Madagascar - Third Social Fund Project

Document of The World Bank Report No. 1 8967-MAG PROJECT APPRAISAL DOCUMENT ONA PROPOSED CREDIT IN THE AMOUNT OF SDR 10.7 MILLION (US$15 MILLION EQUIVALENT) TO THE REPUBLIC OF MADAGASCAR FOR A TH[RD SOCIAL FUND PROJECT FEBRUARY 16, 1999 Human Development IV AFC08 Africa Region  CURRENCY EQUIVALENTS (Exchange Rate Effective 08/27/98) Currency Unit = Malagasy Franc (FMG) 1,000FMG= US$0.18 US$1.00 = 5500.00 FMG 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 (Cr6dit Education) EU European Union FID Social Development Fund (Fonds d 'intervention pour le developpement) IBRD Intemational Bank for Reconstruction and Development ICB Intemational Competitive Bidding. IDA Intemational Development Agency MIS Management Information System NCB National Competitive Bidding NGO Non-Govemmental Organization PCU Project Coordination Unit Vice President Callisto Madavo Sector Manager Arvil Van Adams Country Director Michael Sarris Task Team Leader Eileen Murray  REPUBLIC OF MADAGASCAR THIRD SOCIAL FUND PROJECT TABLE OF CONTENTS Block 1: Project Description ............................................................3 1. The Project Development Objectives and key performance indicators ........................3 2. Project components ...........................................................3 3. Key policy and institutional reforms supported by the project .....................................3 4. Benefits and target population ...........................................................3 5. Institutional and implementation arrangements ............................................................4 Block 2: Project Rationale ...........................................................S. 1. CAS objectives supported by the project ........................................................... 5 2. Main sector issues and Government strategy ............................................................5 3. Sector issues to be addressed by the project and strategic choices ................................5 4. Project alternatives considered and reasons for rejection ..............................................5 5. Major related projects financed by the Bank and/or other development agencies ........6 6. Lessons learned and reflected in the project design ......................................................6 7. Indications of borrower commitment and ownership ....................................................7 8. Value added of Bank support .........................................................7 Block 3: Summary Project Assessments ...........................................................8 1. Economic Assessment ...........................................................8 2. Financial Assessment (see Annex 5) .................................,. 8 3. Technical Assessment .................................8 4. Institutional Assessment .................................8 5. Social Assessment .................................9 6. Environmental Assessment .................................9 7. Participatory Approach ...................................9 8. Sustainability .................................9 9. Critical Risks (see fourth column of Annex 1) ................................ 10 10. Possible Controversial Aspects ................................ 10 Block 4: Main Loan Conditions ................................... 11 1. Effectiveness Conditions .................................. 11 2. Other ......................... ,. . .11 Block 5: Readiness for Implementation ............................. .1. Block 6: Compliance with Bank Policies ............................ I I  Annexes Annex 1. Project Design Summary Annex 2. Detailed Project Description Annex 3. Estimated Project Costs Annex 4. Economic Justification Table 1. Number of Sub-projects financed under FID I and II (up to June 1998) Table 2. Key Performance Indicators for FID lI Table 3. Number of Sub-projects to be financed underthe proposed project Table 4. Key Performance Indicators for the proposed project Table 5. Social Indicators - Madagascar and Comparators Annex 5. Financial Summary Annex 6. Procurement Disbursement and Financial Management Arrangements Table A. Project Costs by Procurement Arrangements Table B. Thresholds for Procurement Methods and Prior Review Table C. Allocation of Loan Proceeds Annex 7. Project Processing Budget and Scheduling Annex 8. Documents in Project File Annex 9. Statement of Loans and Credits Annex 10. Country at a Glance Map: IBRD Reference number 20035R1  INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT INTERNATIONAL DEVELoPMENT ASSOCIATION Africa Regional Office AFC08 Project Appraisal Document Madagascar Third Social Fund Project Date: February 12,1999 1 [] Draft [X1 Final Task Manager: Eileen Murray Country Manager: Michael N. Sarris Sector: Human Resources Sector Manager: Arvil Van Adams Project ID: 18584-MAG Program Objective Category: Poverty Reduction Lending Instrument: Specific Investment Loan PTI: [XI Yes [ I ] No Project Financing Data [] Loan [x] Credit [ I Guarantee [ Other [Specify] For Loans/Credits/Others: Amount (US$m/SDRm): US$15 million/SDR10.7 million Proposed Terms: [x] Multi-currency [ Single currency Grace period (years): 10 years [I Standard Variable [ Fixed [1 LIBOR-based Years to maturity: 40 years Commitment fee: N/A % Service charge: 0.75 % Financing plan (US$m): Source Local Foreign Total Government 1.6 1.6 Cofmanciers N/A IDA 9.4 5.6 15.0 Beneficiary Communities 1.1 1.1 Other (specify) Total 12.1 5.6 17.7 Borrower: Republic of Madagascar Guarantor: N/A Responsible agency(ies): Fonds d'Intervention pour le Developpement (FID) Estimated disbursements (Bank FY/US$M): 1999 2000 2001 Annual 5.0 5.0 5.0 Cumulative 5.0 10.0 15.0 For Guarantees: [] Partial Credit [ Partial risk N/A Proposed coverage: Project sponsor: Nature of underlying financing: Terms of financing: Principal amount (US$) N/A Final maturity Amortization profile Financing available without guarantee?: I[ Yes [ No If yes, estimated cost or maturity: Estimated financig cost or maturity with guarantee: N/A Expected effectiveness date: July 1, 1999 Expected Closing date: December 31, 2001   Project Appraisal Document Page 3 Madagascar Third Social Fund Project Block 1: Project Description 1. The Project Development Objectives and key performance indicators The Project Development Objectives and key performance indicators (See annex 1) (PDO) are similar to those of the on- going Social Fund II Project (Cr. 2778-MAG) which has been highly successful in supporting Madagascar's rural population since 1994. The mid-term review of the on-going project concluded that FID II met or surpassed its development objectives in parallel with disbursements being about a year ahead of schedule, and concluding that FID will now focus on testing approaches involving greater community participation. In this context, Project Development Objectives (PDO) will include: (1) improved access of poor rural population to social and economic infrastructure, (2) on a pilot basis, empowerment of poor rural communities and/or conununes to identify, organize, manage funds and implement sub-projects responsive to community needs in order to achieve decentralized social and economic development; (3) employment creation; and (4) improved capacity of private sector local small contractors and consultants, artisans, skilled labor, and NGOs. Key indicators for development, by type of sub-projects financed by theFonds d'Intervention pour le Developpement (FID) under the project are shown in Annex 1 and in Table 3 of Annex 4. Those figures are only estimates since FID is demand- driven. They will be subject to change according to the requests that the rural conmnunities/comunues themselves submit to FITD during project implementation. The 350 communal infrastmcture projects to be rehabilitated or constructed will benefit about 260,000 people; about 188 permanent jobs and 1,600,000 person/days of temporary employment will be created. Under the 30 capacity building sub-projects to be implemented, about 200 rural communities and communes, 150 micro-enterprises, 75 consultants, 50 artisans and 20 NGOs will be trained. 2. Project components: Component Ca Cost Incl. % of Contingencies (US$M) Total (1) Comunual infrastructure rehabilitation and Physical 15.0 85% construction (2) Capacity building for communes, hural Institutional Building and 0.5 3% communities/local associations, and private "Others" sector (small contractors, consultants, arisans, skilled labor and NGOs)--in very specific cases study tours could be envisaged abroad. (3) FID equipment purchases, operating costs, staff Institution Building and other 2.2 12% training, audits and studies Total 17.7 100% 3. Key policy and institutional reforms supported by the project The project supports the Government's poverty agenda which is summarized in a Letter of Sector Policy dated July 28, 1995. 4. Benefits and target population FID's target population will continue to be poor rural communities. The types of sub-projects to be implemented will: (a) have a direct impact on the well-being of the population while using a participatory community approach to the delivery of basic social services; (b) pilot mechanisms for rural community groups (especially women) and/or communes to identify, plan, and implement sub-projects responsive to community needs (details in Annex 2 para. 14); (c) support the development of rural communities, communes, small contractors, consultants, artisans, and NGOs; and (d) provide temporary and permanent employment opportumities. FID will finance projects only in rural areas where 80 percent of the population is considered poor in Madagascar. SinceFID's operating costs under the on-going project have been less than 15 percent, the assumption is that poor rural communities will benefit directly or indirectty from at least 85 percent of the proceeds of this project.  Project Appraisal Document Page 4 Madagascar Third Social Fund Project 5. Institutional and implementation arrangements Implementation period: 2 years Executing agencies: Fonds d 'Intervention pour le Developpement (ED) Project coordination: Project coordination will be the primary responsibility of the Director General of FlD in collaboration with FID's 6 Regional Directors (1 for each province). FID has made a big effort to abide by thesectoral policies put in place by the various Ministres, and therefore, is in regular contact with all concerned ministries (Health, Education, Public Works, Energy and Mines, and Agriculture). In addition, in the case of the Ministries of Health and Education, there have been formal agreements which have been signed between those ministries and FID which explicitly specify the role and responsibility of FID. At the regional level, FID's Regional Directorates closely follow FID's Manual of Procedures which clearly stipulates that individual FID sub-projects must be approved by the Regional Directorate of the concerned Mnistry prior to the start-up of any field activity. Project oversight (policy guidance, etc.): The project will continue to be implemented by the FID which is a non-profit association forpublic service (une association a but non-lucratifreconnue d'interet public). FU operates in accordance with the provisions of its Articles of Agreement, by-laws, and Manual of Procedures and the Framework Agreement(Convention) between FID and the government. The FID Association operates under a six person Board of Directors. The Board includes one representative each from the Prime Minister, the donor community, NGOs, and socio-professional organizations, in addition to two Directors elected by members of the "Association FID". Accounting, financial reporting and auditing arrangements: At the central level, a Financial and Administrative Director oversee FID's accounting, financial reporting, and auditing arrangements (which are also clearly defined inFID's Operating Manual). The Financial and Administrative Director's responsibility is to oversee FD's entire accounting, auditing, and financial reporting system which has been decentralized in the 6 regional directorates since 1994. Each of theFID's 6 respective regional directorates has an accountant who is responsible for accounting forFID's transactions in each of the regional offices. He/she is also responsible for providing all justification of transactions on a monthly basis to the Centml Level which reviews them and replenishes regional project accounts on the basis of proper accounting documentation. FID has a computerized accounting/financial management system which is being used by all the regional offices and data transmission from FID's regional offices to the central level is done electronically. Therefore, FID can now provide up to date accounting and financial information for each regional office. Given the extensive reliance on the Bank' s 90-day Special Account advance procedures and SOEs, special attention will be paid by the Financial Management Specialist in the resident mission in verifying SOE reliability, and to the internal procedures and controls involved in SOE preparation and the flow of funds from the Central Special Account and the decentralized (regional) accounts. Given the relatively short implementation period, it is not envisaged that LACI type disbursements will be introduced. An assessment of FID's financial management system is in the project files. Monitoring and evaluation arrangements: Monitoring is an integral part of each sub-project implemented. FID has about 1,500 sub-projects which have been completed or which are on-going. FID is hi the process of expanding its data base capabilities to maintain detailed information on each sub-project (stage of the sub-project, total cost, beneficiary contribution, cost per square meter or other unit cost, training issues and modalities, maintenance aspects, in the case of schools enrollment rate information, in the case of health centers attendance figures, impact on women, etc.). This data base has been decentralized to FID's regional offices which will be responsible for entering detailed information on each individual sub- project which in turn will be transmitted electronically from the regional offices and compiled by the central level. Approximately 30 percent of monitoring and evaluation resources will be directed towards the pilot activities which this operation will finance directly through conmnunes/community groups to gather data and information as to how best to design the larger community development social fund (FY00). FID will also continue to undertake detailed evaluations of selected sub-projects semi-annually (at the time of the technical audit and at the time of the annual review with the Bank). In addition, FID will transmit updated information on a quarterly basis to its Board of Directors and to the Bank. It is envisaged that FID will make the effort to also sharesectoral information with concerned line ministries at the central, regional as well as local levels. As a result of FD piloting sub-projects through communes and community groups, FID's monitoring and evaluation system will have to monitor the community, communal aspect as well as selected individual households. Supervision missions will use this information as one of the bases for discussions.  