Senegal - Small Rural Operations Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 14765 PERFORMANCE AUDIT REPORT SENEGAL SMALL RURAL OPERATIONS PROJECT (CREDIT 991-SE) JUNE 29, 1995 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Abbreviations BNDs National Development Bank CER Center for Rural Expansion (Centre d'expansion rurale) CRAs Capital Replacement Accounts DRF Drought Relief Fund GOS Government of Senegal IDA International Development Association MAH Ministry of Human Affairs MDR Ministry of Rural Development MFP Ministry of Forestry and Inland Fisheries MTPE Ministry of Public Works and Water OED Operations Evaluation Department ONCAD National Office for Cooperation and Development Assistance (Office national pour la cooperation et d'assistance au developpement) NGOs Non-government Organizations PAR Performance Audit Report PCR Project Completion Report PMU Project Management Unit PR President's Report RDA Regional Development Agencies SAED River Senegal Delta Management and Development Corporation (Societe d'amenagement et d'exploitation des terres du delta du fleuve Senegal) SAR Staff Appraisal Report SODESP Corporation for Livestock Development in the Sylvo-Pastoral Zone (Societe pour le developpement de l'elevage en la zone sylvo- pastorale) SODEVA Development and Agricultural Extension Corporation (Societe de developpement et de vulgarisation agricole) SOMIVAC Casamance Agricultural Development Corporation (Societe de mise en valeur agricole de la Casamance SRO Small Rural Operations SROP Small Rural Operations Project TSA Technical Service Agencies UA Units of Account FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General June 29, 1995 Operations Evaluation MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Senegal - Small Rural Operations Project (Credit 991-SE) Attached is the Performance Audit Report (PAR) on the Senegal-Small Rural Operations project (Credit 991-SE, approved in FY80) prepared by the Operations Evaluation Department. The project was conceived as a means of funding small sub-projects in rural areas with emphasis on productive activities, developing local capacities to implement such projects, increasing local participation in related decisions and strengthening the creditworthiness of farmers. The project's principal components were: construction, equipment and technical services for small irrigation perimeters growing rice, vegetables and bananas; provision of equipment and funding for a well construction and deepening brigade; and provision of a management structure with appropriate technical and administrative staff and logistical support. Thirty percent of the proposed funding was left unallocated for additional activities to be identified during implementation. Implementation was characterized by delays, and the project period was extended by four years. Institutional arrangements for sub-project implementation were complex and shared responsibilities were unclear. Many implementing agencies treated the sub-projects as unimportant activities. Technical performance of sub-projects was often indifferent, increasing costs to farmers e.g., for pumping. Several perimeters are now partially or completely abandoned, and yields are below projected levels. In consequence, and because of the many other similar donor funded 'free' perimeters, farmers were reluctant to collaborate in cost recovery, and little progress appears to have been made in strengthening farmers' creditworthiness. No additional sub-project activities were developed by the project and there is little evidence of local participation in subproject design. The sustainability of the project institutions and assets is rated as uncertain. The overall ERR is clearly less than the ten percent estimated by the PCR. Institutional development is rated as negligible and project outcome as unsatisfactory. The PCR, prepared in 1991, was more optimistic about both the institutional and physical results of the project, rating both as satisfactory. The principal lessons of the audit are that effective dialogue between farmers and project institutions requires relationships between the parties to be of a contractual rather than of a patron- client nature. The audit also confirms the need for a greater emphasis to be placed at appraisal on the balance between the social/institutional and the technical aspects of a project in order to more precisely determine the measures necessary to achieve a satisfactory project outcome. Commenting on the audit, the Borrower highlights the risks of rainfed farming and the role of risk insurance, information support and farmers' training in ensuring sustainability. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY Contents Preface........ 3 Basic Data Sheet.................................................... 5 Evaluation Sum m ary ................................................ 7 1. Introduction ................................................... 13 Background .................................................... 13 2. Project Design ................................................. 15 H istory ....... . ...... ... .. .. .. .. ........ ... ...... .. ... .. .. .. .. 15 Project Objectives and Description ...................................... 16 3. Implementation ................................................ 19 Context ...... ................................................. 19 Project Results ..... ............................................. 23 Achievements ................................................. 24 Sustainability ................................................. 26 Farmer Participation and Organization ................................ 28 Credit ...................................................... 29 Planning Capacity .............................................. 31 R atings .... .. . ..... .... . . .. .. ........ ... ... ... .. .. . ... ..... . 31 Compliance with covenants ........................................... 32 4. Issues and Lessons .............................................. 35 C onclusions.................................................... 35 Lessons....................................................... 38 Annex 1 - Comments from the Office of the Prime Minister ................... 41 Map - IBRD #14615R1 This report was prepared by John English (Task Manager), and A. Graham Lowe (Consultant) who audited the project in October, 1994. Constance Frye provided administrative support. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  3 Preface This is the Performance Audit Report (PAR) of the Small Rural Operations Project for which Credit 991 for $11 million was approved March 17, 1980. The Closing Date was September 20, 1988, three years later than planned. Final disbursement was on September 20, 1988 leaving $5,500 which was cancelled. The PAR is based on the Project Completion Report (PCR) prepared by the Agricultural Operations Division, Sahelian Department, Africa Regional Office (on the basis of the Borrower's detailed completion report), on the Staff Appraisal Report (SAR), President's Report (PR) and the legal documents, on study of project files, supervision reports and project documents, and on discussions and findings issuing from an OED mission which visited Senegal in October 1994, and met with officials of the Government of Senegal and the Implementing Agency in Dakar and in the field, and farmers affected by the project. The collaboration and assistance of these officials and farmers is gratefully acknowledged. The draft report was sent to the Borrower for comments. The comments received are attached as Annex 1.  5 Basic Data Sheet SMALL RURAL OPERATIONS PROJECT (CREDIT 991-SE) Key Project Data (in US$ million) Item Appraisal Actual Actual as % of Appraisal Estimate Estimate Total Project Costs 14.1 12.8 91 Credit Amount (US$ m) 11.0 11.0 100 Cumulative Estimated and Actual Disbursements (in USS millions) FY81 FY82 FY83 FY84 FY85 FY86 FY87 FY88 FY89 Appraisal Estimate 1.6 4.7 8.3 107 11.0 11.0 11.0 11.0 11.0 Actual 0.79 1.21 2.81 3.82 5.56 7.41 10.03 10.96 10.99 Actual as % of Appraisal 49 26 34 36 51 67 91 100 100 Final Disbursement September 20, 1988 Project Timetable Item Original Actual Appraisal 03/79 07/79 Negotiations 10/79 01/80 Board Approval 12/79 03/11/80 Credit Signing 03/80 04/02/80 Credit Effectiveness 06/80 09/10/80 Credit Completion 06/30/84 06/30/87 Credit Closing 12/31/84 09/20/88 Staff Inputs (in staff weeks) FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY86 FY87 FY88 FY89 Total Preparation 25.9 - - - - - - - - - - 25.9 Appraisal 6.9 35.2 - - - - - - - - - 42.1 Negotiation - - 3.1 - - - - - - - - 3.1 Supervision - 1.4 10.4 10.2 26.7 12.7 7.1 6.5 8.9 1.6 1.3 86.8 Other 1.5 - 0.1 - 0.1 - - 0.9 2.7 - 0.6 5.9 Total 34.3 39.7 10.5 10.2 26.8 12.7 7.1 7.4 11.6 1.6 1.9 143.8 6 Mission Data Project Cycle Month/ No. of Days Specialization Performance Rating Type of Year Persons in Represented Status b Trend PrOblems' Field Preparation 1 12/77 1 5 a Preparation 2 04/78 1 5 a Preparation 3 07/78 1 4 a Preparation 4 10/78 2 9 1,a Preparation 5 01/79 1 3 a Preparation 6 02/79 1 4 a Preparation 7 04-05/79 3 30 a(2),e Preparation 8 06/79 1 4 e Appraisal 06-07/79 5 30 e,a,i Post-appraisal 12/79 2 10 e(2) Supervision I 08/80 2 11 e(2) 2 1 0 Supervision 2 11/80 1 20 e 2 1 0 Supervision 3 03-04/81 1 17 e 2 1 T,O Supervision 4 11/81 1 23 e 2 2 OT Supervision 5 06/82 1 26 e 2 2 M,O Supervision 6' 03/83 4 21 e,a,i,f 3 2 M,T,O Supervision 7 07/83 1 10 e Supervision 8 10/83 2 8 a,i Supervision 9 12/83 2 7 a,i 2 1 M,O Supervision 10 06/84 2 14 a,i 2 3 M,O,T,F Supervision 11 10/84 I 10 a,i - - - Supervision 12 03/85 2 8 a,f 2 1 M,O,T,F Supervision 13 09-10/85 I 10 a 2 2 F,M Supervision 14 04/86 I 10 a 2 2 F,M Supervision 15 10/86 I 5 a 2 2 M,F Supervision 16 04/87 I 7 a 2 2 M,F a. a = agronomist: I livestock specialist; i = irrigation engineer; e = economist; f = financial analyst; o = operations assistant. b. I = no problems; 2 = moderate problems: 3 = serious problems. c. I = improving; 2 = stationary; 3 = worsening. d. T = technical; M = managerial: F = financial; 0 = other. e. Mid-term review. 7 Evaluation Summary Objectives 1. The Small Rural Operations Project (SROP) was conceived as a follow-up project to the Drought Relief Fund (DRF) which had financed rural water supplies, firebreak construction, fire-fighting equipment and livestock vaccination in the early 1970s. The DRF was considered successful and the GOS and the Bank hoped to build on the experience of the DRF by using a similar, "flexible financial procedure." 2. The principal objectives of the project were described by the SAR as: "to encourage local initiatives and thereby broaden popular participation in decision making on investments in the rural sector and to undertake only those small rural projects that are manifestly in the interests of the beneficiaries." In addition it was intended that small groups of producers would be rendered "creditworthy in the judgement of the local banking system", and that "the existing Senegalese capacity for identification and preparation of small rural projects" would be strengthened. Increased crop and animal production was considered a "necessary but incidental project objective." 3. The project was approved in 1980 and was to extend over four years. It included seven major components: construction, equipment and technical services for small irrigation perimeters for rice in the Senegal River valley for vegetables in Thies and northern Sine- Saloum Regions, and for bananas at S6dhiou in Casamance Region; equipment for beekeeping and processing of honey and wax at Bignona in Casamance Region; provision of 20 small fishing boats with outboard motors and ancillary equipment in Louga Region; provision of equipment for an open well brigade, construction of 50 new wells and deepening of 50 other wells in Sine Saloum Region; and provision of a management structure with the required technical and administrative staff and logistical support, including the capability for identifying, preparing, appraising and evaluating small rural projects. In addition 30 percent of proposed funding was set aside for activities to be identified during implementation. 