Ghana - Agricultural Services Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 14500 IMPLEMENTATION COMPLETION REPORT GHANA AGRICULTURAL SERVICES REHABILITATION PROJECT (CREDIT 1801-GH) MAY 12, 1995 Agriculture and Environment Division West Central Africa Department Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CL. RRKENCX JL ! * ALLN I S (Deccmber 1994) Currency Unit = Cedi Exchange Rate: at appraisal: US $1.(0 = Cedi 151.00 average, 1987-1994: US $1.00 = Cedi 422.00 at completion: US $1.00 = Cedi 915.00 WEIGHTS AND MEASURES I millimeter (mIn - 0.039 inches I meter (m) 3.28 feet I square meter = 10.76 square feet I cubic meter = 1.31 cubic yards I kilometer (kni) 0 0.62 inches I square kilometer 0 ).386 square miles I hectare (ha) = 2.47 acres (ac) I kilogramii (kg) = 2.'05 pounds (lb) I metric toin (ton) = 2,205 pounds FISCAL YEAR January I to December 31 ABBREVIATIONS AND ACRONYMS ADC Agricultural Developmenit Corporation of South Korea AED Agricultural Extension Department AtDB African Development Fund (African Development Bank) AMTD Agricultural Mechanisation and Transport Department AP&HD Anitial Production and Health Department APCC Agricultural Policy Coordinating Coiitnittee ASRP Agricultural Services Rehabilitationi Project PCU Project Coordinating lJiit CIDU Crop Services Department of MOFA COCOBOD Cocoa Marketing Board CRI Crops Research Inrstihute CSIR Council on Scientific and Industrial Research DCA Development Credit Agreement DIC Diverstiture Implementation Committee FAO Food and Agricultural Organisation of the tUnited Nations FD Fisheries Departtment GIDA Ghana Irrigation Developmenit Authority GOG Government of Ghana GSC Ghana Supply Commission ICR Implemenitation Completion Report IDA International Development Association KFW Kreditanstalt fur Wiedaraufbau MOFA Ministry of Food and Agriculture MTD Manpower Training Department MTADS Medium Tertm Agricultural Development Strategy NARMvlP National Agricultural Research Masterplan NARC National Agricultural Research Committee NEAP National Ertvironmental Action Plan O&M Operations and Mainteniance Expenditure PCtl Project Coordination Unit PPD Plant Protection Departnent PPF Project Preparation Facility PPMED Policy, Planninig. Monitoring and Evaluation Department of MOFA SAR Staff Appraisal Report SOE State Operating Enterprises TA Technical Assistance UNDP United Nations Development Program VORADEP Volta Region Agricultural Development Program VSD Veterinary Services Department  FOR OFFICIAL USE ONLY REPUBLIC OF GHANA AGRICULTURAL SERVICES REHABILITATION PROJECT (CREDIT 1801-GH) Table of Contents Page # Preface Evaluation Summary.......................................... Part I Project Implementation Assessment 1. Introduction .................................................1 2. Project Objectives ..................................................I 3. Implementation Expenence and Results ..................................................2 4. Summarv of Findings, Future Operations and Key Lessons Learned ......................... 10 Part II Statistical Annexes 1. Summary of Assessments .......................................... 12 2. Related Bank Credits .......................................... 14 3. Project Timetable .......................................... 15 4A. Loan Disbursements: Cumulative Estimated and Actual ....................................... 16 4B. Disbursement by Categories .......................................... 17 5. Key Indicators for Project Implementation .......................................... 18 6. Key Indicators for Project Operations .......................................... 19 7. Studies Included in Project .......................................... 20 8A. Project Costs .......................................... . 21 8B. Project Financing ........................................... 22 9. Status of Lcgal Covenants .......................................... 23 10. Compliance with Operational Manual Statements .......................................... 25 11. Bank Resources: Staff Inputs .......................................... 26 12. Bank Resources: Missions .......................................... 26 Appendixes: A. Mission's aide-memoire B. Borrower's Contribution to ICR C. Borrower (GIDA)'s Evaluation of Twinning Component D. Map This document has a restricted distribution and may be used by recipients only in the performance of their official duties Its contents may not otherise be disclosed without World Bank authorization.   IMPLEMENTATION COMPLETION REPORT GHANA AGRICULTURAL SERVICES REHABILITATION PROJECT CREDIT NO. 1801-GH Preface 1. This is the Implementationi Completion Report (ICR) for the Agricultural Services Rehabilitation Project (ASRP), for which Credit 1801-GH in the amount of SDR13.3 million (US $17.0 million equivalent) was approved on June 30, 1987 and made effective on December 18, 1987. 2. The credit was closed on June 30, 1994, compared with the original closing date of June 30, 1993. It was almost fully disbursed with a small undisbursed balance of SDR 23,918.21 being cancelled at closing. The last disbursement took place on September 8, 1994. Parallel financing for the project was provided by: UNDP (US$1.5 million), to strengthen the Policy, Planning, Monitoring and Evaluation Department (PPMED); AfDB (US$25.0 million), to finance the importation and distribution of fertilizers by the private sector; and, KFW (US$8.5 million), to finance supply of animal vaccines and other veterinary drugs and to provide logistical support to the Animnal Health and Production Department (AHPD). The KfW credit, which closed in March 1994, was fully disbursed. The AfDB credit closed in 1993 and the UNDP grant closed in 1991, both were also fully disbursed. 3. The ICR was prepared by George Alibaruho, Agriculture and Environment Division, West Central Africa Department and reviewed by Cynthia C. Cook, Division Chief, AF4AE, Anil Bhandari, Portfolio Manager, AF41N, and Franz Kaps, Operations Advisor, AF4DR. The contributions of the Borrower and parallel financiers of the project and their written comments are included as appendices to the ICR. 4. Preparation of this ICR was begun during the Bank's final supervision/completion mission, May 23-June 9, 1994. It is based on materials in the project files, field visits to project sites and interviews with officials of government and implementing agencies as well as participating farmers. The Borrower contributed to preparation of the ICR by participating in the completion mission and contributing views and written comments reflected in the mission's aide-memoire (Appendix A) and this ICR, and by preparing its own evaluation of the project's preparation, execution and post implementation operational plan (Appendix B).   IMPLEMENTATION COMPLETION REPORT GHANA AGRICULTURAL SERVICES REHABILITATION PROJECT [Credit No. 1801-GH] Evaluation Summary 1. Introduction. Ghana's Economic Recovery Program, of stabilization and structural adjustment, was launched in April 1983. IDA policy advice and mobilization of donor assistance played a key role in defining the GOG's coherent and sustainable policy framework that underpinned economic recovery. Prior to the ASRP, IDA support was also through a series of reconstruction import credits; export, energy, power, infrastructure, and social sectors rehabilitation credits; and an industrial sector adjustment credit, (all spread over 15 operations). The first IDA commitment specific to the agriculture sector was fcr a US$25 million Oil Palm project in 1984. 2. Project Obiectives and Components. The primary objective was to initiate a process of rehabilitation to enable Ministry of Food and Agriculture (MOFA) to provide effective support to agriculture and this would be through strengthening the institutional framework for the formulation and implementation of agricultural policies and programs and impioving the delivery of public services in the sector. Corresponding components were: (i) strengthening MOFA's capacity for policy formulation, planning, monitoring and evaluation; (ii) carrying out policy reform, focusing on privatization and divestiture of fertilizer and veterinary drugs distriUution, tractor hire services, and other state-owned enterprises in the agriculture sector; and, (iii) improving agricultural support services in research, irrigation and extension [Para. 2.01]. 3. Special Covenants and Agreements Designed to Achieve Objectives. Conditions of effectiveness were: GOG signing a TA agreement with UNDP and selecting TA experts under the UNDP project to strengthen Policy, Planning, Monitoring and Evaluation Department (PPMED) of MOFA. GOG also agreed to: privatize fertilizer distribution by 1991, phase out fertilizer subsidy by 1990; recover 100 percent of cost of veterinary drugs as from 1988; submit to IDA and other donors a masterplan for agricultural research by June 30, 1988; and withdraw from tractor and plant hire services in the agricultural sector by June 10, 1991. Compliance with these covenants, which was on the whole satisfactory, was critical to the attainment of the institutional and policy objectives of the project [Para. 2.02]. 4. Evaluation of Proiect Objectives. The objective of initiating a process for MOFA to provide effective support to agriculture was clearly appropriate and timely. The strengthening of public sector institutions in policy formulation and plaining and delivery of services to the sector was consistent with this objective. 5. Proiect Success. The project succeeded in initiating a process of rehabilitation to revitalize MOFA to provide meaningful support to the agricultural sector. MOFA's policy formulation, planning, monitoring and evaluation processes were substantially strengthened. Policy reform targets were also mostly realized. Success in improving agricultural support services during the project period was only partial: agricultural extension objectives were substantially met; crops research targets were met to a very limited extent; pilot-testing of small, farmer-operated irrigation facilities was not completed but pilot-testing of rain water harvesting was essentially implemented. Studies substantially achieved their objectives, having led to, among  other things the Medium Term Agricultural Development Strategy (MTADS). National Agriclutural Research Masterplan (NARMP) and National Environimiental Action Plan (NEAP). but no study addressed the problem of duplication of extension services between MOFA and Cocoa Marketing Board (COCOBOD) [Paras 3.01, 3201]. 6. Project Sustainability. The sustainahility of manyv of the benefits from completed project activities, appears well tfounded. The capacity improvemiienits of MOFA and Ghania Irrigation Development Authority (GIDA) are likely to be sustained if GOG can provide the required budgetary resources for periodic investment in vehicles, equipment replacement. operations and mairitenance and staff training. Adequate remuneration is a key to staff retention, which will be critical to sustainability. With privatization of the distribution of fertilizers and veterinary drugs there is reduced pressure on the government's budget and its managerial resources, but poor market penetration by the private sector, especially in remote areas, needs further attention. The benefits from the revolving fund for veterinary drugs established under the project (KFW funds) will be sustained only if the local cedi prices for drugs fully reflect the market exchange rate. For project activities not comilpleted or tested, assessment of expected benefits and sustainability is premature [Para. 3.201. 7. Project Costs, Financing and Tim-letables. Project cost at appraisal was $53.3 millioni that was to be finianced in parallel by an IDA Credit of $17.0 million (SDR 13.3 million), Kreditanstalt fur Wiederautbau (KFW), ($8.5 million); the African Development Bank (AfDB), ($25.0 million); and the United Nations Development Program (UNDP), ($1.5 million). The Government of Ghana's share was $1.3 niillion. At completion, the ASRP cost $53.2 million, which was financed in line with original expectations. Implemilenitationi time tables, from credit effectiveness to closing, proved dift'icult to adhere to. Institutional capacity building (apart from 'JiDA where serious delays were experieniced), policy reformiis and the studies components of the project were implemilented with the least delays. The support services improvelmient activities, apart tfl-omi extension, experienced serious delavs and some of the field pilot programs were not completed [Paras. 2.0(2. 3.01, 3.02, 3.-01]. 8. Key Factors Affecting Proiect Performliance. Adverse factors were: () inadecluate appreciation of the weakness ol' the Governmalenit bureaucracy by IDA and lack of comprelhenisioni of project scope and its demnands by GOG; (ii) IDA procuremilenit and disbursemenit procedures took a longer time to internalize within GOG and Project Coordination Unit (PCU) in particular: and (iii) a multiple number of imiiplemilentinig agencies that initially did not relate smoothly to the PCU. On the positive side, particulakrly for the successful implemienitationi of the institutionial developmenit and policy reforimis, were: (i) the availability of goverinmiienlt personnel of a higih academic level; (ii) the willingness of' MOFA to draw upon professionals from outside MOFA; (iii) commilitimienit to the project at the highest levei of MOFA; and (iv) support fromii the Banik's Resident Mission in Accra. [Paras. 3.04. 3.t)5. 3.18. 3. 191. 9. Bank Performanice. Bank performance was satisfactory. However, it over-estimated Ghanaian implementation capacity and could have reduced the number of project activities. Slow commnuniication increased the turn-around time for clearance of documents sent to Washing,tonl: changes in Bank supervision staff tended to underminle continiuitv and sometimes led to different interpretations of the concept and design of project components. However, by financing a higher percentage of project costs, extending the closing date ftor the credit, and transferring the supervision responsibility for the project to the Accra Resident Mission, the Bank helpe] make substantial imnprovements in implementation. Conversely, by acceding to MOFA and GIDA on changes in the concept. design and implementation of the pilot irrigation schemes. IDA m1ay have  obtained Borrower ownership, but it resulted in reduced farimier involvemilent and consequenitly poor implementationi. 10. Borrower Performance. In spite of poor managemenit in the initial years the overall performance of the Borrower could be judged as satisfactory on the basis of the implemientatioii of reformis and institutional strengthieninig. During the start-up phase, key officials caused long delays in decisions, as in the case of the GIDA/ADC technical cooperation contract; counterpart funds were insufficient; the PCU and implementing agencies were unfamiliar with Bank procedures. However, as staff recruitment and training progressed, implementation improved. The GOG also took the exceptional step of exempting the project from using the procurement services of the Ghania Supply Coiimmissioni, to help reduce the turn-around time for procuremient actions. Nevertheless, the initial delays did not allow comipletioni of field activities in research and irrigationi during the project period [Para. 3.071. 