Ghana - First and Second Financial Sector Adjustment Credits

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 16789 PERFORMANCE AUDIT REPORT GHANA FINANCIAL SECTOR ADJUSTMENT CREDITS I AND II (CREDIT 1911-GH AND CREDIT 2318-GH) June 25, 1997 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Currency Equivalents (As of May 10, 1994) Currency Unit = Cedis (0) US$1 = (C)920.75 (C)1 US$0.67 Abbreviations and Acronyms BHC Bank for Housing and Construction BOG Bank of Ghana CDH Consolidated Discount House CEM Country Economic Memorandum ERP Economic Recovery Program FINSAC Financial Sector Adjustment Credit FINSAP Financial Sector Adjustment Project GCB Ghana Commercial Bank GIB Ghana Institute of Bankers ICAG Institute of Chartered Accountants of Ghana IDA International Development Association IMF International Monetary Fund IPO Initial public offer MFEP Ministry of Finance and Economic Planning NIB National Investment Bank NBFI Non-Bank Financial Institution NPART Non-Performing Asset Recovery Trust NSCB National Savings and Credit Bank OED Operations Evaluation Department PAR Performance Audit Report PCR Project Completion Report PFP Policy Framework Paper PNDC People's National Democratic Committee SDR Special Drawing Rights SSB Social Security Bank SSNIT Social Security and National Insurance Trust Fiscal Year Government: January 1 - December 31 Director-General, Operations Evaluation Mr. Robert Picciotto Director, Operations Evaluation Dept. Mr. Roger Slade for Ms. Elizabeth McAllister Acting Division Chief : Mr. Rene Vandendries Task Manager : Mr. Nicolas Mathieu FOR OFFICIAL USE ONLY The World Bank Washington, D.C. 20433 U.SA Offme of the Director-General Opwations Evaluation June 25, 1997 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Ghana Financial Sector Adjustment Credits I & II (Credits 1911 and 2318-GB) Attached is the Performance Audit Report (PAR) on the Financial Sector Adjustment Credits I & II (Credits 1911 and 2318-GH) prepared by the Operations Evaluation Department. The first credit for an amount of US$106.6 million equivalent was ipproved in FY88 and closed in FY93. The second credit of US$100 million equivalent was approved in FY91 and is scheduled to be closed in FY97. The credits were cofinanced by the Overseas Economic Cooperation Fund of Japan and the Swiss Government. There were no cancellations. Financial Sector Adjustment Credit I The credit was to support the first phase of a financial sector adjustment program. The main objectives were to: (i) liberalize the financial sector through the removal of interest rate controls and sectoral credit targets; (ii) enhance the soundness of banking institutions by putting in place a new legal and regulatory framework; (iii) strengthen the banking supervision function of the central bank; and (iv) restructure distressed banking institutions and attempt to recover their nonperforming assets. All major institutional objectives of FINSAC I were reached. Financial infrastructure has been developed and the framework to restructure distressed banks was put in place. The underlying objective of financial sector deepening, however, was not reached, because of the deteriorating fiscal and macroeconomic performance starting in 1992. Although interest rates were liberalized, bank deposit rates remained negative in real terms, and private bank deposits as a proportion of total deposits declined significantly in the early 1990s. , Given the partial achievement of project objectives, the PAR rates the outcome of FINSAC I marginally satisfactory, instead of satisfactory as in the PCR review. In addition, the sustainability of project benefits is rated as uncertain instead of likely in the PCR, because the macroeconomic environment remained unfavorable. Institutional development impact is rated as substantial, which is consistent with the PCR review, since the banking system was successfully restructured. Financial Sector Adjustment Credit II Under FINSAC II, the Government intended tc consolidate the achievements of FINSAC I and move toward further deepening, diversification and streigthening of the financial sector. The objectives of FINSAC II included: (i) encouraging positive real deposit and lending rates by remunerating legal reserve requirements; (ii) increasing competition in the banking system; (iii) pursuing the bank restructuring program, including privatization, and the recovery of nonperforming assets; and (iv) upgrading professional skills of bankers, accountants and insurance professionals. 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. 2 The project has made significant progress in implementing its institutional development impact components. Individual restructuring plans for distressed banks prepared under FINSAC I have been fully implemented and have resulted in significant improvements in both financial and operational performance of the banks since 1991. The performance of the institution established to facilitate the recovery of nonperforming assets was satisfactory, since most of the amounts targeted during the period 1990-95 were collected. The financial sector policies supported by the FINSAC II program, however, did not lead to financial sector deepening because of continuing fiscal imbalances and macroeconomic instability. Private financial resources were diverted to the public sector through onerous government borrowing requirements. The PAR rates outcome of FINSAC II as marginally satisfactory, sustainability as uncertain and institutional development impact as substantial. Lessons One lesson is that financial sector loans with financial deepening objectives and which include policy reforms should be implemented in periods of macroeconomic stability. Moreover, macroeconomic performance conditions from tranche release should be clearly specified and strictly followed. Another lesson is that financial infrastructure and institutions take a long time to restructure. Financial sector institution building projects should precede financial sector policy loans, especially in periods of macroeconomic imbalance. Attachment Contents Preface....... ........................................................ 3 Basic Data Sheets ....................................................... 5 Evaluation Summary ....................................................9 1. B ackground ............................................................................................................................. 17 The Economic Reform Program.......................................17 Background to the Economic Recovery Program.........................17 Economic Performance During FINSAC I and II.........................18 Current Challenges ..............................................19 The Financial Sector Adjustment Project.................................20 Structure of the Financial Sector.................................... 20 Institutions..................................................20 Policy and Regulatory Framework.................................21 2. Bank Strategy and Project Objectives, Design and Relevance ...................27 The Bank's Assistance Strategy....................................... 27 FINSAC I th Ft................. o...................................... 28 Objectives of FINSAC I.....................20...... .............28 Project Design of FINSAC IRe...........................................28 FINSAC I ...................................................... 29 Objectives of FINSAC I ...................28....... .............29 Project Design of FINSAC I...................................... 30 Relevance of Strategy, Project Objectives and Design ................. ....... 32 3. Project Implementation ..... .................................. ........35 FINSAC I........ ................................................. 35 Credit Processing .......................................................................................... ..... 35 Macroeconomic Management ........................................ 35 Liberalization of Financial and Banking Environment ....................... 35 Legal and Regulatory Framework .....................................36 Bank Supervision..............................................36 Restructuring the Distressed Banks..................... .............37 Recovery of Non-Performing Assets................... ..............38 Development of Money and Capital Markets............... ............38 Corporate Restructuring...........................................38 Strengthening the Accounting and Audit Professions......................39 Training Bankers........................... ....................39 FINSAC ....................................................... 39 Credit Processing............................................... 39 Macroeconomic Management ........................................... 40 Nicolas Mathieu (Task Manager) acknowledges with thanks the contributions of: Ashok Khanna and Melanie Johnson (Consultants); and Jasmine Mason-Anderson and Eneshi Irene K. Davis who provided administrative assistance. 2 Removing Policy Distortions ......................... ..... ........ 41 Improving Competition: Divestiture of Banks .......................... 41 Enhancing the Financial Condition and Effectiveness of BOG .................... 42 Continuing Bank Restructuring .....................................43 Development of Non-bank Financial Institutions .................... .....43 Training of Bankers and Accountants ................................. 44 Informal Financial Sector .......................................... 44 4. Project O utcom e ..................................................................................................................... 45 Overall Contribution of the Financial Sector to Economic Growth ..... ..........45 Financial Policy .......................... ................ 46 Financial Infrastructure ........................... ................ 48 Bank Restructuring and Divestiture.............................49 Capital Market Development.......................................51 5. Role of the Bank and Borrower ........................................................................................ 53 Bank Performance ....................................... .......... 53 Borrower Performance ................................................... 53 6. Overall Evaluation and Ratings ........................................................................................ 55 Contribution to Financial Intermediation and Deepening ......................... 55 Contribution to Financial Policy Development........................ 55 Contribution to Financial Infrastructure ......................... ......... 55 Bank Restructuring and Divestiture ..................................... 55 Capital Market Development .............................................. 56 Overall Outcome ................................... ..... .......... 56 7. Sustainability and Lessons Learned.................................................................................. 57 Sustainability .................................................... 57 Lessons of Experience ..................... .................. 57 Annexes: I. FINSAC I - Policy Matrix ........................................... 59 II. FINSAC II - Policy Matrix..................... ................ 65 III. The Bank of Ghana's Comment .............................. ......... 69 IV. FINSAP Secretariat's Comment .......................................75 V. Federal Office for Foreign Economic Affairs' Comment............................ 83 VI. The Overseas Economic Cooperation Fund of Japan's Comment.. ..................... 87 Tables: 1.1. Macroeconomic Indicators, 1988-94 .......................................... 18 1.2. Banking System Indicators, 1988-95 ......................................21 1.3. Inflation, Real Interest Rates and Nominal Spreads in Ghana, 1988-95 ..... .........22 4.1. Financial Sector Indicators, 1988-95....................................... 45 4.2. Operating Asset Ratios, 1988-95: All Banks. ....................... .......... 50 4.3. Capital Adequacy Ratios of Banks, 1989-95............................. 50 4.4. Comparison of Public and Private Sector Banks' Intermediation Costs and ROE...........50 3 Preface 1. This is a Performance Audit Report (PAR) on the First and Second Financial Sector Adjustment Credits (191 1-GH and 2318-GH) for Ghana in the amounts of US$106.6 million and US$100 million equivalent, respectively. The first credit was approved on May 31, 1988 and closed on December 31, 1992, having been fully disbursed. The second credit was approved in December, 1991 and has not yet closed. Two tranches have been released and the disbursement of the third tranche was authorized in December 1996. 2. The Project Completion Report for the first credit was prepared by the Africa Regional Office. It provides a description of the financial sector adjustment project's implementation experience and lessons. The PAR was prepared by the Operations Evaluation Department (OED). It builds on the PCR by looking at the project's impact on broader indicators of financial sector development. An OED mission visited Ghana in June, 1996 to discuss the project's effectiveness with Government officials, business and bank representatives, private enterprises and donors. Their kind cooperation and valuable assistance is gratefully acknowledged. 3. A draft report was sent to the Borrower and cofinanciers for comments. The comments from the Bank of Ghana (BoG), the FINSAP Secretariat, the Swiss Federal Office for Foreign Economic Affairs, and the Overseas Economic Cooperation Fund of Japan (OECF) are incorporated in the report as Annexes III to VI.  5 Basic Data Sheet FINANCIAL SECTOR ADJUSTMENT CREDIT I (CREDIT 1911-GH) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as % of estimate current appraisal estimate estimate Total project costs 240.0 225.3 - Credit amount 100.0 98.3 98.3 Cofinancing 140 127.0 - African Development Fund - 50.0 - Switzerland - 15.0 - Japan - 62.0 - Cancellation - - - Date physical components completed - - - Economic rate of return - - - Institutional performance - - - Cumulative Estimated and Actual Disbursements FY89 FY90 FY91 FY92 FY93 Appraisal estimate (US$M) 75.0 95.0 100.0 - - Actual (US$M) 42.3 48.7 93.9 96.0 98.3 Actual as % of appraisal 56.4 51.3 93.9 - - Date of final disbursement: March 23, 1993 Project Dates Original Actual Initiating memorandum 02/88 02/88 Negotiations 05/88 05/88 Letter of Development Policy 03/88 03/88 Board approval 05/88 05/88 Signing 06/88 06/88 Effectiveness 08/88 08/88 Closing date 09/90 12/92 6 Staff Inputs (staff weeks) FY89 FY90 FY91 FY92 FY93 FY94 FY95 FY96 Total Preappraisal 1.2 15.5 - 75.2 - - - - 91.9 Appraisal - - 4.0 29.2 - - - - 33.2 Negotiations - - - 10.8 - - - - 10.8 Supervision - - - 22.9 29.7 27.0 12.7 4.0 69.3 Other - - - Total 1.2 15.5 4.0 138.1 29.7 27.0 12.7 4.0 205.2 Mission Data Date No. of Staff weeks Date of Report (month/year) persons infield Appraisal 02/88 3 9 03/88 Supervision I 02/89 4 16 03/89 Supervision II 05/89 2 2 06/89 Supervision III 09/89 3 12 10/89 Supervision IV 02/90 1 3 05/90 Supervision V 06/90 8 32 08/90 Supervision VI 11/90 3 15 12/90 Supervision VII 06/92 3 12 07/92 Supervision VIII 08/92 4 4 08/92 Supervision IX 10/92 3 6 11/92 Supervision X 0993 1 2 - Other Project Data Borrower/Executing Agency: Govermnent of Ghana FOLLOW-ON OPERATIONS Operation Credit no. Amount Board date (US$ million) Financial Sector Adjustment Credit II 2318-GH 100.0 12/91 7 FINANCIAL SECTOR ADJUSTMENT CREDIT II (CREDIT 2318-GH) (Note: This credit is not yet closed) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 110.0 110.0 100.0 Credit amount 100.0 70.6 70.6 Cofinancing - Switzerland 10.0 10.0 100.0 Cancellation - - - Date physical components completed - - - Economic rate of return - - - Institutional performance - - - Cumulative Estimated and Actual Disbursements FY92 FY93 FY94 FY95 FY96 FY97 Appraisal estimate (US$M) 50.0 100.0 - - - - Actual (US$M) 20.0 28.9 61.5 65.4 70.0 70.6 Actual as % of appraisal 40.0 28.9 - - - - Date of final disbursement: ONGOING Project Dates Original Actual Negotiations 11/91 11/91 Letter of Development Policy - 11/91 Board approval 12/91 Signing - 02/92 Effectiveness 01/92 05/92 Closing date 03/95 03/97 8 Staff Inputs (staff weeks) FY89 FY90 FY91 FY92 FY93 FY94 FY95 FY96 Total Preappraisal 1.2 15.5 83.2 - - - - - 99.9 Appraisal - - 4.0 29.2 - - - - 33.2 Negotiations - - - 10.8 - - - - 10.8 Supervision - - - 22.9 29.7 27.0 12.7 34.4 126.7 Other - - - - - - - - Total 1.2 15.5 87.2 62.9 29.7 27.0 12.7 34.4 270.6 Mission Data Date No. of Staff weeks Date of Report (monthlyear) persons in field Appraisal 06/22 - 07/91 Supervision I 11/92 - 2.4 - Supervision II 09/93 4 3.4 10/93 Supervision III 11/93 - 2.6 Supervision IV 07/94 4 3.8 07/94 Supervision V 11/94 3 3.2 01/95 Supervision VI 06/95 2 4.2 08/95 Supervision VII 09/95 4 2.6 11/95 Supervision VIII 11/95 3 4.0 03/96 Other Project Data Borrower/Executing Agency: Government of Ghana FOLLOW-ON OPERATIONS Operation Credit no. Amount Board date (US$ million) None 9 Evaluation Summary Macroeconomic Background 1. Ghana once enjoyed a relatively high standard of living compared to other West African countries, but poor economic management during the 1970s and early 1980s led to protracted economic decline. In 1983 the Government adopted an Economic Recovery Program (ERP) under which it devalued the currency, dismantled most price and distribution controls, eliminated many subsidies, broadened the tax base, improved tax collection, and provided more adequately for maintenance and capital expenditure. Following adoption of the ERP, and considerable donor support, GDP growth averaged about 5% per year, compared to losses of 2-3% per year during the preceding decade. On a per capita basis, real income grew by 2% per year. 2. While real GDP growth ranged between 3.3 and 5.6% during 1988-94, inflation and fiscal restraint remained important concerns in the overall macroeconomic management of the economy during the implementation of FINSAC I and II. The overall fiscal deficit (including external assistance) had been turned into a surplus of 1.5% of GDP in 1991. Inflation fell from 40 percent in 1987 to 32% in 1988, 25% in 1989, and was up again to 37% in 1990. To offset the rise in inflation in 1990, the Government undertook measures to restrict monetary growth, including the sale of BOG bills to mop up excess liquidity in the economy and a sharp increase in the discount rate. By late 1991, the inflation rate had dropped to 18%. Private investment had risen continuously, from a very low base, to a ten-year high of 8.1% of GDP. 3. Instead of building on these gains, however, the Government reversed itself in 1992 and undermined macroeconomic stability. An across-the-board wage increase of 80% for civil servants, a decline in tax revenues and a shortfall in foreign project grants resulted in a fiscal deficit of 4.8% of GDP in 1992. In 1993, inflation rose dramatically to 25% and remained at about the same level in 1994. The Government then took new measures to improve the fiscal situation. By 1994, divestiture receipts helped to generate a fiscal surplus of 2.1% of GDP. In 1995, however, it appears that inflation again accelerated to an annual rate of about 60%. Financial Sector Background 4. Historically, the formal financial system has not been an effective vehicle for mobilizing domestic resources in Ghana. In 1993, the ratio of M2 to GDP remained very low at 17 percent. This level is significantly lower than those in other Sub-Saharan African countries. In addition, Ghana has remained a cash-based economy. Currency as a percentage of M2 was at least twice as high as in Kenya and Zimbabwe in 1992 and is at least three times as high as in Indonesia, Malaysia and Thailand. In most economies, individuals hold relatively large cash balances when economic stability is in question. 5. IDA's work on the financial sector in Ghana began in 1985. The financial sector at that time was in severe distress, posing a constraint to growth in the real sectors of the economy. A major financial sector review, carried out in 1986, revealed several deficiencies in the financial system. These included an insolvent banking system dominated by state-owned banks and saddled with large non-performing loans and excessive intermediation costs; a low level of financial intermediation, reflecting a poor record in domestic resource mobilization; inefficient 10 credit allocation, due in part to central bank-directed interest rates and sectoral credit targets; the virtual absence of a money market and capital market; and weaknesses in the BOG supervisory system and regulatory framework. 6. At the time that FINSAC I was conceptualized, the formal financial system in Ghana was dominated by government-owned banks. Three commercial banks (Ghana Commercial Bank, Standard Chartered Bank of Ghana and the Barclays Bank of Ghana) accounted for about two-thirds of the banking system's deposits and assets. Ghana Commercial Bank alone controlled about 44 percent of the banking system's assets and deposits. There were also seven secondary banks (primarily sectoral development finance institutions and a few smaller commercial banks), a small cooperative bank and over 100 rural banks. All but one secondary commercial bank (Bank of Credit and Commerce) were partly or wholly owned by the Government. Thus, the banking system was concentrated and oligopolistic. The Bank of Ghana had oversight responsibility for the financial system and played a significant role in financing government expenditures. The money market was in an embryonic state in the late-1980s and the capital market was very rudimentary. 7. In the non-bank financial sector, institutions included the Social Security and National Insurance Trust (SSNIT) and 11 insurance companies. SSNIT was a government-owned institution which collected social security payments and invested them almost exclusively in government paper and bank deposits. The deficiencies of the banking system such as negative real interest rates had spawned a myriad financial agencies including money lenders, thrift groups, credit unions, etc. The informal financial system is estimated to be quite sizable. Bank Strategy and Project Objectives, Design and Relevance 8. FINSAC I & II supported a comprehensive financial sector reform program with the objective of reviving the financial system, putting the banking system back on a sound footing, improving financial policies and initiating the development of non-bank financial institutions and the capital market. The reform program was implemented under two credits over a number of years. Both credits had substantial sums allocated for technical assistance for implementing the project and to provide training and other systems support. The specific objectives and design of FINSAC I were to: * ensure the soundness of banking institutions through improved prudential regulation and supervision; restructure financially distressed banks; and improving financial policies by deregulating interest rates, bank credit ceilings, charges and commissions. The objectives and design of FINSAC II were to: * support the removal of remaining policy distortions in the financial system by, inter alia, encouraging positive real deposit and lending rates through the payment of adequate interest on the banks' legal reserve requirements, and enhance the effectiveness of the Central Bank to lead the continuing financial sector reform project; 11 * continue the bank restructuring project; pursue the recovery of non-performing assets by Non-Performing Asset Recovery Trust (NPART); enhance the effectiveness of non-bank financial institutions; and upgrade the professional skills of bankers, accountants and the insurance profession. 9. The assistance strategy for the country and financial sector can be gleaned from the project objectives which implicitly assume the Government will maintain macroeconomic stability. Indeed, Structural Adjustment Credits I & II and the IMF ESAF were under implementation during the period. The primary objective of FINSAC I was modest-to restructure distressed banks rather than increase intermediation or promote financial deepening. FINSAC II was expected to result in financial deepening as banks would have been restructured and financial policy reforms implemented. In addition, bank restructuring was expected to be completed, the divestiture of government ownership begun, and entry into banking permitted. At the same time, a regulatory and supervision framework for non-bank financial institutions was to be developed. 