Guinea - Gueckdou Agricultural Development and First Agricultural Services Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 14766 PERFORMANCE AUDIT REPORT GUINEA GUECKDOU AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 1635-GUI) AND FIRST AGRICULTURAL SERVICES PROJECT (CREDIT 1636-GUI) JUNE 29, 1995 Operations Evaluation Department 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. Abbreviations AfDF African Development Fund ASP Agricultural Services Project BCPA Bureau for the Coordination of Agricultural Policy BSD Bureau of Strategy and Development EEC European Economic Community FAC Fund for Aid and Cooperation (France) FAPAs District Collective Farm (Fermes Agro-Pastoral d'Arrondisement) FD Financial Directors IDA International Development Association IFAD International Fund for Agricultural Development IRAG Guinean Agricultural Research Institute MARA Ministry of Agriculture and Animal Resources MDR Ministry for Rural Development OED Operations Evaluation Department ONADER National Rice Development Office (Office national pour le d6veloppement du riz) ORG Gueckedou Rice Project (Operation Riz Gu6ck6dou) PAG Gueckedou Agricultural Project (Projet Agricole de Gu6ck6dou) PAR Performance Audit Report PCR Project Completion Report PNIR National Rural Infrastructure Project (Projet National d'Infrastructure Rurale) PNVA National Agricultural Extension Project (Projet National de Vulgarisation Agricole) SAR Staff Appraisal Report SPCIA Agricultural Input Production and Supply Company (Societe de Production et de Commercialisation d'Intrants Agricoles) TA Technical Assistance UA Units of Account FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General June 29, 1995 Operations Evaluation MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Guinea - Gueckedou Agricultural Development Project (Credit 1635-GUI) and First Agricultural Services Project (Credit 1636-GUI) Attached is the Performance Audit Report (PAR) on the Guinea Gueckedou Agricultural Development and First Agricultural Services projects (Credits 1635/1636-GUI, approved in FY86) prepared by the Operations Evaluation Department, taking comments from the Borrower into account. The two projects were appraised together shortly after the death of President Sekou Toure, a period of great uncertainty in Guinea. The Gueckedou Agricultural Development project was a regional operation, and was cofinanced with the African Development Fund (AfDF) and the International Fund for Agricultural Development. It aimed at increasing smaliholder productivity by introducing new technical recommendations for rice, coffee, groundnut and cassava; establishing an extension service; developing applied research; improving infrastructure; and developing farmer organizations. Little emerged from the applied research program and, with the exception of an innovative approach to micro-level swamp development, the technical recomendations proved disappointing. Project results, while tangible in several respects, were not commensurate with the resources used. A general suspension of lending by AfDF to Guinea disrupted implementation. The institutional development achieved by the project is rated as modest and the sustainability of its structures is deemed uncertain. The project outcome is rated as unsatisfactory. Because of a lack of data, the PCR did not re-estimate the ERR, but rated the project outcome as satisfactory, on the basis of optimistic presumptions regarding actual and expected impact. The objective of the Agricultural Services project was to enhance the capacity of the Ministry of Rural Development in rural sector planning, project evaluation, and policy development, through the establishment of a Bureau for Strategy and Development (BSD) within the Ministry. The project supported BSD through the provision of technical assistance, facilities and equipment, and training, but its design proved overambitious and inflexible. Weak technical assistance, political interference, and lack of counterpart resources, all contributed to minimal achievement of project objectives, and it was terminated early. However, BSD was restructured after project termination and is now a much stronger element of the Ministry. Hencc, the limited benefits of the project are expected to be sustainable but institutional development is rated as modct and the project outcome as unsatisfactory given the disappointing results achieved ii, relation to costs. The principal lessons relate to project design under conditions of great uncertainty and local resource scarcity. First, under such conditions considerable flexibility in the design of institutional support is advisable and complex co-financing arrangements are best avoided. Second, concentrating support on a small element within a larger entity characterized by acute overall resource scarcity may lead to great pressure for the use of project facilities. equipment etc., to support unplanned activities outside the project. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY Contents Preface........................................................... 3 Basic Data Sheets ..................................................... 5 Evaluation Sum m ary ................................................ 9 1. Introduction ................................................... 15 Background .................................................... 15 2. Project Design .................................................. 17 History .. ...................................................... 17 Project Objectives and Description ..................................... 18 Gueckedou - Credit 1635-GUI ...................................... 18 ASP - Credit 1636-GUI .......................................... 19 Review and Negotiations ........................................... 20 Gueckedou - Credit 1635-GUI ...................................... 20 ASP - Credit 1636-GUI .......................................... 20 3. Implementation ................................................. 23 The Context of the Projects ......................................... 23 Gueckedou - Credit 1635-GUI ...................................... 23 ASP - Credit 1636-GUI .......................................... 24 Project R esults .................................................. 25 Gueckedou - Credit 1635-GUI....................................... 25 ASP - Credit 1636-GUI ................ ......................... 30 Compliance With Covenants ......................................... 33 Gueckedou- Credit 1635-GUI ...................................... 33 ASP - Crdit 1636-GUI ........................................... 33 Bank Performance ................................................ 34 Gueckedou - Credit 1635-GUI ...................................... 34 ASP - Credit 1636-GUI .......................................... 34 This report was prepared by John English (Task Mahager), and A. Graham Lowe (Consultant) who audited the projects in September, 1994. Constance Frye provided administrative support. 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 wihout World Bank authorization. 4. Issues and Lessons ......................................... 37 Conclusions .................................................... 37 Overall Assessment of Projects ..................................... 40 Gueckedou - Credit 1635-GUI ............... . 40 ASP - Credit 1636-GUI ............... 41 Lessons .............................................. ........ 42 Annex A - The Use of the 'Tacheron Ainenagiste' ........................... 45 Annex B - Comments From the Public Projects Administration and Oversight Agency ....................................... 47 Figures 3.1: Timeline Evolution of Gu6ck6dou Agriculture Project - PAG Cr. 1635-GUI ........................................... 29 3.2: Timeline Evolution of Agricultural Services Project - PSA Cr. 1636-GUI ........................................... 35 Maps - IBRD No. 18935R, 18936R 3 Preface This is the Performance Audit Report (PAR) of two projects: the Gueckedou Agricultural Development Project (PAG), for which Credit 1635-GUI for SDR 6.7 million was approved on November 26, 1985. The Credit closed as scheduled on June 30, 1992; and the Agricultural Services Project (ASP) for which Credit 1636-GUI in the amount of SDR 7.6 million was also approved on November 26, 1985. This Credit was closed on June 30, 1992 on schedule but with a balance of SDR 2,038,238.17 which was cancelled on agreement by both Bank and Borrower. For the Gueckedou Agricultural Development Project, the Borrower obtained project cofinancing for the PAG project from the International Fund for Agricultural Development (IFAD) for US$ 5.0 million and the African Development Fund (AfDF) for US$ 6.5 million. For the Agricultural Services Project cofinancing was obtained from France (FAC) for the equivalent of US$0.9 million. The PAR is based on the Project Completion Report (PCR) prepared by the Africa Regional Office (with a contribution from the Borrower to the PCR for the ASP), the Staff Appraisal Report for PAG (there was none for ASP), the President's Reports and legal documents, study of project files and discussions with Bank staff. An OED mission visited Guinea in September 1994. The mission met with officials of the Government of Guinea in Conakry and Gueckedou, farmers affected by the PAG, local representatives of the donor community, NGO representatives, and members of the TA teams. The collaboration and assistance of all of these persons is gratefully acknowledged. The draft report was sent to the Borrower and cofinanciers for comments. The comments received from the Borrower are attached as Annex B.  5 Basic Data Sheet GUECKEDOU AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 1635-GUI) Key Project Data (in US$ million) Item Appraisal % Actual Actual as % of Estimate Appraisal Estimate Total Project Costs 25.2 100 n.a. Credit Amount (US$ m) 6.6 26 8.64 130 a Cofinancing - AfDF (US$ m) 6.5 26 n.a. Cofinancing - IFAD (US$ m) 5.0 20 n.a. Government/Farmers (US$ m) 7.1 28 n.a. Cumulative Estimated and Actual Disbursements (in US$ millions) FY86 FY87 FY88 FY89 FY90 FY91 FY92 Appraisal Estimate 0.30 1.50 2.90 4.10 5.20 6.40 6.60 Actual - 1.97 5.37 6.90 8.35 8.64 8.64 Actual as % of Appraisal - 131 185 169 160 135 130 Final Disbursement May 28, 1901 Project Timetable Item Original Actual Preparation 02/84 09/84 Appraisal - 10/84 Negotiations - 07/85 Board Approval - 11/26/85 Credit Signing - 12/27/85 Credit Effectiveness 04/86 08/31/86 Project Completion 06/30/91 06/30/91 Credit Closing 06/30/92 06/30/92 6 Staff Inputs (in staff weeks) FY75- FY83 FY84 FY85 FY86 FY87 FY88 FY89 FY90 FY91 FY92 FY93 Total 82 Preappraisal 1.9 2.0 7.2 5.9 - - - 16.9 Appraisal - - - 76.6 - - - 76.6 Negotiation - - - .1 5.9 - - - 5.9 Supervision - - - - 7.4 14.0 10.1 6.4 10.3 7.3 5.0 9.4 69.8 Other 1.4 - 2.9 3.2 2.5 - - - - - - - 10.0 Total 3.3 2.0 10.1 85.8 15.8 14.0 10.1 6.4 10.3 7.3 5.0 9.4 179.2 Mission Data Project Cycle Month/ No. of Days Specialization Performance Year Persons in Represented" Ratingc Field Identification - - - Preparationd 83/84 - - Appraisal 10/84 4 28 E,A,F,EN - Supervision 1 02/86 2 10 E,F 2 Supervision 2 10/86 2 11 A,A 2 Supervision 3 03/87 2 12 A,F 2 Supervision 4 10/87 2 9 A,F 2 Supervision 5 03/88 2 10 A,F 2 Supervision 6 10/88 2 5 A,F 2 Supervision 7 09/89 2 3 F,A 2 a. Excess is due to change in the dollar/SDR exchange rate. b. A = Agriculturalist, E = Economist; F = Financial Analyst, EN = Engineer. c. I = Problem free or minor problems: 2 = Moderate problems; 3 = Major problems. d. Preparation was carried out by consultants who liaised with Bank staff in Washington and Abidjan. 7 Basic Data Sheet FIRST AGRICULTURAL SERVICES PROJECT (CREDIT 1636-GUI) Key Project Data (in US$ million) Item Appraisal Actual Actual as % of Estimate Appraisal Estimate Total Project Costs 10.1 n.a. n.a. Credit Amount 7.5 7.3 97 Cofinancing - 2.6 n.a. n.a. Cumulative Estimated and Actual Disbursements (in US$ millions) FY87 FY88 FY89 FY90 FY91 FY92 FY93 Appraisal Estimate 1.00 2.80 4.30 5.90 7.10 7.90 - Actual 1.61 3.12 4.77 6.18 6.69 6.93 7.22 Actual as % of Appraisal 161 Ill 1H] 105 94 88 91 Final Disbursement September 28, 1992 Project Timetable Item Original Actual Appraisal 10/84 10/84 Negotiations 07/85 07/85 Board Approval I 1/85 I 1/85 Credit Signing 12/85 12/85 Credit FTectiveness 07/86 07/86 Project Completion I 2/91 06/92 Credit Closing 06/92 06/92 8 Staff Inputs (in staff weeks) FY78 FY83 FY84 FY85 FY86 FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 Total Preappraisal - - 2.6 -- - - - 2.6 Appraisal - - - 5.7 - - .1 - - - 5.7 Negotiation - - - - 4.4 - - - - 4.4 Supervision - - - - 3.7 7.2 6.5 9.9 8.3 7.6 4.3 12.9 3.5 63.9 Other .2 5.7 1.7 .6 .8 - - - - - - - - 9.0 Total .2 5.7 4.3 6.3 8.8 7.2 6.6 9.9 8.3 7.6 4.3 12.9 3.5 85.6 Mission Data Project Cycle Month/ No. of Weeks Specialization Performance Types of Year Persons in Represented' Ratingb Problemsc Field Identification - - - - - Appraisal 1984 1 10 a - Supervision 1 02/86 2 3 a,b - Supervision 2 10/86 1 7 a 1 Supervision 3 03/87 2 11 b,c 2,2,2 F,M,S Supervision 4 06/87 1 4 a 2 F Supervision 5 11/87 Supervision 6 06/88 2 4 a,b 2,2,2,2 F,M,I,S Supervision 7 10/88 2 5 a,b 2,3,2,2 F,M,I,S Supervision 8 09/89 1 3 b 3,3,3,3,2 S,DO,C,M,F Supervision 9 06/90 1 3 b 3,3,3,3,2 S,DO,C,M,F Supervision 10 10/90 1 3 b - a. A = economist; b =imancial analyst; c = agriculturalist. b. I = no significant problems; 2 = moderate problems; 3 = major problems which are being addressed adequately; 4 = major problems not being addressed adequately. c. F = financial; M = management I = development impact; S = overall status; DO = development objectives; C = compliance with legal covenants. 9 Evaluation Summary 1. The Bank started lending to Guinea in 1968 during the regime of President Toure. Lending in the agricultural sector was limited to three small pilot projects in livestock, pineapple and rice production. The latter operation was the precursor to the Rice Project (ORG) based in the region of Gueckedou, which in turn led to the preparation of the two projects under review. 2. At the time of preparation (1981-83) and appraisal (1984) of the two projects audited, the Bank assumed the role of pioneer in the financing of both agricultural and institutional development efforts in Guinea. Other members of the donor community, while assessing their own priorities, were content to see the Bank take the lead. This took place in an emergent, liberal economy with a nascent government structure hampered by limited infrastructure and equipment, as yet fragile free-market traditions in civil servants and farmers, and a dearth of trained, experienced Guinean technicians. Unfortunately, the very difficult conditions in the country at the time limited the numbers of qualified foreigners who were willing to work there. Project Design 3. The two projects were originally one, which as initially prepared, envisaged extending the activities of ORG to further regions of the country. Preparation started in the centrally- planned economy of a socialist regime; appraisal was completed in a newly-emerging liberalized market economy. The approach was hastily reformulated in response to the rapidly changing Guinean political climate of 1984, following the death of President Sekou Tour6. At appraisal in October 1984, the two major components of the proposal were separated into two projects, one a regional project, (the Gueckedou Agricultural Development Project); the other providing support in policy analysis and planning to the newly-established Ministry of Rural Development (the Agricultural Services Project (ASP)). GUECKEDOu AGRICULTURAL DEVELOPMENT PAOJECT (CREDIT 1635 - GUI) 4. Objectives. The project aimed to increase productivity of smallholders by expanding the area of ORG's operation and introducing new technical packages for rice (both upland and 'swamp'), coffee, groundnut and cassava; combining project and Prefectoral Directorate of Agriculture activities into an effective research and extension service; improving technical packages through applied research; and developing farmer organizations. 