Document: of The World Bank FOR OFFICIAL USE ONLY Report No: 22317 IMPLEMENTATION COM/PLETION REPORT (IDA-23880) ONA CREDrr IN THE AMOUNT OF US$ 19.1 MILLION TO THE REPUBLIC OF RWANDA FOR A FOOD SECURITY AND SOCIAL ACTION PROJECT June 19, 2001 Human Development Unit, AFTH4 Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their ] official duties. Its contents may not otherwise be disclosed without World Bank authorization. j CURRENCY EQUIVALENTS (Exchange Rate Effective January 2001) Currency Unit = RwF (Rwanda Franc) RwF 1 = US$ 0.0023211 US$ 1 = RwF430.8155000 FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS AE Agences d'Encadrement (Executing Agencies) LBW Labor-based Works AGETIP Agence d'Ex6cution de Travaux d' Int6ret Publique MINITER Ministere der rint6rieur (Interior Ministry) AGR Activitds Generatrices de Revenus MTR Mid-tenn Review BCP Bureau de Coordination du PNAS NGO Nongovernmental Organization BE Bureau des Etudes (Bureau for Studies) PAM Programmne Alimentaire Mondiale BTOR Back-to-office Report PAN Programme Alimentaire et Nutritionnel CAS Country Assistance Strategy PCD Project Concept Document CECI Centre canadien d'dtude et de coopdration PIU Project Implementation Unit CENA Centres pour les Enfants Non Accompagpds PME Petites et Moyennes Entreprises CEPEX Bureau Central des Projets et du Financement Exterieur PNAS Programme National d'Action Sociale DCA Development Credit Agreement PSR Project Status Report DRC Democratic Republic of the Congo QAE Quality at Entry FRW Franc Rwandais (Rwandese Franc) RWASAP Rwanda Social Action Project FSU F.onds Social d'Urgence (Emergency Social Fund) SAP Structural Adjustment Program GC G6nie Civil (Civil Works) SAR Staff Appraisal Report HIMO Haute Intensitd de Main d'Oeuvres (see LBW) SSC Small Scale Contractor ICR Implementation Completion Review TA Technical Advice DP InternaUy Displaced Person UNDP United Nations' Development Program [LO International Labor Organization UNICEF United Nations' Children Fund WFP World Food Program Vice President: Callisto Madavo Country Director: Emmanuel Mbi Sector Manager: -Arvil Van Adams Task Manager: Markus Kostner FOR OFFICIAL USE ONLY REPUBLIC OF RWANDA FOOD SECURITY AND SOCI]AL ACTION PROJECT CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 1 4. Achievement of Objective and Outputs 6 5. Major Factors Affecting Implementation and Outcomne 18 6. Sustainability 19 7. Bank and Borrower Performance 20 8. Lessons Learned 22 9. Partner Comments 23 10. Additional Information 23 Annex 1. Key Performance Indicators/Log Frame Matrix 24 Annex 2. Project Costs and Financing 27 Annex 3. Economic Costs and Benefits 29 Annex 4. Bank Inputs 30 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 34 Annex 6. Ratings of Bank and Borrower Performance 35 Annex 7. List of Supporting Documents 36 Annex 8. PNAS/CRDP-PVK - (Cr. 2388-RW - Completion Report (Summarized version) 37 Annex 9. Beneficiary Survey Methodology and Results 47 Annex 10. Aide-Memoire - Food Security and Social Action Project (PNAS) 48 Annex 11. Aide-Memoire - Pilot Community development Project (PVK) 50 Map No IBRD 25927R This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization. Pro ect ID: P002261 Pro ect Name: RW - FOOD SECURITY Team Leader: Markus Kostner TL Unit: AFTRS ICR Type: Core ICR Report Date: June 14, 2001 1. Project Data Name: RW - FOOD SECURITY L/C/TF Number: IDA-23880 CountryIDepartment: RWANDA Region: Africa Regional Office Sector/subsector: SA - Social Assistance KEY DATES Original Revised/Actual PCD: 04/23/1990 Effective: 08/05/1993 03/28/1995 Appraisal: 12/01/1991 MTR: 09/30/1995 07/15/1997 Approval: 08/12/1992 Closing: 06/30/1997 12/31/2000 Borrower/lmplementing Agency: GOVT OF RWANDA/BCP (PNAS); GOVT OF RWANDA/BC (PVK) Other Partners: World Food Program (WFP/PAM), UNICEF STAFF Current At Appraisal Vice President: Callisto Madavo Edward Jaycox Country Manager: Emmanuel Mbi Francisco Aquirre-Sacasa Sector Manager: Arvil Van Adams Allain Colliou Team Leader at ICR: Markus Kostner Jacques Baudouy ICR Primary Author: Michelle Cullen 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: M Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: Yes 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: Since the late 1950s, Rwanda has been plagued by both -violent and non-violent conflict, marked by the political manipulation of ethnicity to create tensions between Hutus and Tutsis and the consequent instability, mass migration, and violence. Plummeting coffee prices in the 1980s caused a decline in employment and income levels and GDP per capita, further aggravating tensions in this landlocked and resource poor country. By 1990, Rwanda was one of the world's least developed countries, with an average GNP per capita of US $310. Population density neared 280 people per km2, which was compounded by a fertility rate of 8.5 children per woman, both the highest in Africa at the time. To make matters worse, demographic growth neared 3.7% per year with needs surpassing the abilities of food production. Consequent malnutrition rates were very high, plaguing some 37% of children under five with moderate to severe under-nutrition. Food insecurity was a major concern, mainly due to a lack of purchasing power. Meanwhile, the AIDS epidemic created numerous famnilies without income-earners. In October 1990, the forces of the Rwanda Patriotic Front, a rebel movement composed mainly of exiled Tutsis, invaded Rwanda from Uganda and occupied a large part of the north of the country. The ensuing civil war disrupted economic and social activities and led to population displacements and a large military buildup by the regime in Kigali. The Government initiated a structural adjustment program during the late 1980s with the International Monetary Fund and the World Bank that included a social safety net of public expenditures to address the country's deteriorating economic situation. Per the Government's request, the Bank assisted with the preparation of a social action program to meet the needs of the most impoverished. Although various social sector operations were being supported by the Bank and other international agencies, new targeted social interventions were also needed to respond to the basic needs of the poorest population groups, who had suffered from ten years of economic deterioration and were unlikely to benefit rapidly from adjustment measures. The Programme National d'Action Sociale (PNAS) was conceived as a component of the Rwanda Social Action Project (RWASAP) to help poorer groups cope with the strains of the structural adjustment program. Project pre-evaluation began in October 1990, with a pre-appraisal mission in May of 1991. By November 1991, PNAS had separated from RWASAP to become its own project, with the official appraisal mission in December 1991. The PNAS objectives were to: a) improve the food security and social welfare of the poorest population groups; b) improve the Government's capability to monitor living standards of the population; and c) initiate a long-term poverty alleviation strategy. The Board approved the Credit on June 17, 1992, and the Development Credit Agreement was signed two months later on August 12th. The pre-effectiveness mission took place in September 1992, and about one year later, the project became effective (August 5, 1993). Project preparation took place while civil war was going on in the country. However, peace negotiations between the parties in the conflict (primarily the Government and the RPF) were underway. These talks were supported by the international community. Despite the difficult conditions, the Bank and the donors decided to continue with their programs as a demonstration of support for the peace process. After the project became effective in mid-1993, the civil war conditions delayed project launch and implementation. The signing of a peace agreement (Arusha Accord) between the Government and the RPF in August 1993 brought hope for the end of the war. However, the Arusha Accord was not implemented and in April 1994, massive violence erupted and up to 800,000 people, mainly Tutsis but also moderate Hutus from opposition parties, were killed. Fighting between the RPF and the Government army also resumed. The violence also led to the internal displacement of one million people and the exodus of 2 million into neighboring countries. By mid-July 1994, the RPF achieved military victory, brought the genocide to an end, and drove the soldiers and militia responsible for the genocide out of the country. A new coalition Government of National Unity was formed and sworn in on July 19, 1994. Rwanda's economy, which had already declined during the war, plunged and real GDP fell by 50 percent. During the genocide, some project staff were evacuated, some were killed while others fled the country. Project property was stolen, and project offices (including files) and works were destroyed. - 2 - 3.2 Revised Objective: After July 1994, the international community launched a major effort to reconstruct Rwanda and in that context, the Bank pledged to provide emergency new resources and the to restructure its existing portfolio to support the reconstruction effort. It was in this context that the project was the project was re-assessed in late 1994/early 1995 according to the country's post-genocide needs: emergency assistance with food and social services; reconstruction and rehabilitation of the destroyed infrastructure; immediate poverty relief; identification of the increased beneficiary groups; and capacity building to compensate for absent or poor-performing national partners, such as non-goverrunental organizations (NGOs) and small- and medium- sized enterprises (PME). Accordingly, the Bank responded in this second phase of the project to alter project objectives and components. The project was revised on March 28, 1995 to focus on improving: a) food security by redistributing food to nutrition centers and centers for unaccompanied children; and b) social welfare by undertaking community-selected infrastructure projects built by local small and medium-sized enterprises. New PNAS objectives sought to: a) support the recovery of community activities by rehabilitating social and economic infrastructure; b) promote small-scale productive activities; c) create temporary and permanent ernployment; and d) improve the nutrition of poor and war-affected groups. Project activities recommenced on July 1, 1995. During this initial period following the genocide, Bank and project staff faced extremely difficult in-country conditions. During the period following the genocide (1995-97), the conditions in the country were very difficult. The top govemrnent officials were new to their jobs, some new to the country, and inexperienced in public administration as well as with donor procedures. They were also very distrustful of international donors and NGOs. Dialogue with the Government was difficult. Security in the country was very fragile, and the congregation of the soldiers and militiamen who perpetrated the genocide in the refugee camps in the Democratic Republic of Congo, near the Rwandan border, remained a serious security threat. In 1997, these forces made many violent incursions into Rwanda and perpetrated more deadly violence. These difficult security conditions discouraged travel outside Kigali. The Bank and other intemational agencies struggled to find the appropriate strategy for such a fluid situation, and to mobilize the human and financial resources needed to respond to the challenges in Rwanda. The genocide had tom apart the social fabric and destroyed community solidarity, creating a difficult environment for projects such as PNAS. The Bank's local office was reopened in 1995 but was run by local staff until mid- 1996 when a representative from headquarters was appointed. By 1998, most of the refugees had returned to their communities, security conditions and public administration, while still fragile, had improved, and govenmment had engaged the Bank and the international community in constructive dialogue. A refomi program supported by the Bank and the IMF was launched in 1997. The Bank responded to the country's changing situation and need and per the Government's request, on April 16,1998, a third phase of PNAS was launched with the creation of Category 8 to carry out community sub-projects and local capacity building initiatives by a Project Coordination and Monitoring Unit (PCMU) in the Ministry of Local Govenument (MINALOC), using a new participation and decentralization approach. Category 8 was extended until December 31, 1999 and then December 31, 2000, while the other categories were closed on December 31, 1998. The remaining proceeds from phase II (categories 1-7) were transferred to Category 8, which served as a pilot for the Community Reintegration and Development Project (CRDP) (RW-PE-51931, effective as of March 31, 1999). These phase III pilot activities took place in 4 rural communes up until CRDP effectiveness and in 6 sectors of Kigali city (Prefecture de Kigali Ville-PVK) thereafter, implemented by a coordination office within PVK. The objectives of phase III were to: a) assist r eturnees and other vulnerable groups through a process of community-based reintegration and development; and b) strengthen the capacity of local communities and the administration at the communal and national levels for the implementation of development projects. -3- 3.3 Original Components: The pre-genocide PNAS design included four components: a) food aid and nutrition (Programme Alimentaire et Nutritionnel - PAN); b) labor-based works (LBW); c) micro-enterprise promotion and development; and d) social surveys and studies. (The latter two components were designated under the Programme d'Appui aux Initiatives de Base (PAIB) and Enquetes Sociales.) 3.4 Revised Components: In 1995, for phase II, project components were revised to include: a) food security and nutrition; b) labor-intensive works focusing on rebuilding infrastructure; and c) micro-credit activities. The food distribution and nutrition services of the original PNAS project were retained through PAN. To address the new socio-economic conditions of post-genocide Rwanda, an emergency social fand (Fonds Social d'Urgence - FSU) was established for community-based social and economic infrastructure sub-projects. The FSU included civil works (Genie Civil - GC), which focused on high labor and income generating activities (respectively, Haute Intensite de Main d'Oeuvres - HIMO and Activites Generatrices de Revenus - AGR). The HIMO (formerly LBW) through community-based work was geared to increase purchasing power (and food security). The AGR component (formerly micro-enterprise promotion and development) had a rural area focus in which it was to identify potential entrepreneurs and facilitate the creation of microenterprises by appraising their viability and providing financing for entrepreneurial endeavors. This component was designed only to assist entrepreneurs in obtaining financing from existing financial institutions. Training and studies were conducted across each of these components (PAN, HIMO, AGR) to improve implementation and gather knowledge for future related work. Before the genocide, the United Nations Development Program (UNDP) envisaged to cost share technical assistance to be provided through the International Labor Organization (ILO) for the labor-based works component (HIMO). However, the planned contribution of roughly $3.1 million was never realized. After the genocide, the revised project of 1995 had increased linkages with local development actors while pursuing its endeavors, mainly due to the increased practice of collaboration, the increase of agency involvement in Rwanda, and a lack of Government capacity immediately following the genocide. Local NGOs (DUTERIMBERE, ATEDEC and other executing agencies) assisted with AGR and the Canadian Centre for International Studies and Cooperation (CECI) facilitated some of the FSU sub-projects. Collaboration in PAN activities (which also occurred before the genocide) included links with the World Food Program (WFP), the United Nations Children's Fund (UNICEF), and CARITAS. The WFP provided funding before the genocide, and continued to after fighting ceased, when the need for these services increased. In all, the WFP provided a total grant of US $15.9 million for food security and nutritional activities (including the food itself, logistics, and the delivery of food and services). However, this US $15.9 million was not used specifically for the PANIPNAS, but rather for general emergencies related to the war. UNICEF