Malawi - National Agricultural Research Project

Doamnmt of The World Bank FOR OFICLIAL USE ONLY Report No. 13634 PROJECT COMPLETION REPORT MALAWI NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 1549-MAI) OCTOBER 25, 1994 Agriculture and Environment Division South Africa Department Africa Regional Office This document has a restricted distribution and may be used bv recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS 1985USS 1.0 = MK 1.72 1986USS 1.0 = MK 1.86 1987USS 1.0 = MK 2.21 1988USS 1.0 = MK 2.56 1989USS 1.0 = MK 2.76 1990USS 1.0 = MK 2.73 1991USS 1.0 = MK 2.80 1992US$ 1.0 = MK 3.60 1993US$ 1.0 = MK 4.39 (Av. Jan. -Sept. 1993) WEIGHTS AND MEASURES Metric System GOVERNMENT FISCAL YEAR April 1 - March 31 ABBREVIATIONS ADD Agricultural Development Division AGREDAT Agricultural Economics, Statistics and Data Processing Unit ARC Agricultural Research Council ART Adaptive Research Team ASP Agricultural Services Project CARO Chief Agricultural Research Officer CIMMYT Wheat and Maize Improvement Centre CTL Commodity Team Leaders DAR Department of Agricultural Research DCA Development Credit Agreement DCARO Deputy Chief Agricultural Research Officer DEVPOL Development Policy FAO/CP FAO/World Bank Cooperative Programme GOM Government of Malawi ICRISAT International Crop Research Institute for the Semi-Arid Tropics IFAD International Fund for Agricultural Development IITA International Institute for Tropical Agriculture INIBAP International Institute for Banana and Plantain Research ISNAR International Service for National Agricultural Research LLDP Lilongwe Land Development Programme MAEPS Malawi Agricultural Extension and Planning Support Project MARE Malawi Agricultural Research and Extension Project MOA Ministry of Agriculture NARP National Agricultural Research Project NRC National Research Coordinators NRDP National Rural Development Programme PCR Project Completion Report RMEA Regional Mission in Eastem Africa SACA Smallholder Agricultural Credit Administration SAR Staff Appraisal Report SFFRFM Smallholder Farmers' Fertilizer Revolving Fund of Malawi TA Technical Assistance TRFCA Tea Research Foundation of Central Africa USAID United States Agency for Intemational Development  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation October 25, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Malawi - National Agricultural Research Project (Cr. 1549-MAI) Attached is the Project Completion Report on Malawi National Agricultural Research Project (Cr. 1549-MAI) prepared by the Africa Regional Office. Part II was contributed by the Borrower and is in general agreement with Parts I and III of the Report. The project was the first phase of a longer term effort to develop the agricultural research institutional framework and capacity to provide relevant technology to Malawi's smallholders. The organizational structure was modified along commodity program lines; Research Master Plans and Action Plans were developed; annual review and planning procedures were improved; external evaluations made useful contributions; collaboration with international centers was enhanced; research infrastructure and facilities were improved and rationalized; and substantial training was carried out. The project contributed to significant achievements of the research system in developing hybrid flint maize valued by smallholders, in promoting the use of high-content fertilizers, and in the biological control of the cassava mealy bug. However, serious implementation problems were incurred, including the inability to improve terms of service for researchers, a 30 percent loss of researchers trained under the project, poor management of procurement and planning of facilities resulting in delayed and inappropriate purchases, ineffective adaptive research team activities which constrained farming system diagnostic programs, and a chronic shortage of operating funds for the system. On balance, the project is rated as having a satisfactory outcome with a modest institutional impact. Sustainability of the institutional achievement is uncertain, however, due to the continued budgetary constraints and the lack of a career structure which would promote the retention and encouragement of trained and experienced researchers. The Completion Report provides a good and comprehensive record of project performance. An individual project audit is not planned, but will likely be undertaken in conjunction with the parallel extension project (Cr. 1626-MAI) and possibly with the ongoing Agricultural Services Project following its completion. Attachment 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.   FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT MALAWI NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 1549-MAI) TABLE OF CONTENTS Page No. PREFACE ........................................................ i EVALUATION SUMMARY ........................................... iii PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE ................. 1 1. Project Identity ............................................... 1 2. Background .1 Introduction .1 The Agricultural Sector .1 The Department of Agricultural Research. 2 The Agricultural Development Policy. 2 3. Project Objectives and Description ............................... 2 4. Project Design and Organization. 3 5. Project Implementation .......................................... 3 Project Costs and Disbursement .............................. 10 6. Project Results ............................................... 10 7. Project Sustainabilityl.1 8. Bank Performance ............................................. 11 9. Borrower's Performance ......................................... 12 10. Lessons Learned .............................................. 12 11. Project Relationship ............................................ 14 12. Consultancy Services ........................................... 14 13. Project Documentation .......................................... 14 PART II: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE ............. 15 PART III: STATISTICAL INFORMATION ................................. 24 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.  Page No. TABLES 1. Related Bank Loans and/or Credits .................................. 24 2. Credit Timetable . ............................................. 25 3. Credit Disbursements . .......................................... 26 4. Project Implementation ........................................... 27 5. Project Costs and Financing . ...................................... 30 6. Project Results ............................................... 33 7. Status of Covenants ............................................. 34 8. Use of Bank Resources ........................................... 36 ANNEXES 1. Seasonal Credit Recovery for 1992/93 Season as on 03/12/93 ................. 38 2. Switch of Fertilizer Mix: Estimated Savings in Fertilizer Import Expenditures in 1992/93 39 3. Civil Works .............................................. . 40  PROJECT COMPLETION REPORT MALAWI NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 1549-MAI) PREFACE This is the Project Completion Report (PCR) for the Malawi National Agricultural Research Project (NARP) for which Credit 1549-MAI in the amount of SDR 24.3 million (US$ 23.8 million) was approved by IDA on 19 February 1985. The credit became effective on 19 November 1985. Originally expected to be implemented over a five-year period the credit was actually closed, after three one-year extensions on 31 October 1993. Actual disbursement totalled SDR 24.0 million (US$ 32.1 million)' or approximately 99%, in SDR terms, of the originally approved credit amount. The last disbursement was made on March 14, 1994, and SDR 230,200 was cancelled. Parts I and III of the PCR were prepared by the FAO/World Bank Cooperative Programme (FAO/CP) for the Agriculture and Environment Division of the Southern Africa Department (AF6AE) of the World Bank. Part II was prepared by the Borrower. Preparation of the PCR was started during a visit to Malawi by a team2 from FAO/CP in December 1993, and is based, inter alia, on information gathered in Malawi discussions with project staff and officials of the Department of Agricultural Research within the Ministry of Agriculture and with staff of USAID which participated in project co-financing. The report also takes into account information and data gathered from the Staff Appraisal Report (SAR), the Development Credit Agreement (DCA), Supervision Reports, USAID Agricultural Research and Extension Project (MARE) final evaluation report, correspondence between IDA and the Borrower and internal IDA memoranda. Excludes the revolving fund amount of SDR 67,032 (US$ 89,562) 2 Messrs. R. Suppa, Economist and Mission Leader; D. Corbett, Agricultural Research Specialist (Consultant). i   MALAWI NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 1549-MA) EVALUATION SUIMMRY Objectives The project's general objectives were to provide the first phase of long-term improvements in the research capabilities of the Department of Agricultural Research (DAR), by focusing on the infrastructure and planning, management procedures, and the skill levels of its staff. The main features of the project were: (i) the creation of an Agricultural Research Council (ARC) to advise on research priorities; (ii) the rationalization of DAR's research station network and the relocation of staff; (iii) the reorganization of DAR's research staff into multi-disciplinary commodity research teams and the introduction of a programme and budgeting system; (iv) the introduction of a technical career stream, staff evaluation system, training needs identification procedures, and the provision of training and technical assistance for DAR staff; (v) increased collaboration with International Agricultural Research Centres and Regional Research Institutes; (vi) improvement in research/ extension links; and (vii) improved economy and efficiency in ongoing research activities. Implementation Experience T'he project was originally scheduled to be completed by 1989/90 after an implementation period of five years. This proved to be too ambitious as actual implementation ended in 1993 after three one-year extensions of the original closing date of 31 October 1990. The original credit agreement was amended twice in 1992, first to reallocate SDR 0.6 million to non-salary operating expenses and the second was to enable SDR 6.0 million to be used for smallholder agricultural credit to provide funds for a drought recovery scheme. ARC was established by the Ministry of Agriculture (MOA) internal memorandum in 1985 with a supporting Secretariat in DAR. Effectiveness of ARC was severely compromised as it is regarded as an internal committee, without legal standing and little status. Progress in infrastructure consolidation and improvement was satisfactory throughout the project. A considerable civil works program was undertaken, including the building of a new research facility at Mkondezi. The main research library at Chitedze was provided with a new building and equipment and smaller libraries were revitalized at Bvumbwe and Lunyangwa. Data processing capacity was improved in the Agricultural Economics, Statistics and Data Processing Unit. Relocation, which affected principally professional and some support staff, was severely impeded by the training programme, with many scientists overseas on post-graduate courses. Procurement performance throughout the project was poor; bid document preparation was sub-standard and led to delays, with mismanagement compounding the problem. DAR headquarters, within MOA, was reorganized, with two Deputy Chief Agricultural Research Officers (DCARO) assigned to support the Chief Agricultural Research Officer (CARO). While the DCARO for Research Programmes was made responsible for the planning and conduct of research programmes through National Research Coordinators and Commodity Research Team Leaders, the DCARO for Technical and Administrative Services was made responsible, through Station Administrative Officers, for all operations at research stations. However, the organizational iii  structure developed by the project was complex and exacerbated administrative pressure on senior staff from the CARO downward. A Programme and Budgeting system essential for managing the research was planned, but not developed. The expatriate consultant hired to undertake the task left, for personal reasons, before the end of the contract, and the work he had done did not comply with the project requirements. The CARO did not place high priority on a replacement, and Bank supervision missions failed to monitor the work of the consultant prior to his departure or to insist on the employment of a suitable replacement. Enhanced staff conditions of service, to improve motivation and retain returning post- graduate trainees in the research service, were to be introduced as part of a new scientific career stream, but opposition from GOM prevented this. GOM's opposition was because of the budget implications and the concern that a new career stream would isolate scientists from other civil service career streams. A large training component, co-financed by USAID, led to post-graduate training and much short-term training in-country and overseas. However, this meant that during project implementation large numbers of staff were overseas and several research programmes were impeded. The considerable technical assistance component in the project provided four key long-term TA positions and many other shorter term positions. Their performance was variable with few leaving fully trained and competent counterparts in post when they left. Adaptive Research Teams (ART) composed of an economist and an agronomist were appointed to all eight Agricultural Development Divisions (ADDs) in the country to improve research/extension links, by conducting on-farm trials and identifing socioeconomic constraints affecting small farm productivity. The ARTs did not function effectively, and contributed very little to strengthening the research-extension linkage as their links with DAR were extremely weak. It was found difficult to retain staff, especially economists, and at the end of the project only one ADD had a full ART in place. A few agronomic trials were conducted but the results have never been disseminated. Overall, project design provided very little focus on Malawi's research problems. Results In spite of the implementation problems as described above, it can be stated that the overall project performance was satisfactory, and most of the original objectives were achieved. A Research Master Plan was prepared and now provides strategic direction, Action Plans were prepared to provide tactical guidance to scientists preparing their Annual Work Plans, and a structured and systematic procedure was instituted for annual research planning, following Annual Reviews of the previous season's work. External evaluation of the research and its management was introduced and two Triennial Research Reviews, held in 1989 and in 1992, made valuable recommendations. Collaboration with International Agricultural Research Institutes improved during the project. Contract research funded some research projects at the University of Malawi. A contract research programme on coffee was successfully completed with the Tea Research Foundation of Central Africa (TRFCA). DAR now has