Project Appraisal Document Page 5 Madagascar Third Social Fund Project Block 2: Project Rationale 1. CAS objective(s) supported by the project: Document number and date of latest CAS discussion: Report No. 16249- MAG, discussed by the Board in February 1997. The project is consistent with the Bank Group CAS discussed by the Executive Directors on February 18, 1997 The CAS main objectives are poverty reduction through high rates of broad-based economic growth, and human capital Development (focused on primary education, basic health care and rural infrastructure). To ensure that the poor will indeed be better off, the CAS proposes activities that would improve rural transport, education and health, including nutrition and access to clean water. A key focus of the CAS is also to assist in the empowerment of local communities to tackle their development challenges. The proposed project will address all of those objectives. 2. Main sector issues and Government strategy Poverty has become entrenched in Madagascar's social fabric. In 1993, Seven out of 10 Malagasy were considered to be poor (i.e., they cannot purchase a minimum 2100-calorie food basket and essential non-food items); six out of 10 are extremely poor (i.e. they cannot purchase even the food basket). Based on Govermment estimates in 1997, it appears that only 3 out of 10 Malagasy are now considered to be extremely poor. Govermuent data also indicates that theGini coefficient has increased from 44% in 1993 to 48% in 1997. Over 85 percent of the poor live in rural areas, where poverty is also deeper. About 18 percent of poor households are headed by women, engaged mostly in farming. Poor households have around 5 members, compared to 4 for non-poor households, and are more likely to: be illiterate; not send children to school; have malnourished children; use wood as main source of fuel; and rely on lakes/ponds/rivers for drinking water. It is clear that rural poverty, which has as its main cause the poor productivity and competitiveness of the agricultural sector, has been exacerbated by the lack of productive and social infrastructure and other services, and an over centralized decision-making process for the allocation of resources. The Government's central objective is to steadily reduce the incidence of poverty by half from about 70 percent in 1996 to 35 percent in 2015. This target can only be realized by achieving very ambitious economic growth rates, while reversing environmental degradation and improving the assets of the poor. Pro-poor growth will call for inter alia: improved sector policies promoting rural development; more efficient public investment and restructuring of public expenditures; and encouragement of local initiatives. 3. Sector issues to be addressed by the project and strategic choices Sector issues to be addressed by the project are the widespread incidence of poverty (particularly in rural areas), the failure of the state to deliver basic social services to the population, the lack of access to markets and the lack of facilities for productive activities. By increasing the availability in rural areas of economically-efficient and beneficiary-driven productive and social infrastructure, the project will contribute to the modernization of the rural economy and enhance rural well-being. The project will explore concrete options of supporting the decentralization process in Madagascar by working directly with communes and other decentralized organizations. It will be one of the main instruments which will have a positive impact on poverty reduction in rural areas by empowering beneficiaries to become key actors in the rural development process--this scenario should ensure that beneficiaries who have had a say in the communal development process (and have also made financial contributions) will also follow-up to ensure that investments made are properly maintained and therefore sustainable. 4. Project alternatives considered and reasons for rejection The Government and Madagascar's rural population perceive FID as the main donor financed instrument which has been able to finance a large array of community development projects nationwide. The alternative to this Social Fund mechanism would be to work directly with line ministries. However, FID's comparative advantage is its ability to directly reach out to the population by giving them an opporbtmity to voice their opinions in terms of prioritization of community development activities. FID has also been able to identify a niche for itself by becoming complementary to efforts of line ministries and other IDA and donor financed projects. Therefore, the first step in the identification and preliminary design of the project has been a series of meetings with line ministries and other donors to establish clearly the areas where FID interventions would be needed and welcome, and would be consistent with the sectoral priorities and policies of the Govermment. For example, there is plenty of room for FID sub-projects for water supply in rural areas, and the rehabilitation of micro irrigated perimeters is well  Project Appraisal Document Page 6 Madagascar Third Social Fund Project suited for FID interventions, as opposed to that of small irrigated perimeters taken care of by other projects. Similarly, the financing of schools by FID complement what will be done under the IDA financed Education project CRESED II and by other donors, and FID's interventions in the health sector are most likely to be limited to health posts (postes de sante') which are not covered by other sources. 5. Major related projects fnanced by the Bank and/or other development agencies (completed, ongoing and planned). Sector issue Project Latest Form 590 Ratings (Bank-financed projects only) IP DO Bank-financed Inadequate health infrastructure and low utilization of Health Sector Improvement Project S S health services (Cr. 2251-MG) Inadequate primary school infrastructure Education Sector Reinforcement S S Project CRESED 1 and CRESED 2 (Cr. 2094-MG and Cr. 3046-MG) Lack of access to safe water in rural areas Rural Water Supply Project S S (Cr. 3025-MG) Inadequate urban infrastucture in Antananarivo and Urban infrastructure Project S S selected secondary cities (Cr. 2968-MG) Poverty and food insecurity in rural areas Second Irrigation Rehabilitation S S Project (Cr. 2644-MG) Inadequate transport infrastructure Transport Sector Project (under preparation) Limited support to decentralization process Support to Decentralization Process (under preparation) Other development agencies Inadequate rural infrastructure and poverty alleviation European Union "Projet Micro- realisations" Inadequate communal road network CAP/USAID Inadequate communal road network NORAD/ILO Poor quality of health facilities GTZ Health Project Poor quality of health facilities UNICEF Health Project 6. Lessons learned and reflected in the project design The project builds on the experience of on-going FID, other IDA financed social funds in Africa and in other parts of the world. Lessons learned show that Communities no matter how poor, are generally willing to contribute a share of the costs of a sub- project in kind or in cash, if they are convinced that the subproject will be implemented and will contribute to their well-being. Continued monitoring of project impact is necessary to learn from operational experience and modify the approach and procedures accordingly. Lessons learned from FID (and other on-going projects in Madagascar) are that in order to be successful a social fund needs: (i) political support from high level Government officials who do not interfere with the project's day-to-day management; (ii) consensus as to the goals and integrity of the social fund; (iii) a qualified and decently-paid staff; (iv) the existence of well defined and transparent management and accounting procedures; and (v) regular and' frequent financial, managerial and technical audits. FID's mid-term review in October 1998 concluded that FID is now the key rural development project in Madagascar which will be complemented by the Transport Sector and Micro-finance Projects and other operations. It is clearly perceived as the mechanism which has actually come through and supported rural communes/communities by financing over 1500 conmnunity development sub-projects nationwide. As a result, FID has created a momentum which this project will maintain. One of the main conclusions of this review was that in light of the on-going decentralization process which has given financial as well as managerial autonomy to the communes, FID would pilot sub-projects to support directly  Project Appraisal Document Page 7 Madagascar Third Social Fund Project poor rural communes to identify, organize, manage funds and implement sub-projects responsive to their needs. FID would in essence channel these funds directly to the communes/communities who would in tum take on the leadership role and defme, plan, initiate and follow-up on the bidding process, supervise construction and maintain their sub-projects. This new approach would give comnunes/communities full responsibility as the key decision makers and operators in the community development process. Beneficiary contributions will also be increased to a minimum of 10 percent (20 percent for communal sub-projects) to ensure a higher degree of ownership (Annex 2para. 14). It is anticipated that by the time this project closes, over 50 percent of sub-projects will be implemented by communes/communities. This project should be considered as a "bridge loan" to the larger FID IV Project which will fully support communallcommunity development initiatives on a large scale basis. Lessons leamed from communal/community involvement will be fully integrated into the larger FID project to ensure that communes/communities are the leaders in terms of development activities which will benefit their communities. This is in line with the decentralization process where communes will have financial, administrative, and managerial autonomy to decide how best to proceed in terms of overall community development. 7. Indications of borrower commitment and ownership From the beginning, the Governnent has been very supportive of FID's objectives and operations. Over the years, Government has given FID the necessary counterpart funds. The Prime Minister's Office has exercised general oversight of FIT) activities, but has not interfered in FID's staffmg and day-to-day operations. From recent meetings with all technical ministries working directly or indirectly with FID, it is clear that FITD is perceived as a Malagasy project which supports Govenrnent'ssectoral investment plans. Also, FID is now fully accepted by the population at-large, whose confidence has not been eroded by the fact that FID has not been able to respond favorably to all requests (although by the mid-term review, about 1,500 sub-projects had been completed or were under implementation under the ongoing Social Fund II Project, FITD had been able to address only 20 percent of the requests received). 8. Value added of Bank support The Bank has acquired a significant amount of knowledge during the past decade on the design and implementation of social funds as mechanisms to quickly and efficiently channel funds to the most vulnerable groups and to empower rural communities to take care of their own development This is exactly what has happened with FID, as community groups are becoming full- fledged actors in the development process. In addition, the Bank has "fine-tuned " most of the social funds documents, such as manual of procedures, including guidelines for simplified procurement and disbursement for community-based investments, and therefore the cost and time required for preparation are minimized. The Bank's involvement in the ongoing Social Fund II Project has shown that the Bank can play a key role in efforts that combine poverty alleviation and development. The proposed project complements the Bank efforts in Madagascar in rural development, education, health, infrastructure and private sector development.  Project Appraisal Document Page 8 Madagascar Third Social Fund Project Block 3: Summary Project Assessments 1. Economic Assessment [ Cost-Benefit Analysis: NPV=US$ [ Cost Effectiveness [x] Other million; ERR= % Analysis: See annex IV Given the multi-sectoral nature of the sub-projects, a range of benefits is expected, many of which are not quantifiable. Since the project is demand driven it is difficult to determine a priori the composition of FID's portfolio, and an overall NPV or ERR has not been estimated. The economic benefits include improved standard of living in beneficiary areas due to retmns from sub-project investments in social and economic infrastructure, the creation of income earning opportunities through employment creation, and increased productivity and sustainability through increased ownership by conununes/communities of created conmnunal assets. There are also indirect increases in incomes in beneficiary areas through multiplier effects. The expected social benefits include enhanced community organizational capacity to undertake conmnunity-prioritized projects, as well as improved collaboration between communal leaders, associations' representatives and local sector officials in carrying out a process of cross-sectoral planning coordinated at the local level. Simple screening procedures will be applied to ensure targeting of benefits to poorer and more vulnerable groups and to establish the economic justification of sub-projects. For revenues earning sub-projects, a mmiimum economic or financial rate of return will be required. Non-revenue earning sub- projects will be subject to cost-effectiveness analysis. 2. Financial Assessment (see Annex 5). This being a Social Fund Project, there are no major financial issues. 