Malawi - Agricultural Services Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 20594 IMPLEMENTATION COMPLETION REPORT (Credit 25 14-MAI) ON AN IDA CREDIT IN THE AMOUNT OF US$45.8 MILLION TO THE REPUBLIC OF MALAWI FOR AN AGRICULTURAL SERVICES PROJECT JUNE 13, 2000 Rural Development Operations Eastern & Southern Africa This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQIIVALENTS AtAppraisal USSI = MWK400 Al Cottipiction USS I - MWK46 43 WEIGHTS AND MEASURES Metlic System FISCAL YEAR OF BORROWER January 1-Decemnber 31 ABBREVIATIONS AND ACRONYMS AAU Agri-business Advisor Unit ACB Agnctltural communications Branch ACS Agricultural Communications Section ADS African Development bank ADD Agrncatural Development Drvision ADF Agrncoltural Developmenl Fund ADMARC Agricultural Development & Marketing Corp APATU Agricultural Porlcy Analysis & Training Unit APIP Agricultural Producttvity Investnsent Program APRU Agricultural Pohcy Reseatrch Unit ASC Agricultural Sciences Comnitnee ASSC Agricultural Sector Steering Canmtittee ASP Agricultural Services Project BES Block Extension System BIAS Beneficiary Impact Assessment Smvey CMEU Central Monitoring & Evaluation Unit CRP Contract Research Progrant DAETS Department of Agricultural Extension& Tech Savices DAHi Department of Animal Healli & Industry DARTS Departtnent of Agricullural Research & Tedincal Services DFID Department for Intem;atinal Devdopment DO Development Officer DOI Departmets of Irrigauon DREA Depattment of Research & Environmental Affaira EPA Extension Planning Atea EU European Union FA Field Assistant FAO Food & Agriculture Organtzation FHA Farm Home Assistant F&N Food & Nutrition GOM Govetment of Malawi ICB Intemational Competitive Bidding ICR Implementation Completion Report IDA Intemational Development Association IFAD International Fund for Agricultural Dev IFPRI Intematidnal Food Policy Research Institute LCB Local Competitive Bidding MCOS Motorcycle Owneship Scheme MPTF Maize Productaity Task Force MIPA Malawi Investment Promotion Agency MOAI Ministry of Agriculture & Irrigation MoF Ministry of Finance MRFC Matawi Rural Fhnace Company MTEF Mid- Term Expenditure Framework MTR Mid-Term Review NGO Non-Goverment Organization NRDP National Rural Development Pngranm NRCM National Research Council of Malawi NSCM National Seed Company of Malawi PCR Project Completion Report PD Planaing Division PDO Project Development Objective PEM Participatory Extemion Methods PPF Project Preparation Facility RELO Research Extenston Liaison Officer RDP Rural Development Project SADEC Southem Afncan Devlopnient Comnuunitv SFFRFM SmnalilholderFarmar Fertilizer Revolving Fund of Malawt SFSP Snallholder Food Security Project SMS Subject Matter Specialist SPS Starter Pack Scheme TA Technical Assistance TO Technical Ofricer T&T Transport & Travel T&V Training & Vtsit USAID United States Agency for Intemational Development VA Veterinary Assistant VLPA Village Levd Participatory Approach WB World Bask Vice President Callisto Madavo Country Manager/Director Barbara Kafka Sector Manager/Director: Sushma Ganguly Task Team LeaderlTask Manager Bernard W Van de Poll  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT MALAWI AGRICULTURAL SERVICES PROJECT (Credit 2514-MAI) TABLE OF CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 3 5. Major Factors Affecting Implementation and Outcome 12 6. Sustainability 13 7. Bank and Borrower Performance 14 8. Lessons Learned 16 9. Partner Comments 17 10. Additional Information 17 1 1. ANNEXES Annex 1. Key Performance Indicators/Log Frame Matrix 18 Annex 2. Project Costs and Financing 25 Annex 3. Economic Costs and Benefits 29 Annex 4. Bank Inputs 32 Annex 5. Ratings for Achievement of Objectives/Outputs of Components Annex 6. Ratings of Bank and Borrower Perfo-mance 33 Annex 7. List of Supporting Documents: Annex 7A - Executive Summary of Project Completion Report for ASP, prepared by MoAI (December, 1999); Annex 7B - Aide Memoire of ICR ASP Mission, prepared by FAO (May, 2000) This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization.  I . I I - . . . 1 1 . 1 1 I I  1. Project Data Name: AGRICULTURAL SERVICES PROJECT Credit 2514-MAI Nusmber: Country/Department: MALAWI Region: Eastern & Southern Africa, Sector/subsector: AE - Agricultural Extension; AG - Agency Reform; AM - Agro-Industry & Marketing; AR - Research KEY DATES Original Revised/Actual PCD: 02/19/88 Effective: 01/21/94 01/21/94 Appraisal: 06/25/92 MTR: N/A 05/29/98 Approval: 06/15/93 Closing: 09/30/99 12/31/99 Borrower/lImplementingAgency: Government of Malawi/Ministry of Agriculture and Irrigation Other Partners: Afncan Development Bank (co-financing of selected sub-components Other Partners:eon a parallel basis) STAFF Current [At Appraisal Vice President: Callisto E. Madavo E Jaycox Country Manager: Barbara Kafka _ Sector Manager: Susha Ganguly J . Shivakumar Team Leader at ICR: Bemard Van De Poll l Richard Anson ICR Primary Author: Michael Wales (FAO/CP) _ 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=Hfigh, SU=Substantial, M=Modest, N=Negligible) Outcome: U Sustainability: UN Institutional Development Impact: N Bank Performance: S Borrower Performance: U QAG (if available) ICR Quality at Entry: S S Project at Risk atAny ime: Yes Yes  3 Assessment of Development Objective and Design, and of Quality at Entry Original Objective: 3.1 The Project was conceived as the first phase of a longer-tern exercise that aimed to reform and integrate research, extension and input supply services. The Project was to (a) support institutional and management reforns in research, extension and input supply, with the aim of providing cost-effective and sustainable technologies for different types of smallholders and agro-ecological zones; and (b) enhance smallholder incomes and food security by increasing the productivity of smallholder agriculture, especially that of female and resource-poor farmers. Although there was no Country Assistance Strategy at the time of project design, the project's objectives were appropriate, and congruent with broad Government strategy and development priorities, and it tried to build upon the lessons of earlier projects, such as the National Rural Development Program (NRDP), the Agricultural Extension and Planning Support Project (AEPSP), and the National Agricultural Research Project (NARP). Original Components: 3.2 The IDA contribution for the project components, to be implemented over six years, were: (a) Agricultural Research (US$15.8 million): (b) Agricultural Extension (US$17.2 million), (c) Input Supply (US$4 million), and: (d) Institutional Strengthening (US$8.8 million), including project preparation: 3.3 The project was a successor to the 30 Rural Development Projects (RDPs) which comprised NRDP. It was intended to simplify the provision of support services by consolidating them into national, centrally managed initiatives, with a focus on resource poor farmers. The Research component was designed to encourage scientists to focus on farmers' problems, and to make better use of limited resources. However, the associated reforms failed to allow for intrinsic resistance to change in management procedures within the Ministry of Agriculture, now the Ministry of Agriculture & Irrigation (MOAI). The Extension component was intended to improve the efficiency and effectiveness of the existing extension service, rather than introduce radically new systems. In contrast to the Research component however, the activities prescribed were rather general and relied very heavily for successful implementation on competent management that understood and was committed to change. The Input Supply component was essentially to support liberalization of the trade in inputs, and in particular to encourage the private wholesalers and retailers to enter the sector in an attempt to introduce a bit of competition. The last component -Institutional Strengthening- was an attempt to introduce much needed changes in the operational procedures of the MOAI. Its principal weakness was a very heavy dependence on internationally recruited Technical Assistance, to develop and pilot new procedures. 2  QOiality at Entry: 3.4 A QAG assessment conducted in mid 1998, concluded that the project design was sound, ready for implementation at the time of approval, and that the Government was committed to the project. However, some concems were voiced at appraisal, that the project was too complex for a weak ministry to implement, that Government had played little part in the preparation of the project and showed few signs of "ownership". The ministry at the time was characterized as lacking goals, leadership, commitment, financial and management accountability. The holistic design, was intrinsically more complex than the NRDP, and much more demanding of management. Concern was also expressed at appraisal that the centralization of control over funds might be inimical to responsiveness by the Agricultural Development Divisions (ADD). The implementation experiences suggest that these concerns were valid, and the ICR mission re-assessed the quality at entry as marginally satisfactory. In particular too much emphasis was placed on centralization of management systems. 3.5 At appraisal a number of changes were made to simplify the project. Livestock extension, land resource management and conservation, certain aspects of small-scale irrigation including institutional strengthening, and part of fertilizer supply, were transferred to an African Development Bank (ADB) project; whilst a parallel-financed IFAD project - the Smallholder Food Security Project (SFSP) included group development, burley tobacco and hybrid maize development, soil conservation and fertility, the provision of seeds and fertilizer, small-scale irrigation infrastructure and training. Support for smallholder credit was transferred to the World Bank-funded Rural Financial Services Project, and the IFAD-funded Mudzi Financial Services Project. The result was an overall design in which sequencing and interdependencies were very complex. 4. Achievement of Objective and Outputs Outcome/achievement of objective: 4.1 The project is rated as unsatisfactory. However, its perfornance was influenced by extraneous factors such. as HIV/AIDS and its political and policy environment. Although the project has contributed to a number of policy reforrns, it has not achieved its principal objective of delivering agricultural services to the smallholder sector, in particular the resource-poor. The absence of specific performance indicators over the life of the project, makes it difficult to assess to what extent smallholder productivity, incomes and food security have been enhanced. The Starter Pack Scheme (SPS), which was supported by the project in the 1998-99 and 1999/2000 seasons, was effective in targeting poverty and household food security, and was a contributory factor in the production of bumper harvests in those seasons, but its sustainability is highly questionable. Overall, maize yields have increased, although it is not possible to distinguish the effect of the relatively good rains received in the last two years, and high prices, from project impact. There is evidence of diversification into sweet potato, Irish potato, soybean, pigeon pea, cassava, cabbages and other vegetables. 3  Outpults hy components: 4.2 Agricultural Research: This component is rated as marginally satisfactory. Its objective was to develop a cost-effective and sustainable research system that would generate technologies responsive to farmers' needs. Support was for (a) the formulation and implementation of research action plans; (b) strengthening research- extension-farmer linkages; (c) diversifying participation in publicly-funded research through a contract research program (CRP), and commercialization of some research activities; (d) improving the definition of research priorities and the allocation of funding by strengthening the Department of Agricultural Research (DAR, now DARTS) and activating the Agricultural Sciences Committee (ASC); and (e) improving DAR's management systems and the motivation and accountability of researchers. The component has partially achieved its objectives through the CRP, although its overall impact was limited by difficulties in channelling funds 'to scientists under the regular research program, and there was virtually no involvement of farmers in developing research agendas. Some 30 maize varieties, and around 30 varieties of other crops including sorghum, groundnuts, legumes, cassava and sweet potato were released during the project period. However, it is not clear whether the emphasis on maize breeding, was a sensible priority, given the many other constraints to production by small farmers; and most of the varietal development work was undertaken by private companies or by International Agricultural Research Centres (IARCs). 4.3 Research Action Plan: Research Action Plans were prepared at the start of the project, but the annual planning procedures of DARTS have not changed. The most successful activity supported was the Maize Productivity Task Force (MPTF), jointly funded with the Rockefeller Foundation and the EU, which carried out fertilizer trials nationally. This resulted for the first time in a database of region- specific fertilizer recommendations for farmers. The strong GOM ownership of this initiative, reflected in the fact that it received nearly 50 % of total GOM project counterpart funding of the project, was a key factor in this success, as was the commitment and hard work of the Technical Assistance expert retained to assist the MPTF. 4.4 Research-Extension-Farmer Linkages: Despite the formulation of a set of guidelines, research-extension-farmer linkages remain poorly developed. Although Research Extension Liaison Officers (RELOs) were appointed at research stations, no provision was made by DARTS to cover their operational costs. 4.5 Diversification of Research Bodies: Although a successful CRP procedure for channelling resources to scientists was developed, questions remain about the relevance of some of the research topics supported, and its sustainability. During the project only 33 projects, in one batch in 1995, were approved for funding. Funds went almost exclusively to DARTS scientists, and only one private sector researcher and not a single NGO received grants. Although there was good monitoring of the projects, no research results have yet been translated into extension messages, and not all projects received their budgeted allocation of funds. It was envisaged that the CRP mechanism would be adopted to fund all agricultural research. This has not happened. 4  4.6 Research Priorities, Funding and ASC: The re-establishment of the ASC was temporarily achieved by the secondment of two DARTS officers to the National Research Council of Malawi (NRCM). During the project the ASC, developed and managed the CRP, prepared an operational manual, and a paper on options for a Sustainable Funding Mechanism for research. To date none of the options proposed has proved attractive to donors. After failing to meet several deadlines, an Agriculture and Natural Resources Research Master Plan was finally completed, in September 1999. Although it will undoubtedly serve as a useful reference document, the Master Plan is unlikely to be used as a basis for planning and managing research because it produced an extensive list of activities rather than a stratified and prioritized list, and makes unrealistic assumptions about the availability of resources. ' 4.7 DAR Management and Staff Motivation: The overall management of DARTS has not improved and the flow of funds to scientists remains a major problem. The project undoubtedly augmented the DARTS operating resources, and scientists have benefited from training, even though only two of the 39 slots for short-term overseas courses were utilized. The poor flow of funds, other than CRP resources, the failure to procure most of the laboratory equipment funded under the project until almost the closing date, and a lack of promotion opportunities, has left the Department with low staff morale. Although annual joint review meetings, involving research and extension staff, to select research projects have taken place, the constraints of time and poor attendance have meant that some of the priorities identified by extension have not been reflected in the research program. The weaknesses of DAR/DARTS management were recognized at appraisal, and it was envisaged that a long-term TA Research Management Advisor would be recruited to develop better systems. The failure to hire such expertise partially explains the failure of DARTS to improve its management. Agricultural Extension: 4.8 This component of the project is rated as unsatisfactory. It was designed to continue support for MOA's nation-wide Block Extension System (BES), a modified form of T&V extension. The project supported measures (a) to increase the efficiency of extension service delivery through restructuring and streamlining all operations and staffing levels, functions and capacities; (b) to expand the extension coverage of smallholders, particularly resource-poor and female farmers through separating credit administration from extension work, promoting more relevant and affordable technologies/practices, strengthening research-extension-farmer linkages, integrating technical messages, collaborating with NGOs involved in extension, upgrading all farm home assistants to Field Assistants (FAs) and training all FAs in gender-related issues, particularly nutrition and the labour constraints faced by many female-headed households; and (c) to increase the effectiveness of service delivery through improving field staff training, using more interactive teaching methods, and encouraging better quality and consistency of field supervision by providing The draft Agricultural Research Strategy A.D.2000-2010, prepared by DARTS, dated February 2000, does not refer to ANRRMP or CRP. 5  adequate operational funding, more closely monitoring supervision and rewarding better performance. 