a well constructed and equipped research station network based on three regional stations (Chitedze, Bvumbwe, Lunyangwa) and a number of sub-stations. Some sub-stations and many trial sites have been closed or handed over to other departments of MOA to reduce DAR's overheads. A new research sub-station has been built in a hitherto unrepresented agro-ecological area. The training programme has provided DAR with a body of well trained scientists. The most significant result of the project has been the development of hybrid flint maize varieties valued by small farmers. They are higher yielding than local varieties, whether fertilized or not, and, during the drought year (1991-92) produced some yield when local varieties failed. First released in 1990, by 1992-93 they occupied about 25% of the maize area, increasing rapidly in acreage annually. The iv  collapse of the credit system in 1993 has thwarted this processing trend, but it is anticipated that the hybrid maize area will recover and continue to increase. On a number of important areas, as mentioned above in the Implementation discussion (i.e. the development of a programme and budgeting system, introduction of a scientific career stream, research/extension linkage), progress was not satisfactory. Project Sustainability Sustainability of the activities/programmes supported by the project and the facilities provided is in doubt. For most of the project duration, GOM was unable to provide operating costs as agreed at negotiation. This affected both the conduct of experimental programmes and maintenance of buildings and plant. DAR, despite strong pressure from IDA, was unable to establish a separate scientific career stream that would provide incentives to motivate and retain research scientists. Consequently, numbers of professional staff remain about the same as before despite the large overseas graduate training programme. Thus, the quality of the working environment is affecting staff retention. About 30% of those trained sought employment outside the research service on their return. The situation is particularly evident with the ARTs who have been unable to retain any of the economists whose role was regarded as crucial in the SAR. In view of the fact that this project was the first in a long-term sequence to restructure Malawi's research capacity, the problem of project sustainability should be fully examined by the recently initiated Agricultural Services Project (ASP) under which further assistance for agricultural research in Malawi is being provided. Lessons Learned The major lessons learned include: - Institutional reforms require a long-term commitment and should be phased, build on existing institutions where possible, and based on detailed action plans and broad consensus. - Project scope should take into account realistic budgetary, institutional and human resources constraints; - An independent body with adequate financial autonomy is needed to coordinate research, and such a body should have the authority and influence to set research priorities; - An improved program budget system for recurrent cost funding should be developed and implemented in MOA, this should be coordinated with an ongoing review of staff levels and job description; - In order to improve the relevance of research to farmers' needs, there is a need for a promotion and rewards scheme to encourage scientists to re-orient more of their work towards on-farm research and form closer links with extension, aside from the provision of mechanisms such as ARTs and research/extension workshops etc; Field supervision of financial, technical, and management aspects of the project with IDA donor assistance is extremely important, especially the need to supervise procurement, and strong government procurement arrangements and capacity are also essential; v  More care should be taken in drawing up terms of reference, selection and supervision of technical assistants; Overseas degree training should not be emphasized, especially when incentives are weak in the public sector, in addition, future training programmes should be designed to allow fewer trainees away at one time, if research activities are not to be impeded; and - Sustainability may be improved if the new varieties of crops produced by DAR scientists could generate an income by a formal sale of appropriate material to seed or agricultural merchants under a royalty or similar agreement; The design features of the follow-up project (ASP) have taken into account all of the above lessons. It is important that ASP implement effectively the various action plans prepared to address the key lessons from NARP. Regarding the development of a cost-effective and sustainable research system that is geared to serving the needs of the majority of the farmers in Malawi, ASP action plans give emphasis to: - formulating and implementing research plans consistent with realistic budgetary constraints that will generate technologies responsive to farmers' needs, especially those of the resource poor (many of whom are women), and improving mechanisms to promote research-extension-farmer linkages; - improving the mechanisms to update research priorities, diversifying participation in publically funded research through the commercialization of some research activities, and improving the allocation of funding by strengthening the DAR; and * improving DAR's management systems and the motivation and accountability of its researchers. vi  PROJECT COMPLETION REPORT MALAWI NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 1549-MAI) PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE 1. Project Identity Project Name : National Agricultural Research Project Credit No. 1549-MAI RVP Unit AF6AE (Formerly AF6AG) Country Malawi Sector Agriculture 2. Background Introduction 2.1 The project was identified following the 1981 National Rural Development Programme (NRDP) review which concluded that a strengthening and reorientation of agricultural research was essential for increased agricultural production. An overview of the research system was carried out by the International Service for National Agricultural Research (SNAR). This was followed by preparation by the Department of Agricultural Research (DAR) of the Ministry of Agriculture (MOA) with Bank and consultant assistance, in 1982. The project was appraised by IDA in March 1984 with USA'D participation at selected meetings. The IDA credit was negotiated in January 1985 and the Credit Agreement, signed on May 23, 1985, became effective on November 19, 1985 (Part III, Table 2). The Agricultural Sector 2.2 Agriculture is the backbone of Malawi's economy. Up to 1990, it accounted for about 40% of GDP, and 85% of exports and employment. Maize is by far the most important crop in terms of cultivated area, number of growers and food security. Other major food crops include groundnuts, pulses, cassava and rice. Exports are dominated by three crops: tobacco, tea and sugar. Production is predominantly under rainfed conditions. Malawi's agriculture is characterized by smallholders and estates, differentiated by regulations concerning production, marketing, pricing and land tenure. While estate production focuses mainly on export crops, the smallholder sector accounts for about 80% of food production and only 10% of exports. It involves about 1.8 million families operating under customary tenure on about 1.75 million hectares, producing primarily for subsistence. Land and labour productivity among the majority of smallholders is very low, and nearly half of the population lives below the level of absolute poverty. High population growth, low productivity of I  land and labour in agriculture, limited employment opportunities, and low quality of human capital due to low nutritional levels and limited access to education and health services, are the main causes of poverty. This situation was further aggravated by the absence of high-yielding maize varieties acceptable to smallholder farmers, inadequate supply and use of fertilizers, and a narrow range of high-value cash crops available to smallholders. The Department of Agricultural Research 2.3 The DAR has national responsibility for research on all commodities with the exception of tobacco, tea and sugar. It is also responsible for the production of breeders' seed of agricultural crops, for the propagation and distribution of selected cultivars of fruit and nut trees, the provision of plant quarantine, soil survey and analytical services, and seed inspection and testing services. Information about improved technology is made available to the farming community reasonably regularly. At the time of the SAR, effective management in DAR was hindered by lack of staff continuity and by inadequate support staff. Planning and coordination of research programmes needed improvement to avoid duplication of effort and ineffective cost allocation of expenditures. Other major constraints included (i) the lack of adequate training programmes, (ii) an over-expensive research-station network, (iii) the absence of a policy advisory board or forum, (iv) the inadequate policy covering staff recruitment, evaluation and promotion, (v) the sporadic research collaboration with international and regional programmes; and (vi) the ineffective linkage with the extension system. The Agricultural Development Policy 2.4 As outlined in its "Development Policy (DEVPOL) 1987", the Government accorded first priority to promoting agricultural growth. The primary objectives for the agricultural sector were to ensure food security at the household and national levels, improve rural incomes, and increase export earnings, while protecting the environment. In addition to increasing agricultural productivity, the strategy envisaged policies and investment measures to expand off-farm employment opportunities. The donor community, including the Bank, supported this Government strategy. 3. Project Objectives and Description 3.1 The project aimed to improve the focus and quality of Malawi's smallholder-oriented agricultural research programme; formulate an overall strategy (master plan) for agricultural research; strengthen critical linkages, particularly between the research and extension programmes; and improve the efficiency and economy of DAR's ongoing activities. The ultimate objectives were the development of new varieties, husbandry practices and farming systems aimed at responding to farmers' regionally specific problems and significantly increasing smallholder production and incomes. 3.2 The main features of the project, involving investments, policies and institutional development, were: (i) the creation of an Agricultural Research Council to include the Controller of Agricultural Services, the Chief Agricultural Research Officer, the Chief Agricultural Officer, the Chief Veterinary Officer, the Chief Forestry Officer, the Secretary of the National Research Council, and representatives of the Tea Research Foundation, the Tobacco Research Authority, the sugar plantations, the Estates Advisory Board, and the Economic Planning Division; (ii) the rationalization of DAR's station network, including investments in civil works and equipment; (iii) the reorganization of DAR into multi-disciplinary research teams and the introduction of management and systems improvement; (iv) the introduction of a technical career stream, staff evaluation system and the 2  provision of training and technical assistance for DAR staff; (v) intensified collaboration with International Agricultural Research Centers and Regional Research Institutes; (vi) improvement in extension-research linkages through the establishment Adaptive Research Teams (ARTs) and research- extension workshops; and (vii) improved economy and efficiency in ongoing research activities. 4. Project Design and Organization 4.1 The Bank's interest in strengthening agricultural research in Malawi arose from emphasis of the 1981 National Rural Development Programme (NRDP) Review and the need for stronger research to improve and accelerate technologies for smallholders. USAID had already embarked on this process in 1979/80 by funding technical assistance, graduate training for DAR staff, modest civil work construction and procurement of equipment. The resultant project for co-financing by the Bank and USAID was to be the first (for the Bank) project of a 10-15 year programme to strengthen research for smallholders and reflects investments to deal with the identified deficiencies (para 2.3). One criticism of this approach is the emphasis on organization, management, cost effectiveness and staff development to the exclusion of ideas on how to deal with Malawi's most urgent and important problem, how to improve the productivity of the maize crop, occupying about 70% of the arable area. An innovative design feature was the inclusion of the US$ 15.5 million non incremental recurrent costs that were to account for over 90% of the Government's contribution. 4.2 The Ministry of Agriculture administered by the Principal Secretary for Agriculture was made responsible for implementing the project. The Chief Agricultural Research Officer who reports to the Principal Secretary through the Controller of Agricultural Services, was directly responsible for project management. Some guidance for this purpose was also expected from ARC. Internationally recruited specialists were to underpin financial management and procurement activities. The Ministry of Works had responsibility for the large civil works programme and a firm of architects, funded by the project, was to supervise the programme. The large training programme was to be managed by consultants supporting the MOA Training Unit, which was to be established under the companion Malawi Agricultural Extension and Planning Support (MAEPS) project. The USAID contribution to the project, under its Malawi Agricultural Research Extension (MARE) project, which was approved in July 1985, was managed by the consortium for International Development, involving four universities, with Oregon State University serving as lead institution. 4.3 However, with hindsight, it is possible to identify some design weaknesses. An agricultural research council, as commonly understood, cannot function effectively within a Government department because such a council requires autonomy. In the case of this project, this weakness was evident in two aspects: (i) the Council itself was perceived by members and by DAR as an unimportant, powerless internal committee, which led to poor attendance by members and its advice being ignored by Treasury; and (ii) DAR's inability to develop a separate career stream with better pay and other incentives for scientists, who are still within the civil service. The management structure outlined in the SAR was unworkable, especially in the face of the many staff changes at senior levels in DAR. Proposals described in the SAR, therefore, were not fully implemented. Senior scientists continued to be Officers-in-Charge of stations with senior administrators appointed to lighten their administrative work load. This was not wholly successful, however, and at Chitedze, the largest research station, a Deputy Officer-in-Charge was appointed in 1993 to share the work load. 5. Project Implementation 5.1 Implementation Schedule. The development credit was signed on 23 May 1985 and became effective on 19 November 1985. A delay in implementation was caused by problems in finalizing 3  recruitment of four key consultants. The project, due to end on 31 October 1990, closed on 31 October 1993 after three extensions (Part III, Table 2). Extension of the project was primarily to provide bridging finance for ongoing research programmes until a follow-up project could be launched. 