1 1. Proiect Outcome. Overall project outcome was satisfactory. with variations between componlenits. Institutional rehabilitation and reform, capacity-buildinig and logistics support to MOFA were satisfactory. The studies component was satisractory. Policy reforms, extension. preparationi of MTADS, NARMP. NEAP anid follow-on projects, were satisfactorily carried out. In the early stage implemilenitatioii suffered from the inexperience of the PCU and beneficiary agenicies, which led to unsatisfactory outcome in the case of the pilot and research activities in the field. At project closure, four of the six irrigation pilot schelimes and two of the six valley bottom rice pilot projects had not comimileniced. Correspondingly, the Borrower did not derive the full b'enefits of the techinical assistance under the GIDA/ADC twinniniig contract, which was expected to assist GIDA in improving its services [Para. 3.201. 12. Key Findings of Imiplemilenitation experience. (i) a demoralised bureaucracy is slow to respond to efforts to improve its operations, particularly in an environllment where macroeconomic reform substantially erodes the buy'ing power of salaries; (ii) Bank's procurement and disbursement procedures require muchi forward planining and painstak-inig attention to detail and it takes time to understand firstly, wlhv the procedures are necessarv and, seconidly. actually miake the effort to comlply with themil; the situation is worsened if there is no direct benefit accruinig to individuals involved; (iii) havinig gained training and experienlce, staff become more niarketable as an economy liberalizes and are likely to miove out of the public sector and therefore affect sustainability of activities supported under the project: and (iv) institutions whichi are being strengtihenied through a learning-by-doing process need a longer timie to implemiienit activities/projects identified for this purpose. 13. Plans for Future Operation. Support services initiated under the project are being assisted by IDA under successor projects for agricultural research (FY 91 ). agricultural extension (FY 92), livestock (FY 93), environimilental resOurces maniagemilenit (FY 93), agricultural sector investment (FY 94); and the Fisheries Sector Capacity Building Project (FY 95). The GOG has undertaken to provide funds for the pilot activities that wvere not completed within the project period. On policy, the GOG is coiimmitted to the reforimis carried out under the project. Oin institutionlal rehabilitation, the GOG will provide adequate budgetary fundinig ftor O&M. particularly for vehicles, office and field equipmienit. GOG will also build into the budget and work planis the periodic replacement of equipimienit so as to iiiaiit and i uprove upoin the level of service efficieticy created by the project. GOG will also maintain the PCLI and consolidate administrative services tor as miany IDA-funtded projects as possible under the PCI). [Para. 3.031.  14. Key Lessons for Future Proiects. Where a key project objective is the strengthening of institutions, expectations on their capacity to simultaneously undertake execution of investment activities in the project should be closely examined; where the role of the PCU is crucial in advising on and facilitating the procedural aspects of implementation, especially procurement and disbursement, the PCU should be established, staffed and made fully functional, perhaps through a PPF, before major project activities commence; the benefits to be derived from twinning arrangements for technology transfer can be drastically reduced if suchl arrangements are not well- synchronized with project implementation; to improve Borrower understanding of project concept and implementation demands and ensure overall project ownership, a broad-based project launching workshop should be organized and all key officials concerned with implementation of the project should participate; locating supervision responsibility in the resident mission definitely improves project implementation; where an investment project contains sector or macropolicy adjustment objectives and measures to attain them, their development and social impacts should be studied and monitored closely [Para. 4.02]  IMPLEMENTATION COMPLETION REPORT GHANA AGRICULTURAL SERVICES REHABILITATION PROJECT [Credit 1801-GH] PART I: PROJECT IMPLEMENTATION ASSESSMENT 1. Introduction 1.01 Ghana's Economic Recovery Program, of stabilization and structural adjustment, was launched in April 1983. Policy advice and mobilization of donor assistance, by IDA played a key role in defining the GOG's coherent and sustainable policy framework that underpinned economic recovery. Prior to the ASRP, IDA support was also through a series of reconstruction import credits; export, energy, power, infrastructure, and social sectors rehabilitation credits; and an industrial sector adjustment credit, (all spread over 15 operations). The first IDA commitment specific to the agriculture sector was for a $25 million Oil Palm project in 1984. 2. Project Objectives Proiect Objectives and Components. 2.01 The ASRP was designed to strengthen institutional capacity and services in the Ministry of Food and Agriculture (MOFA) and to support agricultural policy reforms. Specific objectives were: to strengthen capacity for the formulation and implementation of agricultural policies and programs, through technical assistance and staff training, provision of computers and other office, field and laboratory equipment, supply of vehicles, rehabilitation and refurbishing of offices and staff houses, and provision of funds for some project-related operating costs. Departments to be covered were: the Policy, Planning, Monitoring and Evaluation Department (PPMED), to enhance capacity for formulating sectoral policies and programs, monitoring and evaluating their impact; and the Ghana Irrigation Development Authority (GIDA), to strengthen capacity for development of sustainable irrigated agriculture, through a program of technical assistance underpinned by a twinning arrangement; and, the Project Coordinating Unit (PCU), to facilitate project coordination and administration, including procurement. At mid-term the scope expanded to include: MOFA's Head Office, to strengthen policy and program coordination and decision-making; Fisheries Department (FD), to improve mobility and logistics; Plant Protection Department (PPD), to improve logistics and capacity for pest control, seed-testing and staff training; and, the Manpower and Training Department (MTD), to manage and provide ministry-wide training and improve the effectiveness of human resources development services; to carry out policy reform, focusing on divestiture of fertilizer and veterinary drugs distribution, tractor hire services and privatization/reform of agricultural state operat- ing enterprises (SOEs). Departments involved were: Crop Services Department- Input Development Unit (CIDU), to promote privatization of fertilizer marketing; Agricultural Mechanization and Transport Department (AMTD), to privatize  2 tractor hire services; Animal Health and Production Department (AHI&PD)[later renamed Veterinary Services Department (VSD]), to promote privatization of veterinary drugs; to undertake studies and develop a Medium-to-long-Term Agricultural Development Strategy (MTADS), leading to sectoral and sub-sectoral plans, programs and projects. to improve agricultural support services in research, extension and irrigation. This was to involve: the Council on Scientific and Industrial Research (CSIR), to prepare a National Agricultural Research Master Plan (NARMP) and, through affiliated Crops Research Institutes (CRIs), conduct research on cotton, rice and rain water harvesting techniques, in six different ecological zone settings; GIDA, to design, in consultation with farmers, and pilot-test the establishment of farmer-owned and farmer-operated small-scale irrigation schemes at six-to eight ecologically different sites; and, Agricultural Extension Department (AED), to carry out pilot programs In agricultural extension. Covenants and Agreements Designed to Achieve Obiectives 2.02 Key covenants were: for effectiveness by December 31, 1987, the signing of an agreement between GOG and UNDP for TA to PPMED; selection of TA personnel under UNDP project to strengthen PPMED; appointment of head of CIDU, to be in charge of fertilizer privatization program; arrangement by MOFA, satisfactory to IDA, for divestiture of SOEs; establishment of the PCU and of the APCC. Others: PPMED to review with IDA progress of fertilizer privatization program by December of each year, starting 1988; Borrower to reduce fertilizer subsidy to 30 percent of retail prices in 1988, 15 percent in 1989 and eliminate it in 1990; Borrower to privatize fertilizer distribution by: testing private retailing in Volta and Brong Ahafo regions in 1988; selling only at its main distribution stores in 1989; selling at Tema, Swedru, Kukurantumi and Tamale depots only in 1990; and, relinquishilg all import and distribution functions in 1991; Borrower to recover 100 percent of cost of veterinary drugs as from 1988, submit to IDA and other donors a masterplan for agricultural research by June 30, 1988, discontinue wheeled tractor services by December 31, 1987. combine harvester services by February 28, 1989, and land clearing services, by June 10, 1991. Compliance with these covenants, which was on the whole satisfactory, was critical to the attainment of the institutional and policy objectives of the project. 3. Implementation Experience and Results Achievement of Proiect Obiectives 3.01 Realization of the project objectives can be summed up as partial. However, since the project covered many areas, a summary of the components and the extent to which their objectives were achieved follows. Institutional strengthening, including rehabilitation, and logistical support to MOFA and its departments and agencies: objectives substantially achieved; the PCU, to provide administrative services, particularly procurement, accounting and financial reporting services and promoting cohesion among project entities: objective substantially met; privatization of tractor hire services, distribution of veterinary drugs, except vaccines, and import and distribution of fertilizer: objective substantially realized; privatization  3 of agricultural SOEs: this activity was transferred to the Divestiture Implementation Committee, which has a broader mandate to privatize/divest all SOEs, including those outside agriculture; pilot-testing extension modalities: objective substantially realized; crops research: objective partially met for rice but negligible for cotton; pilot-testing the ability of small farmers to establish with their own resources and sustainably manage irrigation facilities: results negligible; pilot-testing rain water harvesting practices (low risk rice cultivation trials), in conjunction with low cost small scale irrigation on some sites (valley bottom rice cultivation): objective partially achieved; and, Studies, including preparation of MTADS and NARMP: objectives substantially achieved, but follow-up is needed to translate them fully into beneficial programs. The benefits from all activities could increase in time, as resources and supervision inputs are made in the post-implementation phase [para.3.03]. Proiect Costs and Financing 3.02 Project cost at appraisal was $53.3 million that was to be financed by an IDA Credit of $17.0 million equivalent. Parallel financing was to be provided by Kre,ditanstalt fur Wiederaufbau (KFW), ($8.5 million equivalent); the African Development Bank/Fund (ADB/ADF), ($25.0 million); and the United Nations Development Program (UNDP), ($1.5 million). The balance, $1.3 million was to have been contributed by the beneficiaries. At completion, the ASRP cost $53.2 million, which was financed in line with original expectations. Expenditures by project categories are shown in Table 4B, by project components in Table 8A, and yearly and cumulative disbursements in Table 4A of the Statistical Annexes. Proiect Sustainability 3.03 The sustainability of many of the benefits from completed project activities, based on the design and performance so far, of follow-on projects, appears well founded. The capacity improvements of MOFA and its service and technical departments, the organizational and structural changes and logistical support that have improved coordination are likely to be well sustained if GOG can generate and allocate to them the required budgetary resources. Maintaining a reasonable level of periodic investment in vehicle and equipment replacement and provision of adequate O&M funds are critical. In terms of retaining the benefits of staff trained under the credit, their environment and working facilities need to remain serviceable and adequate, renovating or replacing them as needed, and their remuneration must be attractive. In the final analysis, these are budgetary matters. On privatization of fertilizers and veterinary drugs, the benefits from the policy changes supported under the credit are debatable, from the point of view of some key officials, due to what is seen as inability of the private sector to supply remote areas at affordable prices; doubts about benefits could undermine policy commitment and its sustainability. The benefits from the refrigeration and field equipment for veterinary services are protected by the operation of the revolving fund established under the project (KFW funds), whose operation will continue beyond the implementation phase. For project activities not completed or tested, it is too early to tell whether expected benefits will accrue, let alone be sustainable. This includes the pilot irrigation program and much of the research, especially on cotton. For rice research, the identified benefits need to be confirmed by more testing and economic analysis to ascertain their sustainability. Follow-up activities, such as re-organization of the cotton industry, need to be pursued to realize and sustain project benefits.  4 Factors Generally Affecting Achievement of Objectives 3.04 Scope. The needs of the Ghana agricultural sector and related institutions were extensive, following more than a decade of neglect. The ASRP sought to build both planning and implementation capacities whilst rationalizing institutional responsibilities in areas that appeared critical to medium-to long-term growth in the sector. The multiplicity of components required coordination and implementation capacity that needed time to be built. Management of the project's field activities consisting of sub-components such as irrigation, research and extension at different sites and locations, proved difficult. Most activities did not get off the ground until up to three years after credit effectiveness. The project's implementation schedule was too ambitious to be completed in a timely manner, in light of institutional weaknesses whose rectification was the genesis of this very project. Implementation of activities in multiple phases over a longer period, but planned at identification, may have achieved more results for the field sub-components. 