4.9 Management and Operation: There is little indication of any improvements in efficiency or effectiveness of extension delivery. Whilst the de- linking of extension from credit has been achieved, FAs are still engaged in the registration and training of groups for the purpose of credit delivery through the Malawi Rural Finance Company (MRFC). Initially this tended to bias extenlon delivery towards farmers in credit groups. Latterly this tendency diminished and has been replaced by a bias towards villages where bi -lateral donor-funded extension initiatives have started. The BES, has virtually collapsed as a result of inadequate staff numbers and operating funds at the field level. The BES suffered from low farmer attendance due to an absence of new and affordable messages, and inadequate supervision. Although most Subject Matter Specialist (SMS) posts were filled, many of the staff were not suitably qualified. This, together with a chronic shortage of operating funds at ADDs, Rural Development Project (RDP) and Extension Planning Areas (EPA), resulted in poor quality of supervision and technical backstopping, with little or no support from Headquarters. Extension planning and management at the ADDs has not improved. Despite the introduction of a log frame approach, the use of workplans and the prioritization of work, extension work plans and budgets continue to be far in excess of resources available. 4.10 Extension Coverage and Content of Messages: The ratio of Front Line Staff to farm households has deteriorated from 1:980 shortly after the start of the project in 1995, to 1:1700 at the end of the project in 1999. This is the result of a failure to recruit new staff following formal and informal recruitment freezes, retirements, an increasing rate of "deaths in service" due for a large part to HIV/AIDS2, , and the fact that the National Rural College no longer offers courses to train FAs. Veterinary and animal husbandry field staff, whose operational costs were to have been funded by ADB, found themselves with extremely limited resources as a result of slow disbursement from the ADB credit. On-Farm Demonstrations (OFDs) became a priority activity and expanded enormously in number from 15,000 to almost 160,000, but this, together with an increase in the number of farmer production clubs to service, has over-stretched the FAs, and the quality of supervision has deteriorated. There was, however, considerable success in propagating low-cost technologies, soil conservation and fertility management, agro- forestry and on-farm root crop multiplication, through SFSP and programs funded by EU and USAID. The transition to a unified extension system, in which the FA is responsible for all farner contacts, including livestock production activities, appears to have resulted in even less attention to livestock extension than before. There was a large expansion in the number of farmer's clubs based on specific crops, especially burley tobacco, or livestock activities, as well as women's groups, and others organized through churches or mosques. The women's program was, however, focused almost exclusively on food and nutrition. The support provided by the 2 For example, it is reported by the Human Resources Department that there were 60 deaths-in-service dtiring 1999 in Machinga ADD, and 43 in Lilongvve ADD. 6  project has allowed the extension service to serve as the vehicle for various extension-type initiatives such as the promising Village Level Participatory Approach (VLPA) of the World Bank, PROSCARP (EU), MAFEP (USAID), Agricultural Productivity Investment Program (APIP), Farmer Field Schools, as well as the activities of a large number of NGOs, all of which are based on a large degree of farmer participation and responsiveness to farmer priorities. The ultimate task in this respect was the distribution of Starter Packs to some 2.8 million farm households in 1998-99 and 1999-2000, in which FAs played a crucial role in registering recipients. 4.11 Training: No training targets were achieved. Funds for training in innovative extension approaches, such as Participatory Extension Methods (PEM), were exhausted at senior levels and few if any FAs benefited. A total of 17 staff undertook Ms. courses overseas and six attended short courses in media and extension management. A limited number of FHAs received training to become FAs, and courses were held to upgrade FAs to the supervisory Technical Officer (TO) level, and to provide them with the necessary skills in animal production. The Food and Nutrition (F&N) subcomponent was delayed by the late arrival of the TA expert. A Community Nutrition and Training Manual was finally prepared, has been printed but not yet distributed and no training associated with it has been conducted. The program at the field level has been handicapped by the fact that there is often only one staff member to undertake both F&N and women's programs. The failure of Bunda College to finalize the socio-economic baseline survey means that it is almost impossible to make any quantifiable assessment of the benefits of much of the project. 4.12 Improved Access to Transport: The motorcycle ownership scheme was successfully implemented with supervisors receiving motorcycles. While this was considered innovative and partially successful, the positive impact that this should have had on supervision was negated by the fact that there were never sufficient funds to settle transport and travel (T&T) claims. 4.13 Civil Works: A program of civil works (total cost US$5.9 million) aimed primarily at the extension service, was substantially not implemented. The original proposal for houses for FAs, and four EPA centers, was modified to accommodate rehabilitation of existing FA houses and construction of two RDP centers. However these works were never undertaken. This was because the Intemational Competitive Bidding documentation was submitted by GOM too near the end of the project to allow for completion of works within the project life. 4.14 Extension Aids Branch: The Agricultural Communications Branch (ACB) -formerly the Extension Aids Branch, was to have been strengthened in support of the overall extension program, in particular, with equipment to modernize the Visual Aids Section for the use of radio to reach resource poor farmers in listening groups. ACB received vehicles and domestic training and, towards the end of the project, equipment. Two staff had long-term training on video production. Radio program and visual aids produced by ACB have had a significant impact in the field, even though its operations were constrained by lack of resources. 7  4.15 Input Supply: This component of the project is rated as satisfactory. Its objective was to support the increased availability of improved seeds and fertilizer to smallholders and to expand use of these inputs by smallholders. The subcomponents were (a) support for the formulation and implementation of policy reforms to further liberalize importation, pricing and distribution of inputs; (b) improved donor co-ordination; and (c) provision of foreign exchange to help finance incremental fertilizer imports on a commercial basis. 4.16 Policy Reforms: The reform with most far-reaching implications was the amendment to the Special Crops Act, which allowed smallholders to begin growing burley tobacco. This, together with support targeted at burley groups by SFSP, resulted in a major expansion in the number of smallholders growing burley from 18,000 to 50,000, and substantially enhanced incomes for these farmers. 4.17 The deregulation of fertilizer imports has been partly achieved, and at one stage there were nine local and intemational companies active. Subsidies, which started to be reduced from 1991/92, were completely removed in 1993/94. Towards the end of the project, however only two private fertilizer companies were operating, and the GOM has again been playing a major role in fertilizer importation thereby reversing the liberalization trend of the fertilizer policy. There has been no success in reducing the cost of fertilizer imports to farmers for a number of technical and policy related reasons. The small size of many fertilizer orders, long haulage distances, poor roads and high freight rates all contribute to high prices. Equally important however is uncertainty about the GOM maintaining liberalization policies. This discourages private suppliers from making the substantial long-term investment that could reduce prices. Problems of late delivery also still exist. The Smallholder Farmer Fertilizer Revolving Fund for Malawi (SFFRFM) operated commercially, selling to farmers and to local companies until the start of the SPS. 4.18 The objective of liberalizing production and marketing of hybrid seed was achieved, and all subsidies on improved seed were removed in 1994/95. Two commercial companies are currently producing or importing almost all hybrid maize seed used. Towards the end of the project there was a substantial increase in the use of hybrid maize which, according to MOAI estimates, now accounts for as much as 40% of the planted area (1.5 million ha in 1999/2000). However, there are indications that substantial numbers of farmers are using recycled seed. The expansion in the use of hybrids is also partially attributed to the free and low cost SPS (around 280,000 ha for about 2.8 million households), and APIP (250,000 ha) programs. The relatively low intervention prices for maize still discourage the use of high cost inputs such as hybrid seed. 4.19 There remains however a substantial agenda of second generation reforms to be undertaken, such as the privatization of the parastatal ADMARC, which controls 50% or more of the fertilizer market and, in remote areas, is the only supplier. The role and sustainability of the maize dual band intervention price support system, introduced after the 1994 drought, in order to protect consumers, also needs to be 8  reviewed. Overall whilst the agreed policy reformns have been largely implemented, this component has not fully achieved its objectives because the overall impact on the availability of inputs to small farmers, and competitiveness in supply, has been very modest. 4.20 Fertilizer Imports: Under the project IDA financed the importation of 12,500 t of fertilizer, through loans to SEDOM and INDEFUND through the Reserve Bank of Malawi for private sector traders. ADB funds were used to procure 19,877 t of fertilizer, without ADB approval, in 1997/98 for SPS, which led to temporary the suspension of the ADB loan. 4.21 Institutional Strengthening: Partly influenced by the poor performance of the Civil Service Reforn, this component of the project turned out to be unsatisfactory. Its objective was to strengthen the capacity of the Ministry to deliver agricultural services in an efficient, effective and sustainable manner. The subcomponents were (a) planning and financial management in MOAI/ADDs; (b) transport management; (c) procurement; (d) agricultural policy analysis training; (e) small-scale irrigation (funded by ADB and IFAD); and (f) sectoral and project monitoring and evaluation. The implementation of all sub-components suffered as a result of the MOAI failing to appoint adequate counterparts to technical advisers. 4.22 Planning and Financial Management: Although some progress has been made in financial management, including the installation of a computerized system (Finays), and related staff training, overall financial recording was very poor. In the main, this is attributed to a lack of interest by MOAI managers, in adopting better financial management systems. Furthermore work plans, budgets, and financial reports were generally produced late. But by the MTR of the project substantial progress was made in prioritizing expenditures, particularly with the Bank suspending disbursements till a work plan satisfactory to IDA was formulated. Arrangements for improved fund flow were also put in place. As a consequence some progress in fund flow to the field for the implementation of priority programs followed, but with a change in the financial management staff at MOAI the system once again collapsed. The most critical constraint to the achievement of the project's objectives was the failure of the MOAI to channel resources to extension field level staff and research scientists on a secure and regular basis. This problem has been compounded by the lack of realistic work plans and budgets at the ADD and RDP levels, and by DARTS. The logical framework approach has been adopted at all levels as part of the Medium Term Expenditure Framework (MTEF) exercise started by the Ministry of Finance with Bank assistance. However, the process of prioritizing activities remained a problem, with budget proposals grossly in excess of available resources. 4.23 Transport Management System: The provision of increased mobility to staff at all levels in support of extension delivery, was a crucial part of the project's design. Although over 20 (4x4) vehicles, 440 motorcycles and 865 bicycles were procured, a transport management system was not put in place and no effective training was conducted. 9  4.24 Procurement3: Although in the SAR, procurement was identified as "the most intractable constraint on timely project execution", the project has failed to improve the situation, despite the retention of an internationally recruited procurement specialist. On the basis of the Mid-Term Review (MTR) assessment, field visits and the experience of ADDs, it appears that procurement was invariably slow and erratic4, often wasteful through the purchase of inappropriate goods, (e.g. average price of cameras, and photocopiers purchased for the extension service as reported by MOAI were US $ 7600 and US$ 14,900 respectively), and that items once procured were often misallocated. Responsibility for the procurement study called for in the SAR was shifted between several Ministries and was eventually completed at the end of 1999. Little or no training was carried out, leaving the Procurement Section at the end of the project without operational capacity. 4.25 Agricultural Policy Analysis Training: The project supported the Agricultural Policy Analysis Training Unit (APATU) at Bunda College of Agriculture, in an attempt to strengthen the policy analysis capacity of the MOAI. Support, which was implemented, included procurement of equipment, and infrastructure development of a 30-bed hostel. The unit successfully conducted around five, 2-3 week courses per year5 against a target of nine, and has served a range of agencies within government and NGOs. The main problem was slow disbursement of funds, which has limited the number of short courses that could be conducted, and inadequate support from the international agency contracted to provide technical assistance support. The main failure has been the inability to attract candidates for the Ms. degree program in agricultural economics, extension and rural sociology, with only one person of the targeted 30 receiving a degree. Six students were enrolled in 1999 but, with the closure of the project, they will not be able to complete the degree course. 4.26 Training: A total of 5 Ph D, 33 M Sc, and 30 diploma and short courses were funded by the project, with the majority of the beneficiaries being from Headquarters. This was far below the targeted training program, and is attributed largely to the decision to suspend the training program pending a training needs assessment, which was not completed until the end of the project6 4.27 Technical Assistance: The project was to have provided 410 person months of TA in 18 different areas, of which 144 months would be long-term and 298 months international, including 100 months for the procurement study. Only half of the technical advisers were hired, with critical gaps in research management, It proved difficult to obtain information on procurement since the TA officer had moved to a different project and was not in the country. 4Laboratory equipment for research stations was not delivered until the final year. Vehicles were not delivered for two years. The bulk of the civil works program was not implemeented. 5This was reduced to 2-3 per year in 1998 and just one in 1999 due to lack of resources. Training Needs Assessnient Report, by Deloitte and Touche, dated 1999. 10  extension management training, agricultural policy, monitoring & evaluation, livestock extension and financial systems. The MOAI reports that the quality and output of much of the international TA that was provided was disappointing. Particular problems include difficulties in obtaining progress and final reports, unduly strict adherence to terms of reference, recruitment of inappropriately qualified personnel, failure to appoint counterparts and, in the case of the TA support to Bunda College, a strong sense that they have not received value for money. 4.28 Monitoring and Evaluation: A major objective of the institutional strengthening component was to improve the M&E function at all levels in the Ministry. Initially there was inadequate staffing, a problem not resolved until the end of the project through recruitment and training. Although the annual work programming process improved with the introduction of the log frame, implementation has remained poor. Performance indicators were re-formulated by the planning department, although not until the end of 1998, and they have not yet been endorsed by the ADDs. Work on improved agricultural data collection, based on the Area Sampling Frame (ASF), and did not proceed beyond the pilot stage. Guidelines for ASP reporting were never issued and the submission of M&E reports was chronically behind schedule almost throughout the project. Overall, there is still no effective system in place to monitor and evaluate the progress of agricultural development in Malawi, and no capacity to integrate M&E management information into the strategic and annual work planning processes. 4.29 MIPA: Support for the Agribusiness Advisory Unit (AAU) in the Malawi Investment Promotion Agency (MIPA) to strengthen the capacity to promote local and foreign investment, disseminate agribusiness technical and market information, and advise entrepreneurs on financial opportunities and accessing advisory services, started in 1996. AAU staff successfully completed a range of induction and specialist training courses, although on-the-job training by the TA did not take place. The AAU achieved its objectives, in so far as establishing the Domestic Investment Promotion Department. However, it was hampered by slow flow of funds from MOAL and failure to procure the necessary equipment for its Business Information Center. 