Madagascar - Community Development Fund Project

Document of The World Bank Report No: 21507-MAG PROJECT APPRAISAL DOCUMENT ON A PROPOSED PROJECT IN THE AMOUNT OF SDR 85.2 MILLION (US$110 MILLION EQUIVALENT) TO THE REPUBLIC OF MADAGASCAR FOR A COMMUNITY DEVELOPMENT PROJECT March 23, 2001 Human Development Department, Group IV AFCO8 Africa Region  CURRENCY EQUIVALENTS (Exchange Rate Effective 12/6/00) Currency Unit = Malagasy Francs (MGF) MGF 1.00 = US$0.000147 US$1 = MGF 6,800 FISCAL YEAR January 1 st - December 31 st ABBREVIATIONS AND ACRONYMS CAS Country Assistance Strategy CRESAN Health Sector Improvement Project (Credit Sante) CRESED Education Sector Reinforcement Project (Credit Education) EU European Union FID Community Development Fund (Fonds d'Intervention pour le Developpement) IBRD International Bank for Reconstruction and Development ICB International Competitive Bidding. IDA International Development Association MIS Management Infornation System NCB National Competitive Bidding NGO Non-Governmental Organization Vice President: Callisto Madavo Country Manager/Director: Hafez Ghanem Sector Manager/Director: Arvil Van Adams Task Team Leader/Task Manager: Eileen Murray  MADAGASCAR COMMUNITY DEVELOPMENT PROJECT CONTENTS A. Project Development Objective Page 1. Project development objective 3 2. Key performance indicators 3 B. Strategic Context 1. Sector-related Country Assistance Strategy (CAS) goal supported by the project 3 2. Main sector issues and Government strategy 4 3. Sector issues to be addressed by the project and strategic choices 5 C. Project Description Summary 1. Project components 5 2. Key policy and institutional reforms supported by the project 8 3. Benefits and target population 9 4. Institutional and implementation arrangements 10 D. Project Rationale 1. Project alternatives considered and reasons for rejection 12 2. Major related projects financed by the Bank and other development agencies 12 3. Lessons learned and reflected in proposed project design 13 4. Indications of borrower commitment and ownership 15 5. Value added of Bank support in this project 16 E. Summary Project Analysis 1. Economic 16 2. Financial 16 3. Technical 17 4. Institutional 17 5. Environmental 18 6. Social 20 7. Safeguard Policies 21 F. Sustainability and Risks 1. Sustainability 21 2. Critical risks 22 3. Possible controversial aspects 23  G. Main Conditions 1. Effectiveness Condition 23 2. Other 23 H. Readiness for Implementation 23 I. Compliance with Bank Policies 24 Annexes Annex 1: Project Design Summary Annex 2: Detailed Project Description Annex 3: Estimated Project Costs Annex 4: Cost Benefit Analysis Summary, or Cost-Effectiveness Analysis Summary Annex 5: Financial Summary for Revenue-Earning Project Entities, or Financial Summary Annex 6: Procurement and Disburserhent Arrangements Annex 7: Project Processing Schedule Annex 8: Documents in the Project File Annex 9: Statement of Loans and Credits Annex 10: Country at a Glance Annex 11: Coordination with other IDA financed projects focusing on rural development Annex 12: Background information on decentralization MAP(S) IBRD 20035RI  MADAGASCAR Community Development Project Project Appraisal Document Africa Regional Office AFC08 Date: March 23, 2001 Team Leader: Eileen Murray Country Manager/Director: Hafez Ghanem Sector Manager/Director: Arvil Van Adams Project ID: P055166 Sector(s): BD - Decentralization, SA - Social Assistance Lending Instrument: Specific Investment Loan (SIL) Theme(s): Social Protection Poverty Targeted Intervention: Y' Project Financing Data [ 1 Loan [X] Credit [ ] Grant [ ] Guarantee [ ] Other: For Loans/Credits/Others: Amount (US$m): Total of $US 110 over 5 years. Proposed Terms: Standard Credit Grace period (years): 10 Years to maturity: 40 Commitment fee: 0.5 Service charge: 0.75% Financing Plan: Source Local Foreign Total BORROWER 7.00 0.00 7.00 IDA 106.80 3.20 110.00 LOCAL COMMUNITIES 20.00 0)00 20.00 Total: 133.80 3.20 137.00 Borrower: GOVERNMENT OF MADAGASCAR Responsible agency: FONDS D'INTERVENTION POUR LE DEVELOPPEMENT (FID) Direction Generale Address: Lot II J - Villa "Alma" Ambodivoanjo - 101 Antananarivo MADAGASCAR Contact Person: Arsene Randriamamonjy Tel: (261-20) 22 42377 Fax: (261-20) 22 24689 Emai[: fid(dts.mg  Estimated disbursements ( Bank FYIUS$M): :;00$l:FY ; 0 i 2002 2003 2004 2005 V2006 X2007 Annual 15.00 20.00 20.00 20.00 25.00 I 0.00 Cumulative 15.00 35.00 55.00 75.00 100.00 110.00 Project implementation per od: January 1, 2002 to December 31, 2006 Expected effectiveness date: 09/15/2001 Expected closing date: 06/30/2007 The program targets poor and, especially, physically isolated rural communities. OCS PAD Fon- R.F Md, 2M) -2 -  A. Project Development Objective 1. Project development objective: (see Annex 1) The project development objective is to improve use of and satisfaction with project-supported social and economic services within poor rural communities. 2. Key performance indicators: (see Annex l) The indicators of attainment of the project development objective are: * A minimum of 80% utilization of designed capacity for: -Enrollment in primary schools -Visits to health centers -Uses of potable water supply -Hectares irrigated and cultivated -Vehicles per week that use rural roads * At least 70% of beneficiaries report they are satisfied with services from project-supported infrastructure. Note: Annex I sets forth the indicators of project outputs. B. Strategic Context 1. Sector-related Country Assistance Strategy (CAS) goal supported by the project: (see Annex 1) Document number: Report No. 16249-MAG Date of latest CAS discussion: 02/18/97 A central CAS objective is to assure that economic growth is broad-based and benefits the poor. Since a large majority of the poor live in rural areas, the CAS aims to reduce rural poverty by improving the productivity and competitiveness of the rural economy. To this end the CAS proposes investments in human capital development in rural areas, and in particular in education, basic health care, water, nutrition, and rural infrastructure. Moreover, the CAS supports strengthening the delivery of basic social and economic services in poor areas through decentralization of power and resources to communities. The Project addresses these objectives through intensive investment in provision of social and economic services in poor rural areas. Over the medium and long term, a sustained flow of improved social and economic services from the investments is expected to improve public health and education in the targeted communities. This increase in human capital would then contribute to higher labor productivity and incomes there. The Project would support the CAS objective of decentralization through significant devollution of powers to communities. Most importantly, all investments would be selected and implemented by community associations or by commune-level governments. The long-term vision is that the transfer of funds to communes will be successful so that communes will play a key role in rural development. Although its role would change, the FID would become a permanent instrument for targeted transfers of matching funds to the community and commune levels and for implementation activities. - 3 -  2. Main sector issues and Government strategy: Poverty in Madagascar. The population of Madagascar, to a large extent, has experienced a long descent into extraordinary misery as average real income per capita fell by a third over 1960-99. The growth of real GDP recovered in the 1990s but this did not reduce poverty rates. Instead, the poverty rate was broadly stable in the 1990s, rising from 70 percent in 1993 to 73% in 1997 and then slipping to 71% in 1999. Physical measures of nutrition are an alternative and basic indicator of poverty. Unfortunately, stunting (age/height) improved only marginally over the decade and so was roughly consistent with the poverty rate. However, there were some significant changes in the profile of poverty in the 1990s. The rural poverty rate rose from 74.5% in 1993 to 76.7% 1999, despite migration of poor rural families to urban areas --which led to a smaller increase in the poverty rate there. The worsening of rural poverty seems associated with physical isolation caused by a breakdown of the road system. The distribution of income among provinces shifted significantly in the 1990s. Toliara was the poorest province in 1993, with a poverty rate of 81%, but poverty fell to 72% in 1999. In contrast, poverty in Mahajanga deteriorated dramatically from 53% in 1993, which was the most moderate poverty rate in the country, to 76% in 1999 --well above the national average. Similarly, poverty in Antsiranana (Diego Suarez) was relatively modest in 1993 at 60%, but the rate deteriorated to 73% in 1999. The poverty status of Fianarantsoa also deteriorated rapidly, becoming the poorest province in 1999 with a poverty rate of 81%. Finally, poverty in Antananarivo moderated over the 1990s and reached 62% in 1999. Other indicators of poverty were more positive. In particular, the school enrollment rate of children 6 to 14 rose from 51% in 1993 to 68% in 1999, despite the decline in consumption. Living conditions were generally quite poor but tended to improve over the decade. Access to electricity in urban areas rose and access to sanitation services rose across the board. Access to clean drinking water rose from 18% in 1993 to 23% in 1999. The Interim Poverty Reduction Strategy Paper (PRSP). The Govemment's interim PRSP outlines a poverty reduction strategy built on three pillars: (a) higher and more sustainable levels of economic growth; (b) govemance and institutional reform; and (c) improved basic service delivery. Government Program. The Government plans a number of short-term actions in the context of its emerging poverty reduction strategy. These actions stress institutional development and good governance to enhance the environment for investment and promote stable growth and efficient delivery of basic services to the poor. Thus, key elements of the program are: (i) improving the budget control for public services, financial monitoring mechanisms and dissemination of information; (ii) strengthening the natural resource management system, especially in the forestry sector; (iii) strengthening judicial systems; (iv) improving rural infrastructure; (v) increasing access to education for the poor; (vi) broadening the health system; and (vii) establishing a poverty monitoring system. Highly Indebted Poor Country (HIPC) Relief and Use of Resources. Madagascar's external debt would be reduced significantly with the delivery of assistance under the enhanced HIPC Initiative. Debt service savings would amount to about US$ 59 million per year (1.1 percent of GDP) on average during the period 2001-2019. Government has provided IDA and IMF staff with a detailed plan for using those resources in support of its poverty reduction program. The plan allocates the resources to four areas: - 4 -  education, health, rural roads, and clean water for the poor. 3. Sector issues to be addressed by the project and strategic choices: The decentralization process is currently underway: the main framework law for decentralization at the provincial level was adopted in August 2000, and elections of provincial councils were held in December 2000. The restructuring of responsibilities will provide effective decision-making powers to local officials, who, in turn, will be fully accountable to the elected provincial councils. The transfer of responsibilities to the six provinces will take place over the next 18 months, on the basis of bilateral agreements with the central government. Relations between provinces, municipalities, and departments will be defined by provincial statutes and administrative tribunals will oversee provincial acts. The objective of decentralization is to improve the mix, flow, and quality of public services by decentralizing powers over the services and resources to the beneficiaries. Hence, the Government will transfer responsibility for delivery of services such as local roads, primary health and eduzation to the provinces, and to some extent justice and policy. The government recognizes that the success of decentralization will depend upon its ability to put in place a viable system of intergovernmental fiscal relations and to strengthen local institutions. The Project is multi-sectoral in nature as it will complement other Bank and donor interventions in education, health, rural roads and water supply, etc. The design has taken into account the Government's strategies in the above sectors and will be implemented in close coordination with the concerned sectoral ministries. As an example, the Government's Programme National pour l'Education has as one of it's main objectives to focus on the developmen,t of primary education in rural areas to which the project will make a significant contribution. C. Project Description Summary 1. Project components (see Annex 2 for a detailed description and Annex 3 for a detailed cost breakdown): This is a community-driven development project that employs two ways ('windows') for transferring funds to poor rural communities and then compares outcomes. * The Project transfers funds through the first window to community associations to finance community sub-projects (investments); * It transfers funds through the second window to commune governments to finance a sub-project which consists of the priority investments in their communal development plans. These two windows will co-exist and will not be two parallel systems: they will be funded from a single special account, and payments will be made by a single execution agency. During the mid-term review, the Bank project team will compare the development outcomes flowing from the transfers through these two windows. The criteria for the comparison will be: 1. Financial management (soundness, transparency); 2. Costs (whether communes build infrastructure that is no more expensive than the cost of similar infrastructures built by FID or by community associations); 3. Quality (similar quality of construction, as assessed by annual technical audits); and 4. Timeliness (the time the commune spends implementing an activity is not significantly greater than for similar activities executed by FID or by community associations). -5 -  If results are within a 10 percent margin of what has been executed in the first window by community associations, a proposal could be made to reallocate funds (at least 20 percent) from the first window to the second during the mid-term review. Building up the capacity of commune government is of great importance given the fluidity with which the decentralization process is moving and the uncertainty regarding fiscal and administrative decentralization. At the time of the mid-term review, any increase in the allocation for direct transfers to communes will also be based, among other criteria, on the level of development of communes and their capacity to manage funds. This project has been designed independently of the success and the timing of the decentralization process to provinces. The communes are in place, have elected officials and are receiving budgetary transfers; the project will enable communes to strengthen their managerial, financial, and developmental capacity. -6 -  Indicative Bank- % of Component Sector Costs % of linancing Bank- ..