Madagascar - National Agricultural Research Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 17588 IMPLEMENTATION COMPLETION REPORT MADAGASCAR NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 2042-MAG) March 31, 1998 Technical Agriculture Cluster II Country Department 8 Africa Region |This document has a restricted distribution and may be used by recipients only in the| performance of their official duties. Its contents may not otherwise be disclosed without |World Bank authorization.  IMPLEMENTATION COMPLETION REPORT MADAGASCAR NATIONAL AGRICULTURAL RESEARCH PROJECT (Credit 2042-MAG) Currency Equivalents Malagasy Francs (FMG) per US$ 1 1989 l 1990 l 1991 l 1992 l 1993 l 1994 l 1995 1996 1997 l 1,603 1,494 1,835 1,863 1,913 3,067 4,265 4,053 5,101 Abbreviations and Acronyms CIRAD Centre de Cooperation Internationale en Recherche Agronomique pour le DOveloppement COFIRA Steering Committee for National Research ESSA Ecole Superieure des Sciences Agronomiques FIFAMANOR Highlands Agricultural Development Project FOFIFA National Center for Applied Research for Rural Development FSR Farming Systems Research GOM Government of Madagascar IARC International Agricultural Research Center IRRI International Rice Research Institute ISNAR International Service for National Agricultural Research MRS Ministere de la Recherche Scientifique MS Master of Science NARP National Agricultural Research Project NGO Non-Government Organization ORSTOM Office pour la Recherche Scientifique et Technique d'Outre-Mer PhD Doctor of Philosophy PIP Public Investment Program SAR Staff Appraisal Report Weights and Measures Metric System Fiscal Year of Borrower January I - December 31 Vice President Callisto Madavo Director Michael Sarris Sector Manager Joseph Baah-Dwomoh Staff Member Yves Wong  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT MADAGASCAR NATIONAL AGRICULTURAL RESEARCH PROJECT (Credit 2042-MAG) TABLE OF CONTENTS PREFACE ........................................................ i EVALUATION SUMMARY ......................................................... j PART 1: PROJECT IMPLEMENTATION ASSESSMENT ........................................................ I A. INTRODUCTION ........................................................ I B. PROJECT OBJECTIVES ........................................................ 2 C. ACHIEVEMENT OF PROJECT OBJECTIVES ................................... 3 D. MAJOR FACTORS AFFECTING THE PROJECT ................................., 6 E. PROJECT SUSTAINABILITY.7 F. IDA PERFORMANCE ........................................................ 8 G. BORROWER PERFORMANCE ........................................................ 8 H. ASSESSMENT OF PROJECT OUTCOME ............ ....................8........ ........... 8 1. FUTURE OPERATION .........................................................11 J. KEY LESSONS LEARNED ......................................................... I PART II: STATISTICAL TABLES ........................................................ 12 TABLE 1: SUMMARY OF ASSESSMENTS ........................................................ 12 TABLE 2: RELATED BANK CREDITS ......................................................... 13 TABLE 3: PROJECT TIMETABLE ........................................................ 14 TABLE 4: CREDIT DISBURSEMENTS: CUMULATIVE ESTIMATED AND ACTUAL ............................................... 14 TABLE 5: KEY INDICATORS FOR PROJECT IMPLEMENTATION.............. ...... 15 TABLE 6: KEY INDICATORS FOR PROJECT OPERATION ........................................................ 18 TABLE 7: STUDIES INCLUDED IN PROJECT ................................ ,18 TABLE 8A: PROJECT COSTS ............................... 19 TABLE 8B: PROJECT FINANCING ............................... 19 TABLE 9: ECONOMIC COSTS AND BENEFITS ............................... 20 TABLE 10: STATUS OF LEGAL COVENANTS ................................................... 21 TABLE 11: COMPLIANCE WITH OPERATIONAL MANUAL STATEMENTS ................................................... 23 TABLE 12: BANK RESOURCES: STAFF INPUTS ................................................... 23 TABLE 13: BANK RESOURCES: MISSIONS ............................................................. 4 ANNEXES Annex 1: Borrower's Evaluation of the Project ............................................................. 25 MAPS IBRD Map No. 21261 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.  i IMPLEMENTATION COMPLETION REPORT MADAGASCAR NATIONAL AGRICULTURAL RESEARCH PROJECT (Credit 2042-MAG) PREFACE This is the Implementation Completion Report (ICR) for the Madagascar National Agricultural Research Project, for which Credit 2042-MAG in the amount of SDR 18.6 million (US$24.0 million equivalent) was approved on June 15, 1989, and made effective on July 12, 1990. The credit was closed on December 31, 1997. The final disbursement is forecast to take place by April 30, 1998. In April 1994, the project was restructured, and SDR 9.5 million (US$12.3 million equivalent) was canceled. The ICR was prepared by Yves Wong of the World Bank Madagascar Resident Mission and Eavan O'Halloran, Agricultural Technical Cluster II of the Africa Region. It was reviewed by Joseph Baah-Dwomoh, Technical Manager, Agriculture Technical Cluster II and Michael Sarris, Director, Country Department 8. Preparation of the ICR began after the Bank's completion mission in January 1998. The ICR is based on the results gathered during this completion mission, on material in the project files and on discussions with the Government of Madagascar and the beneficiaries of the project. The Borrower contributed to the preparation of the ICR by preparing its own evaluation of the project (Annex 1). The ICR was discussed with the Borrower on January 28, 1998, and the Borrower's comments were subsequently incorporated into the report.  ii IMPLEMENTATION COMPLETION REPORT MADAGASCAR NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 2042-MAG) EVALUATION SUMMARY Introduction 1. Agricultural development in Madagascar in the 1970s and 1980s was hampered by the absence of a sound agricultural research base. The departure in 1974 of the large number of separate French-based institutes, which had previously represented agricultural research in the country, had left behind a weak national research system, which was an under-funded skeleton structure that carried out poor quality research. In 1983, at the request of the Government of Madagascar, the International Service for National Agricultural Research (ISNAR) critically evaluated Madagascar's agricultural research effort and concluded that a more efficient, relevant and productive system, which was responsive to farmer needs, was urgently needed. 2. In June 1988, with the assistance of ISNAR under IDA financing, FOFIFA (the National Center for Applied Research for Rural Development) research staff prepared a seven-year National Agricultural Research Project (NARP) which was discussed in September 1988 by an IDA Pre-Appraisal Mission. In December 1988, the National Agricultural Research Project was formulated, which constituted the first 7- year phase of a 15-year prograrn to strengthen national agricultural research in Madagascar. Project Objectives 3. The project aimed at assisting the Government of Madagascar in achieving one of its major objectives of improving the incomes of the majority of its population, including its poorest segments, by the expansion and strengthening of its national agricultural research programs, projects and facilities. Specifically, the project included four major components: (a) institutional strengthening; (b) research quality and adaptability improvement; (c) human resource development; and (d) rehabilitation of physical plant and equipment. 4. The objectives supported by the project were broadly consistent with the needs of the agricultural research services in Madagascar. Project benefits were expected from: (i) accelerated agricultural productivity and growth; (ii) higher returns to farmers and increased employment; and (iii) increased foreign exchange earnings. However, the project faced several risks including: (i) inadequacy of budgetary funding levels and mechanisms; (ii) weak FOFIFA organization and management; (iii) lack of agricultural scientists; (iv) non-adherence to priorities and balance in research programs; and (v) inadequate relevance and adoption of research findings. To reduce these risks, the project included specific measures, backed up by conditionality, with respect to Government funding, donor coordination, linkage of salaries to performance, and measures to improve research/extension linkages.  Project Costs and Financing 5. At appraisal, the project was expected to be implemented over seven years and to cost US$70.6 million, of which IDA was to finance US$24.0 million, Government US$30.5 million, other donors US$10.4 million and FOFIFA US$5.7 million. However, both the project sco]pe and cost, and the contributions of the Government and IDA, were over-estimated and the final total project cost (US$35.9 million) amounted to slightly over half of appraisal estimate. When the project was restructured in April 1993, the amount of US$12.3 million was canceled. At credit closing on December 31, 1997 (except for Category 1, which was extended to April 30, 1998), the amount of US$2.8 million equivalent was still undisbursed, but this is expected to be fully disbursed at the end of the disbursement grace period on April 30, 1998. Donor contributions, which financed specific research programs under parallel financing arrangements, were fully disbursed. Implementation Experience and Results 6. From its inception, the project experienced problems. Effectiveness was delayed by seven months until the statutes of FOFIFA were amended. Despite this delay, the project management succeeded in launching some activities: most key positions were satisfactorily filled; the 1991 workplan and budget were approved; and research programs were launched for rice and other crops, forestry and farming systems. The enactment of FOFIFA's new statutes effectively protected the agency from interference by the Tender Board. However, the project suffered from lack of counterpart funds, and relationships were poor between the Ministry of Scientific Research and FOFIFA. 7. The volatile political situation in the early 1990s affected project implementation, and the Bank in early 1993 rated the project Unsatisfactory in all mandatory rating categories, except for project management performance which was rated Satisfactory. It requested a reappraisal of the project and recommended that disbursements be suspended unless substantial progress was achieved by June 1993. In response, FOFIFA endeavored to fulfill all its agreements with the Bank and took the lead in preparing the project restructuring, including consultations at all levels of the institution. 8. A subsequent Bank mission in August 1993 agreed with FOFIFA on the extent of the restructuring required, which included prioritization and planning of project activities to fit FOFIFA's implementation capacity according to Public Investment Program commitments. These effectively set limits to the recruitment of technical and research personnel and to the creation of new research stations. Coordination of research and extension was made a conditionality of disbursement. A general reduction in project scope was also planned, and changes reflecting Government's regionalization policy were introduced, making the regional centers more responsive to economic and environrnental constraints of producers, and promoting synergy between research and extension efforts and relationships with local NGOs. The prioritization process resulted in cost savings and, as a result, an amount of SDR 9.5 million (US$12.3 million) was canceled from the credit in April 1994. 