3.05 Project Coordination. Administration and Implementation Arrangements. The PCU, which was to provide essential support services to all project entities, including facilitation of compliance with Bank policies and procedures, should have been fully functional before the start of project activities. The Unit was slow in appointing a financial controller and so lacked, in the start up period, the capacity to discharge adequately its accounting, financial and procurement functions. Training of the Unit's key personnel in Bank procedures, particularly procurement and disbursement, took longer to accomplish than was expected. Additionally, the functions of the PCU did not always seem to be well understood by all parties. Some project entities felt that the PCU was being obstructionist, and not supportive, while the Unit tried to maintain central control to ensure that the correct procedures were followed, even though that might cause delays. The relationship between the PCU and project entities was smoothed out by giving more authority, including control of their approved budgetary resources, to the implementing agencies. The PCU was then able to attend more effectively to the procurement and accounting functions. Most of the implementing agencies were able to attend more efficiently to their operations. 3.06 Capacity Building in MOFA: The more successful aspects of the ASRP include the strengthening of capacity for: policy analysis and decision-making processes; formulation of sector strategy and programs; project coordination, implementation, monitoring and evaluation. Contributing to this result were: facilities' rehabilitation and upgrading; and, provision of mobility and logistical support to MOFA and associated departments' agencies; computerization of data storage, retrieval and analysis; and, training. The Agricultural Policy Coordinating Committee of MOFA, established with the support of the project, is functioning as intended. The strengthened PPMED has improved its performance substantially and is functioning well; the staff have received the required training and are putting it to good use in data collection, including conduct of field surveys, and data analysis and publication of results. This improved performance is facilitated by the computing and other office equipment and related facilities and staff support. With PPMED staff attached to all MOFA departments, coordination and linkage with on-going programs and projects in all sub-sectors has improved. On the problems side, refurbishing residential bungalows for the senior staff, did not take place. As these were in the general government housing pool, there was no guarantee that, after renovation, the units would have been allocated to PPMED staff. Furthermore, partly because of poor housing and better opportunities in the local private sector and overseas for  5 trained people, staff turnover has been larger than normal and has tended to undermine the sustainability of project benefits. 3.07 Irrigation Agronomy and Efficiency Trials. This program of work was only partially fulfilled largely due to its late start. A contributory factor was the issue, which was not resolved until two years after credit effectiveness, of whether or not the program should have been the responsibility of CRI or GIDA. The conduct of research field trials, by its nature so heavily influenced by the vagaries of the seasons, requires adequate time to ensure that the results and conclusions are sound and replicable. Nevertheless, the findings so far are showing that the low risk water harvesting, rain-fed, rice farming practices developed under the project appear to offer good prospects for raising efficiency of rice produztion in the north of the country. Farmers are enthusiastic and yields as high as 3.0 to 4.0 tons/ha have been recorded on trial sites, compared with 1.0 ton/ha. in neighboring farms using the same seed variety with unimproved water management. The valley-bottom water management technology development and testing for improved cropping was conducted by CRI in the Kumasi and outlying areas also with promising results. Farmers have been highly impressed with the technology and are organizing themselves to adopt it where possible. As it entails the need to divert perennial streams or use of pumps, especially in the dry season, the cost effectiveness of this technology needs to be further evaluated. 3.08 Rice Research. This program was linked closely to pilot water use in valley bottoms, focused on developing and testing appropriate agronomic practices for smallholder application. It included soil fertility and water use responses, tillage and crop rotation practices, variety trials and pest management methods. The research was managed by the CRI at Besease, near Kumasi. The work was replicated on only four of the planned six ecological sites, because it was started two years behind schedule, following resolution of the issue of responsibility between GIDA and CRI for the conduct of the work. Early results from this work are promising and CRI is well on its way to working out applicable agronomic practices for small farmer rice production in the valley bottoms. There is need to continue the program, to test the findings under other representative ecological zones and to conduct "on-farm" trials to generate realistic costs so as to assess the commercial viability of the responsive production operations. This should be undertaken in the post-implementation phase. 3.09 Cotton Research and Fibre Testing. The program covered a range of agronomic trials to determine the optimum recommendations for smallholder producers for high yielding varieties, and best management practices for soil tillage, its fertility and moisture levels including their conservation, and pests and diseases. It also provided for training of staff of the cotton buying company; and for refurbishing and equipping the laboratory, including provision of cotton fibre testing equipment for quality control. On-station trials were conducted in the vicinity of the station at Nyankpala, Wa in Upper West Region, Manga in the Upper East region and at Damongo in the Northern Region. The on-station yields ranged from 1.0 ton/ha of seed cotton to 2.0 tons/ha. No "on-farm" trials have been carried out to test the findings of the trials on a practical scale and this will be necessary in order to be able to make appropriate judgments on their commercial viability and to formulate sound recommendations to farmers. Fibre testing laboratory equipment was installed by the supplier in October, 1993, two years after orders were made. The equipment had not been made operational by the time of the completion mission's visit. The supplier's follow-up service was needed to rectify the problem. The equipment would assist the cotton buying companies in analyzing the fibre  6 quality essential for the export markets. Also, the training received by the ginnery staff cannot be put fully into use until the fibre-testing laboratory becomes operational. 3.10 Cotton Industry and the Small Farmer. The project paid for consulting services to study the structure and organization of the industry and present a plan and recommendation for its possible restructuring. A key objective of such restructuring and re-organization was to bring the benefits of the research financed under the ASRP to the small farmer. The consultant's report confirms that, without changes, the small-scale cotton farmers will continue to operate as seasonal contract laborers for ginneries, dependent largely on them for the supply of inputs, including credit. They are paid the equivalent of only 7.0 percent of the f.o.b price for their seed cotton, after ginneries have deducted the cost of input supplies and other support services they render to the farmers. Larger farmers, who source their own inputs independently of ginneries and pay a ginning fee, receive 91.0 percent of f.o.b price of lint and seed, which implies that small-scale farmers receive poor prices for comparable grades. At current market prices, the technologies being tested, while commercially beneficial to larger farmers and ginneries, offer relatively little benefits to small farmers unless the industry is re- organized. Re-organization would be needed to increase competition in the supply of inputs to the small farmers, including credit, and to enable them to sell their output competitively as lint and seed and realize higher returns. Follow-up by GOG is recommended so that the benefits of this project activity may be fully realized and sustained. 3.11 Building GIDA's Capacity. This project component, representing a substantial part of the expenditure out of the IDA credit, started very late. In its first decade of irrigation development Ghana's strategy concentrated on establishing costly water impounding reservoirs or pumping schemes for medium to large scale irrigation. It did not pay sufficient attention to the farming requirements and farmers' interests and concerns. The result was very little down- stream development of sustainable irrigated lands which could be productively managed by farmers. Thus the main consideration in the ASRP was a shift to low cost, small scale, sustainable irrigation systems, with direct involvement of farmers in assuming the risks of ownership and operation of the facilities. This new approach called for change in the management and operations of GIDA and commensurate training of the staff in the required skills. Support to building GIDA's capacity was three-pronged: a twinning arrangement with a successful similar irrigation authority in a country with comparable resources and other relevant conditions; pilot testing ol small scale irrigation schemes identified with the participation of farmers; and, logistical support, mainly in the form of vehicles and office equipment. GIDA would train the farmers to prepare them to assume most responsibilities for designing, financing and operating their schemes; 3.12 The Twinning Arrangement. The twinning benefits were modest because of delays in signing the twining contract and therefore a reduced implementation period. Other factors which reduced the impact are discussed below. GIDA twinned with the Agricultural Development Corporation of South Korea (ADC). Communication between GIDA and ADC staff, while adequate at the senior management level, left much to be desired at the lower professional and technical levels. There appeared to be a perception among national engineers that in any event, the designs of the "experts" (meaning the Koreans), who knew less of the local farming and crop cultivation methods, would carry the day, although they were based more on a Korean-type agricultural and socio-economic environment: more capital-intensive, with a focus on rice irrigation in a mono-crop culture. Substantially better results could have been achieved if there was greater cooperation between the technical staff of the partners.  7 Cultural and communication problems, including unfamiliarity with the English language and hierarchical barriers to open discussions, made the training and skills transfer difficult to accomplish. Furthermore, some of GIDA's professional staff felt that the arrangement was conducted in a client/consultant relationship, rather than as a " partnership", aimed at transferring specialized skills and experience, where basic technical knowledge is presumed of Ghanaian engineers. The relationship resulted in a degree of indifference between the two groups of staff. There was also a perception among national staff that in any event, the designs of the foreign "experts" would always carry the day. On the positive side, GIDA staff claimed that the study periods in South Korea and elsewhere (part of the intended training) were beneficial, particularly the level of technical information received. Also, substantial benefit was derived from the practical training of farmers by staff of the twinning partner, in proper irrigation practice on some of the existing schemes. As a result, this part of the program was extended to reach more farmers. 3.13 The Small-scale Pilot Irrigation and Farmer Training Schemes. The conduct of these pilots was associated with the twinning program, and had to await the selection and recruitment of the twinning partners. Decisions on this by GOG proved very difficult and the process became protracted. The result was several years delay in starting the pilot schemes. The performance of GIDA in undertaking these pilot schemes was also poor. They neglected to observe in the exercise the basic concepts underpinning the new policies of GOG to foster irrigation development. The schemes' and GIDA's major deficiencies were: inadequate involvement of the cooperating farmer groups in the identification, planning and designing of schemes; neglect of simple, inexpensive options, which were affordable and manageable by the farmers; use of contractors in implementation rather than farmers; and, failure to provide technical guidance during the construction and early stages of operation and maintenance. While six pilot sites were selected, only two reached the implementation stage (Kikam and Sata). The designs were inappropriate in parts, and unnecessarily complicated. GIDA, contrary to the pilot requirements, chose to implement by contract under local competitive bidding. The selection process was subject to delays in decisions. The contract prices were much higher than expected and threatened the financial sustainability of project activity. Neither contractor showed the required competence for such work and both experienced serious problems. In one case, they relied on heavy earth moving equipment for the canal networks where the ground conditions were too wet for such equipment The contractor on the second scheme was unable to complete the works and GIDA staff were left with the task of completing them. By this stage the cost overruns had become prohibitive and cost cutting measures adopted had to focus on reduction in design standards while maintaining original design area. This led to discontent among farmers who preferred to have concrete-lined, instead of clay-lined, main canals, which required more maintenance. At project closure, the farmers had not gained enough experience in operating the scheme and the technical feasibility of water flow through the canals had not been tested. More time is needed to demonstrate the technical viability of the schemes and the competence of the farmers in operating them and achieving economic results. This project activity would not have suffered so much if: (a) GIDA realized that manual methods were more appropriate and efficient for small scale systems; (b) farmers were involved in their projects from the conceptual stage and undertook as much of the implementation works as possible; and (c) IDA had taken a firmer stand against the changes by GIDA in the agreed implementation guidelines. GIDA disagrees with some of the above findings and these are reflected in its comments in Appendix C.  8 3.14 Pilot Extension System. This was one of the most successful activities of the project. Under the pilot extension activity, eight districts in four regions, reflecting different crop combinations or the dominance of a single crop (as in the case of cocoa), were identified to try out different modalities for reaching the farmer with improved technical packages. MOFA Extension Department and COCOBOD, in the case of cocoa, were the implementing agencies. The mobility of extension staff was facilitated by the provision of motor bikes and pedal bicycles. The point of departure for testing extension modalities was the well-known Training and Visit system but during implementation, modifications were made in training frequency, from a bi-weekly frequency to a monthly cycle. The research/extension/farmer linkage was strengthened. A more efficient Unified Extension mechanism was established, as the demand on the farmers' time by different subject matter extension staffs was curtailed. The system worked well and it was replicated in four other districts during implementation. The objectives of this activity were substantially realized, leading to the nation-wide National Agricultural Extension Project. The dichotomy of an extension system run by COCOBOD, in addition to MOFA's national system, still exists. This duplication needs to be sorted out if there is to be a really unified system. 