredit apex body, Palli Karma-Sahayak Foundation (PKSF), as well as the Infrastructure Development Company Limited (IDCOL)—both agencies with substantial professional capacity and autonomy—should be mainstreamed throughout government. These practices include strict eligibility criteria, transparent guidelines for NGO selection, and monitoring mechanisms to assess performance. Successful contracting of NGOs typically also includes competitive bidding with transparent, time-bound application and approval processes. The newly established NGO Foundation should vi Executive Summary adopt and adhere to such best practices. A system for certifying NGOs would help to improve partnerships and contracting arrangements, as discussed below. Approximately 70 percent of private charitable contributions in Bangladesh go to religious institutions, with educational establishments (15 percent) a distant second. Most development NGOs do not feature prominently among recipients of these donations. However the new provision for tax deductions for charitable contributions introduced in the 2005 budget may increase resources going to development NGOs. In order to reduce the burden on the National Board of Revenue, the certification process for tax deductions could be devolved to an independent professional body. Lessons may be drawn from Pakistan and the Philippines: in both countries the Ministry of Finance has devolved certification for tax purposes to an independent body, the Pakistan Centre for Philanthropy and the Philippine Council for NGO Certification, respectively. The internally generated funds of NGOs mainly consist of their net income from microcredit operations. Several NGOs also generate surpluses from commercial activities, as described below. We find that, by and large, microfinance operations in Bangladesh are cost- efficient relative to international comparators. Analysis of NGOs of varying sizes suggests that on the whole most microfinance programs are financially sustainable. ASA stands out as the industry’s most profitable microcredit operation, largely because of its cost-efficiency but also because of its higher average loan sizes. Profits are used to further expand microcredit and to subsidize access by the poorest. Projections show that the microcredit market is not saturated and that interest income is likely to grow. However, careful attention will need to be paid to portfolio quality as competition for borrowers intensifies. Growing financial sustainability has meant declining dependence on donor funding, and the bulk of revolving loan funds are now from client savings and microcredit surpluses. PKSF is an important source of finance, but NGOs are increasingly exploring ways to reduce their dependence on this apex body now that PKSF has directed that any NGO it lends to should charge no more than 12.5 percent interest. NGOs’ access to commercial funds is limited at present. Independent credit ratings of microfinance institutions, disclosure and benchmarking of performance indicators, and a regulatory framework that allows equity investments in microfinance programs will lead to greater financing by the formal financial sector. Several NGOs have established commercial enterprises whose profit stream is earmarked to support their development programs. This “endowment model” is designed to reduce dependence on donors and other outside funding sources, and is an innovative solution also used by foundations and charities elsewhere. A further innovation is the NGOs’ role as “social venture capitalists,” developing “frontier” markets with commercial ventures such as Grameen Phone. As incorporated businesses, they should pay taxes to the extent that their profits are not plowed back into developmental activities of the parent NGO. Our analysis indicates that the profitability of NGO enterprises varies, and larger NGOs tend to have the edge. The current market share of most NGO businesses is small, with some notable exceptions. Overall, NGO businesses that are separately incorporated from the parent NGOs’ social programs perform better, because separation allows for greater specialization, more strategic business planning, and vii Economics and Governance of Nongovernmental Organizations in Bangladesh improved governance. This will provide greater transparency and create a level playing field on which to compete with the private sector. Looking ahead, the sources of financing for NGOs will vary according to the type of activity the NGO is engaged in. The increasingly mature microfinance industry can sustain much of its activity from self-generated funds and market borrowing. Education and health services in the medium run are best financed by the government budget to ensure greater cohesion with public services. However, the benefits of greater financing of NGO social services through the budget must be weighed against the risks. These include increasing dependence on government and the potential for conflict to arise from advocacy work that challenges official policy. The second risk is that poor contracting procedures may lead to poor-quality NGOs being selected, and thus to unsatisfactory development outcomes. A common theme that emerges from the discussion of the various types of financing for NGOs is that there is a need for regulatory reform and financial accountability improvements. We find that fiduciary weaknesses within several NGOs and the lack of a certification system that distinguishes reliable NGOs from fly-by-night entities are key constraints to a proven service delivery model. The potential to access commercial sources of financing for microfinance programs is also partly impeded by the lack of a modern regulatory framework overseeing this growing sector and the lack of a rating agency or certification body. Profits from commercial ventures cause resentment among the private sector and government due to the lack of clarity regarding tax obligations and the use of profits. There is, therefore, is a clear need to strengthen regulations, regulators, and financial accountability requirements for the modern NGO sector. The regulatory framework and financial accountability The legal framework relating to NGOs is obsolete, with an over-abundance of laws and official agencies with limited capacity. The NGO Affairs Bureau (NGOAB) is widely recognized as the primary regulator of development NGOs, administering the laws relating to foreign donations. 5 Staffing at NGOAB has barely changed since 1990, even though the number of NGOs the agency is supposed to oversee has tripled. As a result, delays are common, financial reports are not scrutinized adequately, and basic data relating to foreign-funded NGOs are not maintained systematically. There are 12 laws for registering and regulating NGOs, but most NGOs register under the Voluntary Social Welfare Agencies (Regulation and Control) Ordinance 1961, administered by the Ministry of Social Welfare. The Registrar of Joint Stock Companies registers “societies and nonprofit companies,” but extremely weak capacity in this agency is compounded by the absence of computerized systems. There are other weaknesses in the legislative framework. Tax exemptions are increasingly complicated by NGO commercial activity to fund development programs. Appropriate appeals procedures and mechanisms for merger or exit of NGOs are also absent. Laws relating to internal governance and financial accountability of NGOs are inadequate. The reporting and accounting burdens placed on NGOs are contradictory, and weak 5 The Foreign Donations (Voluntary Activities) Regulation Ordinance 1978 and the Foreign Contributions (Regulation) Ordinance 1982. viii Executive Summary capacity means that regulators present a feeble deterrent against cases of poor internal governance. Capacity and computerized accounting systems vary and are related to the size of the NGO. PKSF introduced some standardized financial management practices for microfinance programs, but accounting and auditing practices are also influenced by donor financial reporting requirements that are highly variable and that typically focus on the donor-financed project rather than on the whole institution. The quality of audit reports varies significantly, and regulators rarely focus on the comprehensiveness and quality of these reports. There is little oversight of NGO audit standards within the profession. Moreover, audit reports are rarely posted in public spaces, such as on relevant Web sites. NGO boards suffer from weaknesses common in the for-profit corporate sector and typically feature the founder-director, family members, and a number of overstretched individuals who rarely rotate their positions. As the regulatory framework for NGOs is modernized, a first principle should be maintaining—or better, enhancing—the space given to NGOs to innovate and scale up their activities. Regulations should be “facilitative,” and excessive or poorly designed regulation must be avoided at all costs. Furthermore, regulatory focus should change from government micro- management of NGOs’ foreign funds to support for better corporate governance, strengthened accountability, and transparency by NGOs. Relevant regulations need to be harmonized to incorporate standardized requirements relating to corporate governance, financial disclosure, and accountability to regulatory agencies. Other necessary modifications include changes to reflect the contemporary scope of NGO activities, such as NGO commercial ventures, and to simplify registration. New laws could define NGOs as “public benefit organizations,” merging dated definitions of “social welfare agency” and “organization involved in voluntary activity” to focus on organizational purposes and activities. A new draft law governing microcredit activities is currently being considered by government. This new law is likely to address several of the lacunae in the current regulatory framework for NGOs. It will be necessary to monitor implementation of the new law to ensure that the additional regulatory burden is kept to a minimum. Moreover, it will be important to ensure that changes in the regulatory framework covering other NGO programs, are based on the experience with this new microcredit law. The composition, tenure, and functions of NGO boards need to be strengthened, and apex bodies can play a key role in fostering consensus on these issues. Uniform accounting standards are also necessary for the whole sector. The value of audit reports can be improved by setting up an oversight system to monitor the quality of audits through the proposed Financial Reporting Council. Public perceptions about the transparency of NGOs would be improved if financial disclosure were strengthened, such as by ensuring uniform public access to NGO audit reports. Improving NGO regulation also requires addressing institutional capacity weaknesses. One option that the government could consider is setting up an independent NGO Commission that would take over the functions of the current regulatory bodies. The commission could be empowered to deal with the establishment of NGOs and encourage standardized reporting and disclosure practices, similar to the role played by the Securities and Exchange Commission for listed companies. The NGO Commission could be supported by certification bodies that are licensed to rate and certify medium and small NGOs, reducing the regulatory burden of the ix Economics and Governance of Nongovernmental Organizations in Bangladesh proposed commission. The proposed commission would need to build on the practical lessons of establishing other independent commissions in Bangladesh such as the regulatory bodies in energy and telecommunications. Setting up a multi-stakeholder NGO Law Reform Committee mandated to modernize legislation governing NGOs is a recommended next step. This proposed committee could also assess the experience with the new microfinance legislation assuming it is enacted in the next few months. Scaling up through a strategic compact Large development challenges remain in Bangladesh in spite of impressive progress on a number of fronts. For instance, while infant mortality has declined, neonatal mortality is proving stubbornly hard to bring down. Although strides in school enrollment and gender parity have been achieved in primary education, dropout rates among the poor at the early stages of secondary school remain high. Microcredit has reached vast numbers of poor households nationwide and yet there remain significant pockets of the country with limited access to financial services. Female empowerment indicators have undoubtedly improved, due in part to NGO activity, but women still confront significant social challenges in both the domestic and public spheres. These facts demonstrate the need to improve services to the poor and to promote access to existing services. We find that NGOs continue to play a significant role in delivering pro-poor services that have a positive impact on household outcomes. At the same time, the direct provision of NGO services could be weakening the accountability link between elected central and local government representatives and clients. Moreover, the weaknesses in the regulatory framework and in financial accountability raise questions as to the fiduciary risks of further expansion. Does further NGO growth make sense? Is NGO provision at an optimal level, or is there a case for more, less, or a different mix of NGO services? What does this imply for the role of government and accountability to clients? Once again it is useful to assess this question from a sectoral perspective before drawing general conclusions. In microfinance, we argue that government should strengthen regulation and supervision, leaving direct provision to NGOs. Strengthening oversight over NGOs that have mobilized large amounts of savings from poor people will strengthen the accountability link between policy maker and client. The regulatory framework and government policy should facilitate and support microfinance NGOs to: (a) meet the needs of underserved areas (such as the northwest); (b) scale up successful microfinance experiments targeting the poorest; and (c) gradually expand into microenterprise lending, paying close attention to portfolio quality. Programs for the poorest and those in underserved areas may require carefully designed subsidies. Turning to education, we find that NGO schools successfully reach the poor with quality services, and we argue that there is a case for further expansion of NGO primary schools as well as tighter focus on geographically and socially marginal groups. Where government and NGO x Executive Summary primary schools coexist, a voucher system could be piloted to provide consumer choice, strengthening the short route to accountability between provider and client. Other strategic growth areas for NGOs in education are at the pre-primary level, as highlighted in the PRSP, and in disseminating educational best practices in relation to school management, community involvement, personnel management, and teaching methods. There are recent examples of NGOs handing over educational assets to communities, following a period of capacity building and co- financing—an interesting model that, if successful, could bring about large benefits. In health, the main strength of the NGO system is the army of community health workers who provide critical awareness raising within households and operate as intermediaries between government or private facilities and poor clients. Looking ahead, the focus needs to be on second-generation challenges in the health sector that could be more complex to address. To help reduce neonatal mortality, NGOs could be further involved in raising awareness of postnatal care, as well as equipping community health workers to better support home deliveries. Interventions to improve stubbornly poor nutritional outcomes could also be scaled up through the network of community-based health workers. NGOs offer a strong comparative advantage in control of communicable diseases: community health workers can promote awareness, detection, and treatment of such diseases as HIV/AIDS, malaria, and tuberculosis, as has been seen in the success of the directly observed therapy program for TB. Scaling up these NGO disease-control efforts would be a clear public good. As part of a process of realigning priorities, NGOs need to develop exit strategies in areas where quality services are already being provided or where they lack a comparative advantage. Looking ahead, the role of NGOs in building the capacity of community groups to access resources is one that could, in the long run, bring greater benefits to a larger number of people than directly providing services. Is further NGO growth possible? The report analyzes the organizational factors underlying success in the implementation of NGO programs to date and considers whether these provide a basis for further growth. Strong, charismatic leadership contributed to the expansion of large NGOs by providing strategic direction, attracting donor support, instilling a sense of collective purpose, and strengthening staff incentives. The challenges of succession cannot be taken lightly, as loyalty to the founding head is an important part of the unity of many NGOs. However, there are encouraging signs that new managerial structures and processes are being established to smooth the transition to a new generation of leaders. Aside from leadership skills, effective human resource management was vital in scaling up service delivery. We argue that the relative effectiveness of NGO staff is not linked to greater altruism, nor do NGO staff enjoy better salaries and perks than government staff, though incentive systems that tie remuneration to performance certainly help. Close monitoring and supervision supported by “campus-style” living arrangements in the field, along with the ability to dismiss employees for poor performance, offer the most convincing explanation of good staff performance. There are other factors as well, such as investments in staff training, participatory decision making, and the satisfaction of seeing results on the ground. xi Economics and Governance of Nongovernmental Organizations in Bangladesh Rapid growth of the NGO sector was also enabled by a “franchising effect” in which NGOs reproduced organizational structure and program design in hundreds of branches nationwide. The impact was particularly marked among microfinance programs, where a standard range of financial products permitted rapid recruitment and training of staff. This model remains dominant, and the focused expansion recommended for the future will also depend on a franchising model. However, the proposed areas of expansion are more complex (e.g., suitable programs for the poorest, programs to reduce neonatal mortality) and thus will require stronger staff capacity. Greater investments in developing a solid middle management layer will be needed to oversee these more complex products. The franchise model should also strive not to stifle innovation, such as in microcredit and education where the Grameen and BRAC models are so dominant. The strategic compact As we look ahead to scaling up, there is a demonstrable need for a strategic compact between government, NGOs, donors, and clients. The proposed actions are a combination of short- and medium-run measures whose combined impact will improve services to the poor and accelerate Bangladesh’s progress toward achieving the Millennium Development Goals. The main role for government is in modernizing the legal and regulatory framework and ensuring effective oversight of NGO activities. Toward this end, it is important that all parts of government recognize NGOs as allies in development and accept constructive criticism as supportive of a democratic polity. Government should withdraw where NGOs have the clear edge in the delivery of certain services. As the process of decentralization evolves, local governments can play a more active role in delivering services in partnership with NGOs. Government ought to strengthen monitoring and contracting procedures to improve the effectiveness of aid that is channeled to NGOs. The Government-NGO Coordination Council should be revived to provide a forum for regular government-NGO discussions. Donors should continue to support NGO activities in Bangladesh, both to improve pro- poor service delivery and to promote a broad-based civil society, while ensuring that they do not undermine the accountability of NGOs to Bangladeshi stakeholders. Donors should also reduce the transaction costs of their financing by harmonizing reporting requirements, upgrading financial management skills of their own staff, and retaining “institutional memory” within their organizations. In financing social sector activities, donors also need to develop a clear strategy for sustainability, as has occurred in microfinance. This strategy ought to be developed in tandem with government and NGOs, so that in the medium term the budget progressively absorbs the costs of financing various social sector programs. In areas such as advocacy, where the conflicts of interest with government financing are too great, donors should fund NGOs directly and work with NGOs to tap the market for greater private charitable contributions. Moreover donors need to be accountable for the occasions when poor project design or inadequate supervision contributes to unsatisfactory implementation of projects involving NGOs and institute mechanisms to prevent these from reoccurring. xii Executive Summary NGOs also have a large agenda if they are to continue to be effective partners in development. Priorities include the need to (a) strengthen financial management and corporate governance; (b) scale up efforts to provide public disclosure with respect to income sources, expenditure patterns, and programs, along with a public information strategy to address misconceptions about NGOs; (c) make greater efforts to align programs with national strategies, coordinate with local and national government officials, and act as facilitators of government, private sector, and community service provision; (d) revisit strategic directions periodically based on national development needs and internal capacity constraints; and (e) develop a middle management layer to oversee more complex development programs. Finally, clients can also support the strategic compact using mechanisms that improve their capacity to demand and monitor services, for example, vouchers enabling choice between public, NGO, and private providers. xiii Chapter 1: Introduction 1.1 Bangladesh’s achievements in poverty reduction and human development over the past three decades have been credited in substantial part to the mix of public and private service provision, including the pioneering approaches of its development NGOs. 6 A plural approach to poverty reduction and social service delivery in Bangladesh remains necessary because of the scale of the problem, and because of its severe and intractable nature. Acting alone, the government of Bangladesh cannot command the resources, personnel, expertise, or administrative reach necessary to maintain progress on poverty reduction. Between them, government and NGO programs have achieved wider total coverage in terms of services for the poor and have developed innovative, replicable approaches to tackling poverty. Government- NGO partnerships also permit the scope for both to exploit their comparative organizational advantages to enhance service provision outcomes. 1.2 However, the rapid growth and diversification of the NGO sector has also given rise to questions and concerns. Government is worried about alleged NGO involvement in proscribed political activity and about financial irregularities, concerns that led to a freeze on foreign funds to a prominent NGO. This issue has also provoked moves to tighten the regulations overseeing NGOs, which are widely recognized as weak. The commercial private sector regards the growth of NGO commercial activities as unfair competition; there is debate as to whether this market involvement is a useful strategy for achieving program sustainability or an unwarranted diversion from the focus on poverty reduction. As the volume of resources managed by the NGO sector grows, so do concerns about NGOs’ internal and corporate governance, including their relations of accountability with government, donors, and beneficiaries. There is an increasing focus on the modalities and implications of different contracting arrangements between government and NGOs. Finally, questions have been raised about the coverage, impact, and quality of NGO services, given the increasing share of aid resources that are channeled through NGOs. This study aims to contribute to these debates. 1.3 The report is organized as follows. The remainder of this chapter describes the unique characteristics of the NGO sector, setting out current debates about the role of NGOs in Bangladesh and outlining the analytical framework used in the report. Chapter 2 takes a closer look at three key services provided by NGOs—microfinance, education, and health—and at their advocacy role, and summarizes the evidence on the development impact of these activities. Chapter 3 examines the implications of changes in NGO financing, focusing on donor aid, government financing, private donations, and self-financing through microfinance surpluses and commercial enterprises. Chapter 4 uses the analysis of the state of the NGO sector to examine the legal, regulatory, and fiduciary issues relating to NGOs in Bangladesh. Chapter 5 discusses 6 The number of people below the poverty line fell from over 70 percent in the early 1970s to around 50 percent in 2001. Infant and child mortality declined from 140 and 95 respectively in 1972 to about 60 and 30 in 2000, and on these indicators Bangladesh is on course to meet targets set by the Millennium Development Goals. Gross primary enrollment increased from 72 percent in 1980 to 98 percent in 2001; significantly, the share of girls in primary enrollment increased from 37 percent to 50 percent during this period. The fertility rate fell from 6.3 children in 1975 to 3.3 in the mid-1990s, and annual population growth at 1.5 percent per year is lower than in India and Pakistan (World Bank 2003a). Economics and Governance of Nongovernmental Organizations in Bangladesh the case for further scaling up of NGO activity in certain areas, the internal capacity implications, and the roles of government, donors, and clients in this process. 1.1 Characteristics and evolution of the Bangladesh NGO sector 1.4 Bangladesh is unusually well endowed with a dense and sophisticated network of development-oriented NGOs. While other countries may have comparably sized nonprofit or “third” sectors, few if any combine the qualities of scale with a secular, innovative development orientation and a focus on the poor, as is found in Bangladesh. Institutional and definitional variety makes direct data-based comparison with other developing countries difficult. 7 However, by 2000, more than 90 percent of rural communities had some NGO presence. 8 Around 16 million mainly poor women are now reached through microfinance programs, some 8 percent of primary enrollment is provided by NGO schools, and there are nationwide health and sanitation programs that involve NGOs. Approximately 45,000 NGOs are registered with the Ministry of Social Welfare, though the vast majority are not the active development NGOs that are the focus of this report (see chapter 2). As of 2004, the NGO Affairs Bureau (NGOAB) had registered 1,882 NGOs as potential recipients of foreign funds, and we estimate that about 2,000 development NGOs are currently operating in Bangladesh. 9 1.5 These NGOs range from tiny local groups to a small number of vast organizations that are among the largest NGOs in the world. Following unprecedented expansion in the 1990s, these big entities now employ thousands of staff and handle multimillion-dollar budgets. They dominate the NGO sector in Bangladesh, and are nearly comparable to government departments in terms of their budgets, reach, and complexity. The sector undertakes a wide range of activities, with NGOs active in microfinance, health and education services, social safety net programs, agricultural extension, social forestry and environmental protection, safe water and sanitation, disaster management and relief, and legal and human rights education. 10 However, as we discuss in chapter 2, the typical NGO branch focuses on providing three or four services. While some NGOs put an early emphasis on mobilizing poor people to press their claims on local or national government, the sector as a whole has gained prominence in advocacy activities only since the 1990s. 11 NGO commercial enterprises are also growing in importance, including businesses that may contribute to rural economic growth and employment of the rural poor by linking rural and urban markets. 1.6 Bangladesh’s NGO sector is remarkable for the speed with which it grew to its present size and prominence. While community-level welfare activity has a long history in the 7 The most rigorous source of comparative data, the Johns Hopkins Comparative Nonprofit Sector Project, unfortunately does not collect data for Bangladesh. However, to the extent that comparison is possible, the data for Bangladesh suggest that it boasts an unusually large number of active domestic, development-oriented NGOs. So while India has around 1 million nonprofit organizations, these include professional and caste associations, and only a small proportion of the total are likely to be involved in development activities. Of Thailand’s 15,000 nonprofit organizations, up to half are cremation associations, some with associated welfare activities. Only 800 urban nonprofit organizations were identified in Ghana, and most of these were unregistered; however, this figure excludes rural credit, savings, and self-help associations (Salamon and Anheier 1998). 8 Fruttero and Gauri 2005. 9 Information from the NGO Affairs Bureau, October 2004. 10 Zohir 2004. 11 Stiles 2002. 2 Chapter 1: Introduction region, 12 the institutionalized nonprofit sector was insignificant at the time of Bangladesh’s independence in 1971. The leading NGOs of the present cut their organizational teeth on postwar relief activities in the 1970s. Their initially radical edge was blunted early on, as most NGOs began to prioritize service provision over “conscientization” and mobilization work in the 1980s. This new thrust reflected donor preferences for a less radical model of civil society and for more emphasis on service provision; also, NGOs themselves began to recognize that efforts to mobilize the poor without also providing them with economic inputs were unlikely to succeed. This shift meant that NGOs were, with some justification, viewed as aid-dependent service delivery agents supportive of the military regime through the 1980s. 13 1.7 It was during the late 1980s that donors began to fund NGOs on a large scale, particularly a small group that now includes the largest organizations: BRAC, the Grameen Bank (registered as a specialized bank but sharing many characteristics of NGOs), Rangpur Dinajpur Rural Service (RDRS), Association for Social Advancement (ASA), and Proshika. 14 A new phase in NGO advocacy work began with the emergence of multiparty electoral democracy in the 1990s. By the mid-1990s, development NGOs began to build new alliances and develop new strategies for influencing and challenging official policies and practices, in collaboration with other branches of civil society. 1.8 One decisive factor in the rapid growth of the sector has undoubtedly been the availability of foreign aid. Grants to NGOs averaged $160 million annually in the first half of the 1990s and have grown to about $255 million annually since then (see table 3.1). These figures do not include the share of credit funds disbursed by lending agencies that flow through NGOs, as this is not recorded by the NGO Affairs Bureau and is therefore harder to estimate. However, estimates for this study indicate that around $82 million has been channeled annually to development NGOs from concessional loan sources. Adding these two sources, we estimate that total aid to NGOs rose from about 14 percent of total aid to Bangladesh in the first half of the 1990s to about 25 percent since 1995. The larger NGOs are now funding a rising proportion of their development activities from fees for services (primarily microcredit interest income) and from the proceeds of their commercial enterprises, as detailed in chapter 3. 1.9 Population density and the availability of a good transportation infrastructure network were also important contextual factors behind the expansion in NGO services. Other institutional features linked to NGO growth are the presence of strong leadership and management skills, the development of staff incentives and monitoring systems, and the in-built capacity for experimentation (see chapter 5). 15 The regulatory regime also contributed to NGO growth, as we discuss below. 12 Hashemi 1995. 13 Lewis, forthcoming. 14 BRAC, now known by its acronym, was called the Bangladesh Rural Advancement Committee when it was founded in 1972. The Grameen Bank is registered as a specialized bank and is not an NGO. However, its credit delivery model has been widely replicated by the NGO sector in Bangladesh, and for this reason Grameen’s experiences are often included in a discussion of NGOs. 15 For a detailed case study of the institutional development of ASA, now one of the largest NGOs in the country, see Rutherford (1995). 3 Economics and Governance of Nongovernmental Organizations in Bangladesh 1.1.1 The role of government in creating a supportive environment 1.10 The importance of the space made for NGO activity by successive Bangladesh governments is often underplayed, and deserves some attention. In comparative, and certainly regional, perspective, public policy toward NGOs in Bangladesh has been unusually successful in balancing the need for official oversight with the operational autonomy necessary for NGOs to operate and innovate. The absence of heavy-handed constraints on the NGO sector by the government of Bangladesh has been noted to explain why the sector has grown more rapidly in Bangladesh than in other countries. 16 In part this reflects the comparative simplicity of the regulatory architecture: a single layer of government in effect regulates NGOs in Bangladesh, in contrast to the situation in India and Pakistan, where the federal systems create multiple layers of government for NGOs to negotiate. 17 1.11 In the case of NGO-run microfinance programs, it is generally acknowledged that the nonintrusive policies of successive governments of Bangladesh were instrumental to the sector’s growth. Essentially, NGO microfinance programs were permitted to expand as long as they mobilized savings only from members. Instead of direct government regulation and oversight, there is de facto supervision by an autonomous apex body, Palli Karma-Sahayak Foundation (PKSF), whose role includes monitoring standard performance indicators. To date, the regulatory framework does not cover or legally mandate supervision of microfinance activity or microfinance institutions specifically. Also vital to the space for microfinance activity was the stability and economic opportunities ensured by impressive macroeconomic performance over an extended period. 18 1.12 The government’s fiscal policy signals the value placed upon NGO service delivery. The apex body for microfinance, PKSF, has received sizeable budgetary transfers over the past few years. The recent creation of the NGO Foundation and the endowment it has received from the government reflect official willingness to support NGOs and acknowledge their strength. Government included a provision for tax deductions for charitable contributions in the 2005 budget. This is line with tax laws in many countries and has the potential of providing a significant boost to private philanthropic contributions to NGOs. 1.13 Also contributing to the supportive official environment are the long-standing partnerships between government and NGOs, some dating back more than 20 years. One well- known example is the joint effort to promote oral rehydration therapy (ORT), which was used very effectively to tackle diarrheal diseases in a much-studied and emulated model of development management (see chapter 2). Government actors from all levels were engaged in the program, from the secretary of health to thana (subdistrict) 19 officials to representatives of union councils (the lowest level of local government). Even the post office was brought on board, using postage stamps to disseminate messages about ORT. Such long-standing collaborative arrangements may also have helped legitimize the work of NGOs in wider society, 16 Smillie and Hailey 2001, 40; Salamon and Anheier 1998. 17 Smillie and Hailey 2001. 18 Zaman 2004. 19 Thanas are subdistrict administrative units, also known as upazilas. There are 507 in total. 4 Chapter 1: Introduction contributing to the high levels of acceptability and trust these organizations have enjoyed in Bangladesh. 1.14 The Bangladesh Poverty Reduction Strategy Paper (PRSP) points to the significant role assigned to NGOs, a role that is considerably greater than that provided in most other PRSPs. The Bangladesh document discusses the role of NGOs in areas as varied as agriculture, education, health, microfinance, safety nets, nutrition, local governance, anticorruption, and access to justice. 20 1.15 Anecdotal evidence suggests that NGO and government staff have developed effective working relations. 21 This evidence is supported by the 2003 World Bank survey of NGOs, in which the staff of field-based NGOs overwhelmingly report viewing both national and local government officials as helpful in their work (see annex figure 1.1). At higher levels, it seems that government-NGO relations have frequently been smoothed by the close personal relationships between senior officials and NGO leaders, facilitated by their mutual membership in the close-knit national elite. 22 Public policy makers in Bangladesh are well informed about the vast and intractable nature of the development challenge, and are realistic about the capacity of government to tackle the problem alone. 23 1.16 This overall context of the government’s facilitating role for NGOs needs to be borne in mind as we turn to the next section that describes recent tensions between the government and specific NGOs. 1.2 Current debates 1.17 A number of issues are currently at the forefront of the public policy debate on NGOs in Bangladesh. These include (a) perceived weaknesses in the regulatory framework and in the financial accountability of NGOs; (b) the scope, impact, and cost-effectiveness of NGO activities, including trade-offs between service delivery and advocacy; (c) the implications of shifts in NGO financing that have resulted in greater reliance on commercial activities and microfinance revenues; and (d) the nature of government-NGO partnerships and contracting arrangements. 1.18 There is consensus that gaps and weaknesses in the regulation of NGOs need to be addressed as a priority. The organizational forms of NGOs in 2006 are very different from those for which the existing legislation was designed. The need for better external regulation and supervision is particularly apparent for the vast microfinance sector, which now handles a stock of around $560 million worth of members’ savings. Despite the scale of microfinance provision in Bangladesh, the sector is distinctly underregulated when compared to other countries. The 20 Government of Bangladesh 2005. The PRSP matrix includes NGOs along with other implementing agencies in the following areas: improving agriculture extension; promoting quality seed development; improving livestock productivity; disaster management; reaching the ultra-poor; access to pre-primary, primary, and technical education; improving child, maternal, and reproductive health; controlling communicable and noncommunicable diseases; improving maternal and child nutrition; promoting local government; tackling corruption; and improving accountability and access to justice. 21 Goetz 2001. 22 Close personal relations between members of the Bangladesh elite are examined by Hossain (2005). Professor Yunus of Grameen Bank admits that he used such connections to “overcome impossible bureaucratic hurdles” (1999, 168). 23 Hossain 2005. 5 Economics and Governance of Nongovernmental Organizations in Bangladesh potential for financial irregularity within NGOs has also directed attention to the current state of NGO corporate governance. The need for improved external regulation was underlined by the 2002 split in the Association of Development Agencies in Bangladesh (ADAB), the apex body of development NGOs in the country, which formerly played some role in self-regulation of the sector. 1.19 Concern within the government about alleged partisan political activity and financial irregularities has placed the need for improved regulation of NGOs on the public policy agenda. The renewed emphasis since the 1990s on NGO advocacy and, in some cases, on social mobilization has produced its own controversies, highlighting the fine line between civil society activism and partisan politics in Bangladesh. Recently, the government-NGO relationship has deteriorated. Citing concerns about allegedly inappropriate NGO activities, the government in 2000–1 froze the foreign funding of a handful of NGOs, including that of a large national NGO, Proshika. This freeze remains in effect for Proshika as of early 2006. In addition, proposed amendments to regulations governing foreign-funded NGOs have been perceived within the NGO community as heavy-handed. In response to strong criticism the draft legislation has been withdrawn, pending further consultation. In this context, the increasing prominence of NGO advocacy work has stimulated debate about the political space for and limits to NGO activism. 