9. The restructuring improved project implementation significantly. Working relationships between the Ministry of Scientific Research and FOFIFA were improved, and contacts between research and extension staff were strengthened. The regionalization program of research centers was pursued, and regional multi-disciplinary research teams were set up at the centers. For the first time, the Government and donors were presented with a coherent and integrated picture of the overall research effort, which reflected the objectives of a national research program.  iv 10. In 1994, the first research results began to be extended to farmers, including new higher-yielding and more adaptable varieties of rice and other crops, more efficient cultural techniques, and improved tools and farm equipment. Integration with the extension service increased. The impact of research outputs is continuously monitored and evaluated jointly with the extension services, and results indicate a significant impact on production of rice, peanuts, cotton and other crops. 11. In 1995, the project was rated highly satisfactory for development objectives and satisfactory for implementation progress. The project achieved a much higher level of performance as the new Minister of Scientific Research was very supportive of the project and of its staff. 12. During 1996, however, the project continued to suffer counterpart funding problems and financial management difficulties, but was still rated satisfactory for development objectives and for implementation progress. Research developments under the project began to accrue and significantly raised crop yields, increased farmer income, and reduced environmental degradation. In the north-west, the development and release of virus-resistant rice varieties enabled farmers, who had abandoned their lowland paddies to encroach on highland forest land to grow rice, to return to them and thereby spare further forest degradation. A newly-released cotton variety in 1996 virtually covered the whole of the cotton areas in the south west and northwest one year after its release, with a 30 percent increase in yield over the previous varieties. Bank and Borrower Performance 13. The Government did not always cooperate with FOFIFA, and nor did it take ownership of the project during the first few years of implementation. It did not provide adequate budgets to FOFIFA, and the project continually suffered from lack of counterpart funds. There were poor relations between the Government and FOFIFA, and between the Government and the Bank. For its part, the Bank can be criticized for designing an overly-ambitious project and for not conducting a proper assessment of FOFIFA's capacity. However, performance of the Government and the Bank improved during the life of the project. The Government and FOFIFA took some tough and sensitive personnel decisions in the second half of the project which helped to turn it around. The Bank performance during supervision was satisfactory, and supervision of the project was intensified when the Madagascar Resident Mission assumed responsibility for the project in late 1996. The Bank and FOFIFA reacted quickly in carrying out the successful restructuring of the project. Assessment of Outcome 14. The overall assessment of the project's outcome must be rated as satisfactory. The project has satisfactorily achieved its overall objectives of laying down the framework for an effective decentralized research capacity. The objective of creating a regional network of eight research centers has been achieved. These centers are staffed with a core of motivated young researchers who work under satisfactory physical conditions. The institutional impact of the project is also satisfactory: FOFIFA's organization has been rationalized, and its staff numbers trimmed and decentralized. The retrenchment of over fifty percent of its staff has significantly reduced FOFIFA's annual recurrent budgetary needs and made FOFIFA become more financially sustainable. The target for post-graduate training of researchers has been fully achieved. Both female farmers and female FOFIFA staff have benefited under the project as FOFIFA now has many female researchers, some at the highest level of management at director level. Research program planning has been rationalized, and programs prioritized through a participatory approach involving farmer groups and extension services. There is greater accountability of researchers  v 15. Impact on Productivity, Growth and Farmer Income: The project has resulted in the development and dissemination of an important array of improved crop varieties with higher yields, greater resistance to local pests and diseases, and better adaptability to local conditions. Superior crossbreeds of cattle have been produced. Other developments include improved cultural techniques and more efficient tools and equipment. All these have had a significant impact on agricultural production and farmer income. The degree of adoption of improved varieties will increase in time as farmer confidence in the new technologies increases, and in addition, as the extension services increase their national coverage, an increasing number of farmers will have access to these technologies. Thus, the impact of research will increase with time. Summary Findings, Future Operations, and Key Lessons Learned 16. The most important finding of the report is that Government commitment is essential to the success of any project and that, without that commitment, the project had little chance of succeeding. This commitment is particularly valuable in periods of political turmoil. Closeness to clients is critical: the decentralization of the research system has led to the greater participation of all stakeholders, with improvements in the relevance and quality of research programs and adoption of research results. In the same way, transferring supervision responsibility to the Bank's Resident Mission helped support the project management to accelerate decision making and restructuring and hence improve project performance. Finally, the unconventionally long implementation time-frame was entirely appropriate for such an institution-building project. The key element in the project was the development of a critical mass of well-trained researchers which is a time consuming process. 17. Project sustainability is now probable because the investment phase for physical infrastructure and equipment is virtually complete, core teams of trained researchers are in place at most of the centers, and FOFIFA's financial situation has been significantly improved. The decentralizaltion of the research system, the institutionalized participation of farmers and extension services, the improved delivery of new technology through the national extension service, the good management of FOFIFA's resources, and a deliberate policy to contract with the private sector, are sound foundations of sustainability. 18. Government is committed to the implementation of the second phase of the Research Masterplan, which will focus on human resource development and on further strengthening the regional capacity to carry out high quality research. It has requested IDA support for implementation of this phase. Experience from the first phase has shown the need for research development to be well integrated with extension development but completion of the extension project will not take place before December 1999. Until then, the Government has agreed that Cr. 2729-MAG, which supports the IDA-financed National Agricultural Extension Project, will provide financing to FOFIFA to continue its operations for the next two years. 19. The main lessons learned from this project are: (a) Borrower commitment and ownership are essential at all times; (b) The responsibility for financing of agricultural research by the Government should be clearly defined, and funding should be assured based on this; (c) Stakeholder participation is essential for ensuring the relevance and responsiveness of research programs to farmer needs; (d) Decentralization of the research system is critical to its development; (e) A long implementation period is appropriate for this type of human resource development and institution building project;  vi (f) The development of well-trained researchers takes time; (g) Effective accounting and financial management systems are prerequisites of project implementation; (h) The national agricultural extension has not yet achieved a national coverage and is not yet able to provide a full delivery service for research results; (i) An effective and committed project management team combined with availability of staff at the Resident Mission are important elements in improving project performance and impact.  1 IMPLEMENTATION COMPLETION REPORT MADAGASCAR NATIONAL AGRICULTURAL RESEARCH PROJECT (Credit 2042-MAG) PART I: PROJECT IMPLEMENTATION ASSESSMENT A. INTRODUCTION 1. From the 1970s to the early 1980s, Madagascar's agriculture sector, which provided employment to 85% of the population, faced severe decline as a result of ill-conceived economic policies which involved Government interventions in the sector. Up until 1974, the country was self-sufficient in rice, its main staple food, but production steadily decreased thereafter to such an extent that by 1982 record amounts of rice had to be imported. Production of other crops also dropped. Changes of Government policies since 1983 helped to restore production to some extent as they set out to promote a free market system and improve incentives to producers, withdraw Government services from production and commercial activities, and concentrate resources on essential Government services, such as research and extension and rehabilitation and maintenance of rural infrastructure. 2. The absence of a sound agricultural research base in Madagascar affected agricultural development. The departure in 1974 of the large number of separate French-based institutes, which had previously represented agricultural research in the country, had left behind a weak national research system which was an under-funded skeleton structure that carried out poor quality research. It was characterized by: (i) a lack of a national strategy for agricultural research; (ii) low quality and poorly coordinated research programs; (iii) a severe shortage of qualified scientists, eighty percent of whom were based in the capital, and a surplus of low level non-scientific staff; (iv) a research organization based on top-down, disciplinary research carried out mainly on stations; (v) poor linkages with weak extension services; (vi) severe under-funding; and (vii) a centralized, outdated and poorly maintained physical infrastructure. 