Upper Volta - Third Rural Development Fund Project

Document of The World Bank FOR OmCIL USE ONLY RapurtNo. 11706 PROJECT COMPLETION REPORT BURKINA FASO THIRD RURAL DEVELOPMENT FUND PROJECT (CREDIT 1218-BUR) MARCH 5, 1993 Agriculture Operations Division Sahelian Department Africa Regional Office 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 BND Banque Nationale de Developpement BOAD Banque Ouest Africaine de Developpement West African Development Bank CCCE Caisse Centrale de Coop6ration Economique (French Aid Agency) CDR Comite pour la Defense de la Revolution (Committee for the Defense of the Revolution) CNCA Caisse Nationale de Credit Agricole (National Agricultural Credit Bank) CRPA Centre Regional pour la Promotion Agropastorale (Regional Agropastoral Promotion Center) DSA Direction des Services Agricoles (Agricultural Services Department) ERR Economic Rate of Return FAER Fonds d'Assistance a l'Equipement Rural (Rural Equipment Assistance Fund) FAO/CP Food and Agriculture Organization/Cooperative Program FDR Fonds de Developpement Rural (Rural Development Fund) FEER Fonds de l'Eau et de l'Equipement Rural (Water and Rural Equipment Fund) HER Direction de l'Hydraulique et de 1'Equipement Rural (Rural Water Supply and Engineering Department) IDA International Development Association MAE Ministere de l'Agriculture et de l'Elevage (Ministry of Agriculture and Livestock) MDR Ministere du Developpement Rural (Ministry of Rural Development) ME Minist6re de l'Eau (Ministry of Water) MQP Ministere de la Question Paysanne (Ministry of Peasants Affairs) PCR Project Completion Report ORD Organisation R6gionale de Developpement (Development Regional Organization) RAF Loi sur la Reorganisation Agraire et Fonciere Agrarian and Land Tenure Reorganization Law R&D Research and Development RDF Rural Development Fund SAED Societe Africaine d'Etudes et de Developpement (African Company for Development and Studies) SAR Staff Appraisal Report TPR Tribunal PopuLaire de la Revolution (Popular Tribunal of the Revolution) UNDP United Nations Development Program FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES Metric System  FOR OFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation March 5, 1993 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND ThE PRESIDENT SUBJECT: Project Completion Report on Burkina Faso Third Rural Development Fund Project (Credit 121 8-BUR) Attached is a copy of the report entitled "Project Completion Report on Burkina Faso Third Rural Development Fund Project (Credit 1218-BUR)" prepared by the Africa Regional Office. The Borrower contributed Part II. The project aimed at improving rural infrastructure and increasing agriculture production through small-scale works and support services. It also aimed at enhancing the institutional capacity for funding and managing rural investments. While it achieved significant results, the production impact was short-lived and the project failed to develop a reliable and cost-effective capability for managing small- scale rural investments. The project was implemented in a turbulent administrative environment. The Rural Development Fund (RDF), a financial intermediary, had to take on technical tasks because of severe weaknesses in line agencies and poor coordination. On balance, the project is rated as marginally satisfactory, its sustainability as uncertain and the institutional impact as partial. The Project Completion Report is of high quality. An immediate audit of the third project is not intended. The two predecessor projects were audited. Attachment This document has a restricted distribution and may be used by recipients only In the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.   FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT BURKINA FASO THIRD RURAL DEVELOPMENT FUND PROJECT (CREDIT 1218-BUR) TABLE OF CONTENTS Page No. Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evaluation Summary . . . . . . . . . . . . . . . . . . . . . . . PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE A. Project Identity .1.. . . . . . . . . . . . . . . . . . B. Project Background . . . . . . . . . . . . . . . . . . . 1 C. Project Objectives & Description . . . . . . . . . . . . 2 D. Project Design and Organization . . . . . . . . . . . . 3 E. Project Implementation . . . . . . . . . . . . . . . . . 4 F. Project Results . . . . . . . . . . . . . . . . . . . . 5 G. Project Costs and Financing . . . . . . . . . . . . . . 10 H. Project Impact and Sustainability . . . . . . . . . . . 10 I. Bank Performance ................... 12 J. Borrower Performance ................. . 12 K. Project Relationship ................. . 13 L. Consultancy Services ................. . 13 M. Project Documentation and Data . . . . . . . . . . . . . 13 PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE . . . . . . . 1S PART III: STATISTICAL INFORMATION 1. Related Bank Credits .17 2. Project Timetable .19 3. Disbursement of Credit .20 4. Project Implementation .21 5. Project Costs & Financing . . . . . . . . . . . . . . . 23 6. Project Results. 25 7. Status of Covenants .27 8. Use of Bank Resources .28 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.   BURKINA FASO THIRD RURAL DEVELOPMENT FUND PROJECT (CREDIT 1218-BUR) PROJECT COMPLETION REPORT PREFACE This is the Project Completion Report (PCR) for the Third Rural Development Fund Project in Burkina Faso, for which credit 1218-BUR in the amount of SDR 13.6 million was approved on March 30, 1982. The Credit became effective on May 4, 1983 and was closed on April 30, 1990, after three years of extension. The last disbursement was on April 25, 1990 and the Credit was fully disbursed. The PCR was prepared by the Agricultural Operations Division of the Sahel Department, Africa Region (Preface, Evaluation Summary, Parts I and III), and the Borrower (Part II). The PCR is based, inter alia, on the Staff Appraisal Report, the Development Credit Agreement, Supervision Reports, Progress Reports by the Borrower, the first four year report prepared by the Borrower in July 1987, the external evaluation report prepared in March 1988, a Summary technical Synthesis prepared by the Borrower in March 1991, correspondence between the Bank and the Borrower, internal Bank memoranda and interviews with the executing agency officials who were associated with the Project.   - iii - PROJECT COMPLETION REPORT BURKINA FASO THIRD RURAL DEVELOPMENT FUND PROJECT (CREDIT 1218-BUR) EVALUATION SUMMARY obiectives 1. The project -a follow-up to two quite successful Rural Development Fund projects- aimed at developing a reliable and cost-effective capability for managing small-scale rural investments. The specific sub-project types identified were to increase crop and fuelwood production, improve the availability of safe drinking water for rural communities and increase the value of agriculture produce through processing and better storage. 2. Project components were: (i) land development works aiming at limiting soil erosion and water runoff and providing control of flooding and, in some specific cases, irrigation; (ii) support to research, extension and training; (iii) credit to farmers or villagers for productive investments; (iv) village cereal Banks and storage facilities; (v) tree plantations to protect anti-soil erosion works and produce wood, fodder and fruits, and (vi) village water supplies for human and animal consumption. 3. The project was cofinanced by the Dutch Government and the West African Development Bank (BOAD). FAO provided technical assistance. Implementation Experience 4. The project was implemented during a turbulent period in Burkina's history, which saw fundamental changes in institutions and policies and, towards the end of the period, growing economic problems and the launching of an adjustment program. Major changes and events which affected and disrupted project implementation were as follows: - The 1983 Revolution, the aftermath of which gave rise to intense scrutiny of civil servants, suspected -or accused- of working against the interests of the people. Popular Tribunals of the Revolution were established, creating a climate of fear among civil servants, including those linked to the project. - Frequent government reshuffles and changes in the structure of ministries and other government service structures. - Government issuance of a new land tenure law (RAF) establishing that all land belonged to the State and no land could be owned privately. This law discouraged people from making any long term investment in land.  - iv - - Transfer of the Project implementing agency, the Rural Development Fund (FDR), to a different ministry and its merger with the Rural Equipment Assistance Fund (FAER) to create the Water and Rural Equipment Fund (FEER). - Dissolution of one of the key executing central agencies, the Rural Water Supply and Engineering Department (HER). - Redefinition of the mandate and status of the Regional Development Organizations (ORD), responsible for the execution of some components of the Project and providing technical support. As they became Regional Agro- Pastoral Promotion Centers (CRPA) with administrative status, the ORDs lost their autonomy and their commercial status. - Increasing involvement of FDR-FEER in technical matters and in project execution, a major shift in its mandate. Results 5. FDR, HER and the ORDs were identified at appraisal as the key implementing agencies. with the dissolution of HER and the change in the ORD mandate, FDR-FEER found itself assuming also the technical functions (execution and supervision) initially assigned to other responsible executing agencies. Because it did not have a reliable capability in these areas, FDR- FEER had to abandon operations such as irrigation. 