4.30 Project Management: Overall project management, which was the responsibility of MOAI, with the Central Monitoring and Evaluation Unit (CMEU) responsible for day-to-day activities, remained weak throughout the project. The principal mechanism for implementing co-ordination was the Agricultural Sector Steering Committee (ASSC) which was scheduled to meet quarterly, but in fact only met twice during the project life. An ASP Facilitator (on contract) was overwhelmed with work, remained in post for only one year and was never replaced. Despite two start-up workshops, detailed knowledge of the project's objectives was not widespread. The cost tables in the SAR (Annex 6) gave only very broad indicative cost estimates for the individual components. The PER and the Implementation Manual, which provided greater detail on investments, were hardly used by the MOAI. As a result many subcomponent managers were not clear as to what assets/expenditure they were supposed to be responsible for. Management of individual components by different departments was generally inadequate, although 11  ADDs generally fared well under difficult circumstances. Weak management and limited commitment to the objectives of the ASP led to the project funds being used in an ad hoc way to fund a range of initiatives Net Present Value/Economic rate ofreturn: 4.31 A formal ERR was not calculated for the project. However, potential benefits were assessed (Annex 16) using farm budgets and linear program analyses. The models assumed substantial increases in the use of fertilizer and hybrid maize, as well as in burley tobacco production by smallholders. By 2002/3, it was estimated that 84% of smallholders would use flint hybrid maize in pure stands, that burley production by smallholders would increase to 35,000 Kg, and that the overall increase in the real value of smallholder production would be 35%. On the basis of these assumptions an ERR of 23.8% over 20 years was predicted. In terms of the main assumptions adopted in the model, it is unlikely that the project achieved such a high ERR. It is estimated that in 1999-2000, whilst burley production has increased substantially, around only 40% of the maize area is now planted with hybrid seeds, and this mainly the result of the SPS and APIP, rather than the ASP. Institutional development impact: 4.32 Little sustainable institutional development within MOAI was achieved. A contract research procedure was developed, but contrary to expectation, it has not been adopted by DARTS for routine funding of research, and DARTS failed to secure alternative research funding sources. The extension service has formally ceased its involvement in credit delivery but has failed to develop improved extension approaches, and has not undertaken any institutional restructuring. Improved, computer- based, methods of work planning, budgeting and financial recording have been introduced but have yet to be fully adopted. The policy reforms, in particular the liberalization of the markets for hybrid seed and burley tobacco, have encouraged greater involvement of private seed and fertilizer companies, but at the time of project closure there were clear signs that the GOM is becoming increasingly involved in fertilizer distribution again. 5. Major Factors Affecting Implementation and Outcome Factors outside the control of government or implementing agency: 5.1 A major drought in 1994 diverted attention away from routine implementation at the beginning of the project. The decision to have parallel and co- financing, resulted in complex implementation and problems of interdependency, with several components financed by two financing institutions. Finally, an important cause of the high level of staff vacancies and turnover was the increasing number of deaths-in-service, due for a large part, to HIV/AIDS. Factors generally subject to government control: 5.2 Macro-economic instability with high inflation, the devaluation of the Kwacha and high interest rates, had a negative impact upon the profitability of agriculture dependent on imported inputs and the use of credit. Associated with this and the impact of the 1994 drought, was the virtual collapse of the smallholder credit system. The failure of the MOAI to adopt a workable mechanism for the prompt 12  reimbursement of claims between the Bank, MoF and MOAI resulted in serious cash flow problems at the field level which resulted in chronic under-resourcing of the field extension and research programs, and undermined the entire project. In addition to involuntary losses, the high level of staff vacancies due to formal and informal freezes on recruitment as part of the civil service reforn, and resignations to join NGOs and the private sector also affected the project. Factors generally subject to implementing agency control: 5.3 Weak management, limited commitment, poor co-ordination of donor assistance in general were major factors contributing to the poor performance of the project. Inadequate formal and in-service training, lack of resources for field activities, non- and delayed payment of T&T claims, failure to rehabilitate FA housing, and career uncertainty resulting in low staff morale, had a negative impact on the project, especially at the field level. The disproportionate allocation of operating funds given to Headquarters staff through high levels of travel and subsistence claims, and preferential access to training opportunities, overseas study tours, vehicles and equipment, adversely affected implementation capacity and morale in the field. The TA recruited was generally poorly supervised and this undermined the envisaged institutional strengthening. Costs andfinancing: 5.4 In the SAR the total cost of the project was US$ 64.8 million. The IDA share of this was US 45.8 million, with co-financing from the African Development Bank of US$ 12.7 million and a Government of Malawi (GOM) contribution of US$ 6.3 million. The project formally closed on September 30, 1999, as scheduled; although categories 3 and 4 remained open until December 31 1999 in order to allow reimbursement of final claims for procurements in connection with SPS, and research equipment. Total IDA disbursements were SDR 30.98 million (US$43.62 million), which represents 95% utilization of the credit. There was a substantial disbursement lag throughout the life of the project. Originally the GOM contribution was set at 10% of total project costs (US$6.3 million), although after the MTR this was reduced to 5%. Actual GOM contributions, as reported by the MOAI in April 2000 amounted to US$1.32 million, equivalent to 21% or 41 % of the original and final targets respectively. Some 50 % of the GOM counterpart funding were spent on Research, mainly for the MPTF. Disbursement of ADB funds at the time of project closure were UA 5.68 million, equivalent to US$7.87 million. The bulk of the ADB disbursement, (US$6.Omillion), was for fertilizer purchase. 6. Sustainability Rationale for sustainability rating: 6.1 Few aspects of the project are expected to be sustainable. The regular research program and research -extension linkage has virtually ceased following closure of the project and failure of the MOAI to provide funds from its budget, and there are no sources of funding to continue the CRP. The continued operation of ASC is doubtful since the senior staff seconded has returned to the MOAI. The extension service has not developed an acceptable system or approach, and is essentially dependent on ad hoc donor project support. The motorcycle ownership 13  scheme cannot be sustained because reflows into the revolving fund are inadequate to purchase new machines. The management of the MOAI remains very weak, unable to face the challenges of the future in the absence of effective planning and budgeting of work programs, and poorly functioning M&E, financial, procurement and transport management systems. The continuance of the APATU program is likely through tailoring courses to the needs of various donors, especially under the SADEC Regional Food Security Training Program, and offering more specialized courses focusing on agribusiness for graduates Transition arrangement to regular operations: 6.2 Nothing has been done to transfer the operational costs of the project to the regular MOAI budget. In fact following project closure, there has been a major cut in the resources available to the MOAI. The result is that much of the research and extension activity, especially in the ADDs have become dependent upon small flows of "Other Recurrent Transaction" (ORT) funds and or and funding of specific activities by bilateral programs. Extension activities at the field level have been severely curtailed by the cuts that have resulted in resources being concentrated at headquarters and the ADD level rather than in the RDPs and EPAs 7 Bank and Borrower Performance Bank Lending: 7.1 Rated marginally satisfactory. The project design provided a holistic linkage between the core business of the Ministry (Research and Extension), and input supply and agricultural policy. The project's objectives were appropriate, but a centralized, top-down, message delivery extension system was not likely to be effective, and was certainly not fiscally sustainable. The project design, whilst drawing on the lessons from earlier projects, and implicitly recognizing weak management within the MOAI as the key constraint failed, to come up with discrete prescriptive activities to address these weaknesses, particularly decentralized decision making and management. The strategy of relying on TA to develop and introduce better management procedures was very risky, especially given previous experience with TA in Malawi. Supervision: 7.2 Marginally satisfactory. A large number of staff weeks were devoted to supervision of the project. Supervision missions were, for the most part, regular, and identified actual and impending implementation problems. However, although institutional strengthening was a critical part of the project, an institutional specialist was never included in the missions. Despite successive "unsatisfactory" ratings for the extension component, nothing was done to remedy the problems. Fundamental to all implementation problems was the inability, from the beginning to the end of the project, to resolve the problem of flow of funds. The Bank made great efforts to simplify its cost recovery procedures, but these were not adequately reciprocated by the GOM. Although an attempt was made at the MTR to re-orient the project, none of the action plan proposals were implemented and it was only at this stage that the Project Development Objective (PDO) rating was downgraded to "Unsatisfactory". 14  Consequently, at the end of 1998, the project was declared to be "at risk", disbursements against operating expenditures were suspended for five months, and it was decided not to extend the project closing date, even though some US$17.5 million of the credit remained undisbursed. At this point the Bank agreed to the utilization of most of the remaining funds to finance seeds, fertilizers and logistic costs associated with the first & second SPS. Overall Bankperformance: 7.3 Overall, the Bank's performance has been marginally satisfactory. There were three successive Bank Team Leaders during the project cycle and the emergence of serious problems at an early stage was not adequately recognized. Whilst the Bank did identify many of the project weaknesses and problems during implementation it was unable to satisfactorily address these weaknesses, and in particular the weaknesses in the institutional structure and management of the MOAI Borrower Preparation: 7.4 Rated unsatisfactory. Although there was probably some MOAI commitment to the project in the early stages, there was a fundamental lack of ownership, which resulted, in part, from contracting out much of the preparation work to local consultants Government implementation performance: 7.5 Rated unsatisfactory. The lack of ownership contributed to the project problems, by resulting in increasingly ad hoc management and an inability to take corrective action when problems started to emerge. However the main weakness was simply poor overall project direction and management by the MOAI, and a reluctance to introduce institutional reforms. Procedural problems, starting at the field level and stretching to MoF with respect to the timely submission of withdrawal applications, and reimbursement of claims by departments and ADDs, led to a chronic problem of channeling funds to the field. It took two years before the Special Account was allocated to cost centers, and the procedures for reimbursement via MoF were complex and slow, all of which undermined the entire project implementation. Non-salary operating costs were often late and insufficient, even though they were 100% reimbursable. Recommendations of the TA specialists to improve institutional capacity in financial and transport management, procurement and M&E were not adopted. Management capacity in the Ministry was further stretched by acceptance of other donor sponsored extension projects. Indecisive management and poor procurement procedures resulted in the almost complete failure to implement the civil works program (10% of original category). The supervision of TA, including failure to appoint counterparts and react decisively to problems, was also weak. For a period, audit reports were submitted late, and with qualifications Overall Borrower performance: 7.6 Overall, the borrower's performnance has been unsatisfactory. An attempt was made at the MTR to reorient the project by focusing on training at the field level, on strengthening M&E, research and extension and on activities to test 15  alternative approaches to extension delivery and management on a pilot scale. None of the proposed activities were implemented. Instead, all project activities were abandoned in favor of SPS. 8. Lessons Learned The principal lessons that can be drawn frorn the implementation of ASP are: 8. 1 General * For successful implementation all stakeholders must be involved in the conceptualization and design of a project. * Innovative project design, especially when it involves institutional reform, must be based on a realistic assessment of management capacity, and commitment to change. * Problems associated with centralized procurement, and the related excessive demands placed on the CMEU, suggest that decentralization of many project activities would have been preferable. * Successful institutional strengthening of MoAI can not only be addressed by capacity building through training and technical assistance, but will for a large part be dependent on successful Civil Service Reform and the willingness of management to make hard decisions to move the decentralization process. * Investments aimed solely at increasing productivity are unlikely to yield enhanced farmer incomes unless market development-related issues are adequately addressed simultaneously. - Effective flow of project resources to the field level is a sine qua non for successful project implementation. This must be achieved through a combination of simplifying procedures, strengthening financial management, and designing projects in which responsibility is devolved to the lowest level possible. - A dependence on co-ordinating mechanisms such as steering committees only works when such instruments have authority and meet regularly. 8.2 Policy * The policy reforms carried out under ASP alone are not sufficient to yield significant benefits to the sector and a range of "second generation" reforms need to be addressed. 8.3 Research * Improving research funding procedures (e.g. CRP), must be combined with improved research project selection methods * Research-extension-farmer linkages can only be strengthened by farmers participating in the development of research agendas. * Research Master Plans must be prepared in advance of any changes in research funding and management systems, and must be supported by a strong commitment to prioritize the use of funds. 16  8.4 Extension * Dependence on a standard routine, visit-based, message delivery extension system (such as BES) is neither effective nor fiscally sustainable. * Participatory approaches, which empower farmers' groups (e.g. VLPA), are most efficient and cost-effective; however such approaches are multi-sectoral and should not only involve Ministry of Agriculture staff. * In the interests of sustainability and affordability, a clear definition of publicly funded extension priorities is essential, together with partnerships between government, NGOs and the private sector. 8.5 Bank * Effective project supervision is a crucial part of implementation, and, where necessary, additional resources should be allocated to the supervision of projects when implementation begins to deviate substantially from original intentions. * Although the incorporation of a project within an existing ministry (MOAI) avoids dependence upon a Project Implementation Unit, this can only work if the institution itself is strong. 9. Partner Comments Borrower/implementing agency: 9.1 The Borrower prepared a comprehensive draft Project Completion Report in December 1999, which identifies the same project implementation problems and weaknesses as described in this report. The Executive Summary of their report is attached as Annex 7A of this report. The Borrower also carried out a Beneficiary Impact Assessment of ASP in November, 1999, which is available on request. 10. Additional Information 10.1 The Bank initially resisted support for the Starter Pack Scheme (free distribution of fertilizer, hybrid maize and leguminous seeds for 0.1 ha/hh) on the grounds that it was not sustainable (US$30million/year), and could not be satisfactorily targeted to the truly needed. The Bank was subsequently persuaded that it could be considered as a household-level food security mechanism to benefit for the poorest Malawian families and as part of the solution to the problem of soil fertility among the poorest smallholders. It was also recognized as being a more cost effective way of improving food security than simply importing maize at a time when national supplies and stocks were very low. Whilst the scheme undoubtedly contributed to the good yields achieved in the 1998/99 and 1999/00 maize seasons, the Bank remains of the opinion that such measures are essentially short-term or emergency measures that other bilateral donors are in a better position to support and their SPS should be based on a clear safety nets strategy. 