3. In 1983, the Government of Madagascar requested ISNAR (the International Service for National Agricultural Research) to evaluate the national agricultural research system. The ISNAR mission concluded that a more efficient and effective agricultural research system was urgently needed, in addition to improvements in other agricultural services. Such a system had to be responsive to farmer constraints and provide new technology to improve food security, increase and diversify export earnings, and increase small farmer income. Based on ISNAR's recommendations, the internal organization of FOFIFA (the National Center for Applied Research for Rural Development) was modified to include the establishment of a Research and Development Department whose functions were to rationalize research programs and strengthen linkages with extension. In June 1988, with the assistance of ISNAR under IDA financing, FOFIFA research staff carried out its own evaluation of the research system and prepared the first draft of a 15-year Master Plan for Agricultural Research. They subsequently prepared a first seven- year phase of the Master Plan which was discussed with an IDA Pre-Appraisal Mission in September 1988. This phase became NARP (the National Agricultural Research Project), which was thus the first half of a long term support program to strengthen national agricultural research in Madagascar.  2 B. PROJECT OBJECTIVES 4. The project aimed at assisting the Government in achieving one of its major development objectives of improving the incomes of the great majority of its population, which includes its poorest segments, and thereby reducing rural poverty and income disparities. This would be achieved by strengthening national agricultural research to develop improved technology and then apply this technology to increase agricultural productivity. The quality of research would be iimproved and made more relevant by the development of research staff and infrastructure, implementing demand-driven research programs, effectively decentralizing research to the regions, and strengthening links with extension. S. More specifically, the project was to: a) focus on priority multi-disciplinary and nationally coordinated research programs with a significant likelihood of high and early returns to investment; b) develop technologies that contribute to increased farmer income, in particular of smallholders, through a greater responsiveness of research to real farmer constraints; c) increase research efficiency through an increased emphasis on on-farm research, improvements in research quality, a reorganization of FOFIFA and decentralization of the research system, stabilizing research funding, and improving coordination with other donor programs; d) strengthen linkages between research and extension ranging from problem diagnosis to the monitoring of impacts; and e) ensure the long-term institutional viability of FOFIFA. 6. In support of these objectives, the project included four major components: (a) institutional strengthening, through the provision of adequate staff and budgets, and improvement of financial management and accounting, donor coordination, and linkages with universities and external research organizations; (b) research quality and adaptability improvement, through strengthening research design, linkages with extension, monitoring and evaluation of research, and adopting an integrated research approach; (c) human resources development, through establishing long term staff planning, education and training for scientific, technical and administrative staff, and providing incentives to staff to move to the regional centers; and (d) rehabilitation of physical plant and equipment, through rationalization of the research station network and establishment of 8 regional research centers. 7. Over a seven year period, the project would support nationally-coordinated research programs on crop and animal production, forestry and fish culture, agricultural technology, agricultural mechanization, on-farm irrigation, and farming systems research.  3 8. Evaluation of Project Objectives. The objectives supported by the project were consistent with the national goal of improving food security and reducing rural poverty by increasing farm productivity and diversification through better agricultural technology. However, the agricultural research system was totally ineffective at the time, and the very low technical base of agriculture was a major constraint to agricultural development. On the other hand, the objectives of developing the agricultural research services in Madagascar were a key element of the agricultural development strategies of both the Government and the Bank which were developed in a collaborative way by the Government, IDA, and other donors. Project benefits were expected from: (i) improved agricultural productivity and growth; (ii) higher returns to farmers; and (iii) increased foreign exchange earnings. However, the project faced several risks including: (i) inadequate budgets; (ii) weak institutional capacity of FOFIFA; (iii) resistance of staff to change to demand-driven, on-farm research; (iv) uncertainty on recruitment of qualified staff; (v) inappropriate research priorities and insufficient donor coordination; and (vi) inadequate relevance and adoption of research findings. To reduce these risks, the project included specific conditionalities with respect to Government funding, appropriate statutes for FOFIFA, rationalization of FOFIFA staffing, provision of financial incentives to research staff and to field staff in particular, donor coordination, and strengthening of research/extension linkages. 9. However, FOFIFA's institutional capacity was over-estimated at appraisal, while the difficulties of staff decentralization were under-estimated, although incentives were provided. The scope of the staff reduction program needed was under-estimated, and no provision was made to use the credit funds to pay for compensation. Finally, the project design over-estimated the financial scope of the project, as half of the IDA credit was canceled when the project was restructured. 10. Project Costs and Financing. At appraisal, the project was expected to be implemented over seven years and to cost US$70.6 million, of which IDA was to finance US$24.0 million, Government US$30.5 million, other donors US$10.4 million and FOFIFA US$5.7 million. However, both the project scope and cost, and the contributions of the Government and IDA, were over-estimated and the final total project cost (US$35.9 million) amounted to slightly over half of appraisal estimate. After two and a half years, only eight percent of the credit had been disbursed, which was due in large part to project inactivity during the time of political unrest. When the project was restructured in August 1993, the investment program was prioritized in line with research priorities and the implementation capacity of FOFIFA. The reduced project scope and activities meant that US$12.3 million was canceled in April 1994. At credit closing on December 31, 1997 (except for Category 1, which was extended to April 30, 1998), the amount of US$2.8 million equivalent was still undisbursed, but this is expected to be fully disbursed at the end of the disbursement grace period on April 30, 1998. Donor contributions, which financed specific research programs under parallel financing arrangements, were fully disbursed. C. ACHIEVEMENT OF PROJECT OBJECTIVES 11. The project has satisfactorily achieved its overall objectives of laying down the framework for a more effective, decentralized and sustainable research capacity, consistent with the objectives of the first phase of the national research masterplan. The physical objective of creating a regional network of eight research centers has been virtually achieved. The centers are staffed with a core of motivated young researchers who work under conditions of adequate physical facilities and operating budgets. The institutional impact of the project is also satisfactory: FOFIFA's organization has been rationalized, and its staff numbers trimmed and decentralized. About 60 percent of its total staff, mainly at non-scientific lower levels, have left under an incentive system, and a quarter of its researchers have been induced to move to the regional centers. The staff retrenchment program has significantly reduced FOFIFA's annual recurrent budgetary needs and made FOFIFA become more financially sustainable. Research program  4 planning has been rationalized by more thorough screening of priorities, implementation of a participatory approach which fully involves farmer groups and extension services in problem identification and search of solutions, and greater accountability of researchers. Significant results in the development and adoption of improved varieties and technologies are having an increasingly important impact in improving agricultural production and farmer income. 12. Implementation Experience. The credit agreement was signed on September 11, 1989 but effectiveness was delayed by seven months as the last covenant for credit effectiveness - the publication of the statutes of FOFIFA - was not achieved until July 1990. Despite this delay, the project management succeeded in launching some activities: most key positions were satisfactorily filled (three directors and six heads of research departments); arrangements for hiring technical assistance and starting training in biometrics were agreed; the Steering Committee for Agricultural Research (COFIRA) approved FOFIFA's budget and workplan in October 1990; research programs were launched for rice and other crops, forestry and farming systems, and programs in livestock were refined. The enactment of FOFIFA's new statutes was very effective as the agency's decisions on procurement were no longer subject to National Tender Board approval. This enhanced project management's accountability and minimized the risks of price rigging. The establishment of COFIRA was one of the most innovative and positive features of FOFIFA's new statutes. 13. The volatile political situation in the early 1990s affected project implementation, and the Bank supervision mission of February 1993 rated the project Unsatisfactory in all mandatory rating categories, except project management performance which was rated Satisfactory. It recommended that a reappraisal of the project be undertaken and that disbursements be suspended unless substantial progress was achieved by June 1993. FOFIFA endeavored to fulfill all its agreements with the Bank and took the lead in preparing the project restructuring, including consultations at all levels of the institution. 14. A subsequent Bank mission in August 1993 agreed with FOFIFA on the extent of the restructuring required, which included prioritization and planning of project activities to fit FOFIFA's implementation capacity based on Public Investment Program (PIP) commitments. These involved limits to the recruitment of technical and research personnel and to the creation of new research stations. Coordination of research and extension was made a conditionality of disbursement. A general reduction in project scope was also planned, and changes reflecting Government's regionalization policy were introduced, making the regional centers more responsive to economic and environmental constraints of producers, and promoting synergy between research and extension efforts and relationships with local NGOs. The prioritization process resulted in cost savings and, as a result, an amount of SDR 9.5 million (US$12.3 million) was canceled from the credit in April 1994. 