5. Implementation experience. Financing was originally intended to be simultaneously available from three sources: IDA (6.7 m SDR), AfDF (6.65 m SDR), IFAD (5.05 m SDR). However, difficulties were experienced, arising from IDA and AfDF cross effectiveness requirements. While the IDA Credit was effective from October 1986, the AFDF loan was not effective until December 1988, leading to rapid draw down of IDA funds and back-to-back, rather than concomitant, funding. 10 6. A credit scheme funded by the International Fund for Agricultural Development (IFAD) was abandoned in 1990. No credible system had been established by the project, a large part of the IFAD loan remained undisbursed, farmer reimbursement was variable, and those funds which were recovered were misused by project management. At closure, the credit scheme had a zero balance, and an intended revolving fund had never been established. 7. The Project suffered greatly from mismanagement of funds, and the cavalier attitude towards accounting and fiscal responsibility on the part of the project administration had repercussions at both the organizational and operational levels. Particular problems were experienced in the project accounting systems. The PCR blamed this largely on inadequacies of the Financial Directors, who were part of the TA team. However, project design cannot escape its responsibility. The project was virtually the only significant source of funding for government structures in the region. Pressures on the project to 'help out' were inevitable. Also the Financial Director (FD) was a member of the TA team, responsible to the locally directed project. Separate responsibility for financial management would have been preferable. Results 8. The project developed a cadre of staff at the regional level, with a primary focus of support to smallholder agriculture, including research and input supply, where none had existed before. However, field observations showed a lack of motivation or technical objectives among field agents who had been given scant material or managerial support. Data on actual performance does not exist, but it is doubtful that 17,000 families were really contacted as claimed in the PCR. Difficult logistical conditions hampered technical assistance (TA) staffing efforts, mitigated TA impacts and created problems in procurement. Somewhat fortuitously, the project supported an expansion of agricultural production in a region of the country which has borne the brunt of the refugee influx from Liberia since the late 1980s. In turn this local influx increased the local demand for food crops, particularly rice. 9. Varieties and techniques which had been introduced under ORG were extended to more families. The promotion of 'appropriate technology' packages to farmers for paddy rice production under local conditions was particularly noteworthy. This made use of an innovative approach to the management and support of micro-level swamp-development, (using contracted 'land technicians' to supervise work) which has been extensively replicated elsewhere in the country. However, little emerged from the applied research program and, with hindsight, Gueckedou was not in a position to benefit from large scale adoption of the T&V system, and would have been better served by a more limited structure, which could have been better supported. 10. Since closure of the IDA credit, Government has taken steps to privatize some of the activities undertaken by the project (seed production, input supply, and road maintenance) in ways which may also prove replicable, and the extension and swamp development activities have formed a basis for broader national programs. 11. On the other hand, because of the failure of the monitoring system, it is essentially impossible to know what the direct result of the project was. Certainly, the general consensus outside the project is that the achievements claimed by it are overstated. The LAC-23 rice variety, which was the basis of its effort in upland rice production, was developed in Liberia 11 in the late 1970s and was generally available before the project started, so that it is overstating the case to attribute its use to the project, although its support and seed distribution efforts were a positive factor. In addition, the sustainability of the structures created by the project is still uncertain, but they represent a logical step by the Guineans to adapt the structures of the project to the evolving needs of the country. 12. Overall, despite its positive elements, the project is judged to have been unsatisfactory. There was significant leakage of resources and the results, while tangible in several respects, were not commensurate with the resources used. The institutional development achieved by the project was modest and the sustainability of its structures is uncertain. AGRICULTURAL SERVICES PROJECT (CREDIT 1636-GUI) 13. Objectives. The objective of the project was to enhance the capacity of the Ministry of Rural Development through the establishment of a Bureau for Strategy and Development (BSD) within the Ministry. The project supported BSD through the provision of substantial technical assistance, facilities and equipment, and local and foreign training. 14. Implementation experience. Weak technical assistance and BSD management, political interference, poor financial management and lack of counterpart resources all contributed to poor performance. The TA team was terminated early. The PCR characterized this project as being an unqualified failure, on the grounds that no specific studies or planning documents were produced as detailed in the project documentation, and there was significant misappropriation of resources. Certainly, the TA financed by the credit was not cost-effective, and there was substantial leakage of project resources, for similar reasons as occurred under the Gueckedou project. 15. In the view of the audit, this overall performance represents the inadequacy of the project design as much as the failure of the TA team. (a) It was unrealistic to expect that BSD would produce a lot of reports and prepare projects at a time when most donors wished to establish their presence, make their own judgements, and prepare their own proposals. (b) In a rapidly changing situation, the project design was too rigid. Long-term TA should have been reduced, with an increase in funds available for short-term TA. (c) The project should have recognized the lack of local resources elsewhere in the ministry and might have made provision for some level of support for the ministry as a whole, (eg. buildings and equipment). (d) The overly large structure initially created under the project, with 40 positions, has been streamlined, with the hiving off of half the initial units (eg. statistics, agricultural research, and rural infrastructure). This was a desirable move, so that BSD's successor (BCPA) can concentrate on its core role in policy and strategy formulation and in coordination. As a consequence, the Ministry of Agriculture now has a greater capacity in these areas than most of the technical ministries. 12 16. However, the primary objective of the operation, the establishment of a capacity in the agricultural ministry to undertake planning, strategy and policy formulation and project analysis has been achieved, with the help of continuing assistance from other sources. One of the successes of the project was the selection of a high quality local cadre for BSD. Many of these persons are now occupying senior positions inside and outside the ministry and have formed the basis for this improvement. It is generally accepted that the Ministry of Agriculture is now better placed in this regard than other technical ministries. Because of the strength of this local staff, it was unfortunate that the initial proposal for an extensive program of locally based training was not followed up.' 17. Rating. Overall, the audit concurs with the PCR a3sessment of the project outcome as unsatisfactory. While recognizing the difficulties faced, steps should have been taken by the three parties, Government, Bank and TA team to agree and achieve redefined objectives. Given the level of inputs provided, institutional development must be judged to have been modest at best, but the present entity is sustainable. Bank Performance 18. Both projects suffered from clear deficiencies in design. Bank supervision of the Agricultural Services Project, in particular, was inadequate. Despite the emphasis on specific outputs in the Terms of Reference of the TA team, and the use of their absence as a reason for project failure, there is no record in supervision of these elements having been systematically reviewed, or of any recording of outputs against targets, or of steps to revise the overall work program. There is also no indication of efforts being made to implement local training programs. Former staff of BSD who were interviewed complained that supervision missions did not focus adequately on training issues or on the overall lack of collaboration in BSD, and did not follow up on the issues even when raised. Lack of continuity in supervision did not help. Maintenance of project documentation was also inadequate. Lessons 19. Institutional support in times of change must be flexible and well supervised Given the newness of the Ministry structure at the inception of the ASP project, and the radical change in government strategy, it was ill-advised to design the project with such a heavy contingent of long term technical assistance, to define the program in such detail and to tie financing to the defined program. The response should have allowed for a core long-term TA (one or two specialists) with the facility to provide consultative services on an ad hoc basis. Financing could have been on a semestrial or annual tranche basis with the program being presented and reviewed with each tranche report. Supervision did not assist in modifying the projects to evolving circumstances. 20. Islands of "plenty" in the midst of seas of 'paucity' should be avoided. A major difficulty for the ASP project was the 'gravitational pull' generated by the financial mass of the project budget in contrast with the other resources available to the host Ministry. The imbalances caused by the monolithic focus of the project on the BSD created tensions between 1. This para. has been modified in light of comments by the Borrower. 13 the project and the Minister, which could have been foreseen. Some mechanism should have been devised to include benefits to other departments eg. through Ministry wide training sessions. Geographical isolation reduced the impact of this syndrome in the case of Gueckedou, although isolation also led to abuses and mismanagement. 21. Smaller IDA funded operations may be preferable to larger multi-donor operations in conditions of great uncertainty. The Gueckedou project was almost fatally handicapped by the delay in AfDF funding. Under conditions such as those of Guinea in the early 80's the likelihood of one of a group of donors not being able to declare an operation effective is high, with inevitably disruptive effects in conditions, which are already difficult for project management. A smaller IDA funded project, with a more limited extension effort, would have been preferable. 22. The apparent success of the 'Land Technician' suggests scope for more flexible, less monolithic structures for rural development. This approach, in effect subcontracting supervision of field operations, has potential as a means of providing public support but avoiding creation of bureaucracies. Directly hired project staff tend to outlive their usefulness and their transfer from one project to the next is seldom smooth. Effective use of a more contractual approach would require continuity and good overall supervision and checking by the implementing agency. It is not clear that the latter has been achieved in Gueckedou. 23. Steps need to be taken to define responsibility within the Bank for maintaining project documentation, especially when the project is managed from afield office. It is not clear that in this case, material was systematically sent, or copied, to headquarters for filing. In addition, during significant reorganizations, as in 1987, during which personnel and responsibilities were transferred among units, misunderstandings can occur and significant material be lost.  15 1. Introduction Background 1.1 Guinea is a medium sized (246 000 km'), sparsely populated (23 hab/km) country situated on the western coast of Africa. The country may be separated into four major ecological zones: Upper, Lower, Forest and Maritime Guinea (See Map 18936R). The rainfall through these zones varies considerably from as low as 1500 mm in the North bordering on Senegal and Mali, to some 4500 mm on the Coastal Plains close to Conakry. Centuries of prevailing tropical conditions, relatively high rainfall and temperatures, have contributed to the weathering of old surfaces to yield some of the world's richest bauxite deposits and soils which, for all their abundant vegetative cover, are poor in nutrients and rapidly depleted. 1.2 Colonized by France at the beginning of this century, the country saw agricultural productivity reach its peak in the 1940s and early 1950s through plantations of forestry and agricultural products for the European markets. Large rice farms reclaimed from mangrove swamps showed benefits to major infrastructural investments (feeder roads, dams, drainage), economies of scale and commensurate organizational skills. Exporters also enjoyed access to markets yet to be dominated by Asian or North American rice production. 1.3 After Independence in 1958, the country saw the disintegration of the colonial infrastructure, the disruption of colonial market networks and the installation of a centrally planned economy, all of which led to a profound degradation of the country's economic performance. Despite government efforts at nation building and local language education, the population became insular, lost literacy, and reverted to subsistence modes of agricultural production. 1.4 The death of President Sekou Tour6 in early 1984 led to radical changes in the government of the country as the new leaders espoused the principals of a liberalized, market led economy. This new orientation dictated profound changes to the established order of political, economic, social and productive life of the Guineans. 1.5 Following the change of regime, the overall strategy of government toward the rural sector'changed. The new objectives were to: (a) seek to increase small holders' productivity; (b) develop satisfactory incentives for farmers as a part of the first objective; (c) enhance the Government's capacity for policy formulation, investment planning, project identification and preparation; (d) improve overall organization of rural development services; and (e) help create efficient and effective field level support services (research, extension, data collection, monitoring and evaluation, credit and input supply). 