10. The strategy for financial sector development and project design was appropriate, with one caveat. It was risky to implement interest rate liberalization while most banks were insolvent, bank supervision had not yet been upgraded and credit ceilings were in place. Moreover, the use of indirect monetary policy instruments should have been delayed until the central bank had been strengthened. Exposing banks and the central bank to additional risks at a time when both were in financial distress and lacked the skills to manage the risk was inappropriate. The design of FINSAC I and II was grounded in the strategy articulated above and the detailed elements were based on the Financial Sector Review (1986) and a series of technical papers prepared in 1990/91, which could not be located. The projects covered every aspect of financial sector development. They may have covered too much and contained too many performance conditions for tranche releases. A narrower focus on financial policies and infrastructure and restructuring and privatizing the distressed banking system-by far the biggest financial intermediaries-would have been more appropriate. Finally, tranche releases based on unspecified macroeconomic targets and uncertain transactions such as banks' divestiture to a strategic investor were inappropriately open-ended. Nevertheless, the bank strategy embedded in project objectives and project design was clearly relevant for financial sector development to enhance private sector activity. Project Implementation 11. Despite some delays, the components of the two credits were effectively implemented and Ghana's financial policies were dramatically changed. Interest rates were liberalized, sectoral credit targets and credit ceilings removed, banking charges decontrolled and taxation of banks unified with corporations. The legal and regulatory framework for the financial sector was modernized to give BOG appropriate powers to manage the banking and non-bank financial institutions sectors and capital market regulations were put in place. BOG was reorganized and its staff trained to improve monetary management and supervision of the bank and non-bank financial system. Distressed banks were successfully restructured and prepared for privatization. Two banks have offered shares to the public and a third is in a position to follow suit. Strategic investors have been located for these banks, which triggered the release of the third Tranche. Targets for collections from non-performing assets hived off from restructured banks have been substantially met. The accounting and audit professions have been strengthened and facilities for 12 training bankers were established. Finally, studies have been completed on corporate restructuring and the informal financial sector. Project Outcome 12. While the two FINSAC projects have contributed importantly to institutional strengthening of the financial sector, to date they have had little impact on increasing financial intermediation or deepening the financial system. Most of the indicators for financial sector intermediation have either declined or remained stagnant during the period. M2/GDP has remained flat throughout the 1988-95 period. More importantly, currency held outside of banks, which had dropped during 1990-91, has been increasing over the past 4 years. Thus, little financial deepening has occurred during the project implementation period. In addition, the financial system's performance in mobilizing savings remains inadequate. Private savings as a percentage of GDP remained extremely low at 2.6% of GDP in 1994. Gross domestic savings, already low at 7.8% of GDP in 1991, became negative in 1993, and had only recovered to 4.4% of GDP in 1994. Not surprisingly, private sector investment as a percentage of GDP dropped by nearly 50% in 1992 and has not yet recovered. These results suggest that the improvements in the financial sector have been unable to generate the results expected in the face of the deterioration in the country's macroeconomic performance. 13. The reform of financial sector policies has had little impact on the efficiency and effectiveness of financial sector. Although interest rate liberalization was a step in the right direction, it may have been premature in 1988. In fact, in other countries, liberalization of interest rates prior to the establishment of a sound regulatory, legal and supervisory framework had been deleterious, particularly when banks remain insolvent and seek out deposits by offering high interest rates as well as higher and higher risk Borrowers at high lending rates. Although Ghana seems to have escaped that fate, other policies and practices, such as high reserve requirements, the Government's ownership of banks, and moral suasion regarding interest rate spreads and credit targets, have hampered competition and allowed the Government to continue to use the banking sector to finance its expenditures. Thus, the conduct of monetary policy has been driven by fiscal needs and has not contributed positively to macroeconomic management. 14. Ghana has made excellent progress in updating the financial system's legal and regulatory framework and putting in place an appropriate system for monitoring financial institution performance. These efforts provide a solid foundation for the future development of Ghana's financial system. Bank restructuring has been useful in cleaning up the balance sheets of the banks, refocusing their attention on streamlining their operations and initiating the divestiture process. In addition, strategic investors have been found for these institutions. For the larger public banks, such as GCB, a strategic investor is a critical element in changing the management structure and philosophy of the bank. 15. In capital market development, the goals of the two FINSAC projects were appropriately modest. Technical assistance has been useful in developing the stock market and providing an adequate supervisory and legal framework for non-bank financial institutions and capital Bank regional staff notes: "This conclusion is reasonable. However, the policy reforms have been put in place, regulatory and supervisory regime enhanced and institutional capacity strengthened. Without these, the situation may have been worse. Instead, the sectors now poised to take off once the macroeconomic environment improves..." 13 markets. It is unlikely, however, that the capital markets will experience much growth until inflation comes down to more manageable levels. In addition, given the complexity of the two projects, adding capital market development tasks to the project may have diverted attention from banking sector issues.2 Evaluation and Ratings3 Bank Performance 17. Project design was generally appropriate although some issues in sequencing of financial policies and conditions of tranche releases were highlighted above (para. 10). 18. Project Preparation and Supervision. Bank and Borrower cooperation in project preparation and supervision appear to have been good. Appraising FINSAC I took only 6 staff weeks, owing in part to detailed sector work prior to the appraisal mission. As the project was complex, with substantial technical assistance, it needed 104 staff weeks for supervision. FINSAC II is still under implementation. Overall, Bank performance is rated as satisfactory for FINSAC I and II. Borrower Performance 19. The Borrower has implemented the technical aspects of these complex projects in a satisfactory manner. In FINSAC II, an implementation secretariat was established to oversee the task. A small group of senior Ghanaian experts, including a former governor of BOG, assisted by a technical expert from overseas, are implementing the project. Bank privatization is proceeding but locating a strategic investor has proved difficult. The Borrower has made good faith efforts to find strategic investors for the restructured banks but the final tranche of FINSAC II has not yet been released. While glitches have occurred with foreign technical experts (including the quality of their work) the projects have been implemented satisfactorily, despite the delays in the divestiture. Project implementation has primarily been marred by the macroeconomic circumstances not being conducive to the project's success. On balance, Borrower performance is rated as marginally unsatisfactory for FINSAC I and II. Contribution to Financial Intermediation and Deepening 20. The primary objective of financial sector operations is to deepen the financial sector, increase financial intermediation and improve its quality. Several indicators in Table 4.1 (below) suggest that between 1988 and 1995 there was no substantial deepening in Ghana's financial sector and that financial intermediation to the private sector did not increase. Judging by low loan loss provisions made by the banks in recent years, the quality of financial intermediation may have improved. As bank loans have been made at negative real interest rates 2 Bank regional staff notes: "The focus of the first project was largely on specific banking sector issues (bank restructuring, legal framework, supervision). The same focus continued under the second project. Both projects, however, did lay the groundwork for capital markets development, so as not to miss out on this less urgent but nevertheless important objective." As FINSAC II is not yet completed, ratings for its achievements are provisional, and based on the current situation. 14 for several years during the project, however, the quality of financial intermediation remains uncertain. Contribution to Financial Policy Development 21. The financial policy components of FINSAC I to liberalize interest rates, eliminate credit allocation and develop money markets and instruments for monetary policy operations were consistent with good economic policy. The timing, phasing and sequencing of implementation could have been better, however. Liberalizing interest rates during a period of high inflation and large government borrowing requirements was excessively risky. Moreover, liberalizing rates was risky while credit allocation remained in place (until 1991), many banks were insolvent and being restructured and bank supervision needed strengthening. It appears that bank portfolios did not suffer, but money market development was stifled by onerous government borrowing requirements and volatile and negative real rates of interest. Furthermore, overall credit controls and interest rate caps have been reintroduced through moral suasion by BOG. Also, lending to priority sectors at preferential interest rates continues even though no formal requirement has been specified. These partial reversals of macrofinancial policy reforms have been facilitated by the continuing Government ownership of the banks. Finally, the macroeconomic stability requirement was not achieved. FINSAC II's contribution to financial policy development included tax policy reforms and bank reserves management which were successfully implemented. The macroeconomic stability condition has, however, been ignored. Contribution to Financial Infrastructure 22. Prudential regulations and supervision capability have improved markedly under FINSAC I and II, and substantial contributions have been made to developing the legal framework for banking, non-bank financial institutions, accounting and audit standards and banking education. Updating commercial laws and procedures, important for collateral, foreclosure and liquidation, were omitted. Nevertheless, the FINSACs' contribution to this component was a salient accomplishment, particularly in the design and implementation of a complex and large technical assistance project. Bank Restructuring and Divestiture 23. Financial restructuring of the banks was successful and NPART has substantially met its targets for collections from the banks' non-performing assets. The restructured banks now meet capital adequacy and other prudential norms and have achieved financial performance ratios equivalent to that of the main private foreign banks. The performance improvement is a significant achievement. The organizational and operational restructuring of two public banks has been less successful but satisfactory. They continue to have weak management and inadequately trained and excessive staff. 24. Privatization of two banks through a public offering has been successful and a third has been readied for a public offering. Divestiture to strategic investors has not been successful. Sale memoranda have been prepared and merchant banks have been appointed to search for strategic investors, but no serious offer has come forward. With hindsight, sequencing the divestiture differently may have resulted in a more successful outcome. Finding a strategic buyer first would have made banks more attractive in a public offering. On the other hand, given 15 the difficulty in finding a strategic investor, it may have been appropriate to proceed with the public offerings. Overall, the effort made to divest the banks has been significant. Capital Market Development 25. The technical assistance provided for developing regulations and supervision for NBFIs was successfully implemented. The support for capital market development was also successfully completed. Rating of Project Outcome 26. The project designs were appropriate and the implementation of project components was satisfactory. Financial infrastructure has been developed and distressed banks were restructured. The stated objectives of financial institution strengthening have been achieved. The underlined objective of financial deepening, however, was not. Therefore, institutional development impact is rated as substantial, and the overall outcome for both projects is rated as marginally satisfactory. Sustainability and Lessons Learned Sustainability 27. The main successes of the two FINSACs until now have been in developing financial infrastructure and restructuring the banks. The sustainability of these achievements depends on the future improvements in financial deepening, intermediation and policy. In turn, success in these objectives hinges critically on macroeconomic stabilization-lower inflation, positive real interest rates and lower government borrowing from banks. Thus, sustainability depends on improved macroeconomic management and is rated as uncertain. 28. Financial sector reform projects often face an internal conflict. Some elements (e.g. financial policy reforms) can be implemented in a relatively short period while others (e.g. developing financial infrastructure and restructuring banks) take a longer time. In countries where macroeconomic stability is uncertain, it may be better to proceed first with a large technical assistance credit to upgrade financial infrastructure and restructure banks and follow it with a sector loan to improve financial policies when macroeconomic stability is achieved. Lessons of Experience * Financial policy reforms can only achieve financial deepening objectives in periods of macroeconomic stability. Therefore, macroeconomic performance conditions for tranche release should be clearly specified and strictly followed. * As financial infrastructure and institutions take a long time to restructure and develop, financial sector institution building projects can precede financial sector policy loans, particularly in periods of macroeconomic imbalance. 16 * The phasing and sequencing of financial policy reforms should be designed to minimize undesirable outcomes such as a recurrence of negative real interest rates or bank insolvency. * Financial policy reforms such as eliminating credit ceilings, credit allocation, interest rate controls and subsidized interest rates can often be informally reversed, especially where the government owns a substantial portion of the banking system. * Complex projects with many components are difficult to design and implement, especially in countries short of appropriately skilled personnel. A series of smaller, focused projects is probably more appropriate. * Too many conditions for loan tranche release delay the release of funds. A few, clearly specified conditions would be more appropriate. * Tranche conditions that hinge on uncertain transactions over which the Borrower has no control are inappropriate. 17 1. Background The Economic Reform Program Background to the Economic Recovery Program 1.1 Ghana once enjoyed a relatively high standard of living compared to other West African countries, but poor economic management during the 1970s and early 1980s led to protracted economic decline. Expansionary fiscal and monetary policies, high inflation and an overvalued exchange rate caused a substantial real appreciation of the currency, leading to external payments imbalances. Policymakers imposed a range of administrative controls on prices, imports, foreign exchange use and the distribution of goods and services. These policies contributed to a downward economic spiral between 1970 and 1982. 1.2 In 1983 the Government adopted an Economic Recovery Program (ERP) and has since devalued the currency, dismantled most price and distribution controls, eliminated many subsidies, broadened the tax base, improved tax collection, and provided more adequately for maintenance and capital expenditure. Under the ERP, GDP growth has averaged about 5% per year, compared to losses of 2-3 percent per year during the preceding decade. On a per capita basis, real income grew by 2 percent per year. Most farmers and rural workers gained from improved producer prices for cocoa and from liberalization in other cash crops trade. In the initial years of the ERP, government spending on social projects also rose significantly, leading to improvements in basic social services and social indicators. 1.3 Reforms initially focused on stabilizing the economy. Fiscal deficits were reduced through improved revenue collection and monetary growth was restrained. The government lifted price controls, liberalized both interest rates and the exchange rate system and rehabilitated infrastructure. As discussed below, however, it has been difficult to achieve lasting macroeconomic balance, particularly because of difficulties in managing the fiscal situation. 1.4 Before the ERP, the Ghanaian financial system was highly controlled. Interest rates were set by the Bank of Ghana (BOG); banks were subject to sectoral lending guidelines; and the bulk of credit went to the government or state-owned enterprises. Banks were required to hold government instruments that paid little or no interest and real interest rates were negative. As a result of the economic crisis of the late 1970s and early 1980s, the mobilization of financial resources declined precipitously. 1.5 Even by Sub-Saharan African standards, Ghana's recorded savings and investment rates were--and remain-low. Evidence from household surveys suggests that actual savings are higher than records show, but most are kept in nonfinancial forms such as inventories or building materials. In general, individuals are reluctant to save with formal financial institutions, for reasons that will be outlined in this report. 18 Economic Performance During FINSAC I and II 1.6 While real GDP growth ranged between 3.3 and 5.6% during 1988-94, inflation and fiscal restraint remained important concerns in the overall macroeconomic management of the economy during the implementation of FINSAC I and II. The overall fiscal deficit, which including external assistance had averaged 3.5% of GDP during 1982-84, had been turned into a surplus of 1.5% of GDP in 1991 (Table 1.1). Inflation fell from 40 percent in 1987 to 32% in 1988, 25% in 1989, increasing again to 37% in 1990. To offset the rise in inflation in 1990, the Government undertook measures to restrict monetary growth, including the sale of BOG bills to mop up excess liquidity in the economy and a sharp increase in the discount rate. By late 1991, the inflation rate had dropped to 18%. Private investment had risen continuously to a ten-year high of 8.1% of GDP. Money supply growth remained high, but was falling. Table 1.1: Ghana - Macroeconomic Indicators, 1988-94 1988 1989 1990 1991 1992 1993 1994* Real GDP Growth(%) 5.6 5.1 3.3 5.3 3.9 5.0 3.8 Budget Deficit/GDP(%) -5.3 -5.3 -4.8 -4.0 -10.7 -10.8 -6.8 Narrow Budget Deficit/GDP (%) NA NA -4.4 1.5 -4.8 -2.5 2.2 CABOP/GDP (%) -5.1 -6.0 -7.0 -6.6 -8.6 -13.4 -8.6 REER Index 1988= 100 100 106 100 98 109 136 165 CPI(%) 32 25 37 18 10 25 25 Domestic Savings/GDP (%) 5.4 5.6 6.0 7.8 2.0 -1.0 4.4 Public Savings/GDP (%) NA NA NA 3.2 -2.2 -0.9 4.7 Private Savings /GDP (%) NA NA NA 6.1 6.4 2.3 2.6 Public Investment/GDP (%) 8.0 7.8 6.8 7.7 8.5 9.9 11.5 Private Investment/GDP (%) 3.3 5.5 7.6 8.1 4.3 4.9 4.4 * = estimates Source: Ghana: Growth, Private Sector and Poverty, World Bank, May 15, 1995. 1.7 Instead of building on these gains, however, the Government reversed its fiscal stance in 1992 and undermined macroeconomic stability. An across-the-board wage increase of 80% for civil servants, a decline in tax revenues and a shortfall in foreign project grants resulted in a fiscal deficit of 4.8% of GDP in 1992. In 1993, inflation rose to 25% and remained at about the same level in 1994. The Government then took measures to improve the fiscal situation. By 1994, divestiture receipts helped to generate a fiscal surplus of 2.1% of GDP. In 1995, however, inflation accelerated again to an annual rate of about 60%. 1.8 The budget deficit of 1992 was financed mainly through increased borrowings from the banking sector. During most of the earlier ERP years the Government had made net repayments to the banking system. But, in 1992 Government had net borrowings from the banking system equivalent to 3.4% of GDP. As a result, excessive liquidity was introduced into the economy, fueling inflation. Money supply (M2) increased by more than 50% that year. Although the growth in money supply has fallen since, it averaged about 41% per year in 1992-94 against 23% in 1989-91. 19 1.9 Investment had increased substantially during the ERP, moving from 4.7% of GDP in 1982-84 to 14.5% in 1989-91. After the fiscal shock of 1992, however, total investment fell to 12.8% of GDP. Private investment also fell from 8.1% of GDP in 1991 to 4.3% in 1992. The fall in private investment arose because of political uncertainty and macroeconomic instability and has not yet recovered. In fact, public investment had increased its share of total investment from 49% in 1991 to 72% in 1994. Ghana has paid a high cost for this fall in private investment. If private investment had remained at 1991 levels during 1992-94 (and assuming unchanged public investment), gross investment over 1992-94 would have been almost $950 million more. 1.10 Savings also maintained an upward trend up to 1991, doubling from a low of 4.0% of GDP in 1982-84 to 8.0% in 1989-91. During the same period, private savings increased from 4.6% of GDP to 6.1% of GDP. The fiscal shock affected total savings, which averaged only 3.8% of GDP in 1992-94. While the majority of that decline arose because of the fall in public savings, private savings also fell during the same period, from 6.1% of GDP in 1991 to 2.4% in 1994. 1.11 The current account deficit in the balance of payments widened from about 5% of GDP in 1987 to about 7% in 1990, reflecting a significant deterioration in the terms of trade. Nevertheless, because net aid disbursements rose and non-concessional debt amortization obligations declined, a sizable overall balance of payments surplus was maintained between 1987 and 1990. After peaking in 1988 at 67% of exports of goods and services, the debt service ratio, including IMF and arrears payments, dropped to about 39% in 1990. 1.12 As with other components of economic performance, Ghana's balance of payments and external debt situation deteriorated after the 1992 fiscal shock. For example, the overall balance of payments surplus of US$ 171 million in 1991 turned into a deficit of US$ 124 million in 1992. The current account deficit increased because of the increased budget deficits. The capital account worsened in that year because of delays in the disbursement of long-term loans from the donor community (in response to worsening macroeconomic management) as well as substantial private short-term capital outflows. For example, a net inflow of short-term capital in 1991 became a net outflow of US$ 119 million. 1.13 Ghana's external debt situation has worsened and its dependence on external financing has increased. Net foreign borrowings, both public and private, have increased from US$ 291 million in 1989-91 to US$ 458 million in 1992-94. Almost half of the increase was due to higher private borrowings. The external debt to GDP ratio also increased from 61% in 1989-91 to 78% in 1992-94. The debt service ratio, however, has improved from 38% in 1989-91 to 29% in 1992-94, largely because of lower interest rates and smaller principal payments. The exchange rate adjustments have kept pace with relative inflation between 1988 and 1992, and exceeded it in 1993-94. Current Challenges 1.14 During 1993-94, the Government implemented a number of measures to promote private sector growth and reduce the role of government in the economy. It accelerated the privatization of state-owned enterprises, including the Ashanti Goldfields Corporation, and liberalized the regulatory framework for private investment. Nonetheless, macroeconomic instability has reduced the impact of these efforts to promote private investment. The fiscal 20 shock of 1992 highlighted the underlying fragility of the economy, including remaining structural issues (such as the oversized public service). 