17  Annex 1 Key Performance Indicators OUTCOME/IMPACT INDICATORS I ndiWcator iProjectediin SAR Lit est/Actual Minimum average real smallholder income increases 6% No estimates made Reduction in pre-harvest, food-scarce period from 3 to 2 BIAS suggests that the food-scarce period in 1999 reduced from 2 to I months month Adaptation, dissemination and adoption of labour-saving 2 technologies Several technologies have been adopted by smallholder farmers technologies in food processing/transportation/fuel use/collection Three low cost agricultural productivity enhancing By 40,000 farmers technologies Reduction in annual rate of research staff turnover 20% Coverage of smallholder population by extension 60% Only indicator is FLS: farm household ratio - average 1: 1680 but ranges from 1:748 to 1:3087 Ratio of female farmers covered by extension 50% No data available Number of smallholders adopting HYVs of maize 40% As a result of SPS I and II, in 1999, 64-70% of farmers using hybrids Number of farmers using nitrogen-fixing soil conservation 40% Estimated in ASP Semi-Annual Report at 40% techniques Upgrading FHAs to FAs 100% Some FHAs have been upgraded to FAs Incorporation of nutritional messages into FAs work 100% Information spread by leaflets but FAs not trained on how to use Nutrition Manual No. of ha under small-scale irrigation 300 (ADB support) No. of farmers using small-scale irrigation 1000 (ADB support) No. of local, problem-specific surveys undertaken by each 6 to 24 Block Survey and Policy Impact Surveys conducted only for two ADD seasons - subsequently stopped due to lack of funds Reduction in MOA transport costs 12% No data available - but indications that no savings have been achieved No. of GOM staff who have received Agricultural Policy 30 Only I student graduated with M Sc Analysis Training Increase in number of private hybrid maize seed Thd In 1999, two private seed distributors in operation producers/distributors _ ___ __ 18  OUTPUT INDICATORS Indicator- Projected in Latest/Actual SAR Research: Action Plans for 7 research groups Action plans prepared originally but not up-dated. Annual programme of research agreed by consensus but limited time and inputs from extension. Plans to reflect priorities and budget ceilings Budget ceilings not available before planning programme. Large part of budgets spent on travel Plans reflect smallholder needs Research not explicitly geared towards smallholder problems Senior researcher appointed for contact with ADDs Liaison researcher appointed but not replaced upon retirement. Weak response to ADD problems due to lack of travel funds On-farm trials researclher-extension-farner managed On-farm trials conducted but questionable reliability due to limited supervision Increased number of field days Field days at stations and on farmer fields, well attended, Joint committees to review results and plan next year Public seminars held; Joint Committee to review results at annual meeting involving extension staff, university, ASC and farmer representatives Results disseminated in publications Leaflets and posters distributed, but many technologies not publicized Expanded use of contract research 33 projects funded and scientists satisfied with results; projects receive more funds than DARTS projects; funds reach the scientists directly; accountability AS C as advisory board ASC successfully set up and operated during project Technical secretariat to ASC set up Technical Secretariat established National Research Master Plan prepared National Agriculture and Natural Resources Research Master Plan finally issued October 1999 Improved research monitoring Monitoring of research generally good Improved accounting No improvement in accounting Advise on allocation of public research funds Yes Stronger links with regional and international Few systematic links with international centres established research bodies fostered 19  Improved accounting at DARTS No improvement in accounting Professional Interview Board set up Interview Board not set up Studies conducted Action plans prepared Yes Training programme for research scientists DARTS has benefited from training in research programme undertaken preparation and budgeting; inadequate professional training - only 2 of 39 overseas training slots used Improved career structure and promotion Overall poor morale in DARTS due to lack of promotion prospects; opportunities erratic funding; laboratory equipment arrived at end of project TA: Research Management Expert 12 p-m Not recruited TA: Librarian 20 p-mi Not recruited Extension: Food and nutrition programme National food and nutrition survey conducted; training materials produced; F&N Manual finally printed but not distributed TA: Food & Nutrition Expert 12 p-m Recruited late Training: Programme implemented 1994-96, 1997-99 no training took place due to need for Training Needs Assessment; 1999, overseas courses for 35 staff plus in-country courses for 11 staff Diplomas 6 3 M.Sc.s 6 19 Study tours 2 4 short courses overseas + 2 in-country courses De-link extension from credit recovery Extension de-linked from credit recovery butFAs still involved in organising and training credit groups; time released increasingly well used to work with special interest groups Streamline MOA's structure and staffing No progress with streamlining structures; Functional Review not to be implemented until after project Reorientation training of staff Substantially less than planned Upgrading training of FHAs to FA or VA level Limited number of FHAs upgraded More integrated information on main areas covered Not clear whether overall coverage of resource-poor has occurred. Low-cost soil fertility objectives met mainly through PROSCARP. On-farm root crop multiplication benefited resource-poor More targeted infonnation linked to local conditions Location-specific fertilizer recommendations from MPTF 20  and tvpe of smallholder Expand mini-plots Mini-plots (OFDs) expanded greatly in number but dubious qualitv; block gardens declined but not abandoned Reduce block garden demonstrations Introduce topics in response to farmer requests Participatory extension not institutionalized and few FAs trained in Methodology Shift focus from credit groups to wider range through Many natural groups incorporated into extension but needs emphasis churches, mosques etc Greater coordination with local NGOs Some coordination with NGOs, which work with FAs but no contracting-out of services Reorientation training to mainstream women's issues No progress in reorientation of training to women's issues Training in nutrition by F & N Unit Training on soya bean done and preparation of food and nutrition materials Livestock and environmental training to farmers No livestock or environmental training done Improve field supervision, budget and planning - No improvement in supervision, budgeting, planning - allowances transport & allowances increased but often not paid and always late Improved access to vehicles, bicycles & motorcycles: 4x4 double cab 20 7 (13 by ADB) 10 seater minibus I At NRC Motorcycles 440 485 (44 by ADB) Bicycles 865 (50 by ADB) Motorcycle ownership scheme for supervisors Access to vehicles and motorcycles improved through ownership scheme. but negated by late or non-payment of allowances (T&T) Houses for field staff 80 EH 10 houses No improvement in field staff housing; failed to implement substituted FA house rehabilitation programnme Office/training facilities at EPA level: Offices 60 Not constructed EPA centres 4 Not constructed Classrooms 4 Not constructed Storerooms 12 Not constructed Equipment: computers 4 (2 by ADB) Strengthen mass media for resource-poor - focus on No improvement in mass media except "wind-up" radios distributed radio under EU scheme 21  Agricultural Communications Branch (ACB) Upgrade ACB and VAS equipment ACB received equipment TA: Extension Training Specialist 18 p-m Not recruited Input Supply: Liberalising production and marketing of seed to Hybrid maize seed production and marketing liberalised 1993 increase competition, choice and lower prices Remove all subsidies on improved seed __ Subsidies removed 1994/95 Formulate and implement policy reforms: Deregulate imports and make SFFRFM commercial Import deregulated. SFFRFM still government-owned Emphasise high analysis fertilizer import, ICB, Companies free to import whatever fertilizer demanded bagged Single max target retail price for each type No single retail price Liberalise distribution for PS entry into wholesale Distribution liberalised; initially nine companies involved in and retail and freedom to buy from SFFRFM and importation and distribution, but at end of project only two Admarc Admarc seller of last resort to defend target max Admarc still dominates the market and in some remote places, is sole price supplier Foreign exchange for fertilizer - IDA 12,500 t Procured using SEDOM and INDEFUND as revolving fund Foreign exchange for fertilizer - ADF 18,750 t 19,774 t CAN procured Institutional Strengthening: Planning and Financial Management: TA and in-service training to HQ and ADDs for: TA unsatisfactory - no usable outputs produced Portfolio management in PSIP Introduction of MTEF approach on full scale for 1998/99 budget. FY changed from April to July Work plans linked to ADD budget preparation Failure to prepare realistic work plans Project planning anid budgeting Annually Improved accounting Some improvements in financial management TA: Financial Management Expert 27 p-m Recruited but performance poor TA: Financial System Analyst 12 p-m Recruited but performance poor Transport Management: Introduce MOA/ADD transport management system: Not implemented 22  Training in transport and workshop management Workshop training not conducted Private sector maintenance on pilot basis Yes - almost all maintenance by private sector Workshop equipment, computer hardware and Computers not procured software TA: Transport Management Advisor 24 p-m TA unsatisfactory - monitoring systems designed but not operational; poor TOR strictly adhered to Procurement Study to improve procurement and intemalise results 100 p-m Responsibility for the study shifted between different Ministries and was finally completed (under OPC) at the end of 1999 Support for MOA procurement - procurement agent Procurement slow and Items procured not distributed according to to handle civil works and goods plan In-service training Little training conducted by TA; two short course attended TA: Procurement Management Expert 24 p-m Extended to 60 p-m; failed to conduct training, no procurement Policy Analysis Training: manual preparation Non-degree and PG courses, Bunda RD Dept: Commodity, market and price analysis Conducted with reduced frequency Food security and nutrition policy Conducted with reduced frequency PAM Conducted with reduced frequency Environment and natural resource policy Conducted with reduced frequency Project preparation and management Conducted with reduced frequency M&E Conducted with reduced frequency M.Sc degree course 30 1 M.Sc awarded; 6 students completed ISt year only before end of project Support for Bunda: Hostel 20 beds 30 bed hostel completed 1998 Computer training room Constructed 4x4 Pickup 6 Procured (5 by ADB) 4x4 Saloon car _ 2 Procured (I by ADB) 25 seat minibus I Procured TA: Policy Analysis Course _33 p-m Unsatisfactory - contract with IFPRI terminated Other Procurement:- - Motorcycles 12 6 (6 by ADB) Bicycles 150 150 23  Conmputers 66 55 (I1 by ADB) Printers 32 29 (3 by ADB) Photocopier 5 3 (2 by ADB) Facsimile 2 (2 by ADB) Monitoring & Evalhation: Streamlining survey and reporting systems Little success in strengthening M&E; monthly, semi-annual and annual; report chronically late Pilot to introduce Area Sampling Framework (ASF) not followed-up Developing quantitative indicators and targets, RRA Performance indicators still not agreed techniques Strengthening M&E capacity at ADD level - training Upgraded 7 staff to M.Sc level Computers to enhance speed Procured Increased management accountability Work plans based on log frame approach started but poorly utilised Beneficiary and local problem assessment Two beneficiary assessments carried out Disaggregating data by gender and farm size Not achieved TA: M&E Advisor 7 p-m Not recruited Management: Vehicle 1 Procured Computer I Procured Printer 1 Procured 24  Annex 2A Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Actual/Latest Percentage of Estimate Estimate Appraisal Project Cost By Component US$ million US$ million Research 18.10 6.54 36.1 Extension 24.20 9.73 40.2 Input Supply + SPS 10.00 21.68 ' 216.8 Institutional Strengthening 11.30 13.97 123.6 Project Preparation 1.20 0.89 74.2 Total Baseline Cost 64.80 52.81 Total Project Costs 64.80 52.81 Total Financing Required 64.80 52.81 1/ Includes US$6.0 million disbursed by AfDB 25  Annex 2B Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) Procurement Method: Expenditure Category ICB : NCB Other2 N.B.F . Total Cost: 1. Civil Works 0.00 5.90 0.00 0.00 5.90 (0.00) (0.00) (0.00) (0.00) (0.00) 2. Vehicles & Equipment 5.30 0.00 1.00 0.00 6.30 (2.9) (0.00) (0.9) (0.00) (3.8) 3. Technical Assistance 0.00 0.00 5.90 0.00 5.90 (0.00) (0.00) (5.4) (0.00) (5.4) 4. Training 0.00 0.00 13.20 0.00 13.20 (0.00) (0.00) (11. 1) (0.00) (11.1) 5. Fertilizer & Seeds 10.00 0.00 0.00 0.00 10.00 (4.0) (0.00) (0.00) (0.00) (4.0) 6. Operating Costs 0.00 0.00 22.30 0.00 22.30 (0.00) (0.00) (16.3) (0.00) (16.3) 7. Refinancing of PPF 1.0 0.2 0.00 0.00 1.2 (1.0) (0.2) 0.00 0.00 (1.2) Total 16.30 6.10 42.40 0.00 64.80 (7.9) (4.2) (33.7) (0.00) (45.8) 26  Project Costs by Procurement Arrangements 3' (Actual/Latest Estimate) (US$ million equivalent) Procurement Method 1l Expenditure Category JCB NCB Other2 -N.BF. Total Cost 1. Civil Works 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 2. Vehicle & Equipment 0.00 0,00 0.00 0.00 0.00 ..__________________ .(0.00) (0.00) (0.00) (0.00) (0.00) 3. Technical Assistance 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 4. Training 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 5. Fertilizer & Seeds 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 6. Operating Costs 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 7. Refinancing of PPF 0.00 0.00 0.00 0.00 0.00 ________________ __ . .(0.00) (0.00) (0.00) (0.00) (0.00) Total 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 1/ Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies 2/ Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. 3/ Not available 27  Annex 2C Project Financing by Component (in US$ million equivalent) Percentage of Appraisal_Estitmate Actual/Latest Estimate A Appraisal ___________________ ;Bank Govt.: CoF. ::Bank Govt. CoF. Bank Got tCoF. Research 15.80 2.30 5.90 0.70 0.00 37.3 30.4 0.0 Extension 17.20 3.10 3.90 8.30 0.50 0.90 48.3 16.1 23.1 Input Supply + SPS 4.00 6.00 15.70 4 6.00 392.5 0.0 100.0 Institutional 7.60 0.90 2.80 12.90 0.10 1.00 169.7 11.1 35.7 Strengthening I _ _ Project Preparation 1.20 0.90 75.0 0.0 0.0 Total 45.80 6.30 12.70 43.60 1.30 7.90 95.2 20.6 62.2 Note: Co-financing by the African Development Bank 1/ Based on data provided by MoAI (4/26/00) 2/ Includes US$9.7 million for SPS 28  Annex 3 Economic Costs and Benefits Cost Benefit Analysis 1/ 6~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1/ Not Available 29  Annex 4A Bank Inputs Missions: M 4 C rS ;.N .nof 'Perso i an SeiaIt eoiance Rati. SAge obf Project Cycle_____ __________ Impemetatio n Dvlmn : . .: : :i:: : i S i : . E : iei ;,; :. ' W -.- Y m oR ei R-z Month/Year Count specialt Pro.greObectve. Identification/Preparation Appraisal/Negotiation Supervision Supervision (PSR 1) June-93 Initial S S Summary Supervision (PSR 2) April-94 6 Ag Econ (3), Agron (2), Res S S Supervision (PSR 3) 7 Ag Econ(3), Agron (2), Res, Ext S S June-94 Supervision (PSR 4) 4 Ag Econ (2), Ext. Res. S S Dec-94 Supervision (PSR 5) 6 Agron (1), Ag Econ, Res (2), S S Jun-95 M&E, Ext Supervision (PSR 7) 7 Agron (2), Ag Econ, Res (2), Fin S S Jan-96 Anal, Proc. Supervision (PSR 8) 6 Agron (2), Ag Econ, Res (2), Ext S S Jun-96 Supervision (PSR 10) 7 Agron (2), Ag Econ, Res (2), S S Jan-97 M&E, Fin Anal Supervision (PSR 12) 7 Agron (2), Ag Econ, Res. (2), Fin S S Aug-97 Anal. Ext. 30  Supervision (PSR 14) MTR I/June 1998 13 Agron (2), Econ (2), Irrig, Ext, S U Res (2), Ag Econ (2), Fin Anal, M&E, Task Asst. Supervision (PSR 15) 7 Agron (2), Ag Econ, Ext, Fin U U November 1998 Anal, M&E, Task Asst Supervision (PSR 16) 4 Agron (2), Ag Econ, Fin Anal U u June 1999 Supervision (PSR 17) 3 Agron (2), Fin Anal U U October 1999 ICR ICR MissionApril 2000 3 Econ, Agron, Ext U U 1/ Key to type of expertise: Ag. Econ. = Agriculture Economist; Agron. = Agronomist; Ext. = Extension Specialist; Res. = Research Specialist; Fin. Anal. = Financial Analyst; M&E = Monitoring & Evaluation Specialist; Iff. = Irrigation Specialist; Proc. = Procurement Specialist. 2/ Performance Ratings recorded in Form 590s/PSRs and in the ICR S = Satisfactory; U = Unsatisfactory 31  Annex 4B Bank Inputs Staff.- Stag ofproject CyCl 'A tual/Lats Estimat iNo. St ff weks (000) Identification/Preparation 107 318 Appraisal/Negotiation 173 333 Supervision 408 1097 ICR 41 115 Total 728 1863 32  Annex 5 Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Macro policies NA Sector Policies M Physical N Financial N Institutional Development N Environmental NA .Social Poverty Reduction M Gender N Other (Please specify) Private sector development M Public sector management N Other (Please specify) 1/ Marginally satisfactory due to Starter Pack initiative Annex 6 Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending s1' Supervision S! Overall S 6.2 Borrower performance Rating Preparation U Government implementation performance U Implementation agency performance U Overall U 1/ Marginally Satisfactory 33   Annex 7A MINIS'TR OF AGRICULTURE 'GOVERNMENT OF MALAWI AND'IRRIGATION: PROJECT COMPLETION'REPOR FRTHE AGRICULTURAL SERVICES PROJECT( (SP), Cr. 2514 MINIlSTRY OF AGRICULTURE AND IRRIGATION: PLANNING DIVISION P.O. BOX 30134 LILONGWE:3 Decembter, 1999.  