Rapò sou Achevman Aplikasyon: Pwojè Fon Ekonomik ak Sosyal Ayiti

Document of The World Bank FOR OFFICIALUSE ONLY Report No.: 18702 IMPLEMENTATION COMPLETION REPORT HAITI ECONOMIC AND SOCIAL FUND PROJECT (CREDIT 2205-HA) December 15, 1998 Country Department III Latin America and the Caribbean Region Human and Social Development Group This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized CURRENCY EQUIVALENTS Currency Unit = Gourde (G) (December 1990) US$1.00 = G 5.00 US$0.20 = G 1.00 (June1998) US$1.00 = G 16.5 US$0.061 = G 1.00 FISCAL YEAR OF BORROWER October 1 - September 30 ABBREVIATIONS AND ACRONYMS DG Directeur General FAES Fonds d'Assistance Economique et Sociale GoH Government of Haiti IDA International Development Association IDB Inter-American Development Bank MARNDR Ministere de l'Agriculture, des Ressources Naturelles et du Developpement Rural MEF Ministere de l'Economie et des Finances MENJS Ministere de l'Education nationale, de la Jeunesse et des Sports MIS Management Information System MSPP Ministere de la Sante Publique et de la Population MTPTC Ministere des Travaux publics, des Transports et des Communications NGO Non-Governmental Organization PAHO Pan American Health Organization SAR Staff Appraisal Report SNEP Service National d'eau Potable Vice President: Shahid Javed Burki, LCRVP Director: Orsalia Kalantzopoulos, LCC3C Manager: Xavier Coll, LCSHD Task Manager: Elizabeth Waters, LCC3C FOR OFFICIAL USE ONLY Table of Contents PREFACE..................................... .3 EVALUATION SUMMARY .5 Introduction .5 Project Objectives .6 Implementation Experience and Results .7 Future Operations and Key Lessons Learned .8 PART I: PROJECT IMPLEMENTATION ASSESSMENT .10 A. Project Objectives .10 B. Achievement of Project Objectives.12 C. Key Factors Affecting Project Implementation .18 D. Project Sustainability.20 E. Bank Performance.22 F. Borrower Performance.23 G. Assessment of Project Outcome .24 H. Future Operations.24 I. Key Lessons Learned.24 PART II: STATISTICAL TABLES .25 A. Table I Summary of Assessment........................................ 26 B. Table 2 Related Bank Loans........................................ 27 C. Table 3 Project Timetable........................................ 28 D. Table 4 Loan Disbursements: Cumulative Estimated and Actual............ 28 E. Table 5 Key Indicator for Project Implementation.............................. 29 F. Table 6 Key Indicator for Project Operation.................................... 30 G. Table 7 Studies Included Project........................................ 30 H. Table 8A Project Costs........................................ 31 1. Table 8B Project Financing........................................ 31 J. Table 9 Economic Costs and Benefits........................................ 31 K. Table 10 Status of Legal covenants........................................ 32 L. Table 11 Compliance with Operational Manual Statements.................... 33 M. Table 12 Bank Resources: Staff Inputs........................................ 33 N. Table 13 Bank Resources: Missions........................................ 34 This document has a restricted distribution and may be used by recipients only in the perfornance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. 2 PART III: APPENDIXES. A. FAES 11pre-appraisal Mission Aide - Memoire*.................................... 35 B. Borrower's Contribution to the ICR.................................... 52 C. Geographic distribution of FAES Investments.................................... 57 D. Map of Haiti.................................... 58 *As all FAES funds were committed by April 1997, at the time of the last supervision mission, it was deemed appropriate to include in the ICR the Aide-Memoire from the December 1997 pre-appraisal mission for FAES 11. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization,- 3 IMPLEMENTATION COMPLETION REPORT HAITI ECONOMIC AND SOCIAL FUND PROJECT (Cr. 2205-HA) PREFACE This is the Implementation Completion Report (ICR) for the Economic and Social Fund Project in Haiti, for which Credit 2205 - HA in the amount of SDR 7,900,000 or US$11.3 million equivalent was approved on January 17, 1991 and the Credit Agreement was made effective on May 21, 1991. The loan was closed on June 30. 1998, three years after the original closing date of June 30, 1995. It is expected to be fully disbursed, and the last disbursement is expected to take place on October 31, 1998. Cofinancing for the project was provided by the Inter- American Development Bank (IDB). The ICR was prepared by Elizabeth Waters of the Caribbean Country Management Unit of the Latin America and Caribbean Region and reviewed by Orsalia Kalantzopoulos, Director of LCC3C, Xavier Coll, Director of LCSHD, Jamil Salmi, Country Sector Leader of LCSHD, and David Warren, Operations Specialist, LCSHD. Preparation of this ICR was begun after the Bank's final supervision mission in April 1997 and the Bank's pre-appraisal mission of FAES II in December 1997. It is based on material in the project files and interviews in Haiti. The borrower contributed to the 4 preparation of the ICR by commenting on the draft ICR, contributing views reflected in the FAES II pre-appraisal mission aide-memoire (Appendix A), and preparing its own evaluation of the project's preparation and execution (Appendix B). 5 ECONOMIC AND SOCIALFUNDPROJECT (Cr. 2205 - HA) HAITI EVALUATION SUMMARY Introduction 1. Among the major challenges facing Haiti in 1990, after several years of political unrest and economic decline, were the need to instill hope among the Haitian people and to root democracy firmly in institutional practice throughout the country, while simultaneously addressing the urgent needs of the poor via quick relief efforts during the transition period to political and economic stability. 