________________________ __WSW . . ... ..S... Total (US$M) financing 1. Transfers to Community Social Funds 80.00 58.4 64.00 58.2 Associations (Community sub-projects). Transfers of funds to community associations to finance community sub-projects (investments). This component funds sub-projects chosen through a participatory rural appraisal process and executed by community associations (or by FID in the case of rural transport infrastructure and small scale irrigation). IEC as well as service provision activities (such has training of teachers) will be integral parts of these sub-projects. 2. Capacity Building. Decentralization 15.00 10.9 12.00 10.9 This activity will train community associations and communes in project management, accounting, and procurement, etc. It will also support the preparation of participatory communal development plans. -7 -  3. Transfers of funds to communes. Social Funds 25.00 18.2 20.00 18.2 (each annual transfer is a sub-project that funds their priority investments). Funds will be transferred directly to targeted communes to enable them to implement priorities # I or #2 of their participatory communal development programs. Ex-post auditing mechanisms will be put in place to monitor fund usage on a semi-annual basis until the mid-term review. The direct transfers to communes will start off as a pilot as it will target 60 communes and 120 communes for the first and second years of project implementation respectively (out of the country's 1380 rural communes). At the time of the mid-term review, mechanisms have been built in to assess the development outcome and eventually allocate more funds to this component. 4. Administration. Social Funds 17.00 12.4 14.00 12.7 This funds administration and operating costs, social mobilization and community participation, ex-post audits of projects, training of staff, beneficiary assessments, impact evaluations, poverty studies, monitoring and evaluation, and financial and technical audits. Total Project Costs 137.00 100.0 110.00 100.0 Total Financing Required 137.00 100.0 110 00 100.0 2. Key policy and institutional reforms supported by the project: Project contributions to decentralization. The Project supports decentralization by financing investments selected by rural communities through a participatory process and executed by community associations. In addition, the Project supports decentralization at the commune level. The Project favors rural communities at the commune level because these are relatively small units -- comprising 5,000 to 12,000 people -- with potential for widespread participation in small investments. Furthermore, the Central Government has established the legal status of the communes, held elections of commune governments, and made small budgetary transfers. Moreover, the Project will test whether commune governments can successfully execute investments in - 8-  comparison with community associations. If successful, project resources will be shifted from the community association window and made available for transfer through the commune government window. The Project's advantage in promoting decentralization is its success in working with rural communities to realize small-scale community infrastructure sub-projects. The project execution unit is skilled at the engineering aspects of the infrastructure, in supervising construction and procurement, in disbursement, and in training. However, a successful decentralization program requires a much broader effort. Limits to the Project's contribution to decentralization. The methods applied in the Project are expected to improve the fiscal viability of sub-projects. Nevertheless, continued operation of the infrastructure will require the central Government to increase fiscal decentralization and communes to improve tax collection and management. Further fiscal decentralization requires an assessment of central government transfers to communes and the design of a credible system for monitoring these transfers. Contribution of complementary projects. Progress in several areas related to decentralization are outside the domain and strengths of a social fund. Therefore the Project cannot single-handedly implement all the institutional, fiscal, and administrative measures needed to reach the long-term vision of a decentralized state in Madagascar. Instead the success of the Project -- and of decentralization -- requires the support of complementary activities in the country program. Furthermore, fiscal decentralization can be achieved more effectively through a macro-economic dialogue focusing on this issue 3. Benefits and target population: Benefits. The Project benefits Malagasy who inhabit rural areas by: v Primarily, increasing their access to basic social and economic services and by improving the quality of the services; v Building the capacity of community associations and commune authorities to plan, implement and operate development sub-projects; v Providing communities and local authorities with the ability to seek additional funding for development activities in a systematic manner ; and by * Increasing productivity and income over the medium and long-term, as economic investments in infrastructure such as feeder roads improve the profitability of agriculture and as social infrastructure in health and education improve labor productivity. -9-  Targeting the poor population. The Project will select communes at the level of the six provinces to improve coordination with the Rural Development Support Project and to ultimately enhance their joint impact in the field. Even more importantly, the Project will assure that funds are transferred to poor communities through the use of poverty indicators. For transfers to community associations, the Project will transfer funds programmed for each year for use in the six provinces according to their shares in the rural population. (Note that about 90% of the rural population fall below the poverty line). Within each province, about the first 10% of communes serviced will be in the most enclaved quarter of communes --because enclavement is a major cause of severe poverty. The remaining 90% of communes served will be drawn from the poorest half of communes according to an index of school enrollment and visits to health clinics. In order to start transfers to communes quickly, the Project will disburse funds to communes that already have communal development plan financed by other partners (UJNDP, French Cooperation, European Union) during the first year of operation. After the first year of operation, the transfers to communes component will distribute funds across provinces in proportion to their share in the rural population. Within each Province, communes will be selected according to their implementation capacity. More specifically, the Project will give preference to communes which have communal development plans, which have experience in carrying out an investment specified in the plan, and that, at least, maintain administrative accounts. To ensure that this targeting is done objectively, the project will finance a study of poverty in the various communes and seek coherent and complete data across the country. 