Honduras - Third Social Investment Fund Project (FHIS Second Stage)

Document of The World Bank Report No. 14181-HO STAFF APPRAISAL REPORT HONDURAS THIRD SOCIAL INVESTMENT FUND PROJECT (FHIS SECOND STAGE) JUNE 16, 1995 Country Department II Human Resources Operations Division Latin America and the Caribbean Regional Office  CURRENCY EQUIVALENTS Currency Unit = Lempira (L) US $1.00 = 9.30 Lempira (Feb. 1995) SDR 1.00 = US $1.57582 (April 30, 1995) FISCAL YEAR January I - December 31 ACRONYMS AND ABBREVIATIONS AMNHON Honduran Association of Municipalities (Asociaci6n de Municipios de Honduras) FHIS Honduran Social Investment Fund (Fondo Hondureino de Inversion) GOH Government of Honduras IDA International Development Association IDB Inter-American Development Bank Kf4W German Credit Institute for Reconstruction (Kreditanstaltfur Wiederauflbau) ICB International Competitive Bidding MIS Management Information System MOE Ministry of Education (Secrelaria de Educacion Pzihlica) MOF Ministry of Finance (Secretaria de Hacienda y Credito Pfiblica) MOH Ministry of Health (Secrefaria de Salud Pfiblica) NCB National Competitive Bidding NGO Non-governmental Organization OPEC Organization of Petroleum Exporting Countries PRAF Family Assistance Program (Programa de Asignaci6n Familiar) RUTA SOCIAL Regional Unit for Technical Assistance in the Social Sector SECPLAN State Secretariat of Planning, Coordination, and Budget (Secrefaria de Planificaci6n, Coordinaci6n y Presupuesto) SDB Standard Bidding Document SEDA State Secretariat of Environment (Secretaria de Ambienfe) SIF I Social Investment Fund Project SIF II Second Social Investment Fund Project SOE Statement of Expenditure UNICEF United Nations International Children's Emergency Fund UNDP United Nations Development Program USAID United States Agency for International Development  HONDURAS THIRD SOCIAL INVESTMENT FUND PROJECT STAFF APPRAISAL REPORT TABLE OF CONTENTS 1. ECONOMIC AND SOCIAL CONTEXT I A. Economic Reform I B. Poverty Reduction 2 C. Government Strategy 3 D. Bank and IDA Support 3 E. Rationale for IDA Involvement 4 2. HONDURANSOCIAL INVESTMENTFUND S A. Background 5 (1) Objectives 5 (2) Approach 5 (3) Achievements 8 B. Implementation Experience and Issues 10 (1) Targeting 10 (2) Subproject Portfolio 12 (3) Management Information System 15 (4) Collaboration with Line Ministries and Other Govermnent Institutions 15 (5) Resource Mobilization and Needs 16 C. Lessons Learned from Past Operations 17 3. THE PROJECT 22 A. Project Objectives and Description 22 B. Project Components 22 4. PROJECT COST, PROCUREMENT, DISBURSEMENTAND AUDITS 25 A. Project Cost and Financing 25 B. Procurement 29 C. Disbursements 31 D. Accounts and Audits 32 This report is based on the findings of a joint IDA, IDB, and KfW appraisal mission that visited Honduras in February 1995. The IDA team was composed of Messrs./Mmes. Willem Struben (mission leader), Livio Pino (IDA), Steve Maber, Maria Eugenia Zavala, Maria Elena Anderson, and Alfredo Sarmiento (RUTA Social), David Warren, Gillette Hall, Peter Sollis, Luis Schmidt, and Alejandro Cano- Ruiz (consultants). The IDB team was composed ofMs. Beatriz Uribe (mission leader), Mr. Jorge Torres, and Mr. Andres Garrett (consultant). Mr. Nassir Djafari (KfW) joined part of the time. Mmes. Diane Steele, Rosaria Troia, and Lerick Kebeck contributed to the report at headquarters. Peer reviewers are Messrs. Jacques van der Gaag (PHN) and Jean-Jacques de St. Antoine (LAIHR). Messrs Edilberto L. Segura and Kye Woo Lee are Director and Division Chief respectively, for this operation.  ii 5. PROJECT MANA GEMENT AND SUPER VISION 34 A. Project Implementation 34 B. Project Supervision 36 6. PROJECT BENEFITS AND RISKS 36 A. Project Benefits 36 B. Participation of Women and Indigenous Groups 36 C. Environmental Impact 37 D. Project Risks 37 E. Poverty Category 38 7. AGREEMENTS TO BE REACHED AND RECOMMENDATION 38 8. ANNEXES 41 1. Social Sector Data and Requirements 2. FHIS Poverty Targeting 3. FHUS Organization Chart 4. Community Participation 5. Subproject Environmental Assessment 6. Subproject Proposals and 1995 Program 7. Coordination with Line Ministries and Municipalities 8. Infrastructure Subprojects Component 9. Basic Needs Program Component 10. Institutional Strengthening Technical Assistance Component 11. Social Sector Data Technical Assistance Component 12. Environmental Technical Assistance Component 13. Project Costs 14. Project Financing 15. Credit Disbursement 16. Project Implementation 17. Monitoring and Evaluation 18. Supervision Plan and Mid-Term Review 19. List of Documents in Project Files LIST OF TABLES 1. Distribution of Number and Amount of Approved Subprojects by Type of Soliciting Agency 2. Distribution of Resources for Approved Subprojects by Menu Category 3. Investments Approved by Poverty Category of Municipality and Year 4. Amounts and Sources of Financing by Year 5. Project Cost Summary 6. Financing Plan 7. Procurement Methods by Category  111 HONDURAS THIRD SOCIAL INVESTMENT FUND PROJECT CREDIT AND PROJECT SUMMARY Borrower: The Republic of Honduras Implementing Agencies: Honduran Social Investment Fund (FHIS), State Secretariat of Planning (SECPLAN), and State Secretariat of Environment (SEDA). Poverty: Program of Targeted Interventions. The project would address priority needs in the social sectors in accordance with a poverty targeting formula favoring the poorest municipalities. Beneficiaries: Low-income groups in both rural and urban areas targeted on the basis of access to basic social services. Amount: SDR 19.1 million (US $30.0 million equivalent) Terms: Standard IDA terms with 40 years maturity, including 10 years of grace. Commitment Fee: 0.50 percent on disbursed credit balances, beginning 60 days after signing, less any waiver. Financing Plan: See para. 4.5 Net Present Value: Not applicable. Environmental Classification: B Map: IBRD No.27058 Project Identification Number: 37709   HONDURAS THIRD SOCIAL INVESTMENT FUND PROJECT 1. ECONOMIC AND SOCIAL CONTEXT A. Economic Reform 1.1 Although Honduras experienced some economic progress over the last two decades, per capita GNP remains low at an estimated US $580 in 1993. One-third of the population of 5.5 million still lives in extreme poverty and 80 percent of the poor live in rural areas. With the level of poverty closely linked to average wages, particularly in rural areas, the country's economic performance is of primary importance to improving the poverty ratios. However, in recent years, this performance has fluctuated considerably. 1.2 Until recently, Honduras' development strategy was based on import substitution, protected by high tariff barriers and extensive Government intervention. This resulted in a large and inefficient public sector and serious structural imbalances in the economy. By the end of 1989, the parallel exchange rate was twice the official rate, external arrears amounted to US$620 million, and the country had been cut off from international creditors. Starting in 1990, the Government successfully implemented a stabilization and adjustment program, complemented by a series of sectoral reform programs. Combined with a tightening of fiscal controls in early- 1991, these reforms began to show positive results during the second half of 1991. In 1992, GDP growth increased, inflation fell to the lowest level since 1987, and the current account deficit improved. Arrears with the multilateral and bilateral institutions were cleared by mid-1990, after which point Honduras was declared eligible for IDA credits on the basis of its heavy indebtedness and low income level. In September 1991, Honduras was reclassified as an exclusively IDA country. 1.3 However, the promising economic trends of 1991-1992 were not sustained in 1993. As the elections approached, the fiscal deficit, which was targeted to fall to 3.8 percent of GDP in 1993, reached over 10 percent. Inflationary pressure began to build up, and foreign exchange reserves were depleted. The Government, which took office in January 1994, thus inherited serious fiscal and balance of payments problems, but is committed to the resumption of sustainable economic growth through the implementation of a medium-term program that aims to: (a) reduce the fiscal deficit; (b) reform the public sector by reducing its size and increasing its efficiency; (c) reform the judicial system; (d) restructure the social security system to eliminate the existing financial deficit; (e) privatize selected public enterprises; (f) reduce the number of tax exemptions, and (g) strengthen tax administration. At the same time, the Government intends to maintain the present level of social expenditures in the face of fiscal austerity, and continue its program of poverty alleviation and human resource development.  2 B. Poverty Reduction 1.4 The conclusions of a joint Bank-IDB Public Investment Review mission in November/December 1994 confirm that substantial social expenditures will continue to be required to help improve the country's poor social indicators, including an illiteracy rate for the adult population of 32 percent (reaching over 42 percent in rural areas), a total fertility rate of 4.5 per woman of child bearing age, a maternal mortality rate of 221 per 100,000 live births, an infant mortality rate of 45 per 1,000 live births, and malnutrition affecting 46 percent of children under five. (More detail is provided in Annex 1.) 1.5 Although Honduras has made considerable progress in improving education in recent decades, particularly at the primary school level, it still experiences major problems related to quality, efficiency, and equity, including overcrowded and dilapidated physical facilities and limited school access. More than one-third of the rural primary schools are incomplete, i.e., they cover only the first three or four grades of instruction. As a result, there are still 70,000 children with no access to full primary education, predominantly in rural areas of extreme poverty and geographic isolation. According to recent data from the Ministry of Education (MOE), there is a need, over the next four years, for about 3,200 additional classrooms, as well as for the repair and rehabilitation of about 11,800 existing classrooms, meaning a total investment of about US$61 million,' excluding the cost of furniture and supplies. 1.6 Health and nutrition indicators also reflect the acute problems still facing the poor: i.adequate diets lack of sanitation, deficient coverage of the primary health care system, ignorance about hygiene, correct nutritional practices, and preventive health care. Infant and matemal mortality are much higher in rural areas, and are subject to wide regional variations. For example, in three of the poorest regions of the country, maternal mortality is estimated at above 340 per 1 00,000. Nutritional deficiencies are a contributing factor in about 60 percent of infant deaths. By the time they enter primary school, the growth of 35 percent of children is stunted. Undernutrition is also a major health problem of pregnant and nursing women: over half the women attending health centers suffered from mild or moderate anemia and vitamin A deficiency. According to recent data from the Ministry of Health (MOH), primary health care still does not reach about 20 percent of the population. This would require, inter alia an investment in new health centers, as well as in repairing and rehabilitating existing centers, of about US$3 million. This amount, too, excludes furniture and supplies. 1.7 The provision of water and basic sanitation is highly inefficient and biased in favor of urban areas. Although about 70 percent of the population -- 90 percent of the population in urban areas and 54 percent in rural areas -- has access to a water supply system, treatment is deficient, and only 17 percent of the population has safe water. In About US$6.2 million for 290 classrooms is provided for under the Basic Education Project (see para. 1.10).  