1.15 At the beginning of this decade, the Government had made significant strides in improving the prospects for macroeconomic stability and a more rapid pace of growth. The fiscal shock of 1992, however, severely undermined macroeconomic stability, reversed many of the macroeconomic gains and seriously dampened the environment for private sector investment. Ghana's ability to return to the growth and private investment rates of 1991 will depend on achieving macroeconomic stability, which remains precarious. Among other steps, it seems clear that the Government needs to maintain a surplus in the narrow fiscal accounts (largely to be achieved through expenditure reductions) and adopt sound monetary policies for controlling inflation and preventing short-term capital outflows. The financial sector reform projects, which are assessed in this report, were primarily designed to help strengthen the institutional base for the financial system and to increase the quantity and quality of financial intermediation. Although these measures were important contributions to building a more solid foundation for the financial system, the macroeconomic and monetary imbalances have already erased some of these gains and may block further development of the sector. The Financial Sector Adjustment Project Structure of the Financial Sector 1.16 IDA's work on the financial sector in Ghana began in 1985. The financial sector at that time was in severe distress, posing a serious constraint to growth in the real sectors of the economy. A major financial sector review, carried out in 1986, revealed several deficiencies in the financial system. These included an insolvent banking system dominated by state-owned banks and saddled with large non-performing loans and excessive intermediation costs; a low level of financial intermediation, reflecting a poor record in domestic resource mobilization; inefficient credit allocation, due in part to central bank-directed interest rates and sectoral credit targets; the virtual absence of a money market and capital market; and weaknesses in the BOG supervisory system and regulatory framework. Institutions 1.17 At the time that FINSAC I was conceptualized, the formal financial system in Ghana was dominated by government-owned banks. Three commercial banks (Ghana Commercial Bank, Standard Chartered Bank of Ghana and the Barclays Bank of Ghana) accounted for about two-thirds of the banking system's deposits and assets. Ghana Commercial Bank alone controlled about 44 percent of the banking system's assets and deposits. There were also seven secondary banks (primarily sectoral development finance institutions and a few smaller commercial banks), a small cooperative bank and over 100 rural banks. All but one secondary commercial bank (Bank of Credit and Commerce) were partly or wholly owned by the Government. Thus, the banking system was concentrated and oligopolistic. The Bank of Ghana had oversight responsibility for the financial system and played a significant role in financing government expenditures. The money market was in an embryonic state in the late-1980s and the capital market was very rudimentary. 21 Table 1.2: Banking System Indicators, 1988-95 (in percent) 1988 1989 1990 1991 1992 1993 1994 1995 Size of Sector Number of 10 10 12 12 13 14 13 15 banks Banking system 21 22 19 19 23 33 35 25 assets/GDP Asset Concentration (% of total) largest bank 55 57 54 52 44 50 44 29 (GCB) of 4 largest 81 83 82 77 72 74 72 67 banks Source: Bank of Ghana. 1.18 In the non-bank financial sector, institutions included the Social Security and National Insurance Trust (SSNIT) and 11 insurance companies. SSNIT was a government-owned institution which collected social security payments and invested them almost exclusively in government paper and bank deposits. The deficiencies of the banking system and the negative real interest rates had spawned a myriad financial agencies including money lenders, thrift groups, credit unions, etc. The informal financial system is estimated to be quite sizable in Ghana. 1.19 Ghana Commercial Bank was wholly state-owned and the Government had acquired a 40% share in the other two primary commercial banks in 1975. All three primary commercial banks reported small profits and low non-performing assets. Accounting practices related to income recognition, asset classification and bad debt provisions were, however, varied and did not come close to international standards. It was felt that Ghana Commercial Bank was in serious difficulties and should be subjected to a thorough audit. The secondary banks, especially the development finance institutions, were thought to be in dire financial straits although their situation was not fully evident from their accounts, except for the National Investment Bank. About 52% of its portfolio was affected by arrears and principal and interest arrears amounted to 33% of outstanding loans, even when reschedulings were ignored. Furthermore, it had suffered a foreign exchange loss in 1983 of more than two times its net worth. All three development banks had experienced liquidity problems and had difficulty in meeting BOG's reserve requirements. While the rural bank scheme was well conceived and loan collections had not been a problem until 1982, a two-year drought period resulted in about 60% of loans defaulting by end-1984 and the national loan recovery rate declined to below 30%. Policy and Regulatory Framework 1.20 Deposit Mobilization. The formal financial system has not been an effective vehicle for mobilizing domestic resources. In 1993, the ratio of M2 to GDP remained very low at 17 percent. This level is significantly lower than those in other Sub-Saharan African countries. In addition, Ghana has remained a cash-based economy. Currency as a percentage of M2 was at least twice as high as in Kenya and Zimbabwe in 1992 and is at least three times as high as in Indonesia, Malaysia and Thailand. In most economies, individuals hold relatively large cash 22 balances when economic stability is in question. As discussed in the previous section, Ghana has experienced several bouts of high inflation and low levels of output. It appears, however, that savers have avoided the formal financial sector for a variety of other reasons as well. 1.21 First, before 1987, the Government controlled both, deposit and lending rates. Because it maintained artificially low lending rates (to promote various sectors, such as agriculture), real deposit rates were often negative in real terms. As shown in Table 1.3, interest rates (using the 3-month T-bill rate as a benchmark) did not turn positive in real terms until 1991, even after the liberalization of all interest rates in 1988. Traditionally, banks have relied primarily on demand deposits (averaging over 50% of all commercial bank deposits); long-term savings have not entered the banking system in significant quantities. Moreover, despite relatively high and variables rates of inflation, the banking system has not introduced financial instruments offering rates indexed to inflation. Table 1.3: Inflation, Real Interest Rates and Nominal Spreads in Ghana, 1988-95 (in percent) Nominal Real T-bill Interest Rate Nominal Nominal Year CPI Rate Spread Deposit Rates Lending Rates 1988 32 -8.9 6.8 19.25-25.50 22.50-30.25 1989 25 -4.2 10.4 19.25-25.50 22-50-30.25 1990 37 -7.1 8.1 16.25-25.50 22-50-30.25 1991 18 7.1 5.5 21.00-32.00 22.50-36.00 1992 10 13.9 2.0 15.50-20.00 19.00-29.00 1993 25 5.4 5.3 22.00-24.50 25.50-31.50 1994 25 4.5 7.2 14.00-31.00 22.00-37.50 1995 60 -19.5 7.4 14.50-37.00 23.00-47.50 Notes to Table 1.3: 1. CPI Source: Ghana: Growth, Private Sector and Poverty, World Bank, May 15, 1995. CPI growth rate for 1996 is estimated at 34% (source: International Financial Statistics, April 1997). 2. T-bill rate Source: Ghana: Financial Sector Review: Bringing Savers and Investors Together, World Bank, December 29, 1994. 3. Real T-bill rates for 1994-95 and all other real rates calculated using same methodology as that in Table 4.1.3, page 38 of Ghana: Financial Sector Review: Bringing Savers and Investors Together, World Bank, December 29, 1994. 4. Nominal spreads are calculated using same methodology as that in Table 4.1.3, page 38 of Ghana: Financial Sector Review: Bringing Savers and Investors Together, World Bank, December 29, 1994. They are essentially the spread between the posted lending rate for manufacturing and the posted upper range for 3-month deposits. 5. Nominal deposit rates Source: FINSAC PCR for 1988-1993 (Sept); and Bank of Ghana, Quarterly Economic Bulletin, Oct.-Dec. 1995 for 1994-1995. Lower rates are for 3 month deposits and high rates are for 1 yr. deposits. 6. Nominal lending rate Source: FINSAC PCR for 1988-1993 (Sept); and Bank of Ghana, Quarterly Economic Bulletin, Oct.-Dec. 1995 for 1994-1995. Lower rates are for agriculture and exports and higher rates are for manufacturing and "other", which includes trade. 1.22 In addition, until recently, high cash reserve requirements on banks reduced their incentive to mobilize long-term savings at high interest rates. The cash requirement ratio has varied over the past 15 years, ranging between a high of 35% in 1982 and a low of 5% in 1993. Prior to FINSAC I and II, cash reserves with the BOG were not remunerated. Beginning in 1991, BOG began to pay market rates on reserves in excess of the minimum cash requirement and relatively attractive rates on secondary reserves. At the same time, while the cash requirement has fallen, BOG increased the secondary reserve requirement from 10 percent in 23 1987 to 52 percent in 1993. Moreover, because of the attractive rates offered on BOG securities after 1991, banks continue to hold a significant excess over the quantity required. In late 1993, for example, over 60% of total deposits were held in secondary reserve instruments. While this policy has improved the profitability of the banks, it has reduced the incentive to lend to the private sector. 1.23 In addition to the disincentives presented by negative real interest rates and cash requirements, the Government had taken several other measures in the 1980s which shook savers' confidence in the formal banking system. These measures included the demonetization of 50 cedi notes, the freezing of bank deposit accounts, the recall of bank loans for financing trading inventories and the compulsory payment by checks for all business transactions in excess of 1,000 cedis. Businesses and individuals responded to these measures by rechanneling their financial resources into the informal financial sector. Finally, high transaction costs and poor quality service also serve as a disincentive to savers, particularly in rural areas. 1.24 Credit Allocation. One of the most striking aspects of the Ghanaian financial system throughout the 1980s and 1990s is the large proportion of financial flows to the Government. Because of its excessive financing needs, only a small proportion of total credit goes to the private sector, ranging between 7.3 and 3.1 percent of GDP in the 1986-95 period. In contrast, credit to the private sector as a percentage of GDP in Kenya has hovered at about 20 percent and is as high as 70 percent in Malaysia. In addition, the Government has consistently maintained a negative balance between savings and investment between 1986 and 1992. The private sector has also held largely negative balances between savings and investment, but of a lesser magnitude than the Government. Steady increases in foreign savings since 1985 have allowed both the public and private sector to hold negative financial balances, but the economy's reliance on foreign savings may make it more vulnerable to external shocks and world economic conditions. Low levels of credit to the private sector may also impede growth. 1.25 Second, in the 1980s, the BOG used global and sectoral credit ceilings to control the growth and allocation of credit. Sectoral lending targets, other than for agriculture, were removed in 1987. Individual credit ceilings for banks, however, were not lifted until 1991, thus impeding the banks' abilities to make new loans. In addition, increasing capital adequacy and other prudential risk exposure limits also restricted banks' capacity for new lending. Moreover, in the 1990s, risk-free interest from BOG bills and good rates from short-term placements with discount houses reduced the relative attractiveness of loans. Lastly, the legal and information infrastructure in Ghana makes it hard to expect adequate security for loans and to enforce contracts. 1.26 Finally, lending rates in Ghana were also controlled by BOG until 1988, when they were liberalized. Until recently, however, commercial bank rates have remained relatively rigid in responding to changes in government rates and the premium charged for "riskier" projects has at times declined substantially. In an environment with relatively high and variable inflation, it is possible that banks were unwilling to raise lending rates above what projects could reasonably pay. Moreover, banks may have decided in this environment to focus their efforts on their long- standing customers, who are also in a better position to negotiate more favorable lending terms. Discussions with bankers indicate that they did little new lending during this period, preferring to place their funds in treasury bills and rolling over credit to existing clients. 24 1.27 Liquidity Management. As noted previously, the BOG uses two kinds of minimum reserve ratios to control the expansion of credit: a minimum cash reserve ratio and a minimum reserve liquidity ratio or secondary requirement. For many years, the minimum reserve requirements were not effective in controlling liquidity because bank financing of the government deficit was far in excess of the reserve requirements. Consequently, real M2 growth often outstripped real GDP growth, further fueling inflation. Moreover, the maintenance of relatively high liquidity and reserve requirements reduced the incentives for banks to innovate or expand their lending, particularly when they also faced individual credit ceilings. At the time of FINSAC I, the Government had not yet begun to auction government bills to manage liquidity. 1.28 Money Market. The development of money markets is a relatively new phenomenon in Ghana. The first discount house, Consolidated Discount House (CDH), which is owned by banks and insurance companies, opened for business in 1987. After an initial flurry of activity in its first few days, however, few transactions were subsequently handled by CDH, because of an inadequately detailed operational framework, unrealistic expectations by the banks as to the absorptive capacity of CDH and imprecise definitions of the scope of CDH activities. Thus, at the time of FINSAC I, the money market in Ghana was largely quiescent. 1.29 Capital Market. At the time of FINSAC I, Ghana did not yet have a formal stock exchange. The Accra Securities Market LTD had been formed for some years but had not been activated. Nonetheless, 18 companies, owned primarily by foreign shareholders, were converted into public companies in 1975-76 under the Investment Policy Implementation Decree. Only two other issues had occurred between 1976 and 1987: one involving the capitalization of reserves and a second failed rights issue. 1.30 Regulatory Policies. Prior to FINSAC I, several serious omissions in the existing body of legislation, together with weak bank supervision, contributed to the concentration of risk in banks' portfolios, inadequate capital and reserves, inflated profits and unrecognized loan losses. These shortcomings included: * Unspecified limits, (by BOG), for unsecured credit as the total of each bank's paid- up capital and reserves, thereby allowing banks to extend credit to single borrowers far in excess of the bank's capital funds; * Varying treatment of interest on non-performing assets in the absence of uniform accounting standards. Moreover, many banks took inadequate loan loss provisions and reserves; * The lack of appropriate regulation and accounting standards also led banks to have insufficient capital relative to their assets. It was estimated at the time of appraisal of FINSAC I that many banks would be technically insolvent had they been required to recognize the loan losses in their portfolios. Moreover, the existing legislation merely specified a capital adequacy ratio based on the deposit base (5%) of banks thereby ignoring the risk components/contents of assets; * The neglect in treating development banks engaged in commercial banking on the same basis regarding the treatment of retained profits; * Penalties which were too low to act as meaningful deterrents to imprudent behavior; 25 * Audit and regulatory reports to BOG which were insufficient information to conduct off-site supervision or detect early signs of inadequate liquidity or solvency; and * Insufficient numbers of personnel and inadequately trained banking supervision staff. Consequently, no bank in Ghana had been extensively reviewed by BOG.  27 2. Bank Strategy and Project Objectives, Design and Relevance The Bank's Assistance Strategy 2.1 The main elements of the Bank's assistance strategy to Ghana, at the time of the FINSAC I appraisal were: * to assist the Government, through the Bank's economic and sector work, technical assistance and project lending, to improve incentives for production, increase the efficiency of economic management and restore a sound financial basis for growth; * to promote the long-term growth and development of the economy by underpinning structural adjustment lending with infrastructure rehabilitation and sector adjustment operations within a framework of appropriate sectoral policies, the latter encompassing industry and education; and * to contribute to improved aid effectiveness in Ghana by acting as the focal point for aid coordination between donors and Ghana as the Government strengthened its own planning and aid coordination ability. 2.2 During this period, the Bank has supported changes in incentive policies and improvements in economic management through a series of project credits designed to provide critical imports, particularly to export sectors, and supporting economic infrastructure such as transportation. Supporting financial sector restructuring and reform was seen as integral to this overall strategy. It was envisaged that the original action project under FINSAC I would help address some of the sector's shortcomings, such as its institutional weaknesses, limited mobilization of financial resources and deficiencies in credit allocation. It was felt that the project would help increase the efficiency of the financial sector and thus contribute to growth by increasing savings and channeling them to higher yielding investments. Restructuring banks was also seen as a step toward reinstituting public confidence in the formal financial system. Strengthening the financial system's information infrastructure (accounting and audit, bank supervision, etc.) as well as initiating the development of money and capital markets were also seen as steps towards enhancing the sector's efficiency and paving the way for a diversification of markets and instruments. 2.3 At the time of FINSAC II, the Bank's assistance strategy was to help the Government achieve its goals for economic growth, macroeconomic stability and poverty alleviation. The first element in its strategy was an emphasis on sustained growth with an expanded role for the private sector while also achieving macroeconomic stability. The Bank was supporting entrepreneurial development and improvements in regulations through an adjustment credit- Project to Promote Private Investment and Sustained Development. To complement that loan, FINSAC II was prepared to address outstanding institutional and policy reform needed to increase the efficiency of the financial system, building on FINSAC I. 28 FINSAC I Objectives of FINSAC I 2.4 The FINSAC reform project was intended to complement the structural adjustment project initiated by the Government in 1983. The financial sector adjustment project supported by FINSAC I was designed to deal with the more fundamental and urgent problems confronting the Ghanaian financial sector in 1988. The credit was seen as one of a series of operations supporting the reform effort. The main objectives of the Credit were to: * liberalize the environment for banking operations through the removal of interest rate controls and sectoral credit targets; * enhance the soundness of banking institutions by putting in place a new legal, regulatory and prudential framework; * strengthen the banking supervision function of BOG; * restructure financially distressed banking institutions following the adoption of a suitable framework and the formulation of institution-specific restructuring plans; and * attempt to recover, to the extent possible, the non-performing assets from distressed banks. 2.5 In addition, other objectives supported by the Credit included: * initiating the development of the money and capital market; * initiating a study on the need and modalities for corporate restructuring; * strengthening the accounting and auditing profession; and * training bank managers and staff. Project Design of FINSAC I 2.6 The original IDA credit of US$ 100 million was made to the Government at standard IDA terms with 40 years maturity. The credit consisted of 2 components: (a) an institutional and policy reform component; and (b) a technical assistance component. 2.7 Sector Reform Component. This component of US$ 95 million equivalent was designed to support the Government's financial sector policy and institutional reforms, as laid out in its Statement of Financial Policy (see Annex I for policy matrix and time table). It was to be disbursed in three tranches: US$ 45 million in June, 1988, US$ 30 million in February, 1989 and US$ 20 million in December, 1989. The three major tasks supported by this component included: 29 * enhancing the soundness of banking institutions through (1) a review of the legal framework; (2) improved banking supervision; (3) training bankers; (4) strengthening the accounting and auditing professions; and (v) supporting the financial restructuring of the corporate sector; * restructuring financially distressed banks on the basis of full external audits, followed by a thorough review of each bank's prospects for future viability within a liberalized and competitive system, including reduced reliance on GOG funding and other financial privileges. The decisions regarding each bank's future (recapitalization, restructuring, merger, liquidation) was to be based on these reviews and would lead to the preparation of a specific restructuring plan. The implementation of the individual restructuring plans were to be monitored through a performance contract; and * improving resource mobilization and allocation through (1) measures aimed at restoring public confidence in banks; (2) liberalization of interest rates, bank charges and commissions and sectoral credit ceilings; (3) development of money and capital markets through rationalization and strengthening of the CDH and the introduction of policies and institutional vehicles for capital market development. 2.8 Technical Assistance Component. This US$ 5 million equivalent component was designed to finance (1) a comprehensive project of technical assistance to the BOG ; (2) training projects for the Institute of Chartered Accountants; (3) training projects aimed at improving the banking profession; (4) technical assistance to CDH; (5) an assessment of the potential for capital market development and related studies; (6) a study to assess the need for corporate restructuring and the initiation of a pilot project; and (7) a study on exchange risk protection. FINSAC H Objectives of FINSAC II 2.9 In late 1990, after the release of the third tranche of FINSAC I, the Government had made considerable progress in placing financial institutions on a more sound footing. It recognized, however, that several important tasks had not yet been completed. Monetary policy management, for example, was very rudimentary, and needed to be strengthened to help manage the variable and relatively high inflation. The Government remained the primary shareholder in the banking sector, reducing the scope for more even competition between banks and contributing to the high intermediation costs in the system. BOG itself was also in a precarious financial situation, owing to large foreign exchange losses, its quasi-fiscal activities, poor operational capacity and insufficient autonomy from the Government in its management of the money supply. 