3.15 Privatization of Fertilizer Trade. Tractor Services and Veterinary Drugs. The privatization of input supply services (fertilizer, tractor hire services and some veterinary drugs) has essentially been successfully accomplished. Privatization of tractor hire services was completed. Distribution of veterinary drugs, except vaccines, has been privatized. The revolving fund, which supports the vaccine program, is not replenished at levels that will ensure program sustainability because cedi vaccine charges have not been raised in proportion to cedi devaluation in the past. Apart from a small implicit subsidy, commercialization of fertilizer has been carried out. Private dealers import and distribute fertilizers of the types specified by MOFA and only with its permission. Thus the dealers appear to act as wholesale and retail agents of MOFA, not taking the full business risks of trading in the products. And occasionally, MOFA "assists" dealers in the form of transport and sometimes warehouse rental (MOFA still owns the warehouses but leases them to the dealers/agents). There is also an element of guaranteed fertilizer purchase by MOFA from the agents (MOFA's purchases various quantities of fertilizers for what it calls "a strategic buffer stock"), suggesting that some indirect subsidies may still exist on fertilizer trade. MOFA is aware of this, but explained that: (a) the required capital base and level of risk-taking in private fertilizer trade was beyond the means of Ghanaian businessmen; (b) fertilizer was a seasonal product with a short shelf-life, a life made shorter in Ghana by inadequate private storage facilities; and (c) scarce working capital could be tied-up for up to a year, at high commercial interest rates, if deliveries were not made on time for a crop season, leading to dealer bankruptcies that could disrupt agriculture. MOFA regretted that the private sector had not been sufficiently responsive in certain markets for fertilizer and veterinary drugs, particularly the remote rural areas, with possible adverse impact on some farmers. Thus while much progress has been made towards realization of the privatization objectives, it might have helped to identify farmers who were adversely affected by this policy and to design and implement targeted mitigating programs so as to enhance policy commitment and its sustainability. 3.16 Privatization of Agricultural SOEs. This activity was discontinued under the project. Shortly after ASRP start-up, the GOG initiated a program of divestiture of State-owned Enterprises (SOEs), and this led to MOFA transferring the responsibility for privatization of agricultural SOEs (18 enterprises had been identified in the SAR) to the new agency. the Divestiture Implementation Committee (DIC), established for the purpose. Thus ASRP took  9 no further part in the privatization exercise and its progress is therefore not assessed in this report. 3.17 Studies Financed Under the Project. This activity was carried out satisfactorily. The objective of developing a strategy for the development of the agricultural sector through a series of studies leading to sectoral and sub-sectoral plans and specific projects appear to have been substantially realized. The ASRP is generally described as "the seed that produced all agriculture projects", following the resumption of IDA lending to Ghana. The ASRP pilot programs, studies, sectoral and sub-sectoral plans and preparatory activities led to the Medium Term Agricultural Development Strategy and the National Agricultural Research Master plan, in which context the following IDA-assisted projects have been designed and are at different stages of implementation: (a) the National Agricultural Research Project (Cr. 2247 GH); (b) the National Agricultural Extension Project (Cr. 2346 GH); (c) the National Livestock Project (Cr. 2441 GH); (d) the Agricultural Diversification Project (Cr. 2180 GH); (e) the Agricultural Sector Investment Project (Cr. 1802 GH); (f) the Environmental Resources Management Project (Cr. 2426 GH). and (g) the Fisheries Sector Capacity Building Project approved on May 2, 1995. Several development agencies and co-financiers have also based their sectoral assistance strategy and plans on the studies, pilot projects and programs tested and developed under the ASRP. On the other hand, the water resources and food security studies and action plans were not completed. Careful follow-up will be needed to translate all studies into beneficial programs, some of which, like in the restructuring plan for the cotton industry, are urgently needed. 3.18 Bank Performance. Bank performance was, on the whole, responsive and satisfactory. However, staff were overly optimistic about the pace at which the implementing capacity could be built up. Consideration could have been given to a two-phased approach, planned in advance of implementation: rehabilitation/institutional strengthening objectives and their associated activities could have been pursued in phase one, and field/investment activities in phase two, after capacity-enhancing benefits of the project had accrued to the implementing entities. Also, slow communications increased the turn-around for clearance of procurement and other documents sent to Washington; several changes in Bank supervision staff, which tended to undermine continuity at critical times and led to differing opinions on the basic concept and design of the project components. Responding to the need to shorten the turn-around time for procurement and other necessary consultations between implementing agencies and IDA, the task-management of the ASRP was transferred to the Accra Resident Mission. IDA later also agreed to re-design of the project at mid-term, re-allocation of funds between categories, finance a higher percentage of project costs and extend the closing date for the credit. This, and subsequent actions taken at mid-term, led to substantial improvements in the implementation. By conceding to MOFA and GIDA and agreeing to contractor- implemented schemes, the Bank may have obtained Borrower ownership, but this resulted in reduced farmer involvement and poor implementation. The Bank could have taken a firmer stand with GIDA over the design and implementation of the pilot irrigation schemes, considering that a shift towards farmer-owned and farmer-operated small-scale irrigation projects was a key objective for the irrigation sub-sector. 3.19 Borrower Performance. In the initial years of the project, poor management, lack of experience by the PCU and other implementing agencies led to deficient performance. However, as the PCU and other beneficiary agencies became fully staffed and their staff better trained, especially in Bank procurement, disbursement and financial reporting procedures, the  10 Borrower's performance improved. When IDA delegated procurement-related decisions to the Accra Resident Mission, the GOG also took the complementary step of exempting the project from using the procurement services of the Ghana Supply Commission, and further reduced the turn-around time. Some of the start-up delays were due to decisions outside of MOFA, e.g., the long delay in signing the technical co-operation contract with ADC. This was a major factor in the late start-up and non completion of the field irrigation component. An additional factor that partly slowed the pace of implementation was inadequate provision for counterpart funds. Aside from teething problems early in the implementation cycle, the Borrower's overall performance could be judged as satisfactory. 3.20 Assessment of Outcome. The outcome of the project is satisfactory on the whole. Rehabilitation, institutional reform, capacity-building and logistics support to MOFA and to most of its departments and support agencies were satisfactory. The studies cemponent was implemented satisfactorily. Policy reforms, extension, preparation of an MTADS, ARM, and follow-on projects, largely underpinned by studies financed under the project were satisfactorily carried out. Nevertheless, some outcomes were unsatisfactory. The project start- up and early stages of implementation suffered from the inexperience of the PCU, which consequently constrained the implementing agencies. The management bottlenecks led to delays in full implementation of the pilot and research components in the field. The performance of the project's field activities was generally deficient. The need to complete as many of these field activities as possible was a major factor in the extension of the credit's closing date by one year to June 1994 and at project closure, some of the irrigation pilot schemes and the cotton research at Nyankpala, including the fibre testing facilities, were still incomplete. Furthermore, the impact of some of the technical assistance may be less than expected; for example, the T.A for the restructuring of the cotton industry has not yet shown the expected full benefits and the GIDA/ADC twinning contract did not convey all the expected benefits to GIDA. 4. Summary of Findings, Future Operations, and Key Lessons Learned Project's Future Operation 4.01 Support services initiated under the project are being assisted by IDA under successor projects for agricultural research (FY 91), agricultural extension (FY 92), livestock (FY 93), environmental resources management (FY 93), agricultural sector investment (FY 94); and the Fisheries Sector Capacity Building Project (FY 95). The GOG has undertaken to provide funds for the pilot activities that were not completed within the project period. On policy, the GOG is committed to the reforms carried out under the project. On institutional rehabilitation, the GOG will provide adequate budgetary funding for O&M, particularly for vehicles, office and field equipment. GOG will also build into the budget and work plans the periodic replacement of equipment so as to maintain and improve upon the level of service efficiency created by the project. GOG will also maintain the PCU and consolidate administrative services for as many IDA-funded projects as possible under the PCU. The PCU is already providing procurement and accounting services to the IDA-assisted Livestock Project (Cr. 2441-GH) and will perform similar functions for the Fisheries Sector Capacity Building Project (FY 95) [Details are in Appendix B].  11 Key Lessons for Future Operations 4.02 Lessons learned are: where a key project objective is the strengthening of institutions, expectations on their capacity to simultaneously undertake execution of investment activities in the project should be closely examined with a view to simplification of project scope, possibly through phasing or smaller multiple projects; where the role of the PCU is crucial in advising on and facilitating the procedural aspects of implementation, especially procurement and disbursement, the PCUs should be established, staffed and made fillly functional, perhaps through a PPF, before major project activities commence; the benefits to be derived from twinning arrangements for technology transfer can be drastically reduced if such arrangements are not well-synchronized with project implementation; Bank's procurement and disbursement procedures require much forward planning and painstaking attention to detail and it takes time for officials to understand firstly, why the procedures are necessary and, secondly, to actually make the effort to comply with them; to improve Borrower unders.anding of project concept and implementation demands and ensure overall project ownership, a broad-based project launching workshop should be organized and all key officials concerned with implementation of the project should participate; locating supervision responsibility in the resident mission definitely improves project implementation; where an investment project contains sector or macropolicy objectives and measures to attain them, their development and social impacts should be monitored closely and, if necessary, complementary targeted programs should be designed and implemented to minimize adverse social impacts.  12 PART II: STATISTICAL ANNEXES Table 1: Summary of Assessments A. Achievement of Obiectives Substantial Partial Negligible Not applicable Macro Policies it E Li Sector Policies EL 0 Li Financial Objectives E Cli [ Institutional Development El [ Li Physical Objectives [ L Li Poverty Reduction El L Li Gender Issues Ei I El Other Social Objectives El 5 Environmental Objectives 5 5 5 Public Sector Management [ VI Private Sector Development EL L Li Other (specify) EL L L Li B. Proiect Sustainability Likely Unlikely Uncertain (v') (9') (9')  13 Highlv C. Bank Performance satisfactory Satisfactory Deficient (/) (1) (vi) Identification a El Preparation Assistance E el Appraisal E Supervision E El a Highly D. Borrower Performance satisfactory Satisfactory Deficient (~/) (V) (V) Preparation Q E Implementation E I Cl Covenant Compliance 5 IA a Operation (if applicable) 5 5 5 Highly Highlv E. Assessment of Outcome satisfactory Satisfactory Unsatisfactory unsatisfactory 0/ (I) 0/ 0/  14 Table 2: Related Bank Credits Loan/Credit Title Purpose Year of Approval Status Preceding Operations 1. Second Oil Palm Development 1984 Completed CR1498-GH, US$25.0 m of oil palm estate Following Operations 1. Cocoa Rehabilitation Support Policy 1988 On-going CR. 1854-GH, US$40.0 m Reforms and increase cocoa production 2. Forest Resources Conservation and 1989 On-going Management CR 1976-GH Management of forestry US$39.4 million resources 3. Rural Finance Project To provide investment credit 1989 On-going CR 2040-GH, US$20.0 m to farmers 4. Agric. Diversification Development of tree and 1991 On-going CR 2180-GH, US$16.5 m horticultural crops 5. National Agric. Research Rationalization of Agric. 