1.20 The public image of the sector has been affected by what some perceive as an illegitimate shift into for-profit activity. NGOs of all sizes have established commercial activities, though enterprises associated with the Grameen Bank, BRAC, and Proshika are larger, engage a range of sectors, and are therefore more visible. Diversification into profit-making activity enables such organizations to reduce their dependence on donor funds for their development programs. Certain ventures have also had clear positive developmental consequences by entering new markets (e.g., mobile phones for the rural poor) and by linking rural and urban markets. However, the commercial private sector has resisted the entry of NGOs into for-profit activities on the grounds of unfair competition. Questions have been raised about subsidies to the equity base of NGO commercial activity, and about whether the tax-exempt status of NGOs provides an implicit subsidy to their commercial operations. The growth of microcredit and the perception that NGOs are generating large surpluses has meant increased public scrutiny and debate about the level of interest rates. These issues are linked to larger questions about whether the pursuit of financial sustainability diverts attention from serving the poor, and about the impact of microcredit. 1.21 The evidence on service quality and on the development impact of NGO programs remains limited. The scale, geographic coverage, and types of activities carried out by NGOs are often described on an anecdotal basis or by focusing on a single institution. Claims that NGOs in Bangladesh are a “parallel government” have seldom been tested against household and other survey data that are representative nationally and measure scale and access across a range of services. Moreover, there is a paucity of well-designed impact assessments of entire programs, and those that exist are often not disseminated effectively outside the development and academic community. 6 Chapter 1: Introduction Box 1.1 Eight sizes fit all: Lessons from WDR 2004 What works to improve service provision for the poor? The many examples documented by World Development Report 2004 show that success comes in many shapes and guises: pro-poor service provision is found in a range of contexts, with no single template for success. Distilling lessons from this diversity, however, WDR 2004 outlines eight sets of basic service provision arrangements that appear to deliver quality services to the poor in most contexts. The eight sizes that “fit all” depend on a number of factors: the political system, the characteristics of the service-user population, and the difficulty of monitoring the service in question. Under pro-poor regimes whose populations have homogenous service needs, for instance, it may be best for central or local government to finance the service, while provision and regulation are contracted out (size 1) or retained by the public sector (size 2). Whether size 1 or 2 fits best will depend on how difficult it is to monitor provision. Where service users’ needs are more diverse, local government financing is more appropriate, with the choice of contracting out (size 3) or local government provision (size 4), again depending on the difficulty of monitoring contracts. Where the political context does not favor pro-poor outcomes, services are more appropriately delivered in ways that strengthen the power of users over service delivery. These modes are necessarily more experimental and may involve different modes of contracting (size 5), self-monitoring provision (size 6), community control (size 7), or market-type mechanisms (size 8). Source: World Bank 2003d. 1.22 Government contracting of NGOs has costs as well as benefits. The level of financing going to NGOs through government systems has increased since the mid-1990s. This trend clearly has its benefits, as the sustainability of certain donor-dependent NGO programs is less uncertain once the program is absorbed into the budget. But it also poses a few challenges. First, it may serve to dampen NGO advocacy work, as organizations that depend on official financing are less likely to risk criticizing government. Second, the lure of government contracts has attracted more “fly-by-night” NGOs, which in turn have tainted the image of the NGO sector. Third, increasing use of NGOs to deliver public services raises issues of the role of local government and communities in service provision and monitoring, as well as questions about accountability. We elaborate on these issues in the next section. 1.3 Assessing service provision for the poor in Bangladesh: An analytical framework 1.23 The report evaluates the NGO contribution in Bangladesh using an analytical framework designed for assessing ways to improve service provision for the poor. This framework is built around themes developed in World Development Report 2004: Making Services Work for Poor People (World Bank 2003d). Despite the current emphasis on context and diversity, there are some general analytical starting points from which to assess different options for service provision in developing countries. These include: (a) the types and characteristics of services provided; (b) the relationships between providers, policy makers, and service users; and (c) the overall national political context for public and private service provision (see box 1.1). 7 Economics and Governance of Nongovernmental Organizations in Bangladesh 1.3.1 The characteristics of services 1.24 Analysis of the characteristics of NGO services will give empirical substance to debates about the appropriate role of NGOs. The classic case for public intervention rests on the grounds that markets fail to provide enough of the services society deems desirable. Such services may include “public goods” as well as those with positive externalities, those about which information is imperfect, or those that contribute to equity and social justice (see box 1.2). Public intervention does not necessarily mean that government itself provides the service: the intervention may be more effective if government enables others to undertake provision, and then regulates or monitors the arrangement. The appropriate intervention depends ultimately on the capacities of the providers, on the availability of finance, and on how well the resulting arrangements fulfill political goals. The conditions underlying the case for public intervention or NGO provision may change when certain development targets are achieved, or when the capacity of other providers—communities, local government, and the private sector—improves. 1.25 There is also a case for public intervention to involve NGOs in relation to services that are difficult to monitor or regulate. These tend to involve “incomplete contracts”—activities whose process or outcomes cannot be quantified against preset targets. Immunization outcomes can be measured in terms of the population covered, for example, whereas the performance of schoolteachers is less easily quantified. Under such conditions, the incentives for good service provision may not be produced within conventional bureaucratic or profit-oriented organizations. Organizations that feature internally generated incentives to perform, such as the presumed altruism of NGO employees, may be better suited to providing such services. 24 1.26 Chapter 2 analyzes four key NGO activities—microfinance, education, health, and advocacy—to provide a picture of the type, scale, and impact of NGO activity. This chapter adopts a “public finance” approach, describing spending patterns and assessing the equity and efficiency of microfinance, education, and health services. NGO activities are discussed further in chapter 3, which looks at NGO commercial ventures in the context of changing trends in NGO financing. The rationale for intervention is not static, making it necessary to periodically revisit key sectors in which market failure has prompted a role for NGOs and/or the public sector in the past. We explore issues of NGO management, incentives, and organizational culture in chapter 5. 1.3.2 Institutional actors and relationships of accountability 1.27 An assessment of how NGOs can help improve services to the poor requires an understanding of relationships between poor service-users, politicians or policy makers, and service providers. It is the links between these actors that create the relationships of accountability that govern service provision and ultimately determine its quality. Accountability may be exercised through what World Development Report 2004 calls the “long route,” the set of relationships that connect service users to service providers through their links to policy makers, or the “short route,” the relationships that enable users to hold service providers directly to account. As citizens, service users can in theory influence the quality of services by using their political power as voters to sway policy makers, but in practice this rarely results in effective influence over policy. Advocacy efforts by NGOs and other actors can help bridge this gap, and 24 Besley and Ghatak 2001. 8 Chapter 1: Introduction Box 1.2 Market failures and the case for intervention Market failures include instances in which the kind or quantity of a service that society deems desirable is not made available through the market, creating a case for public intervention. This is often the case for public goods—clean air or malaria control, for example. Public goods are defined by two characteristics. The consumption by one person of a public good, once it has been produced, does not reduce its availability to others. And people cannot be prevented from consuming a public good, once produced, without having paid for it. Public intervention into service areas that are not public goods can sometimes be justified on the grounds that they generate externalities. A positive externality is a beneficial side effect that goes beyond that gained by the consumer herself, and usually accrues to the wider society. Goods with positive externalities may be underprovided unless governments intervene. Immunization, for example, generates important positive benefits for the wider society. So the case for public funding of immunization is usually made on the grounds that the public value derived from immunization is greater than what individual households would be willing to pay for themselves. Differences in access to services or imperfect information about services can also cause market failures, for example, by discouraging people from using services at levels that would benefit the wider society or themselves. In such cases, government may need to intervene by regulating the service, publicizing its benefits, or improving access, but it may not need to provide the service itself. some Bangladeshi NGOs have made efforts in this direction. But as chapter 2 documents, there is well-placed skepticism about the capacity of the NGO sector to mount an effective challenge to government in sectors where many of its members seek government support and/or financing. Nevertheless, NGOs have helped communicate the needs of poor service-users to policy makers through the less direct “demonstration effect,” proving that it is possible to design services that reach the poor, and poor women in particular. 1.28 Even when citizens are able to influence policy makers, the other half of the long route to accountability—the “compact” between providers and policy makers—may impede service improvements. This may be because policy makers are too close to providers to regulate or monitor provision adequately, or because educated, urban public sector employees may not be inclined to serve the needs of poor, rural communities. Solutions to these problems in a number of countries have included growing reliance on public-private partnerships. Because efficiency and quality cannot be guaranteed by mimicking market forces and must necessarily rely to some degree on trust, NGOs have tended to be favored as substitute service providers. It is conventionally assumed that NGOs have the edge on other frontline providers because they (a) reach poor, remote communities; (b) are more flexible, responsive, and experimental; (c) feature internal incentives for staff performance; and d) may be more cost-effective than other providers of services to the poor. The analysis here takes a closer look at the empirical evidence about the NGO advantage in the Bangladesh context. In this context, an emergent area for analysis is the increasingly important role of public-private partnerships in subcontracting arrangements. 1.29 The short route to accountability, the direct link between providers and users, tends to be weakest where users are socially and physically distant from providers, as is characteristic of public sector provision in Bangladesh. But it may be strengthened where users can behave as clients exercising choice, or are empowered to participate in provision, as in forms of 9 Economics and Governance of Nongovernmental Organizations in Bangladesh “institutionalized co-provision.” 25 The increasing prominence of NGOs in providing services reflects their comparative advantage in bringing services directly to poor users. However, it also highlights the need for a closer look at the conditions for accountability by NGOs to beneficiaries—conditions that cannot be taken for granted, as discussed in chapters 2 and 5. Chapter 4 analyzes the mechanisms and arrangements for NGO accountability to government, donors, and beneficiaries through its analysis of the regulatory, legal, and fiduciary issues. Contractual relations between government and NGOs are explored in chapter 3. Together these provide an account of the half of the “long route to accountability” that links policy makers and providers. 1.3.3 The political context of public and private service provision 1.30 The political context for public service provision in Bangladesh is mixed. On the positive side, Bangladesh has seen a rising share of public spending allocations going to the social sectors and to pro-poor programs within these sectors. Yet the weakness of reform efforts to enhance efficiency in the public sector or to tackle the serious problem of corruption has resulted in weak delivery systems that have diluted the impact of pro-poor allocations. 26 The literature also suggests that there may be indirect positive and negative consequences of the prominent service provision role of NGOs in the Bangladesh context. The positive side effects of NGO services may include incentives for public sector providers to improve their performance in response to a competitive threat or positive example. 27 On the negative side, to the extent that decisions about the financing of NGO services in Bangladesh take place among donors and their NGO partners, these decisions are divorced from national policy processes. This reduces the possibility of feedback mechanisms creating strong political coalitions in support of public services. This in turn raises questions about the desirability of further NGO expansion, an issue that will be dealt with in chapter 5. 1.31 The political context for NGO service provision is similarly mixed. As discussed in section 1.1.1, successive Bangladeshi governments have permitted and even encouraged a role for NGO service provision in recognition of the overarching goal of poverty reduction. The space permitted to NGO activities, coupled with the rising importance of government-NGO partnerships, suggests a political context generally in favor of pro-poor service provision. However, the essentially voluntary character of NGO service provision means there are no official or NGO channels through which service users may call NGO service providers to account. As is the case in other aid-dependent developing countries, donors in Bangladesh frequently intervene in the compact between providers and policy makers, both through their general influence over public spending and through their direct funding to service providers. This may cut out the role of policy makers in the compact, further reducing the scope for accountability to service users through the “long route.” The long-term political implications may include a stronger reliance on NGO service that diminishes accountability to service users. As discussed below, recent events highlight the vulnerability of service provision arrangements that depend on the occasionally volatile relationship between NGOs and government. Chapters 3 and 5 analyze how this relationship can be made less fragile with the help of a stronger legal and 25 Joshi and Moore 2004. 26 World Bank 2003c. 27 White 1999. 10 Chapter 1: Introduction regulatory foundation and the development of a strategic compact between government, donors, NGOs, and clients. 1.32 An analysis of NGO service provision needs to relate to the overall service delivery framework in Bangladesh, including that of the private sector. There are many impressive examples of successful service delivery outcomes in Bangladesh. As discussed earlier, they have resulted from the pluralist service provision system comprising government, NGOs, and the private sector. At the same time, there are certain clear areas for improvement in both the public sector (discussed above) and the private sector. Private sector services in education, health, and banking have rudimentary corporate governance arrangements, are overseen by an overstretched regulatory framework, and in many cases are linked to the politics of the country. The analysis of NGO activities, their corporate governance, and their financing arrangements needs to be considered in light of this more general framework of service provision in Bangladesh as a relative benchmark of NGO activities. 11 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 2.1 This chapter takes a closer look at the main activities of NGOs in Bangladesh. It focuses on four principal areas of NGO activity: microcredit, education, health, and advocacy. The chapter provides an overview of the key issues, expenditure patterns, and targeting in each of these sectors and discusses available evidence on the impact of these interventions at the individual, household, and community levels. 2.1 Defining and classifying NGOs in Bangladesh 2.2 The development-oriented NGOs in which we are interested are a subset of a larger body of nongovernmental organizations. In Bangladesh, civil society includes approximately 45,000 clubs, local-level organizations, religious organizations, foundations, and development- oriented NGOs that are registered with the Department of Social Welfare. It also takes in national and local trade unions, professional and business associations, and numerous local community-based organizations, the latter including savings, religious, community development, and social welfare groups, many of which may be temporary and informally constituted. 2.3 Although part of the wider body of civil society, development-oriented NGOs in Bangladesh have a distinctive set of features. They constitute a coherent set of organizations, readily identifiable on the grounds that they feature most or all of the following: (a) a tight focus on poverty alleviation and on disadvantaged groups through direct delivery of services to poor people and/or advocacy and social mobilization activities aimed at improving services to, organizing, or establishing the rights of the poor; (b) a formal organizational presence (e.g., a board or formal governing structure, institutionalized reporting and accounting procedures, and legal registration with relevant authorities); (c) receipt of foreign funds, either directly from aid donors or indirectly through contractual arrangements with government agencies or other NGOs; (d) Bangladeshi staff and management in most cases, and a predominance of professional as opposed to volunteer staff; and (e) a secular orientation. 2.4 Size matters: both the sector as a whole, and individual organizations within it, stand out because of the scale of their activities. Registration with the NGO Affairs Bureau (NGOAB) and the Association of Development Agencies in Bangladesh (ADAB) can serve as a reasonably reliable indicator of involvement in development activity. By late 2004, 1,882 NGOs were registered with NGOAB, with approximately 1,100–1,200 of them receiving foreign funds. 28 At the same time, the 2004 list of NGOs compiled by the apex body ADAB shows 2,152 NGOs. Hence, for the purposes of this study we assume there are roughly 2,000 development NGOs in Bangladesh. 28 Information from NGOAB, October 2004. Economics and Governance of Nongovernmental Organizations in Bangladesh Box 2.1 World Bank Survey of NGOs in Bangladesh In order to design the sampling frame for a 2003 World Bank survey of NGOs, lists from the NGO Affairs Bureau and the Association of Development Agencies in Bangladesh (ADAB) and the branch lists of larger NGOs were merged to compile a list of 6,559 stand-alone NGOs, NGO headquarters offices, and NGO branches. Out of a total of 507 subdistricts, 35 were selected randomly, and the NGOs in each were counted. A stratified sampling technique was used to ensure that subsamples of different types of organizations were included. The survey was administered to 310 NGO branches as well as to head offices of small NGOs, where these are directly involved in development work. Two questionnaires were administered, one to the manager/director of the field NGO and one to a focus group in the community in which the NGO was operating. The NGO manager was asked about advocacy and service-delivery activities, partnerships and subcontracting arrangements, coverage and target groups, fees for services, religious orientation, qualifications and evaluation of staff, the NGOs’ assets and revenues, networking with other NGOs, and the organization’s governance structure. The community groups were asked to assess the community’s needs, to list NGOs/activities in their area, to evaluate NGO activities, and to compare the performance of different service providers, including local and national government, NGOs, private sector interests, and community actors. 2.5 Among these, a small group of organizations stand out as having grown to unprecedented size and importance. Among the branches of development NGOs compiled for the World Bank NGO survey (see box 2.1), 40 percent were branches of BRAC and ASA, two of the largest entities. In another sample of 720 NGOs, 90 percent had programs in less than five (out of 64) districts, and only two NGOs (BRAC and ASA) had programs in all districts. The dominance of very large NGOs is even more pronounced within the microcredit sector. There are about 1,200 microfinance institutions (MFIs) in Bangladesh, most of which are NGOs, 29 but the industry is dominated by four large MFIs: BRAC, Grameen, ASA, and Proshika. These four combined serve about 14.2 million clients, or 87 percent of all nongovernment microcredit borrowers. After the “big four,” only 10 other MFIs have more than 100,000 borrowers each. 2.6 There are other ways of defining a large NGO. One review included not only NGOs with large coverage but also small NGOs that have a significant impact on policy through their advocacy work. The 11 NGOs included in this study accounted for 85 percent of all donor funding to NGOs. 30 These very large organizations give the sector as a whole much of its distinctive character. BRAC, Grameen, and ASA employ tens of thousands of employees and handle multimillion-dollar budgets. These big organizations have also provided models for the programs and organizational structures and practices of other, more recent entrants. They continue to provide leadership and direction for the sector as a whole through the NGO associations, sectoral umbrella organizations, and national policy arenas. 29 One notable exception is the Grameen Bank, which is a specialized bank. 30 The review covered BRAC, ASA, Proshika, RDRS, CARITAS, Thengamara Mohila Sabuj Sangha (TMSS), Christian Commission for Development in Bangladesh (CCDB), BURO Tangail, Health Education and Economic Development (HEED), Friends in Village Development Bangladesh (FIVDB), and Nijera Kori. It also included data from Gonoshasthaya Kendra and Grameen Bank, although the latter is not technically an NGO (DFID 2000). 14 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 2.2 The range and emphasis of NGO activities 2.7 The development NGO sector as a whole undertakes a wide range of activities, including microfinance, health and education services, social safety net programs, agricultural extension, social forestry and environmental protection, safe water and sanitation, disaster management and relief, and legal and human rights education. 31 Interventions include the provision of credit, training, inputs, and support for institution building; the provision or co- provision of schools, clinics, or mobile health-related services; the training and supervision of service providers; and the design and production of teaching and training materials. Larger organizations also undertake networking, training, research, monitoring, and evaluation activities. Advocacy and policy dialogue have become increasingly prominent since the 1990s, 32 and commercial enterprises are growing in importance. 2.8 Clients are offered a strikingly Figure 2.1: Main services provided by NGOs homogenous set of services, with 92 percent of development NGOs providing microcredit. While there are many different Drinking w ater Sanitation interventions across the sector as a whole, the Health care types of services on offer to clients are mostly Child education the same (see figure 2.1). The World Bank Credit NGO survey carried out in 2003 shows that most NGOs offer four or five different 0 20 40 60 80 100 services. Credit is by far the most common NGO branches providing service (%) activity, with about 92 percent of all NGO branches providing it. The second most Source: World Bank NGO survey 2003 common area of service activity is health and sanitation, involving slightly more than half of NGO branches, followed by the provision of child schooling and provision of clean drinking water. However, NGOs do differ in their “product range” within these sectors, as the discussions on the various sectoral interventions will illustrate. The 2003 survey results also illustrate that the main services provided by NGOs are in line with the priorities of the community (table 2.1). Annex figure 2.1 illustrates the main methods by which communities are consulted. 2.9 NGOs are also closely involved with advocacy and public awareness work. In the NGO survey, 93 percent of NGOs report attempting to raise awareness of at least one issue, with the focus on sanitation, health, and nutrition, and 42 percent report having been involved in advocacy or lobbying the government over the previous year. 31 Zohir 2004. 32 Stiles 2002. 15 Economics and Governance of Nongovernmental Organizations in Bangladesh 2.3 Whom do NGOs reach and where do they operate? 2.10 Some NGOs have responded to Table 2.1 Focus group rankings of criticism about the sector’s neglect of hard-to- community priorities, May 2003 reach areas and groups. Criticisms of NGOs frequently focus on whether they reach—or even attempt to reach—marginal and very poor areas Groups that Service mentioned it as or groups. The regional imbalance of NGOs has a priority led to concerns about overlapping NGO programs and underserved pockets. 33 Access to credit 251 81% Microcredit coverage remains uneven, varying Health care 224 72% from 100 percent of households below the Safe drinking water 210 68% poverty line in Meherpur district to only 31 percent of poor households in Sunamganj. 34 The Employment generation 182 59% 2003 World Bank NGO survey also shows Non-formal child education 181 58% considerable variation: while 17 thanas have only one NGO, Mohammadpur in central Dhaka Sanitation (waste disposal) 156 50% boasts 192. The average number of NGOs per Agricultural training for thana across the country is 15. adults 96 31% Road construction/repair 87 28% 2.11 Remarkably, over the 1995–2000 period, Non-formal adult education 79 25% the proportion of rural Bangladeshi communities with at least one NGO program almost doubled Electricity system 69 22% (see figure 2.2). At the same time, the average Other 49 16% number of NGO programs within communities tripled, and coverage by BRAC, Proshika, Transportation services 46 15% 35 Caritas, and Grameen Bank more than doubled. Land rights/tenure 35 11% Coverage by other NGOs rose even more Telephones/other dramatically, but from a lower base, from 18 to communications 13 4% 48 percent of all communities. Credit, education, Source: World Bank NGO survey, 2003. and family planning services were the major areas of growth over the second half of the 1990s. Figures 2.2 and 2.3 both highlight the dominance of credit among NGO services. 2.12 New NGO activities were not necessarily located in the poorest or most deprived communities. Evidence from the 2001 census shows that the concentration of NGOs per capita was lowest in the poorest of the 35 subdistricts sampled for the World Bank NGO survey, and highest among the middle quintile. 36 Within subdistricts, NGO concentration was highest among the second-poorest quintile group, but lowest among the poorest.37 While these results are to be treated with caution, as the findings of a random sample of 35 thanas, they are consistent with other research results. Microcredit services have not necessarily been targeted to the poorest 33 World Bank 1996. 34 PKSF 2003. 35 Gauri and Fruttero 2003. ASA was excluded because it would have had to have been categorized differently in the two years. 36 Gauri and Fruttero 2003. 37 It should be kept in mind that NGO location decisions may reflect targeting of the poorest within communities, a possibility that Gauri and Fruttero were unable to assess. 16 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance communities, and the poorest may even have been systematically excluded from social welfare services. 38 2.13 Some NGOs have made special efforts to reach less-accessible groups and regions. Addressing the complex constraints of remoteness, ethnic diversity, and armed conflict has proved challenging, however. Success in extending NGO service coverage to the Chittagong Hill Tracts, for example, has required strong donor support coupled with willingness to experiment and wait for results (see box 2.2). Figure 2.2 Communities with at least one NGO program 1995–2000 Chittagong Barisal Rajshahi Khulna Dhaka Skills training Family Skills Training Education Credit Caritas Proshika BRAC Grameen All NGOs 0 20 40 60 80 100 1995 2000 percent Figure 2.3 Average number of NGO programs per community 1995–2000 Chittagong Barisal Khulna Dhaka Rajshahi Water & sanitation Family planning Skills training Education Credit Caritas Proshika BRAC Grameen All NGOs 0 1 2 3 4 1995 2000 Number of NGOs 38 Amin, Gai, and Topa 2003; Matin and Hulme 2003; Rahman and Razzaque 2000. 17 Economics and Governance of Nongovernmental Organizations in Bangladesh 2.4 A closer look at three services 2.4.1 Microcredit: Facts and issues 2.14 Bangladeshi NGOs are best known for their pioneering, large-scale provision of microfinance services, principally tiny collateral-free loans to poor women. Most MFIs are very small; as table 2.2 shows, Grameen Bank, BRAC, ASA, and Proshika between them lend to 87 percent of the 16.4 million active borrowers in the nongovernment microcredit sector. In the macro context, microcredit loans constitute around 5 percent of total private sector credit in the economy (see annex table 2.1). Table 2.2: The scale of nongovernment microfinance activity Number of active Outstanding loan Members borrowers portfolio Member savings (millions) (millions) (US$ millions) (US$ millions) Big Foura 17.9 14.2 955.9 470.6 Grameen Bank 5.2 4.7 388.9 275.5 BRAC 5.1 4.3 255.9 129.7 ASA 5.7 3.8 238.8 41.3 Proshika 1.9 1.4 72.4 24.0 Other NGO MFIsb 3.9 2.2 124.3 85.9 Total 21.8 16.4 1,080.3 556.5 Big Four as % of total 82.1 86.6 88.5 84.6 Sources: PKSF, Grameen, ASA, Proshika, BRAC, and Credit and Development Forum. Figures are for September–October 2005. 2.15 The growth of the microfinance sector took place in several stages. The origins of microfinance can be traced to action-research undertaken in the 1970s, a time when academics and activists were developing strategies for dealing with the postwar reconstruction needs of the new Bangladesh. The “Grameen model,” built around group-based credit delivery with peer monitoring, was developed by a team led by Professor Muhammad Yunus. By the 1980s, small- scale credit was recognized as an important potential means of tackling the roots of rural economic oppression, empowering poor women, and channeling resources to poor rural “target groups” in ways that would avoid elite capture. In line with this, a number of NGOs began to emphasize the delivery of services, particularly credit, and to reduce their focus on strategies of social and political mobilization. 39 2.16 Through the 1980s and early 1990s, NGOs started to develop the managerial capacity, program design, and organizational features that paved the way for expansion. NGOs opened new branches that simulated practices and organization of existing branches, a process enabled by the simple, standardized nature of the product being offered and the homogeneity and density of the population. Donors contributed to the expansion of revolving loan funds in the early 1990s; around the same time, PKSF was established to provide wholesale financial and technical support to the sector. Expansion of mainstream credit services continued apace through the mid- 39 ASA is the best-known of the NGOs that made this transition (see Rutherford 1995 for a detailed account). 18 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 1990s, while research and experience contributed to the diversification of the Box 2.2 Tackling poverty in remote regions: sector to meet the needs—including the The Integrated Development Foundation noncredit needs—of different client The diverse ethnic minority groups in remote, groups. conflict-ridden Chittagong Hill Tracts are frequently overlooked by NGOs. The Integrated Development 2.17 A number of factors contributed Foundation (IDF) made it its mission to work among these to the success of scaling up in communities, starting microcredit operations with $7,500 from the Grameen Trust in 1993. Now the largest MFI in the Bangladesh. Strong institutions were Chittagong Hill Tracts, IDF covers 45,000 families in 1,300 built on a foundation of effective villages across five districts. An evaluation in 2003 found it leadership and appropriate human had successfully generated additional employment and resource management systems, and enabled borrower households to accumulate capital and practices of experimentation and enter new occupations. institutional learning fed continuous Getting started was difficult. Few NGOs had been improvements. Constructive donor- present in the area previously, and it took more than a year client relations played a role, including and the support of the headman before initial suspicion wore appropriately timed financial and other off and the confidence of local people was gained. But the support. The presence of an enabling need for financial services was great, as interest rates on informal credit were high and the assetless poor lacked macroeconomic environment and a access even to these sources. As a result, once IDF was liberal regulatory regime allowed the established locally, people came from neighboring areas to NGOs room to grow. Population participate. density, ethnic homogeneity, and religious tolerance all contributed to the Donors recognized the extra constraints IDF faced and proved willing to support its activities over a long ease and rapidity with which new period. The Swedish International Development practices could spread, especially with Cooperation Agency (Sida), in particular, supported IDF in respect to rural women taking loans its efforts to test and expand the model over an eight-year from NGOs. Finally, the presence of a plan. IDF has diversified its activities, and in addition to professional autonomous apex body offering a range of financial service products is also active in health, agricultural, and rural livelihood activities, (PKSF) was important. 40 housing and renewable energy, disaster management, and environmental work. 2.18 Microfinance programs are reasonably successful at reaching the Source: IDF 2003. poor, though there are underserved pockets in highly poor districts. The microfinance sector in Bangladesh is the world’s largest, with around 43% of all households in the country accessing microcredit.41 However, a detailed microcredit program mapping exercise by PKSF shows that geographic coverage is varied, with coverage thinnest in the poorer, more remote, and less populous districts of the north and southeast (see figure 2.4). Conversely, the relatively prosperous central districts have a larger share of their population accessing microcredit. This is not altogether surprising. Compared to a targeted safety net program such as a public works program, microcredit is more dependent on the physical and economic endowment of the locality, which helps determine how well 40 More details on the determinants and impact of the scaling up of microcredit can be found in Zaman (2004). 41 When converting the number of clients with access to microcredit (16.4 million) to the number of households with access, we need to take into account the number of borrowers who borrow from multiple sources and the number of households with more than one individual who borrow from microfinance institutions. A recent survey suggests that around 33 percent of households meet either one or both of these criteria. Hence, we estimate that around 10.8 million households in Bangladesh have access to microcredit, which represents approximately 43 percent of the total number of households in the country. 19 Economics and Governance of Nongovernmental Organizations in Bangladesh Figure 2.4 Coverage of poor households by microfinance programs, by district Source: PKSF 2003. borrowers can put the funds to use. Hence both the demand for and the supply of microcredit are lower in the poorer areas. 2.19 A large proportion of extremely poor households, measured by initial landholding size, join microcredit programs. 42 For instance, in the most comprehensive nationwide survey of microcredit borrowers carried out in Bangladesh thus far, 60 percent of the sampled members had less than 20 decimals of land. 43 At the same time it has been estimated that certain groups of extremely poor households do not take part in microcredit programs because of a number of constraints (fewer income sources, worse health and education, demographic challenges, etc.) that dissuade them from borrowing. 44 The past five years has seen a number of new efforts to reach this group by offering smaller loans, more flexible repayment schedules, and more intensive supervision (see box 2.3). 45 42 Hussain 1998; Zaman 1999; Khandker 2003. 43 Khandker 2003. 44 Hashemi 1997. 45 PKSF began a program with 19 of its partner NGOs to pilot various approaches for reaching the very poor in 2002. By March 2004 they had a combined total of more than 24,000 very poor clients, while by July 2005 ASA reported having 105,000 very poor clients. Grameen Bank’s “beggars” program had 56,000 clients as of November 2005, with loan sizes starting as low as 500 taka. 20 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 2.20 Several studies also show that 15–30 percent of microcredit members are from “nontarget” households measured in terms of land ownership. 46 However, these are typically marginal farmers and can be considered part of the “vulnerable nonpoor” group, prone to transient bouts of poverty. 2.21 There is also evidence to suggest that households that join microcredit programs a few years after the village group has been established tend to be less poor than the members who join at the start of the program. 47 This pattern of better-off households joining over time has also been noted in many targeted antipoverty programs worldwide. 48 2.22 Another new area of expansion is microenterprise lending. This entails larger loans than are usually disbursed in mainstream microcredit, made to enterprises that extend beyond the livelihood portfolio of the household. Microenterprise lending is not always provided to the poor, although the enterprises involved may employ poor people. New flexible savings products for the poor are also emerging, as well as various forms of insurance. 2.23 A focus on financial sustainability is now the norm in the industry. As part of this report, we selected a sample of 18 NGOs ranging in size and geographic presence to assess the financial sustainability of their programs between 2001 and 2004. The analysis suggests that the sector has made impressive progress toward financial sustainability and that ASA stands out as the most cost-efficient by far. Portfolio quality has been consistent over time and remains good, partly because under PKSF rules MFIs are required to have sound approaches to provisioning for bad debts (see table 3.2). 2.24 There is now greater competition in the microcredit market. Detailed borrower-level case histories show that it is far easier to switch from one MFI to another, and waiting times for accessing credit are far shorter, than a decade ago. 49 Complementary service provision means that the larger organizations, at least, do not compete solely on the basis of price (see box 2.4). Profits are recycled, fueling growth and providing NGOs with a cushion against the risks entailed in growth. There is, however, some cross-subsidization within microfinance programs, so that, for example, lending to poorer groups and areas is subsidized by the profits of mainstream lending. 2.25 Sources of MFI funding have diversified since the mid-1990s. Dependence on donor support has declined sharply (see detailed discussion of donor trends in chapter 3). PKSF’s total contribution to revolving loan funds varies according to the size of the NGO, averaging about 45 percent for its small clients and 25 percent for its two largest clients. Members’ savings and own capital, much of it from retained profits, are increasingly important sources of financing, and their share grows with the size of the NGO. For instance, the share of savings in the revolving loan fund of very small NGOs is approximately 23 percent, while for Grameen Bank it is over 100 percent. These growing sources, together with the (still latent) potential for increasing access 46 Zaman 1998; Khandker 1998. It is interesting to compare this figure with Copestake’s (1992) evaluation of India’s Integrated Rural Development Project, where the proportion of nonpoor households ranged up to 36 percent. 47 Matin 1998. 48 Lipton 1996. 49 Personal communication with Stuart Rutherford, who is currently conducting field research in selected Grameen Bank villages. 