4. The Project Management Unit (PMU) was to be situated in the Small Rural Projects Department in the Ministry of Human Affairs. This ministry had managed the Drought Relief Project, and many of the NGOs which might be involved in implementation had their primary links to Government with this ministry. It was agreed that the new project should emphasize directly productive activities and include more direct participation by project beneficiaries. The need for cost recovery was addressed by the provision for Capital Replacement Accounts (CRAs) for the amortization of capital assets. Implementation Experience 5. Project execution was typified by delays. By the project's mid-term review (June 1983) all subprojects had begun to bog down: the vegetable and banana perimeters were held up for redesign; beekeeping and fishing were held up by mistakes in equipment specifications; technical difficulties in the rice perimeters were also holding up land works. Initial lump sum 8 payments for the CRAs were "almost all tardy" due to lack of support from technical service agencies (TSAs). Furthermore, though implementation was said to have improved following the mid-term review, the closing date was extended twice covering four years. Much of the delay was primarily a result of the bureaucratic context and modus operandi and reflected a general lack of commitment on the part of the GOS. (a) The institutional environment was cumbersome and complex. Six categories of institution or organization, ranging from the ministerial level to the NGO, grass-roots level, had some involvement with the project, and institutional mandates were unclear and often in conflict. (b) Shared responsibilities were unclear. Most subprojects were implemented through three agencies: the project management unit (PMU), the government's technical service agencies (TSAs), and various producer groups. The responsibilities of each were not adequately spelled out in the appraisal report or legal contract documents. (c) The TSAs treated the subprojects as marginal activities. They frequently discharged responsibility for execution by placing low level technicians at the disposal of the project, whom the PMU was then called on to manage. The frequent efforts of the TSA staff to modify project requirements and the generally lackadaisical attitude to field supervision, indicate that management at the TSA level was, at best. poor. 6. These deficiencies inevitably had an impact on the attitudes and responses of the intended beneficiaries. Beneficiary groups, not surprisingly, resisted project conditions and responsibilities. In particular, the operation of the CRAs was a constant source of friction between the PMU and beneficiaries. Results 7. Rice and Vegetable Perimeters. While the initial targeted numbers were achieved, relations between the PMU and the various groups to be supported by these components were strained throughout. Difficulties in engineering and land works were evident from the start, with the TSA ignoring promises made to the farmers on land development, lot size and cultivation techniques. Farmers criticize the lack of consultation and supervision on the part of the project and the TSAs, and feel that they would be better placed to sign-off for the quality of work performed by the engineers and contractors. 8. Banana Perimeters. The banana perimeters of the Casamance constitute the project component with the greatest degree of success and are probably the primary reason for the Borrower's completion report's affirmation that the fundamental objectives of the project as a whole had been achieved, "with variable degrees of success." After the resolution of early problems with badly chosen pumping equipment, and failure of certain contractors to honor their obligations, the banana perimeters are profitable enough to maintain farmer enthusiasm, to stimulate farmer organization, to engender a marketing orientation in the farmers' perspective with attention being turned to export opportunities, to encourage to establishment of peripheral, rainfed banana plantations, and to stimulate demand from other farmers for similar investment opportunities. 9 9. Beekeeping and Fisheries. Results of both these components were poor because of inappropriate technologies pushed by the project and, in the latter case, by the fact that many of the 'groups' were fictitious and loans were not repaid. 10. Village Water Supply. Delays were caused by staff discontinuity, the slow delivery of construction material and equipment, and fuel shortages. Despite these problems all the planned wells were built and are generally satisfactory. 11. Unidentified Subprojects. Of the almost 300 requests received for financing under the unidentified facility of the project only 14 were approved. All of those approved were similar, and geographically close, to the project activities already under implementation. Four were for banana perimeters, one was for fishing and nine were for small vegetable perimeters. The objective of moving into new and innovative activities was not achieved. As the results have turned out, all subprojects were increased in size, cost more than foreseen and were developed by technical service agencies who applied barely understood "boilerplate" projects to more or less willing populations. 12. Planning Capacity. A major objective of the project was to "strengthen the existing Senegalese capacity for identification and preparation of small rural projects". While it impossible to compare the planning capacity before and after the project, assessment of the projects financed by the flexible "unidentified projects" facility and in the second phase shows very little innovation. No small pilot projects were attempted and no efforts were made to involve farmers in the definition of new projects (on the grounds that farmers cannot know what is possible). At the same time, the monitoring and evaluation component of the PMU has been notably weak and as a result there has been no acquisition of field condition information which could be used to improve the quality of the planning in the "mirror-image" projects. Sustainability 13. The sustainability of the project is judged to be uncertain. The majority of the rice perimeters did not perform well, because of design faults and 40 percent were rehabilitated by the follow-on project (SRO II). Relationships between many of the sub-project beneficiary groups and the TSAs have been poor, interest is limited and the Capital Replacement Accounts (CRAs) have not been successful as a means of accumulating capital to replace the assets financed by the project. The future commitment of the groups is uncertain. Conclusions 14. Overall Assessment. It is the judgement of the audit that the PCR assessment of the project outcome as satisfactory is not justified. While the narrow physical targets were achieved, this was partly at the expense of the funds available for additional subprojects not identified at appraisal. The objectives of the project were stated at appraisal as: "The main specific objectives are to render small groups of producers creditworthy in the eyes of the local banking system, and to strengthen the Senegalese capacity for identification and preparation of small rural projects. Increased crop and animal production is a necessary but incidental project objective." In addition it aimed to "encourage local initiatives and thereby broaden popular participation in decision making on investments in the rural sector." As 10 discussed at length in the audit, the project 'failed to achieve most of its relevant objectives', and was clearly not cost-effective in achieving project objectives. It was, therefore, rated as unsatisfactory on this basis. 15. Farmers' Groups. Farmers' groups have been criticized by project staff on the grounds that they did not understand the conditions and obligations of the project, but this fault may be laid more at the door of the TSA and the PMU. While portrayed as innocent or stubborn, the farmers have consistently acted to protect their own interest, effectively acting as "project traders", where the project benefits become the commodity, and future streams of benefits are calculated and negotiated with each project financier. This approach has been encouraged by the plethora of public and private 'projects' among which they can choose. 16. Institutional Arrangements. A Mid-Term review has been undertaken of SRO II. This was highly critical of project operations and made a number of recommendations. The principal of which was that the project entity should be removed from its position of being under ministerial 'tutelle' and reformulated as an autonomous institution with separate legal standing. Further proposals were made to tighten up financial and contract management and improve collaboration with beneficiaries. 17. The audit concurs with these recommendations. However, there are additional issues and implications which need to be addressed. It is not clear that 'privatizing' the PMU would, by itself, improve the overall effectiveness of the project. In particular, the TSAs have not been effective for field supervision of works. This function will never be a high priority for them, and at times conflicts with their primary responsibilities. The audit concludes that it would be preferable to use private agents, under contractual arrangements, for this work. This could shift the balance towards satisfying the farmers, rather than meeting the target, as the contracted supervisor would have no overall target to achieve, but rather must report to the beneficiaries on the quality of work performed for them. 18. Such a change could, however, have implications for the sharing of responsibilities of this type of project. First, it would be necessary to use the contractual field supervisors only in the implementation of locally proven technologies, thus avoiding disincentives to responsibility taking by private agents inherent in unproven technologies. Second, the government services should assume the responsibility for testing and proving the technologies through the medium of pilot projects which allow for the establishment of local norms and standards, which will then be implemented by private contractors. Third, the private supervisor must not be designated as responsible for the effectiveness of the beneficiary groups. This will be the responsibility of the PMU or TSA who choose the groups. In so distributing the responsibilities, the project will mitigate difficulties attending the introduction of new technologies, or of assisting marginal groups, by making specific provision to absorb the risks involved. 19. Participation. While from the beginning of the SRO project there has been much discussion of the need to improve 'popular participation' in this type of project, time does not appear to have improved performance in this area. Indeed, the follow-on project, Small Rural Operations Project II (SRO II), has been characterized by the same difficulties in achieving participation. This situation in the project is all the more regrettable since the original objectives were to encourage local initiatives and to broaden popular participation in decision- 11 making on investments in the rural sector, and to undertake only those projects that were manifestly in the interest of the beneficiaries. These objectives have been largely ignored throughout the implementation of the project, and, until this is redressed, the results of the project will always be limited. 20. The social, institutional and technical conditions have changed considerably since the SRO projects were initiated 15 years ago. There is evidence that local leaderships have become more development oriented and that networks have developed, often involving migrants, which are potentially effective in raising capital for small projects which have generally perceived benefits. The audit recommends that a review be undertaken of the comparative performance of the different types of small 'projects' now operating in Senegal. This should include the SRO projects and also others promoted by NGOs, other external aid agencies, and those initiated and financed by local groups themselves (some based on remittances of migrants). This should include a review of the history of these groups (i.e. how and under what conditions they were initiated), their methods of operation (including how they operate as groups, and the procedures for managing any jointly held assets), organizational structure, and their results. This study would attempt to assess those factors linked to successful outcomes and whether, and under what conditions, a public agency such as SRO can be expected to be effective. Lessons 21. Major emphasis in the preparation of projects of this type has been placed on technical elements of design and enterprise profitability. However, there needs to be much better understanding of the sociallinstitutional aspects of the operation; i.e. the nature and origin of the groups supported by the different SRO project elements, and their potential to effectively manage the asset developed and to "stand on their own two feet." 22. It is generally agreed that farmers must participate in the identification and preparation of sub-projects if they are to effectively participate in implementation and bear subsequent responsibility. However, this will not be achieved unless the relationships between the farmers and public agencies are changed, so that they are contractual rather than patron-client. This will then require greater emphasis on the ability of the farmers (individually or in groups) to fulfill their contractual obligations. Where this ability is not clear, involvement of third parties (eg. NGOs) to provide necessary support will be required. 23. At the same time there will need to be greater clarity on the nature of the contractual obligations, and the ability to enforce them, including those of third parties such as contractors. Where this is not possible, for example, because of technological uncertainties, the effort should initially be piloted, with specific emphasis on determining the appropriate criteria for use as a basis for the contractual arrangements. 24. Complex, nationwide, compendium projects of this type are extremely difficult to supervise and control and, thus, are costly. The circumstances under which public agency efforts are likely to be effective are limited and, therefore, they need to be closely targeted and monitored.  