supported the nutrition component, provided technical supervision, and provided micro-nutrients. CARITAS assisted with those families affected by AIDS and orphans, and the United Nations Volunteers (UNV) helped manage and coordinate PAN activities through the Ministry of Health and UNDP/UNV. As the project progressed, PNAS continued to evolve. Project adaptation in 1998 added an eighth category, comprised of two main components: a) community initiatives in each of the participating communes/sectors; and b) institution and capacity at the central and local levels. This adaptation attempted to build horizontal social capital within and between communities through increased participation and - 4 - vertical social capital between the Government and communities through decentralization. Improved conditions in the country, including capacity at central and local levels, permitted this community-focused approach to be adopted. The Category 8 design also benefited from increased Bank knowledge and focus on participation, social funds, and post-conflict reconstruction. The key focus of phase III was to listen to the communities themselves as they described their needs, and to give the people what they decided they needed while also attempting to improve the quality of work. One hundred thousand U.S. dollars were allotted to each of the 6 sectors in PVK for development purposes, that they themselves chose and prioritized. An elected community development committee (CDC), the entry point of phase III to local development, engaged the community members in development processes and served as a liaison with central government. Phase III served in particular as a test site for participation and decentralization in an urban environment. Phase I Phase 11 Phase III Effective 8/5/93 3/28/95 4/16/98 Objectives a) improve the food security and a) support the recovery of a) assist returnees and other social welfare of the poorest community activities by vulnerable groups through a population groups; rehabilitating social and process of community-based b) improve the Government's economic infrastructure; reintegration and development; capability to monitor living b) promote public participation, and b) strengthen the capacity standards of the population; primarily considered as of local communities and the and beneficiary contributions in administration at the communal c) initiate a long-term poverty cash or kind, in small-scale and national levels for the alleviation strategy. productive activities; implementation of development c) create temporary and projects. permanent employment; and d) improve the nutrition of elatively poor and vulnerable groups. Components a) food aid and nutrition (PAN); a) provide food security through a) community initiatives in each b) labor-based works (LBW); PAN; and of the participating sectors; and c) micro-enterprise promotion b) an emnergency social fund b) institution and capacity and development; and (FSU) to finance building at the central and local d) social surveys and studies. comrnunity-based social and levels. economic infrastructure sub-projects. The FSU included civil works (GC), which focused on high labor and income generating activities (HIMO and AGR). 3.5 Quality at Entry: The standards for assessing Quality at Entry (QAE) did not exist when the PNAS project was conceived, technically therefore, the assessment of QAE for this project is not applicable. An attempt was made however, to use today's guidelines based on the latest QAG Quality at Entry Assessment Questionnaire and make an assessment as to what the QAE may have been if today's standards had been applicable in 1991. - 5- Based on this exercise, the project would have most likely been rated satisfactory for the following QAE components: concept, objectives, and approach; poverty and social aspects; and policy and institutional aspects. Considering the deteriorating country conditions at the project's inception, PNAS may have been rated marginally satisfactory for fiduciary aspects, implementation arrangements, and risk assessment and management. There is not enough existing documentation on the project from 1991 to evaluate the technical, financial, and economic aspects. Environmental aspects are not applicable, as the project was rated Category C. In terms of development objectives, the project may have been rated either high or moderate. Overall therefore, it is estimated that the project's QAE would have been rated "S" for satisfactory at that time if today's guidelines had existed. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The main objectives of the PNAS project included initiatives geared toward poverty reduction and alleviation through the generation of employment; training and capacity building; improving food security through the distribution of food and improving nutrition education and intake; reaching the most vulnerable groups (including women, AIDS victims, and children); and the rehabilitation of infrastructure to help communities return to some sense of normalcy in terms of daily functions. Despite the complexities associated with project implementation in a conflict-affected country, the overall national impact of PNAS was diverse and widespread, as detailed below. Technical audits of the project were periodically conducted by independent external actors to monitor the progress and outcomes, while the PNAS Coordination Office (Bureau de Coordination du PNAS - BCP) also kept concurrent records and met regularly to discuss progress. Mid-term reviews were conducted in April 1997 to review technical matters and in July 1997 to review financial matters. A final technical audit was done in August of 1998. Once categories 1-7 of PNAS closed on December 31, 1998, a final financial audit was completed and the Goverrment's Bureau Central des Projets et du Financement Exterieur (CEPEX) conducted an inventory review. To enrich and boost these findings, and add the perceptions of those for whom the work was geared, a beneficiary assessment of phase II was implemented in the autumn of 1999. A final evaluation of phase III implementation was completed in December 2000, this included an accompanying beneficiary assessment. The final financial audit was completed March 2001. The Government also completed an independent analysis of the PNAS project, the "Rapport d'Achevement du Projet Programme National d'Actions Sociales (PNAS)," which was prepared by CEPEX (see Annex 8). When evaluating PNAS project outcomes, it is difficult to separate the effects of the conflict from project impact. Despite the difficulties incurred during project implementation, the overall evaluation for the PNAS project was satisfactory for various reasons. PNAS activities had widespread effects that benefited many in need, in terms of providing or improving water, food, health, education, income, and training. The project adequately addressed many immediate and short-term needs in a difficult environment. Some design improvements were made, benefiting from increased Bank knowledge and experience - such as improved structure of social funds, an increased emphasis placed on beneficiary participation, qualitative monitoring and evaluation, and further refinement in the Bank's frarnework and tools for conflict-affected countries. However, this was late in the life of the project. - 6 - Phase I objectives Early during project development, indicators were chosen to monitor and evaluate performance of the infrastructure sub-projects. These indicators included: the number completed, the number under construction, and the number processed for financing. Performance indicators for PAN were the tons of food distributed and the number of beneficiaries. No indicators were originally established to specifically address micro-credit enterprise promotion efforts in terms of number of credits allocated or the rate of repayment because of the uncertainty linked to the capacity of the executing agencies (agence d' encadrement-AE). In the short period between credit effectiveness and the onset of genocide: * The food security and social welfare of Rwandans were certainly improved by PNAS efforts. * Through the various training provided for the BCP / Project Implementation Unit (PIU), the Government's capability to monitor living standards of the population improved. * The PNAS project did not initiate a long-term poverty alleviation strategy. The genocide rendered the full achievement of phase I objectives unattainable. Phase II objectives After the 1995 revisions, key performance indicators changed to: i) the number of FSU financed sub-projects completed; ii) the number of beneficiaries for each sub-project - infrastructure, water supply, health centers, and micro-credit projects; iii) the increase in revenue originating from micro-projects financed by credit; iv) the person/years of work created by small and medium enterprises; and v) the metric tons of food distributed. Estimates of beneficiaries were rough, for it was difficult to accurately assess the numbers of those reached for some of the sub-projects conducted, such as roads and markets. In terms of meeting its set objectives: - The project did provide overall support for the recoveiy of community activities by rehabilitating social and economic infrastructure. m Small-scale productive activities in the form of micro-credit projects were supported. * Both temporary and permanent employment were created. - The nutrition of the general population was temporarily improved. The PNAS project had an impact in rural areas in terms of improved infrastructure (markets, roads, water sources, schools, health centers), poverty alleviation, and improvements in health due to nutrition training and food delivery. The revised PNAS focused mainly on the immediate needs of the population, and thus the sustainability of the improvements due to PNAS would depend on other follow-on actions by the Government. The progress of PNAS was captured to some extent by the indicators selected to monitor performance, while regular monitoring of existing indicators did permit PNAS to fine-tune interventions. As an operation targeting emergency needs, the focus was on output. However, it would have been useful if the indicators had also looked at impact, such as the quality of works completed, a disaggregation of data in terms of type of beneficiary reached (such as by gender, age, etc.), level of participant involvement, potential for sustainability, and whether the sub-projects matched and mnet targeted community needs. Phase III objectives The objectives of this phase focused on community-based development and local involvement in project planning and execution. Whether or not these objectives were reached was to be evaluated by examining - 7 - levels of a) decentralization, i.e. the transfer of decision-making and expenditure authority from the central to the community level; b) participation, i.e. a partnership between the local administration and the local population around sectoral planning and project implementation; c) social capital, i.e. trust and cooperation within and between local government and the local population. Based on its objectives, phase III: * strengthened the capacities of CDCs as the focal point of project implementation; * helped build local infrastructure as identified by the community; and * helped bridge the gap between the administration and the population that had existed as a result of a tradition of a centralized government structure. The achievement of the objectives of this project can be rated as "U" for phase I and as "S" for phases II and III. Overall, project outcome is rated as "S". 4.2 Outputs by components: Infrastructure Sub-projects The main objective of the FSU was to address poverty reduction and alleviation by rehabilitating infrastructure, generating employment, and building capacity through training. The indicators to monitor this activity were the number of infrastructure sub-projects completed - including infrastructure, water supply, health centers, and micro-projects - and the respective amount of beneficiaries. These indicators captured tangible improvements to infrastructure and rough numbers of those reached. By using local contractors and workers, the GC/HIMO aimed to rebuild local implementation capacity. hnfrastructure sub-projects consisted of nine different activities: rehabilitation and extension of schools; furnishing the schools; rehabilitation of health centers; rehabilitation of centers for non-accompanied children (CENA); spring development; rural water supply and sanitation; reforestation; rehabilitation of roads; and construction of markets. Local participation was also encouraged, but for the most part, this meant beneficiary contributions in cash or kind and involvement in AGR and HIMO activities. No data were available regarding performance before the project halted in 1994, i.e. for phase I. When project activities recommenced in the field in September 1995, the project was off to a good start. By October, one sub-project already had been completed, four more were underway, and 15 were about to begin. Achievement in 1996 and 1997 was rapid, which reflected an improvement in management and capacity at the local level. Training and capacity building were carried out at the local level, most likely contributing to this improvement. According to final reports, 393 sub-projects were completed during PNAS phase II, and over 75 percent of funds in this category were spent in the education sector. These sub-projects created roughly 33,000 temporary jobs and totaled some 2.5 million working days. The infrastructure works injected about $US 6,000,000 (RWf 1,875,000,000) into rural areas in terms of pay to contractors and laborers. Estimates of sub-projects completed, which reached an estimated 1,513,000 beneficiaries, are as follows: During site inspections, the audit missions observed that basic working conditions were in compliance to standards, such as finding no children on construction sites and having an increased participation of women (sometimes up to one-third of the labor force), who were paid the same salary as men. This access to the workforce for women was crucial, for in some rural areas females have been estimated to comprise up to two-thirds of the population, and as much as 50 percent of the households were headed by women in some communes. -8 - Category of works Numbers Estimated Amounts % completed beneficiaries disbursed (FRw) Rehabilitation and extension 159 199,000 2.,339,121,589 62.5 of school centers Furniture for school centers 113 174,000 467,867,774 12.5 Rehabilitation and furnished 5 NA 43,435,693 1.2 school centers Rehabilitation of health 15 180,000 78,821,247 2.1 centers Rehabilitation of centers for 5 4,900 66,480,199 1.8 non-accompanied children (CENA) _ Spring development 44 (750 sources) 141,000 110,429,627 2.9 Rural water supply and 16 70,000 336,258,627 9.0 sanitation Reforestation and 26 183,500 133,499,845 3.6 environmental protection Rehabilitation of rural roads 4 190,000 98,049,216 2.6 Construction of village 4 370,700 60,827,876 1.6 markets Other 2 NA 6,330,520 0.2 Total 393 1,513,000 3,741,122,213 100.0 Source: CEPEX final report. The CDC structure employed in phase III supported the process of decentralization, while also addressing urban poverty. The mechanisms that linked the CDCs to local administrations were designed to help facilitate the mobilization and engagement of the population in project activities. Responsibilities were passed on to CDCs and the communities themselves, in an attempt to increase feelings of ownership, improve participation, and help ensure that communities' priorities were met. Participatory research methods (Participatory Rural Appraisals-PRA), awareness-raising, and training were conducted to help ensure beneficiary understanding and participation in project development and implementation. In this sense, the process affiliated with implementation of phase III was as important as the output. The socio-economic infrastructure sub-projects completed during phase III, similar to those undertaken during phase II, were well-needed in the PVK area, which not only was largely damaged by the genocide and consequent displacement but also experienced massive in-migration. Community development plans for the 6 sectors were developed and realized with the involvement of the communities, CDCs, and local administration. In these projects, recurrent costs were provided by the community itself. -9- Sector Sub-project Nyakabanda Rehabilitation of school center Kimisagara Built a maternal school Rehabilitated a primary school Gatenga Constructed and