Ghana - Rural Finance Project

Document of The World Banlk FOR OFFICIAL USE ONLY Report No. 14603 IMPLEMENTATION COMPLETION REPORT GHANA RURAL FINANCE PROJECT (CREDIT 2040-GH) JUNE 14, 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.  CURRENCY EQUIVALENTS (December 31, 1994) Currency Unit = Cedi Exchange Rate (average annual): Appraisal: US $1.00 = 250.0 Cedi Average (1989 - 1994):US $1.00 = 499.0 Cedi Completion: US$ 1.00 = 956.0 Cedi WEIGHTS AND MEASURES Metric System FISCAL YEAR OF BORROWER January 1 - December 31 ABBREVIATIONS AND ACRONYMS ADB - Agricultural Development Bank ARB - Association of Rural Banks BOG - Bank of Ghana BSD - Bank Supervision Department (BOG) GCB - Ghana Commercial Bank CCA - Canadian Credit Association CGF - Credit Guarantee Fund (BOG) CIDA - Canadian International Development Agency CUA - Credit Union Association EFC - Export Finance Company GOG - Government of Ghana IDA - International Development Association MOFA - Ministry of Food and Agriculture NGO - Non-Governmental Organization RBD - Rural Banking Department (BOG) RFID - Rural Finance Inspection Department (BOG) RIR - Reference Interest Rate SAA - Subsidiary Administration Agreement USAID - U. S. Agency for International Development  FOR OFFICIAL USE ONLY GHANA RURAL FINANCE PROJECT (Credit 2040-Gm Table of Contents Pae Preface Executive Summary ................................................................... i Part I: Project Implementation Assessment ................................1......................... A. Project Objectives, Covenants and Financing .................... ............................. 1 Project objectives and components .................................1....................... 1 Covenants to achieve project objectives ....................... ............................. 2 Evaluation of project objectives . .......................................................... 2 Project costs and financing .................................................................. 2 B. Major Factors Affecting Project Performance ............................................... 3 Overview ................................................................... 3 Delays in achieving credit effectiveness ........................... ................ 3 Lack of interest by larger commercial banks ............................................ 3 Low ceiling on subloans ................................................................. 4 W eak capacity of project implementing agency ............... ........................... 4 C. Overall Project Implementation Performance ....................... ............................ 4 Overview ................................................................. 4 On-lending component ............................................................... .. 5 Institution building component ............................................................... 5 Technical assistance financed under the project ............... ........................... 7 Studies financed under the project . .......................................................... 7 D. Project Sustainability ......................... ........................................ 7 Institution Strengthening ................................................................. 7 Rural banks ................................................................. 8 Subsidy Dependence ................................ ................................. 8 E. Bank Performance ................................................................. 9 F. Borrower's Performance .................................................................. 9 G. Overall Assessment of Project Outcome ........................... .......................... 9 H. Future Operations and Key Lessons Learned .................. ............................. 10 Future operations ............................... .................................. 10 Key lessons learned ............................... .................................. 10 This document has a restricted distribution and may be used by recipients only in the perfonnance of their ofricial duties. Its contents may not otherwise be disclosed without World Bank authorization. I  Part 11: Statistical Annexes Table 1 Summary of Assessments ......................................................... 11 Table 2 Related Bank Loans and Credits ....................................................... 12 Table 3 Project Timetable .................... ..................................... 13 Table 4 Credit Disbursements: Cumulative and Actual .............. ...................... 14 Table 5 Key Indicators of Project Implementation ....................... ...................... 15 Table 6 Studies Included in Project ..................................................... .... 17 Table 7A Project Costs by Category .......................................................... 18 Table 7B Project Financing ......................................................... 19 Table 8 Status of Legal Covenants ......................................................... 20 Table 9 Compliance With Operational Manual Statements ............. ...................... 22 Table 10 Bank Resources: Staff Inputs ......................................................... 23 Table 11 Bank Resources: Missions .......................................................... 24 Part III: Sector-Specific Data Table 12 Distribution of Subloans by Beneficiary Category .................. ........ ....... 25 Table 13 Distribution of Subloan by Gender .................................................... 26 Table 14 Summnary Statistics and Subsidy Dependence of Selected Rural Banks ....... .... 27 Table 15 Rural Banks: Total Deposits, Reserves and Advances, 1989 to 1993 ............. 28 Table 16 Repayment of Subloans under the Rural Finance Project ........... ................ 29 Table 17 Rural Banks: Training Courses Offered, 1991 - 1994 ................................. 30 Table 18 Rural Banks: Regional Distribution of Training .................... ................... 31 APPENDICES: A: Mission's Aide-Memoire B: Borrower's Contribution to the ICR C: Maps: No. 21317 - Distribution of Rural Banks  IMPLEMENTATION COMPLETION REPORT GHANA RURAL FINANCE PROJECT [CREDIT NO. 2040-Gi] Preface 1. This is the Implementation Completion Report (ICR) of the Rural Finance Project (RFP), for which Credit 2040-GH in the amount of SDR 15.2 million (US $20.0 million equivalent) was approved on July 28, 1989 and became effective on April 26, 1990. 2. The credit was closed on December 31, 1994, compared with the original closing date of December 31, 1992. Final disbursement was on May 8, 1995 and SDR 0.24 million which remained undisbursed was canceled. 3. The ICR was prepared by Rudolph A. Polson, Agriculture and Environment Division, West Central Africa Department, Africa Region, and was reviewed by Cynthia C. Cook, Division Chief, AF4AE and Franz Kaps, Operations Advisor, AF4. 4. Preparation of the ICR was begun during the Bank's final completion mission, January 18 to February 22, 1995. It is based on material in the project file, field visits to selected rural banks, and on interviews with government officials, participating financial institutions (PFIs), and project beneficiaries. The Borrower's contribution to the ICR was coordinated by the Bank of Ghana, which assigned three senior officials from its Rural Finance Inspection Department to participate in the final supervision mission and to provide information on the economic, institutional and social environment in which the project was implemented. These officials also coordinated the preparation of the Borrower's action plan for the future operations of the project. Evaluation of the project from the Borrower's perspective is attached unedited as Appendix B to the ICR.  !  IMPLEMENTATION COMPLETION REPORT GHANA RURAL FINANCE PROJECT [Credit No. 2040-GH] Evaluation Summary 1. Introduction. At the time of preparing the Rural Finance Project (RFP) in 1988, Ghana had undertaken substantial policy and institutional reforms. Macroeconomic policy reforms initiated in 1983 under the Economic Recovery Program (ERP) had earlier helped to restore fiscal and monetary discipline and to stabilize interest rates. These macro-level reforms were followed, in 1989, by financial sector reforms which also helped to strengthen the sector by improving the regulatory framework, restructuring financially distressed banks and improving the efficiency of resource mobilization as well as the allocation of credit. Because the latter reforms focused primarily on macro-level financial institutions, the purpose of the RFP was to extend financial sector reforms to the rural financial sector with the primary objective of strengthening the network of rural banks, credit unions and rural cooperatives in order to improve the efficiency of rural financial intermediation. The project was therefore consistent with IDA lending strategy in the financial sector which emphasized restructuring of these rural banks to enhance their capacity to mobilize increased local resources and to provide increased credit for rural investments. Proiect Objectives and Covenants to Achieve Obiectives 2. Project Obiectives and Components. The main objectives of the project were to: (i) expand rural productive capacity and employment by providing financing for viable subprojects through the banking system; (ii) implement a financial restructuring program for the 125 rural banks to enable them to become more efficient in mobilizing savings and delivering credit; (iii) strengthen the Association of Rural Banks (ARB) and the Credit Union Association (CUA) to promote intermediation in the informal financial sector; (iv) strengthen the capacity of the Bank of Ghana (BOG) for rural bank inspection and examination; (v) build an enhanced capacity for rural credit appraisal among the rural banks and the national-level banks; and (vi) build local capacity for research in rural finance policy. To achieve these objectives, the project had two distinct components: an on-lending component, through eligible participating financial institutions (PFIs); and an institution building component, comprising technical assistance to the BOG, the ARB, rural banks, rural cooperatives and credit unions [paras. 