5.2 The original credit agreement was for SDR 24.3 million. In 1991-92, during the project extension, Malawi suffered a severe drought and there was an urgent need in the agricultural community for funds for drought recovery. Devaluation of the Malawi Kwacha during the project period led to savings of SDR 6.0 million by October 1992, and the original credit agreement was amended (para 5.24) to enable the SDR 6.0 million (MK 36 million) to be used for smallholder agricultural credit, especially for the purchase of fertilizer and seeds for the small farm sector (Annex 1, Table 1). The original project objectives, therefore, were met for a total actual disbursement of SDR 17.6 million (Part III, Table 5). 5.3 Establishment of an Agricultural Research Council. ARC was established by MOA internal memorandum in 1985 and held two meetings before project implementation. The ARC, chaired by the Controller of Agricultural Services, was to give policy guidance and general support to DAR. A Secretariat was established in DAR, eventually in the Agricultural Economics, Statistics and Data Processing Unit (AGREDAT). Technical and Finance sub-committees were appointed to give detailed consideration to research programmes and contract research proposals before their eventual submission, with appropriate comment, to the Council, which approves the annual research programme and budget before forwarding them to Treasury. 5.4 The effectiveness of ARC was severely compromised by several factors. Its creation by internal memorandum established ARC as an internal committee, without legal standing and little status. Meetings were ill-attended and held irregularly because of competing claims on members' time. The Treasury did not fund ARC's approved programme and budget recommendations fully but, instead, cut them quite severely. These weaknesses were recognized and changes proposed but they were not put into effect under this project. ARC has been helpful to DAR in the provision of policy advice, development of the Master Plan and identifying strategic research issues. 5.5 DAR Infrastructure Consolidation and Improvement. At appraisal, the total investment cost under the three budget heads forming this component (Civil Works, Staff Relocation, Equipment) was US$ 14.9 million. The building programme, which included library, laboratory, office and utility buildings as well as staff houses, was completed almost in line with SAR estimates (Part III, Table 4). An extension programme, with construction of a health centre and drying shed and seed store at Chitedze, was also completed. Final disbursement under the Civil Works component, expressed in SDR, amounted to about 73% of the Development Credit Agreement (DCA) estimate (Part III, Table 5-C). 5.6 However, there were problems in design and quality in the civil works programme. Following economies in house construction, as suggested by the World Bank, the designs of houses were the subject of complaints (too small, proximity of servants quarters leading to lack of privacy, outside access for toilets). Glasshouses were too complex in design, with problems in maintenance and operation under Malawi conditions; the quarantine glasshouse at Bvumbwe did not conform to the minimum requirements for quarantine purposes, with no security, no containment and no provision for sterilizing used plant and soil material. Poor quality was evident at several sites, most notably at Kasinthula where the wiring of the cold room and store shed were not completed and connected to the mains. Surface drainage at Mkondezi was not provided for, making the site vulnerable to floods during heavy rain. 4  5.7 The research station network envisaged in the SAR was varied somewhat after discussion. The three main regional stations, i.e. Chitedze, Bvumbwe, Lunyangwa were retained as proposed, but Makoka's proposed closure was rescinded and it remains a substation, housing the cotton research programme and workers. Mbawa was retained to house the national Malawi Zebu herd, and Bolero was retained as a trial site. Otherwise closures and changes were as proposed at appraisal. 5.8 Relocation, which affected principally professional staff and some support staff was severely impeded by the training programme, with many scientists overseas on post-graduate courses while the relocation occurred. However, a significant achievement was the consolidation of the maize research team at Chitedze. Actual disbursement under this component, which constituted only 0.4% of the total IDA planned contribution, came to about 84% of the DCA estimate. 5.9 Equipment and Vehicle Procurement performance throughout the project was consistently poor and was rated accordingly during supervision. A lack of procurement expertise in MOA and frequent staff changes meant that bid document preparation was sub-standard and led to delays. Mismanagement compounded the problem. The nature of equipment and quantities of supplies ordered were not scrutinised critically, so expensive equipment, for some of which were of little or no use and some of which did not have trained users (e.g., three freezing microtomies, two in store) and unrealistic quantities of supplies (e.g. 3 tons of plasticine, most of which went unused; over 3,000 litres of absolute ethyl alcohol; 1 km of soft rubber tubing) were bought. On receipt no checks were made against delivery notes or invoices, no inventories drawn up or proper storage provided or inventory procedures instituted. 5.10 The main store is at Chitedze and is a disused laboratory with wooden shelving, no safety or fire precautions or equipment and no established stores procedures or ledgers. Some attempt has been made since a particularly critical supervision report to bring order into the store, but delicate equipment (e.g., freezing microtomies) is in the same store room with corrosive material and there is potential danger from co-storage of flammable liquids (e.g., ethyl alcohol, benzene, chloroform) and powerful acids (e.g., concentrated sulphuric acid, concentrated nitric acid). Furniture for houses and offices was not ordered until near the end of the project, so completed houses had to remain unoccupied for months until they were furnished. Some of the furniture supplied was sub-standard and had to be replaced. 5.11 Reorganization. A requirement of the project was the reorganization of DAR HQ, with Chief Agricultural Officer (CARO) supported by two Deputy Chief Agricultural Officers (DCAROs), one for Research Programmes and the other for Technical and Administrative Services. All research planning on commodities were to be vested in multi-disciplinary Commodity Teams, each reporting to a relevant National Research Coordinator, who in turn, would have reported to the DCARO Research Programmes. Also included was a reorganization at the operational level, displacing Directors or Officers in Charge of stations by Administrative Officers having responsibility for all non-research activities (research farm managers, labour, machinery and operations), reporting to DCARO Technical and Administrative Services, who would have also controlled plant protection and quarantine activities, seed services, regulation of pesticides and library services, but not soil analytical services. Funding was also to be split. Funds for research programmes were to be assigned through National Research Coordinators (NRC) to each Commodity or Adaptive Research Team Leader and thence to each project leader who would be accountable for their expenditure. Funds for station activities were placed directly with station Administrative Officers. Thus organization and planning of research programmes as well as their funding were separated completely from their operation in the field. Former Directors of Stations were supposed to be freed of administrative duties by this change and able to devote their attention entirely to research. Adaptive Research Teams were also to be 5  appointed to each Agricultural Development Division (ADD), operating locally but coordinated by NRC (Adaptive Research), and reporting to DCARO Research Programmes. They were principally to do adaptive research trials on farmers' fields, cooperating closely with extension services. 5.12 Separating the management of research planning and research operations at the stations, as well as funding them separately, was clearly going to increase management complexity, which could only exacerbate, not relieve, administrative pressure on senior staff from CARO downwards. This was recognized by a supervision mission as early as February 1986 and later by the First Triennial Review (1989) which recommended reinstatement of scientists as Officers in Charge of stations. Management/administrative overload on senior scientists was greatly increased also by the preparation of the Master Plan, done with the expenditure of much time and effort of senior scientists with some central secretariat TA support. Furthermore, the need to prepare Action Plans to amplify the Master Plan and the preparation for two Triennial Reviews, were additional to the scientists normal annual administrative burden of preparing Annual Plans and Budgets, Annual Reviews and Reports and the updating and revision of the Guide to Production Practices. In addition, Commodity Team Leaders (CTL) were expected to produce monthly, and National Research Coordinators quarterly, reports. Taken together with the staff absences caused by the training programme (para 5.21), these time consuming administrative requirements hampered the research programme and decreased the time available for writing scientific publications. 5.13 Management and System Improvements. The project was intended to provide for a Programme and Budgeting system for managing the research. Although a start was made on such a system, it was not concluded successfully. The expatriate consultant appointed to help in this development provided adequate financial accounting systems, but these were not suitable for a programme and budgeting system which required a cost and management accounting approach. The consultant left prior to the end of his contract for personal reasons and the CARO did not attach a high priority to replacing him. Bank supervision failed to monitor the work of the consultant prior to his departure or to insist on the employment of a suitable replacement. There is, therefore, no system in operation in DAR to enable it to manage its research programmes properly. An effective programme and budgeting system should describe the research work well enough to permit assessment of its scientific soundness as well as providing a means to ensure that the research is in accord with GOM policy and MOA research strategy. Research proposals also need to be costed so that value for money judgements can be made as well as views formed on the balance of research, making sure that the proportion of research funding of different scientific and commodity areas accords with policy. Research programme and budgeting systems are based on cost and management accounting, which differs greatly from normal public service financial accounting. In Uganda, provision was made in the research project SAR for a two year expatriate consultancy, with short follow-up visits in years three to five, for the development and installation of a programme and budgeting system. The Ugandan experience would be worth studying to see if it could be applied in Malawi. 5.14 Programme Planning and Review. The project built on existing programme planning procedures to establish a structured planning and review system based on meetings at three levels. Each Commodity Team Leader (CTL) was to convene working meetings of his team of research workers, including, if possible, a biometrician, to review the previous year's results and propose programmes for the following year. These meetings would be followed by meetings of CTLs with their respective National Research Coordinators (NRCs), at which the CTL would summarize their reports and present their proposals and budgets for the next season. These meetings should then be followed immediately by all the NRCs meeting as a group with DCARO Research Programmes to plan the overall research programme. Presumably, DCARO Technical Services would institute analogous meetings to apportion Plant Protection personnel to research programmes and later, to plan 6  Research Station programmes and budgets to carry out the projects planned by the NRCs. Following these series of meetings, the consolidated programme and budget would be forwarded to the Technical sub-Committee for review and approval before submission to ARC for final DAR approval. After endorsement by the Council, the final programme and budget would be submitted to Treasury as part of MOA's budget estimates for inclusion in GOM's budget for the following year. By starting the series of meetings in mid-August, when the previous season's research results should be available, it was expected that the DAR budget proposals would be ready in time for submission in the normal GOM estimate rounds held in October. 5.15 In addition to the procedure for Annual Reviews and Planning outlined above, the project provided for Triennial Research Reviews in which the whole research programme and its management would be subject to external evaluation by experienced scientists of international standing. Two such reviews were held in 1989 and in 1992 and both reported extensively on research programmes as well as some management issues. While they provided valuable information to help guide the research effort and its management, external reviews need considerable preparatory work by research workers, especially the CTL and NRC, which reduces significantly the time available for research during the period they are preparing for the review. The reviews themselves are also very costly, with international travel expenses and subsistence to be taken into account. Triennial reviews are unusual: the International Agricultural Research Centres in common with some large national research systems (e.g., UK AFRC) have for long settled on reviews every four years for their external evaluation. A similar interval for DAR would save expenses (25% in a 12 year period), reduce the time occupied by preparation and would provide essentially the same information. Future projects, therefore, should move to reviews every four years in place of the present Triennial Reviews. 5.16 Research Collaboration as foreseen at appraisal, was a small component (US$ 0.96 million) but an important one. Collaboration with international institutes, including the International Institute for Tropical Agriculture (IITA), the International Crop Research Institute for the Semi-Arid Tropics (ICRISAT), the Wheat and Maize Improvement Centre (CIMMYT), and the International Institute for Banana and Plantain Research (INIBAP) occurred with several programmes. Valuable flint maize material was received for the maize breeding programme from IITA, following a very successful visit of scientists to IITA programmes in Nigeria and Cameroon. Subsequently cooperation with CIMMYT, which is Malawi's official maize research collaborator, has been good. 5.17 The Contract Research programme funded a successful coffee research programme at the Tea Research Foundation of Central Africa (TRFCA), for which houses and office and laboratory facilities were built on the main tea research station at Mimosa. Macadamia nut research was instituted late in the project with contract funds through the Tree Crops Authority. Several smaller research contracts were awarded for research proposals from Bunda and Chancellor Colleges. There does not seem to be a formal review or evaluation procedure for contract research, and most research contracts have not produced reports. 