2.10 Under FINSAC II, the Government intended to consolidate the achievements of FINSAC I and move toward further deepening, diversification and strengthening of the financial sector. The objectives of FINSAC II included: * supporting policy measures designed to strengthen the institutional framework in the financial sector and improve the financial condition of BOG; 30 * increasing competition in the banking system by encouraging new entrants and divesting the Government's shareholdings in banks; * reducing intermediation costs by aligning taxation on bank profits with levels applicable to other sectors; * improving the operational capacity and institutional efficiency of BOG by reorganizing its structure and strengthening its procedures and capabilities; * supporting the ongoing bank restructuring project to improve the financial and operational performance of formerly distressed banks; * helping pursue the recovery of non-performing assets taken over by NPART; * designing the arrangements for the rehabilitation of potentially viable enterprises in the corporate sector; * formulating a healthy and stable institutional framework and enhance the effectiveness of non-bank financial institutions; * studying the informal financial markets to foster linkages with the formal financial sector; and * upgrading the professional skills of bankers, accountants and insurance professionals. Project Design of FINSA C I 2.11 As with FINSAC I, the second IDA Credit of US$ 100 million equivalent was made to the Government at standard IDA terms with a 40 year maturity. It consisted of 2 components: a sector reform component (US$ 82 million equivalent) and a technical assistance component (US$ 18 million). 2.12 Sector Reform Component. The major objectives of the sector reform component, as detailed in the Policy Matrix (Annex II) are as follows: * support the removal of remaining policy distortions in the financial system by: (1) addressing the disintermediation in the financial system caused by negative real interest rates through a policy of encouraging positive real deposit and lending rates (this included the payment of adequate interest on required commercial bank deposits with BOG and promoting open market operations); (2) increasing competition in the banking system through the divestiture of public sector shareholdings in banks; (3) reducing the high cost of financial intermediation; and (4) lowering the rate of corporate tax on financial institutions to the level applicable to other sectors. * enhance the effectiveness of the Central Bank to lead the continuing financial sector reform project by: (1) developing a sound legal framework for the financial system including legislation for central banking; (2) strengthening the financial condition of 31 BOG and phasing out its non-central banking, quasi-fiscal activities; (3) realigning the organizational structure of BOG to enhance its capacity to formulate and implement monetary policies and financial sector reforms; and (4) improving operational efficiency in the payments system, open market operations, reserves management, bank supervision, research and other central bank activities through computerization and changes in the operating procedures and staff training; * continue the bank restructuring project started under FINSAC I through: (1) preparation for the divestiture of public sector shareholding in banks; (2) formulation of an appropriate project for monitoring the performance of restructured banks; (3) improving the efficiency of management, accounting and information systems in GCB; * pursue the recovery of non-performing assets by NPART, including setting monitorable targets and a timetable; * enhance the effectiveness of non-bank financial institutions by strengthening the prudential regulation function and encouraging competition. Steps would include: (1) reviewing the legal and regulatory framework applicable to all non-bank financial institutions and develop new legislative proposals for their regulation and supervision; (2) development of uniform accounting requirements and prudential regulations; and (3) a review of the linkages and prospects for closer interaction between the formal and informal financial systems; * upgrade the professional skills of bankers, accountants and the insurance profession through training projects, improving the effectiveness of the national banking college, strengthening national accounting and auditing capabilities and upgrading the standards and skills of the Auditor General' Office and of the accounting and insurance professions. 2.13 The proceeds of the sector reform component were to be disbursed in three tranches as follows: * a first tranche of US$ 25 million equivalent would become available immediately upon effectiveness, projected to be January, 1992; * a second tranche of US$ 29 million equivalent would become available following a review of performance about nine months after effectiveness (October, 1992); and * a third tranche of US$ 28 million equivalent would be made available following a review of performance, about six months after second tranche release (April, 1993). 2.14 Technical Assistance Component. This component was designed to finance: * an extensive technical assistance project for BOG, including computerization, training, technical assistance and advisory services in economics research, banking supervision, monetary policy and foreign exchange (including reserve management), organizational restructuring, clearing house, development of market operations, financial system development and data coordination; 32 * computer hardware, software, service and training for the computerization of management information systems in GCB; * continued technical assistance for the institutional restructuring of 7 banks; * review of the legal and regulatory framework for non-bank financial institutions and the development of new legislation, diagnostic study of the insurance industry, training and technical assistance for the new National Insurance Commission, the study of SSNIT's new pension scheme and its role in the capital markets and the study of informal financial markets; * capacity building and training for bankers and accountants, including technical assistance in establishing the banking college, teaching materials and books, faculty development and the cost of foreign instructors and for the Auditor General's Office and ICA; * establishing the FINSAP Implementation Secretariat; * equipment and training for a small FINSAC II liaison unit within the MFEP ; and * a small unallocated amount (US$ 300,000) to be used as needs arise during project implementation. Relevance of Strategy, Project Objectives and Design 2.15 The links between FINSAC I and II, and the bank's country assistance strategy can be gleaned from the project objectives which implicitly assume the Government will maintain macroeconomic stability. Indeed, two Structural Adjustment Credits and an IMF ESAF project were under implementation during the implementation period. The first loan/credit's objectives were to improve the policy environment for banking and put financial infrastructure in place- prudential regulations, bank supervision, accounting and legal issues. Simultaneously, distressed banks commence the restructuring process and the development of money and capital markets was initiated. Bank and IMF staff emphasize that the primary objective of FINSAC I was modest-to restructure distressed banks rather than increase intermediation or promote financial deepening. 2.16 The second loan/credit's objectives focus on strengthening the central bank's organization and financial condition and technical capacity and continue to improve banking policy. In addition, bank restructuring was expected to be completed and the divestiture of government ownership begun and entry into banking permitted. At the same time, a regulatory and supervision framework for non-bank financial institutions was to be developed. Bank and IMF staff suggest that FINSAC II was expected to result in financial deepening as banks would have been restructured and financial policy reforms implemented. 2.17 Generally, the strategy for financial sector development and project design was appropriate, with one caveat. It was risky to implement interest rate liberalization while most banks were insolvent, bank supervision had not yet been upgraded and credit ceilings were in place. Moreover, initiating indirect monetary policy instruments should have waited until the central bank had been strengthened. Exposing banks and the central bank to additional risks at a 33 time when both were in financial distress and lacked the skills to manage the risk was inappropriate.4 Bank regional staff notes: "At that time (early 1990s) the insolvent banks were under restructuring. Actions were clearly designed and sufficiently strong to avoid any such risks. Furthermore, the restructuring bonds (FINSAC Bonds) issued to the banks to clean up the non-performing loan portfolio provided for some of these bonds to be discounted at the Bank of Ghana, if liquidity became a critical problem. This arrangement, thus, addressed both the solvency and short-term liquidity needs of the affected banks, while also avoiding the risks of high-cost borrowing". On the use of indirect monetary instruments, regional Bank staff remarks: "The central bank had the latitude to engage in open market operations to then extent commensurate with its operational capacity, which was assessed by the IMF at the time they recommended the use of such instrument..."  35 3. Project Implementation FINSAC I Credit Processing 3.1 The preparation of FINSAC I stemmed directly from IDA's financial sector work in Ghana. After a preliminary mission in 1985, which identified the shortcomings in the financial sector, a major review of the sector was undertaken in 1986 by a joint working group of Ghanaians and IDA staff. This team formulated an action project which was subsequently discussed and agreed with the government as the policy matrix for FINSAC I (Annex I). The Credit was formally negotiated in April, 1988, approved by the Board on May 31, 1988 and declared effective in August, 1988. The original amount of SDR 72.1 million (US$ 100 million) was subsequently increased to SDR 77.2 million (US$ 106.6 million) through a supplementary IDA credit. Cofinancing totaling US$ 105.7 million was mobilized from Japanese and Swiss donors to support the reform program and the technical assistance. The second tranche (US$ 30 million equivalent) was released in June, 1990, about 17 months after effectiveness, compared to the 9 month interval anticipated at the time of appraisal. This significant delay occurred because IDA staff and the Government underestimated the time necessary and complexity inherent in completing the requirements for second tranche release (10 conditions). In contrast, the third tranche was released in December, 1990, or 6 months after the second tranche release. There were 5 conditions attached to the third tranche's release. Macroeconomic Management 3.2 FINSAC I's President's Report makes no reference to macroeconomic management under the section on monitorable actions. Thus, tranche releases were not tied to macroeconomic targets. Inflation was a high 32% when FINSAC I was made effective and the first tranche released in 1988. Although inflation came down to 25% in 1989, it had picked up to 37% in 1990 when the second and third tranches were released. Other macroeconomic indicators showed no significant change during this period (see Table 1.1). Liberalization of Financial and Banking Environment 3.3 The process of liberalization and deregulation of the financial system had begun prior to the inception of FINSAC I and was substantially completed during it. Control of interest rates was gradually removed and full interest rate liberalization was achieved in February 1988. Because of fluctuating inflation and remaining rigidities in the ways in which banks priced loans and deposits, however, real interest rates continued to fluctuate between positive and negative. BOG also began to decontrol all banking charges (fees and commissions) in October 1990. All sectoral credit targets were gradually phased out beginning in 1987, except for agriculture, for which the lending floor was abolished in November 1990 (third tranche release condition). Credit ceilings for individual banks remained in place until 1991, however, which constrained the growth and credit as well as competition among banks. 36 Legal and Regulatory Framework 3.4 Preparation work for the project had revealed a number of deficiencies in the prevailing Banking Act of 1970, especially in the areas of minimum capital requirements, risk exposure and prudential lending limits for banks, provisions for possible loan losses and methods of interest accrual on non-performing loans. In 1988, the Government undertook a comprehensive examination of the Act and an interim Legislative Instrument was enacted on December 30, 1988. Subsequently, the authorities continued to develop a comprehensive legal and regulatory framework for banking activities through extensive consultations with bankers, accounting and other professional bodies in Ghana and overseas, as well as with IDA. This process led to the enactment of the Amended Banking Law on August 8, 1989, replacing both the Banking Act of 1970 and the Banking and Financial Institutions Decree of 1979. 3.5 The new Act and its regulations provide sound prudential and regulatory bases for the country's banking system. It contains an appropriate set of prudential banking provisions, patterned upon the Basle Guidelines, which require banks to maintain a minimum capital base equivalent to 6% of asset base adjusted for risk, and limits secondary capital. It sets penalties, including restrictions on dividend payments for banks that do not achieve their capital adequacy ratios. Guidelines are provided for determining provisions for bad and doubtful loans and for the treatment of related accrued interest. The Act also establishes for banks risk exposure limits to a single financial group or individual as a percentage of net worth, and sets limits on loans and advances to directors and employees of banks. It also contains restrictions on the extent of direct exposure in commercial, agricultural, industrial and real estate activities. The provisions of the new Act considerably strengthened BOG's ability to manage and oversee the banking sector. Bank Supervision 3.6 FINSAC I also aimed to strengthen banking supervision by BOG. Under the new Banking Act, BOG was required to undertake on-site inspections of each bank at least once a year. Standard auditing and accounting principles were introduced in 1988, which follow International Auditing Guidelines and Standards. Since then, the accounting records of banks must be examined by qualified external auditors at least once a year and both statutory and long- form auditors' reports are required for all banks. The reporting format introduced by BOG in 1988 requires banks to provide considerably more data than had been submitted previously for off-site monitoring, including liquidity, volume and nature of credit granted, composition and maturity of all assets and liabilities, quality of portfolios, profitability and information regarding capital adequacy. BOG has begun to computerize this database and its operations to facilitate monitoring bank performance. 3.7 To strengthen the Banking Supervision Department in BOG, two high-level advisors from the Reserve Bank of India were retained to provide formal as well as on-the-job training to staff working in bank supervision. All the Department's bank inspectors have now participated in a basic training course and manuals for both on-site and off-site inspection have been substantially revised and expanded. Regular on-site inspections were initiated in early 1989 with the direct participation of the advisors and the quality of the resulting reports has improved considerably. The Department has also increased its staffing to accommodate the more intensive work load from 20 bank examiners in 1988 (of whom only 5 had suitable experience) to 63 (with adequate field experience) by 1993. 37 Restructuring the Distressed Banks 3.8 Under FINSAC I, the approach to bank restructuring was quite systematic. The Government's Statement of Financial Policies provided the basis of the framework for bank restructuring. The main guiding principles were that: (1) the restructuring aimed to foster a sound, dynamic banking system in which all banks would compete freely and operate by market rules; (2) pending the completion of bank restructuring, measures were to be taken to arrest the financial deterioration of banks; (3) financial restructuring would not be undertaken without managerial and organizational changes; (4) restructured banks were to have independent and professional management with sufficient autonomy to make day-to-day decisions, and the boards of directors were to include suitable representation from the private sector to broaden the experience base and enhance their capabilities and contribution; and (5) the Government would consider reducing its share bank ownership by seeking private Ghanaian and/or foreign shareholding in the restructured banks. In addition, subject to proper safeguards, entry into the banking system of new, private domestic or foreign banks would be encouraged. 3.9 The Government retained consultants to help develop a framework for restructuring 7 of the 9 major banks. These guidelines were formulated and elaborated in 1989 and subsequently reviewed and approved by the Government and IDA. The framework set out a one-time financial package of measures of each of the seven distressed banks to restore solvency and provide sufficient capital and liquidity for them to operate in a self-sustained manner following their restructuring. The measures were tailored to the needs of individual banks. All non- performing loans made to state-owned enterprises were removed from the banks' portfolios, totaling C 22.6 billion at the end of 1988. To the extent necessary, these loans were converted into equity, Government loans or other claims (deposits). The banks' non-performing loans to the private sector, amounting to about C 36.4 billion at the end of 1988, were to be replaced by the issuance of Government bonds. The non-performing assets of distressed banks were to be transferred to a newly created and wholly-owned Government agency, the Non-Performing Assets Recovery Trust (NPART), whose function would be to realize such assets to the extent possible. To facilitate NPART's work, a special judicial tribunal was appointed and given the necessary powers to speed up asset recoveries. 3.10 Implementation of the bank restructuring project has been generally satisfactory. Individual restructuring plans for six of the seven distressed banks have been implemented and have resulted in significant improvements in both the financial and operational performance of five of these banks (GCB, SSB, NIB, ADB, NSCB) since 1991. These restructuring plans included changes in the banks' management teams, staff reduction, closure of unprofitable branches and reduction of operating costs. For the seven restructured banks, the combined work force has been streamlined by 35% to about 7,000 by the end of 1992. Ratios of operating expenses to operating income have declined substantially and profitability has increased. More importantly, the combined loan portfolio of the restructured banks continues to show improvement, with the ratio of non-performing loans declining from 72.5% in 1989 to 28.5% at the end of September, 1993. While these are important improvements, it is disconcerting that the banks continue to do so little new lending. In addition, one of the seven banks (BHC), continues to need improvement, particularly in its management team. And, the small COOP Bank remained in precarious condition. 38 Recovery ofNon-Performing Assets 3.11 As noted previously, the Government, with IDA's agreement, decided to establish NPART to facilitate the recovery of non-performing assets and allow the banks to get on with other restructuring tasks. In February, 1990, the Government and IDA agreed on NPART's operating policies and guidelines. The law establishing NPART was enacted in February, 1990 and it became operational later in that year. NPART has received extensive technical assistance from IDA, including expatriate experts with previous experience with the Resolution Trust Corporation of the US. 3.12 The Government and IDA agreed in 1990 on an action project to achieve an appropriate level of debt recoveries by NPART. The action project set an aggregate recovery target of C 18 billion (compared with a total portfolio of C 50 billion), based on an account-by-account review of its portfolio. Annual recovery targets were then set: 12% in 1991, 22% in 1992, 26% in 1993, 23% in 1994 and the remaining 17% in 1995. NPART's actual collection performance under FINSAC I was satisfactory, totaling about C 9.0 billion or about 95% of the amounts projected in the first three years. Development of Money and Capital Markets 3.13 FINSAC I set out to strengthen CDH and complete a study of the potential for capital market development in Ghana. Following agreement with IDA in 1989 on an institutional strengthening project, CDH has continued to expand its business and operate profitably. A second discount house, the Securities Discount Company, opened in 1990. The capital market study, completed in October 1989, led to the establishment of the Ghana Stock Exchange in 1990. GSE has created a minimal organizational structure and the requisite basic operational capacity and procedures. The level of its operations to date, however, remains extremely modest. Only 17 companies are listed on the GSE, with only four new listings since 1992. This disappointing performance stems from the small number of market participants and the general lack of public confidence in financial markets. A much increased volume of transactions will be necessary for GSE to operate in a financially-self-supporting manner and ensure its long-term sustainability. Corporate Restructuring 3.14 The Government commissioned a study to assess the magnitude and nature of corporate distress and to recommend a project to facilitate the restructuring of potentially viable enterprises. The first phase of the study, completed in 1989, reviewed a sample of 214 firms to assess their potential for viable restructuring. It concluded that a majority of them had good prospects for viable operations if appropriately restructured from a financial, technical and managerial standpoint. The second phase of the study, completed in 1990, examined the alternative institutional arrangements for carrying out the enterprise restructuring process. This report recommended the establishment of an investment bank to carry out these restructurings. Neither the Government nor IDA accepted this proposal, as 2 new private merchant banks had opened in 1990 in addition to the previously existing Merchant Bank. 39 Strengthening the Accounting and Audit Professions 3.15 Measures to improve the accounting and auditing professions were included under the project, and a comprehensive review of the training and professional development needs of the accountancy profession was completed in 1989. The consultants' report recommended reforming the regulatory framework for accounting and auditing, developing professional standards, improving training and professional development of accountants and strengthening the previously existing the Institute of Chartered Accountants of Ghana (ICAG). ICAG made good progress in implementing these recommendations. In addition, it became somewhat more self- sufficient financially, reducing its dependence on outside financial support and increasing its own revenues from membership fees and dues. Training Bankers 3.16 The technical assistance component of FINSAC I included measures to assist both BOG and the Ghana Institute of Bankers. BOG recruited a senior training specialist (from the Reserve Bank of India) who, at the end of 1989, completed an assessment of training needs in the sector and formulated training projects and course curricula. External training experts also provided courses for both BOG and commercial bank staff on credit management, international banking, internal auditing and other core disciplines. In parallel, the Ghana Institute of Bankers upgraded its physical facilities and received funding for trainers' training. GIB has made good progress developing its own expertise (especially conducting examinations) and reduced its reliance on the London-based Chartered Institute of Bankers, UK. The Institute now has about 3500 students studying for various levels of certification. Nonetheless, the Institute needs to consider ways to increase its financial self-sufficiency, as it has relied heavily on external support from IDA and other donors. 