13 1. Introduction Background 1.1 Senegal is situated at the western limit of the Sahelian zone of West Africa. The total land area is just under 197,000 km2, with a population of 6.7 million persons, of whom 64 percent are in rural areas, with a density of some 21 persons per km2. The country is essentially flat rising gradually from sea level to a little over 100 m at its highest point. The climate varies through the country ranging from sahelian in the North (rainfall at Podor: 330 mm) through soudano-sahelian in the centre of the country (rainfall at Kaolack: 800 mm) to subguinean in the South (rainfall at Ziguinchor: 1500 mm). With mean annual temperatures situated around 28oC, large areas of Senegal depend on subterranean resources for both potable and productive water. In years of drought these reserves become crucial. 1.2 Senegal has received influxes of various ethnic groups at different times in its history. European contact was established in the XVth century when Portuguese voyagers came ashore. However, French interest in the area grew through the XVII and XVIIth centuries until control was established in 1893 and Dakar became the capital of French West Africa. During the French colonial period Senegal became a major exporter of groundnuts, a crop which to this day plays a major role in the country's agricultural exports (70 percent). At the time of the project appraisal 60 percent of the rural population was judged to be concentrated within a 150 km radius of Dakar within the Central Groundnut Basin, where local population densities were seen to exceed 100 persons per km2. The groundnut production of Senegal is rainfed. 1.3 In recent times Senegal has suffered from two major periods of low rainfall which resulted in famine in 1972-73 and 1978. As a result of these difficult periods the Senegalese State is preoccupied with protecting the rural populations from future famine through the exploitation of underground water resources. While low rainfall has left the soils relatively unleached, the same phenomenon limits the productive capacity of rainfed soils to "low desirability" cereals (sorghum, millet) and groundnuts. Diversification of agricultural production has therefore been seen as tributary to the exploitation of the same underground water resources. 1.4 Government's strategies and policies for the agricultural sector at the time of project preparation were cited by the SAR as follows: a) socialize rural society; b) increase home- grown food and make output less dependent on rainfall; c) diversify exports; d) increase rural income and restrain urban-rural income differences; e) generate public savings from the rural sector. However different government strategies have been cited in initial project briefs: "i) increasing groundnut yields and millet production; ii) diversifying production by emphasizing cultivation of new crops and promoting livestock development in Eastern Senegal, and encouraging migration of farmers to this region from the heavily populated areas of the Groundnut Basin; and iii) seeking more secure drought relief by harnessing existing water resources which have only in recent years started being developed" and in the initial preparation report (see para. 2.2). This diversity of strategies indicates some confusion as to 14 which were actually being implemented by the government and possibly implies a lack of clear thinking and commitment on the part of the government. 1.5 Bank loan history was varied in Senegal and at the time of the project preparation, total sectoral lending had amounted to some US$74 million equivalent of which US$67 million in IDA credits had helped to finance twelve agricultural development operations. Three for irrigation in the Senegal River Delta, two to establish and implement an agricultural credit program, two for rice development in the Casamance, two resettlement programs in Eastern Senegal, the Sine Saloum development project for crop improvement and diversification in the Southern Groundnut Basin and a Drought relief program. 1.6 Project Audit Reports for these projects showed patchy results with deteriorating management effectiveness and consequent shortfalls in project benefits. Major shortcomings were identified as: i) Poor quality of engineering design and supervision of construction of project works; ii) scant attention to marketing, unremunerative price and poor input distribution; iii) the lack of thorough investigation of the social aspect affecting project outcomes. These weaknesses were to be addressed in the planning of the Small Rural Operations Project (SROP). 15 2. Project Design History 2.1 The SROP was conceived as a follow-up project to the Drought Relief Fund (DRF) facility which, from the years 1970 to 1975, had financed rural water supplies, firebreak construction, fire-fighting equipment and livestock vaccination. The DRF was considered successful and the GOS and the Bank hoped to build on the experience of the DRF by using a similar, "flexible financial procedure." It was agreed that the new project should emphasize directly productive activities and include more direct participation by project beneficiaries. 2.2 After considerable delay and discussion with the borrower on institutional aspects and consultant profiles, project preparation was started in mid-1978. Under the first proposals, project objectives were said, again, to reflect those of the government, i.e.: (a) raising income of the poorer farm families through introduction of high value crop and livestock enterprises; (b) improving village nutrition through making available increased quantities of vegetables and animal protein; (c) raising rural labor productivity and creating employment for the young; (d) building up the development capacity of the Communaut6s rurales,' and (e) improving water and fuel wood supplies. 2.3 Project components detailed in the initial proposal were: irrigated rice production*; market gardening*; reforestation; village poultry units; pastoral wells; beekeeping*; fishing*; village water supplies*; flexible funding mechanism*; banana plantation*; pig production; sheep fattening; technical assistance.2 2.4 The principal issues in preparation were (a) the central organization and management structure, (b) responsibility for the field execution of subprojects, and (c) means of ensuring adequate cost recovery as an aid to replicability. The Drought Relief Project had been managed by the Ministry of Social Development and many of the NGOs which might be involved in implementation had their primary links to Government with this ministry. Most of the government agencies likely to be involved in implementation, on the other hand, were under the control of the Ministry of Rural Development. However, it was felt that if the 1. The Communautis rurales were set up in February 1972 as a result of the Administrative Reform. Rural community councils (Communaut6s rurales) were to be the grass roots unit of a development organization which was to receive technical guidance from rural development services in agriculture, livestock management, health, water supply, forestry, education etc. The Communautds rurales were assigned some 75 percent of per capita rural taxes. 2. The asterisk (*) denotes elements included in the implemented project 16 project were lodged in that ministry it would be engulfed, and that a degree of independence was desirable. The need for cost recovery was addressed by the provision for Capital Replacement Accounts (CRAs) (which were effectively sinking funds) for the amortization of capital assets. Project Objectives and Description 2.5 The objectives of the project were described by the SAR as: "... to encourage local initiatives and thereby broaden popular participation in decision making on investments in the rural sector and to undertake only those small rural projects that are manifestly in the interests of the beneficiaries. ... The main specific objectives are to render small groups of producers creditworthy in the judgement of the local banking system, and to strengthen the existing Senegalese capacity for identification and preparation of small rural projects. Increased crop and animal production is a necessary but incidental project objective."' 2.6 The project was to span some four years and consisted of eight major components: (a) construction, equipment and technical services for 30 perimeters of 20 ha each for irrigation of rice in the Ngallenka valley in Fleuve Region; (b) construction, equipment and technical services for 18 perimeters of 2 ha each for irrigation of vegetables in Thies and northern Sine-Saloum Regions; (c) construction, equipment and technical services for 15 perimeters of 4 ha each for irrigation of bananas at S6dhiou in Casamance Region; (d) construction, equipment and technical services for beekeeping and processing of honey and wax for four groups of beekeepers at Bignona in Casamance Region; (e) provision of 20 small fishing boats with outboard motors and ancillary equipment in Louga Region; (f) provision of equipment for an open well brigade, construction of 50 new wells and deepening of 50 other wells in Sine Saloum Region; (g) unidentified activities; (h) provision of a management structure with the required technical and administrative staff and logistical support, including the capability for identifying, preparing, appraising and evaluating small rural projects. 3. The Borrower's comments suggest that it would have been helpful if the project had made greater and more explicit provision for training and information for farmer groups (see Annex 1). 17 This Project Management Unit (PMU) was to be situated in the Small Rural Projects Department in the Ministry of Human Affairs. 2.7 The productive elements of the project (a to e) were to benefit some 2,400 families having incomes below the relative poverty line for Senegal of US$120. With the inclusion of estimated families served through the unidentified project facility, the total target population was to be around 4,200 families. The water supply element of the project was to benefit some 10,000 families (or 100,000 people). 2.8 During the review process the principal issues raised related to the need for prefinancing of operations and it was agreed to establish a revolving fund to address this. Continued concerns were expressed over the effectiveness of the provision for credit recovery or capital replacement, and the replicability of the proposed mechanism. A credit of US$11 million was approved on March 10, 1980, with cofinancing equivalent to US$400,000 from FAC to cover technical assistance.  19 3. Implementation Context 3.1 The project was typified by delays. The extended preparation period has already been noted. By the time of the project's mid-term review (June 1983) all subprojects had begun to bog down: the vegetable and banana perimeters were held up for redesign; beekeeping and fishing were held up by mistakes in material specifications; technical difficulties in the rice perimeters were also holding up land works. Initial lump sum payments for establishment of the CRAS were "almost all tardy" due to lack of support from technical service agencies. Furthermore, though implementation was said to have improved following the mid-term review, the closing date was extended twice covering four years. Much of the delay was the result of the bureaucratic context and modus operandi and of the GOS lack of commitment to the project and its objectives. This inevitably had an impact on the attitudes and responses of the intended beneficiaries. A cumbersome, diverse and confusing institutional context. 