1.6 As of 1985, the population of Guinea was estimated at some 5.8 million inhabitants of whom 90 percent had no access to safe water, 80 percent (adult population) was illiterate and 75 percent lived in rural areas, depending on agricultural production. Emerging market 16 networks were rudimentary and feeble as were the nascent national administrative cohort and the ensemble of rural institutional support mechanisms. 1.7 Bank loan history in Guinea dates from FY68 and until 1984 had been primarily in infrastructure (roads, ports, water, and urban development) energy (hydro power, petroleum exploration) and education. Lending in the agricultural sector had been through three projects: Daboya Pineapple Project (FY76), Rice Developmew. Project (FY80) and Livestock Development Project (FY81). 1.8 The Bank's Agricultural Sector Review for Guinea, published in April 1984, identified several areas of priority. Emphasis was placed on a strategy of support to small-holders as producers with improved inputs in the form of fertilizers and seeds, better husbandry techniques, simple water control and irrigation techniques. Recommendations were also made for the restoration of producer incentives (relying on private market forces through trade liberalization), formal decontrol of all priority commodities, institutional pluralism, and the dismantling of administrative trade barriers. The study indicated the lack of capacity for sectoral planning and strategy development in the rural sectors as a key institutional constraint. Sectoral training and support services were required in adaptive research, extension and input delivery. 1.9 Prior to the change in government in April 1984, three ministries were responsible for activities related to rural development in Guinea. These were the Ministry of Agriculture and Forestry, the Ministry of Livestock and Fisheries, and the Ministry of FAPAs (Fermes Agro- Pastoral d'Arrondisement-state farms). The FAPAs and other state agencies had their own projects and were more or less autonomous. Under this system there was really nobody at the center who could provide coordination and leadership. Responsibility for extension activities was shared between the Ministry of Agriculture field staff and the FAPAs but, in practice, there were no effective support services to smallholders. In fact, the impact of government on smallholders was probably negative as, in principle, farmers were required to deliver a 'quota' of their major product to the state. The result was to provide a disincentive to production and an impetus to rural people to shift to non-agricultural and urban activities. In April 1984, the Ministry of FAPAs was mergcd with the Ministry of Agriculture and, in an administrative reorganization in December 1984, a single Ministry of Rural Development was created by merging the Ministries of Agriculture and Livestock. This single ministry was then responsible for all aspects of rural development, including extension. 17 2. Project Design History 2.1 The Rice Development Project (Operation Riz Gueckedou-ORG) was the precursor to the Gueckedou Agricultural Development Project (Credit 1635-GUI) and the Agricultural Services Project-ASP (Credit 1636-GUI). The project was located in the vicinity of Gueckedou, close to the borders of Liberia and Sierra leone, in the forest zone (see Map 18936R). Initially, planning for a follow on operation envisaged two components in a second, expanded phase to the ORG project. The preparation of a 'Rice II' project was started in 1983 as the execution of the earlier, pilot ORG drew to a close. The definition of this second phase ORG was strongly influenctd by the prevailing economic and political situation of the country (still centrally planned) while taking into account the lessons and issues emerging from the preceding experiment. Project briefs indicated the following objectives: (a) the development of ONADER (National Rice Development Agency) into a truly national institution with responsibilities for the oversight of rice production in Guinea (b) the expansi.i of research facilities with stations being established in Upper and Maritime Guinea, in addition to Forest Guinea; (c) the expansion of areas benefitting from bottom-land development (am6nagement des bas-fonds) and from improved rainfed rice varieties and techniques; and (d) the establishment of a large flexibly-administered budget destined for studies, short-term consultancy input and training fellowships. 2.2 By April 1984, as the country adapted to the passing of President Sekou Tour6, the project was evolving along two major axes. The first axis was comprised of the consolidation and reinforcement of the ORG experiences through expansion of the field areas developed by the project in the Gueckedou region and for the extension of support activities to two other areas in Guinea-Conakry-coast and Siguiri (Maritime and Upper Guinea respectively). The second axis involved the creation of a mechanism for supporting the new government structures for the preparation of a National Food Production Policy. 2.3 Concerns over the viability of this approach were confirmed at appraisal. In the case of Maritime Guinea, the EEC was planning activities and it was agreed that they would cooperate with a Bank supported project but proceed separately. The mission determined that farmer interest in the savannah and savannah fringe zones had not been sufficiently demonstrated for them to be included in a project. The Bank and the Government agreed that the project would confine its field activities to Gueckedou. It was also agreed that this would be extended to cover the rehabilitation/replacement of long standing coffee plantations, and the improvement of food crops other than rice (eg. cassava, groundnut, maize). Thus, it would be an area development project and not a sectoral one. It was also agreed that a project with an 18 area component, and a large component of support to the ministerial headquarters, would be unwieldy and that it would be best to create two separate operations. Project Objectives and Description 2.4 From appraisal, therefore, two projects were processed in tandem; the Gueckedou Agricultural Development Project (PAG) and the Agricultural Services Project (ASP). Gueckedou - Credit 1635-GUI 2.5 The objectives identified in the SAR sought to build on experience of the pilot, rice development component of the ORG. Results from ORG had been particularly encouraging with respect to the response obtained from smallholders in the area. Project objectives addressed the need to: (a) increase the productivity of smallholders in rice production by expanding on the scope and area of the first phase project: and the introduction of new technical packages for coffee, groundnut and cassava production in the expanded area; (I' reactivate the Gueckedou Prefectoral Directorate of Agriculture by completely absorbirn ;ir activities and those of the project into a single regional agricultural development service; (c) improve available technical packages for the broader forest region through applied research; (d) help development of farmer organizations. 2.6 The project was to span some five years and consisted of six major components: (a) project management and administration including a Liaison Office in Conakry; (b) extension services for about 16,000 farm families involved in cultivation of irrigated and upland rice as well as other rainfed and perennial crops, including an effort to reach women with the extension message, as well as training for Guinean staff, and agricultural research; (c) seed processing and distribution to farmers from seed produced on a mechanized seed multiplication farm and through contract seed farmers; (d) land development of some 2,000 ha of swamp (bas-fonds) through low cost irrigation works; (e) upgrading of 240 km of primary, secondary and tertiary roads and maintenance of these as well as 260 km of existing roads, and a garage to repair and maintain project equipment; (f) input supply and agricultural credit for project farmers; and 19 (g) provision of technical assistance and consultant services as well as training facilities and fellowship program for the project staff. ASP - Credit 1636-GUI 2.7 Prior to 1984, there was little sectoral and sub-sectoral planning within the technical ministries. They concentrated on project preparation, which was undertaken only when projects were expected to be submitted for donor finance and, generally, with heavy donor involvement. Once the overall national plan, prepared by the Directorate-General of Planning and Statistics, was adopted by the political bodies, projects were examined for conformity with its largely non-economic objectives. 2.8 At the same time as the new MRD was created (para. 1.9), two old entities were merged into the new 'Bureau de Strategie et Developpement' (BSD); the 'Bureau d'Etudes' of the former Ministry of Agriculture; and ONADER, established under the IDA funded Rice Development Project. This new planning unit of the MRD, like those of the other technical ministries, had a very limited capacity to undertake studies and develop strategy. It lacked statistics on the rural sector, and did no systematic gathering of information, which greatly hampered its ability to adequately plan activities. Conditions at the time in Guinea were extremely difficult . The ministry also lacked adequately trained personnel, office space and equipment, basic supplies and reference material, and the means of getting to the field. 2.9 The objective of the Agricultural Services Project was to improve the capacity of the MDR in rural sector planning, strategy and policy formulation and project analysis, through support for the establishment of the BSD within the ministry. This was to serve as the main advisory body of MDR. It was to have five divisions, four technical and one administrative and financial, which were to implement or coordinate the implementation of the project. The project was planned to span seven years and its aims were: (a) development of a sectoral strategy through defined policies and the elaboration of a three-year rolling investment plan; (b) training of Ministry staff in the identification and preparation of new projects and in the analysis and review of proposed investments in the sector; (c) establishment of monitoring and evaluation units within projects through the training of staff, preparation of work programs and the supervision of such programs; (d) establishment of a statistics and data collection capacity within the Ministry with training on methods, surveys and work programs; (e) support of agricultural research and of a small-scale irrigation strategy; (f) logistical support to new projects. 2.10 The major elements of the project were to be; a substantial training program for Ministry staff through seminars and short- and medium-term fellowships; investment in buildings and infrastructure for the Ministry; the provision of technical assistance (in the 20 divisions of the BSD covering the various objectives: rural sector planning; agricultural economy; monitoring and evaluation; agrostatistics; research; rural engineering) and; consultancy inputs for a PPF-financed, ministry reorganization and master plan study, the preparation of an agro-industrial master plan and specialist short term consultancies in other areas such as, training, sociology, fisheries, forestry, and irrigation. Review and Negotiations Gueckedou - Credit 1635-GUI 2.11 During review the major area of concern related to the Gueckedou project was whether the Government would actually follow through on liberalization, and agreements were reached that the coffee export agency would maintain its purchase price at a level of at least 80 percent of the market price, input prices charged to farmers would not be subsidized (in order not to create conditions in which private traders would not handle inputs) and that sales of imported rice would also not be subsidized. Concerns were also expressed over financial autonomy of the project and at negotiations it was agreed that IDA and IFAD would not engage in joint financing, but would use parallel financing. It was considered that this would simplify accounting and allow separate special accounts to be set up. 2.12 The project was approved by the Board in November, 1985 for a Credit of SDR 6.7 million. It was cofinanced by the International Fund for Agricultural Development (IFAD), (SDR 5.05 million), and the African Development Fund (AfDF) for a sum of 6.65 million Units of Account (UA). The IDA credit was to cover the foreign costs of half of the vehicles, equipment and spare parts, the total foreign costs of the resident technical assistance, and the costs of agricultural inputs for the first year. The AfDF credit was to finance, in parallel, the foreign costs of half of the vehicles, equipment and spare parts, the total costs of building construction, and the foreign costs of fuel and lubricants. In addition AfDF included funds for parallel preparation studies in the forest zone, support for certain rural development activities and the equivalent of US$0.7 million in local costs. IFAD was to finance, in parallel, the costs of fellowships, consultants studies and audit, and the remaining foreign costs of the agricultural inputs. ASP - Credit 1636-GUI 2.13 The Agricultural Services Project was treated as a technical assistance project (although it was not so named). Normal preparation stages, including separate appraisal were foregone and only the President's Report was produced. The audit is left with the impression that there was little real dialogue between the Bank and Government on the details of the operation. At the time Government was in the process of major policy and institutional changes and was groping in the dark and, therefore, had limited basis for pressing strongly on details. Within the Bank concerns were expressed on the size of the TA input, the need for a clear program, with a work calendar and specific outputs and training program, and on the potential for overlap with the national statistics agency in developing a capacity for data collection. 21 2.14 The project was approved by the Board at the same time as the Gueckedou Project for a credit of SDR 7.6 million (US$7.5 million equivalent). Co-financing from FAC amounting to US$0.9 million was to cover two of the seven TA positions financed by the project.  