3.17 The technical assistance component of FINSAC I also provided financial support to the Ghana Association of Bankers (GAB) toward the establishment of a Credit Clearing House (subsequently renamed the Interbank Credit Data Bank). This step was envisaged to develop a centralized and computerized data system on credit granted by all Ghanaian banks to all Borrowers. Implementation of this proposal was delayed, however, because of the GAB's ineffectiveness. FINSAC H Credit Processing 3.18 The preparation of FINSAC II was based the work completed under FINSAC I and a series of technical papers (not found in the files). The action project developed in cooperation with the Government is provided in Annex II. The credit of SDR 74 million (US$100 million) was negotiated in November, 1991 and approved by the Board in the same month. The credit was declared effective in May, 1992 and the first tranche (US$ 25 million) was released. At the time of Board presentation, cofinancing of US$ 100 million was being sought from Japan and Switzerland. The Swiss Government agreed to contribute US$ 10 million to support the reform program. The second tranche (US$ 29 million) was released in June 1993, about 8 months behind schedule. As of January, 1997, the disbursement of the third tranche of FINSAC II has been authorized but not yet released. Consequently, this report focuses on the implementation of 40 the loan until end-June, 1996. The Government has made good progress in completing all of the steps necessary for third tranche release (Annex II), other than completion of the divestiture project and attaining macroeconomic stability. Macroeconomic Management 3.19 FINSAC II was predicated on the Government being able to maintain and build upon the macroeconomic improvement that they had achieved in 1990-91. The first tranche was released in 1992 when inflation was 10%, the lowest during the 1988-95 period. Unfortunately, as noted earlier, the political changes in 1992 contributed to a destabilization of the economy, particularly through a large wage increase for public employees. As a result, the fiscal deficit rose to around 6% of GDP in that year, the money supply expanded by more than 50% and inflation increased again in 1993. The current account deficit also widened sharply to 8.6% of GDP. 3.20 In 1993, the Government took some corrective measures to restore macroeconomic stability, following detailed discussions with the IMF and the Bank. At the time of the second tranche release, in June, 1993, inflation was running at 25%. The Government, Bank and IMF expected that an improved fiscal situation would enable the Government to make repayments to the banking system and that BOG would maintain a tight credit policy to bring inflation down. While the Bank and IMF considered that the project in place at that time (monitored informally through quarterly benchmarks in the IMF Staff Report for 1993 Article IV Consultation) would be sufficient to restore external viability and sustained growth, it appears that Bank and IMF staff were over-optimistic in this assessment. As Table 1.1 shows, budget deficits continued to fuel balance of payments (current account) deficits and inflation. 3.21 The links between the maintenance of macroeconomic stability and the tranche release for FINSAC II were vaguely worded. The President's Report states the condition as "...and consistency of the macroeconomic policy framework with the objectives of the project". The Second Tranche Release Memorandum dated June, 1993 states "during the first quarter of 1993, consumer prices rose by 13% and the cedi depreciated by 15 percent. But, reflecting the impact of budgetary measures, the exchange rate and inflation have stabilized since then... .IMF staff consider the project...will be sufficient to reestablish Ghana's strong track record of policy performance". The weak macroeconomic conditionality has had serious consequences for the overall success of the reform project. 3.22 Since the release of the second tranche, FINSAC II's supervision has not dealt explicitly with issues of macroeconomic instability. While other missions reviewed these issues with the Government (for example, in th.e preparation of the CEM), the lack of a review under the FINSAC II missions may signal to the Government that FINSAC II tranche releases will not be conditioned macroeconomic stability. Meanwhile, high inflation and fiscal deficits are negatively affecting the banking sector in several ways (eroding the banks' capital base; allowing the Government to mobilize subsidized credit; continued monetary expansion, etc.). 3.23 Conditions regarding the level of interest rates were also left open. Because interest rates had been liberalized prior to FINSAC I, the Government and IDA had few explicit conditions regarding interest rates under FINSAC II. The only condition related to interest rate policy was the agreement between the Government and IDA that they keep interest rates on bank reserve requirements "under joint review" and make future adjustments as appropriate. Experience during the project's implementation, however, has shown that interest rates have 41 remained negative in real terms for long periods, primarily because of Government ownership of the main banks and control over all the money market instruments. Thus, the Government has been able to borrow at negative real rates from the banking system through its high reserve requirements, reducing the incentive to control expenditures and crowding out private sector Borrowers. Removing Policy Distortions 3.24 The Government has taken some other steps to improve the efficiency of financial intermediation. For example, the differential rates of corporate taxation (50% for banks compared to 35% for other sectors) has been eliminated. The tax rate on financial institutions was first lowered from 50 to 45 percent in the 1992 budget. Subsequently, the Government announced in the 1993 Budget Speech a further reduction of that rate to 40% and that unification with the standard rate of 35% would be achieved by the third tranche of the Credit. The reduction in the rate to 35% has since taken effect. Improving Competition: Divestiture of Banks 3.25 In 1992, the Government appointed a team of financial consultants to prepare in accordance with IDA-approved terms of reference, a specific proposal for implementing a divestiture project. The consultants' review spelled out various technical and financial modalities for divestiture, including rationalizing the seven banks through selective mergers, the valuation of net worth of the banks, marketing steps and other details. The first phase of the project was envisaged to include 3 banks (GCB, SSB and NSCB) which represent over two- thirds of banking system assets. GCB was expected to be privatized and SSB and NSCB were to be merged. (Subsequently, NIB has also been prepared for privatization). In the second phase, the 4 smaller remaining banks were to be privatized along with the new bank resulting from the merger of SSB and NSCB. At the time of the second tranche release, the Government had accomplished the following tasks in this portion of the project: * a public announcement of the decision for divestiture for GCB and the merger of SSB and NSCB (and its subsequent divestiture); * approval of GCB's divestiture project by its Board of Directors; * approval by Government of the consultant's proposal for incorporating GCB's London branch as an independent subsidiary company with 40% Government shareholding; * submission to Parliament of a draft law converting GCB and NSCB from statutory corporations into companies registered under the Companies code to allow private ownership; * Government's approval of the merger of SSB and NSCB prior to the privatization of the newly formed bank; * approval of the merger by the Board of Directors and shareholders of SSB and NSCB; 42 * appointment of the merger team for SSB/NSCB; * starting the formulation of the merger plan for SSB/NSCB including a timetable; and * adoption of marketing strategies as recommended by the consultant. 3.26 Completion of the divestiture of the three banks targeted under the first phase of the divestiture project is a condition of third tranche release. The divestiture has proved to be more complicated and difficult than either the Government or the Bank anticipated. SSB was the first bank readied for the market, due to its relatively strong financial performance and an active management team. SSB launched its initial public offer on the Ghana Stock Exchange in September, 1995. The shares offered represented 30% of the valuation of the bank, and 70% of that amount was subscribed. The search for a strategic investor, however, has been less successful. 3.27 The divestiture of GCB initially moved more slowly, owing to difficulties in finalizing the valuation. GCB launched its initial public offer (IPO) in February, 1996. As with SSB, the divestiture strategy calls for 60% of the shares offered to private investors (30% through an IPO and 30% to a strategic investor). The public offer was successful for both GCB and SSB. As of December 1996, the banks were listed on the stock exchange, SSB with 22% of its shares now held by private investors, and GCB with 41%. Shares of both banks were traded on the stock exchange above IPO level. 3.28 The search for the strategic investor, however, progressed more slowly than anticipated for both GCB and SSB. The difficulty in finding strategic investors arose in part because many international banks decided to consolidate their operations and concentrate on core markets in Europe, the U.S. and East Asia. In addition these banks are believed to be emphasizing transaction based business over deposit banking. In spite of difficulties, an agreement was signed in November 1996 for the acquisition by a foreign industrial and financial growth of 40% of GCB shares. For SSB, the Government received an offer at the same time from an international consortium for the acquisition of 40% to 51% of SSB's shares. In addition, although NIB was not included in the first phase of banking divestiture, the Government has nevertheless decided to initiate its privatization process. The Government remains clearly committed to the completion of the divestiture. Consequently, the closing date for the credit has been extended to March 31, 1997. Enhancing the Financial Condition and Effectiveness of BOG 3.29 The Government has taken a number of steps to enhance the effectiveness of the BOG. A revised Bank of Ghana Act, which adequately spells out BOG's autonomy, role and responsibilities, was enacted in early 1992 as a condition of effectiveness for FINSAC II. In consultation with IDA, BOG has phased out its previous quasi-fiscal activities, such as development finance, subsidized lending, equity participation in rural banks and assumption of the exchange risk on behalf of the Government. For the release of the second tranche, it was agreed that BOG would implement its new organization and management structure, which it did. The new structure consolidates the former 19 departments into 6 and additional suitable personnel are being recruited to fill key positions. Annual training projects have been developed for 1993 and beyond and work is well advanced on the development of a computerization project for BOG. Appropriate technical assistance has been provided under FINSAC I and II, covering 43 advisory and consulting services, as well as equipment, hardware and software for banking supervision, research, foreign exchange management, accounting, computerization, human resource planning and training and general management. Continuing Bank Restructuring 3.30 As noted earlier, substantial work was carried out under FINSAC I to restructure 6 weak financial institutions. In addition, NPART was created to assist with debt collection from the restructuring banks. Prior to negotiations for FINSAC II, the Government developed an action project for the recovery of non-performing assets and for liquidations, including agreed annual recovery targets. In addition, the Government agreed to establish a project including performance indicators for monitoring bank performance. Prior to Board approval, it also developed a project and timetable for the computerization of GCB. For second tranche release, NPART was to complete implementation of its action project for recovering non-performing assets. NPART was successful in this endeavor, collecting about 92% of the targeted amounts over its first 5 years. NPART has remained in operation through 1996, having received a mandate from the Government to make collections for COOP Bank which is now being restructured and to continue to pursue other non-performing assets. NPART has experienced some slowdown in the past year in its recovery efforts, due in part to the economic slowdown (which reduces the debt servicing capacity of its debtors) as well as to the fact that it had dealt with the easier loans in the first three years. The special judicial tribunal attached to NPART started its activities in October 1992, allowing NPART to initiate liquidation proceedings. Development ofNon-bank Financial Institutions 3.31 Until 1993, Ghana did not have an adequate regulatory or legal framework for non-bank financial institutions (stock exchange, insurance companies, insurance brokers and agents, discount houses, building society, export finance company, credit unions, thrifts and savings and loan institutions). In addition, the government-owned State Insurance Corporation of Ghana had been allocated all Government and state-run business underwriting, resulting in over 75% of the non-life business going to one company. For the release of the second tranche, the Government agreed to furnish IDA with the draft legislation for the regulation and supervision of non-bank financial institutions. The Government initiated in 1992 a review of the legal and regulatory framework applicable to all non-bank financial institutions. The review led to the preparation of draft legislation, which was reviewed by IDA. The Act was approved by the PNDC and signed in January, 1993. For third tranche release, the Government was expected to review with IDA the results of a diagnostic study of the financial condition, capital adequacy, profitability and competitive position of the insurance industry. The study has been completed by an international consulting firm and discussed with the Government. The Bank is using it to prepare a project to assist the NBFI sector's development. 3.32 Social Security and National Insurance Trust (SSNIT) is the dominant non-bank financial institution in Ghana. From its inception until the late 1980s, the management of SSNIT's financial assets had been hampered by inappropriate Government policies and directives, which had required SSNIT to invest mainly in Government stocks and bonds with inadequate and often negative returns. For second tranche release, the Government was to commence a review and actuarial assessment of SSNIT's new pension scheme and its role in the capital market. A consulting firm was retained to undertake a study of SSNIT, based on detailed 44 terms of reference agreed with IDA. The study has been completed by an international accounting firm and discussed with the Government. It is also being used to prepare the NBFI project mentioned above. Training ofBankers and Accountants 3.33 FINSAC II continued efforts to train bankers and accountants which began under FINSAC I. As noted above, portions of the technical assistance component have been used for this objective under both credits. Under FINSAC TI, additional measures were specified for strengthening the skills base in these two sectors. First, a Banking College was established in 1994, with an appropriate legal foundation. Courses were first offered in July, 1994. 3.34 Progress in the accounting profession has not been as satisfactory (although vague wording of tranche release conditions would permit funds to be easily released). For example, while the ICAG has increased the number of accountants certified, IDA staff feel that this institution has not functioned as effectively as expected, particularly in regards to its financial self-sustainability. In addition, the Auditor-General has been reluctant or unable to outline how his office intends to use the technical assistance funds to strengthen their operations. Informal Financial Sector 3.35 With the Government's agreement, this study was dropped from the project as the World Bank's research department completed a study of Ghana's informal financial sector. It found that the informal financial sector was large and intermediated almost 50% of savings. The study was discussed with the Government. 45 4. Project Outcome Overall Contribution of the Financial Sector to Economic Growth 4.1 The overarching objective of financial sector projects is to increase the quantity and improve the quality of financial intermediation while deepening the financial system. While the two FINSAC projects have contributed importantly to institutional strengthening of the financial sector, to date they have had little impact on increasing financial intermediation or deepening the financial system. As shown in Tables 4.1 and 4.4, most of the indicators for financial sector intermediation and deepening have either declined or remained stagnant during the period. M2/GDP has remained flat throughout the 1988-95 period. More importantly, currency held outside of banks, which had dropped during 1990-91, has been increasing over the past 4 years. Thus, little financial deepening has occurred during the project implementation period. In addition, the financial system's performance in mobilizing savings remains inadequate. Private savings as a percentage of GDP remained extremely low at 2.6% of GDP in 1994. Gross domestic savings, already low at 7.8% of GDP in 1991, fell as low as -1.0% in 1993, and had only recovered to 4.4% of GDP in 1994. Not surprisingly, private sector investment as a percentage of GDP dropped by nearly 50% in 1992 and has not yet recovered. Table 4.1: Financial Sector Indicators, 1988-95 (in percentage) 1988 1989 1990 1991 1992 1993 1994 1995 M2/GDP 18.0 16.9 13.9 13.2 17.3 16.8 19.5 17.6 Non Bank- held 34.3 34.5 28.3 26.5 35.3 33.6 39.2 41.2 currency/M2 Non Bank- held currency 52.4 52.8 39.4 36.1 54.6 50.6 52.8 69.8 /Total Deposits Total Deposits/GDP 15.1 17.1 15.7 16.7 16.5 21.7 20.9 17.0 Dem. Dep./Ttl Dep. 56.2 65.6 67.4 59.7 55.5 NA 54.3 48.4 PrivSecDep./Ttl. Liab. 42.7 30.6 32.7 30.4 34.7 31.0 30.3 29.9 PrivSecDep/Ttl Dep 66.7 42.3 43.7 40.0 48.7 NA NA NA PrivSec Credit/ GDP 5.1 2.9 2.8 3.1 4.5 NA NA 5.2 PrivSec Credit/Ttl Dep 23.5 32.6 25.0 19.3 28.0 NA NA NA Reserves (Dec)/Ttl Dep NA NA NA NA NA 72.4 80.0 75.5 Net Loans/Ttl Assets 21 13 12 12 17 15.5 16.5 18.3 Sources: 1. Ghana: Financial Sector Review: Bringing Savers and Investors Together, World Bank, December 29, 1994. 2. Quarterly Economic Bulletin, October-December, 1995, Bank of Ghana. 3. Bank of Ghana. 4. Project Completion Report for FINSAC I, April 1995. 4.2 Preliminary indications are that the financial system's intermediation (particularly through banks) to the private sector has not increased. The deposits in banks/GDP ratio has stagnated since 1988 and private sector deposits/total deposits declined to below 50% 1989 and remained below that level until 1992 (last year of data). Overall, banks' total reserves/deposits have exceeded the required level of 57% during 1993-1995, by a substantial margin at times, 46 leaving little for intermediation. Not surprisingly, total loans/assets have remained below 20% and credit to the private sector/GDP declined from 1988 to 1990 and has only recently achieved its 1988 levels. Private sector credit/total deposits varied during 1988-1992, the years for which data are available, averaging a low 25%. High reserves and low net loans/assets (less than 20%) are probably the best indicators of the low levels of intermediation. Equity funding, has grown to about 13% of assets in recent years. The quality of intermediation may, however, have improved, as loan loss provisions have decreased substantially in recent years (Table 4.2). On the other hand, it is difficult to determine whether there has been an improvement in the quality of intermediation, as loans are short term and banks are currently lending at negative real interest rates. 4.3 The record on financial indicators (financial deepening, savings mobilization and resource allocation) cannot be judged independently of the macroeconomic context. It seems clear that the fiscal shock of 1992 had a negative impact on financial sector performance, by diverting credit from the private to the public sector, fueling inflation and shaking investors' confidence. In addition, the unstable macroeconomic conditions, high inflation and prolonged periods of negative and volatile interest rates have reinforced the Ghanaian tradition of holding non-financial assets. The FINSAC projects did not generate the expected results in the absence of more effective macroeconomic management. First, they were designed to be carried out only in the context of macroeconomic stability. Tranche releases could have been delayed to ensure that sufficient steps had been taken to reduce instability. Second, FINSAC II in particular should have addressed the increasing secondary reserve requirements, which have allowed the Government to borrow excessively from the banking sector at negative real rates for substantial periods. Financial Policy 4.4 Interest Rates. The liberalization of interest rates in 1988 appears, in hindsight, to have been premature. Through 1990, real deposit interest rates remained negative. While they moved into the positive range between 1991 and 1992, they have subsequently fallen dramatically, reaching -30% in real terms in 1995. The negative real interest rates have been made possible by virtue of continuing Government ownership of the main banks and control over money market instruments. The combination of largely negative and highly volatile real rates has served to deter both savers and investors from participating in financial markets. Banks had little incentive to increase deposit rates as safe lending opportunities were scarce (particularly in view of macroeconomic instability). In addition, because the Government chose to finance its deficits through bank reserve requirements, the banks again had little headroom and little incentive to seek out other lending opportunities. Finally, interest rates on government paper served as a benchmark for all other interest rates, especially after individuals were permitted to invest directly in those instruments. 4.5 In addition, the new capital adequacy and risk exposure requirements have made the banks even more risk averse. Nearly all banks hold government paper in excess of the reserve requirements. They say that there are few high quality Borrowers or bankable projects. In the pre-1987 era, banks faced stringent cash reserve requirements, sectoral and individual lending ceilings. In the current environment, they face high secondary reserve requirements (which are presently remunerated at negative real rates), stringent (appropriately) capital adequacy requirements and a dearth of viable private lending opportunities. Interest rates play little role in allocating credit in these circumstances. 47 4.6 Recently the BOG tried to control interest rate spreads through moral suasion, made effective because of Government ownership, to about 5% in nominal terms. A more sustainable method for reducing the level of interest rates, however, would be for the government to reduce its borrowing through the banking system and renew its efforts to lower inflation. In addition, it appears that borrowers in some sectors pay lower interest rates, particularly to GCB. In a recent interview, GCB staff indicated they charged 30% interest on loans for agriculture and exports, a lower interest rate than they pay on corporate deposits. Again, such low interest rates are made possible because of Government ownership. 