4. Institutional and implementation arrangements: Executing Agency. FID will execute the Project because it successfully executed earlier social funds in Madagascar. Since the FID is one of the key Government instruments supporting its decentralization policies, it will continue to maintain a regional office in each province. Moreover, the FID is becoming a more community-based funding instrument and therefore has recruited experienced community mobilizers in each of its regional offices. The FID may also have to make other staffing changes as the project evolves such as recruiting experienced IEC and training specialists to ensure it has the appropriate and competent skills mix. Due to the large activity volume, it is envisaged that FID will also have to sub-contract part of the work to NGOs and local consulting firms. The FID is a nonprofit association (une association a but non-lucratifreconnue d'utilit public). It is governed by a six-person board of Directors including representatives from the Prime Minister's Office, the donor community, NGOs, and socio-professional organizations. In addition, two directors are elected by members of the "Association FID." It is envisaged that FID's membership will be enlarged to include mayors of rural communes. The operation of the FID is regulated by its Articles of Agreement, by-laws, and Manual of Procedures and by its Framework Agreement (Convention) with the Government. Program Coordination. The FID Director General will be responsible for coordinating the program in collaboration with the FID Regional Directors. FID will continue to respect sector policies established by the Ministries of Education, Health, Agriculture, Energy and Mines, and Civil works. Moreover, the FID has signed formal agreements (Protocoles d'accord) with the Ministries of Health and Education that specify the rights and responsibilities of each party. Decentralized ministerial staff will participate with - 10 -  FID in all desk and field appraisals of sub-projects to minimize redundant expenditures. Accounting, financial reporting and auditing arrangements. A Financial and Administrative Director in the FID's central office will oversee its accounts, financial reports, and audit arrangements. (The requirements for these are defined in the Manual of Procedures). Each of the FID's six regional offices has an accountant responsible for regional accounts as well as the electronic transmission of regional accounts to the central level, using the computerized accounting and financial management system. For the transfers to communes, it is envisaged that semi-annual ex-post audits of commune expenditures will[ take place and FID is currently working on the modalities, including timing and objectives, and preparing, terms of reference so that these audits can provide considerable input in the evaluation of the transfers to communes component. Monitoring and Evaluation Arrangements. The MIS system has been further upgraded to take into account the further decentralization of project activities to the level of communes and communities. The monitoring and evaluation system (Systeme d'information FID) has become a powerful tool for tracking FID activities. The system includes modules on: (i) management of beneficiary requests from the time of receipt of the request until the completion of works; (ii) management of follow-up documents related to sub-projects described in the manual of procedures; (iii) management of the annual work program; (iv) management of contracts with local consulting firms, small and medium enterprises (SMEs), and NGOs; (v) preparation of the monthly reports; (vi) preparation of quarterly reports required for LACI; and (vii) monitoring performance indicators (outcome as well as impact). FID is in the process of linking electronically technical data from its MIS to data from its accounting system. Also, beneficiary assessments will be undertaken every other year, sub-project ex-post evaluations annually, and technical audits annually. To complement FID's monitoring and evaluation system, communities will monitor and evaluate their own sub-projects. FID regional offices have agreed on output and outcome indicators that comamunities or communes can monitor easily and reliably before, during, and after implementation of a sub-project. Coor(dination in the Rural Areas with other IDA Financed Projects: the Rural Development Support Project (RDSP), Rural Transport Project (RTP), & Second Environment Project. T he Bank CDP and RDSP have agreed on a common design for participatory rural appraisals through which the population will identify social as well as productive investments and rank them by order of priority in all communities where they are working. On the targeting issue CDP and RDSP will intervene in the same geographic areas (based on the above mentioned poverty criteria) to cumulate and strengthen their impact. The four projects operating in rural areas will furthermore agree on compatible annual work programs. CDP will continue to finance small scale irrigation sub-projects in areas where the RDSF' project is not in a position to intervene. Moreover, the CDP will finance rural transport sub-projects in collaboration with RTP, but the latter will take the lead on ensuring proper operation and financing of the road maintenance fund for periodic tertiary road maintenance. In addition, the CDP will make use of the policies and recommendations such as provisionr of basic access, labor intensive works modalities when they become available. The Groupes de Travail ze Developpement Rural (GTDR) will participate in the meetings of regional consultative committees of FID that will approve CDP supported sub-projects (See Annex 11 for detailed coordination modalities). - 11 -  D. Project Rationale 1. Project alternatives considered and reasons for rejection: The main design issue facing the Govemment and the Bank team is how to further devolve power to use program funds to communes and communities. In 1995, an earlier generation of this social fund took a significant step by decentralizing most activities of the executing agency to regional offices. To gain experience in community implementation, FID III has implemented community management of funds and procurement in many sub-projects. The executing agency channels these funds directly to the community associations, which then define, plan, initiate and follow-up on the bidding process and then supervise construction using a participatory process. FID is moving now towards participatory rural appraisal mechanisms for sub-project (investment) selection. Under the FID III Supplemental Credit for the Rehabilitation of the 2000 Cyclone Damages, community associations select and implement all investments in rehabilitation. Moreover, the community as a whole meets to validate --or reject-- all rehabilitation investments selected by community associations. The question is how to devolve implementation to communes in a way that pernits timely delivery of high-quality investments. The constraints on devolving implementation are the communes' ability to increase their capacity through training and leaming-by-doing, the need to respect national sector policies, and Bank requirements for adequate financial management, procurement, and supervision. The Government and the Bank are still learning how to devolve implementation effectively in the context of Madagascar. Therefore, the Government and the project team have jointly decided to advance community-driven development through a mix of approaches: a majority of investments will be implemented entirely by community associations while some will be implemented through commune governments. 