extended a primary school Kimisange Kagugu Built a "salle polyvalente" Gisozi Extended a school center Source: CEPEX final report. Phase III was a small-scale activity that piloted new processes related to the CRDP. Therefore, the manual of procedures for the CRDP was used, causing some implementation problems as urban and rural communes are quite different, and the procedures manual was designed for work in rural areas. Other problems encountered include: a) the procedures for processing the sub-projects moved slowly; b) the demands for non-objections, which were initially designed for a larger project, slowed down the progress of the project; and c) the funds set aside for each sector (US$ 100,000) were not adequate to meet the community prioritized needs, especially in water supply. (In Kimisange, water supply was the first priority but when this sub-project could not be realized because of financial constraints, the community could not agree in time on another priority before project completion.) Had all provisions of the CRDP manual been applied, far greater and time-consuming investements would have had to be undertaken. Other organization problems were incurred in the CDC, such as the fact that CDC members had other functions outside of the CDC role, which restricted the time they had for the project. Overall, however, the quality of works and community ownership were considered higher than under phase II. Infrastructure sub-projects can be rated as "S" for phases II and III. Micro-Credit Activities The AGR's objectives focused on employment generation and capacity building through training. Eight various training sessions were held in regard to AGR. To measure this performance, the increase in revenue originating from micro-projects financed by the credit were to be examined, as were the numbers of beneficiaries reached and number of micro-projects created. Under phase II, there were 3 different types of micro-credit projects: * Type A - A medium-sized credit with a maximum amount of 2,000,000 FRW per project, granted by PNAS and managed by an executing agency (AE). * Type B - A small-sized credit with a maximum amount of 500,000 FRW put into place by the Union des Banques populaires du Rwanda (UBPR). * Type C - A small-sized credit with a maximum amount of 300,000 FRW put into place by PNAS in collaboration with the Ministere du Genre, de la Famille et des Affaires Sociales (MIGEFASO) and - 10 - managed by a consortium of associations and groups; this took place in four neighboring communes of Kigali (Kanombe, Rubungo, Rutongo, and Kanzenze). Feedback on each of the three micro-credit components revealed that most were largely unsuccessful for various reasons in terms of sustained micro-credit projects, rates of repayment, failure rates, and quality. The first type of the micro-credit projects involved contracting local NGOs to manage and oversee the sub-projects and lending (Type A). This type was quite unsuccessful, with initial poor repayment levels. Type A also experienced problems due to the weakness of the NGOs used to support these activities. The component was then restructured a second time in an attempt to improve recovery. The second type utilized a Bank (the UBPR) to manage affairs (Type B). When this scheme failed, due to a UBPR change in policy that decided against continuing with micro-credit projects, a third type was tested. This type C used a consortium of associations and groups to manage the lending process. Consortiums of local commune members oversaw type C, and included respected NGO or local administration representatives from the commune of the debtor. Although this third type met with partial success, the impending project closure (categories 1-7) cut this experiment short. Types of micro-credit Numbers Amounts disbursed Percentage of loans repaid activity Completed (FRw) Type A 27 51,486,070 12% Type B 136 60,000,000 100% (These funds were fully recuperated as they were repaid by the UBPR, not necessarily the debtor.) Type C 48 13,762,500 47% Total 211 125,248,570 58% Source: CEPEX final report. Certain levels of economic success were achieved by AGR promoters, mainly in the areas of handicraft and brick-making. For the most part, none of the AGR was considered very effective, as many loan repayments were not made and many sub-projects failed (this is not aLtypical, as many new enterprises often fail, and conditions were compounded by a post-conflict environment). Of the 211 micro-credit projects financed under phase II, 22 resulted in litigation, mainly due to failure to repay. These failures, in terms of projects and repayments, were partially due to the challenges affiliated with attempting to jump-start micro-credit projects in highly impoverished conflict-stricken environments where purchasing power is severely limited. Low levels of group cohesion, weak supervision from overseeing bodies and Govenmment counterparts, and a lack of adequate training to both those managing and borrowing funds also complicated the implementation of AGR. Under phase III, the population in one sector (Nyakabanda) identified a fund for AGR as one of its priorities. There was not adequate time devoted to fully develop this component, while slow local bank transfers also hindered progress. By the time PNAS closed, only 10 out of 20 planned AGR had been started, too few to judge success or failure of this community-driven approach to micro-credit. - 11 - The micro-credit component can be rated "U" for phases II and III. Food Security and Nutrition Program The objective of PAN was to ameliorate food security by distributing food, and improving nutrition education and intake (primarily aimed at the most vulnerable groups - children, AIDS victims, and widows). This was to be done by providing nutritional activities at each level of service; nutrition education; supplementation of vitamin A, iron, and iodine; and training and supervision at each level of nutritional service. The indicator chosen to assess this was metric tons of food distributed. Before the genocide, PNAS paid for the transportation costs and storage of food, while also assisting with the processing of contracts for private transport. WFP was responsible for buying the food and distributing it to the nutritional centers. After the genocide, PNAS paid for half of the costs of storage and transportation. The food security component of PNAS enabled the delivery of food to 80 nutritional clinics and 40 centers for non-accompanied children in all prefectures of the country, and attempted to target the most vulnerable groups. Training on nutritional activities was provided to 540 staff in these centers. Roughly 8,465 tons of food were distributed, improving (at least temporarily) food security among vulnerable groups. A total of 128,300 benefited from PAN activities. Summary of targeted PAN Services Estimated beneficiaries Nutritional services 81,000 people Services provided for families affected by AIDS 6,4000 families Services for children of families affected by AIDS 2,500 children Nutritional services for under fives in orphanages 2,330 children under five Rations for over fives in orphanages 7,660 children over five Source: Trimester Execution Report PNAS/BCP Jan I - March 1997. The food security and nutrition component can be rated as "S". Training and Capacity Building Some training was provided before the genocide, mainly in project management, procedures for tender, training of surveyors, and training of those involved with the implementation and supervision of PNAS. A more concerted effort was undertaken during phase II, where 943 people benefited from PNAS training (of which 301 were from AE and PME). Various shortcomings in project implementation, especially those related to quality of work, sub-contractor mis-management, and poor repayment rates on micro-credit activities, spurred the development of training programs after the genocide. After the technical audit of 1996, it was decided that training programs were needed for small-scale contractors and consulting engineers to help correct building, planning and design problems noted in monitoring reports. For engineers of small local consulting offices, the following training was provided: labor-based road - 12 - technology (making inventories of rural roads and small structures, design and supervision of road works); building techniques (design in function of locally available materials and supervision of works); rural water supplies and sanitation (design and supervision); and enterprise management (preparation of bids, computation of unit prices and engineers estimate, financial and personnel management). For small-scale contractors (SSC), training was provided on: road construction techniques (practical and theoretical training of supervisory staff); building techniques (in particular wooden roof structures); and enterprise management (training of contractors in preparation of bids, computation of unit prices, financial, equipment and personnel management). These training programs were implemented by local and international consultants, and an expert from the ILO/UNDP project "entreprendre a Madagascar" for training enterprise management. Most of the training activities were classroom based. Though noteworthy, these efforts came too late in phase II to have a substantial impact on the quality of works. People Trained 1995 1996 1997 Total PNAS bodies 12 4 5 16 Bureau d'6tude (BE) 0 0 24 0 PME 0 0 78 21 AE/GC 0 0 4 4 AE/AGR 0 120 156 276 Assistants for AIDS 0 20 20 40 Agents CENA 70 72 120 262 Nutritional Service Agents 35 43 160 238 Total Numbers Trained 117 189 567 943 Source: Figure 1 of the Execution Report PNAS/BCP January 1 - 31, 1997. Phase III tested a new approach to development in Rwanda, requiring intensive upfront capacity building. To ensure that CDCs function properly, it is crucial to provide adequate training not only to improve capacity but also to motivate. When CDCs are able to communicate well with the beneficiaries, the activities of the project are far more likely to succeed. At start-up of phase III, training has been provided to all CDCs. Two training sessions were organized for the 88 CDC members, including for participatory rural appraisals, organizational, project and financial management, and environmental awareness. In addition, 177 beneficiaries of AGR received basic training in micro-credit management. The fact that the CDCs were able to achieve their objectives to the satisfaction of beneficiaries reflects well on the capacity building component and overall design. The training component can be rated as "U" for phase 11 and as "S" for phase III. - 13 - Studies Forty-three various studies were completed under phase II. Most of these were technical works to guide sub-projects on bridges, roads, markets, schools, water pipelines and sources, and a veterinary clinic. These studies were conducted by engineers (affiliated with the AEs and bureau d'etude - BE) and used to inform sub-project design. Studies undertaken after the completion of sub-projects found various problems with implementation, such as: * Building construction: not all contractors designed blueprints or building plans before construction, or provided required technical specifications. Some buildings lacked proper anchoring for the roofs and were plagued by mis-measurements; * Water sources and piping: some contractors used the wrong pipe dimensions, not adequately gauging source levels and population needs, and lacked the technical expertise to adequately address the required filtration layers around wells; * Tracks (dirt roads): a number of contractors did not pre-plan construction, nor did they consider anticipated traffic volume, road usage or requirements according to level and weight of traffic, and the cost of transportation by vehicle. Type of Study Number of studies Communes investigated conducted Tracks (dirt roads) 11 Gikongoro, Gitarama, Kibungo, Byumba Bridges 2 Gitarama, Kigali-rural Markets 4 Gikongoro, Byumba, Kigali-rural, Gitarama Schools 16 Kibungo, Kigali-rural, PVK Water sources and piping 7 Gikongoro, Byumba, Kigali-rural Veterinary Clinic I Kibungo Phase III envisaged the carrying out of selected studies to further the understanding of community development issues in an urban context. Delays occurred at the PVK and its Bureau de Coordination (BC) in identifying such studies, and in preparing terms of reference and bidding documents. By the time the documentation was ready, pending project closing inhibited the carrying out of the studies. The studies component can be rated as "S" for phase II and as "U" for phase III. Achievement from the beneficiaries' perspectives Beneficiary assessments (BAs) were carried out for phases II and III. (BA methods are explained in Annex 9.) For phase II, the BA was undertaken during September - December 1999 to supplement the final technical audit of August 1998 and the initial findings from the CEPEX inventory conducted in January 1999. This process involved interviewing those who provided (contractors, local administration, former PNAS staff, etc.) and received (such as beneficiaries, nutrition center staff, NGOs and cooperatives) services from the FSU and PAN components at a limited sample of project sites. - 14 - Overall, the beneficiaries interviewed throughout this small cross-section of project activity sites appreciated PNAS' efforts under phase II and made it clear that they would like additional work of the same nature to be undertaken within their commune. However, various issues emerged within the areas assessed that should be revisited to help improve future design and implementation of similar Bank projects. Infrastructure Sub-projects In the areas where the BA was conducted, problems with the quality of infrastructure sub-projects were apparent, especially in education. Schools were found to have experienced various faults: leaking roofs, splintering or warped floors, uneven door and window jams that prevented proper closure, construction techniques that prevented or hindered ventilation, rapid erosion of bricks due to improper fabrication, and blackboards on which chalk could not be used. In some situations, classrooms were built, but no adjacent latrines or buildings for school management were constructed. These problems reflected the use of poor quality materials and unqualified contractors and laborers. Some work that was commenced was never finished, while in some areas sub-projects never even started. The difficulties incurred with poor quality mfaterials and unqualified contractors stemmed to some degree from a lack of close supervision and management by the BCP. In some cases, there was a lack of planning and integration with other local development activities or actors, where either the church, local NGOs, or the local administration were side-stepped. The lack of skills held by those building the infrastructure sub-projects, and the lack of management (financial and administrative) by the enterprises involved reveals that more training should have been provided. In other sectors, the impact on the population was perceived much more favorably. Those who were beneficiaries of road projects were extremely satisfied with PNAS' work, as well as some of those who benefited from the water projects (source development and systems). However according to BA findings, some of the water source development did lead to the exclusion of the poorest of the poor from these services. Water source development, while facilitating access to water and alleviating sicknesses affiliated with contaminated water, can in some cases lead to such types of exclusion from services. Although pre-PNAS water sources may have been farther away before source development, they were at least free. After source development, however, fees were required for water usage, resulting in the exclusion of the poorest in the commune. Community involvement in infrastructure sub-project activities was limited for the most part to either work as a laborer or beneficiary contribution in cash or kind. Beneficiaries did not seem to be involved in PNAS sub-project activity selection, design, or monitoring. More often than not, sub-project proposals appear to have originated from the burgomaster rather than the community, as had been envisaged. Micro-Credit Activities Type A of the micro-credit activities was launched in April of 1996 and lasted until April of 1997. Type B ran for 8-12 months after it was initiated in September 1996. While type C began in March of 1998 and ran until a few months after project closing. Beneficiary assessment findings in general noted that each component