Document of The World Bank FOR OFFICLAL USE ONLY Report No: 24017 IMPLEMENTATION COMPLETION REPORT (TF-21609; IDA-28560; PPFI-P9070; PPFI-P907 1) ON A CREDIT IN THE AMOUNT OF US$56 MILLION TO THE REPUBLIC OF MALAWI FOR A SOCLAL ACTION FUND June 27, 2002 Human Development 1 Country Department 2 Africa Region This documcnt has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (Exchange Rate Effective December 31, 1995) Currency Unit = Malawi Kwacha (MK) MK 1.00 = US$ 0.0649 US$ 1.00 = MK 15.4 REPUBLIC OF MALAWI FISCAL YEAR April 1 - March 31 (July I - June 30) (as of July 1998) WEIGHTS AND MEASURES I meter (m) = 3.281 feet (ft) I kilometer (k1m) = 0.621 miles (mi) I square kilometer (sq km) = 0.386 square miles (sq mi) I cubic meter (m3) = 264.2 U.S. gallons ABBREVIATIONS AND ACRONYMS DDC = District Development Committee CAS = Country Assistance Strategy CSP = Community Subprojects DDF = District Development Fund DEC = District Executive Committee EPA = Extension Planning Area GDP = Gross Domestic Product GNP = Gross National Product GOM = Government of Malawi ICB = Intemational Competitive Bidding IDA International Development Association IEC = Information, Education, and communication lIHS = Integrated Household Survey MASAF = Malawi Social Action Fund MEPD = Ministry of Economic Planning and Development MU = Management Unit of MASAF NCB = National Competitive Bidding NEC = National Economic Council NGO Non-Governmental Organization NSO = National Statistical Organization PAP = Poverty Alleviation Program PIM = Project Implementation Manual PMSF = Poverty Monitoring Support Facility PRSP = Poverty Reduction Strategy Paper PSC = Project Steering committee PSIP = Public Sector Investment Programme PWP = Public Works Program SCC = Systematic Client Consultation SOE = Statement of Expenditures TWC = Technical Working Committee VAM = Vulnerability Assessment Monitoring System WFP = World Food Program Vice President: Callisto Madavo Country Manager/Director: Darius Mans Sector Manager/Director: Dzingai Mutumbuka Task Team Leader/Task Manager: N. Mungai Lenneiye MALAWI SOCIAL ACTION FUND CONTENTS Page No. 1. Project Data I 2. Principal Performance Ratings I 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 4 5. Major Factors Affecting Implementation and Outcome 7 6. Sustainability 10 7. Bank and Borrower Performance 11 8. Lessons Learned 13 9. Partner Comments 14 10. Additional Information 23 Annex 1. Key Performance Indicators/Log Frame Matrix 24 Annex 2. Project Costs and Financing 25 Annex 3. Economic Costs and Benefits 27 Annex 4. Bank Inputs 28 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 30 Annex 6. Ratings of Bank and Borrower Performance 31 Annex 7. List of Suppordng Documents 32 Annex 8. Public Works Program Targeting Mechanisms 36 Annex 9. Government ICR 37 MAP: IBRD No. 31950 IBRD No. 31951 Project ID: P001668 |Project Name: SOCIAL ACTION FUND Team Leader: Mungai Lenneiye I TL Unit: AFTH1 ICR Type: Core ICR IReport Date: June 27, 2002 1. Project Data Name: SOCIAL ACTION FUND L/C/TFNumber: TF-21609; IDA-28560; PPFI-P9070; PPFI-P9071 Country/Department: MALAWI Region: Africa Regional Office Sector/subsector: SA - Social Assistance KEY DATES Original Revised/Actual PCD: 10/01/1994 Effective: 05/30/1996 Appraisal: 03/05/1996 MTR: 07/01/1999 03/01/1997 Approval: 05/09/1996 Closing: 12/31/2001 12/31/2001 Borrower/lImplementing Agency: GOVERNMENT/GOM Other Partners: STAFF Current At Appraisal Vice President: Callisto E. Madavo E.V.K. Jaycox Country Manager: Darius Mans Barbara Kafka Sector Manager: Dzingai B. Mutumbuka Roger W. Grawe Team Leader at ICR: Nginya Mungai Lenneiye Norbert Mugwagwa ICR Primary Author: Mirey Ovadiya 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: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The objective of the project was to contribute towards poverty reduction in Malawi t.hrough (i) provision of additional resources for programs targeted to the poor; (ii) promotion of a new approach to rural development by community participation in project preparation and implementation; (Hi) a labor intensive public works program targeted at the poorest (food deficient) districts; and (iv) strengthening the poverty monitoring and assessment system. The project objective was appropriate given pervasive poverty rates, with about 65% of the population living below the poverty line and very poor social indicators including low enrollmnent rates, poor health status, poor access to clean water, food insecurity and poor infrastructure. The project objective and design were in line with the main elements of the Country Assistance Strategy: poverty targeted :nvestments and development of human resources, together with programs of macroeconomic stabilization, fiscal restructuring and deregulation, private-sector led growth, and capacity building and decentralization. Malawi Social Action Fund (MASAF) also complemented Government of Malawi's Poverty Alleviation Program, announced in 1995, whose objective, among others, was to improve access to education through elimination of school fees and uniforms in order to promote universal coverage of prim ary education, and re-orientation of public expenditures to social sectors and service delivery mechanisms which benefit the poor. The "poverty reduction" objective of this project may appear to be quite broad when viewed in isolation. It should be noted that the project was conceived as the first and "learning" phase of a ser_s of such projects, which would be expected to achieve certain development outcomes over a longer period. MASAF was designed to, and did provide a number of lessons on how to implement flexible, process- criented (as opposed to "blue print") projects; how to build social capital at grass roots levels, and how to generate a process of community-driven development. These are all Bank/IDA priorities in the Africa Region. Furthermore, during implementation, MASAF provided a series of lessons which are now fMly incorporated in the on-going MASAF 11 and are being systematically included in the proposed MASAF III which is now under preparation. These lessons concern management of strong community demand for sub-projects; the need to agree on and build monitoring and evaluation systems around poverty reduction key indicators; methods to sustain community based or local govermment supported maintenance of assets; and capacity building techniques to foster inter-agency cooperation and collaboration. The project design supported the creation of community level social and economic infrastructure to benefit the poorest as well as promoting community participation in identification, selection, imp)ementation and maintenance of such infrastructure with assistance from NGOs, private sector and local govemment. This was a new and welcome concept for a country coming out of 30 years of a highly centralized era of a single-party regime. The project was intended to inject cash into the rural economy, supplement incomes of the poorest households, and improve food availability in food deficient areas. It also supported a labor intensive rural public works program targeted to the poorest and most food deficient districts to improve their livelihoods through cash transfers whilst improving economic infrastructure. Given that Malawi is prone to natural and economic shocks, and has large numbers of poor and limited resources, it is crucial to be able to monitor poverty in order to make informed policy and expenditure decisions with limited public resources. The project aimed to strengthen the poverty monitoring and assessment system and capacity in the country. Given the novelty of the project concept, and uncertainty about in-country capacity for implementation, the project design was flexible enough to allow for learning and evolutior of procedures and institutional roles in response to community demands and capacities (p.34, Staff Appraisal Report). Thus, although the development objective was broad in terms of targets, it is assessed as responsive to - 2 - Borrower circumstances and development priorities. Although the project was fairly demanding in geographic coverage and institutions involved, use of an autonomous implementing agency, which improved over time, provided the necessary inpetus for the project Based on the above, the development objectives and design of the project are assessed as satisfactory. 3.2 Revised Objective: Project objectives were not revised. 3.3 Original Components: The project had the following three components. Community Sub-projects (CSP) component financed activities (such as upgrading/construction of comrnunity infrastructure for education, health, water and transport, among others) identified, prepared and implemented by the communities with or without assistance from facilitators of their choice who may be government officials, NGOs, private organizations or individuals; this included financing for institutional capacity building to ensure appraisal and supervisory support by financing the management unit (MU) staff, field offices, transportation, equipment and operational costs. An information, education and communication campaign to inform communities and other participants about the demand-driven processes oftbe project was also financed under this component. cemmurity project committees were trained in project management and implementation. Public Works program (PWP) component financed a safety rnet operation in poor and food deficient areas through labor-intensive public works construction including rehabilitating and maintaining economic infrastructure such as rural roads, minor irrigation, afforestation and land development works. This component intended to provide temporary employment and nrral minimum wage rate equivalent earnings for those poor with no off-farm income eaming opportumnities. The works were to be implemented jointly by local government staff and MASAF MU where private sector participation in the supervision and execution of works was encouraged. Poverty Monitoring Support Facility (PMSF) component financed (i) capacity building in institutions to coordinate, collect, analyze, disseminate and use data relevant to poverty reduction policies and programs; (ii) the development and implementation of an Integrated Household Survey Program at the National Statistics Office; and (iii) a poverty analysis program. The component was to be implemented by a Technical Working Committee for poverty monitoring. The design of all components was clearly linked to the objectives described above and recognized the administative challenges involved in implementing a nationwide demand-driven community development program. 3.4 Revised Components: The project components were not revised. 3.5 Quality at Entry: No formal quality at entry review was done for this project. Based on the following assessment of various elements of project design and preparation, quality at entry is rated satisfactory. Project Concept, Objectives and Approach. As noted above, the design of the project was consistent with the objectives of the CAS and govemment priorities and poverty alleviation strategy. In addition, documentation indicates that MASAF design took on board international experiences from other social - 3 - funds, pilot projects implemented through the Project Preparation Facility, and findings from systematic client consultations conducted during preparation. Although outcome indicators, not included during the design, were introduced at mid-term, the absence of baseline information meant that data collected could not have been analyzed meaningfully. Institutional Capacity. Given the lack of capacity at local govemment levels, and the public sector's inexperience with community driven development, an autonomous MU, under the guidance of a National Steering Committee, was set up. It would coordinate and work with district level technical and political bodies, and NGOs at the local level to inform and assist communities to apply for and implement sub-projects. This approach was appropriate as the project was received enthusiastically by local governments and communities and generated tremendous demand for sub-projects. Although not many NGOs were involved in project implementation, private contractors were involved in implementation of community sub-projects. Readiness for implementation. While the project concept was sound, community demand for the program was underestimated, resulting in an initial inability of the MU and district staff to process sub-projects in a timely manner. However, the design allowed for technical, financial management, operational and supervision capacity to be adjusted over the life of the project to smooth implementation. Although a monitoring and evaluation framework was defined, and the necessary resources were available, a cohesive and effective management information system was not in place until later. Technical and economic assessment. An economic rate of return (ERR) analysis was presented in the Staff Appraisal Report, using economic returns to education, roads and water sectors, and was prepared with the best available data; however, experience shows that it is not best suited to evaluate economic benefits of a social fund. A cost effectiveness (including technical quality aspects) analysis and financial benchmarks would have been more appropriate. Risk Assessment and sustainability. Overall, a fairly accurate assessment of risks involved in implementation was undertaken and mitigating measures were included in the project design. However, expectations with respect to capacity building requirements from resource-poor local administrations for supervision of sub-projects as well as prompt cooperation from line ministries on operation and maintenance were not realistic. This contributed to implementation delays and maintenance problems downstream. These problems were by and large overcome by strengthening the MU implementation capacity, and enhancing sensitization at the district and community level; however, budget support for recurrent requirements of health and roads interventions continued to be erratic. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: Achievement of objectives of MASAF is rated as satisfactory, notwithstanding the absence of outcome indicators initially, and the absence of baseline data to inform on indicators identified during project implementation. This assessment is inferred through the achievement of output targets, qualitative and quantitative information from beneficiary and impact assessments conducted during project implementation. While the satisfactory rating on the outcome depends on inferences on achievement of outputs, these are not contentious. For example, it is assumed that the 2,146 number of classrooms built by MASAF at the request and with the participation of beneficiary communities has resulted in improved educational attainment within the assisted communities, given the fact that these classroom blocks were fully staffed and equipped. Similarly, it is inferred from the construction of about 1,000 - 4 - safe water points that there have been improvements in the quality of life of beneficiary communities including reduced labor for girls and women, and improved health outcomes due to reduced incidence of waterborne diseases. MASAF also generated improved ability on the part of the communities to plan, implement and monitor community based projects and more generally to charge of their own lives. More specifically, with respect to the first and third activities supporting the objective, although adequate statistical data is unavailable on the poverty impact of the project on community and household welfare status, the Beneficiary Assessments conducted (1997, 2000) in 25 representative project sites note the following as project benefits: reduced distances to schools, health centers, and water points resulting in improvements in school enrollment/attendance, access to clean water and health services and employment opportunities created by the sub-projects. Based on the project surpassing its output targets, in most instances, in infrastructure created, person-days of employment created, and number of beneficiary households, these objectives were achieved. With regard to the second activity supporting the objective, the initial slow sub-project implementation/completion trends, and drop-outs, point to weaknesses in the participatory mechanisms under MASAF. Qualitative data suggests that the systematic application of participatory methodologies at the community level for project identification only took root during MASAF implementation due to the initial inadequate outreach, training and monitoring. Nonetheless, a large number of community project committees (> 1,300 with 10 people each) were trained in accounting, project implementation, supervision, and direct contracting through which they were able to manage their own development activities. Experience from other social funds indicate that community led development requires time to take root and is a process in evolution; MASAF has done considerable groundwork in introducing a new way of doing business at the sub-district level. Therefore this objective was largely achieved. As regards to the fourth objective, in support of PMSF, it has only partially achieved its intended objective. As a result of the advocacy under this component, the govemment provided resources for the staff and other recurrent costs of undertaking an Integrated Household Survey (IHS) in 1997/98; consequently most of the IDA and Danida grant funds were used for the purchase of complementary equipment, design and implementation of a series of complementary panel surveys, as well as small policy studies to stimulate the local analysis of data, and intemational technical assistance to train and build capacity in poverty analysis and data use locally. The data and analysis produced under this component has since formed the basis for Malawi's PRSP, the latest public expenditure review, and the development of a more robust targeting system for the safety nets strategy for the country. Nonetheless, implementation start-up was slow, marked by a poor working relationship between the management unit for the PMSF and the MU for MASAF. Furthermore, in regard to the use of the poverty analysis in policy making, while progress has been made as evidenced above, there is much wider scope for promoting the use of this information in policy making across sectors. 