3.2 The project was conceived and executed in a complex environment where institutional mandates were unclear and often in conflict. Six categories of institution or organization, ranging from the ministerial level to the NGO, grass-roots level, had some involvement with the project: a) Policy making and technical institutions: Several ministries could claim responsibility for strategy definition and policy decisions within the rural sector touched by the project. However, while an earlier effort at decentralization had devolved executing responsibility to regional development agencies (see below), both policy and execution were defacto covered by both the institutions (ministry and RDA). The ministries involved included Human Affairs (MAH), Rural Development (MDR), Public Works and Water (MTPE) and Forestry and Inland Fisheries (MFP). b) Regional development agencies: Ostensibly, these parastatal agencies assumed responsibility for support services to crop production in each of their areas: SAED Senegal River Valley irrigation for rice & vegetables SODEFITEX south-eastern Senegal cotton marketing & extension SODESP north-center cattle marketing & extension SODEVA center, groundnut basin extension SOMIVAC Casamance TN south east land settlement The RDAs, constituted in large measure in response to foreign financing, had relatively well paid staff and good equipment, but had varying degrees of financial autonomy. Agency status varied from wholly government owned (SAED, SODESP, SOMIVAC, STN) to 20 mixed government-private enterprise (SODEFITEX, SODEVA). In addition to executing capacity, some RDAs (SAED, SODEVA, and SOMIVAC) had well-staffed planning units. c) Rural Development Centers (CER): The remnants of a nationwide field staff from different government departments (in particular Agriculture and Human Affairs) were grouped together into CERs at the local district level with shared office, warehouse and transport facilities, much of which was dilapidated. These centers were identified in the SAR as ineffective and underfunded but were expected to oversee activities in the vegetable perimeter! d) National Development Bank (BNDS) and the National Cooperative Organization (ONCAD): Mandated to assist development projects in rural activities (agriculture, fisheries, small industries, housing, handicrafts), and to provide inputs in kind to cooperatives, the BNDS effectively avoided small rural producers as high risk and high cost business. Prior Bank intervention led the SAR team to believe that the BNDS operations had been improved and rendered satisfactory. However, in 1989, the BNDS met its demise to be replaced by the CNCAs, which in 1994 was still avoiding small rural producers. ONCAD overall performance was dismissed as poor in the SAR. e) Local government: As a result of administrative reform in 1972, a nationwide structure was implemented with its base in Rural Community Councils or communautis rurales grouping approximately 10,000 persons and rising through District and Departmental Councils to the Regional Council. The project was originally planned to work through the communautis rurales causing them to be strengthened. Symptomatically, "Government considers that it would be premature for communautis rurales to ... guarantee loans made to small groups of community members"' and thereby hampered any effort to strengthen the CR. f) NGOs and agency projects: Project preparation was completed in a context of an increasingly large number of small rural projects financed by bilateral and international agencies, particularly by non-governmental organizations. This environment was one of duplication of effort and competition. Approaches differed particularly in respect of beneficiary participation in asset costs and the amount of technical assistance provided by organizations. This situation has not changed fifteen years later and is still compromising the results of the project's second phase. Shared responsibilities were unclear. 3.3 The subprojects were implemented through three agencies: the project management unit (PMU), the government technical services (RDA, CERs or even government departments) and various producer groups. The relationship between these three units was established by the means of contracts which were to cause difficulties between the signatories. 4. SAR. 21 (a) The PMU: 3.4 Its responsibilities were not detailed in the SAR, but were described in the PCR as follows: a) drawing up a contract for technical services with the relevant technical agency; b) drafting a contract between technical agencies and each producer group; c) acting, wherever appropriate, as procurement agent for the technical agency; d) supervising the project's physical and financial execution; and e) performing adequate monitoring and evaluation. 3.5 The PMU was also responsible for disseminating information on subproject selection criteria, terms and conditions of financing, and procedures to be followed and assisting in subproject preparation by providing guidelines and formal training. The PMU was required to appraise and report on each new subproject proposed to the Interministerial Committee. General administration included annual work plans, budgets, annual and semestrial progress reports, consolidated project accounts and status reports on each Capital Replacement Account (CRA) All reports were to be submitted simultaneously to the Interministerial Committee and to IDA. To carry out these responsibilities the SAR proposed a staff of Project manager, four specialists (of whom two would be expatriates), an accountant, three bookkeepers and support staff. The PMU staff ultimately forewent the services of expatriates and the bookkeepers. 3.6 The evaluation of the PMU by the PCR identified handicaps which were never truly rectified: i) a lack of planning and programming; ii) a lack of technical know-how and, consequently, ineffective supervision and control of subprojects; iii) too much concentration of power [in the hands of the PMU], even if by default; iv) a lack of understanding of the project's administrative and financial requirements; and v) an ad hoc decision-making process. Detailed proposals for PMU improvement were proposed by the mid-term review team which included the recruitment of international specialists for engineering and administration and financial management. These proposals were not adopted and Senegalese nationals were appointed to these positions. Difficulties with the management of the PMU have continued into the second phase of the project, and project activities are presently suspended due to financial irregularities. (b) The technical service agencies (TSA): 3.7 These responsibilities also were not specified in the SAR and were barely touched upon in the PCR. Covered by poorly defined contracts from the PMU, TSA responsibilities varied from direct management of first year plantation of the irrigated banana perimeters (Casamance) to minimal intervention in the Louga fisheries project. The TSA were expected to keep accounts for project-financed expenditures, to verify the CRA maintenance, and to furnish a monthly report to the Project Manager (PMU). 22 (c) Producer groups: 3.8 Their responsibilities were defined during project execution and depended to considerable degree on the defined responsibilities of the technical service agencies. In general however, the groups were called upon to have at least one bank account and maintain a de facto identity with an appropriate operational structure, and to: a) provide, on an unremunerated basis, all unskilled labor requirements during construction ("sweat equity"); b) collect cash deposits necessary for the installation of medium term assets (GMP) and contributions to the annuity (CRA); c) allocate land to group members; d) devise and apply rules for access to common goods and services; e) collect cash and/or provide labor for operation and maintenance of shared facilities. Mismanagement of the project at all levels discredited and undermined the objectives of the project. 3.9 The Borrower's report identifies management problems at several levels starting with the Inter-ministerial Committee which could not meet due to the priorities of the Minister for Social Development who was President of the committee. Further lack of Ministerial support can be implied from comments on difficulties experienced by the project with the Customs Office of the government (who managed to lose tax exoneration requests several times), with the National Commission of Administration Contracts (CNCA) (who countermanded certain contracts with suppliers and contractors while remaining exceedingly slow in their procedures), and with the Treasury (who took an inordinate length of time to pay suppliers and project personnel). All these problems indicate considerable lack of commitment on the part of the many responsible government agencies. 3.10 The management problems in the PMU have been mentioned already, however it is to be underlined that problems of mismanagement have been common to both the first and second phases and not restricted to phase 1. Indications are seen in the power accumulated by the PM, in complaints of lack of travel allowances and late payment, and in comments received in the field from several interlocutors (the non-management of contractors and suppliers is particularly indicative). Executing agencies treated the subprojects as marginal activities: 3.11 Implementation contracts (poorly crafted by the MU) between the TSA and the PMU were ignored, as TSAs avoided contracted responsibilities for execution and instead, placed low-level technicians at the disposal of the PMU/project. The latter was then designed to "manage" the subprojects, taking decisions about procurement, construction, obliged group creation and technical oversight of the works. In certain cases, subproject managers were found to be subverting the objectives of the project (eg. village water supply and fisheries) 23 through mismanagement and spurious groups, while other TSA teams were repeatedly criticized for not applying the most basic of project rules (not least, consultation with the populations!). The constant efforts of the TSA teams to modify the project requirements (labor, down payments), and their almost generalized lackadaisical attitude to field supervision, indicate that management at the TSA level was, at best, poor. Producer groups resisted change, conditionalities and project responsibilities 3.12 While the primary aim of the project was to "encourage local initiatives and ... broaden popular participation" the Completion Report of the Borrower' underlined repeated resistance from, and conflict with the rural populations of the subproject areas. Almost none of the subproject TSAs had constituted producer groups prior to the allocation of funding to the areas. Conflicts arose over the extraction of "free" labor for land preparation, and the injection of a) the 5 percent seed money and b) the contributions to the replacement annuities (para. 3.13). The constant sale of produce outside the group organization led to understatement of receipts. This was used as justification for non reimbursement of credit and/or annuities. Relations between the most farmers and the TSA/Project remain poor. Farmer-group officials interviewed in Ngallenka suggested to the audit team that the Bank should deal directly with them, without intermediaries. On the other hand, the relative success of the banana perimeters in the Casamance has favored a good relationship between the TSA officer and the farmer groups'. Project results 3.13 While the PCR states that project outcomes exceeded appraisal estimates, an evaluation of the implementation difficulties, management problems and cost overruns would yield a less optimistic appreciation of the project. Substantial difficulties were encountered due to poor, badly supervised engineering. Pricing difficulties continue to plague the farming community, complicating the estimation of economic returns to the project, biasing farmer decisions, and limiting response strategies which might allow the absorption of post-devaluation cost increases. Difficulties with fishing communities only serve to show how little social dynamics were taken into account in project implementation. All in all, only the banana perimeters of the Casamance show any sign of farmer motivation and organization.' The physical results of the subprojects are summarized in the table. 5. Completion Report (Credit 991-SE) - Small Rural Operations Project, April, 1988. 6. Farmers' comment "If the project were not here we would not be able to buy medicine or send our children to school ... at first we were very reticent, people thought we were mad to be in the project ... now we want to improve our quality and export." 7. Jammeh, Sidi, AF5AG "SENEGAL : Second Small Rural Operations Project (Cr. 1992-SE) Mid-Term Review" World Bank, Office Memorandum, April Ist, 1993. 24 Achievements Irrigated rice perimeters - Ngallenka (Senegal River Valley). (31 percent of actual project costs) 3.14 Of the 30 perimeters planned in the SAR, 28 were eventually constructed. Early problems in crop rotation on the original 20 ha were solved when perimeters were extended to 30 ha to allow non-rice crops on the extra 10 ha. Difficulties in engineering and land works were evident from the start with the SAED ignoring promises made to the farmers on land development, lot size and cultural techniques. In fact land works were so poorly executed that farmers continue to experience problems in more than half of them.' These perimeters had been constructed with excessively long, non-compacted, primary canals which significantly increased the perimeter demand for water and induced farmers to run the pumps almost around the clock. This in turn reduced the useful life of the pumps from four to three years. Farmers in the zone bemoan the lack of consultation and supervision on the part of the project and the executing agency, and feel that they would be better placed to sign-off for the quality of work performed by the engineers and contractors. Farmers' criticisms of the pumps chosen by the project (too high in fuel consumption, expensive to replace) are similarly dismissed by project officers. Irrigated vegetable perimeters - Thies/Sine Saloum (Central Groundnut Basin). (18 percent of actual project costs) 3.15 The original design for this component was based on "modest additional facilities" to perimeters of 18 groups on areas of approximately 2-3 ha. By the end of the project the "modest" additional facilities had become major tube wells many meters deep, the average area of the group perimeters had been increased to 5-6 ha and the number of groups supported was reduced to 12. The irrigation systems consist of pumping water to single large reservoirs or holding tanks which then feed small field tanks of approximately I m'. Water is then taken and applied to the farmers' plots using watering cans. Other projects in the locality use a sprinkler system and are considered by the farmers to be much more effective and are indeed less labor intensive. 