3. Technical Assessment The design and standards of sub-projects will be consistent with the norms and standards of the technical ministries involved, and will be based on the use of simple, appropriate and environmentally sound technologies, corresponding to the needs and capabilities (for construction, as well as operation and maintenance) of the population. Labor intensive works will be preferred whenever proven as technically and economically efficient and insofar as they generate employment opportunities that the local population can take advantage of. 4. Institutional Assessment a Executing agencies: The project will continue to be implemented by the FID which is a non-profit association for public service (une associafion a but non-lucratif reconnue d'interet public). FID operates in accordance with the provisions of its Articles of Agreement, by-laws, and Manual of Procedures and of the Framework Agreement (Convention) between FID and the Government. The Association FID operates under a six-person Board of Directors. The Board includes a representative each of the Prime Minister's Office (who is Chairman of the Board), the donor community,NGOs, and socio-professional organizations, in addition to two Directors elected by the members of the Association FID. b. Project management: FID is managed by a General Manager (DirecteurGen6ral) who was recruited based on a competitive recruitment process in 1995. His performance has been satisfactory since he has been able to increase substantially the pace of project implementation. His other strength has been the fact that he has been able to coordinate with all technical ministries concerned who perceive FID as an "ally" who is helping them implement their sectoral policies. The mammum amount of FID financing or grant for an infastructure project is US$50,000 equivalent, with the exception of transport infrastructure project such as communal roads for which the ceiling is US$100,000. The modalities for implementation vary, depending on the type and cost of sub-projects. FID has a Manual of Procedures which clearly defines identification, preparation, appraisal, selection and implementation modalities for each type of sub-project In order not to block smaller scale sub-projects, regional directors can approve projects of up to US$20,000 fnancing per project, the General Manager up to US$45,000 (or US$90,000 for transport infrastructure projects) and anything above requires the approval of FID's Board of Directors (for an estimated 10% of sub-projects). In any case, beneficiaries are expected to contribute a mninimum of 10 percent of the cost of each sub-project in cash or in-kind (except for feeder roads where the beneficiary contribution will be 5 percent). Sub-projects presented by communities willing to provide higher contnbutions are given a higher priority for implementation due to the fact that they lead to higher sustainability. Beneficiary assessments are in the process of being launched to determine minimum beneficiary contributions per type of sub-project. All pilot conununity/conununal sub-projects will require a minimum 20 percent contribution.  Project Appraisal Document Page 9 Madagascar Third Social Fund Project 5. Social Assessment Special attention will be paid to the involvement of women in the decision-making process, particularly to give priority to sub- projects which reduce their household tasks (rural water supply, and improved access to local markets and health centers), increase their income (construction of markets) or improve their living conditions (rural health centers and sanitation). Women will also be represented in the users' associations. The construction/rehabilitation of schools will ensure that girls continue to have access to education (female enrollment is not a problem, since girls account for close to 50 percent of enrollments). 6. Environmental Assessment Environmental Category [] A [x] B [ i C Justification/Rationale for category rating: The project could have a limited environmental impact for certain types of sub- projects. As a result, an environmental assessment has been undertaken prior to board presentation to: (a) provide an overview of potential environmental issues for selected types of sub-projects; (b) assess any necessary capacity building within the project unit to examine and evaluate projects which could have a negative environmental impact; and (c) draft the "enviromnental checklist" for sub-projects to be financed by FID to be inserted into the Manual of Procedures. Eligible sub- projects will continue to include reforestation and erosion control. Prior to final project approval, all FID sub-projects will be subject to the environmental checklist. 7. Participatory Approach Identification/Preparation Implementation Operation Beneficiaries/coimmunity groups COL* COL* COL* Intermediary NGOs COL COL COL Academic institutions N/A N/A N/A Local government COL COL COL Other donors COL COL COL Local Consulting Firms COL COL COL Local Contractors) COL COL COL COL: Collaboration* Those contracted are the primary beneficiaries 8. Sustainability The sustainability of FID's sub-projects will be achieved by ensuring maximum community involvement in the identification, preparation, appraisal, selection and implementation of a sub-project. In addition, FID has to ensure that the target beneficiaries (and not political figures) are actually making the in-cash or in-kind contributions. Experience also shows that the more substantial the beneficiary contnbutions, the better chance that the sub-project will be properly and efficiently utilized by the community and the better chances that the proper funds are allocated towards maintenance. During the last year, FID has made a substantial effort in creating awareness among the various communities who have benefited from sub-projects as to the need for maintenance and is allocating funds to train community members in proper maintenance techniques. Cost recovery mechanisms have been put in place for a large number of PID sub-projects mainly in the areas of health where health posts are now charging user fees, certain feeder roads which are charging tolls, etc. In addition, in the cases of primary schools and health posts, agreement has to be reached with the sectoral ministry (prior to the start of the works) that the appropriate personnel (teachers in the case of school and nurse/mid-wife in the case of health post) will be transferred to that facility. All projects financing rural development activities, including FID will have to coordinate cost recovery policies in coordination with concerned line ministries to ensure coherence from one project to another. Lastly, FID has explicitly requested that a maintenance clause be inserted in all civil works contracts which basically stipulates that contractors are also responsible for training community groups in maintenance activities. Regarding rural roads, FID and other concerned parties will have to make a strong case to budget resources from the road fund for road maintenance activities as communal budgets alone will not be able to finance road maintenance activities. This is coherent with the new "CharteRoutiere" in Madagascar which breaks down roads into three different types (primary, secondary, and tertiary). On this basis, the road fund should provide an estimated $300/km per year for the maintenance of teriary roads to be rehabilitated by FID and the other $300 matched by the communes. This is also coherent with the "Charte Routiere" and other legislation pertaining to decentralization whereby tertiary roads are "owned" by communes and therefore have to be maintained by them.  Project Appraisal Document Page 10 Madagascar Third Social Fund Project 9. Critical Risks (see fourth column of Annex 1): Project outputs to development objectives Risk Risk Rating Risk Minimization Measure - The decentralization policy is not Modest - Close follow-up from the Bank and other donors to ensure implemented. the policy is implemented in addition to the fact that the Bank is preparing a Decentralization LL. - Technical ministries to provide additional Modest - FUD management to ensure adequate coordination with line inputs (such as teachers or nurses) to ensure ministries, both at the central and regional levels, including the satisfactory operation of the completed through the Comites Consultatifs Rfgionaux (CCR) of FID. communal infrastructure. Project components to outputs Risk Risk Rating Risk Minimization Measure - Rural communities/communes do not have Modest IDA fnanced Decentralization LIL currently under sufficient capacity to implement sub- preparation will pilot training programs for selected projects. communes/communities. - Inadequate coordination with line ministries, Modest either directly or indirectly through the FID over the past 2 years has invested substantially in Comites ConsultatifsRegionaux of FID. coordination activities with the Ministries of Health, Education, Public Works, Agriculture, etc. and its actions are - Unavailability of a sufficient number of consistent with sector policies. consultants, contractors, artisans and skilled Modest labor, NGOs, suppliers, and trainers/training As part of this project, FID will continue to train these local facilities. partners through formal training activities and on the job. - Government will interfere in FID's day-to- Modest Legal documents and manual of procedures are very explicit day management. as to the role of the Government and any remedial action that could be envisaged. Overall project risk rating Risk Risk Rating Modest 10. Possible Controversial Aspects: N/A  Project Appraisal Docunent Page 11 Madagascar Third Social Fund Project Block 4: Main Loan Conditions 1. Effectiveness Conditions (a) the FID Convention has been executed on behalf of the Borrower and FID; (b) the Revised Manual of Procedures is satisfactory to IDA; and (c) an amount of $100,000 has been deposited into the Project Account. 2. Other The following Board Conditions have been fulfilled: (a) Satisfactory assessment of FID' s financial management system and establishment of an action plan addressing any needed improvements; (b) An environmental assessment has been carried out; and (c) The Government has submitted to the Association a letter confirming the undertaking made by the Government in its letter of Sector Policy dated July 28, 1995 under the Second Social Fund Project. Execution of the Project. The Govenmment/Executing Agencies will: (a) Ensure that the positions of Director General and Regional Directors of FID are filled at all times with personnel whose qualifications and experience are satisfactory to IDA. (b) Conduct a comprehensive annual project implementation review no later than November 30 of each year of project implementation and a mid-term review by June 30, 2000. Financial Covenants (See Annex 6 on procurement and disbursement atrangements) Financial Management, Disbursement and Special Accounts (See Annex 6 on procurement and disbursement arrangements) Procurement and Consultant's Services (See Annex 6 on procurement and disbursement arrangements) Block 5: Readiness for Implementation [] 1. a) The engineering design documents for the first year's activities are complete and ready for the start of project implementation. [] 1. b) Not applicable. Ix] 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 of satisfactory quality. [] 4. The following items are lacking and are discussed under loan conditions (Section G): Block 6: Compliance with Bank Policies [x] roject co h 1 applicable Bank policies. T anager: Eileen Murray Sect Manager: Arvil Van Adams Country Director: Michael N. Sarris  Annex 1 Page 1 of 3 Third Social Fund Project Project Design Summary Note: Since the FID is demand driven, the figures mentioned in this annex on the number and type of sub-projects as well as the number of beneficiaries are indicative only; they are subject to change based on the requests made by the communities themselves. Natfr*~~~~~$w~~as~~~y. K~~~y ai ~ ~ ~ 110, I ,, CAS Objective (CAS Objective to Bank Mission). Poverty reduction through high Outcomes monitored would * Socio and economic rates of broad-based economic include, for example: surveys. * Macro-economic reforms growth in the medium to long and decentralization could term, and human capital * Accessibility of rural roads * Education, health and go according to schedule. development (focused on * Job creation employment statistics. primary education, basic health * Decrease in mortality rates * Continued willingness of care and rural infrastructure) in * Increase of primary * Public Expenditures Government to support the short to medium term. To enrollment rates Reviews (PER). poverty alleviation efforts. ensure that the poor will indeed be better off, the CAS proposes activities that would improve rural transport, education and health including nutrition and access to clean water. A key focus of the CAS is also to assist in the empowerment of local communities to tackle their development challenges. ___ Project Development (Development Objectives to Objectives CAS Objective) (1) Improved access to social a Percentage of children * Project impact monitoring . Maintenance of primary and economic infrastructure. held back every year system (i.e., socio- road network . Increased health center economic impact studies). * Trained teachers available attendance * Trained and quality health * Increased availability of * Quarterly and annual workers available water at the village level progress reports prepared * Drug distibution system a Increased agriculture by FID. operational production at the village level * Annual fnancial and * Increased community technical audit reports income by transportation submitted by fnancial of agricultural production auditors and independent to rural markets. consultants. * Supervision mission reports by Bank staff. . Beneficiary assessments by independent consultants. Baseline and targeted values should be shown, with the latter divided into values expected at mid-term, end of project and full impact.  Annex 1 Page 2 of 3 m= ~~~~~~~~~~~~~ s Project Development (Development Objectives Objectives (continued) to CAS Objective) Improved capacity of private Contractors, NGOs, and local Local employment and training sector: local small contractors consultants available to opportunities contribute to and consultants, artisans, undertake labor based works increase economic activity at skilled labor and NGOs. the community level. Project Outputs (Outputs to Development Objectives) A total of 350 communal * Number of infrastructure * Project impact monitoring * The decentralization infrastructure sub-projects sub-projects completed: system (i.e., socio- policy is not implemented, implemented. * 130 schools; economic impact studies). so that the communes have * 40 health centers; adequate financial * 30 water supply; * Quarterly and annual resources to properly . 