Guinea - Gueckedou Agricultural Development Project

Document of The World Bank FOR OMCLa USE ONLY Repo" No. 12545 PROJECT COMPLETIfON REPORT GUINEA GUECKEDOT AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 1635-GUI) NOVEMBER 24, 1993 MICROGRAPHICS Report No. 12545 Type: PCR Agriculture Operations Division Occidental and Central Africa Department Africa Regional Office This docment has a restricted distibution and may be used by reipients only in the performance of their offlcial duties. Its contents may not otherwise be disclosed without World Bank autorization.  C:RRENCY EOUIVALENT Currency Unit Guinean Franc (GF) US$ 1 =GF365 US$ 1 =GF922 ABBREVIATIONS AND ACRONYMS AfDB African Development Bank AfDF African Development Fund BSD Bureau Strategie DIveloppement (Strategy & Development Unit) EEC European Economic Community IDA International Development Association IFAD International Fund for Agricultural Development IRCC Institut de Recherches du cafe et du cacao (French Research Institute for coffee and cocoa) M & E Monitoring & Evaluation ORG Operation Riz Gueckedou (Gueckedou Rice Projet Management Unit) PAG Projet Agricole de Gudckedou (Gueckedou Agriculture Project) PCR Project Completion Report SAR Staff Appraisal Report SDR Special Drawing Right T & V Training et visit UA Unit of Account FISCAL YEAR January I - December 31  FOR oMCAL U1s ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Oirector-oenerul Operations Evaluation November 24, 1993 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRES)ENT SUBJECT: Project Completion Report on Guinea Gueckedou Agricultural Development Project (Credit 1635-GVI) Attached is the Project Completion Report sin Guinea - Gueckedou Agricultural Development Project (Credit 1635-GUI). Parts I and m of the PCR were prepared by the Africa Regional Office. There is to Part II as the Borrower did not provide comments. The project *bJectives were to increase production of food and cash crops by smallholders through strengthiened agricultural services, and introducing improved seeds and agri- cultural technology. Despite some implementation shortcomings, especially poor management of financial aspects, inadequate local funding, and the difficulty of attracting high cahber consultants for extended assignments, changes in project scope, a flexible and innovative approach to project problems (by the Bank and the Borrower) led to a satisfactory overall outcome. Main results were that an effective extension system was established, improved varieties and techniques were developed and introduced for rice cultivation, coffee planting technology was improved, low cost roads opened up new areas, and a significant number of shifting cultivators were helped to convert to sedentary agriculture. Institutional development impact is rated as partial since the establishment of a viable credit fund could not be achieved and remains a priority. The project investments are likely to be sustained, provided Government allocates the necessary resources. Given Government interest in the project, overall sustainability is rated as likely. The PCR provides a detailed account of the project experience, but is short on outcome data and does not quantify economic impact The PCR was sent to the cofinanciers (IFAD and AfDB) for comments but none have been received. An audit is planned. Attachment This document has a retcted diiton and may be used by ecpea In the p frmaunce of ti* of ficil dutie. Its contents may not othewise be disdosed wtout World Bank autoizaon.  FOR OFFICIAL USE ONLY PROJ1ECT REPO RT REPUaU OF GUINA GtlBCEOU AGRITUIXM Pfl~ PR (Credfit 1635-GUI) TABLE OF CONME= * ~~~~~~~~~~~~~PAGE NO. PREFACE ..................... EVALUATION SUMMARY ......................iii R&lI PROJECr REVIEW FROM BANK'S PERSPECTIVE 1. ProjectIdenity ............................ 1 2. Background ........................... 1 3. Project Objectives and Description ........................... 2 4. Project Design and Organization . ............................. 2 5. Project Implementation. 3 6. Project Results. 6 7. Project Sustinability ............................... 7 8. Bank Performance. 8 9. Borrower Performance ..................................... 9 10. Consulting Seavices. 9 11. Project Documentation and Data. 9 PART II PROJECT REVIEW FROM BORROWER'S PERSPECIVE PART HI STATlSICAL INFORMATION 1. ReIated Bank Loans ..10 2. Project rTmetable ................................... 10 3. CreditDisbursements ..11 4. Use of Bank Resources ..11 A. Staff Inputs ................. 11 B. Mission ................. 11 5. StateofCovenants ................. 12 fr[WJ IBRD No. 18936R 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.  PROJECr COPLETI&ON REPORT GUIN-EA GUECEDOU AGRILTURAL DEVELOPMENT PROJECT (CuREDT 16354}GUID PEF-ACE Ihis is the Project Completion Report (PCR) for the above-named Project which was approved by the Board in November, 1985 for a Credit in the amount of SDR 6.7 million. The Project was cofinanced by the ntaional Fund for Agricultral Development (FAD) (Loan 170 GUI for SDR 5.05 million) and the African Development Fund (AfDF) for a sum of Units of Account (UA) 6.65 million. The closing date for this Projact was June 30, 1992, to which there was no extension. IDA funds were quickly used up. Final disbursement from IDA was made on May 28,1991. There remained a balance of SDR 71,485 which was cancelled. At closure, substantial sums remained undisbursed under the IFAD and AfDB loans. In the latter case, because of late effectiveness, funding continues. An important component was to have been financed solely by IFAD, but it required restrucuring which did not materialize by the date of closure. At the time of this report, no formal cancellation of the IFAD balance had been made. Preparation of this report is based on the SAR, the Credit Agreement, supervision reports, correspondence between IDA and the Government, as well as the cofinanciers. A field mission was carried out in November 1992, during which data were updated, and interiews carried out with local and other officials related to the Project.  - iii - MMJE COMELMON REEQRI GUECEDU AGRICULRAL DEVELOPMENT P1W)ECT (CREDIT 1635-GUI) EVALUATION SUMMARY Poect w bie 1. This Project was the partial continuation of a pilot phase project (Rice Development Project Cr. 952-GUI), and aimed at increasing the productivity of smallholder in food and cash crop production in an expanded Project area through the introduction of improved technical packages, extension services, rural infrastructure and a strengthened Gueckedou Prefectoral Directorate of Rural Development. Some 16,000 farm families were to be assisted in improving their upland and swamp rice production, as well as other cash crops; local staff were to receive training; and applied research on rice, cassava, peanuts and coffee was to be carried out. A seed production and processing unit was envisaged for the supply of improved seed to farmers, also a credit system to provide inputs to farmers, and rural roads constructed and improved to permit the movement of inputs and produce. Implementation Enerience 2. As the first project of its kind in Guinea under the new regime, (post Sekou Toure era), this project can be said to have accumulated and provided much of the experience which is now being applied in succeeding projects elsewhere in the country. Hence its implementation experiences are of immense benefit to Guinea as a whole. It has successfully established an effective extension system in which farmers' participation has been ensured. The research in upland and swamp rice has resulted in the application of improved techniques and introduced improved varieties with a resultant increase in productivity. An innovative introduction of the "swamp development complex" cropping system in which development takes place both within and around the swamp, allows many farmers to concentrate their farming activities around the swamp, thereby replacing the need for shifting cultivation. A further innovation has been the introduction of intercropping of newly planted coffee with a short stemmed and relatively short season variety of upland rice to ensure that coffee is planted at the right time (which coincides with rice planting), thus avoiding a clash in the agricultural calendar, while reducing the spread of farmers' fields. A low-cost approach to road construction and improvement has been effective in opening up enclaves in the area. 3. The financial management of the project left a great deal to be desired, and tainted the very good results being obtained in agriculture. The most disturbing aspect of this was the failu- - to put in place an effective credit system. Repayments of credit over the life of the project was of the order of 50% for a far reduced program of credit than was envisaged. However, no attempt was made to establish a proper revolving fund for the continuation of credit, and the collected funds disappeared in the project's treasury, and were never recovered by the credit unit. As a result of this sad situation, the use of IFAD's  - iv - funds for this component waS delayed pending agreement on new arrangements (a proposed Nationol Ctedit Project). No agreement could be arrived at by the time the project closed, and over SDR 3.0 million of IFAD's loan was left undisbursed. To a large extent, the failure in financial management can be linkced to the failure on the part of the consultant firm to provide a capable expatriae financial director, even after three attempts. In other areas the performance of the consultants (provided jointly by two firms) was mixed. However, technical leadership and guidance in agriculture was lacking, and the successes in this field cannot be attributed to the technical assistance team. 4. From IDA's perspective, the frustration of trying to get the project to improve its financial management combined with the near exhaustion of its funds early in the life of the project resulted in a considerable reduction in supervision activities, and there was a cut-off of valuable technical advice and assistance to the local technical staff. This action also signalled to Government and other observers that the project was not succeeding. Government carried out a study of the project, changed the project management, and made a number of recommendations for improving management and ataining the objectives. This action, although too late to help the project, was useful in assuring that project components would fall under the umbrella of the national departments which were being established mainly through other projects (credit, rural infrastructure, extension and research, seed production and coffee development). Results 5. The project's physical results were mainly achieved and surpassed in some cases. Over 17,000 farmers were in direct contact with the project, and there is ample evidence that as the only project of its kind in the area, its influence has spread beyond dhe project area, and the upland rice variety, LAC 23, introduced by the project, has spread throughout the forest zone. The project's technological introductions in rice production are important in the environment of shifting cultivation, and increasing pressure on the land. Thus, increased productivity in upland rice (more than double) leads to reduced land need, while the irrigated techniques and structures in swamp rice production (estimated at 600 farmers) with increased yields (over 3.5 tons/ha) and multiple cropping, reduces the need for shifting cultivation. The project estimated that some 270 farmers have abandoned upland shifting cultivation in favor of sedentarisation in the swamp complex system. While the project concentrated on new plantings of coffee in order to guarantee effectiveness in the program, and because of the poor quality of existng coffee trees, the system adopted to ensure good farmer participation (early planting with intercropping with rice) has had good results. However, there were many problems with the provision of seed and seedlings, and only 991 ha were planted versus a goal of 1,200. Also, the project encountered technical problems largely attributable to a lack of experience and the withdrawal of technical support through consultancy. The women's activities were late in starting but helped in the promotion of peanut production among women, so that over 1,700 ha have been cultivated. Inproved cassava seed material was introduced too late in the project to permit adequate multiplication and distribution. Only negligible quantities of this material have been distributed, and further diffision will depend on a farmer- to-farmer transfer. Unfortunately, the head of the Monitoring and Evaluation Unit died soon after receiving trainig on the operations of such a unit, and although the project employed a short-term consultant to help set up programs, it was already too late for meaningful comparative studies to be carried out. As a result, the project lacks socio-economic data to more fully assess the impact of the project on the community. However, the technical results attained by the project are being extensively used by newer projects in the country.  