21 Economics and Governance of Nongovernmental Organizations in Bangladesh Box 2.3 Combining financial services with social protection: The IGVGD program At least one major NGO program in Bangladesh has demonstrated that microcredit can be linked to social protection programs in order to tackle extreme poverty. Starting in 1985, BRAC worked with the government of Bangladesh and the World Food Programme to integrate a program of credit and training with a targeted feeding program for the very poorest households through the Income Generation for Vulnerable Group Development (IGVGD) program. At any time, IGVGD covers around 200,000 beneficiaries. Extremely poor households are selected by local elected officials to receive a VGD card entitling them to receive 30 kilograms of wheat per month over an 18-month period. BRAC provides training in poultry rearing and then microcredit to start an income- generating activity. Evidence suggests that the IGVGD program has been highly successful in reaching the poorest and has on average brought permanent improvements in beneficiaries’ incomes and assets. Hashemi (2001) found that on average, these improvements peaked in the period immediately after a household exited the program, when the loss of the wheat inputs had not yet been felt; three years on, however, most households showed sustained improvements in economic status. In addition, by 2000, 66 percent of IGVGD participants were moving into standard microfinance programs; of these, about two-thirds entered the BRAC credit program, but one-third joined other NGOs, highlighting an important “grooming” function of the program. to commercial funds, suggest an overall trend in MFI financing toward increasingly nonsubsidized or commercial sources of funds. 2.26 Despite its scale and prominence, the Bangladesh NGO-MFI sector remains comparatively underregulated in international perspective. Government, donors, and NGOs are increasingly focused on the need to develop a supportive regulatory framework with which to build a stronger microfinance sector that can serve the needs of more poor people. A draft of a new microfinance law is under consideration by the government. While the draft law was produced in a consultative manner, it is unclear whether certain key parts of the proposed law— such as a provision for microcredit banks—will remain in the final version that will be presented to Parliament. Chapter 4 discusses legal and regulatory issues in more depth. 2.27 The government also has a number of microcredit programs in a range of ministries. We estimate that these programs serve around 800,000 borrowers, with the Bangladesh Rural Development Board having the largest share. Interest rates are 10–12 percent flat, slightly less than the average NGO rate. At the same time, the extent to which these programs target the poor is not clear, and while they report repayment rates between 77 percent and 100 percent, it is not obvious how those rates are calculated. In general, these programs have treated microcredit as one of several inputs in a rural development project. Funding is provided from multiple sources: the regular department budgets cover some or all staff and operating costs, usually without any clear way of allocating costs between different functions and programs, while donor funding and annual development budget allocations cover the loan funds and some incremental costs. As a result, most program managers do not know the actual cost of their microfinance programs, nor can they assess the sustainability of the programs except to note that most programs obtain grant subventions from the budget on an annual basis. 22 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance Box 2.4 High interest rates but still a popular microlender: BURO, Tangail At 20 percent flat, BURO, Tangail’s interest rate is higher than that of any other MFI. That it has managed to grow by 65–67 percent (in numbers of loans and amounts disbursed) each year since 2000 testifies to its remarkable ability to retain and attract clients in an area where there are many competing MFIs. From its experimental origins in 1989, BURO, Tangail was different from other MFIs in that it was committed to offering open-access savings to its clients; it now offers one insurance, three savings, and nine credit products. In 2005 BURO, Tangail won CGAP’s Financial Transparency Award for the quality of its financial reporting systems. BURO, Tangail remains a relatively small MFI compared with Grameen, BRAC, and ASA, but it plans to scale up and is actively seeking commercial sources of financing to do so. BURO, Tangail’s experience shows that offering flexible financial services attracts clients who are prepared to pay a higher price for the additional services. Sources: BURO, Tangail 2004; Hossain 2005; Wright and Hossain 2001. 2.4.2 The impact of microcredit in Bangladesh 2.28 Microcredit has had a positive impact on several individual and household outcomes in Bangladesh, most clearly on consumption smoothing and social indicators. Analysis of a large representative survey of microcredit borrowers and an appropriate control group shows that consumption variability is 47 percent lower for eligible Grameen households and 54 percent lower for eligible BRAC households than for the control group. This consumption smoothing is driven by income smoothing, as evidenced by the significantly lower labor supply variability experienced by microcredit members compared to the control group.50 The importance of this result cannot be overemphasized, given the fact that seasonal deficits play a key part in the poverty process in Bangladesh. 2.29 Aside from microcredit’s contribution to stabilizing incomes, there is also evidence that microcredit raises the level of household income on average. A recent panel analysis of households over the 1992–98 period shows that for every 100 taka lent to a woman, household consumption (a proxy for income) increases by 8 taka. 51 Somewhat surprisingly, the impact appears to be greater for households that started off extremely poor (18 percentage point drop in extreme poverty in seven years) than for moderately poor households (8.5 percentage point drop). These results differ from earlier evidence that pointed to moderately poor borrowers benefiting more than extremely poor borrowers because the poorest have a number of constraints (fewer income sources, worse health and education, etc.) that prevent them from investing the loan in a high-return activity. 52 2.30 There has been limited work on the aggregate poverty reduction impact of microcredit at the local or national level in Bangladesh. The panel data discussed above have been used to show some positive externality from microcredit programs, such as additional employment or additional demand for products, but the overall spillover benefits appear somewhat limited. For instance, the net impact of microcredit on moderate poverty for nonborrowers is a small decline of 1.1 percentage points between 1991/92 and 1998/99, compared to a decline of 8.5 percentage points for borrowers in the same village. So while borrowers clearly benefit from microcredit, 50 Morduch 1999. The estimates of labor supply variability are 39–46 percent lower for microcredit members than for a control group. 51 Khandker 2003. 52 Wood and Sharif 1997; Zaman 1998. 23 Economics and Governance of Nongovernmental Organizations in Bangladesh nonborrowers do not gain as much. The impact on extreme poverty is estimated to be somewhat greater, 4.8 percentage points for nonborrowers and 18.2 percentage points for borrowers over this seven-year period. 2.31 The impact of microcredit on nonincome indicators in Bangladesh is also broadly positive. There is now convincing evidence, from a representative household survey that takes into account common methodological problems such as selectivity bias, that access to microcredit empowers women in several ways. Those who participate in microcredit are more able to discuss family planning issues with their spouses, take a greater role in household decision making, have greater access to financial, economic, and social resources, and enjoy greater mobility. 53 These findings are in line with other work on microcredit and female empowerment in Bangladesh. 54 There is also clear evidence that participation in microcredit groups leads to increased enrollment of children in school and better health outcomes. 55 2.32 The discussion of the impact of microcredit would be incomplete without referring to the broader package of interventions that are provided with microcredit. MFIs in Bangladesh vary significantly in terms of the noncredit services they offer, though typically these include training, business development services, and social messages on education, health, and civic rights. One carefully designed study finds that these noncredit interventions raise self-employment profits in rural Bangladesh by 125 percent, while the combined impact of credit and noncredit interventions on self-employment profits is 175 percent. 56 2.33 The discussion above has focused exclusively on benefits, without referring to costs. In fact, there has been very little cost-benefit work done in the microfinance literature. Analysis of the Bank of Agriculture and Agricultural Cooperatives in Thailand concludes that “clearly some nonzero subsidy could be justified.” 57 Cost-benefit ratios for three major microfinance programs in Bangladesh compare favorably with those of other safety net interventions, according to another study. 58 Yet another study assesses the cost-effectiveness of the Grameen Bank and concludes that “Grameen probably was a good investment.” 59 There are a number of reasons why there is such a limited amount of evidence of cost-benefit work, and why the studies cited above are somewhat tentative in their conclusions. Measuring the cost-effectiveness of microcredit is subject to numerous methodological limitations, the principal ones being that program costs vary considerably depending upon the stage of an MFI’s growth, that many benefits resulting from microcredit are difficult to measure (e.g., change in social status), and that the benefits from borrowing vary with time. The difficulty with measuring the discounted value of these benefits and the highly variable cost implications of an MFI expansion strategy mean that the type of cost-benefit work that is possible with building a road, for instance, is far more complex in microfinance. 53 Pitt, Khandker, and Cartwright 2003. 54 Amin and Pebley 1994; Hashemi, Schuler, and Riley 1996. 55 Khandker 1998. 56 McKernan 2002. Selection bias is addressed by jointly estimating program participation and profit equations, and village-level fixed effects are used to control for program placement bias. 57 Townsend and Yaron 2001. 58 Khandker 1998. 59 Schreiner 2003. 24 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 2.4.3 Health: Facts and issues 2.34 NGO health expenditure has expanded over the past decade but remains relatively small compared to private health spending. NGO spending on health and nutrition activities is 9 percent of total health expenditure and around 0.3 percent of GDP, a threefold growth between 1996 and 2002 (see table 2.3). The bulk of total spending on health services is by households on private care (65 percent in 2001–2). The ratio of government to NGO spending on health is around 3:1. Table 2.3 Health spending by NGOs compared to other sources 1996–97 1997–98 1998–99 1999–2000 2000–1 2001–2 As % of total health expenditure Ministry of Health 27.6 28.3 27.2 25.3 23.1 23.3 NGOs 2.9 3.0 4.1 6.9 8.8 9.2 Households 64.1 65.1 65.7 65.1 64.8 64.2 As % of GDP Ministry of Health 0.9 0.9 0.8 0.8 0.7 0.8 NGOs 0.1 0.1 0.1 0.2 0.3 0.3 Households 2.0 2.0 2.0 2.0 2.1 2.1 Source: Second National Health Accounts (NHA-2). 2.35 Larger NGOs are becoming increasingly dominant in health. A survey of 149 NGOs involved in health and nutrition sector activities 60 highlights the growing dominance of the five largest NGOs in the sector. Their spending almost doubled between 1999 and 2002, reaching 73 percent of total NGO health sector spending. 61 In contrast, spending by small NGOs declined in absolute terms, from Tk 819 million (16 percent of NGO health spending) in 1999 to Tk 486 million (6 percent of NGO health spending) in 2002. 2.36 The high burden on poor households of medical expenses makes a strong case for NGO provision of health services. Illness within poor families is a major cause of impoverishment and downward mobility. While NGO facility-based curative health services are a small proportion of such services in total, the community-based health delivery approach contributes to reducing inequities by improving access for poor rural women for whom distance to health facilities is a major determinant of use. Family planning tops NGO health spending, closely followed by general health and maternal and child health (see figure 2.5). Spending by NGO health facilities is more progressive than spending by public or private facilities, with the rural poor in general successfully exempted from fee payments. 60 The sample was drawn from a list of NGOs surveyed under the first National Health Accounts (NHA-1) combined with a list of NGOs involved in health and nutrition activities from the health NGO umbrella organization Voluntary Health Services Society. 61 MOHFW 2003. 25 Economics and Governance of Nongovernmental Organizations in Bangladesh 2.37 On the other hand, spending by public health facilities tends to benefit the rich— surprisingly, even more so than spending by private clinics (see table 2.4). For instance, around 32 percent of public health expenditures went to the top quintile, compared to 27 percent of expenditures by private facilities and 19 percent of those by NGO facilities. Table 2.4 Expenditures on health, by type of facility and income quintile (percent) Quintile Public health facilities Private health facilities NGO health facilities 1 14.7 16.4 20.2 2 17.2 17.2 14.1 3 15.5 18.2 22.5 4 20.6 21.1 23.8 5 32.0 27.2 19.4 Source: Second National Health Accounts (NHA-2). 2.38 NGO facility-based care is relatively micro in scale. Comparisons of NGO and public sector health service provision are difficult because of differences in scale and service delivery modes. NGOs use community health workers to provide door-to-door services. These workers provide health education on family planning, immunization schedules, prenatal care, nutrition education, tuberculosis control, basic hygiene and so forth. Community health workers also provide curative care for basic health problems and refer complicated cases to local public and private health facilities (see box 2.5 on one of the largest NGOs in the health sector, Gonoshasthaya Kendra). NGO facility-based services, on the other hand, cover only a small fraction of the population. The 2000 nationwide Household Income and Expenditure Survey found that around 4 percent of women obtained postnatal care from NGO facilities and that only 2 percent had received immunization services from NGO health centers. 2.39 The provision of water and sanitation services by NGOs is also critical to achieving health outcomes. Around 700 NGOs are active in the water and sanitation sector, with international, national, and small local NGOs commonly working in partnership. The policy context is strongly supportive of NGO involvement, in recognition of the scale of need and the shortfall in present provision, as well as of NGO capacities to stimulate behavioral change and to target and supply credit to poor rural women. The two major policies on water and sanitation, the National Safe Water Supply and Sanitation Policy (1998) and the National Water Policy (1999), both encourage and support a role for NGOs. The target-group approach and practices of beneficiary training and community-based group meetings have been vital for transmitting messages to rural populations about arsenic contamination, open defecation, the use of safe water, and hygiene. Another key intervention is the provision of microcredit for sanitary latrines and shallow tubewells. 26 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 2.40 Government-NGO collaboration Distribution of NGO Expenditures by Areas of is increasingly important in the health Healthcare (1999-2002) sector. NGO involvement in the health Immunization sector to date has been principally financed Family 7% planning by donors, although growing recognition of Training 25% the potential importance of NGO 11% involvement in the sector has led to greater Communicable diseases linkages with government. While direct 17% donor support to NGO health programs rose in absolute terms between 1999 and 2002, General health Maternal and 20% the share as a proportion of total NGO child health health funding declined from 78 to 67 20% percent. This declining share reflects the sharp growth in resources channeled through government to NGOs, which rose from 12 to 26 percent of total funding for NGO health programs between 1999 and 2002. 62 2.41 Public-private collaboration in the health sector has taken a variety of forms. An early successful case was the National Tuberculosis Program. From 1994 on, the government signed annual agreements with BRAC and the Leprosy Coordinating Committee (an umbrella organization of health NGOs) to co-provide tuberculosis curative services through rural community health workers. Cure rates of 85 percent were recorded, and the BRAC program was found to cure three patients for every two cured by the government program. The success of this collaboration is widely recognized and is attributed to the clear definition of the roles and responsibilities of the different actors. 2.42 Examples such as the National Tuberculosis Program have demonstrated the benefits of harnessing NGOs’ advantages with respect to reaching poor rural communities to achieve wider public policy goals. The government has recently indicated its intention to collaborate further with NGOs, including by developing a “strategic framework” for NGO contracting. 63 These new directions in public policy mark the beginnings of a process of separating the financing and provision of health services. A larger role for NGOs in service provision thus has the potential to improve the “long route” to accountability to poor service-users, through the separation of policy-making and provisioning functions. Chapter 3 discusses the experience and future directions with NGO contracting in greater detail. 2.4.4 The impact of NGO health programs 2.43 While no data are available with which to directly compare user satisfaction levels in government and NGO services, there is some evidence regarding overall satisfaction with each. 62 The share of NGO funding attributed to transfers between NGOs averaged around 9 percent during this time (MOHFW 2003). 63 Prominent programs involving contracts with NGOs include the NGO Service Delivery Program, through which 41 NGOs deliver clinical services to 22 million people across 61 districts, including communities not reached by government services. The Bangladesh Population and Health Consortium, supported by the UK Department for International Development, supports NGO community work in training, management, and management information services. A third example is the Urban Primary Health Care Project, supported mainly by the Asian Development Bank, in which responsibility for primary health care has been given to the Ministry of Local Government and Rural Development and urban primary health care centers have been constructed in 16 partnership areas. 27 Economics and Governance of Nongovernmental Organizations in Bangladesh User satisfaction with government services appears to have declined since 2000, with government services having received a rating of “good” from 38 percent of respondents in 1999 but only 10 percent in 2003; meanwhile, “bad” ratings rose over the same period from 38 to 45 percent. 64 A different study based on an exit poll of 817 users of NGO facilities found high levels of approval regarding a number of aspects of the service: more than 90 percent of clients said that the provider had spent enough time and asked questions, and the same proportion said they were willing to recommend the facility to others. 65 Still, NGO services have their share of problems and inadequacies; for example, half of the NGO centers did not have a full complement of basic supplies or equipment. 2.44 Multivariate analysis using the data from the Child Nutrition Survey 2000 shows that the presence of BRAC or Grameen Bank in a village can have a significant effect on malnutrition. Results show that the presence of one of these organizations can lead to a decline of about 20 percent in moderate and severe malnutrition among the poor, 66 controlling for a set of other household and community factors. Explanatory factors include the use of microcredit for consumption-smoothing purposes, the impact of nutritional education during weekly group meetings, and the impact of nutritional supplements and education directly provided by health workers. 2.45 Other credible evidence points in a similar direction. Panel data suggest that between 1992 and 1995, severe malnutrition (as measured by mid-upper-arm circumference) declined significantly for children of BRAC members compared to a control group of nonmembers. 67 Similarly, the survival probability of children belonging to BRAC households is better than that of children from poor nonmember households, primarily because of mortality differences in the immediate postnatal months. The same study also shows that BRAC members had a significantly higher probability (0.57) of using family planning compared to poor nonmembers (0.50), and also compared to wealthy households in the village (0.51). 2.46 Our earlier analysis refers to the impressive partnership between government and NGOs in controlling the spread of tuberculosis (TB). A comprehensive evaluation shows that the prevalence rates of tuberculosis cases in the program intervention subdistricts was half that in the nonprogram comparison subdistricts. 68 One of the most important facets of tuberculosis control is detection, and NGO community health workers carrying out door-to-door identification of potential TB cases played a key role in the effective campaign against the disease. Moreover, given the social stigma associated with TB, there is typically low attendance at government health facilities, and so the doorstep counseling and treatment services provided by community health workers are a crucial part of the strategy. Other important factors cited are the regular availability of free drugs through the community health workers, the availability of sustained 64 The proportion of respondents who rated private and NGO services as “good” rose in the same period from 25 to 37 percent. However, NGO services were not distinguished from private services, and as private for-profit providers are overwhelmingly dominant in the health service market, this figure cannot be treated as evidence of rising approval ratings for NGO health services. 65 AHI 2004. 66 Deolalikar 2004. 67 Chowdhury and Bhuiya 2001. 68 Chowdhury et al. 1998. 28 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance Box 2.5 The Gonoshasthaya Kendra story: Cost-effective health care for the poor Gonoshasthaya Kendra (GK) started in 1971 when a group of young Bangladeshi doctors working in the United Kingdom returned home to provide health care for freedom fighters during the liberation war. After independence, the team moved to implement their dream of providing affordable health care for rural Bangladeshis. Their approach was to cover a selected thana (subdistrict) through a health pyramid with a hospital at the apex; a second tier of health centers in each union, run by paramedics and visited on a regular basis by doctors from the apex hospital; and a foundation tier of village-based community health and family planning workers. The impact of this approach is reflected in the table below, though clearly the figures do not control for potential program placement bias. The idea was to use the health pyramid effectively and to reduce unit costs of health care to a level that could be sustained by the local population. Every household would pay a monthly insurance fee for full coverage, plus a small co-payment of 2 taka (3 cents) for each visit. Today the system covers about 1 million people in rural areas and another 100,000 in urban areas. One original idea that had to be modified was the use of rural health insurance premiums to cover the full cost of care. Based on assessments of the ability to pay, it was determined that in rural areas about 60 percent can be covered by such payments. The remainder has been subsidized by GK through its revenue-earning operations such as GK Pharmaceuticals. GK Pharmaceuticals started as an effort to reduce the high cost of medicines in Bangladesh by establishing a facility to manufacture generic drugs. While helping to reduce the cost of drugs, the pharmaceuticals venture has at the same time been a profit maker, and the profits have been used to finance the deficit in health care operations. The high cost of medicines and the sale of less-than-useful, high-profit prophylactic drugs by manufacturers led to another effort spearheaded by GK, which became the global precursor for the essential drugs policy. The successful efforts in Bangladesh led the World Health Organization and others to successfully promote this in other parts of the world. GK has also spearheaded the anti-smoking movement in Bangladesh and has campaigned for social equality for all. GK’s recruitment policy is that at least 60 percent of personnel at all levels should be women. Vital events rates in Gonoshasthaya Kendra areas compared with national averages, May 2004–March 2005 GK area National average Crude death rate (per 1,000) 3 9 Neonatal mortality (per 1,000 live births) 18 41 Post-neonatal mortality (per 1,000 live births) 11 24 Infant mortality (per 1,000 live births) 28 65 Maternal mortality ratio (per 100,000 live births) 153 320 donor support for this program, and the technical support provided by the government to the NGO intervention. 2.4.5 Education: Facts and issues 2.47 NGO education services are targeted to the poor. Around 46 percent of NGO branches in the 2003 World Bank NGO survey provide child education services, with a further 25 percent involved in adult education. 69 As many as 1.5 million children, or around 8 percent of all currently enrolled primary students, are in NGO primary schools. Of these, 1.2 million alone are enrolled in BRAC’s network of around 35,000 non-formal primary schools. As figure 2.6 69 Gauri and Galef 2004. 29 Economics and Governance of Nongovernmental Organizations in Bangladesh illustrates, the majority of students in NGO schools are from the bottom two socioeconomic quintiles, a much higher proportion of poor students than in the government and other schools. Figure 2.6: Distribution of primary students by facility and quintile 120 Fifth quintile % of primary students 1 100 18 15 16 Fourth quintile 28 80 14 49 25 27 Third quintile 60 25 27 23 23 19 Second quintile 40 20 20 44 14 12 20 17 First quintile 10 14 19 10 10 0 Government Government- NGO-operated Madrassa Private and subsidized Others Source: Authors’ estimates from Household Income and Expenditure Survey 2000. 2.48 The share of nonsalary inputs in NGO primary school expenditure is far larger than in government schools. Unit costs in NGO primary schools, at around $17 per student per year, are slightly lower than in government primary schools. Notably, the composition of expenditures in NGO schools is sharply different. As table 2.5 illustrates, NGO schools spend only about half of their recurrent expenditures on teacher salaries, while the equivalent figure for government schools is 95 percent. This is largely due to the fact that NGO primary school teachers are less qualified and are willing to accept a far lower salary scale. 70 Table 2.5 Composition of education expenditure, by type of school (percent) Total expenditure Recurrent expenditure Expenditure item Govt. Private Madrassa Non- Govt. Private Madrassa Non- formal formal Salaries (teachers and others) 81.8 61.3 81.5 49.6 94.6 87.4 93.9 51.1 Physical facilities 12.4 27.5 11.8 0.9 — — — — Materials/stationery 1.4 3.2 1.8 30.9 1.6 4.6 2.1 31.8 Rent 0.5 0.3 0.5 15.3 0.6 0.4 0.6 15.8 Fixed assets 1.1 2.4 1.4 2.0 — — — — Other expenditure 2.8 5.3 3.0 1.3 3.2 7.6 3.4 1.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: CAMPE 2001. 70 NGO primary school teachers are paid around Tk 1,200 per month (World Bank NGO survey, 2003). Teachers in government schools earn around Tk 2,700 per month, plus benefits and allowances that amount to about 75 percent of their basic salaries. Moreover, government schools generally have teachers with more years of service (an average of 19 years against four years in NGO schools). 30 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 2.49 Among NGOs involved in education, a few large NGOs provide nationwide education programs and services such as teacher training and materials development. At the same time, a large number of very small NGOs operate mainly at the subdistrict level, providing education directly and relying heavily on the larger NGOs for financing and technical support. In between, there is a group of medium-size NGOs that both provide education directly and provide educational services to local-level NGOs, and that have a significant regional presence. 71 Annex table 2.2 shows estimates of the numbers of NGO educational centers and learners under NGO programs. 2.50 NGO education programs in Bangladesh remain best known for their non-formal primary education programs. These are widely understood to be high-quality and cost- effective interventions that reach the poor, and in particular poor girls. Box 2.6 describes the BRAC Non Formal Primary Education (NFPE) program, by far the largest NGO education program in Bangladesh. 2.51 Coordination between government and NGO education programs in primary education remains weak. NGO schools are not formally recognized in official statistics on primary education. Since the closure of the Department of Non Formal Education, there has been no mechanism to coordinate government-NGO activities. CAMPE, the national network of education NGOs, and the government have attempted to coordinate the activities of NGOs involved in education. 72 At the same time, individual large NGOs such as BRAC and Friends in Village Development Bangladesh (FIVDB) are involved in contracting and coordinating the activities of their small partner NGOs. 2.52 NGO education programs depend largely on donor financing. Donor financing has been relatively stable over the last five years and remains vital to NGO education programs (see annex table 2.3). However, dependence on external financing leads to concerns about sustainability. The bar on funding Proshika has led to the closure of most of its schools (see box 2.8), and the Gono Shahajjo Sangstha (GSS) education program has resumed on a limited scale following the withdrawal of donor funding (see box 4.1). Student fees and cost recovery are rarely feasible given that non-formal education programs serve poorer populations and community contributions are relatively small. The most realistic option for NGOs to reduce their reliance on donor funds is by forging stronger relationships with the government. However, problems with government arrangements for contracting NGOs need to be addressed as a first step (see chapter 3 for more details). 2.53 There has been a strategic shift into pre-primary and secondary education by the dominant NGO in education, which has led to greater collaboration with government. The number of NGO primary schools has changed little over the past decade, and there appears to be an increasing focus on other areas. The growing number of BRAC pre-primary schools is designed to meet needs in an area where the state has little presence and where NGO schools are seen as playing an important role. Children passing out of BRAC pre-primary schools are 71 The large national NGOs are essentially BRAC, Proshika, Dhaka Ahsania Mission (DAM), and FIVDB. Smaller or regional NGOs include Jagoroni Chakra, Bangladesh Development Society, Community Development Centre (CODEC), Uttaran, Underprivileged Children’s Educational Program (UCEP), and Concern. 72 Approximately 700 NGOs are affiliated to the Campaign for Popular Education (CAMPE), the umbrella organization for education NGOs set up in 1991. 31 Economics and Governance of Nongovernmental Organizations in Bangladesh admitted into government primary schools. In secondary education the Ministry of Education has given permission to certain NGOs to develop the capacity of secondary school teachers, school management committees, and communities. This role as a facilitator of better-quality education is likely to yield greater results for larger numbers of people than if NGOs focused on becoming direct providers of postprimary education. 2.4.6 The impact of NGO education programs 2.54 Efficiency indicators in NGO schools are noteworthy. Attendance in NGO schools was estimated at 85 percent, compared to 55 percent in government schools. The completion rate for NGO schools tends to be higher than the average of 68 percent in formal schools, with BRAC at 93 percent, Proshika at 90 percent, and FIVDB at 85 percent. However, while the transition rate from NGO primary schools to government secondary schools is impressive, dropout rates in Box 2.6 BRAC’s Non-formal Primary Education Program BRAC’s Non-formal Primary Education (NFPE) program is the biggest such program run by an NGO, with 34,753 centers nationwide and 1.2 million learners presently enrolled. BRAC schools are found in more than 450 thanas across the country, and an additional 200 small NGOs are contracted to run another 2,250 schools on the BRAC model through its Educational Support Programme. Since 1985, BRAC’s NFPE program has focused on bringing non-formal primary education to the children of the poor, targeting girls in particular. The curriculum is learner-centered and includes Bangla, mathematics, and social studies, with English from class 2. In 1998 the improved program was expanded to four years, covering the primary curriculum for classes 1–5. Government textbooks are used in classes 4 and 5. Instruction is given for two and a half hours a day, six days a week, amounting to approximately 268 days a year. The quality of the teaching-learning materials and the emphasis on teacher training contribute to better educational outcomes in BRAC schools than in other types of mass schooling in Bangladesh. Teachers are trained for 12 days initially at a residential BRAC training center, followed by monthly refresher sessions. The 30:1 pupil- teacher ratio in BRAC schools compares favorably with class size in government schools and contributes to effective teaching. The per-student cost of BRAC schooling is about $17, compared with around $20 in government primary schools. Teacher unit costs are much lower than in formal primary schools, and almost 30 percent of the NFPE budget is allocated to effective management and supervision of schools. Along with other NGO non-formal education programs, BRAC’s NFPE is credited with having helped improve the educational prospects of poor girls in Bangladesh, contributing to closing the gender gap at primary level. At least 70 percent of the students must be girls from the poorer households in the community. But girls are also encouraged to enroll by the predominance of women teachers (97 percent of NFPE teachers are local married women) and the presence of parent-teacher associations. The fact that schools are local minimizes the time and trouble of travel, and timetables are set to allow for the domestic responsibilities shouldered by school-age girls. The system has also provided access to formal schooling for a large number of girls from poor families, as around 90 percent of the approximately 2.6 million BRAC school graduates are believed to have entered the formal system subsequently. The NFPE program was renamed BRAC Primary School (BPS) in 2004. Sources: BRAC 2003; Nath 2002; Sukontamarn 2004. the first year of secondary remain high. 2.55 Exposure to NGO schools has a greater impact on female enrollment relative to government schools. There is clear evidence that NGOs set up schools in villages where a high proportion of children have no schooling, and where adults typically have lower-than-average 32 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance schooling. 73 After controlling for a range of individual and household-level characteristics and village-level fixed effects, exposure to an NGO school in a village has a greater impact on girls’ enrollment than on boys’, while there is no gender difference for government schools. This impact is greater for girls from poor households in rural areas than for those from poor households in urban areas. 2.56 The impact of NGO schools on test scores is highly significant. Test score data from the nationally representative Assessment of Basic Competency survey of 3,360 children carried out in 1998 have been used to assess the quality of education in NGO schools. 74 The results show that after controlling for a range of factors that could influence test scores, attending an NGO school increases both the reading and writing test scores significantly more than attending a government school. Reading skills are 22 percent higher for girls attending NGO schools that for girls who have never been to school; girls attending government schools gained only 8 percent. The corresponding gains for writing skills are 31 percent for girls in NGO schools and 10 percent for girls in government schools. The NGO school “premium” for boys is slightly narrower, but still very striking for both reading and writing skills. 75 2.57 While NGO school students on Table 2.6 Literacy rate by type of school (percent) average achieve better educational outcomes than students in other types of Literacy rate Literacy rate schools, achievements on basic School type (classes 4–5) (overall primary) competencies in all types of schools are Government 62 46 still relatively low. In the 1998 survey, only 38 percent of NGO school students Private 63 33 satisfied the basic education criteria, Non-formal compared with 21 percent in (NGO school) 97 60 government schools. However, more Madrassa 65 54 recent evidence suggests that NGO All 65 45 school students score well in both Source: CAMPE 2003. absolute and relative terms on literacy measures (see table 2.6) 2.5 NGOs and advocacy issues 2.58 NGO advocacy activities have intensified since the return to democracy in the 1990s, presenting a more direct—but still cautious—challenge to successive governments. NGO involvement in advocacy work in the 1970s and 1980s was sporadic and mainly limited to the “demonstration effect,” that is, instances when NGOs’ own activities highlighted the need for policy shifts. For example, Proshika’s program of providing livestock to members was suffering from the public sector monopoly on livestock vaccination in the 1970s. Proshika advocated the training and licensing of its group members as “para-vets,” which enabled them to break the monopoly, extend coverage, and help create new livelihoods for members. 73 Sukontamarn 2004. 74 Nath, Sylva, and Grimes 1999; Sukontamarn 2004. 75 Sukontamarn 2004. 33 Economics and Governance of Nongovernmental Organizations in Bangladesh 2.59 By the late 1980s, a number of now-large NGOs began to perceive their service delivery goals as conflicting with their social development strategies. On the one hand, episodes of social conflict and clashes with local representatives of the state endangered programs and loan recovery; on the other, competition between NGO microfinance institutions and pressure from donors to show results led to reduced emphasis on conscientization and group mobilization. The late 1980s saw the beginning of the use of public interest litigation, now an important advocacy tool, to address a growing range of concerns—civil liberties as well as environmental concerns and social and economic rights. Box 2.7 A selection of NGOs involved in advocacy in Bangladesh The Fair Election Monitoring Alliance (FEMA), established in 1995, is an alliance of 180 NGOs and civil society groups involved in electoral monitoring and voter registration. FEMA’s main activity is monitoring elections at the parliamentary, union parishad, and municipal levels. This involves deploying thousands of monitors in polling booths on election days, as well as long-term activities such as voter registration and postelection dispute resolution. Transparency International Bangladesh (TIB) is the Bangladesh chapter of Transparency International, an international NGO that since 1996 has been exposing official corruption, advocating administrative reform to improve transparency, and conducting research on related issues. TIB is engaged in grassroots mobilization through citizen committees that work to generate demand for good governance at the local level. Bangladesh Legal Aid and Services Trust (BLAST) began in 1993 by providing legal aid services to the poor, and has moved on to legal literacy campaigns and networking of NGOs involved in good governance. A main activity is public interest litigation, which it has filed in diverse areas: employment discrimination; public examinations; illegal land acquisition, construction, and property development; treatment of salt to prevent iodine deficiency disease; slum evictions; and abuse of power by the police and other executive agencies. Ain o Shalish Kendra (ASK) also began offering legal aid services in 1986 and has since moved on to dispute resolution activities, legal/rights awareness education, research, networking, and lobbying. With BLAST, ASK has been involved in litigation on slum evictions, torture, illegal detention and arrest, gender discrimination in state and local government institutions, legal rights of children in custody, and factory working conditions. Starting in 1976, Samata has focused on supporting poor people’s rights to khas land (government-owned land that, according to law, is available to the landless poor). Activities include mass gatherings and rallies of the landless to press demands on local government officials, research and dissemination activities around issues relating to landlessness and land rights, and advocacy with national and local level government officials. Through its Institute for Development Policy Analysis and Advocacy set up in 1994, Proshika attempts to influence national policy, primarily through research and publications and its program of advocacy training and capacity building. Proshika has in the past been involved in highly public and confrontational advocacy activity, such as campaigns against harmful pesticides and a controversial voter education program in the mid-1990s. Although BRAC only recently established its own Advocacy and Human Rights Unit, in 2002, it has been involved in advocacy since the 1980s, primarily through “proof of concept” and collaboration with public agencies. BRAC’s main advocacy activity remains oriented to social communication rather than directed at national policy agendas. Rangpur Dinajpur Rural Service (RDRS) is a multisectoral NGO that focuses its advocacy effort on pro- poor rights. RDRS has carried out media campaigning and workshops with local elites to promote women’s rights and gender equality. It also provides legal aid for the poor and supports group members in accessing local resources. 