35 MPI; progress reports prepared operate and maintain * 40 markets; by FI). communal infrastructure. . 50 feeder roads; a 25 bridges and other. * Annual financial and * Technical ministries do technical audit reports not provide additional submitted by financial inputs (such as teachers or auditors and independent nurses) to ensure the A total of 30 capacity building Number of capacity building consultants. satisfactory operation of sub-projects implemented for sub-projects implemented: 30. the completed communal rural communities/communes, * Completion report infrastructure. small contractors, local prepared by FID and Bank consultants, arfisans, skilled staff. labor, and NGOs. * Supervision mission reports by Bank staff. . Beneficiary assessments by independent _______________________ _______________________ consultants. 2 Baseline and targeted values should be shown, with the latter divided into values expected at mid-term, end of project and full impact.  Annex 1 Page 3 of 3 Note: Since the FID is demand driven, the figures mentioned in this annex on the number and type of sub-projects as well as the number of beneficiaries are indicative only; they are subject to change based on the requests made by the communities themselves. Project Components (Components to Outputs) Communal nfastructure * Total cost of US$ 15.0 * Quarterly and annual rehabilitation and construction nmillion, including progress reports prepared * Rural communities! (schools, water supply, health contingencies, by FID. communes do not have posts, markets, micro-irrigated sufficient capacity to perimneters and rural roads). * Bisbursement prepared by implement sub-projecls, FR) and Bank. and remain committed to community development Capacity building for * Total cost of US$ 0.5 * Financial audit reports communes, rural communities / million, including prepared by idependent . Inadequate coordination local associations, and private contingencies. auditors. with line ministries, either sector (small contractors, directly or through the consultants, artisans, skilled * Bank supervision mission Comifes Consultatffs labor, and NGOs). reports. Regionaux (CCR) of FID. Unavailability of a Project Management . Total cost of US$ 2.2 sufficient number of million, including consultants, contractors, contingencies. artisans & skilled labor, suppliers, NGOs, and trainers/training facilities. * Government will not interfere in FID's day-to- day management. 3 Baseline and targeted values should be shown, with the latter divided into values expected at mid-term, end of project and full impact.  Annex 2 Page 1 of 5 Third Social Fund Project Detailed Project Description Project Objectives 1. The Project Development Objectives (PDO) are: (1) improved access of poor rural population to social and economic infrastructure; (2) on a pilot basis, empowerment of poor rural communities and/or communes to identify, organize, manage funds and implement sub-projects responsive to community needs in order to achieve decentralized social and economic development; (3) employment creation; and (4) improved capacity of private sector: local small contractors and consultants, artisans, skilled labor, and NGOs. Detailed monitoring targets by type of sub-projects financed by FID under the project are shown in Annex 1 and in Table 3 of Annex 4. Those figures are only estimates since FID is demand- driven. They will be subject to change according to the requests that the rural communities/communes themselves submit to FID during project implementation. The 350 communal infrastructures to be rehabilitated or constructed will benefit about 260,000 people; about 188 permanent jobs and 1,600,000 person/days of temporary employment will be created. With the 30 capacity building sub-projects to be implemented, about 300 rural communities and communes, 150 micro-enterprises, 75 consultants, 50 artisans and 20 NGOs will be trained. 2. The strategy is to address the incidence of poverty amongst the country's poorer communities through the replenishment of a social fund designed to support demand-driven initiatives developed by such communities and, in doing so, to create employment and improve access to basic social services. The proposed project will support community development initiatives which appear to be the most responsive to the population's needs. FID will address the widespread incidence of poverty in rural areas of all the six provinces of Madagascar, and the primary beneficiaries will be mainly poor rural communities. The proposed project will continue the experiment successfully begun under the first two projects of a participatory approach for decentralized development. The types of sub-projects to be implemented will: (a) have a direct impact on the well being of the population by supporting innovative approaches to the delivery of basic community social services; (b) provide temporary and permanent employment opportunities; (c) provide a mechanism for rural communities to identify and participate in the implementation of sub-projects responsive to community needs; and (d) support the development of small local contractors, consulting firms, NGOs, artisans and skilled labor. Once the sub-projects are completed, communities will be expected to finance their operation and maintenance. Project implementation 3. The proposed project is well prepared and ready for implementation, since it is for the most part a continuation of activities that are carried out satisfactorily under the ongoing project. The only exception is the experiments that FID is about to launch under the ongoing project and will expand under the proposed project to test various options for financing infrastructure sub-projects that will be managed and implemented by the rural communities and/or communes themselves. A preliminary design exists for those experiments, but it is likely to evolve over time as FID draws the lessons of experience. 4. The proposed project will finance the construction and rehabilitation of basic infrastructure and capacity building. FID's procedures are clearly defined in a Manual of Procedures, which is part of the Project Implementation Plan. The Manual of procedures recalls the fundamental objectives of the FID, and establishes: (a) the types of sub-projects, and the selection and evaluation criteria; (b) the beneficiaries; (c) the terms and conditions of financing; (d) the procedures whereby FID receives and  Annex 2 Page 2 of 5 processes requests from beneficiaries, including who has the authority to approve sub-projects and grants; (e) sub-project implementation and supervision, including model contracts between the FID and beneficiaries; (f) the procurement procedures, including sample bidding documents and model contracts for sub-project engineering, execution and supervision; (g) the disbursement and accounting procedures, and the requirements for audit by independent external auditors; (h) the financial management procedures; and (i) the monitoring and evaluation procedures, and the nature and periodicity of reporting by the FID. 5. The Government will continue to provide IDA funds to FID on a grant basis. The maximum amount of FID financing or grant for a sub-project is US$50,000 equivalent, with the exception of transport infrastructure (rural roads, bridges, etc.) for which the maximum is US$100,000 equivalent per sub-project. The criteria for sub-project selection vary by type of sub-project. In all cases, sub-project selection procedures include: (i) formation of an association of beneficiaries responsible for the identification of the sub-project, the provision of the community contribution and the subsequent maintenance of the completed sub-project; (ii) in order to ensure ownership, assurances that the community's contribution would not be less than 5 to 10 percent of the total cost of each sub-project; (iii) an assessment of cost effectiveness as reflected in the cost per beneficiary or other indicator - sub-projects are subject to economic analysis wherever appropriate; (iv) an assessment of environmental impact wherever appropriate; (v) consistency with the sector strategies and programs, and adherence to norms established by the technical ministries; (vi) costs of sub-projects which are within the capacity of the population to pay its share; and (vii) provision for maintenance of infrastructure and sustainability of sub- project operation beyond the project period, including recovery of operating costs where appropriate. To this end, FID will require that communities provide evidence of their ability and willingness to ensure appropriate maintenance of the sub-project beyond the period of FID financing as one of the criteria of sub-project selection. FID does not finance recurrent activities or sub-projects that: are of an emergency nature; already receive funding from other sources; have been eliminated from government or local authorities' investment programs; may have a negative impact on the environment; and are too sophisticated or heavily mechanized. Also, the FID will not finance sub-projects in any community where the beneficiaries of previous FID sub-projects have not complied with their commitments regarding the implementation, management and maintenance of those previous sub-projects. Project Description Total project cost: US$ 17.7 million (IDA credit: US$ 15.0 million, Government contribution: US$1.6 million, and beneficiary contributions US$ 1.1 million). 6. The types of sub-projects to be financed will basically be the same as those financed under the ongoing project, since communities want more of the same. However, as FID does not intend to become a financial institution, the proposed project will not include any income-generating activities. More emphasis will be given to training activities to build the capacity of not only consultants and contractors, but also of local associations of beneficiaries and communes. As part of project preparation, there has been a series of meetings with line ministries and other donors to establish clearly the areas where FID interventions are needed and welcome, and are consistent with the sectoral priorities and policies of the Government.  Annex 2 Page 3 of 5 7. It is estimated that, under the proposed project, FID will finance a total of 380 sub-projects: 350 communal infrastructure sub-projects and 30 capacity building sub-projects. A tentative distribution by type of sub-project is given in Annex 1 and in Table 3 of Annex 4. Those figures are indicative only; they are subject to change based on the requests made by the communities themselves. Project Component 1: Rehabilitation/construction of Communal Infrastructure - Total estimated cost: US$ 15.0 million (IDA credit: US$12.4 million; Government contribution: US$ 1.4 million, and beneficiary contributions: US$ 1.1 million). Context 8. The contribution of community organizations in assessing local needs and capabilities forms an important part of social policy, and community organizations may be best placed to implement certain activities. In particular, the contribution of community groups can be critical to the development of education and health institutions at a local level. After contributing their ideas and financial resources to the construction / rehabilitation of a school or a health post for example, community groups can play a crucial role in the sustainability of the institution by organizing and maintaining a system for operation and maintenance. This is even more true for water supply, micro irrigated perimeters and markets, where proper operation and maintenance of the facilities that have been built is critical to ensure their sustainability. Description 9. Communal infrastructure activities are designed to complement sectoral development strategies. The sub-projects to be financed under this component will be the rehabilitation of essential community services (primary schools, community water supply and health posts), but will also include the rehabilitation of rural roads, bridges, markets, and micro irrigated perimeters. Labor intensive works will be preferred whenever proven as technically and economically efficient and insofar as they generate employment opportunities which the rural poor can take advantage of. 10. It is estimated that, under the proposed project, FID will finance a total of 350 communal infrastructure sub-projects: 130 schools, 30 community water supply, 40 health centers, 35 micro irrigated perimeters, 40 markets, 50 feeder roads, and 25 bridges and other transport infrastructure sub-projects. As mentioned above, those figures are indicative only since the FID is demand-driven. One of the mechanisms with which FID will track poverty incidence will be the use of the 1998 Anthropometric Survey as a Baseline since nutritional status is an excellent proxy for poverty. Implementation arrangements 11. The FID is a financing entity that provides grants to finance sub-projects identified and prepared by the beneficiaries themselves. The FID carries out promotion activities, appraises sub-projects, supervises their implementation, and monitors and evaluates their impact on the target groups. The role of FID in implementation varies depending on the amount of the grant for, and therefore on the size of, the sub- project. 12. For small infrastructure sub-projects (requiring a grant of less than US$7,500 equivalent), local grass-root groups and communities are responsible for implementation and, if required, enter into contracts with consultants, artisans and skilled labor, small contractors and suppliers. Their rights and  Annex 2 Page 4 of 5 responsibilities (including counterpart contribution, procurement procedures and disbursement schedules) are defined in their contract with FID. For those types of sub-projects, FID has a financing and monitoring role. 13. Al other infrastiucture sub-projects are implemented by FID under delegated contract management. As 'Maitre d'Ouvrage Delegu", or the entity authorized to act on behalf of the owners/beneficiaries of the infrastructure, FID enters into contracts with: (i) consultants for the design and supervision of the works; and (ii) small local contractors for the execution of the works. The Manual of Procedures of FID specifies clearly the procedures that FID must follow for procurement and disbursement. It also includes a number of provisions to ensure that beneficiaries are fully involved in the design of the sub-project, participate in its implementation (their contribution in cash or in kind must be made before construction actually starts, with the exception of their contribution in the form of labor), and are informed of the requirements for maintenance and receive some initial assistance to help them to carry out their responsibility for maintenance. 