23 3. Implementation The Context of the Projects 3.1 General Conditions. All interlocutors agree that conditions in Guinea were extremely difficult at the time of appraisal and implementation of the two projects. The Ministry of Rural Development, created from the former ministries, had limited resources of buildings, equipment, or trained personnel. Work traditions of government officials were still influenced by the habits of thirty years of socialist rule. State-owned facilities were inefficient, supplies of goods and food in the country were limited, and services were rudimentary. Not only did the projects have difficulties obtaining material and equipment, but it was also difficult to find experienced technical assistance prepared to undergo such hardships. These day-to-day problems, compounded with the general confusion of working in a context in constant evolution, had a profound impact on the implementation of the projects. However, they also constituted the major justification for a regional concentration of efforts in the outlying region of Gueckedou, and for a substantial support for institutional strengthening of the Ministry of Rural Development in Conakry. Gueckedou - Credit 1635-GUI 3.2 Cross-effectiveness Problems. The financing arrangements described above were overly complex and almost bound to cause trouble. Implementation problems arose immediately when AfDF did not declare its credit effective because of a general cessation of lending. Cross effectiveness conditions in the Credit Agreements were waived in the interests of project implementation, and the IDA credit was effective as of October 1986, while the effectiveness of the AfDF credit was delayed until December 1988. As a result, since IDA funds had to replace AfDF funding in the intervening period, by the time the latter financing came on line the IDA loan was already 70 percent disbursed, so that, essentially, there was no collaboration between the two agencies.' 3.3 Financial mismanagement demoralized technicians. The project suffered greatly from mismanagement of funds. The impact of administrative laxity was felt from the beginning, when funds assumed to be remaining from the first phase were discovered to have already been committed, and some US$800,000 above the amount available from the first project were, in fact, owed to suppliers for goods delivered in the first phase. 3.4 A cavalier attitude towards accounting and fiscal responsibility on the part of the project administration had repercussions at both the organizational and operational levels. In the first case, the lack of a clear management framework or of organization-wide accountability (work programs, reports, paper trails, delivery monitoring) led to the absence of equipment and resources and, over the long haul, to the demoralization and demotivation of the staff at all levels. Indeed, given the limited resources which reached field level technicians I. The Borrower in its comments suggests that it would have been preferable for co-financing arrangements to have been renegotiated at this point. 24 (eg. motorcycles were never delivered to field staff despite payments having been made to two separate suppliers) it is moot to question the efficacy of an extension service (itself swollen with 'extras') which was essentially 'a pied nu'. The difficulties experienced in establishing an operational monitoring and evaluation component for the project are not surprising in this context. The efficiency of the project execution is even more difficult to assess given the lack of firm data on its achievements. Overall, the impression given is that results were not commensurate with the costs incurred. More could surely have been done with the resources (human, financial and material) available in terms of areas developed, roads improved or farm families serviced by the organization. 3.5 Accounting difficulties from the start. Particular problems were experienced in the project accounting systems. According to the PCR, the lack of qualified local accounting personnel was overcome by the use of expatriate Financial Directors (FD). However "none of the three expatriate FDs was able to satisfactorily regularize the management of the project's finances." As a result accounts were closed late and, hence, audits were always late and conditional. Limited accounting experience is cited in the PCR as reason for this situation, and in this case, the use of third party accountants (other than the Borrower and the TA) of international repute, mandated to establish, implement and monitor project accounting systems, could affirm government commitment to the financial responsibility of its officers, and liberate TA project management from pressure to step beyond the confines of the project mandate. It is the judgment of the audit that the organizational arrangements were equally responsible with the inadequacies of the TA personnel, for the failures of the projects. ASP - Credit 1636-GUI 3.6 Weak Technical Assistance could not handle difficult conditions. The Bank expressed reservations about the initial choice of Technical Assistance consulting company made by Government. However, the Borrower was adamant and the Bank finally respected this choice, only to find that the delays caused by these discussions were used by the consultant company to claim that members of the team originally proposed were now no longer available and, as a result, substantial changes were made in its composition. From this point on a distinct lack of continuity characterized the TA delivery. Particularly crucial was the post of Macro- economist/Team Leader which was filled by four different specialists during the period before the project was closed 2-1/2 years later. The impact of discontinuity in two basic areas and posts was a major factor in the unsatisfactory performance of the team as a whole. The ability of the Bank to act decisively in the early period of the project was undermined by the earlier expression of reservations and subsequent capitulation. This had the effect of requiring the Bank to give the TA team the benefit of the doubt when early problems arose. 3.7 Selection process attracted high caliber nationals. Guinean staff for BSD were selected from the existing civil service by means of two tests, one general and one specific. About 600 candidates applied for the 39 initial positions. Given the large number of candidates for the BSD positions, the selection process laid the basis for one of the project's most tangible successes, i.e. the identification and employment of high caliber cadres. To the present day, the impact of these cadres is appreciated and valued by all interlocutors. The only sour note on this level was the lack of responsibility given to Guinean cadres in the original project design, indeed the placement of expatriate consultants in administrative functions swamped the TA in day-to-day management problems, and gave them little 25 institutional structure to build on or to train. The need for a more flexible support mechanism will be discussed later. Project results Gueckedou - Credit 1635-GUI 3.8 Implementation began immediately as the TA and management teams were largely held over from the ORG project. However, the funding hiatus caused some operational delays. Because of the lack of AfDF funding the bulk of the costs of the planned program were initially borne by the IDA credit and, as a result, this was largely disbursed by 1990, when the Bank essentially terminated its supervision inputs. The project limped along until 1992, when a revised agreement was reached between the AfDF and Government, under which a reduced program of operations is still continuing. 3.9 Poor reception for coffee. During colonial times Gueckedou had been a center for smallholder production, but this had fallen away. As a result, when the project started, most of the existing stands were over 30 years old. It soon became apparent that most of them did not warrant rehabilitation, as foreseen by the project proposal, and it was agreed that the project could concentrate on promoting new planting, and the target was raised from 1,000 to 1,200 hectares. Arrangements for seed were made with IRCC in Cote d'Ivoire, but arrangements became entangled in bureaucracy. Supplies were slow, quality was often poor, and germination low. At the same time low coffee prices did not create a conditions in which farmer interest was high. While certain farmers continue to show enthusiasm for coffee, in one case visited, the farmer was incensed by the fact that new coffee plants given to him under a credit mechanism of the PAG had yielded very little, and that he was expected to pay again for other varieties under the AfDF financed 'Rehabilitation Cafe II'. Recent improvements in world prices for coffee have served to heighten interest in RCII and may lead to renewed attention to project promoted stands of coffee. 3.10 Research no longer pertinent. The rice component of the research program is singled out by the PCR for faint praise. The ORG project had begun the dissemination of LAC23, an upland variety, which was well received by farmers, and this program continued. The research effort then entered the almost inevitable plateau phase in which significantly improved varieties are hard to find. About five varieties were shown to outperform LAC23, but interest by farmers was reported to be low. This appears to have been because these varieties were more reliant on fertiliser for good performance, and availability of the latter is not guaranteed, and because of poor feedback in the extension system which did not help clarify farmer needs and preferences to the researchers. Although there were difficulties with the siting of swamp rice research fields, the team managed to study some 600 new varieties, preselected 45, performed multiple site trials on 15, and promoted four varieties to the farmers, and which have elicited interest. 3.11 Other problems affecting the research program included the fact that there was inherently limited potential to new varieties in cassava and peanut. Cassava is a staple reserve, grown after upland rice and other food crops and usually harvested as needed. Higher productivity is, therefore, not a priority. Mealy bug was not reported to be a problem. The area is marginal for groundnuts, due to high rainfall and storage is difficult. Production is for 26 subsistence and some immediate sale. Increased productivity is, therefore, also not a priority. Women are also heavily involved in groundnut production and it was not until the Women's Activities Division concentrated on this crop that significant progress was made. As a result, research efforts in these crops had little relevance to farmer interests. Seed quality may also have been compromised by genetic degeneration of the base stock which was not adequately renewed after its introduction. 3.12 Extension effort shows few results. Partly for reasons indicated above, there is little evidence to support the PCR statement that the extension effort was successful. Indeed the PCR itself shows that very few of the proposed 'technical improvements' have been implemented by the farmers. At the initiation of the project, ORG had already introduced improvements in rice production, especially LAC23. Because of market limitations and cultural factors (such as those related to groundnuts), there were no breakthrough opportunities in the alternative crops. Some progress was made with the promotion of the integrated system known as 'complexe bas-fonds' (see below). However, the research extension effort was faced with a situation where improvement reached somewhat of a plateau, in the absence of any research breakthrough and with little farmer incentive for change. In the field, extension officers reported difficulties in transferring technical packages based on improved seed and fertilizers. With hindsight, the view at appraisal was overoptimistic. Use of the T&V system was in vogue, and the results of the ORG suggested that there might be further opportunities for widespread extension. These hopes proved false and the extension effort was left with little to extend and fell back on formulaic approaches. It is indicative that certain extension agents interviewed reporting having difficulties in getting farmers to 'semer en ligne' (sow in rows) at a time when their superior officers reported that this theme had been dropped from the extension package. Extension results were also criticized in the proposal for reformulation of the Gueckedou Project prepared by the BSD in 1992.' In this document results were said to be limited and agents demotivated with almost half the personnel in one of the zones having quit their posts.' 3.13 Evolution of an approach to development of 'bas-fonds'. The Gueckedou region is well suited to rice production, with a short dry season and an annual rainfall of about 2,000 mm. One of the successes of the project has been the evolution of an approach to the development of 'bas-fonds', primarily for rice production. These low-lying areas are waterlogged for long periods during the rainy season and have a high water table at all times. The project's approach had three significant elements; concentration on small areas; the development of a 'subcontracting' method of works supervision; and the encouragement of intensification of cultivation on the area adjacent to the bas-fonds. (a) Size. The project focussed on the improvement of small areas, generally 5-10 hectares or below. This had the advantage of simplifying the works required and reducing the 2. "Proposals for the Refornuladon of the Gueckedou Agricutural Project", June 26, 1992: para. 2.45 'The extension system (Training and Visit) requires very close supervision and great rigor in its application. Because of a lack of necessary resources and also because of negligence, these requirements are far from being respected and the impact has been limited.' 3. 16id "...the organization (171 agents and managers) is difficult to manage and ... is often paralyzed by lack of resources. Absenteeism is substantial ... almost half of the agents in zone II are recorded as having quit their positions." 27 number of farmers who had to collaborate with the project and each other (often only one was involved). (b) Supervision. The problem of supervising and recording work on small sites was overcome by the development of the concept of the 'Tacheron Amenagiste' (see Attachment). This person has a responsibility somewhat akin to that of a resident engineer on a large engineering project. Most commonly he was a former public employee with agricultural experience. Once the project had inspected a proposed site and agreed to support its development the 'tacheron' was hired, for a fixed per hectare fee, to supervise the required work. He surveyed the site, agreed with the farmer(s) the extent of the work required, ordered any materials, and validated the satisfactory completion of the work, subject to final inspection by the project. This approach has been adopted generally in Guinea for bas-fonds development efforts. (c) Technology. An effort has also been made to combine the improvement of bas-fonds with that of the immediately adjacent areas in an approach referred to as the development of the 'complex de bas-fonds'. In effect this has been an effort at watershed management on a micro-scale, and has involved the planting of oil palms (or other water tolerant crops) immediately adjacent to the bas-fonds where the water table is high and upland rice, cassava, groundnuts or coffee, higher up the slopes. The circumstances were propitious for this emphasis on intensification, given the influx of refugees from the turmoil in Liberia and eastern Sierra Leone. Those farmers who have adopted this approach are producing for the market and have been much more interested in further improvements, and are willing to spend for fetilizer and use varieties which require them. 3.14 Agricultural credit scheme was a fiasco. In this context it is perhaps not surprising that the history of agricultural credit in the zone was chequered to say the least. The scheme envisaged by the SAR was to have been supported by IFAD to the extent of some three and a half million dollars. An initial sum of $300,000 dollars was distributed from IDA funds. Early ORG attempts had resulted in repayment rates of around 50 percent and the Gueckedou Project was to improve on this by using a system based on individual credits. While initial repayments were of the order of 70 percent, the lack of supervision and repayment management led to the degradation of repayment rates to some 55 percent in 1991, having seen an all time low of 51 percent in 1989. Record keeping was also poor. Despite warnings from supervision missions, corrective measures were not applied and, in fact, more than 102 million FG recovered (US$170,000) was used by the project for operational costs. After freezing funds, IFAD decided to reallocate the major part of the original funds to a national project. Under this project, a convention was signed in 1990 with a locally established Credit Mutuel for the use of some US$800,000 to cover the set-up and running costs of the project for two years. Results from this project seem encouraging. However, the principles on which the Credit Mutuel is based and the smallness of the sums received on deposit and subsequently loaned by it mean that seasonal agricultural credit is still not available. The attempts to set up revolving funds in the Credit Unit of the project were to no avail. 