Actual Project Name : Rural Financial Services Project Country: Ghana US$M): Project Costs (US$M

Transcription

1 Public Disclosure Authorized IEG ICR Review Independent Evaluation Group 1. Project Data: Date Posted : 07/09/2009 Report Number : ICRR13120 Public Disclosure Authorized Public Disclosure Authorized PROJ ID : P Appraisal Actual Project Name : Rural Financial Project Costs (US$M US$M): Services Project Country: Ghana Loan/Credit (US$M US$M): Sector Board : FPD Cofinancing (US$M US$M): Sector(s): Banking (64%) Micro- and SME finance (32%) Central government administration (4%) Theme(s): Small and medium enterprise support (40% - P) Rural markets (40% - P) Legal institutions for a market economy (20% - S) L/C Number: C3374 Partners involved : IFAD Evaluator : Panel Reviewer : Group Manager : Group: Tim L. De Vaan James Sackey Ismail Arslan IEGCR Board Approval Date : 06/08/2000 Closing Date : 12/31/ /31/ Project Objectives and Components: a. Objectives: Public Disclosure Authorized The Rural Financial Services Project (RFSP) sought to promote growth and reduce poverty in Ghana by broadening and deepening financial intermediation in rural areas through the following measures : i. ii. iii. iv. strengthening operational linkages between informal and semiformal microfinance institutions and the formal network of rural and community banks in order to expand services to a larger number of rural clients; building capacity of the rural and community banks, the principal formal financial intermediaries operating in rural areas, in order to enhance their effectiveness and the quality of services they provide; supporting the establishment of an apex structure for the rural banking system to provide the economies of scale needed for these unit rural banks to address generic constraints related to check clearing, specie supply, liquidity management and training, etc. which have impeded growth of the rural finance sector; and strengthening the institutional and policy framework for improved oversight of the rural finance sector. b.were the project objectives/key associated outcome targets revised during implementation?

2 No c. Components (or Key Conditions in the case of DPLs, as appropriate): A. Capacity building for the informal financial sector (Appraisal cost: US$5.43 million). This component included the following: (i) developing, testing and disseminating innovative products and instruments suitable to the needs of rural clients, often comprising small community self -help groups, associations with limited organizational skills, assets and information; (ii) supporting the organization, training and empowerment of informal financial institutions, community self -help groups and associations to enhance their ability to mobilize savings and access credit from formal sources; (iii) investing in various mechanisms for more effective linkages between informal and formal financial sectors for better integration; (iv) improving the regulatory and policy environment under which informal financial sector entities operate to enable them to exploit new market opportunities; and (v) training and capacity building of microfinance institutions that serve rural clients and often help link them to more formal sources of finance. Component A was revised at mid-term in order to focus on training and capacity building proven most likely to achieve results in improving performance and outreach of rural microfinance institutions (MFIs) and linking them to Community-based organizations (CBOs) and Self help groups (SHGs). The expectation was that the revised component would: improve and expand training for rural MFIs; restrict training of CBOs to those directly concerned with micro-finance; launch a microfinance support initiative (MSI) to build the capacity of Rural and Community Banks (RCBs) and apex MFIs to deliver and sustain microfinance services to the rural poor in their catchment area; and update the feasibility for capacity building fund to draw lessons from the pilot. B. Capacity building for rural and community banks (Appraisal cost: US$4.31 million). This comprised: (i) undertaking a thorough and in-depth assessment of the financial and management performance of each rural bank in order to determine the extent of the constraints affecting their profitability and operational effectiveness and to design appropriate policy response (liquidation/restructuring, capacity building needs, etc.); (ii) supporting a human resource development program for key staff and personnel (e.g., managers, accountants, rural credit analysts ) in order to augment the overall quality of rural bank personnel; (iii) upgrading internal controls and procedures for greater transparency of operations needed to enhance depositors' confidence; (iv) investing in various technologies needed to promote rural bank -to-rural bank communication, operations and