2. Major related projects financed by the Bank and/or other development agencies (completed, ongoing and planned). = - 0 : : 00 t ~~~~~~~~~~~~Lateost Supervision Sector issue Project : (PSR) Ratings ____________________________ _____________ I: (Bank-financed projects on@} Implementation Development Bank-financed Progress (IP) Objective (DO) Improved health status of the Second Health Sector Support S S population and quality health services. Project (Cr. 3302-MAG) Increased agricultural productivity Rural Development Project Provision of primary education, Education Sector Development S S particularly in rural areas. Project (Cr.3046-MAG) Delivery of safe water in rural areas Rural Water Supply Project U U (Cr.3025-MAG) Improved rural transport infrastructure Rural Transport Project--APL S S 3 (Cr. (Cr.3364-MAG). Reduced poverty and food insecurity in Community Nutrition II Project S S rural areas. (Cr.3060-MAG) - 12 -  Second Irrigation and S S Rehabilitation Project (Cr. 2644-MG) Other development agencies Improved rural infrastructure European Union "Projet Micro-realisations" Improved communal roads NORAD/ILO Broader access to primary education French Cooperation--PEM Broader access to primary health and UNICEF education facilities IPIDO Ratings: HS (Highly Satisfactory), S (Satisfactory), U (Unsatisfactory), HU (Highly Unsatisfactory) 3. Lessons learned and reflected in the project design: The Project design incorporates lessons from social funds supported by the Bank and by other agencies in many regions of the world as well as from the Bank's experiences with two earlier social funds in Madagascar and with an on-going social fund -- FID III (Implementation Completion Report for the Food Security and Nutrition Project, Cr.2474-MAG and Implementation Completion Report for Social Fund II Project Cr. 2778-MG). The Implementation Completion Report for the Social Fund II concluded that the FID is now a key rural development project in Madagascar, and is a valuable complement to the health, education, micro-finance, nutrition, transport, agriculture and other operations. The public and the Government perceive the FID as the mechanism that has actually provided support for rural communes and communities by financing close to 1,800 community development sub-projects nationwide. As a result, it is assumed that the Project will maintain the momentum created by the earlier projects. Effectiveness of an independent execution unit. This Implementation Completion Report summarizes several specific lessons learned. Projects designed and managed outside of Government can recruit more competent staff, and better paid staff get better results; but they cannot be completely isolated from political interference at the center. This will be addressed in the new Project by further ernpowering commune governments and communities so that they can take the lead at all stages of the project cycle. The component that transfers resources to communes will take this step further as communes will become responsible for project implementation and therefore for countering political interference from central government structures. Also, FID will make additional investments in community participation mobilization. This will probably increase FID's operating costs as it strengthens its comtnunity outreach and training functions. Respect of strategies of sector ministries. The activities supported by the Project will conform with strategies of sectoral ministries, not only at the national level but within the decentralized structure. Continued efforts will be made by FID at the central level as well as regional levels of concerned ministries. For instance, FID signs agreements with various line ministries regarding its intended activities in the sector, and in return, the concerned Ministry agrees to provide adequate staffing, training, and support when necessary. Line Ministries on the whole have been honoring their commitments regarding the provision for necessary inputs for operation of sub-projects. There have been some instances where FID schools lacked teachers due to the imposed hiring freeze from the Bank and IMF on civil service recruitment. Some health facilities also lacked trained nurses, but this is in the process of being corrected - 13 -  since the MOH has reopened its nurse training program. FID will also maintain regular contact with ministerial staff at the decentralized level and will not only involve them in the project approval process but will ask them to participate in desk and field appraisals of sub-projects as another means to ensure that they remain involved in the long-term. Training. As a result of FID training programs for contractors and site supervisors, the quality of sub-projects continued to improve significantly. Additional training requests have been made for micro-irrigation. Senior technicians and management may also be trained. Courses will be offered for sociologists and other employees of consulting firms dealing with community development. FID, as part of the CDD initiative, will expand its training activities by training communal staff and communities in project management, selection, implementation, accounting, procurement, disbursement, etc. Benefits of investments in basic infrastructure. Beneficiary assessments in 1999 and 2000 showed that: (a) construction of schools was associated with better working conditions for teachers, increased enrollment rates, decreased repetition of grades, and increased parent involvement in the schools; (b) health investments were associated with increased visits to health posts; (c) small irrigation projects showed a positive rate of return; and (d) investment in feeder roads was associated with an increase in the revenues for truckers and farmers selling agricultural products. General lessons learned from experience around the world (which is not exhaustive) applied to the Community Development Project are: Financial Management. Another important lesson learned is to establish incentives for good management of project funds. Specifically, the executing agency should be subject to: (i) guidelines, independent financial and technical audits, and a management information system that promotes transparency; (ii) simple procurement and disbursement procedures; (iii) competitive salaries to enable the recruitment and retention of qualified staff; (iv) targeting mechanisms for sub-projects; (v) integration of the environmental dimension into sub-projects; and (vi) monitoring and evaluation mechanisms, including beneficiary assessments. A Quality Enhancement Review took place in February 2000 and the following recommendations are taken into account in the design of this project. Community mobilization and sustainability. The most important lesson learned from earlier social funds in Madagascar --and elsewhere is probably that true community participation maximizes the sustainability of sub-projects. Adequate investments in community mobilization are essential before disbursing funds to communities, to ensure that communities agree on and choose priority activities, feel a stake and ownership towards them and maintain them over time. The proposed Project budgets for these mobilization activities. Regarding maintenance, the new financing agreement between FID and the beneficiary includes a maintenance clause regarding specific obligations of beneficiaries to maintain sub-projects once they are completed. However, sustainability in social funds can only be fully achieved if the roles and responsibilities of the central and local governments are clearly defined and fulfilled in the context of the project. Harmonization of Policies. The Project is evolving toward closer cooperation with beneficiaries and with the agencies and authorities working in the same communities. This will be done by creating a formal mechanism for community groups and communes for selecting, in a participatory manner, sub-projects to be requested. This is already taking place through meetings to develop priorities for the community in - 14 -  general (and not only tied to FID). Targeting and Resource Allocation. In order to ensure that the poorest communities benefit, the Project will identify remote areas that have not yet benefited from a sub-project and ensure that promotion efforts are applied. The Project will also finance some capacity-building activities and provide local consultants to assist in project design. Investment Cycle. The Project will continue its efforts to ensure that all sub-projects (investments) are identified in a participatory manner. To the extent possible, it will ensure that marginalized groups, such as women and children, are brought into the decision-making