34 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 2.60 The emergence of a multiparty electoral democracy made space for the scaling up of NGO advocacy work in the 1990s. Good governance became a leading focus of NGO advocacy: a donor-coordinated initiative, the Development Partnership, was established to address local governance. The Fair Election Monitoring Alliance (FEMA) was set up as an independent election watchdog in 1995, and 1996 saw the establishment of the Bangladesh chapter of Transparency International, focused on anticorruption activities. Proshika was the earliest of the large NGOs to institutionalize its advocacy activities, with the establishment of the Institute for Development Policy Analysis and Advocacy in 1994. It was not until 2002 that BRAC followed suit with its own separate advocacy unit. 2.61 Most NGO branches participate in public education campaigns designed to promote mass awareness of issues such as sanitation, nutrition, rights, and gender (93 percent of NGO branches, according to the World Bank NGO survey). In addition, 42 percent report lobbying national or local authorities. Of these, the majority reported having met with the authorities in order to lobby for financial support for their own or other local NGOs, with only 20 percent reporting that their lobbying goals included policy change. This suggests that NGO advocacy— in the sense of lobbying for policy change—is carried out by selected NGOs and mainly at the headquarters level. NGO advocacy activities are increasingly prominent, institutionalized, and coordinated in the areas of good governance, state compliance with constitutionally guaranteed laws and universal human rights, and developmental rights. Box 2.7 summarizes the advocacy strategies and practices of seven organizations that can be taken as representative of NGO advocacy activity. 2.62 The scope and nature of NGO advocacy work is influenced by (a) official control over NGOs through formal registration requirements and other legislation, (b) perceptions of the fine line between party political activism and advocacy from within civil society, and (c) an NGO’s ability to protect its political space through support from donors, NGO champions, and influential individuals within government, and through media activism. 2.63 Unlike other areas of NGO activity, advocacy often cuts across sectoral spaces, and where it involves challenges to government it has the potential to conflict with NGO service delivery goals. The conflict arises because NGOs need government permission for their service delivery activities. As a result, many NGOs were hesitant to involve themselves in pro- democracy and related campaigns through the 1980s and beyond, and this in turn caused them to be treated with mistrust by other parts of civil society. There remains lingering skepticism about the scope for NGOs to combine effective pro-poor advocacy with service delivery. 2.64 Understanding the role and impact of NGO advocacy requires an understanding of the political space in which they operate, and the challenges and constraints they face. Where the “political” line is drawn is a matter of continuous negotiation and conflict between NGOs, other civil society actors, political parties, and the state. This has been demonstrated most clearly since 2001 in the form of official actions to deregister or halt the activities of five NGOs (including Proshika) for their allegedly partisan bias during voter education campaigns in the run-up to the 1996 election (see box 2.8). 35 Economics and Governance of Nongovernmental Organizations in Bangladesh Box 2.8 The controversy over Proshika One of the largest NGOs in Bangladesh, Proshika has scaled back its service delivery programs significantly following a government clampdown on donor funding of Proshika in 2002. Accusations against Proshika range from serious financial irregularities, uncovered by a government audit, to sedition. On the issue of fraud and misuse of resources, the donors who had been instrumental in Proshika’s rapid growth in the 1990s rallied behind the organization and questioned the validity of the audit report, contrasting it to the clean bill of health given to Proshika by the international auditing firm PricewaterhouseCoopers in successive annual audits. Meanwhile, the coalition government led by the Bangladesh Nationalist Party (BNP) accused Proshika officials of plotting with the main opposition party, the Awami League, to overthrow the government. Specifically, Proshika officials are accused of encouraging thousands of group members across the country to assemble in the capital to launch an opposition platform. The time frame of the alleged plot coincided with the April 30, 2004 deadline the Awami League had issued for dislodging the BNP-led government. Proshika’s president was arrested along with a number of other employees in May 2004. As of early 2006, he was out on bail as the legal processes continued. Proshika has been treading a fine line for about a decade. In the mid-1990s the Awami League launched a movement for a neutral caretaker government and Proshika played a prominent role in this campaign, drawing criticism from within and outside the NGO sector for its involvement in a highly sensitive political issue. During the 1996 elections Proshika was heavily engaged in voter education programs along with other NGOs. The material in Proshika’s program, and that of five other NGOs, was considered to be pro-Awami League by the BNP, which lost that election. The same perception also prevailed in the 2001 election that the BNP won. At the same time the president of Proshika also headed the NGO apex membership body, ADAB, and the BNP government’s misgivings about Proshika threatened to spill over to the rest of the sector. Draft legislation was prepared to strengthen controls over NGOs, but the legislation was strongly opposed by donors and NGOs and has not yet been discussed in Parliament (see chapter 4). A separate NGO apex body, the Federation of NGOs in Bangladesh, has been formed by NGOs who seek to distance themselves from Proshika’s perceived involvement in politics. Ultimately, whatever the merits of the case for or against Proshika, it is the poor who have lost out. The lack of funding has resulted in a sharp squeeze on Proshika’s education programs, which at one point were second only to BRAC’s in size. In 2000–1, Proshika operated 10,100 non-formal primary schools and 11,586 adult literacy centers. By 2004–5, the number of primary schools had dropped to 450 and the adult literacy program had closed. The number of microcredit borrowers who received loans in 2004–5 is around 40 percent of the 2000–1 number. The social forestry program, one of Proshika’s flagship programs, at one time had 2,419 part-time caretakers, but in 2004–5 there were only 54 left. Looking forward, this story has lessons for government, donors, and NGOs, as the discussion on the “strategic compact” in chapter 5 will further develop. NGOs ought to refrain from engaging in any form of partisan political activity and should also be aware that advocacy campaigns can jeopardize their service delivery activities. Donors need to retain institutional memory and further develop the capacity to properly monitor NGO activities and financing. Finally, the government ought to develop mechanisms that to the extent possible protect service delivery to the poor while legal action against an NGO, or selected NGO staff, is in progress. 2.65 Given the broad range of issues addressed by NGO advocacy, quantifying its impact is difficult. Nevertheless, the following section attempts to assess the impact of advocacy in bringing about electoral reforms, curbing corruption, protecting civil liberties, and strengthening the rights of the poor, particularly women. 2.66 FEMA has regularly campaigned for a range of electoral reforms. Several of its recommendations have been implemented in full, including allowing poll observers and monitors in polling booths and raising the security money for candidates to Tk 1 million. Those that were 36 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance only partially heeded include empowering the Election Commission to stop polling in a particular constituency. 2.67 Transparency International focuses on advocacy related to anticorruption and is best known for its country corruption index. The ranking of countries draws considerable attention and Bangladesh’s ranking in particular has been the subject of much debate and controversy. Less well known is Transparency International Bangladesh’s analysis and advocacy regarding specific ministries and themes. For instance, the police are allowing TIB to monitor the governance of selected police stations. TIB’s report on corruption in Chittagong ports was noted by a parliamentary standing committee, resulting in an internal investigation by the Chittagong Port Authority. An initiative to introduce citizen report cards in local-level service points (schools, health clinics, local administration offices) is putting the spotlight on key issues such as absenteeism and resource leakages, though its tangible impact on the status quo has thus far been limited. 2.68 We examined advocacy on civil liberty issues that has resulted in public interest litigation. In the vast majority of cases, it appears that the court withheld a final judgment, issuing various forms of interim rulings such as stay orders. However, there are occasional examples where public interest litigation has led to proactive policy changes. For instance, five cases contesting the practice of holding female abuse victims in police custody resulted in two amendments to legislation, so that the female victim’s consent is now required before she is placed in police custody. The Bangladesh Environmental Lawyers Association has also had some success in protecting the environment through public interest litigation. 2.69 The largest NGOs involved in advocacy issues regarding women’s rights are the Mahila Parishad and Naripokkho. Together with other NGOs and civil society groups, they have carried out sustained advocacy and awareness-raising campaigns that have led to legislation on violence against women (specifically, acid throwing) and the setting up of women’s courts. 2.70 While the above examples point to areas where advocacy has had some positive effects, another example related to rights of the landless illustrates how slow implementation of reforms can limit the impact of advocacy. Samata, whose main goal is improving landless people’s access to idle government land (khas land), has been effective in lobbying for amendments to the Vested Property Act to allow the landless access to such properties. However, the implementation of these amendments is proving slow because of deficiencies in land records, compounded by lethargy in the local Khas Land Distribution Committees. 2.71 While the impact of advocacy for pro-poor rights is hard to measure, it is critical to set targets for various outputs associated with these activities and make the targets publicly available. As table 2.7 indicates, in 2004 the Rangpur Dinajpur Rural Service met or exceeded most targets with respect to a range of goals set as part of its strategy to raise awareness of various social and legal issues facing the poor, particularly women. 37 Economics and Governance of Nongovernmental Organizations in Bangladesh Table 2.7 Advocacy and awareness raising: RDRS’s progress in 2004 Indicator Annual target Achieved Women’s rights protection and promotion cells functioning (at federations) 260 257 Federation leaders trained in mediation skills and village court 225 245 Group members had legal education 93,600 93,954 Adolescents trained in gender awareness and reproductive health 750 687 Pregnant mothers given orientation on family laws and birth registration 2,340 30,818 Mediations organized/handled at federation level 520 1,739 Teachers trained in human rights and family law 200 175 Popular theaters organized on gender issues 161 167 Source: RDRS Annual Report 2004. 2.6 Cross-cutting themes 2.72 This chapter focused on three key services provided by NGOs—microfinance, health, and education—and on NGO advocacy activities. The empirical evidence presented here shows that these services are by and large successfully targeted to poor households, though there is some room for improvement, particularly in terms of the geographic concentration of NGO activity. The impact of these services is also clearly positive, and as box 2.9 illustrates, is valued by the communities that the NGOs serve. 2.73 However, while the impact of NGOs on their clients has been largely positive, the wider “macro” impact has been limited by the fact that the services are still relatively “micro” in certain aspects. For instance, while microcredit has a macro reach, which has resulted in important benefits for borrowers, the micro loan size has limited the extent of the employment that such credit has generated. Similarly, while non-formal primary education programs have contributed to reducing gender disparities in access to schooling, and students in non-formal schools have better educational outcomes than those in government schools, the impact on national educational quality is limited by the fact that only 8 percent of primary school students are in NGO schools. The Gonoshasthaya Kendra experience shows that in areas where NGO health interventions are intensive, there have been clear improvements in outcomes. Nevertheless, quantifying the NGOs’ contribution to rising national health indicators is difficult, particularly as a significant part of NGO work centers on raising awareness of health issues— such as treatment for diarrhea, sanitary practices, and family planning—and on facilitating linkages with the public sector in efforts such as the tuberculosis program. All of these are hard to measure. Moreover, tackling the emerging priorities in the health sector, such as curbing neonatal mortality, will require more facility-based care, an area where NGO activity is still relatively micro. Hence, a key issue that will be explored in the final chapter concerns the mechanisms by which NGOs can scale up further, or shift directions, to enhance their impact. 38 Chapter 2: Key Activities and Their Impact: An Assessment of NGO Performance 2.74 One issue that is central to the discussion in this chapter is the joint and separate provision of NGO programs. There are NGOs that focus on one core activity—for example, Transparency International focuses entirely on anticorruption advocacy, and ASA provides one service, microfinance. Most NGOs, though, are multipurpose. Our overall support for joint provision of social services by NGOs, such as health and education, stems from the economies of scale in senior management oversight when both functions are provided by one NGO. The important part is to keep accounts for different programs separate and all forms of cross- subsidies transparent. In the case of microfinance, there may be greater opportunities in the future for a few large NGOs to spin off their microcredit operations into a specialized financial entity such as microcredit banks. However, for the majority of NGOs, joint provision of the various services is the most practical arrangement, given the synergies between programs. Box 2.9: What affects the community perception of NGO services? A nationally representative survey of 310 NGO branches was conducted by the World Bank from March to May 2003. Survey enumerators applied a questionnaire to NGO branch managers and they also conducted, in the vicinity of each NGO, a structured group interview with randomly selected community members. The structured group interviews asked the community members to collectively ascribe performance ratings to the local NGO branch. Most of the groups interviewed reported substantial satisfaction with the work of the NGOs and NGO branches. Some 92 percent of the groups were satisfied with the performance of their NGO, 72 percent felt the NGO was quick to respond, and 81 percent felt that the NGO representatives were good at what they did. The interview participants were also asked, without prompting, to name the priority needs in their communities. Access to credit, health care, and safe drinking water were rated as the top priority needs. The groups were asked to participate in an imaginary willingness-to-pay exercise for NGO services, to rate NGO performance along eight dimensions (which were aggregated into a final score), and to evaluate the altruism of NGOs. The data were aggregated, and various NGO attributes were included as right-hand-side variables in regressions that attempted to identify aspects of NGOs related to high levels of community satisfaction. The regressions included, as controls, observable characteristics of the focus groups thought to be related to high ratings, including the percentage of focus group members who were female, NGO employees, NGO members, and union parishad members. The sector of work had no effect on NGO performance scores. Branches of big NGOs (BRAC, ASA, Proshika, and Caritas) received ratings similar to those of other NGOs in the country. Older and established NGO branches were seen to be performing better than more recently established branches, perhaps because older NGOs have had time to develop relationships with community members and build satisfactory service provision programs. The average salary of professionals in the NGO was positively related to all three measures of community satisfaction. 2.75 One area that is less clear-cut is whether an NGO that provides services can also engage in advocacy work. The earlier discussion illustrated the broad range of advocacy activities and the extent of emphasis various organizations place on advocacy. It is clear that multipurpose organizations such as BRAC and RDRS focus on providing services, and their engagement in advocacy is relatively low-key and deliberately nonconfrontational. RDRS strives to involve local government officials, elites, religious leaders, and the media on issues that are clearly relevant to the poor, so its advocacy efforts are seen as part of its overall poverty reduction mandate. Particularly where NGOs receive government financing or are in other types of 39 Economics and Governance of Nongovernmental Organizations in Bangladesh partnerships with government in delivering services, it is likely that multipurpose NGOs will be compelled to limit their advocacy activities to issues that do not directly challenge the state. 2.76 In general, separating these functions and having specialized organizations—those that deliver services and those that engage in advocacy—makes more sense than having organizations that mix the two. However, we also recognize that there are organizations that have combined the two successfully by ensuring that their advocacy work does not “overstep the mark” and remains focused on issues that are important for the poor, such as legal awareness and violence against women. We now move to examine the sources of finance for NGOs, and as part of this will assess how potential shifts in these sources may affect the debate on the trade-offs between service delivery and advocacy. 40 Chapter 3: Financing NGOs: Trends and Prospects 3.1 This chapter examines issues surrounding various ways of financing NGO activity, highlighting trends over time and their implications. The chapter first looks at sources of funds that are external to the NGO, including direct donor financing and financing through government contracts and private charitable contributions. It then examines issues related to self-generated sources of funds, focusing on interest income from microfinance and revenues from commercial enterprises. 3.1 External financing 3.2 The share of aid to NGOs as a portion of total aid to Bangladesh has risen from 14.4 percent in the first half of the 1990s to 24.6 percent since then. We estimate that total aid to NGOs rose from an average of $233 million (0.7 percent of GDP) in 1990–95 to $343 million (0.7 percent of GDP) in 1996–2005. Meanwhile, total aid to Bangladesh fell from $1.62 billion (4.9 percent of GDP) to $1.39 billion (2.9 percent of GDP) during this period (see table 3.1). Table 3.1 Aid flows to NGOs (US$ millions except where specified) Official development assistance to central Share of Estimated Share of government NGO funding to total aid to Total grant grants in NGOs Total NGOs in o/w o/w Grants aid to total from loan aid to total aid to Period Total Grant Loan to NGOs Bangladesh grant aid sources NGOs Bangladesh 1990–91 1,716 831 885 106.6 938 11.4 73.4 180.0 10.5 1991–92 1,574 817 756 121.6 939 13.0 71.8 193.4 12.3 1992–93 1,607 818 789 195.7 1,014 19.3 71.7 267.4 16.6 1993–94 1,530 710 820 171.0 881 19.4 67.9 238.9 15.6 1994–95 1,682 890 791 209.5 1,100 19.1 73.1 282.6 16.8 1995–96 1,346 677 669 259.3 936 27.7 65.8 325.1 24.2 1996–97 1,469 736 733 250.1 986 25.4 80.9 331.0 22.5 1997–98 1,209 503 706 206.9 710 29.1 76.1 283.0 23.4 1998–99 1,490 670 821 273.5 944 29.0 69.6 343.1 23.0 1999–2000 1,545 726 819 182.4 908 20.1 92.0 274.4 17.8 2000–01 1,327 531 796 250.9 782 32.1 71.1 322.0 24.3 2001–02 1,198 481 717 208.3 689 30.2 73.1 281.4 23.5 2002–03 1,428 510 918 279.6 790 35.4 111.3 390.9 27.4 2003–04 1,114 380 734 310.0 690 44.9 69.4 379.4 34.1 2004-05 (est) 1,734 361 1,373 324.0 685 47.3 158.9 482.9 27.8 Source: Author’s estimates based on data from Economic Relations Division, NGO Affairs Bureau, and International Monetary Fund. Economics and Governance of Nongovernmental Organizations in Bangladesh 3.3 A sizeable portion of external assistance to NGOs comes through government agencies from funds disbursed by lending agencies. These funds are typically part of larger projects that combine funding for government agencies as well as for NGOs. These projects are recorded in the government’s Annual Development Programme, but since the portion of the projects that are channeled to NGOs is not recorded in the budget it is difficult to provide accurate estimates of these flows. There has been a large infusion of multilateral resources to microfinance since 1995, and a number of large multi-donor projects in education and health. Our estimates suggest that on average 20 percent of the external assistance that is provided to NGOs comes through government from lending agencies, averaging about $82 million a year from loan sources. Table 3.1 shows that the average share of funds going to NGOs since 1995 is approximately 24.6 percent of total aid flows to Bangladesh. 3.4 Financial sustainability of microcredit operations has led to shrinking dependence on external assistance for the typical NGO. While the volume of external assistance to NGOs has grown in absolute terms over the past decade, the relative importance of donor resources has declined as microcredit interest income has grown rapidly. This can be illustrated in several ways. First, during the 1995–2000 period, when the number of communities with at least one NGO program doubled (see figures 2.2. and 2.3), the volume of external assistance to NGOs grew at a far slower pace (table 3.1). Second, a study of 11 large NGOs showed that while donor grants accounted for 94 percent of the combined income sources of these 11 NGOs in 1989–90, they constituted only 35 percent by 1998–99. A large part of this difference was made up by microfinance interest incomes, members’ savings, and commercial funding. 76 3.5 An approximate breakdown by sector suggests that health (31 percent), microcredit (12 percent), and education (12 percent) receive significant shares of external resources channeled to NGOs. These percentages give a rough approximation, as they are compiled from a range of sources. Moreover, they are three-year averages that mask significant variations from year to year. Apart from these three sectors, agriculture and rural development (agricultural extension, fisheries, forestry, etc.), social safety net programs, and social development/advocacy programs also receive significant amounts of donor aid, though precise breakdowns are difficult to quantify. The total volume of bilateral donor funding for microfinance has declined sharply over the past 10 years, and there has been a trend by bilaterals to move away from direct provision of revolving loan funds, at least through the use of grants, toward a focus on technical assistance, capacity building, and, increasingly, innovation in the sector. Resources for revolving loan funds are now being channeled almost entirely through PKSF for on-lending to NGOs. 77 3.6 Transfers to NGOs as a share of total resources transferred varies significantly across development partners. A recent survey of five donor agencies shows that multilateral agencies such as the World Bank provide the bulk of their resources to the government of Bangladesh. An average of 14 percent of total International Development Association (IDA) disbursements over the past decade were channeled to the NGO sector through government line agencies or autonomous bodes such as PKSF (see annex table 3.1. for details). Bilateral 76 In 1998–99, members’ savings accounted for 12 percent (up from 0 percent in 1989–90); service charges, bank interest, and own resources accounted for 31 percent (5 percent in 1989–90); and bank and other borrowings accounted for 22 percent (1 percent in 1989–90) (DFID 2000). 77 The World Bank disbursed $282 million through PKSF for expanding microcredit outreach over this period. 42 Chapter 3: Financing NGOs: Trends and Prospects agencies’ shares to NGOs are higher: the UK Department for International Development (DFID) and the Canadian International Development Agency (CIDA) direct approximately 29 percent and 20 percent to NGOs respectively, while the European Union estimates that it spends up to 40 percent of its resources in Bangladesh through NGOs. 78 Moreover, the sectoral focus varies significantly by donor. Around 21 percent of DFID’s funds to NGOs went to advocacy activities in 2003–4 and around 10 percent to “enterprise development.” IDA, by contrast, did not channel any resources to advocacy activities, while spending around 70 percent of its resources to NGOs on microfinance. 3.1.1 Issues with direct donor-NGO financing 3.7 There are a wide variety of donor-NGO partnerships in Bangladesh. These are shown in annex table 3.2. 3.8 The simplest form of funding consists of project funding to provide very specific services. Such funding involves no support to the wider institution, and its objectives are usually set by the funding agency. Donors often pool their funds if the projects they are financing are large and involve large NGOs. In some cases a donor liaison office provides an interface between donors and the NGO, promoting harmonization and reducing transaction costs. 3.9 Other types of financing involve broader institutional support, based on a shared vision of change. In a partnership agreement, an NGO may agree on a common vision with a donor or group of donors, which then finances the whole range of the NGO’s activities. The consortium funding of Proshika phase VI was an example of this. So is DFID’s funding of Nijera Kori and Samata, the World Bank’s funding of PKSF, and funding of Rupanter by the US Agency for International Development (USAID). This type of funding has very different parameters from project financing, as the donor invests heavily in capacity building for the whole organization. A more extensive form of this funding also includes an institutional development process that sees the NGO transformed into a different legal entity by the end of the funding period—for example, CIDA’s funding of the Palli Daridro Bimachan Foundation (PDBF) and USAID’s funding of the Aranuk Foundation. However, experience shows a mixed picture with both partnership agreements and institution building. Some partnership agreements have tried to upscale the work of an institution too quickly, with resulting NGO management strains, expansion problems, and resultant overfunding. 79 Predictably, institutional transformation has also not been easy, as the NGO and funder sometimes have different visions of change; in any case these processes typically take a long time to be implemented. 3.10 There is a re-emergence of project-based financing and greater use of wholesale funds for smaller NGOs. Donors have found that dealing with small and medium-size NGOs carries high transaction costs relative to the scale of funding. 80 This has led to creation of a number of wholesale vehicles in which donors set the objectives of the program, with an agency 78 Matsaert 2005. 79 DFID 2000. 80 These costs are mainly for staff time, accounting, and financial reporting, as well as capacity-building needs. 43 Economics and Governance of Nongovernmental Organizations in Bangladesh managing a pool of money for smaller NGOs. 81 The managing agency could be an international NGO or a consulting firm. The European Commission (EC) has also recently stipulated that any NGO contract must be competitively tendered, which has resulted in the creation of large number of EC outcome-based funds. At the same time, the amount of funding of partnership agreements is being reduced due to the mixed experience described above, with a return to a project-based approach for large NGOs. 3.11 NGOs face a high staff turnover in donor agencies and a wide array of different donor financial stipulations. Staff turnover in donor agencies results in lapses in institutional memory regarding the NGOs that the agencies fund. 82 The varied quality of financial reporting also stems from the many different formats imposed by donors, the concentration on cash accounting rather than accruals, and a focus on specific projects without attention to what is happening to the wider institution. 83 A review of a 10-year program of support to NGOs in the agriculture sector found that only 50 percent of NGOs fully adhered to financial obligations put in place by donors. 84 The situation for smaller NGOs appears worse, as most donors acknowledge that smaller entities are generally unable to comply with donor requirements and with all accounting and auditing standards. 85 Clearly, several donor financial stipulations are unrealistic, and there is an urgent need for harmonization in this area (see chapter 4 for more details). 3.1.2 Issues with government contracting of NGO services 3.12 As discussed above, NGOs also receive resources from the Bangladesh government. The largest share is through donor-financed projects implemented through a particular line agency, with a smaller proportion coming from government’s own revenues. A recent example of the government using its own resources to contract with NGOs is the $17 million allocation to PKSF in the 2004/05 budget, to be channeled to NGOs to scale up a pilot program with the extreme poor. 3.13 There are clear merits in having partnerships and contracting arrangements for social service delivery in Bangladesh. As neither government nor NGOs command sufficient resources and capacity on their own, different modes of partnership have evolved over time. These range from formal collaborative and contracting arrangements to informal “gentlemen’s agreements.” There is cross-country evidence that contracting NGOs has resulted in rapid gains in service delivery outcomes, showing how non-state provision can go to scale cost-effectively. 86 Financing NGOs through the government budget improves the comprehensiveness of the fiscal framework and leads to more rational and coordinated public policy choices. 81 An example is Manusher Jonno, a DFID-funded rights promotion fund administered by CARE, which screens, funds, and provides technical assistance to smaller NGOs. Other examples are the various wholesale funds run by USAID (such as its health and human rights funds) and CIDA’s Canada Fund. 82 Sultan 2004; DFID 2003. 83 S. R. Khan 2005. 84 ASIRP 2003. 85 Khan 2005; Irish and Simon 2005. 86 Loevinsohn and Harding 2004. 44 Chapter 3: Financing NGOs: Trends and Prospects 3.14 There are some notable success stories in contracting NGOs. A current example of successful partnership between donors, government, the private sector, and NGOs is in solar energy. An IDA-financed project is financing 50,000 solar energy home systems over five years. To implement the project, government contracted the Infrastructure Development Company (IDCOL), a government-owned company governed by a board with public-private representation and managed with considerable autonomy. IDCOL has the operational autonomy to select NGOs and other partners; it also provides technical assistance, ensures equipment standards, and monitors progress. The Rural Electrification Board, a highly regarded government body, helps NGOs select areas in which to promote the solar home systems. 3.15 NGOs raise awareness about solar systems and train household members in their use and repair. Microcredit loans are used to pay for these systems, and staff are trained to provide maintenance support during monthly repayment collection visits. The private sector manufactures and supplies the system components, with quality monitored by IDCOL. 3.16 The project’s successes to date include distribution of as many as 30,000 solar home units in the first two years, more than twice the targeted coverage of 14,000 households. These achievements depend on a number of factors: (a) learning from earlier initiatives with renewable energy; (b) the independence of IDCOL to select partners and monitor program activities; (c) use of the country-wide NGO network of microfinance organizations to channel financing to rural households; and (d) regular interaction among the various partners. 3.17 There are other successful examples of contracting NGOs. PKSF, an autonomous nonprofit apex body for microfinance, offers a number of lessons about how to design a successful partnership between donors, government, and NGOs. Among other things, PKSF shows the importance of sound internal governance of an independent foundation and sets strict eligibility criteria for NGOs to receive funds (see box 3.1). 3.18 At the same time, there have been a range of practical problems with government contracting of NGOs. Key problems relate to delays, corruption in assigning NGO contracts, and the dampening of advocacy efforts by NGOs. The lack of transparency in selecting NGO partners has led to the emergence of dubious NGOs and has driven out a number of competent NGOs from the bidding process. These are related to broader governance problems with public procurement in Bangladesh. NGOs also voice concerns that government finance might imply greater political demands, and potentially a patron-client relationship with government. 3.19 Experiences with contracting in non-formal education offer clear lessons in what not to do. Future collaborative arrangements need to avoid repeating the contracting errors made under the Integrated Non-Formal Education Project, when NGOs were contracted to provide non-formal education programs. A selection process evolved in which proposals from NGOs were evaluated and selected through an inter-ministerial Subvention Committee that lacked specific evaluation criteria. Efforts to avoid domination by some NGOs meant that numerous smaller organizations were contracted and unqualified “fly-by-night” NGOs emerged. Weak monitoring systems also made it difficult to ensure that programs were implemented according to agreed contracts. The Department of Non Formal Education was established in 1995 to facilitate 45 Economics and Governance of Nongovernmental Organizations in Bangladesh Box 3.1 Using an autonomous foundation to contract NGOs: Lessons from PKSF Bangladesh’s famed microfinance sector owes part of its success since the 1990s to the support provided by the Palli Karma-Sahayak Foundation. Set up in 1990 as an autonomous nonprofit organization, PKSF channels funds to NGOs to expand their microcredit operations, carries out advocacy for the microcredit sector, and helps build the capacity of the industry through hands-on assistance and formal training courses. PKSF currently lends to 222 NGOs, charging 4.5 percent interest to small and medium NGOs and 7 percent to large NGOs. PKSF is also widely credited for sharpening the focus of many NGOs on financial sustainability and for setting appropriate standards that will ease the way for a strengthened regulatory structure for microfinance. An important factor behind these achievements is that the organization is widely acknowledged to be governed in a transparent, nonpolitical manner. PKSF’s internal governance is structured to ensure that government has representation on its board while also permitting PKSF full operational autonomy. The governing body provides for a balance of interests, with representatives from government, civil society, and the private sector. The high profile and nonpolitical stance of respected public figures on the governing body are central to PKSF’s autonomy and legitimacy. PKSF employees earn on average 70 percent more than employees in public sector banks, creating better incentives for performance. PKSF lends to NGOs only after a strict screening process that includes examination of the NGO’s track record of implementing microfinance programs, its basic standards of internal governance, and its geographic location. The application process for becoming a “partner organization” is simple, and advice and information are freely available from PKSF’s Counseling Cell. The process of evaluating prospective partners is time-bound, and feedback is given to partners at each stage of the application process, including after rejections. Every year, PKSF receives hundreds of applications from small NGOs desiring to become partner organizations of PKSF. Thus far, PKSF has found that around 10 percent of these NGOs meet PKSF’s strict eligibility criteria. The Counseling Cell offers advice on the changes an NGO needs to undertake in order to become a PKSF partner. Once an NGO becomes a PKSF partner organization, staff undertake field visits to assess the amount of its loan application based on its track record and capacity to absorb funds. Standard monitoring formats are introduced to maintain comparability across NGOs. Repayment rates are almost 100 percent due to strict monitoring and incentives for NGOs to repay in order to access new loan tranches. government-NGO collaboration; however, following problems with contractual processes and reports of corruption, the department was shut down in 2003. 3.20 A recent audit of short-listed NGOs in a new phase of an ongoing World Bank–financed education project 87 found that 34 percent of the listed NGOs had dubious credentials, including one or more of the following: they (a) lacked a functional office; (b) had less than two years of experience in non-formal education; (c) submitted no audit reports, or fake reports; and (d) used fake experience certificates from local government officials. NGOs will now be selected from a revised, shortened list, but this recent experience reflects the fact that contracting NGOs remains problematic in the education sector. 3.21 A comprehensive assessment of contracting NGOs in the health sector has been carried out. 88 Contracting of NGOs by government is increasingly important in the health 87 The Post Literacy and Continuing Education Project. 88 World Bank 2005. 46 Chapter 3: Financing NGOs: Trends and Prospects sector, and efforts to gather lessons about what works and why are beginning to be made. 89 Comparative analysis of different contracting models shows that most lacked systems for monitoring performance or providing staff performance incentives. 90 3.22 These experiences offer a number of lessons for contracting NGOs. Contracting systems need to use competitive bidding processes with simple, transparent application and approval processes and guaranteed processing times. The requests-for-proposal ought to specify the consequences of providing incorrect information and should ensure that the final selection will only be made following third-party inspection of ongoing programs. The selection criteria for NGOs should include (a) a solid track record of past performance; (b) a management information, a robust accounting system, and appropriate fiduciary arrangements; (c) accreditation systems and mechanisms for monitoring performance and accountability; (d) effective systems for payment, retention of user fees, and performance-related pay; and (e) mechanisms for social accountability through community participation. 3.1.3 Issues with private charitable donations 3.23 Private charitable contributions are largely related to religious traditions. A survey of 1,062 households in 2001 showed that around 70 percent of private charitable contributions go to religious and social welfare institutions (mosques, madrassas and orphanages). Educational institutions are a distant second (15 percent), followed by recreational events (9.5 percent). 91 People cite a number of reasons for donating, the most common relating to religious beliefs, humanitarian needs, and desire to promote a local organization. The Islamic tradition of individuals donating a certain percentage of their savings to worthy causes (zakat) appears to be a key driver behind these charitable contributions. Development NGOs that receive foreign funding do not appear to feature prominently among the recipients of private charitable contributions. This is partly due to the fact that except in times of national emergency, these NGOs do not actively seek local donations. 3.24 Private transfers to “Islamic NGOs” are now under greater scrutiny. A spate of security incidents linked to religious extremism has led to a greater focus on so-called “Islamic NGOs”. In this context it is worth noting that there are a good number of faith-based NGOs with a proven track record of delivering services that are valued by the poor. Box 3.2. highlights the work of one such NGO, – Anjuman Mufidul Islam. 