2. The Bank has supported economic and social sector projects in Haiti since 1956. Prior to the Economic and Social Fund (FAES: Fonds d'Assistance Economique et Sociale) project, the Bank financed three projects in the health, education and water supply/sanitation sectors in Haiti: the Education IV-A Project (Cr. 1592-HA approved in 1985 for US$10 million), the Health and Population Project (Cr. 2085-HA approved in 1990 for US$28.2 million), and the Water Supply and Sanitation Project (Cr. 2052-HA approved in 1990 for US$15.2 million). The first loan helped finance the GoH's basic education reform, initiate a reform of the vocational/technical training system, and strengthen the management of the education sector. The purpose of the second project, which is still active, is to improve access to basic health care services and support interventions to address the impact of the TB and AIDS epidemic. The third project, which is also still active, is strengthening water management, improving the supply of drinking water in Port-au-Prince, and developing a sanitation plan for the capital's metropolitan area. 3. Concurrent with the Economic and Social Fund Project, the Bank-funded Emergency Economic Recovery Project (Cr. 2670-HA approved in 1994 for US$40 million) and Employment Generation Project (Cr. 2765-HA approved in 1995 for US$50 million) also supported the GoH's efforts to combat poverty. The former financed rehabilitation and maintenance for the agriculture, energy and infrastructure sectors, and provided employment and income opportunities for the poor. The latter provided short- term employment to people in extreme poverty, improved seriously deteriorated infrastructure, and strengthened the capacity of institutions to manage small projects. 6 Two projects currently under preparation will perpetuate such support to the GoH: the Basic Education Project (Cr. 29389-HA, for US$28 million), which will improve the quality and equity of basic education services; and FAES II (US$25 million), which will continue to finance small projects in poor communities, with an emphasis on promotion, targeting and the sustainability of investments (Table 2). Project Objectives 4. As defined in the Development Credit Agreement dated February 4, 1991, the main objectives of the Economic and Social Fund project (Cr. 2205. - HA) were to assist the Borrower in: (i) establishing FAES as a means of responding efficiently and in a decentralized manner to the basic needs of the nation's poor through non-governmental organizations, including cooperatives, organized community groups and other grass-roots organizations; (ii) strengthening health care, nutrition and education services (primarily for infants and pregnant and lactating women) and providing physical infrastructure addressing the needs of the poorest population; (iii) providing an effective channel for required donor financing and coordination of scattered domestic efforts; and (iv) strengthening the capabilities of FAES and delegated entities for project preparation and implementation. The project combined such diverse activities as: civil works; provision of equipment and supplies; support to epidemic control, immunization, and adult literacy training programs; construction of water supply systems, sewerage systems and latrines; and advisory services to communities, for a total project cost of US$23.7 million, of which US$1 1.3 million equivalent (SDR 7,900,000) were to be creditedL by IDA. 5. The project objectives, though clearly stated and consistent with the GoH's economic and social policies as well as with the Bank's country strategy for Haiti, were broad in scope and the strategies to achieve them resulted in a complex and very demanding project, considering that FAES was a new agency with limited capacities at the outset. 6. Key loan covenants expected to promote the achievement of project objectives included: (i) employing qualified fixed-term staff satisfactory to IDA; and (ii) identifying, appraising and implementing subprojects under terms set forth in the Operational Manual. Notwithstanding the fact that audits were consistently overdue, the Borrower's compliance with all loan covenants was highly satisfactory. No major gaps in key positions occurred during the project. The steering committee met regularly and fostered inter-institutional collaboration in addressing implementation obstacles, which was one of the few examples of successful public-private collaboration in Haiti. However, the targeting and impact studies were seriously postponed. 7 Implementation Experience and Results 7. As FAES was basically demand-driven, there were no quantitative targets for numbers of different kinds of projects at the outset. Nearly all civil works, equipment and supplies financed were completed or procured: out of the 349 projects financed by IDA at completion date, 310 were completed, 34 were in final stages, and only 5 had been abandoned. In the health and nutrition sector, 10 nutrition projects were implemented, 5 health facilities were rehabilitated and 9 were constructed, equipment was provided for 5 health centers, 4 preventive health programs were supported, and 24 small water supply systems as well as 69 small sewerage systems and latrines were built or repaired. In the education sector, 10 primary school facilities were rehabilitated and 24 were constructed, furniture and equipment were provided for 149 school facilities, education materials were provided for 21 schools and 2 training centers, and support was given to one literacy and one training program. In the infrastructure sector, 9 road projects (bridges) and 5 agricultural projects were implemented. Overall, the project had a positive impact on the welfare of most client communities. Technical assistance was provided to FAES for several key studies, as planned; however, funds for technical assistance to weak grass- roots organizations were not disbursed. 