1991 On-going CR. 2247-GH, US$22.0 m research resources 6. Agric. Sector Adjustment Reform of agricultural pricing 1992 On-going CR. 2345-GH, US$80.0 m and marketing; improvement Agric. Sector Adjustment of resource allocation and to 1994 On-going CR. 2345-1-GH, US$5.74 m foster private initiative 7. National Agric. Extension Provision of efficient 1992 On-going CR. 2346-GH, US$30.4 m extension services 8. Environmental Resources To support National 1992 On-going Management Project Resource and Environmental CR. 2426-GH, US$18.1 m Planning and Management 9. National Livestock Increase livestock production 1993 On-going CR. 2441-GH, US$22.5 m and services 10. Agric. Sector Investment Provision of low-level 1993 On-going CR. 2555-GH, US$21.5 m investment for rural infrastructure  15 Table 3: Project Timetable Steps in project cycle Date planned Date actual Preappraisal 10/30/86 10/30/86 Appraisal 11/22/86 11/25/86 Negotiations 04/13/87 04/13/87 Board presentation OS/21/87 06/30/87 Signing 06/22/87 06/22/87 Effectiveness 07/30/86 12/18/87 Midterm review 06/20/91 07/05/91 Project completion 12/31/92 06/30/94 Loan closing 06/30/93 06/30/94  16 Table 4A: Credit Disbursement: Cumulative Estimated and Actual US $ Million FY 88 FY 89 FY 90 FY 91 FY 92 FY 93 FY 94 Appraisal Estimate 1.5 2.6 8.3 12.9 16.1 17.0 17.0 Actual 1.1 2.5 6.0 9.8 14.0 15.9 16.8 Actual as % of Estimate 73.3 96.2 72.3 76.0 87.0 93.5 98.8 Date of Final Disbursement: September 8, 1994.  17 Table 4B: Disbursement by Categories Appraisal Estimate Actual/Latest Estimates Cat. Description Amount Percentage Amount Percentage (US $M) _ (US $M) 1. Civil Works 100% of Foreign, 100% of Foreign, A. Irrigation 1.1 85% of Local 0.4 85% to 14/8/92, B. Other Items 0.1 Expenditures 0.2 95% to 30/6/94 of Local Expenditure 2. Vehicles, Plant, 100% of Foreign, 100% of Foreign Equipment & Spare and 90% of Local & 90% of Local Parts Expenditures Expenditures A. Irrigation 0.6 0.1 B. Other Items 2.5 6.5 3. Consultants' Services 100% 100% Studies & Training A. Irrigation 3.0 3.9 B. Other Items 4.8 4.0 4. Operating Costs 80% 80% to 14/8/92, (Excluding Local 90% to 30/6/94 of Salaries) Local Opt. Cost A. Irrigation 0.5 0.1 B. Others 2.5 1.6 5. Refunding of Project 0.5 0.5 Preparation Advance 6. Unallocated 1.4 TOTAL: 17.0 16.8  18 Table 5: Key Indicators for Project Implementation Page 1 of 2 Project Component Expected as at Completion Actual A. STRENGTHENING OF PPMED Fully operational a. Strengthening of PPMED Reorganized and strengthened PPMED. PPMED. Refurbishing b. Refurbishing PPMED offices Refurbished offices. completed c. Housing for PPMED staff Full completion. Discontinued under the project as no guarantees could be given that only PPMED staff would occupy houses. B. STUDIES a. Review of agricultural credit Strategy for improved credit Completed system. b. Food security study Food security policy in place. Not completed. c. Action program of food security Food security and programs in Not completed. place. C. RESEARCH a. Review of agricultural research Research program/plan in place. Completed b. Prepare master plan for research Research program/plan in place. Completed c. Conduct research on cotton, Recommendation on useful Partially completed. rice, rrigation. technologies. D. EXTENSION a. Review agricultural extension Process of reform in place/unified Partially carried out. nation-wide. extension in place. Completed b. Pilot T & V Extension program Experience to be used for expanding the program nationwide.  19 Page 2 of 2 Proiect Component Expected as at Completion Actual E. IRRIGATION a. Preparation of National Water Master plan in place. Not completed. Master Plan b. Re-organization of GIDA Smaller and efficient GIDA. Completed c. Pilot small-scale irrigation Six schemes (300 ha) fully Four out of six completed. schemes completed. d. Training (overseas) of staff. Overseas and on-the-job training Essentially carried for GIDA staff. out. F. ANIMAL HEALTH a. Construction of cold storages Improved logistical support. Completed b. Establishment of a revolving Financially sound scheme. Completed fund for veterinary services. G. PRIVATIZATION OF FERTILIZER MARKETING a. Phased program of Completely privatized marketing. Completed privatization. No subsidy. Completed b. Phased elimination of subsidies. H. REFORM OF SOEs Phased program of reform of Activity Ghana oil Palm Dev. Co., Ghana discontinued Seed Co., Ghana Cotton Co., under this project. FASCOMs; Irrigation Co. (UR); Grains Dev. Board (UR); Ejura Farms, Crown Rice Mill; and Pomadze Poultry. 1. PRIVATIZATION OF Termination of government Completed TRACTOR SERVICES services; elimination of subsidies. Table 6: Key Indicators for Project Operations No indicators for project operations were established for the project period. For future operations, please refer to Borrower's Evaluation Appendix B (Attachment as Appendix 6B).  20 Table 7: Studies Included in Project Study Purpose as defined at Status Impact of Study appraisal/redefined 1. Agric. Extension Integration of services, budgetary National Agricultural and training requirements and Completed Extension Project is one linkages with adaptive research. outcome. 2. Agric. Research Review Review of agricultural research National Agricultural services and preparation of master- Completed Research Project is one plan, and research on cotton and outcome. rice. 3. Valley Bottom Rice Completed Research, and Water To improve rice yields. on four out Rice yields have improved. Resources Harvesting of six sites. and Management. 4. Medium Term Agric. Agenda for Sustained Growth Follow-on IDA and other Development Strategy and Development. Completed donors' assistance is within (MTADS). framework of MTADS. 5. Preparation of Designing of Livestock Services Completed National Livestock Services Livestock Project. Project was born out of the Development Strategy. study. 6. Pilot Study on Farmer To test the feasibility of sustainable Not completed, designed and operated small scale irrigation schemes attempted on None y et. small-scale irrigation. operated and owned by small 2 of 6 sites. farmers. 7. National Water To define a strategy for the medium- Resources Utilization to-long term development Not completed None Study/Plan. and use of water resources. 8. Agricultural Credit To identify steps to improve credit Completed Rural Finance Project Review to the sector. was born out of this study. 9. Food Security Study To identify actions needed to Not completed None improve food security. 10. Cotton Research To develop strains that are high yielding using minimum purchased Not completed None yet. inputs. 11. Study of Cotton To identify a strategy and actions for Industry possible reorganization of industry. Completed None  21 Table 8A: Project Costs Appraisal Estimate Actual/Latest Estimates (US $M) (US $M) Item No. _ Local Foreign Total Local Foreign Total Cost Cost Cost Cost 1. PPMED 0.2 2.0 2.2 0.2 1.0 1.2 2. MOFA Head Office 0.0 0.0 0.0 0.2 0.8 1.0 3. Agric. Res. 0.4 2.6 3.0 0.6 2.0 2.6 4. Fisheries 0.0 0.0 0.0 0.1 1.3 1.4 5. Agric. Extn. 0.7 1.5 2.2 0.5 1.3 1.8 6. Irrigation 1.0 4.2 5.2 1.7 4.7 6.4 Services 7. Veterinary 0.4 1.2 1.6 0.5 2.5 2.5 Services 8. Plant Prot. 0.0 0.0 0.8 0.1 0.4 0.5 9. CIDU 0.1 0.3 0.4 0.1 0.6 0.7 10. VORADEP 0.0 0.0 0.8 0.2 0.6 0.8 1I. P.C.U. 0.5 0.7 1.2 0.3 0.5 0.8 12. Ref. Project 0.1 0.4 0.5 - 0.05 0.05 Prep . Advance 13. Contingencies 0.6 1.3 1.9 - - 14. Vaccines & - 8.5 8.5 - 8.5 8.5 Drugs & KFW 15. Fertilizers & - 25.0 25.0 - 25.0 25.0 Chemicals AfDB TOTAL: I/ 4.2 49.1 53.3 4.5 48.7 53.2 I/Totals may not add up due to rounding  22 Table 8B: Project Financing Appraisal Estimate Actual/Latest Estimate (US $M) (US $M) Source Local Foreign Total Local Foreign Total Cost Cost Cost Cost IDA 2.9 14.1 17.0 3.1 13.7 16.8 Government of Ghana 1.3 1.3 1.4 - 1.4 UNDP - 1.5 1.5 - 1.5 1.5 KFW (Parallel) 8.5 8.5 8.5 8.5 AfDB (Parallel) - 25.0 25.0 - 25.0 25.0 TOTAL: 4.2 49.1 53.3 4.5 48.7 53.2  23 Table 9: Status of Legal Covenants Paze 1 of 3 Credit Agreement Reference Covenants Status 6.01 (a) I Signing of agreement between Government and Done UNDP on TA for PPMED, as condition of effectiveness. 6.01 (b) 2 Selection of TA personnel under UNDP Project to Done strengthen PPMED, as condition of effectiveness. 6.01 (c) 3 Appointment of Head of CIDU (also in charge of Done fertilizer privatization program), as condition of effectiveness. 6.01 (d) 4 Arrangement by MOA, satisfactory to IDA, for Decision taken to divestiture of SOE's (including establishment of a unit transfer this activity to coordinate the divestiture, and appointment of staff to Government's to the unit), as condition of effectiveness. overall divestiture program under SAC. 6.01 (e) 5 Establishment of the PCU, as condition of Done effectiveness; and maintain it thereafter. 6.01 (f) 6 Borrower to deposit in the Project Account 50 million Done credit initially, as condition of effectiveness. 3.01 (d) 7 Borrower to establish and maintain an Agricultural Done Policy Coordination Committee (APCC) by December 31, 1987. Sch. 5 8 The APC to meet at least once a year. Done Sch. 5 9 Borrower to continue to maintain PPMED, the Agricultural Research and Development Advisory Done Committee, the Crop Inputs Development Units, the Crops Production Department. Sch. 5 10 PPMED to review with IDA progress of fertilizer privatization program by December I of each year Done starting 1988. Sch. 5 11 MOA to review with IDA each year its annual public Done expenditure budget. Sch. 5 12 PCU to submit to IDA quarterly progress reports on the project by February 15, May 15, August 15, and Done November 15.  24 Page2of3 Reference Covenants Status Sch. 5 13 Borrower to reduce fertilizer subsidies to 30% of retail prices for 1988, 15% of 1989, and eliminate the Done subsidies starting 1990. Sch. 5 14 MOA to review with IDA, each year, prices of Done tobacco, cotton and palm oil. Sch. 5 15 PCU to submit to IDA by September 30 of each year Done annual work program and budget for the project. Sch. 5 16 Borrower to privatize MOA's fertilizer operations according to following schedules: a) test private retailing in Volta and Brong Done Ahafo regions in 1988; b) sell fertilizer only at its main distribution Done stores in 1989; c) sell fertilizer at its depots only in Tema, Done Swedru, Kukurantumi and Tamale in 1990. d) relinquish all its import and distribution Done programs in 1991. Sch. 5 17 Borrower to recover 100% of cost of veterinary drugs Done as from 1988. Sch. 5 18 Borrower to submit to IDA and other donors a master Done plan for agricultural research by June 30, 1988. Sch. 5 19 Borrower to complete study of agriculture extension Done by December 31, 1990. Sch. 5 20 MOA to discontinue its tractor services according to following schedule: a) wheel tractor services by December 31, 1987; Done b) combine harvester services by February 28, Done 1989; c) land clearing services by June 10, 1991. Done  25 Page 3 of 3 Reference Covenants Status Sch. 5 21 GIDA to prepare a national water master plan and Decision taken by submit this to IDA by December 31, 1990. MOFA and IDA to carrv this out later in the framework of Government's MTADP Sch. 5 22 Borrower to submit to IDA audit reports, including Complied, save for audit reports on project and special accounts, SOEs by some delays as in the June 30, of each calendar year. case of the 1989 report. Sch. 5 23 Borrower to carry out Agricultural Extension Study Unification of extension services of MOFA and Cocobod, which activity was to have been based on this study, was not effected. Sch. 5 24 MOA to carry out Food Security Study by December Study not completed. 31, 1989. First draft of the consultant report submitted on June 15, 1991 but not found acceptable. Sch. 5 25 MOA to submit to IDA action program on food Dependent on security by June 30, 1990. outcome of study on food security. Table 10: Compliance with Operational Manual Statements There was no significant lack of compliance with any applicable Bank Operational Manual Statements.  - 26 - Table 11. Bank Resources: Staff Inputs Stage of Planned Revised Actual Project Cycle Weeks USS2 Weks USS weesS Through appraisal na. n.a. n.a. n.a. 181.3 n.a. Appraisal-Board n.a. n.a. n.a. n.a. 37.7 n.a Board-Effectiveness n.a. n.a. na. n.a. 00.5 n.a Supervision 123.5 n.a. 235.3 n.a 173.8 n.a. Completion 20.0 23.8 20.0 23.8 24.4 23.8 TOTAL n.a. n.a. n.a. n.a 417.7 n.a. Table 12. Bank Resources: Missions Stage of Month/ Number Days Specialized Staff efirmance ti Types of Project Cycle Year of in Skills Implementa- Develop- Problems Persons Field Represented tion Status ment Persons Field Represented Objectives Through Appraisal 04/85-12/86. 24 151 REIAG/EX/FA/ * AEIMFAID Appraisal through 01/87-05/87 9 21 RE/FA/LEG/ _ Board Approval EXIAG/AE Board Approval through 06187-12187 2 0 RE/AE - - Effectiveness Supervision 03188g8193 22 145 RE/AG/EX/FA/ 1-2 1 Implem. AE Delays Completion 05/94-12/94 1 23 EC 2 2 I Also includes Bank-financed and trust fund consultants. 2 Available for FY 94 only. 3 Available for FY 94 only. 4 Key to specialized staff skills: RE - Rural Engineer, AG - Agriculturalist; EX - Extension Specialist; FA - Financial Analyst; AE - Agricultural Economist; EC - Economist; Me -Monitoring and Evaluation; ID - Institutional Development; LEG - Lawyer  APPENDIX A AGRICULTURAL SERVICES REHABILITATION PROJECT [ASRP] [CREDIT NO 1801-GH] IMPLEMENTATION COMPLETION MISSION AIDE-NMEMOIRE 1. Introduction. A mission consisting of Messrs George Alibaruho (World Bank), Allassan Salifu (Ghana Government [GOG], Ministry of Food and Agriculture [MOFA], ASRP, Project Coordinating Unit [PCU], Mark Owusuansah and John Poku (Consultants) undertook final implementation supervision of the ASRP project in conjunction with preparation of the Implementation Completion Report (ICR), from May 23, 1994 to June 16, 1994. This Aide-memoire is a record of: the main findings of the mission; understandings reached with ihe Government and implementing agencies on the main achievements of the project; steps to address some important problems identified in the field; the agreed framework and content of the ICR and its processing, including the key elements of the Government's post-implementation project operational plan, its contribution to the ICR and the consultation process with other financiers of the project. Prior to the mission's departure from Washington, parallel financiers of the project were informed by IDA of the forthcoming ICR mission. KFW contributed to the mission's field investigation program. No response was obtained from AfDB. Contact with UNDP was established in the field. 