4.2 Outputs by components: Community Sub-Projects (CSP) (US$48.03 million). This component is rated satisfactory; 1,392 sub-projects were approved (original target was 800), and 93% of these were completed with the incompletes being carried over to MASAF 11. Of these, 46 % were in the southern region (where 47 % of the population reside), 36 % in the central region (where 41% of the population reside), and 18 % were in the northem region (where 12% of the population reside); with the majority of the projects benefiting the southem region which is the poorest in the country. This component also supported - 5 - training for 1,131 community project committees on project implementation. According to project progress reports, an estimated I million people benefited from CSP projects, with 52% of these being females, as well as 30% of project management committee members being female. Projects under CSP were heavily skewed towards the education sector (54%), and this was a reflection of the needs in that sector following the launch of Govemment's Free Primary Education Policy in 1994. Public Works Program (PWP) (US$19.47 million). This component is rated satisfactory. PWP built/rehabilitated over 5,000 miles of roads, provided over 13 million persondays of employment, benefiting over 136,000 households (over 680,000 people). Both geographic and wage rate targeting were used to direct benefits to the poorest and food deficient areas of Malawi. Firstly, Extension Planning Areas (EPAs) where household food shortages were most severe were identified using the Vulnerability Assessment and Mapping (VAM) system, utilizing information gathered by agricultural field officers, WFP and USAID Famine Early Warning System,, and district level key poverty indicators. Secondly, district teams were to identify projects in consultation with communities in EPAs with the highest poverty incidence. At the third level, a daily wage rate lower than minimum rural wage rate was to be used to self-target poorest households in the selected EPAs. About 1.2 million are reported to have benefited directly (through cash transfers) and indirectly (through access and use of infrastructure) from PWP projects. On average, wages for unskilled labor constituted about 48% of project costs, versus the target of 36%, the average achieved by other public works programs in Africa. Over 680,000 people (188,000 actually employed) benefited from the program, again, above the estimated targets of the program. Available data suggest that about 60% of PWP projects were implemented in the Southem Region where poverty incidence is highest in the nation; although not an output indicator, about 30% of unskilled labor employed were women. Poverty Monitoring Support Facility (PMSF) ($1.51 million). This component (to strengthen the poverty monitoring and assessment system to collect and analyze and use poverty data to improve effectiveness of targeted interventions) is rated unsatisfactory. Between 1996 and 2000, over eight policy studies, one national Integrated Household Survey, and a complementary panel survey series were carried out. Technical and financial support were provided to the national Technical Working Committee for poverty monitoring, and for training of staff from the National Economic Council, National Statistics Office and the Center for Social Research (CSR). However, overall, output targets under this component have not been achieved. Achievement of outputs was adversely affected by a number of factors: disconnect within the Bank implementation team where the "main" project supervision missions did not incorporate supervision of the PMSF, which was regarded as a component supporting not only the project, but the CAS as a whole and supervised separately for the most part; staff turnover both in NEC and on the Bank team, and poor working relationship between MASAF and NEC. 4.3 Net Present Value/Economic rate of return: The assumptions used to calculate the economic rate of return to primary education, water and road (earth/gravel) sectors in the Staff Appraisal Report (SAR) were not found to be realistic as the analysis fails to capture the magnitude of immediate benefits accruing to communities as well as the positive externalities involved due to the public nature of the infrastructure created or maintained. Hence, it was agreed that the ICR will focus on a cost effectiveness and efficiency analysis, comparing MASAF use of resources with Government agencies and other donors for creation of various infrastructure. The economic analysis is attached in Annex 3. 4.4 Financial rate of return: No financial rate of return was calculated for the project at appraisal and the ICR did not compute the - 6 - FRR 4.5 Institutional development impact: The project's institutional development impact was substantial. The SAR indicates that "there will be continuous trial and upgrading of procedures as the community driven development approach takes root as well as continuous evolution of institutional roles" which is the essence of the workings of a social fund to allow flexibility and maximum impact. MASAF was launched as a participatory, demand-driven approach to community development that empowered communities, and to some extent district administrations, in voicing their needs and formulating solutions, and Malawians responded enthusiastically. MASAF re-established a link of trust and started a dialogue on development between the govemment and the people of Malawi. It has also, to some extent promoted and achieved cooperation across sector ministries through implementation of sub-projects. However, building social capital and engaging communities, private sector, and local govemments in development planning is a process that evolves and improves over time with continued financial and human resources. MASAF planted a seed of partnership between communities, district administrations, private sector and others in development planning and implementation. Different elements of its institutional development impact include: * Over 13 million person days of employment were created through labor intensive works, which relied on, and cultivated development planning and management capacity at different levels of govemment and communities. Examples include communities leaming negotiating skills, local authorities and contractors becoming more accountable to communities, sector ministries becoming more responsive to community demands in allocation of recurrent expenditure budgets, Malawi Broadcasting Corporation oriented itself towards rural reporting, and the "sub-project implementation cycle" becoming the way of doing business at the district level. * The MU, through its working relationship with other government agencies and ministries, has advanced, and continues to advance, the community development, decentralization, and social risk management/social safety net agenda in Malawi. * Community development has been brought to the forefront of Govemment's Poverty Reduction agenda and fostered (i) a service delivery culture at the local govemment as well as responsiveness, transparency, and accountability towards communities where there was practically none prior to the program; (ii) ownership, culture of self-help and taking charge of one's life at the community level; and (iii) improved understanding, at the central level, of community driven development, working across sectors, and devolution of powers to lower levels of govemment for effective service delivery. * Lessons/experiences under MASAF will enable Govemment to make informed choices and expand the range of social safety net interventions offered. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: HIV/AIDS. The HI V/AIDS incidence among adult population has increased (currently at about 19 percent) eroding further the human capital base of Malawi, increasing the pressures on extended families to take care of increasing number of dependents. Structure of the economy and external shocks. During the period of project implementation, - 7 - Malawi's economic performance has been highly variable mainly because of its excessive dependence on weather pattems and agriculture, mainly tobacco, tea and sugar as its main exports. There were floods and droughts which deteriorated the food security situation in the north and south of the country During this period also, Malawi remained vulnerable to exogenous shocks like decreases in intemational commodity prices for its exports; i.e. tobacco, coffee, tea and sugar. Macroeconomic instability, high rates of inflation, coupled with sustained depreciation of the kwacha had adverse impacts for large scale agriculture and small-holders. These adverse conditions resulted in (i) greater demand for sub-projects which was much higher than envisaged during project preparation, resulting in initial delays in project approvals, completions and poor technical quality of sub-projects; (ii) depreciation of the Kwacha which put pressure on local currency denominated budgets and required periodic adjustments to sub-project costs to keep up with depreciation and inflation; (iii) PWP wage rates not always keeping up with inflation thereby constantly eroding program impact; and (iv) slower implementation as a result of weather conditions, especially in Southem and Northem regions. 5.2 Factors generally subject to government control: Making budgetaryprovisionsfor recurrent expenditures to support social and economic infrastructure created. Social expenditures per capita and as a percent of total govemment expenditures have decreased from 97/98 to 99/00 whereas the Govemment with programs similar to MASAF has been building social infrastructure in response to increased demand. This, in tum, means that Govemment is not providing commensurate resources (recurrent budget) in support of the infrastructure built; i.e. teachers, fumiture, training materials, etc. As an example, only 5 out of 56 of health posts (about 10%) built under MASAF have medical staff although 100% of the school blocks built are staffed. (MASAF switched to other sectors once the problem became evident; see p. 10) Delays in completion and lapses in quality in PWP and CSP projects indicate lack of consistent and adequate technical support by sector ministry staff at the district level on sub-project appraisal and implementation, lack of adequate resources to operate and maintain assets post-completion, and inability to manage staff tumover at the district level; most of these can be attributed to inadequate budgetary resources being made available. Poor technical quality of infrastructure. Until 1999, there were shortcomings in desk and field appraisal, project costing, and supervision during and after completion of sub-projects due to inadequate technical capacity at the MU and district levels. There was a mismatch between existing and required institutional capacities at different levels of the project cycle. MASAF MU took a number of measures to strengthen its technical capacity, including hiring of technical supervisors, providing training in project costing, and hiring additional staff at the MU towards the end of MASAF to improve technical quality and implementation of sub-projects. Subsequently, there have been substantial improvements in costing, implementation completion (from an average of 18 months in the beginning to 8 months at the end) and on1 the structural quality of projects. Coordination and coherence between community level interventions. Govemment needs to ensure community level targeted interventions are not following multiple and conflicting procedures that give conflicting messages to communities. Misunderstandings with respect to other Govemment or donor supported development programs such as food-for-work and starter-pack programs [a national program that distributed farm inputs to rural households partly to mitigate the effects of removal of subsidies and price increases] adversely impacted implementation especially since CSP was designed around self-help, targeting of poorer communities and community contribution concepts as opposed to - 8 - having universal coverage and no access criteria. 5.3 Factors generally subject to implementing agency control: Limited Technical and Operational Capacity at MASAF MU. Overwhelming community demand, beyond initial projections, caused implementation delays. This was due to limited capacity at the MASAF MU especially for procurement, technical services and operational staff such as zone officers. There were delays in disbursements for community sub-projects as well as wage payments under the public works program. MASAF MU adjusted its organizational set-up and staffing to speed up processing of larger volumes of sub-projects and to facilitate disbursements. Effectveness of outreach and awareness campaign. Overwhelming demand from communities attest to the extent of the outreach that MASAF was able, from the start, to feed back its successes into its messages and national image. Despite the successes of the Information, Education and Communication campaign, Beneficiary Assessments and discussions during field visits suggest that there is still lack of understanding on key issues such as community asset maintenance, responsibilities for project management and supervision, targeting principles of MASAF and environmental guidelines. Adequacy of communityparticipation. Beneficiary Assessments and project documentation suggest that systematic application of participatory methodologies at the community level for project identification has taken root mid-stream during MASAF as a result of inadequate outreach, training and monitoring. Some of these shortcomnings seem to have been rectified over the life of the project Ineffective management information systenm Two information/motoring systems (an effective financial management system, and a poor project tracking system) exist in MASAF; it does not have an integrated management information system. These cannot provide combined data on physical and financial indicators. Key performance indicators are cumbersome to obtain, and this diminishes MASAF's capacity to respond to changing needs in a timely manner. Effectiveness ofpoverty targeting andpoverty impact. The PWP was intended to target the poorest and food deficient but questions have been raised by district officials and communities about the reliability, accuracy and transparency of VAM data. More attention should have been devoted to confirming data with district officials and communities as well as wider dissemination of targeting criteria. Use of a higher program wage rate than minimum wage rate, the second level targeting tool for the "poorest" to self-select into the program, has probably diluted the poverty impact of the program. However, in the absence of baseline data and systematic data collection on changes in welfare of beneficiary households through impact evaluations of MASAF programs, it is not possible to ascertain the effectiveness of poverty targeting, and poverty impact Working relationship between the imptementing agencies for MASAF: There was poor communication and tensions between the MU, a semi-autonomus/semi-public agency, for the CSP and PWP components and the National Economic Council, a government agency, responsible for the implemnentation of the PMSF. Tensions were due to different pay-scales and rules and regulations goveming the two. There was also high staff tumover with four different heads of the Poverty Monitoring Unit during the five years of project implementation. This meant that key decisions had to keep being revisited. 5.4 Costs and financing: MASAF was designed utilizing PHRD funding, and a Project Preparation Facility supported the piloting of activities, in order to inform on the design of the Project. The project cost at appraisal was -9- US$71.3 million, to be financed through an IDA credit of US$56 million, counterpart funds from the Government of Malawi in the amount of US$5.99 million, in-kind/cash community contributions equivalent to US$8.3 million, and US$1 million to be financed by the World Food Program. IDA disbursements amounted to US$51.7 million, as a result of exchange losses. Government's counterpart funding amounted to US$2.7 million, and communities contributed US$8.3 million. Support from the World Food Program was provided at a later date, and as parallel financing. DANIDA pledged co-financing of US$1.8 million in support of the PMSF; US$1.64 million was disbursed. IDA funds were fully committed by December 2000, and disbursed by November 2001. Total project cost at the end of the project equaled US$63.7 million. MASAF was able to deliver more than the projected number of sub-projects as a result of the depreciation of the kwacha. Reallocations were also made from the PMSF component, as DANIDA co-financing (as well as IDA funds from the Fiscal Restructuring and Deregulation Project) was available to support that component. Substantial delays were encountered in the provision of counterpart funds which affected the disbursements to contractors and completion of works at times. By the end of the project, the full amount of counterpart funds required in kwacha terms was received; however, it was less than the appraisal estimate in dollar terms due to the depreciation of the currency against the US dollar. Also, community contributions (in kind and cash) were not systematically reflected in overall project accounts thus actual contributions appear lower than reported at sub-project appraisal. Nonetheless, community contributions were usually 20% or more (in some cases up to 40%) of sub-project costs as reported in project proposal documents. 6. Sustainability 6.1 Rationale for sustainability rating: The project overall enjoyed commitment from the Government of Malawi. Community sub-projects and public works components were designed to support Malawi's Poverty Alleviation Program to improve services and delivery mechanisms which benefit the poor. The majority of education and water facilities rehabilitated or built are reported to be in working condition (staffed and equipped) by supervision missions even though the understanding of maintenance requirements has not been consistent across the board. There are, in general, operations and maintenance (O&M) committees and/or parent-teacher associations to ensure day- to- day operation of these facilities. The track record with roads, bridges and health facilities is not as good due to weak community ownership when in use by several communities ("free-rider problem"), inadequate technical and financial support from local administrations, and weak coordination and linkages with line ministries. As a result of concems about sustainability, MASAF has avoided health projects since it could not ensure the sustainability of assets, and has, in its social mobilization, placed greater emphasis on project implementation, including O&M. MASAF sub-projects also initially suffered from lapses in technical quality due to shortcomings in desk and field appraisal, project costing, and supervision during and after completion of sub-projects. However, the MU took a number of measures including hiring of technical supervisors, providing training in project costing, and hiring additional staff at the MU to improve technical quality and implementation of sub-projects. Through the afore-mentioned measures, the viability of community initiatives improved over time even though maintenance of assets remained a weaker area of the MASAF program. Thus, overall sustainability of assets created under the program is rated likely. - 1 0 - With respect to PMSF, a number of studies have been produced, quantitative analysis done and considerable capacity has been built in various institutions. The govemment has chosen to implement this component using existing govemment structures and staff within the NEC and the NSO funded by the government's budget It also financed the IHS from the budget rather than relying on IDA or other external funding. While these features of the PMSF are encouraging signs regarding the sustainability of the activities under the component, the challenge will be to continue the work in the absence of extemal funding. Now that the PRSP is finalized, the implementation of the PRSP may consolidate the commitment of the Ministry of Finance and National Economic Council to sustaining the poverty monitoring system since monitoring and evaluation is a crucial pillar of the PRSP program. Overall, the sustainability is rated as likely, especially since the Bank is supporting two follow-on operations (one under preparation) designed on the basis of lessons leamed from this project. It is expected that the operation under preparation will build on the MASAF demand driven and participatory planning framework to strengthen linkages between central and local governments and communities moving closer to a decentralized governance system. 