EXECUTIVE SUMMARY As the Agricultural Services Project (ASP) came to a close on 30 t' September, 1999 a Project Completion Report (PCR) has been compiled by the Ministry of Agriculture and Irrigation (MoAI) as a preparatory work for the donor consultants who will produce the final PCR in January, 2000. This report covers ASP (IDA) components and a summary of the ASP (ADF) components. ASP's key objectives were: a) to support institutional and management reforms in research, extension and input supply, with the aim of providing cost-effective and sustainable technologies for different types of smallholders and agro-ecological zones; and b) enhance smallholder incomes and food security by increasing the productivity of smallholder agriculture, especially that of female headed households and resource-poor farmers. The project's main aims were to: a) improve focus, cost effectiveness and impact of agricultural research and extension activities; b) support seed and fertilizer marketing reforms to increase supply and use, and provide foreign exchange for importing incremental fertilizer; c) strengthen planning and financial management, transport management, procurement, agricultural policy analysis, small-scale irrigation development, and monitoring and evaluation in MoAI and constituent ADDs. All loan conditions tied to the ASP were fulfilled by the Government of Malawi (MoAl) between 1994 and 1995. During the early stages of the project implementation was smooth but got stagnated towards the end of the project. The Total Estimated Cost was adjusted to take care of inflation and newly introduced activities but cash floats remained low, and delays in reimbursements of funds was experienced because of late submission of claims by cost centres to MoAI and then to Treasury. Although a Financial System Analysts was recruited in 1995, no improvement was experienced in the Accounts Department and this resulted into poor performance of the TA as most of the work stipulated in his Terms of Reference (TOR) were not achieved. The mandate of the Procurement component under the ASP-IDA was to provide institutional support and management improvement through the purchase of equipment, and, to help in building capacity of the procurement unit in the ministry. The Technical Assistant was recruited in 1994 to facilitate the implementation of the procurement activities. Some notable achievements were made in the equipment purchase sub-component where equipment was purchased and distributed to the project beneficiaries. Close to 80% of the funds planned for this activity were utilized. Under the procurement of agricultural inputs component, the major achievement was the purchase of inputs for the Starter Pack Scheme for the 1998/99 and 1999/2000 seasons. The total input purchases for the scheme exceeded the planned (US$4.0 million) estimate for the component by over 100% as balances from other sub components of the project covered for the financial increase. The Technical Assistants who were recruited under the other project components utilized over ix  30 % of the allocated funds allocated under the TA sub component. The performance of most of these TAs was not impressive as some of them left at the end of their contracts without achieving most of what was stipulated in their TORs and also without leaving contract completion reports behind. Nothing was achieved under capacity building for the procurement unit as no training was organized and as the project was closing the unit had no staff capable enough to handle procurement activities. The same was the case with civil works, as no infrastructure was constructed as planned in the appraisal. The fundamental objective of the research component was to establish a cost effective and sustainable research system that would generate technology responsive to farmer's needs, especially those of the resource poor. During the project implementation period, all research stations except for Kasinthula, Baka and Mkondezi established their own cash offices. The Department of Research has managed to release new varieties of various crops and make recommendations on cultural practices. Although a number of technologies were released, uptake of these technologies has been very low due to several factors, among them, being the inavailability of farm inputs mainly due to escalating prices of inputs caused by the liberalization of the economy, the weak research- extension-farrner linkage and the erratic flow of funds. The Agricultural Sciences Committee (ASC) was mandated to strengthen the National Agricultural Systems (NARS) structurally, technically and financially in order to increase agricultural productivity and promote sustainable utilisation of the renewable resources. The Committee achieved most of the activities it was mandated to undertake. The Contract Research Programme was initiated, implemented and managed leading to generation of twenty-five (25) high impact technologies. The Malawi Agricultural and Natural Resources Master Plan (MANRRMP) was produced and launched in September 1999. An effort to co-ordinate the research agenda has been very successful and most research institutions are now familiar with steps followed when appraising research proposals and funding the research projects. The development of a sustainable funding mechanism for research and consultations was not successful, This was due to donors not willing to lump their resources in one account but rather fund specific programmes of their interest. However, researchers have indicated that funding through Agricultural Sciences Committee was good as it funded trials directly and usually in line with proposed budgets. The Extension and Crops departments had a mandate to i) improve the efficiency and effectiveness in delivery of agricultural extension services to smallholder fanners, ii) development of appropriate management system for the services, iii) to improve crop production and productivity through increased farmner use of improved varieties and fertilizer uptake, iv) to improve the availability of seed through the involvement of smallholder farmers in seed multiplication programmes; and v) encourage farmers participation in various agricultural activities through the use of On-Farm - Demonstrations (OFDs) rather than government owned demonstrations. Maize Productivity Task Force (MPTF) Action Group 1, which was funded within the ASP- IDA, generated and disseminated technologies that promoted the use of area specific inorganic fertilizers etc. x  All the activities to be carried out under these technical areas were to be implemented at the ADDs, but co-ordinated as a technical area under new departments namely Land Resources and Conservation Department and Irrigation Department. New technical areas too were introduced under the Extension and Crops Component and these included Agricultural Communications Branch (ACB) and Natural Resources College (NRC). The performance of the department in delivery of extension services can be termed as marginally satisfactory. The group contact approach and the participatory extension methods adopted during the implementation period was not fully achieved as few extension staff were trained on these. The motorcycles ownership scheme improved farmer coverage and staff morale. The emphasis on group dynamics and management has assisted farners to have access to credit from various lending institutions. A good number of field staff underwent upgrading courses at NRC apart from those who benefited from offshore short and long- term courses. This has and will obviously improve their capacity. The department re-enforced the group approach and collaborated with non-governmental organizations in order to extend coverage to a wider range of smallholder farmers through informal groups such as natural, special interest groups and strengthening group dynamics and mobilization. In order to strengthen the block extension system, some extension approaches were initiated and implemented during the project period. These approaches include Participatory Extension Methods (PEM), Village Level Participatory Approach (VLPA), Priority Setting and Diagnostic Grid. There has been very little improvement, in Research-Extension-Farmer (REF) linkages due to poor coordination between research and extension departments, and, lack of adequate funding, Most of the researchers expected extension department to bear the cost whenever they attend a meeting though both departments had a budgeting sub programme of the REF linkage. Due to inadequate funding during the implementation period, extension department failed to meet these costs to sustain the linkage. Lack of adequate funding also affected the implementation of MTRWs so much that they were only implemented during the initial phase of the programme. Joint field supervision were carried out smoothly during the initial phase of the project but were never carried out during the course of the project implementation as departments expected either Planning Division or Department of Agricultural Extension Services to provide allowances and transport for the joint field supervision. Very little achievement has been made under the Food and Nutrition Unit though an expert was recruited. The activities of this unit still could not be differentiated as in the past, as most of the field staff who handled the Food and Nutrition issues were Women's Programme Officers in the ADDs. Most of the field staff lack knowledge in special nutrition issues because not much training was carried out although the Nutrition Technical Assistant developed training manuals for Food and Nutrition staff. xi  The Motorcycle Ownership Scheme was implemented as a way of reducing the burden the ministry had in maintaining staff motorcycles and in order to improve mobility of field staff, to achieve wider farmer coverage. While the beneficiaries are quite happy with the scheme and that it has been very successful in actual acquisition and distribution of motorcycles, problems have been faced when it came to payment of travel and transport claims to supervisory staff. Production of the legal framework forms for the motorcycle ownership scheme took too long as they were to be done in conjunction with Ministry of Justice. Lack of continuity of the scheme demotivated those staff who did not benefit from the scheme. Under the Training a number of courses have been conducted during the project life, both staff and farmers have benefited from these courses. Several members of staff benefited from offshore courses, however, internal training for both staff and farmers has been less than those planned at both headquarters and ADD levels. Poor flow of funds has contributed to the poor performance of the component. The Beneficiary Impact Assessment Survey (BIAS) for the ASP, which was carried out, involved interviewing a total number of 480 smallholder farmers and 216 ADD staff. The survey findings indicate that farmers really benefited from the project. 70 % of the respondents reported to have been using improved hybrid maize seed over the years due to the increase in number of players in input supply. About 50% of the farmers keep livestock for home consumption while others keep them as source of income, draught power and to meet social obligations i.e. lobola. Farmers access to extension services has remarkably increased over the project life while less than 50 % of them had access to credit facilities through farmer groups. The most common sources of credit are MRFC, ADMARC and FFC. Farmers faced a number of problems which reduced their participation in credit schemes and these include - i) high interest rates stringent credit requirements, ii) lack of collateral and; iii) lack of adequate and timely information about sources of credit. Increase in input prices due to direct response to changes in foreign exchange rates has led to some farmers applying organic manure as an altemative. However, some farmers indicated that they do not use manure because of the following problems: - transportation of manure, shortage of labour and source of manure. Nearly 50% of the farmers indicate that their food security status has improved. This achievement could not be attributed to improvement in the provision of extension services alone, but rather to the many micro and macro-economic effects encountered following policy reforrns during the project implementation period and also due to other projects which were implemented parallel with ASP i.e. APIP, Starter Pack Scheme, Smallholder Food Security Programme etc. During the initial implementation of the project, the Agricultural Communication Branch (ACB) was not very much emphasized until it was realized that the communication branch be taken on board to improve technical message delivery and coverage. During the project period, modernization of the equipment in all ACB sections was done and this xii  has improved capacity especially in the printing section, thereby enabling it to print and circulate various agricultural booklets and leaflets. Additional equipment has been provided to all ADDs and there has been expansion in use of radio by farrmers in listening to Agricultural Programmes. Co-ordination between the Departments of Extension and Crops and the ministry headquarters up to the ADDs was good. The department benefited a lot in terms of finances from NGOs however, inadequate funding during the project implementation period has been a major stumbling block to full implementation of the planned activities in order to achieve more. The Natural Resource College (NRC) benefited from ASP through strengthening the college activities, provide college material support, and provide external training for members of staff at the college. At least some equipment and material were purchased and up grading courses and other training sessions were conducted. Most of the sub component activities were not done especially under college strengthening (exchange programmes and student attachments). The major problem was lack of funds and claims of expenditure not being honoured on time. The Training component was implemented through the Hunman Resources Development and Management unit of the ministry. The main activities of this component were to i) assist in recruiting local personnel to fill the vacant posts in the ministry not useful for specific posts, ii) recruitment of Technical Assistants which was done in conjunction with the Procurement unit, iii) to coordinate training plans for the ministry for both in country and offshore courses, iv) coordinate the production of the Legal framework for the Motorcycle Ownership Scheme (MCOS), and, carry out a Training Needs Assessment for the entire Ministry. Most of these activities were achieved though there were some problems, which affected their implementation. There are still a lot of vacant posts in the Ministry caused by deaths, retirements, resignations etc. Though some members of staff benefited from training, most of the beneficiaries were from headquarters. The Training Needs Assessment affected the ministry-training plan as the consultants took time to come up with a report. The Institutional Strengthening component consists of five sub-components namely: - Planning & Financial Management, Transport Management, Agricultural Policy Training Unit (APATU), Central Monitoring and Evaluation Unit (CMEU) and Malawi Investment Promotion Agency (MIPA). Under the Planning and Financial Management, a Planning and Financial Management Specialist was recruited in 1994 to develop yardsticks for linking workplans to budget allocation in the Ministry. However, the performance of the TA was below standard and nothing was achieved as required under his Terms of Reference. Efforts have been made to come up with a proper format that would link annual workplans to budgets and several formats have been developed but no link has been provided. The efforts were overtaken xiii  by Treasury's introduction of the Medium Term Expenditure Framework (MTEF) which all ministries and departments adopted. The objective was to achieve optimal resource allocation particularly to priority areas thereby reducing government expenditure. Not much was achieved under Transport Management. A Transport Advisor was recruited in 1995 to manage transport system in the Ministry and to improve the efficiency and effectiveness of service delivery. This was to be achieved through training in transport operations and workshop management. A monitoring system was devised for the motorcycle ownership Scheme and use of vehicles. However, little progress has been made on the implementation of action plans for transport management during the project period. The Agricultural Policy Analysis Training Unit (APATU) of Bunda College of Agriculture has conducted several short courses and workshops since the project began. The unit co-ordinated and conducted courses initiated by government departments such as Ministry of Agriculture and Irrigation and National Economic Council and other Non- Governmental Organizations. Only one student has been trained under ASP at MSc level and at present six (6) students from the ministry are studying for MSc degree in Extension, while other six (6) students from the ministry resented the MSc Degree in Agricultural Economics at the college due to:- i) lack of specific time frame for the course; ii) lack of international exposure. The proposal to hire PhD level staff to strengthen capacity of lectures in the Rural Department and the introduction of an attachment programme to an overseas university for a specified time period never materialized but APATU managed to hire IFPRI staff at US$90,000 a week to support the implementation of activities, however, it was felt that one week period was too short for the impact to be seen. It was then proposed to hire a local member of staff who was cheaper and ready to support the unit for longer period. The Monitoring and Evaluation component was to come up with a detailed M&E system which would be evolutionary and responsive to the requirements of implementing managers, planners, policy makers and donors. Several achievements were accomplished which included: - development of Monitoring Indicators, guidelines for the collecting and reporting on progress of ASP implementation of activities, formulation of a standard reporting system which was abandoned and replaced by a Semi Annual and Quarterly reporting formats, Annual Work Plan and Budgets System which are currently used by the ministry. Evaluation Working Party meetings were carried out where Evaluation Officers meet and planned how they would carry out monitoring and evaluation issues and at least two extension monitoring surveys were conducted and reports produced during the project period. Several staff training were carried out and these were: - A Participatory Rural Appraisal (PRA) course in conjunction with OUTREACH -India; a Monitoring and Evaluation Training in conjunction with American Institute for Research. Four (4) graduate Officers from the CMEU/ADD were trained overseas at MSc level in Agricultural Economics. The CMEU and ADD evaluation units got equipment like computers and printers under this project. xiv  The CMEU carried out the Beneficiary Impact Assessment Survey (BIAS) on ASP and has produced the Project Completion Report (PCR) for the project. The Area Sampling framework which was planned to be funded under ASP never, took off. The Starter Pack Scheme (SPS) which was initiated in 1998 had the following objectives: - i) to assist fill the food gap; ii) to promote crop diversification; and iii) to promote the concept of soil fertility improvements. The programme was set to address Malawi's persistent food gap than providing free food relief, as has been the case in the past. The inputs were planned to be in a form of a package