1, 2]. 3. Specific Covenants to Achieve Project Objectives. Specific covenants for achieving project objectives were: for credit effectiveness, signing of Subsidiary Administration Agreement (SAA) between the Borrower and BOG as the implementing agency; submitting to IDA the operational policies and procedures for the Rural Finance Inspection Department (RFID), the department within BOG responsible for project implementation; signing of a participatory agreement between the BOG and at least two PFIs; recruiting a Rural Finance Advisor and a Rural Bank Restructuring Advisor; submitting to IDA the restructuring plans for the Agricultural Development Bank (ADB) and at least 20 of the 125 rural banks; and establishing a rural bank Recapitalization Fund within the BOG with an initial sum in Cedis equivalent to US$300,000. Other agreements addressed issues of sectoral policy and overall project implementation [paras. 3,41. 4. Evaluation of Proiect Objectives. The project's objectives, as stated in the SAR, were clear, concise and realistic and were consistent with government's priorities for restructuring the rural  ii financial sector. Given the dismal state of the rural banks, these priorities consisted primarily of strengthening the 125 rural banks as the vehicle for promoting savings mobilization and credit delivery in the rural economy, while using the line of credit as an added incentive for participation in the institution-building program [para. 51. Project Implementation Experience and Results 5. Factors Affecting Achievement of Major Objectives. Key factors affecting the project were: (i) inability of the implementing agency to timely meet effectiveness conditionalities, thereby delaying project start for nearly a year after credit approval; (ii) lack of interest by the larger commercial banks to participate in the on-lending program, due in part to their traditional preference for trade and industry subsectors, the high costs of project funds (as measured by the reference interest rate), and the inherently high risks associated with agricultural credit; (iii) the low ceiling of Cedis 50.0 million (about US$50,000 at current exchange rate) on subloans, which effectively excluded larger borrowers; and (iv) slowness of the implementing agency to develop the capacity for effective project management during the initial two years [paras. 7 - 11]. 6. Project Sustainability. Some key project activities -- training of rural bank staff and directors, capacity enhancement at the RFID for inspection management audits of the rural banks, capacity enhancement at the ARB, and the development of rural cooperatives -- yielded benefits that have high probability of being sustained, and for which the Borrower, through its post-credit operational plan, has indicated a commitment for continuing support beyond the completion date. The project succeeded in conducting a comprehensive diagnosis of the constraints faced by the rural banking sector and developed an action plan for the restructuring of each rural bank. Although 55 of the 125 rural banks are currently rated satisfactory in accordance with the PNDC Law - 225, a majority of rural banks (including some from among the "satisfactory" category) are weak, with shallow capital base, poorly trained staff and weak internal controls [paras. 21, 22, 23]. 7. Proiect Costs, Financing and Timetable. At appraisal, the total project cost was estimated at US$38 million, comprising the on-lending component (US$28.million), institution building (US$9 million), and the Rural Finance Inspection Department (US $1 million). The financing plan was: IDA (US$20 million), GOG (US$1.1 million), PFIs (US$5 million), beneficiaries (US$3 million) and various co-financiers (US$8.9 million). However, co-financing of US$8.9 million did not materialize and project costs were accordingly adjusted by the amount of the expected co-financing, to US $29.1 million. By completion, the actual (latest estimate of) project costs was US$28.4 million, which was consistent with the revised financing plan established at credit effectiveness [para. 6]. 8. Bank Performance. On the whole, Bank performance, from project identification through completion, was satisfactory. Overall, the Bank was responsive to problems identified during project implementation, to which it responded adequately through increased supervision and consultation with the Borrower [para. 12]. 9. Borrower Performance. Despite the initial problems encountered by the implementing agency, overall assessment of Borrower's performance was satisfactory. The major implementation delays occurred prior to the project mid-term review (MTR), after which the implementing agency emerged as a cohesive team with shared vision and purpose [para. 13].  iii 10. Project Outcomes. On the whole, the outcome of the project was satisfactory. The project succeeded in channeling needed resources to the rural economy that helped, to a large extent, in increasing smallholders' incomes and in spurring rural investments, especially of community-level enterprises. More than 15,250 beneficiaries, consisting of 1,550 groups and 13,700 individuals benefited from the estimated Cedis 9.8 billion (about US$19.8 million at the average project exchange rate of Cedis 500.0). The main activities funded were poultry (Cedis 3.9 billion), trading (Cedis 2.1 billion), crop production (Cedis 1.9 billion) and cocoa marketing and estate rehabilitation (Cedis 1.2 billion). The average subloan size was Cedis 4.7 million for groups and Cedis 180,000 for individuals, indicating that loans were not concentrated among a few beneficiaries, such as parastatals. The project led to increased deposit mobilization and increased credit to the rural economy that would not have otherwise occurred. Total deposit mobilized by the rural banks increased from Cedis 4.6 billion in 1989 to Cedis 13.2 billion by 1994, the completion of the project. Loans and advances by the rural banks nearly doubled, from Cedis 3.7 billion in 1991 to Cedis 6.8 billion in 1994, compared to the Cedis 1.7 billion of external project resources channeled through the rural banks. The project strengthened key rural financial sector institutions: the RFID, in inspection management audits of the rural banks; the ARB, in training of rural bank staff and directors; and rural cooperatives, in inventory credit. Overall performance of technical assistance was satisfactory [para. 20]. Summary of Findings. Future Operations and Lessons Learned 11. Key Findings of Implementation Experience. The implementing agency was created specifically for the project, and lacked experience with Bank procedures. It also experienced frequent changes in senior staff. Given this, expectations of its initial performance were optimistic, especially with regard to timely compliance to effectiveness conditionalities. A delayed project start was the result. The well-balanced combination of training and regular inspection and examination helped in the successful use of technical assistance for strengthening rural financial institutions [paras. 11 & 22]. 12. Plans for Future Operations. The Borrower is committed to seeing the development of a strong rural financial sector, the centerpiece of which is the rural banking system, and has therefore, in its post-credit operational plan, identified the priority areas and the resources to ensure continuity in key project activities. Because the project focused on strengthening sector institutions, it single greatest contribution was that it helped to lay a good institutional foundation upon which future rural finance operations can build. However, in order to maximize development impact and to effectively address the needs of both formal and informal sector clients, follow-up rural finance interventions should be preceded by comprehensive diagnosis of the constraints, issues and priorities for addressing the credit needs of the rural economy [para. 24]. 13. Key Lessons Learned. The key lessons for future projects in the rural finance sector are: (i) training of rural bank staff and directors, perhaps more than any single technical assistance activity, has emerged as the most important ingredient for the development of a viable rural banking sector; (ii) in order to be effective monitoring tools, inspection mangement audits of the rural banks should be combined with regular feedback to the rural banks, with clear guidelines for correcting observed deficiencies, a time frame for addressing exceptions noted, and possible sanctions for failing to rectify deficiencies; and (iii) inventory credit has emerged as an innovative marketing tool crucial in linking the formal and informal sectors [para. 25].   