6.2 Transition arrangement to regular operations: Seeing the gains made in a short period through the community driven development approach, and realizing the work remaining to be done at the local level, the Government requested a follow-on operation, MASAF 11, which became effective in February 1999. It built on the localized development process initiated under MASAF, and increased emphasis on improved local governance, participatory processes, improved technical quality and sustainability of assets created as well as targeted interventions for vulnerable groups. MASAF Ill, currently under preparation, is expected to support capacity building at the district level with a view towards decentralization. These interventions derive from the Poverty Alleviation Program, the PRSP and the National Safety Net Strategy which focus on long term poverty reduction through growth, improving the quality of service delivery in the social sectors through effective deployment of qualified staff coupled with provision of adequate recurrent budgets and basic supplies, and improving the quality of life of the most vulnerable. 7. Bank and Borrower Performance Bank 7.1 Lending: Bank performance during project preparation is rated satisfactory. The project was designed in response to a government request, in line with the CAS and Government priorities, for a targeted program to improve services and service delivery mechanisms which benefit the poor. It was prepared in 14 months (from identification to Board presentation) by a team with diverse technical skills including sector and operational expertise which assisted the Govemment with major design issues and ensuring consistency with Bank policies. Many of these activities were supported by a PHRD Grant. The Bank staff also facilitated the use of a Project Preparation Facility which was used to set up the project management unit, pilot CSP and PWP projects, and to conduct systematic client consultations and workshops all of which contributed to project design and are reflected in the SAR. Risks associated with the project were well articulated and, for the most part, counter-balanced with mitigating measures included in the design. However, better guidance could have been given during project preparation on (i) adequacy of staffing and skill mix of the Management Unit (MU) especially on operational and technical staff required for processing and supervision of sub-projects, (ii) setting up a management information system, and (iii) monitoring and evaluation of outcomes rather than output targets. -11 - 7.2 Supervision: Bank performance for supervision is rated satisfactory. The Bank team spent over 170 staff-weeks for supervision of the project with a team composed of operational and technical staff . Eleven supervision missions were undertaken over the life of the project, and reasonable continuity among the task team and the task team leader. Supervision paid particular attention to financial management and procurement, and provided regular guidance on sub-project implementation including on communication strategy. The team made use of the flexible project design and allowed for re-allocation among components as well as improvements in institutional set-up to maximize use of credit resources and allow for the project processes to evolve over time. More timely and better advice could have been given on re-defining the outcome indicators, balancing responsiveness to demand with quality, adherence to the sub-project cycle, technical quality of infrastructure, and establishing an effective MIS system. Technical audits of MASAF suggest that quality-and therefore sustainability-- has been compromised in favor of speedy implementation and maximum number of beneficiaries. Nevertheless, for the most part, Project Status Reports (PSR) have identified implementation bottlenecks, and suggested and followed-up on solutions. Supervision of the PMSF appeared to be disconnected from that of the rest of the project; even though there were dedicated staff in charge of this component. Towards the end of the project period, there was also staff tumover in the sub-team supervising the component. The major concem, however, is that the findings from the supervision missions probably did not receive due attention from the rest of the Bank's task team. This resulted in delayed resolution of implementation bottlenecks, especially with regard to areas (such as procurement) where the MASAF MU and the NEC needed to work together. NEC, the implementing agency for this component, also noted this disconnect as a major factor for delays in implementation. Thus supervision of this component is rated unsatisfactory. 7.3 Overall Bank performance: Overall Bank performance is rated satisfactory in view of the intensive preparation effort and fairly comprehensive supervision effort as reflected in the PSRs and Aide Memoires . Borrower 7.4 Preparation: Government of Malawi prepared and appraised MASAF jointly with the Bank. Along with trust fund (PHRD) resources, the Govemment requested a PPF to carry out systematic client consultations and stakeholder workshops, and to pilot sub-projects and test project design before project approval. The pilot experiences provided both Govemment and Bank teams with lessons to improve project design before project launch. The preparation team included staff of relevant sector ministries and agencies. The preparation of the credit by the borrower is rated satisfactory. 7.5 Government implementation performance. Overall, Govemment of Malawi has been supportive of the MASAF program through, although sporadic, provision of counterpart funds and recurrent expenditures. The program has been strongly embraced by the President and political leadership as a national program benefiting the poor. Support at the district level for project appraisal and implementation has not been consistently strong primarily due to resource constraints, but also lack of a clear national local governance framework although it has improved through project implementation. Govemment's performance is therefore rated as satisfactory. - 12 - 7.6 Implementing Agency: In general, MASAF MU operated with a high degree of efficiency, and was able to remain largely protected from political influences over the life of the project. However, high level political endorsement and high visibility of the program, and the national IEC campaign preceding program launch, heightened public expectations leaving MASAF MU confronted with vast demand which it could not cope with effectively due to staff constraints at headquarters and local levels. Initial demand put enormous pressure on the MU to move very fast which resulted in compromising technical quality, and adherence to principles of participation-community and local govemment-in implementation. However, over the life of the project adjustments in staffing and procedures to improve implementation and project effectiveness were made successfully. Therefore, the MU's perforrnance is rated highly satisfactory on CSP and PWP. On the PMSF, the MU and NEC, responsible for the implementation of the PMSF component, were not able to establish a good working relationship which adversely impacted the outputs and the achievement of the outcome. 7. 7 Overall Borrower performance: Borrower's performance is rated satisfactory in view of the achievement of the targets set for the program. 8. Lessons Learned Need to strike a balance between objectives of reaching the maximum number ofpoor and creating sustainable assets. Social funds always face a trade off between aiming for quality of assets, and thorough dissemination and application of community driven development concepts or responding to as many communities needs as possible within the shortest time period possible. However, sustainability of assets created and community driven development taking root depends on success in creation of adequate social capital and understanding of participatory development processes at different levels. Ensure an accurate understanding of existing capacity and capacity building requirements during project preparation. The extensive lEC campaign created very high expectations on the part of communities as well as parliamentarians and other elected officials. This put a lot of pressure or. the MU to deliver rapidly without due attention to agreed processes at all times, i.e., field appraisal, ensuring sector ministry support, insufficient focus on post-completion maintenance and the essence of the project was compromised. Socialfunds should make use of Information Educationi and Communication (IEC) campaigns as tools to rationalize community demand, and at the same time foster participatory community development. Under MASAF, the lEC strategy started out as more of a public relations tool. It, however, became progressively an education and outreach tool disseminating community development principles and program access criteria more systematically. Need to improve technical quality. There is need to improve technical quality of infrastructure by providing increased technical support and training to communities for planning, design, supervision, contracting and accounting of funds. Lack of an effective and responsive management information system throughout project implementation has hampered the program's ability to identify implementation problems in a timely manner, to react to changing conditions rapidly and monitor and evaluate program impact promptly. - 13 - To be able to draw definitive conclusions on the economic effectiveness and impact of the program social funds should ensure that (i) a comprehensive management information system is in place; (ii) baseline data are collected; (iii) outcome indicators and scope of impact evaluations are well defined prior to project effectiveness; and (iv) collected data is analyzed and fed back to management at regular intervals. Need to ensure strong linkages between central, local government and socialfund management units. Social fund operations need to ensure continuous focus on maintenance of assets and availability of resources (e.g. teacher salaries, furniture, teaching materials, health staff, drugs, etc) to operate assets effectively through memoranda of understanding with govemment ministries, and support from district administrations. Requirements for effective operation and maintenance of assets should be outlined in the operational manual and be part of the sensitization and capacity building for communities and district administrations. Also, increased training and sensitization for district level staff improves the viability of community assets created through more consistent support from district staff. Choice of institutional set-up may determine extent and timing of development impactfor social fund operations. Social funds also face a trade off between extensive sensitization, capacity building targeted at communities and district administrations; adhering to participatory methods thoroughly; giving more responsibility to district govemments; and speed in implementation through more centralized functions of a large management unit. MASAF operated closer to the first model where management unit was the center of all processes whilst local govemments were marginally involved in sub-project cycle processes partially due to the absence of a clear local govemance framework at the time, and partially for sake of speedy implementation. Ensuring effective targeting and wide-spread understanding of eligibility criteriafor safety net programs to maximize impact with available resources. Justifying targeting in a country with the majority of its population living below poverty, especially in rural areas, is difficult. If there are certain groups of people identified as more vulnerable in face of certain shocks (i.e. drought) and as deserving govemment support, then every effort should be made to ensure effective targeting and wide dissemination of eligibility criteria to dispel any questions of favoritism. Supportfor Poverty Monitoring and Analysis as a component under socialfunds does not produce long-term results unless there is strong leadership and commitmentfrom the Government. Poverty monitoring and analysis needs commitment and oversight from Government; otherwise, Bank efforts usually associated with social fund operations do not go farther than supporting household surveys, equipping national statistics offices and training a few staff 9. Partner Comments (a) Borrower/implementing agency: Summary of MASAF Government ICR 1.1. Introduction The trickle down approach that Malawi had assumed would reduce poverty did very little to have meaningful impact as evidenced by the 1993 Situation Analysis of Poverty. Social and economic indicators were among the worst in the world. Low access to basic necessities including safe drinking - 14 - water and food supplies in most parts of the country were some of the basic indicators of high levels of poverty. Following the 1994 general elections, government raised the poverty alleviation agenda to the centre-stage in its development policy framework. This lead to the formulation of the Poverty Alleviation Program (PAP), and subsequently an implementation framework in which, MASAF was one of the key instruments for operatianalising the program. The MASAF became effective on May 15, 1996, as a result of a long series of consultations. It was envisaged that MASAF would contribute to the goal of poverty alleviation through: (a) the upgrading of community infrastructure by rehabilitating and constructing education, health, water, and transport and communication assets, (b) improvement to the delivery of and access to social and economic services, (c) fostering grass root initiatives (social action) intended to enhance the technical and managerial capabilities and attitudes to self-managed development, and (d)reorienting and building capacity of actors in community development including communities, extension workers, NGOs, the private sector and religious groups to equip them with skills in participatory methods for planning and implementing poverty alleviation interventions. 1.2. MASAF 1 Components The MASAF 1 objectives were to be achieved through three operational components: the Community Sub-Projects (CSP) component, the Public Works Component (PWP) and the Poverty Monitoring Support Facility (PMSF). 1.2.1 Community Sub-Projects (CSP) component The CSP component was the largest and accounted for 71 percent of the funds under MASAF I, was designed to provide financial and technical resources for programs targeted at the poor by funding the creation of community assets and provide an opportunity for communities to effectively participate in the identification and implementation of their own development projects in collaboration with relevant collaborators. The project processes would empower communities by creating opportunities for communities to revive the spirit of self-help, enabling them to take control over, and be accountable for their own development. 1.2.2 Public Works Program (PWP) The PWP component accounted for 27 percent of the financial resources under MASAF I and was designed to transfer income, through the creation of employment opportunities by funding labour intensive public works, and in the process, build economic infrastructure that would be beneficial to communities. 1.2.3 Poverty Monitoring Support Facility (PMSF) The PMSF component accounted for 2 percent of MASAF I financial resources and was managed by - 15 - the National Economic Council (NEC) to support the poverty reduction policies and programs by bridging poverty-related gaps and developing analytical skills in government and private institutions. The aim was to enhance collection and analysis of data relevant to track progress in poverty reduction. In addition, there were support divisions and units such as Administration Division, Finance Division, Internal Audit Unit, Information Technology Unit, Technical Services Unit, and the Information, Education and Communication (IEC) unit. All these played an integral part in the realisation of MASAF 1 goals. 1.3 Analysis of Achievements 1.3.1 Design Assumptions MASAF I objectives were met satisfactorily in that they addressed issues that have a direct bearing on poverty, at the centre of government and community priorities. They were relevant to and consistent with the Malawi Government's poverty alleviation policy fiamework (PAPF) for achieving poverty reduction through asset and economic empowerment, gender balance and environmental protection. In general, the design of the project could be rated as highly satisfactory and appropriate for attaining the set objectives. Transparency, accountability, community participation, gender sensitivity and direct community financing that MASAF embraced in its design were principles that would contribute to a program geared towards empowering the poor. The design of the MASAF assisted to enhance decentralisation by applying community level decentralised planning processes and collaboration. The geographical and individual or household targeting mechanisms and the labour intensity principles enshrined in the design of the PWP component were appropriate for a safety net operation. As a result, the component was able to effectively reach its intended beneficiaries. The design of the PMSF component was to a large extent appropriate. It emphasised capacity building efforts that would be critical for sustained analysis of poverty related data in the long term. The overall Project objective was to contribute to poverty reduction. Although the level of contribution to poverty reduction was not quantified, the Project development objective was substantially achieved. An estimated 2.0 million people (or 20 percent of the Malawian population) directly benefited from investments in the education, health, water and sanitation, transport (roads and bridges) and afforestation sectors; and the transfer of income through the PWP component. In total 1,864 sub-projects were approved, of which 1,392 projects were under the CSP component and 472 projects under PWP component. Under PMS, training for key personnel at the National Economic Council has been undertaken In addition, equipment for data analysis has been procured for the National Statistical Office (NSO) and National Economic Council (NEC). - 16 - 1.3.2 Project Impacts There is ample evidence that shows significant improvements in the welfare of a common man especially in the rural areas. * Reduction in incidences of water borne diseases. * Increased enrolment in primary and secondary schools. * Improvement in Socio-economic Status through safety nets transfers and assets created. * Improvement in communication and access to social and economic facilities. * Stimulation of the rural economy. * Fostering of a Collegial Spirit towards community development efforts. * Building capacity for local level planning and development management. 