14. In line with the Government policy of decentralization, FID intends to progressively transfer more responsibility for the implementation of sub-projects to rural communities and communes. Under the ongoing project, FID is about to start 12 to 20 pilot sub-projects of different types and sizes to test various alternative arrangements for the management and implementation of infrastructure by the beneficiaries themselves. Those experiments will be continued and expanded under the proposed project. According to the preliminary design for those experiments, transport infrastructure sub-projects (essentialy bridges at this stage) will be managed and implemented by communes, and other infrastructure sub-projects by associations/communities. The contribution of beneficiaries to the construction costs will be at least 20 percent. FID will enter into a financing agreement with the association/community and/or commune spelling out the rights and obligations of each party. As required, the association/community or commune will recruit a consultant for design and supervision. FID will approve the design. Construction will be either by the beneficiaries themselves with the assistance of artisans and skilled labor, or by contracts with small contractors, or by a combination of both. Very simplified bidding documents wil be used for contracting. FID financing will be transferred to the bank account of the association or commune, with the first payment after beneficiaries have provided their own contribution and subsequent payments linked to the progress of construction. FID will supervise those pilot sub-projects with its own personnel, paying close attention to monitoring and evaluation. Based on the lessons from those experiments, other arrangements may also be tested. Project Component 2: Capacity building - Total estimated cost: US$ 0.6 million (to be financed by the IDA credit). Context 15. Under the ongoing project, there has been an emergence of local contractors and consultants which is attributable to FID's increased work program and, therefore, to the increasing number of contracts for consultancies and civil works. The works to be financed under the proposed project will create even more business opportunities for contractors and consultants. Both contractors and consultants have weaknesses which could have a negative impact on the quality of the works financed by FID, and jeopardize their subsequent maintenance. Also, local associations of beneficiaries and rural communities/communes need to be strengthened to be able to play a more active role in the identification, design, implementation and maintenance of the sub-projects. In particular, some rural communities / local associations and communes will need to be trained to be able to manage and implement satisfactorily the pilot  Annex 2 Page 5 of 5 infrastructure sub-projects included in the proposed project, and participate in a meaningful manner in the monitoring and evaluation of those experiments. Description 16. The objective of this component is to promote an increased participation of beneficiaries and rural communities/communes in the development of communal infrastructure and to ensure better quality and maintenance of the infrastructure sub-projects. The capacity building sub-projects will assist FID's partners (communes, rural communities, local associations, NGOs, small contractors and consultants) to improve their effectiveness in identifying, preparing and implementing sub-projects. This will be done through training programs tailored to the needs of the participants. 17. For local contractors and consultants, the training will be similar to the one done under the ongoing project. It will be both theoretical and practical, and mostly technical in nature, dealing with the use of labor-intensive techniques, the preparation of cost estimates, the preparation of bidding documents, bid evaluation, and the management and supervision of works. It will also cover the management of the firms themselves, including accounting and financial management. 18. For rural communities/local associations and communes, the training will cover project preparation, evaluation, selection of consultants, procurement (including the purchase of local materials and the recruitment of small local contractors and artisans), management of works and contract administration, simplified accounting, and the monitoring of some basic implementation indicators. 19. Itis estimated that, underthe proposed project, FID will finance atotal of 30 capacity building sub- projects for rural communities, local associations, communes, local contractors, consultants, artisans and NGOs. Under the 30 capacity building sub-projects to be implemented, about 300 rural communities and communes, 150 micro-enterprises, 75 consultants, 50 artisans and 20 NGOs will be trained. As mentioned above, those figures are indicative only since the FID is demand-driven. The IDA credit will finance up to 100 percent of the tuition costs for the training and, where necessary, a percentage (to be determined) of the transportation and subsistence costs of participants. Implementation arrangements 20. For this component, the FID will be essentially a financing organization, since it does not have (and should not attempt to acquire) the capability to conduct training. FID will be involved in the selection of the participants and the preparation of the training curriculum in order to ensure that it is fully responsive to the needs of the participants as identified by FID in its operations and that it will make those participants better partners of FID. The training itself, however, will be contracted to consultants or training organizations, including those financed by the Bank and other donors, in Madagascar or in neighboring countries. Note: A Project Implementation Manual, which includes the Manual of Procedures of FID, can be found in the project file. These documents will be used as a basis for field supervision during the project implementation phase.  Annex 3 Page 1 of I Third Social Fund Project Estimated Project Costs Rehabilitation of Community Infrastructure 13.0 0.6 13.1 Capacity Building 0.5 1.0 FID Management & Operations 1.7 0.3 2.0 Total 15.2 0.9 16.1 Total Baseline Cost Physical Contingencies 1.5 0.1 1.6 Price Contingencies Total Project Cost 16.7 1.0 17.7  Annex 4 Page I of 10 Third Social Fund Project Economic Justification I. Introduction 1. The FID is an instrument to address infrastructure needs of poor segments of the population. The FID relies on community participation (at least 5 percent contribution to costs) to assign ownership of investments to beneficiaries and create incentives at the ground level for rapid project implementation and subsequent maintenance. Sub-projects support government policies in those sectors. 2. The social fund FID will assist poor communities to finance a wide range of investments: school construction! rehabilitation, health posts, markets, rural roads, potable water supply, and micro irrigated perimeters. Since the FID is demand-driven, it is difficult to determinea priort the composition of the FID's portfolio. Based on experience with FID I and FID II, communities have highest demand for construction/rehabilitation of schools (35 percent of sub-projects), followed by rural roads (15 percent of sub-projects), health centers (13 percent of sub-projects), micro irrigated perimeters - MPI (11 percent of sub-projects), markets (9 percent of sub-projects) and water supply (8 percent of sub-projects). 3. After a presentation of FID's achievements to date and the performance targets for the proposed project, the analysis is organized in the following sections: (a) the section onpoverty characteristics, targeting and monitoring identifies those with greatest need to ensure funds are appropriately directed; (b) the section on project benefits presents the benefits that can be expected from the project; and (c) the section on sustainability and fiscal impact discusses the relevant factors at the community and central government levels. II. FID Achievements to Date and Key Performance Indicators for the Proposed Project 4. FID has achieved significant results. Table 1 shows the number of sub-projects completed or under execution through June 1998, and Table 2 shows the key performance indicators for FID II. Regarding the proposed project, Tables 3 and 4 below shows a range of indicators which will be used to guide FID during project implementation. These have been developed based on the experience of the on-going project. However, it should be mentioned once again that FID is demand driven and therefore these figures are subject to change as the political, social, and economic context changes and communities request different types of support.  Annex 4 Page 2 of 10 Table 1 Number of sub-projects financed under FID I and II (Up to June 1998) Infrastructure Schools 124 134 130 Health centers 45 54 41 Water supply 29 34 27 Other 1 2 3 Sub-total social infrastructure 199 224 201 Micro irrigated perimeters (MPI) 54 22 48 Markets 8 47 41 Other 2 1 4 Sub-total productive infrastructure 64 70 93 Rural roads 56 48 58 Bridges and other 23 25 31 Sub-total transport infrastructure 79 73 89 Environmental protection 6 - 3 Total Infrastructure 348 367 386 Income Generating Activities 36 63 232 Private Sector Support (training) I 10 12 GRAND TOTAL 394 442 618 Table 2 Key Performance Indicators for FID II Person-days of temporary employment 1,250,000 1,135,000 90.8 Number of permanent jobs created 700 345 49,4 Number of contracts with artisans & skilled 50 38 76.0 workers Number of contracts with Micro, Small and 300 359 119.7 Medium Enterprises Number of contracts with consulting firms 100 179 179.0 Number of contracts with NGOs 100 151 151.0 Number of schools rehabilitated 30 94 313.3 Number of health centers rehabilitated 20 36 180.0 Kilometers of rural roads 250 281 112.4 Hectares irrigated 7,500 2,397 32.0 Hectares protected 4,000 20 0.5  Annex 4 Page 3 of 10 Table 3 Number of sub-projects to be financed under the proposed project Communal Infrastructure sub- projects Schools 130 52,000 Health centers 40 40,000 Water supply 30 6,000 Other _ Sub-total social infrastructure 200 98,000 Micro irrigated perimeters 35 8,750 Markets 40 80,000 Other _ Sub-total productive infrastructure 75 88,750 Rural roads 50 50,000 Bridges and other 25 25,000 Sub-total transport infrastucture 75 75,000 Environmental protection - - Total Infrastructure 350 261,750 Capacity building sub-projects 30 600* GRAND TOTAL 380 ; Rural communities/communes, local contractors, consultants, NGOs, and artisans. Table 4 Key Performance Indicators for the proposed project Person-days of temporary employment 1,600,000 Number of permanent jobs created 188 Number of contracts with artisans & skilled 50 workers Number of contracts with Micro, Small and 350 Medium Enterprises Number of contracts with consulting firms 175 Number of contracts with NGOs 20 Kilometers of rural roads 300 Hectares irrigated 2,500 Hectares protected  Annex 4 Page 4 of 10 Ill. Characteristics of Poverty, Targeting of Activities and Performance Monitoring Poverty 5. The key finding of the poverty assessment of June 1996 is that the majority of Madagascar's 13.5 million people live in a state of significant deprivation. The country is one among few experiencing declines in education, nutrition and life expectancy, with health and nutrition indicators at or below highly unsatisfactory averages for Sub-Saharan Africa (SSA). Lack of access to safe water, poor sanitation and low nutritional standards are the leading causes of illness and death. Few seek health care; public facilities are ill-equipped, and private health care is too costly. Schools are poorly managed, with under- paid, ill-trained teachers. Public spending on health and education has been falling in real terms and does not benefit the poorest layers of the population. Also eroding living standards are problems of security in rural areas and isolation, combined with the lack of physical access to social services, financial services and agricultural markets and the lack of facilities for the development of productive activities. Finally, slash-and-bum agriculture continues to cause extensive deforestation and loss of topsoil, threatening biological diversity and the agrarian economy 6. Poverty has become entrenched in Madagascar's social fabric. Seven out of 10 Malagasy are poor (e.g., they cannot purchase a minimum 2100-calorie food basket and essential non-food items); six out of 10 are extremely poor (e.g., they cannot purchase even the food basket). 1997 Government data indicate that the situation has improved with only 3 out of 10 persons considered to be extremely poor. Over 85 percent of the poor live in rural areas, where poverty is also deeper; the remaining 15 percent poor are urban, accounting for nearly half that population. Regionally, poverty incidence is highest in Toliara. Farmers are the poorest: 80 percent of poor households are headed by farmers, herders and fishermen. In contrast, wage-earners fare best (only 14 percent of poor households). About 18 percent of poor households are headed by women, engaged mostly in farming. Poor households have around 5 members, compared to 4 for non-poor households, and are more likely to: be illiterate; not send children to school; have malnourished children; use wood as main source of fuel; and rely on lakes/ponds/rivers for drinking water. It is clear that rural poverty, which has as its main cause the poor productivity and competitiveness of the agricultural sector, has been exacerbated by the lack of productive and social infrastructure and other services, and an overcentralized decision-making process for the allocation of resources. 