6. As a result of the project, crop production yields increased, in most cases, during the first two years but declined subsequently; the performance of tree plantations remained very low; the supply of safe drinking water benefited hundreds of communities; the processing and storage facilities for cereal production, though very much in demand by the community, faced management and maintenance problems, and credit repayment rates were low. (i) Under the land Development component, the project supported the following developments: for the anti-erosion measures sub-component, over 58,000 ha, nearly 300% of the area targeted; for simple management of bottomlands, 677 ha or 107% of the area targeted; for the improvement of existing bottomland developments and the small scale irrigation sub-components, 19% and 14% respectively of the targeted objectives, mainly as a consequence of the closing down of HER. As a result of these operations, crop yields increased significantly the first two years, but started declining for a series of reasons. The earth bunds technique utilized for the anti- erosion measures was found inadequate as bunds had a very short life span. These were later replaced by stone bunds which did not require heavy machinery, had a longer life span and needed less maintenance. In the bottomlands and the irrigated areas, decline in yields was mainly due to the limited or non-use of additional production inputs such as fertilizers or improved seeds. (ii) Support services such as training and extension were provided at a rate, in person-days, above the targeted objectives: training of extension agents (107%), extension (105%), workshops (181%), and training of tractor drivers (128%). On-farm trials and demonstration plots (applied research) were established at over 201 sites. In addition, literacy courses -not planned for under the project- were organized. (iii) Rural credit was made available for the acquisition of grain mills, animal drawn implements and rice dehullers. The number of units actually financed through the credit represented 63%, 18% and 0% respectively of the targeted figures. The demand for grain mills was high but poor management and maintenance, and low credit  repayment rates forced the project to down size the operation. Except for the cotton area, credit granted for animal drawn implements was low, as farmers could not finance the initial down payment. With regard to the rice dehullers, there was no interest from the community for such equipment. A separate line of credit was set aside to promote the establishment of animal drawn implement production units but could not be utilized as the agency designated to administer the funds, FAER, could not meet the conditions (restructuring) requested. (iv) The program for the construction of cereal banks was fully completed and provided the country with an additional storage capacity of 5,000 tons. Credit was provided to help village groups, in charge of the cereal banks, acquire initial cereal stocks but the credit repayment rate was low (68%). Cereal banks faced some difficulties in financial and inventory management, stock disposal in the surplus regions, and pest and disease control. (v) The construction of village supply stores was successfully completed and credit was made available for the purchase of the initial stock of goods. Poor financial and stock management resulted in very low credit repayment rates. (vi) The tree plantation sub-components of the project were implemented and the execution rates, in terms of hectares planted and in comparison with the targeted areas, were between 103% to 423%. The results in terms of survival and growth, however, were very low due to the prevailing soil and climatic conditions. (vii) The rural water supply component of the project, consisting of digging new -or deepening existing- wells and drilling tubewells was successfully completed, providing safe drinking water to hundreds of thousands of people and to their livestock. The free supply of the well pumps by the project was conditional on the community's commitment to finance the maintenance and operating costs. Because this condition could not be met, pumps were not provided. 7. Although most of the targetted objectives were physically attained (number of trees planted, wells dug, hectares of land treated, etc.) the operations were generally not very successful from the beneficiaries' point of view - the communities. Many of the investments faced difficulties or were abandoned after project financing was discontinued. Sustainability 8. The question of sustainability must be addressed at two levels: (a) the sustainability of FDR-FEER as "a reliable, cost-effective capability for managing small-scale rural investments" as this was defined as the main objective of the Project, or (b) the sustainability of the investments financed by FDR-FEER, namely the small-scale rural investments as such. Sustainability at the level of FDR-FEER is basically irrelevant if most of the investments it was financing are not sustainable, as it was in fact the case under the Project. Moreover, many of these investments were subsidized. Of the total project disbursements, credit to rural communities represented only three percent. The rest constituted subsidized costs for investments, support services and administration. Even for the small amount on-lent, the credit repayment rates were low. The project relied, therefore, heavily on grants. In addition the actual costs of FDR-FEER (CFAF1,887 million or 21% of total costs) were too high, and exceeded appraisal estimates (CFAF552.2 million or 9% of total costs). Long-term sustainability was therefore not achieved under this project, with a few exceptions such as the stone bunding operations.  - vi - 9. At the time of appraisal, economic rates of return (ERR) were estimated for some investment components. These were: the Erosion Control Measures (13%), the Simple Management of Bottomlands (60%), the Improved Management of existing Bottomland Developments (12%), the Small-Scale Irrigation (12%), and the Tree Planting (15%). For the purpose of the PCR exercise, the calculation of the Economic Rate of Return for Erosion Control Measures was disregarded since the unsuccessful technique of earth bunds was abandoned. The rate for the Simple Management of Bottomlands was 40%, and 11.3% for the Improved Management of existing Bottomland Developments. The ERR for the Small-Scale Irrigation component was negative (-7.1%). However, it excludes significant external benefits such as the development of vegetable gardens by women, fishing activities, the use of water resources for drinking purposes (human and animals), and for the production of mud bricks. The rate of return for the Tree Planting component could not be calculated as exploitation of these trees has yet to begin. However, it is expected to be very low or even negative, based on actual survival rates and the slow growth monitored. 10. From the communities' point of view, investments with negative or low economic rates of return can still be considered attractive if (i) the investment are free of cost because they are subsidized, (ii) labor constitutes their in-kind contribution and has in fact no real opportunity cost, and (iii) there is an incremental production, acceptable enough for them to justify their present level of input. In other words when their financial rate of return is positive. Findings and Lessons Learned 11. The multiple role played by FDR-FEER as financier, implementor and supervisor hampered its performance. This, and the lack of proper internal monitoring mechanisms allowed FRD-FEER, for example, to pursue its erosion control program using inappropriate techniques (earth bunds) for five years before switching to stone bunds. 12. Another lesson learned was that the execution, on the other hand, of technically and economically sound investments does not, by itself, constitute a sufficient condition for success if a proper enabling environment is not created. Problems not related to technologies were encountered, which prevent some operations from meeting success. The findings indicate a pressing need for: - a clear and appropriate legal framework to provide security over land and other natural resources, and guarantee community rights over the benefits generated by the investments; - an adequate training and extension to enable people to efficiently maintain and manage the investments and maximize the results. Irrigated crop production, for instance, cannot be sustainable if the scarce and expensively obtained water resources are not utilized in a rational manner and if traditional production systems are not improved (use of production inputs such as fertilizers, pesticide and/or high yielding crop varieties); and - a technical back-up and efficient supervision, monitoring and feed-back system to be set-up after the investment is realized, in order to identify problems and take corrective measures when necessary. 13. Emphasis was put, at appraisal and during implementation on quantitative achievements (number of wells to be dug, trees to be planted or mills to be installed). A better method would have been to establish, as a principal objective, the development of sustainable rural production systems, based on the most rational use of the existing renewable natural resources and using a holistic approach. The specific actions to be carried out, would then be decided at the implementation stage with the community members. As a  - vii - the investments tended to be scattered and unrelated, with little cumulative or synergetic effect. 14. The basic responsibility for reporting, identifying problems, and proposing solutions acceptable to IDA for projects, rightly must lie on the Borrower. While there is no overwhelming evidence to suggest that Bank supervision missions under this project fell short in providing effective problem identification and recommendations, the borrower's comments to the effect that "supervision rarely went beyond observations and/or criticisms" is a cause for concern. So is the comment that financial recommendations were provided only in English. There are indications that, especially given the complexity of the project, more supervision inputs and more focus on field visits, would have been most useful. 15. The concept of participation was applied in a restrictive way. People were informed about possible investments the project could finance, had some choice between these investments, and demonstrated their interest and commitment by providing in-kind labor and, in some cases a financial contribution. They were then trained on how to operate, maintain and replace the investments. In practice, the decision making process remained rather supply driven and top-down. A true active participation implies that the beneficiaries occupy a central role in deciding what their needs and priorities are in terms of financial and technical support. 16. FDR-FEER became a large and centralized rural development agency, too remote from the field to be able to make sound investment decisions at the micro level. Such large institutions have a tendency to become bureaucratic and their operating costs too high. The presence of decentralized, local, relatively autonomous services is essential for good project implementation. 17. The Economic rates of return as defined and calculated in the Appraisal Report for various components, do not fully reflect the significance of this type of projects for the rural communities. To be fully reflective, cost-benefit calculations should be made, based on a variety of point of views (such as the Government, the collective, the targeted communities and, more importantly, the individual beneficiary). Furthermore, there are external benefits that are difficult to accurately measure and qualitative results that are not measurable. 