15. The restructuring improved project implementation significantly. Working relationships between the Ministry of Scientific Research and FOFIFA improved, and contacts between research and extension staff strengthened. The regionalization program of research centers was pursued, and regional multi- disciplinary research teams were set up in several centers. Training was carried out in rural economics, communications, phytopathology, integrated pest management, procurement and management of research centers. The timely procurement of computers allowed training in statistics conducted under a subsidiary agreement by the Ecole Superieure de Sciences Agricoles (ESSA). For the first time, Government decision makers and donors were presented with a coherent and integrated picture of the overall research effort, which reflected the objectives of a national research program. 16. In 1994, the first research results began to be extended to farmers, including new higher-yielding and more adaptable varieties of rice and other crops, more efficient cultural techniques, and improved  5 tools and farm equipment. Integration with the extension service increased, and the impact of research outputs was continuously monitored and evaluated jointly with the extension services. Results indicated a significant impact on production of rice, peanuts, cotton and other crops. 17. In 1995, the project was rated highly satisfactory for development objectives and satisfactory for implementation progress. The project achieved a much higher level of performance as the new Minister of Scientific Research was very supportive of the project and its staff. 18. During 1996, however, the project continued to suffer counterpart funding problems and financial management difficulties, but was still rated satisfactory for development objectives and for implementation progress. Research developments under the project began to accrue and significantly raised crop yields, increased farmer income, and reduced environmental degradation. Increased benefits from improved rice varieties and cultural techniques were recorded. In the north-west, the development and release of virus-resistant rice varieties enabled farmers, who had abandoned their lowland paddies to encroach on highland forest land to grow rice, to return to them and thereby spare further forest degradation. The second phase of FOFIFA's staff retrenchment program was started in 1997 and was completed in early 1998. 19. Institutional Strengthening. Achievement of this component's objectives is satisfactory. The restructuring of FOFIFA's commodity departmnents into regional multi-disciplinary centers has been completed. Staff reduction, mainly of the lower non-scientific levels, has been successful (the total staff number is now 34 percent of that at appraisal), resulting in a leaner and more efficient structure. The regional centers have full autonomy in research programming and implementation, and have their own operating budgets. FOFIFA now has management and financial autonomy, and has reinforced its management capacity and accountability. It has developed strong links with national research organizations of other countries, as well as with the international agricultural research centers (IARCs). Donor coordination is much improved, although under informal arrangements due to the small number of donors still supporting research. Despite all these improvements, the institutional strengthening process is not complete. While the institutional framework for an effective research system has indeed been laid down, the numbers of multi-disciplinary teams of trained researchers are not yet complete at all the centers. 20. Research Quality and Adaptability Improvement. Research quality significantly improved through the adoption of the participatory approach of demand-driven research program planning and evaluation involving regional research centers, farmers and the extension services. Thus, research priorities now reflect more closely real farmer needs, and not mainly research staffs personal agendas, through three main programs: (a) agricultural intensification in high potential areas; (b) yield stabilization of crop and animal production systems in marginal areas; and (c) agricultural diversification into high value crops for export. The focus has also been shifted from station to on-farm research, which is more integrated and multi-disciplinary. Scientific rigor has been reinforced by training researchers in biometrics and improving the quality of research protocols, and accountability for results has been established by requiring each researcher to provide appropriate performance indicators in the research protocol and to report on them at the end of the year. The training input facilitated by the project together with the firm support of the Director of FOFIFA led to improved skills among research and extension staff in participative on-farm systems diagnosis. 21. Institutional linkages between research and extension are excellent. Through regional committees, researchers participate with extension staff and farmers in the diagnosis of problems and in planning and setting up field tests and demonstration plots. They also train extension services providers  6 under the monthly workshops to review agricultural technologies developed under the project and to adapt them to local conditions for adoption by farmers. An important aspect of the cooperation is the joint annual review of the performance and impact of their work programs. 22. Human Resources Development. This component has substantially achieved the project's original objectives, although it is now realized that the number of trained researchers is insufficient for today's needs. A total of 47 researchers out of 50 planned at appraisal (94 percent) have received post- graduate training (MS and PhD) under the project, mainly under bilateral aid arrangements. However, FOFIFA recruitment of trained researchers has been constrained by the freeze on civil service recruitment. Thus, the lack of a critical mass of trained researchers at two of the eight centers will continue to be a constraint for some time. The young age of the regional researchers raises the issue of research leadership, which is lacking at some centers and needs to be addressed by FOFIFA if it is to become a top class institution. 23. Rehabilitation of Physical Plant and Equipment. This program is virtually complete. Resources (equipment, buildings, budgets) have been allocated in priority to the eight regional centers and their satellite stations, as opposed to headquarters in Antananarivo. Priority was given to the improvement of the social infrastructure (adequate housing, running and drinking water, electricity supply) to attract researchers to stay at centers. D. MAJOR FACTORS AFFECTING THE PROJECT Factors Outside of Government Control 24. The political turmoil of 1991/92 had a negative impact on the project as all activities within the country basically came to a halt and paralyzed project execution from June 1991 through January 1992. Most of the research trials for the first season could not be harvested, and those of the second season could not be laid out. The development of institutional and policy aspects was greatly impaired, and field activities were suspended. This meant that in 1993, three years after Board approval, the project was still struggling to take off. Another problem outside Government control which constrained FOFIFA's capacity to carry out research was the shortage of agricultural scientists in the country. Factors Under Government Control 25. The Government did not provide full support to the project and to FOFIFA at the beginning of the project. In 1992, the Ministry of Scientific Research (MSR) interfered with FOFIFA's draft statutes, and conflicts between the Ministry and FOFIFA on responsibility for procurement iinterfered with the progress of the project. All the high level officers of the Ministry who were familiar with the project were dismissed, FOFIFA's statutes were changed unilaterally, and FOFIFA's Director was replaced without prior consultation with IDA. FOFIFA researchers were told to engage in commercial operations, which was contrary to the Government policy of disengagement from commercial activities. The stand- off between MSR and FOFIFA paralyzed project implementation until the replacement of the Minister in mid-1993. The strengthening of FOFIFA's administrative and financial management was not completed until the project was restructured in 1994 when the agency was given autonomy for these two functions. 26. The expansion of the new FOFIFA headquarters from 1800 sq m to 3000 sq m was out of proportion to the country's and the agency's needs, and inconsistent with the policy of decentralization. This building absorbed scarce financial resources that could have been more constructively used elsewhere and exacerbated the tendency for research staff to remain in the capital rather than accept  7 assignments up-country to regional research centers and nearer to the clients. 27. Only three out of fifteen title deeds have been issued to FOFIFA, due to delays in Government approval and to the presence of squatters. However, processing of the remaining title deeds is well- advanced. At first, the lack of title deeds delayed the construction of regional centers but with written assurances from the Ministry of Agriculture on the transfer of lands, FOFIFA proceeded with the rehabilitation and construction program. 28. Donor coordination by the Government could have been improved. The national research program, which was initially characterized by donor-driven agendas, did not provide the early leadership for donor coordination. Coordination was subsequently reinforced, and there was close coordination with USAID, which funded the development of improved rice varieties under a joint FOFIFA-IRRI (International Rice Research Institute) program, and with French Cooperation-financed programs (CIRAD). 29. Counterpart Funding. Many of the project's problems stem from insufficient counterpart funding because of over-optimistic expectations at the time of project design and appraisal, the political upheaval in 1991 and 1992, and the ensuing economic recession. In spite of its stated commitment to research development, the Government was not able to meet its financial obligations to provide an adequate budget for FOFIFA. The project suffered funding delays and low budgets, as evidenced by its PIP allocations. In 1993, the availability of counterpart funds was reviewed in the context of the PIP, and project activities were tailored accordingly. However, up until 1997, the Government did not meet its obligations of providing adequate funds to pay wages, arrears in social benefits and staff pensions, and back taxes. E. PROJECT SUSTAINABILITY 30. Project sustainability is now probable because the investment phase (for physical infrastructure and equipment) is virtually complete, core teams of trained researchers are in place at most of the centers, and FOFIFA's financial situation has been significantly improved (reduced recurrent budget needs due to its now much leaner structure, its debts have been paid, and additional revenues are collected from research contracts with the private sector). The decentralization of the research system, the institutionalized participation of farmers and extension services, the improved delivery of new technology through the national extension service, the good management of FOFIFA's resources, and a deliberate policy to contract with the private sector, are sound foundations of sustainability. 