8. The project was implemented over a seven-year period, including a three-year extension due exclusively to a complete suspension of activities in October 1991 because of a military coup. Following the relaunching of the project in April 1995, smooth project implementation was affected temporarily by factors such as: (i) the limited institutional capacities of FAES; (ii) deficiencies of several firms contracted to manage civil works activities; (iii) delays in the disbursement of IDA funds; and (iv) the limited institutional capacities and commitments of key line ministries. However, as management systems rapidly became more efficient and effective, the project was able to disburse completely over a three-year period, faster than initially foreseen when the credit was reactivated in 1995, and FAES became a critical source of funding for poor communities. Overall, given the special circumstances of Haiti during the last seven years (the extreme poverty, the destructive impact of both the coup and the international embargo, the restoration of a fragile democracy, extremely weak institutional capacity at every level of the Government, etc.), the project has been quite successful. 9. Bank Performance. Overall, Bank performance was marginally satisfactory at the identification. preparation and appraisal stages, and highly satisfactory at the supervision stage. The Bank responded quickly and effectively to the GoH's request for assistance in identifying project components and procedures, and several characteristics of the project were the result of its conditionalities. While appraisal was thorough in some aspects, technical studies and cost estimates were weak, and insufficient attention was paid to (i) promotion and targeting strategies, (ii) subproject sustainability, recurrent costs and maintenance issues, and (iii) the need for intensive mronitoringand follow-up during project execution. The Bank paid a relatively high level of attention to the project after appraisal, with over 40 weeks of staff inputs in the 12 months between appraisal and 8 effectiveness. Supervision missions involving 101 weeks of staff inputs were consistently conducted from December 1994 until project completion; five missions were conducted in 1995 alone, during the relaunching period after the return of the legitimate Government from exile. Bank supervision was effective in diagnosing the most visible underlying constraints, and project inputs were closely monitored; however, funds were not always made available on time, and project impact and policy targets were unevenly evaluated. 10. Borrower Perbrmcance. The performance of the Borrower during project preparation was satisfactory, and initial commitment to the project was strong. With the arrival of President Aristide in 1991, however, the launching of activities was postponed due to a disagreement that arose concerning the position of DG. Borrower performance varied during the course of the project, but was generally satisfactory, as FAES was allowed to operate with a minimum of interference from GoH. While interinstitutional coordination improved gradually and FAES was able to sign memoranda of understanding with several line ministries and para-public bodies, it remained generally weak, and Board meetings focussed more on identifying and addressing common obstacles to the efficient delivery of project outputs and less on the linkages between the subprojects and achieving policy targets. 11. Project Outcome. The overall project results were highly satisfactory, as practically all output objectives were met. Although specific outcome/impact indicators were not built into subproject designs, there is evidence of reduced overcrowding and improved learning in schools, better health care due to upgraded facilities, reduced incidence of infectious diseases due to vaccination campaigns and new or rehabilitated water supply systems, and strengthened capacity of FAES to manage multisectoral initiatives. Future Operations and Key Lessons Learned 12. The GoH has retained FAES to act as implementing agency for several new projects, or project components. The IDB continues to support civil works and other activities under a US$23 million FAES II project, and some bilateral donors have also indicated interest. The Project Concept Document for IDA's FAES II Project has been prepared. It is recognized that there has been a total breakdown of social services in Haiti in recent years, and that institutional strengthening efforts to reconstruct the social sector ministries are likely to take years; in this context, FAES is now serving a vital need in Haiti as a large source of community-based financing and the only institution that directly addresses poverty. As a result, the rationale for a follow-up FAES project is clear. 13. Several key lessons emerge from the project: (i) project design must incorporate adequate institutional measures aimed at strengthening the implementing agency from the beginning ; (ii) thorough integration of project implementation units and activities (e.g., promotion, evaluation, control and monitoring) within the implementing agency increases the probability of objectives being achieved and sustained; (iii) given the poor 9 performanceof severalfirms in implementing or supervisingcivil works, new optionsfor the monitoringof works and ex-ante/ex-post evaluationof firms must be consideredand piloted;(iv) specificproject outcomeand impact measuresin additionto inputand output indicatorsshouldbe establishedduringdesign,consistentlymonitored,and evaluatedas part of the projectto determineprojectachievementand impact;(v) project successin a communitysettingis consistentlydue to shared leadershipat the project identification/preparation stage,strong communityparticipation at all phasesof the projectcycle,and the capacitiesfor self-relianceto be found within the community. 