18. With respect to technical assistance, the lesson is that the presence of a large group of expatriate experts stifles the initiative of national project staff and failure to designate local counterparts undermines the anticipated transfer of knowledge and skills. 19. The lessons on policy, technology and institutional issues learned through this project, and described above, were instrumental for the design and development of a new series of more promising projects in the Sahel, such as the Burkina-Environmental Management Project and the Mali-Natural Resources Management Project, for which credits were approved by the Bank in FY91 and FY92 respectively. A Niger-Natural Resources Management Project is currently under preparation.   PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE A. Project Identity Name: Third Rural Development Fund Project Credit no.: 1218-BUR RVP Unit: Africa Region, Country Department V Country: Burkina Faso Sector: Agriculture Subsector: Rural Development B. Project Background 1. Sector Development Obiectives. At the time the project was appraised in 1981, Agricultural production accounted for about 45% of GDP and nearly 65% of exports. Fifty per cent of export revenues were earned from the sale of cotton lint and oil seeds, while the export share for livestock and meat products was decreasing. Production of coarse grain was sufficient in good rainfall years but deficits were foreseen due to population growth trends. The urban population's demand for rice was rising, and most of rice was being imported, since irrigated production was lagging behind its potential due to several constraints. 2. The government's top priority at the time of project appraisal was to achieve national self-sufficiency in food production translated into the following proposed measures: (a) maintain, and if possible improve, the productivity of rainfed crop production through extension advice and inputs; (b) develop bottomland agriculture and irrigated production schemes to counter the effects of insufficient and unreliable rainfall; and (c) promote migration to the under-populated southwest and specially the Volta valleys where rainfall is adequate and reliable and soils more fertile. 3. The Bank provided financial support to three successive Rural Development Fund projects --RDF, RDF II and RDF III-- for a total amount of USS27.6 million, over a 17-year period between 1972 and 1990. The second and third projects were also supported by loans from the West African Development Bank (BOAD) and Dutch grants. In addition, the implementing agency, FDR-FEER, received grants from Canada and the European Development Fund (FED) to finance other projects in the country, and technical assistance from the French government and UNDP-FAO. Earlier Projects 4. Credit 317-UV in the amount of US$2.2 million for the first Rural Development Fund project (RDF) was approved in June 1972 and the final disbursement was made on June 23, 1977. The objective of RDF was to finance a variety of small productive investments which, because of their size and scattered locations, were otherwise not suitable for conventional Bank Group lending. The project, which was conceived as the first experimental phase of a longer program, was to consist of: (i) 500 wells to improve village water supply; (ii) 700 small village storehouses for farm inputs; (iii) the development of 1,500 ha of bottomlands for rice cultivation; (iv) erosion control works on 1,000 ha to improve crop yields (sorghum, cotton, groundnuts); (v) 200 ha of small irrigation schemes downstream of the existing dams, to produce rice; (vi) the upgrading of 270 km of feeder roads; and (vii) a number of unidentified investments. An essential point in investments selection was the interest of the people concerned, expressed through their willingness to provide free labor for investments, implementation, subsequent operations, and maintenance.  5. Concentrating on the central plateau, Credit 640-UV for the Second Rural Development Fund Project (RDF II), in the amount of US$9.4 million was approved in May 1976 and financed the same type of components and investments as under the first project with the exception of feeder roads. In addition, it provided for training and technical assistance to overcome staff constraints and financed studies of development potential in adjacent areas. Other components included agricultural credit and the creation of a maintenance brigade for the wells. The project was to be executed over a five year period, but was extended by one year as activities in the first, and partly the second, year were still financed under RDF. Additional financing provided by Dutch Aid (DFL5.15 million equivalent to about US$2.0 million) was to ensure that no substantial reduction of the work program will be necessary in the last project year. 6. According to the Staff Appraisal Report for the RDF III project, the major lessons learned from the first Rural Development Fund project were: (a) the need to strengthen staff in the technical and executing agencies (HER and ORDs); (b) the problem of ensuring maintenance of assets created with RDF financing, especially feeder roads; and (c) the absolute necessity for ensuring the availability of credit to enable farmers to acquire complementary inputs and equipment in the absence of which the full benefits of the land development schemes could not be realized. Overall, however, the project was described in the PPAR issued in September 1979 as a "highly innovative project which was successful in implementing diversified and scattered subprojects". Beneficiaries were estimated at some 230,000 people. 7. The follow-up project, RDF II, incorporated the lessons of RDF. It continued with the sub-projects proved successful under its predecessor, included elements which had been revealed to be essential (credit and strengthening of staff of technical and executing agencies) and excluded feeder roads for which a more competent national body had meanwhile been created, thus ensuring better maintenance. Results, under RDF II, were generally satisfactory, especially for the erosion control and water supply components. Credit, primarily medium-term loans for animal traction and agricultural equipment, was a popular program although repayment rates at farm level could not be determined. 8. Poor results recorded for the two projects in bottomland development were the result of: (a) land tenure problems, (b) the heaviness of bottomland soils which made land preparation difficult, (c) poor design and quality of works; and (d) an undeveloped marketing system for paddy. The major successes of these earlier projects, however, were creating an efficient organizational tool for appraising small productive rural investments, providing sound financial control during their execution, and building up a capability for the monitoring of such projects. The Third Rural Development Fund project (RDF III) was based on a full utilization of these capabilities. C. Project Objectives and Description 9. Project Obiectives. The project aimed at completing the process pursued under the preceding two projects, of developing a reliable and cost-effective capability for managing small-scale rural investments. The specific sub-project types identified were to increase crop and fuelwood productions, improve the availability of safe drinking water for rural communities and increase the value of agricultural produce through processing and better storage. 10. Project Components. The project included the following components: (a) land development comprising: engineering design services, implementation and supervision of a program of (i) anti-erosion bunds to protect about 20,000 ha of dry land farms, (ii) simple bunding on 640 ha bottomlands to provide partial control of flooding, and construction of earth dams, intake structures, spillways and flood protection dykes to provide for complete control of flooding and supplementary irrigation on 240 ha of existing bottomland developments, and (iii) small-scale irrigation works  -3- to provide full water control irrigation on 160 ha downstream of existing reservoirs or from ponds and shallow wells; (b) support to ORDs consisting of training and providing initial funds for the departments responsible for training of extension staff and farmers; credit administration and input supply; and research and demonstration; (c) credit (i) to farmers for the purchase of ox-drawn implements and oxen to be administered by CNCA; (ii) to groups of farmers, to finance small rice dehullers, small grain mills and initial stocks of cereals and general merchandise for group-owned and operated village stores, to be administered by RDF; and (iii) to the Rural Equipment Assistance Fund (Fonds d'Assistance a l'Equipement Rural-FAER) to finance the purchase of inputs required for the manufacturing of animal drawn implements; (d) villaqe storage sheds which would be used either to store inputs and implements in which case they would be operated by ORD staff, or as community-owned and operated stores for traded cereals (cereals bank) or general merchandise; (e) tree plantations consisting of investments in 6 nurseries and about 400 ha of trees to provide wood, fruits, fodder and the equivalent of a further 160 ha in trees to fix bunds on soil erosion sites; (f) village water supplies consisting of the establishment of 60 new wells and deepening of 200 existing wells; the drilling of 150 boreholes; a provision for the equipment and operating funds for wells and boreholes maintenance brigades; and a stock of spare parts for pumps which would be sold for cash to villagers; and (g) proiect management costs consisting of a headquarters building for RDF and other costs associated with managing the investment program outlined above. (h) The project was also to provide funds for the financing of other small- scale rural projects which would be identified in the course of project execution. Such small projects would be selected using substantially the same criteria of technical and economic feasibility as for the already identified investments. D. Proiect Design and Organization 11. The project was designed to be flexible in its implementation, thus the different components were described in a broad manner. The choice of villages or communities and the type and size of investments were to be decided at the local level at the time of implementation. Emphasis was put at appraisal on the selection criteria for the investment which were defined as follows: (a) need as evidenced by a request from potential beneficiaries; (b) effective demand as evidenced by a willingness on the part of such beneficiaries to contribute in kind or in cash to the investment; (c) technical feasibility which implies a design simple enough to be realized by local farmers under the technical supervision of staff of project executing agencies (ORD, HER, etc.); (d) economic efficiency, which is deemed to be satisfied by a minimum economic rate of return of 12% on productive projects and, for infrastructure investments by a design judged to produce the lowest cost per beneficiary; and  - 4 - (e) social acceptability, which requires RDF to examine proposed subprojects taking into account the larger social context to ensure their acceptance by target populations. 