process as a means to ensure cven greater sustainability. It envisages expanding the process to take into account indicators and issues such as participation, environment, maintenance, and sustainability issues. All contracting will be done by the beneficiaries with technical assistance from FID, which will assist in the areas of design Df sub-projects, in cost effective and sustainable technology, and improving management. The beneficiary associations or communes will monitor maintenance and service delivery performance. Service Provision. The Project would expand the options of the transfers to community associations component to finance the direct provision of certain basic social and economic services. The list of options will be kept as broad as possible so that the Project remains open to innovative types of investments in services. The team has explored options for infrastructure and service provision activities to be integrated into one sub-project, and infrastructure development will be complemented by programs such as teacher/nurse training/retraining, community strategic planning, etc. FID will not be responsible for funding both components but can require that communities receive information on possible partners. Some service delivery activities such as training of health workers will be standardized under partnership agreements with other development organizations. Procurement. The experience of the implementation of community sub-projects where [DA funds were transferred directly to beneficiary association bank accounts has been very positive. To the team's knowledge, there has been no misuse of funds, probably because of all the safeguards that were built into the procedures manual to ensure complete transparency (FID giving the first check in a public ceremony, informing all association members of all contract awards, amounts, and expenditures, in addition to the control which was exercised by FID staff). Similar procedures will be used for piloting the "Transfers of Funds to Communes" component. 4. Indications of borrower commitment and ownership: The Government, and especially the Prime Minister, strongly support the social funds established with the Bank and have also supported the FID. Furthermore, the Office of the Prime Minister acts as the supervising Ministry (Ministere de tutelle) of the Project. Moreover, the social funds have been supported by communities because the communities have been able to control the development process and become less dependent on the state. The Govemment has demonstrated its commitment over the years by always paying counterpart funds for the social funds in a timely manner, which is not the case for all projects in Madagascar. Some further evidence of commitment is that neither the Prime Minister nor other cabinet members have interfered in the past four years in the day-to-day management of the FID. From meetings with technical ministries, it is clear that the social funds are perceived as Malagasy projects, supporting the Government's various sectoral policies. Also, social funds executed by the FID are perceived by the rural population as projects helping the community. - 15-  5. Value added of Bank support in this project: The Bank has acquired a significant amount of knowledge and experience of Social Funds in many countries. The Bank has produced and refined systems for operating social funds that are expressed in manuals of procedures, and simplified procurement and disbursement policies. Finally, the Bank is acquiring experience in decentralization and the modalities for fully empowering the lowest tiers of Government. Although several donors in Madagascar contribute to poverty alleviation projects, resources are inadequate to finance the increasing development needs of the rural populations. The Bank's contribution can be significant because it is by far the largest donor in the sector. E. Summary Project Analysis (Detailed assessments are in the project file, see Annex 8) 1. Economic (see Annex 4): O Cost benefit NPV=US$ million; ERR = % (see Annex 4) * Cost effectiveness O Other (specify) Research into social funds has shown that cost-effectiveness analysis is more feasible and reliable than cost-benefit analysis or calculation of rates of return. That is because it is difficult to measure the social benefits of social funds. Therefore, the economic evaluation of the Project focuses on a comparison of the unit costs of rehabilitation and construction by the executing agency with that achieved by other agencies and by the Government. Some limited information on the benefits of social funds is nevertheless available from beneficiary assessments, mid-term reviews, and annual technical audits. Some partial indicators of benefits are drawn from these sources and used to supplement the cost-effectiveness analysis. 2. Financial (see Annex 4 and Annex 5): NPV=US$ million; FRR = % (see Annex 4) Cost sharing and cost recovery. The purpose of community cost-sharing is to assure that the community places a high priority on the sub-project, to establish an incentive to maintain the sub-project, and to develop the community's capacity to mobilize its own resources to operate and maintain the sub-project. The community contribution may be in any combination of cash, construction materials, and or labor. Communities will be expected to contribute at least: * 15% of the cost of investments (sub-projects) that communities select and implement; * 10% for rural roads (of which 5% is a contribution to the cost of rehabilitation and 5% is set aside in a financial institution for road maintenance). Through supervision, the Bank team will take all the necessary steps to ensure that these contributions are actually made for all sub-projects. Beneficiary contributions for the on-going project are 20% and primary indications show a greater degree of ownership from beneficiary communities which will hopefully lead to better maintenance of sub-projects. For many sub-projects, part of the maintenance costs will be borne by community associations and the remainder will be paid from commune budgets. - 16 -  Furthermore, the Project will finance capacity building activities in accounting and financial management to assist communal staff in managing their funding allocations, making and forecasting payments, tracking expenses, and dealing with other financial matters. Financial Controls and Fiduciary Aspects. The manual of procedures of the FID addresses the fiduciary aspects of transfers of funds to community associations and communes. This manual covers the procedures that will have to be followed by community associations and communes for procurement, accounts and financial reporting on the use of funds. In the case of direct transfers of funds to communes (which will be piloted until the mid-term review), semi-annual audits of the use of funds will be carried out. Fiscal Impact: N/A 3. Technical: The annual FID technical audits report that the quality of infrastructure has continued to improve. Therefore, FID now has a track record of building quality infrastructure in most cases al: competitive prices and more quickly than other projects. (Schools and health posts take four to six months to be completed.) After each technical audit, the consultant wrote an action plan identifying the weaknesses of each type of sub-project and defining measures to remedy the weaknesses over the next calendar year. Whenever possible, FID uses architectural drawings approved by the various line Ministries, and when these are not available, works with concerned Ministries to develop standard drawings. In addition, the executing agency is now examining options to use cement blocks, which technically are at par but are less expensive and involve less deforestation and pollution than traditional bricks. Labor intensive works are preferred where proven technically and economically efficient. 