IMPLEMENTATION COMPLETION REPORT GHANA RURAL FINANCE PROJECT [CREDIT 2040-GH] PART I: PROJECT IMPLEMENTATION ASSESSMENT Project Objectives. Covenants and Financing 1. Project Obiectives and Components. The objectives of the Rural Finance Project (RFP) were to: expand rural productive capacitv and employment by financing eligible subprojects in crop production, livestock, fisheries, artisanal agro-processing, input supply, etc. through eligible participating financial institutions (PFIs); implement a financial restructuring program for the 125 rural banks and strengthen the organizational and financial management of credit unions to enable them to become more efficient in mobilizing savings and delivering short and medium-term credit, especially to smallholders; * strengten the Association of Rural Banks (ARB) and the Credit Union Association (CUA), two key non-governmental institutions active in promoting financial intermediation in the informal financial sector; * build an enhanced capacitv for rural credit appraisal among rural and national-level banks; - strengthen the capacity of the Rural Finance InsMection Department (RFID) of the Bank of Ghana (BOG) to provide regular and systematic inspection management audits of the rural banks to ensure their compliance with the Banking Laws and other prudential guidelines; and build local capacitv for research in rural finance policy within the BOG and other sector institutions. 2. To achieve these objectives, the project consisted of two distinct components: (i) an on-lendin, wnw=, to support viable rural investments in the agricultural sector that would contribute to rural income enhancement and employment; and (ii) an institution building component, comprising mainly technical assistance for strengthening sectoral institutions to enable them to improve the efficiency of rural financial intermediation. Specific institution building objectives were: restructuring 80 of the 125 rural banks; strengthening the ARB and the RFID and rationalizing their roles; establishing capacity for rural credit appraisal among the PFIs; and establishing a pilot program of rural cooperatives in the Upper West Region. In order to implement the project, the BOG transformed its Rural Banking Department into a Rural Finance Inspection Department (RFID) and gave it an expanded mandate for rural finance and the statutory responsibility for inspection oversight of the rural banks, as opposed to the narrowly-defined objective of rural banking. Appointment of the director of the newly created agency was a condition for project negotiation.  2 3. Covenants to Achieve Project Objectives. The legal covenants of the project and their status are summarized in Table 8. Specific covenants for credit effectiveness were: signing of Subsidiary Administration Agreement (SAA) between the Borrower and BOG; submitting to IDA the operational policies and procedures for the RFID; signing a participatory agreement between the BOG and at least two financial institutions; recruiting a Rural Finance Advisor and a Rural Bank Restructuring Advisor; submitting to IDA the restructuring plans for the ADB and at least 20 of the 125 rural banks; issuing deposit guarantees up to a limit of Cedis 100,000 per depositor for rural banks depositors whose claims could not be met during the restructuring process; and, establishing a rural bank Recapitalization Fund in the BOG with an initial sum in Cedis equivalent to US $300,000. These covenants were largely met. Credit effectiveness was delayed for six months as the Borrower needed more time to meet these conditions, and in particular for completing the restructuring plans for the initial 20 rural banks. 4. Covenants specific to sector policy and project implementation required the Borrower to: abolish the Cedis 25,000 and 125,000 ceilings on shareholdings in the rural banks by individuals and companies, respectively; limit shareholdings in particular rural banks by individuals and companies to not more than 5 percent and 10 percent, respectively, of their share capital; abolish sectoral quotas imposed on rural bank operations, while maintaining initially a minimum of 20 percent of rural bank portfolio for lending to the agriculture sector; assign to special collection accounts all non-performing loans of the rural banks and require the rural banks to make commensurate provisions for bad debts; prepare a shortlist of consultants to provide loan support services to the rural banks; and complete the nomination of representatives of BOG to the boards of rural banks on which BOG was not fully represented. Other agreements required the Borrower's implementing agency to: prepare a monitoring system for the operations of the rural banks; complete staff development plans for the RFID; and recruit a Rural Bank Inspection Advisor. These covenants were also largely met. Based on implementation experience, two key covenants were abolished or substantially modified: that limiting shareholdings by individuals and companies in the rural banks to 5 and 10 percent, respectively was abolished to allow the rural banks to increase their share capital; and that requiring the BOG to appoint its representatives on the board of rural banks was repealed due to increasing conflict of interests of BOG representatives in their dual roles as individual shareholders (in some cases) and as officials of the oversight institution, the central bank. 5. Evaluation of Project Obiectives. The project objectives, as stated in the SAR, were clear, concise, and unambiguous. The project objectives were consistent with, and clearly reflected the government's priorities existing at the time of appraisal: to extend financial sector reforms to the long neglected rural finance sector; to strengthen the rural banking system comprising the 125 rural banks as the key to increased rural savings and credit delivery; and to address sectoral policy constraints -- such as the requirement that at least 20 percent of rural bank portfolio be lent to agriculture -- that had in the past, affected the efficiency of rural financial intermediation. 6. Project Costs and Financing. At appraisal, the total project cost was estimated at US $38.0 million, comprising the line of credit (US $28.0 million), institution building (US $9.0 million) and the RFID (US $1.0 million). The financing plan was: IDA (US $20.0 million), Government (US $1.1 million) the PFIs (US $5.0 million), Beneficiaries (US $3.0 million), and Co-financiers, notably CIDA and USAID (US $8.9 million). However, by project effectiveness, no firm commitment had been made by these agencies to co-finance the project, and the total cost was reduced by the amount of the unrealized co-financing. At effectiveness, the revised project cost was therefore US $29.1 million. By completion, the total actual (latest estimate of) project cost was US $28.4 million,  3 consisting of IDA (US $20.4 million), GOG (US $1.1 million), PFIs (US $3.9 million), and Beneficiaries (US$ 2.9 million), which was consistent with the revised financing plan established at credit effectiveness. Cumulative estimated and actual disbursements are in Table 4. The estimated and actual project costs are in Table 7A, with the estimated and actual financing plans in Table 7B. Major Factors Affecting Project Performance 7. Overview. The major factors affecting the project were: (i) delayed project start-up as the Borrower took nine months to meet conditionalities for effectiveness; (ii) lack of interest by the national-level commercial banks; (iii) the low ceiling on subloans, which effectively led to the exclusion of larger borrowers from participating in the project; and (iv) the weak capacity of the RFID, the implementing agency. Factors (i) and (iv) were largely under the control of the Borrower; factors (ii) and (iii) were problems inherent in the design of the project that were addressed during the Mid-Term Review (MTR) and through subsequent amendments to the Credit Agreement. 8. Delays in Achieving Credit Effectiveness. The credit was approved on July 28, 1989, but it did not become effective until April 26, 1990. Seven effectiveness conditionalities were stipulated in Sections 6.01 (a) to (g) of the Development Credit Agreement (see Table 8). Most of these conditions were satisfied not long after credit approval. The major delay was caused by Section 6.01 (e) which required the Borrower to submit, in a manner satisfactory to IDA, the financial restructuring plans for at least 20 of the rural banks. The RFID was slow in assembling a team of consultants to prepare the restructuring plans for the initial set of 20 rural banks, thereby delaying credit effectiveness for nearly nine months. Part of the reason was that the RFID itself had just been established and, lacking the experience and critical mass of staff, was therefore not fully operational at this stage to deal effectively with these conditionalities. 