3.16 By September 24, 1994 a PMU report showed activity in only four of the first phase producer groups (25 percent), with one group's perimeter lying "fallow"(!) and another perimeter having been sown to millet. Visits in the region in early October (during the rainy season) found farmers using hardly half their perimeter areas and citing difficulties in organizing contributions for the seasons purchase of fuel oil for the pumps, in controlling out- of-group parallel sales, and in generally motivating group activity. The relations between the PMU and the various groups to be supported by this component were strained from the start. Pumps were confiscated from three groups and CRA contributions were never readily forthcoming. Technical assistance was planned to be provided by the local CER, which had suffered from a lack of resources and mobility even at appraisal. 8. Twelve were selected in the second phase as being badly in need of rehabilitation. 25 Irrigated banana perimeters at Sidhiou (Casamance). (20 percent of actual project costs) 3.17 The banana perimeters of the Casamance constitute the project component with the greatest degree of success and are probably the primary reason allowing the Borrower's completion report to affirm that the fundamental objectives of the project had been achieved, with variable degrees of success. The PCR reports that an additional 7 percent of surface area was covered by the perimeter constructions over the 66.4 ha planned at appraisal. Despite initial reticence on the part of the farmers, difficulties with badly chosen pumping equipment and failure of certain contractors to honor their obligations, the farmers groups are the most dynamic and exciting of the sites visited by the audit team mission. After the resolution of earlier problems, the banana perimeters are profitable enough to maintain farmer enthusiasm, to stimulate farmer organization,' to engender a marketing orientation in the farmers' perspective (with attention being turned to export opportunities), to encourage to establishment of peripheral, rainfed banana plantations, and to stimulate demand from other farmers for similar investment opportunities. Beekeeping - Bignona (Casamance) (2 percent of actual project costs) 3.18 This component of the project suffered from a lack of communication with the beneficiary beekeepers by the appropriate government department. As a result the modern hives were not initially appreciated, CRA contributions were not provided, and appropriate "modern" techniques were not transmitted to the participating beekeepers. Of the 500 hives planned for distribution, all were placed, though by the time of preparation of the PCR only 476 operated, of which 209 (42 percent) were populated. There are no data on production. Fishing boats - Louga (1 percent of actual project costs) 3.19 The fisheries component of the project was, like the beekeeping component, overseen by the local government department. The subproject manager (SPM) allowed the distribution of boats (of inappropriate design) to fishing groups which were later shown to have a fictitious membership. Needless to say, contributions to the CRA were not forthcoming. A further complicating factor contributed to the poor showing of this component. Given the pressure for the project and the need for groups, the SPM arranged for the distribution of boats to fishing communities who were migratory, and who promptly migrated when it came time to pay into the CRA, or who declared that their outboard motors had been "lost at sea".'o Well construction - Sine Saloum (15 percent of actual project costs) 3.20 After initial delays all planned wells were constructed (50) and rehabilitated (50). The operation was severely disrupted by the resignation of the first sub-project head, which was followed by a lengthy search for a replacement. Further delays were caused by the slow 9. Groups within the project have come together in a regional association which meets regularly on a monthly basis (the audit team was fortunate to observe one of these meetings). Furthermore, the regional association has established a Federation of banana producers of Senegal with the banana farmers of the Tambacounda region. This Federation is able to supply some 80 percent of the countries needs in bananas. 10. Completion Report (Credit 991-SE) - Small Rural Operations Project, April, 1988. 26 delivery of construction material and equipment, and fuel shortages. Largely because of fuel problems, the inventory of wells was only 75% completed. Despite these problems all the planned wells were built and are generally satisfactory. A very limited number run dry between March and June in comparison with traditional wells, which almost all dry up. Projects unidentified at project preparation 3.21 Of the almost 300 requests received for financing under the unidentified facility of the project 14 were approved (5 percent), all similar and geographically close to the project activities already under implementation. Four were for banana perimeters, one was for fishing, and nine were for small vegetable perimeters. Some 1,200 families benefitted from the additional projects to bring the project total to 2,900 benefitting families out of a projected 4,200 (70 percent). Sustainability 3.22 Rice-pricing and the future. The major uncertainties in these subprojects relate to national pricing policy and the marketing system. Prices for rice are still fixed at official levels which have not risen to the same extent as the cost of imported inputs and goods. In 1992, a World Bank research paper on national and regional advantages in cereal and export crop production" showed that rice production in the Valley region was comparatively disadvantaged with respect to world prices for rice f.o.b. Dakar (Domestic resource cost-DRC: 1.87).12 Improvements to this competitive position will require management adjustments which the SAED has thus far been unable to introduce and which the farmers apparently resist." Marketing channels are still rudimentary as SAED has only stopped purchasing paddy, and processing and selling rice this year. The ultimate effectiveness of the merchant network will depend upon the improvement of comparative prices through the above mentioned management improvements. 3.23 Further squeezes are in the pipeline for the farmers as the government subsidies to the local tomato processing facility are phased out. Tomatoes are the primary dry season crop produced on the "polyculture" 10 ha. Comparative disadvantages were shown to be prohibitively high in the 1992 price study (DRC from 2.7 to 3.46 depending on farm management efficiencies (8)) and removal of subsidies may only end in closure of the facilities. Diversification has yet to be explored by the project farmers as they are still seeking to improve rice-related income through demands for better prices, cheaper credit and guaranteed marketing from the government. Despite repeated suggestion from project personnel, the farmers of the region have not succeeded in organizing a Farmers' Association. I1. Hollerman CF and Jones D. "Senegal's structure of protection and comparative advantage in cereal and export crop production" Agriculture Division AF5AG, Sahel Department, World Bank, 1992. 12. A DRC greater than unity, shows that the value of domestic resources used in production is greater than the value of foreign exchange earned or saved, i.e. no comparative advantage. 13. Derrick Knight "A Burning Hunger" 208 pages PANOS/Christian Aid 1994. 27 3.24 Vegetables - technology, manpower and off-farm income. The general disaffection of the farmers does not bode well for the sustainability of the project infrastructure: field-tank technology is onerous and demanding of manpower, the pull of nearby Dakar is considerable and takes away surplus manpower4 from a region considered at high risk from climatic variations, farmers recognize little benefits from input use and generate insufficient income to justify early seasonal contributions for fuel bills and pumping. "Crop production constraints are severe in this zone and households are clearly moving into other sectors of the economy." (IFPRI, p. 39) The primary issue here is why groups which utilize the perimeters have not been able to evolve to the point where they can obtain capital necessary to invest in the next step in improving their production. Are returns from vegetable production, or off farm incomes, too low to provide adequate cash flow to finance further improvement? Or are some of the perimeters located on such poor land that they are not worth improving? The facilities provided by the first phase of the project are quite obviously failing to respond to farmers' strategies and needs in the vegetable perimeters. 3.25 Bananas - excessive construction costs, market saturation, exports and quality control. A recent review" of engineering installations for phase two of the project concluded that "the banana perimeters of the Casamance region have been well designed and well constructed." However, the "construction of these perimeters has been extremely costly. The projected perimeter development cost was 5.4 million FCFA/ha (US$ 10,800/ha) ... actual realized costs have been on the order of 12.1 million FCFA/ha (US$ 24,200/ha)." The Mid-Term Review also stated that "It is also likely that these systems could be implemented/replicated for a cost considerably less than the real costs incurred." Given the nature of the soils in the perimeters visited by the audit team, it is appears that water losses through percolation are large, and greater surface areas could be irrigated if farmers constructed low cost channels to irrigate individual plants. This technology was not considered in phase I when irrigated bananas were unheard of in the zone. The prohibitively high costs of this component, as it was originally designed, may well compromise it's replicability. 3.26 As for the sustainability of the present producer groups and banana production the indications are positive. However, efforts in market orientation of the Producers' Federation must be focussed on quality control and possible export to neighboring countries. Present market domination (80 percent of the national market) is achieved through government protection from Ivory Coast imports (import quota regulation). If this protection is removed (liberalization) the producers must be able to compete with produce equal to or better that the bananas imported from Ivory Coast. The farmers visited were aware of this challenge and seemed motivated to improve the quality of their bananas. 14. International Food Policy Research Institute "Consumption and Supply Impacts of Agricultural Price Policies in the Peanut Basin and Senegal Oriental" USAID/Senegal August 1993. 15. Jammeh, Sidi, AF5AG "SENEGAL : Second Small Rural Operations Project (CR 1992-SE) Mid-Term Review" World Bank, Office Memorandum, April Ist, 1993. 28 Farmer Participation and Organization Farmers were excluded from project planning and implementation: 3.27 One of the clearest messages to come out of the project experience is that farmers who are not involved in project planning will not readily comply with project conditions, notably conditions which are designed to increase the level of commitment and participation of the farmer, without giving him/her an input to decisions (eg. "sweat equity" in land levelling; original 5 percent CRA contributions and subsequent annuity contributions). This approach created a context in which the farmer was a beneficiary to a project provided for and for him/her, but was in no sense a partner in the project. Farmers were told what to do (in the case of the beekeeping in Bignona this went as far as the definition of hives and their use, in Ngallenka they were told to level the land, in S6dhiou to clear the land) in order to receive the project investments. In the worst case scenario the farmers do not use the investment (vegetable perimeters in the Groundnut Basin) or begin to change the enterprise parameters (downgrading of pumps and modification of irrigation system design in Ngallenka). In these circumstances the first casualty in working subprojects is often the credit facility which is then given the same degree of respect by the farmers as they themselves feel they have received. 3.28 This situation in the project is all the more regrettable since the original objectives were to encourage local initiatives and to broaden popular participation in decision-making on investments in the rural sector, and to undertake only those projects that were manifestly in the interest of the beneficiaries. These objectives have been flagrantly ignored throughout the implementation of the project, and until this is changed, the results of the project will always be limited. This issue is highlighted by the experience in the Casamance where the technical service has assumed a service provider-client relationship with the farmer groups and with the Federation of Producers. The satisfaction on both sides is palpable. Farmer groups 3.29 Two cases in the SROP are interesting contrasts which have been underlined in many of the project documents: farmers' groups in Ngallenka, based on rice and tomato production, and the banana growers' groups in S6dhiou. It has been suggested that cultural differences account for the lack of association in the north and for the associative vigor of farmers in the south. However, analysis of conditions in the two regions shows that returns would be minimal in the north in comparison with the south. 