5.18 Library and Data Processing. The research library at Chitedze was moved into a new building constructed under the project and now functions effectively as a central library for the whole research system. The smaller libraries at Bvumbwe and Lunyangwa were improved. Inter-library loans, current contents and reprint photocopy services are provided to scientists at other institutes on demand. CD-ROM readers are available and used to extract information on demand from several databases. The salaries offered to librarians in government service, and lack of career prospects do not attract suitable candidates. The salary and career structure issues, of course, cuts across DAR. Nevertheless, there is a librarian trained by the project in post working as counterpart to the expatriate technical assistant. However, the cataloguer in training during the project has not yet 7  returned, and this post at present remains unoccupied with a large backlog of material to be catalogued. Data processing capacity was improved, but, AGREDAT, HQ, in common with other parts of DAR has lost key staff. 5.19 Publications Development. Provision was made to equip the library with facilities for publishing and binding. These are now in operation and have improved the standard of presentation of reports. Scientific publication of results in journals and other media has been slowed down by the absence of scientists on training or preoccupied with management reporting. The Guide to Agricultural Production in Malawi, a handbook for use by extension staff, was regularly updated by research staff to include the latest research results: the 1993-94 revision has just been published. Although extremely valuable as reference material and much used by extension staff, the way the ART system actually operated, prevented a systematic supply of results of on-farm regional research trials, which devalued the book somewhat. All research stations now produce timely Annual Reports on research done and these are available to Subject Matter Specialists and other extension staff. 5.20 Career Development components of the project required comprehensive job descriptions for all professional staff, an improved staff evaluation system and the implementation of a scientific career stream for DAR which would provide sufficient incentives to retain scientists within the system and enhance promotion prospects for the best scientists. Despite strong pressure from IDA, the establishment of a scientific career stream outside the main Civil Service grade structure was not achieved due to GOM opposition. The opposition was based on budgeting implications and concern that is would isolate researchers from other civil service career streams. Job descriptions and a personnel evaluation scheme appropriate to research workers have not been implemented. However, there has been some improvement in promotion procedures recently in that Scientific Interview Boards have been instituted, the members of which are proposed by MOA and not the Public Service Commission, as was done before. Three or four of the best researchers have received accelerated, or special merit promotions to keep them within the service. Conditions of service, including pay and normal promotions, still remain within the province of the Public Services Commission. The recently introduced retirement age of 50 years is unrealistically low and will deprive the research system of its most experienced managers as they reach their most effective age. 5.21 The Training component of the project was substantially supported by USAID. Total overseas graduate training during the project period included four M.Sc. candidates and nine Ph.D. candidates out of IDA funding and 27 M.Sc. and four Ph.D. candidates with USAID, supplemented by considerable short-term training for both professional and research support staff. Overseas short-term training consisted of 374 person-weeks (of which USAID supported 104) out of a total 1,728 planned weeks. In-country short-term training, for which 3,102 person-weeks were provided, led to 715 persons being trained during a total of 2,500 person-weeks (Part III, Table 4). This big investment in training meant that during the project period large numbers of staff, especially graduates, were overseas at the same time. This had the unintended effect of impeding several research programmes during the project. Less overseas training should therefore be needed in the future. The training to be eventually considered necessary and its timing should form part of a manpower development plan to be included within a comprehensive personnel strategy for the research system. Problems in the management of the trainees' allowances were alleged: apparently trainees were not all paid the same subsistence rates, allowances to attend approved conferences were not forthcoming and frequently maintenance grants were paid late. Many returning trainees resigned from DAR to take better jobs elsewhere, so despite the large training programme, DAR has about the same number of trained scientists as before. 8  5.22 The Technical Assistance element of the project provided for four key long- term technical assistants who were appointed early in 1986 and others later. Their performance was variable, few leaving fully trained and competent counterparts in post when they left. Agro-forestry, library and AGREDAT support was very good, and adaptive research weak: the horticulturist TA took over the running of this programme during the absence or graduate training of all the professional horticulture staff. The Research Adviser, appointed for three years in the first instance with an option to renew his contract for a further two years, was not reappointed. The TA accountant did not develop a programme and budgeting system as required, leaving DAR without this essential tool in research management. As indicated in Part III, Table 4, long-term TA included a total of 625 (244 IDA and 381 USAID) person-months or 116% of SAR estimate. Under the short-term TA programme, 40 person-months (70% of SAR estimate) were provided with USAID assistance. 5.23 Research/Extension Linkages. Adaptive Research Teams were appointed to all Agricultural Development Divisions (ADDs) to improve the link between research (DAR's responsibility) and extension (ADD responsibility) links. Each ART was composed of an economic and agronomist and they were to conduct on-farm trials and identify constraints affecting small farm productivity. Their level of performance, however, was below expectation, partly because of the excessive work load, and also because the economists, who were needed for diagnostic work on farmers' constraints found more lucrative employment and left. At close of the project, only one ADD (Kasungu) had a fully staffed Adaptive Research Team and three (NADD, LWADD, LADD) had no adaptive research personnel. The remaining four ADDs had only one ART member each. Before discarding this approach, however, an investigation should be launched into the reasons for its failure and a revised system installed and tested. Scientists at various research stations should regard adaptive work as the final step in the research process and should be held responsible for its success. Until a system is formalized whereby adaptive research results are fed systematically into the recommendation making process, it would be difficult to obtain from the adaptive research team concept, successful results. 'he teams may need to be more selective in their approach, only taking on a workload that is achievable. This would require some management training so that the adaptive research specialists would know how to rank tasks in order of priority. Research/extension workshops were required under the project. The first one was held in PYI and others annually thereafter. Research stations held annual field days, highly regarded by extension personnel, at which research findings were presented to extension officers. In addition, less formal meetings in which extension and research staff participate were held from time to time. Extension staff were welcome to visit research stations informally to seek advice or specialist information. Although the research and extension services do not form a seamless, integrated organization, relations between the two were thought to be good, with an excellent working relationship ensuring that research results were made available to extension services with little delay. 5.24 Support for Drought Recovery. In early 1992, the Smallholder Agricultural Credit Administration (SACA), which provides agricultural credit to smallholders, faced a financial shortfall of about US$ 31 million to finance urgently needed agricultural inputs for the 1992/93 agricultural season. This gap was due to the low loan recovery rate experienced by SACA as a result of the drought in 1991/92 and of increases in the price of fertilizer. Given its severe financial constraints, the GOM requested IDA help to cover SACA's financing gap. As indicated in para 5.25, the legal document was amended and approved in 1992 in order to help finance drought recovery activities and a total of SDR 6.0 million was made available for this assistance. Under the overall programme carried out by SACA during the 1992/93 season, some MK 144 million, of which MK 36 million or SDR 6.0 million equivalent from IDA, were disbursed benefitting about 384,000 smallholders representing about 15,600 farmer clubs. It is estimated that some 110,000 farmers benefitted from the IDA-assisted programme. 9  According to data provided by SACA (Part III, Annex 1, Table 1) the seasonal credit recovery rate for the 1992/93 season was about 16%, well below the traditional recovery rate of 88% experienced by SACA between 1985/86 and 1990/91. A recovery rate of 38% was recorded in 1991/92. A Smallholder Credit Repayment Study, recently carried out by the Centre for Social Research of the University of Malawi, indicates the following main reasons for the current smallholder loan repayment crisis: (i) breakdown of the recovery system; (ii) credit design and administration; (iii) low gross margin; (iv) political dispensation; (v) 1991/92 drought; and (vi) delinking extension work from credit. These issues are being addressed through the recently appraised Malawi Rural Financial Services Project. Project Costs and Disbursement 5.25 The project was scheduled to be implemented over five years starting from 1985, and expected to cost US$ 49.9 million to be financed by a IJS$ 23.8 million (SDR 24.3 million) IDA credit, a US$ 9.2 million USAID grant and a US$ 16.9 million equivalent GOM's contribution, mainly to cover DAR ongoing programmes. There were two amendments to the legal document for the project approved in 1992 (Part III, Table 5-C). The first, as advised by the July 1992 supervision mission, to reallocate SDR 0.6 million to non-salary operating costs. The second amendment was approved in order to help finance urgently needed drought recovery activities in the form of agricultural inputs to smallholder farmers for which the GOM had requested IDA assistance. Under the restructured project, a total of SDR 6.0 million were made available for this assistance. To date, against an actual project cost of US$ 58.2 million, IDA disbursed US$ 32.1 million (SDR 24.0 million)' and USAID disbursed US$ 9.2 million. Government contribution included mainly DAR ongoing programmes and at the end of the project amounted to an estimated total of US$ 16.9 million (Part III, Table 5-A and B). While actual project costs in US$ terms exceeded SAR estimates by about 16% , total IDA disbursement in SDR reached 97% of the total foreseen in the DCA (Part III, Table 5-C). 6. Project Results 6.1 In spite of the implementation problems as described above, it can be stated that the overall project performance was satisfactory, and most of the original objectives were achieved. The large training programme has now given Malawi a highly trained body of scientists able to efficiently utilize the resources furnished under the project, provided that staff can be retained within the research system on their return from training. A new research sub-station has been built in a hitherto unrepresented agro-ecological area; many superfluous sub-stations and trial sites have been closed with consequent savings in overheads; new houses, laboratories and offices have been built on the main research stations; and older buildings have been refurbished. Library buildings and services have been improved; new laboratory equipment has been installed; and supplies of needed laboratory chemicals provided. New and more efficient research management procedures have also been institutionalized. A Research Master Plan, first prepared in the early years of the project, has been extensively revised during the project and now provides strategic guidance for priority research. 6.2 Benefits (Part III, Table 6) are difficult to quantify in this type of project since benefits arising from adaptive research cannot be separated from those arising from the associated use of additional inputs. Therefore no attempts were made at appraisal to calculate and economic rate of return for the project. The success in breeding the consumer acceptable and high yielding flint hybrids should bring about considerable benefits comparable to those of hybrid maize in Kenya and Zimbabwe and on a Excludes the revolving fund amount of US$ 89,562 (SDR 67,032). 10  smaller scale to HYV dwarf rice and wheat in Asia. There has been some doubt about the returns to research expenditures in Africa. In the case of DAR, there are at least three instances in which research staff brought about massive benefits, far above the roughly US$ 2 million per year operating costs and probably worth more than the entire project cost to the economy. The first was the work on high content (HC) fertilizers. It is estimated that the switch to HC fertilizers, as recommended by research, saved the country some US$ 14 million in 1992/93 if comparison is made with the 1987/88 agricultural season (Annex 1, Table 2). The second was the release of parasites and predators which successfully controlled cassava mealy bug infestation. While controls were introduced by IITA, it nevertheless needed DAR staff to collaborate and follow-up. Finally a hybrid maize breeding programme to produce flint type grain of high yield and acceptable to local farmers' tastes has been outstandingly successful. Two new hybrid flint varieties (MH17 and MH18) released in 1990 outyielded local varieties by more than 30% even when unfertilized. When fertilized, yields on farm doubled to about 3 t/ha although official statistics to support this performance are not yet available within MOA. The new hybrid flint varieties are very popular with farmers and by 1992-93 occupied 25% (about 100,000 ha) of the area planted to maize'. They have the potential to make Malawi self sufficient in maize. Other programmes (e.g., rice, sorghum, cotton) have new varieties ready for release. It should be mentioned that the extension service under MAEPS, contributed less than expected towards the successful introduction of hybrid flint maize varieties in Malawi. 