9. Lack of Interest by Larger Commercial Banks. The major commercial banks such as Barclays Bank, Standard Chartered Bank (SCB) and Ghana Commercial Bank (GCB) did not participate, to a significant degree, in the project, a risk that had been identified during project preparation. There were two main reasons for the low level of commercial bank participation. Firstly, the major commercial banks have traditionally focused on the trade and industry subsectors, with a relatively small share of their overall portfolio going to agriculture, given the high administrative costs of such lending, the risks inherent in rainfed agriculture, and the nature of clients (small, with little or no collateral, and widely dispersed over a large geographical area). Secondly, the commercial banks paid little or no interests on current accounts, which constituted a major source of investment funds, and therefore considered the reference interest rate (RIR) charged on project funds, which averaged 18 percent, as representing an unacceptably high costs of funds. Conversely, the rural banks depended largely on savings mobilized as the major source of investment funds. Since rates charged on these saving deposits averaged 16 to 18 percent, the average RIR of 18 percent represented an acceptable cost of funds and provided a spread of between 10 to 12 percent to the rural bank, given the on- lending rate of 26 - 30 percent. The availability of external resources also gave the rural banks more flexibility in allocating credit and in expanding their client base. By completion, the Cedis 9.8 billion disbursed was distributed as follows: Barclays Bank (0.0 percent), SCB (0.6 percent), GCB (1.0 percent), the Export Finance Company (6.3 percent) compared to the rural banks (18.0 percent) and the ADB (74.1 percent). 10. Low Ceiling on Subloans. The project established a ceiling of Cedis 50.0 million (about US $50,000 at the current exchange rate) on subloans in an effort to prevent the crowding out of small borrowers by larger, private sector firms and parastatals. The low ceiling effectively directed credit  4 to specific size (smaller) operations and denied access to larger enterprises. The exclusion of larger borrowers, coupled with the small number of rural banks participating at any one time, resulted in slbw disbursement of the line of credit: by the original closing date of December 31, 1992, only 45 percent of the line of credit of US $15.0 million had been disbursed. The slow disbursement coincided with increasing demand for funds from produce marketers, input supplier and poultry producers who had benefited from the agricultural reforms initiated under the Agricultural Sector Adjustment Credit (Cr. 2345-GH). The Development Credit Agreement was therefore amended, removing the ceiling on subloans, thereby allowing individual banks to independently set their own loan limit, based on the merits of the application, the bank's own prudential guidelines, and the assessment of risks associated with the potential investment. 11. Weak Capacity of Project Implementing Agency. The Bank was over optimistic in its estimation of the time it would take the project implementing agency -- itself established under the project -- to emerge as an effective implementing unit. The RFID had no prior experience implementing Bank projects, and though its first director was selected because of his knowledge of Bank procedures, his tenure was relatively short, about a year. There was, in general, a high turnover in the senior management of the implementing agency, with four directors heading the agency in as many years. Although the high turnover itself was due to the restructuring underway within the BOG, it nevertheless affected the effectiveness of the unit, especially the continuity needed to strengthen the newly created agency. The RFID was also staffed by employees transferred from other departments of the BOG, and it therefore took some time for the unit to emerge as a team, a situation exacerbated by the fact that most of the training and technical assistance programs aimed at strengthening this capacity, were also delayed. For example, it took nearly two years for the RFID to procure computers needed for the processing of numerous small applications submitted by the rural banks, relegating the unit to the manual processing of applications that was highly inefficient. It was therefore not uncommon to have delays of 3 to 6 months, from the time applications were received by the RFID and to the time the loan was approved and the account of the rural bank credited with the proceeds. With experience, performance of the RFID improved after the MTR, especially in loan administration and inspection management audits of the rural banks, such that by project completion, the RFID had established the capacity to inspect all 125 rural banks at least once a year, to deal more expeditiously with procurement matters, and to reduce loan processing time to 2 to 3 weeks. Overall Project Implementation Performance 12. Overview. The overall project performance can be regarded as satisfactory as the credit line of US $15.0 million was fully disbursed, albeit two additional years being needed for project completion; 55 of the 125 rural banks could be rated as satisfactory, in terms of capital adequacy; the ARB has established the capacity to provide a wide range of services to member banks, including training, management information system and advisory services; the RFID has established the capacity for regular inspection and examination of the rural banks; and the pilot program in rural cooperatives has demonstrated that rural groups can acquire the technical know-how to profitably operate rural enterprises. However, the lack of co-financing prevented implementation of activities to strengthen credit unions. 13. On-Lending Component. The project on-lent, over a five year period, an estimated Cedis 9.8 billion (about US $19.8 million at the average project exchange rate of Cedis 500.0) for financing investments that benefited an estimated 15,250 individuals and groups. (Details of subprojects financed under the credit are in Tables I & 2 of Part III). A total of Cedis 7.3 billion (75 percent) went to 1,550 groups (farmer associations, cooperatives, traders and producers associations) while  5 Cedis 2.5 billion (25 percent) was utilized by 13,700 individual beneficiaries. The average loan size was Cedis 180,000 for individuals and Cedis 4.7 million for groups, indicating that a large number of smaller beneficiaries benefited and that the subprojects were not skewed towards a few groups or individuals. Of important note is that major beneficiary groups were mainly rural cooperative societies and associations and not large government-owned parastatals. The activities of group beneficiaries were also directly linked to the rural economy, in terms of marketing of inputs or purchasing commodities produced by smaller outgrowers. Distribution of subloans by activity was as follows: poultry (Cedis 3.9 billion), trading (Cedis 2.1 billion), crop production (Cedis 1.9 billion) and cocoa marketing and estate rehabilitation (Cedis 1.2 billion). Other activities supported were livestock, fisheries, oil palm, cottage industries and cotton. About 2,500 or 18.0 percent of the 13,700 individual beneficiaries were women who utilized 16.0 percent of the Cedis 2.5 billion disbursed. The number of female beneficiaries, though representing a small proportion of the total, shows a significant increase over the number of female beneficiaries (248 in all) that had accessed the program prior to project mid-term review. Given the continuing low rate of female participation, it is important to further address the specific needs of such beneficiaries during future project preparation in light of the different sets of constraints they face. In general, access by women would be limited by factors such as the lack of land ownership and/or acceptable collateral. 14. The project helped to increase savings mobilization, mainly from small, first time rural clients who were encouraged to open accounts with the rural banks as a condition for participating in the on- lending program, and who have continued to maintain these accounts long after the end of their commitments to the rural banks. The number of rural depositors has increased steadily since 1989, with total deposit mobilized increasing from Cedis 4.6 billion in 1989 -- a year before project effectiveness -- to Cedis 13.2 billion in 1994, with the 26 participating rural banks accounting for 37.0 percent of the total deposits mobilized in 1994. The total loans and advances from the rural banks nearly doubled, from Cedis 3.7 billion in 1991, to Cedis 6.8 billion in 1994, compared to project resources of Cedis 1.7 billion channeled through the rural banks during the past five years, equivalent to about Cedis 340.0 million per annum. 