1.3.3 IMPLEMENTATION RECORD Community Sub-projects Achievement of objectives under the CSP component has been highly satisfactory. Against the projected 800 projects to be approved under the CSP in its entire life, 1,392 projects were approved. The component made significant contribution in the areas of health, education, bridges, water and sanitation and capacity building. The sectoral distribution of the approved projects was as follows: education 735; water 307 involving the drilling of 921 boreholes and construction of 80 water kiosks; transport and communication 120 bridges; health 56 (under-five clinics/maternity units and health centres); and other projects 13. Projects in the education sector involved the construction of 2,146 new classrooms, 854 staff houses; 56,445 school desks and 4,478 VIP latrines. Public Works Component Achievement of the safety net objective was highly satisfactory. The (USD13.7 million) PWP component financed district level safety net operation that has benefited poor people in food deficient Extension Planning Areas (EPAs). A total of MK185 million (US$2.5 million) I US$ = MK73.00 in cash has been transferred to 188,000 people through creation of temporary employment. This has been achieved through the financing of a total of 472 labour intensive projects such as roads, community dams and afforestation projects. The benefits arising from the PWP component have directly reached approximately 680,030 people or about 7.6 % of the Malawian population. However, considering that utilization at the household level would benefit more people than those directly participating on the project, the benefits reach more people than captured here. The projects approved under the PWP have translated into the following outputs: 188,000 employment opportunities created, 5,064 kilometres of earth roads constructed; 252 bridges, 17 earth dams, and (59) community-owned forests have been established where a total of 1,500,000 - 17 - trees have been planted. The community earth dams that have been rehabilitated or constructed across the country are acting as reservoirs for 100,000 cubic meters (1 l1km) of water in the rural areas. Poverty Monitoring Support Facility (PMSF) Achievement of the poverty-monitoring goal through the PMSF has been satisfactory. The PMSF has been instrumental in monitoring the poverty trends, gathering and analysing the necessary information for informed action among development agencies. It has also been key in establishing a PMS through provision of technical and financial support to the TWC and training of staff from institutions involved in poverty monitoring such as the NSO and NEC. Between 1996 and 2000, over eight (8) studies have been carried out, key among which were the following: (a) the Integrated Household Survey (I.H.S) which was a situation analysis of poverty in the country, (b) Indicators on Poverty and Living Conditions, (c) Determinants of Transition out of Poverty and (d) Smallholder Supply Response to Agricultural Policy Changes. The studies have been useful in bringing to the fore issues that need to be accorded priority in the targeting and designing of development programs, or poverty related initiatives. 1.3.4 IMPLEMENTATION EXPERIENCES In general, the Project enjoyed support from both communities, sector ministries and local and national leaders. MASAF I one was e very good learning ground in that most of its experiences are the backbone to the success on subsequent phases. Notable issues are summarised below: i. Multiple Approaches to financing community based development initiatives tended to generate confusion and anxiety amongst communities and at times, led to temporary delays of sub-project implementation. ii. Exclusion of some categories of the vulnerable people such as the aged, orphans, those infected and affected by HIV/AIDS who could not organise themselves into a distinct group proved to be unwelcome in MASAF I. iii. Low Participation of women CSP and PWP. iv. Delays in production of final accounts for completed projects. Experiences by Component a. Community Sub-projects * Overwhelming Demand for Project Assistance was a direct result of the demand driven principle of the Project, which was articulated by the IEC campaign. As a result, 50 percent of the funds under the comrponent were committed one and a half years after Project effectiveness. * Capacity of Community versus size of projects: It was found that appraisal of sub-projects did - 18 - not adequately consider the capacity of communities to implement the project, there were long delays (up to 24 months) in completing the projects. Large sub-projects tended to take long to complete. * Technical support to the communities: supervision for quality assurance for construction of facilities to communities was initially found to be inadequate. * Concentration of projects in the education sector: the projects under CSP were heavily skewed towards the education sector, which accounted for 54 percent of all the projects approved under the CSP component. * Drop out of PMC members: There was considerable drop out rate in PMC members. This was due to the fact that initially, only four members of the PMC were trained. * Delays in servicing completed facilities on Health Sector Projects. * Low Uptake of Projects in the peri-urban and urban areas. b. Public Works Program * Targeting: Geographical targeting using the vulnerability assessment mapping (VAM) methodology in some instances was found not to be reliable in determining vulnerable area due to time lags between the generation of the data and its actual use. * Remuneration on PWP projects: throughout the implementation period, beneficiaries have perceived the wage rate to be low especially in areas where well paying alternatives, though inadequate, managed to pay slightly better. * Predominance of road projects to be funded under PWP presented an implementation challenge of accessibility especially during the rainy season. * Time spent by beneficiaries on the project: it was found that initially, when the beneficiaries were spending close to eight hours working on the project, they did not have time to work in their fields. * Institutional Issues: the implementation of the PWP relied heavily on a collective effort between key departments at the district level, capacity of whom varied between districts. 1.4 FACTORS AFFECTING PROJECT IMPLEMENTATION Factors affecting implementation can be summarised as controllable, uncontrollable and other factors. 1.41 Controllable Factors * Limited Capacity in MASAF due to the overwhelming demand for project assistance especially at zone level. * The Justification Process was delayed due to district and community level capacity inadequacies. * Difficulties in mobilising community contribution in some projects dues to distances from source and community cohesiveness. * Lack of capacity at District Level to timely service the projects and beneficiaries. * Ineffective Computerized Project Tracking System that makes project tracking difficult. - 19 - 1.4.2 Uncontrollable Factors * Depreciation of the Malawi Currency during the entire project life time. * Breakdown in Security leading to loss of property, equipment and some service providers closing rural offices and making project management a bit difficult * Inadequate Capacity for Borehole Drilling:. * Weather and Poor Road Conditions also affected the speed of implementing sub-projects. * Staff Turn Over, inadequate staffing levels and lengthy government recruitment procedures resulted into a slow start to the implementation of the PMSF component at the National Economic Council (NEC). * Effects of the lilV/AIDS Pandemic Malawi continues to experience loss of productivity through deaths of the mostly productive people at all levels. * Multiple and Conflicting Procedures among community development Agencies: Misunderstandings with respect to other development initiatives such as the starter pack This was a targeted Govemment that distributed farm in -puts in the rural areas. Inevitably, some community members did not benefit from the program, hence the controversy and attendant negative impact on community participation it generated. or food for work, affected implementation of some projects especially under the CSP, which thrives on the self-help spirit of communities. 1.4.3 Other experiences (a)FINANCIAL MANAGEMENT ISSUES In general, the system has operated well and successfully. Accountability levels have been very high, especially on community sub-projects. However, some aspects that affected the arrangement which needed improvement. Firstly, the volumes of tranches, especially the first tranche was so small that it could only achieve a small proportion of the works. Secondly, changes in the macro-economic environment such as the fluctuation of the Malawi Kwacha against the US dollar and high inflation rates, brought about considerable pressure on the sub-project budgets which were denominated in Kwacha. This led to some of the sub-projects' targets being reduced in order avoid incurring overruns on the sub-project budgets. Thirdly, the counterpart contribution from Government was coming in drips and would not match the speed of funding requirements in MASAF. Fourthly, PMC s took a long time to account for the funds once disbursed. Fifthly, although the level of community contribution was captured during sub-project appraisal, it was not reflected in the Project accounts during implementation. The actual contribution was only captured at the end of each sub- project. (b) PROCUREMENT - 20 - In general, procurement at both the MU and community level was satisfactory done. Procurement under community participation, despite some operational challenges such as PMCs being cheated by unscrupulous traders, the judgement by PMC s on what to buy and when, was highly satisfactory However, there was inadequate capacity at MU level to handle procurement issues. There was only one procurement officer who was also responsible for other technical functions in his office and this usually delayed submission of requests for the processing of procurement related documents such as requests for proposals and contracts. The result was delays in carrying out critical steps that were a requirement of procurement. (c)SUSTAINABILITY MASAF I facilities are likely to be sustained. However, there are selected reported cases where some assets especially roads have suffered maintenance problems once MASAF funding ceases. This is because of lack of community ownership influenced by free riders and the fact that some roads pass through several villages. 1.5BANK AND BORROWER COUNTRY PERFORMANCE (a) Bank Performance The Bank's performance can be rated satisfactory. It has responded to specific issues at a fairly good pace. Bank staff have been instrumental in providing technical advice in regard to project management and procurement issues through their supervisory visits. The Bank staff accorded MASAF a conducive environment that allowed management to apply Project procedures flexibly, in line with a process oriented project. (b) Borrower performance The Government of Malawi's performance in the implementation of the MASAF Project is rated as satisfactory, as evidenced by its full commitment to the Project. Government has shown full commitment in project preparation and fulfilling its obligations for counterpart funding. The high profile that the MASAF Project assumed is also attributable to His Excellency, the State President Dr. Bakili Muluzi's support. The capacity and commitment of the district assemblies has facilitated smooth implementation of the projects at district level. However, in some district there were experiences of low commitment by Government staff which undermined the implementation speed of projects and accountability of Project funds. The MU staff demonstrated throughout MASAF I their ability to work towards the achievement of set targets. In fact, the set targets were overachieved, which was indicative of the level of commitment of the MASAF staff. The performance based job evaluation system that was introduced - 21 - in MASAF contributed to the high level of staff delivery of intended accountabilities, and in turn the Project Goal. 1.6 MAJOR LESSONS LEARNT i. The need for flexibility in the application of procedures, and the adoption of continuous learning as an integral strategy of project management. ii. Need for supporting mechanisms for community led development. iii. Need to balance the demand driven principle with Poverty Reduction iv. Need to strike a balance between accountability requirements and realizing speed in the implementation pace v. Need to Match Government's Capacity to Service Project Outputs vi. Need for Robust Procedures to ensure financial accountability vii.Role of an effective IEC in supporting a demand driven approach. viii. Need for Effective Quality Assurance and project activities tracking Mechanisms:. ix. The need for an expanded menu of projects to be funded under safety net in integrated and holistic manner. x. Need to take into account other household or individual productive activities when designing a safety net operation xi. Application of national level data for Targeting purposes: xii.Need for a special focus on peri-urban and urban areas: 1.7FUTURE OPERATIONS The overlap between MASAF I and MASAF II operations has meant that lessons learnt would not be separated by period of implementation. However, most of the lessons that have arisen from implementing the MASAF I Project have been actively included in the Project processes under MASAF II. i. Future operations should focus on actively integrating environmental issues. ii. The issues of sustainability should be adequately addressed for some types of projects such as roads whose benefits are not localised. iii. Monitoring and evaluation mechanisms should be enhanced. iv. Adequate capacity for the Project Procurement function should be put in place. v. Community contribution levels need to be captured and reflected in project accounts during project implementation. 1.8 Conclusion Annexes will include detailed tabulations and support information. Since this is only a summary of the ICR, a detailed ICR will be available for more information. (b) Cofinanciers: (c) Other partners (NGOs/private sector): - 22 - 10. Additional Information - 23 - Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: Anld'1tQfMatrix PVQjrtSd In last PSR Actual[atest Estimate Reduced poverty levels through higher 280 education subprojects implemented 735 literacy and awareness levels especially for girls and women - longer stay in school, - through primary and secondary school attendance - Improved quality of life through better health, 320 water and 80 health projects 477 water and 56 health low morbidity and mortality, higher life implemented expectancy, low fertility through access to safe water and health facilities Access to markets and economic activity 120 subprojects implernented 120 economic infrastructure projects through improved roads and bridges Improved skills and community organizabon, 800 CPCs trained and 25% of CPC 924 CPCs trained, 307 CBMs trained particpation and involvement through members are women training of Community Project Commttees and indusion of women in CPCs Poverty mapping and better targeting, - 5 BAs carried out 2 BAs carried out surveys done, reports prepared, numbers of the poor reached, changes in poverty levels, depth and distribubon, clients using servies. Note 1: End of project as projected in the SAR Note 2. These indicators were introduced in agreement with the Government mid-stream during the project and do not accurately reflect outcomes that can only be attributable to MASAF. Thus, outcome indicators monitored during project implementation (second column) are not necessarily reflective of the outcomes (first colum). The ICR team believes that the choice of outcomes reflect unfairly on the performance of MASAF as they are very general and certainly contingent upon many many exogenous factors other than MASAF. Output Indicators: lndIcatofIMatrli Projected In lastPSR ActuatiLatest Estimate Number of CSP subprojects approved 800 subprojects approved 1,392 Miles of rural road equivalent 6,000 5,036 builtVrehabilitated Persondays of employment created 12,096,000 persondays 13,600,634 Percentage of unskilled labor cost to total 36% 48% subproject cost Number of households benefiting from 120,960 136,006 employment Populabon to benefit from the employment 604,800 680,030 created Total wage income transferred MK252,000,000 MK165,000,000 Number of studies (PMSF) 14 13 Review of impact of CSP and PWP 10 3 Integrated Household Survey 2 1 End of project Note 1: End of project as projected in the SAR. - 24 - Annex 2. Project Costs and Financing Project Cost by Component (in US$ mnillion equivalent) Appralsal Actual/Latest IPercentage of Esttmiate Estimate Appralsal Project Cost By Coomponent US$ miiion US$ million Community Subprojects 48.03 47.80 99.5 Public Works Program 19.47 12.50 64.2 Poverty Monitoring Support Facility 1.51 0.40 50.6 Administration 2.31 3.00 129.9 Total Baseline Cost 71.32 63.70 Total Project Costs 71.32 63.70 Total Financing Required 71.32 63.70 Due to exchange losses, the IDA credit amount was in US dollar terms was less than envisaged at Appraisal. Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) Procurement Method' Expenditure Categoly NCS NC N..F. Total Cost 1. Works 0.00 3.00 14.47 2.00 19.47 (0.00) (2.85) (13.95) (0.00) (16.80) 2. Goods 0.00 0.22 0.09 0.00 0.31 (0.00) (0.22) (0.08) (0.00) (0.30) 3. Services 0.00 0.00 2.81 0.00 2.81 (0.00) (0.00) (2.81) (0.00) (2.81) 4. Subprojects 1.00 7.32 27.23 10.33 45.88 (0.95) (6.95) (26.05) (0.00) (33.95) 5. Incremental Operating 0.00 0.00 2.14 0.71 2.85 Costs (0.00) (0.00) (2.14) (0.00) (2.14) 6. Miscellaneous 0.00 0.00 0.00 0.00 0.00 .______________________ (0.00) (0.00) (0.00) (0.00) (0.00) Total 1.00 10.54 46.74 13.04 71.32 (0.95) (10.02) (45.03) (0.00) (56.00) Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equival nt) Proctwement- Mdethd Expenditure Categoiy 1C PM N.B.F. Total Cost 1. Works 0.00 0.00 12.50 0.00 12.50 (0.00) (0.00) (I 1.90) (0.00) (1 1.90) 2. Goods 0.60 1.70 0.20 2.50 (0.60) 0 (1.60) (0.20) (2.40) 3. Services 0.00 0.00 2.90 1.07 3.97 - 25 - (0.00) (0.00) (2.90) (1.07) (3.97) 4. Subprojects 0.00 0.00 35.00 8.00 43.00 (0.00) (0.00) (34.30) (0.00) (34.30) 5. Incremental Operating 0.00 0.00 3.00 0.50 3.50 Costs (0.00) (0.00) (2.70) (0.50) (3.20) 6. Miscellaneous 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) Total 0.60 0.00 55.10 9.77 65.47 (0.60) (0.00) (53.40) (1.77) (55.77) Note 1: Project cost includes US$ 1.8 million parallel financing from DANIDA for the Poverty Monitoring Support Facility. "Figures in parenthesis are the amounts to be financed by the IDA Credit. 