10 PART I - PROJECT IMPLEMENTATION ASSESSMENT A. Project Objectives I. Among the major challenges facing Haiti in 1990. after several years of political unrest and economic decline, were the need to instill hope among the I-laitianpeople and to root democracy firmly in institutional practice throughout the country. while simultaneously addressing the urgent needs of the poor via quick relie' ef'forts durilngthe transition period to political and economic stability. In this context, the Economllic and Social Fund (FAES) was established by the Government of Haiti (GolI) as an administratively and financially autonomous project financing body to achieve a mission whichi,as stated in the decree creating the agency, is "to finance short-term projects with a highllabor component to improve the living conditions of low-income populations in urban and rural areas and increase their productive potential". 2. The FAES project was designed as a US$23.7 million operation, to include a subproject component (US$21.3 million) and an institutional development component (US$2.4 million). The project, for which Credit 2205-HA in the amount of SDR 7,900,000 (US$11.3 million equivalent) was approved on January 17, 1991 and signiedon 1-F'ebruary 4, 1991 (three days prior to PresidenitAristide's inauguration), sought to enable the newly-elected Gol-Ito meet its poverty alleviation and political objectives. As definiedin the Development ('redit Agreemlient dated February 4,1991, the objectives of the project were to assist the Borrower (Gol l) in: (i) establishing FAES as a means of responding efficiently and in a decenitralizedmaninerto the basic needs of the nation's poor through non-goverinmenital organiizationis. including cooperatives, organized communlitygroups and othlergrass-roots organizations; (ii) strengthening health care, nutrition and education services (primilarily lor intfantsand pregnant and lactating women) and providing physical intrastructure addressing the needs of the poorest population; (iii) providing an effective channiellor required donor financing and coordination of scattered domestic eltorts; and (iv) strenigthleniilng the capabilities of and delegated entities for project preparation and implemiientLationi. 3. Tlheapproaclhesoutliniedullder these objectives were broadly concordant with the Bank's country strategy tor 1-laiti, whiichi aimed to: (i) alleviate poverty by providing assistance via selected investmilenits in critical social services, such as health, nutrition and education directed towards the low-income population, emphasizing project management and institutional development; (ii) provide limited support for the productive sectors to arrest the decline in the productive base; and (iii) finance projects designed to prevent fuirtherdeterioration of physical facilities and the environment. However, the objectives, though clearly stated. were broad in scope and the strategies to achieve them resulted in a complex and very demanding project: in addition to an institutional development component, the project comprised a subproject component involving activities to be I1 implemented in several sectors by a new agency with limited capacities. This component combined such diverse activities as: civil works; provision of equipment and supplies; support to epidemic control, immunization, and adult literacy training programs; construction of water supply systems, sewerage systems and latrines; and advisory services to communities. 4. The IDA credit for the subproject component (US$10 million) was to finance the following: (i) health projects (subtotal: US$6.4 million), including construction and rehabilitation of health centers with an emphasis on primary health care facilities in rural areas, provision of equipment and essential drugs, support to epidemic control, immunization, day care and nutrition programs for preschool children and pregnant and lactating women, construction of small water supply systems, and small sewerage systems and latrines; (ii) education projects (subtotal: US$2.5 million) including primary school construction and rehabilitation, primary school furniture and equipment, textbooks, and support to informal training programs and adult literacy programs targeting women involved in businesses; and (iii) rural roads (subtotal: US$1 million) to provide access for poor communities to markets, schools and health centers. Advisory services (sub-total: US$0.1 million) were also to be provided to weak grass-roots organizations in the poorest communities lacking the institutional capacity to identify and carry out projects. No objective was stated relating to project sustainability, or more specifically to the establishment of user committees or recurrent cost financing. The estimated sectoral distribution in the project cost table reflected needs as demonstrated by the 182 proposals already received from requesting agencies at the time of project appraisal, as well as estimated institutional capacity in the various sectors to implement sound projects. Expectations were expressed not in terms of the number of subprojects to be implemented in each sector or subsector, but of average subproject or unit cost. The costs were estimated at US$45,000 equivalent in health and education; it was estimated that rehabilitation of rural roads would cost approximately US$1 0,000/km and that construction could vary from US$20,000/km to US$40,000/km depending on the technique used. 5. The portion of the IDA credit for the institutional development component (US$0.9 million) was to finance salaries and travel expenses of selected fixed-term FAES staff, consultant services, office equipment and supplies, vehicles and microcomputers. Technical assistance was also to be provided to FAES for: (i) the establishment of a price data base on unit costs for construction materials and labor, as a basis for procurement under direct contracting; (ii) the design and implementation of a management information system, including monitoring and evaluation systems; (iii) the completion of an operational manual including project evaluation criteria; (iv) the efirnementof poverty targeting mechanisms; (v) the design of an impact study; and (vi) the establishment of an accounting system. 