V 7. The extension services started under the project are to be under the supervision of the national department, similarly, the research seglrent will be merged into the regional structure of the national service, thereby guaranteeing some degree of continuity in this activity. This is important particularly because research has begun to focus on reducing the adverse impact on the enviroinment of the traditional system of shifting agriculture, and finding methods to reduce the need for chemical fertilizer which is so costly, and which, in the absence of an effective credit scheme, could make it difficult to maintain the high yields obtained under the project. The continued use of swainps, particularly for multiple cropping is an effective and economic way of exploiting this environn1nt, and farmers are therefore likely to continue this practice as long as markets for rice continue to be favorable. As a cash crop, coffee Is in great demand, and the Project's cautious approach of limiting the area of each new farmer ensures that farmers will have only that area that they can effectively manage. The expansion of coffee production is scheduled to be continued under the Coffee II Project financed by "Caisse Francaise" (formerly Caisse Centrale). The National Rural Infastructure Project has already set up a regional office in Gueckedou with experienced Gueckedou Project staff. Technical support in this field can therefore be continued. Findings and Lessons Learnt 7. The principal lessons gained from the project may be summarized as follows: (a) Greater attention must be paid to the selection and performance of technical assistance, and the solution to a lack of local expertise in some cases may be to employ short-term consultants rather than long-term technical assistance staff; (b) The hierarchy of project management should include an overseeing body that can review project planning and progress from the viewpoint of regional, administrative and national interests; (c) Careful planning and adequate staffing are necessary before the introduction of a credit system which must also be separate and indepene -t, while being in close liaison with the field extension staff. The establishment of the credit system should also take into account the reality of the context in which the project would be implemented. In this case, the appraisal mission failed to examine and specify how many farmers were to be reached with credit; how credit would be delivered and recovered and, how farmer groups would be organized and strengthened so that farmers would become more responsible for their own affairs; (d) Government must seek ways to reduce the need for imported inputs such as fertilizers for such staples as upland rice, and should therefore actively encourage research into cropping systems which can bring this about; (e) There is still a lot to be learnt about farmers' role in, as well as attitude towards the production and sale of upland rice seed; farmers cannot yet put a high value on this type of seed, and may be reluctant to buy it at the costed value; and (t) Now that there is a more cohesive approach and appreciation of the value of monitoring and evaluation, it is essential for this activity to be well planned, and staff trained and provided early in the implementation of the project, in order to ensure that it becomes a valuable management tool for project implementation.  GUINEA fiUEKED?OU Ag;RIa,JTURAL DEVELQMENT PRQIE (Credit 1635-GUI) PART I: PROJECT REVIE EROM BANK'S PERSPECJ 1. Pita i Project Name : Gueckedou Agricultural Development Project Credit No. : 1635-GUI Country Republic of Guinea Sector Agriculture Subsector : Area Development 2. RadWQoMW 2.01 Between preparation and appraisal of this project, there was a ihange of government in Guinea, and, with this, an entirely new philosophy for economic development of the country was being worked out. In the meantime, the principal objectives for the rural sector declared by the new regime, were to promote cereal production for food self sufficiency, to increase export production and to raise the standard of living of the rural population. These objectives had been recommeaded in an Agricultura Sector Review Report prepared by the Bank in 1983. In previous years, Government had neglected many basic issues essential to rural sector development, including: inappropriate pricing and marketing systems for agricultural products; the absence of agriculural services such as extension, research, input supply and credit; an overstaffed ministry with poorly trained and poorly motivated persomnel, with ill-adped structures and a serious lack of resources to permit a properly fnmctioning ministry. Although Government was still in the process of evolving a strategy for the sector, it had already proposed to: izr.prove smaliholder agriculture by providing services and incentives to farmers accompanied by freedom to mv. -tet their produce; reorgnize and streamline agricultral services with emphasis on smallholder; develop the planning and investment capability of the central ministry and establish a proper functioning central ministry. Government, with the help of the Bank and other donors, was in the process of determining programs for rural development. 2.02 Government policy under the Sekou Toure regime, which existed at the time of project preparation, was to promote collective flrming, and there was fierce resistance from the regame to any depature from this. Since 1980, the Bank had sought to influence Government to change this policy, and to concentrate on individual smallholder, in line with which a pilot phase project, the Rice Development Project (Cr. 952-GUI), was funded by IDA. The straegy worked well, and even the Sekou Toure Government had commenced accepting it as a viable approach. As a result, preparation of the present project was agreed to, and a rural incentives study was carried out by the Bank which laid the groundwork for the comprehensive agricultural sector review which was carried out in 1983. The succeeding Government regime was committed to focusing on the smallholder as a motor for bringing about the country's development, and hence, the Gueckedou Project was timely in demonsg this policy.  -2- 3. Erojec Objectives and Descrnigion 3.01 Qbijfrv,s. Based on the implementation experiences of the pUot r3ueckedou Rice Development Project (ORG) component of the IDA- and AfDB-financed Rice Development Project, (Cr. 952-GUI), the PAG Project aimed at: (i) increasing ihe productivity of smallbolder in rice production by expanding the scope and area of the first phase project, and introducing new technical packages for coffee, groundnut and cassava production in the new project area; (ii) reactivating the Gueckedou Prefectoral Directorate of Agriculture by integrating the two services into a single one; (iii) improving available techical packages in- the forest region through applied research and -(iv) helping to develop farmer otganizations to facilitate the provision of activities and services to the smallholder. 3.02 Comnnt. Project implementation was to take place over five yeas and was to consist of: (i) Project management and administration, including a liaison office in Conakry; (ii) extension services for about 16,000 farm families involved in the cultivation of irrigated and upland rice as well as other rainfed annual and perennial crops, with the introduction of activities for women farmers, staff training and agricultural research; (iii) seed processing and distribution to farmers from a seed multiplication farm and through contact seed farmers; (iv) land development of some 2,000 ha of valley bottom swamps through low cost irrigation works and upgrading of 240 km of agricultural roads, and maintenance of these as well as 500 km of existing roads, and the construction of a garage to repair and maintain Project equipment; (v) agricultural input supply and credit for iroject farmers; and (vi) provision of technical assistance and consultant services as well as training facilities and a fellowship program for the Project staff. 