12. The preparation cycle for the investment micro-projects at local level was conceived in order to involve the beneficiaries in the identification and design of these projects with the back-up of technical agencies at regional (ORD- CRPA) and central (HER, FDR-FEER) level. Pre-appraisal was done by the ORDs and appraisal by FDR-FEER. The latter deciding on the financing, supervising the implementation and carrying out ex-post evaluations. The FDR-FEER Board approved the annual program to be executed by the competent technical agencies and the beneficiaries providing basic labor in kind. 13. The organization setup proposed for the project involved several agencies: at regional and local levels, FDR-FEER called on the existing institutions (ORD, HER etc.) to execute project sub-components. Over the years however, FDR-FEER began executing directly some of the tasks and studies in addition to being funding institution with a financing, monitoring and supervision role. This prevented the institution from having an objective feed- back on its own performance. E. Project Implementation Modification in Project Implementation and Financing 14. The Credit was approved by the Board on March 30, 1982 for an amount of SDR 20.2 million, equivalent to USS 23.8 million. This amount included a Dutch grant for an amount of SDR 6.6 million equivalent to US$7.8 million to be administered by IDA. Soon after Board approval, the Dutch Government decided to administer directly its funds and to use parallel financing. This decision delayed until October 28, 1982 the signature of the Credit which became effective May 4, 1983. These delays did not, however, seriously affect project implementation as the Dutch Government provided an interim grant to FDR. Factors that Affected Proiect Implementation 15. The 1983 Revolution led to a series of measures, some of which had an adverse effect on all the government staff including the project staff. These measures were: (i) set-up of Committees for the Defense of the Revolution (CDR), creating a tense working environment for civil servants suspected of exploiting the workers; (ii) attempts to "moralize" the civil service by establishing Popular Tribunals of the Revolution (TPR). The fear created by these TPRs paralyzed civil service employees who became too scared to do anything without clearance from the highest level; and (iii) suspension of allowances, including travel allowances, which hampered the work of field staff whose regular interaction with the rural communities was essential to project success. There were, at the same time, budget cuts on means for transportation (field vehicles) and office supplies. 16. In 1984, the Ministry of Rural Development (MDR) was dissolved and two new ministries were created: the Ministry of Agriculture and Livestock (MAE) and Ministry of Water (ME). The Regional Development Organizations (ORDs), formerly with the MDR, were put under the MAE and the FDR, also formerly with the MDR, was attached to the ME. Because ORDs and FDR ended under different ministries, with FDR in the ME controlling the funds, ODRs became less committed to the project and were brought to focus more on agricultural extension at the time the Bank started to support the agricultural extension program in Burkina.  17. The government adopted in 1984 a new agrarian and land tenure law (RAF) giving the ownership of all lands to the state. Although never fully implemented, this law constituted a disincentive for long term investments in rural areas. 18. In 1986, the Ministry for Peasants' Affairs (MQP) was created and later renamed Ministry of Peasants' Cooperative Action (MACP). The Rural Water Supply and Engineering Department (HER), in the Ministry of Water was abolished and part of its activities were transferred to FDR, which was renamed Water and Rural Equipment Fund (FEER) and put under the ME. The key role expected to be played by HER in project implementation (to assure quality control for all interventions related to water erosion control and water harvesting, collection, storage, and distribution) could not be effectively assumed by FEER since it did not have the necessary expertise. As a result, the quality of the work declined. In addition FEER was granted a status of an Administrative Public Corporation (EPA) with much less administrative and financial autonomy than FDR had as a Specific Public Corporation (EPS). 19. The Government took the decision in 1986 to abolish the ORDs but for about one year the decision was not implemented and ORD staff remained employed but not knowing what their future would be. Under these circumstances, little work was done and project activities slowed down. In 1987 ORDs were re-baptized Regional Centers for the Promotion of Agro-Pastoral Activities (CRPA) but lost in the process their autonomy with regards to signing contracts with agencies such as FEER. The CRPAs ceased to administer the Agricultural Credit at local level. F. Pro-ect Results 20. The project's original duration was four years with December 31, 1986 as completion date. The credit closing date was then extended to December 31, 1989, which allowed some of the project components to be completed beyond appraisal estimates, although other components could not fully implemented. Results are summarized below. (a) Land Development 21. Anti-Erosion Measures using contour bunds were carried out over a total area of 58,251 ha (nearly 3 times the estimated area of 20,000 ha) at the cost per ha of CFAF 61,274 (appraisal was CFAF 77082). The technique of earth bunds was found inadequate as these were damaged by water erosion and animal stamping, and became inefficient after three to five years while their expected life span, at the time of appraisal, was 10 to 15 years. The project staff was not sufficiently trained for the proper design of these bunds; the grassing along the bunds was not done and the works maintenance by the communities was neglected because of the amount of labor required. In addition, the farmers used to break open the earth bunds themselves during good rainfall years to avoid waterlogging. As an adverse effect, open bunds increased rill and gully erosion. Crop yields increased by 60 to 80% (90 kg net in the North and 250 kg in the central zone) the first year, but declined rapidly when the bunds started to deteriorate. After five years, the yields were back to their initial level. From the beginning, the implementing agency, FDR-FEER, had a tendency to put more emphasis on the execution of physical investments and less on the impact of these investments. The situation was partly corrected in 1987: earth bunds were replaced by stone bunds that allowed water seepage, reduced waterlogging and flooding, and controlled soil erosion. Leguminous trees (acacia) were planted on both sides of the bunds and other soil conservation techniques were utilized. Initial results indicate that yields increased and remained stable over a longer period of time ( 5 years). It is considered that stone bunds, if regularly maintained, will last indefinitely. Precise data from project experiments with stone bunds are not yet available. 22. Management of Bottomlands was executed on a total area of 677 ha, slightly larger than projected (640 ha) and at the cost of CFAF150,567/ha. Total  production during the first four years of the project, and over a total cropped area of 392 ha attained 1837 tons of rice, 76 tons of sorghum and 1.1 tons of soya. Gross yields for rice increased sharply from 0.3 to 1.3 t/ha during the first two to three years, then declined progressively over the years (0.8 t/ha in year-10) mainly due to soil fertility decline (no use of fertilizers) and limited maintenance. The technique utilized (stone bunds) gave good results and bunds had a long lasting effect. One source of disincentive, however, was the very small size of individual plots allocated to each family, (0.09 ha) because of a larger than expected demand from farmers. 23. Improvement of Existing Bottomland Developments. This was carried out, using more sophisticated techniques involving the construction of small dams of 50,000 to 500,000 m3 reservoir capacity. The area developed (46 ha) was small and represented only 19% of the appraisal estimates, as HER --designated during appraisal for the execution and supervision of this component-- closed down. Investment costs (CFAF574,152/ha) were lower than appraisal estimates. Gross yields for rice, the main crop, attained 2 t/ha in the first production year but declined slowly, mainly because of decline of soil fertility in the absence of fertilizer to replenish soil nutrients. Over the first two campaigns (84-85 and 85-86), 125 tons of paddy rice and 9 tons of corn were produced. Given the limited size of individual family plots, 0.05 to 0.1 ha, the value of the incremental production per family was small. This production played, however, a positive role in solving, locally, the food shortages that followed the 1984-85 drought. The life span of the works (small dams) is presently estimated at 15 years, although good maintenance is expected to expand that period. 24. Small Scale Irrigation Works involving construction of dams similar to the above but with larger water capacity, were originally planned over 160 ha spread over 10 sites. This component suffered from the closure of HER and the subsequent lack of technical back-up and supervision. Only 23 ha or 14% of the estimated area were developed. Investment costs were CFAF 3.06 million per ha, nearly 6 times higher than those of improved bottomlands, because of the larger structures (dams) involved. As a result of better supply and control over water, crop yields were higher (about 2.5 t/ha/year) but, as for the managed bottomland areas, yields declined after the first year in the absence of fertilizers. 