experienced problems with repayment and failing micro-credit activities due to various factors. First and foremost, the dismal economic conditions in the rural areas were not favorable at the time for business start-ups. *-15- In addition, problems for the micro-credit component stemmed from: the late provision of training, for both those overseeing the projects or taking loans. In general, training was provided after the micro-credit projects conmmenced; the repayment period was too short (varied between 8 and 12 months) and the interest rate was high, considering that inflation was rapidly decelerating; and those who took loans often did not have enough business experience. As a result of these problems, many stopped their activities after they repaid the loan and not many of the groups formed for the AGR still exist, either the implementing agencies or beneficiary groups. In fact, in post-genocide Rwanda, many groups had been formed for the sole purpose of accessing donor funds rather than for implementing AGR. The poor repaymnent rate for type A stemmed from a dearth of qualified personnel and more competitive micro-credit programs funded by other donors that had lower interest rates, thus causing the sense of importance of and interest in the PNAS program to fall (attracting the most risky borrowers). Furthermore, several AE provided credit to farnily members and friends. Type B had the best repayment rate. However, the consultants who conducted the beneficiary assessment felt that this was due to the fact that the bank would not give the loan for a micro-finance activity unless the potential debtor had a guarantee. This meant that by default, the poorest groups within the communes where type B was implemented did not have access to these loans. Not far behind type B in terms of repayment rate, type C (60% repayment) was deemed by the beneficiary assessment consultants as the most probable model for successful micro-finance activities in Rwanda. They found that many of the type C projects were successful, however had to cease due to the overly high interest rates or PNAS project closure. Certain type C activities that were commenced during PNAS still exist, as do some of the groups who formed to participate in micro-credit activities. Despite the more encouraging results of type C, given the highly complex country context, AGR should have been discontinued in the post-genocide period, as suggested by the Bank. Food Security and Nutritional Program At the time of PNAS, there were (and in some areas still are) high levels of malnutrition; thus, those who did receive aid appreciated PAN efforts. As the PAN project continued, food aid decreased in quality and quantity. It was requested that recipients make contributions if they could. Aid was initially based on a weight/age ratio; this was later switched to a weight/size ratio - effectively disqualifying many malnourished for aid. As PAN progressed, this decrease in the quality of service given translated into fewer people seeking food relief from PAN. Although some health center staff were trained, most beneficiaries stated that they did not receive nutrition training, and that some sort of education about how they could have used local resources to better feed their children would have been appreciated. These observations reflect on the general nature of the delivery of emergency operations where the focus is on short-term output rather than on longer-term impact. Phase III According to findings from the phase III BA, these community-oriented efforts, in general, resulted in improved quality of socioeconomic infrastructure and capacity building. Most sub-project activities undertaken were completed and were of good quality, while procedures for implementation were also respected. Yet in a few cases, infrastructure sub-projects were unable to be finished or reach their full potential due to inadequate funding, inadequate communication between the CDC and beneficiaries, inadequate time allotted for implementation, or numerous changes in community priorities. Some delays were also experienced in jump-starting the sub-projects. In terms of the micro-credit component, which had been identified by the population of one sector as a priority, beneficiaries at times did not understand the - 16 - basics of the program. Overall, beneficiaries appreciated the capacity builiding that occurred with sub-project implementation, and felt that their increased participation would help increase the sustainability of the sub-projects. The community development plans established by the CDCs and communities were also helpful to the local administration in terms of development within the commune. Some CDCs have already contacted other donors to finance remaining activities contained in these plans and partially succeeded in raising funds. Yet, as this was a new approach to development in the PVK, for which many in the pilot communities were unfamiliar, more time should have been allotted to ensure understanding on the part of the CDCs, communities, and local administration. The lessons from the phase III BA are already being applied in CRDP implementation. Summary ratings The following table provides an overview of ratings for phases II and III for each component: Component Phase II rating Phase III rating Infrastructure S S Micro-credit activities U U Food security and nutrition S Training and capacity building U S Studies S U Taking into account the very difficult and evolving country situation and the satisfactory rating for the most important component (infrastructure), the overall output rating for the project is "S". 4.3 Net Present Value/Economic rate of return: N.A. 4.4 Financial rate of return: N.A. 4.5 Institutional development impact: PNAS phase II encouraged institutional development through the training of national and local government to help execute PNAS and its various sub-projects. In addition, local AEs and PMEs benefited from the training conducted for and experience gained from the micro-credit and infrastructure sub-projects. However, such institutional strengthening happened late in phase II. Phase III also used training to help improve institutional development at the local level, but took this a step further by involving the local population and administration in decision-making processes. Community input was sought for prioritizing local needs, developing community development plans, and allocating * 17- funds from a set budget to fulfill and meet those needs, while local contractors were used in the execution of these local sub-projects. Thus phase III not only trained local groups, but also attempted to empower them to take part an active role in local development. The CDC is now an integral part of Government policy and is used for other development interventions. New Bank projects, including the Human Resources Development, the Rural Water Supply and Sanitation, and the Rural Sector Support have used these structures in their community-focused components. The institutional development impact is rated "M" for phase II, "SU" for phase III, and "M" overall. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: Conditions internal and external to Rwanda affected project development, impact, and success. Difficult social and economic conditions in Rwanda were compounded by the start of civil war in October 1990 and by the genocide in April-July 1994. Throughout the project's life, Rwanda experienced political instability, internal and regional insecurity, and suffered from a lack of institutional capacity. Another key problem was the continual mass migration of refugees (exiting and retuming) and internally displaced persons (IDPs). In 1994, more than half a million old caseload returnees, 200,000 new caseload returnees, and I million IDPs flooded the countryside. Roughly 1.5 million refugees returned to Rwanda in 1996/97. It is unlikely that more rapid alterations in project design could have been better able to deal with these issues of such immense proportions. 5.2 Factors generally subject to government control: Most of the documentation recording early project history of PNAS was lost during the genocide, in particular those regarding the procurement of goods from the beginning of the project until April 1994. Therefore Government and implementing agencies' roles are only evaluated in the post-genocide era. Administrative problems in phase II, including supervision and general project management, stemmed from a lack of capacity in the new BCP installed after the genocide. In addition, frequent changes in Government positions affected project implementation, leading to changes in sub-project focus. New officials tended to push a change in project emphasis. In many cases, access was stressed over quality (to some extent however, Rwanda was not institutionally organized to tackle quality in 1995 and 1996, and even 1997). The Government could have also stepped-in in several areas and assisted with ensuring the sustainability of PNAS activities (in terms of covering recurrent costs, etc.). Implementation of PNAS phase II might have been improved through a different choice of parent ministry. PNAS was attached to the Ministry of Finance, a ministry devoid of people skills, with an urban bias, no links to the rural communities, and no construction, engineering and agricultural back-stopping capabilities. The initial choice may have been because of the high visibility of the ministry and the cross-sectoral nature of the project. However, this institutional choice may have led to missed opportunities to build capacity both at the central and local levels. Furthermore, it may have contributed to a certain lack of government ownership (linked also to the fact that many in government saw PNAS as associated with the pre-genocide government). It appears thus more appropriate in post-conflict settings to house a project in the relevant sector ministry and use the project to build much needed capacities, even if this delayed implementation somewhat. - 18- This lesson has been taken up in phase III, which was implemented under the Ministry of Local Government. Furthermore, the overall policy environment for community-based urban development was favorable and support from the PVK and the parent ministry was readily forthcoming. Nevertheless, during phase III, delays occurred in particular regarding procurement through the National Tender Board. 5.3 Factors generally subject to implementing agency control: During phase II, the BCP experienced financial and administrative management difficulties, mainly in instances involving sub-projects and working with sub-contractors. Sub-projects experienced delays, were left uncompleted, used poor quality materials, or were completed in a sub-standard manner. Many of these problems were within the control of the sub-contractors hired to complete the sub-projects. Problems with sub-project management stemmed from the overall weakness and shortage of well-trained and technical personnel at the communal level but also from sub-contractors operating in bad faith - by misusing funds, not paying workers, or not completing (or sometimes not even beginning) work. Tighter BCP supervision may have prevented the use of poor quality materials and helped ensure better quality of end-products. Specific guidelines on how to select sub-contractors may have helped the BCP avoid the retaining of unqualified and poor quality firms. As the project developed, the BCP did begin to address this problem and started a "black list" of agencies who had done poor work or had trouble implementing the sub-projects they were contracted to do. However, this list could not be found in the files, and it can, thus, not be ascertained to what extent it was applied. Nevertheless, judging from the problems incurred with sub-projects, a more stringent policy of evaluating and assessing contractors before hiring them could have been beneficial to the project. Implementation of phase III experienced certain delays at the beginning, linked to the need to set up a new albeit small project office in PVK and the unfamiliarity of BC staff with the new approach to participation and decentralization. 5.4 Costs andfinancing: The total amount of the IDA credit was US$ 19,100,000, or SDR 14,000,000. As of January 22, 2001, approximately 97.00% of the funds had been disbursed. A total of US$ 18,919,150.35 had been disbursed (SDR 13,575,773.98), leaving a total of US$ 551,408.98 undisbursed (SDR 424,226.02). The Borrower obligation, as of December 31, 2000, was US$ 17,712,312. 6. Sustainability 6.1 Rationale for sustainability rating: The PNAS project was established to address the immediate social needs linked to the implementation of the structural adjustment program. Following the genocide, PNAS phase II took on some activities similar to those provided by relief, which met local needs but were not focused on long-term development. To this extent, it may not be appropriate to evaluate the project's sustainability, especially when considering the number of times project objectives and components changed throughout the project's lifetime. PNAS activities after the genocide were initiated on the ground often before the line ministries were effectively engaged in interventions. Thus, PNAS activities were not able to be fully integrated into established government structures at least up until 1997. Difficult in-country conditions, such as insecurity and widespread abject poverty, prevented the establishment of cost recovery systems for most of the sub-project activities, thereby also diminishing project sustainability. To counteract this, however, some - 19- local administrations have put their own cost recovery systems into place to keep infrastructure sub-projects operational. In addition, beneficiaries in some areas participated in the management and financing of infrastructure sub-project maintenance. PNAS did, however, leave a long-term mark in terms of the infrastructure it left in place. Schools, health centers, and water sources constructed by PNAS are still in use today, and beneficiaries are still reaping the rewards. Another sustainable benefit from PNAS has been the increase in local (and to a lesser extent national) knowledge and know-how from training and capacity building activities. National and local Government actors involved in the implementation of PNAS gained experience from the project in terms of financial and administrative management. The micro-credit component, which was integrated into the FSU to perpetuate the momentum in economic and social growth left by PNAS after its closing, was for the most part unsuccessful, and therefore did not leave continued economic growth and development in the communes. The local capacity building efforts under phase III are likely to have a more pronounced, longer-term effect on development. The Government held local elections to establish CDCs in all communes throughout Rwanda, signifying a commitment to local level involvement. This local involvement will continue through the actions of the CDCs and their work with communities in establishing their own development plans, making their own decisions in regards to their needs, and allocating funds to these prioritized needs as they see fit. This decentralization and increased participation in development, as spearheaded under PNAS phase III, could lead to an increase of social capital within communes, while also contributing to a regained trust in both the local and national level administration. In particular because of the importance of CDCs in the Government's development approach, sustainability is rated as "L". 6.2 Transition arrangement to regular operations: The Government discontinued phase II-type operations in favor of phase III-type development interventions geared toward greater decentralization and participation. The CDCs have become part and parcel of the Government's new approach. Their capacities and functions are being enhanced in a wide range of sectors and activities, including other Bank-financed projects. 7. Bank and Borrower Performance Bank 7. 1 Lending: Lack of pre-genocide information does not allow for a detailed assessment of Bank performance during the lending and early supervision stages (files at headquarters pertaining to phase I were lost in the handover of task managers in 1996). The PNAS project design reflected similar other social programs that accompanied the structural adjustment process in the early 1990s and is based on the findings of the 1992 poverty assessment. Available information, thus, suggests that Bank performance can be rated as "S", given in particular the challenges of project preparation in a civil war context. 