31. The other aspect of sustainability, based on the continued access to public funds at present levels, is uncertain for the specific areas of agricultural research where the private sector may be willing or able to pay. The pressure to reduce public funding levels will increase during periods of fiscal restraint, when the public agricultural research system has to compete for scarce Government financial resources with other sectors. For the moment, Government commitment to support of agricultural research is underlined in its "Document Cadre de Politique Economique" agreed with the Bank and the International Monetary Fund, but it is clear that as agriculture becomes more developed and market-oriented, the private sector will be in a better position to pay for its own specific needs for agricultural technology (such as for high- value export crops). Research, such as on staple foods like rice and other foodcrops, that benefits the whole population and is less amenable to exclusion mechanisms, and that can be considered as a public good and is unlikely to be financed privately, will probably continue to be funded by Government. However, as to research for the private sector, FOFIFA has already made it one of its priorities and has entered into contracts with firms for the development of new rice, mango and vanilla varieties for export.  8 A potential important source of funding for FOFIFA might be revenues from property rights for the new varieties developed by it. However, this is an area that is still uncharted in Madagascar. F. IDA PERFORMANCE 32. Bank performance was mixed. While the appraisal mission over-estimated FOFIFA's institutional and financial capacity, it correctly identified the main institutional and technical issues facing the sector and the long term time-frame needed for developing an effective national agricultural research system. The project was technically innovative and consistent with the Government strategy for the sector, as well as with the Bank's country strategy. The Bank reacted firmly to the poor progress after three years of project implementation by carrying out the successful restructuring of the project in 1994: it developed an action plan with FOFIFA, reallocated funds to finance a larger share of operating costs, and canceled half of the credit. The frequent change of Bank task manager (five times) for the project affected continuity and hindered the smooth execution of the project. Overall, however, the Bank performed satisfactorily during project supervision through its skills mix, and the handing over of responsibility for project supervision to the Resident Mission in Madagascar as from late 1996 was significant in advancing the completion of institutional reforms (staff reduction) which had been lagging for many years, in reorienting research priorities, and in developing a performance monitoring and impact evaluation system. The proximity of Bank technical specialists also meant that they could, on a prompt and regular basis, provide international expertise to FOFIFA. G. BORROWER PERFORMANCE 33. The Borrower's performance can be rated as marginally satisfactory at the beginning of the project but satisfactory near the end. At the beginning, the poor working relationship between the Ministry of Scientific Research and FOFIFA and interference of the Ministry in FOFIFA management delayed project implementation. In addition, the Borrower did not provide adequate counterpart funding to FOFIFA and delayed the payment of staff salaries, social security charges, and taxes. Compliance with credit covenants was unsatisfactory during the first years of project irnplementation, but subsequently improved. 34. FOFIFA management's performance was satisfactory, although it did not have the full support of the Ministry of Research to implement some of the more sensitive staff reduction programs. In spite of this, it successfully completed the main aspects of its institutional strengthening program and pressed on with the decentralization of staff to regional centers with some success. In particular, the Government and FOFIFA must be commended for taking some tough and sensitive personnel decisions in the second half of the project which greatly helped to turn it around. Without the staff retrenchment program, the sustainability of FOFIFA would have been at very great risk. 35. The inter-ministerial Steering Committee for Research satisfactorily fulfilled its role at the beginning of the project, but this became less important as many bilateral donors withdrew from the subsector, leaving USAID and Swiss Cooperation as the only other donors remaining and cooperating in the national research program. H. ASSESSMENT OF PROJECT OUTCOME 36. Institution Development. The project has succeeded in establishing an effective, decentralized national agricultural research system which is, therefore, nearer to its clients and more responsive to their  9 needs. The system is based on a network of regional centers, equipped with adequate physical facilities and staffed with core teams of researchers. 37. Research programming has been strengthened by adopting a bottom-up planning process, fully involving extension services and farmers, and implementing multi-disciplinary, on-farm research programs. The research strategy was also modified during 1997 to give priority to market-oriented agriculture in high potential areas, including diversification towards high value crops for exports, while maintaining yield levels in marginal areas. Genetic improvement remains the underlying thrust for yield improvement and increasing adaptability to local conditions and resistance to pests and diseases. To improve the quality of research programs, each researcher is now obligated to include monitorable indicators for his research work and to monitor field impacts at farmer level together with extension staff. 38. FOFIFA has successfully implemented its staff reduction program, mainly of its lower level staff, as its total staff number has been reduced from 1,156 at appraisal to 393 at project completion. As a condition of effectiveness, FOFIFA was to reduce 110 low level staff before end 1989 and another 140 staff by December 31, 1990, to bring the total staff number down to about 900. This target has been substantially exceeded. This first action was completed with funding provided by IDA Cr. 1967-MAG for payment of unemployment benefits and training costs for the staff laid off. Continued efforts by FOFIFA brought the total staff to 877 in January 1997, and in early 1998 FOFIFA was able to implement the voluntary departure and early retirement of another 484 staff (57 percent of total staff), mainly of lower level administrative staff. Financial incentives to leave were funded under the credit supporting this project. Benefits accruing to FOFIFA include: an improvement in the ratio of research to non- research staff from 1:7 to 1:3, and an immediate budgetary savings of US$220,000 per year which, over 5 years including the cost of employing 46 new researchers, fully cover the cost of compensation. The staff reduction program will also result in a reduced budgetary cost per researcher from FMG 72 million to FMG 45 million per year. 39. The redeployment of headquarters staff to the regional centers has also met with reasonable success as a total of 31 researchers (40 percent of the headquarters research staff) have been redeployed. The ratio of researchers at the regional centers as compared to those at headquarters has improved from 27:73 at appraisal to 49:51. Obviously, there are still too many researchers at headquarters, and FOFIFA needs to continue its efforts (8 researchers are scheduled to move to the centers in 1998). Under the center rehabilitation program, emphasis is placed on improving basic housing conditions, such as providing access to running water and electricity, as an inducement for staff to move to field stations. 40. FOFIFA is also developing its links and cooperation with the large private sector as a result of its work on export-oriented higher value agriculture (cotton, vanilla, soybeans, temperate and tropical fruits, etc.). Eight research contracts are being implemented, contributing about 20 percent to FOFIFA's total budget. 41. Development and Dissemination of Improved Agricultural Technology. The first results of the project were obtained in 1992 with the release of 2 improved varieties of rice (potential yields which were double prevailing yields). In subsequent years, other improved varieties and technologies developed by FOFIFA were released for adoption by farmers. Among the most notable results are rice varieties resistant to red-yellow mottle virus released in 1994 and 1996 (which had drastically reduced rice production in the northwest and forced farmers to open up thousands of hectares of forest areas, with a negative impact on the environment); higher-yielding rice, cassava, cotton (89 percent adoption of D388-8 by farmers in the southwest and northwest over 15,200 ha in 1997, with 30 percent increase in yield over previous varieties, one year only after release of the variety in 1996), beans, and corn varieties;  10 high-vanillin vanilla varieties in 1995 (two to three times as high in vanillin as compared to present varieties); and an improved breed of cattle (Manja Boina) for milk production in hot regions. Improved cultural techniques, which increase yields, developed by FOFIFA include the use of young rice seedlings (adoption by farmers amounts to as much as 40 percent in certain regions, with an increase in yields of 1 t/ha of paddy); covering seeds with a phosphate fertilizer coat for rapid seedling growth; and a rapid multiplication method for cassava stakes. A natural insecticide developed by FOFIFA from a local tree is replacing the use of less environmentally-friendly chemical insecticides. Positive results obtained from the testing of vaccines against Teschen disease in pigs have now made possible the rearing of pigs, which were decimated country-wide by that disease. Finally, ongoing variety development programs will continue to produce increasing numbers of superior crop varieties for adoption by farmers. 42. The new technologies developed by FOFIFA have to be transferred to farmers for adoption, and this responsibility rests with the national agricultural extension service, which is under the Ministry of Agriculture. The rate of adoption of FOFIFA's new technology by farmers will depend upon both the quality and the relevance of the technology to them and the effectiveness of the extension services. The National Agricultural Extension Project (NAEP) is underway only as from October 1995, although an agricultural extension pilot project had been implemented as from 1990. Only 9 out of the 29 agricultural districts in the country are at present effectively covered by the extension service; this limits NAEP's effective delivery of research results at the national level (although agents are now in place in all the agricultural districts). Many farmers, therefore, do not yet have full access to FOFIFA's technologies, but their numbers will decrease as NEAP intensifies its activities country-wide. Linkages between research and extension are excellent at all levels. 