25. Besides cereal production, bottomlands and irrigation works provided additional benefits as they permitted the development of vegetable gardens (mostly managed by women) and fishing activities. Irrigation made also water available for humans and animals, and in some cases for the production of mud bricks for housing. 2-6. The management of Bottomlands and the Small Scale Irrigation Works components were executed with the help of the communities, however, most of these works were done on land belonging to community leaders, not on family or community plots. There was therefore little incentive for people to participate actively in the operation and maintain the investments. Even when works were done on community land, people were cautious about investing much of their time and efforts as there was no assurance that they would get their fair share of the output. As per Section 3.04 of the Credit Agreement, the government was supposed to transfer property titles to members of the community who participated in the program, but this clause was not respected. The lack of security on land tenure increased and extended to family holdings when, in 1984, the Agrarian and Land Tenure Reform Law (RAF) was promulgated, stating that all land belonged to the State. The law was never really implemented and it was amended to recognize land rights to communities and individuals. (b) Support to ORD Services 27. Funds were made available under this component for training, applied research and demonstration programs, and for administering rural credit funds. 28. Training. The ORDs (later CRPAs) of Ouagadougou and Khaya were strengthened; offices were built or refurbished, audio-visual equipment and vehicles acquired, and recurrent costs were financed by the Project funds.  ORD/CRPA extension agents were trained in order to provide better assistance to farmers. A total of 11,075 person-days (pd) of training (appraisal estimate was 8550 pd) were organized for the extension agents who, in turn, provided training and extension services to farmers in the form of (i) visits (7318 pd organized for 7000 scheduled), (ii) workshops (2172 pd organized for 1200 scheduled), and (iii) literacy courses (69612 pd organized). The latter, not scheduled at the time of appraisal, was to enable rural groups to better manage cereal banks, mills and other investments for which basic management and accounting skills were required. In addition, tractor operators received 562 pd of training (440 scheduled). A long-term intermediate technician training program, scheduled for 3 people, did not take place as no candidates were found. Training and Extension had a positive effect on the quality of the work carried out by community members, and the design, construction and maintenance of land management works improved. Farmers also learned how to produce and use organic fertilizer more efficiently and protect crops. The quality of this training was not uniform and regular. Communication between farmers and government services were improved through the program of visits. 29. Applied Research and Demonstration. The objective of this component was to establish a link between Agriculture Research and Agriculture Extension Services in order to develop and promote techniques suitable for sustainable agricultural production and for natural resources management. On the basis of recommendations provided by the Agricultural Research institutions, on-farm trials were established with farmers' participation, on lands that were subject to anti-erosion measures, managed bottomlands, and irrigated sites. The principal techniques tested were the production and utilization of organic fertilizer, use of local rock phosphate, tie-ridging, transplanting of rice and use of improved varieties of rice, niebe and sorghum. A total of 201 small sites (16 large sites originally planned) were established, spread over the five ORDs of the Central Plateau where R&D units were created. The project provided financial support for vehicles purchase, travel costs and for the acquisition of necessary inputs. Field experiments and demonstrations were successful in soil improvement technologies and in the use of organic fertilizer and phosphate at anti-erosion measures sites. The bottomlands experiments, with improved varieties of rice produced higher yields but results were not conclusive because of high rainfall variability. (c) Credit 30. The objective was to provide, through the National Agricultural Credit Bank (CNCA), credits to farmers for the purchase of animal drawn Implements, rice dehullers and grain mills, and for the acquisition of initial stocks for the cereal banks and village stores. Credits were also to be made available to the Rural Equipment Assistance Fund (FAER) for the establishment of production or assembling units for the animal drawn implements. 31. Animal Drawn Implements. By the end of 1989, CNCA provided credit for the purchase of 1,479 animal drawn implements while the total scheduled was 8,800 scheduled. This low performance was due, the first year, to a slow start-up of the credit program and, the subsequent years, to poor rainfall and to the priority given by farmers to the use of the credit for additional purchases of seeds. While the needs for animal traction were properly estimated at the time of appraisal, farmers could not mobilize the funds required for the down payment and could not fulfil the principal credit condition, to install a composting unit. It was however found that the use of animal traction was more popular in the cotton producing areas (cash crops). 32. Credit Line for FAER. The credit line for the Rural Equipment Assistance Fund (FAER), later rebaptized CNEA, was never open as Government could not provide the guarantee required in the Credit Agreement with regard to the financial and organizational restructuring of FAER and its proper staffing. 33. Rice Dehullers. No credit was granted under this sub-component as no interest was expressed for such equipment.  - 8 - 34. Grain Mills. The credits for grain mills were to benefit mainly women. Delays occurred during the first year in the supply of equipment. At the end of the third year, 100 mills, or 50% of the total number scheduled, were supplied and installed. People paid to have their grain milled in order to offset the cost of maintenance and repair and to renew the equipment. Mills operated without major problems during the first and second year but, because a large number of them was poorly run and maintained, and credit recovery was very low (34%), the program was slowed down. At the time of completion, a total of 125 mills were completed or 62.5 % of the appraisal estimate. Wherever people were properly trained for equipment maintenance and financial management, mills were able to operate in a sustainable manner. Freed from manual pounding, women could devote more time to other productive (such as vegetable gardens) or family activities. (d) Village Storage Facilities 35. Seventy five village cereal banks or merchandise stores and thirty input and equipment storage sheds were to be established in forty selected villages. RDF was to finance the cost of building materials and the skilled labor for the construction through grants, and the cost for acquiring the original stock of cereals or general merchandise through credit. This program was very popular and a total of 162 cereal banks or merchandise stores and 32 input and equipment storage sheds were established. Cereal banks established had a total capacity of 5000 tons. Credit recovery rate averaged 68% with very high variability (from 99% to 32%). These banks enabled communities to constitute food security stocks. In addition, their presence in some villages made it possible for small farmers to sell locally, and at a good price, while enabling the local consumer to find a source of supply close-by, hence avoiding long journeys to the cities. Merchandise stores made available to villagers and rural communities, essential products at reasonable prices, on a non-profit basis. These stores were unfortunately poorly managed, in terms of financial and stock management and many had to close down. No precise data was available on the number of stores still in operation. (e) Tree Plantations 36. The need to promote rural tree plantations for fuel, construction timber and soil protection was strongly stressed at the time of appraisal and the volumes achieved exceeded appraisal estimates. In total, 16 new nurseries were established and four rehabilitated, producing 1.4 million seedlings between 1982 and 1989. These were utilized for the establishment of 1,392 ha of community and farm woodlots, and 472 ha of erosion control plantations. In addition, 529,000 seedlings were planted on farmers' lands representing the equivalent of another 846 ha. Compared to the appraisal estimates, this represents achievement rates of 348%, 103% and 423% respectively. 37. The results of these tree plantations were not very encouraging. Trees were planted on poor soils and heavy losses were recorded, also due to harsh climatic conditions and lack of maintenance and protection. Seedlings were browsed by livestock or suffered from drought. In some cases (intercropping with food crops), the survivors constituted a problem of competition for light and water. Although fast growing species were utilized, the yields obtained (0.3 to 1 m3/ha/year) were low compared to the 5 m3/ha/year yields expected at appraisal and were in fact similar to those obtained from the natural forest vegetation existing in the area. Tree seedlings were supposed to be sold to farmers but because of the meager results obtained, farmers were unwilling to pay, forcing the project to distribute them for free. No data is available on the exact growth obtained from these plantations as trees have not reached their exploitation age. (f) Rural Water Supply 38. The objective of this component at appraisal was to (i) dig 60 new wells, (ii) deepen 200 existing wells, (iii) equip 175 with hand pumps and (iv) drill and equip 150 boreholes. In 1983 these objectives were modified as follows: 250 new wells to dig, 100 old wells to deepen. The hand pumps and boreholes programs  remained unchanged. While the purchase cost of the hand pumps was to be borne by the project, the maintenance costs were to be met by the community. 