3 addition, those who have no access, mainly the poor, pay up to 77 times as much for water as those who have access. To correct this, and to provide access to an additional 300,000 families in rural areas alone, would cost at least US$27 million. In addition, there is a need for major investment in basic sanitation. Sewerage covers only 25 percent of the population, concentrated in urban areas. Waste removal in most of the country, covering about 37 percent of the population, is handled through latrines. For latrines alone, there is a need for more than 50,000 new and replacement ones yearly, estimated to cost about US$25 million over a four-year period. C. Government Strategy 1.8 The Government has a clear understanding of the social needs, and is assigning high priority to overcoming the country's education and health problems. However, although Government spending in education and health amounted to about 7 percent of GDP, corresponding to about one-third of central government spending over the period 1985 - 92, inefficiency ard inequity of spending have been ongoing problems. The Government is making an effort to strengthen the project planning and implementation capacity of the responsible ministries and agencies, but in the meantime, considers its social safety net programs of critical importance. The Honduran Social Investment Fund (FHIS) and the Family Assistance Program (PRAF) were both established in 1990, and targeted at the poorest and most vulnerable groups; FHIS to generate emergency employment and to build and rehabilitate small-scale infrastructure quickly and efficiently, and PRAF to provide food coupons to poor, pregnant and nursing women, children under five, and primary school children in departments with the highest malnutrition rates. Both programs have been quite successful, and have become the conduit for substantial donor support. Until the line ministries have overcome ongoing weaknesses in procurement, disbursement, civil service systems, and planning and budgeting, the Government wishes FHIS to complement the ministries' activities. FHIS will have the authority to manage the implementation of small and medium-scale subprojects, while the ministries will focus on policy and norm setting, as well as monitoring and evaluation responsibilities. D. Bank and IDA Support 1.9 The social sector has received, and will receive, considerable Bank and IDA support. MOE efforts to improve the efficiency of service delivery were assisted through the Administration of Rural Primary Education Management Project (Ln. 2804-HO in 1987; US$4.4 million), followed by the recently-approved Basic Education Project (Cr.2694-HO in 1995; US$30 million). The latter aims to improve primary school student academic achievement levels, to reduce high repetition rates, to provide for bilingual instruction in primary grades, and to improve sector management efficiency. In health and nutrition, the Nutrition and Health Project (Cr. 2452-HO in 1993; US $25 million) assists PRAF in improving the nutrition status of children and pregnant and nursing women among the poorest segments of the population, supports the development and  4 implementation of a longer-term nutrition assistance strategy, helps to improve the quality of services provided by MOH and to control the spread of AIDS, and assists in strengthening the institutional capacity of MOH and PRAF for sector planning, program formulation, monitoring and evaluation, as well as for the procurement of drugs. A proposed joint IDA-IDB Water Supply and Sanitation Sector Adjustment Project, expected to be appraised in mid-1995, would help improve the water supply and sanitation policy environment, and an ongoing sector study, funded by the Bank and UNDP, would provide much useful information on water and sanitation sector needs (see Annex 8). The recently-approved Environmental Development Project (Cr.2693-HO in 1995; US$10.8 million) aims to strengthen the capacity of the State Secretariat of Environment (SEDA) and key sector agencies in environmental and land use planning, policy development, and development of a sound legal and regulatory framework; to develop the capacity to carry out environmental impact assessments and audits, to monitor for environmental quality and environmentally sustainable resource use, and to enforce environmental laws and regulations; and finally to develop the capacity of selected municipalities and local conmmunities for environmental management, as well as to identify, design, and implement priority small-scale environmental investments. The implementing capacity of MOE and MOH would also be improved under the proposed joint IDA-IDB Public Sector Modernization Project, which is expected to be submitted to the Executive Directors in September 1995, and with the assistance of a recent IDF grant in the amount of US$198,000 for Institutional Development of Social Sector Ministries. 1.10 Under all of the above recently-approved projects, the civil works components are being managed by FHIS, including US$5.4 million under the Nutrition and Health Project, US$6.2 million under the Basic Education Project, and US$2.9 million under the Environmental Development Project. Under the Environmental Development Project, FHIS is also receiving US$0.7 million for incremental operating expenditures and the adrninistration of environmental subprojects. FHIS has been directly supported under the First and Second Social Investment Fund (SIF) Projects (Cr. 2212 - HO in 1991; US$20 million and Cr. 2401-HO in 1992; US$10.2 million), which both aimed to help protect and improve the standard of living of marginal social groups in both rural and urban areas through the provision of employment and improved services, mainly through the financing of labor-intensive social infrastructure subprojects. The Government has now asked that various donors, including IDA, IDB, and KfW, continue to assist FHIS in addressing poverty alleviation through priority subprojects at the municipal and community level. E. Rationale for ]IDA Involvement 1.11 Continued assistance to FHIS is consistent with the Bank Group's country assistance strategy for Honduras, which was discussed by the Board in August 1993 in conjunction with the presentation of the Agricultural Sector Adjustment Credit (P-5923- HO of July 14, 1993). This strategy aims at reducing poverty, modernizing the economy and the public sector, and prudently managing the country's natural resources. An OED Performance Audit Report (No. 13839, December 30, 1994) on the two IDA credits  5 supporting FHIS confirmed that FHIS has succeeded in producing a large number of small, widely-dispersed projects targeted to the poor. Further IDA involvement would be critical to help the Government mobilize additional resources for poverty alleviation to accompany its economic reform program. FHIS would serve as a powerful mechanism to mobilize and coordinate donor assistance for the social sectors. Equally important, the Country Economic Memorandum/Poverty Assessment (No. 13317 of November 17, 1994) stated that FHIS has had a significantly positive impact on the development of small-scale infrastructure, as well as upon improving the implementation capacity of small private construction firms, non-governmental organizations (NGOs), and municipalities, strengthening their institutional capacity, and laying the groundwork for decentralization. 2. HONDURAN SOCLAL INVESTMENT FUND A. Background (1) Objectives 2.1 FHIS was established in 1990 as a temporary agency, part of the Government's response to the potentially serious impact of the economic adjustment program on the country's poor. Through the financing of a large number of subprojects in social and economic infrastructure, social services, and informal sector activities, FHIS aimed to generate emergency employment, and to protect and improve the standard of living in poor communities. 2.2 Recognizing the success of FHIS as an institution that has become a key element in implementing the Government's poverty alleviation strategy, the Government recently extended its mandate to 12 years. Under this new mandate, it is expected that FHIS will continue to play an important role over the medium term as the implementing arm of the line ministries in the management of critical small and medium-scale social and economic infrastructure subprojects, as well as in the funding of priority programs in basic needs, until those ministries have overcome the weaknesses mentioned in para. 1.8. (2) Approach 2.3 Subprojects. FHIS finances subprojects in accordance with eligibility criteria and evaluation parameters established in its Operational Manuals, which have recently been updated, and which will be further revised under the proposed project. Investments have been financed in the areas outlined below, according to a standardized menu of subprojects:2 2 The menu and categorization of subprojects (particularly in the Basic Needs program) will be updated under the proposed project (see para 3.6).  6 (a) Social and Economic Infrastructure: This program includes (i) rehabilitation, maintenance, and expansion of preprimary, primary, and junior secondary schools, health centers, and water and sanitation systems, as well as maintenance of rural roads; (ii) simple works such as repair and construction of canals and ditches, and construction of bridges, gutters and sidewalks; and (iii) rehabilitation of archeological and historical sites aimed at the preservation of cultural heritage and ethnic minorities. (b) Basic Needs: This program provides funding for priority social programs such as nutrition, childcare, training of midwives and other health workers, care for the elderly and disabled, and school and health center furniture and teaching materials. (c) Micro Credit:. Under the Program of Support for the Informal Sector (Programa de Apoyo al Sector Inlformal - PASI), FHIS makes funds available to NGOs for on-lending at market rates to help establish and expand small and micro-enterprises in the productive sectors. Terms and conditions for financing of informal sector subprojects are spelled out in the Operational Manuals.3 2.4 Targeting. Several mechanisms have been put in place to ensure that subprojects are properly targeted to the poor. First, FHIS allocates its funds on the basis of a poverty map, ranking all 18 departments and 293 municipalities of the country based on population, and adjusted for levels of child malnutrition and access to potable water and basic sanitation. These indicators have been used to produce a departmental and municipal index of poverty and to assign additional resources to the poorest departments and municipalities. The formula is currently being revised, and, starting as of 1996, the formula will also include illiteracy, and be adjusted for differentials in the cost of materials across departments. (More detail is provided in Annex 2.) Second, FIUS' emphasis on social infrastructure and social sectors, and its concentration on subprojects with the highest social return (primary education, primary health care, and water and sanitation) also contribute to the proper targeting of funds toward the poor. Third, wages paid on FHIS-financed subprojects are aligned with the market rate for unskilled labor, to ensure participation of individuals in the lowest deciles of the income scale. Finally, FHIS makes special efforts to ensure that its targeting goals are attained, including the use of subproject eligibility criteria, promotion in poorer areas, and technical assistance to those who lack project preparation capacity. 2.5 Sponsors. As shown in Table 1, subproject financing can be requested by government ministries, municipalities, NGOs, and community groups. The subprojects are 3 A separate program funded by the European Development Fund (EDF), the Program of Credit and Technical Assistance for Rural Micro-enterprises (Programa de Credito y Asistencia Tecnica a la Microempresa Rural - PROCATMER), provides credit for rural microenterprises. PROCATMER also lends through NGOs, but operates independently of FHIS management and has a codirector appointed and paid by the EDF. PASI. too, is now being reorganized to operate independently.  7 executed by private sector contractors or NGOs, and supervised by FHIS or a third party with the necessary technical qualifications. Subject to the municipal allocations, F-US evaluates each subproject based on a site inspection, strict cost specifications, and social benefits in accordance with its Operational Manuals agreed upon with IDA. Table 1. Distribution of Number and Amount of Approved Subprojects by Type of Soliciting Agency (1990-94) Soliciting Agency | No. of % US$'000 % ______________________________________ l Subprojects I l Ministry of Education 188 3 3,813 3 Ministry of Health 121 2 3,439 3 Municipalities 3,133 54 48,177 40 NGOs 145 3 5,768 5 Other 2,206 38 60,573 50 HTOTAL 5,793 100 121,771 : 100 Source: FHIS, January 1995. 2.6 Institutional and Managerial Aspects. FHIS was created by law of February 22, 1990, with an initial life span of three years. On April 1, 1994, the FHIS mandate was extended for two more years, and on December 24, 1994, until the year 2002. In addition to stipulating FHIS' life span, the law specifies that FHIS: (a) is to be a financing, and not an executing, entity; (b) is subject to external audits; and (c) uses special streamlined procedures for public contracting. 