4. Institutional: 4.1 Executing agencies: The Program will be executed by FID and commune governments. However, the skills mix of the FID staff will be re-visited in view of the different approaches that this program will take toward community development and capacity-building of communes. FID management is in the process of revising its organizational structure to ensure that competent staff with the appropriate skills mix are in place. Prior to effectiveness, an IEC specialist, an MIS specialist and a training specialist will be recruited at the central level as well as an IEC and a training specialist in each of FID's 6 regional offices will also be recruited. 4.2 Project management: Project management on the whole has been satisfactory. Provided that there is no political interference, FID staff have the capacity to satisfactorily implement this project. 4.3 Procurement issues: Technical audits have been undertaken every year and as part of those audits contract award procedures were examined for selected projects. In addition, a procurement audit was undertaken in 1999 where a sample of about 100 contracts were reviewed in-depth. Following these audits a number of improvements have been made, which include: (a) the elimination of bracketing for the purpose of bid awards in order for - 17 -  contracts to be awarded to the lowest evaluated bidder without any reference to the concept of "remaining capacity;" and (b) procedures being followed for the recruitment of consultants are completely in line with bank guidelines. Consultants and contractors have also been given more time to undertake studies and to prepare their proposals and bids which is beginning to improve the quality of works. The procedures to be followed under the proposed project take into account the above recommendations. They have been tested by FID for sub-projects implemented by communities, and have proven to be the most economic and efficient method of procurement. 4.4 Financial management issues: A computerized accounting and financial management system based on internationally accepted accounting principles is operational. This system operates on a decentralized basis with the six (6) regional directorates. Each regional Directorate maintains separate financial records for all transactions under its responsibility and sends, on a monthly basis, the financial information to the General Directorate for consolidation. The Financial and Administrative Director, at the central level, is in charge of overall financial management aspects, including budgeting, administration of the special account, consolidation and production of the project's financial statements. To strengthen the project's financial management system, an agreed action plan by the borrower has been proposed following the assessment carried out by a Bank/IDA Financial Management Specialist. The main measures to be taken include: i) updating the current chart of accounts to reflect project activities and satisfy reporting requirements; ii) updating the manual of procedures section on accounting; iii) the recruitment of an accounting firm acceptable to IDA to perform the project financial statements audit. Prior to credit effectiveness the Bank/IDA Financial Management Specialist will evaluate progress made by the project in the implementation of this action plan. 5. Environmental: Environmental Category: F (Financial Intennediary Assessment) 5.1 Summarize the steps undertaken for environmental assessment and EMP preparation (including consultation and disclosure) and the significant issues and their treatment emerging from this analysis. N/A 5.2 What are the main features of the EMP and are they adequate? N/A 5.3 For Category A and B projects, timeline and status of EA: Date of receipt of final draft: 5.4 How have stakeholders been consulted at the stage of (a) environmental screening and (b) draft EA report on the environmental impacts and proposed environment management plan? Describe mechanisms of consultation that were used and which groups were consulted? The execution agency recently conducted an in-depth consultation with 27 rural community groups. 5.5 What mechanisms have been established to monitor and evaluate the impact of the project on the environment? Do the indicators reflect the objectives and results of the EMP? The manual of procedures was revised as a result of the environmental study of February 1999. The manual now requires the signature of an annual framework agreement between the FID and the Office National pour l'Environnement (ONE) for each type of sub-project. This framework agreement defines: - 18-  (i) the training of FID staff on environment-related matters; (ii) a plan to train and create awareness on environmental concerns for the agency's partners (community associations, local consulting; firms, communal authorities, the regional consultative committee of the agency, etc.); (iii) the best procedure to follow for joint approval of sub-projects; and (iv) monitoring and evaluation systems to fol low-up on the impact of the proposed measures on each type of investment. - 19-  6. Social: 6.1 Summarize key social issues relevant to the project objectives, and specify the project's social development outcomes. The project has been designed to take into account the structure of the country's rural society, and in particular the structure of rural communities -- where for example notables, such as teachers or doctors, lead the decision-making process of their communities. The awareness-creation campaigns will ensure that these notables promote actions that render investments sustainable, through the operation and maintenance phases. 6.2 Participatory Approach: How are key stakeholders participating in the project? This Project is unusually participative in the context of a country where community participation is not commonplace. The Project was developed in part through the feedback of beneficiaries in annual survey assessments. In addition, consultations took place with deputies, rural mayors, NGOs, decentralized ministerial structures, and civil society to solicit their views as how best to address issues in rural areas. A summary of the consultation process which took place is available from FID. The Project will support participatory rural appraisals through which the population will identify sub-projects and rank them in order of priority in a communal development plan. The communities will also be asked to participate by co-financing and by executing sub-projects proposed in the communal development plans. The Project, like the earlier projects, relies on beneficiary assessments for feedback on its impact. Results of these assessments are used to improve the way the executing agency works with beneficiaries and other stakeholders and in the way it implements its sub-projects. 6.3 How does the project involve consultationls or collaboration with NGOs or other civil society organizations? NGOs will execute many Project activities and will receive training. NGOs will be responsible for promoting community participation, training and technical assistance in areas where they have expertise. Also, in some cases NGOs will work with the communities in a participatory manner to implement sub-projects such as wells and other small sub-projects. 6.4 What institutional arrangements have been provided to ensure the project achieves its social development outcomes? The Project aims to achieve its social development outcome --of reinforcing community institutions-- through its work with community associations and commune governments, and especially through the participatory process for selecting projects. These aspects of the Project have been described above. 