3.15 Road repairs soon washed away. The project site was visited towards the end of the rainy season. Roads were found to be variable in condition: most were fairly badly rutted but passable, and in two cases they were in serious disrepair (one due to a broken culvert, the other, which was impassable, due to waterlogging and rain damage). Road and track 28 improvement, said to have been done early in the project, has not had a particularly long life span. Project engineers explained that, by design roadworks were somewhat superficial (from concerns to keep costs down), had a life span of only some four years, and required regular upkeep. The BSD document (1992) evaluates the road improvement as being more akin to regular road maintenance, with the subsequent maintenance works being of a consequently lower standard. While costs for such interventions are considerably lower than major road works, the costs of maintenance still exceed the available resources of the responsible regional services. Road maintenance is still undertaken on an ad hoc basis through financing by various projects in the area, and sustainable improvement and maintenance will depend upon national strategies for the regional and local networks and, possibly, on the transfer of responsibilities and tax resources to the local, elected government bodies ('collectivites decentralisees'). 3.16 Privatization and Decentralization. With the reduced level of funding available since the end of the 1980s the government has encouraged the project to hive off or privatize some of its activities, and is continuing to incorporate, others into national programs.' Figure 3.1 summarizes this evolution. Some of these efforts have been noted above, the development of the Credit Mutuel and the coffee producers associations, and of the 'tacherons amenagistes'. This process has been taken a stage further with the privatization of seed production. The management of the seed farm has created a company (SPCIA), which has taken over the assets of the seed farm, with the objective of continuing seed production on a commercial basis and the supply of inputs. The road maintenance unit has similarly been hived off into a self financing Mechanical and Public Works Group which is undertaking road maintenance and renovation on a contractual basis. This is no longer part of the project. If these steps are successful, they could be replicated on a broader scale, as has been done with land development. 3.17 Trained cohort of Guinean technicians. The primary project benefit identified by the Guinean field team was that of the training of team members either through on-the-job efforts, local seminars or out-of-country training facilities. While this impact was mitigated by the demoralizing style of project management (as mentioned in para. 3.2), the quality and presence of the team remaining in 1994 would attest that the learning opportunity was seized by a good number among them. Present service directors were all trained by the project and are now spearheading the diversification efforts of the project. 4. This transfer of responsibility is stressed in the Borrower's comments (see Annex B). Flgure 3.1: imeline Evolution of Guékédou Agiriculture Project - PAG Cr. 1635-U péainRcPuékdrucéojctua National Program 1[ Governrnt Services K Privatised ServIces Decentrellsed Governmient pe tvarie te usSPCIA TaExtensionvii gNVA Swamian devlopentTächerons aménageistes (National Program fr .... Consultng c mpanies Rural Intrastructure) ;:iiijiii Rurai roads and tracks ......(ehnical Bureau of Pb G ouP ) reh abilitationp la ntation PrdcrAsa Research andsaitc (Guinean Agricultral Researchy Credit Crédit Mutuel .... . .Crédit Rural Phase l Phase 1l Present (1994) Emnerging outcomes - 30 ASP - Credit 1636-GUI 3.18 The project became effective in July 1986. Because of the controversy over the selection of the consultant team, TA personnel did not arrive until late 1986 and early 1987. Their focus in the first few months was in preparing for a Donors' Conference in mid-1988, for which documentation was prepared. Following this meeting, there is a general consensus among Bank officers that the team 'drifted', and little of the work plan contained in the President's Report was completed. The Bank undertook a Mid-term Review in late 1988, which was critical of performance, but proposed little of substance to tackle the basic problems noted below. Following a visit by Bank Divisional management in April 1989, the TA team was terminated in September, 1989 and for all intents and purposes the project came to an end, although all financial loose ends have still not been reconciled. 3.19 Mission impossible was not an unmitigated failure. The general view of the project has been that it was a failure. Box 3.1 summarizes the PCR description of the project results. This listing rather overstates the 'failure' of the project, for reasons which will be elaborated below. Apart from more specific questions, the lack of achievement of specified results is a reflection not only of the performance of those involved during the project, but also of the realism of the initial design and expectations, and of the impact of subsequent events. 3.20 Inappropriateness of Work Programs. It soon became apparent that it was not appropriate or feasible to attempt to prepare outline project proposals for other donors. Each donor more or less showed up and wanted to 'plant its flag' and often made only cursory contact with the ministry (and BSD). This made it very difficult for BSD to monitor what was going on, and also rendered highly problematic any effort to prepare projects for potential donors. A great deal of time was taken up responding to the various initiatives and trying to create some order out of the chaos. 3.21 Inappropriateness of prescribed role for TA team. In retrospect it was a mistake to place all of the TA team in administrative positions. Some of them had limited prior management experience in a bureaucracy and they knew little of the workings of the Guinean system. Thus, their ability to interact with the rest of the MRD, and with other ministries, was limited. On the other hand the Guinean members of BSD, who had been selected from the bureaucracy, and who already had a network of contacts around the ministry, were shut out of this management role. This placed a heavy burden on the Guinean Director of BSD, who was ill equipped to handle it. What might have been better was to have had, at most, two permanent staff to assist in managing the work program for BSD, and greater provision for short term inputs, which could have been tailored to the evolving needs of the unit. 31 Box 3.1: PCR Summary of Project Results Project results have not been satisfactory. One of the key tools for the planning, implementation and control of project activities, the elaboration of annual work plans was not used. The inadequate project performance can be seen through a review of individual components: * No sector strategic initiative or specific policy effecting; inter alia, prices, marketing, food and export production, input distribution, livestock and forestry development seem to have been produced as a result of the project. Further, BSD does not seem to have provided any impetus in the preparation of the three year agricultural investment program as envisaged at appraisal. - BSD does not seem to have identified or prepared any project. The few studies that BSD claims to have realized are, on closer analysis, a recycling of existing information or studies. Donors continued to identify and prepare their own projects, although they occasionally associated with some BSD staff. BSD did establish summary data sheets on about 90 projects and helped organize a national seminar on monitoring and evaluation. It is not clear, however, that BSD helped establish Monitoring and Evaluation units within the projects, assist those units to prepare annual work programs and supervise the execution of these programs. * No visible results were achieved in statistics and data collection, the funding for which was transferred to a different project called Permanent Agricultural Statistics Service (PASS) which was jointly financed by FAO. , * Agricultural research and small scale irrigation TA inputs were financed by French bilateral aid funds and were under the responsibility of the relevant departments with MDR. They have achieved positive results: their Government's policy on research and small scale irrigation is taking shape and a number of project ideas are being generated. One small irrigation project has been identified and is being processed. * No studies were undertaken that could be attributed to the project. The agricultural sector strategy document (LPDA) was essentially conceived outside of BSD, although it contributed to its elaboration, like other services of the Ministry of Rural Development. An office building was erected and staff houses built as planned. The office structure has become the home of the Ministry of Agriculture. 3.22 Cumbersome TA Team had difficulty responding to needs. The largest single input was long-term technical assistance, with six persons being provided (in addition to the Finance director): sectoral agricultural planning and team leader; project preparation; monitoring and evaluation; statistics; agricultural research (supplied by FAG); rural engineer (supplied by FAG). With the exception of the technical specialists supplied by FAC (who had previously been working in Guinea), the experts supplied were, in general, not fully equal to the demands of their positions. The late arrival of the team leader, and his subsequent death and 32 replacement, all hindered the development of effective teamwork, which is essential in a program of the type envisaged. 3.23 Training. The borrower's comments on the PCR underline the perceived importance of the training courses provided to the BSD national team. In all some 16 professionals were trained overseas (USA, France, Germany) and ten agents were given data processing training in Conakry. The Borrower states that on-the-job training was emphasized by the project, but the Bank's report indicates certain deficiencies at this level eg. lack of detailed training duties for TA specialists and supervision measures to ensure that training did in fact take place. Indeed, it would have been useful to identify a separate component for training, given the different requirements (as discussed elsewhere). This would also have helped focus the attention of supervision missions on this issue, which tended to be ignored. As a result no concerted training program was created or followed. 3.24 Built-in imbalance of support to the Ministry. The PCR notes that there was constant pressure on BSD to obtain use of resources for the ministry as a whole, and that leakage of funds occurred. This is true, but the project should have made provision for some broader support to the Ministry. It should have been clear that local funding for the ministry would be minimal. A position where a small unit is well endowed while the structure as a whole (and especially it's head) has minimal resources, is not acceptable. Some provision for the Minister's office (eg. for equipment and its operation) might have provided project management with stronger grounds for resisting pressure for misuse. 3.25 Project design failed to allow for the circumstances in the new ministry: (a) the various resources had to be brought together in a new unit/team, (b) the staff had to reorient themselves to the new policy and role of the ministry following the change of government and the shift to liberal economic policies. This whole process involved considerable change in habits and ways of thinking and BSD was clearly called upon to play a leadership role within the ministry. Outside the basic policy and strategy function of the BSD, the other areas of emphasis (rural infrastructure, research) owed more to the provenance of the project than to the overall needs of the ministry. While some focus was desirable, it was necessary to provide some support mechanism which would have assisted the ministry's structure as a whole to begin to pull together. 3.26 This could have included: (a) reduction in fixed TA inputs and increase in provision for short term inputs to provide assistance to the individual departments in developing work programs, policies and project proposals- (b) joint training or sessions (eg agricultural symposia, with rotating sectoral emphases) which could have brought all on board, and broadened the perspectives on agriculture; 33 (c) study tours to neighboring countries to learn how problems now faced by Guinea were tackled. (Guinea had been effectively cut off from these contacts for two decades) 3.27 Institutional Rationalization. The initial structure of BSD, developed under the project, was overly large and uneven, with six divisions and a staff of over forty. It should have been clear that this structure, and the costs it imposed, were out of all proportion to the country's resources. Since the termination of IDA financing, a number of steps have been taken to rationalize this structure (see Figure 3.2). The core of the BSD function has been taken over by a smaller unit, the Bureau for Agricultural Policy Coordination (BCPA). This has a staff of about fifteen in three divisions; policy and planning; studies (including monitoring of preparation; and monitoring and evaluation. BCPA carries out few studies itself but is primarily focussed on liaison with the departments of MARA, which carry out project preparation and implementation themselves. The other functions of BSD during the project period have been hived off into separate units or taken over by others, as indicated in Figure 3.2.' The result has been to create a smaller, more focussed unit, operating in close collaboration with the Secretary-General. 