synergy; (v) rationalizing agency banking and linking the number of agencies to each individual rural bank's capacity; and (vi) addressing residual policy constraints (geographical restrictions, secondary reserves, classification standards, etc.) which affect the smooth operations and functions of the rural banking sector. C. Technical assistance for the APEX bank (Appraisal cost: US$8.69 million). This included: (i) sensitizing of member rural banks, their shareholders, and clients on the proposed institutional changes which would be introduced under the apex initiative; (ii) developing and clarifying the institutional and legal framework under which the apex would operate; (iii) supporting technological innovations that would permit more efficient exchange of information between the member rural banks, the apex, and the oversight departments of the Bank of Ghana, the central bank; (iv) strengthening the responsible departments of the BOG for more effective oversight of the activities of the apex bank and the member rural banks; (v) supporting training and other capacity building activities needed for the effective operations of the apex; (vi) establishing a decentralized clearing and support system of the apex consistent with the regional clearing systems of the Bank of Ghana; and (vii) financing technical assistance to put into place appropriate operational procedures and policies needed for transparency and accountability of apex operations. D. Institutional support to the Bank of Ghana and Ministry of Finance (Appraisal cost: US$1.4 million). This component aimed at strengthening the Banking Supervision Department (BSD) for effective monitoring and supervision, as well as strengthening the capacity of the Rural Finance Inspection Department (RFID) for rural finance policy and strategy formulation. These two departments are the focal points for monitoring of the rural and community banks. The investment in new technologies will facilitate improved reporting, standardization, and timely analysis of rural banks' returns to more effectively guide decision makers and provide an early warning system of changes in rural banks' performance. It will also reinforce the capacity of the Ministry of Finance (MOF) to coordinate various programs supporting rural, informal and microfinance institutions and to collaborate in the development of capacity building programs. The appraisal cost included US$3.13 million for physical and price contingencies, bringing the total

3 appraisal cost to US$22.96 million. d. Comments on Project Cost, Financing, Borrower Contribution, and Dates: The ICR does not provide information on the total actual costs of the project, the actual costs of the components or IDA's share in the components. IDA's actual costs reported in this review, US$ 5.67 million, are taken from Operations Portal. RFSP was cofinanced with IFAD, government counterpart funds were provided, while AfDB provided parallel funding and GTZ provided technical assistance. The magnitude of actual cofinancing is not listed in the ICR, although the share of each donor during appraisal was provided. According to the Government's ICR, actual cofinancing of US$ million was made up of US$11.02 million from IFAD, US$5.01 million from AfDB, US$0.75 million from the Government, and US$1.95 million from beneficiaries. There were substantial delays between approval (mid-2000) to effectiveness (end-2001), which the ICR attributes to the inauguration of the new government (that wanted to ensure commitments from the previous regime were consistent with its agenda). The closing date, December 31, 2007, was one year later than anticipated in the PAD. The ICR was put on hold because IFAD support to the project continued until September 2008, which allowed for a joint World Bank-IFAD ICR. 3. Relevance of Objectives & Design: The project's objectives were highly relevant, and were consistent with the Bank's strategies as evidenced by subsequent CASs. These indeed highlight private-sector led economic growth, agricultural development and the need to address lagging regions in particular the Northern areas of Ghana. Rural financial institutions are the primary vehicles for providing critically-needed financial services to Ghana's rural areas where an estimated 70 percent of the population live (at appraisal). Lack of access to purchase agricultural inputs, processing equipment, and working capital not only constrained agricultural productivity and value -addition, but simultaneously stymied ancillary services in rural -based agricultural and non-agricultural sectors. The rural sector was serviced mostly by informal organizations that were financially weak, fragmented and operated with outdated technologies--while oversight was thin and over-extended. Broadening and deepening the coverage of rural financial services required institutional strengthening, and RFSP's objectives