7.2 Supervision: Bank supervision during phase II did troubleshoot some of the administrative and financial management problems incurred by PNAS staff, and it is evident by certain project changes that Bank staff tried to accommodate to changing conditions and respond to feedback from the technical reports. However, - 20 - supervision missions between late 1995 and early 1997 were few, and an inadequate skill mix can be observed throughout phase II. At the same time, the involvement of the Bank's office in Kigali remained limited. Closer supervision during phase II may have helped: a) increase the capacity of the BCP to ensure better use of moneys and more qualified contractors for infrastructure and micro-credit sub-projects; b) recognize earlier the need for training on basic infrastructure and micro-credit activities, while also better matching training to meet required needs; and c) recognize the need for an improved monitoring and evaluation system. Under phase III, the skill mix was more adequate and the field office was closely involved in supervision. It should be noted that mission travel until 1997 was risky and sometimes outright dangerous. Moreover, an extraordinarily limited supervision budget had a negative impact on the level of supervision and the skill mix of teams. Given these circumstances, Bank supervision efforts during the immediate post-genocide years were laudable. Nevertheless, Bank supervision perfonmance is rated as "U" during phase II and "S" during phase III. Overall, Bank supervision perfonnance is rated as "U". 7.3 Overall Bank performance: Within a most complex environment, the Bank has been able to respond appropriately to events and needs throughout the life of the project, to learn from the evolving project, and to apply increasing knowledge accumulating institution-wide. The overall Bank performance is, therefore, rated "S". Borrower 7.4 Preparation: Missing documentation complicates an assessment of Government performance during lending. The PNAS project stemmed from a Govemment and Bank effort to alleviate negative economic conditions from the structural adjustment program and agricultural decline. The Government assisted with targeting the country's most immediate social needs (in the fonn of food security, employment, and quick injections of funds at the local level), as well as the most vulnerable groups and regions. This also was designed to address some of the consequences emerging from the civil war. An independent consultant found that the government was interested in creating social expenditure programs that would help address the social and economic problems facing Rwanda at the time. Borrower performance during preparation is, therefore, rated as "S". 7.5 Government implementation performance: The Government's role in the PNAS project was limited by the difficult conditions in the country including the low levels of capacity after the genocide. To aL lesser extent, frequent changes in Government officials also hindered project performance under phase II and National Tender Board procedures unnecessarily delayed phase III implementation. Consultant findings show that the Government effectively oversaw PNAS' implementation of phases II and III. However, the problems encountered especially during phase II suggest that closer attention would have been warranted. Consequently, Government implementation performance is rated as "U". 7.6 Implementing Agency: The PNAS implementation unit (BCP) of phase II accomplished quite remarkable achievements, - 21 - re-launching activities rapidly in late 1995 and accommodating many of the needs emanating from the massive return of refugees in 1996/97. The speedy start-up, however, came at the cost of less than adequate sub-project supervision also due to the fact that supervision staff were few and capacities were limited. Furthermore, financial and administrative management problems were encountered. Performance of the implementation agency during phase II is, thus, rated as "U". In this light, the "HS" ratings for development objectives and implementation progress during 1996 and 1997 may not have been justified. The coordination office in PVK had only limited staff, which caused some delays in the start-up phase. However, it implemented phase III 8 diligently and following strict financial management procedures. Its performance is rated as "S". Overall, the performance of the implementing agency is rated as "U". 7.7 Overall Borrower performance: The overall Borrower's performance is rated "S". Special note should be made that this project was implemented under complex circumstances, and despite this, was able to set out and achieve, to a fair extent, its objectives. The Borrower's performance contributed to this achievement. Furthermore, the Government very strategically re-oriented the project in phase III toward an as of yet untested approach to development, using phase III methods of decentralization and participation also to demonstrate to other development partners the appropriateness of the new approach. 8. Lessons Learned Considering the pre- and post-genocide states of Rwanda, and the continuing insecurity that plagued the country throughout the project's existence, the PNAS project was effective in meeting its established goals and objectives. Both Bank staff and Government representatives operated under difficult conditions, especially in the years immediately following the genocide. Based on the experience of PNAS, there are various recommendations on project design and management that can be made to help ensure the quality of similar projects in conflict-affected countries. Post-conflict environments necessitate malleable project designs in order to be able to adapt to the often rapidly changing conditions as to what is feasible and what is desirable. PNAS was a standard investment project, which is inherently inflexible despite the ability to restructure, with lengthy implementation periods. At the same time, the ability to adapt a project to the emerging situation in the country can help facilitate, and accompany, the difficult transition from short-term emergency to longer-term reconstruction activities. PNAS has proven that a project can indeed evolve and adjust to shifting priorities, closing also the gap between what is desirable and what is feasible in a post-conflict context. Speed and ownership are a regular trade-off for interventions in a post-conflict environment. In an emergency period, implementation through an independent agency (such as a social fund) may be most efficient as long as local capacities are too weak to carry out such activities. However, the most effective and efficacious implementation strategies may be those more participatory in nature that actively involve the community and the local administration. This may help improve sub-project activity selection and targeting sub-projects to match beneficiary needs. An integral part of this involvement includes capacity building activities that seek to train and empower local administrators, the beneficiaries, sub-contractors and NGOs. Furthermore, sustainability options should be built into each project to the extent possible, such as cost recovery systems that are identified and agreed upon by the population and the local administration. Monitoring and evaluation, including quantitative and qualitative indicators, should be conducted regularly - 22 - so as to be able to make snap decisions and respond expediently to matters in a changing and sometimes volatile environment. Organization is quite important for projects in conflict areas, which require good documentation systems, regular supervision missions with adequate skill mix (especially regarding fiduciary responsibilities), and keeping on top of the rapidly changing social, political and economic environments. Where Bank supervision from headquarters is constrained for security reasons, staff in the field office should be involved closely. If this were not feasible either, alternative arrangements could be sought with NGOs and/or donor/UN partners. Projects in post-conflict countries should pay particular attention to financial management. The emergency situations typical to the post-conflict environment necessitate the quick disbursement of funds. However, weak Government capacity and the lack of adeq[uate financial controls require the close monitoring of funds. Care should also be taken to ensure the equal allocation of funds to projects that are inclusive and target all needy groups, and do not focus solely on one particular group (so as to avoid exacerbate inter-group tensions). Micro-credit activities, which by nature require some level of stability within the environment they are implemented, may not be appropriate in post-conflict environments. Local level capacities are usually drained, and economic conditions may not be the most favorable for business start-ups. If micro-credit activities are implemented, community ownership seems essential and great care should be taken in selecting executing agencies, which should be experienced, trustworthy groups with a history of work in the area. 9. Partner Comments (a) Borrower/implementing agency: A summary report by the Government's Bureau Central des Projets et du Financement Exterieur (CEPEX) is provided in Annex 8. The complete report is available in the project files. (b) Cofinanciers: (c) Other partners (NGOs/private sector): 10. Additional Information * Annex 9 provides a summary of the Beneficiary Assessment methodology applied for assessing project impact under phases II and III. * Annex 10 provides the aide-memoire of the IC]R preparation mission (June 2000). * Annex 11 provides the aide-memoire of the ICR review mission (March 2001). - 23 - Annex 1. Key Performance Indicators/Log Frame Matrix Outcome I Impact Indicators: Indicator/Matrix Projected in last PSR ActuallLatest Estimate Infrastructure sub-projects Rehabilitation and extension of school 156 156 centers Furniture for school centers 122 122 Rehabilitation and fumishing of school 13 centers Rehabilitation of health centers 13 13 Rehabilitation of centers for 4 4 non-accompanied children (CENA) Spnng development 750 (44 sources) 750 (44 sources) Rural water supply and sanitation 15 15 Reforestation and environmental protection 26 26 Rehabilitation of rural roads 3 3 Construction of village markets 3 3 Other 0 Amount of employment generated 2,500,000 2,500,000 (personfday) Number of beneficiaries Schools 199,000 199,012 School Furniture 174,000 174,048 Health Centers 180,000 180,000 CENA 4,900 4,856 Spring development 141,000 141,120 Water supply and sanitation 70,000 70,000 Reforestation and environmental protection 183,500 190,000 Rehabilitation of rural roads 190,000 190,000 Construction of village markets 370,700 370,222 Micro-credit actvities Numbers of micro-projects completed 211 209 Beneficiaries 815-1150 - 24 - Food security component Metric Tons of Food Distributed 8,465 8,465 Beneficiaries 128,300 128,300 Training and capacity building Number of training sessions provided 108 108 Beneficiaries 943 943 Studies Number of studies conducted 43 40 Original objectives (phase 1) Improve the food security and social welfare Yes Yes of the poorest populaton groups Improve the Govemments capability to Yes Yes monitor living standards of the population Initiate a long-term poverty alleviation strategy Yes No Revised objectives (phase 11) Support the recovery of community activities Yes Yes by rehabilitating social and economic infrastructure Promote small-scale productive activities Yes Yes Create temporary and permanent Yes Yes employment Improve the nutrition of relatively poor and Yes Yes vulnerable groups Adapted objectives (phase l1l) Openness of dialogue achieved in the All CDCs discuss project-related issues in a 5 out of 6 CDCs discuss project related Community Developmnent Committees of frank manner and resolve problems issues regularly and openly. participating communes harmoniously. Extent of transparency and accountability All CDCs share information regularly with the All CDCs share information with the within the local decision making process of population and respond to possible population and the concems and complaints partcipating communes complaints by the population directly and of the population are taken into account properly. during sub-project execution. Level of satisfaction of beneficiaries with A significant majcrity of the populabtion The beneficiaries are very satisfied with the community-level project outcomes expresses clear satisfaction with the way the results obtained. They appreciated the project is designed and implemented, opportunity to identity their needs, translate regarding process (participation and these needs into sub-projects, and decentralization) and products participate fully in the execution of these (sub-projects). sub-projects. Assessment regarding the replicability of the A significant majorty of the population fully A majority of the population fully supports the project activities on a wider scale. supports the project (processes and project and most procedures have proven outcomes) and financial, technical and appropriate. Certain procedures can be administrative procedures at the central and improved without much effort. This approach local levels are ccnsidered appropriate is, thus, replicable in all sectors of PVK. (understandable, effective) by the population. - 25 - Output Indicators: IndicatorlMatrix Projected in last PSR Actual/Latest Estimate la. Number of sub-projects financed. At least 75% of planned sub-projects have 27 out of 38 sub-projects have been been executed. implemented (71%). 6 out of 7 infrastructure sub-project have been executed. 10 out of 20 micro-credit sub-projects have not been financed. lb. Comparison of community development At least 75% of funds are disbursed Not applicable. The implementabon manual plans and sub-projects executed. according to the community development has been adapted so that financial plan. management was handled by the Bureau de Coordination (BC) of PVK. 2a. Number of training activities undertaken Start-up training has been provided to all Two training sessions have been organized and of people trained. CDC members. for the 88 CDC members, including participatory rural appraisals, organizational, project and financial management, and environmental awareness. In addition, 177 beneficiaries of AGR received basic training in micro-credit management. 2b. Rate of approval of sub-projects. At least 90% of all proposed sub-project 38 out of 43 proposed sub-projects have have been approved by the CDCs. been approved (88%). All infrastructure (7) and capacity building (1) sub-projects have been approved. 