4.7 Credit Allocation and Credit Controls. Formal mechanisms for allocating credit by sector were removed in 1987 and individual bank ceilings were removed in 1991. As indicated above, however, the Government continues to "review" overall credit targets with the banks and discourages them from exceeding these targets. Finally, GCB (and perhaps other public sector banks) continue to direct credit toward sectors such as agriculture, paying more attention to policy directives than to commercial priorities. 4.8 Money Market Development. The development of money markets in Ghana has reaped benefits for the banks and financial institutions, but has not contributed positively to monetary policy. Government borrowing from the banking system has been driven by fiscal needs. Trading in government paper has not been used to influence interest rates. Instead, as noted previously, the government has relied on moral suasion with the banks to influence interest rates. In addition, the discount houses are increasingly becoming marginalized. In the earlier period, when prudential regulations and supervisory framework were inadequate, the discount houses played a useful role in catalyzing the interbank market. At this point, however, the banks deal with each other. Discount houses simply absorb government paper that has not been purchased at auction (at auction price). Without a discount on those purchases, the discount houses have no incentive to make a market. Consolidated Discount House's profit after tax plummeted in 1995 to only 10% of 1994 in nominal terms. Instead, the discount houses have turned to fee-based services and credit operations, for which they are not necessarily prepared. For example, one of the two discount houses encountered serious financial difficulties when it diversified its operations and BOG was forced to intervene. 4.9 Competition. The reform project has done little to change the competitive structure of Ghana's banking sector. While several new banks have entered the market in the past few years and the share of the 4 largest banks has declined somewhat, the banking sector remains dominated by the same large public sector banks. In addition, the market remains highly segmented. The established foreign banks (Standard, Barclays) have a secure client base of multinational companies and blue chip local firms. Government banks primarily serve the retail depositor, Government, public enterprises and some domestic firms. Merchant banks are wholesale banks, focusing on fee-based and trade transactions. The liberalization of interest rates and changes in regulatory structure has not stimulated much innovation in product offerings or significant improvements in the quality of service. Instead, the current array of incentives in the banking sector may discourage innovation, given the small amount of funds available after fulfilling reserve requirements and relatively stringent capital adequacy ratios. It appears that only the market for low cost deposits shows effective competition in the banking sector at present. 4.10 Overall Assessment. In summary, the two FINSAC loans have had a negligible impact on the efficiency and effectiveness of financial sector policies. Although interest rate 48 liberalization was a step in the right direction, it may have been premature in 1988. Other policies, such as high reserve requirements and the Government's ownership of banks and moral suasion regarding interest rate spreads and credit targets, have hampered competition and allowed the Government to use the banking sector to finance its expenditures. Thus, the conduct of monetary policy has been driven by fiscal needs and has not contributed positively to macroeconomic management. Financial Infrastructure 4.11 Prudential Regulations and Supervision. The FINSAC projects have successfully supported a comprehensive reform of Ghana's prudential regulation and supervisory framework. Ghana's prudential framework is at international standards, setting out stringent requirements in the areas of capital adequacy, risk exposure, income recognition, asset classification and provisioning for loan losses, etc. Experience in other countries, such as Bangladesh, has demonstrated that failure to put such a framework into place permits "business as usual" to continue and provides few real incentives for the banks to improve the quality of their operations. Both on-site and off-site supervision in Ghana have improved. The Banking Supervision Department has trained and increased the number of staff to fulfill its expanding functions. It has requested a further increase in staff (40) to monitor foreign exchange operations and off-balance sheet risks. The Banking Supervision Department has also successfully merged and/or closed a couple of banks. It now has credibility among financial institutions, which adds clout to the regulatory framework and also has helped develop the interbank market. Finally, the Government has now put in place appropriate regulations for non-bank financial institutions and has established a supervisory unit in BOG. This unit has already successfully handled problems in two finance companies and one discount house. 4.12 Accounting and Audit. Efforts in this arena have been mixed. On a positive note, appropriate accounting and audit standards have been established. In addition, the banks are now audited annually according to those standards by international firms. Unfortunately, efforts to develop a strong core of indigenous accounting professionals have been less successful. While the ICAG has received technical assistance from IDA under both credits and has increased the number of accountants it has certified, the quality of work done by indigenous firms remains uneven. International accounting firms continue to conduct the audits of the banks, financial institutions and blue chip companies. Small and medium-sized firms use indigenous accounting firms. Because the quality of work by the local firms is uneven, however, these smaller firms often have difficulty securing financing from the banks. Therefore, additional efforts are necessary to develop the local accounting profession to remove barriers to financing for small and medium-sized firms. 4.13 Banking Education. Efforts to train banking professionals under both credits have been largely successful. The Bankers Institute was assisted under the credits and the National Banking College was established under FINSAC II. After some delays in its start-up, it is now running smoothly. As a result, the technical skills within the banking profession are improving. 4.14 Legal Framework. As indicated earlier, the FINSAC projects have supported important revisions in the legal framework for the financial system. New banking and central banking acts have been promulgated, along with appropriate regulations for non-bank financial institutions. Commercial laws, however, remain obsolete. In particular, it remains difficult to secure collateral for loans, foreclose on defaulted loans or register land. While the NPART tribunal has 49 helped to expedite judgments and execute decisions in foreclosure cases, it remains unclear whether these procedures will spill over into the judicial system. Moreover, the NPART tribunal has a limited term and only covers loans that it has taken over in the bank restructuring project. 4.15 Summary. Overall, Ghana has made excellent progress in updating its legal and regulatory framework for the financial system and putting in place an appropriate system for monitoring financial institution performance. These efforts provide a solid foundation for the future development of Ghana's financial system. Bank Restructuring and Divestiture 4.16 Bank Restructuring. From the banks' perspective, the financial restructuring was largely successful. It removed the non-performing loans mainly from the public banks as well as some from Barclays and Standard and improved their financial position. They have been able to delegate collections from those loans to NPART while they concentrated on revamping their operations, rebuilding their capital bases and preparing for divestiture. In addition, NPART has succeeded in meeting its collection targets over the past 5 years. As of October 31, 1996 it had recovered an aggregate amount of 18.7 billion Cedis, slightly above the target set by the Government in consultation with IDA. It also took on additional non-performing loans from the COOP Bank. The other components of the restructuring-reorganization, technical upgrading and staff training-have also moved ahead. Two of the public banks (GCB and NIB), however, remain overstaffed and have weak management. 4.17 Financial Performance of the Restructured Banks. As a group, banks are now operating profitably after restructuring. As shown in Tables 4.2 and 4.3, both the condition of their loan portfolios and operational efficiency have improved. The removal of the non-performing loans from the banks' portfolios improved the overall quality of their portfolios and reduced the need for loan loss provisions. In the past few years, banks have engaged in little new lending, and instead, as indicated previously, have devoted a large proportion of their assets to government paper. This move helps the banks to maintain a relatively high quality portfolio and reduces the need to replenish capital (as treasury bills carry a lower risk weight for capital adequacy purposes). Data on bad debt provisions and capital adequacy indicate that bank portfolios are reasonably sound. Under the prudential norms adopted by Ghana, assets are classified and bad debt provisions are made automatically on the basis of the Borrower's repayment record. Provisions are a charge against income and directly impact capital adequacy. Thus, improvements in bad debt provisions and capital adequacy strongly indicate improved loan portfolios. Moreover, compliance with these regulations is checked by the Bank of Ghana and in audits done by international accounting firms. Finally, loan portfolios are a small proportion of bank assets which are mainly government paper, suggesting that, by conventional measures, Ghana's banks are financially sound. Only COOP Bank has continued to experience difficulties, and was taken into receivership by BOG. 4.18 Most of the improvement in operational efficiency (as measured by a reduction in operating expenses) occurred between 1988 and 1990. Lending spreads have also increased in nominal terms. As shown in Table 4.4, a comparison of public and private sector banks shows that their performance indicators are now very similar, indicating a successful restructuring of the public banks. It is important to note, however, that the high returns on equity shown in 1994 and 1995 are somewhat illusory. With the rapid escalation in inflation, nominal interest rates have risen, thus boosting nominal income. The equity base, however, has not been pegged to 50 inflation. Also, with inflation at 60% in 1995, a return on equity of 50.7% for banks represents a poor return. Table 4.2: Operating Asset Ratios, 1988-95: All Banks (Percentage of Total Average Assets) Net Operating Operating Operating Loan Loss Profit Return on Year Income Expenses Income Provisions Taxes after Tax Equity 1988 8.2 6.9 1.3 5.3 0.8 -4.8 NA 1989 7.5 7.0 0.5 6.6 0.3 -6.4 NA 1990 9.2 5.2 4.0 3.5 1.2 -0.6 NA 1991 10.9 4.9 6.0 2.8 1.9 1.3 NA 1992 9.0 4.9 4.1 1.5 1.2 1.5 17.1 1993 9.3 4.8 4.5 1.2 1.7 1.6 22.0 1994 15.0 5.2 9.8 1.6 3.1 5.1 52.5 1995 16.0 6.3 9.7 0.6 2.5 6.6 50.7 Source: Bank monitoring reports, World Bank staff estimates and FINSAP Secretariat estimates. Table 4.3: Capital Adequacy Ratios of Banks, 1989-95 Number ofBanks 1989 1990 1991 1992 1993 1994 1995 Below0% 6 0 1 2 0 0 0 1-6% 1 0 1 0 0 0 0 7-12% 3 4 6 6 6 6 4 13-18% 0 3 2 3 2 1 4 19-24% 0 3 1 1 3 3 2 25% and above 0 0 2 1 0 1 2 Source: Bank of Ghana. Table 4.4: Comparison of Public and Private Sector Banks' Intermediation Costs and ROE (% of Total Assets) 1992 1993 1994 1995 Public Banks Operating Expense 4.1 3.4 4.3 6.4 Loan Loss Provisions 1.6 0.7 2.1 0.2 Taxes 0.7 1.2 3.4 4.4 Total Intermediation Cost 6.4 5.3 9.8 11.0 ROE 8.8 9.2 63.2 57.1 Private Banks Operating Expense 7.2 7.7 7.3 7.7 Loan Loss Provisions 0.3 0.9 0.5 0.1 Taxes 2.5 3.1 2.5 2.9 Total Intermediation Cost 10.0 11.7 10.3 10.7 ROE 31.6 43.0 52.6 46.9 Source: FINSAP Secretariat. 51 4.19 Divestiture. The search for a strategic investor for these banks has been more difficult than anticipated. Potential investors have cited a number of disincentives to participating in these banks, including: (1) the unstable macroeconomic situation; (2) Ghana, and more generally, the Sub-Sahara region was not a priority for them; (3) the overall volume of banking business is small and, at best, stagnant; and (4) they consider Ghana's overall medium term economic prospects to be mediocre. Moreover, many countries have offered government banks for sale at the same time, increasing the competition for strategic investors. Increased competition has also come at a time when global banks are undergoing consolidation to focus more on core markets. In spite of this relatively unfavorable environment, the larger Government owned banks found strategic investors and are in the process of completing their privatization: GCB, SSB (in which NSCB has been merged) and NIB. 4.20 Summary. Overall, the restructuring has been useful in cleaning up the balance sheets of the banks, refocusing their attention on streamlining their operations and initiating the divestiture process. It remains to be seen, however, whether strategic investors will be found for these institutions. For the larger public banks, such as GCB, a strategic investor is a critical element in changing the management structure and philosophy of the bank. Moreover, despite the improvements in financial performance, the bank restructuring has not facilitated an increase in financial intermediation. New loans are stagnant and most of the banks' funds go into government paper. More ominously, in the medium-term, if the banks continue to invest in government paper at negative real interest rates, they will ultimately erode their capital base once again. Capital Market Development 4.21 The goals of the two FINSAC projects in this arena were appropriately modest. Technical assistance has been useful in developing the stock market and providing an adequate supervisory and legal framework for non-bank financial institutions and capital markets. It is unlikely, however, that the capital markets will experience much growth until inflation comes down to more manageable levels. In addition, given the complexity of the two projects, adding capital market development tasks to the project may have diverted attention from banking sector issues.  53 5. Role of the Bank and Borrower Bank Performance 5.1 Project Design. The FINSAC projects were based on the Financial Sector Review (1986) and a series of technical papers prepared in 1990/9 1, which could not be located in the files. The projects covered every aspect of financial sector development. If anything, they covered too much and were too broad and contained too many performance conditions for tranche releases.5 A narrower focus on financial policies and infrastructure and restructuring and privatizing the distressed banking system-by far the biggest financial intermediaries-would have been more appropriate. In addition, FINSAC I & II were clearly one comprehensive financial sector reform project. Whether they should have been designed as two loans/credits with six tranches is open to question. Finally, tranche releases based on unspecified macroeconomic targets and uncertain transactions such as banks' divestiture to a strategic investor were inappropriately open-ended. Nevertheless, bank strategy embedded in project objectives and project design was clearly appropriate to enhance private sector activity. 5.2 Project Preparation and Supervision. Bank and Borrower cooperation in the process of project identification, preparation, appraisal and supervision appear to have been good. Appraising FINSAC I took only 6 staff weeks, owing in part to detailed sector work prior to the appraisal mission. As the project was complex, with substantial technical assistance, it needed 104 staff weeks for supervision. FINSAC II is still under implementation. 5.3 Overall, project design conformed with Bank country assistance strategy and project implementation of individual institutional components was satisfactory. Bank performance is therefore rated as satisfactory for FINSAC I and II. Borrower Performance 5.4 The Borrower has implemented complex projects of institutional change in a satisfactory manner. In FINSAC II, an implementation secretariat was established to oversee the task. A small group of senior Ghanaian experts, including a former governor of BOG, assisted by a technical expert from overseas, are successfully implementing the project. Bank divestiture to the public is proceeding but locating a strategic investor has proved difficult. The Borrower found strategic investors for the restructured banks and the disbursement of the third tranche of FINSAC II was authorized in December 1996. While glitches have occurred with foreign technical experts (including the quality of their work) the projects have been implemented satisfactorily, despite the delays in the divestiture project. Bank regional staff notes: "...In regard to the privatization of banks, there was indeed an initial under-estimation of its difficulties (particularly the search for the strategic investor) and in particular its time-consuming nature, but the relevant conditionally did contribute to accelerating the privatization process, even at the cost of delaying the third tranche release." 6 Bank regional staff notes: "...Although macroeconomic targets are necessarily "unspecified" in the project's credit agreement, the Bank and IMF have, during the entire life of the two projects, been monitoring the macroeconomic situation, and the project supervision was accordingly guided." 54 5.5 Project implementation has primarily been marred by the macroeconomic circumstances not being conducive to the project's success. The project was carried out in spite of high and varying inflation, volatile interest rates which were negative in real terms for substantial periods, and huge amounts of government borrowing from banks. On balance, Borrower performance is rated as marginally unsatisfactory for FINSAC I and II. 55 6. Overall Evaluation and Ratings' Contribution to Financial Intermediation and Deepening 6.1 The primary objective of financial sector projects is to deepen the financial sector, increase financial intermediation and improve its quality. Several indicators in Table 4.1 suggest that between 1988 and 1995 there was no substantial deepening in Ghana's financial sector and financial intermediation to the private sector did not increase. Judging by low loan loss provisions made by the banks in recent years, the quality of financial intermediation may have improved. As bank loans have been made at negative real interest rates for several years during the project, however, the quality of financial intermediation remains uncertain. Contribution to Financial Policy Development 6.2 The financial policy components of FINSAC I to liberalize interest rates, eliminate credit allocation and develop money markets and instruments for monetary policy operations were consistent with good economic policy. The timing, phasing and sequencing of implementation could have been better, however. Liberalizing rates was risky while credit allocation remained in place (until 1991), many banks were insolvent and being restructured and bank supervision needed strengthening. It appears that bank portfolios did not suffer, but money market development was stifled by onerous government borrowing requirements and volatile and negative real rates of interest. Furthermore, overall credit controls and interest rate caps have been reintroduced through moral suasion by BOG. Also, lending to priority sectors at preferential interest rates continues even though no formal requirement has been specified. These partial reversals of macrofinancial policy reforms have been facilitated by Government ownership of banks. Finally, the macroeconomic stability requirement for tranche releases was not specified or appears to have been ignored. Contribution to Financial Infrastructure 6.3 Prudential regulations and supervision capability has improved markedly under the FINSAC projects and substantial contributions have been made to developing the legal framework for banking, non-bank financial institutions, accounting and audit standards and banking education. Updating commercial laws and procedures, important for collateral, foreclosure and liquidation, were omitted. Nevertheless, the FINSACs' contribution to this component was a salient accomplishment, particularly in the design and implementation of a complex and large technical assistance project. Bank Restructuring and Divestiture 6.4 Financial restructuring of the banks was successful and NPART has substantially met its targets for collections from the banks' non-performing assets. The restructured banks now meet 7 As FINSAC II is not yet completed, ratings for its achievements are provisional, and based on the current situation. 56 capital adequacy and other prudential norms and have achieved financial performance ratios equivalent to that of the main private foreign banks. That performance is a significant achievement. The organizational and operational restructuring of two public banks has been less successful but still satisfactory. They continue to have weak management and inadequately trained and excessive staff. 6.5 Bank divestiture of two banks through a public offer has been successful and a third has been readied for a public offer. Divestiture to strategic investors is well advanced. Sale memoranda have been prepared and merchant banks have been appointed to search for strategic investors, and offers have come forward. With hindsight, sequencing the divestiture differently may have resulted in a faster outcome. Finding a strategic buyer first would have made banks more attractive in a public offer. On the other hand, given the difficulty in finding a strategic investor, it may have been appropriate to proceed with the public offers. Capital Market Development 6.6 The technical assistance provided for developing regulations and supervision for NBFIs was successfully implemented. The support for capital market development was also successfully completed. Overall Outcome 6.7 The project designs were appropriate and implementation of project components satisfactory. The financial infrastructure has been developed and distressed banks have been restructured. The stated objectives of financial institution strengthening have been achieved while the underlined objective of financial deepening was not. Therefore, institutional development impact is rated as substantial, and the overall outcome for both projects is rated as marginally satisfactory. 57 7. Sustainability and Lessons Learned Sustainability 7.1 The main successes of the FINSACs until now have been in developing financial infrastructure and restructuring banks. The sustainability of these achievements depends on the future improvements in financial deepening, intermediation and policy. In turn, success in these objectives hinges critically on improvements in macroeconomic stabilization-lower inflation, positive real interest rates and lower government borrowing from banks. Thus, sustainability depends on improved macroeconomic management in Ghana, and is rated as uncertain. 7.2 Unless a prolonged period of macroeconomic stability prevails, financial sector projects have an built-in conflict. Financial policy reforms can be implemented in a relatively short period while developing financial infrastructure and restructuring banks takes a longer time. In countries where macroeconomic stability is uncertain, it may be better to proceed first with a large technical assistance credit to upgrade financial infrastructure and restructure banks and follow it with a sector loan to improve financial policies when macroeconomic stability is achieved. Lessons of Experience * Financial policy reforms can only achieve financial deepening objectives in periods of macroeconomic stability. Therefore, macroeconomic performance conditions for tranche release should be clearly specified and strictly followed. * As financial infrastructure and institutions take a long time to restructure and develop, financial sector institution building projects can precede financial sector policy loans, particularly in periods of macroeconomic imbalance. * The phasing and sequencing of financial policy reforms should be designed to minimize undesirable outcomes such as a recurrence of negative real interest rates or bank insolvency. * Financial policy reforms such as eliminating credit ceilings, credit allocation, interest rate controls and subsidized interest rates can often be informally reversed, especially where the government owns a substantial portion of the banking system. * Complex projects with many components are difficult to design and implement, especially in countries short of appropriately skilled personnel. A series of smaller, focused projects is probably more appropriate.8 Bank regional staff notes: "This statement disregards staff constraints and other operational constraints encountered by the Bank in preparing and processing a series of small projects, as well as the fact that such sequencing may unduly slow down the overall pace of reforms. Furthermore, reforms are often mutually reinforcing and complementary and therefore, require simultaneous implementation in order to be effective..." 