3.30 Ngallenka: The farmers of the Senegal valley have been operating within a monopsonic situation selling their paddy to the SAED and their tomatoes to the local processing plant. While association might be expected to give them some advantage when negotiating with these buyers, it does not, since the price structures of both organizations have been established by government decree, in the first case to protect consumer prices and in the second case to protect the processing plant. Furthermore, the bank providing seasonal credit is a state organization, the organization distributing inputs and technical information is state administered, and other funding agencies (eg. NGOs) in the area are limited by the same pricing structures. Within these constraints, the farmers' best strategy is to complain about pricing (to the purchasing organization), to seek to minimize the costs of credit (by demanding reduced interest rates, repaying late or not repaying at all, or avoiding any real cash outlays to 29 CRAs) and to demand free or subsidized inputs. Association will not produce any better conditions, indeed "dissociation" is favored, and "parallel marketing" achieved through itinerant merchants. Given the lack of comparative advantage for the rice produced in the region (Hollerman and Jones, 1992) and it would seem that there will be no real farmers' association until this system is dismantled, as is planned by the government. 3.31 Sedhiou. The pricing and marketing context surrounding the banana farmers of the south are far more conducive to association. Prices for bananas are not regulated directly by the government, but are set by the market in Dakar. Competition comes from imports from Ivory Coast which are brought in through import licenses to a few traders. At present, imported bananas command a premium over local produce, but local prices remain profitable. Local bananas are sold through intermediaries retained by the producers' association to distributors in the capital. These same distributors may buy from the importing traders. Government intervention has been at the level of investment in irrigation systems and in the regulation of imports to make up shortfalls of production delivered by the Federation of Banana Producers of Senegal formed between the producers of Sedhiou and those from the region of Tambacounda. Parallel sales to itinerant merchants are rare, given the advantageous rates of the association, and the association is able to apply pressures to under-producing farmers to improve yields and eventually to improve quality. 3.32 Farmers' groups are criticized by project staff on the grounds that they do not understand the conditions and obligations of the project (a criticism which might apply more to the TSA and the PMU). However, the farmers have consistently acted to minimize their own costs of operations, minimize their risks of investment in the project, extend the conditions of credit, increase their own leverage and maximize their returns to investment. While portrayed as innocent or stubborn, the farmers actually act as "project traders" where the project benefits become the commodity, and future streams of benefits are calculated and negotiated with each project financier. It is not for nothing that the complex and competitive environment induced by the plethora of project funding agencies should be invoked as playing an important role in the motivation of the farmer groups. What is even more clear is the message on the reverse side of the coin, that civil servants do not make good farm managers. Whether it is in the choice of technology (beekeeping, fishing), the supervision of contractors and land development works (irrigation perimeters), or simply the presence in the field, the reports and documents show that the PMU and the agencies responsible for oversight were consistently deficient. It is imperative that mechanisms be found to allow farmers to assume responsibility for the work that is done on their behalf. This can only be done through a clear communication of project conditions and contractors' responsibilities and through the implication of farmers in the design of projects and land development, so that they are able to evaluate the quality of work performed, both by contractors and by technical service agents. Credit Innovative financial arrangements found little response from producers or TSA. 3.33 The major feather in the conceptual cap of the project as appraised was the setting up of the Capital Replacement Accounts (CRA), which sought to avoid problems experienced in Senegal in the recuperation of medium-term credit. In essence the CRA required the injection 30 of 5 percent of the value of the medium-term asset, into the producer group's bank account at the outset. This account was then to be maintained with additions coming from the production and sale of each year's production. This CRA was then to be used for the replacement of the GMP when needed (which was estimated to be after about four years). From the start of the project the innovation was rejected by the farmers' groups: a) rice perimeters. A battle of wills between farmers, SAED (TSA for the Senegal River Valley) and the PMU over the farmers' execution of field levelling and other land improvements, as well as the injection and maintenance of the CRA, resulted in an impasse which was only broken when the 1983 rainy season failed and the farmers had no production alternative. The problem was resolved through a bargaining process where there land works were paid for in return for CRA injections. Notwithstanding this, there were constant problems in maintaining the CRA and to this day the farmers' groups in Ngallenka are bitter about the conditions of the project; b) Vegetables. Resistance to the payment of the initial sum led to the suspension of construction works in Thibs. In 1985 the mean rate of deposit to the CRA in three producer groups was at 27 percent and in 1986 the pumps were removed from these groups; c) Bananas. It was necessary to apply quotas to the various villages to force them to clear the necessary land. In the end the down payments advanced by the farmers was taken as collateral for the negotiation of the land clearing quotas which were applied. Again, notwithstanding these agreements, the payment of contributions continued to be irregular; d) Beekeeping. The group at Bignona was particularly resistant to the required down payment when the TSA imposed its selection of hives on the group and the appropriateness of the "modern" hives was questioned by the beekeepers. The contributions to the CRA were very poor due to the failure of the technology to give good commercial results. e) Fisheries. There was no problem with down payments for the majority of the fishing groups since most of them were fictitious. Only group "Presidents" had paid the down payments in order to receive the material and then refused to pay the contributions to the CRA. This subproject dramatizes the poor supervision of the PMU and the lack of respect for the project objectives by the TSA in general. Resources have dried up: 3.34 The first of two main specific objectives of the project was "to render small groups of producers creditworthy in the judgement of the local banking system." As the project stands in 1994 the farmers would appear to be less creditworthy than they were at the start of the first phase. Changes in credit regulations, delays in liberalizing prices, and the lack of preparation for private marketing channels have squeezed farmers, who have yet to establish 16. Most frequently this was for diesel pumps ["groupe moto-pumpe" (GMP)]. 31 strategies which will allow them to absorb these impacts. Recent negotiations between the PMU and the CNCAS have seen credit conditions tightened with additional demands that: each group open a CNCAS account of at least 25,000 FCFA (US$100); charges of an initial 25,000 FCFA be levied on each dossier; and that personal downpayment be increased (15 to 20 percent, up from 10 percent). In addition, interest rates have been substantially raised (to around 17 percent from little or no interest), and imported input prices have increased, due the devaluation of the FCFA early this year. 3.35 The application of these conditions has reduced the total amount of credit received in the region of Ngallenka from around 22 million FCFA in 1993/94 to 7 million FCFA for 1994/95. Farmers in other subprojects have voiced concerns over similar conditions which represent enormous increases in a year when they are being called on to absorb some of the impacts of devaluation. Since the financial risks are not assumed by the CNCAS (a Guarantee Fund is to be put in place by IFAD which will cover defaults and costs only 2 percent per annum to the CNCAS), these interest rates also appear high. After fifteen years the situation attached to agricultural credit is worse today than at project appraisal, with farmers no more responsible toward credit institutions and less likely to use these formal credit facilities. Planning Capacity Identification and preparation of small rural projects is repetitive and lacks innovation. 3.36 The second specific objective of the project was to "strengthen the existing Senegalese capacity for identification and preparation of small rural projects." While it is impossible to compare the planning capacity before and after the project, assessment of the nature of projects financed by the flexible "unidentified projects" facility (both in the first and second phases) shows very little innovation. In general, these second generation projects are "similar to and geographically close to" earlier identified projects, notwithstanding the apparently limited impacts of the first phase activities. At the same time, the monitoring and evaluation component of the PMU has been notably weak and, as a result, there has been no acquisition of field condition information which could be used to improve the quality of the planning in such "mirror-image" projects. No small pilot projects have been attempted and no efforts have been made to involve farmers in the definition of new projects (on the grounds that farmers cannot know what is possible). As the results have turned out, all projects have been increased in size, have cost more than foreseen, and have been developed by technical service agencies who have applied barely understood boilerplate projects to more or less willing populations. Ratings 3.37 The PCR viewed the project as being satisfactory. It is the judgement of the audit that this assessment is not justified. The objectives of the project were stated at appraisal as: "The main specific objectives are to render small groups of producers creditworthy in the eyes of the local banking system, and to strengthen the Senegalese capacity for identification and preparation of small rural projects. Increased crop and animal production is a necessary but incidental project objective." In addition it aimed to "encourage local initiatives and thereby broaden popular participation in decision making on investments in the rural sector." As 32 indicated in this report, the project 'failed to achieve most of its relevant objectives', and was clearly not cost-effective in achieving project objectives. It was, therefore, rated as unsatisfactory on this basis. Institutional development was minimal and sustainability is uncertain. It is clear that Government commitment left a good deal to be desired, but a reviewer is left with the question as to whether the objectives for the project were realistic and, in particular, whether they are realistic today. This is discussed further in Chapter 4. 3.38 The difficulty of obtaining a realistic assessment of the rate of return for the project was discussed by the PCR. It's estimate of the economic rate of return for the directly productive subcomponents was 13 percent, compared with the appraisal estimate of 17 percent. The audit has not attempted a further calculation, because of the absence of hard data. However, it is the judgement of the audit that the PCR estimate is high, for a number of reasons: (a) the rice, and particularly the vegetable perimeters, are not fully utilized (para. 3.16); (b) the operational costs are higher than forecast. For example, pumps on the rice schemes have a life of three years or less, compared with four years assumed (para 3.14), and more pumping is required because of poor design of the canals; (c) production is below estimates. For example, the consensus figure given to the mission was three tons or less per hectare for rice, compared with four tons assumed. Therefore, it is clear the PCR estimate of 13 percent for the rate of return is too high and it is the judgment of the audit that the real figure is close to (or even below) 10 percent. This would mean that the estimate including attributable overhead costs (10 percent in the PCR) would be clearly less than 10 percent. Compliance with covenants 3.39 The PCR lists the credit agreement covenants and an assessment of compliance with those covenants which were all complied with or substantially complied with apart from covenants on Producer Group contributions to CRAs (Session number 3.04) and independent verification of accounts (Session number 3.10 (c)). The following comments were gleaned from the completion report of the Borrower and other documents consulted. 