3.25 The new provision for tax deductions for charitable contributions introduced in the 2005 budget may increase resources going to development NGOs. This provision permits deductions for donations to “philanthropic and educational institutions,” and is expected to increase the flow of funds to the nonprofit sector as a whole. NGOs will need to submit a written application to the National Bureau of Revenue describing their activities, and the bureau will then decide whether contributions to the NGO in question will qualify for a tax deduction. It will 89 Recent assessments include Cummings 2004, Loevinsohn 2001, and Loevinsohn and Harding 2004. 90 The modalities of contracting NGOs include (a) direct contracting by government or a governmental body; (b) contracting through or as a nonprofit registered company; (c) using a contracted manager to manage arrangements with NGOs and provide monitoring and evaluation functions; (d) setting up an independent trust; (e) direct funding from donors, usually through a contracting arrangement (see World Bank 2005). 91 Ahmad and Jahan 2002. 47 Economics and Governance of Nongovernmental Organizations in Bangladesh Box 3.2 Anjuman Mufidul Islam: A welfare-oriented “Islamic NGO” Anjuman Mufidul Islam (AMI) is one of the oldest NGOs in Bangladesh. Established in Calcutta in 1905 by a group of Muslim philanthropists, the organization began operating in Dhaka in 1947 and eventually became an independent organization registered under the Societies Registration Act. AMI is primarily urban and has 42 branches in cities and towns all over the country. AMI is unique in that it is the only known welfare organization in Bangladesh involved in burying unclaimed corpses of poor people and assisting those without adequate resources to bury their dead in accordance with Islamic strictures. In 2004–5, AMI buried 1,656 people. It also provides necessary medical assistance through a mobile medical unit. Over the years, AMI has expanded to include activities directed toward the welfare of poor people irrespective of their religion. The charity operates two orphanages, a junior high school, a technical institution, and a college. It has a program that gives cattle and rickshaw vans to the ultra-poor, abandoned wives, and widows, and it also distributes clothes to the poor during Eid festivals and in winter. Following natural calamities, AMI assists disaster-affected people with cash or relief. In addition to normal relief operations, financial assistance is provided in the form of grants to the destitute and disabled, “marriage relief” is given to poor parents to meet the cost of their daughters’ weddings, and grants are extended to poor meritorious students. A notable feature of AMI is that it does not accept foreign donations but relies entirely on local funds. AMI's sources of income are contributions from its members; collection and sale of hides and skins during Eid-ul- Azha; donations given by the general public, often during Eid as part of the zakat tradition; government grants; and income from AMI's own property. AMI’s budget rose from $162,000 in fiscal 2000 to almost $280,000 in fiscal 2005. Its centenary year was marked by messages of support from political figures, NGO leaders, private sector representatives, and various members of the public. be important to monitor this process carefully as it involves considerable discretion as to what constitutes an eligible philanthropic or educational institution. 3.26 The certification process for tax deductions could be devolved to an independent professional body. In Pakistan and the Philippines, the Ministry of Finance has devolved certification for tax purposes to the Pakistan Centre for Philanthropy’s certification program (see box 4.3) and the Philippine Council for NGO Certification, respectively. Organizations receive a “seal of good housekeeping” if they meet defined standards of organizational effectiveness, based on an independent evaluation of their internal governance, financial management, and program delivery. The advantage of such devolution for the government is that it avoids stretching existing capacity within the ministry and does not have to deal with complaints from nonprofits that are not certified. Certification bodies are discussed in greater detail in chapter 4 in the context of proposed improvements in the legal and regulatory framework for NGOs. 3.27 The expatriate Bangladeshi community is a potentially large source of private donations. The bulk of remittance transfers to Bangladesh (around $3.4 billion in 2003–4) are from relatively poor migrant workers who remit funds to sustain their families. However, there is a smaller group of relatively wealthy Bangladeshis who can be tapped for charitable contributions to NGOs. In order to facilitate this transfer, NGOs could explore the possibility of obtaining nonprofit status in the United States and Europe, or could establish tax-exempt foundations, so that income tax deductions could be obtained for contributions. This is likely to be a resource-intensive process feasible only for the larger NGOs, at least initially. The 48 Chapter 3: Financing NGOs: Trends and Prospects government could also facilitate this resource transfer by ensuring that the laws against money laundering are widely known and applied in a streamlined manner. NGOs will also need to make greater efforts to publicize their activities to target expatriate communities, as well as to improve corporate governance, in order to encourage these private charitable contributions. 3.2 Internally generated funds 3.2.1 Microfinance interest income 3.28 Microcredit surpluses are being generated with low-cost operations. As microfinance scaled up in the 1990s, there was a growing emphasis on financial sustainability as individual branches became profitable within a few years. Financial analysis of NGOs of various sizes carried out for this study suggests that many generate surpluses from their microfinance operations (table 3.2). 92 All the NGOs surveyed had positive adjusted returns on assets, with most at about 3.5 percent and ASA much higher at 10.5 percent. There have been consistent efficiency gains for NGOs of all sizes over the past five years, as the number of borrowers and total portfolio per credit officer have been steadily increasing (see annex table 3.3). ASA is known worldwide as an industry leader in cost-efficiency, and all the NGOs surveyed show high levels of efficiency when compared to programs in other countries. High population densities are a significant contributing factor, though NGOs have also worked hard to develop low-cost, efficient, decentralized management systems (see chapter 5). 3.29 There is a trade-off between reducing interest rates and increasing access for the poorest. Microcredit profits are used to finance lending to groups and geographic areas that are underserved as well as to build the equity bases of these institutions. There is scope for reducing interest rates, but this could come at the cost of serving poorer clients, as currently part of the spreads are used to cross-subsidize small loans. BRAC’s average loan size is about 40 percent lower than ASA’s, reflecting the fact that a higher proportion of BRAC loans currently go to the poorest (see annex table 3.4). 93 A few NGOs, such as TMSS, use microfinance profits to cross- subsidize their social programs, though the majority appear to only cross-subsidize the poorest borrowers within their microfinance programs. Moreover, it is important to put the criticism of NGO interest rates in context: the real alternative for poor households is not a commercial bank, but the village moneylender whose rates average 120 percent per year. 3.30 Interest income is projected to grow with microcredit expansion, provided portfolio quality remains unaffected. Annex table 3.4 shows that there has been rapid growth in client numbers in recent years, and there is also clear evidence that the market for mainstream microcredit is not saturated. Chapter 2 illustrates the geographic areas that have relatively little coverage and shows that currently around 43 percent of households access microcredit. Given that about 50 percent of households are poor and another 10 percent of households are part of the vulnerable nonpoor group, there is clear scope for market growth, even allowing for a sizeable 92 The analysis included two large NGOs, BRAC (3.8 million active borrowers) and ASA (2.5 million active borrowers); two medium NGO MFIs, Jagoroni Chakra (JC) and TMSS, with between 50,000 and 500,000 active borrowers by June 2004; eight small NGO MFIs, with between 10,000 and 50,000 active borrowers by June 2004; and four very small NGO MFIs, with fewer than 10,000 active clients by June 2004. 93 BRAC’s average loan size in 2003 was about Tk 6,100, while ASA’s was almost Tk 8,600. 49 Economics and Governance of Nongovernmental Organizations in Bangladesh number who may not wish to borrow. However, it is critical that portfolio quality be maintained at currently high levels, as risks grow with larger loans and in older branches. Table 3.2 Profitability of NGO microfinance institutions, 2001–4 (percent) NGO or category Indicator 2001 2002 2003 2004 Adjusted financial self-sufficiency 168 122 119 121 Very small NGOs Adjusted return on assets — 3.8 3.4 4.0 Adjusted financial self-sufficiency 128 113 122 127 Small NGOs Adjusted return on assets — 1.7 3.4 4.8 Medium Adjusted financial self-sufficiency 144 122 122 119 NGOs Adjusted return on assets — 3.5 3.6 3.2 Adjusted financial self-sufficiency 113 169 173 — ASA Adjusted return on assets — 10.0 10.5 — Adjusted financial self-sufficiency 107 111 118 — BRAC Adjusted return on assets — 2.3 3.4 — Note: — = not available. Financial self-sufficiency is defined as (total annual income/(total operating + financial expenses)) * 100. The adjustments made to calculate these ratios were made to adjust the cost of funds borrowed from PKSF to commercial market rates of borrowing (i.e., from 4 percent for smaller NGOs and 7.5 percent for ASA and BRAC in 2004, for example, to 10 percent, the average lending rate of nationalized commercial banks to NGOs), and add an opportunity cost of capital equivalent to the annual rate of inflation. 3.31 Recent initiatives by some NGOs to securitize their portfolios and borrow from commercial banks indicate a growing integration with the formal financial system. The interest in finding alternative sources of funds grew after PKSF capped the interest rate at which its partner organizations could lend to clients in mid-2004. There are still significant problems to be overcome in accessing commercial funds, such as the type of collateral to offer, but this source of loan portfolio funds will grow if the costs of borrowing from banks decline and if steps are taken to help commercial banks overcome their reluctance to lend to NGOs (see box 3.3). 3.32 Social investment funds have not yet played a large role in the financing of the microfinance sector in Bangladesh, in contrast to their rapidly growing use in Eastern Europe. Equity investments in NGOs will be facilitated if an appropriate regulatory framework to set up microfinance banks is established or if more NGOs register under the Companies Act of 1994. CARE is considering setting up a social investment fund that will combine commercial investments with the use of grant-funded technical assistance, a model employed by at least two of the recently established international social investment funds. BURO, Tangail is one of the few NGOs in Bangladesh to actively seek out international investment funds. However, these types of transactions need to address foreign exchange controls that limit how much can be repatriated for loan repayments. 50 Chapter 3: Financing NGOs: Trends and Prospects Box 3.3 Linking NGOs to commercial bank financing NGOs have made relatively little use of commercial bank funds, as the contributions of other cheaper and more easily accessible sources (e.g., member savings, retained earnings, and loans from PKSF) have been sufficient for microfinance growth. However, commercial lending has the potential to become more important in the future as NGOs seek to diversify funding sources. The Swiss Development Corporation (SDC) has been a pioneer in trying to increase access to and use of commercial bank funding by NGOs in Bangladesh. The first project was a challenge fund with Sonali Bank, the largest nationalized commercial bank. Loans to about 10 NGOs at 10 percent interest were guaranteed by SDC, with independent ratings of the NGO MFIs being a prerequisite. Collateral was required in the form of personal guarantees of the NGO directors. Sonali Bank continued this program on its own after the SDC project was completed and now has 10 to 15 NGO partners, though it has dispensed with the requirement for independent credit ratings. More recently, SDC has partnered with Pubali Bank. A fund of Tk 75 million was formed to lend to NGOs, with SDC contributing one-third of this amount. SDC charges 3 percent for its loan and in turn Pubali Bank lends to NGOs at 10 percent, with NGOs required to have an independent credit rating done before applying. Initial demand from NGOs has been so high that SDC is considering increasing its contribution. In one case where a very large loan was made to a single MFI, Pubali Bank required collateral in the form of a lien on property owned by the NGO, but in most cases only the personal guarantee of the directors is required. Basic Bank has had access to capacity-building support from SDC, along with support to pay for credit ratings done for NGOs that qualify to apply for funding, but no direct support for their loan fund. It has 25 to 30 NGO partners to which it charges 9 percent from a Tk 20 million fund. The Credit and Development Forum (CDF) funds about 40 very small NGOs through a facility funded by two nationalized commercial banks. The NGOs are charged 12 percent for these funds: the banks get 10 percent, and CDF gets 2 percent to administer the program. The facility of Tk 50 million has disbursed approximately Tk 20 million to date, and is personally guaranteed by CDF’s directors. CDF claims that there is considerable NGO demand for larger amounts of funding, something it cannot provide, at least for now. One reason for the willingness of these very small NGOs to use commercial bank funds is that while they aspire to become partners of PKSF, they have not yet been able to qualify. A recent decision by PKSF that its partners may not charge clients more than 12.5 percent has spurred the partner organizations to make renewed efforts to diversify their funding sources, including the use of bank funds, to reduce their dependency on PKSF funding over time. Even so, there are still constraints to be overcome, primarily the relatively high cost of commercial funds, the difficulties of working with banks to meet their requirements, and the fact that some NGOs do not have a legal status that allows them to directly borrow from banks. It appears, though, that bank rates might decline in future. If banks can also become more knowledgeable about microfinance and more convinced that it is good business, as has happened in some other countries, opportunities to link NGO MFIs with commercial banks will increase. 3.33 It is likely that independent credit ratings will be required for NGOs wishing to access commercial funds. Credit ratings are one more level of transparency that gives comfort to investors. PKSF could also make use of published performance indicators and credit ratings as part of its NGO assessment process. The assessments and credit ratings that have been done to date in Bangladesh were carried out by M-CRIL, an international specialized microfinance rating agency, but it would be useful to establish a local rating organization. This could happen by linking an experienced international rating agency with a local organization that wants to become a rating agency or with an existing rating agency that rates regulated financial sector institutions and wishes to develop a capability in microfinance. 51 Economics and Governance of Nongovernmental Organizations in Bangladesh 3.2.2 NGO-managed commercial enterprises 3.34 NGOs run a variety of business operations. The scale and type of operations vary considerably, and annex table 3.5 provides examples of some of the different segments in which NGOs operate. Broadly, there are three types of business: (a) those that have grown as an extension of credit operations and the needs of poor microentrepreneurs (for example, Gano Kallayan Trust maintains hatcheries to produce day-old chicks for poor poultry rearers); (b) those that link rural with urban consumers (for example, Grameen Phone created a market for rural mobile telephone users, described in box 3.4); and (c) those set up without a direct poverty reduction element, but that generate revenues for NGO development programs. 3.35 There is considerable debate about the legitimacy of NGOs operating private sector businesses. The dominant view is that NGOs should be required to separate all their for-profit businesses from their development work, that they should not be allowed to use subsidized capital to enter or operate businesses, and that they should pay taxes on their for-profit activities at the same rate as other businesses. There are also concerns that NGOs may be drifting from their core mission of poverty reduction as management resources are diverted to overseeing commercial activities. There are also contradictory views of NGOs regarding pricing of inputs and outputs. Some argue that NGOs “entrap” their members into adverse pricing relationships, while others say that NGOs unfairly subsidize markets by paying artificially high prices. 94 3.36 On the other hand, it is recognized that NGOs have the potential to stimulate rural growth. There are obvious benefits from NGOs providing poor rural producers with input, output, and marketing linkages. Large NGOs in particular have nationwide networks, and there are private sector actors who are interested in tapping into partnerships with them through subcontracting arrangements or as part of their corporate social responsibility mandate. Donors also see the potential of NGOs to act as facilitators for rural market development and private sector linkages, and regard commercial ventures as a means of reducing dependence on donor financing. It is common practice in many countries for NGOs to operate business ventures to generate revenues (see box 3.5). 3.37 NGO businesses are registered in a variety of ways that have significant effects on their objectives, independence, and operational freedom. Box 3.6 gives examples of different types of NGO business registration. Project support enterprises typically are not separately registered and generally form an integral part of an NGO, although they may operate as “profit centers” within the overall institutional structure. Their accounts are not generally separated from those of the overall NGO. 95 Not-for-profit businesses, the second category, have separate registration and hence attain a level of independence, but their aims may not be explicitly profit- oriented. The governance of these types of businesses is also closer to that of the sponsoring NGO. Still other businesses are characterized by separate incorporation as “for-profit” entities under the Companies Act, with shareholders and a board. These businesses are still generally majority-owned by the NGO, but in the case of financial institutions, a requisite number of external shareholders and minimum share capital are stipulated. The most developed type of NGO business is a publicly limited company floated on a stock exchange. The Gano Kallayan 94 McKennie, Toomey, and Hossain 2003; Alamgir 2004. 95 Some larger NGOs, such as BRAC, do provide segmental reporting on project support enterprises. 52 Chapter 3: Financing NGOs: Trends and Prospects Box 3.4 Moving out of poverty using microcredit and a mobile phone Grameen Telecom introduced mobile phone services in rural areas in 1997 and began an innovative business initiative with Grameen Bank borrowers. The Village Phone Program selects, trains, and provides credit to Grameen members who purchase a phone using the loan and then provide mobile phone services in the village for a fee. Village Phone members are charged a lower per-call rate by Grameen Telecom, and the difference between the user fee and this negotiated rate is the profit. Jharna Begum, now one of 125,000 “phone ladies” operating around the country, has been a member of Grameen’s Shonarong Tongibari branch for the last 13 years. She first joined Grameen to provide some financial support to her father. She was in great financial hardship when she married Habibur Rahman, a hired hand in a rickshaw workshop. Her husband’s meager earnings were barely enough to provide for their two children and they were struggling to make ends meet. Over the years, Jharna took a number of general loans from Grameen Bank. Eventually the family was able to set up a rickshaw garage in their house and start a confectionary. Jharna’s economic situation improved significantly when she took a Tk 12,000 loan from the Grameen Bank to purchase a cell phone under the new program. She conducts her cell phone business from a rented shop in the village bazaar from 9:30 in the morning to 9:00 at night. With an additional monthly income of Tk 5,000–6,000 coming from the mobile phone business, Jharna is now saving every month, something she was not able to do before she bought the cell phone. She makes a regular monthly deposit of Tk 800 in her savings account with Grameen and has also opened a separate savings account for her husband. In stark contrast to her struggling days, she is now the proud owner of a plot of land which she purchased from her savings for Tk 40,000. She has also bought a refrigerator for the confectionary and some furniture for the house. The village phone business has helped her acquire the bakery premises for her husband. According to Jharna Begum, the village phone business has significantly improved her life. Becoming a village phone operator has not only raised her earnings but has also enhanced her socioeconomic status in the community. The local elites of the village now visit her household regularly. When asked about her future plans, Jharna said her primary goal was to educate her son and daughter and further expand her business, starting with buying her own mobile phone shop. chick hatchery is an interesting example in which microcredit clients’ savings were converted into equity, and these members own the majority shares of the business. 3.38 The profitability of NGO businesses appears mixed. An analysis of commercial enterprise financial data for the past several years suggests that financial returns vary significantly across sectors (see annex table 3.6). For instance a comparison of BRAC’s commercial enterprises shows that its returns on assets can vary from 1 percent to 11 percent. Returns for handicrafts, a traditional market for NGO businesses, also exhibited a mixed picture: returns on assets varied from 0.62 percent for TMSS in 2003–4 to 5.62 percent for BRAC. 3.39 From the data available it is also clear that BRAC and Grameen commercial ventures operate very much in a league of their own in terms of scale and market power. They have higher levels of commercial expertise and market share than the bulk of smaller NGOs, which operate less professional enterprises that are generally more capital-constrained. While a larger sample is required to make firm conclusions, it appears that a large set of NGO businesses are managed with limited capacity and variable profitability. 53 Economics and Governance of Nongovernmental Organizations in Bangladesh 3.40 A number of factors explain the widely different returns to NGO businesses. First, few but the large NGOs can afford to allocate specialist staff to operate businesses in a professional manner. There are concerns about businesses diverting management resources away from poverty-reducing activities, so the businesses may be left to run without strong professional oversight. Second, the continued operation of efficient businesses depends on the organizational strength of the NGO. This is by no means assured, especially in smaller NGOs where much responsibility rests on a few talented individuals who oversee operations. Third, many of these businesses, especially in the handicraft sector, remain relatively undercapitalized due to frequent cash transfers to social activities. This may result in low levels of reinvestment, leading in turn to low levels of business growth. From the existing evidence, it may be tentatively concluded that NGO businesses that are separately incorporated from an NGO’s social activities perform better. Box 3.5 NGO-linked commercial ventures in other countries The “endowment model” is an integral part of NGO financing strategies worldwide. NGOs are increasingly trying to raise funds from internal sources to cover basic operating costs, in addition to approaching external donors for their program costs. For example: • Organização de Ajuda Fraterna, a center for street children in Brazil, covers all of its annual expenses from the sale of furniture and hospital equipment, while simultaneously providing job training and income for its constituents. • Child Relief and You (CRY), in India, generates nearly 50 percent of its annual income through the sale of greeting cards, paper products, and children’s toys in the Indian and US markets. • FUNREDES, in the Dominican Republic, covers over 70 percent of its operational expenses by offering Internet and Web site design services to for-profit companies. • The environmental NGO Adonis, in the Czech Republic, has started a for-profit tourist office to attract ecotourists to the region. • In Georgia 55 out of 291 randomly selected NGOs were involved in commercial activities. A variety of laws govern NGO commercial activities in various countries. In Switzerland, the legal principles governing the establishment and functioning of nonprofit associations are found in articles 60–79 of the Swiss Civil Code. These stipulate that while the aim of the association must be nonprofit, this does not imply that the association must refrain from generating profit. However, any profit made cannot be distributed to members and must be used by the NGO in order to achieve its aims. Numerous Bulgarian NGOs have already launched strategies to initiate commercial activities. According to Bulgarian law, NGOs (as separate legal entities) may conduct economic activity, provided it complies with specific legal criteria: • The economic activity must be supplementary, meaning that the volume of the for-profit activity must not exceed the amount of basic nonprofit activity. • The economic activity must be relevant to the basic mission of the NGO, that is, the business activity must be a means of achieving organizational goals. • The revenue generated by the economic activity must be reinvested, that is, used for achieving the objectives set forth in the statutes or the incorporation act of the NGO. • The scope of the economic activity must be defined in the statutes of the NGO. In the Central Asian Union (Kazakhstan, Kyrgyz Republic, and Uzbekistan), commercial activity has emerged as an increasingly important source of revenue for NGOs. Under the Public Associations Laws and the CAU Civil Codes, a nonprofit organization has the right to carry out business activity only insofar as such activity furthers the purposes for which the nonprofit organization was created. 54 Chapter 3: Financing NGOs: Trends and Prospects Separation seems to allow greater specialization, more strategic business planning, and improved governance. 3.41 There is no evidence that NGOs are taking advantage of members through the pricing of inputs or outputs. A 2004 study examined the extent to which NGO members may be “trapped” into adverse pricing relationships by comparing prices that NGOs offered their members for outputs, and prices charged for inputs, with prices offered by private sector vendors of products of similar characteristics. The study conclusively argues that NGOs offer fair prices that do not overly distort markets. 96 Another study shows that BRAC’s program support enterprises often set a price floor even when market conditions change to protect poor producers. 97 The market distortion through this subsidy is justified on equity grounds, given the organization’s overall poverty reduction mandate. 3.42 Looking forward, NGOs ought to be more systematic about market entry and exit. Rationales for starting a commercial venture vary: the objective may be to correct a market failure in order to benefit the poor, or it may be to earn profits to subsidize social programs. The best success stories in Bangladesh have come from NGOs taking innovative steps in areas where private sector actors were too risk-averse to enter (e.g., rural telecommunications). If the objective is to create new markets, then once private sector players have entered the market the NGOs could look to divest in that market segment and reinvest in new “frontier” markets that benefit the poor. In this way the NGO behaves like a social venture capitalist. However, if the objective is to generate revenues for the social activities of the NGO, then remaining in profitable markets clearly makes most sense. Box 3.6 A typology of NGO businesses and their legal status Project support enterprises. Parent NGOs are normally registered under either the Voluntary Social Welfare Agencies (Regulation and Control) Ordinance or the Societies Registration Act. The business activities are undertaken as programs or projects of the main NGO and do not have a separate legal identity. Typically, separate books of accounts are maintained and financial reports are prepared according to the type of business. Examples include all the businesses of Jagoroni Chakra and TMSS and the program support enterprises of BRAC such as seed production and processing, tissue culture, and poultry hatching. Not-for-profit businesses. The not-for-profit company is also widely used as a legal arrangement in which the Companies Act 1994 (sections 27 and 28) is used to create a company limited by guarantee. Examples of such organizations are Grameen Shakti, Grameen Telecom, Grameen Mothsha Foundation, Grameen Uddog, and Grameen Shamogree. For-profit companies. The Companies Act 1994 has been used to create private limited companies. Most are majority-owned by the NGO, sometimes as joint venture companies with private sector participation. Examples include BRAC Industries Ltd., BRAC Concord Ltd., and BRAC Concord Lands Ltd. Banking institutions. The Banking Act 2001 has been used to create banking institutions, such as BRAC Bank. Public limited companies. Gano Kallayan Trust used the Companies Act 1994 to float its poultry hatchery to establish it as public limited company; BRAC Bank plans to take this step in the near future. These companies are overseen by the Securities and Exchange Commission and are subject to its reporting requirements. 96 Louis Berger Group 2004. 55 Economics and Governance of Nongovernmental Organizations in Bangladesh 3.43 In short, NGO-run commercial enterprises can generate significant rewards, but there are considerable risks involved. The analysis here indicates that there are varying financial returns to NGO-run businesses. The type of return is likely to be a function of the NGO’s capacity to employ professionals to run these operations. Hence, small NGOs in particular ought to think carefully before setting up such ventures, given that their capacity to run businesses profitably is likely to be weak. At the same time, there have been clear examples in which NGOs entered markets that benefited the poor and acted as catalysts for private sector entry. This social venture capital role ought to be encouraged, particularly given the comparative advantage NGOs have in linking rural producers with the wider market. We also argue that NGOs should to the extent possible register their businesses as independent entities. This will provide greater transparency to these ventures and will create a more level playing field with the private sector. Moreover, any business profits that are then used for the NGO’s development activities ought to be recorded clearly in the financial records and could qualify for tax exemptions. Clarifying these tax incentives is likely to encourage other private sector entities to channel resources to development activities. 3.44 A common theme that emerges from this discussion on financing NGOs is the need for regulatory reform and financial accountability improvements. Fiduciary weaknesses within several NGOs and the lack of a system to distinguish reliable NGOs from fly-by night entities are constraining external assistance to a proven service delivery model. The potential to access commercial sources of financing for microfinance programs is also partly impeded by the lack of a modern regulatory framework overseeing this growing sector and the lack of a rating agency or certification body. Profits from commercial ventures cause resentment among the private sector and government due to the lack of clarity regarding adherence to tax laws and the use of profits. For this reason there is a clear need to strengthen the regulations, the regulators, and financial accountability requirements for the modern NGO sector—issues that we turn to next. 97 McKennie, Toomey, and Hossain 2003. 56 Chapter 4: The Regulatory Framework and Financial Accountability Issues 4.1 The weaknesses in the regulatory framework governing the NGO sector in Bangladesh have not impeded, and have perhaps contributed to, the growth of the industry. As discussed in earlier chapters, a laissez-faire regulatory stance, an outdated legal framework, and the lack of capacity on the part of various regulatory agencies have provided considerable flexibility for NGOs to undertake new activities, expand their programs, and diversify their funding sources, including by entering into commercial ventures. 4.2 In spite of the regulatory weaknesses there have been few incidents of scandal or debacle in the NGO sector, relative to its size. Several reasons can be cited for this. First, one must give credit to inherent altruism, which has brought together groups of people for a greater goal and which sustains the sector. Peer pressure among NGOs is an important factor, and the role of apex bodies in training, harmonizing rules, and indirectly ensuring compliance with legal requirements is related to this. Oversight and training provided by the donor community have also contributed to the relative lack of scandals. Nevertheless, one must not dismiss the fact that lack of capacity on the part of implementing agencies to enforce rules and investigate complaints could also lead to an underreporting of actual problems. 4.3 There is a growing focus on NGO financial accountability and corporate governance and on the regulatory framework. NGOs in Bangladesh are increasingly being subjected to questions from government, political parties, the business community, beneficiaries, donors, and the public. These include charges of tax avoidance, taking advantage of not-for-profit status to start businesses, collecting saving deposits without legal authority, and involvement in politics, among others. 98 The emergence of fly-by-night NGOs, as described in chapter 3, has also caused concern. Strengthening NGO governance, developing a certification system to screen out dubious NGOs, and modernizing the regulatory framework is, therefore, all the more important if the sector is to effectively counter such attacks. 4.1 An overview of the laws affecting NGOs 4.4 The legal framework for NGOs has not gone through any substantive reform in recent times. This is surprising given the size of the sector described in chapters 1 and 2. The earliest and one of the most important laws dates back to 1860, and the most recent relevant regulation was issued by the NGO Affairs Bureau in 1982. 99 Overall, the legal framework is outdated, obsolete, and in some cases not in sync with modern NGO regulatory concepts. 98 M. M. Khan 2003. 99 The only other development on the regulatory front is the introduction of the Hindu Religious Welfare Trust Ordinance of 1983, the Christian Religious Welfare Trust Ordinance of 1983, and the Buddhist Religious Welfare Trust Ordinance of 1983. Economics and Governance of Nongovernmental Organizations in Bangladesh 4.5 There are a plethora of laws and government agencies dealing with NGOs in Bangladesh. NGOs are registered and regulated under 12 different laws. 100 The most important ones for the registration and operation of the development NGOs examined in this report are the Societies Registration Act 1860, Trust Act 1882, Voluntary Social Welfare Agencies (Regulation and Control) Ordinance 1961, Foreign Donation (Voluntary Activities) Regulation Ordinance 1982, and Companies Act 1994. These are implemented by four government agencies: the Social Welfare Ministry, the NGO Affairs Bureau, the Ministry of Commerce, and the Ministry of Women and Children Affairs. 4.6 The Societies Registration Act 1860 is one law that is used to register NGOs. The Registrar of Joint Stock Companies (RJSC) in the Ministry of Commerce is responsible for registering any society formed for any “literary, science, or charitable purpose.” 101 Approximately 4,900 societies have been registered under this act, including several large development NGOs. The Societies Registration Act is one of the most lenient with respect to fiduciary requirements. Other than filing the annual list of the governing body with the registrar and receiving prior consent of the government in case of dissolution, there is no other formal requirement that NGOs have to abide by. Crucially, the act does not even specify whether the society is expected to maintain accounts and audits, although in practice the RJSC will likely not register the society unless there is provision for accounts and audits in the society’s governing rules. 4.7 The Trust Act 1882 is another law providing legal basis for the formation of NGOs. A trust can be established “for any lawful purpose,” whether private or public, and can be created by any person competent to form a contract. No government registration is required and the trustee is legal owner of the property and personally liable for breach of trust. 4.8 The Companies Act 1913, amended 1994, provides legal status to not-for-profit companies. Not-for-profit companies can be registered with limited liability provided that the company applies any income to promotion of its objectives and prohibits payment of any dividends to its members. As with the Societies Registration Act, the Registrar of Joint Stock Companies is the implementing agency. The Companies Act provides the most rigorous legal framework for NGOs, with clear directives on their fiduciary obligations. These include holding of annual general meetings, maintenance of accounts, and appointment of a registered chartered accountant as an auditor. Noncompliance of directives may result in fines or imprisonment. The number of NGOs registered under this law is significantly lower than under other laws, possibly reflecting a lack of awareness of this option as well as its more rigorous legal requirements. 4.9 A large number of NGOs are registered under the Voluntary Social Welfare Agencies (Regulation and Control) Ordinance 1961. This mandates registration for any formal 100 The Societies Registration Act 1860, Trust Act 1882, Companies Act 1994, Cooperative Societies Ordinance 1964, Charitable Endowments Act 1890, Wakf Ordinance 1962, Voluntary Social Welfare Agencies (Regulation and Control) Ordinance 1961, Foreign Donations (Voluntary Activities) Regulation Ordinance 1978, Foreign Donation (Voluntary Activities) Regulation Ordinance 1982, Hindu Religious Welfare Trust Ordinance 1983, Christian Religious Welfare Trust Ordinance 1983, and Buddhist Religious Welfare Trust Ordinance 1983. 101 Such purposes include literature, fine arts, education, libraries, museums, and so on. Environment, human rights, and historic preservation qualify only under a modern, progressive concept of “charitable,” and there are no provisions for social or sporting clubs, self-help groups, contemplative societies, and the like. 58 Chapter 4: The Regulatory Framework and Financial Accountability Issues Box 4.1 Suspension and rebirth of a large NGO: The case of Gono Shahajjo Sangstha In 1999 the government of Bangladesh took over the administration of a large NGO after investigations revealed misuse of foreign donations and gender discrimination. This was the first time the government had stepped in and taken over the running of a big NGO in Bangladesh. Gono Shahajjo Sangstha (GSS), set up in 1983, provided village education programs and engaged in social mobilization and advocacy. By 1999, GSS, with funding primarily from the European Union and the British, Canadian, and Danish development agencies, was one of the largest NGOs in Bangladesh. At that time GSS’s education program reached 200,000 poor children in 750 primary schools across the country. The allegations against GSS and its executive body, brought primarily by GSS staff members, led to an investigation by the NGO Affairs Bureau and a High Court order for the dissolution of the executive body in 1999. These allegations concerned the misuse of funds relating to the development of the NGO’s new headquarters office, as well as gender discrimination. The management of GSS was transferred to the Ministry of Social Welfare, which in turn appointed a three-member caretaker body. The High Court order was contested in court by the previous executive body, which claimed that allegations against GSS were politically driven. During the four-and-a-half-year legal battle over GSS, GSS programs closed and about 6,400 employees, including 4,000 teachers, were made redundant. In September 2003, the High Court cleared GSS of all the charges. The management and administration together with GSS’s assets and liabilities were handed over by the caretaker body to a newly formed executive committee. GSS started operating again in 2004, following its registration with the NGO Affairs Bureau. Since then, BRAC, under a project that subcontracts smaller NGOs to implement education programs, has given a $0.5 million grant to GSS’s education program for a period of three years beginning January 2004. As of 2006, 350 schools are operating, with approximately 50,000 students enrolled. or informal organization formed to render welfare services to children, youth, and women, do social work, or coordinate social welfare agencies. There are several reasons for its extensive use by NGOs. First, social welfare department officials are accessible at the union level, making it easier for local organizations to liaise with the regulatory authority. Second, organizations must register under this ordinance to receive financial support from the Ministry of Social Welfare. And third, under the Foreign Donations (Voluntary Activities) Regulation Ordinance 1978, a local organization receiving foreign funding for channeling to other organizations can do so only if those organizations are registered under the 1961 ordinance. 