4. Proiect Design and Organiztion 4.01 Preparation of the project took place between 1983 and 1984, with a final report being issued in February, 1984. However, soon after, there was a change of Government, and the Preparation Report was revised to reflect some of the declared policies of the new Government. The final report was issued in September, 1984. The project was originally conceived as a rice development project which would cover the four ecological zones for rice production as had been implemented under the ORG project. (The ecological zones were swamp, upland, riverain and mangrove). There was no coffee component included at preparation; this was added at appraisal at the request of the new regime. Also, there was an agricultural services component which was principally concerned with the provision of a planning and statistical capability to the Ministry. This became a free standing project financed by IDA (Cr. 1636-GUI). In the judgement of the appraisal mission, no proven technology had been established for the riverain ecology, and farmer interest in this technology had not been clearly demonstrated enough during the pilot phase to warrant its inclusion in a project at this time. In the case of the mangrove in  -3- the South Eastern part of Guinea, the European Community (EEC) was funding project activities and planned to concentrate on this ecology in partnership with the Bank. Procedural difficulties delayed their involvement, and, at appraisal, the extent to which the EEC would cofimce this component cold not be determiined. The Bank and Government therefore agreed that the project would confine its adivities to Gueckedou, thereby making it an area development project. 4.02 It was somewhat difficult to put together a project which would fully reflect an agreed strategy with Government, when Government itself was In the process of detrmining policies and strategies for all sectors. Also, although by December 1984 the three ministries whia were concerned with agriculture and rural development had been merged into a single Ministry of Rural Development, there was not yet in place a mechanism for correcting the cumbersome and ineffective structure which existed at the field level. A study was financed by IDA in 1985 which was to propose an organizational structure for the ministry to cover field level operations as well as to create a fully functional headquarters with competent services. The study was also to propose wris of privatizing as many of the agricultural services as possible, and recommend methods for integrating externally funded projects into the administrative and organizational structure of the Ministry. This PAG Project anticipated these changes and --rovided for a merging of the existing regional structure with that of the project, covering the entire admm istrative region of Gueckedou (prefecture), with a small part of the southern Kissidougou region whic.. is adjacent to and has similar ecological conditions as Gueckedou. While the SAR indicated that the project's organization could be modified if Government accepted proposals from the study mentioned above which were different from that of the project, this was not reflected in the Credit Agreement. The project design remained that of an area development project which gave financial and administrative autonomy to the project and direct access of the Project Director to the Minister. This unsettled organizational state existed in the ministry at the beginning of the implementation phase of this project. 4.03 Given the then low level of effectiveness of the agricultural services as a whole, the project was designed to utilize the existing core staff left over from the ORG Project, supplemented by selected local personnel to take account of the expanded project area. A consu!tancy firm was selected to provide a team of technical assistants to direct the technical aspects of the project, under a local Project Director. While this arrangement was intended to inject high implementation standards and on-the-job training, in hindsight, it is questionable whether this aspect of the design was indeed completely appropriate given the difficulty of obtaining high calibre staff to work in Guinea at that time, and the high costs of installing and maintaining expatriate staff at Gueckedou. It should also be noted that after selection, personnel changes were proposed to the list of technical assistance staff submitted. In view of this, perhaps it would have been better to arrange for a consulting firm to plan and supervise some of the project activities with a team of experts making twice yearly visits. Although the SAR anticipated structural changes during the life of the project, greater support could have been given in its management had there been provision for the appointment of an overseeing committee by the Minister, which could have reviewed periodically the progress of the project. 5. Project Implementation 'c01 The IDA Credit became effective in October, 1986. While, for the greater part, the project was successfully implemented, particularly from a technical standpoint, the use of IDA funds was adversely affected by the late effectiveness of the AfDB loan, and although there was a cross effectiveness condition in the Credit Agreement, it became necessary for both IDA and IFAD to waive this condition in the interest of project implementation. As it turned out, the AfDB loan was not made effective until December, 1988, by which time, IDA bad already utilized more than 70% of its fimds. This situation  -4 - was partly caused by the delay in effectiveness of the AfDB loan, from which 50% of the purchases relating to vehicles, equipment and spare parts were to have been met, and patly by the need to meet payments amounting to nearly one miltion dollars for equipment and materials which had been ordered under the ORG project. Orders had been plac3d for this aquipment and materials just prior to the closure of the ORG, on the basis of an overestimation of the fimds available under that project. All of the equipment and materials received were used by PAG. Of the five mllion dollars funding from IFAD, three and a half million were to finance the credit scheme for the provision of aricultural Inputs to project farmers. The funds for the first year's credit were provided by IDA ($300,000), and these were administered by the project's Credit Unit. However, there were plans for a national credit scheme (Projet de Credit Mutuel) which it was subsequently intended would take over responsibility for credit activities of the project. While IFAD funded the establishment of a branch office in Gueckedou of the national scheme, by *he end of 1990, an acceptable credit system which would overcome the reimbursement difficulties experienced by the Credit Unit had not yet been worked out. At this late stage of the project, and in view of the managerial problems which led to important staff changes, IFAD took the decision not to disburse against this category. At closure, there was an undisbursed sum o, 'DR 71,485 under tue iDA Credit, while SDR 3,769,328 remained undisbursed under the IFAD Loan. The sum of 3,134,152 Units of Account remained undisbursed on the AfDB loan at November 30, 1992. 