2. Staff and Officials' Responsibility. The draft Aide-memoire was prepared by George Alibaruho, Agricultural Operations Division of the Africa region, assisted by other members of the mission referred to in paragraph 1. It was provided .o the heads of all implementing agencies, the Hon. Minister of Food and Agriculture and concerned officials of the Ministry of Finance, Development and Economic Planning for their comment. The Aide-memoire was finalized, taking into account comments by Ghanian officials and other World Bank staff. the draft ICR will, in addition, be reviewed by the Project Advisor, Africa Department IV, World Bank. The officials of KFW, AfDB and UNDP will be given an opportunity to comment. 3. Work Plan and Data Sources. Preparation of this Aide Memoire was begun in Washington as a step in drafting the terms of reference for the final supervision and implementation completion mission. The terms of reference for each member of the mission were discussed and agreed between the staffs of IDA and GOG and the two consultants to the mission. They are attached in the annexes. The Aide Memoire is based on: material in the project file; field visits to project sites; interviews of beneficiary individuals and groups; responses to a structured questionnaire administered to all implementing agencies; one-on-one and group discussions with officials of implementing agencies and selected major civil works contractors and the project's auditors. Discussions were also held with the staff of UNDP and FAO, Accra field offices, who are responsible for project affairs. At FAO, the staff who were responsible for executing the UNDP-financed technical assistance were still in post. The mission's program of field visits and schedule of interviews and meetings are attached in the annexes. The consultants are expected to complete their work and submit their draft report to GOG with a copy to IDA on July 15, 1994. The GOG's contribution to the ICR is expected to be sent to IDA a month later. The completed draft ICR, including GOG contribution, will be sent by IDA to GOG, AfDB, KFW and UNDP for comment and a final date for reply will be indicated then.  -2- 4. Project Cost and Financing Plan. The ASRP was costed at $53.3 million at appraisal, of which $17.0 (SDR 13.3 million) was to be provided by IDA as Credit 1801-GH, which was approved by the Board on April 29, 1987. Parallel financing was to be provided by Kreditanstalt fur Wiederaufbau (KFW), ($8.5 million equivalent); the African Development Bank/Fund (ADB/ADF), ($25.0 million); and the United Nations Development Program (UNDP), ($1.5 million). The GOG was to contribute $1.3 million equivalent as part local cost financing. The project was re-designed at mid-term, as the data in paragraphs 6 and 7 and appendix 1 tables show, to support additional departments of MOFA (Fisheries; Plant Protection; VORADEP; and MOFA head office) and to reallocate funds between expenditure categories). 5. Financing and Disbursement Status. The IDA credit became effective on December 18, 1987 and closed on June 30, 1994, instead of the scheduled date of June 30, 1993. The credit will be fully disbursed by the closing date (appendix 1, p.2 and p.3). The KFW credit closed early in 1994 and was fully disbursed. The ADBIADF loan and credit as well as the UNDP grant closed in 1993 and 1991 respectively, and were fully disbursed. At completion of implementation and final disbursement of external loans and credits, the ASRP cost $53.2 million, just about what was estimated at appraisal. 6. Actual Proiect Costs and their Allocation between Components. At the closing date, actual costs are estimated at $53.2 million equivalent (appendix 1, p.1). The shares of each component, compared to appraisal estimates in the SAR/Credit Agreement (in parentheses), are: Policy, Planning, Monitoring and Evaluation Department [PPMED], $1.2m ($2.2m); MOFA Head Office, $1.0m ($0.0m.); Veterinary Services Department [VSD], $2.5m ($1.6 million); Fisheries Department [FD], $1.4m ($0.0m); Plant Protection Department [PPD], $0.5m ($0.Om ); Agricultural Extension [AED], $1.8m ($2.2m); CSIR-Agricultural Research, $2.6m ($3.0m); Ghana Irrigation Development Authority [GIDAJ-Pilot Irrigation Schemes, $6.4 m ($5.2m); Volta Region Agricultural Development Project [VORADEP] -On-farm adaptive trials, $0.8m ($0.0m); Crops Services Department [CIDU], $ 0.7m ($0.4m); Project Coordination Unit [PCU], $0.8m ($1.2m); refund of Project Preparation Advance [PPF], $0.05m ($0.5m); and, Contingencies, $0.Om ($1.9m). The costs of animal vaccines and drugs, financed by KFW, and of fertilizer and chemicals, financed by ADB/ADF were $8.5m and $25.0m, respectively, within the loan limits. 7. Expenditure Categories. The project financed: technical services, including training, $7.9m compared to $7.7m in the Development Credit Agreement [DCA]; equipment, mainly vehicles, motor bikes, computers and other office and field equipment, laboratory, diagnostic, measurement, training and cold storage equipment, $6.6m, compared to $3. lm in the DCA, the increase being largely on account of vehicles, 115 vehicles at a cost of $1.42m; incremental operating costs, excluding local salaries but including consumable items such as office supplies and fuel, $1.69m, against $3.Om in the DCA; works, including physical rehabilitation of office, staff residential facilities, laboratories, irrigation canals and land preparation, $0.6m, compared to $1.3m in the DCA. Procurement of technical services and equipment, largely vehicles, accounted for over 80 percent of the total IDA credit. Procurement of works was less than half of what was planned; and for incremental O&M, only 56 percent of planned expenditure took place (Appendix l, p.4).  -3- 8. Overall Project Objectives. The ASRP was one in a set of operations in support of institutional reformns in the Government's economic recovery program, whose anchor was a series of Structural Adjustment IDA credits. The basic objective of the ASRP was to initiate a self-sustaining process of reforms and rehabilitation so that the Ministry of Food and Agriculture could give more effective support to agriculture through: strengthening the institutional framework for the formulation, implementation, monitoring and evaluation of agricultural policies and programs; rationalizing the range of public sector services and activities, focusing on privatization of fertilizer, tractor hire services and veterinary drugs and privatization/reform of agricultural state operating enterprises (SOEs); developing a medium-to-long-term development strategy for agriculture, through a series of studies leading to sectoral and sub-sectoral plans, programs and specific projects; improving the quality and delivery of public sector services to the farming community, focusing on research, extension, irrigation and veterinary services. 9. Proiect Component-specific Obiectives. Beneficiary agencies and their specific objectives were: Policy. Planning. Monitoring and Evaluation Department, to enhance its capacity for formulating sectoral policies and programs, monitoring and evaluating their impact; MOFA head office, to strengthen its role in policy and program coordination and strengthen decision-making process; Manpower and Training Department, to facilitate and. provide ministry-wide training and improve its effectiveness in rendering human resources development services; Crop Services Department, to promote privatization of fertilizer marketing; Council on Scientific and Industrial Research. to prepare a master plan for agricultural research and, through the Crops Research Institute, conduct research on cotton, rice and irrigation water use on inland valleys bottoms, in six different ecological zone settings; Ghana Irrigation Development Authority, to improve its capacity for effective development of sustainable irrigated agriculture, taking advantage of the experiences of a similar organization with a good track record in another country with comparable conditions through an appropriate twinning arrangement; and to design, in consultation with farmers, and pilot-test the participatory establishment of farmer-owned and managed small-scale irrigation schemes at six sites in different agro-ecological zones; A2ricultural Extension Department. to carry out pilot programs in agricultural extension; Livestock Department. to improve mobility, field logistics, availability and privatization of selected veterinary drugs; Fisheries Department, to improve mobility and logistics; Plant Protection Department, to improve mobility and logistics, pest control and seed-testing capabilities, and staff skills enhancement; and, the ASRP Project Coordinating Unit (PCU). to facilitate project coordination and administration, including procurement. Several Studies were also included in the project. Their execution was assigned to different project entities, depending on the subject, and were to formulate future projects and programs. 10. Realism and Consistency of Obiectives. The rehabilitation and institutional development objectives of the project were well conceived and consistent, focussing on the then key activities of MOFA. During the early implementation years, the changing country circumstances brought additional activities into focus in the MOFA, thus making it necessary  at the mid-tern review to expand the project objectives and redesign it to address adequately the additional priority activities. On the implementation aspects, the tendency, at first, was to centralize critical operational decisions at the PCU level, primarily because of inexperience or lack of consideration for basic measures to ensure effective and timely implementation decisions. This was resolved by allocating to the implementation units all the implementation decisions, and allowing them the authority to control their approved budgetary resources. The PCU was then able to attend more effectively the procurement and accounting functions. Most of the implementing agencies were able to attend more efficiently to their operations. The result was that the expected follow-up from the pilots materialized as expected in most cases. 11.Achievement of Project Objectives: Overview. Overall, achievement of the project objectives has been good. However, in some programs the benefits would increase in time, as resources and supervision inputs are made in the post-implementation phase. Sustainability of project benefits for components completed during the implementation stage appears generally to be firmly placed. Although the project had fairly disparate components, the major objective was institutional strengthening. The institutional strengthening and capacity building objectives were largely achieved, although implementation of some of the components was not wholly satisfactory. Component-specific objectives and the extent to which they were realized are presented in more detail below. 12. Realization of Component-specific Obiectives. Substantial success was achieved in realizing the objectives of the rehabilitation, institutional development and logistical support to MOFA and its technical and services departments. Privatization and commercialization of selected agricultural services and enterprises has been partial; the fertilizer subsidy was officially eliminated in 1990, and in practice, only a very small element of indirect subsidy is implied in the retail marketing arrangement; tractor hire services have been privatized; supply and distribution of veterinary drugs, except for vaccines, was also fully privatized; privatization of agricultural SOEs ceased to be a project objective as the activity was transferred to the Divestiture Committee, which has broader mandate to privatize/divest all SOEs, including those outside agriculture. The crops research objectives have been partially realized, with rice research component clearly more successful than cotton. Realization of the objectives of pilot-testing extension modalities has been substantial. In contrast the implementation of the small-scale pilot irrigation schemes was unsatisfactory. The objective of testing the ability of small farmers to establish with their own resources and successfully manage sustainable irrigation facilities was poorly executed by GIDA. The objectives of pilot-testing rain water harvesting practices (low risk rice cultivation trials), in conjunction with low cost small scale irrigation on some sites (valley bottom rice cultivation), have been substantially achieved. Studies supported under the project appear to have largely fulfilled expectations but some follow-up is needed to translate them fully into beneficial programs. The objectives underlying the setting up of the PCU, to provide administrative services, particularly procurement, accounting and financial reporting services and promoting cohesion among project entities were substantially met.  -5 - Project Activities, Implementation Record and Major Factors Affecting the Project General Factors Affecting the Proiect: 13. Scope and Design. While the original design of the ASRP provided for capacity building in several departments and sections of MOFA, the project start-up and early stages of implementation would not would not have suffered if the respective implementation agencies were not unduly constrained by the PCU. By the time of the mid-term review, sufficient progress had been made to allow some attention to the new priority areas of MOFA, including the Head Office; Departments of Fisheries, Plant Protection and Regulatory Services, and Human Resources Development. However the early implementation bottlenecks caused by an inexperienced PCU led to delays in full implementation of the pilot components which were to be in the field. In the case of the small scale irrigation pilots, the circumstances were more serious because of the inability of GIDA, the implementing agency, to undertake the program as intended. This could have stemmed from difficulties in accepting policy changes and lack of confidence in adopting new operating parameters for developing sustainable irrigation practices. The need to complete as many of these field activities was a major factor in the extension of the credit's closing date to June 1994. While this compares favorably with other projects where more than one extension was granted, it should be noted that some of the irrigation pilot schemes and the cotton research at Nyankpala, including the fibre testing facilities, were still incomplete at project closure. 14. Bank/Borrower Responsibilities. The delay in credit effectiveness and start-up activities and the relative slow pace of implementation were a result of: lack of familiarity by the Borrower and implementing agencies, particularly the nascent PCU, with Bank procedures, including managing procurement procedures; long turn-around time for procurement and other documents sent to Washington for clearance or other action; and inadequate provision of counterpart funds. The several changes in Bank supervision staff added to the problems, with lack of continuity, and differing opinions on the basic concept and design of the project components. 