responded to these needs. The project lays the foundation and paves the way for subsequent operations, from both the Bank and other development partners. RFSP incorporated lessons learned from the Rural Finance Project (closed in 1995) and from IFAD and AfDB projects that had, until RFSP, largely been uncoordinated. The objectives and components formed a clear whole, with the exception of the microfinance component (A). The latter was added after the design of the project was essentially completed and was not well integrated into the overall project, and this was reflected during implementation. This omission was also mirrored in the composition of the design team which could have benefitted from inputs from social development specialist. Specifically, the aim of integrating the informal and formal actors proved futile as there was no demand for such linkage because the actors essentially compete for the same customers. Stakeholders had not been consulted. Around the time of the 2005 MTR there was a clear understanding of this design flaw. The revision of the first component at mid term particularly aimed to correct the lack of specificity in the PAD's formulation of the target groups for end -user training. Training had initially been offered to Community-based organizations that had little to do with microfinance. This was properly corrected, but by then, the Training Fund had already been exhausted. The results framework had initial flaws (e.g., no benchmarks or targets were included ), which was mostly corrected in 2004 (discussed at greater length in section 10). 4. Achievement of Objectives (Efficacy): The achievement of each objective is analyzed below : i. Strengthening operational linkages between informal and semiformal microfinance institutions and the formal network of rural and community banks: Efforts were made at training end-users of rural MFIs, but results fell considerably short of the target set in the MTR : only 10,687 end-users were trained instead of the target of 44,926--mostly due to the exhaustion of the Training Fund. A total of 465 MFIs received training under the project, just short of the intermediate outcome target of 500. Of these MFIs trained, 348 benefited from repeat

4 training. A total of 17 good practice manuals were developed. The number of clients of rural MFIs increased substantially, from 1.3 million in 2001 to 3.2 million in 2006 (excluding clients of susu collectors ), which reflects an average annual increase of 17 percent, just short of the project development objective (PDO) indicator of 20 percent. This total constitutes an estimated 15 percent of the population, 45 percent of which are women and 30 percent of which are in the lower two poverty quintiles. The number of clients of susu collectors and their short -term deposits increased by 261 percent and 410 percent respectively from In the same period, membership of Credit Unions rose by 60 percent, and total assets increased by 248 percent. Average annual inflation between was about 20.5 percent (GDP deflator). In order to make a meaningful assessment of the financial performance of the MFIs, comparisons were made with the performance of reporting West African MFIs in the Microfinance Information Exchange (MIX): these show that MFIs in Ghana have deeper outreach in lending than those in neighboring countries. On the other hand, at the end of RFSP, MFIs in Ghana still depended on external financial support. Also, the operational self-sufficiency (OSS) index for Credit Unions in Ghana (98) is lower than that of their peers in West Africa (118). Achievement: Modest ii. Building capacity of the rural and community banks : Capacity was built through training of over 8,000 staff in customer care, treasury and credit management, anti -money laundering, internal controls, and check clearing. The AfDB project provided hardware support in the form of computers, safes etc. The number of Rural and Community Banks (RCBs) increased from 115 to 125 and there are now 564 networked business points across Ghana. The project missed the intermediate outcome target of having 100 percent of RCBs rated satisfactory. At about 75 percent in 2001, the biggest jump occurred between 2002 and 2003, when 12 RCBs moved from unsatisfactory to satisfactory, and the satisfactory rate went up to 90 percent. It stayed around that percentage until the end of the project (88 percent) in In addition to PDO indicators mentioned above, other project achievements are, for instance : the reduction in check clearing from two weeks to five days, and the approval of advances to salaried workers from 14 days to 3 days. An over-the-counter withdrawal now takes less than 9 minutes compared to almost 20 minutes previously. Non-performing loans are less than 12 percent of total lending portfolio (down from 20 percent at the start of the RFSP) and 99 percent of RCBs are