5.02 Further evidence of mismanagement of the financial aspects of the project can be seen in the use of the revenue from the credit scheme. The project was to strengthen farmer associations with a view to facilitating credit administration and more importantly to make farmers more responsible for their own affairs. Credit terms for short term credit were for a one year period with a 10% interest rate. Medium term credi were for a three year period with a 25% down payment with equal installment payments at the end of each cropping season and a rate of interest of 10%. Despite these favorable on- lending term dispositions, the credit scheme failed to achieve its objectives due to ineffective credit recovery efforts (which, in turn, are explained by unavailability of transport facilities for Credit Unit agents) and failure to establish an efficient accounting system and credt repayment schedules. Supervision missions did notice these problems and made recommendations to improve the situation of the credit scheme, but no specific corrective measures were introduced by project management. A reported total recovery of more than 102 million FG has been used by the project to meet some of its operational CoSts. This was done because of a shortage of local funds caused by the non- or late receipt of Government's budgetary funding. However, the level of Government's funding, although late, exceeded 80% of the budgetary requirements in the first three years of the project. Reimbursements from farmers were to have constituted a revolving fund to permit the continuous supply of agricultural inputs. At closure, the credit scheme had a zero balance, and the revolving fund was never established. 5.03 All project accounts were placed under the supervision of the Financial Dirctor. However, while it was agreed that the shortage of qualified local personnel necessitated the recruitment of an expatriate, none of the three expatriate Financial Directors was able to satisfactorily regularize the management of the project's finances. The closing of the accounts was always late, and inaccurate, for which reason audits were both late and conditional. The last audit was carried out in 1990; the report was presented in November 1992. Although accounting systems had been prescribed for the project, none of the Financial Directors seemed able to see to their proper implementation. It is now felt that it was perhaps more important to have an experienced expatriate Chief Accountant to head the Finance Department rather than a Financial Director with limited accounting experience.  5.04 Project implementation In the field was successful. Extension staff applied a modified version of the Training and Visit system in reaching some 17,000 farmers eventhough throughout the initial years of the project, staff were handicapped by transport problems. The procurement of motorcycles for the project had followed the ICB procedure requested by the Bank, procedure in which three firms were involved. However, after receiving the advance payment, the winring supplier failed to deliver the first lot of motorcycles. An advance payment was then paid to the supplier. Delays were experienced in the delivery of the motorcycles which finally were never supplied to the project inspite of threats of legal actions from the government. The unavailability of transport also affected the task of the Credit Unit agents in the loan recovery process. However, an effective training program was established for extension staff who used the traditional working group system for the diffusion of technical themes. This extension approach continued to provide experience in a system which gave the farmers freedom to produce and market his crops as an individual, based only on the advice of the extension service (in contrast with the collectivization system from which the new regime was emerging). The research program was well implemented in respect of rice, although there were some initial difficulties with regard to swamp rice based on-a poor choice of field site. This was easily resolved early in the project by abandoning the site and concentrating on the one which could be easily drained. There were some delays in respect of cassava and peanuts research, which in turn led to a late start in the inclusion of improved methods of growing these crops by farmers. Implementation of the seeds component was based on the principle of reducing the dependence on the project's seed farm and increasing production of seed by the farmers. Although, initially, there was some resistance to the need to reduce upland rice seed production by the project, on the grounds that there was a lack of experience in the large-scale production of this form of seed, and that the inefficiencies of this lack would be reflected in high costs, it became clear that managing a 100 hectare upland rice seed farm was a difficult proposition. The project eventually settled on 50 hectares, although even at this level, it is yet questionable that a proven system of large-scale production of upland rice seed has been confimed. 5.05 In the case of the coffee component, the SAR foresaw the possibility of rehabilitating some 460 hectares of old coffee, and 1,000 hectares of new plantings. The introduction of the coffee componeur represented an entirely new element in the project, and it became clear that though coffee production was a well established crop during colonial times, after independence it progressively declined in importance because of the government pricing polici& which changed after the fall of the Sekou Toure regime. There was no further promotion of its production, and, as a result, there was a lack of available experienced and skilled personnel to implement this component. An IRCC coffee expert was recruited as a consultant to help to plan and supervise the implementation of the component. He concluded after inspection that the old coffee plantings were under excessively heavy shade, sparse in density, of dubious quality and for the most part over 30 years old (a peculiarity of the Gueckedou area). He estimated that the amount of training of personnel and farmers combined with the high degree of labor necessary to prune back these trees could