12 B. Achievement of Project Objectives 6. The project gradually achieved its broad objective of assisting the GoH in: (i) establishing an effective mechanism - the FAES - as a means of responding efficiently and in a decentralized manner to the basic needs of the nation's poor through NGOs and grass-root organizations; (ii) providing an effective channel for required donor financing and coordination of scattered domestic efforts; and (iii) strengthening the capabilities of the FAES and delegated entities for project preparation and implementation. Efficiency was quite low during the first full year of operation, chiefly because of the limited promotion, evaluation and supervision capacities of FAES, and of the deficiencies of several firms contracted to manage civil works activities. Major constraints were subsequently overcome, however, as various departments were reorganized, capabilities for project promotion, preparation and implementation were developed, and major bottlenecks in the project production line were reduced. From an administrative organizational perspective, FAES did not respond in a decentralized manner to the needs of the nation's poor, as all promotion and evaluation activities and all operations were managed from Port-au-Prince, due to major telecommunication, transportation and other logistical constraints throughout the country. Because FAES was established by the GoH basically as an administratively autonomous body to finance short-term projects, it was never made responsible for the coordination of "scattered domestic efforts," as such coordination has understandably remained the responsibility of the GoHiitself, under close scrutiny from Parliament. 7. The extent to which sectoral objectives were achieved is difficult to assess for three main reasons: (i) specific outcome/impact indicators for the project were not described in the Staff Appraisal Report (SAR), nor measured during or subsequent to implementation, with the exception of sample surveys carried out in July 1996 by an outside IDB consultant and in the fall of 1997 at FAES' request by Quisqueya University's Institut de recherche, de fbrmation el d 'expertise-conseil en gestion d'entreprise (IRFEC); (ii) it is often difficult to isolate the effects of a project from those of other sectoral interventions; and (iii) implementation delays resulted in most activities being completed only in the last two years, and thus their impact cannot yet be accurately determined. With the above caveats, available information reveals that the project was increasingly effective in assisting the GoH with the provision of physical infrastructure and equipment addressing the needs of the poor population; despite implementation delays in the first full year of project implementation (1995-1996), nearly all civil works were completed, and almost all equipment and supplies were procured. However, funds earmarked for technical assistance to weak grass-roots organizations were not disbursed as envisaged. Health Sector (Original Base Cost: US$6.4 million; Actual Total Commitment: US$5.5 million) 8. Nutrition programs. The SAR envisioned support for an unspecified number of nutrition programs for preschool children, and pregnant and lactating women. Due to concerns about the complexity of administration, sustainability and accountability of 13 nutrition programs in Haiti, FAES decided to channel efforts into facilitating distribution and logistical support to school feeding programs, as well as nutrition programs within child care services. Projects (worth US$204,000) presented by a large religious NGO (Mission Alpha) specializing in literacy programs were implemented in all nine regions, and an NGO caring for the handicapped in Port-au-Prince was provided with kitchen equipment. 9. Health care facilities. The SAR anticipated the rehabilitation or construction of an unspecified number of primary health care centers in rural areas. At project closing, five facilities were rehabilitated (total cost: US$175,000) and nine were constructed (total cost: US$405,000). The rehabilitation and construction of the facilities were intended to provide access to modern health care and to reduce the demand for service at higher levels of care, thereby increasing efficiency and lowering overall costs. As no impact assessment was implemented, patient satisfaction with the upgraded or new facilities has not been confirmed. It can only be assumed that the heavy constraints to adequately staffing and managing new health facilities in rural Haiti were at least partly overcome, and that the civil works contributed to the improvement of primary health care in all selected areas. 10. Medical equipment and supplies. The project successfully procured and delivered furniture and medical equipment (worth US$228,000) to five health centers. Although an important contribution, the amount of medical equipment financed by the project was very small compared to the overall stock of old and obsolete equipment in need of replacement. The project also provided pharmaceuticals (worth US$39,500) to a large NGO in Port-au-Prince, and supported UNICEF (four projects worth a total of US$439,000) immunization programs in five regions. 11. Water Supply and Sanitation. Potable water supply has top priority, especially for rural communities, and sanitation measures are also considered to be very important in the prevention of communicable diseases. The main subsector objectives were to improve operation and maintenance of the existing water supply and sanitation systems, and to expand coverage. The project built or repaired 24 small water supply systems (total cost: US$1.135,212) and 69 small sewerage systems and latrines (total cost: US$2.9 million). Although impact of all these projects is difficult to assess as noted above, there is evidence of expanded coverage and improved quality of service. Inasmuch as maintenance plans are adhered to by the beneficiaries, the projects could contribute considerably to the prevention of waterbome and communicable diseases in the targeted communities. Education Sector (Original Base Cost: US$2.5 million; Actual Commitment: US$3.8 million) 12. Primary schoolfacilities. The SAR envisioned the rehabilitation or