profitable. However, only 80 percent of RCBs met capital adequacy ratio requirements. Under the MSI, strategic business plans were developed for the 15 selected RCBs and three training manuals developed. Ten RCBs reported putting their plans into action with a resultant increase in the number of microfinance clients and size of microfinance portfolios. At the end of FFSP, all participating RCBs were operationally self sufficient, with 12 exhibiting an OSS score in excess of international norms; similarly all RCBs were financially self-sufficient (FSS) and all in excess of the international norm, suggesting that extending product lines into microfinance did not undermine financial stability. The ICR does not (directly) follow up on several of the sub-components listed in the PAD, such as providing an in-depth assessment of financial and management performance of each rural bank, supporting a human resource development plan, and upgrading internal controls and procedures for greater transparency. Achievement: Substantial iii. Supporting the establishment of an apex structure for the rural banking system : This component successfully supported the establishment of a professional and functioning Apex bank that provides the institutional framework and facilitates the provision of common services. Offices were constructed, a system of money transfers between members was introduced, and specie was provided to all 127 RCBs. The Apex Bank Regulations were passed by Parliament, providing a legal framework for the network of rural banks. RFSP initiated a computerization program which has subsequently been picked up by Millennium Challenge Account (MCA). Experience this far suggests that while this has generated labor -saving efficiencies and makes oversight by the Apex Bank easier, the full management potential has yet to be realized by RCB managers. BoG is expecting the Apex Bank to be responsible for inspection of RCBs and the ICR reports that it remains unclear who will bear the costs of this function: estimated costs are too high for individual RCBs to bear from current profit margins.

5 However, both the Apex Bank's OSS and profit have been declining significantly, posing some threat to the sustainability of services provided to the RCBs. The OSS index fell from 108 in 2004 to 69 in far below the minimum acceptable threshold of 120 for sustainable financial institutions, while profit dropped by almost 80 percent in the same period. At project closure, the Apex bank still depended on government and donor funding, and faces tension between its public good function and its commercial aspirations. Achievement: Modest iv. Strengthening the institutional and policy framework : Training and equipment were provided to the Banking Supervision Department (BSD) of the Bank of Ghana, and to the Microfinance Unit (MFU) of the Ministry of Finance. Knowledge and skills of BSD staff are said to have improved, allowing examination of each RCB at least once annually. Remedial measures have been taken to address distressed RCBs (including the closure of some failed RCBs). According to the ICR, this would have taken longer to put in to place prior to project interventions. A Facilitating Agency supported the MFU in setting down procedures, standards and best -practice techniques for MFIs (as discussed under Component (i)). However, the microfinance sector still lacks a clear policy framework as the Government failed to approve the Ghana Microfinance Policy Framework (GHAMP), which means that policy coordination has to take place in the absence of a sector strategy or policy. Achievement: Substantial 5. Efficiency (not applicable to DPLs): The PAD did not include specific ERR or FRR targets. However, several indicators were included to measure or serve as proxies for cost -effectiveness in the informal and microfinance sector, although the lack of sufficient baseline data was a complicating factor. Some usable targets are: an increase in deposits by 100 percent (exceeded), and an increase in the share of credit in RCB assets from 33 to 40 percent (exceeded). The PAD also aimed to increase the cost-recovery of training from 25 to 75 percent, but the ICR does not follow up on this indicator with clear data. The financial analysis in the PAD included simulations for the Apex Bank and RCB performance. The ICR argues that the relevance of these simulations has been undermined by the rapid transformation of cost structures and revenue streams, particularly due to a lower than expected return on treasury bills. a. If available, enter the Economic Rate of Return (ERR ERR)/Financial Rate of Return (FRR FRR) at appraisal and the re-estimated estimated value at evaluation : Rate Available? Point Value Coverage/Scope* Appraisal % % ICR estimate % % * Refers to percent of total project cost for which ERR/FRR was calculated. 