7. Project Sustainability 7.1 Sustainability of the activities/programmes supported by the project and the facilities provided is in doubt. For most of the project duration, GOM was unable to provide operating costs at the level agreed to at negotiation. This affected both the conduct of experimental programmes and maintenance of buildings and plant. DAR, despite strong pressure from IDA, was unable to establish a separate scientific career stream that would provide incentives to motivate and retain research scientists. Consequently, despite the large overseas graduate training programme, the professional staff remains about the same as before the project, with many seeking employment outside the research service on their return. Thus, the quality of the working environment is affecting staff retention. About 30% of those trained sought employment outside the research source upon their return. The situation is particularly bad with the ARTs, who have been unable to retain any of the economists whose role was regarded as crucial in the SAR. However, as it was recognized that this project was the first in a long-term sequence to restructure Malawi's agriculture research capacity, the problem of project sustainability should be properly addressed by the recently initiated Agricultural Services Project, under which further assistance for research in Malawi is being provided. 8. Bank Performance 8.1 Bank supervision was good with 15 supervision missions from October 1985 (pre- effectiveness) to May/June 1993. Actual Bank staff inputs from preparation to PCR work, amounted to 287 staff-weeks (Part III, Table 8). One key feature of the Bank's performance was the intensive supervision, amounting to technical assistance, provided by its Agricultural Department. This led to considerable improvements in research management, notably Action Plans to operationalize the Master Plan proposals. The Bank played a particularly active role in supporting the hybrid maize breeding programme. Based on information by DAR of MH18 and MH17 seed sales estimated at 2,500 tons in 1992/93. 11  8.2 In retrospect, however, Bank supervision should have paid closer attention to the monitoring of the work of the consultant responsible for developing the programme and budgeting system prior to his departure, and should have insisted that a suitable replacement be hired. 9. Borrower Performance 9.1 The Borrower showed commitment to the ideals, objectives and substantive goals of better research planning by the development of a Research Master Plan, and research action plans. Borrower support for hybrid maize programme was instrumental to its success. GOM complied with 15 of the 22 covenants requiring action, however, a number of covenants have been in default, some consistently: (i) the MOA did not maintain payment of operating costs of research during the whole project; (ii) the ARC secretariat was not adequately staffed until near the end of the project, when improvements were made; (iii) there were weaknesses in procurement for the whole project duration as reflected by sub-standard bid document preparation, mismanagement, insufficient DAR scrutiny and evaluation of procurement requests and poor procedures for recording delivery, receipt, inventory, storage and distribution of goods delivered; (iv) DAR did not introduce a scientific career stream with enhanced prospects for scientists because GOM was adamantly opposed to this because of the budget implications and concern that scientists would be isolated from other civil service career streams, consequently IDA agreed to "waive" this condition; (v) DAR was unable to implement a programme and budgeting system as required leaving it without an essential tool in research management; (vi) reporting was irregular and often overdue, especially submission to IDA of the reviews of the previous year's work and prioritized proposals for the following years; (vii) project account audits were frequently delayed and submitted long after their due date; and (viii) DAR was unable to comply fully with requirements to maintain project installations adequately because of its limited operating budget. 10. Lessons Learned 10.1 The major lessons to be learned from this project are: - Overseas degree training should not be emphasized, especially when incentives are weak in the public sector, in addition, future training programmes should be designed to allow fewer trainees away at one time, if research activities are not to be impeded; - More care should be taken in drawing up terms of reference, selection and supervision of technical assistants; - Sustainability may be improved if the new varieties of crops produced by DAR scientists could generate an income by a formal sale of appropriate material to seed or agricultural merchants under a royalty or similar agreement; - Field supervision of financial, technical and management aspects of the project with IDA donor assistance is extremely important, especially the need to supervise procurement, in addition, strong government procurement arrangements and capacity are essential; An independent body with financial autonomy is needed to coordinate research, and such a body should have the authority and influence to set research priorities; 12  An improved program budget system for recurrent cost funding should be developed in MOA, this should be coordinated with an ongoing review of staff levels and job description, in addition, a reputable local accountancy firm can be used on a short term consultant basis for the introduction of such system; In order to improve the relevance of research to farmers' needs, there is a need for a promotion and rewards scheme to encourage scientists to re-orient more of their work towards on-farm research and form closer links with extension, aside from the provision of mechanisms such as ARTs and research/extension workshops etc; Project scope must be kept simple and adjusted to realistic budgetary, institutional and human resources constraints; Institutional reforms require a long-term commitment and should be phased,build on existing institutions where possible, and based on detailed action plans and broad consensus. 10.2 Other lessons include: - More rigorous inspection is needed by the line ministry to certify completion of civil works before payment of contracts; - Building (house and glasshouse) design should be suitable for the purpose, the climate and other environmental variables; - The overriding importance of suitably built stores, proper storage and efficient stores accounting/management procedures to ensure the proper storage and inventory of consumables and equipment; - Triennial Reviews, are very costly and entail much preparation, four yearly reviews which would provide information of the same value at much less cost and with less interference in the research programme should be adopted; - The need exists for MOA to continuously maintain at least a building engineer as well as transport management to ensure the proper management and maintenance of buildings and equipment, if the considerable investments structures and vehicles as a result of the project are to be safeguarded; and - Training in financial management, research management, operation of equipment and research station management are essential to ensure the sustainble benefits of the project. 10.3 The design features of the follow-up project, ASP, have taken into account all of the above lessons. It is important that ASP effectively implements the various action plans prepared to address the key lessons from NARP. Regarding the development of a cost-effective and sustainable research system that is geared to serving the needs of the majority of the farmers in Malawi, ASP action plans give emphasis to: - formulating and implementing research plans consistent with realistic budgetary constraints that will generate technologies responsive to farmers' needs, especially 13  those of the resource poor (many of whom are women), and improving mechanisms to promote research-extension-farmer linkages; improving the mechanisms to update research priorities, diversifying participation in publically funded research through the commercialization of some research activities, and improving the allocation of funding by strengthening the DAR; and improving DAR's management systems and the motivation and accountability of its researchers. 11. Project Relationship 11.1 There was good Borrower/Bank collaboration, due largely to the constant dialogue between MOA and the Bank's Field Office. In addition, as pointed out elsewhere (para 8.1), supervision missions provided considerable implementation assistance (especially on technical and policy matters), which was effective, and appreciated by the Government. There were times, both in the early and closing stages when there were some differences between Bank and MOA specialists (e.g., hiring a replacement for the expatriate consultant hired to develop the programme and budgeting system), but the discussions always remained constructive. 12. Consultancy Services 12.1 These were described in the context of the various components. The Government liked the USAID recruitment method, under which USAID financed management identified candidates and brought MOA managers to the USA for selecting the seven MARE financed consultants. The result of consultancy services was mixed, some did good work, while four of the seven multi-year consultancies gave less than satisfactory results. More care is advisable in drawing up terms of reference for consultants and, in some cases, their selection and supervision. 13. Project Documentation 13.1 The Legal Document gave clear guidance on the Borrowers' obligations, but the SAR lacked detailed cost tables and consultant terms of reference, which made implementation, supervision and even PCR preparation more difficult. The cost tables, for instance, are the only record of proposed investment details. The DAR progress reporting was reasonably regular and kept the Bank informed of progress, but annual research results reporting was either very late or given in raw detail (trial by trial) that was difficult to digest. DAR was addressing this aspect, with the objective of establishing punctual and effective annual reporting, but had not yet been achieved at project closure. Due to incomplete data on actual project cost by category and by year, the PCR mission was not in a position to provide an accurate total project cost table. In particular, figures on GOM's contribution, which under the project was supposed to meet mainly the operating costs of DAR's new activities, were not available. 14  PROJECT COMPLETION REPORT MALAWI NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 1549-MAL) PART II: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE Introduction: 1. Following the revised preparation of the PCRs, Part I and 3 have been prepared by the Consultants hired by the World Bank, as a lender of funds to the Malawi Government. Part 2, on the other hand has to be prepared by the borrower, who in this case is the Malawi Government (MG). Within the borrower, there were other components of the project. For example, Agricultural Extension was also a beneficial of the project, besides the Agricultural Research, (AR). This report covers the AR component. 2. To write this report, visits to major research stations and interviews of Senior Scientists were made between December 7 and 15, 1994. Two Malawian counterparts, Dr. A.S. Kumwenda of the Department of Agricultural Research and Mr. Chanza of the Department of Extension and Training accompanied a three man team of the Food and Agricultural Organization (FAO) hired as consultants of the World Bank. The report is further based on some experience by the author during the implementation stages of the project in 1989 to 1992. AR Stations Visited and Interviews Made. DateStation Scientist Interviewed 7/12/93 Chitedze Dr. B.T. Zambezi, Dr. C. Matabwa, Miss C. Wern 8/12/93 Dr. M. Theu - Plant Protection Mr. H.N. Soko - NRC Legumes and Oil Seeds 10/12/93 Bvumbwe All Senior Staff Members Makoka Mr. M.H. Banda, Officer-in-Charge 11/12/93 Kasinthula Mr. E.M. Chinthu and Mr. Kauta 14/12/93 Chitedze Mr. A.W.C. Zimba - Livestock 15/12/93 Mkondezi Mr. C. Chizala, officer-in-Charge Lunyangwa Mr. M.S.L. Kumwenda, Officer-in-Charge 20/12/93 MOA Wrap up meeting with Ministry Officials 15  3. The following is a consolidated factual account gathered during the above visits. The report further incorporates the critical comments made by scientists interviewed during the visits. The report is by station followed by a summary of common issues across stations. Chitedze Agricultural Research Station 4. The overall feeling of the few scientists interviewed was that the project was beneficial to the stations and the DAR overall. This was clearly expressed by the Head of the station who is also an active research scientist as a Maize commodity Team Leader. This feeling was further amplified by the comments made by the individual scientists interviewed at the station. 5. The station had overall benefit in buildings and research facilities. These included the dwelling houses, the internal road network, library, renovations to buildings, laboratory building complex, health unit and vehicles. Through NARP, Chitedze was completely transformed. Dissatisfaction was nevertheless expressed on various issues of the Project as follows: 6. Operational Funds: Although the concept of NARP was to strengthen the AR Institutes of the MOA through infrastructure and manpower development, an acute shortage of operational funds was felt during the implementation of the project. Through NARP, a large consignment of laboratory equipment and vehicles was acquired. Unfortunately this was not matched with adequate maintenance funds on revenue accounts. Often times vehicles broke down and could not be properly maintained. Some of them became obsolete quicker than expected. This also applies to laboratory equipment. 7. Equipment: Along with the laboratory equipment, there were several other farm equipment acquired through the project. However, there were some which could not be used because they were of wrong types. Ploughs acquired were too heavy to be pulled by the tractors available at the station. There were also wrong laboratory equipment which could not also be used because they were wrong designs e.g. in the Virology Laboratory which could not be utilized. What became apparent during the course of the Project was that there was poor planning and coordination in ordering such equipment. The absence of Professional Procurement Officer and poor interaction between the users and the buyers aggravated the usefulness of some of the ordered equipment. 