construction of an unspecified number of preschool centers and primary schools, especially in rural areas. At project closing, 10 facilities were rehabilitated (total cost: US$590,000) and 24 were constructed (total cost: US$1.7 million). The rehabilitation and construction of the 14 facilities were intended to increase access to primary education, with added space for approximately 8,000 students; satisfaction with the upgraded or new facilities, as well as increased enrollment, have not been confirmed, although there is reason to believe that the rehabilitation projects have significantly enhanced the quality of existing education services in selected communities, as overcrowding was reduced and learning environments were improved. As for the 24 new school facilities, it can only be assumed that some of the many constraints to adequately staffing and managing primary schools in rural Haiti were overcome. 13. Provision of equipment and supplies. The project successfully procured and delivered furniture and equipment (worth US$1.1 million) to 148 pre-school and primary school centers, in addition to one informal training center for teachers. It also provided pedagogic material and textbooks to 21 schools and two teacher training centers (total cost: US$260,000), as well as to the Ministry of Education for its adult literacy programs (US$150,000), and supported an informal training program for teachers. No impact evaluation of the textbook and training components was included in the project, but anecdotal evidence would confirm the positive benefits on student achievement. Infrastructure (Original Base Cost: US$1.0 million; Actual Commitment: US$243,000) 14. The SAR anticipated the rehabilitation and/or construction of rural roads and related drainage works, providing access for poor communities to markets, schools and health care facilities. Original base cost allowed for the repair or construction of up to 100 km of roads. Actually, at project closing, as few as nine small road projects were completed (total cost: US$107,000) and these were in the vicinity of Port-au-Prince and in the North-West only. All projects were aimed at repairing or constructing bridges. No project was envisioned at the outset, under IDA financing, in the agricuilturalsector; however, five projects (total cost: US$136,000) dealing with irrigation, soil conservation and stockage, were implemented in this sector. No impact study dealing with these subprojects was conducted. Advisory Services (Original Base Cost: US$0.1 million; Actual Commitment: US$300) 15. The SAR envisioned the provision of advisory services to weak grass-roots organizations in the poorest communities lacking the capacity to identify, prepare and carry out development projects. As it turned out, such services were provided to a certain extent in the course of regular FAES promotion activities, following the establishment of a mechanism for prioritizing subprojects, and the adoption of promotion strategies to target underrepresented regions and demanders requiring technical support. Practically no IDA financing was specifically allocated to such activities. Institutional Support (Original Base Cost: US$0.9 million; Actual Commitment: US$974,000) 16. The SAR depicted an institutional development component to finance salaries and travel expenses of selected fixed-term FAES staff, consultant services, office equipment 15 and supplies, vehicles and microcomputers. At project closing, all equipment, furniture and vehicles required for FAES to be optimally operational had been supplied as needed, although at a third of the cost that was envisaged at the outset. However, salaries paid to selected fixed-term FAES staff over the years were 28 percent higher than envisioned. 17. Technical assistance was provided to FAES for several key studies, as planned. An accounting system was established and developed, and gradually became operational. This task was very demanding, as several components of the initial financial management system were more adapted to institutional management than to project monitoring and management requirements. An operational manual including standard ex-ante project evaluation criteria was completed in 1993 and revised in 1995, and was under revision again at the time of project closing. A formal and quantitative management information system (MIS) was successfully set up in 1995 with support from the Ecuadorian Social Fund; all FAES staff received training on how to feed and use the information in the system, and the MIS was subsequently improved and adjusted by local technicians to meet the specific needs and conditions of the FAES; however, the system was expected to include monitoring and evaluation subsystems, and the Fund has not yet strengthened its capacity to use monitoring information on a regular and systematic basis. A unit cost study was conducted by a local engineering firm, and a dynamic price data base for construction materials and labor was established in March 1997 as a basis for procurement under direct contracting; the system allows for accelerated project evaluation, increased transparency, and better cost management. The establishment of a poverty map and the implementation of a planned targeting study were seriously postponed, which delayed important promotion activities aimed at ensuring a more equitable distribution of projects by sector, type, region and demander; the study was finally conducted by Quisqueya University and targeting mechanisms were designed and gradually refined as a basis for prioritizing projects. The impact study was also seriously postponed, and information on the impact of FAES-supported projects was not readily available until the report of a sample survey, conducted again by Quisqueya University (IRFEC), was finalized in April 1998; in addition, supervisors produced very few final project reports, and follow-up visits to confirm that FAES-supported schools and health centers are operational were not carried out. 18. The FAES was not in a