4.10 The purpose of this ordinance is to control voluntary organizations, and hence it gives extensive powers to the government with respect to NGOs. The ordinance is administered by the Social Welfare Department of the Ministry of Social Welfare, which has broad discretion to suspend or dissolve an NGO without provision for judicial appeal. For example, if after making an enquiry the ministry determines that a registered NGO has failed to comply with the provisions of the ordinance, or has any other irregularities, then it may suspend the NGO’s governing body and appoint in its place an administrator or caretaker body. Box 4.1 describes how the government in 1999 took over the running of Gono Shahajjo Sangstha, the seventh- largest NGO at the time. Though the governing body may appeal to the central government, the final decision, following the appeal, shall be final and shall not be called into question in any court. Curiously, the Ministry of Women and Children Affairs registers and oversees NGOs dealing with women, even though it has not been empowered to do so under the Voluntary Social Welfare Agencies ordinance. 59 Economics and Governance of Nongovernmental Organizations in Bangladesh 4.11 The Foreign Donations (Voluntary Activities) Regulation Ordinance 1978, amended 1982, oversees donor flows to NGOs. This ordinance requires that any organization carrying out voluntary activity using foreign donations be registered with the NGO Affairs Bureau. Each foreign grant then must be approved and monitored by NGOAB. The Foreign Contributions (Regulation) Ordinance of 1982 expands the scope of the law to apply to all contributions to NGOs from abroad. While NGOAB has not actively used its powers under the 1978 and 1982 ordinances to take action against noncompliant NGOs in the past, it has recently started to do so, with one legal case already instituted and two more in process. 4.12 The taxation regime for NGOs in Bangladesh is unclear. The 1984 Income Tax Ordinance stipulates that income used for “charitable” purposes is exempt from taxes. Thus, under the law as it is written, NGOs would not be required to pay tax on income from any source, including income from commercial activities. However, the Finance Acts of 1998 and 1999 made NGOs subject to tax on their income, though the income tax manual does not reflect these changes. At the same time, microcredit surpluses are exempt as long as the NGO is registered with the NGO Affairs Bureau. There are other exemptions too. 102 4.13 The term “charitable purpose” is further complicated by the increasing number of NGOs in Bangladesh that are undertaking commercial activities to fund their development programs. It is not clear whether an organization should still be exempt from tax on the surpluses generated by these operations. NGOs remain confused about the taxation regime, and BRAC is currently in litigation with the National Bureau of Revenue regarding these issues. 4.14 The 2005 budget has made donations to NGOs in Bangladesh tax-deductible in principle. Before 1992/93, the Income Tax Act provided for 100 percent deduction of the amount contributed to charities. However, rampant tax abuse led to this being withdrawn, with specific exceptions—if the beneficiaries are certain health facilities, registered organizations benefiting persons with disability, or the Aga Khan Development Network. These clearly were ad-hoc exemptions and the 2005 budget has widened this list to cover all philanthropic and educational institutions approved by the government. As discussed in section 3.1.3, the implementation of this “approval” process could be delegated to an independent certification agency. 4.15 Another grey area in the legal framework for NGOs in Bangladesh is the definition of political activities and the extent to which an NGO may engage in political activities. Political activities, broadly defined, comprise a range of activities from public education to electioneering. Electioneering, which means supporting or opposing candidates for office, and campaign funding fall at the extreme end of the spectrum and are clearly the most controversial forms of political activity. 4.16 Another shortcoming is the absence of modern and appropriate procedures for censure and appeals. The Societies Registration Act, for instance, has no provisions for fines, 102 Income from “house property” held under trust or other legal obligation wholly for religious or charitable purposes is exempt, but this does not apply to NGOs registered with NGOAB. Another interesting feature is that there is tax exemption for income that is not applied to charitable or religious purposes during the year but is accumulated or set aside for such purposes for not more than 10 years, provided the taxpayer notifies the deputy commissioner of taxes and invests the money in government securities or with the Post Office Savings Bank. 60 Chapter 4: The Regulatory Framework and Financial Accountability Issues penalties, or involuntary dissolution. The Department of Social Welfare (DSW) may “suspend the governing body” of an NGO if it has committed any irregularity with funds or any maladministration in the conduct of its affairs, or has failed to comply with the provisions of the ordinance. No NGO can be dissolved on its own motion, though. The DSW may recommend that the government dissolve an NGO because it has violated its constitution, the ordinance, or rules made under it, or acted “in a manner prejudicial to the interest of the public.” Before recommending dissolution, the DSW need only give the NGO an opportunity to be heard “as it thinks fit.” An NGO facing dissolution has 30 days to appeal to the government, but there is no right to appeal to a court. Any person who contravenes the ordinance or any rule under it or makes a false statement to the DSW can be fined or imprisoned for up to six months. Where an offence is committed by an NGO, every director or officer of the organization is deemed guilty of that offence unless he proves that the offence was committed without his knowledge or consent. 4.17 Under the Foreign Donations Regulation Ordinance, for “reason to be recorded in writing,” NGOAB can inspect or, acting under the Code of Criminal Procedure, seize accounts and other documents after providing the NGO a reasonable opportunity to be heard. NGOAB can cancel the registration of an organization or stop any activities for a failure to submit required declarations, willful submission of false declarations, or any other contravention of the ordinance. Anyone who receives or uses a foreign donation in contravention of the ordinance or any rules under it is subject to a fine of double the amount of the donation or imprisonment for up to three years. 4.18 Other than an appeal to court for relief against actions of various implementing agencies, there is no internal or quasi-judicial forum for appeal or interim relief. Any appeal against NGOAB has to be filed in court. The court process takes its due course, and in the interim, irreversible damage can happen to the NGO as donor sources dry up and programs wind down, eroding the NGO’s image in the community. The foregoing also increases the incidence of rent seeking in some cases. Having either an appeal mechanism within the implementing agencies or a joint appeal venue prior to court appeal will help to establish transparent, cost- effective, and accessible relief to aggrieved NGOs. 4.2 Implementing agencies 4.19 The NGO Affairs Bureau was conceived as a one-stop shop for NGOs. The rationale for setting up NGOAB in 1991 was a large backlog of projects pending government approval under the existing system, which required multiple levels of government review. 103 Though NGOAB was conceived as a one-stop shop for NGOs administering foreign funds, it can never be one in reality. This is because all NGOs have to first establish themselves under the various acts discussed above that are administered by other agencies; in addition, registration with NGOAB requires the approval of the Ministry of Home Affairs, a line ministry, and, recently, approval from local government. NGOs are now required to get clearance certificates from concerned local government officials before applying for clearance of subsequent funding for a project. The certificates are not always easy to obtain, and some NGOs reported being forced to pay rent seekers for these. 103 World Bank 1996. 61 Economics and Governance of Nongovernmental Organizations in Bangladesh 4.20 NGOAB has been regarded as the primary regulator, as most development NGOs were supported by foreign funds in the 1990s. NGOAB undertakes a diverse set of functions. The main ones are registration of NGOs, approval of project proposals, release of funds, and monitoring of NGO projects. 104 Prior to registration by NGOAB an NGO must be approved by the Home Ministry and at least one line ministry—and in practice both these steps take longer than their stipulated time. NGOAB had the same number of staff in 2005 as it did in the early 1990s, yet the number of registered NGOs has tripled. 105 There are about 1,200 NGOs currently receiving foreign funds, though almost 1,900 are registered with the NGO Affairs Bureau. At present, approximately 150–250 projects are inspected annually following complaints registered with the bureau. 4.21 Accounting and audit requirements exist, but the NGO Affairs Bureau seldom assesses the quality of these audits. Under law, organizations registered with NGOAB are expected to provide annual reports and audits with regard to the projects funded. Unless these are accepted and approved by NGOAB, subsequent applications for additional fund clearance for the same project will not be released. The amended 1978 ordinance states that only one bank account can be maintained for receiving foreign funding. Accounts are to be maintained following NGOAB guidelines and books of accounts must be audited by a chartered accountant selected from a list of audit firms eligible to audit NGOs. This list is not particularly selective, as around half of the 264 registered auditing firms are authorized to conduct audits by NGOAB. 4.22 The Department of Social Welfare registers and deals with the NGOs whose purpose is to render welfare services as defined in the Voluntary Social Welfare Agencies (Regulation and Control) Ordinance 1961. There were a staggering 45,536 organizations registered with the Department of Social Welfare as of late 2004, though only a small share are active NGOs. In practice the department is only able to fulfill its registration mandate. It is unable to carry out its monitoring function, except on an occasional basis when investigations are launched following disputes within an NGO’s governing board or complaints from the public. A survey of 83 NGOs carried out in 2005 reported frequent delays and problems with registration. 106 The department also provides limited funds to NGOs to implement various social welfare programs, with a maximum grant amount of Tk 100,000. However, the monitoring and auditing of these grants is weak. 4.23 The Registrar of Joint Stock Companies carries out registration functions under both the Societies Registration Act 1860 and the Companies Act 1994. About two-thirds of the approximately 100,000 registered entities are companies, and only about 5 percent are societies. A review of 2004 records shows that around 84 societies were being registered each month, but the current recording system makes it effectively impossible to determine the stock 104 Other subsidiary but time-consuming functions include receiving information on foreign travel by NGO personnel, approving expatriate consultants, and overseeing all other matters relating to NGOs’ receipt of foreign funds. 105 NGOAB started operations with 66 staff positions, with an additional post created in 1998. There are a total of 15 officer-level staff members including the director general and directors. Five officers are involved in field-level work that entails examining projects and providing fund clearance. One officer serves as a liaison with different ministries and international agencies, and another supervises NGO registration and renewal. A total of 1,882 organizations had registered with the bureau as of August 2004. Of these, an estimated 1,100–1,200 were receiving foreign funding. About 100 new NGOs are registered every year, while the number of renewals ranges from 300 to 500. 106 Irish and Simon 2005. 62 Chapter 4: The Regulatory Framework and Financial Accountability Issues and flow of nonprofit companies. There are few professional staff in the RJSC and no clear separation of staff who deal with societies or companies. The registrar’s office has neither the funds nor the manpower to effectively follow up on registered organizations. They can ask for information and clarifications, but rarely are investigations or charges drawn. The absence of computerized systems makes it difficult to verify different information, such as how many or which organizations have submitted audited financial reports, which societies have provided the annual list of governing body members, or even the number of not-for-profit companies. 4.3 Corporate governance and financial accountability 4.24 The legislative framework relating to internal governance of NGOs, including accounting, disclosure, and transparency, is rudimentary. As discussed earlier, the Societies Registration Act includes a requirement for an annual meeting but does not require the submission of annual accounts or audits. The Trust Act requires clear accounts to be maintained, but there are no requirements for external audit. The Companies Act is the only law that contains any fiduciary responsibility rules, and not-for-profit companies face extensive bookkeeping, audit, and report-filing requirements. However, as discussed earlier, the RJSC is unable to enforce these requirements. Interestingly, the administrator of waqfs and other religious trusts also requires the submission of audited accounts. 4.25 Not only are the laws inadequate in this area, but there are notable differences in the regulatory requirements of the various laws. Some of them refer to the maintenance of accounts and books but do not give any specific guideline as to what accounting method should be used. NGOAB calls for double-entry accounting, while the Companies Act suggests the use of accrual basis. The accounting requirements in the Companies Act lack clarity and at times are incompatible with international accounting standards. 4.26 There are risks with the current structure and composition of NGO boards. Most NGOs in Bangladesh are still first-generation institutions. They face the same corporate governance problems associated with the country’s for-profit sector, as they tend to be founder- dominated and are typically governed by the founder’s friends and/or family. 107 The founder will typically head management as well as hold important board positions such as chair or secretary. These dual roles threaten to create conflicts of interest and to compromise organizational accountability. 4.27 Board and general members typically do not have term limits. Some of the more established Dhaka-based civil society personalities (such as NGO professionals, social workers, academics, and activists) tend to be represented on numerous NGO boards. While these individuals are perceived to bring a level of respect and accountability to an organization, their multiple roles raise doubts as to whether they can effectively govern that many institutions. Board elections, when they are held, appear to take place on a consensus basis, with members tending to rotate posts; this limits the scope for new members. NGO boards therefore tend to remain static and may often be reluctant to challenge decisions of the founder. This lack of understanding of the roles and responsibilities of board members can become a serious liability 107 Saifullah and Mawla 2003. 63 Economics and Governance of Nongovernmental Organizations in Bangladesh when the founder or other key individuals eventually move on and the organization must deal with issues of succession and continuity. 4.28 Financial delegation authority and payment procedures vary considerably across NGOs. Smaller NGOs tend to have board members as signatories to the organization accounts. In the case of large organizations, the board delegates authority at different levels. Certain NGOs set up purchase committees and require submission of comparative quotations for purchases above a certain amount, which can range from Tk 5,000 in one NGO to Tk 100,000 in another. 4.29 Accounting capacity in most NGOs appears reasonable though the use of technology is sporadic. The demand for accounts-related information can be very large due to the multiplicity of funding sources, projects, and areas of operation. Most organizations hire qualified personnel in the head accountant’s post; for larger organizations, trained accountants are recruited at the regional or area offices to oversee the accounts of multiple branches. Large national NGOs like BRAC and Proshika have to a large extent computerized down to the branch level, while ASA is automated centrally. The smaller organizations are more likely to have full manual systems due to resource constraints. PKSF has recently started rolling out an accounting software program for its partner organizations. 4.30 Donor financial reporting requirements are questionable. Smaller NGOs with different donors supporting separate projects have to devote more time and resources to meeting multiple reporting requirements. Another common practice of donors that raises transaction costs is their insistence that NGOs open separate accounts for their funded projects. Discussions with NGO accountants indicate that standardizing donor financial reporting formats would significantly reduce their workload. The perception of a sample of NGO accountants is that donors use a range of lax accounting principles for their reporting requirements. These include dependence on cash basis accounting, disregarding trial balance as a tool for verifying the double-entry system of accounting, and a propensity to focus only on their funded projects as opposed to consolidated accounts and audits. 4.31 PKSF is credited with introducing some standardization of financial management practices. PKSF has instituted certain international accounting standards for financial reporting and as a result, PKSF-funded NGOs are fairly uniform in their financial reporting, audits, and presentations. PKSF partners previously operated on a cash basis but have gradually moved toward accounting on an accrual basis. Their experience demonstrates that adequate capacity building and training can help NGOs attain satisfactory accounting standards. PKSF recently reduced the number of audit firms eligible for auditing partner organizations, to maintain quality. 4.32 In short, financial management practices of NGOs have been shaped by donor requirements, by absence of a coherent and uniform accounting standard within the regulations, and by lack of awareness of the benefits of adopting sound accounting practices based on international accounting standards. As a result, there is no standard basis of accounting, and NGOs follow a range of systems—cash, accrual, modified accrual, hybrid and so forth. The Bangladesh Accounting Standards are normally not applied, and differences exist in presentation, disclosures, terminology, and accounting policies. 64 Chapter 4: The Regulatory Framework and Financial Accountability Issues 4.33 Internal audit practices vary significantly by size of NGO. Some of the smaller organizations do not have internal audit teams or an internal auditor, while larger institutions with multiple branches do. The internal audit team in BRAC reports directly to the executive director and is the largest among NGOs, with 100 auditors employed. The Grameen Bank’s internal audit team is required to check whether targeting criteria for borrower identification has been followed by visits to a sample of client households. 4.34 While annual external audits are common, their quality varies significantly. Analysis of a range of audit reports shows that the quality and comprehensiveness are mixed at best. Many NGOs have separate audit reports for different projects without a consolidated one for the entire organization. Donors are partly responsible as a result of their focus on the specific projects that they finance. The auditor to be appointed and the audit fee approved in the annual general meeting. However, fees are such that audit firms often cannot afford to spend quality staff time on NGO audits. 4.35 Public access to audit reports is limited, which undermines accountability. An extensive Web search failed to find any NGO other than BRAC that posted its audit report on the organization’s Web site. While the laws under which NGOs are registered do not require public disclosure of audit reports, voluntary disclosure would improve public accountability and transparency. In an effort to promote transparency, the Washington-based Consultative Group to Assist the Poor has instituted a yearly award that recognizes microfinance institutions with exemplary financial reporting practices (see box 4.2). 4.4 Improving the institutional framework for NGO regulation 4.36 Human resource development, computerization, and improvement in procedures need to undergird any regulatory reform in the NGO sector. Training efforts so far have concentrated on building the capacity of NGOs, and there has been less focus on training regulatory agencies. This needs to be rectified. An appropriate data collection mechanism should be introduced to help in devising plans for the oversight and development of the sector. The use of complaints from the public, or deviation in use of funds from approved purposes, as the triggers for investigation by the regulatory bodies also needs to be revisited. 4.37 An independent NGO Commission that assumes the functions currently performed by the Department of Social Welfare, NGOAB, and the Ministry of Women and Children Affairs is an option that could be considered. The commission’s aims would be to maintain an effective framework within which nonprofit organizations could operate and work with NGOs to encourage good practice. The commission ought to be empowered to deal with the incorporation or establishment of NGOs, apex bodies, and licensed certification bodies. Lessons from the Charities Commission in the United Kingdom could be used as a basis for this proposed independent NGO Commission (see annex 2). 4.38 The proposed NGO Commission would need to build upon the lessons of other independent commissions in Bangladesh. Lessons from regulatory bodies in energy and telecommunications, as well as the recent experience with setting up the Anti-Corruption Commission, need to be absorbed before setting up a similar body for NGO oversight. The 65 Economics and Governance of Nongovernmental Organizations in Bangladesh proposed commission should report directly to Parliament in order to ensure its autonomy, and the commission’s board should be composed of commissioners selected from both the public and private sectors. The commission needs to be adequately funded and must be able to retain its fee and fine income. Donors could provide the initial seed capital for the commission in order to get it off the ground. It is crucial that the commission be able to determine its staffing and salary structure independently. Furthermore, it ought to comply with international good practices on governance and disclosure. 4.39 The proposed NGO Commission can encourage uniformity in accounting standards. Uniform standards of accounting (e.g., uniformity in NGO accounting principles, reporting, and presentation of NGO financial statements and audit reports) need to be introduced for the sector, with contradictory systems phased out. These standards do not necessarily have to be new ones, as there are international standards that can be applied to the local context. 108 The lead for this undertaking should be taken by the proposed NGO Commission in consultation with NGOs, donors, and accounting bodies such as the Institute of Chartered Accountants of Bangladesh (ICAB). The Securities and Exchange Commission is the country’s only regulatory body that is working to enforce accounting standards for listed companies. 109 The NGO regulatory body could play a similar role for the NGO sector and could also encourage donor harmonization of financial management requirements. 4.40 Certification bodies can reduce the regulatory burden of the proposed NGO Commission. The new commission will need to use its scarce resources to deal with big NGOs and apex bodies, and to concentrate on overall policy issues and capacity building of the sector. Thus the commission will need to be assisted by certification bodies that are privately owned Box 4.2 Microfinance Transparency Award In 2004 the Consultative Group to Assist the Poor (CGAP) announced a Financial Transparency Award for microfinance institutions. The announcement states that “Transparency is essential if the microfinance industry is to reach scale. Investors, donors and, most importantly, domestic savers, will only place their funds where they can reliably assess risk and return on the basis of standardized and comparable information. In a further effort to promote transparency, the award aims to highlight financial reporting disclosure and encourage microfinance institutions to adopt International Financial Reporting Standards (IAS1 on Disclosure) and the industry-specific CGAP Disclosure Guidelines for Financial Reporting by Microfinance Institutions. Compliance will indicate that an MFI’s financial statements contain the information necessary to fully understand performance.” This level of disclosure is also a prerequisite to calculating ratios that allow MFIs to track their performance accurately over time and to compare it with what other MFIs are doing. BURO, Tangail won the award in 2005. Information about the award is available at www.cgap.org/fintransaward/index.shtml. corporate entities with sector-specific expertise, which can be licensed to certify and sanction medium, small, and very small NGOs. This delegated legal authority is superior to the oversight exercised by membership-based self-regulatory organizations such as ADAB or Federation of NGOs in Bangladesh (FNB), as experience has shown that they have problems enforcing rules. Certification bodies can rate NGOs and provide references to donors and the public for a fee. A 108 The background paper for this chapter (Khan 2005) has a detailed review of the applicability of international accounting standards for NGOs in Bangladesh. 109 Noncomplying audits are referred to ICAB for action. The Securities and Exchange Commission allows 60 days for ICAB to look into the matter. If the steps taken are unsatisfactory, the commission has the authority to directly contact the auditors in question and to delist them from auditing other listed companies for up to five years. 66 Chapter 4: The Regulatory Framework and Financial Accountability Issues seal of approval from a certification agency can be used by the Ministry of Finance to determine the list of NGOs to which donations are tax-exempt. They can also help in facilitating NGOs’ work and relations with relevant ministries and regulatory authorities. The NGO Commission can also be the first-level appeal avenue for complaints against certification bodies. Box 4.3 discusses the experience of a certification body in Pakistan. Box 4.3 NGO certification in Pakistan The Pakistan Centre for Philanthropy’s (PCP) nonprofit organizations certification program is the first of its kind in the South Asia region, although it is based on similar initiatives elsewhere, particularly one in the Philippines. The objective is to help nonprofit organizations diversify their resource base and expand their programs while at the same time demonstrating their commitment to best practices in governance and management. Certification is a “seal of good housekeeping” for organizations that meet defined standards of organizational effectiveness based on an independent evaluation of their governance, financial management, and program delivery. PCP staff spend two to three weeks reviewing an NGO’s documents and examining its operations before deciding whether or not to recommend the NGO to the certification panel that is made up of a majority of private sector representatives along with two government officials. The process is voluntary and does not come under the control of any government ministry, and organizations that do not qualify the first time but would like to try again are offered help to create and implement an improvement plan. Certification is recognized by the authorities, and NGOs that are certified also obtain not-for-profit tax- exempt status from the Central Board of Revenue, something that must be renewed every second year. While the certification process is primarily aimed at improving NGO performance and transparency so that donors can better judge which NGOs are reliable, it has the added effect of heightening the credibility of NGOs in general and thus improving the reputation of the NGO sector. Although the process only began last year and is entirely voluntary, it has already attracted the better NGOs as no one wants to be left behind. More information can be found on the PCP’s Web site at www.pcp.org.pk. 4.41 The proposed commission would also require powers to deal with troubled NGOs, including dissolution through proper process, due process for sanctions, and installation of appropriate appeal procedures for all NGOs, apex bodies, and certification bodies. The law needs to also introduce conflict resolution mechanisms for NGOs in disputes with other NGOs, donors, government, and clients. 4.42 The proposed certification system would be voluntary and would focus on creating incentives for better corporate governance and on using moral suasion rather than legal enforcement. In order to increase incentives for certification, donors could request certification before releasing funds above a certain threshold. Certification bodies can finance themselves from public donations and fees from NGOs certified by them. A certification body ought to be required to report to the commission periodically on its activities and compliance, refer noncompliance with commission regulations to the commission for action, and refer conflicts that cannot be resolved at its level to the commission. A certification body’s license can be revoked if there are deficiencies in its practices. 4.5 Improving the laws relating to NGOs 4.43 The focus of regulation ought to be shifted to fostering better governance and accountability in the NGO sector, rather than micro-managing foreign funds. All laws and regulations related to the NGO sector need to be harmonized and streamlined to create a uniform regulatory framework that avoids duplication, overlapping, and contradictory requirements. This 67 Economics and Governance of Nongovernmental Organizations in Bangladesh does not necessarily mean that existing laws and regulations should be repealed and replaced with a new, all-encompassing statute. Instead, the existing laws can be modified by incorporating standardized requirements for board structure, holding of annual general meetings, financial reporting, accountability to regulatory agencies and so forth. The framework needs to move away from the stereotyped welfare approach to one that reflects the present-day scope of development activities. Arbitrary powers of the state that allow it to interfere with NGO structures (e.g., by dissolving or suspending the governing body and replacing it with an administrator) should be removed, and the right to legal appeal against government actions should be instituted. Finally, the process of registration needs to be simplified and the steps made publicly available. 4.44 In the process of legal reform, a number of key principles need to be ensured. Generally, broad discretion should be given to NGOs to structure their organization in a way that suits their individual goals and purposes. NGO activities in Bangladesh span a number of activities, and it is important that the law not be written with a particular group in mind, in order to allow for flexibility in application. It should feature tiered regulation based on the size and activities of NGOs. All penalty provisions should be structured so as to provide a gradation of sanctions, with fines and other intermediate sanctions being applicable before the ultimate penalty of board suspension or dissolution can be used. Administrative discretion should be reduced, reasons for imposition of penalties should be spelled out more clearly, and judicial appeals should be possible to contest any adverse ruling or penalty. 4.45 In consolidating the laws for the formation of NGOs, one could merge the definitions of “social welfare agency” and “organization involved in voluntary activity” into a single, new definition of “public benefit organization” (PBO). These can be defined as entities organized and operated exclusively for public benefit purposes, such as education, poverty, health, environment, art, science, and so forth. In this context, the definition of an organization as a PBO ought to be based on its purposes and activities, without regard to whether its funding is foreign or domestic. Certification as a genuine PBO by the type of professional certification body described above should be a condition for receiving significant tax and other benefits (such as foreign grant eligibility, tax exemption, and deductible donations). 4.46 Apart from direct electioneering, a broad range of political activities ought to be permitted. Banning NGOs from the wide range of activities that could possibly be defined as “political” would not be appropriate in a country that is democratic, the main exception being the funding or support of (in favor or against) candidates for public office. These restrictions reflect a balance between the guaranteed rights to freedom of speech, expression, and the press and laws regulating elections, political parties, and campaign funding. NGOs in other countries are generally prohibited from electioneering and campaign funding, and limits are often placed on direct and grassroots lobbying too. 4.47 The proposed new law overseeing microcredit activities could provide the basis for regulatory reform of the rest of the sector. The microcredit sector in Bangladesh lacks the regulatory framework that an industry of its size merits. As discussed in chapter 2, a laissez-faire regulatory framework has contributed to the growth of the sector. However, now that the sector is past its infancy and close to $500 million of savings has been mobilized from poor households, 68 Chapter 4: The Regulatory Framework and Financial Accountability Issues it is necessary to modernize the laws governing microcredit activities. Currently only the Grameen Bank has a special law to govern its activities, because it is registered as a specialized bank. A multi-stakeholder committee has developed a legal framework for NGOs involved in microfinance and this is currently under the government’s consideration. Following on from the committee’s work, the Microfinance Research and Reference Unit (MRRU) within the Bangladesh Bank has been collecting key data from NGO microfinance programs, and building their capacity, in order to ease the transition to a new regulatory framework. 110 It may well be the case that this new microfinance law will cover a large number of key lacunae in the legislation governing the overall NGO sector if it is enacted. 4.48 Modern procedures for dealing with distressed NGOs and the mergers and exit of NGOs need to be developed. An enabling legal framework that permits the merger of NGOs will help smaller NGOs to continue their mission by joining forces with bigger ones. In the long run, economies of scale could help to improve governance, accountability, and disclosure of the smaller and medium-size NGOs. This will also help in the oversight of the sector. Similarly, a legal mechanism for managing distressed NGOs and closure of NGOs will help to improve the development of the sector. 4.49 The Societies Registration Act ought to be amended based on the current state of the sector and in line with international good practices. The new law would need to address, among other things, issues such as (a) categorization of different types of NGOs; (b) regulatory exemptions of small NGOs that are not engaged in microfinance, with an option to bring them under regulation when certain triggers go off; (c) permitted and prohibited activities of NGOs; (d) internal governance requirements; (e) audit and accounting requirements; (f) tax treatment; (g) appeals, sanctions, and penalties that are commensurate with offences; and (h) transition provisions for exiting societies. 4.50 In addition to the Societies Registration Act, other laws would need to be repealed and amended. A number of changes are suggested. The Trust Act 1882 could be amended to update it on duties of trustees and failure to comply with fiduciary duties, incorporating case law developments and bringing it under the purview of the NGO Commission. The Companies Act 1994 could be amended in line with main NGO law. The Wakf Ordinance 1962 and other religious trust laws could be reviewed to modernize and improve governance. The Voluntary Social Welfare Agencies (Regulation and Control) Ordinance 1961 and the Foreign Donation (Voluntary Activities) Regulation Ordinance 1982 could be repealed. 4.51 Bangladesh needs to review its tax regime for NGOs and clarify the scope and applicability of exemption. In line with international good practices, NGO income in the form of donations, grants, membership dues, and microcredit revenues ought to be exempted from taxation. However, passive investment income and income from an active trade or business must be taxed in order to create a level playing field with other private sector competitors. Deductions for donations to NGOs have recently been reintroduced in the 2005 budget, though the process of 110 The MRRU is collecting annual balance sheet and income statements and information on program size, areas of operation, loan and savings balances, management, and staffing composition every six months. The first analytical report based on this information is posted on the MRRU Web site (http://www.bangladeshbank.org.bd./pub/annual/mrru/ngomfisbd.html). 69 Economics and Governance of Nongovernmental Organizations in Bangladesh granting these exemptions could be publicized more and in the medium term devolved to professional certification bodies. 4.52 The Income Tax Ordinance 1984 has made it mandatory for all NGOs working in Bangladesh to submit income tax returns. However, many small and medium-size NGOs assume that their not-for-profit nature exempts them from this requirement. Moreover, whether the income of NGOs is exempted from tax or not is decided only after submission of their returns. This problem is further compounded by the opaque rules and procedures regarding taxation and tax exemptions for not-for-profits, as discussed earlier. Hence, the tax laws related to not-for- profits need to be simplified and compiled as a separate section, with focus on providing greater incentives for local resource mobilization for the sector. There also needs to be greater awareness and enforcement of the requirement that NGOs file income tax returns. 4.6 Improving corporate governance and financial accountability practices within NGOs 4.53 Based on our earlier analysis, we suggest a number of broad areas where corporate governance and financial accountability of NGOs can be improved. These measures internal to NGOs complement the proposed legislative and regulatory improvements to strengthen financial accountability that were covered above. They are in many respects applicable to the corporate sector as well, as they address weaknesses that exist in both the for-profit and not-for-profit sectors. 4.54 Boards and general bodies should ideally be constituted with a diverse group of individuals bringing different skills and expertise. The size of the board is an important consideration for effective decision making: smaller boards may not engage in sufficient debate, while large boards can be unwieldy and ineffectual. 111 There should be term limits on board members to allow for rotation. 112 During the early stages of an NGO’s development it may be necessary for family members to be present on the board to maintain unity of purpose and reduce the chances of conflict. However, as an NGO grows in size, it is important to allow non-family members to join the board. Management and board should ideally be separated so that the chair (board) and the executive director (management) posts do not overlap. Board members should be oriented on their roles and responsibilities and should understand the consequences of their decisions. 113 Involvement of an organization’s beneficiaries on the board and/or in management and policy decision making can be beneficial, helping the NGO better understand the needs of the clients it purports to serve. This would also enable the organization to be more accountable to its constituents. 4.55 NGOs should develop management and financial policies, service rules, and reporting and documentation guidelines at an early stage of their development. Financial policies should provide details on accounting convention, depreciation, revenue recognition, taxation, recognition of government grants, loan provision, accounting for expenses and so forth. 111 Bangladesh Enterprise Institute 2004. 112 For organizations such as privately funded family foundations or family trusts, the issue of family/friends membership or term limits may not be appropriate. If, however, such organizations become dependent upon or start attracting public funding, than there is a stronger case to be made for the above restrictions. 113 For example, PKSF loan conditions require that three board members act as guarantors. Failure to repay may result in the seizure of their assets, even if they are registered with limited liability. 