2c. Rate of sub-projects with deviations. At least 90% of all approved sub-project have There has been no deviation in the been implemented without unjustified implementation of the 38 sub-projects. deviation. End of project - 26 - Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal ActuallLatest Percentage of Estimate Estimate Appraisal Project Cost By Component US$ million US$ million Civil Works 8.50 12.60 100 Equipment, Materials 4.50 0.81 100 Training & Studies 0.30 0.30 100 Consultants' Services 1.50 2.29 100 Operating Costs 0.90 1.20 100 Refund of P726 0.81 Comm Reintegr Activ 0.73 SPAC $600,000 IN -0.05 CRDP-Category 8 0.22 SP/AC@BNR 0.00 Unallocated 2.10 0.00 PPF 1.30 300 Total Baseline Cost 19.10 18.91 Total Project Costs 19.10 18.91 Total Financing Required 19.10 18.91 Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) Procurement Method Expenditure Category ICB NCB Other2 N.B.F. Total Cost 1. Works 0.00 4.70 4.30 0.00 9.00 (0.00) (4.70) (4.30) (0.00) (9.00) 2. Goods 3.50 1.60 0.10 15.40 20.60 (3.50) (1.60) (0.10) (0.00) (5.20) 3. Services 0.00 0.00 6.80 0.00 6.80 Studies, Technical (0.00) (0.00) (2.70) (0.00) (2.70) Assistance, Training 4. Miscellaneous 0.00 0.00 1.00 5.40 6.40 Incremental Operating (0.00) (0.00) (0.90) (0.00) (0.90) Costs 5. Miscellaneous 0.00 0.00 0.00 2.00 2.00 Micro-enterprise funds (0.00) (0.00) (0.00) (0.00) (0.00) 6. Miscellaneous 0.00 0.00 1.30 0.00 1.30 PPF Refinancing (0.00) (0.00) (1.30) (0.00) (1.30) Total 3.50 6.30 13.50 22.80 46.10 (3.50) (6.30) (9.30) (0.00) (19.10) - 27 - Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent) Procurement Method Expenditure Category ICB NCB Other2 N.B.F. Total Cost 1. Works 0.00 4.70 4.30 0.00 9.00 (0.00) (4.70) (4.30) (0.00) (9.00) 2. Goods 3.50 1.60 0.10 15.40 20.60 (3.50) (1.60) (0.10) (0.00) (5.20) 3. Services 0.00 0.00 6.80 0.00 6.80 Studies, Technical (0.00) (0.00) (2.70) (0.00) (2.70) Assistance, Training 4. Miscel(aneous 0.00 0.00 1.00 5.40 6.40 Incremental Operating (0.00) (0.00) (0.90) (0.00) (0.90) Costs 5. Miscellaneous 0.00 0.00 0.00 2.00 2.00 Micro-enterprise funds (0.00) (0.00) (0.00) (0.00) (0.00) 6. Miscellaneous 0.00 0.00 1.30 0.00 1.30 PPF Refinancing (0.00) (0.00) (1.30) (0.00) (1.30) Total 3.50 6.30 13.50 22.80 46.10 (3.50) (6.30) (9.30) (0.00) (19.10) Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies. 2 Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. - 28 - Annex 3: Economic Costs and Benefits Not Applicable - 29 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, I FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation September 1991 6 - Jacques Baudouy, TTL - Eileen Murray, Financial Analyst - Guislaine Delaine, Statistician - Qaiser Khan, Economist, - Kees Tuinenburg, Economist - Marco Ferroni, Economist Appraisal/Negotiation December 1991 8 - Jacques Baudouy, TTL - Eileen Murray, Financial Analyst - Guislaine Delaine, Statistician - Qaiser Khan, Economist, - Sven Hertel, Engineer - Kees Tuinenburg, Economist - Marco Ferroni, Economist - Harley Johnson, Community Development Specialist September / 6 - Qaiser Khan, TTL, October 1992 - Eileen Murray, Financial Analyst - Guislaine Delaine, Statistician - Sven Hertel, Engineer - Tonia Marek, Consultant - Raymond Martin, Consultant Supervision February 1993 2 - Jacques Baudouy, TTL U HS May 1993 2 - Jacques Baudouy, TTL U IHS - Eileen Murray, Financial Analyst - 30 - July 1993 6 - Eileen Murray, TTL S HS - Christian Rey, Implementation Specialist - Jayshree Balachander, Nutritionist - Kishor Uprety, Lawyer - Qaiser Khan, Economist - Thierry Brun, Economist December 1993 2 - Eileen Murray, TTL S S - Thierry Brun, Economist PSR Update - Eileen Murray, TTL HU HU without site visit on June 1994 PSR Update I - Eileen Murray, TTL HU HU without site visit on September 1994 November / 4 - Eilleen Murray, TTL U U December 1994 - Francois Munyantwali, Operations Officer - Hoang Dam Van, Consultant - Marc Van Imschoot, Consultant April 1995 1 - Eileen Murray, TTL U U September 1995 2 - Eileen Murray, TTL S S - Frangois Munyantwali, Operations Officer April 1996 3 - Menahem Prywes, TTL S S - Eileen Murray, Operations Officer - Fran.ois Munyantwali, Operations Officer PSR Update I - Menahem Prywes, TTL HS HS without site visit on June 1996 February 1997 1 - Menahem Prywes, TTL HS HS June 1997 - I - Menahem Prywes, TTL HS HS Mid-Term Review December 1997 1 - Menahem Prywes, TTL HS HS - 31 - April 1998 2 - Menahem Prywes, TTL S S - Toni Kayonga, Operations Officer August 1998 2 - Menahem Prywes, TTL S S - Toni Kayonga, Operations Officer November 1998 2 - Menahem Prywes, TTL S S - Toni Kayonga, Operations Officer April 1999 2 - Markus Kostner, TTL S S - Michelle Cullen, Social Analyst September 1999 3 - Markus Kostner, TTL S S - Toni Kayonga, Operations Officer - Prosper Nindorera, Procurement Specialist PSR Update I - Markus Kostner, TTL S S without site visit on March 2000 June 2000 2 - Markus Kostner, TTL S S - Toni Kayonga, Operations Officer October 2000 5 - Markus Kostner, TTL S S - Toni Kayonga, Operations Officer - Marie-Claudine Fundi, Disbursement Assistant - Antoinette Kamanzi, Team Assistant - Habdul Haji, Financial Management Specialist ICR September 1999 4 - Markus Kostner, TTL S S - Toni Kayonga, Operations Officer - Michelle Cullen, Social Analyst - Habdul IHaji, Financial Management Specialist - 32 - June 2000 3 - Markus Kostner, TTL S S - Toni Kayonga, Operations Officer - Michelle Cullen, Social Analyst March 2001 5 - Markus Kostner, TTL S S - Toni Kayonga, Operations Officer - Marie-Claudine Fundi, Disbursement Assistant - Antoinette Kamanzi, Team Assistant - Juvenal Nzambimana, Disbursement Analyst (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 68.00 141.00 Appraisal/Negotiation 80.10 158.30 Supervision 136.40 291.50 ICR 12.30 24.50 Total 296.80 615.30 - 33 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating O Macro policies O H OSUOM O N * NA Oi Sector Policies O H OSUOM O N * NA Z Physical O H *SUOM O N O NA El Financial OH OSUOM ON *NA N Institutional Development 0 H 0 SU * M 0 N 0 NA OI Environmental O H OSUOM O N * NA Social N Poverty Reduction O H OSUOM O N O NA O Gender O H OSU*M O N O NA M Other (Please specify) O H * SU O M O N O NA Social: The use of participatory rural appraisals greatly improved the ownership of the activities carried out under phase III. F Private sector development O H O SU * M O N O NA Oi Public sector management O H O SU O M O N * NA Fl Other (Please specify) OH OSUOM ON *NA - 34 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfaczory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating E Lending OHS OS OU OHU O Supervision OHS OS *U OHU Oj Overall OHS OS O U O HU 6.2 Borrower performance Rating O Preparation OHS OS O U O HU El Government implementation performance O HS O S * U 0 HU E Implementation agency performance O HS O S 0 U 0 HU O Overall OHS OS OU O HU -35 - Annex 7. List of Supporting Documents 11/91 Final Executive Project Summary 1992 Staff Appraisal Report 8/95 Budget-Program BCP/PNAS 9/95 CDC first meeting notes 8/95 FSU manual of procedures 2/96 FSU manual of procedures 1/96 PNAS Trimester Report 3/96 PNAS Trimester Report 5/9/96 Back-to-Office Report 6/96 PNAS Trimester Report 8/96 PNAS Trimester Report 9/96 CDC meeting notes 10/96 Trimester Report 1996 PAN Annual Report 1/97 Trimester Report 2/97 CDC meeting notes 4/97 Socio-economic Review 4/97 Aide-memoire 5/97 Mid-course review aide-memoire 5/14/97 Back-to-office report 7/97 CDC meeting notes for mid-project review 11/97 CDC meeting notes 4/98 CRDP implementation manual 7/98 PNAS Trimester Report 8/98 Overall Results on PNAS and outline of AGETIR 12/98 CEPEX Inventory 9/99 Aide-memoire 6/00 Aide-memoire 10/00 Aide-memoire 3/01 Aide-memoire 12/99 Beneficiary Assessment for phase II 01/01 Beneficiary Assessment for phase III (evaluation par les beneficiaires) 01/01 Final implementation review (evaluation de la mise en oeuvre) 4/01 Rapport d'Achevement du Projet Programme National d'Actions Sociales (PNAS), CEPEX 5/01 Rapport d'Achevement du Projet Programme National d'Actions Sociales (version resumee), CEPEX Forms 590 and PSR Audits for 1994-2000 Phase III quarterly progress reports (June 1999 to December 2000) - 36 - Additional Annex 8. PNAS/CRDP-PVK - Cr. 2388-RW - Completion Report REPUBLIQUE RWANDAISE Mirstie des Finances et de la Planificatlon lconomique Bumeau Central de& Projets et du Fhnancement Ext6rkur (CEPEX) PROGRAMME NATIONAL D'AiCTiONS SOCIALES (PNASI PROTET "COMMUNITY REINTGRATION AND DEVELOPMENT PRO!ECTI lCRDP-FVXK (Cr6dit IDA 238:RW) RAPPORT ACHIEVE (Version r6sum6e) :. Contextt. Obiceifs et Mise en euvre Le Programme National dActions Sdiales (PNAS) a k6 concu par le Gouvernement Rwandais pour )utter conbte la situation de dktrioration des conditions sociales du pays. Dans ce cadre, le Gouvernewwnt a nftoci6 et obtonu de l'IDA un credit d'un montant de £TS 14 000 000 dont raccord a t signe le 12 aolat 1992. Le Projet "PNAS" (categories I A 7) a ete si6ture le 31 d&eenbre 1998. Apres la guerre et le g4nocide de 1994, les ob0ectifs du Projet ont dtk reformules et ont 6t6 d6finis cLomnme suit: (a) arneliorer I'ttat nutritionnel de certains groupes dkmunis et %Mln6rables de la popuIation; (b) soutenir des initiatives de redemarrage des activit6s communuautaires en rChabilitant les infrastructures sociales et economiques; (c) encourager l.a participation des populations a des activit6s productives a petite Cchelle; (d) creer des emplois aussi bien temporaires que durables. Ces objectifs etaient tres valables et correspondaient tou.t a fait A la situation catastrophique que coruiaissait le Rwanda pendant cente p6riode dapr*s le genocide et ils constituaient des conditions suffisantes pout redemarrer le developpement du pays. 11 !tait en outre devenu necessaire d'qtendre la zcne d'action du Projet A tout le pays et la population beneiiciaire devait itclure Cgalcment les personnes affecteeq directement ou indirectement par la guerre. Par aifleurs le besoin de repondre & des deinandes importantes de reconstruction ou de rehabilitation dinfrastructures et equipements sociaux se faisait sentir. Le Projet redefini est rests avec deux composantes. la Securite Alimnentaire et Nutrition et le Foonds Social d Urgence. La composante Skcurite Alimentaire et Nutrition avait pour groupe-ible les enfants mal nourris de 0 A 5 ans et leurs families, ains que les femmes enceintes et allaitanLes A faible - 37 - revenu. ls 6taient touches A travers le r4seau des services nutritionnels. LAs faxnilles victines du SIDA ont ete Cgalement assistees. La composante Fonds Social dUrgence devait financer tout d'abord la rhabflitation dtes infrastructures soaales (ecoles, centres de sante, am6nageaent de sources et adduction d'eau) mais egalement la rehabilitation de pistes rurales, de ponts, de marches, de reboiseuwnts etc. (Volet "Infrastructures de Base") Lk Fonds a finance en outre des activitts generatrices de revenus en se faisant appuyer par des Agences d'Encadrement (Volet -AGR') Dans le cadre du Volet "Infrastructures de Base", qui d consttue le gros du travail du PNAS, le Proet a realis6 393 sous-prjts (277 de rehabilitation et equipement scolaire; 44 d'am*r gement de soumces d'eau potable; 15 de rehiabilitation de centres de santc; 26 de reboisemnent; 16 dadduction d'eau; 5 de r6habilitation de centres d'enfants non accompagnes -CENA-; 4 de r&habilitation de rnarches ruraux et 4 de r6habilitation de piates) pour un montant total de 3.741.122.213 Frw. Danm le cadre du volet 'ActivitOs Generatrices des Revenus", 211 petits projets ont bWn6ficie d'un finanement (cr#dits remboursables) du PNAS. Le monitant total des crddits octroyes s'elevait A 12S.248.570 Frw dent seulement 82.469.526 Frw (58% du rnontant total d(i y compris les int4rits) Ctaient rembourses jusqu'en septembre 1998. It est ai noter que ce pourcentage est atteint grAce au remboursement total du montant accorde A I'Union des Banques Populaires du Rwanda (UBPR) pour lexecution d'un progranume de petits cTedits. Le Progrannw Alimentaire et Nutrition a distribue 8-465 tonnes do vivres A SO services nutnitonnels et A 40 Centres pour enfants non accoumpagnes (CENA), et a dispense des fornmations A des agents de ces services. Le Bureau de Coordination du PNAS (BCP) a en outre organis6 des sessions de formation a l'intertion des cadres des Petites et Moyennes Entreprises et des Buzeaux d'Etudes qui travaillaient en etrite collaboration avec le PNAS aisi qu'A l'intention de ses propres cadres at dls aniniateurs d'associations villageoises. Au total, '19 personnes ont pu beneficier des formatins du PNAS. La qualite et I utilit6 des infrastructures construites sont jugees en ,n&eal tres satls(aisates. LIs enquetes ene1es et les visites effectuecs ont confirmn quo ies responsables et les benefidaires sont unanines a se declarer satisfaits des travaux entrepris et, plus particulieremenit, de l'appui S rnisterdiaii qui leur a 01t; octroye. 11 y a pourtant certains cas oui les ouvrages auraient dti rtre adaptcˇs A la situation du nilieu (par exemple le marche de Rurare). Tn est en outre regrettable que le plus souvent des gouttieres ainsi que citernes pour recueiltir 1'eau des pluies n'aient pas 4dtC prevues dans les constructio0s. De m6me, les bureaux des directeurs des centres scolaires ain6i que des cdotures n'avaient pas Otis prCvues. LUs audits techniques ont m-ontre que les normes de 4coit par classe cornstruite cu r&habilit-, pax source aenagsee, par hectare de foret rehabihitee sont relativement faibles et qu'ils faisaient pensex que les marges bt&ficiaires des MIE (Petites et Moyennes Entreprises) Ctaient plut6t faibles: I'accession aux appels d'offre du PNAS etait pour la pitipart des PME le ii - 38 - moyen de se faLre connaitre et dftnerger sur le naurchb du BTP (Batiments et Travaux Publics) que de faire dec bMEnfices substantiels. La part des htudes et de lencadrement (incluant le travail des cadres du PNAS devant se substituer A des AE defaillantes en 1996 et 1997) Otait retadvemnent faible: 7% du coot total. D'une fa(on g4nkrale, on constate que les sous-projets s6lectionrns ou rejetes par le PNAS rondaient respectivement aux crithres d'ligibiite et de rejet dEfinis dans le Manuel de Procdures. Tl est vident que certaines autorit*s coanuunales dynaniques ont quand mOme pu en profiter plus que d'autres. La passabon des marches a OgaIement suivi lee rtgles ainsi que les recommandations trranant des difterents audits 11 est canstat en outre que le codt par sous-projet et les coots unitaires pour les diff6rentes catgories d'infrastructures est tres apprciable gr6ce aux procdures du PNAS (forte concurrence et pateent rapide des attachements). Le suivi des travaux par le BCP a ete tres regulieT et cela explique aussi la qualite des r6alisations. Plusieurs bln6ficiaires ont manifesat leur satisfaction particulierewaunt a ce sujet, d'autant plus qu'ils faisaient une comparaison avec certains autres projets ayant effectu6 des travaux sindlaises. Les se9sions de formation dispensee par le PNAS aux cadres des BE (Bureaux d'Eudes), PME et AE (Agences d'Encadrement) ont sensiblelment contribu4 A I'amslioration de l'ex&cution des travaux. Le PNAS a pgalernent forne lees promoteurs bneficiant du programme de petits cr6dits nus en place avec le MIGEFASO (Minist*re du Genre, de la Famille et Affaires Sociales). Mais malgr6 la formation additonnefle dispens6e en- comptabUite simplifi&. Ia tenue des livres d'achat de pToduits grAce au cr&dit octroy6 et de recettes est reset faible. R1 est regrettable que le BCP n'ait pas 4labore unl rapport final englobant toutes les ralisations du Irojet ainsi tousles autres &l6ments en rapport avec son execution. A la dOture du Projel "Programme National d'Actions Sociales", categories I a 7, Ie Gouvernement Rwandais a demandb A la Banque Mondiale d'utiiser IY reliquat de son cr6dit pour lancer le projet-pilote 'Communlty Reintation and Development Project of PVK (Prtfecture de a ViUle de Kigali), celui-ci devant sealenent op4rer danIs la Prfecture de la Viule de Kigali. Le nouveau projel a etW approuv4 le 16 d6cembre 1998 pour Atre finance par les reliquats du PNAS d'un montant de DTS 8K.717 @catgorie 8 du credit IDA 2388-RW). Le CRDP-PVK a etc dotur6 le 31 d6cembre 2000. Les objectifs du CRDP-PVlC etaient d'aider les rapatnes et autres groupes vulnerables par un processus de r6insertion et de developpement communautaire et de mettre les communautbs locales et l'administration au niveau communal et prefectoral naeux A m6me d'ex6cuter des projets de d&veloppement. A co6t d'un programme de formation intensive basee sur la "MOthode accldr6e% de recherche participative4 (MARP) destinCe surtout aux membres des "Comidtes de DeveIoppement Communautaire' (CDC) des 6 secteurs de la FVC r,etenus comnie zone d'intervention du iii - 39 - Prjet, A savoir Nyakabanda, Kimisagara, Gisozi, Kagugu, Gatenga et Kimisange, le Projets a rtaHis6 Jes sous-projets d'infrastnuctures socio4onoriques dans anq des 6 sectmes ainsi qu'un programme de finan-ement d'activi*s gn6raties de revenus dads le secteur Nyakabanda. Le6 Ssous-projets d9nfrastructures ont tt r#alis6s ummre suit: - Constructon d'une salie polvvalente A Kagugu; - Extension de lE'cole Primaire de Gasave; - Construction dune Ecole Matemelle et Extension de l'Ecole Prinaire de Kamuhoza - Extension de l'Ecole Primaire de Gatenga; et - Rehabilitation de l'Ecole Primaire de Kabusunza. Dans tous les 6 secteurs participants, des sessions de formation ont ete organisees en faveur des membres des CDC. Ainsi, 88 personnes ont reCu respectivement une fonation en kiucation cooptrative, en organisalion sodale, en entreprenariat ainsi qu'en comptabilim! de base. En plus, 173 Wnkficuaires du programme d'appui aux activit6s gCn6ratrices de revenus (micro-crkdit) ont reiu une formation en gestion de micro credit. Des plans de developpernent cowununautaire ont t Claborks pour tous [eS 6 secteurs participants. La methodologie utitisee pour lelaboration de ces plans, la MARP, a pernis la participation active de la population qui s'est sentie reunie et solidaire autour d'objectifs cornmuns Ces plans sont naintenant disponibles et peuvent etre prOsentes aux differents bailleurs de fonds pour requte de financement. D'une fa*on particuhere, cette activit6 de "renforcement des capacites" entreprise par le CRDP-PVK a permis aux CDC des 6 secturs participants de sortir d'un certain anonymat el de mieux pouvoir assurer la mission qui leurs est confiee. Cependant, i aurait fallu que les cadres de planification et les consultants expliquent davantage aux participants que tea plas de dveloppement communautaire OlaborA ne constituaient pas les premi6res pnorites devant ktre financks absolument dans le cadre du Projet CRDP-CVK, particulierement pour les secteus qul avaient comme IP-priorite lapprovisionnement en eau potable. Les travaux de construction et de fournilture de mobilier ont gts realises par des entreprises locales apres appels d'ofrfes. Les CDC y ont eglenent particip. D'une fa.on gen&ale, tes travaux ont Lte solidement effectues. 