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, workshops, studies and incremental operating costs related to (i) managing the project. It also includes conmnunity participation activities under works and subprojects. Project Financing by C mponent (in US$ million equivalent) Percentage of Appraisal Component Appraisal Estimate Actual/Latest Estimate IDA Govt. CoF, II.A Govt. CoF. IDA Cvt, CoF. Community Subprojects* 36.10 11.93 39.80 10.00 110.2 83.8 Public Works Program 16.80 1.67 1.00 12.50 0.60 74.4 35.9 0.0 Poverty Monitoring** 1.00 0.51 0.40 0.00 1.80 40.0 0.0 Administration 2.10 0.21 3.00 0.40 142.9 190.5 TOTAL*** 56.00 14.32 1.00 51.70 11.00 1.80 92.3 76.8 180.0 * Govemment contribution in the SAR under the Community Subprojects component includes USS8.33 million community contribution. GOM provided all the required counterpartfunding (US$2,775,303) over the life of the Credit; however, the amount of actual community contribution has not been thoroughly tracked and accounted for over the life of the project. Thus, we are unable to report an exact additional amount for community sub-projects government/community contribution even though it is estimated at about 20 percent of CSP costs. ** A World Bank administered DANIDA trust fund (USS 1.8 million) financed the Poverty Monitoring Facility under the CSO. *** Due to the high demand for CSP projects through the life of the project, some bottlenecks with the implementation of PWP projects in certain areas, and grant funds having become available for the PMSF, more funds were allocated towards the CSP projects - 26 - Annex 3. Economic Costs and Benefits A cost effectiveness analysis (" Position Paper on Project Cost Effectiveness in MASAF -Analysis and Comparativeness", MASAF Technical Services Unit, August 2001) done by the Technical Service Unit of MASAF indicates that MASAF funded infrastructure; i.e. school blocks, pit latrines, staff housing, dispensary block, roads, bridges, etc. is 20 to 70 percent less expensive than government or other donor supported infrastructure. However, this analysis does not provide cost per square meter comparisons, but rather unit cost comparisons at not always constant prices. Thus, the results have to be interpreted in light of these limitations. More accurate data was not available at the time of this report. Technical audits indicate quality of infrastructure to be "fair" with deficiencies in technical input (engineering designs, plans), supervision, infrastructure. Although, there is no quantitative data to confirm these observations, quality in some instances may have been compensated with lower costs. Based on international experience, overhead costs [Administrative or overhead expenditures include staff costs, consultants and other services, equipment, transportation, rent, utilities, supplies, communication and other non-subproject expenditures. Some socialfunds categorize capacity building and training as investment costs whereas other do not.] of social funds range between 5 to 15 percent which is usually lower than other (government agency or donor supported) programs (Letting Communities Take the Lead, A cross- Country Evaluation of Social Fund Perfornance, Laura Rawlings et al, 2001.) MASAF operating costs were about 6 percent of total project cost (not including capacity building and training costs) which places it at the lower end of the spectrum among social funds in terms of efficiency. In terms of other measures of program efficiency-speed of execution and leverage of community resources-- MASAF has performed moderately well. Completion of projects took up to 24 months (according to MASAF progress reports) in the first two-three years of the project although it has been reduced to 8 months towards the end of the program. Sub-project implementation period was foreseen between 12 to 18 months in the SAR. Community contributions (cash or in-kind) were foreseen as 20 percent of total sub-project cost for CSP projects; however, were not monitored accurately during sub-project appraisal and implementation. Thus, overall community contribution is estimated at about 13 percent of total MASAF program costs as opposed to 11.7 percent as foreseen by the SAR. Overall, MASAF has functioned fairly effectively and efficiently compared to its peers in the world and other national programs. - 27 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Perforrnance Rating (e.g. 2 Economists, I FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation 12/94 9 Economist (2), Operations S S Officer, OA, social scientist, TTL Appraisal/Negotiation 11/95 TTL, Economist (2), 00, S S OA, IEC specialist, social scientist, Financial Mnrngt. specialist Supervision 7/96 4 Economists (2), OA, VTL S S 12/96 4 NGO coordinator, economists S S (2), TTL 3/97 11 Operations officer, OA, FA, S HS social economist, NRM, IEC, architect, sociologist, economist, VT L 10/97 4 FM, economist, architect, VTL S HS 12/98 5 FM, economist, architect, IEC, S HS MIS, TTL 10/99 8 Operations officer, economist, S HS IEC, AG specialist, FMS, procurement analyst, architect. MIS 4/00 4 lEC, architect, FMS, S HS procurement analyst 11/00 3 IEC, RD, procurement analyst S HS 3/01 4 HIV/AIDS specialist, S HS environmental specialist, RD, PH ICR 11/01 4 TIL, Op.Officer, Fin. S S Mngmt Spec., Agric. Serv. Spec., 02/02 3 Op. Officer, Agric Serv. S S Specialist, Fin. Mngmt. Specialist (b) Staff: Stage of Project Cycle [ Actual/Latest Estimate No. Staff weeks USs ('000) - 28 - Identification/Preparation 115 350 Appraisal/Negotiation 80 200 Supervision 173 480 1CR 6 25 Total 374 1,055 -29- Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Fl Macro policies O H OSUOM O N * NA L Sector Policies O H OSUOM O N * NA O Physical * H O SU O M O N O NA O Financial O H O SU O M O N * NA O Institutional Development 0 H 0 SU 0 M 0 N 0 NA O Environmental O H O SU * M O N O NA Social O Poverty Reduction O H O SU * M O N O NA O Gender OH OSU-M ON ONA O Other (Please specify) O H * SU O M O N O NA Participatory/Demand driven development L Private sector development 0 H O SU O M 0 N 0 NA OJ Public sector management 0 H O SU O M 0 N * NA 1 Other (Please specify) O H O SU * M O N O(NA Local Government Capacity - 30 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bankperformance Rating E Lending OHSOS OU OHU E Supervision OiHS*S O u O HU El Overall OHS OS Ou O HU 6.2 Borrowerperformance Rating F Preparation OHS OS O U O HU ? Government implementation performance OHS OS O U 0 HU II Implementation agency performance O H1S 0 S 0 U 0 HU 1 Overall OHS *S OU OJ HU - 31 - Annex 7. List of Supporting Documents Achievement of PMSF Achievement I Indicator Description Time The Malawi National Social Collection of Infornation on indicators that measure te 1995 Indicators Survey general well being of the children of Malawi. Indicators on health, water and sanitation and education. Indicators on Poverty and Living A report that presents a set of various indicators of poverty 1995 Conditions and living conditions in the period 1990-96 The PMS newsletter Bi-annual newsletter on current poverty issues 1995 Vulnerability assessment and A targeting document of poor people done with the PMS 1996 mapping technical committee, FEWS and WFP. Determiinants of the transition out Policy study 1996 of poverty The social economic effect of Policy study 1996 structural adjustment policies. Small-holder Supply Response to Policy study 1996 Agricultural Policy Changes in Malawi. Development of a Research Policy Policy study 1996 and the operational Definition of Poverty in Malawi The integrated household survey Situation of poverty in Malawi 2000 The Vital Registration System A system to ensure generation of data on births and deaths 2000 (VRS) on a continuous process. Harnonisation of data Collection A standard reporting procedure for all collectors of 2000 information on poverty. Annual Report on poverty Annua I Statistical Booklet on Poverty A pocket on statistical indicators of poverty 2000 Capacity Building for PMS I Capacity building on data collection, monitoring and 2000 TWC members analysis on poverty Construction of a central Data A central Data Bank of all poverty indicators to be located at 2000 Bank NEC Table 10: Studies Conducted Study / Consultancy Date Details By whom impact Cost Management and Functional review 1997 _ MIM Rejected Financial management Review 1997 Iman Inproved Develop- Financial ment Manage- _ment Production of MASAF quarterly 1997 Shire Wider newsletter Conrnuni- Disemina- cation tion and feedback Organisation and automnation of MASAF 1997 H. Gausi Improved publications and records and W. filing and Dzoweda access to - 32 - records IEC consultants 1997 Shire IEC Communi- outreach to cation stake- holders Borehole drilling supervision Manjoo Improved Consulting SPN and quality Quarterly audits 1997 Various Trans- films parency, account- ability lEC Radio consultancy 1998 MBC lmproved awareness of project processes Review and Evaluation of MASAF safety 1997 Wadonda Improved net programmne Consult targeting Needs assessment of MASAF peri-urban 1997 R. Fatch Prep. study community and team for MASAF 2 CSP impact on the social economic 1997 Gill Prep. study status of beneficiaries Enterprise for Cons. MASAF2 Policy and regulatory framework in 1997 C. Prep. study urban Simwaka for MASAF 2 Service provider study 1997 A.J.C. Improved Kawonga service delivery Supervision of borehole drilling 2000/2001 Establish- programme ment of systematic SPN schedules Ground water management system. 2000 Improved boreholes planning and drilling Architectural and quantity surveying 1999/2000 Khorandji Reviewed consultancy an & drawings, Macharia and BoQs Kamal Technical assistance, MASAF website 2000 WiSS Wider development. dissemina- tion and feedback _ Project Tracking system 1999 Fady System installed and pilot tested - 33 - Technical assistance in MASAF MIS 2000 Kamal Production Siblini of designs Technical assistance in MASAF Local Afcor Introduced Area Network the LAN MASAF lEC Impact Assessment 2000 Chilowa Wider and dissemina- Associates tion and feedback Technical assistance in MASAF 2000 MIM Impact on Beneficiary assessment. benefi- ciaries PWP Beneficiary Assessment 10/7/00 - Wadonda Same as $86,320 20/10/00 above CSP Beneficiary assessment 1/11/99 - Bunda Same as $94,219 30/6/00 Team above Environmental Impact Assessment Millenniu Produced m ElA guidelines Client Consultation study Apprecia- tion of the baseline informa- tion of benefi- ciaries Baseline survey for PWP Wadonda Apprecia- tion of baseline info for post- evaluation Bi-Annual Audits Consultancy Graham Project Carr, audits for Mwenelup improved embe, financial Delloite & manage- Touch ment Management Development Review 10/3/00 - Improved $49,736 30/7/00 manage- ment and staff appraisals Radio Consultancy MBC Airing of radio pro- gramrnes on MBC Radio 1, 7 times a ___ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .w eek Video Consultancy Malawi Not Institute of finalised Education - 34 - MASAF 1 Review Consultancy Graphic Profes- McCann sional review of the project MASAF News Consultancy Champion Wider Publica- dissemina- tions; tion and Ministry feedback of lnforma- tion Television Programme Consultancy TVM Production of IEC materials Montfort Press; Litho Publishing Co. Dzuka Printing .- - 35 - Additional Annex 8. Public Works Program Targeting Mechanisms To ensure distribution of program benefits to the poorest and achieve maximum impact, public works programs, in general, rely on "self-targeting" of beneficiaries by using a wage rate which is slightly below or around the minimum wage rate. This wage rate would only attract those who do not have better employment opportunities elsewhere, and would not attract people away from other more productive economic activities. The impact and success of the program, transferring income to the poorest, will depend, among other things, on setting the appropriate wage rate. The PWP program wage rate was revised a number of times to adjust for inflation and changes in living conditions. However, throughout the life of the project, the program wage rate seems to have been higher than the national weighted average poverty line, and the minimum wage rate. The initial program wage rate 7.50 MK per day (1996 prices) was lower than the wage rate paid to unskilled workers by Ministry of Works, but higher than the poverty line, 6.82MK (1996). In 1997, program wage rate was revised to 10.50 MK per day.. In 1999, program wage rate (task rate for about 4 hours per day) was revised to 15.80 MK whereas national weighted average poverty line was about 15.0 MK and minimum wage rate was at 15.90 MK (for an 8 hour work-day). Program wage rate at closing (2001) was 25.0 MK (for a four work-day) whereas poverty line was at about 20.6 MK (2001 prices) and minimum wage rate was at 37 MK (for an 8 hour work-day). Thus, program wage rate has consistently been above the poverty line and about twice the minimum wage rate. It is argued by MASAF MU that the minimum wage rate has not been reflective of the rapidly deteriorating living conditions thus was not used as a benchmark. This, in turn, is likely to have adversely impacted the poverty profile of individuals/households self-selecting into the program; i.e. beneficiaries were not always the neediest. As a result, in the absence of a systematic survey of poverty incidence of beneficiaries, it would be difficult to say with confidence that the PWP only benefited the poorest of the poor. In rural areas, with wage rates almost double the poverty line and the minimum wage rate (due to task rates as opposed to daily rates), PWP probably attracted more than just the poorest. - 36 - Additional Annex 9. Government ICR (full text) * In addition to borrower comments included in the main text, the full Government ICR is attached as an annex because of the quality and comprehensive nature of it GOVERNMENT OF MALAWI OFFICE OF THE PRESIDENT AND CABINET MINISTER OF STATE RESPONSIBLE FOR POVERTY ALLEVIATION IMPLEMENTATION COMPLETION REPORT (1CR) FOR MALAWI SOCIAL ACTION FUND PHASE 1 (MASAF 1) 1996 TO 2001 Dated: Monday, 10 June 2002 - 37 - The Ministry of Finance The National Economic Council P.O Box 30049 P.0 Box 30136 Capital City Capital City Lilongwe 3 Lilongwe 3 Tel: (265) 789 355 Tel: (265) 789 390 MASAF MU Private Bag 352 Lilongwe 3 Malawi Tel: 755 666 Email: masaf(namasaf.org 1.0 BACKGROUND Since 1964, when Malawi attained independence, the country emphasised maintaining macroeconomic stability, with a lot of emphasis on the agriculture sector as the lead industry in the development strategy. Poverty was to be addressed through a trickle down effect. However, this approach did very little to reduce poverty, as economic growth stagnated starting from the late 1970's through the 1980's. Drought and unfavourable prices for agricultural products in the world market greatly affected the performance of the economy, and led to instability in the economy. Against this background, stabilisation and poverty reduction had emerged as central objectives of economic management. In 1993, a situation analysis of poverty revealed that poverty in Malawi was very pervasive and deep. The country's social and economic indicators were among the worst in southern Africa and the world in general. Low access levels to safe drinking water (at 53 percent of the population), high infant mortality rates (134 per 1000 per births), poor socio-economic infrastructure and widespread deprivation reflected the deep levels of poverty in the country. In addition, the lack of alternative off-farm income earning opportunities and persistent food deficiencies in most parts of the country, and particularly those in the rural areas meant that people were in a severe state of distress as they could not easily meet their basic needs for the sustenance of life. An autocratic environment, with development approaches that did not adequately allow for the participation of communities in the development process also held in check the social energies that would lend themselves to sustainable development at the local level. Following the 1994 general elections and open acceptance of the extent of poverty in the country, the incoming government raised the poverty alleviation agenda to centre-stage in its development policy framework. This led to the formulation of the Poverty Alleviation Program (PAP), and subsequently an implementation framework in which MASAF was one of the key instruments for operatianalising the program. The MASAF design was the result of a long series of consultations among which were the - 38 - participatory poverty assessment (PPA) with communities and a systematic client consultation (SCC) with major stakeholders such as NGOs and communities in early 1995. A pilot phase was undertaken from mid 1995 and generated a wealth of experiences that fed into the design of the mainstreamed phase of the Project. The Project was appraised by the World Bank, then the sole source of funding for the Project, in November /December 1995 and became effective on May 15, 1996. 1. OEvolution of MASAF It was envisaged that MASAF would contribute to the goal of poverty alleviation through: (a) the upgrading of community infrastructure by rehabilitation of old structures and construction of new ones in education, health, water and transport and communication sectors, (b) improvement to the delivery of and access to social and economic services, (c) fostering grass root initiatives (social action) intended to enhance the technical and managerial capabilities and attitudes to self-managed development, and (d) reorienting and building capacity of actors in community development including communities, extension workers, NGOs, the private sector and religious groups to equip them with skills in participatory methods for planning and implementing poverty alleviation interventions. 2.0 MASAF I OBJECTIVES AND COMPONENTS The interventions under MASAF I were perceived to fall into the two main categories (a) Social Infrastructure and Assistance: that would involve the construction and rehabilitation of primary schools, distant education centres, teachers houses, water supplies, sanitation, improving sewerage and drainage systems in urban and semi urban centres, construction and repair of health centres, clinics and staff houses; and (b) Economic Infrastructure: that would boost the productive capacities and create employment opportunities for the poor. Projects included small-scale irrigation schemes, village access. The objectives of the Project were thus set as follows: * To provide additional resources for programs targeted at the poor, that is, fund the creation of village level assets, which will be directly beneficial to the poor through investments in primary education, peripheral health services and safe water. * To promote a new approach to rural development by involving communities in project preparation and implementation, while encouraging government agencies, NGOs and private institutions to assist communities when needed. To support a district level program of labour-intensive construction to be targeted at the poorest districts as safety net operation.. * To strengthen poverty monitoring and assessment systems through financing a Poverty Monitoring Support Facility (PMSF). 2. INSTITUTIONAL SET-UP AND PROJECT COMPONENTS Implementation of the Project was to be done through a Management Unit, comprising of a head office and zone offices that would work with district level staff. A Project steering Committee - 39 - comprising of key sectoral ministries, one traditional chief, and independent members, provided overall policy guidance. Sub-projects would be implemented at community level through a community project committee (PMC), which was delegated functionary of the community. The MASAF 1 objectives were to be achieved through three operational components: the Community Sub-Projects (CSP) component, the Public Works Component (PWP) and the Poverty Monitoring Support Facility (PMSF). In addition, there were support divisions and units such as Administration Division, Finance Division, Internal Audit Unit, Informnation Technology Unit, Technical Services Unit, and the Information, Education and Communication (IEC) unit. The PMSF was coordinated by National Economic Council (NEC) through the Poverty Monitoring System (PMS) Technical Working Committee (TWC) composed of data users and producers. All these played an integral part in the realisation of MASAF I goals. 