15. Institution Building Component. Overall, performance of the institution building component was satisfactory. A review of specific subcomponents is given below: Financial Restructuring of the Rural Banks. Achievement of this objective was satisfactory. Preparation of an acceptable financial restructuring plan for 20 rural banks delayed project implementation by six months. The financial restructuring of rural banks consisted of three integrated steps: (i) diagnosis, including assessment of the financial and prudential status of each rural bank; (ii) an action plan establishing a monitoring system for performance targets such as the rate of deposit mobilization and loan recovery ratios; and (iii) skills development of staff and management to ensure attainment of these targets. Specific target of the project was to complete restructuring of 80 of the 125 rural banks. However, by completion, only 55 rural banks (44 percent) were rated satisfactory by the BOG, 51 banks (41 percent) were rated as less than satisfactory (mediocre or weak), and 19 banks (15 percent) were rated as distressed. While on face value this may represent reasonable achievements -- only two rural banks were rated satisfactory at project effectiveness -- in reality, the rural banks are still weak. The official classification of "satisfactory" is rather narrow, and only takes into account achievement of capital adequacy of 6 percent of the banks' risks assets. Therefore, banks rated as "satisfactory" could still be characterized by weak management, weak internal controls and a high proportion of  6 non-performing loans. The detailed diagnosis of the rural banking system, which was undertaken after credit effectiveness, revealed, among other things, that these banks were in worse shape that initially believed. Had the diagnosis been undertaken as part of project preparation, it would have influenced the scope and nature of the technical assistance program for the rural banks. The single most important accomplishment of the restructuring program has been that it helped to lay a foundation for subsequent operations to continue with the strengthening of the rural banking system. Strengthenin, the Association of Rural Banks. The outcome of this component was satisfactory. The project enabled the ARB to firmly establish the capacity for developing and managing the training program for the rural banks, that has benefited an estimated 2,000 staff over the past five years. More importantly, the ARB has taken concrete steps to ensure sustainability of these activities, including: (i) decentralization of the training program to the regional chapters of the association in order to reduce costs, expand coverage to member banks and increase local ownership; (ii) training of trainers, which utilizes managers and directors from the better rural banks to assist in the training of managers and staff of weaker rural banks; and (iii) increasing cost recovery: gradually increasing the fees charged the individual rural banks for training services so that in the near future these banks would eventually pay the full costs of these services. From a fledgling organization, the ARB has emerged as a strong association with permanent headquarters and Secretariat and, is planing to undertake, with the support of the BOG, a study that would help it to clearly define its relationship as an apex institution for the rural banks. Strengthening Capacity for Rural Bank Examination. Performance of this component was satisfactory. However, the project design was generally too optimistic as to how quickly the RFID, itself created by the project, could evolve into an effective implementing agency, given its expanded mandate. The RFID has so far overcome the initial teething problems, and has establish good capacity for inspection management audits. By project completion, all the 125 rural banks are being visited at least once a year by the RFID -- a feat not possible only two years ago. A total of 120 staff, including some from other departments of the BOG and from the PFIs, have benefited from local training while 16 staff have participated in various external training, including opportunities to witness successful examples of rural banking in the Philippines and the Netherlands. Capacity for Rural Credit Appraisal. The outcomes of this component was unsatisfactory. The project provided a total of 96 man-months of consultancy to the rural banks for credit appraisal and the provision of 100 motorbikes for logistics to enable rural banks to focus on developing and investing in small investments at the community-level, thereby providing important linkages to downstream operations. Similarly, the project also provided for the hiring of 60 project officers to be based within the larger commercial banks who would also help these commercial banks, traditionally lukewarm to agriculture sector lending, to expand the volume of their sector portfolio. These programs were not implemented and most of the banks interviewed during the ICR process reported that they had no knowledge concerning this aspect of the project. It is therefore conceivable that it may have been overlooked as the RFID became inundated with the requirements of the larger, more  7 visible components of the projects such as the on-lending program and rural bank restructuring activities. Strengthening Rural Cooperatives. Performance of this component was rated satisfactory. The pilot program was conceptualized to test the basic hypothesis that efficiency of rural financial intermediation could be enhanced by organizing rural groups to enable them to achieve the desired economies of scale in their operations. The pilot program of rural cooperatives in the Upper West Region tested and substantiated this hypothesis. The inventory credit program developed under the pilot has emerged as a simple but innovative instrument for not only promoting rural financial intermediation, but also for providing important linkages between the formal and informal financial sectors. The program has benefited more than 750 farm families through community-level agricultural cooperatives that are profitable and that are contributing to enhancing the incomes of these families. The success of this scheme has led to its widespread adoption in other parts of the country, especially by the ADB in maize marketing in the Ashanti and Brong-Ahafo Regions with annual gross sales exceeding Cedis 2.0 billion. Strengthening the CUA. This component was dropped due to the lack of co- financing from USAID and CIDA. 16. Technical Assistance Financed under the Project. The implementation of technical assistance provided in support of project objectives was satisfactory. Technical assistance enabled the Borrower to procure the services of advisors in rural bank restructuring, inspection and management who worked with counterparts from the BOG. The scope of the technical assistance program with the rural banks included: preparing a restructuring plan for the rural banks; preparing guidelines for rural bank inspection management audits; preparing management information systems for the rural banks; identifying skills gaps for training of BOG and rural bank staff; and preparing the initial plans for the liquidation of the 19 distressed rural banks. 17. Studies Financed under the Project. The studies financed under the project had direct relevance for designing the pilot program in rural cooperatives in the Upper West and in putting into a place a program for monitoring and evaluation their impacts. All studies were completed in a timely manner and their objectives were met. The first five of the studies provided a diagnosis of the rural cooperative subsector, the strategy for developing pilots and the impact analysis of these pilots. The sixth study, a review of the rural financial sector, was an important input into the broader Ghana Financial Sector Review, the latter report helping to identify the key constraints, priorities and future strategies for strengthening the financial sector. Project Sustainability 18. Institution Strengthening. Most of the benefits of institution building appear sustainable. These include: capacity enhancement of the RFID to provide regular inspection and examination of the rural banks; training of rural bank staff and directors; and inventory credit. Training: The impact of the training program on rural bank performance has been the most visible aspect of the institution building efforts in the rural finance sector. Sustainability is encouraged by the fact that rural banks are now increasingly paying a larger share of training costs and to budget for such programs in their annual work plans, based on identified skills gaps. Other indicators of sustainability are the training of trainers program, which operates within the regional chapters of the ARB and the decentralization  8 of training to the regional chapters. Rural bank examination and inspection. Sustainability is enhanced by the fact that the BOG is committed to supporting the RFID in these activities as detailed in the action plan for the post-credit operational phase of the project. Rural cooperatives. Pilot cooperatives established under the project have evolved into profitable rural enterprises, sustainable aspects of which include local ownership through strong grassroots organizations trained to apply modern business principles in managing these cooperatives. 19. Rural banks. Loan recovery rates, profitability and the subsidy dependence are three criteria used to assess the sustainability of the rural banks, and in particular, to measure the extent of subsidy dependence among the 26 rural banks utilizing project resources. Loan recovery: rural banks are, in general, saddled with huge non-performing loans, rising overdues and weak internal control mechanisms increase their vulnerability to internal fraud. The average recovery rate has declined for all rural banks, from 70.0 percent in 1986 to only 59.0 percent in 1994. The deterioration is attributed to the rapid expansion of rural banks through the creation of agencies and deposit mobilization centers. With few and poorly trained staff and with little or no capacity for project appraisal, the rapid move to create agencies have left most rural banks highly susceptible to fraudulent acts by branch managers who are often loosely supervised by the head office. Performance was better among the 55 rural banks rated as satisfactory, with 73 percent recovery rate, compared to only 44.0 percent for the 51 banks rated as either mediocre or weak. Profitability: Most rural banks continue to make inadequate provisions for bad and doubtful debts, thereby overstating their profitability: 96 of the 106 active rural banks made inadequate provisions for bad debts amounting to Cedis 469.8 million in 1994. Profits among the 26 participating rural banks was higher, but varied widely, ranging from Cedis 0.2 million to Cedis 49.0 million in 1993 (the last year for which comprehensive data is available). 