2.7 The continued operation of FHIS has been declared a high priority by the President of the Republic. Clearly, FHIS has solid support at all levels of the Government, and is given priority by government agencies. FHIS has a Board of 10 members, chaired by the President of the Republic, and including top government and congressional officials and representatives of the private sector, cooperatives, and NGOs. It meets quarterly to: (a) approve general policies and rules of operation; (b) approve the annual budget (once a year); (c) authorize FHIS to negotiate loans; and (d) review FLIS' performance. Subproject approval is left to the operations committee, which is headed by the FHIS Executive Director and includes key FHIS department directors. The current Executive Director, who comes from the private sector, was appointed in early 1994, and has the rank of cabinet minister. 2.8 Following a recent organizational restructuring, FHIS is now organized into five main departments: Projects, Legal, Control and Supervision, Planning, and Administration. (See Annex 3 for the new organization chart.) A new Deputy Executive Director assists the Executive Director with day-to-day operations and line manages the support units, including the newly formed environment and community participation advisory units; she was Vice Minister of Planning, and will contribute to the effort to ensure that FHIS' activities meet the macro and sector priority criteria. As of March 31, 1995, FHIS had a total staff of 305, of whom 134 were professionals, including civil  8 engineers, economists, sociologists, credit specialists, computer engineers, and other technical specialists. Most department heads have graduate degrees, and their competence is comparable to the highest levels in the public sector. (3) Achievements 2.9 Subproject Processing. As of January 1,1995, FHIS had approved 5,793 subprojects for a total of about US$121.7 million. As of the same date, a total of 976 subprojects were in execution, with 4,817 completed. The average subproject size of about US$21,000 is comparable to the size of subprojects in other social investment funds in the region. On average, informal sector (credit) subprojects have been considerably larger. 2.10 Operational Efficiency. In general, FHIS has been able to process and approve subprojects rapidly and efficiently. Average time of subproject processing, from reception of a formulated project request to approval, has been about two months. FFHS-financed subprojects have also been implemented reasonably well. The average implementation period for FHIS subprojects has been about 6-9 months, depending upon the sector and type of project. 2.11 Total FHIS administrative costs have represented about 10 percent of total investments. This figure seems reasonable based on the number of subprojects processed and their average cost; it is also consistent with the administrative costs in other regional social investment funds. 2.12 Sectoral Accomplishments. FHIS has had a significant development impact by acting quickly and providing infrastructure of generally good quality. By January 1, 1995, FHIS had provided financing for the construction or refurbishing of 4,925 school classrooms (20 percent of all classrooms in the country), 372 health centers (41 percent of all health centers); and 84,246 latrines (34 percent of all latrines), and had provided access to drinking water and sanitation systems to 88,700 and 467,000 people, respectively. FHIS' basic needs program had provided 335,526 school desks (68 percent of the total), as well as training for 4,632 midwives (about 70 percent of the total). It is also expected that FHIS' efforts in basic sanitation subprojects will have a major long-term impact on health, especially on infant and child mortality. They have helped contain the propagation of cholera and lower the incidence of diarrhea, especially in the southern departments and in the densely populated poor areas in the outskirts of major cities. Because of its special procedures, higher quality of operations, promotion, technical assistance, and beneficiaries' involvement, FHIS has delivered quickly a much larger program than the line ministries could have done with the same budget, since they do not have the institutional capacity to plan, implement, and supervise a high volume of decentralized infrastructure subprojects. 2.13 As shown in Table 2 below, the largest share of FHIS financing went to social and economic infrastructure--75 percent of commitments from 1990 to January 1, 1995,  9 compared to a 67% target at appraisal of SIF 2. The vast majority of this has been in social infrastructure projects, including construction, expansion, and rehabilitation of primary and preprimary schools, health centers, and water and sewerage systems; this result confirms FHIS' commitment to supporting infrastructure projects with a high social return. Other infrastructure subprojects included the areas of drainage, road construction and repair, markets, and other community facilities. The basic needs program received 18 percent of FHIS financing--a share close to the appraisal target of 22 percent; the school desks program has accounted for about three-fifths of this total (see para. 2.12). Micro credit subprojects account for about 7 percent of total commitments, compared to an appraisal target of 11 percent. Table 2. Distribution of Resources for Approved Subprojects by Menu Category (percent) Program SIF I SIF I Actual Results I Appraisal Targets Appraisal Targets 1990-Jan. 1995 Social Infrastructure 43 54 65 Economic 27 13 10 Infrastructure Subtotal 70 67 75 Basic Needs 22 22 18 Micro Credit 8 11 7 Total 100 100 100,0 Source: FHIS, January 1995. 2.14 Employment Generated. Through 1994, FHIS subprojects generated approximately 689,000 person-months of temporary employment, equivalent to about 57,000 person-years, at an average cost per employee-year of US$2,120, a cost that is in line with other social funds in Latin America.4 Credit programs are estimated to have created about 24,000 additional person-years of employment of a more permanent nature. Although these figures are small as percentages of the total labor force, representing only about 1 percent on an annualized basis, they are more significant as percentages of open unemployment, representing 20 to 30 percent per year. 2.15 Institutional Impact. FHIS has had a positive impact on the institutional development of executing agencies. NGOs have improved their subproject appraisal methodologies and implementation strategies. FHIS subprojects have had a large impact on private contractors, some 900 of whom have worked on FHIS-financed subprojects to date. The number of private contractors has increased and several of them, who were 4 Figures regarding employment creation need to be interpreted with some caution. These data are derivative; they are based on calculations using subproject contractor payrolls and average execution tinme of projects.  10 individual contractors, have expanded into small firms. Another interesting development has been the improvement of the municipalities' capabilities in project preparation. Several of them, which were institutionally weak at the start of FHIS, have contracted qualified engineers who can put together good-quality project proposals. The poorest municipalities have received special assistance from FLUS in project preparation. FLIS is also starting to have an impact on line ministries, which recognize FLIS' comparative advantage in getting subprojects implemented. They are looking at the FHIS model, and considering ways to improve their methods of operation. B. Implementation Experience and Issues (1) Targeting 2.16 FHIS resources have been allocated to all 293 municipalities, which is consistent with its goal of achieving national recognition for the organization, and developing general political support for the economic reform program with which FHIS is associated. Since there are poor pockets even in wealthier municipalities, nationwide coverage is justified. 2.17 Experience in several social funds has shown that, if no targeting mechanism is used to allocate investments, a larger proportion of funds will be absorbed by wealthier municipalities, which have a greater capacity to prepare subproject proposals because of the larger number of private firms, engineers, and NGOs operating in these areas. The purpose of FHIS' targeting mechanism has been to ensure that funds would be more fairly distributed. To a significant extent, this was attained, although the goals set by FLITS were not reached. Table 3 presents a breakdown of FHIS' resource allocations from 1990 until January 1, 1995 to municipalities by poverty level and by year, along with a breakdown of allocation targets for the period as a whole. FHIS' achievements with respect to targeting have improved over time, with allocations to "very poor" municipalities increasing from 8 percent in 1990 to 20 percent in 1994, while allocations to "regular" municipalities shrank from 44 percent to 36 percent over the same period. Achievements were still well below FHIS' ambitious targets, however: poorer municipalities (i.e., the total of the three groups "very poor", "poor", and "deficient", representing 65 percent of the population) received 64 percent of total resource commitments, versus a target of 73 percent, while wealthier municipalities received 35 percent, versus a target of 25 percent.5 s Since some of FHIS' funds are specifically earmarked by donors for investment in wealthier municipalities, the results in table 3--which are based on the totals of all funds invested by FHIS each year--somewhat distort the results of FHIS' targeting mechanisms.  I1 Table 3. Investments Approved by Poverty Category of Municipality and Year 1990 - 1994 Investment % 1 Poverty % Total Cum. tTa CategoryaPp 1990 11991 1992 .1993 1994 Avg. agv a Very Poor 20 8 19 21 20 20 19 27 .73 Poor 21 12 17 25 24 19 22 24 .92 Deficient 24 28 21 24 22 14 22 23 .97 Subtota 65 47 57 70 66 53 64 73 .86 Regular 26 44 36 26 25 36 29 20 1.46 Acceptable 9 3 8 4 7 5 6 5 1.13 Subtota 35 47 43 30 32 41 35 25 1.39 Various 6 0 0 2 6 1 2 . , TOTAL 100 100 100 100 100 100 100 100 I Source: FHIS, January 1995. a Map at end of this report uses different nomenclature and combines "regular" and "acceptable". bAs of mid-year 1994. 2.18 FHIS targeting has actually been better than shown by the above numbers, because the current poverty mapping does not take into account the wide disparities in poverty levels within metropolitan areas. For instance, within Tegucigalpa--a municipality categorized as "regular"--an estimated 152,000 people (some 2.8 percent of the national population) live in conditions of poverty comparable to the poorest municipalities in the country. According to FHIS staff, FHIS investments within the Central District have been concentrated in these poorer zones; if investments within Tegucigalpa and other metropolitan areas were disaggregated by smaller geographical units--which is planned under the proposed project--the coverage index for the "very poor" category would be closerto 1.0. 2.19 Other factors contributing to a weaker than desired record on targeting are: (a) poorer municipalities generally have a lower capacity to prepare subprojects; (b) NGO- sponsored subprojects tend to be located in "regular" and "acceptable" municipalities, where there are better developed informal sectors; (c) promotion has been insufficient in poorer municipalities; and (d) subprojects are usually smaller in poorer areas. Also, the lack of a unified and reliable data bank regarding social needs (see para. 2.46) reduces the ability of FIUS to efficiently target its resources.  12 (2) Subproject Portfolio Quality, Sustainability, and Implications for FHIS' Institutional Capacity 2.20 Although FHIS' subproject quality is generally good, there are some problems related to implementation and sustainability, which got worse during the rapid expansion of subproject approvals in the election year 1993. FHIS has recently made significant progress in addressing these "problem" subprojects. Nevertheless, the remaining problems point to the need for FHIS to further improve its institutional capacity as it relates to a number of critical steps during the subproject cycle. The underlying issues are discussed briefly below. 2.21 Planning. Up until now, FHIS has responded to project requests more or less on a first-come, first-served basis, filling up its bi-annual sectoral and municipal allocations on an ad hoc basis. Under the proposed project, FHIS would establish a more systematic methodology for planning, implementing, and monitoring its work program, based on the banks of raw requests and of formulated projects already in the pipeline, and on estimates of when projects will enter the different departments. This planning system would enable FFHS to determine whether it is meeting its production goals, and to make appropriate adjustments regarding staffing and other inputs. 