8. Re-organization of the Commodity Groups: Through NARP, the DAR was organized into seven commodity groups. While this proved beneficial in centralizing the research plans, some groups turned out to be too big. For example the Technical Services is currently large. Further more, the attachment of some Commodity Teams to this group had its own drawbacks. Plant Pathologists who are in the Technical Group are in a way isolated from the commodity teams in which their research activities are based. In a way it does not fit the multi-disciplinary approach of activities. Most commodity teams feel they had better had their own pathologists and not those in the service group. Resources permitting, some of the major commodities should have a complete set of staff from a breeder, agronomist, pathologist to an entomologist and indeed some commodities deserve this set up while others do not. 9. Training: As indicated by the scientists interviewed at Chitedze the feeling of the Department is that long term-training through NARP was adequate. The project has transformed the level of trained manpower from few PHD holders to several of them now. 16  10. What was felt lacking in the Project was the level of short term training. Technical staff did not benefit as much as the professional staff. This was felt in Soils and Engineering Commodity where many of the Technical staff who handle the laboratory equipment were not trained to use them. 11. Library: There are several achievements that had been noted in Library through this project. More relevant books were ordered and access to information has greatly improved as a result of these improvements. Indirectly, as a result of trained library manpower, sub-station library units also benefited. The trained library manpower at the Central Library have trained some individuals at the sub-stations. This was not possible without NARP. Furthermore, books on specific topics for stations such as Makoka and Lifuwu were transferred to the appropriate stations where they could be utilized more than at the central station. The library is now considered a Ministry of Agriculture institution, within which there are four departments. These do not all contribute to the costs of running the library. Whether or not it was through NARP, the situation would be better if other components of the MOA shared the running costs. Consolidation of Research Stations 12. To save overhead costs of running stations, there were plans to consolidate the stations. There were to be three major stations and about eight sub-stations. Some stations were closed down, either because they were in the wrong areas or because it was economical to close them. The centralizing of researchers in major stations was hoped to enhance interaction among scientists. Overall this has improved. But as to whether or not overhead costs have been saved is another question. Per capita expenses of several scientists at a major station is lower than per capita expenses of few scientists at a small station. Nevertheless stations like Chitedze have become very large with its own associated social problems. Likewise, coordination of activities by the station management can sometimes become difficult for such a large institution. 13. Another problem in the process of consolidating stations, was that small stations were overlooked in the numbers of buildings, they could have benefited from NARP. Soon after the project was launched it was found out that there were far too many infrastructures for the major stations and almost none for some of the sub-stations. As a result, improvements which would have been achieved through NARP of such stations were little achieved. It was completely forgotten that major stations on their own would not conduct research effectively. They need the support of small stations. Bvumbwe Agricultural Research Station 14. Again the overall impression was that NARP was a success. However, as lessons for the future, the following observations were made: 15. Civil Works/Houses/Laboratories: The planning of what buildings needed to be built where, was poor. Consequently some of the needed buildings which could have been covered under NARP were not included. For example, a Horticultural laboratory is needed at Bvumbwe and yet it was excluded in the plans of buildings for the station. For the buildings that were finally agreed upon, the designs were unsuitable. A good example was the Quarantine Glass House which will probably never be utilized. The main problem is that it is not meant for the tropics and it lacks the essential components of a suitable quarantine glass house in the tropics. A cold room which was not completed at the end of the project was due to poor planning and co-ordination. 17  16. Regarding laboratory equipment, some sections e.g. the pathology section had not received some of the ordered equipment. Again as expressed at Chitedze the co-ordination of the procurement of equipment was poor. There was need for a professional procurement officer. Procurement of wrong equipment was ascribed to the absence of a procurement officer - conversant with scientific equipment. 17. The station also benefitted from the project through houses, but rooms of the houses were too small and had no furniture. During the project, some houses were rehabilitated using NARP funds. Whether or not with the view of saving costs, the local contractor was poor and did a substandard job. In addition, senior staff houses which were planned for at the beginning were eventually not constructed. This further reflected on the initial poor planning. The general feeling among the staff interviewed was that limited scientists were involved in the planning. While NCRs were supposed to consult their commodity group scientists, often times, they did not and the input from them was missed. 18. Library: The library at Bvumbwe was again upgraded through NARP and it is serving a useful purpose. Access to journals and needed information had equally been improved. However, there is no trained librarian. This has its associated problems e.g. assessment of the needs of the scientists at the station is inadequate. Some key journals are missing as a result of this. 19. Training: On the question of training, the feeling was that it was good. Depending on the country where the training took place, NARP funded students were not equally administered. British trained NARP students were receiving lower rates of per diem than their colleagues funded by the same World Bank. Regardless of where the training took place, the sponsor, in this case, World Bank through the NARP should have had standard rates of allowances for their students. Generally, as reflected in the rate of attrition, the level of investment in training mismatched the level of investment in career development of the trained staff. Through the project implementation, NARP highly trained scientists were frustrated with their careers upon return to their countries. Their salaries and conditions of service were not commensurate with their level of training - consequently some of them left the department. 20. Vehicles: Bvumbwe station acquired some NARP vehicles. A general feeling was that the pickup vehicles were unsuitable because no three scientists could travel together because it meant the third being seated at the back of the vehicle. Pajero type of vehicles were suitable for big stations like Bvumbwe. With these type of vehicles scientists could combine their field trips. 21. Contract Research: Regarding contract research, there were strong opinions expressed against the whole rationale of allotting more money to contract research than to the commodities. A comparison was made within the coffee commodity team. The Coffee Research Unit in Mulanje was more funded that the sister components of the commodity at the station. This could have been better planned. 22. Although the Coffee Research Unit (Breeding) was working collaboratively with its sister components (pathology) this was not enough. Often-times the two units, the Bvumbwe and Mulanje, were working as two separate entities. This was undesirable. Other commodities in DAR who had part of their research programmes under contract research felt the same generally. 18  Makoka Agricultural Research Station 23. Makoka Agricultural Research Station was one of the beneficiaries of the NARP. The station acquired some vehicles, houses, motor cycles, office and laboratory equipment. This was contrary to the original agreements. The station was supposed to be closed down and staff move to Kasinthula where cotton research would have been based. As it turned out, local management found this agreement unsuitable because the basis of the decision that the station was not in a cotton growing area meant that the station was there solely for cotton. The station had other commodity teams based there or conducting research there. There are now pigeon peas, roots and tubers, other grain legumes, maize commodity work at the station. Towards the end of the NARP, the station had also an SADC/ICRAF Agroforestry Project based there. It has therefore turned out to be a more useful station to the Department than it was when NARP was planned and initiated. The idea of closing down Makoka was in the end found not necessary. 24. Like at Chitedze and Bvumbwe, the house built at Makoka were found to be of poor finish. In addition, the electrification system was poor. Once again indicating that the firms which were given contracts of some major works under NARP lacked supervision and were generally perhaps cheap types. Kasinthula Experiment Station: 25. Kasinthula Research Station with a team of 4 professional staff members benefitted from NARP through several civil works, laboratory equipment and training. Eight offices, 3 laboratories, 1 open shed, a cold room and water boreholes were completed under the NARP project. In addition a large quantity of laboratory equipment for Irrigation and Drainage Commodity were acquired. Based on the accounts by the Officer-in-Charge and the irrigation agronomist who was interviewed and observations made, it was apparent the following were not properly planned for and done: 26. Civil Workws: The finish of the offices was very poor. The inside walls were not plastered. One problem which was location specific were the damage to the ceilings by the bats. It was so obvious that it will not be long before the ceilings collapse. The type of ceilings used in NARP buildings are made of cemented compressed straw and are heavier than the normal ceiling board. Bat secretions in the cavity between the ceiling and the roof apparently make the ceilings more heavy and will easily fall off the roof. 27. Another problem that was noticeable again reflecting deficiencies in planning, was the recently installed air conditioners. Whether or not it has to do with the weather conditions of the area, the conditioners had not functioned properly since they were installed. Constant requests by the station management to have them serviced by the company that installed them had borne no fruit. 28. Besides the air conditioners, the cold room which was built under the project was not completed because it was not connected to electric power. The reason for this was never known and hence is simply idle up to now. The large consignment of laboratory equipment received at the station did not match with the work load of researchers. There were far too many equipment which had not been utilized since they were received. Crammed in a limited laboratory space were leaf area meters with TV monitors which nobody seemed to have known how to operate. There were piles and piles of metal brief-case type containers of soil core samples which might not be used for a long time. There were other small items which were in their original packing containers clearly indicating they were not used. Overall, the obvious impression was that the station was over-supplied with the equipment and could be re-distributed/re-allocated to other stations who could perhaps use them better. 19  29. The domestic water supply which had to be rehabilitated was eventually found expensive and the alternative was to dig up two bore-holes. These unfortunately had salty water and have been abandoned. The initial plans to rehabilitate the supply would have involved replacement of water pipes, old water pumps, purification of the water from the Shire river and provide a new water tank. Having not done this, a new water tank which was installed was thus connected to the old piping system. Consequently the high maintenance costs of the water supply system which was initially planned to be eliminated still exists. This again demonstrated that the initial plans in most of the NARP civil works were not based on adequate baseline data and that there was inadequate interaction between the end users and the contractors. It seems there were hasty plans without enough information on costs e.t.c. and hence plans could not match the resources in the project. Unfortunately in most cases, the quantity was not adjusted and quality of the work was traded off. 30. On a happy note regarding the station, training under the project was indeed useful to the Sorghum and Pearl Millet and the Irrigation and Drainage Commodity teams. This was evident through the scope of work that was on the ground. It was quite sensible and quite applicable to the nature of the crop production problems in question. Mkondezi Experiment Station 31. This is one station that could be classified as a complete baby of the NARP. It was established with the thinking that Lunyangwa Agricultural Research Station would move from Mzuzu City to the station site. The reason was that the city was growing and would not be ideal to have a station within the city. Lunyangwa was in the end not moved to Mkondezi and it had to become a sub-station of Lunyangwa. The station was opened in 1989 and it has one professional officer, one technical officer and 14 field assistants. A total of five commodity teams conduct their experiments at the station. The main commodities being cassava, tropical fruits, spices and rice. 32. Civil Works: As a new station, all the civil works, most equipment and other amenities are quite new. They include 5 offices, 2 laboratories, 1 work-shop, 1 cold room, 2 farm input stores, 14 class F, 1 class C and 4 class D houses. Again under NARP an access road was improved by grading and graveling. The water supply also improved under NARP by connecting the station supply system to the Nkhata-Bay Township grid which is quite good and reliable. Electrification was done well and is also functioning quite reliably. 33. Problems cited at Mkondezi regarding the dwelling houses were also cited at Bvumbwe and Makoka. The class F houses had no kitchen and the toilets can not be accessed easily from the house. The class D houses had a small bath-room which could have been bigger. The class C house's bedrooms were poorly ventilated. This was improved by constructing an additional window. 34. As at Kasinthula, some houses within the station have an incomplete electrification work. They have been wired but no transformer has been installed for 6 field assistants houses, an office block and a working shed. Another incomplete supply of equipment was the tractor without implements. A trailer and plough were not bought together with the tractor which was expected. The office block has no window burglar bars and hence liable to break in. 35. Overall, the station has substantial civil works which are being utilized. The workshop shed is the only facility that does not match with the single vehicle and tractor. The shed is too large and it could have been utilized by a bigger station. Once again this is a clear indication of the errors of the initial planning exercises. 