20. Subsidy Dependence. Resources to the rural banks were subsidized in two ways: (i) through the low interest rate on project funds, which averaged below the rate rural banks paid on term deposits; and (ii) technical assistance mainly for training, which was largely a grant. Project resources were on-lent at a RIR that was lower than the rate on term deposits of the rural banks. Unlike the larger commercial banks, rural banks do not participate in the interbank lending program of the central bank and they therefore depend largely on savings mobilized for investments. The relevant cost of funds to the rural banks was therefore the interest paid on these term deposits. During the past five years, the rate on term deposits which averaged between 18.0 and 22.0 percent, compared to the average project RIR of 18.5 percent. Technical assistance to the rural banks was largely a grant, averaging Cedis 8.4 million per annum per rural bank over the life of the project. The extent to which subsidized resources affect the sustainability of the participating rural banks is measured by the subsidy dependence index (SDI)'. The estimated SDIs are in Table 3, Annex III and they show, among other things, the widely varying level of capacity among the rural banks. Six of the 26 banks (Nyakrom, Akyempim, Nsoatreman, Braka-Breman, Nwabiagya and Bosomtwe) had negative SDIs, indicating that, not only have these banks achieved fully an acceptable level of 1 The Subsidy Dependence Index (Yaron, 1992) can be represented as S = A(m - c) + [(E * m) - P] + K, where S is the value of annual subsidy received by the PFI, A is the average concessional borrowed funds outstanding, m is the interest rate the PFI would have paid for borrowed funds in the absence of the line of credit, c is the weighted average concessional interest rate, E is the average annual equity, P is before tax profit, adjusted for provisions for bad debts, and K is catchall for all other subsidy, in this case, the average annual value of technical assistance. The SDI ratio is estimated as SDI = S / (LP *i), where LP is the average outstanding loan portfolio of the PFI and i is the weighted average interest rate earned on the loan portfolio. An SDI of zero and less indicates sustainability while positive and increasing SDIs show correspondingly increasing level of dependency.  9 sustainability, but that their annual profits exceeded the value of any subsidy (including technical assistance). Fourteen banks had positive, but low SDIs while seven banks had SDIs in excess of 50 percent, indicative of the level of higher degree of dependency (low sustainability). One implication of these results are that classification of rural banks on the basis of satisfactory and unsatisfactory does not show the true state of these banks, as banks rated satisfactory may not be sustainable in the true sense of the word. Bank Performance 21. Bank performance, from project identification through completion, was, on the whole, satisfactory. The Bank responded quickly to the slow implementation performance, such as that described in para. 11 above, by increasing field supervision and the level of dialogue with the Borrower. The Bank also responded by employing the appropriate skills mix in addressing specific problems, as in inspection management audits (Table 11). Bank performance was complemented by strong support from the Resident Mission which provided advice to the implementing agency on procurement and other related matters and helped to maintain continuity in policy dialogue with the Borrower. Borrower Performance 22. On the whole, Borrower's performance was also satisfactory. The project was consistent with the Borrower's objective of strengthening rural financial intermediation and putting the rural banking system on sound financial, technical and institutional footing. The project therefore had strong Borrower commitment and ownership. Prior to the project, the Borrower had implemented substantial policy and institutional reforms at the macro-level, including financial sector reforms, which helped to provide an enabling institutional and policy environment for project implementation. However, at the implementing agency level, there were practical problems in translating strong Borrower commitment into effective implementation performance: lack of prior experience in implementing Bank projects, high turnover in senior management, delays in procurement of vehicles and computers critical to carrying out statutory functions of the department such as inspection and examination. However, the RFID was able to transcend the earlier problems, such that by the mid-term review, it had developed significant capacity for project implementation. Overall Assessment of Project Outcome 23. Based on implementation performance, achievement of physical targets set in the SAR, sustainability of key project benefits, and the Borrower's action plan for the post-operational phase of the project, outcome of the project is satisfactory. By financing eligible subprojects in the rural economy, the project accomplished a key objective of expanding productive capacity and employment in agriculture. It was also successful in improving the efficiency of financial sector institutions. Assessment of outcome of specific project components is as follows: On-Lending: satisfactory, despite the initial implementation delays. Institution building: satisfactory, but with varying degrees of success in specific subcomponents as noted: (i) financial restructuring of the rural banks -- satisfactory, but with low probability of sustainability; (ii) strengthening of the ARB -- highly satisfactory; (iii) enhancing capacity for rural credit appraisal -- unsatisfactory; (iv) strengthening RFID capacity for inspection management audits -- satisfactory; and (v) strengthening rural cooperatives -- highly satisfactory.  10 Future Operations and Key Lessons Learned 24. Future Operations. GOG has indicated, through its post-project operational plan, a strong commitment to continuing support for key project activities beyond the closing date. An action plan and the resources committed for these priority areas are contained in Appendix B to the ICR. Specifically, the Borrower would continue with the following project activities: training of rural bank staff and directors through the RFID; regular rural bank examination and inspection through the RFID; development of rural cooperatives; and expanding rural investments through the recycling of repayments under the line of credit component through the PFIs. However, based on project implementation experience, comprehensive diagnosis of the rural finance sector is needed to inform the design and scope of future interventions. Accordingly, follow-up rural finance operations should be preceded by broad consultation, studies and in-depth analysis in order to better understand the sector and to better target proposed interventions. The Bank's Financial Sector Review (FY95) includes an analysis of rural finance issues. Another step in this direction is the ongoing collaborative sector work with the Borrower on Rural Institutions. 25. Key Lessons Learned. The key lessons learned from project implementation are: (i) training of rural bank staff and directors is a key factor in the development of viable rural banks as it empowers management to adapt more innovative instruments and approaches to deposit mobilization, credit delivery and loan recovery; (ii) rural banks largely depend on savings mobilized for investments; commercial banks, on the other hand, use deposits on which they pay little or no interests, and which they invest in less risky activities in trade and industry. In determining the RIR for an untargeted credit line, it is important to take these factors into account to ensure that all banks can participate; (iii) in order to be effective monitoring tools, inspection mangement audits of the rural banks should be combined with regular feedback to the rural banks, with clear guidelines for correcting observed deficiencies, a time frame for addressing exceptions noted, and possible sanctions for failing to rectify deficiencies; (iv) as demonstrated by the ARB, the emergence of a strong apex institution for the rural banks is vital to the long-term viability of these banks as it provides a sustainable, non-governmental alternative to the provision of vital services (training, financial advisory and promotional); government's role should be limited to its statutory responsibilities of inspection management audits and examination; (v) inventory credit has emerged as an innovative marketing tool crucial in linking the formal and informal sectors; and (vi) rural cooperatives are more viable if they are organized as rural businesses with profits, instead of subsidy, as the bottomline of their operations. Such rural cooperatives, when viable, provide the critical linkage between the formal and informal sectors and the initial technical assistance for training such groups in modern business management has high returns.  -11- PART II: STATISTICAL ANNEXES Table 1: Summary of Assessments Substantial Partial Negligible Not Applicable A. Achievement of Objectives (1) (1) (1) (1) Macro policies [2] [2] IE Sector policies el E l E Fiumncial objectives [l [2] Institutional development [ 5 I]1 Physical objectives E] I E E Poverty alleviation El Gender issues E E IA Other social objectives Q E E IA Environmental objectives f E E IA Public sector management E E El Private sector development I E E E Likely Unlikely Uncertain B. Project sustabnablilty (1) (1) (1) Highly Satisfactory Satisfactory Deficient C. Bank Performance (1) (I) (1) Identification E l E] Prepadtion assistance El El Apprisal EI El Supervision E IA Highly Satisfactory Satisfactory Deficient D. Borrower performance (1) (1) (1) Identification E IA E Implernentation O IA E Covenant compliance E IA E Opertion (if applicable) E E E Highly Highly Satisfactory Satisfactory Unsatisfactory Unsatisfactory E. Assessment of Outcomes (1) (1) (1) (1) IA El El  -12- Table 2: Related Bank Loans and Credits Year of Loan/Credit Title Purpose Approval Status Preceding Operations: 1. Cr. 1911-GH & 19 11-1, Address fundamental and urgent 1988 & Completed, US$ 100.Om & $6.6m problems of the financial sector 1989 ICR of June Financial Sector Adjust I 1994 & Supplement 2. Cr. 1996-GH, US$30.Om Support small & medium SME Development entrepreneurs in the private 1989 Completed sector Following Operations: 3. Cr. 2180-GH, US$16.5m Agric. Diversification Develop non-cocoa tree crops 1991 On-going and horticultural crops 4. Cr. 2247-GH, US$ 22.Om Rationalize agricultural research 1991 On-going National Agric. Research resources 5. Cr. 2345-GH, US$80.Om Reform agric. policy to 1992 On-going Agric Sector Adj strengthen sector coordination Cr. 2245-1-GH; US $5.74 and management 1994 On-going AGSAC supplement Supplement to agric. policy reforms 6. Cr. 2346-GH, US $30.4 Provide more efficient extension 1992 On-going Nat Agric. Extension services 7. Cr. 2426-GH, US $18.1 m Support national resource and 1992 On-going Env. Resources Mngt env. planning and mngt. 8. Cr. 2441-GH, US $22.5m Increase livestock production 1993 On-going National Livestock Services and services 9. Cr. 2555-GH; US$21.5m Provide development of rural 1993 On-going Agric. Sector Inv. Project infrastructure in support of development  -13- Table 3: Project Timetable | | I Date actual/ Steps in project cycle Date planned | Revised | latest est. Identification (EPS) Preparation 11/06/87 04/02/88 04/02/88 Appraisal 10/12/88 10/12/88 10/12/88 Negotiations 02/20/89 03/13//89 02/13/89 Board Presentation 03/28/89 06/14/89 06/14/89 Signing 07/28/89 n.a. 07/28/89 Effectiveness 04/28/89 n.a. 07/28/89 Mid-Term Review 12/31/91 06/24/92 06/24/92 Project Completion 04/30/93 04/30/94 04/30/95 Credit Closing 12/31/92 12/31/93 12/31/94  - 14 - Table 4: Credit Disbursements: Cumulative and Actual (US$ million) Bank Fiscal Year FY90 FY91 FY92 FY93 FY94 FY95 (Ii) Appraisal Estimate 1.6 6.0 7.0 5.4 Cumulative 1.6 7.6 14.6 20.0 Actual (a) 0.0 3.2 3.7 7.3 5.2 1.0 Cumulative 0.0 3.2 6.9 14.2 19.4 20.4 Actual as % of Estimate 0.0 42.0 47.0 71.0 96.0 102.0 Date of Final Disbursement: April 30, 1995 (a) Credit was extended for two years beyond the original closing date of December 31, 1992. (b) For first half of FY95 only, i.e., July I to December 31, 1994.  -15- Table 5: Key Indicators for Project Implementation Page 1 of 2 l1. Key lInplemnentation Indicators in SAR Expected/Estimnated Actual at Project Completion I. Institution Buildingy (A) Financial Restructuring Rural (i) Restr-ucturing of 125 rural All 125 rural banks completed l Banks banks which: a) have adequate diagnostics phase of financial capital base; b) make profits; restructuring and action plans were c) have sound financial control prepared, but only 55 banks satisfied mechanisms; d) have low capital adequacy requirements under arrears/high recovery; e) have the Banking Law 225, 56 are weak good mechanisms for appraisal and 19 are distressed. The distressed and supervision of loan banks have been shut down, pending portfolios. liquidation. However, among the 106 operating rural banks, sound financial control is lacking. Little capacity for appraisal and loan supervision exists. Internal control mechanisms not in place in most banks. Under provisioning for bad debts among 96 of 106 rural banks based on 1994 management audit reports. (B) Strengthening the Association Emergence of an apex that has ARB has established an excellent of Rural Banks (ARB) sound financial base, provides program for training staff and sound program and services to directors of rural banks. Sustainability members including public of the program rated high due to local relations, financial ownership and increased subscription management and training. from member banks. ARB now has a permanent head office and Secretariat and undertakes good PR with BOG on behalf of the member banks. Apex institution study planned for FY96. C. Strengthening Credit Union Emergence of a CUA that has Component dropped due to lack of co- Association (CUA) sound financial base, is financing. respected by its members, and provides sound technical advice and program for members.  -16- Table 5: Key Indicators for Project Implementation Page 2 of 2 1. Key huplementation Indicators in SAR Expected/Estimated Actual at Project Completion 1. Institutional Building (cont'd): D. Strengthening the RFID, Bank Emergence of strong Reorganization and staffing of RFID of Ghana department with trained staff completed and ample logistics and logistics to provide regular provided. Several staff have inspection management audits completed both local (120 staff) and of the rural banks and to assist external (16 staff) training. RFID has in the development of rural developed capacity to provide regular finance policy. examination of all 125 rural banks at least once a year. E. Strengthening of Rural Establishment of three (3) Three FSCCs established as Pilots in Cooperatives under a Pilot financially and institutionally the Upper West with over 750 farm Program in the Upper West viable Farmer Service families as members. FSCCs Region Cooperative Centers (FSCCs) diversifying into shea brokering and that provide good services to processing in order to spread risks members and that would from single crop. Use of inventory establish economies of scale in credit as innovative tool in marketing inventory credit. of QPM maize. Program has received positive evaluation from local officials for its impact on poverty alleviation in the resource-poor areas of the Upper West. II. Line of Credit Lending of US $20.0 million Total line of credit filly disbursed, equivalent, including US $15.0 equivalent in local currency of Cedis million of IDA resources for 9.8 billion to finance crop production, eligible subprojects in the marketing, processing, transport and agriculture sector. other rural industries with more than 1,550 groups and 13,700 individuals benefited. Estimated Cedis 1.7 billion channeled through 26 participating rural banks.  -17- Table 6: Studies Included in Project Study Purpose Status Impact of Study 1. Development of To help prepare a pilot Completed The study enabled the Department of Farmer Cooperatives program for assisting 1989 Cooperative (DOC), through an NGO, to rural cooperative establish a Pilot Program of Rural societies in becoming Cooperation in three villages in the Upper financially viable and West Region. sustainable 2. Impact Study - Farmer Evaluate the impact of Completed Study assessed the impact of the pilot on Cooperatives the pilot program on 1992 individual groups and farm households and beneficiary villages in enabled the DOC to better design its the Upper West Region technical assistance to these cooperatives. 3. Regular Monitoring and To monitor key variables Regular Regular monitoring provided crucial Evaluation Studies related to the target Quarterly information to DOC and NGO on the group of cooperatives, Report financial viability of the pilots and the such as farm cash flow, impact of technical assistance. Enabled cost recovery, increased implementing agency to develop inventory credit worthiness of credit scheme for servicing members. members, and linkage with formal financial sector. 4. Diversification Study Study to diversify the Completed Study led to creation of diversified FSCCs of FSCCs revenue base of the 1992 through creation of Processing Service FSCCs. Centers (PSCs) for sheanut, cowpeas and grains. PSCs fully operational for all three FSCCs. 5. Review of the Conduct detailed review Completed Study completed but impact unknown. Cooperative College of the existing program 1993 Major recommendations not implemented. for training of cooperatives and recommend changes 6. Rural Finance Sector Conduct a thorough Completed Results of study used to establish priorities Review review of the rural 1994 for future interventions in the rural finance finance sector as an input sector, especially in linking the formal and into the Ghana Financial informal sectors. Results would be used in Sector Review. designing rural finance component of an integrated Subsector lending program in Ghana.  -18- Table 7A: Project Costs by Category Appraisal estimates Actual/latest estimates (US $M) (US $M) Local Foreign Local Foreign Category/item costs costs Total costs costs Total 1. On-Lending 8.0 15.0 23.0 6.8 15.2 22.0 II. Institutional Building: 2.5 2.5 5.0 3.3 2.1 5.4 (a) Restructuring of rural banks 1.3 0.5 0.8 2.1 0.7 2.8 (b) Association of Rural Banks (ARB) 0.3 0.7 1.0 0.5 0.5 1.0 (c) Rural credit appraisal 0.5 0.3 0.8 0.1 0.0 0.1 (d) Rural banks examination 0.3 0.7 1.0 0.4 0.6 1.0 (e) Rural cooperatives 0.1 0.3 0.4 0.2 0.3 0.5 III. Rural Finance Inspection Department 0.3 0.7 1.0 0.4 0.6 1.0 Total 10.8 18.3 29.1 10.5 17.9 28.4  -19- Table 7B: Project Financing Appraisal estimate Actual/latest estimate (a) Source (US$ Million) (US$ Million) Local Foreign Local Foreign costs costs Total costs costs Total IDA 1.4 18.6 20.0 2.5 17.9 20.4 GOG 1.1 - 1.1 1.2 1.2 Participating Banks 5.0 5.0 3.9 3.9 Beneficiaries 3.0 - 3.0 2.9 2.9 Total Project 10.5 18.6 29.1 10.5 17.9 28.4 (a) Actual as at December 31, 1994. Note: Project cost reduced by US $8.9 million due to unrealized co-financing from USAID and CIDA.  -20- Table 8: Status of Legal Covenants Page I of 2 Agreement/ Covenant Present Fulfillment Date Section Type Status Original Revised Description of Covenant Comments Project Agreement (PA)/ 2.08(a) 2, 12 C 09/30/89 Abolition of the Cedis 25,000 and Cedis 125,000 ceiling on Done shareholdings in rural banks by individuals and companies, respectively. 2.08 (b) 2, 12 C 09/30/89 09/30/89 Limiting shareholdings in particular RBs by individuals and Done; lifted in June 1990 to allow companies to not more than 5% and 10%. respectively, of equity mobilization under the share capital. restructuring program. 2.09 (a) 2, 3, 5 CP 09/30/89 Requirement that rural banks assign all non-performing Collection agencies not in existence loans to special collection accounts and make commensurate so no bad loans assigned; about 96 provisions for bad debts. rural banks not making adequate provisions in 1994. 2.09 (b) 2. 12 CD 09/30/89 12/31/90 Abolition of sectoral quotas imposed on rural banks' lending Done in December 1990. operations, while maintaining initially a minimum of 20% for agriculture. 2.11 5 C 09/30/89 09/30/89 Preparation of a shortlist of firms by BOG for purposes of Short list prepared by RFID in providing loan support services to RBs. December 1990. 2.13 5, 12 C 12/31/89 Nomination by BOG of its representative on the Boards of BOG has established a new policy of RBs in which BOG is not represented. removing its staff from the Boards of rural banks to avoid conflict of interest 2.14 (a) 1I 5, 9 C 12/31/89 Preparation by BOG of a monitoring system Done of the rural finance operations of all banks, including operations under this project. 2.14 (b) 5, 10 C 12/31/89 12/31/89 Completion by BOG of a staff development program for the Done. RFIDby December31, 1989. 2.14 (c) 3, 5, 9 C 09/30/89 annually Borrower to cause the ARB to submit to IDA, not later than Done. ARB submitting annual September 30, 1989 and September 30, 1990, its annual workplans to RFID and IDA work program for the following year. 2.15 5 C 12/31/89 Recruitment by BOG of a RBs Inspection Advisor by RB Inspection Advisor arrived in December 31, 1989. March 1990. 3.12 1, 5 C 09/30/89 annually BOG to carry out examination of each RB at least once a All RBs examined at least once a year. year beginning in 1994.  -21- Table 8: Status of Lega Covenants Pane 2 of 2 Agreement/ Covenant Present Fulfilbnent Date Section Type Status Original Revised Description of Covenant Comments Credit Agreement (DCA) 4.01 (b) 1, 2, 3 C Twice a year Borrower shall have all records and accounts for each fiscal year Audits current and up-to-date. audited in accordance with appropriate auditing principles by independent auditors. 6.01(a) 9, 10 C Condition for Signing of Subsidiary Administration Agreement (SAA) between the Done; letter of approval dated effectiveness Borrower and BOG. November 1, 1989. 6.01(b) 9, 10 C Condition for Submission to IDA by the BOG, the operational policies and procedures Policies submitted on time. effectiveness for the RFID. 6.01(c) 9, 10, 12 C Condition for Signing by BOG of Participation Agreement with at least two PFIs. First Agreement signed with effectiveness ADB, Barclays and Standard Chartered Banks 6.01(d) 5, 9 C Condition for Recruitment by BOG of a Rural Finance Adviser and a Rural Bank Done effectiveness Restructuring Adviser. 6.01(e) 1, 2, 5 CD Condition for Submission by BOG the restructuring plans for the ADB and at least 20 Delayed by 6 months due to effectiveness rural banks. problems in recruiting consultants. 6.01(f) 2, 4, 11 C Condition for Issuance of a statement by the Borrower that it would provide a deposit effectiveness guarantee up to a limit of Cedis 100,000 per depositor for rural banks depositors whose claims cannot be met during the restructuring. 6.01(g) 2, 4 C Condition for Establishment by the Borrower of a RB Recapitalization Fund in the effectiveness BOG with an initial sum in Cedis equivalent to US $300,000. Covenant types: Present Status: I = Accounts/audits 8 = Indigenous people C = covenant complied with 2 = Financial performance/revenue 9 = Monitoring, review & reporting CD= complied with after delay 3 = Flow & utilization of project funds 10 = Project implementation not included in 1-9 CP = complied with partially 4 = Counterpart funding 11 = Sectoral budgetary/resource allocation NC = not complied with 5 = Management aspects of the project 12 = Sectoral policy/regulatory/institutional 6 = Environmental covenants 13 = Other 7 = Involuntary resetdement  -22- Table 9: Comploance with Operational Manual Statements The project was in compliance with all applicable Bank Operational Manual Statements. Table 10: Bank Resources: Staff Inputs (in Staff Weeks) Actual Stage of Project Cycle Weeks US$ Through Appraisal 57.4 n.a. Appraisal - Board 26.7 n.a. Board - Effectiveness 6.8 n.a. Supervision 106.6 n.a. Completion 4.5 n.a. Total 202.0 n.a. Note: n.a. = not available at the time of the ICR.  -23- Table 11: Bank Resources - Missions Month/ No. of Days Specialized Performance Rating Stage of Project Cycle Year Persons in Field Staff Skill Implementation Development Problems Appraisal 10/88 5 16 1, 1, 1, 4, 5 n.a. n.a. n.a. Appraisal to Board 6/89 1 0 1 n.a. n.a. n.a. Board to Effectiveness 4/90 1 0 1 n.a. n.a. n.a. Supervision 10/90 1 1 2 2 Restructuring delays 2/91 1 1 3 2 Poor implementation capacity 6/92 4 1, 1, 3, 4 3 2 Slow disbursement 2/93 2 1, 1 3 2 Slow subloan processing 7/93 3 1, 2, 5 3 2 Slow training take-off/Disb. 11/93 1 2 2 2 Second Extension 4/94 1 2 2 2 No major problems identified Completion 2/95 2 30 1, 2 2 2 Note: n.a. = not applicable Key to Staff Skills 1 = Agricultural economist 2 = Economist 3 = Agriculturist 4 = Financial analyst 5 = Banking specialist  - 24 - PART III: SECTOR-SPECIFIC DATA Table 12: Distribution of Subloans by Beneficiary Category Total Loans Gr ups Individuals Distri ution (%) Ave. loan Ave. loan Subsector No. of Total Loan No. of size No. of size Beneficiaries ('000 Cedis) Beneficiaries ('000 Cedis) Beneficiaries ('000 Cedis) Group Individuals Agro-processing 144 33,459.0 10 575.5 134 206.7 83 17 Cocoa marketing 393 972,127.2 13 66,187 380 293.9 89 11 Cocoa Estate Rehab 1,411 230,030.1 39 1,024.1 1,372 138.6 83 17 Rural Infrastructure 137 93,508.9 58 1,057.7 79 407.1 66 34 Cottage Industries 1,099 238,006.9 13 1,254.6 1,086 204.1 93 7 Cotton 145 42,543.1 7 2,018.6 138 205.9 67 33 Crop Production 8,574 1,852,870.9 411 957.4 8,133 175.9 77 23 Farm Inputs 701 93,187.1 - - 701 132.9 0 100 Fisheries 167 24,160.0 - - 167 144.7 0 100 Livestock 216 33,046.7 20 123.8 196 155.9 93 7 Oil Palm 482 97,455.8 20 1,186.8 462 159.6 76 24 Poultry 550 3,896,109.4 101 23,736.7 449 207.8 98 2 Trading 1,103 2,157,547.7 821 2,524.3 282 301.8 96 4 Transport 142 33,180.8 4 382.5 138 229.4 95 5 Total/Average/% 15,264 9,797,233.8 1,547 4,733.8 13,717 180.4 75 25 Source: Mission estimates based on PFI data.  - 25 - Table 13: Distribution of Individual Subloans by Gender Total Loans No. of Beneficiaries Comparative Loan Size Share of Investments ('000 Cedis) ( No. of Total Loan Subsector Beneficiaries ('000 Cedis) Male Female Male Female Male Female Agro-processing 134 27,704.1 96 38 193.5 240.1 67 33 Cocoa marketing 380 111,692.2 274 106 300.7 276.4 74 26 Cocoa Estate Rehab 1,372 190,089.6 1,075 297 140.7 130.9 80 20 Rural Infrastructure 79 32,163.9 49 30 523.2 217.6 80 20 Cottage Industries 1,086 221,696.9 898 188 202.8 210.7 82 18 Cotton 138 28,413.1 104 34 215.8 175.5 79 21 Crop Production 8,133 1,430,657.8 6,735 1,398 180.8 152.1 85 15 Farm Inputs 701 93,187.1 617 84 133.5 128.8 88 12 Fisheries 167 24,160.0 154 13 145.5 135.4 93 7 Livestock 196 30,570.3 163 33 155.5 158.4 83 17 Oil Palm 462 73,719.1 369 93 174.5 100.2 87 13 Poultry 449 93,294.0 341 108 216.5 180.1 79 21 Trading 282 85,103.7 232 50 312.2 253.4 85 15 Transport 138 31,650.8 113 25 229.8 227.2 82 18 Total/Average/ % 13,717 2,474,102.7 11,220 2,497 184.2 163.0 84 16 Source: Mission estimates based on PFI data.  - 26 - Table 14: Summary Statistics and Subsidy Dependence of Selected Rural Banks (cedis '000) Rural bank Region Total Total Loans Subloans