2.22 Promotion. The first steps to ensuring that there will be sustained interest in the implementation, operation, and maintenance of an investment are (a) the verification that subprojects are responsive to genuine local demand, i.e., on the part of beneficiaries, and (b) the involvement of beneficiaries early on in the identification and design stages of subproject development. Therefore, FHIS promotion staff should facilitate the identification of priority subprojects with the participation of communities, on the basis of a series of selection criteria. FHIS promoters do carry out these activities, which they call "direct" promotion; however, about 85 percent of the subprojects implemented since FHIS' creation have been identified and formulated on the basis of what FHIS refers to as "spontaneous" promotion, which is where FHIS staff are not directly involved in overseeing or facilitating subproject development. FHIS has now embarked on a major community participation effort (see Annex 4), and, in that context, is addressing both staffing and methodological deficiencies in its Promotion Unit to improve its capacity to carry out direct promotion and to address community participation issues in the early stages of subproject development. 2.23 Appraisal. In accordance with FHIS' Operational Manuals, the Evaluation Unit appraises formulated subprojects on the basis of a number of criteria, including community involvement, feasibility, social impact, and sustainability. However, the actual process of subproject evaluation could be improved. As is the case in the Promotion Unit, the staffing and methodologies of the Evaluation Unit will have to be improved to ensure that FHIS has the capacity to evaluate whether subprojects accurately respond to the community's demands, and that proposed solutions are appropriate, efficient, and sustainable.  13 2.24 Contracting. Despite the ongoing need for direct contracting for subprojects costing less than US$35,000 (see para. 4.8), FHIS will have to ensure that this mechanism is properly controlled. Direct contracting helps: (a) speed up implementation of the project cycle and stimulate a wide distribution of smaller projects; (b) compensate for the weak technical capacities of many municipalities to carry out feasibility studies; and (c) overcome the small size of the contracting sector as an obstacle to FHIS' program development. However, the incentives provided under direct contracting could contribute to a reduction of the role beneficiaries are to play in the processes of subproject identification and formulation. Thus, although direct contracting should continue to be a substantial procurement mechanism, FHIS needs to somewhat reduce its reliance on this method of procurement, and to strengthen safeguards against possible misuse. 2.25 FHIS also needs to ensure that its policies and practices regarding contractor guarantees and subproject advance payments provide proper incentives for contractors to respect their responsibilities in subproject implementation. During the 1993 run-up to the presidential elections, contrary to its own Operational Manuals, FHIS, on occasion, relaxed its own requirements for the application of bank guarantees to secure subproject progress in favor of the much less reliable promissory note, sometimes without requiring any cosigner. FHIS also, sometimes, increased the percentage of total project cost that could be paid in the form of a project advance. However, FHIS has recently corrected these practices, and started legal action against non-performance. 2.26 Supervision. The implementation problems mentioned above point most significantly to ongoing deficiencies in FH1S' capacity to supervise subproject execution. FHIS' original idea regarding supervision was that the subproject sponsor would have the direct responsibility for supervising the construction phase, with FIUS inspectors supervising these local supervisors on a semi-monthly basis. However, this system has proved difficult, since the majority of supervisors simply did not have sufficient technical skills to satisfactorily perform this function, and subproject sponsors did not have sufficient resources to pay these individuals. Moreover, FHIS' Tegucigalpa-based inspectors have been overwhelmed by the numerical and geographical explosion in the number of FHIS subprojects financed during 1993, and have been unable to visit some subprojects even once a month. In the past, FHIS hired additional inspectors, but that was not sufficient. Therefore, FHIS will now consider more innovative solutions than simply hiring additional staff for its Tegucigalpa headquarters, such as contracting subproject supervision out to regionally-based firms or individuals, to be supervised in turn by FHIS inspectors, and increasing the involvement of beneficiaries in supervision activities. FHIS will also improve the capacity of the Supervision Department to oversee the increased community participation components of subprojects, as well as the basic needs program, and include the cost of supervison in the subproject cost. 2.27 Ex-post Evaluation. Systematic ex-post evaluation of FHIS subprojects is important to ascertain whether the program is achieving the objectives for which it was designed, whether subproject sponsors are meeting their commitments to maintain and  14 operate the subprojects financed by the FHIS, and whether beneficiaries are actually making use of FHIS investments. Monitoring and evaluation also facilitate the drawing of lessons from the program for the improvement of planning. However, at present, FHIS does not have a coherent system of subproject impact evaluation. While some information regarding financial and physical aspects of subprojects is gathered in the course of supervision, and in the context of periodic external audits, this information is not yet sufficient to permit the evaluation of impact, subproject quality, utilization, maintenance, and the perceptions and levels of participation of beneficiaries. The proposed project would support corrective action in this area (see para. 2.39). Other Issues in the Subproject Portfolio 2.28 Basic Needs Program. This program aims to improve the living conditions of vulnerable groups, complement MOE and MOH actions through activities of direct benefit to vulnerable populations, and support social assistance subprojects executed by NGOs that target vulnerable groups. Priority attention is given to women and children, especially female-headed households, and indigenous groups. Although the program has achieved its goal of focusing on the county's most marginal groups--indeed, the country's 118 poorest municipalities received 100 percent of planned financing--the program lacks some definition. As mentioned above, three-fifths of FHIS' investment under this program has gone toward the provision of school desks, and much of the rest toward the training of midwives. New needs should be identified in close consultation with the social line agencies, NGOs, and communities. Also, FHIS' capacity to supervise the program will need to be improved. At present, supervision tends to parallel infrastructure subproject supervision, which focuses on the quantifiable, physical dimension of subprojects rather than their social dimension (e.g. community participation). FHIS is also weak in monitoring these subprojects' success in achieving their objectives. The basic needs program would be supported under the proposed project through a technical assistance component (see para. 3.6 and Annex 9). 2.29 Water and Sanitation. As mentioned in para. 1.7 above, the provision of water and basic sanitation in Honduras is highly inefficient, and biased in favor of urban areas. Although FHIS has the potential to make a significant impact in these sectors, to date, the water supply and sanitation component of FHIS' portfolio has been somewhat weak. Of 5,793 subprojects approved as of January 1995, only 8 percent were in the water supply and sanitation sector--including 3. 1 percent in construction of latrines. Smaller-scale water projects, particularly in rural areas, have been especially under-represented. Sustainability of water and sanitation subprojects has also been identified as an area needing improvement. These subsections would be supported under the proposed project through a technical assistance component (see para. 3.6 and Annex 8). 2.30 Environment. FHIS' activities with respect to the environment fall into two areas:  15 (a) For all subprojects, FHIS carries out environmental impact assessments. Its Operational Manuals stipulate that subprojects not adversely affect the environment and indigenous cultures. FHIS evaluators use a checklist of questions to undertake a short environmental impact assessment for each subproject. The main factors analyzed are the potentially negative effects that may result from: (i) location of infrastructure; (ii) generation of sediments; and (iii) alteration of waterways. With the help of technical assistance under the Environmental Development Project, FHIS is improving its environmental evaluation criteria (see Annex 5). (b) FHIS also finances subprojects in environmental infrastructure and prolectiont. To date, FHIS has financed relatively few subprojects in this area; such subprojects have accounted for only 0.5 percent of all investments (in forestation/reforestation, erosion protection, and irrigation). However, subject to the outcome of the technical assistance provided under the Environmental Development Project, FHIS intends to expand its activities in this sector, and has already established an Environmental Advisory unit to oversee its activities in the sector. (3) Management Information System 2.31 FHIS has developed a substantial microcomputer-based management information system (MIS), which has facilitated subproject administration and monitoring by enabling the staff to rapidly access data on the current status of any subproject, especially in discussions with sponsors. The MIS has also helped to improve subproject supervision and financial analysis, especially in the areas of cash flows and monitoring of administrative costs. However, FHIS has been slow to implement improvements in the system, necessary to provide integration between financial and operational modules. The proposed project would address this (see para. 2.38.e). (4) Collaboration with Line Ministries and Other Government Institutions 2.32 Since its inception, FIHS has actively promoted and achieved good working relations with line ministries and other government agencies, which has contributed to improved quality and sustainability in FHIS-financed subprojects. Specifically, FHIS has signed letters of understanding and inter-institutional agreements with different entities. The general objective of the agreements has been to guarantee the participation of sectoral institutions in the development of concrete subproject activities and provide subproject supervision and continuity. However, there is a need for stronger collaboration at the identification and completion stages, to ensure that only priority subprojects get carried out and that maintenance is provided for. This is even more important in view of the current decentralization effort, with more of the Government's responsibilities expected to be passed on to the municipalities. As part of the preparation and implementation of the  16 proposed project, considerable corrective action has been and will be taken (see para. 2.44). (5) Resource Mobilization and Needs 2.33 Since 1990, FHIS has mobilized financing in the amount of about US$121 million, of which US$99 million, or 82 percent, has come from the international community. Table 4 provides a breakdown of amounts and sources of financing. Total uncommitted funds available from all donors as of January 1, 1995 were nil, but a loan from OPEC in the amount of US$5 million has since become available. Table 4. Amounts and Sources of Financing by Year (in US$ million) Donor i 1990 1991 1992 1993 11994l Total Govermment of Honduras 1.9 2.1 8.0 6.5 3.1 21.6 IDA - 10.1 10.0 5.7 2.0 27.8 IDB - 3.7 19.4 5.5 28.5 Government of Germany (KfW) - 1.7 7.5 8.1 4.2 21.5 USAID 3.7 4.5 0.3 2.8 1.6 12.9 Others 1.3 1.0 2.1 1.3 2.6 8.3 Total 6.8 19.4 31.6 43.8 1j9.0 1026 Note: This amount does not include the contribution which FlS has received from subproject beneficiaries in the fonn of money, materials, land, labor, etc. Source: FlI-S, January 1995. 