6. Outcome: The outcome rating for the project is moderately satisfactory. The Rural Financial Services Project has generated some substantive results. With the RFSP's assistance, the number of clients of rural MFIs increased substantially, as did the total deposits. Other PDO indicators also showed good progress, although not always meeting targets. Good practice manuals have been developed to sustain best practice, and enhanced inspection and regulation strengthen financial performance and facilitate early response. The project suffered somewhat from the initial poor integration of the microfinance component, and its initial focus to link formal and informal financial sector entities--which proved to be flawed. The Apex Bank is fully operational, but its OSS and profit have declined. Moreover, the microfinance sector still lacks a clear policy framework. Somewhat disturbing is the 2006 Beneficiary Assessment finding that there was generally an improved outreach to clients, in terms of loans, but bank managers were reluctant to give loans to farmers due to the high default of loan repayments by farmers. As a proxy for the impact on the agriculture sector, this raises the question to what extent the main perceived beneficiaries were reached. a. Outcome Rating : Moderately

6 7. Rationale for Risk to Development Outcome Rating: Strong Apex organizations have been established among MFIs, susu collectors and Credit Unions with clear roles and mandates and strongly supportive constituencies. Manuals were delivered to continue best practice, while supervision capacity was enhanced. Among the RCBs, 88 percent was rated satisfactory in Ongoing support by the Bank's Rural and Agricultural Finance Program and those of other development partners reflect a continuation of capacity building. That said, some risks appeared. The Apex Bank's profitability has been declining as well as its score for operational self-sufficiency (see section 4). As the ICR points out, it faces an uncertain future as it is seeking to reconcile its role as an association of RCBs with a degree of social function on the one hand with ambitious commercial aspirations comparable to those of commercial banks on the other. Related risks are the transfer of responsibility for the supervisory functions of RCBs from the BoG to the Apex Bank without commensurate resources, and the continued financial viability of the Apex associations which lack core funding. Adding some risk is the absence of a comprehensive policy for microfinance passed by the Government, and the latter's continued deployment of subsidized credit that distorts incentives and undermine financial discipline among borrowers. a. Risk to Development Outcome Rating : Significant 8. Assessment of Bank Performance: The project incorporated experience from prior (Bank) operations, and set out a clear structure for institutional capacity building for RCBs under an Apex umbrella, with enhanced supervision. However, the first component was initially not well integrated into the design of the entire project, which led to e.g., the flawed notion of linking the informal with formal financial sector entities. Also, the lack of a clearly formulated target group initially led to training that was offered to entities that has little to do with microfinance. (Also see section 3.) The Bank undertook regular supervision missions and in a majority of these IFAD representatives joined in, which--according to the ICR--contributed to implementation support and helped build a strong unity of purpose between partners. The ICR reports difficult cooperation with AfDB supervision missions due to their lack of country presence and ad-hoc timing of supervision missions, while a solid rapport with the GoG, BoG and other counterparts occurred. The M&E structure was revised in 2004, following up on issues raised in ISRs. The MTR took place in 2005, two years later than scheduled due to the initial delays in project effectiveness, and led to a revision of the first component. a. Ensuring Quality -at at-entry Entry:Moderately Unsatisfactory b. Quality of Supervision : c. Overall Bank Performance :Moderately 9. Assessment of Borrower Performance: The country team reported strong overall government commitment for the operation. However, three factors prevent a higher rating: there were initial delays following the inauguration of the new government, the government persisted in directing subsidized credits, and GHAMP was not approved. The ICR argues that the decision to locate the project implementation unit within the BoG proved to be a good one. BoG was committed to the transformation of the rural finance system and proved to be a strong supporter of the network of rural banks and the Apex Bank throughout implementation. However, the ICR also argues that support to the microfinance component was less forthcoming as this component was implemented by the Microfinance Unit (MFU) in MoFEP. Reportedly, the use of a Facilitating Agency (FA) to manage implementation of this component could not offset the limited capacity and experience of the MFU. a. Government Performance :Moderately