39. At the Credit closing date, a total of 196 new wells were dug and 115 existing ones deepened. All the 150 boreholes were drilled and equipped between 1986 and 1988, supplying water to 142 villages at an average flow of 2.2 m'/hour/borehole with a flow over 5m3 in 20% of the cases. The digging of wells and drilling of boreholes was successful as this was done on a contract basis by efficient private or state firms (SAED), financed through project funds, with no contribution expected from the community. Of the 175 planned, only three hand pumps were installed on community wells as communities were not willing to pay for the maintenance costs which implied an initial investment of CFAF 50,000 per pump. This component contributed, however, to the supply of drinkable water to hundreds of thousands of people (331,000 in 1986) who previously used water from ponds and were subject to waterborne diseases. Water, readily available in the village, reduced women's workload and provided extra water for them to raise small farm animals and grow vegetable gardens. FEER indicated the occurrence of sanitation problems near and around the water points and, in 1989, a separate sanitation program was launched for 59 wells. (g) Other Accomplishments 40. Beekeeping. The beekeeping program, identified after appraisal, aimed at: installing 600 modern beehives in 3 years, training about 30 beekeepers to manage the hives and organize the marketing of the honey. Each hive was expected to produce 30 kg of honey per year. Between 1983 and 1986, 295 hives were installed in 13 villages in the provinces of Houet, Kenedougou and Comoe with the participation of 95 beekeepers. The July 87 project report from FEER (page 87, para g-l) mentions that at the end of 1985, 50% of the hives were colonized by bees and producing up to 20 kg honey per year sold at CFA 700 per kilo. The beekeepers had to reimburse the cost of the hive or CFA 14,850 (hive and base). 41. Small Stock Raising Units. The objective was to help individual farmers, especially women, earn additional income by raising small animals on their own farms. The units were to consist of 4 sows and one boar, 20 hens and 2 roosters, or 10 sheep. Forty-seven farmers obtained grants to build animal shelters, credits (total of CFA 12.6 million) to purchase animals and feed, and technical support and veterinary services. Credit recovery rate was low (19%) as incomes generated were lower than expected. Because of lack of technical back-up from the livestock department, FDR was forced in 1985 to abandon this program for which it had no expertise. 42. Vegetable Gardens. Twenty six small perimeters of about 1 ha each were established between 1983 and 1988 in bottomland areas where water was available from shallow wells. The main vegetables grown were cabbage, onion, carrots and potatoes. The July 87 report indicated low yields (8 to 12 tons/ha), due to shortages of irrigation water, late seed supply and farmers' insufficient experience in vegetable gardens. 43. Village infrastructure. Acute need for basic facilities in many villages made it necessary for the project to provide financial support for the construction of 10 village stores, 17 schools, 6 houses for teachers, one maternity and one medical center. 44. Extension Services. A decision was made in 1985 to help the province of Yatenga reorganize its extension program and test the Training and Visit system promoted by the Bank. Financial support was provided for the training of extension agents and for field operations in the areas of erosion control and compost production. Erosion control measures were carried out over an area of 1,048 ha and 1514 compost pits were built. The effect of the erosion control measures on crop yields was very positive as these yields doubled in two years. This test program, which was not identified at the time of appraisal, took place between 1985 and 1987. Further extension operations were taken over by the IDA financed Agricultural Services Project that started in 1989.  - 10 - G. Project Costs and Financing 45. Total project cost was estimated at appraisal at CFAF7.8 billion equivalent to US$28.9 million at the rate of CFAF270 per USS at the time of appraisal. The actual project expenditures at the time of credit closing were CFAF9.128 billion equivalent to US$26.1 million with an average exchange rate of 349 for the 82-89 period. IDA's contribution, estimated at SDR13.6 million (US$16.0 million) at appraisal, was fully disbursed. The Dutch contribution of DFL20 million, equivalent to US$7.8 million or SDR6.6 million (CFAF2.106 billion) at the time of appraisal, was partly disbursed (CFAF1.739 billion) and the BOAD loan for an amount of CFAF1.0 billion equivalent to US$3.7 million was fully disbursed. In addition the Government contributed an amount of CFAF 0.388 billion (appraisal estimates were CFAF0.378 million). Higher exchange rates for the SDR during project implementation provided the borrower with a larger CFAF mass of money and a three-year credit extension period, and allowed FDR-FEER to achieve larger targets. H. Project Impact and Sustainability 46. Project investment and operating costs were largely financed with donor grants and loans with limited government funding (5%). At appraisal, the Credit component represented only 15% of the estimated costs (credits to community members individuals and groups). The largest share of the costs went for subsidized investments and operating and management costs. In terms of actual expenditures, the Credit component represented as little as 3% of the total. In addition, credit repayment rates were very low: 17% for the Animal Drawn Implements component and 60% for the other components of the credit. 47. The Project results, in terms of direct physical achievements such as number of trees planted or wells dug, were generally significant and in many cases higher than appraisal estimates. A few, however, could not be attained, particularly when greater community involvement and financial participation were called for. However, when we try to determine whether these investments, successfully implemented, had a positive impact on the targetted beneficiaries in terms of significant increase in crop yields or rural income, or in terms of improving the general living conditions of the rural communities, the findings are less bright. As described in earlier paragraphs: (a) Anti Erosion Measures gave successful but short-lived results, as the earth bunds were destroyed after five years. Use of additional production inputs by farmers was marginal and erratic rainfall patterns had a negative effect on the results; (b) Management of Bottomlands was popular, but the size of the family plots were so small that the impact, in terms of additional income per family, was not significant. Limited use of fertilizers and improved seeds and poor maintenance of the works caused the yields to decline after a first good year or two; (c) Animal Drawn Implements program was abandoned as people could not reimburse the credit. Credit repayment rate was higher with farmers growing cash crops, like those in the cotton area; (d) Efficient management and maintenance of grain mills was often lacking, to the extent that many units stopped operating; (e) Cereal Banks and Merchandise Stores were very popular, but in several cases their management by community members was so poor that most of them closed down; (f) Tree Planting at the village level did not give the results that would have encouraged people to pay for more seedlings and to protect and maintain the plantations; and  - 11 - (g) Rural Water Supply received unanimous approval from the communities but their financial contribution for the maintenance of the wells and the pumps could not be always obtained. 48. In terms of sustainability being measured at the sub-components level (erosion control works, tree planting, grain mills construction, etc.), we can conclude that most operations were not sustainable. 49. In economic terms, rates of return (ERR) calculated at the time of appraisal for the four Land Development subcomponents --Erosion Control Measures, Simple Bottomlands Management, Improved Bottomland Management, and Irrigation-- and for the Tree Plantations component were estimated at 13%, 60%, 12%, 12% and 15% respectively. 50. Based on the data presently available and on reasonable projections (see Part III, section 6 (a) of the PCR), the following Economic Rates of Return were obtained: - Erosion Control Measures using earth bunds and traditional cultivation techniques: not applicable as these unsuitable techniques were abandoned. - Management of Bottomlands using simple bunds: 40.0% (0.12% of total project costs) - Improvement of Existing Bottomland Developments: 11.3% (0.27% of total project costs) - Small Scale Irrigation Works: -7.1% (0.7% of total project costs) - The ERR for Forestry Plantations for wood and fuelwood production could not be determined as trees have not yet been harvested. However, the ERR is not expected to be positive due to the high mortality rate recorded and to the very slow growth of the surviving trees. 51. The main objective of the project was to develop a reliable and cost- effective capability for managing small-scale rural investments. It can be said that FDR-FEER succeeded in being perceived as a performing agency, capable of mobilizing all the services to implement complex programs and disburse funds rapidly. Its cost-effectiveness becomes however questionable when project management costs reach 21% of the Project total costs and when final outputs are taken into consideration. Because of its focus on achieving project's physical goals and faced with the inefficiency of some of the existing executing agencies, FDR-FEER moved progressively away from its principal role as a funding agency or a bank to become an executing agency. By doing so, FDR-FEER grew too big, embracing too many functions, became a less manageable and very centralized organization (FDR-FEER had no regional offices). Sustainability did not appear to be the institution's main priority as long as its sources of funding were assured. It is important however to stress the fact that, at that time, no better alternative existed. 