3.40 Maintenance of an Interininisterial committee: At the beginning of the project the management committee did not meet regularly due to the fact that the Minister for Social Development who was the President of the committee was often occupied by other obligations. 3.41 Quarterly deposits to the project account: The requests for transfer of counterpart Senegalese funds from the Treasury at the beginning of each quarter were processed so slowly that the Project personnel and creditor/suppliers were penalized. 3.42 Contractual relations with technical service agencies: Contract relations were not clearly specified and even in Phase II contracts with technical service agencies "lacked performance indicators, details on financial, material and technical responsibilities of all 33 parties, and [did] not specify the frequency and content of supervision and performance evaluation." (Mid-Term Review) These problems were surely inherited from the first phase. 3.43 CRA contributions: The CRA contributions were universally and consistently avoided, late or contested. The project experienced problems both in receiving the 5 percent personal contribution of the producer groups and subsequently in maintaining contributions in the form of annuities. Since this aspect was dependant on the response of the farmer groups to the project, it is dubious whether the CRA conditions should have been included in the covenants or governmental commitments.  35 4. Issues and Lessons Conclusions 4.1 The PCR noted three lessons from the performance of the project: (a) the scattering of heterogeneous subprojects over very disparate areas makes implementation difficult; (b) executing agencies for identified operations should be closely involved in the preparation and appraisal of subprojects for which they are later responsible; (c) attention must be paid to farmer training in management of collective assets and in making provision for renewal of medium-term assets. 4.2 A follow-on project, the Second Small Rural Operations Project (Credit 1992-SE) is currently under implementation. This project has been dogged by a continuation, and in some respects a deepening, of many of the problems noted in this audit, and the lessons have still not been fully incorporated. A mid-term review was undertaken in 1993. This noted that "the history of financial mismanagement, delay in implementation, intrusion of personal and political motives in project operations, and the unstable and dysfunctional environment clearly support the termination of the IDA and IFAD credits as a means of limiting investment losses for the government and people of Senegal. Despite the high interest of national authorities and the widespread support of project objectives by the population, the project as it is now structured will not be able to assure the responsible use of resources, nor will it be able to attain project objectives. The sole alternative to cancellation of the credit agreements is comprehensive institutional and organizational restructuring." The review concludes by making a series of recommendations geared towards such a restructuring. 4.3 The principal recommendations were that: - The project should be removed from its position under the 'tutelle' authority of the Ministry of Women, Children and the Family, and should be reformulated as an autonomous institution with private legal standing, under the general supervision of a Steering Committee with ministerial and private membership; - Accounting and auditing functions should be obtained under a contractual arrangement, and that a full audit of operations should be undertaken immediately; - Technical assistance should be sought to improve the contract management system; - All contracts with project beneficiaries should be negotiated rather than imposed, and should contain the responsibilities of each party and, as far as possible, should be in local languages as well as French. 36 4.4 The audit concurs with these recommendations. However, there are additional issues and implications which need to be addressed." It is the judgement of the audit that the second and third of the 'lessons' of the PCR do not get to the core of the problem faced by this project and others of this type." Involvement and training will not generate adequate performance if the overall structure of relationships and mutual obligations between the parties is not conducive to this end. Such relations may not be easy at the beginning of a project, but problems should have been identified and removed over the space of fifteen years. The poor performance and implementation experiences of the project, and its successor, indicate that there are institutional, social and policy problems related to this type of operation in Senegal which need to be better understood.9 These relate particularly to the effectiveness of groups through which the beneficiaries participate, and the nature of the relationships between the various parties. "Participation" and the effectiveness of groups 4.5 Over the past 15 years there has been a considerable development of small perimeters (particularly in the Senegal valley), with support from NGOs and others. The mid-term review of the Second SRO Project notes that, in the Senegal valley, the perimeters (PIVs) developed under the project effectively constitute a third phase of development in the valley. "The PIVs being presently constructed are often found at significant distances from the river. This phenomenon is largely a result of "saturation" of land areas immediately adjacent to the river by "first generation" PIVs." These necessitate conveyance of water over longer distances from the river and, thus, have higher capital costs. Thus, these areas will be economically less attractive than early developments and, assuming that the land irrigated first was of higher quality, less productive also. Thus, it might be hypothesized that these areas will only be of interest to those individuals or groups who have few alternatives. These are likely to be those who have not been able to fully participate in the earlier developments and the groups formed are likely to be less well focussed and/or tightly knit than those of the 'early adopters', who participated in the earlier projects or developed their own land. As such, the groups' ability to manage the asset is likely to be reduced. 4.6 There is also evidence that the available supply of 'improved land' exceeds the capacity of local households to effectively use it. The result is that farmers may sign up for more areas than they expect to use. They then make their decisions on what to do on their individual plots on the basis of the physical characteristics of each, the climatic and expected economic conditions in the season, and what sort of 'deal' they get from the scheme sponsors. The effect of this is that schemes may not be fully utilized, and that farmers resist obligations in respect of them (eg. to maintain CRAs, to undertake maintenance of canals, etc.), since the 17. The principal point made in the Borrower's comments is that a "Calamity Fund" to provide risk insurance is essential to ensure the sustainability of these types of small investments by farmers in regions with a high risk of adverse climatic conditions (see Annex 1). 18. Disbursements under SRO 11 are currently suspended and discussions are ongoing on memo to transfer responsibility for implementation to farmer groups. 19. Some of these issues were discussed by Elinor Ostrom "Investing in Capital, Institutions and Incentives." Paper presented at the Conference on Economic and Political Institutions for Sustainable Development, Washington, D.C., October, 1994. 37 'groups' formed to 'support' a scheme may be purely 'flags of convenience', and have a share of members with only a passing interest in the project. 4.7 While from the beginning of the SRO project there has been much discussion of the need to improve 'participation' in this type of project, time does not appear to have improved performance in this area. The experience of this project suggests that one of the difficulties is that very little is understood about the nature of the groupings through which the beneficiaries 'participate'. Project documentation at appraisal dealt at length with technical issues of the proposed sub-projects and institutional relationships, but had virtually nothing to say about the beneficiaries, either in general or in relation to the groups through which they were to participate. Relatively little is known about them. How have these groups been formed? Why (what are their major objectives)? Who are their leaders and do their major interests lie in the operation of the type of asset being created or elsewhere (eg. are they farmers, politicians, or religious leaders)? How did they select and obtain rights over land for the project? Did they have access to capital other than that obtainable under the project?20 Institutional arrangements 4.8 It is not clear that 'privatizing' the Project Management Unit (as proposed by the mid- Term Review of SRO II) would, by itself, improve the effectiveness of the project. The overall complexity and cumbersome nature of the institutional structure in which the project operated, has been noted (para 3.2).Under the present arrangement, a project of the SRO type inevitably results in a target orientation at the field level, with quantities of output, rather than quality of service, being the perceived major performance criterion. The pressure is on the TSA to implement target objectives, with limited sanctions applied if quality is lacking. The TSA performance was far from satisfactory (eg. group mobilization, technical support and supervision) and, in particular, the TSAs have not been effective for field supervision of the quality of land development works. This function will never be a high priority for them, and at times conflicts with their primary responsibilities. 4.9 The audit concludes that it would be preferable to use private agents, under contractual arrangements, for this work. Use of private agencies could shift the balance towards satisfying the farmers, rather than meeting targets, since the contracted supervisor would have no overall target to achieve (eg. 30 rice perimeters), but rather must report to beneficiaries on the quality of work performed for them. 4.10 The contractual approach would, therefore, have implications for the sharing of responsibilities of this type of project. First, it would be necessary to use the contractual field supervisors only in the implementation of locally proven technologies, thus avoiding disincentives to responsibility taking by private agents inherent in unproven technologies. Second, the government services should assume the responsibility for testing and proving the technologies through the medium of pilot projects which allow for the establishment of local norms and standards, which will then be implemented by private contractors. Third, the private supervisor must not be designated as responsible for the effectiveness of the beneficiary groups. This will be the responsibility of the PMU or TSA who choose the groups. In so 20. The Borrower's comments also suggest that a systematic audit be carried out of NGOs operating in rural areas. 38 distributing the responsibilities, the project will mitigate difficulties attending the introduction of new technologies, or of assisting marginal groups, by making specific provision to absorb the risks involved. 