7. The Government's central objective is to steadily reduce the incidence of poverty by half from about 70 percent in 1996 to 35 percent in 2015. This target can only be realized by achieving very ambitious economic growth rates, while reversing environmental degradation and improving the assets of the poor. Pro-poor growth will call forinter alia: improved sector policies promoting rural development; more efficient public investment and restructuring of public expenditures; and encouragement of local initiatives. Targeting 8. The overarching objective of the project is to alleviate poverty. The vast majority of beneficiaries are expected to be from the population below the poverty line. Since the Madagascar poverty assessment has recommended a rural based strategy for poverty reduction, the proposed project will target primarily rural areas where more than 85 percent of the poor live. Sub-project proposals must demonstrate clearly that they will lead to a reduction in poverty or vulnerability of the community. The impact of the project on the incomes and living standards of the poor will be assessed through beneficiary surveys and studies of socio-economic impact, as is being done for the ongoing project.  Annex 4 Page 5 of 10 Monitoring 9. Monitoring of implementation progress has been robust under FID I and II. An elaborate system has been developed and is being implemented which keeps track of the type and location of sub-projects, investment cost, community contribution, and number of beneficiaries. The existing capacity would serve the proposed project quite well. A beneficiary assessment of activities under FID I has been undertaken; it has been used for the final design of the FID II and will also serve as a checkpoint for the implementation of the proposed project. Monitoring of targeting effectiveness would be aided by the fact that FID interventions are mainly in rural areas, where the vast majority of the poor live. 10. Monitoring of outcomes is an area where the proposed project could make progress over FID I and II. The costs associated with monitoring are not that critical, given the size of FID operations. It remains important to obtain some objective measures, in addition to the favorable comments circulating by word of mouth in Madagascar, about the impact that individual sub-projects have on the communities that requested the investment. The FID is undertaking a study of the socio-economic impact of its sub- projects, adopting a sampling approach and monitoring a small pool of each type of sub-projects. For education sub-projects, enrollment at the village level, examination performance, or wages in subsequent years, are suitable outcomes to measure. For health posts, number of visits, number of patients, and medications purchased or distributed are reasonable indicators of impact. Finally, for water sub-projects, agricultural output for irrigation perimeters, or reduced incidence of illness for potable water supply will be considered. Since the objective is always to ascertain the marginal impact or benefit individual sub- projects have on the community, some data gathering is required before a sub-project is undertaken. IV. Project Benefits 11. Given the multi-sectoral nature of the sub-projects, a range of benefits is expected, many of which are not quantifiable. In any event, since the project is demand driven it is difficult to determini priori the composition of FID's portfolio which will depend on communities' priorities, and an overall NPV or ERR has not been estimated. Reduced poverty of poorer communities is expected to be achieved through a mixture of economic and social benefits. 12. The proposed project aims at alleviating poverty while building local community capacity for development. Specifically, it is expected that the project will: (a) develop communities infrastructure such as schools, health centers, markets, rural roads and irrigation systems; (b) finance 350 contracts with micro-, small- and medium-enterprises and 175 contracts with consulting firms, as well as another 20 contracts with NGOs and 50 with artisans and skilled workers, and create about 3 millions person-days of temporary employment on infrastructure sub-projects, and about 300 permanent jobs; and (c) train about 300 rural communities/communes, 150 contractors, 75 consultants, 20 NGOs, and 50 artisans. Economic benefits 13. The economic benefits include improved standard of living in beneficiary areas dueto returns from sub-project investments in social and economic infrastructure, the creation of income-earning opportunities through employment, and increased productivity and sustainability of created communal assets with improved in-service delivery. There are also indirect increases in incomes in beneficiary areas through multiplier effects. 14. Simple screening procedures, using data on average cost per beneficiary for example, are applied to establish the economic justification of sub-projects. Technically non-viable sub-projects are not eligible for financing. Also, community contributions as a share of sub-project cost are taken as an indicator of  Annex 4 Page 60flO community ownership and commitment to maintain the productive life of the assets. Prudent management dictates that sub-projects with a low chance of sustainability not be undertaken and communities with poor track records of sustainability be denied financing for additional sub-projects. This is indeed reflected in the manual of procedures of FID. FID sub-projects are subject to economic analysis wherever appropriate. Economic internal rates of return are usually high because most of the sub-projects consist of the rehabilitation/reconstruction of existing infrastructure and include substantial sunk costs. In the case of rural roads, the potential traffic volume is used to justify the investment and to assess whether the design is appropriate. In the case of micro irrigated perimeters, the expected increase in agricultural production is taken into account. 15. Social infrastructure sub-projects are not required to meet a rate of return test. In any event, school and health post sub-projects consist mainly of the replacement or rehabilitation of existing structures; the underlying rate of return is expected to be quite high for these sub-projects even though not explicitly calculated. 16. A case can be made that for education and health sector sub-projects financed under a social fund which are in support of a well-prepared governmental master plan for each sector, economic analysis on a sub-project basis for the social fund may not be necessary. This assumes that an optimal strategy for each sector has been worked out and economic analysis has been done at the national level when the strategy was conceived and within it, all investments were shown to have positive economic NPVs. 17. Complete economic and financial benefits to education are often difficult to capture because a substantial portion of the benefits to education are extemalities - they are benefits to society above those that accrue to individuals from their own education. Empirical studies for Madagascar are currently unavailable, but based on discussions with education and community officials or experiences in other countries, three types of benefits have been identified for this category of sub-project. First, better facilities contribute to a drop in repeater rates for the school. The second source of benefits is the expected increase in farmer income as a result of receiving better schooling. Finally, in some cases students are expected to save time by attending school in their own community, rather than undertaking a long commute to a nearby village. Regarding costs, communities bear part of the construction cost, with the remainder born by Government through FID. The distributional impact of the investment (between the local economy, imports and taxes) also vary. Recurrent costs to the sub-project are born jointly by the community and the Government. The community is responsible for upkeep of the school building and the Government for providing a teacher for the school. Generally, studies done in other countries show that the benefits accruing to society as a whole are rather small, relative to the cost of the investment. Usually, the largest benefit comes from incremental income gained by individuals who were not previously participating in the school system. Several reasons can be given for the relatively low return to this investment. First, failure to consider broader benefits and externalities of education (such as better health for girls, etc.), due to lack of data; second, absence of data on earnings differentials between literate and illiterate heads of households. Finally, the pay-off from education may be low due to the lack of economic opportunities. If jobs valuing education are few, then the return to education is low. In this case, the high demand for education can only be seen as having some sort of utility rather than market value, if parents believe that education is necessary, but not sufficient to guarantee, higher future income. 18. Similar analyses on health infrastructure suggest that their financing and operation may be tenuous because several sources of financing must come together for the infrastructure to be operational and sustainable. A high level of coordination may be required on the part of bilateral and multilateral donors, government and NGOs for a sub-project to meet its objectives. In health, the proposed project is expected to finance the construction of primary level infrastructure (rural health posts), which tend to carry out  Annex 4 Page 7 of 10 interventions in the higher end of the cost effectiveness spectrum (EPI, re-hydration therapy, malaria treatment, TBA ante-natal and delivery care, etc.). Such interventions, which are nornally carried out at the primary level of care, tend to favor primarily women and children and are expected to yield substantial gains in life-years saved. 19. As all the sub-projects are small in size (a maximum FID financing of US$100,000 for transport infrastructure sub-projects and US$50,000 for other sub-projects), shadow prices are not used in the analysis. However, in view of the fact that these projects are highly labor intensive and in view of the high unemployment/ underemployment rate in the country, it is likely that the rates of return would be higher if shadow pricing were to be used. For those sub-projects requiring an economic internal rate of return justification, a very simple sensitivity analysis is used. Social benefits 20. The expected social benefits include enhanced community organizational capacity to undertake community-prioritized projects, as well as improved collaboration between communal leaders, associations' representatives and local sector officials in carrying out a process of cross-sectoral planning coordinated at the local level. FID is helping communities to take charge of their destiny and empowering them. By doing so, FID is helping to accelerate the process of effective decentralization, which is a major objective of the Govemnment. 21. On the institutional side, the proposed project provides additional financing for an ongoing social fund which has been successful. FID has a very strong management which is to a large extent politically insulated. Some problems, particularly regarding the active role that beneficiary associations should play and the capacity of FID partners to perform satisfactorily, have been identified and measures have been taken to correct them. These include training for contractors and consultants, as well as associations of beneficiaries to ensure adequate maintenance of sub-projects. IV. Sustainability and Fiscal Impact Sustainability 22. Regarding the institutional sustainability, a social fund itself is generally not sustainable in the long run without external financing. The proposed project is an additional phase of FID requested by Government, proving that an autonomous institution such as the FID remains the best option to implement small-scale sub-projects at community level. In the future, financial assistance is also likely to be forthcoming from other donors. It is worth noting that FID does support the sustainable development of communities, in particular by helping build local capacity to prepare and execute development sub- projects, for example, at the community and micro-enterprises level. 23. For the Government, FID is considered an integral part of Madagascar's rural development strategy and is an essential instrument to strengthen the capacity of the communes and rural communities to maintain their assets and, therefore, ensure the sustainability of the sub-projects. The sustainability of FID's sub-projects will be achieved by ensuring maximum community involvement in the identification, preparation, appraisal, selection and implementation of a sub-project. In addition to contributing a share of sub-project costs, as a condition of sub-project approval communities will be requested to submit a plan for the maintenance of the completed asset. This plan will show responsibilities and financial contributions to be made by community members or users. Sub-project proposals are also reviewed to ensure that sectoral ministries will provide for recurrent expenditures generated by the sub-project and for  Annex 4 Page 8 of 10 which the sector is responsible. FID will ensure that the target beneficiaries (and not political figures) are actually making the in-cash or in-kind contributions. Experience shows that the more substantial the beneficiary contributions, the better the chance that the sub-project will be properly and efficiently utilized by the community and the better the chance that adequate funds will be allocated towards maintenance. Even under the ongoing Social Fund II Project, FID is exploring ways to increase those beneficiary contributions. Recently, FID has made substantial efforts to create awareness among the various communities who have benefited from sub-projects as to the pros of maintenance and is allocating funds to train community members in proper maintenance techniques and organizational matters. A maintenance clause inserted in all civil works contracts stipulates that the contractor is also responsible for training a few persons designated by the community in maintenance activities. Regarding rural roads, FID has been working with decentralized public officials such as mayors to ensure that the necessary funds are budgeted and made available for maintenance activities. 