position to conform its activities to the macroeconomic policies determined by the GoH, or even to the sector policies determined by the line ministries, as such policies had yet to be drafted and submitted to elected bodies for approval.' At the time of credit approval, there was no broad social and economic program to combat poverty in Haiti, and both FAES and the GoH were acting as "ad- hocracies," long on short-term and incremental planning and short on long-term development strategy. FAES' financing remained exclusively demand-driven for well over a year, as targeting mechanisms based on poverty maps and indicators had yet to be The unelected Government's macroeconomic objectives at the time of loan approval were basically those of an IMF stand-by program. The Government's current macroeconomic objectives are still those of a stabilization program, and have not been approved by elected bodies. The Ministries of Education and Health were able to make their sectoral policies known only in 1997, and such policies have yet to be approved by Parliament. The MTPTC is still in the process of delineating its own policies. 16 used, and it could not reasonably be expected - in the short term, at least - to become an intrinsic part of a broad social and economic program to combat poverty. In addition, it is difficult to establish to what extent the composition of the Fund's project portfolio reflected - at least, until recently - community preferences, as opposed to the interests of religious leaders, politicians, NGOs, contractors, or even suppliers, as surveys of community needs, priorities and preferences, as well as needs analyses, were not conducted. The GoH's financial objectives were at least partially met, however, as no significant counterpart funding was required from the Ministry of Finance, and practically no commitment was required from the line ministries concerning the recurrent costs associated with maintenance of the facilities to be constructed or rehabilitated. 19. The project's contribution to the institutional development of FAES itself, and more specifically to its establishment as a mechanism capable of assisting the Government in responding flexibly and efficiently to the basic needs of selected poor communities through grass-roots organizations, was quite substantial. However, FAES' contribution to the institutional development objectives of line ministries was only partial, as this was not a specific objective of FAES. The Fund has managed to transfer some components of its operational technology to ministries such as MENJS and MTPTC, and - notwithstanding the fact that some of the country's ablest civil servants have found a "better" job with the Fund - there is no indication that the creation and expansion of FAES has undermined the capacity of government institutions. The fact remains, however, that while FAES has entered into formal agreements with four ministries as well as with SNEP, there has been little genuine coordination between the subproject programs of the Fund and those of the line ministries and of other public bodies. If only because the efficiency and effectiveness of their management teams are so weak, the Fund has been led to act, as a rule, on its own. In addition, FAES has implemented projects that GoH ordinarily would not. This is either because of a lack of resources, because the projects are requested by municipal governments, or because they are grass-roots projects requested by private sector entities such as NGOs, religious groups or community organizations, who have learned over decades how not to depend on the State. 20. Achievement of the physical objectives of the project was substantial. Whereas rough estimations at appraisal were that approximately 200 health and education subprojects could be implemented under IDA financing, subprojects in these sectors actually numbered 335, including 141 social infrastructure subprojects. In addition, 14 economic infrastructure subprojects were implemented. The quality of physical outputs (buildings and other structures) is variable, but in most cases it is acceptable, and in some cases (e.g., clean water supply systems) it is very high. Several problems stem from the fact that projects were often not comprehensive and had been planned in an incremental fashion, as components of larger projects. Some structures indicate negligence on the part of the contractors as well as a lack of vigilance on the part of the engineers hired by FAES to supervise construction. Some schools have ventilation and lighting problems which could have a negative effect on the health and learning abilities of children. Some still lack a proper water supply system. Some were equipped with electrical outlets, even when communities had no intention of using them or no power supply was to be made 17 available. Several dry latrines are mislocated, and lack of maintenance and/or proper drainage could generate health problems for user communities. 21. While poverty reduction was understandably negligible, as this was not an explicit objective, the project had a positive impact on the welfare of most communities where subprojects were executed, and succeeded in mitigating some of the most adverse effects of the economic crisis brought about by the military coup and the international embargo. It provided many poor communities with infrastructure, equipment and supplies. Short-term employment was quickly generated, and this allowed the communities to buy goods and services that were not readily accessible before. 2 Clean water supply systems and latrine have most probably contributed to a decline in the incidence of infectious diseases. Thousands of children were provided with better school facilities and basic school furniture and supplies. And thousands of farmers were provided with better feeder roads and protection against soil erosion. 