15. Proiect Coordination and Administration. The PCU, which was to provide essential support services to all project entities, including facilitation of compliance with Bank policies and procedures, should have been fully functional before starting the project activities. The staffing and training the unit's key personnel in Bank procedures, particularly.procurement and disbursement, took longer to accomplish than was expected. Additionally, the terms of reference of the PCU did not always seem to be well understood by all parties. Some project implementation agencies felt that the PCU was being obstructionist, and not facilitative or supportive, while the Unit tried to maintain central control with the object of ensuring that the correct procedures were followed regardless of any delays which may occur as a result. The unit was slow in appointing a financial controller and so lacked, in the start up period, the capacity to discharge adequately its accounting, financial and procurement functions. Component-specific Activities and Factors Affectine them 16. MOFA and its Services Departments: The more successful aspects of the ASRP include the policy analysis and decision-making processes, programs and project coordination, implementation , monitoring and evaluation, facilities rehabilitation and provision of mobility, office and field equipment to MOFA and its services departments and other units. The  -6 - Agricultural Policy Coordinating Committee of MOFA established with support by the project, is functioning as intended. The strengthened PPMED has improved its performance substantially and is funcfioning very well at MOFA headquarters; the staff have received the required training and are putting it to good use in data collection, including conduct of field surveys, and data analysis and interpretation and publication of results. This improved performance is facilitated by the computing and other office equipment and related facilities and staff support. With PPMED staff attached to all MOFA departments, coordination and linkage with on-going programs and projects in all sub-sectors has improved. On the problems side, refurbishing residential bungalows for the senior staff, did not take place. As these were in the general government housing pool, there was no guarantee that after renovation, the units would have been allocated to PPMED staff. Furthermore, partly because of poor housing and better opportunities in the local private sector and overseas for trained people, staff turnover has been larger than normal and has tended to undermine the sustainability of project benefits. 17. Irrination Agronomv and Efficiency Trials. This program of work was only partially fulfilled. This has been due partly to the lateness in starting. A contributory factor to this situation was the issue of whether or not the program should have been the responsibility of CRI or GIDA. The issue was not resolved until 1989, two years after credit effectiveness. The conduct of research field trials, by its nature so heavily influence of the vagaries of the seasons, requires adequate time to ensure that the results are replicable and would allow sound conclusions. Nevertheless, the findings so far are showing that the low risk water harvesting, rain-fed, rice farming practices developed under the project ajre offering good prospects for raising efficiency of rice production in the north of the country. Farmers are enthusiastic and yields as high as 3.0 to 4.0 tons/ha have been recorded on trial sites, compared with 1.0 ton/ha. in neighboring farms using the same seed variety with unimproved water management. The valley bottom water management technology development and testing for improved cropping was conducted by CRI in the Kumasi and outlying areas also with very promising results. Farmers have been highly impressed with the technology and are already organizing themselves to adopt it where possible. As it entails the need to divert perennial streams water, especially in the dry season, or use of pumps, the cost effectiveness of this technology needs to be further evaluated . 18. Rice Research. This program focused on developing and testing appropriate agronomic practices for smallholder application. It included soil fertility and water use responses, tillage and crop rotation practices, variety trials and pest management methods. The research was managed by the CRI at Besease, near Kumasi. Thus this activity was closely linked to pilot water use in valley bottoms. The work was replicated on only four of the planned six ecological sites, because it was started 2 years behind schedule due to the an then unresolved issue on responsibility between GIDA and CRI for the conduct of the work. The results of this early work shows good promise and CRI is well on its way to working out applicable agronomic practices for small farmer rice production in the valley bottoms. There is need to continue the program, to test the findings under other representative ecological zones and conducting "on-farm" trials to generate realistic costs and assess the-commercial viability of the responsive production operations. This should be undertaken in the post-implementation phase. 19. Cotton Research. The program covered a range of agronomic trials to determine the optimum recommendations for .smallholder producers for high yielding varieties, and best  -7- management practices for soil tillage, its fertility and moisture levels including their conservation, and pests and diseases. It also provided for training of staff of the cotton buying company; and for refurbishing and equipping the laboratory, including provision of cotton fibre testing equipment for quality control. On-station trials were conducted in the vicinity of the station at Nyankpala, at Wa in Upper West Region, Manga in the Upper East region and Damongo in the Northern Region. The on-station yields ranged from 1.0 tonlha of seed cotton to 2.0 tons/ha. No "on-farm" trials have been carried out to test the findings of the trials on a practical scale and this will be necessary in order to be able to make appropriate judgements oti their commercial viability and to formulate sound recommendations to farmers. 20. Cotton Fibre Testing Laborator . The computerized fibre-testing laboratory equipment was ordered in December, 1991, and delivered and installed by the supplier in October, 1993. It was not operational at the time of preparation of this report but it is understood that the equipment operated one week after it was installed and has not had required follow-up service by the suppliers. The cotton buying company confirmed that the fibre-testing service is needed so that the ginneries could specify the fibre characteristics of their product and more and attend to proper labelling of the cotton bales prior to selling, especially in the export markets. The training received by the ginnery staff cannot be put fully into use until the fibre-testing laboratory becomes fully operational. 21. Or2anization of the Cotton Industry. The project paid for consulting services to study the structure and organization of the industry and present a plan and recommendation for its possible restructuring A key objective of such restructuring and re-organization was to bring the benefits of the research financed under ASRP to the small farmer. The consultant's report confirms that, without changes, the small-scale cotton farmers will continue as seasonal contract laborers for ginneries, dependent largely on them for the supply of inputs, including credit. They are paid for their seed cotton the equivalent of only 7.0 percent of the f.o.b price for the combined lint and seed products after ginneries have taken into account the value of input supplies and other support services rendered. Analysis of returns per/man-day is currently being undertaken. However, the widespread practice of diverting ginnery-supplied fertilizer to non-cotton crops (rather than the intended cotton) may be an indication that returns per man-day are lower in cotton than in other crops where effective competition is ecologically feasible. Larger farmers, who source their own inputs, independently of ginneries and pay a ginning fee (about 120 cedis/kg) receive 91.0 percent of f:o.b price of lint and seed products, or thirteen times what small-scale farmers receive for comparable grades. Even when the cost of purchased inputs is removed from the large farmers' revenues from the crop, inputs, the disadvantage of the small farmer in the arrangement is indisputable. At current market prices, the technologies being tested, while commercially beneficial to larger farmers and ginneries, offer relatively little benefits to small farmers unless the industry is re-organized to enable them to sell their output as lint and seed, and pay the ginneries the at the going rate for ginning. Follow-up by GOG is recommended so that the benefits of this project activity may be fully realized and sustained. 22. Capacity Building of GIDA. This project component, representing a substantial part of the expenditure out of the IDA credit, started very late. It involved a two-pronged- approach to strengthen the skills of GIDA staff to adopt more appropriate pblicies and practices for promoting the development of sustainable irrigation programs. In its first decade of irrigation development Ghana concentrated on establishing water impounding reservoirs or  -8 - pumping schemes for medium to large scale irrigation schemes and neglected to pay the necessary attention to the farming requirements and farmers' interests and concerns. The result was costs'schemes, with very little down stream development of sustainable irrigated lands which could be productively managed by farmers. Thus the main consideration in the ASRP was a shift to low cost small scale systems, with direct involvement of farmers in assuming the risks of ownership and operation of their irrigation facilities. This new direction in policy called for substantial change in the management and operations of GJDA, requiring commensurate training of the staff in the required skills. The program for GIDA capacity building was to be two-pronged. One was a twinning arrangement with a similar irrigation authority which had demonstrated a good track record in a country with comparable resources and other relevant conditions. The other prong was the pilot testing of farmer-owned and operated small scale irrigation systems - giving GIDA staff the opportunity to work with farmers in identifying potential schemes; organizing and training farmer groups for assuming the responsibilities for developing and operating their schemes; drawing up the designs and agreeing with them on the implementation, operating and financing procedures. 23. The Twining Arrangement. The twinning benefits were only moderately successful. Communication with staff, while adequate at the senior management level, left much to be desired at the professional and technical staff level. Substantially better results could have been achieved if there was greater cooperation between the technical staff of the partners. Cultural and communication problems, including unfamiliarity with the English language and hierarchical barriers to open discussions, made the training and skills transfer difficult to accomplish. Furthermore, GfDA professional staff intimated that the arrangement was conducted in a client/consultant-type relationship, rather than as a " partnership" aimed at transferring specialized skills and experience, where basic technical knowledge is presumed of all engineers. The relationship resulted in a degree of indifference between the two groups of staff. There was also a perception among national experts that in any event, the designs of the "experts", whc knew less of the local farming and crop cultivation methods, carried the day, although their designs were based more on a Korean-type agro-sociological environment: more capital-intensive and with a focus on rice irrigation in a mono crop culture. On the positive side, GIDA staff claimed that the study periods in South Korea and elsewhere (part of the intended training) were beneficial, particularly the level of technical information received. Nevertheless the professional environment described above did not seem to have promoted GIDA "ownership" of designs that may have been prepared by ADC staff, but which had to be supervised later by GIDA staff. On the other hand, substantial benefit was derived from the practical training of farmers by staff of the twinning partner, in proper irrigation practice on some of the existing schemes. As a result, this part of the program was extended to reach more farmers. 24. The Pilot Small-scale Schemes. Since the conduct of these pilots was to be associated with the twinning program, it had to await the selection and recruitment of the twinning partners. Decisions on this by GOG proved very difficult and the process became unnecessarily protracted. The result was several years delay in starting the pilot schemes. The performance of GIDA in undertaking these pilot schemes was poor. They neglected to observe in the exercise the basic concepts underpinning the new policies of GOG to foster irrigation development. The major deficiencies were (a) inadequate or no involvement of the cooperating farmer groups in the identification, planning and designing, (b) neglecting simple, inexpensive systems which are affordable and manageable by the farmers, (c) the implementation being undertaken by contractors and not by farmers, (d)  9 GIDA staff not providing technical guidance during the construction and in the early stages of operation and maintenance. While six pilot sites were selected, only two reached the implementation stage. The designs were inappropriate in parts, and unnecessarily complicated. GIDA, contrary to the pilot requirements, chose to implement by contract under local competitive bidding. The selection process was subject to delays in decisions. The contract prices were much higher than expected and threatened the financial sustainability. Neither contractors showed the required competence for such work and both experienced serious problems. In one case, they relied on heavy earth moving equipment for the canal networks. The Fround conditions were too wet for such equipment which were also old and in poor mechanical condition, resulting in frequent break-downs and loss of time. Eventually, they had to swit^h to manual labor. At project closure, the farmers had only one year's experience in op-erating the scheme. They will require a few more years to demonstrate their competence in operating it and achieving economically viable results. The contractor on the second scheme was unable to complete the works and GIDA staff were left with the task of completing them. By this stage the cost overruns had become prohibitive and cost cutting measures adopted had to focus on reduction in design standards while maintaining original design area. This led to discontent among farmers who preferred to have concrete lined instead of clay lined main canals which required more attention to their maintenai,ce. The lessons to be drawn from the experiences with these two schemes include (a) manual methods are more appropriate and efficient for small scale systems (b) farmers must be involved in their projects from the conceptual stage and must undertake as much of the implcmentation works as possible and (c) that IDA should have taken a firmer stand against the changes by GIDA in the agreed implementation guidelines. 