containing cereal and legume seed, and fertilizer to plant on a 0.1 ha of land and was to cover 2.8 million smallholder farmers. Funding for both the 1998/99 and 1999/200 seasons for SPS was as follows: i) Government of Malawi (GOM) provided fertilizer, ii) seed and other logistics for the entire exercise was funded by DFID - UK, European Union (EU), World Bank (WB) and the Republic of China. World Bank funded SPS through the Agricultural Services Project with US$ 1,192,912.00 in the first season and US$ 8,824,600.00 in the second season. The programme had a positive impact on the National Crop Production as production of maize for the 1998/99 season was 2,479,406 MT compared to 1,772,392 MT for the 1997/98 season. But the increase in production can not only be attributed to SPS alone because there were other programmes which were implemented parallel to this programme and had the same objective of increasing crop production. The Ministry of Agriculture and Irrigation, through the Agricultural Services Project supported the activities of the Agri-business Advisory Unit ( AAU ) under the Malawi Investment Promotion Agency. The establishment of the Unit followed the recommendations that were made by the December, 1994 World Bank supervision mission. The Unit was mandated to promote and support the capacity of Malawian entrepreneurs to develop viable investment proposals. Over the project life, the Unit made a number of activities like the establishment of the Business Infornation Centre where entrepreneurs, firins and associations are able to access cost effective, technical and marketing information. Through the establishment of the centre, the Unit has managed to raise awareness among business information beneficiaries about the financial services available. The Unit also managed to send both farmers and staff on study tours and courses overseas. The Principal Secretary for the MoAI chaired the Agricultural Sector Steering Committee (ASSC) which was setup in 1994 to coordinate the implementation of the project activities. The committee met only twice during the project life because of a lot of adhoc work within the ministry which involved most people who were members of the committee. Though a Project facilitator was recruited during the same year, not much was done by his office as he was all alone and could not co-ordinate activities as expected. This led to his contract not renewed and coordination of the ASP activities xv  were left to be done by Planning Division - CMEU. The unit still had problems in coordinating the ASP activities due to reasons beyond their control. During the initial stage of the ASP, the Land and Resources Conservation component was implemented under the Land Resources Unit, which was, then under the Department of Agricultural Extension Services (DAES). After the first year of the project, the Unit was established as a department. This component is being funded by the African Development Bank (ADB) and will phase out in December 2000. The Land Resources and Conservation Department was mandated to enhance soil conservation and fertility improvement in the gardens of resource poor farmers. Some achievements were registered in addition to farmer training with an intention of equipping farmers in all aspects of land resources and conservation, making and use of A-Frame for pegging contour ridges, reclaiming gullies using simple techniques and establishing agro-forestry nurseries and transplanting in seedlings. The land resources activities were implemented in all ADDS and farmers in general were receptive to the soil conservation technologies put forward to them, but the major problem has been on supervision of the adoption of these technologies due to lack of funds caused by the suspension of disbursement. The Small-Scale Irrigation component is also being funded by ADB and the component is focussing on institutional development of staff, development of self-help gravity fed irrigation and the construction of the headquarters of the Irrigation Department. This component too will phase out in December, 2000. Achievements under this component include: - training of staff both at MSc and BSc level, procurement of fifteen motorized pumps and treadle pumps and Irrigation policy and Action plan has been produced. The performance of this component at the ADD level has been poor due to poor flow of funds caused by suspension of disbursement. The Livestock extension component was funded by the ADB and is to improve the food security and poverty alleviation of resource poor farmers through animal production and the promotion of simple livestock extension services through Field Assistants and specialized extension messages through the Veterinary Assistants. Achievements have been made in addition to supervision of the livestock husbandry activities in the field, which was successful. The major problem faced is poor flow of funds. Though ASP was rated unsatisfactory by the donor during the Malawi CPPR meeting which was held in April, 1999 in Lilongwe, it was felt by implementers of the project that ASP was a success as it improved several aspects in the Ministry which include: equipment (computers, printers, photocopiers etc) vehicles, training ( offshore and in- country), improved mobility of field staff, improved extension and research activities, and operating cost. These would not have been achieved using revenue funds which mostly are inadequate and also not adequate to sustain most of the ASP activties now that project has closed down. xvi  Data collection for this Project Completion Report (PCR) exercises was carried out from 25th October to 3rd December 1999 and report writing was done during the remaining days of December 1999. The team involved in this work comprised of technical experts from within the Ministry, and these were: Miss. M.J. Nyekanyeka - Economist (M&E - MoAI HQ.) - Team Leader Mr. T. Fatchi - Economist (M&E - MoAI HQ) Mr. H. Mdyetseni - Economist (Budgeting - MoAl HQ) Mr. R. Mchenga - Senior Economist (SLADD) Mr. I. Mwale - Economist (M&E - MoAI HQ) Mr. G. Chande - Economist (M&E - MoAl HQ) Mr. M. Kasamba - Principal Accountant (MoAI HQ) Mr. H. Lozo - Accounts Personnel (MoAl HQ) xvii  Annex 7B IALAWI AGRICULTURAL SERVICES PROJECT (ASP) (IDA Credit 2514 MAI) IMPLEMENTATION COMPLETION REPORT MISSION AIDE MEMOIRE Introduction 1. A missiont from the FAO World Bank Cooperative Programme (FAO/CP) visited Malawi between 3 and 14 April in order to compile ari Implementation Completion report (ICR) for the IDA-funded components of the Agricultural Services Project (IDA Credit 2514 MAI). The mission work-ed closely with the staff of the Ministry of Agriculture and Irrigation (MOAI) and made field visits to research stations and to Lilongwe, Salima, Machinga and Blantyre ADDs. Valuable information was gathered from a meeting with ADD Programme Managers and research station heads in Lilongwe. The mission also met officials from the Ministry of Finance (MOF) and with a number of donors active in the agricultural sector. A draft Project Completion Report (PCR) and Beneficiary Impact Assessment Survey (BIAS) prepared by MOAI, were reviewed by the mission and made an important contribution to its work. 2. The mission would like to express its appreciation to the staff of MOAI at Headquarters and in the ADDs and research stations for their insights and assistance. Project Objectives and Components 3. The Project was conceived as the first phase of a longer-tenn exercise to reform and integrate research, extension and input supply services. The Project was to: (a) support institutional and management reforms in research, extension and input supply, with the aim of providing cost-effective and sustainable technologies for different types of smallholders and agro-ecological zones; and (b) enhance smallholder incomes and food security by increasing the productivity of smallholder agriculture, especially that of female and resource-poor farmers. A secondary objective was to provide a policy and operational framework for channelling and coordinating assistance to agricultural services from other donors, in order to overcome the pattern of fragmented assistance and inconsistent strategies. 4. In order to achieve these objectives, the project was to be implemented over six years, through four components: - Agricultural Research (USS15.6 million): to develop a cost-effective and sustainable research system that would generate technologies responsive to farmers' needs, especially the resource-poor and women, through providing vehicles, research equipment, training, and support for operating expenses. - Agricultural Extension (US$211.1 million): to continue sapport for MOA's national extension system, building on the experience of the IDA-supported pilot T&V initiative, in order to make service delivery more efficient and effective, and to expand coverage to smallholders previously by-passed by the extension system, by providing training, improved mobility and housing for field staff, equipment and support for operating expenses. M.L.Wales, mission leader/economist (TCIR); C.Bevan, agronomist (TCEP) J.Cook, extension specialist (consultant) 4  Input Supply (USS9 million): to support the increased availability of improved seeds and fertilizer to smallholders through supporting policy refonms and providing foreign exchange to help finance incremental fertilizer imports. Institutional Strengthening (USS9.9 million): to strengthen the capacity of MOA to deliver agricultural services in an efficient, effective and sustainable manner, in order to ensure the realisation of benefits to smallholders from extension, research and input supply systems. Overview of Project Timing, Financing and Costs 5. The project was identified in early 1988, prepared in 1991-92 with a US$700,000 Project Preparation Facility (PPF) approved in November 1990, and appraised in 1992, with a post-appraisal in late 1992. At this stage it was agreed that IFAD would parallel finance certain aspects unider the umbrella of ASP throtugh its Smallholder Food Security Project (SFSP) (Loan No. 35-SRS-MW), building upon its area-based projects in Kasungu and Dowa West RDPs; and subsequently, that ADB would co-finance certain elements. The total cost of the project was estimated at US$64.8 million, of which US$45.8 million was to be funded by IDA, US$12.7 million by ADB, and US$6.3 million by Govemment of Malawi (GOM). The IDA Development Credit Agreement (DCA) was signed in October 1993 and the project became effective three months later, in January 1994. A Mid-Term Review (MTR) was conducted in May/June 1998, at which point the prqject was de facto restnictured to accommodate support for a Starter Pack Scheme (SPS) which delivered a 0.1 ha package of hybrid maize seed, fertilizer and legume seed to practically all farm households. There were 11 IDA supervision missions. The DCA was amended xx times, and disbursements were suspended against certain categories for a period of five months, in 1998-99 due to the failure of MOAI to present an annual work plan and budget for FY 1998/99 following the MTR. The original closing date of 30 September 1999 was adhered to, except for categories 3 and 4, which were extended to 31 December 1999 to allow final procurement of seeds for SPS 11 and equipment for research and extension. The ADB loand is scheduled to close on 31 December 2000 and in November 1999 was 62%, disbursed. By end March 2000, at which point all withdrawals had been made from IDA, US$43.6 million of the IDA credit had been disbursed, that is, 94% of the total. Achievement of Ohjectives and Outputs 6. The project is provisionally rated as unsatisfactory. Although the project has made significant achievcments in policy reforms, contract researchi and improved input supply, it has not achieved its principal objective of delivering agricultural services to the smallholder sector, in particular the resource-poor. The absence of specific performance indicators over the life of the project- makes it difficult to assess to what extent smallholder productivity, incomes and food security have been enhanced. The SPS, which was supported by the project for two seasons (1998-99 and 1999/2000)), was effective in targeting poverty and household food security and was a contributory factor in the production of bumper harvests in those seasons. Overall, maize yields have increased, although it is not possible to distinguish the impact of good rainy seasons frorm project impact. There is evidence of diversification into sweet potato, Irish potato, soyabean, pigeon pea, cassava, cabbages and other vegetables. The deregulation of burley tobacco growing by smallholders. together with support targeted at burley groups by SFSP, resulted in a major expansion in the number of srnallholders growing burley from 18,000 to 50,000, and substantially enhanced incomes for these farmers. Agricultural Research 7. The objective of this component was to develop a cost-effective and sustainable research system that would generate technologies responsive to farmers' needs, especially the resource-poor and women. It was designecd to support: (a) the formulation and implementation of research action plans; (b) strengthening research-extension-farmer linkages; (c) diversifying participation in publicly-funded research and commercialisation of some research activities; (d) improving the definition of research priorities and the ADF Loan No: F/MLW/AGR/93/26 5  allocation of funding by strengthening the Department of Agricultural Research (DAR, now DARTS) and activating the Agricultural Sciences Committee (ASC); and (e) improving DAR's management systems and the motivation and accountability of researchers. The project has partially achieved its ob.jectives, especially through the contract research programme, but its overall impact has been limited by difficulties in channelling funds to scientists tinder the regular research programme. Some 30 maize varieties were produced and released, and around 30 varieties of other crops including sorghum, groundnuts, legumes, cassava and sweet potato were produced. However, it is not clear whether breeding, in particular for maize, was a sensible priority, given the many other constraints to production by small farmers. Part of the critical research work was conducted by external funding agencies including IARCs. The prqject has provided core funding for research which would have otherwise received lower levels of funding. S. Research Action Plans. Research Action Plans were successfully prepared, but there is little evidence to suggest that the annual planning procedures of DARTS have changed. The most successful activity has been the work of the Maize Productivity Task Force (MPTF), jointly funded by the project and Rockefeller and EU, which carried out simple fertilizer trials and has resulted in the compilation of a database of region-specific fertilizer recommendations, as well as other husbandry practices such as the use of legum es. The strong GOM ownership of this initiative, reflected in the fact that it received 9%, of total GOM counterpart funding of the project, was a key factor in this success. 9. Research-Extension-Farmer Linkages. Despite the best efforts of both DARTS and DAES, and the formulation of a set of guidelines, research-extension-fanner linkages remain poorly developed. Altlhouglh Research Extension Liaison Officers (RELOs) were appointed at research stations, their operational funding was always inadequate. 10. Diversification of Research Bodies. The main mechanism for diversification of participation in publicly-funded research was the contract research programme (CRP). Although this has been successful as a mechanism for channelling resources directly to scientists, questions remain about the relevance of some of the topics undertaken and its effectiveness in terms of funding research topics which address the priority needs of resource-poor farmers. During the project only 33 projects, in one batch in 1995. were approved for funding. Funds went almost exclusively to DARTS scientists, and only one private sector researcher and not a single NGO received grants. Although there was good monitoring of the projects, no research results have been translated into extension messages. 11. Research Priorities, Funding and ASC. The re-establishment of the ASC was achieved by the secondment of two DARTS officers to the National Research Council of Malawi (NRCM). ASC operated effectively during the project, developed and managed the CRP, prepared a manual on its mandate and a paper on options for a Sustainable Funding Mechanism (SFM). To date, the proposal for a general Research Fund has not proved attractive to donors. The main task of ASC, to prepare an Agriculture and Natural Resources Research Master Plan (ANRRMP), was delayed. After failing to meet several deadlines, this document was finally completed at the end of the project, in September 1999. Although it will undoubtedly serve as a useful reference document, the ANRRMP is unlikely to be used as a basis for planning and managing research because it produced an extensive list of activities rather than a stratified and prioritised list, and makes unrealistic assumptions about the availability of resources'. Although annual joint review meetings to select research projects have taken place, the constraints of time and poor attendance have meant that some of the priorities identified by extension have not been reflected in the research programme. 