6.5 How will the project monitor performance in terms of social development outcomes? The key measure will be whether the communities have been able to implement top priorities outlined in their communal development plans. - 20 -  7. Safeguard Policies: 7.1 Do any of the following safeguard policies aly to the project? P9tlcy Applicabilit Environmental Assessment (OP 4.01, BP 4.01, GP 4.01) 0 Yes * No Natural habitats (OP 4.04, BP 4.04, GP 4.04) 0 Yes 0 NIo Forestry OP 4.36, GP 4.36) 0 Yes 0 No Pest Management (OP 4.09) 0 Yes * No Cultural Property (OPN 11.03) 0 Yes 0 NIo Indigenous Peoples (OD 4.20) 0 Yes 0 No Involuntary Resettlement (OD 4.30) 0 Yes * rNo: Safety of Dams (OP 4.37, BP 4.37) 0 Yes * No_ Projects in International Waters (OP 7.50, BP 7.50, GP 7.50) 0 Yes 0 No Projects in Disputed Areas (OP 7.60, BP 7.60, GP 7.60) 0 Yes 0 No 7.2 Describe provisions made by the project to ensure compliance with applicable safeguard policies. During the environment study that took place two years ago, FID agreed with the Office National pour l'Environnement (ONE) that all of the agency's sub-projects are category "1" sub-projects vwhich, according to the new version of the national legislation, will require only a limited impact environmental study. A provision has been made in FID's manual of procedures that during the appraisal process, a limited impact environmental study will be undertaken for each sub-project. F. Sustainability and Risks 1. Sustainability: To summarize, design features which promote sustainability are: * Devolving the power to select sub-projects (investments) to communities to assure thar they assign a high priority to the investments and are thus more likely to operate and maintain them; * Requiring community contributions to sub-project costs, as a further guarantee that the investments are community priorities, and to provide an incentive for communities to operate and rnaintain sub-projects to avoid losing their own contribution; * Devolving implementation (including financial management, procurement, and mainte:nance) to community associations and commune governments, to more deeply involve communi 'ies; * Training of community associations and commune governments in financial management, procurement, and in the prioritization, selection, operation and maintenance of investments in order to provide them with the capacity and ability to sustain the sub-projects; and * Linking the activities of the executing agency to sectoral development strategies and involving decentralized ministerial structures in desk and field appraisals of individual sub-projects to ensure that line ministries will honor their agreements to provide staff, equipment, and a recurrent budget for the sub-project by the time of completion. Nevertheless, the sustainability of the Project depends on increased fiscal decentralization from the central government to communes and on the successful execution of complementary investments in rural development, as explained in the section immediately below. - 21 -  2. Critical Risks (reflecting the failure of critical assumptions found in the fourth column of Annex 1): Risk Risk Rating Risk Mitigation Measure From Outputs to Objective The Government does not provide N This has not been a problem in the past. counterpart funds on schedule. However, the Bank Team and the executing agency will work with the Government to ensure that adequate funds are budgeted. The central Government does not follow S The country team is planning a macro-economic its policy of fiscal decentralization and credit that will focus on fiscal and other aspects therefore the communes are not able to of decentralization, and this will help ensure that finance their share of the cost of the communes receive adequate funds from the construction or rehabilitation of their center. In addition, the executing agency is infrastructure. working with community groups so that, if necessary, they can assume some responsibilities for maintaining and operating their investments. Moreover, community experience in raising contributions should reinforce their own capacity to step in to replace any shortfall in central government funding. Sub-projects are not coordinated with line M The executing agency will obtain assurances ministries, either directly or through FID's from concerned Ministries prior to starting Comites Consultatifs Regionaux. works. To ensure adequate staffing by health workers and teachers, the executing agency will continue to sign annual agreements with concerned line ministries to further ensure that the line ministry will fulfill its part of the agreement. Also, decentralized ministerial structures will participate in the desk as well as field appraisals of sub-project. Finally, the agency will stop works in the relevant sector if the problem persists. An adequate volume of consultants, M For the past two years, the executing agency has contractors, artisans, skilled labor, NGOs, not been able to further accelerate and suppliers is not available to work implementation of sub-projects due to the with the project. absence of skilled partners. In the proposed Project, the agency will continue to conduct training programs for its partners. An adequate volume of trainers and of M The proposed Project will train trainers while training facilities is not available to work the development of training centers falls within with the project. the scope of the education program and projects. From Components to Outputs The Government does not provide N See above counterpart funds on schedule. - 22 -  The Government intervenes in the M This has not been a major impediment to project day-to-day management of the execution execution in the past. The devolution of agency. increased decision-making powers of use of Project funds to community associations and to communes is expected to reduce the scope of central government intervention in sub-projects. Overall Risk Rating M Risk Rating - H (High Risk), S (Substantial Risk), M (Modest Risk), N(Negligible or Low Risk) 3. Possible Controversial Aspects: N/A G. Main Loan Conditions 1. Effectiveness Condition 1. The Convention (subsidiary agreement) has been executed on behalf of the Borrower and FID; 2. The Director General, Technical Director, Financial and Administrative Director, Communications Director, training specialist, IEC specialist, and monitoring and evaluation specialist cf the central office, as well as the Provincial Director, IEC specialist, and training specialist of each regional office have been duly recruited under terms, conditions, and terms of reference acceptable to the Association. 3. FID has appointed the financial auditors under terms, conditions, and terms of reference acceptable to the Association. 4. The Procedures Manual has been duly approved by the Association and adopted by Fll); and 5. A project account has been duly opened, and an initial amount of the equivalent of $500,000 deposited therein. 2. Other [classify according to covenant types used in the Legal Agreements.] H. Readiness for Implementation D 1. a) The engineering design documents for the first year's activities are complete and ready for the start of project implementation. 0 1. b) Not applicable. X 2. The procurement documents for the first year's activities are complete and ready for the start of project implementation. 1 3. The Project Implementation Plan has been appraised and found to be realistic and ofL satisfactory quality. D 4. The following items are lacking and are discussed under loan conditions (Section G): - 23 -  1. Compliance with Bank Policies Z 1. This project complies with all applicable Bank policies. X 2. The following exceptions to Bank policies are recommended for approval. The project complies with all other applicable Bank policies. /