43. Impact on Beneficiaries. The project impact is measured by the effect of the technologies developed by FOFIFA on agricultural production and farm income, and integrates the efforts of both research and extension services, as these cannot be dissociated from each other, to bring these technologies to farm level and promote their adoption by farmers. Estimates of production increases of rice and other crops due to research results were first obtained in 1993. In 1997, the increased paddy adoption was estimated at 150,000 tons, or about 6 percent of the national paddy production. Farmers are also adopting improved varieties of other crops. The degree of adoption will increase in time as farmer confidence in the new technologies increases, and in addition, as the extension services increase their national coverage, an increasing number of farmers will have access to these technologies. Thus, the impact of research will increase with time. 44. While there is no project activity specifically designed to promote women, they are an integral part of the farming community and benefit to the same extent as men from the improved technologies generated under the project. However, vegetables are cultivated more by women, and the improved varieties developed by FOFIFA would benefit them more specifically. More adaptable and higher- yielding vegetable and other crops have a direct impact in enriching and diversifying diets. Female FOFIFA staff have benefited under the project as FOFIFA now has many female researchers, some at the highest level of management at director level. 45. Environmental Impact. The environmental impact of the project is both direct and indirect. The development of rice varieties resistant to red-yellow mottle virus had a direct positive impact on the environment as it enabled farmers to return to their paddy fields which they had abandoned to encroach on large areas of virgin forest lands. Farmers are also adopting the natural insecticide (Melia azederach) developed by FOFIFA to control bugs on rice and which substitutes economically for environmentally more harmful chemical insecticides. The development of sustainable farming systems is aimed at reducing environmental degradation, and agricultural intensification in the lowlands may lead to reduced  11 pressure to cultivate on the environmentally fragile hillsides. I. FUTURE OPERATION 46. The project supported only the first 7-year phase of the 15-year Master Plan for the development of agricultural research. Government is committed to the implementation of the second phase, which will focus on human resource development, on firming up participatory approaches for research programming and execution, and on further strengthening the regional capacity to carry out high quality research. Experience from the implementation of the first phase has shown the need for research development to be well integrated with extension development. The Government has requested IDA support for the second phase: this support will be provided under the proposed Agricultural Sector Investment Program. Until such a project is implemented, the Government has agreed that Cr. 2729-MAG, which supports the IDA- financed National Agricultural Extension Project, will provide financing to FOFIFA to continue its operations for the next two years. J. KEY LESSONS LEARNED 47. The main lessons learned from the project include: (a) Borrower commitment and ownership are essential at all times; (b) The responsibility for financing of agricultural research by the Government should be clearly defined, and funding should be assured based on this.; (c) Stakeholder participation is essential for ensuring the relevance and responsiveness of research programs to farmer needs; (d) Decentralization of the research system is critical to its development; (e) A long implementation period is appropriate for this type of human resource development and institution building project; (f) The development of well-trained researchers takes time; (g) Effective accounting and financial management systems are prerequisites of project implementation; (h) The national agricultural extension has not yet achieved a national coverage and is not yet able to provide a full delivery service for research results; (i) An effective and committed project management team combined with availability of staff at the Resident Mission are important elements in improving project performance and impact.  12 IMPLEMENTATION COMPLETION REPORT MADAGASCAR NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 2042-MAG) PART II: STATISTICAL TABLES TABLE 1: SUMMARY OF ASSESSMENTS A. Achievement of objectives Substantial Partial Negligible Not Applicable Macro policies _ x Sector policies X _- Financial objectives X_ Institutional development X Physical objectives X Poverty reduction X _ Gender issues X Other social objectives X Environmental objectives X Public sector management X _ Private sector development -X Other (specify) _- B. Project sustainability Likely Unlikely Uncertain x ____- C. Bank performance Highly Satisfactory Deficient Satisfactory Identification X Preparation Assistance X Appraisal X _ Supervision X D. Borrower performance Highly Satisfactory Deficient Satisfactory _ Preparation _ X_ Implementation X Covenant compliance -X E. Assessment of outcome X  13 TABLE 2: RELATED BANK CREDITS Credit title Purpose Year of Status approval l Preceding operations l Agricultural Institutions Improvement of agricultural 1982 Completed Technical Assistance Project institutions, sector financial resource (Cr. 1249-MAG) management, agricultural sector planning, and strengthening of policy and economic analysis Second Agricultural Improving the management of 1986 Completed Institutions Development FOFIFA. Funding an Agricultural Project (Cr. 1709-MAG) Extension Pilot Program Agricultural Credit Project Provision of agricultural credit 1987 Completed (Cr. 1804-MAG) Irrigation Rehabilitation Testing of new system of irrigation 1985 Completed Project (Cr. 1589-MAG) management and maintenance Forests Management and Conservation of Madagascar's 1988 Completed Protection Project (Cr. natural resource base and bio- 1878-MAG) diversity Agricultural Sector Liberalization of rice marketing and 1986 Completed Adjustment Credit (Cr. elimination of the monopsony of 1691-MAG) state rice marketing agencies Public Sector Adjustment Liberalization of agricultural export 1988 Completed Credit (Cr. 1941 -MAG) trade and restructuring of sector public enterprises Following operations Environment Project (Cr. Support to the first phase of National 1990 Completed 2125-MAG) Environmental Action Plan Livestock Sector Project Support to milk and beef production 1991 Ongoing (Cr. 2243-MAG) l Second Irrigation Rehabilitation of irrigation schemes 1994 Ongoing Rehabilitation Project (Cr. and transfer of maintenance 2644-MAG) responsibility to water user groups National Agricultural Establishment of national agricultural 1995 Ongoing Extension Project (Cr. 2729- extension service MAG) l Second Environment Support to second phase of National 1996 Ongoing Program (Cr. N009-MAG) Environmental Action Plan I  14 TABLE 3: PROJECT TIMETABLE D)ate Steps on Project Cycle Date Planned actual/latest estimate Identification (Executive Project Summary) _ December 20, 1988 Preparation/Pre-appraisal _ January 1989 Appraisal January 1989 February 1989 Negotiations April 1989 April 25, 1989 Board presentation June 1989 June 15, 1989 Signing September 1989 September 11, 1989 Effectiveness December 1989 July 12, 1990 Mid-term review September 1993 October 1993 Project completion December 31, 1996 December 31, 1997 Credit closing December 31, 1997 December 31, 1997* * Except for Category 1, which will close on April 30, 1998 TABLE 4: CREDIT DISBURSEMENTS: CUMULATIVE ESTIMATED AND ACTUAL Estimated Actual Cumulative Actual as a % of Cumulative Disbursements estimate Fiscal Year Disbursements (US$ million) (US$ million) 1990 1.4 0.2 14 1991 3.8 0.7 18 1992 7.2 1.5 21 1993 10.6 2.3 22 1994 13.9 3.3 24 1995 18.0 5.4 30 1996 21.6 7.0 32 1997 24.0 8.6 36 1998 24.0 13.4 100* Cancellation US$ 12.3 million in April 1994 Final Disbursement April 30, 1998 * this equals 100% of revised cumulative disbursement of available credit of US$11.7 million.  15 TABLE 5: KEY INDICATORS FOR PROJECT IMPLEMENTATION Key implementation indicators Estimated Completed Institutional Development l Directorate of Support Services to Research I I established l FOFIFA's financial management and accounting I l system strengthened l Operational Manual prepared I I Research Management Biometry Unit established 1 I Programming and Monitoring and Evaluation I 1 Units established New research programming cycle implemented I lIl Regional research teams established 8 8 Regional research programs prepared 8 8_l National Strategy for Research prepared I lIl Human Resources Development Human resource development plan completed IlI Lower level personnel reduced (number) 240 724 (initial total staff: 1,156) (final total staff no: 393) Total number of researchers 99 114 Researchers deployed in field stations 27 56 Researchers transferred from headquarters to field 31 stations (40 % of HQ researchers) Percent researchers at: Headquarters 73 51 Regional centers 27 49 Foreign researchers at FOFIFA 19 9 Training: - short term (staff months) 60 32 - long term (no. of degrees obtained) 50 47 MS and DEA 34 PhD 13 Research Infrastructure No. of regional centers established 8 8 No. of substations established 10 10 Title deeds obtained 15 3  16 Release of Type of Technology Number and Characteristics Important New Technology 1992 . Improved rice varieties * 3 for upland rice in highland areas (yield: 3t/ha) 1993 * Improved cotton varieties * I for north/south west regions (yield: It/ha) 1994 * Improved rice varieties for * I tolerant to red-yellow mottle virus (yield: 4t/ha); north-west region I short cycle upland rice (yield: 3.5 t/ha); I high-yielding (yield: 4.5 t/ha) * Improved maize varieties for * 3 higher-yielding (yields: 2.6-3.5 t/ha) southwest, midwest and lowlands * Improved milk breed for hot * Manja Boina (1,600 litres per lactation) regions 1995 * Improved rice varieties for * 1 long grain, photoinsensitive var. (yield: 4.5 tlha); mid-east, midwest and 10 rainfed varieties (yield: 3.5 t/ha) highland regions . Improved vegetable varieties * 3 high-yielding vouandzu varieties (2 t/ha); I bean variety resistant to rust (2 t/ha) . Improved vanilla varieties * 2 high vanillin content (4% against 2% actual) and disease-resistant * Strategic treatment of * reduces infestation parasites in cattle . Improved eucalyptus * higher-yielding coppicing * Improved acacia species * higher-yielding * Management of pine * higher-yielding plantations 1996 * Improved rice varieties for * 4 higher-yielding (yield: 4.5-5.0 t/ha) highlands under good water control * Improved rice varieties for * 3 more tolerant to red yellow mottle virus northwest * Natural insecticide * Melia azedarach developed for farm use to control rice bugs * Improved cotton varieties * I high-yielding and good technological characteristics (1.2 t/ha) . Improved cassava varieties . 