2.1.1 Community Sub-projects (CSP) The CSP component accounted for 71 percent of the funds under MASAF I, was designed to meet the first two objectives.. Firstly, it would provide financial and technical resources for programs targeted at the poor by funding the creation of community assets and secondly, it would promote a new approach to community development by providing an opportunity for communities to effectively participate in the identification, preparation and implementation of their own development projects in collaboration with government agencies, NGOs and private institutions. The project processes would empower communities by creating opportunities for communities to revive the spirit of self-help, enabling them to take control over, and be accountable for their own development. 2.1.2 Public Works Program (PWP) The PWP component accounted for 27 percent of the financial resources under MASAF I. It was designed to transfer income, through the creation of employment opportunities by funding labour intensive public works, and in the process, build economic infrastructure that would be beneficial to communities. The activities included construction, rehabilitation and maintenance of economic infrastructure such as rural roads; small-scale irrigation systems; improved natural resource management through afforestation, terracing and rainwater harvesting. 2.1.3 Poverty Monitoring Support Facility (PMSF) The PMSF component accounted for 2 percent of MASAF I financial resources and was managed by the National Economic Council (NEC). The PMSF supported the poverty reduction policies and programs by bridging poverty-related gaps and developing analytical skills in government and private institutions. It financed the essential building blocks for collection and analysis of data relevant to track progress in poverty reduction and to help guide poverty alleviation policies and guidelines. 3.0 ASSESSMENT OF THE PROJECT This section assesses the MASAFI Project objectives, design and implementation. The assessment of - 40 - the objectives and design of the Project are done with reference to their relevance to Government development priorities, and the prevailing conditions at the time of design. The assessment of implementation is done against the extent to which the Project was able to meet its stated objectives. 3.1Assessment of project objectives The MASAF I objectives as stated in the Staff Appraisal Report of 1995 did not change in the course of implementing the Project. The objectives would be rated as satisfactory. They were relevant to and consistent with the Malawi Government's poverty alleviation policy framework (PAPF) for achieving poverty reduction. The community development financing objective addressed shortfalls in the provision of social sector services such education, water and health, which are critical for poverty alleviation. The benefits arising from investments in the social service sectors would accrue mostly to the very poor, given the pervasive nature of poverty in Malawi. Furthermore, facilities such as water points (mainly boreholes) and school blocks were envisaged to benefit women and children directly, who are amongst the groups identified as being most adversely affected by poverty in the PAPF. The safety net objective aimed at ameliorating hardships faced by individuals and households in food insecure EPAs, was relevant and sound in an environment where there was limited alternative opportunities and chronic food shortages. It would contribute to meeting the basic requirements such as food for the very poor who otherwise would not be able to weather the impact of extreme poverty and within a difficult economic environment. The poverty monitoring objective directly addressed gaps in the systematic collection and analysis of poverty related data and build capacity in policy analysis. It was therefore consistent with, and would contribute to strengthening of the Poverty Monitoring System under the PAPF. 3.2Project Design In general, the design of the project could be rated as highly satisfactory and appropriate for attaining the set objectives The quality of the design was greatly improved by input from international and domestic experience with social investment funds, and the lessons learnt in the pilot phase of the Project. Transparency, accountability, community participation, gender sensitivity and direct community financing that MASAF embraced in its design were principles that would contribute to a program geared towards empowering the poor. In particular, under the CSP, a strategy that emphasised the participation of the poor in a country emerging from the shadows of a generally inhibitive environment was consistent with a people centred development approach and the spirit of the PAPF. Financing communities directly and allowing them to identify their needs and prepare project requests allowed for the communities' effective participation in the Project activities and development efforts. In addition, district level processes such as desk and field appraisal and monitoring and supervision offered a framework through which Government, Project beneficiaries, NGOs and MASAF would work together to meet community development needs. Under the PWP, the geographical and individual or household targeting mechanisms and the labour intensity principles enshrined in the design of the component were appropriate for a safety net operation. As a result, the component was able to effectively reach its intended beneficiaries. - 41 - The design of the PMSF component was to a large extent appropriate. It emphasised capacity building efforts that would be critical for sustained analysis of poverty related data in the long term. It would contribute to program learning and lead to improvements in the design of poverty related interventions. 4.OACHIEVEMENT OF OBJECTIVE The overall Project objective was to contribute to poverty reduction. Although the level of contribution to poverty reduction was not quantified, the Project development objective could be said to have been substantially achieved. An estimated 2.0 million people (or 20 percent of the Malawian population) directly benefited from investments in the education, health, water and sanitation, transport (roads and bridges) and afforestation sectors; and the transfer of income through the PWP component An approach that actively involved beneficiaries also fostered a collegial spirit (which was hitherto adversarial) in community development efforts between communities and Government agencies. In total 1,864 sub-projects were approved, of which 1,392 projects were under the CSP component and 472 projects under PWP component. Under PMS, training for key personnel at the National Economic Council and various studies have been undertaken. In addition, equipment for data analysis has been procured for the National Statistical Office. In regard to promoting a new approach to development, there have been discernible effects, particularly in shifting the perceptions of the major stakeholders in the Project and the attendant trust the approach has generated. The Project demonstrated that it is a viable and effective approach for channelling project assistance to communities. For instance, although there is no information to indicate that the MASAF approach has been replicated within the country, commercial banks such as then National Bank of Malawi and the National Bank of Malawi have channelled funds through the Project to finance water kiosks in Blantyre and Lilongwe districts. The direct financing principle, along with transparency and accountability requirements have been key in the promotion of this new paradigm in the country. 4.1Project Impacts A socio-economic impact study of the CSP (MASAF 1997 ) and the PWP (MASAF 2000) on the beneficiaries and communities showed that project effects were to be seen in the following areas: Reduction in incidences of water borne diseases: in communities where water facilities such as boreholes have been completed, beneficiaries have reported that outbreaks of cholera and diarrhoea. For instance, in Chatha community in the outskirts of Blantyre, cholera which was yearly occurrence in the community was never experienced again after the completion of the piped water project in the community. Other communities country-wide have reported similar experiences. * Increased enrolment: in the second CSP BA, in 9 of the 25 study sites, there was increased enrolment of pupils in schools. Beneficiaries cited reduced distances and improved learning environment as a result of the new class rooms. - 42 - Improvement in Socio-economic Status: The PWP second BA revealed that participation on the PWP safety net operation led to improvements in the quality of life of the beneficiaries. Seventy percent of the sampled beneficiaries felt that their economic status had improved due to their paTticipation in PWP. Descriptions of the improvement in economic status included from ability to purchase basic necessities, saving up some of the proceeds from the wages, owning small businesses and owning more assets than was the case before participation in the PWP. * Improvement in communication and access to social and economic facilities: under the CSP and the PWP, the construction of roads and bridges has improved communication and access to schools, markets, postal agencies and other services, which was not possible before the completion of the projects. This has faciliitated business and education among others. * Stimulation of the rural economy: project activities under both the CSP and PWP are stimulating the rural economy through employment. Although not quantified, the Project activities create demand for goods and services in the rural economy. For instance, construction activities are generating employment and demand for goods such as timber. The income earned from these activities brings about a multiplier effect as it in turn generates demand for other goods such as food and clothing. Fostering of a Collegial Spirit towards community development efforts: the project processes, re-enforced by the IEC has engendered a collegial spirit and strengthened the social bonds between communities, their leaders and Government officers in development efforts. Communities, and Government are now better able to work together in a joint effort towards satisfying community development needs than was the case before. In the past, and as a result of an autocratic environment and the top- down approach to community development, the relationship between communities, leaders and Government was one of mutual suspicion and adversarial Improved poverty data collection and analysis: the project has for the first time assisted government to collect up to date information on poverty status of Malawians. It has also imparted technical and analytical skills in analysing the data as input into policy formulation. Further, the project has built capacity of the institutions under the PMS through short and long-term training. 5.01MPLEMENTATION RECORD This section assesses the implementation record of the MASAF I Project by component. Overall, achievement of objectives on all the three component was satisfactory. Table A below is a summary of achievements on end of project output indicators. Details of achievements on each component are offered in annex 9. Table A: Achievement on Project End Output Indicators by Component Notes: Under PWP, employment created refers to that of unskilled workers Exchange rate at time of World Bank Project Appraisal was $l=MKl5.0 An average household was assumed to consist of 5 members. The total population benefiting is derived from that size of the household. -43 - Beneficiary Assessments were envisaged to be done under PMSF. However, in the course of implementation, they came to be done by the MASAF MU. Component/ Indicator Project End Achievement Achievement (% B Target (Absolute of A) (A) Terms)OfA (B) A. Community Sub-projects (CSP) Number of projects Approved 800 1392 174 Number of completed projects 800 1288 161 B. Public Works Program (PWP) Kilometres of Rural Roads 4,563.2 5063.9 110.9 Completed in '000 Number of households benefiting 120.96 136.0 112.4 from employment created in '000 Population to benefit from 604.8 680.0 112.4 employment created Total Wage income transferred (MK 252 184.7 73.3 million) C. Poverty Monitoring Support Facility Beneficiary Assessments (number) 5 2 40.0 Studies Completed 14 13 92.9 Source: World Bank (1996), Staff Appraisal Report for the Malawi Social Action Fund Project and the Current MASAF I Status Reports. 5.1 Community Sub-projects Achievement of objectives under the CSP component has been highly satisfactory. Against the projected 800 projects to be approved under the CSP in its entire life, 1,392 projects were approved. The overachievement is attributed to four main factors: (a) As a result of the overwhelming demand, the number (20) of projects under implementation in a district at any one time was surpassed during implementation with the number rising to an average of 60 projects per district, (b) The estimated average cost for a typical CSP sub- project varied from district to another depending on the component package, sources of materials and distances from the main trading centres. There is evidence that that more sub-projects were implemented that originally estimated. - 44 - (c) The depreciation of the Malawi Kwacha on the US dollar led to an accumulation of addition funds on the original amount earmarked for the CSP, and enabled the component to approve more projects, and (d) Funds carried forward from the pilot phase and other expenditure categories in the mainstreamed phase that were likely not to be spent before the end of the project were transferred to the financing of additional sub-projects. The sectoral distribution of the approved projects was as follows: education 735; water 307 involving the drilling of 921 boreholes and construction of 80 water kiosks; transport and communication 120 bridges; health 56 (under-five clinics/maternity units and health centres); and other projects 13. Projects in the education sector involved the construction of 2,146 new classrooms, 854 staff houses; 56,445 school desks and 4,478 VIP latrines. The benefits accruing to the beneficiaries from the assets financed under the CSP include improved learning and increased enrolment rates in the project sites than was the case before construction of new classrooms; non-occurrence of water and sanitation related diseases such as cholera in project areas; and improved access to other social and economic facilities as a result of roads and bridges. These have been beneficiaries' perceived effects as revealed by both the two Beneficiary Assessments (BA s) and the Socio-economic Impact Studycarried out on the CSP component. In addition, a cadre of enlightened individuals has been created at the community level for local management of development initiatives. The successful implementation of sub-projects, with high levels of financial accountability by PMCs is testimony to this. In total, 19,140 people at the community level have been trained in project implementation management and community based management of water facilities. Fifty percent of this figure is composed of women. In PMCs, the representation of women is 30 percent while in Community Based Management committees for water facilities, women account for 60 percent of the members. 5.2Public Works Component Achievement of the safety net objective was highly satisfactory. The PWP component (USD13.7 million) financed district level safety net operation that has benefited poor people in food deficient Extension Planning Areas (EPAs). A total of MK185 million (US$2.5 million) 1 US$ = MK73.00 in cash has been transferred to 188,000 people through creation of temporary employment. This has been achieved through the financing of a total of 472 labour intensive projects such as roads, community dams and afforestation projects. The benefits arising from the PWP component have directly reached approximately 680,030 people or about 7.6 % of the Malawian population. However, considering that utilization at the household level would benefit more people than those directly participating on the project, the benefits reach more people than captured here. The projects approved under the PWP have translated into the following outputs: 188,000 employment opportunities created, 5,064 kilometres of earth roads constructed; 252 bridges, 17 earth dams, and (59) community-owned forests have been established where a total of 1,500,000 - 45 - trees have been planted. The community earth dams that have been rehabilitated or constructed across the country are acting as reservoirs for 100,000 cubic meters (11 lkm) of water in the rural areas. The dams have enabled communities to establish small-scale irrigation schemes, which have facilitated agricultural activities such as vegetable gardening in the project areas. The dams have also provided the areas with water for domestic use and for livestock consumption. These dams also check water run off and thereby control soil erosion. See Annex 9 for details of achievements under PWP. The second PWP BA, which focused on the socio-economic impact of the PWP component on beneficiaries and communities found that over 80 percent of the beneficiaries used the income eamed from working on PWP projects on basic necessities like food and health. Thus, the PWP represented a major source of additional income that was accessed by poor people who would not otherwise meet their food requirements. In addition, 70 percent, of the sampled beneficiaries in the second BA felt that their economic status had improved due to their participation in PWP. With the income earned from participating in the PWP they were now able to purchase basic necessities, saving up some of the proceeds from the wages, owning small businesses and owning more assets than was the case before participation in the PWP. Facilities such as roads have also greatly improved access to markets and schools that were hitherto inaccessible. Benefits from afforestation and community dam projects will also be significant Apart from improving soil conservation and replenishing the stock of trees in the project areas, woodlots will be sources of wood-fuel at community level at convenient distances. 