24. Although empirical studies of community financing in Madagascar do not exist, FID experience and field visits have revealed the ability of communities to gather financial resources for priority projects. Overall, the financing requirements that sub-projects place on communities, for both capital and recurrent expenditures, are considered reasonable and moderate. Fiscal impact 25. The project is part of an overall package meant to help the poorer sections of the country's population survive the current difficult period until sustained economic growth reduces poverty. Nevertheless, as mentioned earlier, beneficiary contributions are required for all sub-projects to ensure ownership and sustainability. The government contribution is expected to be about US$ 2.0 million, almost all of it to cover taxes on civil works contracts and consultancies, since those will not be exempt from domestic taxes; so that the net Government contribution will be negligible. On the other hand, in the case of infrastructures built by contractors, government gains revenues from taxes paid by those contractors on all the items that they import. Tax revenues from anticipated higher income streams to sub-project beneficiaries are not considered as part of the fiscal impact, because those benefits are expected to accrue to rural communities that, based on reality in Madagascar today, are outside the Government's tax base. Financialpredicament 26. In recent years, the Government has had difficulty meeting its obligations for education and health; for those two sectors, the budgetary allocations have decreased in real terms, and have been insufficient to meet their requirements. This situation almost "reverses" the question of fiscal impact of the social fund; rather than draw government resources, the social fimd often complements government expenditure, and sometimes serves as a substitute when communities assume responsibility for some basic social services. Capacity and financial constraints of the Govermuent have led to a demise of basic social services to the poor. Herein lies, in good part, the justification for designing Bank assistance in the form of a social fund. 27. Comparisons with other countries in the region and of similar size and income are revealing (Table 5). Although comparable budgetary expenditures for countries of similar size are not readily available, some outcome comparisons can be made. They indicate that Madagascar lags behind or is barely better off than the lower income groups, and is much worse off than the next highest income group.  Annex 4 Page 9 of 10 Table 5 Social Indicators Madagascar and Comparators Madagascar 13,101 230 93 114 51 2.9 Cape Verde 381 910 115 40 68 2.3 Comoros 485 510 86 89 56 2.7 Djibouti 566 780 39 115 49 4.7 Guyana 825 530 112 48 65 0.6 Maldives 246 900 n/a 55 62 3.1 Mauritius 1,104 3,180 106 17 69 1.0 Low Income* - 390 101 73 62 1.8 SSA* - 500 66 99 52 2.9 -/ Country averages  Annex 4 Page 10 of 10 V. Conclusion 28. Madagascar is among the poorest countries in the world. Despite repeated efforts, the Government continues to face enormous challenges in providing basic social services to the population. In this environment, FID can be an effective instrument for delivery of social infrastructure to the poor when they are ready to assume part of the responsibility for investments. All sub-projects support sectoral plans, and in doing so alleviate some fiscal pressure on the central government. 29. Given the multi-sectoral nature of the sub-projects, a range of benefits is expected, many of which are not quantifiable. In any event, since the project is demand driven it is difficult to determinea priori the composition of FID's portfolio which will depend on communities' priorities, and an overall NPV or ERR has not been estimated. 30. The economic benefits include improved standard of living in beneficiary areas due to returns from sub-project investments in social and economic infrastructure, the creation of income-earning opportunities through employment, and increased productivity and sustainability of created communal assets. There are also indirect increases in incomes in beneficiary areas through multiplier effects. The expected social benefits include enhanced community organizational capacity to undertake community- prioritized projects, as well as improved collaboration between communal leaders, associations' representatives and local sector officials in carrying out a process of cross-sectoral planning coordinated at the local level. 31. Simple screening procedures will be appliedto ensure targeting of benefits to poorer and more vulnerable groups and to establish the economic justification of sub-projects. Despite the merits of the social fund as an instrument of choice, there remains a need for selectivity and stringent evaluation on the part of FID for individual sub-projects so that in every case the social fund is the best instrument to meet communities' objectives and sustainability of the sub-project can be assured.  Annex 5 Page 1 of 1 Third Social Fund Project Financial Summary Years Ending December 31 in US$ million Project Cost 5.4 5.4 5.4 16.2 Investment Costs Recurrent Costs 0.5 0.5 0.5 1.5 Total 5.9 5.9 5.9 17.7 Financing Sources (% of total project cost) IBRD/IDA 85% 85% 85% Co-financiers Govmmnt9% 9% 9% Goavernment Central Local User Fees/Beneficiaries 6% 6% 6% Others Others Others Others Total 100% 100% 100%  Annex 6 Page 1 of 7 THIRD SOCIAL FUND PROJECT Procurement Disbursement and Financial Management Arrangements 1. Procurement of goods and works required for the project and to be financed out of the proceeds of the credit shall be govemed by the "Guidelines for Procurement under IBRD Loans and IDA Credit" published by the Bank in January 1995 and revised in January and August 1996, September 1997 and January 1999. Procurement of consultants' services shall be governed by the "Guidelines for the Selection and Employment of Consultants by World Bank Borrowers" published by the Bank in January 1997 and revised in September 1997. The following paragraphs describe the procurement arrangements under the project. Capacity Assessment of FID for Procurement 2. FID which will be the executing agency for this project has been operating since 1993 under the Food Security and Nutrition Project (Cr. 2474-MAG) and the on-going Second Social Fund Project (Cr. 2778-MAG). It is fully experienced in procurement, in particular procedures applicable to small contracts for communal infrastructure. During those projects, technical audits took place in 1996, 1997 and 1998 and selected sub-projects were thoroughly reviewed including the procurement procedures--no major discrepancies were discovered in particular with the provisions of FID's Manual of Procedures. Most of the procurement under the project relates to the sub-projects financed by FID. The Development Credit Agreement (DCA) for the project states that all procurement for those sub-projects shall be done in accordance with the Manual of Procedures of FID. The Manual of Procedures of FID is acceptable to IDA, and any amendment to such manual will have to be also acceptable to IDA. 1. Procurement Methods for Goods and Works (see Table A) 3. Civil works and goods contracts for infrastructure sub-projects involving the construction, repair, and rehabilitation of community infrastructure including primary schools, health posts, community water supply, feeder roads, rural markets, etc. and the purchase of school desks and similar type furniture for other types of sub-projects costing more than US$7,500 but less than US$60,000 (US$120,000 for rural road sub-projects) per contract up to an aggregate amount of $5.0 million would be procured through NCB procedures described in the Manual of Procedures of FID. Civil works and goods contracts for sub-projects costing less than US$7,500 per contract up to an aggregate amount of US$ 2.5 million would be procured through obtaining at least three bids from small local contractors (civil works) and through local shopping procedures (goods), as described in the Manual of Procedures. Smaller works and goods contracts involving full community participation for an aggregate amount of US$ 5.0 million would be carried out in accordance with procedures acceptable to the Bank Such procedures are described in details in the Manual of Procedures. 4. Procurement methods for goods for FID. Contracts for goods such as vehicles, office equipment and furniture for FID national and regional offices costing more than $100,000 per contract would be purchased through ICB in accordance with the Bank's Guidelines. Contracts for goods available locally for an aggregate amount of US$ 100,000 which cost more than $30,000 but less than or equal to $100,000 per contract will be procured through NCB procedures acceptable to IDA. Small quantities of office supplies and consumable materials which are normally available off the shelf at economical prices and cannot be grouped in packages of $30,000 would be procured through national shopping based on price quotations obtained from three reliable suppliers provided that the aggregate amount does not exceed $100,000.  Annex 6 Page 2 of 7 Prior review thresholds (Table B) 5. IDA financed contracts above the threshold of $75,000 for works and goods will be subject to IDA's prior review. All other goods and works contracts will be subject to post review. However, a technical audit will be performed during each year of project implementation. About 10 percent of the on-going or completed sub-projects (selection will be on a random basis) will be subject to an in-depth review as part of the annual technical audit. This review will look into detail into compliance with the procurement procedures utilized to ensure conformity with FID's Manual of Procedures. Procurement of Consultancy Services, Technical Assistance and Training 6. Consultants' Services for infrastructure and capacity building sub-projects. These types of services, for an aggregate amount of US $600,000 would be for: (a) design and supervision of FID's infrastructure sub-projects (US$500,000 million); and (b) technical assistance and training support for the execution of FID's capacity-building sub-projects (US$100,000 million). Consultant selection for sub- projects will be addressed through competition among short listed firms in which the selection will be based on Quality and Cost Based Selection (QCBS). For contracts with firms estimated to cost less than US$50,000 per contract, the short list may consist exclusively of national consultants. For the financial audits, which will involve firms, the least cost selection will be used. Some smaU contracts for technical audits and overseas training, not exceeding US$ 50,000 per contract, will be procured on a single-source basis. Consultants' services consisting of technical assistance and training support to FID which would involve individuals, including recruitment of local FID staff will be selected on criteria based on qualifications, experience, and competence for their assignments. Prior Review for Consultancy Services, Technical Assistance and Training 7. Prior IDA review for consultants' services will be required for: (a) all contracts for individuals costing more than $30,000 per contract; (b) all contracts (for any amount) for FID's high level personnel including the national director, the regional directors, directors and division chiefs; (c) all contracts (any amount) for the financial and technical audits; and (d) all contracts with firms costing more than $50,000 per contract. The technical evaluation report for the recruitment of consulting firms will be subject to prior review for contracts exceeding US$ 100,000 each. All other contracts wiU be subject to post review. Once again, the technical audit will look at selected random contracts for the recruitmnent of consultants. Financial Management 8. A computerized accounting and financial management system in conformity with internationally accepted accounting has been established. It has been assessed by a Bank/IDA Financial Management Specialist and practical measures have been recommended (annex 7) to improve its performance and ensure its adequacy for an orderly financial implementation of the project. Given the short duration of the project (2 years), disbursements will be made in accordance with the SOE-based method. This decision notwithstanding, the project will be required to submit, in addition to its basic financial statements, five quarterly reports, namely; a Summary of Sources and Uses of Funds, a Contract Expenditure Report - Goods & Works, a Contract Expenditure Report - Consultants, a Procurement Management Report - Goods & Works, and a Procurement Management Report - Consultants. All procurement documents, contracts, and invoices would be maintained separately by FID and be made accessible to supervision missions and auditors.  