3.28 The 'downsizing' of BSD has resulted in a dispersal of its staff. This was not necessarily undesirable. As the Guineans expressed it, BSD acted as a 'pepiniere' (nursery) for the ministry. The overall high quality of the staff selected by the rigorous selection process (para. 3.6) has meant that several of those who have moved, have done so positions of responsibility elsewhere, eg. to head the Guinean Institute of Agricultural Research, and the National Agricultural Statistics Agency. Others have moved elsewhere within the ministry or been promoted internally. The current Secretary-General of MARA and the Director of BCPA were both initially with BSD.' Thus, the original members of BSD have functioned rather as a group of 'Young Professionals', and have helped in the overall process of creating a ministry with a coherent purpose. As noted earlier, (para. 3.26) other steps could have been taken by the project which would have further strengthened this process. Compliance with covenants Gueckedou - Credit 1635-GUI 3.29 Overall adherence to covenants was poor. Accounting and progress reporting was inadequate (para. 3.5) and input prices to farmers were subsidized in contravention to the agreement. Agreements on counterpart funding were seldom met and procurement irregularities also were frequent. ASP - Credit 1636-GUI 3.30 Similar problems were faced in this project, with the addition that most of the specific agreements on outputs and the preparation of agreed plans were not respected. 5. The Borrower comment notes that modest additional external support has also been provided. 6. The Director of BCPA has just been promoted to the position of Secretary-General of the Ministry of Commerce. 34 Bank Performance Gueckedou - Credit 1635-GUI 3.31 Bank supervision reports indicate that Bank officers tended to focus on the narrow aspects of project management and technical execution." This was particularly the case in the PAG where Bank agronomists spent considerable time tinkering with the coffee component and the technological packages. However, the period of active Bank involvement was only four years (1986 through 1989) and research in that period yielded little of interest to farmers. Little effort was spent checking the veracity of the figures advanced for the achievement of objectives. As a result, few members of the donor community in Guinea believe the overall figures which were quoted in the PCR. 3.32 The national project team complained about the number of Bank officers assigned to supervision. In all, Bank staff visited the projects (there is no indication of how long was spent on the project site) for a total of 120 person days over seven supervision missions (an average of 17 person days or 8.5 days per mission of two officers). Given the difficulty of movement about Guinea at the time of project execution, these averages do not allow for much time in the field (i.e., farmer visits). It is, therefore, necessary to view project indicators and figures given in the PCR as optimistic. ASP - Credit 1636-GUI 3.33 Bank supervision of the Agricultural Services Project was inadequate. Lack of continuity in supervision did not help. Despite the emphasis on specific outputs in the Terms of Reference of the TA team, and the use of their absence as an indication of project failure, there is no record in supervision reports of these elements having been systematically reviewed, or of any recording of outputs against targets, or of steps taken to revise the overall work program or modify outputs. In fact, until the time at which the TA team was terminated, the project was consistently rated as having only moderate problems (2 on a scale of 1 to 4). Former staff of BSD, Guineans and consultants, who were interviewed complained that supervision missions did not focus adequately on training issues or on the overall lack of collaboration in BSD, and did not follow up on the issues even when raised. 3.34 Perhaps more basically, when it became apparent, after the slow and fitfull start, that the detailed work program was not appropriate, the Bank did not step in to provide guidance. The 1988 Donor Round Table Meeting initially took center stage, but it is clear that, beyond that, the other parties were at a loss. The Guinean Director was not up to the task, and the consulting company was also not clear at that stage as to what the priorities should be. Given the background of the project the onus was clearly on the Bank at that stage to step in and provide leadership, but it failed to do so. 7. The Borrower's comments suggest that IDA should have held a Mid-Term Review (although this was not formally called for). 3.2: ~1mne Evo~ution of Agr~lural s em Projet PSA Cr. i u de au de satwge et développ~met Bueau do ~~inallen etd. Govermet Svios (Oiki Monlicdn and ai t ) Insitun~ de recerh Agriclturl resarchagricole de Gulnée )....... Div......s.........i.....N de génie niurale Phull l Phame il Present Phase 36 3.35 Similarly, during the Bank review process, the potential overlap of the planned statistical work with the role of the Central Statistical Service was raised (para. 2.13), but ignored. However, by the time the project became effective, the Ministry of Plan had, with Bank support, already initiated an exercise to undertake a census of agriculture. In order to avoid duplication the statistics division of BSD did not start its own data collection exercise immediately. There followed a period of over a year during which there was communication with Plan to determine how BSD might best support the census. That was never adequately clarified and in mid-1987 BSD, with Bank agreement, decided not to formally support the census. 3.36 Maintenance of general project documentation was also inadequate. Until late 1990 the projects were supervised from the regional office in Abidjan. This undoubtedly gave rise to some problems in ensuring that relevant material was lodged in the central files. The audit discovered that headquarters files are incomplete. Some material was recovered from Abidjan, but this also had significant gaps. Other material is still missing, eg the Bank response to the agreement of Government with the verbal proposals of Bank staff members to terminate the TA team. 37 4. Issues and Lessons Conclusions 4.1 The results of these projects reinforce the findings from elsewhere, especially in SSA and underline the importance of beneficiary participation; borrower ownership; government responsibility; and project sustainability. While it would be a simple affair to criticize the projects on these counts, it is important to underline that there were extenuating circumstances which made corrective project design difficult. 4.2 Beneficiary participation. In the first instance the relationship between the government and the rural population was extremely poor at the time of the appraisal. If beneficiary populations are to participate in project definition and supervision, it is essential that some level of confidence exist between the farmers and civil servants. This confidence also depends on the self-confidence of each of the parties. This has built up over the intervening years for the farmers, but is considerably less for the civil servants, since the latter have largely sought to deliver 'goodies' to the populations rather than supply services for which the user pays (even extension was built on imported seed and fertilizer). When, at the end of the project, these goods were longer available, government agents were at a loss as to how to address the farmers. It will be necessary for the civil servants to modify their approach to one which responds to farmers requirements. This will require a more 'market oriented' attitude on their part, whether this be in the field of extension, in research, or in privatized services (SPCIA - seeds and fertilizer). 4.3 Borrower ownership was extremely difficult to ensure as the projects were brought on line. This was particularly the case with respect to the ASP, where the proto-ministry was at a loss to know what it required or what might reasonably be identified as necessary. At the same time there was clearly pressure on the Bank to get a program of assistance to the Ministry going. Supporting a unit within a ministry which itself barely exists requires a broad based and flexible approach. 4.4 Government responsibility. Both projects were subject to financial abuses, with year- end audits being late (up to two years late at times) and auditors making their reports "conditional". The use of project funds for purposes outside the project led to the scope of changes in staff on the TA teams (as staff withdrew or were removed to avoid problems) and national teams (as those responsible for abuses were replaced). These abuses indicate that, under the conditions existing at the time of project execution, the government did not exercise its responsibilities with respect to the use of borrowed funds. While certain covenants were established, these were not always met, with consequent frustration accumulating on both the Borrower's and the Bank's side. Given the loose structure of the emerging ministry, the old habits of a centrally managed economy, and the relatively new experience for the Ministry of managing external credits, it would have been advisable to place responsibility for the financial management of the project under the control of the Ministry of Finance, which had already had experience with the Bank under other credits and loans. Also, the use of tranches for funding would at least motivate project managers to prepare financial reports on time. 38 4.5 Sustainability. The sustainability of the project efforts will depend on the ability of local entities (primarily Governmental) to assume the tasks undertaken by the project. In both cases Government has sought to rationalize technical and financial support for the different components. Figures 3.1 and 3.2 illustrate the evolution of the two projects from the original source project (ORG (Credit 932)) to possible future diversification and privatization. 4.6 Figure 3.1 shows the three potential mechanisms for pick-up of the Gueckedou components, i.e. assimilation into national government programs (PNVA; PNIR; IRAG) or local projects (RC II), privatization of certain activities (SPCIA; Consulting engineers; producer associations; Credit Mutuel; land technicians (see Annex) and transfer to local government institutions (collectivites decentralisees). Each mechanism is progressively further in the future and/or less probable. The national programs seem able to assume responsibility for their respective project components, but this also requires a reduction in personnel attached to the zone and a consequent increase in efficiency of the remaining agents. 4.7 Privatization is a road with both considerable potential and considerable risk for the component/service sustainability. The least risky element of the Gueckedou project is likely to be the assumption of the credit role by the Credit Mutuel. However, it must be underlined that this organization has benefitted from almost one million dollars in IFAD funds. The land technicians are private individuals (contractual) but are still dependent on publicly land development programs. They may be employed by the consulting company issuing from the possible privatization of the activities of the BTGR, although this would seem to be well in the future. Producer associations are being attempted for the coffee growers, but evidence of organization for other activities is conspicuous by its absence (particularly since this was one of the original objectives of the Gueckedou Project). 4.8 Figure 3.2 shows the services which are assimilating the original responsibilities of the BSD. Given the institutional nature of the original project, it is hardly surprising that these services are all governmental. The strategic element of the BSD mandate has now been assumed by the Bureau for the Coordination of Agricultural Policy (BCPA). Having picked up some of the ex-staff of the BSD, the BCPA is presently assuming national responsibility for the coordination of project strategy, studies for project preparation, and oversight of investment projects. BSD responsibilities for agricultural statistics, research and engineering have been hived off to the relevant services. 4.9 The projects were subjected to a multitude of changing critical conditions. Given the enormous changes that have occurred in Guinea over the past twelve years it is essential to make some assessment of their impact on project implementation and the outcome of the project. In many respects project design and objectives were overtaken by political, institutional, social and economic changes. Some of these external factors may be listed as follows: (a) The radical change in government ideology following the death of Sekou Tour6 led to a quantum leap in donor interest in the country and in the pressure/will increase to support the new, liberalizing regime. In uncertain times, the Bank was called upon to take the lead in donor lending while other donors were content to bide their time. With time, the stability and commitment of the new government to the liberalization process were well demonstrated and donor confidence in Guinea increased. This led to a certain amount of 39 jockeying for position between donors and projects with commensurate competition between various development philosophies. The leadership shown initially by the Bank was no longer taken for granted. (b) The changes in the administration of government services placed many ministries in a state of flux. Structural modifications to the organization of the ministry responsible for rural development saw the erstwhile Ministry of Agriculture become the Ministry for Rural Development and later the Ministry for Agriculture and Animal Resources (MARA). These changes had an impact on the ASP project as the Ministry was better able to define the structure it needed and the strategies to adopt for the support of the rural sector. (c) Social and political instability in neighboring Liberia and Sierra Leone have driven refugees into the Gueckedou project area creating pressure on land (the population of Gueckedou has risen from an estimated 100,000 in 1989 to some 300,000 in 1994). This has increased the local demand for rice and reduced competition from farmers in Liberia and Sierra Leone. This demand side influence may have had more impact on rice production and marketing than the efforts of the project. On the other hand, the influx of relief supplies for the refugees may have depressed local market prices, since distributed food is often resold on local markets. (d) With the liberalization of the economy and the much slower process of decentralization and democratization of the country, farmer confidence and entrepreneurship have increased. The confidence in market forces and the lack of government surveillance of profits may well have increased the 'without project' revenues, thereby diminishing incremental benefits to the project. (e) After having been isolated from non-communist countries for more than a quarter century, Guineans were very anxious to travel and study in W. Europe and N. America. This led to some resistance among the Guineans to suggestions of in-country or regional training. As contact and experience grew, this emphasis changed, and local staff now express the view that more efforts of this type would have been useful. 4.10 This list of conditions, which is by no means exhaustive, should serve to illustrate the extremely volatile environment in which the projects were conceived and implemented. For example, in the ASP project, given the evolving needs of the Ministry of Rural Development, it is probable that a more structured approach to institutional support should have been followed. This could have allowed for annual reviews and agreement on subsequent work programs and been linked to replenishment of special accounts. 