5.3Poverty Monitoring Support Facility (PMSF) Achievement of the poverty-monitoring goal through the PMSF has been satisfactory. The PMSF has been instrumental in monitoring the poverty trends, gathering and analysing the necessary information for informed action among development agencies. Another funding from the Danish government complemented MASAF assistance to NEC. It has also been key in establishing a PMS through provision of technical and financial support to the TWC and training of staff from institutions involved in poverty monitoring such as the NSO, CSR, Ministry of Health and NEC. Between 1996 and 2000, a number of activities have been carried out with the support from the PMSF. Over thirteen (13) studies have been carried out, key among which were the following: (a) the Integrated Household Survey (I.H.S) which was a situation analysis of poverty in the country, (b) Indicators on Poverty and Living Conditions, (c) Determinants of Transition out of Poverty and (d) Smallholder Supply Response to Agricultural Policy Changes. In addition, the PMS Technical Working Committee members were oriented on poverty monitoring related issues. See table 12 for a list of reports and studies under the PMS. The studies have been useful in influencing policies and also bringing to the fore issues that need to be accorded priority in the targeting and designing of development programs, or poverty related initiatives. - 46 - 6. IMPLEMENTATION EXPERIENCES In general, the Project enjoyed support from both communities, sector ministries and local and national leaders. This section highlights implementation experiences in MASAF I: 6.1 Overall Project Experiences Multiple approaches to financing community-based development initiatives: this tended to generate confusion and anxiety amongst communities and at times, led to temporary delays of sub-project implementation. In some instances, there was duplicate funding with other donors. Exclusion of some categories of the vulnerable people: groups such as the aged, orphans, those infected and affected by HIV/AIDS due to the fact that could not organise themselves into a distinct group that would be assisted under MASAF. In view of this, the Project included as one of the components the Sponsored Sub-projects to specifically target interventions to these groups under the second phase of MASAF. Low participation of women CSP and PWP: under the CSP, it is estimated that although women held senior positions in the project committees (e.g. chairperson, treasurer, secretary) they constituted only up to 30 percent of the PMC membership, with similar levels recorded under PWP, especially on road works. The Project embarked on targeted IEC to encourage women to participate in the Project. Results have been minimal though, especially under the CSP where the proportion of women participating in the PMCs has not changed. Low participation by members in TWC meetings (PMSF): over time there was poor participation of members in TWC meetings. NEC, NSO and CSR remained core institutions of the PMS and other members took back sit role in PMS activities. Delays in production of final accounts for completed projects: considerable delays were experienced in the production of final accounts for completed projects under both the CSP and PWP. This in turn led to delays in declaring the projects completed, although physically the projects had been completed and in use. Under PMSF delays were also experienced in the disbursement of funds and procurement of equipment. Transfer of funds between Citi Bank in Washington and Central Bank in Malawi in most occasions took long time to process and access. Further, there were delays at MASAF office in the procurement of equipment such as computers and vehicles. The project also experienced late recruitment of Technical Assistance from IFPRI. 6.2Experiences by Component 6.2.1 Community Sub-projects Overall, communities were able to effectively manage the implementation process, including - 47 - accounting for funds. Activities under the component enjoyed support from the communities and their leaders, both traditional and political. Although this translated into capacity constraints and pressure on MASAF to maintain its reputation and trust from communities, it contributed to the smooth and quickened pace in implementation of the CSP component. The following is a summary of experiences under the CSP component: Overwhelming Demand for Project Assistance: This was a direct result of the demand driven principle of the Project, which was articulated by the IEC campaign. The Project found itself inundated with project applications within the first year of its operations in the mainstreamed phase, which it could not fully meet given the resources available at that time. As a result, 50 percent of the funds under the component were committed one and a half years after Project effectiveness. * Capacity of Community versus size of projects: It was found that when appraisal of sub-projects did not adequately consider the capacity of communities to implement the project, there were long delays (up to 24 months) in completing the projects. Large sub-projects tended to take long to complete. In this connection, subsequent approvals were based on the demonstrated effective demand for a Project. This was determined by the actual amount of community contribution available Technical support to the communities: supervision for quality assurance for construction of facilities to communities was initially found to be inadequate. The assumption that technical skills would exist at community level did not hold for the CSP program. In order to overcome this problem, technical supervisors based in each district were engaged and the quality of the structures improved to expected acceptable standards. * Concentration of projects in the education sector: the projects under CSP were heavily skewed towards the education sector, which accounted for 54 percent of all the projects approved under the CSP component. The second BA for the CSP component established that this was a reflection of the needs in that sector which was largely occasioned by Government's Free Primary Education Policy launched in 1994. * Drop out of PMC members: There was considerable drop out rate in PMC members. This was due to the fact that initially, only four members of the PMC were trained. This created despondency and alienation among the members that did not go for training, and led to their dropping out of the PMC. The effect of this was delayed implementation of sub-projects as the remaining four mernbers became overburdened. The project switched to training all the ten members of the PMC, and the drop out rate was greatly curtailed * Delays in Servicing Completed Facilities on Health Sector Projects: delays were experienced in operationalising most of the completed health projects such as dispensaries and maternity units. Once completed, the Ministry of Health and Population (Mop) could not immediately be staffed, reflect severe capacity limitations in that sector. This was despite endorsement by sector representatives the projects should be approved and that they would be serviced. The Project had to temporarily suspend the approval of facilities such as dispensaries and maternity units, pending resolution of the difficulties in the sector. The Mop has now put in place a strategy to have these - 48 - facilities staffed. * Low Uptake of Projects in the peri-urban and urban areas: there were very few projects that were implemented in the urban areas. Projects approved for the urban areas under MASAF I were less than 2 percent of all the projects approved. The main reason for low uptake was the low level of social cohesion in the urban areas, as compared with the rural areas. Mobilization for creation of public facilities was therefore problematic. 6.2.2 Public Works Program The PWP component has been instrumental in targeting the poorest of the poor thereby having a direct impact on poverty reduction. More importantly is the fact that PWP, besides the cash, has contributed towards the creation of physical assets such as access roads, community owned forests, community earth dams and bridges. The implementation experiences under the PWP included the following: * Targeting: overall, the targeting mechanism was efficient, and most of the beneficiaries participating in the project were indeed vulnerable. According to a baseline survey carried out in 1998,seventy-nine percent (79%) of the beneficiaries cited poverty and lack of alternative income earning opportunities as reasons for participating in the PWP projects (MASAF, 1998). Geographical targeting using the vulnerability assessment mapping (VAM) data in some instances was found not to be reliable in determining vulnerability due too the fact that there was time lag between the generation of the data by the famine early warning systems (FEWS) and its actual use. Where this was the case, District Implementation Teams (DTs) were allowed to choose alternative EPAs that were deemed to be more vulnerable than the VAM data indicated. * Remuneration on PWP projects: throughout the implementation period, beneficiaries have perceived the wage rate to be low. The initial policy was to keep the wage rate below the official Government minimum wage rate, which was consistent with the principle of a safety net. However, due to the depressed nature of the wage rate coupled with high inflation rates, the wage rate had to be set above the official minimum wage rate. * Menu of projects to be funded: a predominance of road projects on the menu of project to be funded under PWP presented implementation challenges. For instance, during the rainy season, it proved extremely difficult to implement road projects in most parts of the country due to heavy rains, which occasioned stoppage of works due to rains. However, the period during which works were suspended was when there were food deficiencies and when most poor people needed the income from PWP. The menu of projects has had to be expanded with the diversification of projects into afforestation, community dams and valley tanks. * Time Spent by Beneficiaries on the Project: it was found that initially, when the beneficiaries were spending close to eight hours working on the project, they did not have time to work in their fields. There was therefore the possibility that there vulnerability would worsen as they would not have food in the following season. Instead of doing an eight hour job, the procedure was modified so that the work became task based as opposed to time based and beneficiaries worked on the project - 49 - only for an average of three hours. This arrangement gave them time to work in their fields. * Institutional Issues: the implementation of the PWP relied heavily on a collective effort between key departments at the district level in its design. This meant that capacity gaps in terms of human and financial resources obtaining at that level affected implementation of the component. Although capacity building efforts such as training of the district teams were undertaken, areas such as financial accountability problems were only partially resolved. With a system where subsequent funding to district teams was made in tranches, speed in implementation was held back. 6.2.3 Poverty Monitoring Support Facility (PMSF) The PMSF has been instrumental in monitoring the poverty trends, gathering and analysing the necessary information for informed action among development agencies. It has also been key in establishing a PMS through provision of technical and financial support to the TWC and training of staff from institutions involved in poverty monitoring such as the NSO, CSR, Ministry of Health and NEC. 7.OFACTORS AFFECTING PROJECT IMPLEMENTATION While the implementation of MASAF in general could be rated as highly satisfactory, there were a number of factors that affected smooth implementation. Some of the factors were within the control of the Project, while some were not. This section highlights the factors that negatively affected the implementation of the Project. 7.1 Controllable Factors These largely related to capacities within the MASAF Management Unit, the district assemblies and the communities. These are summarized below: * Limited capacity in MASAF: due to the overwhelming demand for project assistance, there were capacity limitations in processing and the management of operations under both the CSP and PWP components. The number of zones under the CSP was thus increased from 9 to 12, while four zones were introduced under PWP. This assisted in speeding up the processing and implementation of sub-projects. * The justification process: there were delays in replenishing sub-projects accounts arising from capacity constraints at MASAF and community levels. Delays in processing justification reports by PMCs and District Teams, coupled with inadequate staffing levels at MASAF MU to verify authenticity of sub-project expenditures in turn led to delays in the implementation of projects. Improvements in the training content and an addition of staff in the Justification unit helped in resolving the problem. * Difficulties in generating community contribution: In some communities, it was difficult to generate materials that are ordinarily perceived to be locally available such as stones and construction sand, which would normnally be found outside the project area. This had varying effects on projects including delayed implementation. The Project had to allow for transportation of such materials - 50 - where this proved to be the case. * Lack of capacity at district level: PWP projects experienced delays in funding. Beneficiaries in some projects waited up to a month before they getting their wages. This was closely connected to the problems of lack of transport by the District Assemblies. This also contributed to problems of inadequate supervision by the supervisors and therefore quality assurance lapses in some projects. The Project had to provide funds for running the District Assembly vehicles in a bid to limit the problem. * Ineffective computerized project tracking system: the development of a computerised project tracking system took long, and was affected by long periods of lack of staffing. This meant that trends in project progress could be monitored effectively on an aggregate level. The Project had to have recourse to departmental data bases for storing data that would assist in monitoring progress. * Management problems under PMSF: a lot of management problems were experienced both in NEC and World Bank offices in Washington and Lilongwe mainly due to staff turn over, break down in communication and lack of feed back and follow-up on important issues. 7.2Uncontrollable Factors The following were factors outside the control of the Project which affected implementation: * Depreciation of the Malawi Currency: The exchange rate of the Malawian Kwacha has been depreciating since 1995. This resulted into pressure on the Malawi Kwacha denominated budgets and in some instances, led to the cutting off of agreed components on sub-projects when the problem was first experienced. This also had a deleterious effect on the income earned by PWP beneficiaries, which remained constant for some time in the face of general rise in prices as a result of this problem. PMSF project implementation activities and procurement of equipment were also affected by the depreciation of the Malawi Kwacha. * Breakdown in Security: This also had a strong negative affect on implementation of MASAF 1 projects. A spate of armed robberies saw commercial banks closing down in some districts. As the banks are key in facilitating project transactions, closure of the branches slowed down implementation pace of the projects in those areas considerably. In addition, the Project lost two of its zone vehicles to these armed robberies, and for a time, this affected project operations in the two affected zones (Karonga and Zomba Zone). Inadequate capacity for borehole drilling: there were severe capacity limitations to drill boreholes in the country. In the first one and a half years of the Project, boreholes on some approved stayed up to a year after approval without being drilled. Weather and poor road condition: Poor road conditions in some parts of the country, particularly in the Northernmost district of Chitipa and parts of Machinga and Mangochi in the Southern Region of the country also affected the speed of implementing sub-projects. During the rainy season, this problem was compounded and projects had to be stopped until the rains subsided - 51 - as it was difficult to bring construction materials to the project sites by vehicle. * Staff turnover: The implementation of the PMSF was particularly affected by this. The high staff turnover and inadequate staffing levels at the National Economic Council (NEC) resulted into a slow start to the implementation of the PMSF component. Lengthy Government recruitment process did not assist in resolving the problem. In addition, MASAF I the decentralisation process which was also being implemented at the same time as MASAF saw a number of district staff being transferred from one district to the other. This brought uncertainty and instability at the district level which affected the implementation of the PWP projects which relied heavily on district staff. Effects of the HIV/AIDS pandemic: Just like any other Sub-Saharan country, Malawi has experienced loss of productivity through HIV/AIDS related deaths of the mostly productive people at all levels. This has had devastating effects on project implementation at national, district and community levels. Multiple and conflicting procedures among community development agencies: Misunderstandings with Tespect to other development initiatives such as the starter pack This was a targeted Government that distributed farm in -puts in the rural areas. Inevitably, some community members did not benefit from the program, hence the controversy and attendant negative impact on community participation it generated. or food for work, affected implementation of some projects especially under the CSP, which thrives on the self-help spirit of communities. Community members who did not benefit from the programs demanded that those who received the starter pack for example, should do the work. Misunderstandings were also prominent in areas where some donors had a different approach to project financing and implementation. This was slowly sorted out with more vigoTous IEC on MASAF activities. 