52. The approach adopted by the project remained top down, although the conclusion drawn from FDR 1 and FDR II was that communities must be involved, a message that had then been repeatedly conveyed by the Borrower and the implementing agency itself. It seems that community participation was interpreted narrowly as to communities being "consulted", and "associated" with the implementation of project components, essentially by providing unpaid labor and undertaking training to carry out maintenance and management tasks. In fact, they were not fully part of the decision making process. At best, they were given a menu of investments to choose from. 53. The project was designed to use a multidisciplinary approach. Several operations were therefore launched (anti-erosion measures, wells, grain mills, seed banks, beekeeping, etc.) but independently from each other and in an  - 12 - unrelated manner. As a result, project interventions were rather scattered, focussing on themes across the region (water supply or beekeeping programs for example) rather than based on a holistic approach to natural resources management or to community development, where communities are seen as the central point and where all actions are inter-related and complementary, and converging towards a common goal. I. Bank Performance 54. Project Preoaration and Appraisal. Preparation started in August 1979 and was undertaken by RDF staff under the supervision of the Regional Mission in West Africa (RMWA). It was based on a careful analysis of, and lessons learned from, the two previous RDF projects and other related projects supported by the Bank in Burkina Faso. Successive drafts of the Preparation Report were reviewed by RMWA staff during the fall of 1980 and the full report, incorporating most of the suggestions, was completed before the end of 1980. An Appraisal mission visited the country in March - April 1981. It was composed of four Bank staff, two West African Development Bank (BOAD) staff, two Dutch government representatives and one consultant. The Staff Appraisal Report and the 8 working papers constituted a solid base for all subsequent work carried out. 55. Supervision. The Bank fielded 14 supervision missions over a period of 6.5 years, representing an average of two to three missions per year. Supervision reports were carefully prepared and follow up action carried out. There was however insufficient supervision of the project's financial management and accounting aspects. A total of 95.7 staff weeks were spent on project supervision, with 88 days in the field (missions). Considering the project complexity, supervision missions should have been carried out by multi- disciplinary teams with a large part of these missions devoted to field supervision. Joint donor supervision was very difficult to achieve, particularly with the BOAD. J. Borrower Performance 56. During the first year of the project, FDR/FEER was consistently rated by the supervision missions as a performing institution, capable of mobilizing the different services involved in the project. It had an important technical assistance staff, particularly at the start-up of the project (9 expatriates in 1983), financed by the Dutch and French cooperation and FAO/UNDP. 57. Surviving political turmoil, ministerial changes and restructuring, FDR- FEER operated successively under the umbrella of the National Development Bank (BND), the Ministry of Rural Development (MDR) and the Ministry of Water (ME). Under the BND it was an autonomous financial institution responsible for coordinating, mobilizing and managing funds to finance small rural operations. After being taken out of the BND, the institution became increasingly involved in technical activities for which it had no technical capabilities or competent staff. The obvious reasons for such changes in the institution's focus were FDR's transfer to a technical ministry (MDR then ME), the merging of FDR with another institution (FAER), the closing down of a technical agency (HER), and FDR's decision to take over the execution of some of the tasks that could not be handled efficiently by ailing technical services and executing agencies. 58. On the human resources side, FDR-FEER suffered from lack of specialized technical staff to supervise important components (such as anti-erosion measures and investments geared towards the promotion of women), technical staff overburdened by administrative responsibilities, absence of national counterparts to the foreign technical assistants (FAO/UNDP experts left without having trained local staff to take over) and high staff turnover due to lack of incentives (staff recruited on contractual basis at low civil service salaries). 59. On the financial side, lengthy disbursement procedures delayed project implementation. While IDA financing and financing procedure were clearly defined, the approval process of, and type of expenses to be covered by, the Dutch  - 13 - financing were unclear and raised a series of problems. In addition, FDA-FEER had poor financial management and accounting systems and had long delays in producing financial reports and other essential financial documents. This was repeatedly highlighted in financial audit reports. The stock inventory was also poorly managed. It is only in 1988 that the accounting system was efficiently organized and computerized. 60. FDR-FEER produced detailed technical reports during the first four years of the project (82-86) and an extensive, though unclear, final report in July 87 (the original completion date of the project was December 31, 1986). The IDA credit was extended for 3 more years (86-89), however no final technical report was produced at the end of the extension period, in spite of repeated requests from the Bank. In 1991, a sketchy summary of FDR III technical achievements was submitted to the Bank and financial reports (annual and final) for the 87-90 period are still missing. The uncooperative attitude of FEER during the last years of the credit coincides with the time the Bank announced it would not support a fourth FDR project and insisted on FEER focusing on its original mandate (under FDR), giving up its current role as technical and executing agency, and reducing its staff drastically. K. Project Relationshio 61. Bank staff and Borrower staff involved in the implementation of the project had a very good professional relationship in spite of the difficult situation that followed the 1983 political events. The situation began, however, to cool down when the Bank decided in 1988 at the beginning of the extension period not to consider a fourth RDF project (proposed by the Borrower as FEER I) and recommended that FEER should return to its original mandate as a fund, and dismantle its heavily staffed and centralized technical services. Bank's decision was shared by the Dutch, French and German Governments and the EEC. FEER relationship with other donors was fairly good until 1987. At that point, several key donors started raising questions about current approaches to rural development and recognized the failure of centralized, top down, technology- driven, rural development projects. In other words it was a change of paradigm to which FEER responded reluctantly. FEER's insistence - backed by ME and the then Ministry of Planning and Cooperation - in maintaining a heavy handed role in the implementation of NRM and rural development projects in the country excluded the institution from being associated with the new generation of Natural Resources Management and Terroir Management projects developed in Burkina with the support of the Bank and the donors mentioned above. FEER was not associated either to the National Environmental Management Project ("Projet de Gestion des Terroirs") co-financed by Germany, Norway and France and IDA. L. Consultancy Services 62. Full-time consultants were assigned to the project under UNDP/FAO technical assistance. Similarly, Dutch cooperation provided full-time technical assistance. In addition, short-term consultants were utilized. These consultants and technical assistants could not, however, transfer their knowledge and skills to nationals due to lack of local counter-part staff. M. Project Documentation and Data 63. The Staff Appraisal Report (SAR) was very well prepared, with detailed annexes. Eight Working Papers were produced to guide project staff during the implementation phase. The SAR was mainly based on the Preparation Report, produced by the staff of the implementing agency with the support of Bank staff, and two other documents: "Subproject Appraisal Procedures" and "First Evaluation Report on Land Development Schemes". Other documents available prior to the appraisal were the SARs and Development credit Agreements (DCA) for RDF and ROF II and all the special, annual and final reports related to these two projects.  - 14 - 64. The SAR, Working Papers, and DCA constituted a very useful framework for the Borrower and the bank for the implementation and supervision of the project. The Borrower tried as much as possible to prepare technical and financial reports reflecting the format of Bank's documents while trying at the same time to meet other donors and government reporting requirements. As a result, reporting was a difficult exercise for the borrower while the results achieved (in terms of reporting format) did not necessarily meet each donor's specific requirements. For the Bank, for example, comparison between appraisal estimates and actual achievements, on the basis of data provided by the borrower, was very uneasy. 65. Data relevant to the preparation of the PCR was very difficult to gather and synthesize. The key document, the final report was never produced by the Borrower, except for a very short and sketchy technical summary submitted by the Borrower to IDA in March 1991, after repeated Bank requests. This report summarizes the physical achievements of the project but contains no information on project costs and project impact in terms of investments output or economic benefits. It was also found to contain incorrect data. 66. One of the two most useful documents for the preparation of the PCR was the report dated July 1987 and produced by the implementing agency at the end of the original four year period of project implementation, which was then extended to seven years. Although not very well structured and lacking clarity, this document contains an impressive amount of data on project's physical achievements and costs. It covered, unfortunately, the period from project start, until December 1986 only. The second useful document was the external audit report prepared jointly by a team of donors and government representatives in March 1988. This latter report was based on the earlier one but was better structured and more analytical and covered the same project period.  - 15 - PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE 67. Because of the way this project was designed, its implementation was dependent on action by a group of donors, who contributed to it in varying degrees. Donors 68. West African Development Bank (BOAD). This institution's contribution to the project was strictly limited to the component it financed, which still awaits final evaluation. The fact that technical design work had not taken full account of local conditions meant that major difficulties were encountered, with a consequent adverse impact on the execution process. The rigid nature of BOAD procedures made it impossible to introduce the necessary rectifications. The terms of the Agreement were scrupulously observed by BOAD. 69. World Bank. As lead project donor, the World Bank played its role fully, by monitoring both the technical and financial sides of the project at regular intervals. The Bank's advice on management matters in conjunction with annual audits by Cooper and Lybrand, facilitated sound project management. Although the overall Bank contribution constituted significant support for the project, we wish to note that: - Washington took too long to approve subproject proposals, thereby often jeopardizing their execution; - technical supervision of physical works rarely went beyond observations and/or criticisms, to the formulation of specific recommendations. - Bank's financial documentation and recommendations were produced only in English, making it difficult for project accounting personnel to take appropriate actions. 