7 b. Implementing Agency Performance :Moderately c. Overall Borrower Performance :Moderately 10. M&E Design, Implementation, & Utilization: The PAD's results framework and M&E arrangements had significant design weaknesses. No benchmarks or quantitative targets were included, and some of the qualitative targets proved confusing. The Bank undertook semi-annual supervision missions, which resulted in moderately unsatisfactory ratings for the M&E component in two ISRs. A 2004 workshop helped initiate improvements in the range and quality of indicators, as well as the structure and design of the data collection. However, project development objective indicators (PDOs) were operationalized with annual targets (proportionate increases and a proportion of savings ) and end-of-project (absolute) targets without reference to a baseline value. This meant that the absolute targets became inappropriate for the use as end-of-project goals. In addition, the three main PDOs did not specify between RCBs and more informal financial sector entities, but instead, seem to cover only RCBs. In that sense, the lack of integration of the component supporting the informal financial sector was reflected in the M&E system. The project strengthened the monitoring and information system by adjusting key modules, forming the foundation for on-going monitoring. The core implementing agencies were reportedly adequately collecting and monitoring data related to their respective component activities. That said, the ICR mentions that the extent of evaluation and in-depth analysis of the data has been weak, in part because of the inability of the M&E office to lead and commission profound data evaluation. The Bank undertook and a 2005 MTR that led to e.g., a revision of the first component on the capacity building for the informal financial sector. Other qualitative evaluation reports have been prepared, such as the 2006 Beneficiary Assessment. Also, the introduction of the Microfinance Information Data Exchange (MIX) established an internationally recognized performance monitoring system for MFIs. a. M&E Quality Rating : Substantial 11. Other Issues (Safeguards, Fiduciary, Unintended Positive and Negative Impacts): The project did not trigger any of the Bank's safeguard issues. The ICR mentions that at inception the project s Financial Management (FM) assessment noted that the Rural Finance Inspection Department lacked the requisite capacity for implementation. An action plan was developed to address these weaknesses, which was successfully implemented. The projects FM rating has consistently been satisfactory including ISRs for the last two years of the project (December 2006 and 2007). The Bank of Ghana ensured a sound control environment and all financial covenants were complied with. All project audits were submitted within the deadline of six months following the end of the fiscal year. 12. Ratings: Risk to Development Outcome: Outcome: Bank Performance : Moderately ICR IEG Review Moderately Reason for Disagreement /Comments Capacity building among end-users was incomplete, sustainability of the Apex Bank is in question, incomplete impact on agriculture due to reluctant lending to farmers, among other reasons. Also see section 6. Moderate Significant Apex Bank's role and financing (declined OSS and profit) face an uncertain future, among other reasons. See also section 7. Moderately

8 Borrower Performance : Moderately Quality of ICR : Moderately NOTES: - When insufficient information is provided by the Bank for IEG to arrive at a clear rating, IEG will downgrade the relevant ratings as warranted beginning July 1, The "Reason for Disagreement/Comments" column could cross-reference other sections of the ICR Review, as appropriate. 13. Lessons: IEG supports the lessons learned in the ICR, including the importance of a well designed M&E system with well defined quantitative targets. 14. Assessment Recommended? Yes No 15. Comments on Quality of ICR: The quality of the ICR is satisfactory. It provides a candid account of the activities undertaken under RFSP, although it does not neatly follow up on each of the stated goals under the four project components. Also, there is some discrepancy between judgments made about the Microfinance Unit (MFU) in the Ministry of Finance in sections 2.1 and 5.2 of the ICR. It would be informative to analyze how interest rates on loans have evolved. The ICR should contain the actual total costs of the project, and how these varied by component. In addition, a table detailing the planned and actual costs of each component, broken down by development partner (and counterpart), would have been welcome. a.quality of ICR Rating :