4.11 Such a change would, however, have implications for the primary objectives of this type of project which is seen as responding to a widely perceived need; the construction of small scale capital assets in low-income areas where other capital sources are deemed to be scarce. Social Investment Funds have been successful as means of financing the construction of small infrastructure, eg. in Honduras and Bolivia, using private contractual arrangements of this type. They have also permitted effective targeting towards poor and disadvantaged regions of these countries. However, they have involved the use of well known techniques and the subsequent operation of the clinic, school, etc. is usually assured by line agencies, so that the subsequent operation of the asset created is not dependent upon the management and technical skills of relatively poor and underprivileged groups to operate it. If a contractor is to ensure the satisfactory construction of a perimeter and be responsible for it's operating satisfactorily on completion, he will seek to ensure that the technology being used is well proven and the group for whom the perimeter is being constructed are experienced in the relevant skills and are well organized. This will mitigate against the use of this approach to support the introduction of new technologies or assist marginal groups, unless specific provisions are made to absorb the risks in project costs. Social conditions changed while project structure and approach did not: 4.12 It is the judgement of the audit that much more needs to be learned about the answers to these questions if the performance of SRO is to be improved. The social, institutional and technical conditions have changed considerably since these projects were initiated 15 years ago. There is evidence that local leaderships have become more development oriented and that networks have developed, often involving migrants, which are potentially effective in raising capital for small projects which have generally perceived benefits.2 It may be that such groups are effective at implementing projects of the type promoted by the Small Rural Operations Project and that external support may have limited benefits for them. 4.13 It is the recommendation of this audit that a review be undertaken of the different types of small 'projects' now operating in Senegal. This should include the SRO projects and also others promoted by NGOs, other external aid agencies, and those initiated and financed by local groups themselves (some based on remittances of migrants). This should include a review of the history of these groups (i.e. how and under what conditions they were initiated), their methods of operation (including how they operate as groups, and the procedures for managing any jointly held assets), organizational structure, and their results. This study would attempt to assess those factors linked to successful outcomes and whether, and under what conditions, a public agency such as SRO can be expected to effectively support local groups. 21. Derrick Knight, op. cit. 39 Lessons 4.14 Major emphasis in the preparation of projects of this type has been placed on the technical elements of design and enterprise profitability. However, there needs to be much better understanding of the nature of the groups being supported by the different elements of the SRO project and of their potential to effectively manage the asset developed and to "stand on their own fet. " Little appears to be known about the nature of these groups, how they operate and how they relate to other actors in the rural and agricultural scenes. Many of the groups may be marginally active. If so, it is not clear that additional state approved investment in this type of asset is justified. 4.15 It is generally agreed that farmers must participate in the identification and preparation of projects if they are to effectively participate in implementation and bear responsibility. However, this will not be achieved unless the relationships between the farmers and public agencies are changed so that they are contractual rather than patron-client. This will then require greater emphasis on the ability of farmers (individually or in groups) to fulfill their contractual obligations. Where this ability is not clear, involvement of third parties (eg. NGOs) to provide support will be required This lesson is not new, however it is repeated here as it is as yet extremely difficult for local technical services and international consultants to find methods which will allow farmer participation. Surveys, questionnaires and consultations do not fulfill the spirit of the word, neither does free labour constitute participation. Farmer can participate in project identification through problem identification, prioritization and solution assessment. Farmers can also propose solutions to questions such as "How can we best supervise the work of the contractors? How can we make sure that the extension officers do their job? How can you guarantee that credit advances will be repaid?" Finally, farmers can and should participate in the design of their land development: if engineers cannot explain their proposals to farmers then the works are too complicated, and if they do explain their designs to the farmers ... the maintenance and repair will be better assured. Technical planners must not be afraid of allowing farmers to ask the "Why?" of their plans, if farmers are dubious it would be better not to go ahead with the project, if they are convinced there will be less need for "sensitization". 4.16 At the same time there will need to be greater clarity on the nature of contractual obligations, and the ability to enforce them. including those of third parties such as contractors. Where this is not possible, for example because of technological uncertainties, the effort should be initially piloted, with specific emphasis on determining the appropriate criteria for use as a basis for the contractual arrangements. These examples indicate two quite different areas in which an SRO type project might operate. The first is the pioneering phase in which a technology involving the use of capital assets is being tested. Several of the components of the SRO fell into this category. If the tests prove successful, in most cases the technology is likely to 'take-off, supported by farmers themselves or by other groups better suited to management of small dispersed operations than a bureaucratic structure like SRO. Beyond this stage, an 'SRO' may find itself responding to groups who are unable to exploit opportunities themselves, i.e. unable to gain access to land, capital or whatever else is required to participate in use of the technology. Such groups may have significant operational limitations and require much 'nursing'. The 'financial' return to investments in them may be low, although there may be reasons to judge that the 'economic' benefits are higher. It appears that a significant share of the Second SRO Project has concentrated on this type of 40 operation. Whether this can be justified in the current Senegalese context may be questioned. Privatizing the operations of the project may also work against this emphasis unless specific incentives are provided to consultants etc. to service such 'clients'. 4.17 Complex, nationwide compendium projects are excessively difficult to supervise and control and, thus, are costly. The circumstances under which public agency efforts are likely to be effective are limited and, therefore, they need to be closely targeted and monitored It, therefore, follows that such projects will be more prone to abuse than localized, technician- intensive projects. In such cases project design should allow for technical as well as financial auditors who will be responsible for the monitoring and evaluation of the project. It is salutary that while farmer resistance to CRA contributions was mentioned in the various reports, difficulties with contractors and land developers and with technical service agencies were less apparent. The use of external, private monitoring agents, with sufficient budgets for in-field supervision, would allow governments to better protect the development investments of their projects. 41 Annex 1 REPUBLIC OF SENEGAL OFFICE OF THE PRIME MINISTER THE TECHNICAL ADVISER Dakar, May 17, 1995 Mr. Roger Slade Chief, Agriculture and Human Development Division Operations Evaluation Department World Bank Washington, D.C. Subject: Performance Audit Report on Small Rural Operations Project (Cr. 991-SE) Dear Mr. Slade; Thank you most sincerely for sending me the draft version of the Performance Audit Report on the Small Rural Operations Project financed by Credit 991-SE of US$11 million approved on March 17, 1980. After particularly attentive study of this draft report I would like to make the following comments and suggestions: 1) On page 20, referring to the background to the project, it should be noted that the Drought Relief Fund was set up to finance the basic infrastructure (water points, firebreaks, livestock vaccination facilities, etc.) necessary for countering drought on a sustainable basis. The success of this undertaking created the minimum essential conditions capable of favoring execution of small projects aimed at further improving the living standards of the rural populations. This background information is absolutely necessary as basis for an exhaustive and objective evaluation of the implementation of the Small Rural Operations Project. 2) On page 22, the description and definition of the project objectives provide confirmation of the design approach required for this new operation aimed at: - arousing and encouraging private enterprise in rural areas, and - promoting financing arrangements suited to agricultural and rural activities. Annex 1 42 3) The design of the project failed to include establishment and financing of a Disaster Fund to provide coverage for all the small operations envisaged. A fund of this type is needed because of ..........* on the one hand, and the considerable risks associated with the agriculture sector (rainfed crops for the most part) in a sahelian country (climatic uncertainties) on the other. ** 4) Another significant omission was the failure to include an "Information and Training" component for the farmers whom this small operations project sought to encourage to develop local initiatives. There was a need to assist them in the management of their undertakings and the use of their resources by means of technology-transfer sessions conducted by real professionals. 5) The effect of these "original" shortcomings was felt throughout the implementation of the project, causing the writers of the report to note (page 28) that "shared responsibilities were unclear", (page 31) "mismanagement of the project at all levels discredited and undermined the objectives of the project", and (page 33) "producer groups resisted change, conditionalities and project responsibilities". 6) A detailed analysis of the results of the project is presented on pp. 34-52 and the writers observe that "the project failed to achieve most of its relevant objectives and was clearly not cost-effective in achieving project objectives. It is, therefore, rated as unsatisfactory. Institutional development was minimal and sustainability is uncertain." I share this assessment to the effect that "sustainability is uncertain". The fact is that lasting positive results are unattainable in agricultural development without a fully operational disaster fund. Conclusion: In conclusion, I would like to offer the following suggestions: - make a detailed and thorough audit of the physical and financial accomplishments of all the nongovernmental organizations (NGOs) operating in rural areas in order to develop a reliable data bank on the professional capabilities and expertise of the cooperation and development assistance organizations; - encourage and contribute significantly to the creation of a Disaster Fund, either at the level of the West African Economic and Monetary Union (WAMU) or else at individual country level, as a support measure for sustainable development. This fund could be Phrase or line apparently omitted in French original (Tr.) Noted in para. 4.4. 43 Annex 1 financed initially by means of specific resources mobilized in the context of the conventions on climate change, desertification and biological diversity. Implementation of this suggestion is absolutely vital for the sahelian countries in their efforts to achieve social development, rational natural resource management and promotion of the private sector in their respective economies; systematically incorporate an information and training component in agricultural development projects, in the same way as is done for monitoring and evaluation, in order to develop and consolidate the professional capabilities of the parties involved in the projects. If effectively carried out, this component will induce sustainability of the social and economic impacts of the operations implemented. Reiterating my availability to act in conjunction with the Bank in the interests of the Senegalese people, I remain, Yours sincerely, s/Alioune Gassama  г о � � ш и а а �  IBRD 14615R1 SEN EGAL M A U R I T A N" I A w-)os rs.; by SMALL RURAL PROJECTS Zz Ronso ~our,In,a SMALL RICE PERIMETERS SMALL VEGETABLE PERIMETERS oThlleobcar I SMALL BANANA PERIMETERS R,co d Ter SIENBGA Lrkma Beh. Bo,. k- C) BEEKEEPING Be ho G -Fso" FISHING BOATS - N ger a VILLAGE WATER SUPPLY BASE ,Sero Leone e n - ~~A t / a r POPULATION DENSITY Km2 Eq . o n j L Go -cs 5 - 20 ®.] O,rosso '" 2 - 50 C.k, Co Tiou Ngourone Dohr Unguere Mbr-u-MrMen L O U G AF,1 VIb-u e kh. Daro Sci-. Ibahhby mboro -rane D.,ruMosy M'boyack T o.a... " B.ab n G 9.4 rogo" n ODo,ro. Bogo NMoyé's M.bo Ngor ý mbole NDoo Mb --- DAKAR Ruf,S C) O aT RPM~~iAI., VHRoai rl.l TIbe . .r CAPS PongTaif TIkene "* SnleD'akh.oGos. Doro, Kid,r. 'Bour MNdeua.e-- K,", 0 To,bere B.f.a PAVED RADS ck0D rkos Dlb ---E-GRAVE I ROADS GoNhR- - --EARTH ROADS Fou,,nd Cogne, Mac l nKf odr -g -- -.PAVED ROADS UNDER NDnas.B,rkel.onei CONSTRUCTION P.ssyora- RAILAY5Doffone ýIbo ---hau -_-_--_-_RA _LWAYS _'__gone r yk2o- D N' No g,ndo. - k FERRiES C MdaeGP drnB-mb. CoEkr~o~jn. ~ ~ -• MassembEbAE N E GSA AJRPORTS Nio-oduR,p Sobacou-dQ .,L PROVINCE BOUNDARIES ,K N a n - INTERNATIONAL BOUNDARIES T H E GBt " BANJUL ---_ Frfn b.,Io TOm.o,ouo J.boroh - Dolof-n .. cA S A M A N C E D .... Tm : NNokolo Kolba K-m obMrcud er Ko-nkon* s gr Nd. Mmaako KI--Kd. Kounta.b.O Tougo,CO K..oA Gen Bignona avn o - inaB 5edh. Sann 5.,or T-ngu u N kuo , lp lp 1, I 6, 1, lp 1ý lý - - - R - -r C.lomSt,r,,. gucodougo S- G U I N E A - B-SAU G U l N E . JULY lo0   D C Z mT -C