2.34 Total disbursements of FHIS since its creation until January 1, 1995 have been US$111.6 million. The first and only IDB loan in the amount of US$31.5 million is fully committed, and 95 percent disbursed. With regard to the IDA Credits, SIF I was closed on March 31, 1994, three months beyond the original schedule; the last disbursement was made on July 29, 1994. Under SIF II, US$9.0 million had been disbursed by May 4, 1995, and credit closing is expected by June 30, 1995, six months beyond the original target. Disbursements in 1994 were slower than expected, because commitments during that period were reduced. This was due primarily to uncertainties surrounding the administrative transition of FHIS, although these problems were exacerbated by Hurricane Gert, which hit Honduras in September 1993, and by the severe shortages of cement and electricity that developed in the last months of 1993. However, FHIS has used this slow down to correct some of the problems created in the election year, and to demonstrate its renewed ability to channel considerable amounts of resources into a solid subproject portfolio. As of April 28, 1995, FHIS had developed an active pipeline of 1,483 infrastructure projects and 279 basic needs projects, amounting to the equivalent of US$42.0 million and US$3.8 million, respectively (See Annex 6). In light of FHIS' proven implementation capacity, Honduras' continuing investment needs in the social sectors, and the Government's firm commitment to continue to provide adequate  17 counterpart funds, it is expected that FHIS could easily channel resources of US$100 million over the course of the next four years. C. Lessons Learned from Past Operations 2.35 There has been considerable experience with social investment funds in other countries and under the first two IDA Credits for FHIS. The lessons of this experience have been discussed in numerous World Bank and FHIS reviews and reports (see Annex 19). The principal lessons, and how they are incorporated in the proposed project, are described in this section. 2.36 Limited Roles and Objectives. Limiting social investment funds' roles to the tasks of contracting for the construction of social and economic infrastructure, related goods and services, and a limited range of social services has promoted professionalism and efficiency in the use of funds and prevented duplication of functions with the sector ministries. The proposed project will continue to limit the use of funds to specific social subsectors for well-defined subprojects. Although environmental subprojects may be added, this would only be supported after the successful implementation of pilot subprojects under the Environmental Development Project (see para. 3.4). 2.37 Government Commitment. An important aspect of FHIS' success in achieving its objectives was that it became an integral part of the Government's high priority economic and social policy; i.e., a program to cushion the social costs of the economic adjustment program. This policy framework provided FHIS with sustained, high level political support and the budgetary resources to start the program even before external financing could be mobilized. This, in turn, attracted donors from the international community. The new administration, which took office in early-1994, continues to view FlUS' work as a priority. Under the proposed project, the President of the Republic would continue to serve as the Chairman of the Board, and the Government has committed to the continued provision of adequate counterpart funding. 2.38 Operational Efficiency. Under SIF I and II, Fl-HS succeeded in processing and approving subprojects--and disbursing funds--relatively rapidly, while sustaining subproject quality reasonably well; furthennore, FHIS was able to keep administration costs within acceptable levels. The factors that seem to be most important in ensuring operational efficiency in social investment funds, and a description of the measures FIUS would take under the proposed project to maintain and improve its capacity in this area, are described below: (a) A high degree of autonomy. Freedom from normal government budgeting, procurement, and disbursement restrictions enables social investment funds to award contracts to private contractors simply and quickly, and makes the SIF's financial resources independent from the Government's annual budget cycle and cash management procedures,  18 which means that contractors can be paid in a timely manner. Under the proposed project, FHIS would continue to enjoy freedom from these restrictions. (b) Professionally competent, hard working, and well-motivated staff, particularly at senior levels. High quality management teams enable SIFs to be managed like private enterprises. In FHIS, efficient management is reflected in clearly defined products, large-scale production, quality control, accountability, the capacity to attract and retain quality staff, and an implementation strategy based on decentralization and privatization. The most critical factors in ensuring the quality of the staff have been good salaries, lack of tenure, carefully detailed operating procedures, and an executive director giving priority to technical competence and efficiency. FHIS would implement further improvements under the proposed project, which includes provisions for management training courses, and reforms in the system for recruiting and remunerating staff on the basis of technical competence, including the establishment of a personnel evaluation system. (c) A sound organizational structure. FHIS has recently simplified its organizational structure, and made it better suited to new strategic emphases, such as community participation and attention to the environment. This includes the establishment of the position of Deputy Executive Director, thus relieving the Executive Director of some of his day-to-day activities. Further changes could result from technical assistance to be provided under the proposed project, and from a proposed study of the future role of FHIS and its relationship with the line ministries, to be funded under the proposed Public Sector Modernization Project. (d) Efficient subproject processing mechanisms, including (a) the use of a standardized menu of subprojects, which contributes to efficient subproject implementation by: (i) limiting the range of specialists to be recruited by FHIS, (ii) providing standard parameters and specifications used by sponsors in subproject preparation, and (iii) facilitating appraisal and supervision of subprojects by FHIS staff; (b) comprehensive operational manuals with clear eligibility, appraisal, procurement, implementation, and supervision criteria; and (c) close supervision of subprojects during implementation to ensure quality control. To improve the efficiency of subproject processing, as well as to enhance the quality of subprojects, FHIS has already begun to address a number of methodological and staffing weaknesses in the units responsible for project promotion, appraisal, and supervision. The project would also provide technical assistance to improve management of the project cycle, and to further improve and simplify operational manuals.  19 (e) A well-designed management information system to facilitate automated subproject processing, and to enable line managers to obtain quick access to information necessary for decision-making. The project would provide technical assistance and equipment to help FHIS establish a better integrated MIS. 2.39 Monitoring and Evaluation. Monitoring and evaluation mechanisms provide feedback on operational processes as well as on subproject quality and impact, and can thereby enable SIFs to make institutional and methodological improvements during the course of program implementation. Under the proposed project, FHIS would receive technical assistance (para. 3.6) to improve its capacity to monitor program performance and to evaluate its subproject portfolio through: (a) adjustments in the methodologies used for ex-ante evaluation, in order to improve the basis upon which FHIS selects its subprojects for financing, and to better enable FHIS to conduct ex-post evaluation of subprojects; (b) elaboration of the indicators FHIS uses to provide information on targeting, operational efficiency, and subproject quality and impact; (c) compilation of semi-annual progress reports based on the monitoring indicators; and (d) carrying out of periodic in-depth evaluations of the impact of the FHIS program. 2.40 Transparency/Accountability. The insistence of external donors on the need for full transparency has contributed to generally proper use of funds, as well as reduced the temptation of political interference in FHIS operations. Transparency of operations has been an essential element in attracting external resources and projecting a good public image in Honduras. Under the proposed project, FHIS would continue its system of quarterly and annual audits, as well as strengthen its internal auditing and control systems. 2.41 Targeting. Targeting mechanisms in social investment funds are important to improve equitable distribution of resources, as well as to increase transparency in the allocation of resources. In general, FHIS has targeted its subprojects well. Furthermore, since the number of proposals has grown to exceed budgets for each municipality, and FlI-S plans to put greater emphasis on community participation in the formulation of projects, it is likely that targeting will improve in the future. Nonetheless, with the help of technical assistance, a number of measures have been, or would be, taken under the proposed project to improve FHIS' targeting. These include the following: (a) The process by which budget allocations are derived from the population data and the poverty index has been reviewed and improved. This includes the adjustment of formulas to take into account illiteracy and differing local costs of factors of production (i.e., labor, materials, equipment). (b) The indicators used to calculate the municipal poverty index would be updated and improved. This would include the incorporation of new information regarding existing social infrastructure in each municipality and likely beneficiaries of such infrastructure.  20 (c) Larger municipalities would be broken down into smaller "zones," which can be categorized and targeted appropriately. 2.42 Emphasis on sustainability. In Poverty Alleviation and Social Investment Funds: The Latin American Experience,6 the authors point out that subproject sustainability continues to pose problems for many social investment funds. As mentioned in para. 2.20 above, FHIS is no exception. Although FHIS' involvement in subproject processing technically ends with the completion of construction and the legal closing of a subproject account, there are a number of areas in which FHIS has a strong influence on the prospects for the sustained operation and maintenance of its subprojects. Indeed, FIlS has attempted to structure its subprojects to ensure the sustainability of investments and benefits. Elements of this emphasis on sustainability include: supporting subprojects with low recurrent costs; in cases of construction, requirements that line ministries create and budget necessary staff positions; provisions to ensure adequate cost recovery and maintenance of social infrastructure (including organization and training at the community level'); selection of basic needs subprojects sponsored by NGOs with proven track records; and expansion of already successful projects, with significant cost sharing requirements from sponsors, and in consultation with line ministries and municipalities. 2.43 Under the proposed project, FHIS would make additional efforts to improve the prospects for subproject sustainability. Probably the most important effort FHIS intends to make is to ensure that beneficiaries are involved in subproject development at an early stage. However, this increased emphasis on community participation should not end once the project is formulated but, rather, needs to be instilled in FIUS' operations at every stage along the subproject cycle, including through the construction and operation phases. To this end, FH-S has established a Community Participation advisory unit. FHIS also intends to increase its emphasis on components that are complementary to its infrastructure and social services subprojects, such as community training, and the development of user boards and committees. To assist this effort, KfW has provided a grant in the amount of DM650,000 to fund a Sustainability Pilot Project, to be carried out by a local NGO. Its results should be available in early-1996. FHIS would arrange a workshop with the participation of IDA, IDB and KfW following the completion of the various pilot projects to discuss all ongoing efforts to promote community participation, and to agree upon follow-up action. Under the proposed project, technical assistance would be provided for studying cost-sharing and other financial arrangements to improve the prospects for subproject maintenance and sustainability. Finally, increased monitoring and evaluation of post-implementation quality, sustainability, and impact may also provide valuable inputs regarding other ways FHIS can improve prospects for subproject sustainability. 