25. Extension. Under the pilot extension activity, eight districts in four regions, reflecting different crop combinations or the dominance of a single crop (as in the case of cocoa), were identified to try out different modalities for reaching the farmer with improved technical packages. MOFA Extension Department and COCOBOD, in the case of cocoa, were the implementing agencies. The mobility of extension staff was facilitated by the provision of motor bikes and pedal bicycles. The point of departure for testing extension modalities was the well-known Training and Visit system but during implementation modifications were made in training frequency, from a bi-weekly frequency to a monthly cycle. The research/extension/farmer linkage was strengthened. A more efficient Unified Extension mechanism was established, as the demand on the farmers' time by different subject matter extension staffs was curtailed. The system worked so well that it was replicated in four other districts during implementation. The objectives of this activity were-substantially realized, leading to the nation-wide National Agricultural Extension Project. The dichotomy of an extension system run by COCOBOD, in addition to MOFA's national system, still exists. This duplication needs to be sorted out if there is to be a really unified system. 26. Privatization of Fertilizer Trade. Tractor Services and Veterinary Drugs. The rationalization of services and production activities of MOFA, through privatization of input supply services (fertilizer, tractor hire services and some veterinary drugs), has had mixed success. The privatization of tractor hire services was completed but that of fertilizer trade was not. Whereas only the private sector imports fertilizers of the types specified by and with the permission of MOFA, they appear to act as wholesale and retail agents of MOFA in the local distribution and do not take the full business risks of trading in the products. MOFA explained that the capital base and level of risk-taking in private fertilizer trade is beyond the means of Ghanaian businessmen. Thus private traders have not been able to supply the  - 10 - product to all farmers at affordable prices, considering the returns to small farrm agriculture, especially in remote areas. Thus occasionally, MOFA "assists" dealers in the form of transport and sometimes warehouse rental (MOFA still owns the warehouses but leases them to the dealers/agents). There is also an element of guaranteed fertilizer purchase by MOFA from the agents, to build MOFA's strategic buffer stock, suggesting that some indirect subsidies may not have been eliminated on fertilizer trade. NWhile much progress has- been made towards realization of project objective, input supply services are yet to be fully privatized and the sustainability of the benefits and actions taken so far has not been fully tested. 27. GOG Concerns on full Privatization of Fertilizer. MOFA expressed anxiety at the "negative" impact of the fertilizer privatization policy, explaining that the private sector has not been as responsive in providing the product as was anticipated. Being a seasonal product with a short shelf-life, especially in light of inadequate storage facilities of potential dealers, and considering the risk of tying-up working capital at high commercial interest rates if dealers were to miss one crop season and have to wait for as long as one year to off-load stock on hand, the fertilizer trade could be very risky. Hence the claim that the private sector has not been responsive. It would also appear that net incremental financial benefits of fertilizer use have been seriously eroded by recent fertilizer price increases in local currency due to devaluation and inflation, relative to domestic price increases of rice and other non- export crops. Evidence was cited that in the last three years, market fertilizer prices have risen more than twice as fast as those of rice, hence the reduction in fertilizer use and yields. Additional data to support further analysis of these issues, including nation-wide fertilizer use before and after privatization, are being compiled by the Borrower, through private consultants. 28. Privatization of SOEs. Shortly after ASRP start-up, the GOG initiated a program of divestiture of State-owned Enterprises (SOEs), and this led to MIOFA transferring the responsibility for privatization of agricultural SOEs (18 enterprises had been identified in the SAR) to the new agency established for the purpose. Thus ASRP took no further part in the privatization exercise. 29. Studies. The objective of developing a strategy for the development of the agricultural sector through a series of studies leading to sectoral and sub-sectoral plans and specific projects appear to have been substantially realized. The ASRP is generally described as "the seed that produced all agriculture projects", following the resumption of IDA lending to Ghana. The ASRP pilot programs, studies, sectoral and sub-sectoral plans and preparatory activities led to the Medium Term Agricultural Development Strategy and the National Agricultural Research Master plan in which context the following IDA-assisted projects have been designed and are under implementation: (a) the National Agricultural Research Project (Cr. # 2247 GH); (b) the National Agricultural Extension Project (Cr. # 2346 GH); (c) the National Livestock Project (Cr. # 2441 GH); (d) the Agricultural Diversification Project (Cr. # 2180 GH); and (e) the Agricultural Sector Investment Project (Cr. # 1802 GH). Other follow-on proposals in the current investment pipeline include a National Fisheries Project. Several development agencies and co-financiers have also based their sectoral assistance strategy and plans on the studies, pilot projects and programs tested and developed under the ASRP.  - 11 - 30. Project Sustainability. The sustainability of benefits from project activities, based on the design and performance so far, of follow-on projects, appears well founded. The capacity improvements of MOFA and its service and technical departments, the organizational and structural changes and logistical support that have improved coordination are likely to be well sustained if GOG can generate and allocate to them, the required budgetary resources. Maintaining a reasonable level of periodic investment in vehicle and equipment replacement and provision of adequate O&M funds are critical. In terms of retaining the benefits of staff trained under the credit, their environment and working facilities need to remain serviceable and adequate, rerovating or replacing them as needed, and their remuneration must be attractive. In the final analysis, these are budgetary matters. On fertilizers, the benefits from the policy chang;!s supported under the credit are debateable, from the point of view of GOG. The same is true, though to a lesser extent, about the privatization of veterinary drugs, due to the inability of private sector to penetrate remote areas where the industry is not commercialized. The benefits from the refrigeration and field equipment for veterinary services are protected by the operation of the revolving fund established under the project (KFW funds), whose operation will continue beyond the implementation phase. For project activities not completed or tested, it is too early to tell whether expected benefits will accrue, let alone be sustainable. This includes the pilot irrigation program and much of the research, especially cotton . Fu,r rice research, the identified benefits need to be confirmed by more testing and economic analysis to ascertain their sustainability. Follow-up activities, such as re-organization of the cotton industry, need to be pursued to realize and sustain project benefits. 31. Bank Performance. While its performance was, on the whole, responsive and satisfactory, it appeared to have been somewhat optimistic about the pace at which the implementing capacity could be built up. The objectives and scope of the project were perhaps ambitious, given that proven capacity to implement was weak at the start and it had to be built up by this very project. Staff could have given greater consideration to phasing of project activities over a longer period or designing two separate projects, with the institutional development/technical capacity build-up coming before the field/pilot investment activities. However, IDA realized later that the Borrower's implementation capacity growth was over-estimated and maintained flexibility with the borrower, with re-design of the project at mid-term, re-allocation of funds between categories, financing a higher percentage of project costs and extending the closing date for the credit. Shortening the turn-around time for procurement and other necessary consultations between implementing agencies and the Bank became necessary to accelerate implementation. IDA responded by transferring the task-management of the ASRP to the Accra Resident Mission. while this did not immediately have the expected results, the actions IDA at the mid-term and subsequently led to substantial improvements in the implementation performance. 32. Borrower Performance. Performance of the borrower was at start-up only marginally satisfactory. The PCU and the implementing agencies were initially overwhelmed by Bank procurement, disbursement and financial reporting procedures. It took a while to overcome these teething problems. As the PCU became fully staffed and its staff trained-in the various skills, the Borrower's performance in project administration and implementation improved substantially. when the Bank delegated procurement-related decisions to the Accra Resident Mission, the GOG also took the exceptional step of exempting the project from using the procurement services of the Ghana Supply Commission, and further reduced the turn-around time. Some of the local delays were due not only to the PCU and the implementation  - 12 - agencies but to decisions outside of MOFA. It appeared that the Cabinet was not adequately briefed about the project. After negotiations and signing of the Credit Agreement, the GOG (or some senior officials of GOG) became somewhat uncomfortable with some of the provisions of the project. including the- proposals for capacity building in GIDA by a twining arrangement and privatization of import and distribution of fertilizers and other production inputs. These issues could have been raised earlier in the project cycle but were not brought t- the attention of IDA . Thus the long delay in signing the technical co-operation contract w,I ^.l in; construed to have been due to bureaucratic delays at the level of the GOG's chief legal officer, appears to have been more the result of discomfort at higher levels with the proposals. 3i. Assessment of Outcome. The outcome of the project is good. The rehabilitation, institutional development, capacity-building and logistics support activities in MOFA were successful in most of its service and technical departments and support agencies, the exception being GIDA and CRI whose programs were unduly delayed and incomplete at the end of the project. The impact of some of the technical assistance may be less than expected; for example, the restructuring the cotton industry has not yet shown the expected full benefits. The extension components were particularly well executed and showed early results. The least successful aspects of the project were the pilot irrigation program and some of the research. 34. Proiect's Future Operation. The GOG has indicated that the principles underlying the project's future operational plan are: to continue funding promising project activities that have not been completed as planned and are not funded under on-going successor projects; to continually review those activities whose end results may not justify additional investment, considering the resources sunk in the activity, the length of time the activity has been on-going and prospects for cost recovery, in the case of those activities where cost recovery was an objective: irrigation and water use; to provide adequate budgetary funding for O&M to maintain the value of the investment made, particularly vehicles, office and field equipment; to build in the budget and work plans, the periodic replacement of equipment so as to maintain and improve upon the level of service efficiency created by the project; consolidate administrative services for as many IDA-funded projects as possible under the PCU, which would be strengthened in a manner and time-frame to be agreed with IDA. Further details of the plan are attached (Appendix 5). 35. Key Lessons Learned. A major lesson learned is that simplification of project scope through adequate phasing or smaller multiple projects, should always be considered as an option where a key project objective is the strengthening of those institutions charged with the execution of investment activities in the project. The inclusion of expensive technical assistance arrangement in small dollar-value projects should be looked at carefully. In order to increase Borrower "ownership' of projects, as much formal and informal consultations should be carried out with all affected constituencies early in the project cycle. The Task Manager's counterpart should provide written confirmation that the green cover documents, including the draft DCA, have been reviewed by concerned government agencies and that the negotiating team is conversant with the project. The communication should also confirm that the team's positions at negotiations would represent institutional positions. 36. Appendixes. Appendixes contain: project costs, disbursement and financing tables (Appendix 1); the written contribution of KFW to the field investigation (Appendix 2); the  - 13 - ICR mission member's terms of reference (Appendix 3); the mission's schedule of field visits and meetings (Appendix 4); and, the GOG's draft post implementation Project Operational Plan (Appendix 5). 37. Issues Reguiring GOG Follow-up and Action. The suppliers of cotton fibre-testing equipment for the laboratory at Nyankpala or their agents, should be asked to correct the present deficiencies in the equipment in good time for the next cotton crop. Farmer groups at the Sata pilot irrigation scheme will need substantial technical guidance to manage their operations, and particularly to be able to meet the additional debt burdens brought on by the loans from the Agricultural Development Bank to continue the unfinished construction left by the contractor, on the understanding that they would be able to farm year round to derive the required additional income. Agricultural Operations Division Western Africa Dept. IV World Bank. August 15, 1994   APPENDIX B IMPLEMENTATION H COMPLETION REPORT GHANA AGRICULTURAL SERVICES REHABILITATION (