12. DAR Management and Staff Motivation. The overall management of DARTS has not improved and the flow of funds to scientists remains a problem. The prqject has undoubtedly augmented its operating resources, and scientists have benefited from training, even though only two of the 39 slots for short- term overseas courses were utilised. The poor flow of funds, the failure to procure most of the laboratory equipment funded under the project until almost the closing date, and lack of promotion opportunities, has left the Department with low staff morale. The long-termn TA Research Management Advisor was not recruited. Agricultural Extension The draft Agricultiral Research Strategy A.D.2000-20 10, prepared by DARTS, dated Fcbniary 2000, does not even refer to. ANRRM or CRP. 6  13. This component of the project was designed to continue stupport for MOA's national extension system, building on the experience of the IDA-supported pilot T&V initiative, in order to make service delivery more efficient and effective, and to expand coverage to smallholders previously by-passed by the extension system. The project supported measures: (a) to increase the efficiency of extension service delivery through restructtring and streamlining all operations and staffing levels, functions and capacities; (b) to expand the extension coverage of smallholders, particularly resource-poor and female farmers through separating credit administration from extension work, promoting more relevant and affordable technologies/practices, strengthening research-extension-farmer linkages, integrating technical messages. collaborating with NGOs involved in extension, upgrading all farm home assistants to FAs and training all FAs in gender-related issues, particularly nutrition and the labour constraints faced by many female-headed households; and (c) to increase the effectiveness of service delivery through improving field staff training, using more interactive teaching methods, and encouraging better quality and consistency of field supervision by providing adequate operational funding, more closely monitoring supervision and rewarding better performance. Overall, the prqject has failed to perform as expected. The main elements of T&V extension have disappeared, but there are indications that extension coverage has broadenled through the formation and traininig of farmers' clubs, especially those for burley tobacco which were supported directly by SFSP. The project has also provided core funding for the extension service which has facilitated the operation of a wide range of extension-type initiatives funded froim other sources. However, there is no evidence that women farmers have benefited preferentially from extension services. 14. Management and Operations. There is little indication of any improvements in efficiency or effectiveness of extension delivery. Whilst the de-linking of extension from credit has been achieved, FAs are still engaged in the registration and training of groups for the purpose of credit delivery through MRFC. This continues to bias extension delivery somewhat towards farmers in credit groups. Although this tendency diminished towards the end of the project as alternative extension methodologies were adopted, it has been replaced by a bias towards villages where donor-funded initiatives have started. The Block Extension System (BES). a modified T&V approach, has virtually collapsed as a result of inadequate staff numbers and operating funds at the field level. The BES suffered from low fanner attendance due to an absence of new and affordable messages, and inadequate supervision. Although most SMS posts were filled, many of the staff were not suitably qualified. This, together witlh a chronic shortage of operating funds at ADDs. RDPs and EPAs, resulted in poor quality of supervision and technical backstopping, with little or no support from Headquarters. Extension planning and management at the ADDs has not improved despite the introduction of a log frame approach, with extension work plans and budgets far in excess of the actual resources available. 15. Extension Coverage and Content of Messages. The ratio of Front Line Staff (FLS) to farm households has deteriorated from 1:980 shortly after the start of the project in 1995. to 1: 1700 at the end of the project in 1999. This has been the result of a failure to recruit new staff as a result of fonnal and infornal recruitment freezes, retirements, an increasing rate of "deaths in service'", in part the result of AIDS, and the fact that NRC no longer offers courses to train FAs. Veterinarv and animal husbandry field staff, who were to have been funded by ADB, found themselves with extremely limited operating resources as a result of slow disbursement against the ADB loan. On-Farm Demonstrations (OFDs) became a priority activity and expanded enormously in number from 15,00() to almost 160,000, but this, together wvith an increase in the number of clubs to service, has over-stretched the FAs, and the quality of supervision has deteriorated. There was considerable success in propagating low-cost technologies, soil conservation and fertility management, agro-forestry and on-farm root crop multiplication, through SFSP and programmes funded by EU and USAID. The transition to a unified extension system, in which the FA is responsible for all fanner contacts, including livestock production activities, appears to have resulted in even less attention to livestock extension than before. There was a large expansion in the number of farner clubs based on specific crops or livestock activities, as well as women's groups, and others organised through churches or mosques. The xvomen's programme was, however, focused almost exclusively on food and nutrition. The support provided by the project has allowed the extension service to serve as the vehicle for various extension-type initiatives such as the VLPA (World Bank), PROSCARP (EU), MAFEP (USAID), APIP. FFS, and SG200(0, as well as the activities of a large number of NGOs, all of which are based on a large degree of fanner participation and responsiveness to fanner priorities. The ultimate task in this respect was the successful distribution of Starter For example, it is reported by the Human Resources Department that there were 60 deaths-in-service during 1999 in Machinga ADD. and 43 in Lilongwe ADD. 7  Packs to some 2.8 million farm households in 1998-99 and 1999-2000. in which FAs played a crmcial role in registering recipients. 16. Training. Capacity-building appears to have had little impact, with all training targets substantially under-achieved, especially for field level staff. Funds for training in innovative extension approaches, such as PEM, were exhausted at senior levels and few if any FAs benefited from this. A total of 17 staff undertook M.Sc courses overseas and six attended short courses in media and extension management. A limited number of FHAs received training to become FAs, and courses were held to upgrade FAs to the supervisory TO level, and to provide them with the necessary skills in animal production. The Food and Nutrition (F&N) subcomponent was delayed by the late arrival of the TA expert, but there were substantial achievements in enabling three staff to obtain M.Sc degrees, and conducting a training needs assessment throughout the country. A Community Nutrition and Training Manual was finally prepared, has been printed but not yet distributed and no training associated with it has been conducted. The programme at the field level has been handicapped by the fact that there is often only one staff member to undertake both F&N and women's programmes. A major lacuna in knowledge has been the failurc of Bunda College to finalise the socio-economic baseline survey. 17. Improved Access to Transport. The motorcycle ownership scheme was successfully implemented with supervisors receiving motorcycles. However, the positive impact that this should have had on supervision was negated by the fact that there were never sufficient funds to settle transport and travel (T&T) claims. 18. Civil Works. A programme of civil works (total cost US$5.9 million) aimed primarily at the extension service, was substantially not implemented. The original proposal for houses for FAs, and four EPA centres, was modified to accommodate rehabilitation of existing FA houses and construction of two RDP centres. However, despite these works being designed and tendered, the works were not undertaken, which has had an adverse impact upon field staff morale. The failure of GOM to supply counterpart ftnding for the civil works programme was undoubtedly part of the reason why it was not implemented. 19. Extension Aids Branch. The Agricultural Communications Branch (ACB) (formerly the Extension Aids Branch) was to have been strengthened in support of the overall extension programme, in particular, with equipment to modernise the Visual Aids Section for the use of radio to reach resource-poor farmers in listening groups. ACB received vehicles and domestic training and. towards the end of the project, equipment. Two staff had long-term training on DTP and video production. Radio programmes and visual aids produced by ACB have a significant impact in the field, even though its operations are constrained by lack of resources. The Functional Review proposed abolishing ACB. Input Supply 20. The objective of this component was to support the increased availability of improved seeds and fertilizer to smallholders and to expand use of these inputs by smallholders. The fertilizer subcomponent was to: (a) support the formulation and implementation of reforms to further liberalise importation, pricing and distribution; (b) improve donor coordination; and (c) provide foreign exchange to help finance incremental fertilizer imports on a commercial basis. 21. Policy Reforms. Although the agreed policy reforms have been largely implemented, this component has not achieved its objectives because the overall impact on the availability of inputs to small farmers, and competitiveness in supply, has been negligible. The reform with most far-reaching implications was the amendment to the Special Crops Act, which allowed smallholders to begin growing burley tobacco. The deregulation of fertilizer imports has been achieved, and at one stage there were nine local and international companies operating. Subsidies, which started to be reduced from 1991/92, were completely removed in 1993/94. The Smallholder Farmer Fertilizer Revolving Fund for Malawi (SFFRFM) operated commercially, selling to farmers and to local companies, until the start of SPS. It manages the fertilizer buffer stock which has been practically depleted as a result of SPS I and II. Towards the end of the project, the number of private fertilizer companies shrank to two, inevitably reducing competition. There has been no success in reducing the cost of fertilizer imports to farmers, due to long haulage distances, poor roads and high freight rates, and the problem of late delivery still exists. For hybrid maize seed. the objective of liberalising 8  production and marketing of seed to increase competition was achieved. witli two commercial companies producing or importing all requirements. All subsidics on improved seed were removed in 1994/95. There remains a substantial agenda of second generation reforms to be undertaken, such as the privatisation of ADMARC, which remains a parastatal, controls 50% or more of the fertilizer mark-et and, in remote areas, is the only supplier. The role and sustainability of the maize price band, introduced after the 1994 drought in order to protect consumers, must also be reviewed. 22. Improved Maize Seed. Towards the end of the project, there was a substantial increase in the use of hybrid maize which, according to MOAI estimates, now accounts for as much as 40)% of the planted area (1.5 million ha in 1999/2000). However, there are indications that substantial numbers of farmers are using recycled seed (F2). The expansion in the use of hybrids has been achieved by the SPS (around 280,000 ha), APIP (250,000 ha), as well as the fact that hybrids are now more widely available. Supplies of all other types of improved seed, which it is not attractive to commercial seed companies to produce, are also increasing through farmer multiplication schemes operated by EU and various NGOs. After the MTR, the project financed the purchase of hybrid maize and groundnut seed, and logistics, as part of SPS I and II, utilising around US$6 million of the funds remaining in the credit. 23. Fertilizer. IDA financed the importation of 12,500 t of fertilizer, through loans to SEDOM and INDEFUND through the Reserve Bank of Malawi for private sector traders. ADB procured 19,877 t of fertilizer which was used, without ADB approval, for SPS 1, and led to the suspcnsion of the ADB loan. Institutional Strengthening 24. The objective of this component was to strengthen the capacitv of MOA (niow MOAI) to deliver agricultural services in an efficient, effective and sustainable manner, in order to ensure the realisation of benefits to smallholders from extension, research and input supply systems. The subcomponents are: (a) planning and financial management in MOA/ADDs; (b) transport management; (c) procurement; (d) agricultural policy analysis training; (e) small-scale irrigation (funded by ADB and IFAD); and (t) sectoral and project monitoring and evaluation. The project did not achieve its institutional strengthening objectives, a failure which proved to be critical to the performance of the project as a whole. 25. Planning and Financial Management. Although some progress has been made in financial management, including the installation of a computerised system (Finasys) for Revenue Account, work plans and budgets, and financial reports are still delayed, and little training was conducted. However, the most critical constraint to the achievement of the prqject's objectives has been the inability, throughout the duration of the project, to channel resources to field level staff and research scientists on a secure and regular basis. This problem has been compounded by the lack of realistic work plans and budgets at the ADD and RDIP levels, and by DARTS. The logical framework approach has been adopted at all levels as part of the Medium Term Expenditure Framework (MTEF) exercise started by the Ministrv of Finance with WB assistance. However, the process of prioritising activities remains a problem, with budget proposals grossly in excess of available resources. 26. Transport Management System. The provision of increased mobility to staff at all levels in support of extension delivery, was a crucial part of the prqject's design. Although over 20 4x4 vehicles, 440 motorcycles and 865 bicycles were procured, a transport management system was not put in place and training was not conducted. However, the private sector now carries out almost all maintenance. 27. Procurement'. Although in the SAR, procurement was identified as 'The most intractable constraint on timely project execution", the project has failed to improve the situation. On the basis of the MTR assessment, field visits and the experience of ADDs, it appears that procuremenit was invariably slow and erratic2, often wasteful through the purchase of inappropriate goods. and that items once procured were often misallocated. Responsibilitv for the procurement study was shifted betveen several Ministries and was It proved (dit ficult to obtain information on procurement since the TA officer had moved to a different project and was not in the countrv. 2 Laboratory equipment for research stations was not delivered until the final year. Vehicles X ere mot delivered for two years. Thie bulk ot'the civil w%jorks programme was not implemented. 9  eventually completed (under OPC) at the end of 1999. Little or no training was carried out, leaving the Procurement Section at the end of the project without operational capacity. 28. Agricultural Policy Analysis Training..The project supported the Agricultural Policy Analysis Training Unit (APATU) at Bunda College of Agriculture, to strengthen MOAI policy analysis capacity. Procurement of equipment, and infrastructure development, including a 30-bed hostel, were substantially implemented. It has successfully conducted around five, 2-3 week courses per yearl. against a target of nine, and has served a range of agencies within government and NGOs. It has successfully attracted supplementary funding from outside. The main problem has been the slow disbursement of funds, which has limited the number of short courses that can be conducted, and inadequate support from the institution which was contracted to provide technical assistance support. The main failure has been the inability to attract candidates for the M.Sc degree programme in agricultural economics, extension and rural sociology, with only one person of the targeted 30 receiving a degree. The course itself is perceived to be less attractive than one undertaken at an overseas institution, there are too few teaching staff, it takes too long to complete, and involves no intemational exposure. Six students enrolled in 1999 but, with the closure of the prqoect, they have not been able to complete the degree course. 29. Training. A total of 5 Ph.D., 33M.Sc., and 30 diploma and short courses were funded by the project. with the majority of the beneficiaries being from Headquarters. This was far below the targeted training programme, largely due to the decision to unidertake a training needs assessment, which resulted in a two-year hiatus in the programme. The report was not completed until the end of the project2. At the MTR it was agrced that only field level training should take place, but even this was not carried out. 30. Technical Assistance. The project was to have provided 410 person-months of TA in 18 different areas, of which 144 p-m would be long-term and 298 p-m intemational, including 100 p-m for the procurement study. Only half of these slots were filled, with critical gaps in research management, extension management training, agricultural policy, M&E, livestock extension and financial systems. There has been dissatisfaction in MOAI with the quality and output of much of the intemational TA provided under the project. There have been difficulties in obtaining progress and final reports, unduily strict adherence to terms of reference, recruitmnent of inappropriately qualified personnel, failure to appoint counterparts and, in the case of the TA support to Bunda College, a strong sense that they have not received value for money. 31. Monitoring and Evaluation. A major objective of the institutional strengthening component was to improve the M&E function at all levels in the Ministry. Whilst there has been some progress towards this goal, throughout the project, M&E was unsatisfactory. Initially there was inadequate staffing, a problem not resolved until the end of the project through recruitment and training. Although the annual work programming process improved with the introduction of the log frame, implementation has remained poor. Perfonnance indicators were not formulated until end 1998 and have still not been cndorsed by MOAI, and work on improved agricultural data collection, based on the Area Sampling Frame (ASF), did not proceed beyond the pilot stage. Guidelines for ASP reporting were never issued and the submission of M&E reports was chronically behind schedule almost throughout the project. Overall. there is still no effective svstem in place to monitor and evaluate the progress of agricultural development in Malawi, and no capacity to integrate M&E management information into the strategic and annual work planning processes. 