58 * Too many conditions for loan tranche release delay the release of funds. A few, clearly specified conditions would be more appropriate. * Tranche conditions that hinge on uncertain transactions over which the Borrower has no control are inappropriate. Tranche conditions should be reasonably implementable. 59 Annex I REPUBLIC OF GHANA FINANCIAL SECTOR ADJUSTMENT CREDIT I (CREDIT 1911-GII) POLICY MATRIX ISSUES/ACTIONS IMPLEMENTATION I. Measures to Enhance Soundness of Banking Institutions A. Prudential Regulations 1. (i) Establish a minimum capital adequacy 1-5: Legal Instrument (L.I. 1389) satisfactory ratio for prudential purposes; (ii) increase to IDA passed in December, 1988, covering minimum amount of paid-up capital to inter alia minimum capital, capital adequacy establish a bank (to be revised from time-to- ratio, development banking lending limits and time by BOG); (iii) apply capital adequacy auditing requirements (long-form audit). provisions to development finance institutions Enforcement of its disposition has started engaged in commercial banking; (iv) limit under close monitoring by BOG. dividends by banks that do not meet capital adequacy guidelines; and (v) empower BOG to mandate greater capital for prudential purposes for individual banks on case-by-case basis. Draft of new Banking Law superseding the Banking Act 1970 (Act 339) and the Banking and Financial Institutions Decrees, 1979 (AFRCD 17) under review by IDA. The new law, embodying the dispositions introduced under L.I. 1389, was approved by the PNDC on August 8, 1989. 2. Require development finance institutions, where engaged in commercial banking, to provide for statutory reserves of at least the levels required of commercial banks. 3. Place limits on secured as well as unsecured loans to any single Borrower (individual or enterprise). Place stricter limits on secured as well as unsecured loan to companies in which a bank's director has an interest exceeding a certain percentage. 4. Review and increase penalties for violation of banking regulations. 5. Require audited reports on banks submitted to the BOG to be in long form format in accordance with other BOG guidelines, based substantially on international standards. 60 Annex I 6. BOG to require banks to classify their portfolios New accounting and auditing standards for according to criteria for substandard, doubtful and banks were prepared by mid- 1988 with support loss assets (to be reported on a periodic basis), of an international auditing firm, reviewed by together with the specific provisions established). IDA and subsequently incorporated in L.I. These standards to be developed with the assistance 1389. of an international accounting firm. Implementation of the full range of remedial actions to await the appointment of two technical advisors in bank supervision. 7. Introduce improved reporting requirement from New, improved reporting standards satisfactory banks to BOG, off-site analytical capabilities and to IDA were introduced in 1989. Compliance indicators, as well as logistical support. by banks since have been generally satisfactory. 8. Support establishment of a credit clearing house The credit clearing house (to be renamed for banks. interbank credit data bank) initially expected to start operating by end- 1989, is now to be operational in 1994, following a delay on account of the Ghana Bankers' Association. B. Banking Supervision 1. Enforce, through guidelines and monitoring Has been in effect since 1989 as part of the action, internal controls and in-house audits of banks. restructuring of state-owned banks. 2. (i) Issue mandatory minimum terms of reference New audit requirements enacted under L.I. expanding scope of external audits of banks and (ii) 1389. The new standards have been enforced hold periodic discussions between BOG and the for all banks as of calendar year 1988, under auditors. BOG's close supervision. Compliance is generally satisfactory. 3. Strengthen examination methodology and skills. A new supervision manual has been introduced Review staffing needs of BOG's Examination since 1990. Department, incentives and staff training, and set up a permanent training project for examiners. 4. Establish computerized system for prudential Software for the processing of the related returns for banks and analyze the financial situation returns, including an early warning system, was of banks based on a set of performance indicators. developed (based on the new reporting Provide feedback to bankers. standards) in 1989 and became fully operational by December 31, 1989, following acquisition of required hardware in October 1989. 5. Draw up and implement an annual inspection In 1992, Bank Supervision Department of BOG plan, including full and targeted examination (each carried out on-site inspections of all banks, on bank to be visited at least once a year). the basis of a pre-established annual inspection plan. This was to be repeated in 1993 and thereafter. 6. Repeat external audits of banks with assistance of Done for state-owned restructured banks. international auditors. 61 Annex I C. Accounting and Auditing 1. Institute of Chartered Accountants (ICA) A review of training needs and other measures has agreed to be the focal point for efforts to for improving the accounting profession was upgrade and strengthen the accounting carried out with ODA financing. The report's profession in Ghana. recommendations, including an action plan for strengthening ICAG, has been implemented since 1990. II. Banking Restructuring 1. Carry out full diagnostic studies by Completed by 1989. international auditing firms. 2. Implement safeguard measures to prevent Directive issued by BOG in February 1989, further deterioration. which also has been closely monitoring compliance by banks undergoing restructuring. 3. Recruitment of an advisor on banking Restructuring advisory team started work in restructuring. January 1989 under Swiss grant. 4. Formulation of framework and timetable General Framework document approved by satisfactory to IDA specifying the modalities Government in July, 1989. Implementation for the restructuring of banks, in particular (i) modalities included (i) enabling legislation for measures for dealing with banks portfolios of NPART and the Special Tribunal; (ii) non-performing loans, including loans to state- operational guidelines for NPART; and (iii) owned enterprises, and (ii) transitional measures to arrest deterioration rescheduling/conversion of Government's and achieve a return to profitability of loans to banks. restructured banks. 5. Agreement with IDA on specific proposals The cleaning up of banks' portfolios through and targets for a reduction of banks' non- replacement of non-performing assets by performing portfolio (including off-balance Government bonds and/or other offsetting sheet items). operations was achieved by 1991-92 for six of the seven restructured banks, expect for the COOP Bank. 6. Finalization of specific restructuring plans Restructuring plans for six of the seven acceptable to IDA for commercial and distressed banks were completed by end- 1990, development banks. reviewed by IDA and implemented since. The one exception was COOP Bank. 7. Undertake study on needs for corporate The Time-phase study on corporate sector restructuring. restructuring was completed by external consultants in May 1990. It was subsequently followed in 1991-92 by an updating study, by local consultants. A number of consultants' recommendations were acted upon under a separate IDA investment project (PEED). 8. Carry out study on exchange risk It was subsequently jointly decided by protection. Government and IDA to abandon this study, which no longer appeared relevant following the abolition of exchange controls. 62 Annex I III. Deposit Mobilization and Credit Allocation A. Deposit Mobilization 1. Review provisions in laws and regulations relating to disclosure of information. 2. Allow banks to set scale of charges; Deregulation of banking charges completed by undertake survey of charges on a semi-annual October 1990. basis and publish results. 3. Amend regulations to permit banks to vary No such regulations existing in Banking Act of working days and business hours of branches. August 9, 1989. 4. Review system for collecting and clearing Status unknown. checks. 5. Allow banks to set own interest differential Deregulation of banking deposit rates achieved on demand, savings and time deposits. by February 1988. 6. Design and introduce package of measures Completed. to attract emigrant remittances. B. Credit Allocation 1. Allow banks to set their own interest rates Deregulation of banking lending rates achieved on loans. by February 1988. 2. Discontinue sectoral credit targets. Last sectoral credit target, for agriculture, was removed in November 1990. 3. Establish mechanisms and criteria for Status uncertain. There appears to be no need determining the use of special line of credit for action as the only lines of credit in Ghana and other special schemes to be used only in are from the Bank Group (for SMEs, Export special circumstances. Development, etc.). 4. Review efficacy of Credit Guarantee This scheme has been dormant and is now Scheme. inoperative, thus obviating the need for its review, It is being addressed under the PEED project. C. Efficiency of Banking Operations 1. Prepare and circulate standard set of BOG has been collecting, computing and average efficiency measures for banks. analyzing relevant operating ratios for banks (operating cost, profitability, interest margin, etc.) for comparison purposes. IV. Money and Capital Markets A. Money Market 1. Determine modus operandi for CDH. Consultant report on strengthening CDH, Establish nature and extent of liquidity support financed by ODA-UK, completed by April, by BOG. Identify prudential reporting and 1990 and implemented. monitoring requirement of BOG. Define operational limits through acceptance by CDH and BOG of a CDH Statement of Operational Policies, to be developed with Technical Assistance. 63 Annex I B. Capital Market 1. Conduct study of (i) investors, (ii) corporate Study undertaken by IFC in 1989. Its findings issuers and (iii) secondary trading to determine confirmed desirability to proceed with caution demand. in developing regulatory and institutional (i.e., Stock Exchange) framework since volume of business was expected to be thin in early years. 2. Review experience in other countries, Capital and debt market evaluation being possibly with IFC assistance, and make continued under PEED project. decision on applicability in Ghana. 3. Discuss available incentives, review likely effects and implement. 4. Develop with technical assistance a suitable regulatory framework depending upon outcome of study (I.V.B.I) 5. Give financial and logistical support to ICA ICAG received financial and logistical support to review and upgrade auditing standards for under FINSAC I, which enabled it to start corporate sector. work toward developing a uniform set of accounting and auditing standards for the corporate sector. This work is being supported under FINSAC 1I. 6. Amend Social Security Decree Regulation Previous regulations restricting investments by to allow some of social funds generated to be SSNIT to Government-issued securities were invested in long-term securities of credit- rescinded by mid-1990. worthy finance institutions and companies.regulato 4 65 Annex II REPUBLIC OF GHANA FINANCIAL SECTOR ADJUSTMENT CREDIT II (CREDIT 2318-GH) POLICY MATRIX ISSUES/ACTIONS IMPLEMENTATION A. Measures to remove policy distortions in the financial system. 1. Reduce the implicit taxation of banks by Before negotiations, the Government agreed to remunerating required reserves at a rate of keep interest rates on bank reserve interest closer to the average cost of deposits requirements under joint review with IDA and of the banking system, in the context of the to make future adjustments as appropriate. overall liquidity management policies. 2. Lower the high cost of financial Before Board approval, the Government intermediation through lowering corporate tax confirmed that tax rates on financial rates for financial institutions to levels institutions would be made comparable to applicable to other sectors. other sectors, beginning in 1992. The tax rate was first lowered from 50 to 45 percent in the 1992 budget, then further reduced to 40 percent in the 1993 budget and finally unified with the standard rate of 35% in the 1994 budget. 3. A project for the phased divestiture of Before Board approval, the Government public sector ownership of banks, to promote submitted an acceptable proposal for the competition through increased private sector divestiture of public sector shareholdings in ownership. banks and issuance of invitation for proposals by consultants. For second Tranche release, the Government agreed to commence implementation of the first phase of the project, which included the privatization of GCB and the merger of SSB and NSCB, prior to privatization of the merged institution. The Government made sufficient progress in these two tasks to comply with IDA's expectations for second tranche release (1993). SSB made a public offer in 1995 for 30% of its shares and 22% were subscribed and GCB made a public offer in 1996 for 30% of its shares and 41% were subscribed. As of December 1996, strategic investors were identified for SSB and GCB, and agreements were signed for the purchase by these investors of 40% of GCB fshares and 40% to 51% of SSB shares. 66 Annex II B. Enhance the financial condition and operational effectiveness of the BOG. 1. Develop a sound legal system for the The Bank of Ghana Act was enacted in 1992. financial system. 2. Strengthen the financial condition of BOG Prior to second tranche release, BOG phased by removing revaluation losses and non-central out its previous quasi-fiscal activities, such as banking activities. development finance, subsidized lending, equity participation in rural banks and assumption of the exchange risk on behalf of the Government. 3. Institutional strengthening by realigning As a condition of second tranche release, the BOG's organizational structure to enhance its BOG implemented its new organization and capacity. management structure. The new structure consolidated 19 departments into 6 and several other tasks aimed at enhancing managerial effectiveness were well underway. 4. Improve operational efficiency through As of second tranche release, work was well computerization, changes in operating advanced on the development of the procedures, training to upgrade staff skills and computerization project for BOG and annual technical assistance. training projects had been developed for 1993 and beyond. Appropriate technical assistance has also continued under FINSAC II for the implementation of BOG's institutional strengthening project. C. Bank Restructuring and Loan Recovery by NPART. 1. Commence liquidation of non-viable Before negotiations, the Government, in enterprises in accordance with NPART's consultation with IDA, developed an action action project. project for the recovery of non-performing assets and for liquidations, including agreed annual recovery targets. These targets have been largely met throughout the implementation of the loan. 2. Consolidate, deepen and streamline the Before negotiations, the Government agreed to ongoing bank restructuring project started establish a project, including performance under FINSAC I. indicators, for monitoring the performance of banks. For the release of the second tranche, the Government agreed that the BOG would comply with the requirement of annual inspection of banks in accordance with the Banking Law. During 1992-93, annual on-site inspections were carried out for all banks operating in Ghana (commercial, development and merchant banks). 67 Annex H D. Enhance the effectiveness of non-bank financial institutions. 1. Strengthen the legal/regulatory framework, In 1992 the Government initiated a review of prudential regulation, and the supervision of the legal and regulatory framework for non- non-bank financial institutions. bank financial institutions. The review led to the preparation of draft legislation incorporating IDA's comments. The Act was approved by the PNDC and signed in 1993. 2. Improve competition in the insurance For the release of the second tranche, it was industry. agreed that SSNIT would commence the review of its asset/liability position and the actuarial assessment of its pension scheme as well as its role in the capital market. Consultants were appointed to complete the study under terms of reference agreed with IDA and the review was completed in 1993. E. Develop capacity in the banking and accountancy professions at all levels. 1. Upgrade the professional skills of bankers The Banking College has been established and through the continuation of training projects, began its project in 1994. Progress has been and the development of the national banking achieved in the development and college. implementation of training projects for bankers, accountants and insurance professionals. 2. Strengthen national accounting and auditing Despite ongoing technical assistance, work capabilities and help upgrade the standards and with the ICAG and the Auditor General's skills of the Auditor General's Office. Office has progressed more slowly than expected. F. Informal Sector 1. Study the informal financial markets to This study was dropped from the project as the investigate prospects for closer interaction World Bank's research department completed between the formal and informal financial a study on Ghana's informal financial markets systems. and discussed it with the Government. G. Administration/Implementation 1. Establish the Project Implementation Unit The FINSAP Implementation Secretariat was as a focal point to provide guidance, oversight, established prior to Board approval and it has coordination and monitoring for all FINSAC been fulfilling its functions. related activities. GovernmentIinitiatedarevi  69 Annex III P.O. Box 2674 Bank of Ghana Accra-Ghana Tel: 233-21-664719 Fax: 233-21-668241 GOVERNOR'S OFFICE, BANK OF GHANA, ACCRA FACSIMILE MESSAGE Date: 17/1/97 To: Mr. Manuel Penalver Company: I.M.F. Fax: 202-522-3124 From: H.A.K. Wampah Department: Research Department Number Of Pages: 1 + 5 If transmission is unsuccessful please contact us immediately on 233-21-664719 Comments: GHANA: FINANCIAL SECTOR ADJUSTMENT CREDITS I & II - DRAFT PERFORMANCE REPORT Please find attached our comments on the above report. I hope you will find them useful in the preparation of the final report. _Ust Regards .A. KW PAH for: Head, Research Department Fna 70 Annex III COMNTS ON PERMDP4AXCX AUDIT REPORT: GHANA: FINANCIAL SECTOR ADJUSTMENT CREDITS I p II. On the whole, the report well researched and written. We believe the analysis and conclusions will help significantly in the design and implementation of future projects. However, we have a few comments which, we hope, will help in the preparation of the final the report. Borrower/Bank Performance Rating The report rates Bank performance as satisfactory while rating Borrower performance as marginally unsatisfactory. In our view, these two ratings are not consistent since the report admits that some of the problems encountered in implementation were due to the inappropriateness of some performance criteria put in place by the Bank. For example, the use of the location of appropriate strategic investors as a performance criteria has hampered the smooth disbursement of the loan through no fault of the Borrower. The report also admits that some foreign technical experts have been inefficient in the quality of their work. Admittedly, the borrower could not provide the right macroeconomic environment for the complete success of the FINSAC I & II, although the delays in disbursements due to 4nappropriate performance criteria have also contributed to this. Rating of the Project Outcome The rating of FINSAC I & II as marginally satisfactory is a bit harsh as the financial system in Ghana has undergone 71 Annex III significant improvements which should warrant a more positive evaluation. As discussed in the Performance Audit Report, the financial system has been strengthened through the new legal framework provided for both the banking and non-bank institutions. Benefits of such improvements though difficult to measure, can easily be underestimated, especially when they impinge directly on transaction costs for business. This argument also holds for uniform accounting and audit standards that have been put in place. On those measures of evaluation that can be quantified, it is clear that FINSAC I was successful as banks now meet capital adequacy and other prudential requirements. Financial performance ratios have also been shown to have improved. The successful recovery of loans by NPART is a pointer to the need for further improvements in management of the banks and banking philosophy. Banks need to improve their collections function. FINSAC I & II have been assessed as having little impact on increasing financial intermediation or deepening ie through measures such as M2/GDP ratio or private savings/GDP ratio as well as on currency outside banks. These measures of financial deepening highlight a significant feature of the macroeconomic situation throughout most of the implementation period of the projects. With the benefit of hindsight, it seems as if FINSAC would have been more successful with financial deepening it macroeconomic targets for inflation had been clearly 72 Annex III specified and followed. Hence we agree with the writers that the need for macroeconomic stability for financial restructuring should be stressed. However, we would argue that although high reserve requirements especially secondary reserve requirements of government paper, and the government ownership of banks hampered competition and allowed the continued use of the banking sector to finance fiscal expenditures, they also in part helped improve performance of banks, when financial ratios such as return on equity (ROE) and return on assets (ROA) are used. This is especially the case when the Ghanaian economy itself has not yet undergone the technical and productivity (enhancing) transformations required to allow the banking sectors to create loans with high returns. Summary To summarize, inappropriate macroeconomic conditions prevented the FINSAC I and II projects from achieving the objective of financial deepening in the Ghanaian economy, although the institutional conditions and bank performance as measured by ROA and ROE improved. Nevertheless, it is important to recognize that these improvements, in ROA and ROE for banks depended directly on bank portfolios that were heavily concentrated with government paper, as they were the most attractive assets on the money and capital markets. Also, it must be noted that an isolation of financial deepening measures from growth in stock markets can be misleading. Although the M2/GDP ratio has not changed significantly, Ghana's stock exchange has performed well over the period of implementation of FINSAC. If financial resources can be used to acquire shares, rather than saved 73 Annex III in deposit money banks, then an evaluation of savings mobilization must not isolate stock market developments.  From: PHONE No. Feb. 05 1997 L: 45PM PS1 75 Annex IV February 5, 1997 Mr. Manuel Penalver, Division Chief Country Policy, Industry and F'inance Division. 0. 1. ). Pax : 202..522-3124 Dear Mr. Penalver, DRAFT PERFORMANCE AUDIT FOR CREDITS 1911 - GII AND 2318 - tII. Pursuant to my earlier faxed letwr of January. 10. 1997. on the abovc subject please he advised that the comments from the Bank of Ghana have been forwarded to you directly. 