70 Chapter 4: The Regulatory Framework and Financial Accountability Issues Procurement policies and the roles and responsibilities of audit committees and internal audit teams need to be detailed. There should be budget allocation for capacity building of employees, including accounts staff. All of these guidelines must be available and communicated to all staff. NGOs should invest in automating their accounting systems as the benefits of computerization include increased efficiency, easier information sharing, and a reduction in duplication, errors, and fraud. 4.56 Building the capacity of the NGO sector in accounting and auditing is essential. PKSF’s experience of training its partners will be of use to apex bodies like the FNB that are planning to run similar programs. It is advised that resources be allocated for hiring accounting professionals or that bodies such as ICAB be involved in this process to ensure due diligence in designing the training programs. Lastly, ICAB or another similar institution will need to play a more active role in developing trained professionals to effectively serve the not-for-profit sector. 4.57 Annual audits and financial transparency can help to strengthen public accountability. Public misconceptions about NGOs would be limited if there were a consistent set of disclosure requirements and greater information on NGO programs, funding, and expenditures. The laws under which NGOs are registered do not require public disclosure; nevertheless, ensuring uniform public access to NGO audit reports would be an important part of strengthening public accountability. In cases where NGOs handle public savings deposits, obtain financing from the Bangladesh government, or receive benefits such as tax-deductible donations from citizens, public access to audit reports should be mandatory. This could be done by posting the audit report on the organization’s Web site, making copies freely available in the head and branch offices, making copies available at public libraries and so forth. 4.58 A next step could be to establish a committee with a mandate to modernize legislation governing NGOs. The members of the committee would need to be carefully selected from a pool of experienced professionals who have the necessary experience and understand the NGO sector well. This high-level committee could be tasked with building upon the analysis in this chapter by undertaking a detailed review of the relevant laws and proposing new legislation and amendments to existing laws. One option could be to have this committee be the same as the one charged with developing the microfinance laws; the latter already includes government and industry representatives, though its composition would need to be broadened to include non-microcredit practitioners. Since the new microcredit law may well be approved in 2006, this proposed committee would need to weigh the advantages of changing existing laws or creating a new NGO law to address issues that are not covered by the microcredit law. 4.59 There are risks associated with amending laws and setting up new oversight institutions. The main risk is that overregulation or poorly designed regulation may stifle the NGO sector. As the regulatory framework is modernized, the first objective should be to enhance the space given to NGOs to innovate and scale up their activities. Regulations should be “facilitative,” and excessive or poorly designed regulation should be avoided at all costs. These risks can be mitigated if changes are made in a fully consultative manner, as suggested above, and various options are explored with adequate time and resources provided to evaluate these choices. A second risk concerns the proposed independent NGO Commission. Independent regulatory commissions in Bangladesh have thus far had a mixed track record, with difficulties 71 Economics and Governance of Nongovernmental Organizations in Bangladesh in recruitment, funding, and mandates. Lessons from these experiences need to be absorbed before setting up a new institution. 72 Chapter 5: Scaling Up through a Strategic Compact 5.1 Despite the striking progress made on many fronts, there remain large development challenges and unmet needs in Bangladesh. For instance, while infant mortality has declined, neonatal mortality is proving stubbornly hard to bring down. Strides have been made in school enrollment and gender parity in primary education, but dropout rates among the poor at the early stages of secondary school remain high. 114 Microcredit has reached vast numbers of poor households nationwide, and yet there remain significant pockets of the country with limited access to financial services, particularly ones targeted to the poorest. While female empowerment indicators have undoubtedly improved, due in part to NGO activity, significant social challenges still confront women in both the domestic and public spheres. 5.2 Hence, there is a clear need to provide improved services to the poor and promote access to existing services. Chapter 2 showed that NGOs are playing a significant role in service delivery and advocacy and that these interventions are generally well targeted, with a positive impact on clients. However, the wider “macro” impact has been curtailed by the relatively “micro” nature of some of these activities. For instance, the growth and employment impact of NGO loans has been limited by the small size of the loans. Another example is that the NGO impact on neonatal mortality is constrained by the relative lack of specialized post-delivery care. 5.3 That said, an unchecked future expansion of NGO services across the board may not necessarily prove the best means of ensuring pro-poor service delivery. There are a number of reasons. First, questions may be raised as to whether such expansion will limit the capacity of government to deliver these services in the long run and whether it will weaken the accountability relationship between provider and client. Second, advocating an expansion in NGO services leads to questions about the capacity of NGOs to deliver quality services while retaining their focus on the needs of the poor. Third, and related to the capacity issue, are questions of how growth in NGO programs can best be managed, particularly in light of the weaknesses in financial accountability and traditional contracting mechanisms already discussed. This chapter examines these issues by drawing on the findings from the earlier chapters in an effort to answer the questions: “Is further NGO growth desirable, is it possible, and how is it best managed?” 5.1 Does further NGO growth make sense from a public policy standpoint? 5.4 The Bangladesh Poverty Reduction Strategy Paper articulates the role of NGOs primarily in the context of improving service delivery. 115 The issue is whether the services currently being provided are at an optimal level in terms of size and type, or whether there is a case for a change in scale or mix of activities. How does this relate to the role of central and local government? In certain cases it may be better for NGOs to be facilitators of better service provision by the public or the for-profit private sector, rather than direct providers of services. Given the broad nature of 114 World Bank 2003b. 115 Government of Bangladesh 2005. Economics and Governance of Nongovernmental Organizations in Bangladesh these questions, we focus this discussion in turn on each of the key activities delivered by NGOs, and then draw some general conclusions. 5.5 In microfinance, we believe that the role of government ought to be in strengthening the regulation and supervision of the sector. In line with best practice, the government or dominant funding agencies such as PKSF should not attempt to set final borrower interest rates but should leave that decision to each individual NGO. Given the questionable sustainability of government microfinance programs and the clear comparative advantage of NGOs in this area, we argue that government ought to move away from the direct provision of microcredit. Once an appropriate regulatory framework is established, the larger NGOs could convert into specialized microcredit banks. 5.6 The analysis in chapters 2 and 3 points to a number of strategic directions for microcredit NGOs: • Microcredit programs ought to focus on targeting geographic areas, particularly in northwest Bangladesh, where access to credit is limited. Carefully designed subsidies may be required to generate the incentives to set up microcredit branches in areas where operating costs are likely to be higher than the norm. • The microcredit industry ought to scale up successful experiments that provide microcredit to the poorest. There will remain a group of extremely poor households who do not wish to borrow to sustain their livelihoods and for whom noncredit interventions are more appropriate. However, for a significant share of households, adapting the current microcredit model, and perhaps combining microcredit with other services, will lead to welfare gains. • NGOs that provide microcredit ought to continue with their judicious entry into providing larger loans. Many NGOs have already ventured into this market, though with a few exceptions the number of larger borrowers is still relatively small. While from a public policy standpoint there is a temptation to scale up these programs rapidly, given the potential growth-generating impact, we argue that the current cautious expansion strategy is the right one. Providing larger loans has its benefits but also involves considerable risks. There are a limited number of former microcredit borrows who want to “graduate” into this market, and new wealthier clients do not have a credit history with the lender and are less likely to be socially pressured into repaying. 5.7 The education sector currently has a very different mix of government and NGO provision when compared to the microcredit sector. In contrast to microcredit, where the NGO sector dominates the market, education has limited NGO involvement. Students in NGO schools make up less than 10 percent of primary enrollment (and as chapter 2 shows, these are largely poor children). This figure has by and large remained constant over the past decade. 5.8 We showed in chapter 2 that learning outcomes in NGO schools are superior to those in government primary schools, and that the transition rate of NGO students to government secondary schools is high. Based on these findings, we argue that in order to achieve 74 Chapter 5: Scaling Up through a Strategic Compact Bangladesh’s Millennium Development Goals in education, an expansion of NGO primary schools is warranted. The PRSP states that “access is low for ethnic and linguistic minorities and people living in char-land, lowland, baor, and coastal areas where there is a disproportionate concentration of the extremely poor.” 116 NGO schools could focus on these groups and on reaching the large numbers of poor students who do not enroll, or who drop out of school. In areas where both government and NGO primary schools exist, a voucher system could be piloted to provide consumer choice and strengthen the short route to accountability between provider and client. This expansion of NGO schools could be done by transferring poorly performing government primary schools to NGOs (for which there is precedent), or by financing the construction of new schools, an option discussed later in this chapter. Another approach is to have government or community management of schools, with NGOs acting as facilitators to ensure targeting and teaching quality. The recent Reaching Out of School Children project targets the enrollment of 500,000 poor children, with money going to communities that can contract NGOs to help in running schools. 5.9 At the same time, the PRSP sets forth the government’s intention to create space for NGOs and local communities in pre-primary schooling and calls for NGOs to make a strategic shift into these areas. Whether this shift comes at the expense of or in addition to a growth in NGO primary programs will depend on individual NGOs’ preferences and financing options. Finally, with respect to post-primary education, NGOs could provide financial incentives for NGO-trained primary graduates to remain in secondary schools and could also leverage the experience gained in primary provision to disseminate best practices in community involvement, personnel management, and teaching methods. The PRSP also clearly articulates the role of NGOs in raising literacy levels by providing and supporting continuing education programs. Experiences from Nepal and other countries illustrate the merits of community management of schools and educational facilities. As we look ahead, the role of NGOs in supporting school management committees and community involvement can pave the way for handing over educational assets to communities, thereby strengthening accountability mechanisms (see box 5.1). 5.10 In the health sector, the discussion in chapter 2 highlighted the very different ways that health services are delivered by government, NGOs, and private providers. The main strength of the NGO system is the vast number of village-based community health workers. These workers raise awareness of health and hygiene issues among members of individual households—a crucial role, as the achievement of health outcomes is crucially dependent on such awareness, especially among women. Apart from awareness raising, NGO workers are critical intermediaries between government and private facility-based services and poor clients. In assessing the role of NGOs in primary health care and nutrition services, the draft PRSP calls for scaling up these initiatives and learning from their experiences. While reviewing the fact that immunization coverage expanded from less than 1 percent of the population in 1981 to 60 percent in 2000, the draft PRSP points out that “the immunization programme in Bangladesh is by and large a success story largely because of the collaborative effort with NGOs.” 117 116 Government of Bangladesh 2005, 114. Chars are riverine islands and baors are a type of wetland. 117 Government of Bangladesh 2005, 122. 75 Economics and Governance of Nongovernmental Organizations in Bangladesh Box 5.1 The transition from NGO to community management While the prevalent NGO model is one of direct service provision, there are also notable examples of NGOs building the capacity of communities to manage their own resources. The Rangpur Dinajpur Rural Service (RDRS), a prominent NGO operating in northwestern Bangladesh, has set up a unique federated group structure. At the lowest tier, “primary groups” consisting of around 20 poor individuals are set up and receive a range of services directly from RDRS. Following an evaluation of their effectiveness and initiative levels, these primary groups may be allowed to transition to the second tier. The members can then join other groups from their own or nearby villages to form a union-level federation. RDRS has consistently promoted the emergence of these apex bodies or umbrella organizations, and has encouraged them to gradually take over management tasks from RDRS. The union federations represent the interests of primary groups, provide support services, and act as vehicles for carrying forward development activities by and for the poor. At present there are 260 active union federations, each run by an elected office holder and made up of elected representatives. The federations are required to have written constitutions, hold regular committee and annual general meetings, plan and budget their activities, and carry out annual audits. Federations that have proven to be successful and sustainable can move up one step further and graduate as autonomous bodies no longer reliant on RDRS for resources and support. With encouragement from RDRS, 100 federations have registered separately with the government’s Social Welfare Department and are recognized as independent NGOs. This has given them legal recognition for their development work as well as the right to receive resources, cooperation, and advice from the Social Welfare Department. This recognition also generates enthusiasm, interest, and confidence among the members. Increased interaction with the government means there is a greater opportunity for the voices of the poor to be heard on a wide range of issues. Community management of schools and related educational facilities is another way of strengthening the links between provider and client. In its adult literacy and continuing education program, BRAC has established over 800 learning centers (libraries with information technology facilities) on the premises of government secondary schools or at the center of a union (the smallest administrative unit). In order to assess a community’s interest in this facility BRAC requires a community contribution and the formation of a committee that will oversee the center. Within two years, a learning center is transformed into a registered autonomous trust. As of early 2004, almost 80 percent of learning centers had been transferred from BRAC to the community through these trusts. 5.11 Looking ahead, the appropriate role of government, NGOs, and the private sector in health ought to be shaped by efficiency and equity arguments as well as by the ongoing and emerging needs in the sector. For instance, health indicators show that infant mortality—and more specifically neonatal mortality, which accounts for about two-thirds of infant deaths— remains a key problem. 118 Critical interventions to address this problem include ensuring access to emergency obstetric care, increasing skilled attendance at birth, and improving awareness of postnatal care. As the PRSP suggests, vouchers could be issued and made redeemable at both private and government clinics to increase the use of modern facilities. NGOs could raise awareness of postnatal care and improve the skill levels of their community health workers so that they are better equipped to help during home deliveries and can act as facilitators by referring pregnant women to specialized care. 118 Infant mortality refers to deaths in children up to one year of age; neonatal mortality is in children up to 28 days of age. 76 Chapter 5: Scaling Up through a Strategic Compact 5.12 Malnutrition remains stubbornly high in Bangladesh, with 45 percent of children under age 5 stunted and 48 percent underweight. 119 Interventions that play to the strengths of community-based health workers, such as promotion of breastfeeding, growth monitoring, some nutritional supplementation, and awareness of better nutritional practices, ought to be scaled up by NGOs in partnership with government along the lines of the National Nutrition Programme. 5.13 Another area where there is a clear public good and a comparative advantage to be gained in scaling up NGO activity is control of communicable and vector-borne diseases. The large network of community health workers will be able to promote mass awareness of HIV/AIDS, malaria, and tuberculosis detection and cure. NGOs are currently operating the Directly Observed Treatment, Short-Course (DOTS) program for tuberculosis in over half of the country’s subdistricts, registering cure rates that are higher than those achieved by government health complexes. This suggests that this type of program ought to be contracted out to NGOs to implement on an even wider scale. 5.14 In short, there are existing health programs, such those related to tuberculosis, nutrition, family planning, and immunization, where the case can be made for even more scaling up of NGO activity, as argued by the PRSP. However, there are also emerging areas where NGOs can at best play a facilitating role. One such area is neonatal mortality, where NGOs can help ensure access to specialized emergency obstetric care that can only be provided by either the government or the private sector. 5.15 Turning to advocacy and social programs, the discussion in chapter 2 illustrates the difficulty in measuring the impact, let alone the cost-effectiveness, of these programs. Nonetheless, whatever limited evidence there is suggests that awareness of issues such as corruption, violence against women, land rights and so on have improved due to the work of advocacy NGOs. Moreover, on a more local level, advocacy for pro-poor access to services and efforts to promote participation of poor people in decision-making bodies have also been successful to a certain extent. RDRS’s effort to encourage communities to federate and network directly with local government officials to access resources is another model of facilitating pro- poor advocacy (see box 5.1). The PRSP endorses the active involvement of NGOs in improving governance, specifically in promoting social awareness against corruption, promoting legal literacy among the poor, and improving information flows. We argue that all these activities constitute an integral part of a well-functioning democracy and that the one area that NGOs ought to stay away from is involvement in partisan politics. 5.16 The discussion above highlights some of the areas where NGOs could consider focusing their activities over the medium term, either as direct providers or as facilitators of government, private sector, and community provision. As part of this process, NGOs could consider exiting, or limiting their activities in, certain areas in order to maintain the focus on areas where their comparative advantage is greatest. These decisions will vary from one NGO to another and over time, and are therefore internal management decisions that need to be made periodically. 5.17 We now turn to the capacity of NGOs to implement this strategy of a focused scaling up by assessing NGO leadership, incentives, capacity building programs, and organizational norms. 119 Government of Bangladesh 2005. 77 Economics and Governance of Nongovernmental Organizations in Bangladesh 5.2 Is further NGO growth possible? 5.18 Size matters in the evaluation of NGO performance in Bangladesh, as the scale of service delivery that has marked Bangladeshi NGOs to date is unprecedented. In the late 1980s, “scale was the stick that NGOs were most often beaten with . . . they could never be more than small exemplars of good practice and innovation.” 120 By the 1990s, at least, this could not have fairly applied to NGOs in Bangladesh. This was a group of organizations whose impact lay significantly in the sheer numbers of poor people they reached. This section focuses on what has been learned about the organizational factors underlying success in the implementation of NGO programs, whether these provide a basis for further growth, and how any further scaling up ought to be financed. 5.2.1 Management, incentives, and systems 5.19 Concerns about domination of the sector by the founding generation of NGO leaders may have been overstated. Leaders of the prominent NGOs are in most cases the original founders of their organizations, and they make convincing advocates of their organizational philosophy. Where the original founders are no longer the leaders, they are likely to have been replaced only recently, and in some cases they have retained a managerial or governing presence in the organization. Charismatic leadership was vital in the development of the country’s NGOs and undoubtedly contributed to the expansionary phase in a number of ways. Particularly in the early days, leadership skills helped inspire staff to believe in the possibility of change, motivated staff to perform at high levels, and provided organizations with a sense of collective purpose. The ability of NGO leaders to attract the support of donors and other important audiences, such as government officials, has also been vital. 121 5.20 Concerns are frequently voiced about the domination of the big Bangladeshi NGOs by their founder-leaders. 122 A recent study raised fears that this domination would create difficulties for Bangladeshi NGOs in the transition between leaders, and some NGOs have already floundered in the handover from founder to second-line leaders. 123 It is evident that many NGOs will face turbulent times as they are forced to rely more on the soundness of their formal management structures and processes than on a single individual’s leadership. However, comparative analysis of management practices in South Asian NGOs shows that the link between strong, personality-driven modes of leadership in Bangladeshi NGOs and undesirably authoritarian forms of management has been highly exaggerated. 124 The study acknowledges the existence of hierarchies in South Asian NGOs, but finds that leaders in successful NGOs are characterized by their ability to listen to those below them in the management hierarchy and to respond to what they hear; they are skilled at managing multifunctional teams and at delegating 120 Smillie and Hailey 2001. 121 NGO leaders needed to be able to present convincing visions and new ideas about how to “do” pro-poor development in ways that appealed to donors who may have sought to promote radical champions of the poor, but who also needed formal reporting procedures and accounting (see Sanyal 1991; Stiles 2002). In the early years it was also vital that the same individuals could transmit their vision and organizational goals to senior government officials for approval or support (see Yunus 1998, 168–70). 122 See Smillie and Hailey 2001. 123 Siddiqui 2001. 124 Smillie and Hailey 2001. 78 Chapter 5: Scaling Up through a Strategic Compact to others in decentralized operations. Patterns of NGO growth also suggest that qualities of charismatic leadership are less likely to be vital to future expansion, given that the organizational capacity to replicate with an NGO branch manager—a “franchise holder”—may be more significant than the capacity to innovate. 125 5.21 NGO salaries and incentives cannot compete with the far greater attractions of public sector employment. While it is a common perception that NGO staff enjoy higher salaries and better perks than government staff, there is little evidence to support such a claim. The situation with respect to doctors is illustrative. A study of medical professionals in the public sector found that the average total income was $460 per month, with more than half of respondents earning more than their government salary from private practice on the side. 126 By contrast, the consolidated salary of a doctor in the NGO sector ranges from $180 to $300, including an allowance for not engaging in private practice. The story with teachers is similar. Teachers in the government primary system receive an average of $45 a month (plus an additional 75 percent in allowances and benefits), compared with $20 a month for equivalent teachers in NGO primary schools. 5.22 Cross-country analysis of public sector performance suggests that the wider the gap between the highest- and lowest-paid employee (“salary decompression ratio”), the greater the incentives for better performance. The 10:1 ratio in the public sector is similar to that in most NGOs. A recent survey of 62 NGOs showed that around 60 percent had a compression ratio that varied from 8:1 to 10:1, and only 15 percent of NGOs had a ratio greater than 10:1. 127 NGOs report high staff turnover levels, with staff leaving to enter public sector employment, which remains attractive in large part because of the prestige associated with it, the scope for additional earnings, and the security of employment. 5.23 NGO management of human resources in the health sector presents some difficult challenges, as well as promising prospects. A health facility survey conducted in 2004 found that while most NGO medical staff (doctors, nurses, and midwives) were “happy” with their jobs and found the workload acceptable, about half were dissatisfied with their salaries and benefit levels. 128 More than two-thirds of medical professionals in NGO health centers also felt that the potential for career development and promotion in their organization was poor. Not surprisingly, medical staff turnover rates in NGOs are high, mainly as a result of poor promotion prospects and the lure of public sector employment. At the same time, the success of individual campaigns—child immunization, tuberculosis treatment, and oral rehydration therapy, for example—that have involved NGO frontline community health workers suggests there may be lessons about how to structure incentives to improve performance among public sector medical professionals (see box 5.2). 5.24 There is little evidence to support the assumption that NGO staff performance is linked to higher levels of altruism. Research on women working in NGO rural branches and government credit programs explicitly rejects the notion. Fieldworkers perceived their work as 125 See subsequent section on the franchising system. 126 Cited in World Bank 2003b. 127 This survey was conducted as background work for the Irish and Simon (2005) study. 128 AHI 2004. 79 Economics and Governance of Nongovernmental Organizations in Bangladesh “just a job,” and reported that altruism and progressive motivations played a very small role in decisions to work in rural development. 129 First-generation NGO workers are more likely than their successors to have been inspired by commitment to altruistic or progressive goals. Unemployment rates among secondary school and university graduates—the groups from whom the majority of NGO staff are drawn—are such that the motivations for entering NGO employment may very well reflect the lack of alternatives, instead of or in addition to altruism. 130 5.25 Direct participation—or the perception of being directly involved—in organizational decision making may help keep commitment high. In Nijera Kori, for example, salary levels and benefits are known to be lower than in other NGOs. However, staff members’ motivation is enhanced because they acquire a sense of “belonging” to the organization and a stake in organizational goals through frequent face-to-face meetings and systematic information flows up and down the system, which ensure staff are informed of and contribute to the direction of programs. 131 NGO microfinance staff are encouraged to feel that their work is “special” by NGO efforts to share messages about the success, household impact, and wider importance of microfinance. 5.26 Close monitoring of employee performance and the ability to dismiss employees for poor performance offers the most convincing explanation of NGO employees’ superior performance. Regular and systematic monitoring appears to help explain why NGO schoolteachers are more likely than better-paid government primary teachers to show up for work. School supervisors visit regularly, and they in turn are monitored by higher-level staff. Monitoring of employee performance is facilitated by systems that highlight poor performance. This is clearest in the microfinance sector, with its simple and uniform accounting practices: poor performance will quickly show up through higher default rates. Campus-style living arrangements also lead to “peer monitoring” among staff, with the immediate supervisor (the branch manager) also sharing the same facilities. 132 Finally, the ability to dismiss employees relatively easily, without having to deal with powerful unions or cumbersome dismissal procedures, is an important incentive to perform well. 5.27 Staff training and linking pay with knowledge retention help develop staff capacity. NGO training programs vary significantly across organizations in terms of delivery, scope, and emphasis. Certain NGOs have their own training centers, while other organizations appear to place less emphasis on formal training and more on programs of immersion or on-the-job training through which the young middle-class graduates typically employed by NGOs gain early experience with village life. 133 Apex bodies such as PKSF also offer a range of training courses to various levels of NGO staff. A large part of the training provided by NGOs and apex bodies is training of trainers, which helps disseminate knowledge to wider groups of staff as well as to clients. There are also courses that have been designed to build the capacity of senior and mid- 129 Goetz 2001. 130 The Labour Force Survey for 1999–2000 (Bangladesh Bureau of Statistics 2002) reported unemployment among those with a secondary or higher secondary certificate at 11.5 percent, and with a college degree and above at 7.8 percent. Unemployment figures are generally believed to understate the situation. 131 Kabeer 2003. 132 Jain and Moore 2003. 133 See, for example, the description of Grameen Bank’s training methods in Yunus (1999, 160). 80 Chapter 5: Scaling Up through a Strategic Compact Box 5.2 Managing NGO health professionals: Are there lessons for the public sector? One reason frequently cited to help explain why NGO health programs appear to be effective is that staff—mainly community health workers—are based or recruited locally, unlike government doctors who tend to live at a distance from their workplace. However, local community membership is not always incentive enough. One study found that in the government system, family welfare visitors recruited from the local community were reluctant to leave the health facility to perform their outreach duties in villages (Chaudhury and Hammer 2003). More convincing is the argument that NGO staff performance is influenced by stricter monitoring systems and the incentive structures these put in place. Both regular and random supervisory visits by higher-level staff to hear views from the community on the quality of services provided ensure that problems like absenteeism are minimized. At the same time, monthly meetings with national or regional staff provide staff in the field with opportunities to air their concerns and make suggestions to improve management and performance. NGO health programs have also pioneered financial incentives to improve staff performance. Success in generating knowledge about the use of oral rehydration therapy for the treatment of diarrheal diseases is linked to a system in which BRAC community health workers who trained women to use ORT were compensated on the basis of how well women had retained and used the new knowledge. Similarly, high success rates in tuberculosis treatment completion and recovery through the DOTS program have been attributed to the system by which community health workers receive half of a Tk 100 bond payment when patients complete their treatment. level management. These are critical, as most NGOs prefer to have experienced former fieldworkers in management positions with relatively few lateral entrants recruited directly from management positions elsewhere. Finally, incentives to retain the knowledge learned matters, as box 5.2 illustrates. NGO programs have used knowledge retained by clients, and by implication knowledge retained by staff in training programs, in performance-related pay schemes in the health sector. 5.28 The staff salary bill may grow with an aging staff and more complex programs. Experience from the Grameen Bank is illustrative. Grameen realized that its staff costs would continue to increase unless more staff turnover could take place, and it introduced a scheme offering incentives for staff to leave after 10 years. This can work for mainstream microfinance since it is not difficult or expensive to induct and train a significant number of new fieldworkers every year, though experience becomes more valuable at the middle management levels and higher. But as NGOs mature and as they add products, management costs will inevitably rise, and high staff turnover at the field level might not be so desirable. 5.29 Clearly, capacity across NGOs is far from uniform, though small NGOs can be a source of innovation. Most NGOs are small and have limited managerial and staff capacity to expand on a large scale. While small NGOs are likely to be able to refocus their activities along the lines suggested above, only large and possibly some medium-size NGOs could go to scale with new services. Small NGOs have demonstrated their ability to pilot innovative approaches: for instance, 19 small NGOs, supported by PKSF, have developed products that address the needs of the poorest in recent years. The larger NGOs that can go to scale will need to make a concerted effort to absorb the lessons from these innovations, particularly in education, where BRAC’s non-formal education model is dominant. 5.30 Expansion of NGO activity was based on a “franchising system” in which new branches copied the procedures and norms of existing branches. The period of rapid growth 81 Economics and Governance of Nongovernmental Organizations in Bangladesh of microfinance programs is closely correlated with the growth of the NGO sector as a whole, as NGOs reproduced organizational structure and program design in thousands of small units across the country. 134 Franchising of branches from within organizations was possible because in the early years, at least, there was a strong operational focus on a narrow, standardized range of microfinance products offered to all clients. It was quick and easy to recruit and train staff to offer a simple product at a time when branches were being opened at a rapid rate. This enabled an efficient—and easily replicable—program, and the performance of individual branches could easily be monitored because the narrow range of similar products could be compared using disbursement and repayment rates. 135 5.31 The focused expansion recommended for the future will also use this franchising system, though the products may be more complex. In microfinance, scaling up financial products for the poorest will require a greater period of learning for managers and staff in branches where these programs do not exist. Similarly, in poorer, more remote areas where even the basic microcredit product is in short supply, branch managers and staff may need to provide greater flexibility to the standard savings and loan terms in order to attract clients. In health, the basic “franchising units” are the village-based community health workers. Clearly, investments in building their capacity are essential in order to improve their effectiveness during childbirth and postnatal situations, create awareness of communicable diseases, administer tuberculosis detection and treatment, as well as carry on their more routine and yet vital tasks of family planning, nutritional awareness and so forth. In education, the basic unit of the franchise is the non-formal school, and in both pre-primary and primary education, these programs can be scaled up using existing procedures. However, it is important to balance the pressures to adopt a uniform model while scaling up against the risk of stifling innovative learning methods developed by small NGOs. In short, while the franchising system remains relevant for the future, human resource management will need to address issues of capacity building of field staff and building a solid management layer that can oversee this growth and the addition of more complex products. 5.3 Managing a larger and more focused NGO program 5.32 Chapter 4 discussed the trends and issues related to donor, government, and self- financing of NGO programs. We find that overall levels of external assistance channeled to NGOs directly or through the government have risen over the past decade, particularly with new sectorwide programs in education and health and the financing of microcredit programs through PKSF. As we look forward, issues surrounding the increasing financing of NGO programs through the government budget will become more important. 5.33 Greater financing of NGO social services through the budget has merits as well as some risks. For most NGOs, education, health, and other social programs remain largely dependent on external subsidies, in the form of resources channeled through the budget or direct aid from donors. Financing of NGOs through government has increased over the past few years and in our view this is a trend that ought to be supported. Transferring resources to NGOs through the budget improves the comprehensiveness of the public expenditure framework, 134 Zaman 2004. 135 Jain and Moore 2003. 82 Chapter 5: Scaling Up through a Strategic Compact strengthens coordination with government spending, and in principle leads to better policy outcomes. 5.34 There are two main risks associated with increasing dependence on government financing. The first is the potential for conflict with any advocacy work that challenges prevailing government policy. In this situation one option for multifaceted NGOs is to limit their advocacy activities to issues that are important for the poor yet not highly controversial, such as raising awareness of marriage laws or pressing claims to government services. A second option is for NGOs to separate their advocacy and service delivery work into two different legal entities with different financing sources. A third option is to limit the dependence on government financing to a level that would not jeopardize the whole program should funding be suspended. 136 5.35 The second risk of increasing government financing is the emergence of fly-by-night NGOs seeking to exploit weak government contracting processes. It is clear that government- NGO contracting has not escaped the governance problems associated with public procurement in general. Funding of NGOs through government ought to increase only on the condition that best-practice models of contracting are replicated widely. In short, contracting appears to work best when it is done by autonomous bodies (e.g., PKSF) or by independent third-party firms, and not when it is carried out directly by a government agency. 5.36 Interest income and commercial borrowing can be used to finance microcredit activities. The microfinance industry has now matured to the point where many NGOs can sustain a large portion of their lending activities from self-generated funds and borrowing at market rates. PKSF has an important role to play in building the capacity of NGOs to manage their microfinance portfolio, in promoting innovations within the industry, and in facilitating links to commercial sources of funding. One way of promoting links with commercial banks would be to gradually align PKSF’s own lending rates to commercial rates without any restriction on final on-lending rates from NGOs to borrowers. However, the expansion of microcredit in remote areas or for the extreme poor may well require strategic subsidies. 5.37 A strategic compact between government, donors, NGOs, and clients is required. Finally, we return to the conceptual framework set out in World Development Report 2004 and discussed in chapter 1 of this report. We argue that in Bangladesh, there is a need for a strategic compact between government, NGOs, donors, and clients, in which each party would work to strengthen its current practices and all would demand more from each other in order to improve the quality of NGO interventions. The strategic compact could be structured as follows. 