2 higher-yielding for general release * Improved plantain varieties * 3 higher-yielding * Improved maize varieties * I higher-yielding * Improved techniques of fish * for royal carp feeding and transport of fingerlings * Development of adapted farm * Ploughs, puddlers, weeders for rice cultivation tools 1997 * Improved rice varieties * 5 high-yielding (yield: 4-5 t/ha), I tolerant to high salinity (yield: 4 t/ha); 2 tolerant to submerged conditions (yield: 4 t/ha) . Improved cultural techniques * direct seeding for marginal areas  17 Impact Production Increase (tons) 1994 Paddy: 8,400 tons Peanuts: 2,680 1995 Paddy: 30,105 tons Peanuts: 3,800 Maize: 1,995 Fish: 12 1996 Paddy: 150,000 tons Peanuts: 21,800 Cassava: 5,490 Fish: 34 1997 Paddy: not available Peanuts: " Cassava: 16,900 tons Maize: 8,100 " Cotton: 4,080 " Fish: 37  18 TABLE 6: KEY INDICATORS FOR PROJECT OPERATION Not Applicable TABLE 7: STUDIES INCLUDED IN PROJECT Study Purpose as defined in Status Impact of Study SAR/redefined 1. Reorganization of Kianjasoa Dismantling of Completed Station activities Experiment Station production activities reorganized to focus on research work 2. Rehabilitation of irrigation system Permit year-round rice Completed Irrigation system at Lac Alaotra Rice Experiment cultivation operational Station 3. Topographical study of Mahela For preparation of title Completed Title deed issued Substation deed 4. Topographical studies of CALA, Physical masterplan Completed Physical facilities of Kianjasoa, and Ampandrianomby preparation stations improved 5. Drilling studies at Toliara Station Develop own water Completed Installation completed supply 6. In-depth study of the research and Strengthen laboratory Completed Three-year development service functions of FOFIFA systems and agricultural programs prepared technology research 7. Physical development plans of the Develop regional centers Completed Physical development centers and stations in the south-west, completed mid-west, and north-west 8. Study FOFIFA's administrative and Strengthen procedures Completed Improved system in place financial management 9. Update national research program Reset research priorities, Completed Relevance of research and balance national programs to user needs with local priorities improved 1O. Review of technical specifications Upgrade laboratory Completed Laboratory operational for diagnosis laboratory for animal research 11. Study FOFIFA's central Adapt management to Completed More operational management organizational structure 12. Study rehabilitation of Improve services Completed More operational Research/Development laboratories 13. Review FOFIFA's technical, Improve sustainability Completed Staff reduction program financial and economic viability implemented  19 TABLE 8A: PROJECT COSTS Appraisal estimate Actual/latest (US$ million) estimate (US$ million) Component Local costs Foreign costs Total Total Investment Costs 12.41 13.24 25.65 23.93 Recurrent Costs 26.55 4.86 31.42 11.94 Baseline Cost 38.96 18.10 57.07 Physical Contingencies 1.48 _ 1.32 2.80 Price Contingencies 7.55 3.20 10.75 TOTAL Project Costs 48.09 22.62 70.62 35.87 TABLE 8B: PROJECT FINANCING Appraisal estimate Actual/latest estimate (US$ million) (US$ million) Source Total Total IDA 24.00 13.35 Co-financiers (donors) 10.44 7.23 FOFIFA 5.65 1.89 Government 30.53 13.40 Total 70.62 35.87  20 TABLE 9: ECONOMIC COSTS AND BENEFITS As in most Bank Group-assisted agricultural research projects, no attempt was made to quantify the economic rate of return ex ante due to: (i) inherent uncertainties regarding the timing and value of research findings and the timing and extent of adoption of research findings by farmers; and (ii) difficulties in separating potential benefits of research and complementary investments such as extension and training.  21 TABLE 10: STATUS OF LEGAL COVENANTS Agreement Section Covenant Present Original Revised Description of covenant Comments type status fulfillment fulfillment date date DCA 2.02 C Establishment of Special Account None 3.01 (b) 2 C Make proceeds available to FOFIFA None 3.01 (c) 2 C Make proceeds available to ESSA None 3.02 3 C Adherence to IDA procurement practices None 3.03 3 C Follow procurement guidelines in General None Conditions 3.04 5 C Establishment of the Steering Committee None 3.04 (i) 5 CP Appointment of three scientists of international None repute in Steering Committee 3.04 (ii) 5 C 11/93 10/93 Date of mid-term review None 3.04 (iii) 5 C Operating procedures of Steering Committee None satisfactory to IDA 3.04 (iv) 5 C Coordination and approval of cofinancing by None Steering Committee 3.05 5 C Establishment of appropriate research funding None mechanism to replace stabilizing funds 3.06 5 C ESSA to recruit a professor of biometrics None 4.01 (a)-(c) 5 C Maintain adequate records and accounts and None have accounts audited annually 5.01 (d) 5 C Amendment of Statues of FOFIFA None 6.01 (d) 5 C Reduction of number of lower level staff in a None manner acceptable to IDA Project 3.01 (b) 5 C Appropriate staff for all managerial positions None 3.01 (c) 5 C Recruit a consultant for two years as director of None administration and finance 3.03 5 C Adequate insurance None 3.04 5 C A national research program coordinator for None each operational research program 3.05 5 C Date of submission by FOFIFA of revised None procurement procedures 3.06 (a) (i) 5 CP FOFIFA to provide additional benefits to field Not in compliance staff 3.06 (a) (ii) 5 NC FOFIFA to award a maximum of 5 cash prizes Not in compliance to reward scientific work 3.06 (b) 5 NC FOFIFA to prepare proposals for additional Not in compliance  22 benefits and cash prizes by January 11, 1996 4.02 lC Date of submission of regional research None I I extension committees Keys: C = Complied with CD = Complied, but with delays CP = Complied with partially NC = Not complied with Covenant Types: I - Accounts/audits 6 - Environmental covenants 11 - Sectoral or cross-sectoral budgetary or other resource allocation 2 - Financial performance/revenue 7 - Involuntary resettlement 12 - Sectoral or cross-sectoral generation from beneficiaries policy/regulatory/institutional action 3 - Flow and utilization of project funds 8 - Indigenous people 13 - Other 4 - Counterpart funding 9 - Monitoring, review and reporting 5 - Management aspects of the project or 10 - Project implementation not covered by executing agency categories 1-9  23 TABLE 11: COMPLIANCE WITH OPERATIONAL MANUAL STATEMENTS Statement number and title Describe and comment on lack of compliance No major lack of compliance was observed TABLE 12: BANK RESOURCES: STAFF INPUTS Stage of project cycle Actual Weeks US$ Preparation to appraisal 29.1 72.8 Appraisal 37.4 90.4 Negotiations through Board approval 2.7 7.0 Supervision 188.4 592.7 Completion 4.8 14.1 TOTAL 262.4 777.0  24 TABLE 13: BANK RESOURCES: MISSIONS Performance Rating Stage of project Month/ Number Days Specialized staff skills Implement Development Types of cycle year of in represented ation Objec tives problems persons field Status Preparation 03/88 2 08 A, OA - - Pre-Appraisal 09/88 3 05 A, AE, OA - - Appraisal 01/89 8 15 5A, FA, FE, LE - - Board approval 07/89 4 10 2A, FA, IT - - through effectiveness Supervision I 11/89 2 10 FA, A 2 1 P, LC Supervision 2 03/90 2 12 A, OA 2 1 P, FM Supervision 3 10/90 4 10 2A, FA, OA 2 2 P, FM Supervision 4 04/91 3 12 A, FA, OA 2 2 FM, C, I Supervision 5 03/92 2 16 A, AE 3 3 All Supervision 6 02/93 1 10 A 3 3 All Supervision 7 - 08/93 5 15 3A, FA, OA 3 3 FM, LC, I Restructuring Supervision 8 12/93 4 10 2A, FA, OA 3 2 FM, LC, I Supervision 9 04/94 3 12 A, FA, OA 2 2 FM, LC Supervision 10 11/94 3 07 A, FA, OA S HS Supervision 11 04/95 3 10 A, ES, OA S HS =_- Supervision 12 11/95 2 08 A, AE S S LC Supervision 13 06/96 2 15 A, ES S S FM Supervision 14 11/96 2 31 2A S S FM, I Supervision 15 05/97 1 10 A S S FM, I Completion 01/98 I 07 A S S I Staff Skills Types of Problems A - Agriculturist P - Procurement AE - Agricultural Economist LC - Legal Covenant OA - Operational Analyst FM - Financial Management FA - Financial Analyst C - Country related problems FE - Forestry Expert I - Institutional LE - Livestock Expert ES - Extension Specialist IT - Information Technologist  25 Appendix I Borrower's Comments PROJET NATIONAL DE RECHERCHE AGRICOLE (PNRA) Credit 2042-MAG - Contribution a l'elaboration du rapport d'achevement du projet 1. Introduction Le Projet National de Recherche Agricole (PNRA), objet de I'accord de credit et de l'accord de projet 2042-MAG, signes le 11 septembre 1989, a ete cloture le 31 Decembre 1997. Comme prevu pour les projets finances par l'IDA, un rapport d'achevement a e prepare des le mois de Decembre 1997. Le FOFIFA a e associe dtroitement a la preparation de ce rapport d'achevement, en particulier lors de la seance de travail en date du 28 janvier au cours de laquelle ont e revus et discutes le rapport de la mission de preparation du rapport d'achevement. Outre les commentaires foumis A cette mission, la presente evaluation a e faite A titre de contribution au document final. 2. Objectifs du projet et resultats obtenus Les objectifs initiaux du projet ont ete: * mise en ceuvre des programmes de recherche pluridisciplinaire coordonnee au niveau national; * developper des technologies devant contribuer a l'augmentation des revenus des producteurs, en particulier les petits paysans, A travers l'adequation des recherches aux contraintes reelles des paysans; * augmenter l'efficience des recherches A travers le developpement des recherches hors des stations, I'amelioration de la qualite des recherches, la reorganisation du FOFIFA et la decentralisation du systeme de recherche, la stabilisation du financement et l'amelioration de la coordination des donateurs; * renforcer les liens entre recherche et vulgarisation A travers les recherche en milieu reel et les demonstrations menees en collaboration avec les services de vulgarisation * et assurer la viabilit6 institutionnelle du FOFIFA A long terme. Bien que ces objectifs initiaux ont et6 gardes dans leur globalite, les consequences des evenements de 1991-1993, ont amene a restructurer le projet sur la base de deux reorientations majeures: * Ia decentralisation de la programmation et de l'execution de la recherche A partir des priorites regionales; * la regionalisation de la recherche par l'op6rationnalisation des equipes regionales pluridisciplinaires. 