70. Netherlands. The major financial contribution from the Netherlands, in subsidy form, facilitated enormously project implementation. Backed up by technical assistance in the socioeconomic arena and financial monitoring of project expenditures, this contribution was a significant element in the project results achieved. The financing plan devised by Netherlands technical assistance personnel, and its acceptance by all the other project donors, facilitated the utilization of funds. However, we would note that the Netherlands took no part in the monitoring and technical supervision of project works. 71. UNDP/FAO. UNDP/FAO technical assistance has been an important element in FDR projects, and particularly FDR III. Through it, the projects had access to the services of highly qualified experts with wide experience of development projects, and also to assistance for the directorate-general (in the form of contributions of office equipment and means of transport). 72. Government. Apart from the assignment to FEER of necessary local project management personnel, the Government's other commitments, namely release of counterpart funds and gradual assumption of responsibility for field agents, were met only partially. 73. Consultants/Experts. The full-time experts assigned by UNDP/FAO and the Netherlands provided very important technical support with the execution process, given their numerical strength and the high quality of the services they rendered. In addition, they contributed to some degree to the personnel training effort. However, the presence of such a large group of expatriate experts somewhat stifled the initiative of Burkina professionals, while the failure to designate local counterparts, undermined the anticipated transfer of knowledge and skills.  - 16 - 74. The part played by BOAD and the World Bank in monitoring project execution was limited to the dispatch of consultative missions. There was evidence of failure to coordinate these missions, the usefulness and effectiveness of which were not always apparent. The confusion in the terms of reference of the BOAD consultant made his services of almost no value. 75. National Agencies (ORDs, technical services). In its role as funding agency, FEER had to rely, for the execution of the project, on national agencies whose intervention capacity was limited. FDR III had to compete not only with other projects which those agencies were already executing, but also with their own programs of activities. In addition, they did not always fully understand of the role they were expected to play for executing project components, and this had an adverse impact on project results. 76. High turnover among local staff was an added difficulty for both FEER itself, and the other agencies, a situation not conducive to the stability and continuity necessary for sound project execution. 77. It must be recognized however that, without the participation and availability of these agencies, it would have been impossible to carry out the project. Their contribution materialized despite the numerous constraints under which they operated at the time and continue to operate now. 78. We wish to draw attention to the need for and importance of an agency such as HER with responsibility for technical design and supervision, since experience has shown it to be essential to the successful development of programs and their execution on a sound technical and scientific basis. 79. Beneficiary population groups. Community participation was present in the execution of project components. It must be emphasized however, that this participation, which was expected at all stages of the project, from inception to completion, was limited in fact essentially just to the execution stage. This is clearly insufficient to make the project investment operations sustainable. 80. These various lessons, learned from the experience provided by FDR III, were taken into account in the design and preparation of the five-year FEER I project, which is now in the course of execution.  - 1 7 - PART 111. STATISTICAL INFORMATION 1. Related Bank Loans and/or Credits Credit Purpose Year of Title Approval Status Cr. 225 To develop cotton production 12/30/70 closed West Volta Cotton Project Cr. 317 To raise rural standards of living 06/26/72 closed Rural Development Fund and improve income and health Cr. 442 To assist people in the drought- 12/07/73 closed Drought Relief affected areas redevelop and improve their farms and herds Cr. 496 To provide extension and other 07/19/74 closed Bougouriba Agric. Devt. support services, and improve water supplies, roads, etc. Cr. 557 To construct a livestock market 06/18/75 closed Livestock D;vt. Project and holding ground, and provide for training, field trials, project evaluation and monitoring, and future project preparation Cr. 640 To provide a line of credit to the FDR 06/21/76 closed Second Rural Devt. for investment in small-scale rural projects, Fund Project in order to increase crop production and improve village vater supplies Cr. 706 To provide production inputs, storage 05/23/77 closed West Volta Agric. Devt. and processing facilities, infrastructure and support services to rural communities and promote Government's role Cr. 982 To provide comprehcnsive program to 04/07/80 closed Forestry Project strengthen the operational capacity of the Forestry Administration with staff and equipment to carry out its assigned functions of protecting, developing and exploring the country's forest resources within the framework of the National Forestry Plan. Cr 1013 A pilot phase of a possible three 05/08/80 closed Niena Dionkele rice phase program to convert swampland to Devt. Project more productive use  - 18 - Credit Purpose Year of Title Approval Status Cr. 1097 To consolidate the experience 03/06/81 closed Bougouriba of the first Bougouriba Agricultural Development Project Cr. 1284 To raise domestic food production, 08/26/82 closed Volta Noire Agric. Devt. increase cotton exports and improve the quality of rural life Cr. 1293 To identify & put into place 09/21/82 closed Koudougou Pilot agricultural system capable of Agricultural preventing or stopping soil degradation Cr. 1550 To make possible more widespread and 02/26/85 under SPN Fertilizer Project efficient use of fertilizers and lay a sound basis for increasing cereal production Cr. 1896 Strengthen research capacity by 04/19/88 under SPN Agricultural Research improving research/extension linka,es Cr. 1979 Improve agricultural support 01/17/89 under SPN Agricultural Services services to farmers & hearders; improve functional literacy Cr. 2229 Improve natural resource 04/25/91 under SPN Environmental management at community level Management throughout the country  - 19 - 2. Proiect Timetable Original Revised Actual Phases Date Date Date Appraisal Mission 03/11/1981 03/16/1 981 Credit Negotiation 12/16/1981 12/16/1981 Board Approval 03/30/1981 03/30/1982 Credit Signature 10/28/1 982 10/2811982 Credit Effectiveness 05/04/1983 05/04/1983 Credit Completion 12/30/1986 06/30/1 988 12/31/1989 Credit Closing 06/30/1987 12/30/1988 04/30/1990  - 20 - 3. Disbursement of Credit CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS FY Qtr Cum. Estimate Actual Actual Cum. Actual Actual Cum. Act/Est US$ SDR SDR US$ US$ 82 1 2 3 350 4 700 83 1 1,400 2 2,100 640,000.00 640,000.00 693,472.00 693,472.00 33% 3 2,800 158,181.11 798,181.11 165,761.15 859,233.15 4 3,500 632,235.00 1,430,416.11 661,188.11 1,520,421.26 43% 84 1 4,300 263,506.39 1,693,922.50 275,461.24 1,795,882.50 2 5,100 188,974.83 1,882,897.33 196,435.66 1,992,318.16 39% 3 5,900 552,505.21 2,779,537.15 556,403.43 2,888,227.57 4 6,700 344,134.61 2,435,402.54 339,505.98 2,331,824.14 35% 85 1 7,700 28,659.46 2,808,196.61 27,942.77 2,916,170.34 2 8,700 1536,557.16 3,344,753.77 530,593.60 3,446,763.94 40% 3 9,700 0.00 3,344,753.77 0.00 3,446,763.94 4 10,700 0.00 3,344,753.77 0.00 3,446,763.94 32% 86 1 11,700 452,291.11 3,797,044.88 493,730.02 3,940,493.96 2 12,700 0.00 3,797,044.88 0.00 3,940,493.96 31% 3 13,400 457,011.98 4,254,056.86 534,234.85 4,474,728.81 4 14,100 757,512.14 5,011,569.00 910,602.31 5,385,331.12 38% 87 1 14,800 444,543.91 5,456,112.91 563,088.50 5,948,419.62 2 15,500 1,126,372.50 6,582,485.41 1,458,939.80 7,407,359.52 48% 3 15,800 674,065.29 7,256,550.07 873,035.17 8,280,394.69 4 16,000 861,118.90 8,117,669.60 1,148,646.70 9,429,041.39 59% 88 1 1,041,818.58 9,159,488.18 1,432,844.40 10,861,885.79 2 566,237.09 9,725,725.27 773,867.54 11,635,753.33 73% 3 736,126.98 10,461,852.25 954,268.52 12,590,021.85 4 955,282.59 11,417,134.84 1,295,794.63 13,885,816.49 87% 89 1 764,946.9412,182,081.78 993,41 5.91 14,879,232.40 2 569,696.49 12,751,778.27 708,284.15 15,587,516.55 97% 3 467,422.62 13,219,200.89 579,344.31 16,166,860.86 4 260,057.86 13,479,258.75 343,566.08 16,510,426.94 103% 90 1 106,709,09 13,585,967.84 141,465.31 16,651,892.25 2 14,032.17 13,600,000.01 18,232.43 16,670,124.67 104% Discrepancy between Credit dollar amounts is due to SORiUSS exchange rate fluctuations Credit closed on Dec 31, 1989. Final disbursement made on April 20. 1990. Original Credit amount SDRs = 1 3.600.000 Amount actually disbursed SDRs = 13.600,000.01 Actual/Estimated SDRs 1 Amount canceled SDRs = 0  - 21 - 4. Proiect Imolementation Appraisal Achievements Achievements Indicators Units Estimates as per 1 2,3 1/86 as per 12/31/89 (Volumes) (Volumes) I%) (Volumes) (%) a) Land Development Anti Erosion Measures ha 20,000 35,809 179 58,251 291 Management of Bottomlands - simple Bunding ha 640 423 66 677 106 - improving existing developments ha 240 46 19 46 19 Q Small Scale Irrigation Works ha 160 23 14 23 14 Treatment of 'Ravines" 0 135 NA 135 NA b) Support to ORDO'CRPA Services Training - Extension Agents p.d.- 8,500 9,110 107 11,075 130 -Tractor Drivers p.d. 440 520 128 520 128 - Technicians Nbr 3 0 0 0 0 - Visit to farmers p.d. 6,000 7,318 122 7,318 122 - Workshop for farmers p.d. 1,200 2,172 181 2,172 181 - Literacy p.d. 0 50,112 NA 69,612 NA