not be justified by their relatively low yield. It was therefore decided that the project would concentrate on new plantings only. The newly trained and inexperienced staff was therefore able to concentrate on new plantings. Farmers who on their own wished to carry out rehabilitation could receive advice only from the project. With this decision, the project's new planting program was increased to 1,200 hectares. Stafftraining and seed procurement for setting up and operating nurseries were obtained through IRCC in Cte d'Ivoire. These arrangements could only be concluded with the approval of the Minister of Agriculture in COte d'Ivoire, with considerable procedural difficulties which demonstrated the political nature of assistance in coffee development. It became increasingly difficult to obtain seed from C8te d'Ivoire, and though not necessarily related, the quality of the se-d when received was generally poor, necessitating on one occasion complete replacement. Seed was subsequendy obtained from Togo, but remained a major obstacle to the attainment of project goals. The  -6- eventual recommendation for overcoming this problem was the establishment of a budwood garden (*parc a bois") from which vegetative planting material could be obtained, thus the project would eventually be self sufficient. The establishment of project nurseries eventually led to the start of village nurseries as a means of reducing the cost of seedlings to the fariners and of getting them involved in the production of seedlings for their own use. There were many difficulties in the implementation of this compoi .it, largely at-lbutable to the inexperience of the staff. In the latter years of the Project when funds were exhausted and the IRCC consultancy could no longer be supported, some technical decisions were made by the project which have received substantial criticisms by the consultant on a visit in connection with a Caisse Centrale coffee project. There was a need to continue to provide technical support in the implementation of this component. A successful aspect of implementation of the coffee component has been its integration into the farning system of project farmers. An innovative approach to establishing new coffee fields involves the planting of coffee seedlings intercropped with upland rice, of a short- stalked early variety. This ensures that she coffee receives the full benefit of an entire rainy season, and is taken care of at the same time as the important upland rice crop. In a large number of cases, this system is combined with swamp rice cultivation, with the coffee occupying the adjacent slope or slopes to constitute the swamp complex ("complex de bas fond") development. In other similar projects, it has been found that farmers, by devoting the greater part of the rainy season to their food crop, delay planting coffee until mid- to late rainy season, thereby retarding the establishment of the seedlings, with the risk of losing some of these young plants during the proceeding dry season. 5.06 The swamp rice component was well implemented. The participation of the farmers in opening up and carrying out the engineering works on the bottomlands in a phased manner guaranteed their continued interest. The rural engineering division of the project was well organized, the staff well trained and they developed good relations with the farmers. The project introduced the system of simultaneous cropping of the adjacent slopes, as a means of stabilizing farming activities around the valley bottom ("complex de bas fond"), and particularly facilitating the crop activities of the women. In some cases, coffee development on these slopes has been included. The same division executed the road program,gaining valuable experience in the use of heavy machinery. There was considerable delay in the implementation of women's activities. Although there existed a core of women in service in the project area, it took several months before they could be organized into a unit. Tshey were subsequently handicapped by a lack of transport. As with other projects, imp'ementation of monitoring atd evaluation took some time before the concepts could be fully understood and applied. There was no assistance possible from the BSD as was expected, and both training and consultancy assistance were necessary before the project could finally have a program. Unfortunately, the Head of the division died in a motor vehicle accident soon after he had completed a training workshop. The failure to select a competent technical assistant as head of monitoring and evaluation in the BSD had a negative impact on the implementation of this component. 6. Proiect Results 6.01 As the forerunner of agricultural projects which sought to exploit the individuality of the smallholder, positive results were obtained in the research and extension aspects of the project. As a result of this, there has been a positive impact on most succeeding agricultural projects in Guinea, in that they have used the results of this project in their planning and execution. Also, the objective of increasing rice production in the region has been achieved, both in upland and in swamp rice production. However, because of reduced fallow periods (5 to 6 years versus 10 to 15 years) there is a need for fertilizer use to maintain the high production levels attained. The research section of the project continued the program commenced under the ORG phase and was able to release a short-stalk variety of  7 - upland rice which also matured in less time than LAC 23. This variety has been successfully intercropped with coffee. In addition, recent on-farm trials indicate that some five or so new varieties have surpassed LAC 23 in production. At present, however, LAC 23 still remains the staple variety, upon which increased production has been based. In the case of improved swamp rice production, at the end of the pilot phase, IR-5 was the principal recommended improved variety, which was considerably superior to the existing local ones. At closure, three more improved varieties were in general used by farmers, among which was the variety Suakoko 8 which is tolerant to the iron toxic conditions commonly found in the area, and which had previously led to farmers abandoning this type of swamp because of low productivity. In summation, available information show that average productivity in both upland and swamp rice productivity has been at least doubled. The initiative of introducing the swamp complex system has become established, and the project estimated that as a result, more than 250 farmers have permanently abandoned shifting in favor of sedentarised cultivation around the swamps. The National Rural Infrastructure Project, in addition to using the implementation experiences and staff of the project, has adapted this system.throughout the country. Although, at Gueckedou, all farmers do not have access to swamps, should those who do totally adopt this practice, there would be a considerable reduction in environmental degradation attributable to shifting cultivation. Research was commenced by the project to find crop systems using leguminous species which are capable of adding nitrogen to the soil in the upland enviromnent to reduce the need for fallow, and therefore for shifting. There is a need for this work to continue. 