4.11 Conditions for development offarmer organizations. While the promotion of farmer's organizations was one of the four objectives of the Gueckedou project, there is now, no evidence of such organizations in the zone. Farmers' organizations are essential if the service elements of the project are to become sustainable. Such organizations will be better able to formulate and communicate the real needs of the local farming community and will allow the vestiges of the project team to become more market oriented in their response to the producers. It must be noted that such organizations are not described as 'cooperatives', as this term has become politically and emotionally charged. At the present time, cooperative and precooperative groupings may enjoy some minor tax benefits on the purchase of production 40 inputs. However, since these inputs are not freely available, many still being imported by government bodies (i.e. not at all), these advantages are insignificant. As traditional registration of the cooperatives requires the adoption of relatively complex rules and regulations, which involves government evaluation and supervision, the majority of nominal cooperatives tend to be empty shells. Less formal producers' associations may require less administrative intervention and allow for more communication between farmers, but such organizations must be established around profitable farm enterprises which can sustain the cost of representation, information gathering or consultative technical advice. This may be possible in the coffee sector if world prices continue to be favorable. In rice production, until the number of swampland farmers, with suitably developed land in a given zone, reaches the necessary critical mass (as yet unknown), the viability of rice farmer associations would appear to be weak. 4.12 Private enterprise needs protection from governmental constraints (Import barriers, customs abuses, police interference with transport etc.). The most risky privatization attempt from the Gueckedou project will be the launching of the Agricultural Input Production and Supply Company (SPCIA), the enterprise which will assume the role of seed producer and fertilizer distributor on a commercial basis. This enterprise will require assistance (training, finance) in management and marketing if it is to succeed in responding to the local market potential for seed and fertilizer. Also, if government does not free up the importation of fertilizers, etc., SPCIA is likely to have great difficulty in obtaining the inputs sought by farmers, and this venture will wither on the vine, vitiated by customs officers and informal import barriers. Overall Assessment of Projects 4.13 The assessment of this audit varies somewhat from that of the earlier Completion Reports of the two projects. This review sees the Agricultural Services Project in a somewhat more favorable light than the PCR (and is, in fact, closer to the views expressed by the Government in their comments). Conversely the audit is less sanguine about the outcome of the Gueckedou Project. Gueckedou - Credit 1635-GUI 4.14 The PCR assessed the project as satisfactory. It undoubtedly brought benefits to the region. As noted (para. 3.17), it developed a cadre of staff within a program with a primary focus of support to smallholder agriculture, including research and input supply. This has reportedly formed the basis for progressive expansion of extension efforts of this type across the country. Somewhat fortuitously, the project supported an expansion of agricultural production in a region of the country which has borne the brunt of the refugee influx from Liberia since the late 1980s. In turn this local influx increased the local demand for food crops, and rice in particular. The project also developed the 'tacheron amenagiste' approach to the management and support of micro-level swamp-development, which has been extensively replicated elsewhere in the country, and it has (since closure of the IDA credit) taken steps to privatize some of the activities undertaken by the project (seed production, input supply, and road maintenance) in ways which may also prove replicable. 41 4.15 On the other hand, because of the failure of the monitoring system, it is essentially impossible to know what the direct result of the project was. Certainly, the general consensus outside the project is that the achievements claimed are overstated. The LAC-23 rice variety, which was the basis of its effort in upland rice production, was developed in Liberia in the late 1970s and was generally available before the project started, so that it is overstating the case to attribute its use to the project, although its support and seed distribution efforts were a positive factor. In addition, the sustainability of the structures created out of the project is still uncertain, but they represent a logical step by the Guineans to adapt the structures of the project to the evolving needs of the country. However, given the lack of government service structure, trained personnel or resources at the time of appraisal, it is arguable that the integrated rural development model applied to the area of Gueckedou was an appropriate response, even though execution was poor. This is not to say that the conditions in the region have not evolved to make this no longer the case. ASP - Credit 1636-GUI 4.16 The PCR characterized this project as being an unqualified failure. Certainly, the TA financed by the credit was not cost-effective and there was substantial leakage of project resources, as has been discussed above. However, the primary objective of the operation, the establishment of a capacity in the agricultural ministry to undertake planning, strategy and policy formulation and project analysis has been achieved. At the present time it is generally noted that, in several respects, the Ministry of Agriculture is the best performing of the technical ministries. Particularly, respondents in the aid community indicated that, in their view, MARA has a clearer strategy and objectives than others, and that project or other proposals and reports receive a more coherent and useful response there than elsewhere. This situation is clearly a reflection of many things which have happened over the past decade, but it cannot be totally divorced from the Agricultural Services Project and the establishment of BSD, since the objectives of the project were precisely to create the type of capacity which now exists. 4.17 The PCR is highly critical of the Guinean Government and of the consultant team for the inadequate results of the project. It is the view of the audit that the Bank cannot escape its share of responsibility. Many of the difficulties faced by the Guineans and consultants resulted from design problems, a process in which Bank views clearly predominated because of the circumstances of the time. In particular, the ASP was handicapped by an overly large structure. This was inappropriate to the needs of the ministry; its work program was too rigid in design and application; and the provision of TA was excessive and also inflexible. This largely reflected inadequate preparation and consultation with Government over this project. Supervision did not take steps to tackle these problems. 4.18 Over the past five years Government has streamlined this structure, hiving off half the initial units (para. 3.27), and reducing the staff from 40 to 15, focussing on the core functions which formed the initial objective. This structure should have been adopted from the outset. The PCR concluded that the BSD was unsustainable, but the present structure clearly is sustainable and, over the decade, significant institutional development has also been achieved. 42 4.19 Nationals note that the major legacy of the project is the Ministry headquarters building, which houses its senior staff. This fact is a further illustration of the incongruity of this building being ostensibly constructed by the Project just for the ministerial planning unit, while the ministry itself had to make do with markedly inferior quarters. There appears to be no reason why this structure could not have been designated as a 'Ministry Headquarters' from the beginning. In fact, failure to do so resulted in a poisoning of the atmosphere which lingers to this day. 4.20 Both projects were handicapped by basic design problems and also suffered from two of the principal 'fads' of the time. With hindsight, Gueckedou was not in a position to benefit from large scale adoption of a full T&V system, and would have been better served by a lighter structure. Similarly, the approach of developing large planning teams and units and the heavy emphasis on statistics and data collection (even under conditions where obtaining good survey based data was likely to be a major task) was a principal element in the flawed design of the BSD. 4.21 Ratings. As indicated, both of the projects have had positive elements in their outcomes. However, while discrete measures of outcomes cannot be obtained, it is clear that their performance was not commensurate with the resources used. In both cases, there were significant leakages of resources and more modest projects would have been appropriate. Both projects are rated unsatisfactory and the institutional development they achieved as modest. Lessons 4.22 Institutional support in times of change must be delicate, flexible and well supervised Given the newness of the Ministry structure at the inception of the project, (recently combined FAPA and Ministry of Agriculture), and the radical change in government strategies, it was ill- advised to design the project with such a heavy contingent of long term technical assistance, to define the program in such detail or to tie financing to the defined program. This was particularly the case where, as in the case of the ASP, there was limited opportunity to develop a real dialogue with the borrower. The response should have allowed for a core long- term TA (one or two specialists) with the facility to provide consultative services on an ad hoc basis. Financing could have been on a tranche basis (perhaps annual) with the subsequent program being presented and reviewed with each tranche report. In the highly volatile conditions of the time design should have made provision for change, in the spirit of the "new project cycle" as now being proposed. 4.23 Islands of "plenty" in the midst of seas of 'paucity' should be avoided: A major difficulty for the ASP project was the 'gravitational pull' generated by the financial mass of the project budget in contrast with the financial resources of the host Ministry. The imbalances caused by the monolithic focus of the project on the BSD created tensions between directions which could have been foreseen. Some mechanism should have been conceived to include benefits to other directions eg. through Ministry wide training sessions. Geographical isolation reduced the impact of this syndrome in the case of Gueckedou although isolation also led to abuses and mismanagement. 4.24 Smaller IDA funded operations may be preferable to larger multi-donor operations in conditions of great uncertainty. The Gueckedou project was almost fatally handicapped by the 43 delay in AfDF funding. Under conditions such as those of Guinea in the early 80's the likelihood of one of a group of donors not being able to declare an operation effective is high, with inevitably disruptive effects in conditions which are already difficult for project management. A smaller IDA funded project, with a more limited extension effort, would have been preferable. 4.25 Regionally rather than sectorally focussed projects can be appropriate under certain circumstances. Within the context of political change, sectoral policy fluidity, weak administrative structures and untrained personnel, the integrated rural development model was appropriate. Such a model would be less appropriate under present day conditions. 4.26 The apparent success of the 'Tacheron Amenagiste' suggests scope for more flexible, less monolithic structures for implementation rural development programs. This approach, in effect subcontracting supervision of small works carried out by farmers, has potential as a means of providing public support but avoiding creation of project and/or departmental bureaucracies (see Annex). Directly hired project staff tend to outlive their usefulness and their transfer from one project to the next is seldom smooth. Effective use of contractual implementation would require some continuity of operation, otherwise the 'contractors' will drift off to other activities. This approach also requires good overall supervision and monitoring by the implementing agency. It is not clear that the latter has been achieved in Gueckedou. 4.27 Steps need to be taken to clarify responsibility within the Bank for maintaining project documentation, especially when the project is managed from afield office. It is not clear that, in this case, material was systematically sent, or copied, to headquarters for filing. In addition, during significant reorganizations, as in 1987, during which personnel and responsibilities were transferred among units, misunderstandings can occur and significant material be lost.  45 Annex A The Use of the 'Tacheron Amenagiste' The notion of the 'Tacheron Amenagiste' (which may be loosely translated as a 'land technician') was evolved during the implementation of the Gueckedou Agricultural Development Project. It essentially provides a means of subcontracting out supervision of land development field works. In this particular case the method was used for the improvement of small 'bas-fonds' for paddy rice production. These 'bas-fonds' are low lying areas which flood for significant periods of the rainy season and have a high water table at all times. The areas selected for improvement were generally small (1 to 3 hectares), at the headwaters of small streams, and would be operated by a small number of farmers (often only one). Works required would be simple and require limited levelling to create small rice paddies facilitating retention or drainage of water as required. The steps in the process of developing a 'bas-fonds' using the amenagiste appear to be the following: (a) When a farmer wishes assistance in developing a 'bas-fonds' he goes to the local office of the implementing agency. (b) A member of the agency's planning cell will then inspect the site and determine whether it is suitable for the development proposed. If suitable, the site is surveyed and a plan prepared. (c) The implementing agency contacts a "tacheron amenagiste" and arranges for him to go to the field. (d) He then discusses the necessary work with the farmer and agrees on what needs to be done and how, when, and by whom, it will be done. (e) The "tacheron amenagiste" finalizes the contract with the implementing agency and proceeds to lead the work in the field, laying out the field, and advising the farmer. (f) If, as is occasionally the case, some more permanent structure is required, and material costs are covered by the project, eg. for a small concrete water retention structure, the amenagiste estimates the required quantities of cement etc., and arranges for them to be supplied through the project. (g) When the work is reported to have been completed, the monitoring section in the implementing agency is supposed to visit the site and confirm that the work has been satisfactorily completed and to follow up later to ensure that there are no problems. The current payment to the amenagiste is 100,000 Guinean Francs per hectare (equivalent to about US$100). While this is not cheap, it almost certainly is a lower cost than using the traditional project bureaucracy. 