6P.J. Glaessner, Kye Woo Lee, A.M. Sant'Anna. and J.J. de St. Antoine. 1994. PovertyAlleviation and Social Investment Funds: The Latin American Experience. World Bank Discussion Paper 261, Washington, D.C. For example, FHIS has developed training activities to accompany latrine construction, and introduced user comnittees for water and sanitation projects.  21 2.44 Collaboration with Social Ministries and Municipalities. Coordination with social sector ministries has been essential for the effective delivery of priority social services to beneficiaries, and will be facilitated through the institutional strengthening of MOE and MOH pursued under the Nutrition and Health, Basic Education, and Public Sector Modernization Projects. Such coordination, as well as coordination with the Honduran Association of Municipalities (AMHON), will grow increasingly important as FHIS enters the next, longer-term phase (see Annex 7). Under the proposed project, FLITS' good record in collaboration with these agencies would be maintained and improved through: (a) inclusion of the Ministers of Education and of Health on FHIS' Board of Directors; (b) renewed letters of understanding / inter-institutional agreements with relevant entities; (c) cooperation with line ministries and agencies in the establishment of broad targets for priority spending within each of the social subsectors (education, health, water and sanitation, etc.); and (d) involvement of the line agencies in the annual discussions between the Government, FHIS, and the donors on FHIS' work program, budget, and recurrent costs (see para. 5.3). New agreements between FL-US and MOE and MOH. as well as a model agreement between FHIS and the municipalities. including assurances regarding operations and maintenance, were discussed and agreed during negotiations, and their incorporation in the Subsidiary Agreement would be a condition of effectiveness of the credit (see paras. 4.6 and 7.2(a)).8 2.45 One aspect of the coordination with social sector agencies would be the establishment of broad targets for priority spending within each of the social subsectors (education, health, water and sanitation, etc.). These broad targets would not be intended to diminish the capacity of FHIS to respond to local demand for priority subprojects. However, in light of the increasingly important role FHIS has come to play in the national poverty alleviation strategy, and in light of social spending limits established in the context of the macroeconomic reform program, these targets will enable the line ministries and FHIS to set parameters--based on national priorities--within which local demand can more rationally be addressed. 2.46 As a necessary step to set these parameters, however, Honduras would have to develop a unified and reliable data bank regarding national social needs. At present, bits and pieces of information regarding population and social infrastructure can be found in various institutions and agencies in Honduras. Yet, in order for the Government to establish national priorities, and for FHIS to effectively and efficiently target its investments (see also para. 2.41), these data would have to be collated and analyzed systematically. This will be supported under the proposed project through a technical assistance component (see para. 3.7 and Annex 11). 8 Collaboration with the municipalities would also be helped by the proposed technical assistance for decentralization and municipal development, financed by RUTA Social and coordinated by SECPLAN, FHIS, and the Presidential Comrnksion for Modernization of the State, as well as by the local capacity building technical assistance component included in the project (see para. 3.6).  22 3. THE PROJECT A. Project Objectives and Description 3.1 The proposed project would continue to assist the Government in strengthening its poverty alleviation efforts and in maintaining social cohesion, while the Government attempts to regain fiscal balance and the line ministries strengthen their institutional capacities. It would also support the Government's decentralization strategy, development of the local contracting industry, the sustainability of subprojects, and the targeting of scarce resources to the poorer areas. To achieve this, it would focus on meeting priority social and economic infrastructure and basic needs, decided in close consultation with relevant government departments, municipalities, NGOs, and communities, within the parameters of municipal allocations decided on the basis of a poverty targeting formula. 3.2 The project would support FHIS' investment program for a four-year period, as well as efforts to establish a better social data base and to assist municipalities on environmental matters. It would provide for (a) a range of small-scale, social and economic infrastructure subprojects; (b) a priority basic needs program; (c) a comprehensive program of technical assistance to help FHIS overcome some institutional weaknesses; (d) the establishment of a social data mapping system and strengthening of the design and implementation of the national household survey as the major tools to target the poor and to monitor progress of social development programs; and (e) subject to trials carried out by FHIS under the Environmental Development Project, a small program of technical assistance to municipalities in assessing environmental issues and in preparing programs for corrective action. B. Project Components 3.3 The project would have the following investment, technical assistance, and project management components: 3.4 Infrastructure Subprojects (US$95 million or 84 percent of total project costs). Subprojects sponsored by municipalities, community organizations, and NGOs, dealing with (i) rehabilitation, maintenance, and expansion of preprimary, primary, and junior secondary schools, health centers and posts, and small water and sewerage facilities, as well as maintenance of rural roads; (ii) construction and repair of drainage and irrigation canals and ditches, public markets and other public facilities, and construction of bridges, gutters, and sidewalks; (iii) rehabilitation of archeological and historical sites aimed at preserving cultural heritage, as well as programs focused on ethnic minorities; and (iv) subject to the outcome of the pilot subprojects carried out under the Environmental Development Project in 20 municipalities over the next year, small-scale environmental investments, such as urban and community parks, solid and liquid waste management, protection of water sources, soil protection and erosion control, biodiversity, and energy  23 generation and conservation. These subprojects would be executed by private contractors, NGOs, municipalities, and public sector agencies. A more detailed description of the water supply, sanitation, and environmental projects is given in Annex 8. 3.5 Basic Needs Programs (US$5 million or 4 percent of total project costs). Programs dealing with nutrition, training of midwives and other health workers, childcare and early development programs, care for the elderly and disabled, teaching materials, and community development for ethnic minorities.9 They would be executed by NGOs and private and public sector agencies, and require significant promotional efforts from FHIS. The programs would address priority social needs, identified and designed in consultation with MOE and MOH. A more detailed description of this program is given in Annex 9. 3.6 Institutional Strengthening of FHIS (US$0.8 million or 1 percent of total project costs). This component would include: (i) 16 months of technical assistance to help strengthen FHIS' policies and practices for management of subproject cycle / production line (planning, identifying, appraising, contracting, supervising, and evaluating Subprojects); this would include, inter alia, analysis and recommendations on organizational structure, manuals, staffing, and training requirements; (ii) training programs for FHIS' management and staff, (iii) four months of technical assistance to help improve FHIS' systems for recruitment, training, evaluation, and remuneration of personnel; (iv) six months of technical assistance to help improve the MIS system; (v) two months of technical assistance to help strengthen FHIST' basic needs program; (vi) two months of technical assistance to help promote community participation in all stages of the subproject cycle; (vii) a promotional training program to make US Peace Corps volunteers and their Honduran counterparts aware of how to access FHIS funding for priority community development subprojects; (viii) six months of technical assistance to improve FHIS' accounting and internal auditing practices; (ix) six months of technical assistance for analysis of financing mechanisms for maintenance of subprojects; (x) 12 months of technical assistance to help improve FHIS' water supply and sanitation policies and practices; and (xi) workshops and materials organized by the Program of Support to the Informal Sector (PASI) to assist its micro-enterprise program in general and female entrepreneurs in particular. 3.7 Social Data Collection and Analysis (US$0.1 million or 0.1 percent of total project costs). Because Honduras has no single information tool with which to assess overall deficiencies in coverage of the social sector needs, this component would assist with creating a social data mapping system and improving the household survey sample, questionnaire, and execution. With input from the line ministries, coordinated through the State Secretariat of Planning (SECPLAN), the mapping system would allow FHIS and the responsible line ministries to better assess priority needs in given communities, and to evaluate progress over time. In addition, to obtain more useful data on poverty and social indicators, there is a need to update and improve the household survey sample, redesign the questionnaire, and improve field work practices. A modest program of technical 9School and health center furniture, which previously were part of the Basic Needs prograin, would be moved to the Infrastructure subprojects category.  24 assistance, managed through SECPLAN, in consultation with FHIS, would be provided under the project. A more detailed description is given in Annex 11. 3.8 Environmental Assistance to Municipalities (US$0.9 million or I percent of total project costs). Subject to the outcome of the technical assistance to municipal environmental management in 20 municipalities under the Environmental Development Project, the proposed project would include a modest environmental awareness and training program for municipalities not covered under the earlier assistance. This component should help the municipalities to improve management of their natural resources and to address their environmental problems, as well as to increase public participation in the solution of the environmental problems of the communities. It would be managed by SEDA, in close cooperation with FHIS. A more detailed description is given in Annex 12. 3.9 Project Management (US$10.8 million or 10 percent of total project costs). Project administration during the four-year implementation period is expected to cost the equivalent of US$9.5 million. Other subcomponents would be the procurement of some vehicles and equipment, and the services of the external auditor. A more detailed description is given in Annex 13.  