32. MIPA. Support for the Agribusiness Advisory Unit (AAU) in the Malawi Investmnent Promotion Agency (MIPA) was included in the project following the Joint Review mission in 1994. The Unit, to strengthen the capacity to promote local and foreign investment, disseminates agribusiness technical and market information, and advises entrepreneurs on financial opportunities and accessing advisory services, started in 1996. AAU staff successfully completed a range of induction and specialist training courses, although on-the-job training by TA. did not take place. AAU substantially achieved its objectives, and especially the creation of the Domestic Investment Promotion Departmnent. However. it was hampered by slow flow of finds from MOAI, and failure to procure the necessary equipment for its Business Information Centre. This was reduced to 2-3 per year in 1998 and just one in 1999 due to lack ol resources 2 Trtining Needs Assessment Report, by Deloitte and Touche, dated xxx 11)  33. Project Management. Overall project management, which was the responsibility of MOAI, with CMEU responsible for day-to-day activities, remained weak throughout the project. The principal mechanism for implementation coordination was the Agricultural Sector Steering Committee (ASSC) which was scheduled to meet quarterly, but in fact only met twice during the project life. An ASP Facilitator (on contract) was overwhelmed with work, remained in post for only one year and was never replaced. Despite tNwo start-up workshops, detailed knowledge of the project's objectives was not widespread. There were no detailed cost tables in the SAR and the Implementation Manual was not widely available so that many subcomponent managers were not clear as to what assets they were supposed to acquire. Management of individual components by different departments was generally inadequate, although ADDs generally fared well under difficult circumstances. Major Factors Affecting Implementation and Outcome 34. The main factors which positively affected project inipleinentation were: - political stability, with the transition to democratic systems achieved comparatively smoothly; - the commitment and dedication of field level staff operating under difficult circumstances and with inadequate resources; - policy reforms whiclh liberalised trade in fertilizer and seeds: - expanded access to improved seed and fertilizer under SPS I and 11, - FAs released from the task of recovering credit; - expansion in the number of farmers' groups and clubs, especially those unrelated to credit delivery, serving as a focus for extension delivery; and - expansion in the number and level of activity of NGOs serving farmers, often involving field level staff in their programmes and introducing participatory approaches. 11  35. The main factors which adversely affected project implementation were: - macro-economic instability with high inflation. devaluation of the Kwacha and MRFC lending rates reaching a high of 50%, which had a negative impact upon the profitability of agriculture using imported inputs and the viability of credit; - the virtual collapse of the smallholder credit system in 1994 due to high interest rates and poor repayments following the drought; - a major drought in 1994 which diverted attention away from routine project implementation; - failure to devise a workable mechanism for the prompt reimbursement of claims between World Bank, MOF, MOAI Headquarters, and the ADDs, whiclh resulted in chronic under- resourcing of the field extension and research programmes: - high level of staff vacancies, especially at field level, due to formal and infonnal freezes on recruitment as part of the civil service reform and latterly. increasing losses due to retirement and deaths-in-service, partly due to AIDS: - low staff morale, especially at the field level, resulting from inadequate formal and in- service training, lack of resources for field activities, failure to pay T&T claims. failure to rehabilitate FA housing, and career uncertainty under restructuring; and - undue retention of project operating resources at Headquarters through high levels of travel and subsistence claims, and preferential access to training opportunities, overseas study tours, vehicles and equipment. Project Costs and Financing 36. The project formally closed on September 30, 1999, as scheduled. However, categories 3 and 4 remained open until December 31 1999 in order to allow reimbursement of final claims for procurements in connection with SPS, and for equipment for research to be processed. Total disbursements were SDR 30.98 million (US$43.62 million), which represents 94%, utilisation of the credit. There was a substantial disbursement lag throughout the life of the project, which was finally 24'S,. Although GOM counterpart contribution to the project was set at 10% of total project costs (US$6.3 million), after the MTR this was reduced to 5%. Actual GOM counterpart contributions (Development Budget Part 11) amounted to US$0.72 million, as compared to the US$6.3 million envisaged, revised to US$3.2 million. Research received 50% of the counterpart funding, procurement 25%, and the ADDs 10%. ADB co-financing of the project, which is due to close on December 31, 2000, amounting to UA 9.21 million (US$12.7 million), was only 62% drawn down (UA 5.68 million, equivalent to US$7.87 million) by November 1999. The decision to extend the ADF loan had not been taken by April 2000. Project Sustainability 37. Although few aspects of the project can be expected to be sustainable. in some cases, such as extension, changes in methodology and approach make this unimportant. Following closure of ASP, there has been a major cut in the resources available to ADDs and field level staff for extension, with operations dependent upon small flows of ORT and funding of specific activities by bilateral programmes. The motorcycle ownership scheme cannot be sustained because reflows into the revolving fund are inadequate to purchase new machines. The regular research programme under DARTS has been similarly affected and there are no sources of funding for CRP. Scarcity of funds will undoubtedly result in the cessation of even the limited research-extension-farmer linkages. The continued operation of ASC is doubtful since the senior staff seconded to it has returned to the Ministry. MOAI itself is in a weak position to face the challenges of the future in the absence of functioning M&E, financial, procurement and transport management systems. Although MIPA and the AAU are in a poor financial position and have been adversely affected by the 12  cessation of funds under the project, the sustainability of AAU's activities, which are generally recognised as being valuable, could be guaranteed through a proposal to raise funds from a levy on certain commodities. The continuance of the APATU programme is likely through tailoring courses to the needs of various donors, especially under the SADEC Regional Food Security Training Programme, and offering more specialised courses focusing on agribusiness for graduates. Overall, the fact that there is no World Bank-funded project in the pipeline for the agricultural sector limits the sustainability of MOAI activities. Bank and Borrower's Performance Bank Performance 38. Overall. the Bank's performance has been unsatisfactory. The preject's objectives were appropriate, focusing on research and extension, and on institutional strengthening aimed at bringing effective agricultural services to all small farmers, especially the resource-poor and women. However, in retrospect, a top-down, message delivery extension system was not likely to be effective and was certainly not fiscally sustainable. It tried to build upon the lessons of earlier projects, especially prqjects under the National Rural Development Programme (NRDP), the Agricultural Extension and Planning Support Project (AEPSP), and the National Agricultural Research Project (NARP). In particular, the conclusion that earlier prqjects had primarily benefited the better-off farmers, underlay the design. However, in a laudable attempt to allow flexibility in implementation, the design proved to be insufficiently prescriptive to facilitate implementation. In part, this was reflected in concerns expressed at the time that the project was too complex for a weak MOA to implement, and that because Government had played little part in the preparation of the project, it showed few signs of "ownership". The design, which created a project with national coverage as the successor to the 30 Rural Development Projects (RDPs) which comprised NRDP, financed by several different donors, was intrinsically more complex than the RDPs and more demanding of management. In particular, there were concems that the centralisation of control over funds that would result might be inimical to responsiveness in the ADD extension services. 39. There were three successive Bank Team Leaders during the project cycle, and a large number of staff weeks were devoted to supervision of the prqject. Supervision missions were, for the most part, regular, and identified actLal and impending implementation problems. However. although institutional strengthening was a critical part of the project, an institutional specialist was never included in the missions. Despite successive "unsatisfactory" ratings for the extension component, it appeared to be impossible to remedy the problems. Fundamnental to all implementation problems was the inability, from the beginning to the end of the project, to resolve the problem of flow of funds to the field (ADDs and beyond) which resulted from inadequate and late delivery of GOM counterpart funds, and difficulties with the processing of reimbursement claims. Guidelines on the use of the Special Account were issued late and frequent modifications were made to procedures, right up to the end of the project. Although an attempt was made at the MTR to reorient the project, none of the action plan proposals was implemented and it was only at this stage that the PDO rating was downgraded to "Unsatisfactory". Consequently, at the end of 1998, the prqject was declared to be "at risk", disbursements against operating expenditures were suspended for five months. and it was decided not to extend the prqject closing date, even though some US$17.5 million of the credit remained undisbursed. At this poiiit, when it was deemed that none of the project's development objectives was likely to be achieved, the Bank agreed to utilise almost all remaining funds to finance seeds and logistic costs associated with first, SPS 1, and subsequently SPS II. Borrower's Performance 401. Overall, the borrower's performance has been unsatisfactory. Althlough there was probably commitment to the project in the early stages, the fundamental lack of ownership resulted in increasinglyad hoc management and an inability to take corrective action when problems started to emerge. Procedural problems, starting at the field level and stretching to MOF with respect to the timely submission of witlhdrawal applications, and reimbursement of claims by departnents and ADDs, led to a chronic problem of channelling funds to the field. It took two years before the Special Account was allocated to cost centres, and the procedures for reimbursement via MOF were complex and slow, all of which undermined the entire prqject implementation. Non-salary operating costs were provided late and below the required level, even though they 13  were 1 00%/o reimbursed by the project. Failure to act decisively wheni efforts to improve institutional capacity in financial and transport management, procurement and M&E were failing. meant that the delivery of improved agricultural services could not be achieved. Management capacity in the Miiistry was weakened as other donor projects attracted senior staff to join special PlUs. Although piecemeal introduction of altemative extension approaches (such as PEM, SEC, KAP) without associated training, as well as programmes funded by other donors (such as PROSCARP, MAFEP, APIP) caused confusion and diverted efforts at the field level, it actually opened the door to new participative extension methods. It appears that indecisive management and poor procurement procedures resulted in the almost complete failure to implement the civil works programme (10% of original category). The supervision of TA, including failure to appoint counterparts and react decisively to problems, was also weak. For a period, audit reports were submitted late, and with qualifications. 41. An attempt was made at the MTR to reorient the project by focusing on training at the field level, on strengthening M&E, research and extension, and on activities to test altemative approaches to extension delivery and management on a pilot scale. None of the proposed activities wvas implemented. Instead, ali project activities were abandoned in favour of SPS. Lessons Learned 42. The principal lessons that can be drawn from the implementation of ASP are: General - The participation of all stakeholders in the conceptialisation and design of a project are fundamental to securing full commitment to implementation. - The problems arising from complex project design are compounded by the involvement of co-financing. - Innovative project design must include a realistic assessment of management capacity, and ambitious institutional strengthening (especially in financial management, procurement and M&E) cannot be carried out during operational implementation. - Although centralised procurement may apparently offer benefits of efficiency, this is offset by the difficulties of consulting end users in the process in order to ensure that the correct items, appropriately specified, are procured and allocated. - Decentralisation of project activities, including the channelling of operating funds directly to the lowest possible level in the administrative hierarchy, together with capacity to manage them and accountability for their use, is more effective than central administrative control; the management structure placed excessive demands upon the CMEU, which suffered high staff turnoverl and weak technical capacity. - The process of decentralisation, the changing demands upon MOAI, and the need to ensure sustainability of service, indicates the need for substantial restructuring of the Ministry to reflect clearly-defined core functions. - Investments aimed solely at increasing productivity are unlikely to yield enhanced farmer incomes unless market development-related issues are addressed through improving access to markets, promoting farner associations, and developing market linkages for inputs and outputs by providing a conducive environment for the private sector. Policy There were no less than 7 different Chiefs/Controllers of Planning during the course of the project 14  The policy reforms carried out under ASP alone are not sufficient to yield significant benefits to the sector. Second generation reforms, such as the privatisation of Admarc, changing the maize pricing policy', defining the role of the National Food Reserve Agency, the tobacco Intermediate Buyer system, resolving the problem of rural finance, and addressing this issues of land policy. Research - Whilst securing funding for scientists is important (as under CRP), it does not necessarily ensure the relevance of the work carried out. The involvement of NGOs and the private sector in research has to be actively sought and the areas for public- and private-funded research clearly defined. More rigorous screening of research proposals is needed, and proposals for agreed research topic priorities should be solicited, possibly through a competitive grants system. - Research-extension-farmer linkages are unlikely to be strengthened unless farmers participate more actively in developing the research agenda. - A Research Master Plan is of little value unless it is produced ahead of research funding and represents the consensus of the main stakeholders Extension - Uniform national coverage of a routine, visit-based, message delivery extension system (such as BES) is neither effective nor fiscally sustainable'. - New participative extension approaches require longer term testing to assess sustainability. In the mean time, a plurality of extension approaches is likely to be a better course of action than adopting any single approach. - Participatory approaches, which empower farmers' groups and give them greater responsibility for deciding what their priorities are and to choose the level of technology which suits them best, are most efficient and cost-effective. - In the interests of sustainability and affordability, a clear definition of publicly-funded extension priorities is essential, together with partnerships between govemment, NGOs and the private sector in meeting the information needs of fanners. Bank - Effective project supervision is a crucial part of implementation. Additional resources should be allocated to the supervision of projects in which implementation deviates substantially from the plan, in particular providing skills to address institutional issues. However, excessive "micro-management" which debilitates management of the project by the borrower, and diverts attention from the resolution of major issues, should be avoided. - Although the insertion of a project within an existilg institutional structure (MOAI) avoids dependence upon a PIU, this can only work if the institution itself is strong and all parties involved in implementation feel ownership and are fully familiar with its objectives. In this context, start-up workshops, wide distribution of implementation manuals and cost tables, and regular review workshops are vital, especially if staff turnover is high. The use of coordinating mechanisms such as steering committees comprising those directly involved in implementation, should be rigorously applied. The operation of the mEaze price band cost governuent USS 19.5 million in 1999. 15  Effective flow of project resources to the fieid level is a sine qua non for successful project implementation. This must be achieved through a combination of simplifying procedures, strengthening financial management, and designing projects in which responsibility is devolved to the lowest level possible. 16   lORD 29374 To Dour Es tidoio '0 ID~~~~~~~ r x~~~~P 0e"-'\ E f#~MAAWI rhbrnopwosproourso mt 5ed , ADMINISTRATIVE DIVISIONS Th on'r4colors, X-1 dnomtinfotioneand any REGIONS AND DISTRICTS othe informato ahow impton thi t of RUMPH) Th. World Bonk Group, ony iudgmnt on the ego(- startue of any territory Ktm or any endorement Rum - PRiNCIPAL ROADS or acptance of suchPPNIAROD bouncories. ,Ytr- - RIVERS aderoEI;nd ni Lak XeS \