8.00ther experiences: 8.1 FINANClL> MANA GEMEWNT ISSUES MASAF I was funded to the tune of USS 56 rnillion. The IDA contributed US$ 51.4 million and the Malawi government contributed US$8.3 million through community contribution and US$2.9 million in cash. Disbursement from IDA was done to MASAF in tranches into the special account. As at June 2001, a cumulative total of USD34.2 million was disbursed out of the total allocation of USD34.1 million on the CSP, representing 98.5%.. A total of USD13.3 million had been disbursed under PWP representing 96 % of the total component allocation of UDS 12.8 million. The initial float from the IDA into the MASAF Special account was US$1.5 million. However, due to the high rates of disbursement to the to sub-projects, this amount had to be adjusted to USD3.0 million in order to meet the demonstrated disbursement capacity of the Project. Disbursement of project funds to community sub-projects and PWP projects was done in tranches with the subsequent tranche coming after funds in the previous tranches have satisfactorily been accounted for. Funds for PWP were disbursed to assemblies while under the CSP, funds were channelled directly to the benefiting communities through elected PMCs. Initially, the tranches were - 52 - done as follows: 25 percent first tranche, 40 percent second tranche, 25 percent third and 10 percent fourth tranche with a possibility of a supplementary tranche where necessary. At the MU, financial management was done using a computerized system. A Justification Unit, supported by an Internal Audit system worked to ensure that accountability requirements were met. Due to delays experienced as a result of the justification process, the number of tranhes was reduced to three, and their volumes changed to 30, 40, and 20 percent as first, second and third tranches respectively. In general, the system has operated well and successfully. Accountability levels have been very high, especially on community sub-projects. However, some aspects that affected the arrangement which needed improvement. Firstly, the volumes of tranches, especially the first tranche was so small that it could only achieve a small proportion of the works. This was compounded by the fact that the time between submission of reports by PMCs and receipt of next tranche took on average, a period of two to three months during which projects would be at a standstill. In order to ensure sustained flow of incomes to projects with a view to speeding up implementation pace, the tranching arrangement was changed to be as follows: 30, 40 and 30 percent representing first, second and third tranches. Secondly, changes in the macro-economic environment such as the fluctuation of the Malawi Kwacha against the US dollar and high inflation rates, brought about considerable pressure on the sub-project budgets which were denominated in Kwacha. This led to some of the sub-projects' targets being reduced in order avoid incurring overruns on the sub-project budgets. To resolve this, the budget at approval was denominated in dollar terms so that they could be cushioned from the deleterious effects of these changes. Thirdly, the counterpart contribution from Government was coming in drips and would not match the speed of funding requirements in MASAF. Although the Government fully met its obligations in the end, there were occasions when the Project could not receive the full funding requirement from the World Bank due to inadequate matching funds from government. This affected implementation of the Project. Fourthly, PMC s took a long time to account for the funds once disbursed. This posed serious problems when the accounts for sub-projects that were completed needed to be closed. Determination of completion of sub-projects was therefore problematic and IEC through messages to PMC s, District Teams and communities on the need to promptly account for funds was deployed as a strategy to resolve this. Fifthly, although the level of community contribution was captured during sub-project appraisal, it was not reflected in the Project accounts during implementation. The actual contribution was only captured at the end of each sub- project. Total funding under the PMSF was US$lmillion. These resources were to be used for supporting capacity building in small policy studies, training and procurement of equipment and vehicless. However, disbursement was dome through MASAF and Treasury.Due to late start of the project, the project was extended for another year in order to finalise outstanding activities. - 53 - 8.2 PROCUREMENT Procurement under MASAF I was done using various methods and in conformity with the World Bank guidelines for procurement of goods, works and selection of consultants. Project Equipment such as computers and vehicles was done centrally at MU using international competitive bidding. In regard to sub-projects, procurement was done using a method called community participation. Procurement under PMSF was also done using MASAF procedures.. In general, procurement at both the MU and community level was satisfactory done. Procurement under community participation, despite some operational challenges such as PMCs being cheated by unscrupulous traders, the judgement by PMC s on what to buy and when, was highly satisfactory However, there was inadequate capacity at MU level to handle procurement issues. There was only one procurement officer who was also responsible for other technical functions in his office and this usually delayed submission of requests for the processing of procurement related documents such as requests for proposals and contracts. The result was delays in carrying out critical steps that were a requirement of procurement. 8.3 SUSTAINABILITY MASAF I facilities are likely to be sustained. Communities under the CSP are aware of their operation and maintenance (O&M) responsibilities as revealed by the second BA. Increased ownership created by the project processes and the substantial contributions to the projects by communities will guarantee sustainability of the assets created. In addition, sustained service delivery from the assets created will be assured by the participation of sector ministries who endorse the creation of the facilities at sub-project appraisal. The existence of committees such as the parents and teachers associations (PTAs) and water point committees (WPCs) responsible for O&M at the community level in the education and water sectors is also mechanism that will assure sustainability of the facilities created. Experience with such committees in the past has indicated that they are able to operate and maintain community level facilities such as school blocks staff houses, boreholes and wells. In addition, MASAF provides community based management training (CBM) of water facilities as part of the sub-project package. This is done in order to increase the capacity for long term O&M of water facilities. Under PWP, natural resources management committees (NRMC) exist at community level and are responsible for the O&M of assets such as community woodlots and dams. In regard to roads bridges under PWP, district assemblies have the responsibility for O& M, but only for those roads that are gazetted. In view of this the PWP component stopped the funding of projects that are not gazetted and fall within the purview of the district assemblies. This means that the sustainability of road and bridge projects that have been funded outside the responsibility of the communities is suspect. However, some chiefs such as Group Village Headperson (GVH) - 54 - Mang'omba on the Chigumula - Mang'omba Road project are able to mobilise communities to maintain roads. In this connection, more targeted IEC on sustainability issues would help re-enforce this spirit in communities. PMSF played a very important role of analysing the poverty situation in the country. However, the PMSF could have been sustained if there was a sense of ownership and full participation by all the TWC members. 9.OBANK AND BORROWER COUNTRY PERFORMANCE 9lRBank Performance The Bank's performance can be rated satisfactory. It has responded to specific issues at a fairly good pace. The Bank moved quickly to appraise the proposed MASAF Project in December 1995, and again in 1998 to appraise a second phase of the MASAF Project. following a request by the Government of Malawi. During implementation, Bank staff have been instrumental in providing technical advice in regard to project management and procurement issues through their supervisory visits. The Bank staff accorded MASAF a conducive environment that allowed management to apply Project procedures flexibly, in line with a process oriented project. Response to procurement related issues such as review of procurement activities and responding to requests for "No Objections" has also been quick, and facilitated speedy procurement of goods and services for the Project. The Bank's support to the Project has, therefore, contributed to the success that the MASAF Project has achieved. 9.2Borrower performance The Government of Malawi's performance in the implementation of the MASAF Project is rated as satisfactory, as evidenced by its full commitment to the Project. During Project preparation in 1995, it prepared the Project Concept Document (PCD) which was the basis for the World Bank's appraisal of the Project. It also participated in the subsequent appraisal process with the World Bank officials. In 1998, the Government again prepared a PCD for the second phase of the Project. In addition, the Government was able to provide counterpart contribution to the Project and allowed the MASAF MU operate without improper influence from those in power. The high profile that the MASAF Project assumed is also attributable to His Excellency, the State President Dr. Bakili Muluzi's support. The capacity and commitment of the district assemblies has facilitated smooth implementation of the projects at district level. However, in some district there were experiences of low commitment by Government staff which undernined the implementation speed of projects and accountability of Project funds. - 55 - The MU staff demonstrated throughout MASAF I their ability to work towards the achievement of set targets. In fact, the set targets were overachieved, which was indicative of the level of commitment of the MASAF staff. The performance based job evaluation system that was introduced in MASAF contributed to the high level of staff delivery of intended accountabilities, and in turn the Project Goal. 10.0 MAJOR LESSONS LEARNT The key lesson that was learnt throughout the implementation of the MASAF I Project was the need for flexibility in the application of procedures, and the adoption of continuous learning as an integral strategy of project management. In a heavily process and people oriented project, the rigid application of procedures would have certainly derailed the successful implementation of the Project. In this connection, challenges provided the learning points in the course of project implementation. The following is a summary of the main lessons learned from implementing the MASAF I Project: Need for supporting mechanisms for community-led development: in order to assist communities realise their development needs, efforts such as training and technical supervision are vital. Not only did they contribute to the successful sub-projects, but also contributed to quickening the project implementation pace of sub-projects. Need to balance the demand-driven principle with poverty reduction goal: the implementation experience of MASAF I demonstrated that communities that are less endowed in terms of natural resources and leadership skills could be left out of the benefits from the project if the demand driven principle is strictly applied. There is need to deliberately target areas that do not take up projects in order to achieve national equity in the distribution of Project benefits. Need to strike a balance between accountability requirements and realizing speed in the implementation pace: the justification process proved to be a potent mechanism for ensuring that Project funds were used for their intended purpose. However, it was also the cause of considerable delays in project implementation. The changes that were instituted to reduce the number of tranches and to increase volume of each tranche represented a mechanism that recognized the need for maintaining the high level of accountability, while at the same time allowing for quick implementation as a result of increased flow of funds to sub-projects. Need to match government's capacity to service project outputs. The experience with the health units that took time to be staffed brought to the fore the need to match Government's capacity to service project outputs. This is recognising that the construction of new facilities necessarily represent an additional demand on its recurrent budget. Need for robust procedures to ensure financial accountability: In a project where a bulk of funds is channelled directly to communities where there were no formal mechanisms for financial accountability required a robust system that would ensure that funds were satisfactorily accounted for. The justification system, along with other sub-project processes such as the launch and the internal and external audits have ensured that lapses are detected in time and resolved. A computerised financial tracking system has also been key in ensuring that sub-project specific - 56 - financial issues are resolved in time. Role of an effective IEC in supporting a demand driven approach: the IEC under the Project played a critical role in ensuring that communities countrywide knew of the funding opportunities in MASAF. It played a catalytic role in mobilising communities and other stakeholders for community development with MASAF assistance. It also assisted in clarifying project processes and procedures that contributed to the smooth implementation of the Project. Need for effective quality assurance mechanisms: where quality assurance mechanisms are inadequate as was the case in the early stages of the Project, facilities tend to be of sub-standard quality. Under the CSP, supervision capacity was assumed to exist in the communities and in sector ministries at design. Implementation experience revealed that that capacity was lacking or severely limited. The recruitment of 27 Technical Supervisors and deployment of consultants to supervise borehole drilling assisted in removing these gaps and led to significant improvements in the quality of the facilities constructed.. The need for an expanded menu of projects to be funded under safety net: Projects under the PWP consisted only of roads. Occasions when roads could not be done such as during the rainy season or when in the targeted area there are no roads to be done, implementation of the safety net operation could not be done. The PWP menu had to be diversified in order to overcome this challenge. Need to take into account other household or individual productive activities when designing a safety net operation: In the initial stages of the Project, it was found that beneficiaries were spending most of their time working on PWP projects and this meant a reduction in the amount of time they spent in their own fields. This led to low yields and a perpetuation of the beneficiaries' vulnerability. Norms and amount of time spent on PWP by beneficiaries was reduced in order to afford them time to work in their gardens. Application of national level data for Targeting purposes: while the vulnerability assessment mapping (VAM) methodology was a good guide for geographical targeting under PWP, the implementation experience revealed the need to support such data with district level knowledge of vulnerability to increase the effectiveness of targeting. Need for a special focus on peri-urban and urban areas: the low uptake of projects in the urban areas point to the need for a special focus on these areas. This especially considering that there are unique problems in those areas that render them unable to access MASAF assistance, although the level of services such as water are low. 11.0 FUTURE OPERATIONS There is need for improved communication and management between World Bank and NEC. Need for more TA on policy analysis. This is important in the sense that the data generated under PMSF need to feed into policy formulation. The overlap between MASAF I and MASAF II operations has meant that lessons learnt would not be separated by period of implementation. However, most of the - 57 - lessons that have arisen from implementing the MASAF I Project have been actively included in the Project processes under MASAF 1. Suggestions for the future operations have also inlcuded experience from PMSF activities. The suggestions for future operations are summarised below. 13.1 Future operations should focus on actively integrating environmental issues. The environmental impact assessment procedures that were introduced after the mid-term review were not fully utilised partly due to lack of capacity both in MASAF and district levels. Secondly, other vulnerable groups require programs that specifically target them. In this regard, under MASAF II, a sponsored sub-project (SSP) component that would target the vulnerable groups such as orphans and people with disabilities was introduced. It targets marginalized and vulnerable groups who were not adequately addressed by MASAF 1. 13.2 The issues of sustainability should be adequately addressed for some types of projects such as roads whose benefits are not localised. The experience under MASAF I showed that such projects became "no body's" baby after completion if mechanisms are not worked out before hand about how they will be operated and maintained. In addition, there should be support to a process of instituting local level O&M of facilities through capacity building efforts such as training. The IEC campaign should vigorously project community level O&M as a key issue in project implementation. 13.3 Monitoring and evaluation mechanisms should be enhanced. Under MASAF I this aspect, has been found to be weak. The computerized project tracking system should represent a major aspect of developing a comprehensive monitoring and evaluation system. Such a monitoring system should be able to aggregate project related data that shows the impacts of the Project activities. 13.4 Adequate capacity for the Project Procurement function should be put in place. This would ensure that delays arising from inadequate staffing and job overloading of staff responsible for procurement are minimised, and speed in procurement of goods and services achieved. 13.5 TWC composition and objectives (PMSF) need to be reviewed in order to reflect interests of all the members. 13.6 Finally, community contribution levels need to captured and reflected in project accounts during project implementation. A mechanism that captures community contribution either on a monthly or quarterly basis should be developed. - 58 - MAP SECTION 32' t 34' 36- To D., Es a~7.7N., K ToDar E s CX 9zuzu ZONE OFFICES (9) -tPa -i o sekenge - ZONE BOUNDARIES K-NGA R PRINCIPAL ROADS MALAWI Nkhow 5 RIVERS SOCIAL ACTION Thilum TOWNS (D DISTRICT CAPITALS FUND (MASAF) Livingstonia * NATIONAL CAP TAL COMMUNITY SUBPROJECTS: --K--o- DISTRICT BOUNDARIES ZONES, DISTRICTS AND - - INTERNATIONAL BOUNDARIES PLACEMENT OF ZONE OFFICES V E sofio'M,b i Lake TANZANIA ,,~~~~~~E MZUf oL Ag. e_ /) g ~~ 2 .~~~~Bay - MZIMB, NsKHATA / J_ - 12° 6 ' Chizum/10 12' ~~~~~~~1 ~ ~ ~ ~ w AX X 'M w,w: \M twi ˘ J MOXalwi o20 40 60 80 100 V 5 ( Chuluo 0-tf W q i 2 KILOMETERS Nkhotokoto ZAMBIA c ,J,S I 3- ( MOZAMBIQUE X ~~~ ~ ~ ~ ~ ~~To Losolsol L Mcini -1 , 2 ' 1 > { XY k \s v 14'U-1k _ - - I I _ X / \ S 9 X c a h k e t1 DEM0OCRATIC B WAKNDj-~ XOLIA 6OAB5tE~ FIEPUBEIC OF / N,~~~~~~~~~~~LIONWE5 . - CON GO TANZANA INDIAN 20 A r S k , i, 2, OCrN t toTee _ / ^ BE 'n ?' PHAOMB ANGOLA >' MX\s>(|˘ ; X - ; jJ 9 Chikwawac n ULA NJEA DEMOCRATIC ~ ~ ~~~~~~~~~~~~Chkosa C fZO ILNDE 6 BU ZAMBIA QUE? oKone Thypo BTmss dd - ,./t_< ! 93 < RE° f X \ i4. § MaP DesigMUBLIC of The Wold Bank 'r r \ 21MBAEWE / g t ) P/ eo, v 3t~~~~~~~~~~~~~'I this maJp do nOt imply, on the porf of ffi I D } l j l ~~~~~~~~~~~~~~~~< \ Mvm ~~~~~~~The World Bonk Group, any 1odgment z r~ BTWN WAA A on the legol stalus of any territory, or i \~ * \ sassnfe any endorsement or acceptance of < / SOUTH '7 ,5 i rJ ) ( such boundaries C G AFRICAA JUNE 2002 IBRD 31951 32 ,3 36' 7o Dor Es Solorr M k UU ZONE OFFICES f12) rlipo) 0etPa,_8 ogelenge ZONE BOUNDARIES - i h RPRINCIPAL ROADS MALAWI NXkhwwo RIVERS SECOND SOCIAL Chilumba ° TOWNS ® DISTRICT CAPITALS ACTION FUND (MASAF IIJ) vnsoi @ DSRCC>TL ACTION FUND (MASAF 11) RUMPHI I vngstontiO NATIONAL CAPITAL COMMUNITY SUBPROJECTS: KotoH ' Mu ----- DISTRICT BOUNDARIES ZONES, DISTRICTS AND h --- INTERNATIONAL BOUNDARIES PLACEMENT OF ZONE OFFICES Eu E0^sLlmbo /F>E%enile& Lake TANZANIA X zr ? (\ {H -)-Xt ~~~St. A fguWneT 12' ) 12° X _t~~~~~- ); 7 g~~~ (L?koma Ma wi 0 20 6o 60 80 100 MCrtINJI ' DO A Dowo A~~A L/LO M ( Chulunkey Sff,A, iKILOMETERS ') \ ! s>42> /glN khotc X~~~~~~~ khotao ZAMBIA J) - t . ( D 8/n Noc MOZAMBIQUE --~ 32'+ - > Malocnbo Chlulor RWANDAkS - KENY IMI / < |)ACI-INGAj-- DEMOCRATIC -~~~~~~~~~~~t MOZ MBIQL/E~~~~4 le a REPUELIC OF z MZMIlaE <_ i ok ANZANIA t7 INDIAN I r S t rg X t; OCEAN 0 / + t { tht? X~~~~~~~~~~ 10 PI-ALOM8E< 1 A?O<' ;--/>< Bay~/ 14 6 / 11' - iE, (' g ZAMBIA to;, $3>MOZAMDIQUE ,, T Mc ouna reC, ,olors, denominations a r \ ~~~~ZIMBABWE r,{ ,i Chr7 X ramoa d oronyather informao;;on shossnonl zNr- goTswANA z< \ MADAG CAR on the terrtry, or ZI' ROTTH WAN _~~'( * s> <