5.38 The government needs to modernize the legal and regulatory framework for NGOs while recognizing NGOs as an ally in the development process. As part of this modernization the government should rationalize the number of agencies that oversee NGOs and consider establishing an independent regulatory body. It needs to be more tolerant of critical views and recognize that NGOs that engage in pro-poor advocacy activities have an important role in all democratic environments. 136 Action Aid has explicit policies to limit funding from institutions whose policies and programs they might want to campaign against. 83 Economics and Governance of Nongovernmental Organizations in Bangladesh 5.39 The government ought to withdraw from areas where the public-good rationale for its intervention is missing, or from areas where NGOs have a comparative advantage (for example, direct provision of microcredit). At the same time, there are areas where the government may wish to become more actively involved in the long run. These could include (a) strengthening the capacity of local government bodies, which can then choose to contract NGOs to implement services; (b) improving public expenditure and procurement systems so that aid to NGOs channeled through the budget is recorded properly and disbursed in a transparent manner; (c) reviving the Government-NGO Coordination Council and monitoring that it acts as an effective vehicle for dialogue between Government and NGO representatives (d) helping to mainstream certain NGO activities such as non-formal primary education by providing formal recognition to these schools; and (e) widely disseminating lessons from successful partnerships with NGOs and also lessons on what has not worked, with respect to both implementation issues and impact. 5.40 Donors should continue to support NGO activities in Bangladesh, both to improve pro- poor service delivery and to promote a broad-based civil society, while ensuring that they do not undermine the accountability of NGOs to Bangladeshi stakeholders. While financing social sector activities donors also need to develop a clear strategy for sustainability, as has occurred in microfinance. This strategy ought to be developed in tandem with government so that in the medium run the budget progressively absorbs the costs of financing many of these social sector programs. In areas such as advocacy, where the conflicts of interest with government financing are too great, donors should fund NGOs directly and work with NGOs to tap the market for greater private charitable contributions. 5.41 Donors need to streamline and harmonize their financial reporting requirements, as current practices often create unnecessary duplication of efforts. At the same time, staff in donor agencies that interact with NGOs should upgrade their own financial management skills in order to interpret best-practice reporting formats. Donors also need to set in place systems that allow the transfer of “institutional memory” during frequent staff rotations. In general donors need to be accountable for the occasions when poor project design or inadequate supervision contributes to unsatisfactory implementation of projects involving NGOs and institute mechanisms to prevent these from reoccurring. 5.42 While there is a clear case for a focused expansion of NGO activity, and there are NGOs that have invested in building the capacity to expand, any discussion of greater financing requires a quid pro quo on the side of NGOs. The NGO part of the compact could involve the following: (a) strengthening of financial management and corporate governance arrangements; (b) a systematic effort to provide public disclosure with respect to income sources, expenditure patterns, and program information, along with a public information strategy to address misconceptions about NGOs; (c) greater efforts to align programs with national strategies, coordinate with local and national government officials, and act as facilitators of government, private sector, and community service provision; (d) revisiting strategic directions periodically based on national development needs and internal capacity constraints; and (e) developing a middle management layer to oversee more complex development programs. 84 Chapter 5: Scaling Up through a Strategic Compact 5.43 Clients also have an important role in this proposed strategic compact. Clients can improve the quality of services provided to them by demanding better service and by monitoring service provision. Mechanisms that improve client power also need to be put in place so that poor households can exercise these rights. For instance, vouchers can be introduced so that clients can choose between public, NGO, and private providers. NGOs ought to remove any implicit or explicit barriers to clients moving from one NGO service provider to another. Monitoring mechanisms can also be set up through, for instance, user groups or management committees to provide systematic feedback to NGO officials and regulatory bodies. 5.44 In conclusion, we argue that on the basis of the evidence presented in earlier parts of this report, there is a case for a focused scaling up of activity by NGOs, both as direct providers of services and as facilitators of pro-poor interventions. This scaling up can build on investments in capacity and systems that have already been developed, though clearly this capacity is by no means uniform and will need strengthening. 5.45 The financing required for the focused scaling up can be obtained from a range of sources, depending on the activity in question. We argue that the microfinance sector has matured to a point where the sector ought to move toward commercial sources of funding through linkages to the formal financial system. In the social sectors, there is a strong case for financing through the government budget, provided contracting procedures improve. The source of this financing could be external assistance or domestically raised revenue; while this mechanism involves some risks, we argue that the benefits of greater coordination in public policy outweigh these costs. For advocacy work, financing sources ought to be independent of government, raised from private donations or external grants. 5.46 Finally, in order to achieve these goals, the report argues for a “strategic compact” between government, NGOs, donors, and clients. The implicit compact would commit all the actors to strengthen various areas that are within their own mandates. The combined impact of these improvements will go a long way toward achieving the common goal of improving services to the poor in Bangladesh and attaining the Millennium Development Goals. 85 Annex 1: Supplementary Figures and Tables Annex figure 1.1 NGO staff perceptions of relations with government counterparts 90 80 70 60 percent 50 40 30 20 10 0 Help Hindrance Both Neither National government Local government Annex figure 2.1: How NGOs assess community needs workshops with Participatory communityOther 0 10 20 30 40 50 60 70 80 90 percent Source: Gauri and Galef 2004. Economics and Governance of Nongovernmental Organizations in Bangladesh Annex table 2.1 Microfinance as share of total private sector credit FY02 FY03 FY04 Outstanding (in billion takas) Banks 675.7 776.6 902.2 Non-banks 24.6 31.6 40.2 a Microfinance institutions 36.3 43.1 53.0 Total 736.6 851.3 995.4 Microfinance as % of private sector credit 4.9 5.1 5.3 Source: Bangladesh Bank. a. Includes PKSF, Grameen Bank, BRAC, ASA, and Proshika. Annex table 2.2 NGO education programs No. of No. of learners NGOs No. of Education program involved centers Female Male % Female Total Early child development 41 1,615 16,901 13,707 55 30,608 Pre-primary education 119 9,450 163,970 110,281 60 274,251 Primary education 559 44,372 943,120 522,776 64 1,465,896 Adolescent education 200 4,340 77,071 33,205 70 110,276 Adult education 251 5,701 107,922 46,738 70 154,660 Post-literacy and continuing 219 14,964 487,607 124,130 80 611,737 education Other education 14 1,500 33,268 4,674 88 37,942 Total 81,942 1,829,859 855,511 68 2,685,370 Sources: Author’s calculations based on CAMPE’s Directory of NGOs with Education Programmes (2004) and BRAC Annual Report 2003. Note: Figures are estimated by collating data from NGOs with significant education programs with information from CAMPE. 88 Annex 1: Supplementary Figures and Tables Annex table 2.3 Donor financing of education programs of big NGOs (Tk millions) 1999– US$ 1998–99 2000a 2000–1 2001–2 2002–3 Total millions VERC 0.7 1.8 1.0 1.1 0.8 5.5 0.1 FIVDB 37.5 33.5 27.2 — — 98.3 1.6 BRAC 902.4 2,226.9 854.5 1,052.6 1,394.2 6,430.6 107.2 CAMPE — 5.0 7.7 35.8 47.9 96.3 1.6 PLAN — — 212.9 245.9 274.0 732.8 12.2 RDRS 9.8 6.0 4.9 — — 20.6 0.3 Proshika 468.3 627.4 — — — 1,095.7 18.3 CMES 44.2 45.3 94.1 48.3 50.2 282.2 4.7 DAM 16.5 3.4 16.8 8.4 19.4 64.5 1.1 Total 1,479.5 2,949.3 1,219.0 1,392.1 1,786.6 8,826.5 147.1 US$ millions 24.7 49.2 20.3 23.2 29.8 147.1 Source: Staff estimates from 2004 NGO Affairs Bureau data. a. The rise in funding in 1999–2000 was an anomaly as BRAC received a large infusion from its donors. 89 Economics and Governance of Nongovernmental Organizations in Bangladesh Annex table 3.1 Estimated IDA disbursement through NGOs, 1995–2004 (US$ millions) Sector and program Disbursement Energy (as of Dec. 2004) 1 Rural electrification & renewable energy development 6.7 2 Renewable energy development 3.1 Energy subtotal 9.8 Microfinance 3 Poverty alleviation (Microcredit I: disbursed as of Sept. 2004) 131.0 4 Microfinance II (disbursed as of Dec. 2004) 149.5 5 Microfinance Financial Services for the Poorest 1.6 Microfinance subtotal 282.1 Health 6 Bangladesh Integrated Nutrition Project (FY97–FY02) 41.9 7 Health and Population Program 3.0 8 National Nutrition Program (disbursed as of Dec. 2004) 6.4 9 HIV/AIDS prevention (disbursed as of Dec. 2004) 2.4 Health subtotal 53.7 Water 10 Arsenic mitigation water supply (disbursed as of Jan. 2005) 6.0 11 Water Supply Program Project 0.0 Water subtotal 6.0 Education 12 Female Secondary School Assistance Program (1993–2001) 5.6 13 Non-formal Education Project (1996–2001) 7.4 14 Post-literacy and continuing education (2001–2005) 13.3 Education subtotal 26.3 Rural development 15 Silk Development Pilot Project (1997–2003) 10.2 16 Social Investment Program Project (2003–2005) 1.7 Rural development subtotal 11.9 Total IDA disbursement through NGOs (1995–2004) 389.8 World Bank disbursement (1995–2004) 2,788.7 IDA disbursement to NGOs as share of World Bank lending (%) 14.0 Source: Staff estimates from project data files. 90 Annex 1: Supplementary Figures and Tables Annex table 3.2 Donor financing mechanisms Type of Funding from international NGO or large relationship Funding direct from donor domestic NGO Contractual For a specific short- to medium-term task. For a specific short- to medium-term task. Project For a specific area of programming. Often Acting as a subcontractor to an INGO, or as donors coordinate funding, e.g., for non-formal part of a network, e.g., Grameen Trust, education, water and sanitation, etc. Proshika, etc. Wholesale Contractor manages (usually via a tender) a INGO manages a facility for donors to finance management pool of money or “facility” for a specific other small NGOs around an objective, e.g., (usually via a theme. Smaller NGOs apply through the Oxfam (disaster relief), CARE (Income III); tender) facility. Most are based in Bangladesh, e.g., BRAC (microfinance), Samata (LAND Manusher Jonno (ASIRP, with government of network). Bangladesh); Katalyst (DFID); Health and Human Rights Funds (USAID); Canada Fund (CIDA). Some are headquarters-based, e.g. Civil Society Challenge Fund (DFID). Partnership Donors pool funds around agreed outcomes INGO funds a partnership with a domestic agreement that relate to the entire operations of an NGO, NGO across all areas of its operations, e.g., e.g., Ashroi, Proshika phase VI, Nijera Kori, NOVIB and Proshika, Oxfam and BRAC. Samata, Rupanter. Institution Donors co-fund the entire organization, which INGOs co-fund the entire organization, which building emerges as a different legal entity. This is emerges as a different legal entity—e.g., early usually part of the “traditional exit,” e.g., funding of Grameen by Ford Foundation. PDBF (CIDA); Aranuk Foundation (USAID); Manusher Jonno Trust (DFID). Sources: Matsaert 2005; Sultan 2004. 91 Economics and Governance of Nongovernmental Organizations in Bangladesh Annex table 3.3 Microfinance efficiency indicators 2001 2002 2003 2004 Very small NGOs Borrowers per credit officer 148 161 167 207 Portfolio per credit officer (taka) 570,000 573,000 633,000 811,000 Cost ratio (%) 7.6 9.9 9.9 9.2 Small NGOs Borrowers per credit officer 190 200 209 207 Portfolio per credit officer (taka) 506,000 576,000 622,000 677,000 Cost ratio (%) 10.5 10.3 9.9 9.1 Medium NGOs Borrowers per credit officer 209 215 220 249 Portfolio per credit officer (taka) 686,000 733,000 740,000 898,000 Cost ratio (%) 8.8 11.4 10.1 9.6 ASA Borrowers per credit officer 345 448 461 461 Portfolio per credit officer (taka) 1,310,000 1,700,000 1,970,000 2,110,000 Cost ratio (%) 4.1 3.4 3.3 3.1 BRAC Borrowers per credit officer — 298 364 — Portfolio per credit officer (taka) — 881,000 1,160,000 — Cost ratio (%) 13.5 12.6 11.5 — Note: Cost ratio is the cost per 100 taka lent out, calculated by dividing total expenses by total disbursements in a year. Source: Staff estimates for this study. 92 Annex 1: Supplementary Figures and Tables Annex table 3.4 Outreach and average loan size by NGO size 2001 2002 2003 2004 Very small NGOs Annual growth in clients (%) — 37 27 25 Average loan size (taka) 6,580 6,256 6,435 7,742 Small NGOs Annual growth in clients (%) — 20 18 16 Average loan size (taka) 5,622 6,610 7,510 9,979 Medium NGOs Annual growth in clients (%) — 19 21 24 Average loan size (taka) 5,420 5,960 5,504 6,289 ASA Annual growth in clients (%) 25 40 8 11 Average loan size (taka) 6,722 7,524 8,589 — BRAC Annual growth in clients (%) — 5 15 — Average loan size (taka) 4,967 5,886 6,088 — Source: Staff estimates for this study. 93 Economics and Governance of Nongovernmental Organizations in Bangladesh Annex table 3.5 An illustrative range of NGO businesses Business sector Businesses Sponsor Poultry hatchery GKT poultry hatchery GKT BRAC poultry hatchery BRAC Feed mill BRAC poultry feed mill and marketing BRAC Seed production and processing BRAC seed production and processing BRAC BRAC tissue culture (potato) GKT seed processing GKT GKT tissue culture (potato) Fish hatchery and fish production BRAC fish hatchery BRAC JC fish farming JC Grameen Mothsha Foundation Grameen TMSS fish production TMSS Horticulture BRAC horticulture and nursery and BRAC vegetable exports Dairy BRAC dairy products BRAC TMSS dairy and livestock TMSS Handicrafts/garments/handloom TMSS Uthpaddan TMSS Chakra handicrafts JC Aarong BRAC Grameen Shamogree Grameen Grameen Uddog Grameen Rural cell phone Grameen Telecom Grameen Internet services BRAC BD Mail BRAC Citechco.net Grameen (joint venture) Private banking BRAC Bank BRAC (joint venture) Land development BRAC Concord Land BRAC (joint venture) Housing financing BRAC-Delta Housing Finance Corp. BRAC (joint venture) Private higher education BRAC University BRAC Renewable energy (solar Grameen Shakti Grameen photovoltaic systems) BRAC solar energy program BRAC TMSS solar energy program TMSS Knitwear Grameen Knitwear (in export processing Grameen zone) Cell phone Grameen Phone Grameen (joint venture with Telenor and other investors) Source: Alamgir 2004, 10. 94 Annex 1: Supplementary Figures and Tables Annex table 3.6 Average profitability of four NGOs with commercial activities, 2001–4 Sector NGO Return on sales (%) Return on assets (%) Agro-based products Jagoroni Chakra 9.00 n.a BRAC 4.17 1.51 GKT 0.04 n.a Handicrafts/dressmaking Jagoroni Chakra 21.90 — BRAC 8.10 6.20 TMSS 1.99 2.48 Printing BRAC 7.12 11.29 Dairy products BRAC 5.81 6.99 Non-agro-based businesses BRAC 7.21 5.34 Property (house) BRAC 24.84 0.94 Source: Staff estimates for this study. 95 Annex 2: The Charities Commission for England and Wales The Charity Commission is established by law as the regulator and registrar for charities in England and Wales. The Commission is independent from the political process, with powers given by the Charities Act 1993 (although it dates back to 1601). This authority is exercised by five commissioners with different areas of expertise. Charity in Britain is defined on the basis of public benefit. Thus charities in Britain are voluntary bodies whose purpose is to benefit the community. Charity includes, inter alia, relief of poverty or sickness, education, and community relations. The Commission has three objectives: to maintain an effective framework within which charities can operate, to work with charities to encourage good practice, and to eradicate bad practice and abuse. A core function of the Commission is to determine whether an organization meets the requirements of a charity and to register it in the public Register of Charities. Charities may take a number of forms, including trusts, associations, or charitable companies. Although responsibility for running the charity rests with its trustees, ultimate accountability is to the High Court. The Commission’s role is to act as a streamlined administrative alternative to the oversight of the High Court. Regulation and accountability: The Commission provides a framework of accountability for charities. This includes meeting requirements for registration, reporting, and maintaining proper financial accounts. The board of trustees are accountable to their members. Charities in a particular field, particularly ones that have a substantial public policy interest, are often accountable to a particular specialist body. For example, legal pro bono work needs to comply with the Bar requirements. The role of the Commission is to provide a framework for that accountability, cooperating with other forms of accountability. Charities are expected to be transparent because they serve the public interest. The Commission by law is not allowed to interfere in the manner in which trustees administer their charities in seeking to fulfill their purposes. The Commission gives advice and guidance and has the power to intervene in cases of mismanagement or misconduct. Registration and monitoring: The Commission maintains a register of charities. At the registration stage the Commission is essentially concerned with determining that the body is charitable and that it is properly formed in accordance with the requirements of the law. The status of registered charity is valued, not only for the tax relief but for the credibility it brings. With registration come accountability and monitoring responsibilities. Accountability varies according to the size of the charity, ranging from simple reporting of activities and receipts and payments accounts for small charities up to sophisticated reporting and accounting for large charities. The Commission sends a return to all charities, but charities below the 1 million pound a year threshold are only asked to provide basic details. The 5 million–plus charities above the 1 million pound a year threshold must complete a more detailed return that includes financial accounts. The aim of the accountability framework is to support good management and to make the charity accountable to its members, beneficiaries, and the public. The 1993 Act gives the Commission the power to scrutinize the activities and financial circumstances of the charity as the basis for regular supervision. Economics and Governance of Nongovernmental Organizations in Bangladesh Advice, guidance, investigation, and professional support: The Commission’s offices have Support Divisions that are responsible for the day-to-day relationship with charities. They are responsible for, among other things, providing charities with a legal service to amend or update their constitutions and giving advice and guidance on a range of legal, governance, management, and financial issues. Each of the Commission’s offices has an Investigation Division responsible for using the Commission’s intervention powers where problems have arisen that cannot be dealt with on a cooperative basis by the Support Division. The Commission’s powers to intervene are used where there is suspected misconduct or mismanagement or when a charity’s resources need protecting. The powers include suspending trustees, freezing bank accounts, and appointing a receiver and manager to act in place of the trustees. Although the Commission does not have the power simply to deregister the charity, it has on occasion advised the dissolution of a charity, transferring its resources to a comparable charity. In addition to its registration, support, and investigation functions, the Commission also has a responsibility to develop the interpretation of the law and to disseminate it to the charitable sector. The Charities Bill 2004: A new Charities Bill was published on December 21, 2004. This legislation came in response to a strategic review of the charitable and voluntary sector. The bill promotes greater empowerment of the charitable sector and a more defined role for the Charity Commission in order to enhance both compliance and accountability. The bill: • provides greater freedom for charities to operate within a more flexible legal framework that will help them respond to changes in society; • supports trusteeship by providing simpler arrangements for obtaining relief from liability for honest mistakes, and paying trustees for providing services; • supports greater choice by establishing a new structure for charities, the Charitable Incorporated Organization, which will simplify administration and reduce risk for charities that want to become companies; • provides new descriptions for charitable purposes; • makes benefit to the public a priority. 98 Bibliography ADAB (Association of Development Agencies in Bangladesh). 2004. Directory of NGOs in Bangladesh 2003–2004. Dhaka: ADAB. AHI (Aus Health International with ADSL Bangladesh). 2004. “An Evaluation of NGO Contracting in Bangladesh: Contracting NGOs for the Delivery of Health Services.” Report prepared for the World Bank. Dhaka. Ahmad, M., and R. Jahan. 2002. “Investing in Ourselves.” Asian Development Bank, Manila Ahmad, A., and M. Rafi. 1999. “NGOs and Evaluation: The BRAC Experience.” Paper presented to the World Bank Conference on Evaluation and Poverty Reduction, Washington, DC. Ahmed, S. 2004. “Aid and NGOs in Bangladesh..” In Revisiting Foreign Aid: A Review of Bangladesh’s Development 2003, ed. R. Sobhan. Independent Review of Bangladesh’s Development Series. Dhaka: Centre for Policy Dialogue Bangladesh/University Press Limited. Alamgir, D. 2004. “Review of Businesses Owned and Managed by NGO-MFIs in Bangladesh.” Background paper for this study. World Bank, Washington, DC. Amin, S., and A. Pebley. 1994. “Gender Inequality within Households: The Impact of a Women’s Development Programme in 36 Bangladeshi Villages.” Bangladesh Development Studies 22 (2– 3), “Special Issue on Women, Development and Change,” ed. S. Amin, 121–55. Bangladesh Institute of Development Studies, Dhaka. Amin, S., A. S. Rai, and G. Topa. 2003. “Does Microcredit Reach the Poor and Vulnerable? Evidence from Northern Bangladesh.” Journal of Development Economics 70 (1): 59–82. ASIRP (Agricultural Services Innovation & Reform Project). 2003. “Working with NGOs in Bangladesh: A Ten Year Review: Volume 3, 1992–2003.” Agricultural Support Services Project (ASSP) & Agricultural Services Innovation and Reform Project (ASIRP). Dhaka. Bangladesh Bureau of Statistics. 2002. Report of the Labour Force Survey Bangladesh, 1999–2000. Dhaka: Bangladesh Bureau of Statistics, Ministry of Planning. Bangladesh Enterprise Institute. 2004. The Code of Corporate Governance for Bangladesh. Dhaka: Bangladesh Enterprise Institute. Besley, T., and M. Ghatak. 2001. “Government versus Private Ownership of Public Goods.” Quarterly Journal of Economics 116 (4): 1343–72. BRAC. 1980. “The Net: Power Structure in Ten Villages.” Dhaka. ———. 2003. Annual Report 2003. Dhaka. Economics and Governance of Nongovernmental Organizations in Bangladesh BURO, Tangail. 2004. Annual Report 2003. Dhaka. CAMPE (Campaign for Popular Education). 2001. Education Watch 2000: A Question of Quality: State of Primary Education in Bangladesh. Vol. 1. Dhaka: CAMPE/University Press Limited. _______. 2003. Education Watch 2002: Literacy in Bangladesh: Need for a New Vision. Dhaka: CAMPE/University Press Limited. CDF (Credit and Development Forum). 2003–4. CDF Microfinance Statistics. Dhaka: CDF. Chaudhury, N., and J. S. Hammer. 2003. “Ghost Doctors: Absenteeism in Bangladeshi Health Facilities.” Policy Research Working Paper 3065, World Bank, Washington, DC. Chaudhury, N., J. Hammer, M. Kremer, K. Mularidharan, and H. Rogers. 2004. “Roll Call: Teacher Absence in Bangladesh.” World Bank, Washington, DC. Chowdhury, A. F., S. Delay, N. Faiz, I. Haider, B. Reed, P. Rose, and P. D. Sen. 2005. “Bangladesh: Study of Non-State Providers of Basic Services.” Report prepared for UK Department for International Development, London. Chowdhury, A. M. R., J. P. Vaughan, S. Chowdhury, and F. H. Abed. 1998. “Demystifying the Control of Tuberculosis in Rural Bangladesh.” In Tuberculosis: An Interdisciplinary Perspective, ed. J. Porter and J. Grange, 393–412. London: Imperial College Press. Chowdhury, A. M. R., S. R. Nath, and R. K. Chowdhury. 2002. “Enrolment at Primary Level: Gender Gap Disappears in Bangladesh.” International Journal of Educational Development 22: 191–203. Chowdhury, A. M., and A. Bhuiya. 2001. “Do Poverty Alleviation Programmes Reduce Inequities in Health? The Bangladesh Experience.” In Poverty, Inequality, and Health: An International Perspective, ed. D. Leon and G. Walt, 312–32. London: Oxford University Press. Chowdhury, S. 2001. “Educating for Health: Using Incentive-Based Salaries to Teach Oral Rehydration Therapy.” Viewpoint Note 235, World Bank Private Sector and Infrastructure Network, World Bank, Washington, DC. CIET (Canadian Institute for Energy Training). 2004. “Bangladesh Health and Population Sector Programme 1998–2003: The Third Service Delivery Survey 2003: Final Report.” CIETcanada and Bangladesh Ministry of Health and Family Welfare, Dhaka. Copestake, J. 1992. “The Integrated Rural Development Programme.” In Poverty in India, ed. B. Harris, S. Gukon, and R. Cassen. Mumbai: Oxford University Press. 100 Bibliography CPD (Center for Policy Dialogue). 2003. Employment and Labour Market Dynamics: Annual Review of Bangladesh’s Development 2002. Dhaka: University Press Limited/Centre for Policy Dialogue. Cummings, C. 2004. “Scoping Study of Potential Mechanisms for Strategic Financing of NGO Provision of Health and Education Services.” Report prepared for UK Department for International Development. Options, London; Cambridge Education Consultants, Cambridge, UK. Deolalikar, A. 2004. “Attaining the Millennium Development Goals in Bangladesh: How Likely and What Will It Take to Reduce Poverty, Infant Mortality, Child Malnutrition, Gender Disparities and to Increase School Enrollment and Completion?” South Asia Human Development Department, World Bank, Washington DC. DFID (UK Department for International Development). 2000. “Partners in Development: A Review of Big NGOs in Bangladesh.” London. ———. 2003. “DFID Influencing Study: NGO Case Studies.” London. ———. 2005. “Achieving the MDGs in Bangladesh: Rethinking the Parameters for Working with NGOs.” London. Duncan, A., I. Sharif, P. Landell-Mills, D. Hulme, and J. Roy . 2002. “Supporting the Drivers of Pro- Poor Change in Bangladesh.” UK Department for International Development, London. Fruttero, A., and V. Gauri. 2005. “The Strategic Choices of NGOs: Location Decisions in Rural Bangladesh.” Journal of Development Studies 41 (5): 759–87. Gauri, V., and A. Fruttero. 2003. “Location Decisions and Nongovernmental Organization Motivation: Evidence from Rural Bangladesh.” Policy Research Working Paper 3176, World Bank, Washington, DC. Gauri, V., and J. A. Galef. 2004. “NGOs in Bangladesh: Activities, Resources, and Governance.” World Development 33 (12): 2045–65. Goetz, A. M. 2001. Women Development Workers: Implementing Rural Credit Programmes in Bangladesh. Dhaka: University Press Limited. Goetz, A. M., and R. Sen Gupta. 1996. “Who Takes the Credit? Gender, Power, and Control over Loan Use in Rural Credit Programs in Bangladesh.” World Development 24 (1): 45–63. Government of Bangladesh. 2005. “Poverty Reduction Strategy Paper: Unlocking the Potential: National Strategy for Accelerated Poverty Reduction.” General Economics Division, Planning Commission, Dhaka. 101 Economics and Governance of Nongovernmental Organizations in Bangladesh Haque, S. 2002. “The Changing Balance of Power between the Government and NGOs in Bangladesh.” International Political Science Review 23 (4): 411–35. Hashemi, S. 1995. “NGO Accountability in Bangladesh: Beneficiaries, Donors and the State.” In Non-Governmental Organisations—Performance and Accountability: Beyond the Magic Bullet, ed. M. Edwards and D. Hulme, 103–10. London: Earthscan. ———. 1997. “Those Left Behind: A Note on Targeting the Hardcore Poor.” In Who Needs Credit? Poverty and Finance in Bangladesh, ed. G. Wood and I. Sharif. Dhaka: University Press Limited. ———. 2001. “Linking Microfinance and Safety Net Programs to Include the Poorest: The Case of IGVGD in Bangladesh.” CGAP Focus Note 21, Consultative Group to Assist the Poor, Washington, DC. Hashemi, S. M., S. R.. Schuler, and A. P. Riley. 1996. “Rural Credit Programs and Women’s Empowerment in Bangladesh.” World Development 24 (4): 635–53. Health Economics Unit. 2003. “Bangladesh Health Accounts, 1999–2001.” Report prepared by Data International for Government of Bangladesh, Dhaka. Hossain, M. 2005. ‘The Case of BURO, Tangail.” Paper presented at INAFI Bangladesh National Workshop on Innovating Demand Driven Products and Delivery Mechanisms, Dhaka. Hossain, N. 2005. Elite Perceptions of Poverty in Bangladesh. Dhaka: University Press Limited. Hossain, N., and I. Matin. 2004. “Engaging Elite Support for the Poorest? BRAC’s Experience with the Ultra Poor Programme.” Working Paper 4, BRAC Research and Evaluation Division, Dhaka. Hussain, A. M., ed. 1998. “Poverty Alleviation and Empowerment: The Second Impact Assessment Study of BRAC’s Rural Development Program.” BRAC Publications, Dhaka. IDF (Integrated Development Foundation). 2003. Annual Report. www.idfbd.com. Irish, L. and K. Simon. 2005. “Legal and Regulatory Environment for NGOs in Bangladesh.” Prepared by Irish/Simon/Firoze Consultancy for Local Consultative Group, Dhaka. Jack, W. 2001. “Public Policy toward Nongovernmental Organizations in Developing Countries.” Policy Research Working Paper 2639, World Bank., Washington, DC. Jain, P., and M. Moore. 2003. “What Makes Microcredit Programmes Effective? Fashionable Fallacies and Workable Realities.” Working Paper 177, Institute of Development Studies, University of Sussex, Brighton, UK. Joshi, A., and M. Moore. 2004. “Institutionalized Co-Production: Unorthodox Public Service Delivery in Challenging Environments.” Journal of Development Studies 40 (4): 31–49. 102 Bibliography Kabeer, N. 2002. “We Don’t Do Credit: Nijera Kori, Social Mobilisation, and the Collective Capabilities of the Poor in Rural Bangladesh.” Nijera Kori, Dhaka. ———. 2003. “Making Rights Work for the Poor: Nijera Kori and the Construction of ‘Collective Capabilities’ in Rural Bangladesh.” Working Paper 200, Institute of Development Studies, University of Sussex, Brighton, UK. Kairy, S. N. 2004. “Role of BRAC’s Programme Support Enterprises on Development.” Holiday (Dhaka), June 18, 2004.. Khan, M. M. 2003. “Accountability of NGOs in Bangladesh.” Public Management Review 5 (2): 268–77. Khan, S. R. 2005. “Review of NGO Financial Management and Accountability Issues in Bangladesh.” Background paper for this study. World Bank, Washington, DC. Khandker, S. 1998. Fighting Poverty with Microcredit: Experience in Bangladesh. New York: Oxford University Press. ———. 2003. “Microfinance and Poverty: Evidence Using Panel Data from Bangladesh.” Policy Research Working Paper 2945, World Bank, Washington, DC. Lipton, M. 1996. “Successes in Anti-poverty.” Issues in Development Discussion Paper 8, International Labour Office, Geneva. Lundberg, S., and R. Pollak. 1993. “Separate Spheres Bargaining and the Marriage Market.” Journal of Political Economy 101 (6): 988–1010. Loevinsohn, B. 2001. “Practical Issues in Contracting for Primary Health Care Delivery: Lessons from Two Large Projects in Bangladesh.” World Bank, Washington, DC. Loevinsohn, B., and A. Harding. 2004. “Contracting for the Delivery of Community Health Services: A Review of Global Experience.” HNP Discussion Paper, World Bank, Washington DC. Louis Berger Group. 2004. “Evaluation of BRAC Group Pricing Analysis: Mid-Year Report on First Survey.” Louis Berger Group Inc., Washington, DC. Mahmud, W. 2002. Popular Economics: Unpopular Essays. Dhaka: University Press Limited. Matin, I. 1998. “Mis-targeting by the Grameen Bank: A Possible Explanation.” IDS Bulletin 29 (4): 51–59. Matin, I., and D. Hulme. 2003. “Programmes for the Poorest: Learning from the IGVGD Programme in Bangladesh.” World Development 31 (3): 647–65. 103 Economics and Governance of Nongovernmental Organizations in Bangladesh Matin, I., and R. Yasmin. 2004. “Managing Scaling Up Challenges of a Program for the Poorest: Case Study of BRAC’s IGVGD Program.” In Scaling Up Poverty Reduction: Case Studies in Microfinance, ed. Consultative Group to Assist the Poor. Washington, DC: CGAP/World Bank. Matsaert, F. 2003. “Enterprise Development Assessment of DFID Bangladesh’s programme 1997– 2002.” UK Department for International Development, Dhaka. ———. 2005. “Donor Funding of NGOs: Survey Summary.” Background paper for this study. UK Department for International Development, Dhaka. McKennie, F., F. Toomey, and I. Hossain. 2003. “BRAC Business Strategy Review.” Paper prepared for UK Department of International Development by Shorebank Advisory Services, Chicago. McKernan, S. 2002. “The Impact of Micro-Credit Programs on Self-Employment Profits: Do Non- Credit Program Aspects Matter?” Review of Economics and Statistics 84 (1): 93–115. MOHFW (Ministry of Health and Family Welfare). 2003. “Bangladesh National Health Accounts, 1999–2001.” Health Economics Unit, Ministry of Health and Family Welfare/Data International Ltd., Dhaka. Moore, M., and J. Putzel. 1999. “Thinking Strategically about Politics and Poverty.” Working Paper 101, Institute of Development Studies, University of Sussex, Brighton, UK. Mukherjee, R., O. Gokcekus, N. Manning, and P. Landell-Mills. 2001. “Bangladesh: The Experience and Perceptions of Public Officials.” World Bank Technical Paper 507, World Bank, Washington, DC. Morduch, J. 1999. “The Microfinance Promise.” Journal of Economic Literature 37 (December): 1569–1614. Morduch, J., and B. Armendáriz de Aghion. 2005. The Economics of Microfinance. Cambridge, MA: MIT Press. Nabi, R., D. Datta, S. Chakrabarty, M. Begum, and N. J. Chaudhury. 1999. “Consultation with the Poor: Participatory Poverty Assessment in Bangladesh.” NGO Working Group on the World Bank, Dhaka. Nath, S., K. Sylva, and J. Grimes. 1999. “Raising Basic Education Levels in Rural Bangladesh: The Impact of a Non-Formal Education Programme.” International Review of Education 45 (1). Nath, S. R. 2002. Education Watch 2002: Literacy Situation in Bangladesh: Instruction Manual for Interviewers. [In Bangla.] Dhaka: BRAC Research and Evaluation Division. Oguta. J. 2003. “Microfinance Governance Survey.” Finance Local Consultative Subgroup, Dhaka. 104 Bibliography Pitt, M. 1999. “Reply to Jonathon Morduch’s ‘Does Microfinance Really Help the Poor: New Evidence from Flagship Programs in Bangladesh.’” Department of Economics, Brown University, Providence, RI. Pitt, M., and S. Khandker. 1998. “The Impact of Group-Based Credit Programs on Poor Households in Bangladesh: Does the Gender of Participants Matter?” Journal of Political Economy 106 (5): 958–96. Pitt, M., S. Khandker, and J. Cartwright. 2003. “Does Micro-Credit Empower Women? Evidence from Bangladesh.” Policy Research Working Paper 2998, World Bank, Washington, DC. Pitt, M., S. Khandker, O. Choudhury, and D. Millimet. 2003. “Credit Programs for the Poor and the Health Status of Children in Rural Bangladesh.” International Economic Review 44 (1): 87–118. Pitt, M., S. Khandker, S. McKernan, and M. A. Latif. 1999. “Credit Programs for the Poor and Reproductive Behaviour in Low Income Countries: Are the Reported Causal Relationships the Result of Heterogeneity Bias?” Demography 36 (1): 1–21. PKSF (Palli Karma-Sahayak Foundation). 2003. “Maps on Microcredit: Coverage in Bangladesh.” Dhaka. Rahman, A., and A. Razzaque. 2000. “On Reaching the Hardcore Poor: Some Evidence on Social Exclusion in NGO Programmes.” Bangladesh Development Studies 26 (1): 1–36. Rasmussen, S. 2004. “The Economics and Governance of NGOs in Bangladesh: The Microfinance Sector.” Background paper for this study. World Bank, Washington, DC. Rhyne, E. 2005. “Perspectives from the Council of Microfinance Equity Funds.” Small Enterprise Development Journal 16 (1): 8–16. Rutherford, S. 1995. ASA: The Biography of an NGO. Dhaka: Association for Social Advancement. Saifullah, A. K. M., and A. Q. M. G. Mawla. 2003. “A Study on Legal Structure, Governance and Ownership in NGO-MFIs in Bangladesh.” Microfinance Research and Reference Unit, Bangladesh Bank, Dhaka. Salamon, L. M., and H. K. Anheier. 1998. “Social Origins of Civil Society: Explaining the Nonprofit Sector Cross-Nationally” Voluntas 9: 213–48. Sanyal, B. 1991. “Antagonistic Cooperation: A Case Study of Nongovernmental Organizations, Government and Donors’ Relationships in Income-Generating Projects in Bangladesh.” World Development 19 (10): 1367–79. Schreiner, M. 2003. “A Cost-Effectiveness Analysis of the Grameen Bank of Bangladesh.” Development Policy Review 21 (3): 357–82. 105 Economics and Governance of Nongovernmental Organizations in Bangladesh Siddiqui, T. 2001. Transcending Boundaries: Labour Migration of Women from Bangladesh. Dhaka: University Press Limited. Smillie, I., and J. Hailey. 2001. Managing for Change: Leadership, Strategy, and Management in Asian NGOs. London: Earthscan. Sobhan, B. 1997. “Partners or Contractors? The Relationship between Official Agencies and NGOs: Bangladesh.” Occasional Paper Series 14, International NGO Training and Research Centre, Oxford, UK. Stiles, K. 2002. Civil Society by Design: Donors, NGOs and the Development Circle in Bangladesh. Westport, CT: Praeger. Sukontamarn, P. 2004. “The Entry of NGO Schools and Girls’ Educational Outcomes in Bangladesh.” Economic Organisation and Public Policy Programme, London School of Economics. Sultan, M. 2004. “Key Experiences, Issues and Recommendations on Partnerships—Synthesis of Care Bangladesh Workshop Discussions with Selected INGOs & PNGOs (January & June 2004).” CARE Bangladesh, Dhaka. Toufique, K. A., and C. Turton, eds. 2002. Hands Not Land: How Livelihoods are Changing in Rural Bangladesh. Dhaka: Bangladesh Institute of Development Studies; London: UK Department for International Development. Townsend, R., and J. Yaron. 2001. “The Credit-Risk Contingency System of an Asian Development Bank.” Economic Perspectives (Federal Reserve Bank of Chicago) Q3: 31–48. USAID (United States Agency for International Development). 1998. “Involving Communities: Participation in the Delivery of Education Programs.” Prepared by Creative Associates International, Inc. for USAID, Washington, DC. Wood, G., and I. Sharif, eds. 1997. Who Needs Credit? Poverty and Finance in Bangladesh. Dhaka: University Press Limited. World Bank. 1996. Pursuing Common Goals: Strengthening Relations between Government and NGOs in Bangladesh. Dhaka: University Press Limited; Washington, DC: World Bank. ———. 2003a. “Bangladesh Development Policy Review: Impressive Achievements but Continuing Challenges.” Report 26154-BD, Poverty Reduction and Economic Management Sector Unit, South Asia Region, World Bank, Washington, DC. ———. 2003b. “Private Sector Assessment for Health, Nutrition and Population (HNP) in Bangladesh.” Sector Report 27005, World Bank, Washington, DC. 106 Bibliography ———. 2003c. “Bangladesh: Public Expenditure Review.” South Asia PREM, World Bank., Washington, DC. ———. 2003d. World Development Report 2004: Making Services Work for Poor People. New York: Oxford University Press. ———. 2005. “NGO Contracting Evaluation for the HNP Sector in Bangladesh: Evidence and Policy Options.” South Asia Human Development, World Bank, Washington, DC. World Bank and Asian Development Bank. 2002. “Poverty in Bangladesh: Building on Progress.” South Asia PREM, World Bank, Washington, DC. Wright, G., and M. Hossain. 2001. “BURO, Tangail’s Approach to Product Development: A Case Study.” MicroSave-Africa, Nairobi. Yunus, M. 1999. Banker to the Poor: The Autobiography of Muhammad Yunus, Founder of the Grameen Bank. With A. Jolis. London: Aurum Press. Zaman, H. 1998. “Can Mis-targeting be Justified? Insights from BRAC’s Micro-credit Program.” IDS Bulletin 29 (4): 59–66. ———. 2004. “The Scaling Up of Microfinance in Bangladesh: Determinants, Impact and Lessons.” Policy Research Working Paper 3398, World Bank, Washington, DC. Zohir, S. 2004. “NGO Sector in Bangladesh: An Overview.” Economic and Political Weekly, September 4, 2004. 107 World Bank Bangladesh Development Series publications Paper No. 1 Bangladesh PRSP Forum Economic Update – Recent Developments and Future Perspectives (Nov. 2005). Paper No. 2 End of MFA Quotas: Key Issues and Strategic Options for Bangladesh Ready Made Garment Industry (Dec. 2005). Paper No. 3 Bangladesh Country Water Resources Assistance Strategy, (Dec. 2005). Paper No. 4 Comparative Advantages of Health Care Provision, (Dec. 2005). Paper No. 5 Targeting Resources for the Poor in Bangladesh (Dec. 2005). Improving Trade and Transport Efficiency – Understanding the Political Paper No. 6 Economy of Chittagong Port (Dec. 2005). Paper No. 7 Revitalizing the Agricultural Technology System in Bangladesh (Dec. 2005). Bangladesh Integrated Nutrition Project, Effectiveness and Lessons Learned Paper No. 8 (Dec. 2005). Paper No. 9 Social Safety Nets in Bangladesh: An Assessment (Jan. 2006) Paper No. 10 Bangladesh Country Assistance Strategy (April 2006) All Bangladesh Development Series papers are downloadable at www.worldbank.org.bd/bds 108