6.02 The late start in determining technological packages and in identifying seed sources for cassava and peanuts resulted in a shortfall in project objectives. However, the Women's Activities Division concentrated on promoting peanut production among women and achieved in a short time a reemarkable response, with some 1,740 hectares being cultivated (versus 2,000 hectares targeted in the SAR). The difficulties encountered in promoting coffee planting has led to a shortfall in the area established. While combining rice and coffee have in the past resulted in poor and slow establishment of the coffee, by intercropping coffee with upland rice, a higher percentage of farmers were able to realize a commercial yield (250-300 kg./ha) in the third year after planting, and these sam farmers have been able to peak between 800 and 950 kg./ha at the sixth year of production. 6.03 The project has initiated many technical improvements, which can be applied elsewhere in the country, but has equally begun to search for long term solutions to the major problem of the wastage and degradation of land under shifting cultivation. It is essential that this research be continued in the interest of best utflizing the land resources of the forest zone which is becoming increasingly limited with the influx of refugees from the Liberian crisis. 7. Project Sustainability 7.01 The establishment of national services for extension, rural engineering, seeds supplies, research and coffee production are intended to provide sustainability to activities which have been started under this project. These services are being promoted through projects with external funding, and regional arrangements are being made for the coverage of those activities in the project area. Already, as proposed in the Government's document "Proposition pour la Reformulation du Projet Agricole de Gueckedou* each department/project has put forward plans to take over responsibilities for activities in the project area in the post project period. The influx of Liberian refugees into the area has considerably increased the need for food, and Gueckedou has become an important marketing center. There is evidence that apart from the continued demand for rice, there is also an increasing market for cassava, peanuts, cowpeas and vegetables. Thus, there is enough stimulus for the continuous growth of these  crops in relation to the system of the swamp complex. The likely difficulty that farmers will face is in relation to the maintenance of fertility levels in the soil in view of the increasing pressure on the land (upland), and the needs of the improved varieties. The project faifed to establish an efficient credit system with which farmers could obtain much needed inputs, particularly fertilizers. ITis could lead to undesirable adjustments in the cropping system with reduced productivity. The research started in the project which seeks to resolve some of these problems through the manipulation of crops (rotation, fallow in legumes etc.) must continue. 7.02 Seed production at a national level is being reviewed under the National Seeds Project, but a rational system of seed production will be necessary to guarantee the farmer a renewal of seed periodically. The future of the seeds processing unit of the project will be determined by the review. The complex of project buildings built under the pilot phase (ORO) located a few kilometers outside of Gueckedou, and though well equipped, is not ideally suited for day-to-day contact with farmers who would normally visit Gueckedou. Staff and workers generally reside in Gueckedou and have to find transport to get to the project complex. Also there are senior staff quarters located on the complex, which were ideal for the expatriate staff, but remain largely unoccupied since their departure, and, because of the distance from the town are not being used by local staff. Construction of new staff housing is underway in the town, financed by AfDB. Government will have to decide on the disposition of these project structures. 8. Bank Performance 8.01 The responsiveness of the Bank to the changing circumstances of the sector in the light of the then uncertain policies of the new regime was a positive factor in the evolution of this project. The Bank made adjustments to the project concept during and after appraisal which resulted in an area project, rather than a purely rice development project as the pilot phase was, and allowed implementation to be concentrated in the general Gueckedou area. The project, in effect, became the pilot operation for agricultural development in the country's forest zone, and the lessons learnt from it are now being implemented elsewhere. The project was strongly influenced during implementation by the constructive dialogue and advice offered by the experienced Bank staff. Because of the enclave nature of the project and the lack of exposure of local staff to the then current research and technological developments, technical monitoring of the project was essential, and contributed positively to the implementation of the agricultural component of the project. Tle proximity of supervision staff (Abidjan) in the early years of implementation was also a beneficial factor. 8.02 The Bank withdrew its supervision involvement in the project in late 1990 when its financial management was in disarray, and pardy because most of the Bank funds had been prematurely used up. Unfortunately, this act sent a signal to Government that the Bank wished to disassociate itself from the project, leaving the impression that the project had failed. This status remained even after Government had replaced key management staff, and had proposed that the cofinanciers of the project should get together with Government representatives to determine the future of the project, given that substantial amounts of IFAD's and AfDB's funds were yet to be disbursed. In addition, since 1987, the J3ank's technical supervision was reduced as a financial analyst was put as the task manager of the project. With no funding leverage, and a policy commitment to national projects at the expense of area projects, the Bank's disinterest was evident. Nothing came of Government's proposals. The coffee component of the project suffered the most as a result of this. Serious technical decisions were made by the still inexperienced project staff without the influence of either the supervisor or the IRCC consultant with adverse effects for the program. In spite of this false impression of faflure, all succeeding  -9 - agricultural projects have drawn heavily on the positive lessons of the project, thereby confirming the high value of project results. 