22. Achievement of the objectives relating to gender issues was partial. The FAES has yet to identify and address women's issues within its programs, to adopt a policy regarding gender equity, to support deliberate initiatives aimed at increasing women's participation in various decision-making processes and improving women's income levels and socioeconomic conditions, and to build its institutional capabilities so that gender considerations are fully integrated into policies, programs, projects and activities. The employment-generating function of the infrastructure projects has tended to bypass women, who represented less than 10 percent of paid workers on a typical construction site. Characteristically, women have been expected to fetch pail after pail of water as required by paid laborers, without being paid the usual minimum daily wage granted to other men and women; this bias has been unfortunate, as women are usually hard workers, who also happen to head 50 percent of all households in Haiti, and are known to spend quite judiciously what earnings come into their hands. At least, women benefit more than men as users of the health infrastructure delivered; health centers are visited more by women, who need pre- and post-natal care and are responsible for their children's health. Clean water supply systems are also used primarily by women and girls, and it is expected that the new facilities will alleviate their hardships. 23. Achievement of environmental objectives was negligible. Current concern with environmental problems in rural Haiti is focused on the alarming rate of deforestation and soil erosion: no subproject was implemented with the objective of either protecting the remaining forests or increasing wood production. While two soil conservation subprojects were implemented, satisfaction with these projects was not documented, and no environmental impact assessment was conducted. Subproject proposals were to include environmental impact assessments in relevant areas; health facilities were to be designed so as not to contaminate the environment, and equipped with septic tanks or improved latrines; similarly, a plan for improvement of sanitation of schools was to be required, including latrines. However, no environmental impact assessment was ever 2 It is estimatedthat between80 and 100 person/months of short-term employmentweregeneratedby each of the 155 infrastructure projectsunderIDAfinancing. Locallaborerswere paidminimumdaily wage(currently,approx. US$2.25). 18 conducted, and this makes it difficult to establish the extent to which environmental objectives were achieved. While most latrines and drainage systems were built according to specifications, it was observed that some dry latrines are poorly located, and that lack of maintenance and/or proper drainage could generate environmental problems in several areas, especially near schools and health centers. 24. Achievement of objectives associated with public sector management was partial, as already discussed (para. 19). Achievement of the objectives associated with private sector development was also partial. FAES used a large number of small engineering firns or contractors, who earned supervision fees (5 percent) and profits (10 percent) estimated at well over US$ 1.0 million (IDA financing only). As the vast majority of project inputs were procured in the domestic market, employment generated directly or indirectly by the subprojects in the short-term (e.g., provision and transportation of local inputs) was significant, and a large proportion of the total income generated by these subprojects wound up in the hands of importers and traders. Local and international NGOs were also widely used as executing agencies, or as "intermediary agencies" in the case of three large organizations. However, FAES did not fund one single project belonging to the "directly productive" category (e.g., rotating funds for small enterprises, breeding stock projects, agricultural and marketing cooperatives). While directly productive projects were eligible for financing at the outset and as such appeared on the project menu, and over 150 requests were submitted by small farmers or entrepreneurs, this category was removed from the menu in the fall of 1995, as it did not fit with FAES' dominant "production-line" strategy. C. Key Factors Affecting Project Implementation 25. Project implementation was affected by two main factors: (i) a military coup, which led to a complete suspension of activities in October 1991, for a period of three years; and (ii) the limited institutional capacity of FAES during the first full year of activities (1995-1996), coupled with deficiencies of several firms contracted to manage civil works activities. Project implementation was also partially affected by (i) delays in the disbursement of IDA funds; and (ii) the limited institutional capacity and commitments of key line ministries. 26. Factors Not Subject to Government C'ontrol. FAES was established by the GoH in 1990, under the Ministry of Economy and Finance (MEF), and a DG was named in June 1990 by the unelected Head of State, for a two-year period. The Credit Agreement was signed in February 1991 and pilot projects were soon envisaged, but with the arrival of President Aristide, the Fund's launching was postponed due to a disagreement that arose between the financiers and the President: the latter, in opposition to the IDB and IDA, wanted to name a new DG. By the time of the military coup in September, 1991, the Fund had not been able to begin operations. The project was suspended until the return of President Aristide from exile in the fall of 1994, and subproject activities started in April 1995. Because of the coup and the international embargo that followed, the 19 project was thus implemented over a seven-year period, including a formal three-year suspension, instead of the four-year period originally envisaged. 27. There were delays in the disbursement of IDA funds during the first half of 1996. The delays were so prolonged that FAES was forced to use IDB funds to cover some expenses that would have been covered by IDA financing under normal circumstances. Whereas approx