Annex 6 Page 3 of 7 Disbursement Method of Disbursement 9. It is envisaged that based upon the outcome of the financial management assessment, disbursements will be transaction-based (traditional mode) as they are currently done under the on-going project (Cr. 2778-MAG) in accordance with procedures outlined in the Bank's Disbursement Handbook. Given the short disbursement period of this credit (2 years) it is not foreseen that LACI type disbursements will be introduced. Allocation of loan proceeds (Table C) 10. Table c indicates the disbursement schedule, the amounts (in US$) for each expenditure category and the disbursement percentage applicable to each. The disbursement percentages have been calculated on a tax-inclusive basis such that, when applied to invoices denominated in local currency, the percentage not financed by the Bank is sufficient to cover Government counterpart contributions and eliminate any Bank financing of local taxes and duties. I1. The credit closing date is forecast for December 31, 2001, about six months after the end of the disbursement period. Category 5 (operating expenses for FID) includes expenditures for local contractual support staff salaries, employment related benefitis, travel expenditures and other allowances for travel related to the Project, rental and maintenance of equipment, vehicle maintenance and repair, vehicle insurance, office rental and maintenance, materials and supplies, utilities and communications expenses, banking charges, and insurance for FID goods and equipment related to the project. No high level salaries will be financed under operating costs. Use of statements of expenses (SOEs) 12. A simplified format will continue to be used for SOEs in which expenditures are summarized by category. A sample of the simplified SOE form is contained in the project implementation manual. The documentation for withdrawals under SOEs would be retained at FID (National and Regional Offices) for review by IDA staff during supervision missions and for annual audits. SOEs will be used for payments of contracts of less than US$ 75,000 for goods and works and consultant contracts of less than US$ 50,000 and US$ 30,000 for firms and individuals respectively. SOEs will likewise be utilized for all training and operating costs. Special account 13. The special account will be opened in a commercial bank acceptable to IDA and receive an advance of US$3,000,000 covering IDA's share of 4 months of estimated expenditures. Upon effectiveness, an amount of $1,500,000 will be advanced to the special account with the balance of the initial deposit ($1,500,000 being advanced once SDR 3,000,000 has been disbursed, or earlier if IDA so agrees. Flow of Funds 14. The "special account 90-day advance procedure" especially designed for decentralized agencies will be used to facilitate the operation of FID's regional offices and to ensure prompt payment of  Annex 6 Page 4 of 7 contractors and suppliers. This procedure is described in "Bank-Financed Projects with Community Participation: A Manual for Designing Procurement and Disbursement Mechanisms" and is currently utilized under the on-going project. Under this procedure, FID will advance funds sums covering no more that 90 days estimated expenditures (in local currency) to regional FID accounts based upon submission of satisfactory, budgeted work plans. Regional FID offices will submit quarterly expenditure reports indicating sources and uses of funds upon which basis, justifying the use of funds, and accompanied by reconciled bank statements. FID will then aggregate this data, prepare reconciliation statement for the main special account, prepare replenishment applications accompanied by the bank statement for the main special account. Audit 15. The independent auditor of the on-going Social Fund II Project will be retained, on a modified and improved terms of reference, as the auditor of the proposed project for the fiscal years ending December 31, 1999 and 2000. The project accounts will be audited in accordance with intemational audit standards acceptable to the Bank/IDA. The auditors will: (a) submit their reports and the related project accounts to the Bank/IDA within 6 months after the end of Government fiscal year; (b) in addition to their standard short form report with the opinion, the auditors will be required to: (i) carry out a comprehensive review of all the SOEs as well as the intemal control procedures governing their preparation for the relevant period under audit, and express a separate opinion thereon; and (ii) review the management and utilization of the 90-day special account and express a separate opinion thereon as well; and (c) finally, the auditor will carry out an in-depth review of the intemal control system of the project with a view to identify the major weaknesses and shortcomings and proposing practical recommendations for improvement. The results of this review would be documented in a Management Letter to be submitted along with the audit report. Revised terms of reference will be approved for the 1999 audit.  Annex 6 Page 5 of 7 Table A: Project Costs by Procurement Arrangements (in US$ million equivalent) G~~~~~~~~~~~ 1. Works and Goods for sub-pro.ject 3.5 10.5 14.4 (3.0) (9. 1) (12.5) 2. Consultancy services for infrastructure and capacity building sub-projects 0.8 0.8 (0.6) (0.6) 3. Consultant Services Technical assistance, training 0.4 0.4 and studies (0.4) (0.4) 4. Equipment, vehicles, fumiture, and materials 0.1 0.1 0.1 (0.3) (0.1) (0.1) (0.1)* (02)* 5. Miscellaneous Operating costs 1.8 1.8 (1.2) (1.2) Total 0.1 5.4 12.2 17.7 (0. 1) (5.1)* (9.8)* (15.0)* Note: N.B.F. = Not Bank-financed. Figures in parenthesis are the amounts to be financed by the Bank loan/IDA credit Figures may not add up due to rounding  Annex 6 Page 6 of 7 Table B: Thresholds for Procurement Methods and Prior Review . . 1. Works Sub-projects >$7,500 and<$60,000* NCB >$75,000 / $ 1.0 million <$7,500 Local Shopping None 2. Goods Sub-projects >$7,500 and <$60,000 NCB >$75,000 / $ 0 million <$7,500 Local Shopping None Equipment, vehicles, >$100,000 ICB >$100,000 / $ 0.1 million fumniture, & materials for FID >$30,000 and <$100,000 NCB >$75,000 / $ 0,05 million <$30,000 Local Shopping None 3. Services Consultants & Training Firms QCBS, LCS >$50,000 / $ 0.45 million Individuals Short-list >$30,000 / $ 0.1 million FID high level cadres Short-list All / $ 0.15 Technical audits Single source All / $ 0.05 million Financial audits LCS AU / $ 0.05 million Overseas training Single source All / $ 0.05 million 4. Miscellaneous Operating Costs N/A N/A N/A Total value of contracts subject to prior review: US$ 2.0 million. * $120,000 forrural road sub-projects. Note on selection method for consultants: QCBS = Quality- and Cost-Based Selection. LCS = Least-Cost Selection.  Annex 6 Page 7 of 7 Table C: Allocation of Loan Proceeds 1. Goods and works for sub-projects 12.1 86% 2. Consultancy services and training under sub- projects 0.5 100% foreign, 85% local expenses 3. Technical assistance, training and studies 0.4 100% foreign, 85% local expenses 4. Equipment, vehicles, furniture and materials 0.3 100% foreign, 90% local expenses 5. Operating costs 1.0 90% 6. Unallocated 0.7 Total 15.0  Annex 7 Page 1 of 1 Third Social Fund Project Project Processing Budget and Scheduling A. Project Budget (US$000) Planned Actual (At final PCD stage) Document write-up and processing $20,000 $20,000 B. Project Schedule Planned Actual (At final PCD stage) Negotiations January 1999 January 1999 Board Presentation March 1999 March 1999 Time taken to prepare the project (months) 2 months 2 months First Bank mission (identification) N/A N/A Appraisal mission departure N/A N/A Negotiations 1/11/1999 1/19/99 Planned Date of Effectiveness 6/30/1999 6/30/1999 Prepared by: Fonds d'Intervention pour le D6veloppement Preparation assistance: N/A Bank staff who worked on the project included: Eileen Murray, Task Team Leader; Gervais Rakatorama, Financial Analyst; Bertrand Ah-Sue, Senior Procurement Specialist; Moctar Thiam, Senior Transport Specialist; Harisoa Rafaell, Task Team Assistant; T. Mpoy Kamulayi, Country Lawyer. Sidi Boubacar, Country Lawyer; David Freese, Disbursement Officer; Hilda Emermwa- Creppy, Office Administrator; Manorama Gotur, Operations Analyst, Nathalie Lopez-Diouf, Language Team Assistant. Lead Adviser: Mr. Qaiser Khan, MNHSD Peer Reviewer: Mr. Hans Binswanger, Director, Rural Development and Environment.  Annex 8 Page 1 of 1 THIRD SOCIAL FUND PROJECT Documents in the Project File* A. Project Inplementation Plan Manuel de Procedures FID, dated December 1998 Mid-term review documentation of on-going project B. Bank Staff Assessments Staff Appraisal Report FID II Cr. 2778-MAG C. Other Draft documentation on maintenance issues for FID II Evaluation of FID's Financial Management System *Including electronic files.  Annex 9 Page 1 of I Status of Bank Group Operations in Madagascar Operations Portfolio ....,.. .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~,. . .. Number of Closed Projects: 57 Active Projects MG-PE-1559 1998 GOVT OF MADAGASCAR EDUCATION SECTOR DEV 0.00 65.00 0.00 66.74 6.60 0.00 MG-PE-1564 1998 GOVERSMENT RVRAL WATER SEC.PILO 0.00 17.30 0.00 17.31 8.11 0.00 MG-PE-1568 1998 GOVERNMENT NUTRITION II 0.00 27.60 0.00 27.55 1.31 0.00 MG-PE-56487 1998 REPUBLIC OF MADAGASCAR MINING PROJECT 0.00 5.00 0.00 5.32 .34 0.00 MG-PE-1537 1997 REPUBLIC OF MADAGASCAR ENVIRON. II 0.00 30.00 0.00 19.26 -2.46 0.00 MG-PE-1555 1997 GOVERNMENT PRIV SECT DEV & C.B. 0.00 23.80 0.00 19.68 19.59 0.00 MG-PE-40019 1997 GOVERNMENT CAPACITY BUILDING 0.00 13.83 0.00 5.84 1.90 0.00 MG-PE-48697 1997 GOVERNMENT O0 MADAGASCAR URBAN INFRASTRUCTURE 0.00 35.00 0.00 33.16 8.25 0.00 MG-PE-1533 1996 GOVT OF MADAGASCAR SCAR ENERGY SECTOR DEV PR 0.00 46.00 0.00 37.26 12.82 0.00 MG-PE-35669 1996 GOVT MADAGAS/FOND D'INTER SOCIAL FUND 2 0.00 40.00 0.00 8.00 -4.55 0.00 MG-PE-1522 1995 GOVT. OF MADAGASCAR IRRIGATION II 0.00 21.20 0.00 11.36 4.31 0.00 MG-PE-1563 1995 GOVERNMENT AG.EXTENSION PROJECT 0.00 25.20 0.00 9.40 3.93 0.00 MG-PE-1583 1994 MTP/MUT URBAN WORXS PILOT 0.00 18.30 0.00 0.00 -1.07 0.00 NG-PE-1550 1993 GOVT OF MADAGASCAR FINANCIAL INSTITUTIO 0.00 6.30 0.00 2.30 2.15 0.00 MG-PE-1552 1992 GOVERNMENT VOC. EDUCATION 0.00 22.80 0.00 1.81 .98 .94 MG-PE-1520 1991 GOVERNMENT NAT HEALTH SECTOR 0.00 31.00 0.00 6.18 4.43 0.00 MG-PE-1549 1991 GOVERNMENT LIVESTOCK 0.00 19.80 0.00 3.00 3.03 0.00 MG-PE-1512 1990 GOV.OF MADAGASCAR TANA PLAIW DZV 0.00 30.50 0.00 15.56 13.68 0.00 MG-PE-1540 1990 GOVT. FIN SECTOR/APEX 0.00 48.00 0.00 11.75 7.90 0.00 Total 0.00 526.63 0.00 301.48 91.25 .94 Active Projects Closed Projects Total Total Disbursed (IBRD and IDA): 226.94 1,138.76 1,365.70 of which has been repaid: 0.00 108.32 108.32 Total now held by IBRD and IDA: 526.63 1,010.69 1,537.32 Amount sold 0.00 6.86 6.86 Of which repaid : 0.00 6.86 6.86 Total Undisbursed : 301.48 39.30 340.78 a. Intended disbursements to date minus actual disbursements to date as projected at appraisal. Note: Disbursement data is updated at the end of the first week of the month and is currently as of 31-Jan-99. Page I  Annex 10 Madagascar at a glance Page lof 2 Sub- POVERTY and SOCIAL Saharan Low- Madagascar Africa income Development dhamond' 1997 Population, mid-year (mflilons) 14.1 614 2,048 Life expectancy GNP per capita (Atlas method, US$) 250 500 350 GNP (Atlas method, US$ billions) 3.5 309 722 Average annual growth, 1991497 Population I%) 2.8 2.7 2.1 Labor force (%) 2.8 2.6 2.3 GNP Gross per primary Most recent estmate (latest year available, 1991-97) capita enrollment Poverty (% of population below national poverty line) 75 Urban population (% of total population) 28 32 28 Life expectancy at birth (years) 58 52 59 Infant mortality (per 1,000 lIVe births) 86 90 78 Child malnutrition (% of children under 5) 32 .. 61 Access to safe water Access to safe vwater (% of population) 29 44 71 Illiteracy (% of populathon age 15+) 54 43 47 Gross primary enrollment (% ofschool-age populaton) 72 75 91 Madagasrcar Male 73 82 100 Low-income group Female 70 67 81 KEY ECONOMIC RATIOS and LONG-TERM TRENDS 1976 1986 1996 1997 Economic ratlos GDP (US$ bllons) 2.2 3.3 4.0 3.6 GrossdomesticinvestmenVGDP 8.1 9.0 11.1 12.1 Trade Exports of goods and services/GDP 15.2 12.1 20.5 22.4 Gross domestic savings/GDP 5.8 6.1 5.8 4.3 Gross national savings/GDP 5.8 6.1 4.9 6.8 Current account balance/GDP -3.1 -3.0 -6.2 -5.3 Domestic Inv Interest payments/GDP 0.2 1.5 1.5 1.4 Savings estment Total debt/GDP 9.2 92.1 112.3 121.3 Saig Total debt service/exports 3.8 47.6 27.8 24.4 Present value of debt/GDP .. .. 79.6 Present value of debt/exports .. .. 347.8 Indebtedness 1976-86 1987-97 1996 1997 1998-02 (average annual growth) GDP -0.1 0.9 2.1 3.7 5.5 Madagascar GNP per capita -3.4 -1.4 0.2 1.6 2.4 Low-income group Exports of goods and services -4.4 5.5 3.9 -1.5 6.7 _ STRUCTURE of the ECONOMY 1976 1986 1996 1997 Growth ratesofoutputandlnves,nent(%) (% of GDP) 0- Agrcuiture 33.3 36.8 31.7 31.7 100 Industry 16.1 12.9 13.5 13.4 so Manufacturing .. 10.6 11.5 11.1 o__ Services 50.6 50.3 54.8 54.9 - 92 93 94 9 s 97 Private consumption 83.0 85.0 87.8 88.4 -1os General govemment consumption 11.2 8.8 6.3 7.2 GDI - -GDP Imports of goods and services 17.5 15.0 25.8 30.2 197646 1987-97 1996 1997 Growth rates of exports and imports (%) (average annual growth) 20 Agricusture 1.5 1.9 2.5 2.4 2- Industry -2.3 0.7 2.0 3.5 10 Manufacuring . 0.1 1.2 0.9 o5 Services -0.6 1.1 2.0 4.5 93 94 95 96 97 Private consumption -0.5 1.1 1.4 4.7 -2D General govemment consumption 2.1 -1.6 -2.1 8.0 Gross domestic investment -1.7 -0.7 9.5 7.7 -30 Imports of goods and services -4.2 3.0 3.4 7.0 Exports 0 Imports Gross national product -0.7 1.4 3.3 4.7 7 Note: 1997 data are preliminary estimates. The diamonds show four key indicators in the country (in bold) compared with its income-group average. If data are missing, the diamond will be incomplete.  Annex 10 Page 2 of 2 Madagascar PRICES and GOVERNMENT FINANCE Domesfc pnces 1976 1986 1996 1997 Inflation (%) (% change) 60s Consumer prices 8.3 12.4 19.8 4.5 408 Implicit GDP deflator 9.9 14.2 18.1 7.3 30 / Govemment finance 10 (% of GDP, includes current grants) 0 , Current revenue .. 12.7 9.4 11.7 92 93 94 99 96 97 Current budget balance .. 1.9 -1.1 0.7 GDP deflator *CPI Overall surplus/deficit .. -3.5 -8.4 -5.7 _ TRADE (UIS$ millIons) 1976 1986 1996 1997 Exportand import levels (USS millions) Total exports (fob) .. 326 555 531 800 Coffee .. 139 62 37 Other food 48 20 15 60 Manufactures .. 57 348 366 Total imports (cift) . 356 758 799 400 Food .. 52 60 54 21 Fuel and energy .. 58 106 105 Capital goods 94 164 152 o l L L 91 92 93 94 95 96 97 Export prce index (1995=100) .. 97 86 82 Import prie index (1995=100) .. 68 100 92 m Exports U mports Terms of trade (1995=100) .. 143 85 88 BALANCE of PAYMENTS (US$ rrdilions) 1976 1986 1996 1997 Current account balance to GOP ratio (%) Exports of goods and services 330 396 821 795 0 Imports of goods and services 393 490 1,033 1,071 2 Resource balance -63 -95 -212 -276 Net income -17 -155 -158 -108 -1 Netcurrenttransfers 13 152 121 196 Current account balance -68 -97 -249 -188 Financing items (net) 79 163 384 243 -12 1 Changes in net reserves -12 -66 -135 -55 91 92 93 94 95 9a 97 Memo: 1_1 Reserves including gold (US$ millions) 0 41 241 295 Conversion rate (DEC, local/US$) 239.0 676.3 4,054.6 5,093.4 EXTERNAL DEBT and RESOURCE FLOWS 1976 1986 1996 1997 (US$ millions) Composition of total debt, 1997 (USS millions) Total debt outstanding and disbursed 201 3,003 4,498 4,307 IBRD 26 31 7 3 A:3 G:110 IDA 68 422 1,147 1,212 Total debt service 13 211 255 224 b: 12 IBRD 2 4 5 4 IDA 1 5 17 18 E: 1,781 Composition of net resource flows Official grants 22 88 117 179 C: 73 Official creditors 18 149 -5 110 Private creditors 1 47 -5 -1 Foreign direct investment 1 14 10 14 Portfolio equity 0 0 0 0 D: 1,128 World Bank program Commitments 36 103 60 177 A -IBRD E - Bilateral Disbursements 16 95 78 130 B-IDA D - Other multlateral F - Private Principal repayments 0 3 13 14 c - IMF G - Short-term Net lows 16 92 65 116 1 Interest payments 2 6 9 9 Net transfers 14 87 55 107 Development Economics 10/1/98  IBRD 20035R1 }2' ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~12' Antsiiranana 2 MADAGASCAR 4A Pooed rood. Routes goud-rones _____Alt-oaotdw rood, R alt-s s,,les ieS tout, oO NOSY-BE Ambilobe ' -___ Non p.na,on.ot roads H 4ll-'ilb+ ohimorino ----.----Rail-oads . Chon- e' fe, r Ambonjo Flero St 6,r,w.'eO 7k ANTS RANA A .Airpo,ts /r p- Sombara y Notioral capital So.bom '~capt.ai. dfEtot ;B:o Ioo. Faritaoy (P-oiocel capitals A-oioaoa 0I Of>/ C Chef-iu de Foritony Ant ohihy/ 4- ; 4'Andopo ' F _ nitony boundories '\ Antalr ho iMites des For/tony F-'-- \n r l-otornotiool bo-odories /B