20  Lunyanewa Research Station 36. Lunyangwa Research Station is one of the 3 regional research centres in the country. It has 5 sub-station under itself in the Northern Region. It has 600 ha of land, 326 ha are under natural woodland and 230 ha as arable land. There are 7 professional officers and 25 technical assistants. At the station are also based the following Research Development Projects: GTZ Biological Control Programme, FAO Cassava Research and Development and the FAO small-holder Coffee Research Unit. Besides these donor funded projects, there are cassava, maize, oil seeds and livestock commodity teams working there. 37. The station benefited from NARP through the acquisition of vehicles, field implements, laboratory equipment and civil works - including 5 vehicles, 6 class C houses, 10 class D houses, office block of 11 offices, 1 library, 2 laboratories, 1 glass house, 1 screen house, a spraying equipment for livestock. One of its substations, Ntchenachena also benefitted through the building of a crop storage, 1 class D house, 1 laboratory, 2 class E houses and 2 offices. Several houses were also rehabilitated using NARP funds at the station and at Bolero, Baka, Meru and Mbawa. 38. Civil Works/Laboratory: Problems cited by the station management were not at variance with what was cited at other stations and they included: (1) Wiring of houses was poor and switches had already been burnt. (2) Roofs of some houses were leaking (3) Ground surface drainage was poor and there were water pools around houses after rain. (4) Class E houses inside walls were not plastered (5) Class C houses had boys quarters attached to the houses and had poor stoves difficult to use for cooking (6) Furniture for the houses and offices were of very low standard - made of weak wood and for Lunyangwa the total furniture was not adequate for the house built. (7) Glass houses have wood bases on the concrete slab and they are already rotting 2 years after starting using it. 39. One of the recommendation under NARP was to consolidate research stations by upgrading Bvumbwe, Chitedze and Lunyangwa as regional research centers. Lunyangwa is not upgraded to that level until now. The officer-in-charge post is not upgraded to DCARO level. 40. Training: Regarding training, management felt strongly that the station benefitted little from this programme. There was nobody sent, using NARP funds, on neither long nor short-term training. Even if the training was claimed to be inadequate, the quality of work going on in the laboratories shows appropriate training of technical support staff had been obtained otherwise. 41. Considering their large number of civil works and several rehabilitations of old buildings, the station benefitted significantly from NARP. 21  Across Research Stations and DAR Issues 1. DAR Research Stations Consolidation 1.1 NARP generally strengthened the infrastructure development of many stations with emphasis, of course, on the 3 main stations. Through NARP improvements are evident through improved office and library facilities at the main stations. Some sub-stations like Kasinthula and Chitala also benefitted in similar manner quite significantly. Other sub-stations have, of course, indirectly benefitted from the main station improvements through improved research programmes of highly trained scientists. 1.2 The most significant error, which can serve as a lesson for the future, was the little input from scientists on the ground who could have suggested appropriate adjustments and changes to the design of the dwelling houses, laboratories and glass-houses. 1.3 Another significant error was that sub-stations benefitted much too little compared to major stations. What was overlooked was that major research stations need improved research facilities at the minor stations for improved results of out-reach trial sites. 2. DAR Re-Organization 2.1 With the exception of the composition of the Technical Services group, that of the remaining 6 commodity groups is acceptable. As originally conceived, the grouping of the research programmes into groups of seven has turned out to be very acceptable. The only notable flaw is that the grouping has reduced the interaction between different commodity team scientists which, at a departmental level is not ideal. Scientists in the DAR should interact more and promote the cross-fertilizations of ideas. 2.2 While multi-disciplinary approach was emphasized under NARP and thereafter, commodity team scientists in pathology, entomology and the like (Plant Protection) having been included in the Technical Services group, appear isolated from the commodity in which their research activities exist. For few selected commodity teams, a full package of scientists including breeders, agronomists and relevant plant protection scientists would be most ideal. 3. Master Plan and Action Plans 3.1 Following the agreements with the Bank, the DAR prepared the first and second editions of the Master Plan. Specific to commodities were the Action Plans which had also been prepared towards the end of the NARP. Although the second edition is not bound - it serves as a good guide for DAR on policy issues. What is clearly a guide to research programmes are the Action Plans which have been reviewed by the World Bank and are now being used by scientists. 4. Preparatory Exercises of the Project 4.1 As evidenced by the findings at the research stations, contractors of civil works were working in isolation of the end users of the work they were contracted for. In some cases, because of cost saving motives, some houses were built in wrong areas of the stations. Unfortunately they were not as flexible in their plans as would have been desired. This was further indicated by the resistance to improve on the dwelling houses which were later found unsuitable. Had there been consultations with station staff coupled with an in-built flexibility, some of the changes that were requested later on 22  would not have taken much of the extra money - e.g. an addition of a single window to a class C house to improve its ventilation. S. Training 5.1 This was a successful component of the project, especially with regard to the long-term type of training. It has transformed the trained manpower level of the DAR. 5.2 Unfortunately, the high expectations accruing from the high calibre trained manpower was not accompanied by improved working conditions. The final result was that rate of attrition increased which the department had to bear with. 5.3 The impact of such trained manpower has already started being felt in the DAR. There is a significant improvement in reporting - particularly in timeliness. The impact on crop and livestock production is difficult to assess now. One hopes however, that with improved research capabilities of the scientists, improved results will be obtained which should contribute to increased crop and livestock production. Comments on Part I of PCR Item 2 - Under Summary Recently initiated ASP is mostly concerned with the provision of operational funds. The project does not provide maintenance for the infrastructure procured under NARP. Item 5.11 While this is true, who else could have written the reports. Past experience showed that it is better to engage local scientists plan for the department. Item 5.12 Scientific soundness of research proposals is intended to be achieved at the annual project meetings Item 5.16 In addition to the absence of a formal review and perhaps because of this, most research contracts have not produced reports. Item S.17 The question of salaries and career structure cuts across the board in the Department, not just library staff. Item 5.21 Because DAR has about the same number of trained scientists as before the project, long term higher training programme under ASP is therefore justified for the remaining or newly recruited staff members. 23  PART m: STATISI'ICAL INFORMATION Table 1. Related Bank Loans and/or Credits Loan/Credit Title Purpose Calendar Year of Status Comments I I Approval I I (1) National Rural The project was designed November 1978 Closed The project included co- Development to assist GOM to June 30, 1986. financing from CIDA, EDF, Programme implement the first five- PCR dated UK and Fed. Rep. of NRDP Phase I year phase of the June 1987. Germany. (Credit 857-MAI) programme. Completion of Lilongwe Land Development Programme (LLDP). (2) National Rural Very similar components October 1981 Closing date Third IDA project to support Development to NRDP IV. Main September 1986 rural development in Karonga Programme objectives to increase extended to Agricultural Division (ADD) NRDP Phase ElV agricultural production, September 1988. in North Malawi. (Credit 1183-MAM) income and welfare of the PCR dated smallholder sector. It also May 1990. included forestry. (3) National Rural The project included April 1983 Closed Fourth IDA project to support Development components to strengthen April 30, 1991. rural development in Programme and support extension PCR dated Lilongwe ADD. Extension NRDP Phase IV services, training and April 1991. services satisfactory and an (Credit 1343-MAI) adaptive research and effective mechanism for the made provision for issue and recovery of seasonal improved credit and credit was established. marketing facilities. (4) Agricultural Improving extension work September 1985 Closed The project formed part of Extension and nationally and research- June 30, 1993. the fifth phase of NRDP. It Planning Support extension linkages. PCR dated succeeded in changing Project February 1994. extension methodology but the (Credit 1626-MAI) impact on improved staff training, technical competence and widening farmers' contracts had not yet occurred at the time of PCR. (5) Agricultural Services Designed to help formulate 1993 Under Total project cost: US$ 79.2 Project and implement Institutional implementation. million. Co-financing (Credit 2514-MAI) and Management reforms Launched in provided by IFAD and in Research, Extension and December 1993. African Development Bank. Input Supply Systems. NRDP 11 was a Wood Energy Project. 24  Table 2. Credit Timetable Item Date Planned Revised Date Actual Date - Identification' - 1981 - Preparation2 - September 1983 - Appraisal3 - March 1984 - Credit Negotiations January 7-11, 1985 - January 7-11, 1985 - Board Approval February 19, 1985 - February 19, 1985 - Credit Agreement May 23, 1985 - May 23, 1985 - Credit Effectiveness November 19, 1985 - November 19, 1985 - Credit Closing October 31, 1990 April 30, 1991 October 31, 1993 October 31, 1992 I Following the 1981 NRDP review. 2 By the Department of Agricultural Research (MOA), following an overview of Malawi's agricultural research system carried out in April 1982 by ISNAR. The World Bank's Regional Mission in Eastern Africa (RMEA) undertook regular reviews during preparation. 3 With USAID participation at selected meetings. 25  Table 3. Credit Disbursements Cumulative Estimated and Actual Disbursanenis (IDA) IDA Fiscal Year 1985/86 1986/87 1987/88 1988/89 1989/90 1990/91 1|91/92 1992/93 | 1993/94 ...................................... (US$,million)...................................... A. In USS Appraisal Estimates 1.8 7.3 14.9 20.5 23.0 23.8 23.8 23.8 23.8 Actual 1.5 4.5 7.0 10.3 14.8 17.6 19.6 30.2 32.1 Actual % of Estimats 83 62 47 50 64 74 82 127 135 ....................................... (SDR'million).. B. In SDR Appraisal Estimates' 1.8 7.5 15.2 20.9 23.5 24.3 24.3 24.3 24.3 Actual 1.3 3.7 5.6 8.1 11.6 13.5 15.0 22.7 24.0 Actual as % of Estimates 72 49 37 39 49 56 62 93 99 Date of Final Disbursement: March 14, 1994: SDR 160,751 (US$ 225,153.62) SDR 230,200 was cancelled. Mission's Estimates: Figures in SDR were not given in SAR.  Table 4. Project Implementation (Page 1) Indicators Appraisal Estimates | Actual (or PCR Estimates) A. Civil Works' Adm./Offices Blocks n.a. 10 Library 820 m2 4 Laboratory Blocks n.a. 12 Greenhouses n. a. 3 Glasshouses n.a. 62 Houses: - class C 47 - class D 115 41 - class E J 63 Houses Rehabilitation 62 n.a. Crop Sheds n.a. 6 Stores n.a. 6 Dispensary 1 Other n.a. 43 B. Relocation of Applied To be relocated by end PY 4 Impeded by the Training Research Staff Programme but most Conmmodity Teams were satisfactorily staffed at project end. C. Establishment of ART's - HQ PY I PY 1 - LADD PY I PY 1 - KADD PY I PY 1 - BLADD PY I PY 1 - MZADD PY 3 PY 2 - SLADD PY 3 PY 3 - NADD PY 4 PY 4 - KRADD PY 5 PY 5 D. Master Plan PY 2 PY 2 (draft report) PY 4 (lst Edition) PY 6-PY 8 (Revised Editing) See details in Annex 1, Table 3. 2 Including one quarantine glasshouse. 3 Including workshops, insectary and access road. 27  Table 4. Project Implementation (Page 2) Indicators Appraisal Estimates Actual (or PCR Estimates) E. Technical Assistance ARC Adviser By effectiveness PY I P&B Specialist/Fin. Controller By effectiveness PY 1 Procurement Specialist By effectiveness PY 1 Librarian By effectiveness PY 1 Training Adviser July 1985 October 1986 F. Staff Evaluation Process PY 1 Not Achieved G. Scientific Career Stream PY 3 Not Achieved H. Training Malawi Counterparts Financial Controller September 1985 June 1987 Librarian September 1986 June 1987 I. Publications - Agricultural Handbook of Malawi PY 2 Considered unnecessary duplication of comprehensive guidelines revised and published annually. - Annual Report PY I Annual Commodity Team reports produced regularly but DAR consolidated reports years out of date. J. Workshops and Reviews - Annual Planning & Reviews . DAR PY 1 PY I . ARC PY I PY 1 - Triennial Review PY 3 PY 4; PY 7 - Nat. Res. - Ext. Workshops PY 1 PY I - ADD Res. - Ext. Workshops 6/annum/ADD Not imnplemented K. Contract Research First Contracts to University PY 1 PY 5 Coffee Research Facility PY 2 PY 3 L. Library Utilization System Established PY I PY 2 Cataloguing Completed PY 2 Not completed (Target was unrealistic) 28  Table 4. Project Implementation (Page 3) Indicators T Appraisal Estimates T Actual (or PCR Estimates) M. Vehicles Pick upl4WD n.a. 47 Saloon cars n.a. 10 Bus n.a. 1 Trucks n.a. 12 Motorcycles n.a. 42 N. TA Persons/Months - Long-term . IDA 264 244 . USAID 273 381' - Short-term . USAID 57 40 (Used for two triennial reviews and Short-term T.A. for Master Plan priority assessment and DAR restructuring) 0. Training (i) Degree Training - PhD (3 years) 6 132 - MSc (2 years) 38 313 (ii) Short-term Training (person/weeks) - In-country research 3,102 2,5004 - Training abroad 1,728 3745 (iii) Study Tours 1 Including 2 training advisers for an estimated 100 person/months. 2 Of which 4 financed by USAID. 3 Of which 27 financed by USAID. 4 Total of 715 persons were trained. 5 Of which 104 financed by USAID. 29  Table 5. Project Costs and Financing A. Project Costs SAR Estimates Revised Actual" 2 l l I ~~~~~~~~~~~~~~~~~~~~~~~~~~~~Estimates' Local Costs j Foreign Exchange j Total ............................. (US$'000) . 1. Civil Works 6,633 4,277 10,910 9,480 9,322 2. Staff Relocation 53 51 104 141 115 3. Equipment 2 Vehicles 23 3,840 3,863 4,962 5,245 4. T.A. & Training 407 16,700 17,107 16,3723 15,5713' 5. Recurrent Costs 11,144 6,770 17,914 19,4064 19,5984' 6. Refunding of Project Preparation Advance - - - 326 313 7. Special Account - - - 568 8. Eligible Expend. for Incremental Seasonal Credit - - - 8,317 8,010 9. Unallocated - - - 212 Total 18,260 31,638 49,898 59,784 58,174 Figures in Local Costs and Foreign Exchange Costs not available. 