11 est particulierement remarquable que ]es travaux ont it6 le mieux ex6cut#s dans les secteurs w4 les CDC 4taient les plus dynamiques. Certaines imperfections qui ont W constat6es sent de moindre importance et pournaient Otre corrigOes pendant la p6riode de garnmnie. De toute facon, il est remarquabie que les travaux executes sont nettement de meilleure qualitE que ceunx du Projet PNAS (catfgories I - 7). l.es travaux de rehabilitation ont ete egalement rdaIis4s par des entreprises locales apres appels d'offres ct les CDC y ont egalernent participe. Pourtant pour les deux sousprojets, les bnneficiaires ne se sont pas sont montrfs tout a fait satisfaits des realisations. Apparemnunt ils n'avaient pas etc associ6s A la prration du cahier des charges et ills ne l'ont pas suffisamient Ws lots de l'ex4cution des travaux. -0v - 40 - Concernant ractivit6 'Financement de projets ginbtateurs de revenus (Fonds de Credit Rotatif)"un seul secteur, celit de Nyakabanda, a bWnOfici6 de cette composante relative A I'octroi dc micro-cK-dits. En effet, c'est le seul secteur qui lavait place daLs ses priorit6s. 11 s'agit dbn type de cr#dit rotatif destins aux petiles assciations et goupements avec un systeme de cau lionnenient soliaaire. 11 est regrettable que la sonune prevue a cette actviti n'ait ete enti6rement liberec, d'autant plus que le micro-cr&it constitue un domaine dactivit#s pouvant toucher rtellement les persoiines vulntrables, et de cette facon pouvailt jouei un r6le preponderant au niveau de la rMintegration. Appareinment il y a eu des malentendus i difforents niveaux et le choix des associations bentficiaires du niicro-cridit par le CDC du secteur il! se serait pas effectue sans probl6mes. Cependant Ia condition du CRDP-PVK formulde A 11'adresse du CDC de Nyakabanda de d&bluuer la deuxieme tranche apTes justification de l'utlisation de la lee ne paralt pas convainquante et realiste, d'autant plus qu'il restait trOs peu de temps jusqu'a la cl6ture du Projet. 11 est vrai que le micro-credit constitue une activit6 a risques dans beaucoup de projets de d6veloppement, particulierement A cause de la mauvaise comprehension de ce systeme paT la population rwandaise. Cependant, grAce A la coullaboration du Projet avec ulne nouvelle structure de gouvernance d¢ralis&e telle que les CM, ce type dXactiviWe deviait plutOl prendre de l'ampleur et etre mnme itendue sur tous les autres secteurs. Selon les documents disponibles, le budget total utiWsb par le CRDP-PVK s'dlIve a Frw 39.406.056 equivalent A USD 982.241 sans compter les dettes du PNAS aux fournisseursm Le montant utilis6 pour la categorie "sous-projets" est de Frw 198.685.000 (USD &30.746), constituant ainsi 64,2% du budget total, ce qui en fait la composante budgttaire prin4ipale. La repartition du montant a affecter A chaque seteur destin6 A Pex&!ution des sous-projets avait ete fix6 A une moyenne de USD 100.000. Pourtant les diiftrences entre les secteurs, m6me s'ils font tous partie de la "Ville de Kigali" est Lnonne A differents niveaux. infrastructures existantes, ressources huinaines, potentiel 6conurruque, etc. Certains secteurs couime Kimisange tw sont pas trfs peuplas et n'ont pratiquement rien de la "Ville"! Ce facteur aurait do etre pris en consideration avant de &eterauner le monitant A affecter A chaque secteur pour ]'ex6cution des stus-projets. Dan l'ensemnble, la performance du Projet CRDP-PVK dans la rnalisation de ses activites est jugee satisfaisante. surtout si on tient compte du fait que le ternps qui lui etait imparti pour la rt.alisation de toutes ces activitds elait seulemen£t de 20 rnois y comnpris une longue p6riode de sensibilisation de la population et des autorit6s administratives A 1a nouvelle approche - ce qui etait par ailleurs absolument incontournable. v - 41 - 2. ItQbILges maieurs rencontris tars de l'ex&udacn du urgrpatnin * Certaines entreprises n'ont pas pu honorer leur contrat et ont dal atre remplacees. D'autres ont certes ex6cut# les travaux mais n'ont pas correctement r4mun6r* leurs employfs; n 1I a 6tE constat6, surtDut au d6but, que peu de PME et BE marlbisaient les systSmes de gestion des contrats, de gestion d'entreprises, des techniques de constructior, dl'adducdon d'eau et routiEres, C'est pour cette raison que des sessions de formation ont e dispentses aux cadres; * Des faiblesses ont 06 constat6es: mecanismes de concertation aux niveaux central, prMfectoral et communal peu structur6s; manque d'expertise technique des niinisteres et des collectivitfs locales au niveau du trraln; pauvre coordination de l'ensemble des baiLleurs de fonds et des ministlres prenant au Projet; * D'une facon particulibre, une defaillance a & constat6e chez la majorite des Agences d'Encadrement (ONG internationales et locales) charges d!appuyer les promioteurs dans le cadre des AGR. La consequence de leurs faibles capacites techiiques caractrskes par un manque de suin des realisations et de la performance des ricro- entreprises cr&es a eti le faible taux de realisation de prcets d'AGRI le faible taux de rmboursement de credits et le dinournernent d'affectation des credits. jStgre 8 * Le manuel de procedures utilise pour le CRDP-PVK est celii qui avait 4tk pr6pare pour le CRDP-Minaloc. Cependait, nnmme si lea ob$ectifs de ces deux projets sont idenfiques, it y a beaucoup de dissemblances qui rendaient difficiles l'application des diffrentes procdures dte ce umanuel la demande de noK-objection auprks de la Banque Miondiale dtant n&cssaire tars de chaque cac d'inapplicaoion * 11 avait ele pr6vu que l'Agent de DNveloppement Communautaire (ADC) soit affectW auprbs du CDC, mais la r#aIit6 a fail que l'ADC soit au service du Projet. Ceci a eu comme consequence que les rapports prisentks par I'ADC soient faits pour le compte du Projet plutOt que du CDC; * Plusieurs activit6s ont ete rkaliskes avec des retards considtrables surtout A cause des tongues proc6dures de passation de marchk au National Tender Board (NTB). * Concernant le syst-ime de passatiuons de marche particutiOrernent, ia Banque Mondiale rnavait pas fixO un montant minimal pour lequel la non-objection n'dtait pas obligatoire comme cela avait in* le cas pour le PNAS. Corollairenient A cette Situation, les attributions du march6 devaient recevoit sa non-objection; et comme il ny a pas de spOcialiste en passation de march6s a la Mission Rtsidente de Kigali, to"s les dossiers devaient etre transnis a Bujumbura. 3. Performance de la Danque Mndiale * Plusieurs fois par an, la Banque a envoyk des missions de supervision. Celles-ci ont effectu4 des visitos de tenrain, ont mene des discussions avec les autoritls du pays et diff6rents concern6s, se sont rendu compte de l'ktaL d'avanement du Projet et ont donne leurs observations et reconmnandations a travers des aide-memoire, ce qui a itn d'une grande utilite pour le bon avancemenit des activites d u Prqet; vi - 42 - * Le suivi du Projet par la Banque a travers les audits financiers et techniques ainsit que l'(valuation A mi-parcours et l'etude sur lPimpact socio-econonique du PNAS ont 6th egalement d'une grande ulilil pour le bon avancernent et lamdlioration des actiitis du Projet; * La Banque a tacilitW au BCP le dkcaissement des fonds, celui-.i sest effectue sans probmnws unmjrs; * A la reformulationi du Projet apres la guerre et le gfnocide de 1994, la Banque aurait du donner des orientations plus claires quant A la part quhe devaient avoir les diff6rents types de sous-prets d' 'lnfTastructufe de base" en pr6cisant la part des sous-profet3 sociaux et celle des sous-projets 6conomiques" (routes, march&-%, etc.). cat6Gorie B * Sur demande du Gouvernement Rwandais, la Banque Mondiale a releve le plafond du compte special qui est passe de USD 70.000 A USD 250.000; * Apres la date de cloture du Prlet, la Banque Mondtiale a accord6 un dilai de grace de 4 mois au Bureau de Coordination, ceci lw a pernis d'avoir le temps suffisant de cltrer tous les dossiers; * La Batque Mondiale a donu sa non-ot ection pour le recrutewnent dun comsultant assistant a la gestion. celui-ci est entre en fonction au mois d'aodt 2000. Ce consultant a reiforce I'quipe de coordination pour accomplir les thches urgentes de la p6riode de cl6ture du Prc4et; - La date de cl6ture du Projet etant fiuxe au depart au 31 d&eembre 1999, Ia dur6e du Proet a connu une extension et la cl6a,re a eu lieu le 31 cl&embre 2000. Etant donnc l'ampleur des activit4s et la nouvelle approche du Projet qui devait etre suffisamrent expliquee A la population, ta Banque Mondiale aurait dd ac-order une extension supplenientaire d'au moins une annee, d'autant plus que la 1- date de cifture nYtait pas du tout r6aliste. 4. Perfornmance de V'EtipruntsA4 * Le Gouvememnent Rwandais a mntitu6 le cadre legal du PNAS par 'ArrOt4 du Premier Ministre du 21 avnl 1993; * Le Gouvernement a mis sur pied i un Conseil de Coordination du PNAS qui etait charg4 de d4finir et d'orienter la politique generale du Projet et d'en assurer la supervision. Le Conseil a bien rempli ses fonctions; Le Gouvemement a payr comnme contrepartie la sonmme de 65 millions de Frw en 1993, depassant la sormn'e de 116 000 DEn qui devait Otre pay6e d'aprfs le demnier amendement de I'Accord de Credit (16. 12.1998); * I-es autoites du pays ont engage des poursuites judiciaires envers certains entreprepeurs qui n'ont pas honord correctement leur contrat, particuliremient envers ceux qui riont pas remun6re les ouvriers. Le CEPEX suit de trLs pres T'evolution des differents dossiers. QC116garie 8 * Le Gouvernement rnvandais a aide A la mise en pLace des conitWs de developpement cornmunautaire qui sont des organes decentralis6s de Itadministration A I'4cheWon local. Les CDC des secteurs ont W la base nmme des realisations des activit6s du Projet CRDP- PVK; * T.a PVK a nnis A la disposition des CDC et de la population b6ndficiaire des parcelles pour l'nstallation des infrastructures (Extension de l'Ecole Primaire de Gasave, Construction vii - 43 - de l'Ecole Matemelle de Kamuhoza) et eUle at fsdlitt au secteur Kagugu l'acquisition du terrain ot la salle polyvalente a tt costruit.; * Avec l'aide du Ministe chargg des Travaux Publics, la PVK a amenage la route qui m6ne A la nouvelI 'cole matemelle de Kantuhoza et elle a participe aux travaux de terrassernent powu la contruction de cette kcole; • Le Gouvernement rwandais, A travers le Ministere des Finances tA de la Planif)cation Economique (Minecofin), le Minasthre de FAdministration Locale et des Affaires Sociales (Minaloc), le Bureau Central des Projets et du Fi,iancement Exterieur (CEPEX) et particulirenment I'Administration de la Prefecture de la Ville de Kigali (IJK), a suivi de pr&s et nigulierement lexecution du Projet, et a dorune des suggestions et conseils visaL le bon avancementt des activites. S. Durabilite et Viabilite des g6alisatiorm du Projet D'une faKon grn#rale, le PNAS a r6alise solidement ses travaux, particulikrement dans le cadre du volet "Infrastructures de base". Les r6alisations du PNAS, vu leur qualite, pourrant slitement Otre utlisables pendant plusicurs anrees. Pourtant il sera n&essaire que la population berfiaire continue A Otre organiske et sensibilis6e de faqon b supporter les charges r&urrentes et l'entretien r6gulier des infrastructures: ereation de comit6s de parents actifs pour les koles, creation de regies et de comit6s de consonmuateurs pour les adductions d'eau et les sources am6nag&es, etc. Dans la plupart des communes, ces systemes existent et sont operationnels. Parni les petites entreprises qui ont particip6 aux diffOrents travaux de r6habilitation, il y en a celles qui ont le potentiel de devenir de moyetnes entreprises et de la contribuer davantage au developpement econoniique du pays. De plus, les cadres des PME et BE form6s ont acquis des connaissances qui leur ont permis d'amrliorer durablement leur perfornance technique. Comme cela est exphque ailleurs danm ce document, 1'&hec relatif du systeme "Activites GMenratrices des Revenus1 d0it trouver tune solution A une echelle plus Mlevee: des rffle,ions, des concertations devraie.nt Qtre menkes au niveAu natioual pour mnettre sur pied un systeme efficace et efficient de petit credits A octroyer au monde rural. Cot6aorie 8 * L'approche appliqu&e lors de rexecution est celle de la dcentrahisation et de la participation communautaire qui est prOnOe par le Gouvernement rwandais. A travers cette approche, la population b6ndficiaire participe directement a la prise de d&cision et a la r6alisatdon de ses propres prWets de developpement, et par-IA elle se sent plus motivke et responsabilis&e que dans lc pas. En plus, les fonnations -de bonne qualit6 et bieo approprikes pour ic developpement A la base- que les mnembres des CDC ont reiues devraient etre utiles A long terme movennant quelques sessions de recyclage et de rafratchissenient; * Les infrastructures nouvellement construites sont solides at les travaux ont 6W en gneral bien ex6rutes. Les ben6ficiaires rnalrisent dejk le systwne de gestion des kcoles primaires, les comit0s des parents et de direction Mant en place. Par exemple pour lh4cole maternelle de Kamuhoza. les bWnficiaires ont JAjA nris sur pied un syst4me de gestion tout A fait particulier et qui apparemment, fonctionne tr6.s bier. Quant A la salle polyvalente de Kagugu, il faudra ahsolumient que lea responsables administratifs a tous viii - 44 - les &helons veiTlent A ce que les b&i4ficiaires accomplissent leurs pronesses, en particulier il faudra qulun syst6me d'approvisionnement en eau pour la salle soit installe au plus vite, sinon la salubritk nkessaire pour un le] edifice ne sera pas garantie. Conune dans d'autres projets similaires, un systeme de micro-credit r4ellement fonctionnel n'a pas pu Otre mis en place et, avec la cl6ture du projet, il ne paralt pas probable que e Londs de d4veloppement comnmunautaire puisse voir le jour. 11 faudra des efforts exceptionnels de la part des membres des CDC pour recouvrer les credits d6ja octroyes et les red istribuet. Les autorit&s adn-inistratives sup6rieures devront y jouer Uil grand role. Dans 1'ensernble, la viabil;ite des r6alisations du aCDP-PVK danm les 5 secteurs semble garantic, la population binificiaire ayant rMellement contribuw aux activit6s de le-ur misw Pn place. Particuliarement les infrastructures socio4conomiques seront utiles A long terme, leur systWme de gestion etant bien maItrisi par les binficiaires. Les acquis des sessions de formation semblent egalement pouvoir #tre utis pour d'autres activites de planification - msme domestiques, comme cela nous a et indiqal& moyennant quelques sessions de rafraichissement. Lc domaine du micro-credit devrait vcuir le systime am6lIorM et, cornme il a et4 propose pour d'autres projets. des concertations devraient etre men4es au niveau national pour mettre sur pied un systeme efficace et efficient de petits credits a odtroyer au monde rural et aux petites associations, 6. Leeons tirecs et Recomandations • Un Fonds Social d'Urgence tel quewK celui qui a 'tA mis sur pied par le PNAS n'est pas necessairement l'instrument le plus approprik pour faire le type de crit "AGR' s'il ne trouve pas de partenaires de rel"is capables, comp6tents, motvAs et viables. De plus, il n'est pas conseilk de faire ce programme de credit en rAgpe, parce que d'une part cela n6cessite heaucoup de personnel pour assurer un suivi rapprochA indispensable pour reussir et que, d'autre part, les credits sont en principe beaucoup m;ieux geres par des institutions plus proches de ta population tlles que les anues Populaires et les Conmit6s de Developpement Coinnunautaire (CDC). L'efficacite et Putilite des Agences d'Encadrement devraient Otre serieusement evalu&es avant de leur confier des thches telks que celles du domaine des "ActivitAs GMerratrices des Revenus'; * Ie PNAS a pu Atablir un bilan globalement positif en particulier dans les doamunes dinfrastructures communales de petite et moyertne taille faisant appel aux PME et aux BE locaux. Partant de cette experence, une nouvelle structure permanente du type "Agence d'Ex&ution de Travaux d'InterAt Public" (AGETIP) devrait etre mise sur pied pour continuer dans la r-nme voie, Une alternative