25 4. PROJECT COST, PROCUREMENT, DISBURSEMENT AND AUDITS A. Project Cost and Financing 4.1 Project Cost. The total project cost, including taxes and duties, is estimated at about US$1 12.6 million equivalent, of which an estimated US$22.9 million would be in foreign exchange (20 percent of project cost); taxes and duties amount to about US$4.4 million. The table below summarizes the estimated cost by component. Investments amount to 88 percent of total costs, technical assistance to 2 percent, and management to 10 percent. Project costs were estimated at February 1995 price levels. Since the project is essentially a financial intermediary type operation, and physical targets are not the main objectives of the proposed project, no contingencies have been provided for the subprojects. Any increase in the average subproject cost would be reflected in a reduced number of subprojects executed. Since most technical assistance would be provided in the first project year, no price contingencies have been provided for this category either. However, for project administration, vehicles and equipment, and the external auditor, price contingencies were calculated at 1.8 percent for 1996, 2.4 percent for 1997, and 2.5 percent annually thereafter, according to the Bank Group's estimated Index of Unit Value of Manufactured Exports, October 1994. These rates have been applied to local and foreign costs, since Honduras has adopted a floating exchange rate system which results in periodic market adjustments, that tend to compensate for the difference between international inflation on a US dollar base and local inflation rates. No costs for environmental subprojects have been included, since these will depend on the results of the pilot subprojects funded under the Environmental Development Project, which are expected in about one year (para. 4.14). However, subject to the outcome of these pilot projects, environmental subprojects would be eligible for funding under the credit. In that case, the allocation to other sectors would be reduced. Also, except for a modest amount of technical assistance to PASI, no costs have been included for FHIS' special operations, such as the micro credit schemes under PASI and PROCATMER; these two operations have sufficient funding for credit at present.  26 TABLE 5. Project Cost Summary (USS 000) |__ _ __ Local Foreign: | Total A. FlIS Portfolio 1. Infrastructure Subprojects 74,250.0 20,750.0 95,000.0 2. Basic Needs Programs 4,250.0 750.0 5,000.0 Subtotal 78,500.0 21,500.0 100,000.0Q B. histitutional Strengthening of FHIS 1. Basic Needs, Community Participation and Cooperation with US Peace Corps 57.0 43.0 100.0 2. Personnel and Management Training, MIS, 105.0 349.0 455.0 Subproject Cycle, Accounting, and Maintenance 3. Water Supply and Sanitation 30.0 123.0 153.0 4. Micro Credit (PASI) 89.0 - 89.0 Subtotal 281.0 515.0 797.0 C. Project Management 1. Administration 9,500.0 - 9,500.0 2. Vehicles and Equipment 230.0 689.0 918.0 3. External Auditor 250.0 - 250.0 Subtotal 9,980.0 689.0 10,669.0 Total FHIS 87,047.0 24,560.0 111,607.0 D. Social Sector Data Mapping and Household Survey 96.0 49.0 145.0 E. Envirorunental Assistance to Municipalities 860.0 30.0 890.0 Grand Total 89,753.0 22,889.0 112,642.0 a Includes taxes and duties estimated at about US$4.4 million. 4.2 The total cost projections of the FHIS subprojects reflect both the demand in the various social subsectors, described in paras. 1.5 - 1.7, and the country's limited budget prospects. Estimated construction costs for classrooms and health centers are based on unit prices per square meter, provided by FHIS and derived from current contracts and unit costs for similar standards of construction. Estimated school and health center rehabilitation and repair costs are based on unit prices, provided by FHIS and derived from current contracts. Estimated construction costs for water supply, sanitation, and other facilities, as well as the estimated average basic needs subproject size, too, were provided by FHIS, and derived from current contracts. There is also a provision for supervision at 3-6 percent and, for preinvestment studies at up to 7.25 percent of total subproject cost, depending upon the type of subproject. Estimates for technical assistance are based on current rates for local and foreign experts, and estimates for operating costs, too, are based on current costs of FHIS. Annex 13 provides detailed cost tables. 4.3 Incremental Recurrent Costs. As detailed in Annexes I and 13, over the period 1995-99, there is a need for the construction/reconstruction of about 2,900 classrooms, the rehabilitation/repair of about I11,800 classrooms, the rehabilitation/repair of about 2,600 school bathrooms, and the rehabilitation/repair of about 6,300 school latrines, as well as some minor work on health centers. If fully addressed by FHIS, this would mean incremental recurrent expenditures for operation and maintenance of about US$0.6 million  27 in the second project year, US$1.6 million in the third year, US$2.4 million in the fourth year, and US$3.0 million thereafter. However, FHIS' contracting capacity is limited, and the incremental recurrent expenditures can be expected to be much less. Furthermore, additional teachers for new classrooms would require about US$0.6 million in the second year, US$1.8 million in the third year, US$3.2 million in the fourth year, and US$4.5 million thereafter. Based on the 1994 MOE budget of about US$141 million and projected budget increments of 4 percent per year during the project period, and even assuming that all incremental costs would have to be borne by MOE, the impact of these costs on the MOE budget would be 0.2 percent in 1996, 0.8 percent in 1997, 1.7 percent in 1998, 2.4 percent in 1999, and 2.7 percent in 2000, thus remaining manageable. 4.4 MOE and MOH are responsible for operation and maintenance costs in their respective sectors. However, in the context of the Government's decentralization drive, the Government has decided that the municipalities should start to bear at least part of these costs. Consequently, the model agreements to be signed between FHIS and the municipalities (see para. 2.44) provide for the municipalities to pay 60 percent and the beneficiaries 40 percent of the maintenance cost of completed education and health subprojects. To help specify the operational arrangements among the various actors, the project will provide technical assistance, and a proposal on the improved funding of maintenance costs will be submitted to IDA by the Government no later than March 31. 1996 (para. 7.1 (a)). Teacher salaries are the responsibility of MOE, and this would be reflected in the new agreement between FHIS and MOE (see para. 2.44). The cost of operation and maintenance of water and sanitation subprojects would be borne by the beneficiaries, which reconfirms the importance of establishing a water user board already at the time of project identification. The early establishment of "ownership" is of critical importance in ensuring beneficiary contributions. 4.5 Project Financing. The proposed project would be financed by IDA, IDB, the Government of the Federal Republic of Germany through KfW, various other donors, and the Government of Honduras. In addition, subproject beneficiaries would contribute in the form of cash, labor, materials, or land. The proposed IDA credit of US$30.0 million equivalent (SDR 19.1 million) would finance 27 percent of total project costs, net of taxes and duties; it would finance 30 percent of foreign exchange expenditures and 26 percent of local expenditures. The IDB loan of US$40.0 million, on soft terms, and grant of US$619,900 would finance 36 percent of total project costs, net of taxes and duties. The KfW soft loan of about US$ 10.0 million equivalent (DM 15.0 million) would finance 9 percent of total project costs, net of taxes and duties. The amount of local cost financing by IDA is in line with other social projects elsewhere in Latin America. The IDA credit would be made to the Repulblic of Hotnduiras, which would make the funds available to FHIS and SEDA for the execution of the project. The Government of Honduras would finance US$11.2 million equivalent, or 10 percent of total project costs, including taxes and duties estimated at US$440,000 equivalent. The proposed financing plan is summarized in the table below, and expanded in Annex 14.  28 Table 6. Financing Plan (US$ million) ____________________ Local | Foreign i TotalJ Government of Honduras 11.2 0.0 11.2 IDA 23.2 6.8 30.0 IDB 33.1 7.5 40.6a Government of Federal 6.7 3.3 10.0 Republic of Germany (KfW) OPEC 3.3 1.7 5.0 Other Donors 7.1 3.5 10.6 Subtotal 82.9 24.6 107.6 Beneficiaries b 5.0 0.0 5.0 fTota I 89.7 22.9 112.6 a Includes USS619,900 million in grants for technical assistance b Contributiom in cash labor, land and/or materials. 4.6 During negotiations. agreement was reached that the Govemment would provide annual budget allocations for the project. as well as ensure the funding of the recurrent costs of completed FHIS investments, and that these allocations would be reviewed with the Bank. IDB. and KfW during annual project implementation reviews (paras.5.3 and 7.1 (b)). During negotiations, the Government also provided assurances that financing of US$40.0 million would be available from IDB. DM 15.0 million from KfW.and US$15.6 million from other donors, and that the respective funding agreements would be signed b March 31. 1996:, if not. alternative financing would have been arranged (para. 7. l(c)). As a condition of effectiveness, a FHIS Agreement for making the IDA funds available to FHIS. as well as describing the new arrangements between FHIS and MOE and MOH. satisfactory to IDA. would have been entered into by the Government and FHIS (para. 7.2(b)).  29 B. Procurement 4.7 A Country Procurement Assessment Report for Honduras was issued in September 1990. Honduran procurement legislation and procedures are generally satisfactory, except where participation of foreign contractors and consultants are concerned, due to the preference provisions for local contractors and consultants. However, at its establishment, FHIS was exempt from the Government's standard, complex procurement procedures. FHIS' special procurement procedures have proved successful in expediting project implementation and in ensuring competitive prices, and would thus continue to be applied under the proposed project. Procurement of civil works and goods would be done following the Guidelines for Procurement under IBRD Loans and IDA Credits of January 1995 (Procurement Guidelines). The hiring of consultants would be done following the Guidelines for Use of Consultants by World Bank Borrowers and by the World Bank of August 1981 (Consultant Guidelines). For international competitive bidding (ICB), FIlS would use the standard bidding documents (SBDs) issued by IDA, with such modifications as agreed to by IDA to be necessary for the purpose of the project. For purposes of bid evaluation under ICB, a margin of preference for local manufacturers may be applied, in accordance with IDA guidelines. For national competitive bidding (NCB), SBDs, satisfactory to IDA, would be used. Table 7 shows estimated project costs by procurement methods. Table 7. Procurement Methods by Category' (US$ million) ._ Category Procurement Method . - ICB NCB Otber - NoI[A -'-Od .FhlGlCed: FIRS Subprojects Civil Works 4.2 20.0 61.3 85.5 (4.0) (19.2) (23.2) Goods 1.4 1.4 6.7 9.5 (1.3) (1.3) (2.6) Services - 2.0 3.0 5.0 (2.0) (2.0) Vehicles and Equipment for 0.7 0.4 - 1.1 FHIS (0.5) (0.3) (0.8) Consultants and Auditor 0.4 0.6 1.0 Services for FHIS (0.4) (0.4) Training and Learning 1.0 0.1 1.1 Materials (1.0) (1.0) Salaries and Operating Costsc - 9.5 9.5 Total 0.7 5.6 25.2 81.2 112;6 (0.5) (5.3) (24.2) (3 a Figures in parentheses are amounts estimated to be financed by IDA. b Non-ICB/LCB are aggregated as Other, including local shopping and direct contracting and hiring of consultants. c Procurement methods not applicable. d Figures may not add due to rounding.  30 4.8 Civil Works. Because of the small size of the civil works in dispersed locations (see para. 2.24) to be funded under the Credit, there will be no ICB for the civil works costing approximately US$24.2 million equivalent. NCB would be used for contracts between US$50,000 and US$500,000 equivalent. To the extent possible, similar types of works in the subprojects would be combined in bid packages of at least US$50,000 equivalent, to minimize administrative costs and take advantage of contractor specialization. Works estimated to cost between US$35,000 and US$50,000 equivalent, up to an aggregate amount of US$6.7 million equivalent, would be procured under lump sum fixed price contracts awarded on the basis of quotations obtained from a least three qualified domestic contractors in response to a written invitation. To encourage competition and increase interest by contractors whenever feasible, contracts for repairs and rehabilitation would be grouped into contract packages expected to cost no less than US$35,000 equivalent. Procurement of contracts of works and goods, or a combination thereof, in small subprojects with a cost less than US$35,000 equivalent, up to an aggregate amount of US$13.3 million equivalent, would be done through direct contracting of local contractors where competitive proposals cannot be obtained. In these cases, to ensure fair pricing, FHIS has a well-tried system of regional, standard unit prices, which are being updated periodically, to be used as the basis for negotiations with the contractors. During negotiations. FHIS provided assurances that it would continue to periodically review its standard regionalized unit prices of civil works and goods. and change prices promptly. in a form and substance satisfactory to IDA. if so recommended by such a review (para. 7. 