2 Actual Disbursement as provided by World Bank. 3 Including US$ 9,220,000 USAID Grant. Total USAID Grant of US$ 15.0 million was available for both NARP and MAEPS projects. A total of US$ 14.8 million was actually disbursed for both projects but not break down was available. It is therefore assumed that US$ 9.2 million were disbursed for NARD in line with SAR's estimates. ' Including US$ 16,884,000 as GOM's contribution to cover ongoing DAR programmes.  Table 5. Project Costs and Financing B. Project Financing SAR Estimates' | Revised2 Actual3 Actual as a % l l | of SAR Estimates ................. .(US$'OO) . A. IDA Financing 1. Civil Works 9,390 9,480 9,322 99.3 2. Staff Relocation 100 141 115 115.0 3. Equipment & Vehicles 3,360 4,962 5,245 156.1 4. Technical Assistance and Training (a) Other than (b) below 5,766 4,851 4,818 83.6 (b) Contract Research 980 2,301 1,533 156.4 5. Oper. Costs of new Activities 1,195 2,522 2,714 227.1 6. Refunding of Project Preparation Advance 587 326 313 53.3 7. Special Account 401 568 - - 8. i. Unallocated 2,015 - - 8.ii. Eligible Expend. for Incr. Seasonal Credit - 8,317 8,010 9. Unallocated - 212 Total IDA Financing 23,794 33,680 32,070 134.8 B. USAILD Financing' 9,220 9,220 9,220 100 C. Government Contribution 16,884 16,8845 16,8845' 100 Total Project Financing 49,898 59,784 58,174 116.6 As given in SAR; IDA Financing as per DCA amounted to SDR 24.3 million equivalent as indicated in part C below. 2 For IDA rinancing, figures are derived from the amended DCA expressed in SDR and converted into US$ using the exchange rate of 1 SDR = 1.336 (Average 1992). 3 IDA financing derived from actuAl World Bank disbursement equivalent to SDR 24.0 million (US$ 32.1 million) as indicated in part C below. 4 Mainly for Technical Assistance and Training. s Actual figures not available: assumed to be the same as SAR estimate.  Table 5. Project Costs and Financing C. tDA Financing Planned' Revised' Revised3 Actual Actual as a % Categories Value Value Value Value of Planned (SDR'000) (SDR'000) (SDR'000) (SDR'000) I. Civil Works 9,590 39.5 9,590 39.5 6,840 28.2 6,977.4 29.1 72.8 2. Staff Relocation 102 0.4 102 0.4 102 0.4 85.9 0.4 84.3 3. Equipment & Vehicles 3,430 14.1 3,430 14.1 3,580 14.7 3,925.8 16.4 114.5 4. Consultants and Training (a) Other than (b) below 5,890 24.2 5,890 24.2 3,500 14.4 3,606.2 15.0 61.2 (b) Contract Research 1,000 4.1 1,000 4.1 1,660 6.8 1,147.1 4.8 114.7 5. Oper. Costs of new Activities 1,220 5.0 1,820 7.5 1,820 7.5 2,031.0 8.5 166.5 Introduced by Project 6. Refunding of Project Preparation Advance 600 2.5 600 2.5 235 1.0 234.6 1.0 39.1 7. Special Account 410 1.7 410 1.7 410 1.7 - - - 8.i. Unallocated 2,058 8.5 1,458 6.0 - - - - 8.ii. Eligible Expend. for Incr. Seasonal Credit - - - - 6,000 24.7 5,994.8 25.0 9. Unallocated - - - - 153 0.6 - - - Total IDA Financing 24,300 100.0 24,300 100.0 24,300 100.0 24,002.8 100.0 98.8 Original DCA. Approved in by the IDA in September 1992. Amended DCA.  Table 6. Project Results A. Direct Benefits SAR Estimates PCR Estimates SAR assumptions that comprehensive benefits could accrue soon, were generally optimistic. i. Inputs Optimize the use of inputs. Research, on request, proved that high content (HC) fertilizers were as effective per unit nutrient as low content materials and developped recommendations for HC use. The switch to these fertilizers, based on a comparison of the fertilizer mix imported by SFFRFM in 1987/88 and in 1992/93 saved the country some US$ 14.4 million in 1992/93 (Annex 1, Table 2). ii. Improved Varieties Selection of improved varieties of The release of flint maize hybrids should important crops. Resistance to bring about substantial benefits by way of pests and diseases. Good storage higher productivity, and higher returns to qualities. fertilizers and labour. iii. Crop Husbandry Systems To ease major constraints of the The switch to hybrid maize would enable small farmer such as labour farmers to produce as much as before from a shortages at peak periods. smaller sown area. iv. Range of Options To enable farmers to maximize Flint hybrids maize sould serve as an for Small-scale Farmers return to labour and land important step in raising returns from every resources and increase income. resource applied or applicable to farmers. v. Land Improved farming systems for No special programmes for this objective underutilized land. were launched other than irrigation related trials in the Shire Valley which are at an early stage. vi. Natural Resources Conservation and improvement of Dynamic Agro-forestry Research has led to the natural resources of the findings that are now in the large-scale field country (soil, water and forestry). trial stage and some measures, notably planting Feidherbia albida as bio-fertilizer method and vetiver grass for erosion control, are in the implementation stage and form part of extension workers' advisory programmes. vii. Number of Beneficiaries n.a. At least 250,000 farm families to date, and likely to benefit the majority of Malawi's farmers in future. However, under the Short- term Credit assisted programme introduced in 1992 as a result of the amendment of DCA, some SDR 6 million were made available to SACA. It is estimated that some 110,000 farmers benefitted under this programme (Annex 1, Table 1). B. Economic Impact Not Applicable C. Financial Impact Not Applicable 33  Table 7. Status of Covenants (Page 1) Covenants Subject Status Section 3.01(b) Borrower to sustain its annual contribution GOM had difficulty in complying with this to DAR's operating expenditures at the requirement for some years. MOA made level in real terms reached during 1984/85. allocation as specified by IDA for 1992/93, but whether this can be maintained in future remains to be seen. Section 3.01(c) ARC Secretariat to be adequately staffed. The original proposals did not prove very satisfactory. The Borrower made some progress towards establishing a more effective and modified staffing arrangement. Section 3.01(d)(i) Borrower to employ a qualified and Initially MOA did employ an expatriate experienced procurement officer. procurement officer and did fully comply with this covenant. Following the end of the expatriate procurement officer's contract, the staffing of the procurement unit with Malawian officials has been very weak. Although new proposals were made, as a condition of extending the Closing Date, for strengthening the unit, they were only partially fulfilled. Section 3.01(d)(ii) DAR to appoint a deputy chief agricultural Complied with. research officer. Section 3.01(d)(iii) DAR to employ qualified and experienced Complied with. administrative officer at three main research stations. Section 3.01(e) DAR to introduce a scientific career stream This has not been complied with and because by April 1987. Government had difficulty meeting this requirement, IDA agreed to 'waive' this condition. Section 3.01(f)(i) DAR to prepare research master plan by DAR has prepared a research master plan. March 1988. The original 1988 version has been updated twice. Section 3.01(f)(ii) Borrower to implement the agreed research DAR is implementing a maize action plan, master plan. which is an important element of the overall research program; and has prepared other action plans for priority programs identified by the Master Plan. Section 3.01(g)(1) DAR to appoint counterpart financial This has been complied with. However, controller. accounting and financial planning and management, remained weak. The accountant after returning from graduate training and attending ISNAR training to introduce improved accounting and information systems, has left DAR for better pay elsewhere. 34  Table 7. Status of Covenants (Page 2) Covenants Subject Status Section 3.01(g)(ii) By April 1987 DAR to review with IDA the The expatriate left DAR for personal reasons need to extend the contract of the expatriate and his place was taken by a Malawian financial controller. accountant with IDA's concurrence, but the purpose for which the expatriate had been hired, to introduce a suitable research costing and information system was not attempted, and IDA should have insisted on employment of a replacement. Section 3.02 DAR to employ consultants satisfactory to Complied with. IDA using acceptable selection procedures. Section 3.02(a) Goods to be insured in transit. Complied with. Section 3.03(b) Goods to be used exclusively for the Complied with. project. Section 3.04(a) Borrower to provide IDA with project Complied with. plans, specifications, etc. Section 3.04(b)(i) Maintain satisfactory records. Complied with. Section 3.04(b)(ii) Enable IDA staff to have free access to the Complied with. project. Section 3.04(b)(iii) Regular reporting and review of project Routine project progress reports have usually progress. been submitted but were often late. The annual reporting requirements require DAR to submit to IDA a review of the previous year's research and a prioritized research programme for the following year. This was complied with for the first time in 1991, although the format of the report was not satisfactory. TA has since been used to improve the format - also an Editorial Committee has been established for that purpose. The first triennial research review was completed to IDA's satisfaction; the second took place punctually. Section 3.04(d) Borrower to prepare project completion Part II of PCR is to be prepared by the report. borrower. Section 3.05 Borrower to provide all land needed for Complied with. project. Section 4.01(a) Borrower to maintain satisfactory records Complied with, although finalization of and accounts. accounts frequently delayed. Section 4.01(b) Borrower to maintain adequate records for Compliance believed to be satisfactory. expenditure claimed against statements of expenditure. Section 4.01(c) Borrower to have project accounts audited. The audit has frequently been delayed. Section 4.02 Borrower to maintain project installations DAR has had difficulty complying with this adequately. covenant due to its limited operating budget. 35  Table 8. Use of Bank Resources A. Bank Staff Inputs Planned Actual . . . . . . . . . .(Staff-w eeks) . Preparation n.a. 23.6 Appraisal n. a. 32.2 Negotiations n. a. 4.7 Supervisions n. a. 226.4 PCR n. a. 0.3 Total n.a. 287.2 36  Table 8. Use of Bank Resources B. Mission Data Stage of Project Cycle | Monthi/Year Number of Specialization Days in Field2 Performance Rating | Typcs of l j Persons T Represented' Status3 Problems4 Pre-preparation n.a. n.a. n.a. n.a. Preparation n.a. n.a. n.a. n.a. Appraisal February/March 84 4 n.a. n.a. Final Appraisal June/July 84 4 n.a. n.a. Supervision I October 85 2 R, E 9 Pre-effectiveness Supervision 2 February 86 2 R, T 12 1 Supervision 3 June 86 2 R, E 11 I Supervision 4 February/March 87 2 R, A 20 (10) 2 PM Supervision 5 October/November 87 45 A, A, E, R 26 (10) 2 PM Supervision 6 May/June 88 55/ A, A, E, R, AR 26 (7) 2 PM Supervision 7 October/November 88 35' A, A, E 31 (8) 2 AF Supervision 8 February/March 89 3 A, E, R 31 (10) 2 AF, PP w Supervision 9 November/December 89 25' A, A 25 (10) 2 AF, PP Supervision 10 June/July 90 1 E 26 (8) 3 DO, LC, AF, PP Supervision 11 December 90/January 91 2 E, R 30 (11) 2 AF, PP Supervision 12 April/May 91 ,3 A, R, P 22 (7) 2 PP Supervision 13 October/November 91 35/ A, A, T 26 (6) 2 LC, PP Supervision 14 June/July 92 35/ A, A, R 28 (6) 2 PP Supervision 156 May/June 93 2 A, R, 18 (9) n.a. 6 n.a.6/ 'Specializations are: R = Research Specialist; A = Agriculturist; E = Economist; T = Training Specialist; AR = Architect; P = Procurement Specialist. 2 From the forth Supervision mission, the World Bank combined the supervision of other IDA assisted projects; e.g. MAEPS, NRDP III, NRDP IV, Agricultural Marketing and Estate Project, Dowa West Rural Development Project, Kasungu Agricultural Development Project and Smallholder Agricultural Credit Project. Figures in brackets are attributed to NARP. 3 Performance rating status: I = Problem free or minor problems; 2 = Moderate problems; 3 = Major problems. 4PM = Project Management; AF = Availability of Funds; Pr = Procurement Progress; DO = Development Objectives; LC = Compliance with Covenants. Including consultants. 6 No Form 590 was prepared.  MALAWI: Nationai Agricultural Rescarch Project (Credit 1549-MAI) - Project Completion Repon Part Ill: Annex 1: Seasonal Credit Recovery for 1992/93 Season as on 03/12/93 Amon.unt Anmout Amount Borrowers ADD Loaned Recovered X Recovery Batance Clubs | Paid Off Not Start Out Starding Beneficiaries (MHK'000) (K('000) X( . . . . . . . . . . . . K'000)........ . (No.) ................. ..... ........... KRADD 1,772 746 42.08 1,026 610 99 167 511 6,779 MZADD 15,739 1,675 10.64 14,064 2,002 70 700 1,932 38,384 KADD 43,892 4,066 9.26 39,826 3,772 68 2,620 3,704 95,931 LADD 48,001 6,193 12.90 41,808 3,571 43 2,406 3,528 116,419 SLADD 8,076 714 8.84 7,362 1,151 35 698 1,116 24,559 LWADD 14,316 3,874 27.06 10,442 2,001 280 733 1,721 50,125 BLADD 10,963 1,901 17.34 9,062 1,786 113 635 1,673 40,040 NADD 1,528 845 55.34 683 741 138 99 603 11,981 Total 144,287 20,014 13.87 124,273' 15,634 846 8,058 14,788 384,218' 00 Source: SACA. ' Of which MK 36 million (SDR 6 million) from IDA. 2 Of which 110,000 benefitted from the IDA assisted programme.  MALAWI: National Agricultural Research Project (Credit 1549-MAI) - Project Completion Report Part IlIl: Annex 2: Switch of Fertilizer Mix: Estimated Savings in Fertilizer Imports Expenditures in 1992/93 Fertilizer (Title) Quantities' Nutrient Content. Prices 1992/93 of Values of Fertilizers 1987/88 1992/93 ~~~~~Fert iIi zerS2 III 11987/88 1992/931987/88 12/93 1987/88 1992/93 . . . . . . . . . . . . . . . . . . (Tons) . . . . . . . . . . . . . . . . . . (MK/kg) . . . . . . (MK million) . . . . . S/A (21%) 8,500 - 1,785 - 1.000 8.5 - CAN (26%) 30,000 25,000 7,800 6,500 1.390 41.7 34.8 20:20:0 (40%) 20,000 - 8,000 - 1.300 26.0 - 23:21:0:4S (44%) 5,000 18,751 2,200 8,250 1.400 7.0 26.3 UREA (46%) 9,000 55,000 4,140 25,300 1.216 10.9 66.9 DAP (64%) 5,000 37,500 3,200 24,000 1.323 6.6 49.6 Total 77,500 136,251 27,125 64,050 100.7 177.6 Average Price of Fertilizer in 1987/88: MK 1.3/kg (MK 100,700,000: 77,500,000) N.B. in 1992/93 terms. Average Price of Fertilizer in 1992/93: MK 1.3/kg (MK 177,600,000:136,251,000) NP.. in 1992/93 terms. '.0 Average Price of Nutrient in 1987/88: MK 3.712/kg (MK 100,700,000: 27,125,000) N.B. in 1992/93 terms. Average Price of Nutrient in 1992/93: MK 2.7m7/kg (NK 177,600,000: 64,050,000) N.B. in 1992/93 terms. Total Savings in Fert. lIports Expend. in 1992/93 as coupared to 1987/88: ((64,050,000 x 3.712) - (64,050,000 x 2.773)] = MK 60.2 million (USS 14.4 million) 1 Source: MOA (SFFRFM Fertilizer Imports). 2 Source: MOA.  MALAWI: National Agricultural Rescaerh Project (Credit 1549-MAI) - Project Completion Report Part III: Amex 3: CiviL Works Number Research | Adm. Library Laboratorie Houses Crop Access of Stations Blocks/ Blocks s Blocks Greenhouse Glasshouses Sheds Hostet Dispensar IJorksho Stores Insectory Roads Contract Offices s Class C|Class DIClass E Y P 1 Chitedze/Makoka 2 2 3 2 22 16 - 2 1 1 - - - - (8 rooms) 2 Chitala & Salima 2 - 2 - 6 2 10 - 2 - 1 - - 3 Lunyangwa, 3 1 4 1 11 15 2 2 - - 1 2 Mkondezi, Karonga, Ntchenachena, Baka 4 Kasinthula, 2 - 2 - 2 - - 1 - - - 2 - Ngabu, mimosa 5 Bvumbwe, 1 1 1 - 6 8 2 - - - - 2 1 1 Liwonde, (1 quarantine) Blantyre Total 10 4 12 3 47 41 6 6 1 1 1 6 1 1 Source: Norman & Dawbarru (Contractor).