9. Borrower Performance 9.01 As has already been mentioned, the Borrower's performance could have been improved had there been in place a committee consisting of a mixture of local administration, farmers and ministerial personnel with responsibility to advise the Minister for overseeing project operations. It would have been necessary for such a committee to meet regularly in view of the vast distance between the project area and Conakry. In this manner, it may have been possible to curb the shortfalls of project management. There have been delays and inadequate funding on the part of Govermment, as well as some procurement problems which may be attributed to a lack of experience. These deficiencies affected the project adversely. However, Government retained a high level of interest in the project, and the dialogue between the Bank and Government has been open and good. 10. Consulting Services 10.01 The technical management of the project was entrusted in the hands of a consulting firm who supplied six expatriate experts. Their performance has already been discussed, and the observation was that they were effective in training, rural engineering and garage management. Staff changes had to be made for non-performance. However, it was evident that there were difficulties in obtaining high calibre staff to work in rural Guinea at that time. Short-term consultants were used for training in the T&V extension system, and for monitoring and evaluation. These were satisfactorily executed. 11. Project Documentation and Data 11.01 The Credit Agreement and the Appraisal Report were adequate bases for the execution of the project. However, project reports were not satisfactorily prepared during the life of the project, and were often late. There is no doubt that an understanding of monitoring and evaluation continued to elude project staff, so that it was not until late in the project that useful data were collected for use by project management. Also, because the IFAD and AfDB-fnanced components are not yet fully implemented, the project management has not yet prepared Part II of the PCR (Project Review from the Borrower's Peropective).  - 10- PROJEa=CMIS]I1 REPORT GUECKEIDU AGRICULTURAL DEVELOPMENT PROJET fCreait 1635-GUI ?ART m: STATISTICAL INFORMATION Tbtle 1: Related Bak Loans Credit Title Near of Amoroval Purcse Status first Agriculturat 1636-wi 1986 Improvement of Completed Services Project plam.ing capability Livestock Sector 175-WI 1987 Restructuring of Under Rehabilitation livestock services implementation Nationat Seeds 1864-GWI 1988 Seed production Under Implementation National Agrlc. 1955-6UI 1969 Restructure and Under Research & Extension improve research 8 Implementation extension Forestry and 2068-WI 1989 Improve forest Under Fisheries Management ngemet Implementation National Rural 2106-GUI 1990 Rural roads and Under Infrastructure asriculturat implementation structures Mational Agric. 2407-GUI 1992 Export of Agric. Approved export PromotIon produce Table 2: Project Timetable Ithm RhIDate-PA Date Actwt Identification 1983 Preparation Feb. 1984 Sept. 1984 Appraisal Oct. 1984 Negotiations July 1985 Board Approval Nov. 26 1985 Credit SIgnature Dec. 27 1985 Effectiveness April 1986 Aug. 31 1986 Credit Ctosing Jtme 30 1992 Juie 30 1992 At the completion of the preparation report by the consultants (Feb.1984), there was a change of government, and the report was revised to reflect some of the new government's policies. It was reissued in September 1984. The delay in effectiveness was due mainly to the need to await the cross effectiveness of the AfDB Loan Agreement; in the end, this condition was waived in the interest of implementation.  - 11 - Table 3: .Credit Disbursements Cunulative Estimated and Actual Dlsbursements (USS *000) Actual as Aoaraisal Estimate Aetual X Estimate FY86 300 FY87 1,500 1,970 131 FY88 2,900 5,370 185 FY89 4,100 6,900 168 FY90 5,200 8,350 160 FY91 6,400 8,640 135 FY92 6,600 8,640 (1) 130 Date of final disbursement: May 28, 1991 (1) Excess of Actual over Estimate accounted for through appreciation of the SDR. Table 7: Use of Bank Resources A. Staff Inputs Stace of aroiect Cvcle Total S1 Through Appraisal 93.6 Appraisat through Soard Approvat 6.0 Board Approval through Effectiveness 9.9 Supervision 52.2 B. Missions Stage of Specialization Performance Project Cwcte Month/Year No. of Persons Dams in Field Reoresented !/ Rating Status Identification N/A N/A U/A Q/A N/A Preparation g/ 83/84 - - Appraisal 10/84 4 28 E,A,FF,EE Supervision 02/86 2 10 E,F 2 2 10/86 2 11 A,A 2 3 03/87 2 12 A,F 2 4 10/87 2 9 A,F 2 5 03/88 2 10 A,F 2 6 10/88 2 5 A,F 2 7 09/89 2 3 F,A 2 af Key to Specialization: A = Agriculturist, E Economist, F u Financial Analyst, EN = Engineer b/ Key to Status: 1 = Problem free or minor problems; 2 a Moderate problems; 3 a Major problems c/ Preparation was carried out by consultants who liaised with Bank staff in Washington and Abidjan.  - 12- Table 6: Status of Covenants Covenant SbIect Sta Loan Asreement Section 3.01(b) ({) open project account with $300,000 equivalent; Net (iii) ensure project has adequate foreign exchange to finance inputs in years 4 and S. Not met Section 3.02 Nerge ORG and DPDRG; appoint local director; set up liaison office, with approved head of liaison at Conakry. Net Section 3.03(b) Consultants and specialists to train local steff. Net Section 3.04(a) Establish Agricultural Credit Unit Met Section 3.04(a) ACU to open and maintain an Met, but account (credit fund) for funds used for deposits of farmers' repayment other purposes Section 3.05 Inputs sate price to reflect actual costs. Not met Section 3.06Ca) Prior approval of fellowship candidates by IDA Met Section 3.06(b) Take measures to ensure recipients of fellowships remain in service at least 3 years after training. Not met, but trainees remained section 4.01(a) maintenance of records and Not satisfactorily accounts according to sound met because of accounting practices. irregularities Section 4.01(b) Annual audit of project Met up to F.Y. '90, accounts, with report but always late uithin 6 months. Section 5.01(a) Permit private trading of products throughout Guinea Met Section 5.01(b) Review annually with IDA levels of agricultural prices especially rice and coffee. Section 5.01(c) No subsidy on imported rice. Section 5.01(d) Permit traders to freely purchase farmers' coffee. met  - 13 - Studies Studies Status Imact Study Agricultural credit study Conpleted An agricuLtural credit system was was set up for alt all of Guinea. Preparation of foltlou-up project Not done Comments Preparation of folLow-up proJect was abandoned because of Government's policy to Integrate all project activities Into properLy established national structures.  ,9 9 T - -g_. S E N E G A L X, . - ~~~~~~~~~~~~~~~~~~~~~~~~~~~iBRD 18936 4 'ALG;ounkoon M A L A 4