46 Annex A Preference for 'amenagiste' positions has been given to laid-off government personnel with appropriate experience. They are given some initial training as necessary in the basic technology and are then registered. This approach, like most activities, is potentially open to abuse. Successful use will require effective (and cost-effective) supervision of the 'amenagiste' by the implementing agency. Supervision might be aided by an appropriate use of reporting by staff of the implementing agency and the amenagiste. For example, use might be made of small disposable cameras to record the site at the first inspection by the implementing agency, and the amenagiste be required to provide one or more photos showing progress from agreed positions, noted on the site plan. This could be repeated at intervals specified in the contract, eg. annually for the first three seasons, to ensure that the works are not defective, eg. not holding water, or vice versa. A portion of the fee might be withheld, pending satisfactory operation. In the long run, the true test of the effectiveness of this approach will be if individual farmers and communities are willing to contract directly for the services of individuals (such as the Tacheron Amenagistes) with the necessary skills to assist in small scale investments in land improvement. 47 Annex B Republic of Guinea Office of the President of the Republic PUBLIC PROJECTS ADMINISTRATION AND OVERSIGHT AGENCY Conakry, May 8, 1995 Our ref: No. 404/ACGP/CAG/1995 Mr. Roger Slade Chief, Agriculture and Human Development Division Operations Evaluation Department World Bank Washington, D.C. Dear Mr. Slade: Subject: Performance Audit Report on Gueckedou Agricultural Development Project (Credit 1635-GUI) and First Agricultural Services Project (Credit 1636-GUI) On behalf of the Public Projects Administration and Oversight Agency (ACGP), I should like to acknowledge receipt of the Performance Audit Report on these two projects and also to say that the Department places great value on the content of the document and the useful lessons to be drawn from it by both the World Bank and the Government of Guinea. The Report concludes that the projects did not achieve the objectives set for them, an opinion shared by the Government as a result of its own review of agricultural projects, which it launched in October 1990. The conclusions reached as a result of this review - which were passed on at the time to all donors, including IDA - provided the basis for restructuring not only the Gueckedou Agricultural Project itself (Credit 1635) but also the Bureau for Strategy and Development (Credit 1636), which was converted in 1994 into the Bureau for the Coordination of Agricultural Policy. ACGP, in its capacity as contracting authority for these two major public projects, has drawn the following lessons: 1. Projects need to be soundly formulated, in the sense that they should be focused on the benefits that will accrue to target population groups, and on the sustainability of the development Annex B 48 actions proposed - this sustainability being a matter of choosing effective and affordable strategies through which to achieve the objectives in view. 2. A permanent local mechanism is needed to ensure external monitoring, evaluation, supervision and inspection of public projects. ACGP already performs these functions, among others; the aim is early identification of any need for corrective measures in project implementation, and thus better overall project management. ACGP is ready not only to serve as both the partner and representative of donors but also to coordinate their contributions to sound program execution and project development. 3. Mechanisms need to be set up which will enable the various partners (Ministry of Agriculture, project management, contractors, ACGP and donors) to consult with one another, and which will also conduct periodic evaluations of project results and problems, recommending remedial measures and identifying the individual responsibilities of partners in each case. Annexed to this letter is a summary of comments and observations from ACGP staff. It would be appreciated if you would kindly ensure that they are taken into account in the final version of the Performance Audit Report. Trusting that what has been learned so far will guide our current and future actions, I remain Yours truly, /s/ lbrahima Kassory Fofana Administrator-General 49 Annex B Republic of Guinea Office of the President of the Republic Public Project Administration and Oversight Agency Agriculture Division Summary of Comments and Observations on the Performance Audit Report on the Gueckedou Agricultural Development Project and the Bureau for Strategy and Development PART I: GUECKEDOU AGRICULTURE PROJECT (PAG) Credit 1635-GUI 1. Project Preparation 1.1 The study of the agriculture sector in Guinea released by the Bank in April 1994 identified several priority concerns. One was adoption of a strategy to give small farmers access to improved inputs (fertilizer and seed), improved livestock breeding techniques, and simple irrigation and water management techniques. This type of strategy was not reflected in the IDA funding for PAG, which was allocated as follows: Equipment/vehicles 42.5% Agricultural inputs 4.6% Technical assistance 45.4% Unallocated 7.5% Total 100% 1.2 It was not the intention of the project designers to create a permanent, replicable development mechanism, especially not where credit and inputs were concerned, since it has been demonstrated that when these two requirements are handled directly through projects there are always major shortcomings. In the particular case of PAG, these inherent difficulties were accompanied by poor financial management and gradual exhaustion of working capital funds. Annex B 50 2. Project Execution 2.1 The way project financing was structured meant, at least, that in the interests of good management the first IDA disbursement should have been accompanied by entry into effect of the AfDF and IFAD loans. In the absence of such an arrangement, the success of the project depended on changes to the initial allocation of Credit proceeds so that resources would be available to finance not only the original components of the project but also operation of a vehicle fleet that included a high proportion of heavy equipment, and creation of satisfactory living conditions for the project experts. By departing from Credit Agreement clauses governing effectiveness dates and not making any formal provisions for introduction of the desired changes, IDA created conditions that actually made management of the project more difficult. While there were good reasons why IDA acted as it did, supporting arrangements should have been made by restructuring use of the proceeds of the Credit. 2.2 The review of agriculture projects undertaken by the Ministry of Agriculture in October 1990 (a copy of which was forwarded officially to the World Bank's resident representative in Guinea) was part of preparations for issue of an Agriculture Development Policy Statement. The major conclusions arrived at were as follows: PAG ACHIEVEMENTS AS OF JUNE 1990: Smallholder Extension Services Actual Estimated Rainfed rice 8,230 ha 19,200 ha Irrigated rice 1,420 ha 2,000 ha Planting of coffee trees 324 ha 1,000 ha Pruning of coffee trees 114 ha 480 ha Development Works Road maintenance 600 km 500 km Road construction 499 km 240 km Low-lying ricegrowing areas 1,357 ha 2,000 ha 51 Annex B 1986 1987 1988 1989 Areas equivalent to selected seed distributed through PAG: rainfed rice 1,100 ha 1,500 ha 1,800 ha 1,900 ha irrigated rice 15 ha - 250 ha 330 ha Areas equivalent to fertilizer distributed through PAG 2,200 ha 2,800 ha 700 ha 3,000 ha Machinery and a sizeable fleet of vehicles and mobile equipment was available for PAG purposes. In 1990, a survey revealed that there were 471 civil servants assigned to the project, including other seconded public sector personnel. Simultaneously, there were 757 contract employees on the project books. As of June 20, 1990, financial implementation figures under the various credit agreements were as follows: IDA: 99% AfDF: 26% IFAD: 9% In other words, 75% of the funding resources disbursed by that date had come from IDA. The findings obtained from the Ministry of Agriculture review of projects prompted the Government to restructure the PAG and promote a replicable system of development. The chief measures decided on and actually put into effect by the Government were: (a) Thinning out of the ranks of project personnel: 97% of contract staff were dismissed, and half of the group of civil service staff were redeployed. (b) Promotion of small and medium contracting or service enterprises, set up by former PAG staff who opted to move to the private sector. Two examples are: (i) GMTP (Guin&nne de maintenance et de travaux publics - Guinean Maintenance and Public Works Company); (ii) SPCIA (Socigtd pour la production et la commercialisation des intrants agncoles - Agricultural Input Production and Supply Company). Annex B 52 (c) Opening of nine local branches of the Cr9dit Mutuel in Gueckedou Prefecture. (d) Execution of the extension and research component by the appropriate Guinean government agencies. (e) A very light management structure, responsible for coordinating a range of contractors. These measures would have been introduced earlier if IDA supervision missions had been carried out correctly. Similarly, prompt restructuring of the PAG would have made it possible to avoid cancellation of the remainder of the IFAD loan and to advance further with development of the project area. The chief concerns of IDA supervision missions were (a) disbursement capacity and (b) technical performance results in the low-lying paddy complex. With regard to the latter, it may be noted that the main techniques developed over the course of the PAG have now been disseminated country-wide. In conclusion, then, problems were encountered throughout the design and implementation phases of the project. COMMENTS ON SPECIFIC POINTS MADE IN THE IDA PERFORMANCE AUDIT REPORT Paragraph 3.6: Lack of adequate local external monitoring and evaluation, and ineffective monitoring and evaluation in general. Where the Government's main strategic lines of action were concerned, these shortcomings meant that major problems were not identified and rectified promptly. Paragraph 3,7: Laxness on the part of successive IDA project leaders. Missions became progressively shorter as project implementation advanced, focused chiefly on technical and financial concerns, and consisted mainly of agricultural and financial specialists. The mid-term review was never carried out, although it was due at a point when major changes were taking place. Paragraph 3.8: Failure of the original project design to provide for the post-project phase. Over 25% of the cost of the project was to be met by the Government and beneficiaries. Quite apart from this considerable financial burden, the project was executed 100% on force 53 Annex B account with excess personnel, a large operating budget, and a lack of administrative and financial autonomy. The main details of the project should be completed by presenting project results as of June 1990, as noted in paragraph 2.2. PART H: FIRST AGRICULTURAL SERVICES PROJECT (Credit 1636-GUI) 1. Mention should first be made of the weaknesses of the original approach to the project, which called for a major team of international experts to be assembled and given direct responsibility for: - formulation of an agricultural policy; - preparation of a rolling three-year investment plan. These objectives rapidly took precedence over others that were no less important, namely: - personnel training; - collection of statistical data on the agriculture sector; - support for agricultural research and small-scale irrigation. During the implementation process, serious problems affected all components of the project except the last, which had the support by two members of the French cooperation team, who assisted senior local professionals. As far as policy formulation was concerned, lack of knowledge of local circumstances quickly led the team of experts to summarize the general lines of action recommended in the study for Guinea's master plan for rural development and reorganization of agricultural services. Although the Bureau for Strategy and Development (BSD) organized the technical aspects of the first conference of donors to the agriculture sector, held in October 1987, it did not succeed in putting into place the means of monitoring operations in the sector, although this would have enabled it to developed subsectoral policies based on field experience. However, organization of the second rural development conference, held in April 1989, did result in establishment of a small planning division, the first step toward creation within the Ministry of Agriculture of a capacity for reflecting on aspects of planning. Annex B 54 It is for this reason that the following substitution should be made in paragraph 16 of the Performance Audit Report: Replace "However, the primary objective of the operation, the establishment of a capacity in the agricultural ministry to undertake planning, strategy and policy formulation and project analysis has been achieved." with "However, the primary objective of the operation, to equip the Ministry of Agriculture better for planning and for strategy and policy formulation, was achieved after the departure of project technical assistance personnel, thanks to better coordination among development partners (Ministry of Agriculture, UNDP, FAO, the French Ministry of Cooperation and the EEC, principally) and to greater participation by Guinean professionals." 2. Given these institutional changes within the Ministry of Agriculture, now is a good time to enunciate a number of lessons for the future: 2.1 Flexible institutional support and monitoring of government departments and agencies with responsibilities in the agriculture sector should be part and parcel of effective programs of services for farmers, especially input supplies, farm credit, and extension and training. Institutional support for these government departments and agencies, in the form of agricultural services projects, needs to be based on very clear definition of the objectives in view at each stage. The end result should be that these entities are much more effective in fulfilling their ongoing functions in the development process. 2.2 One of the mandates given to the Public Projects Administration and Oversight Agency (ACGP) is to facilitate project execution in the field by taking all possible steps to eliminate obstacles. ACGP should aim at relieving donors in the external monitoring and evaluation of projects, by creating conditions under which timely decisions can be made in assessing the management of projects and making the kinds of financial improvements in them that will benefit the country. 2.3 Mechanisms - project steering committees, for instance - should be created that will enable project partners to consult together, jointly examine project successes and problems, and propose prompt corrective measures as necessary. IBRD 18935R 긱  -BRD 18936R 汰   IMAGING y[prt No: 1476 Type: PPARP