2. Resultats obtenus D'une maniere globale, le projet a atteint ses objectifs, tant institutionnel que technique et economique, et cela de maniere satisfaisante malgre les retards enormes au demarrage. Outre les aspects evoqu6s avec  26 la mission de pr6paration du rapport d'ach6vementl, les acquis fondamentaux du PNRA concernent: La regionalisation de la recherche: Malgr6 les reticences et les difficult6s enregistrees au depart, la r6gionalisation de la recherche agricole est maintenant une realite, cela grace a: * 1'existence au niveau des huit (8) Centres r6gionaux de recherche d'infrastructures de recherche fonctionnelles (bureaux, laboratoires de terrain, equipements, materiels informatiques, etc...); * une masse critique de chercheurs suite au redeploiement des 31 chercheurs a partir d'Antananarivo; et * la mise en (euvre de programmes de recherche definis A partir des priorites regionales. La mise en euvre de nouvelles approches: Contrairement aux demarches tres descendantes et lineaires pratiquees dans les annees 80, de nouvelles approches sont maintenant adoptees et intemalisees au sein de FOFIFA, favorisant ainsi le rapprochement avec les utilisateurs et la diffusion des resultats. Ils s'agissent entre autres de: * I'adoption de l'approche participative et pluridisciplinaire. * l'institutionnalisation des analyses d'impact au niveau de tous les programmes; * le developpement des experimentations et tests en milieu reel en etroite collaboration avec la vulgarisation et les paysans; et * enfin le d6veloppement du partenariat avec le secteur prive et associatif. L'Itaboration des strategies et des priorites de recherche: Le changement du contexte macro-economique et l'6volution de la situation socio-economique au debut des annees 90 ayant amen6 des modifications sur la configuration des problematiques de developpement agricole et rural, le Plan Directeur de la Recherche Agricole (la reference premiere pour l'execution du projet) a dGe &re actualise (1996). Avant la cl6ture du projet, 1'exercice de planification a ete continue et a abouti A la finalisation actuelle de: * Plans Strategiques de Recherche; - assortis des Plans Operationnels definissant les programmes de recherche pour 1998-2002. Concemant la priorisation des recherches, il faudrait mentionner les decisions arretees et appliquees en matiere de redefinition des priorites et qui portent sur la mise au point de technologies visant A: * intensifier I'agriculture dans les zones a hautes potentialites; * maintenir les rendements dans les zones marginales, y compris ceux des systemes d'e1evage extensif; * diversifier les filieres de production vers des cultures A haute valeur destinee a 1'exportation. La reorganisation du dispositif de recherche et la rationnalisation des effectifs: La regionalisation de la recherche evoquee auparavant et la recherche d'une viabilite du FOFIFA avaient amen6 A redimensionner l'ensemble du dispositif du FOFIFA, A revoir son organisation et A renforcer la formation de son personnel. Ainsi, parmi les acquis majeurs dans ce domaine, on peut citer: * Ia reduction des effectifs non chercheurs de pres de 480 agents qui ont et soit mis A la retraite anticip6e, soit redeployes vers d'autres secteurs d'activite, ramenant ainsi le ratio chercheur/non chercheur de 1/7 a 1/3; * Ia formation dipl6mante d'une cinquantaine de chercheurs environ, pour leur conferer le niveau de DEA ou Master's (33) etlou Doctorat ou PhD (13). 1 Commentaires fournis lors de la seance en date du 28 Janvier 1998  27 La valorisation et la diffusion des resultats: Durant les quatre derniere anndes du projet, un accent particulier a dte mis sur le transfert des technologies, c'est-A-dire, la diffusion des rdsultats de recherche aupres des utilisateurs et leur exploitation dans le cadre des systemes de production. Cet aspect a d'ailleurs permis le renforcement et l'intensification des liens entre recherche et vulgarisation, car si vers la fin des annees 80, les vulgarisateurs avaient peu de nouvelles technologies disponibles A offrir aux producteurs, actuellement une gamme assez variee de solutions techniques sont A leur disposition. Ainsi, ont ete liberees entre 1992 et 1997 (avec assurance de fourniture des semences de pre-base et/ou de base pour les secteurs s'occupant de la production de semences), diverses varietes plus performantes adaptees aux differentes conditions etlou contraintes de plusieurs especes: riz, mais, coton, vanille, haricot, manioc, vouandzou, cafe, etc.... En matiere d'elevage, deux nouvelles races bovines, dont une laitiere specialement adaptee aux zones chaudes permettant de developper la production laitiere de ces regions jusque la marginalisees, ont ete diffus6es. Diverses techniques culturales, dont celle permettant l'economie d'engrais mineraux et celle relative au traitement phytosanitaire, ont dgalement dte vulgarisees. Enfin, il faut inclure dans cette partie "valorisation et diffusion des resultats" le developpement de la "communication en matiere de recherche agricole" avec la parution reguliere de la revue trimestrielle "KAROKA" et des autres supports type fiches techniques et autres brochures de vulgarisation. Le renforcement institutionnel: Le premier element A mentionner concernant cet aspect porte sur la mise en place de mecanisme permettant la validation des actions de recherche, notamment celles se rapportant a la mise en auvre des nouvelles approches, la definition des priorites de recherche, la valorisation des resultats ainsi que le renforcement des liens recherche-vulgarisation mentionnees plus haut: A cet egard, les reunions de campagne r6gionales, les visites commentees (field-days) et les restitutions periodiques des resultats ont montre leur efficacite et leur utilite. Sur le plan inteme, la mise en ceuvre du systeme de programmation et de suivi-evaluation des recherches fait aussi partie des acquis non negligeables, bien que l'operationnalisation du systeme reste A etre rod6e et A parfaire actuellement. II en est de meme de l'informatisation du suivi budgetaire et comptable qui a ete durant longtemps sous-employee et dont l'exploitation est en cours de revision pour tenir compte de la decentralisation des gestions suite A l'autonomisation des centres regionaux apres la regionalisation. En matiere de renforcement de la capacite institutionnelle du FOFIFA, de nombreuses avancdes ont ete realisees (e.g.: la releve des chercheurs expatries au niveau de nombreux programmes, participation aux recherches en reseau au niveau regional, etc...). 3. Analyse sur les facteurs ayant affecte le projet II n'est plus necessaire de rappeler les impacts des evenements de 1991-1992 sur le deroulement du projet. Neanmoins, cette experience avait montre d'une part, l'importance de I'autonomie de fonctionnement de ['agence d'execution, et d'autre part, le risque encouru pour un bon achevement d'un long projet (7 ans) dans un contexte socio-politique en perpetuel changement. Outre ces aspects globaux, la contrainte majeure rencontree se rapporte aux respects des engagements pris dans l'Accord de credit, en particulier le niveau de contre-partie et les appuis pour la realisation des mesures d'accompagnement definis au demarrage du projet (ex.: obtention des titres fonciers, postes budgetaires pour le recrutement de chercheurs, solutions pour le paiement des taxes et droits de douane,  28 etc...). Sur un autre plan, il faudra souligner que: * Ia lenteur et la lourdeur des procedures administratives nationales n'ont pas toujours permis A I'agence d'ex6cution d'op6rer avec toute la c6lerite voulue; * et du c6te de la Banque, la rigueur de certaines procedures -malgre les quelques assouplissements dejA faits - ne permet pas de faire face A temps aux problemes qui se posent. 4. "Performance de l'IDA" Le FOFIFA a toujours trouve appui et comprehension aupres des differents "task-managers" qui ont suivi le projet. Leur changement frequent l'a neanmoins quelque peu gene. Vers la fin du projet, il a enormement apprecie la designation d'un charge de projet resident sur place. Cette decision a beaucoup contribue A acc6lerer la r6solution des problemes ou du moins A faciliter les echanges de point de vue et la comprehension mutuelle des problemes entre le bailleur, 1'emprunteur et l'agence d'execution. Ce qui nt eait pas toujours possible malgre l'existence de moyens de communication type messagerie electronique ou autre. Sans cette disposition, la finalisation du projet aurait d'ailleurs ete des plus difficile. Enfin sur ce point, le FOFIFA a apprecie la compr6hension dont, suite A ses requetes, l'IDA a fait preuve pour accepter dans des delais raisonnables la modification de certaines dispositions qui s'imposaient dans l'accord de credit. Un demier aspect positif A signaler porte sur l'effectivite du co-financement du projet qui traduit bien le souci de coordonner les appuis A la recherche (participation de l'USAID, les Cooperations bilaterales francaise et allemande). La mise en cruvre de ce co-financement gagnerait cependant A etre harmonis6e davantage, notamment A travers un fonctionnement mieux organise du Comite de Financement de la Recherche Agricole cree A cet effet. 5. "Performance de l'Emprunteur" L'Etat malgache: Malgre ses problWmes et obligations, I'Etat a aid6 de son mieux le FOFIFA en lui accordant certaines derogations, par exemple le benefice de l'achat symbolique pour regler les problemes d'obtention des titres fonciers. Du fait de son statut, le FOFIFA n'a cependant pu obtenir la detaxe pour les importations qu'il a fait dans le cadre du projet, et ce n'est que vers la fin du projet qu'une solution a ete trouvee. Dans le meme esprit, il faut rappeler que les fonds alloues au titre de la contre-partie n'ont pas toujours ete en rapport aux allocations du credit IDA. Enfin, I'insuffisance du credit reserve aux paiements des frais de personnel et aux charges de fonctionnement des structures a handicape le FOFIFA durant de nombreuses annees. Sur ce plan, I'impact de la rationalisation des ef-.ectifs avec 1'effort deploye par le Ministere de la Recherche Scientifique et celui du Budget et Finances pour integrer dans le Budget general une partie des chercheurs devrait A l'avenir rendre plus soutenable les charges recurrentes du FOFIFA. L 'agence d'execution du projet: le FOFIFA: Le FOFIFA, qui a herite d'une longue tradition de recherche tres sectorisee (filiere et discipline), a dfi se depenser pour faire admettre et internaliser dans le cadre du  29 PNRA: * I'approche regionale et pluridisciplinaire; * Ia budgetisation par programme; * la rationalisation des effectifs avec la reorganisation des dispositifs de recherche; * et l'adoption des mesures d'impacts (obligation de resultats) en compl6ments des criteres scientifiques habituels pour evaluer les programmes. Les acquis actuels sont certes importants, neanmoins, avec la regionalisation effectuee, un grand ddfi reste A faire au niveau de l'autonomisation totale sur le plan du fonctionnement des centres regionaux et de leur gestion administrative et financiere. 6. Soutenabilite future et lesons a tirer En guise de conclusion, on peut dire que la dimension et l'organisation du FOFIFA apres le projet PNRA sont en rapport avec les fonctions normalement devolues A une institution de recherche agricole qui se veut avoir un mandat national. Les mecanismes mis en place pour tenir compte de l'ensemble du Systeme National de Recherche Agricole (SNRA) renforcent ces perspectives, notamment pour capitaliser les acquis de tous les intervenants du systeme pour eviter les duplications d'actions, mais aussi pour pouvoir repondre A toutes les sollicitations que FOFIFA seul ne peut assumer. Cet aspect devra dans le futur proche faire l objet d'une formalisation de la "coordination des actions de recherche" pour aboutir A une meilleure synergie et pour eviter la balkanisation des interventions. Le r6le et les initiatives de la nouvelle direction de recherche agronomique et zootechnique du Ministere de la recherche scientifique dans ce domaine serait des plus attendus. Le corollaire du point precedent se rapporte a la redynamisation du fonctionnement du Comite de Financement de la Recherche Agricole qui ne devrait pas se cantonner uniquement au niveau des revues des allocations de fonds pour la recherche mais, veiller aussi A la coherence des financements avec les priorites des recherches pour le developpement rural. A ce titre, la mise en place et l'operationalisation du "Fonds de la Recherche Agricole", qui aurait dfi etre terminee durant le projet, devrait etre finalis6e dans le court terme pour assurer la perennisation du soutien A la recherche, telle que prevue initialement. Enfin, comme il est evoque plus haut, les futurs projets ou programmes d'appui du type PNRA ne devrait pas etre definis sur un pas de temps trop long du fait de leur sensibilite aux conjonctures vecues par le pays. Fevrier 1998.  42' 44' 4L6 48' 50 -1 2