au systome d'exkcution de travaux A travers uime AGEI'IP serait une impfication plus Intensifie des CDC des communes, en leur pourvoyant de moyens de travail ainsi que d'une formation n&essaires; * La formation, l'information et la concertation doivent Atre utilisees pour rendre plus efficaces les operateurs pnves et pour ama*liorer leurs relations de partenariat avec les admninistrations; * Certains sous-projets devraient etre adaptes a la realitA du milieu, par exemple le type de marchOs ruraux ne devrait pas 6tre absolument le meBme que celui des marches construits en ville, 45 - * Au niveau national les responsables communaux devraient etre encore rnieux &clairts et informes tout au long du cycle d'un tel projet pour qu'ils puissent en btn6fider d'une faoon plus equitable; * 11 serait tr*s utile de prevoir des installations pour recueillr les caux de pluie, plus particuliOrement dans les regions arides comme l'Umutara. Egalement les bureaux des directeurs des centres scolaires devraient Otte prevus dans les plans des constructions, * Une comnission devrait atre mise sur pied pour faire le suivi des litiges et de tous les autres dossiers non achevts, du remboursement des petits credits octroy6s dans le cadre de la collaboration avec le MIGEFASO (les sommes rembours6es devraient fte vers*es au Tresor Public) - le CEPEX a d4ji commenLe k entreprendre cette action. Cat6gone 8 * Les fornmations dispensees aux nmembres des CDC ont W apprkclies par les b6n6flciaires et par les autorit6s. les r6sultats se sont fait sentir au niveau de l'elabora lion des plans de d4veloppement. II est souhaitable que ces fornations soient complMtees par des sessions de recyclage et de rafrajchissement en y incluant des modules compl6mentaires; * Les meilleurs resultats ayant etc constatt dam les secteurs oCi les membres des CDC travaillent avec beaucoup d'abnfgation et de d6vouement. les autorits admninistatives sup6rieures devxaient davantage en prendre compte lons des cainpagnes de sensibilisation de ia population et du choix de candidats aux diffkents postes; * II est souhaitable que les documents et plans de d6veloppement comnmunautaire soient traduits en kinyarwanda afin qu'iLs soient cotnpris de la mfte faqon pat tons les b6n6iiciaires; * Les membres des CDC devenant de plus en plus sollicies dans le cadre de l'execution des projets de d&velopperrent, un syst6me de travail les rendant plus disponibles devrait etre etudie par le Gouvenement, 6ventuellenent un systeme qui leur accorderait des hmoluments pour le travail effectue. 1Yautre part, vu l'indisporubilitd souvent observoe des mnembres des CDC, la possibilite de mettre en place un "secr6tariat' permnanent des CDC devrait stre 0tudike; * Dans la mesure du poosible, los ~ikes devraient ftre plus impliques dans toutes les Gtapes de 1'ex&ution de sous-projets des la phase de preparation (cahiers des charges, passation de marches) pusqu'A la phase de l'execution des travaux. Etant associes, ils comrprendraient rnieux les u*canismes et les complexit,s de Nexkution des projets, et cela pourrait aussi les inciter i accroltre leur niveau de participation; * Le Prejet CRDP-.IK ayant OtX r6alise comne profet-pilote ceuvrant seulement dans six secteurs, et en considerant les resultats atteints particuliWereent concemrant la prcparafion des plans de developpement cornrmunautaire et le 'reveil" des CDC des secteurs concernes, la PVK, avec l'appui du Gouvemernent, devrait prendre contact avec les bailleurs de fonds pour 1'ex6cution d'un projet sinilaire couvrant tous has secteurs de la PVK. -46 - Additional Annex 9. Beneficiary Survey Methodlology and Results Methodology The Beneficiary Assessment (BA) brought together project stakeholders and initiated dialogue based on a predetermined set of issues. For the PNAS beneficiary assessment, stakeholders included beneficiaries - those using or receiving services, or have done so in the past - in addition to commune elders, leaders, local administration, and those associated with NGOs, HIMO, or PMEs, etc. BA techniques included participant observation and conversational interviews. The BA examined whether the PNAS project fulfilled the basic minimum needs of the population (outlined below), in relation to the objectives established by the project. This was assessed by reviewing the project's effect on: * Agriculture: grazing, arable cropping, agro-forestry, irrigation (in relation to food security and AGR). * Use of natural resources like forests, swamps, valleys and water resources for agricultural purposes, other income-generating activities (AGR, reforestation/infrastructure). * Social and economic infrastructure: schools, rural health centers, water supply and sanitation, roads, bridges, etc (infrastructure/public works). * Any other, like requirements for small-scale enterprises or income generating activities (AGR). Information on basic minimum needs was collected by utilizing ranking (benefits, problems, wealth) and mapping exercises (community, personal, institutional) and trend analysis (history of events). The key issues assessed whether PNAS was able to meet the needs of the community in an effective, expedient, and sustainable way; and if it was able to target the mostl vulnerable groups (i.e., women, widows, orphans, victims of AIDS, etc). The BA tried to solicit recommendations from the commune members for future project improvement. The BA also assessed the level of involvement of the beneficiaries in the projects. Criteria for Selection To facilitate site visits and the implementation of the beneficiary assessment (BA), sites were selected where there was a general overlap of project activity i.e., in neighboring communes where all three types of activities occurred: infrastructure, micro-credit, and PAN (see Annex 1). An effort was made to choose a broad cross-section of prefectures, thus giving the beneficiary assessment more national coverage (out of 12 prefectures, BAs were conducted in 5). The number of each project type to be evaluated was chosen according representative numbers of each project type completed (see Annex 2), and to gather a broad perspective of each of the activities covered. Thus, for the 20 sites visited, there is the following breakdown among the three major components: * Ten infrastructure sub-project sites were visited * Six Micro-credit activities sites were visited * Four site visits were made on PAN activities. - 47 - Additional Annex 10. Aide-Memoire - Food Security and Social Action Project (PNAS) Aide-Mkmoire Food Security and Social Action Project (PNAS) Kigali, June 9, 2000 From May 20 to June 9, 2000 a World Bank mission visited Rwanda to advance preparation of the Implementation Completion Review (ICR) of the Food Security and Social Action Project (PNAS). The mission also addressed issues outstanding from the financial audit for 1998. The mission was led by Mr. Markus Kostner (Task Manager) and comprised Mr. Toni Kayonga (operations) and Ms. Michelle Cullen (social analysis). The mission would like to thank staff at CEPEX, the Government counterpart responsible for the project's review, for their collaboration and joint effort. 1. Financial audit for 1998 All outstanding issues raised in raised in the Aide-Memoire of September 17, 1999 have either been, or are being resolved. The mission commends CEPEX on the work achieved. Specifically: I. The Government has acknowledged that documentation for 427.4 million FRw spent during the year ending December 31, 1994 was destroyed during the war and will thus not be found. This file has been closed. 2. The audit report 1998 states that counterpart funds for the year ending December 31, 1996 remain to be paid. However, according to the last DCA amendment (December 16, 1998), the only counterpart contribution was 20% for local expenses under category 2, which closed on December 31, 1998 after spending about SDR 580,000. This would imply a maximum contribution of SDR 116,000. According to the 1995 audit report, the government paid FRw 65 million in 1993 as counterpart contribution, easily exceeding the necessary amount. There thus does not appear to be any counterpart contribution outstanding. This file will also be closed. 3. The auditors identified some outstanding debits incurred to suppliers in 1994. Government could verify the claim of only one supplier (La Rwandaise), worth 4,219,351 FRw. This claim will be paid through category 8 as agreed upon between the Govermment and the Bank in the amendment of December 16, 1998. 4. The PNAS project duly paid the contracting agencies in Gikongoro, who in turn did not pay local workers. MINECOFIN has brought this file to court. A decision is pending. 5. Regarding the loans for income-generating activities that have not been repaid, 22 cases have been handled by MINECOFIN's lawyer. Judgements have been passed on ten of these cases, in favor of MINECOFIN. A decision for two cases is pending. Twelve files are for debtors whose address is not known; CEPEX continues to investigate. I/ The findings and recommendations of the mission are subject to senior Bank management review. Any changes will be communicated to the Government without delay. -48 - 6. The PNAS project agreed with WFP to pay 50% of transport costs under the PAN component. An amount of approximately US$ 25,000 is still outstanding from an advance made by PNAS. CEPEX submitted relevant documentation to WFP and awaits their response. 2. Government review by CEPEX CEPEX has constituted a team to review the PNAS project. The team's activities are, among others, to: visit various project sites; conduct a basic inventory of project materials and works completed; review remaining files; and conduct interviews with those involved. This review is being conducted independently of the Bank's review, and a draft will be submitted by early July. Given the lack of documentation both in the field office and in Washington, the mission requested CEPEX to pursue a number of outstanding issues, including: a review of studies completed for PNAS infrastructure and AGR projects; the provision of copies of the training manuals to other projects; an examination of how many of the infrastructure projects that were built are still functional, and how these are functioning; and an examination of how many of the NGOs and cooperatives that were formed during the project are still operating. 3. Next steps Categories 1-7 a CEPEX to submit PNAS project review draft for categories 1-7 by July 15, 2000. a MINECOFIN to continue to pursue outstanding legal and financial matters as highlighted above. * By July 31, 2000 category 8 to settle accounts with La Rwandaise in the amount of 4,219,351 FRw, and to transfer to the Treasury 9,094,528 FRw for salaries not claimed from before the war. * Bank to complete ICR draft for categories 1-7 by July 15, 2000. * Bank to harmonize drafts by August 30, 2000. Category 8 * The PVK Bureau de Coordination as implementing unit of category 8 to recruit two consultants (implementation review, beneficiary assessment) by October 1, 2000. * Consultants to complete evaluation reports by December 15, 2000. - 4'3 - Additional Annex 11. Aide-Memoire - Pilot Community Development Project /PVK Aide-Mkmoire Pilot Community Development Project /PVK Kigali, March 19, 2001 A World Bank mission visited Rwanda between March 12-19, 2001 to undertake the implementation completion review mission of the PNAS project, including: (i) a review of activities carried out under category 8 during the final quarter of operation (October-December 2000) and of the final evaluation of category 8 undertaken by independent consultants, (ii) a discussion of the final steps for closing the project, and (iii) a discussion of the draft ICR. The mission comprised Mmes./Messrs. Markus Kostner (mission leader), Toni Kayonga and Antoinette Kamanzi (operations), and Juvenal Nzambimana and Claudine Fundi (financial management/disbursement). The mission would like to thank CEPEX and PNAS staff for their close collaboration. The main observations are as follows. This aide-memoire is subject to Bank management review. Any changes will be communicated to the Government without delay. I . Category 8 The mission and CEPEX undertook an initial review of category 8 activities and concurred with the consultant's findings that it achieved its objective of testing participation and decentralization approaches in an urban setting. Although community mobilization proved more complex than in rural areas because of constraints of CDC members in terms of availability, progressive participation is taking root. The population in the six pilot sectors in PVK, in which the level of poverty is equal to if not higher than in many rural areas, very much appreciates the way it was involved in the process. The capacity building activities undertaken at inception proved most appropriate. All but one infrastructure sub-project were executed on time and generally in good quality. Problems with sub-projects occurred in only one sector, which also implemented the only micro-project activity. Project implementation encountered several difficulties emanating from the use of the CRDP implementation manual, several provisions of which were not applied because of the small scale and limited duration of category 8. It was acknowledged that otherwise, far greater investments into capacity building would have had to be undertaken. This limited to some degree community ownership over both the process and outcome. For instance, financial management.and sub-project design were undertaken by the PVK rather than the CDCs. Further problems arose from the fact that the amounts allocated per sector ($100,000) were insufficient to cover their priority needs, which were mainly in water. As a result, CDCs had to reprioritize, which not only took time but also partly diminished their motivation. Allocations should also have differed by a sector's absorption capacity. In addition, some communication problems arose between the Bureau de Coordination/PVK and the CDCs. Nevertheless, all those involved felt that the general approach was appropriate. - 50 - 2. PNAS Closing The mission noted with satisfaction that closing procedlures as set out in the aide-memoire of October 13, 2000 have been followed properly: (i) the salaries and bills outstanding to staff and contractors from before the war have been paid; (ii) CEPEX has reached agreement with CARE and WFP for the payment of workers in Gikongoro and of transport costs incurred for the PAN component, respectively; (iii) the external evaluations were undertaken before closing as planned, and the final audit report is expected for mid-April; and (iv) recovery procedures for the special account have been discussed and agreed upon. 3. ICR Draft CEPEX deemed the Bank's ICR draft well researched, exhaustive, and well written. CEPEX felt that the evolution of the PNAS project was appropriate and responded to the emerging country situation (mitigating social effects of structural adjustment [91-93]; post-war reconstruction [95-97]; decentralized, community-based development [98-00]). CEPEX provided technical comments to the mission which will be duly taken into account in the revised ICR. 4. Next Steps * CEPEX will merge its two reports on categories 1-7 and category 8 into a single assessment of the PNAS project; April 15, 2001. * The Bank will revise and finalize the ICR on the basis of comments received; May 31, 2001. - 51 - MAP SECT'ION -20au 29o00 2930' j 3200' 3230' _w=oo )_ /J-= UGANDA >~~~~ ~ ~~~~~~~~ A N Z A \ AZN I A RWANDA DEM. REP iDD /. NATIONAL CAPITAL OF CONGO j - N (AKAGERA PREFECTURE CAPITALS WRUNA akeNAL COMMUNE BOUNDARIES / HAll NY, N NYByumba PARK MAIN ROADS /'AYYYYYAA TEA NA FSFjF' INTERNATIONAL BOUNDARIES A A A CA O ) A N f / f~~~~~~~~~ \ CSNK OATlF 'LI AJASX ONCO C, NlB NLr0rrA 5"Y' La~~~~~~~~~~~~~~~~~~~~~ke K (W X KAYOV CASEY N F1NFN s \ 7ur N4GIroL EYCI JIKR NLake LAKE iKPI E ' I N YOFONC - 00' I `C -NA A YYNF2 30' COY CO E5 _Cy tW;onAMBIQu Gii YES On goro r:,r - 2130' "RE IN.GEM ~NGWE 29°00' ' - ND FOREST H~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~YAOC GISHO RUWIN~~~~~~M(OOAYA BURUN D D TAN GiSKNY ''~N REAPORS CONG Taegaaika DEM. REP This trap -asAreard bythe hMop 'a I - , ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~N~0Desig Unit.oTh World Bak. ZAMBIA .n~~~w OF CONGO 1..~~~~........ . ' KILOMETERS 5 5 10 15 20 20 20 25 40ETERS 45 50 and0'2-30"3540 colorTh. den-nmino,sconi-ti >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ MILES 0 2 10 15 2'0 25 30 The.oldBn Group, anyjuget any endo-sement or accopt-e of such '0 YY~~~~~~~~~~~~~~~~~~~~~~~9oYY 29230' 7 2900' 39a' o telglsan fnytriyo