1 (d)). In addition, FHIS agreed to gradually reduce the share of direct contracting from the historical average of 75 percent of the amount of contracted subprojects to no more than 40 percent in 1998. Monitoring procedures would ensure that: (i) contracts are not divided into smaller packages for the purpose of awarding smaller contracts under direct contracting; and (ii) there is no unjustified concentration of contracts awarded to the same individuals or contractors. 4.9 Goods. Procurement of vehicles, equipment, furniture and other goods would be done through procurement procedures satisfactory to IDA. Vehicles and equipment, estimated to cost US$ 1.1 million equivalent, would be procured through ICB for bid packages of US$150,000 equivalent or more, through limited international bidding (LIB) for bid packages of US$50,000 to US$150,000 equivalent, in an aggregate amount not to exceed US$0.2 million equivalent and through international shopping for bid packages of US$25,000 to US$50,000 equivalent, in an aggregate amount not to exceed US$0.1 million equivalent. Contracts for equipment, valued at less than US$25,000 equivalent, would be awarded on the basis of a comparison of price quotations solicited from at least three suppliers, in an aggregate amount not to exceed US$0.1 million equivalent. School and health center furniture and other goods, estimated to cost US$2.8 million equivalent, would be procured through NCB for contracts between US$25,000 and US$150,000 equivalent, up to an aggregate amount of US$1.4 million, and through national shopping (based on quotations from at least three eligible suppliers) for contracts of less than US$25,000 equivalent (up to an aggregate amount of US$0.9 million equivalent). In cases where it would not be possible to obtain competitive proposals for furniture,  31 contracts estimated to cost less than US$8,000 equivalent, and up to an aggregate amount of US$0.5 million equivalent, may be awarded through direct contracting. 4.10 Technical Assistance. Individual consultants and consulting firms that would carry out technical assistance, amounting to about US$0.4 million, would be hired following the Consultant Guidelines. During appraisal, the time tables for technical assistance under the proposed project were reviewed, and found satisfactory. The appointment of consultants would be done using standard contract documents for larger contracts, and standard documents, to be agreed with IDA, for smaller assignments. 4.11 Prior Review. Prior review by IDA would include: (a) procurement arrangements in accordance with Appendix 1, paragraph I of the IDA Guidelines; (b) all procurement by ICB and LIB; (c) the first two NCB contracts for each works and goods, and all works contracts equal or above US$150,000 equivalent, with respect to subprojects; and (d) all amendments to works contracts under direct contracting raising the contract value above US$35,000 equivalent and under contracts based on quotations obtained from at least three domestic contractors raising the contract value about US$50,000 equivalent. Prior review by IDA of the selection and hiring of consultants would not be required for (i) individual consultants when the contract is valued at less than US$25,000 equivalent, and (ii) consulting firms when the contract is valued at less than US$50,000 equivalent. This exception does not apply to prior review of terms of reference and single source selection. The level of IDA prior review of procurement is relatively low, but is considered acceptable in light of the audit arrangements for the project, which include quarterly and annual auditing by a qualified external firm, including physical audits of subprojects. Periodic project physical performance evaluations would also look at procurement issues in the course of random review of subprojects, including field visits. In addition, prior review by IDA will be supported by ex-post review of procurement documentation during regular and frequent supervision missions (Annex 18). C. Disbursements 4.12 The proposed project would be implemented over a four-year period, and is expected to be completed by June 30, 1999 and closed by December 31, 1999. The proceeds of the credit would be disbursed over a 4 1/2 year period as follows: (a) subprojects, 96 percent of disbursements by FHIS approved on or before March 31, 1999, according to FHIS' project selection and evaluation procedures; (b) vehicles and equipment: 100 percent of foreign expenditures for imported goods, 100 percent of total expenditures for locally-produced goods (ex-factory cost), and 93 percent of other local expenditures; (c) technical assistance: 100 percent of the cost of eligible training, workshops, learning materials, and consultant services, and (d) external auditors and special independent reviews for FlUS: 100 percent. Retroactive financing of up to US$3.0 million equivalent would be permitted for eligible expenditures under subprojects incurred as of January 1, 1995. This is justified since the second IDA credit was fully committed as of the end of 1994, and also other donor funding in early-1995 has been in short supply.  32 4.13 Proceeds of the proposed credit would be disbursed against expenditures that would be presented to IDA and fully documented for civil works and goods contracts of US$150,000 or more and for consultant contracts of US$25,000 or more for individuals and US$50,000 or more for firms, subject to prior review (para. 4.11). Claims for expenditures under contracts below these amounts would be disbursed against presentation of Statements of Expenditure (SOEs), for which the supporting documentation would be retained by the relevant implementing agencies - FHIS and SEDA - for periodic inspection by IDA and by external auditors. In order to facilitate project implementation, the Government would establish a special account in US Dollars at the Central Bank with an authorized allocation of US$ 2.0 million, representing the average of four months of expenditures that are expected to be made from the account. The possible need for a separate Special Account for SEDA, to help implement the environmental technical assistance component, would be reviewed when the condition for this component has been met (para. 4.14(i)). Withdrawals from the Special Account would be solely for eligible expenditures under the credit, supported by the required documentation, and at the exchange rate prevailing on the date of withdrawal. The minimum amount for withdrawal applications submitted outside the Special Account would be US$400,000 (20 percent of the authorized allocation). 4.14 During negotiations. agreement was reached that no disbursements would be made under the credit against any environmental technical assistance managed by SEDA until: (i) the technical assistance provided under the Environmental Development Project was being implemented successfully, and (ii) detailed terms of reference (including design of methodologies and materials) had been drafted, and were acceptable to IDA. Similarly, no disbursements would be made against any environmental subprojects until the environmental subprojects carried out by FHIS on a pilot basis in 20 municipalities under the Environmental Development Project were judged to be progressing well (paras. 7.3(a and (b)). Also, no disbursements would be made against food expenditures (unless the relevant IDA policy were to be changed) (para. 7.3 (c)). D. Accounts and Audits 4.15 FHIS and SEDA would maintain records and separate accounts adequate to reflect, in accordance with sound accounting practices, the resources and expenditures of the project. During negotiations. agreement was reached that FHIS and SEDA would: (a) have the records and accounts of the project for each fiscal year. including the Special Account, audited by independent and qualified external auditors, in accordance with generally accepted auditing standards and procedures. and terms of reference satisfactory to IDA: (b) furnish IDA, no later than four months after the end of each year. a certified copy of the agreed audit reports: and (c) furnish IDA such other project financing information as IDA would from time to time reasonably request (para. 7. I(e)). For all expenditures with respect to which withdrawals from the Credit Account were made on the basis of Statements of Expenditures (SOEs), and for all expenditures with respect to  33 which withdrawals were made from the Special Account, FHIS and SEDA would provide access to the independent auditors to all records, accounts, and the supporting documentation necessary to allow the timely performance of their audit obligations. The audit reports would include separate opinions on the eligibility of expenditures included in the SOEs and on the accuracy and reliability of the SOEs themselves. 4.16 FHIS' intemal auditing unit is currently being strengthened through staff training; additional technical assistance is provided for under the project. With a stronger unit, cooperating closely with the external auditor, the role of the external auditor could probably be reduced over time. However, during negotiations. it was agreed that, for the time being. FHIS' extemal auditor would continue to also carrv out quarterly performance audits, combining a financial review with physical inspection of works in the field (para. 7.1 (f)). Quarterly, the auditor would examine a sample of subprojects, review the standard regionalized unit prices, review the individual contracts, ascertain whether the terms of these contracts are complied with, inspect the progress of works, reconcile the physical progress with the financial expenses incurred, and state whether procurement procedures specified in the Credit Agreement have been properly followed. The auditor would prepare quarterly reports for the Board of FHIS and for IDA (to be submitted within thirty days after each quarter) and an annual summary for IDA, to be submitted together with the audit report on the project financial statements.  34 5. PROJECT MANAGEMENT AND SUPERVISION A. Project Implementation 5.1 Project Preparation. The project was prepared with technical support from the Regional Unit for Technical Assistance in the Social Sectors (RUTA Social) and financing from a Government of Japan grant. The FHIS components were designed by EFIIS with the help of consultants, the social data component by a group of SECPLAN, FHIS, and statistical officials under the coordination of the Vice-Minister of SECPLAN, and the environmental component by SEDA with the help of consultants and in consultation with FHIS. High-level MOF, SECPLAN, MOE, and MOH officials were consulted at all stages of the project cycle, ensuring that the Government's strategies for the sectors are reflected in the project design. The experience gained by FHIS from the implementation of the earlier IDA, IDB, and KfW-assisted projects and by SEDA in helping to prepare the Environmental Development Project were instrumental in helping to prepare their respective components. Similarly, the experience of SECPLAN and USAID helped to design the social data component. The findings and recommendations of the Public Investment Review mission in November/December 1994 set the framework for the total budget and social sector investment projections. Based on the review of technical and institutional components, and the organizational and staffing changes made in recent months, the appraisal mission concluded that project preparation has been completed to an advanced stage of readiness for implementation. In addition, with grant funding from IDB, some of the consultants required for further strengthening of FHIS' management and operations are already being recruited, including a Technical Assistance Coordinator to be based in Honduras. Because of the importance of ensuring continued good management and operations, FHIS has now filled the remaining vacant senior management positions, agreed upon a detailed investment program for 1995 with relevant central and line ministries, and revised its Operational Manuals in line with the recommendations made during project preparation. During negotiations. FHIS provided assurances that any future appointment to the positions of Executive Director. Deputy Executive Director, Director of Planning. Director of Projects. Director of Control. Director of Supervision. and Director of Administration would be subject to prior IDA consultation (para. 7. 1(g)). In addition, a strengthened and simplified Operational Manual would be introduced by no later than June 30, 1996. A detailed project implementation schedule is provided in Annex 16. 5.2 Project Monitoring. Project implementation of the FHIS components would be monitored by FHIS according to performance indicators agreed with IDA (see Annex 17). '