1. herby, forward the comments from this office (FINSAP Secretariat) to you. I trust that you will find thcse comments helpful. Yours Sincercly. C.D C C : Serge Michailof, l-;q. Country Director for Ghana World Bank. Dr Elic Salecby Task Managcr 14,NSAP 11 Project From: PHONE No. Feb.05 1997 1:45PM P02 76 Annex IV COMMENTS.ON FINSAP I AND II DRAFT PERFORMANCE AUDIT REPORT The document brings together in one write up a good deal of the factual presentation on FINSAPs I and II. The analyses are quite good but a few comments are presemted below on certain aspects of the document. GENERAL COMMENT: There appourt to be a mix in the use of the acronyms "FINSAP" and "FINSAC". Bank (IDA) staff see everything through the eye of the CREDIT they grant for the implementation of their projcQts. Consequently, the word FINSAC is used even where it should be FINSAP. As an offielal document it would be proper to use the appropriate acronym. When activities are being referred to we should use FINSAP rather than FINSAC which refers to the resources nadc available. EVALUATIONS AND RATINGS (P. 13): 1. It seems that a more realistic assessment of the performance of the projects should be based on the examination of the specific objectives set out at start. In this respect the objectives of FINSAP I and II as listed on pages 28, 29 and 30 of the document could be taken one by one and assessed or ranked in terms of 'good'- 'average'- or 'poor'- performance. Alternatively, each objective could be classified as 'achieved' or 'not achieved' to determine a total ranking of the project performance. A matrix of performance indicators could have been developed. If such an assessment and ranking methodology were adopted, nearly all the specific objectives would be seen to have been largely met. This view is conveyed by your project implementation assessment on page 11 paragraph II of the document. The paragraph begins "Despite some delays, the components of the two credits were effectively implemented and Ghana's financial policies were dramatically changed ...". The paragraph lists various achievements under the programme. It is therefore inconsistent to rank. On balance, the borrower From PHONE No. Feb.05 1997 1:46PM P03 77 Annex IV performance on page 13 paragraph 19, as marginally unsatisfactory for FINSAP I and 11. The ranking connotes poor performance at a glance. We believe that for policy follow-up performance should be distinguished from outcome. 2. In our vicw. apart .from the methodology issue, this conclusion has been arrived at based on the combination of FINSAP I and II. While in some cases it would appear appropriate to look at long term effects of policy to determine their effectiveness or cfficiency, in certain cases it would appear better to trace the implementation process to assess the performance and output. In this case, it would help policy formulation to separate the assessment of FINSAP I and FINSAP II for a number of reasons- Firstly, at the end of FINSAP I (1991), a significant macroeconomic reversal (your document p.9 pp 3 ) relating to a sizable budget deficit financed by the banking sector, occurred. Secondly, a new project FINSAP 11 came on stream under a new civilian government. Thirdly, FINSAP I is still under implementation and its assessient would better be left tu its conclusion. The statistical trends emerging from your table 4.1 on page 45 relating to (M2/GDP), (Non Bank-held Currency/M2), (Non Bank-held Currency/Total Deposits), (Total Deposits/GDP) and (Domestic Savings/GDP) and (Private Investment/GDP) of table 1.1 of page 18 were predictable and encouraging and re-enforce the need to look at the FINSAP I and II separately. The trend up to 1991 has been steady. With regard to the implementation of FINSAP 1, tahle 4.1 indicates a desirable trend and therefore ranking of performance and output should be more positive. 3. FINSAP 11 which is still under implementation. has run into some problem because the basis of the project design (see page 40 paragraph 3.19) has been undermined with the macroeconomic reversal of 1992. Notwithstanding the problem, implementation of the project has continued with the best of efforts in the face of the reversal overhung. The project objectives of FINSAP II are being achieved, albeit with a significant delay. In terms of classifying as "achieved" or "not achieved", quite a lot of the objectives have been achieved. Here, however, it is conceded that even though the project specific objectives have been largely met, the fundamental requirements of financial sector deepening as well as increased and imprdved financial intermediation have not. been realized. That so much progress has been made in the face of such difficult macroeconomic environment must be a credit to the project management team. Overall ranking should also consider where we had come from and where we are now. Frooi PHONE No. Feb.05 1997 1:46PM p84 78 Annex IV In addition, the point has to be made also that a sector project which aims at specific objectives cannot attain certain levels of achievement if the macroeconomic environment is repressive. Specifically, inadequate financial intermediation and financial deepening could be due to inadequate growth in the real sectors. Moreover, coming from a regime in which banks were blamcd for laxity and imprudence in lending, the tendency is to be more strict and avoid lending in a transitional phase. V BANK RESTRUCTURING AND DIVESTITURE: Sequencing the activities of the divestiture process in terms of a public floatation first before the SI search or vice versa received an active debate. In principle, identification of a good SI in advance confcrs significant advantages in a subsequent initial public otTer (IPO). However in thu two recent bank divestitures, public floatations were undertaken first because the SI searches had not been successful up to a point. As a policy, in the two cases SI searches were in fact initiated first but as the periodic monitoring of progress dictated, it became necessary to open a parallel track for both activities and to allow the faster or more successful activity to be implemented first. This was to avoid being labelled as a hesitant and unwilling vendor. In one case, SI negotiations commenced ahead of the IPO, but the failure of the particular negotiations resulted in an IPO ahead of the SI find. Our experience is to keep options open and assess the ground situations from time to time and take appropriate decisions. PROJECT OUTCOME: COMPETITION (P. 47): The claim that "the reform project has done little to change the competitive structure of Ghana's banking sector" (page 47 pp 4.9) appears a bit too strong. At the start of FINSAP I there were about 12 formal banks but by June 1996 this number bad risen to about 18 in addition to a host of new non-bank financial institutions which all compete for the available financial services. OVERALL ASSESSMENT OF FINANCIAL POLICY : The detailed review of components of policy under 'interest ratcs', '4rUdit allocation and credit controls', 'money market development', and 'competition' have generally given sufficient information on positive developments and associated problems. Perhaps, a fairer assessment could have been that in spite From: PHONE No. Feb.05 1997 1:47PM Po5 79 Annex IV of the fact that output expectations have not becn realized because of macroeconomic problems, sufficient groundwork has been done which could facilitate macroeconomic management as the economy stabilizcs. Such an assessment recognizes chat, in principle, the policy designs are appropriate except for the reverberations of the 1992 reversal. -- FINANCIAL INFRASTRUCTURE (P48): ACCOUNTING AND AIMIT : The report recognizes the positive contributions in terms of establishment of STANDARDS as evidenced in the regular annual audit of banks. The concern that insufficient indigenous accounting professionals have been trained leading to the use of international accounting firms in the conduct of audits needs to be clarified. Firstly, most of the 'so--called' international firms are either entirely manned or largely manned by indigenous staff. Secondly, as a matter of historical expediency to gain international recognition and respect, these firms have been set up as associates of the international firms. Thirdly, in order to confer respectability on their documents, the local banks typically recruit mostly these International accounting firms as their auditors. Fourthly, in view of the ]link of thc World Bank with the restructuring of the Banks the auditors of these banks have to be seen as the internationally respected accounting firms. Even though In recent times a lot of indigenous accounting and auditing firms have been set up, it remains that for respectability sake, the major assigments such as audits of banks are done by these 'international accounting' firms. The issue of concern should be whether we have increased the numbers of internationally accepted local accountants and auditors since the FINSAP 1 ? DIVESTITURE (PS1 PP4.20) : The suggestion that Government could have divested its shares in Barclays and Standard Chartered (40% and 25 %) as a way of making a market in bank shares and providing a benchmark to facilitate the banking divestiture appears unnecessary. In this case, the Standard Chartered is already listed on GSE and Barclays Bank has been bidding Lime to let the three banks' (SS3, GCB, and NIB) Shares he sold or let the process bc sufficiently advanced to let a successful Barclays Bank Share offer come on stream. Therefore, to let. the Govermcnt's current minority shares in BBPLC and Stanchart come ahead . .4 Fror; PHONE No. Feb.05 1997 1:48PM POM 80 Annex IV of the selling of the threc listed sliares would have complicated the issue of finding sufficient share takers and as well delaying the target date for the three listed banks. The introduction of this issue at this point appears as an unnecessary pushing of a decision on when to sell the remaining BBPLC and Stanchart shares. Fortunately, the process is well under way now without resort to such strategies. . - 9 CAPITAL MARKET DEVELOPMENT (P51 PP 4.22): The establishment of the capital market in 1990 should be seen as timely because the strength of the financial sector depends on a strong and healthy complenentarily of the banking and non-banking sector. The timing of the setting up of the OSE has proved to be well considered and the GSE has supported the broadening of the financial sector. It has been helpful in increasing the share ownership base and was particularly instrumental in the divestiture of Ashanti Gold fields Company. In the process it helped to increase foreign investment in Ghana by allowing inflows from non-resident Ghanaians and foreign fund managers. Moreover, many local enterprises have accessed long term capital without necessarily producing costly collateral which banks would have requested. In all, the time and resources diverted from the banks into supporting OSE could be said -to have brought increased benefits and efficiency to the financial system and the economy. OVERALL EVALUATION AND RATINGS (P55-56): The overall evaluation and ratings on pages 55 and 56 touching on. Financial policy development, Financial infrastructure, Bank restructuring and divestiture, and Capital market development, are appropriate. The institutional deve!cpment impact rated as substantial is also appropriate. The overall outcome for both projects, howeveir, is rated as marginally satisfactory. If FINSAP I and II were rated separately, one would expect a higher rating for FINSAP 1. Then a marginally satisfactory rating for uncompleted FINSAP 11 as of Junc 1996, when some of the significant divestiture successes have not been achieved, would be considered appropriate. FINSAP II could show a marked improvement in outcome by the closing date when we expect to secure ..5 From PHONE No. Feb.05 1997 1: 4ePM Poy 81 Annex IV SI for the banks on the divestiture list as well as improve other elements of the programme. OTHER COMMENTS: P 10 PP 8 Line -8 :-- For consistenc5the rendering of specific objective one, should read: ensure the soundness of banking ........and supervision; restructure financially distressed banks; and improe ..... P 24 PP 1.30 Line 5 : Need for consistency in the listing of shortcomings. These shortcomings include - . unspecified limits, (by-BOG), for unsecured credit ..... and reserves, therehy allowing banks capital funds; . varying treatment of interest on non performing assets in the ahence of uniform accounting standards ..... Moreover, many hanks.. . fourth line of the third sentence : Moreover, the existing legislation merely specified a CAR based on the deposit base (5 %) of the banks thereby ignoring the risk components/content of assets. . the neglect in treating development banks engaged in commercial bunking on the same lines/basis regarding the treating of retained profits; . penalties which were too low .... . audit and regulatory reports to BOG which provided ........ . insufficient numbers of personnel and inadequately trained banking supervision staff ........ P 37 PP 3.10 Last Line: The last sentence could be deleted without much loss. In any case, COOP Bank has not been merged yet. Plans are afoot to merge it with one of the un divested agricultural and rural - bascd banks. P 56 PP 6.7 : The first sentence appears incomplete.  17/01/97 09:01 BAWI ENTWICKLLNGSDIENST - 202 522 3124 NO.514 901 83 Te a Annex V Telefax Fedeml Office for Foreign Economic Affairs Bundeshaus Ost / Effingerstrasse 1/3 3003 Berne Bcrno, 16 January, 1997 W 0041-31-324 08 18 Fax 0041-31.324 09 62 Mr. Manuel Pehalver Division Chief Your ref.: 1911-GH and 2318-GH Country Policy, Ihdustry and Finance Division,. OED The World Bank Our ref.: 3/84-GH - mni/tis Washington D.C. Number of pages: 3 (including cover sheet) Fax no.: 001-202-522 31 24 Ghana / Financial Sector Adjustment Credits I and It: Draft Performance Audit Report Dear Mr. Pefialver, We acknowledge with thanks receipt of the above mentioned docuTrent. The preparation of the PAR by the Operations Evaluation Department is a very cruclal exercise to ensure that the past experiences are properly reviewed and analysed in order for them to be capitalized upon in the preparation and implementation of new operations. This is particularly truc for IDA Adjustment Credits to Ghana, as the Btructural adjustment process in this country has generally been considered to be one of the most successful in Africa. The experiences gained in Ghana are consequently of gendral interest for other developing countries. The Swiss Federal Office for Foreign Econoiic Affairs (FOFEA) therefore very much welcomes the preparation of this performance audit report on FINSAC I and 11 for Ghana. As a co-financier of FINSAC I and, in fact, the only co-financier of FINSAC II, we appreciate and thank you for the opportunity given to us to comment your draft report. We generally appreciate the candid analysis and assessment present6d in the PAR. In the same spirit, we submit herebelow for your consideration our main cpmmenL and remarks on the draft document. 1. PefuranMe ratng of Project Outcome The overall outcome of both projects is rated as marginally satisfactory in the draft PAR: we would like to question this assessment. Indeed, while fully appreciating the nature and scope of the achievements in terms of (i) setting-up the regulatory framework for a 17/01/97 09:01 BAWI ENTWICKLUNGSDIENST -* 202 522 3124 NO.514 902 84 Annex V market-based financial system in Ghana, (ii) developing the financial infrastructure, and (iii) restructuring the banks, we consider that an objective evalUation of the overall performance of both projects should better reflect the effective results and impact of this long-standing (7 years) and Nery substantial (more than 200 millions US$) operation vis-a-vis the actual "raison d'ftre", the primary justification, tht aim of these two projects. In this context, we note first of all that the PAR recognizes that the sustain- ability of the achievements in developing financial infrastructure *nd restructuring the banks is uncertain. Secondly, and more importantly, we considet that, based on the available information, the orojects have definitely failed to meet thir primary objective which was to deepen the financial sector, to increase financial intermediation and improve its quality. It's a fact that, by 1996, much remains to be done in Ghana for the financial sector to play an effective role in mobilizing domestic saving and prolviding increased bank lending, for the development of liquid and active financial markets, 4nd for the implemen- tation of an effective monetary policy. Our assessment is that the overall imact of the projects on the efficiency and effective- ness of the financial sector in Ghana has been very limited. and aculally not satisfactory. We believe that the PAR should acknowledge in a more straightforward manner this critical state of affairs, which could then lead to the undertaking of additional - and constructive - analysis on possible causes, in the projects' design anid implementation, for such a failure in achieving crucial objectives. Obviously, the deficient link between the financial sector reform and the macroeconomic framework, as w1ll as the inadequate liming of the interest rate liberalization, are crucial elements to be taken into account But additional factors could have been highlighted in the draft PAR as r4quiring some further investigation. Without aiming at any comprehensive review, we recommend that the PAR should also address issues such as the following : (i) what was tie effective impact of the training programmes ? (ii) were the accompanying technical assistance programmes adequate to face the needs, properly sequenced and effective 7 (iii) was the design of the development of money markets adequate (in particular the role of the discount houses) ? or (iv) should (and if yes, how ?) the informal financial sector (representing nearly 50% of domestic saving) have been better - and at an early stage - inttgrated in the overall reform process ? etc. The discussion of such issues could certainly have a relevance not only for better understanding the situation in Ghana, but also for providing other developing countries, the Bank and other donors, with a useful set If lessons learned and experience gained. 2. Bank Perfurnnce Rating The performance of the Bank for both projects is rated as satisfactory in the draft PAR. When considering the failure of the projects to achieve primary and crucial objectives, we assess less positively than you the overall performance of the Bank. Although we certainly agree that the final responsibility of the projects lay with the Borrower, we do believe however that the Bank has been playing actually a key rote in providing policy guidance, advice and technical assistance during the process of design, implementation and supervision of the projects. In this context, we consider that the Bank's self-evaluation should be more critical as regards its own performance in the steering and supervision of the operations. The draft PAR mentions some shortcomings; we assess quite a few of those as actual serious deficiencies. For illustrative purpose, we can highlight three such examples : (1) the effective lack of direct correlation between the financial sector reform and the macroeconomic framework - no explicit and firm macroeconpmic conditionality in Pederal Office for Foreign Economic Affairs 1"/01/97 09:02 BAWI ENTWICKLUNGSDIENST - 202 522 3124 NO.514 903 85 Annex V FINSAC I and It - is hardly understandable in retrospect. Moreover, pven more difficult to understand is that no corrective measures have actually been taken in this regard during the 7 years period of implementation, despite very substantial sipervision (104 staff weeks for FINSAC T), and obvious negative consequences of macroconomic instability in the implementation of financial sector reforms; (2) the release of ttie second tranche of FINSAC U in June 1993 at a time when several macroeconomic indicators were "red" (inflation had surged to 25 %) demonstrates a definite lack of vulnerabiity/sensitivity analysis in the course of project implementation and supervision; (3) the lack of explicit interest rates policy conditions - and the subsequent absence of action by the Bank to introduce corrective measures - during FINSAC II has led to severe distortions in the structural adjustment process, the Government being in a position to borrow during long periods at negative real rates from the banking system, and thery crowding out the private sector. 3. Other rmnarks As per Article 3/b/(v) of the Procedural Arrangements between the Government of Switzerland and IDA for the Cofinancing of Projects and Programmes, signed on 9th May. 1990, the Federal Office of Foreign Economic Affairs (FOFEA) thould have received regularly the FINSAC supervision reports, as soon as available, as this is being practised for similar operations in other countries. As a matter of fact, such reports have only been transmitted to FOFEA by the World Bank on a rather erratic bakis. Taking the most recent example : in November 1995, following a request by the World Bank, FOFEA agreed to finance a specialized consultant for his participation as a f611-fledged member of the FINSAC IT supervision mission which was being organized. Althbugh we had received explicit confirmation that a supervision report (or back to office repprt) would be sent to us, and despite several reminders in the course of 1996, we have until now not yet received such a report. Finally, we would like to draw your attention to two small errorg in the "Basic Data Sheet" where the co-financing by Switzerland has been omitted. We thank you in advance to take into consideration the above-mentioned elements when you finalize the PAR on the FINSAC I and II. We shall be very grateful if you could kindly send us a copy of the finalized PAR, as soon as available. Wilh best regards. Yours sincerely Development Policy pivision Matthias Meyer Vice-Director Copy to: - Swiss Embassy in Ghana - Swiss Executive Director at the World Bank - SDC, Economic Issues Section - SDC, Bretton Woods Section - imb, mey, fon/tri, mni, ger, fis, FSI1 FederAl Office for Fupign Economic Affairs  87 FACSIMILE MESSAGE Annex VI TO: Mr. Manuel Penalver, Division Chief FACSIMILE NUMBER: 522-3124 DATE: 02/07/97 THIS IS PAGE ONE OF 4 SUBJECT: Comments on the Draft Performance Audit Report for Financial Sector Adjustment Credits I and II (Ciedits 191 1-GH and 2318-GH), Ghana REMARKS: SENDER: Taisei Matsuki THE OVERSEAS ECONOMIC COOPERATION FUND OF JAPAN 2100 Pennsylvania Ave, NW, Su le 535, Washington, DC 20037-3202 Tel: 202.463.7492 Fax: 202.463.7496 Internet: oecwsh@dgsys.com 02/07/97 16: 43 T202 463 7496 OECF OF JAPAN IO002/004 88 Annex VI February 7, 1997 Country Policy, Industry and Finance Division Operations Evaluation Report The World Bank Attention: Mr. Manuel Penalver, Division Chief Re: Comments on the Draft Performance Audit Report for Financial Sector Adjustment Credits I and I (Credits 1911-GH and 2318-GH), Ghana Dear Mr.Penalver, It is my great pleasure to comment on the above report. I would like to express our thanks for your patience. The Overseas Economic Cooperation Fund of Japan (OECF) financed the Financial Sector Adjustment Program (L/A No. GH-C1), as part of a cofinaicing program with the Bank's Financial Sector Adjustment Credit I(C/A No. 1911). The Loan Agr:ement was finalized on February 2, 1989 and the Loan Amount was 12,558 million Japanese Yen, which was divided in three tranches, V6,000 mil., V 3,500 mil. and Y3,058 mil.. The OECF disbursed the 1' tranche in March 1989 and the 2nd d3 tranches in December 1990 and February 1991 respeztively. We would like to comment on the Report as follows: 1. Program design for FSAC I, II a) Lack of the linkage between macro econond stabilization and tranche release The fact that the tranche releases were not linkel with macro-economic stabilization was a serious deficiency in the design of the program. As a resul, the programs did not achieve financial deepening nor did they improve the efficiency of the financial smctor, as originally expected. From past experiences with financial sector refo -ms in developing countries, the OECF has learned that financial sector reforms can be realized only af er a country's financial market is secure enough to implement sophisticated financial policies. Only after macro economic stabilization is achieved, should financial reforms be implemented in the latter phase of reforms, including trade and price reforms. This is because the speed of transition to a market base interest rate system depends on the macro economic situation, the distortion of the real sector, compet tive environment of the financial system, and the existence of the appropriate legal and institutional frimeworks. 02/07/97 16: 43 U202 463 7496 OECF OF JAPAN W03/004 89 Annex VI b) The timing of interest rates liberalization We agree with the point that "Exposing banks and the central bank to additional risks at a time when both were in financial distress and lacked the skills to manage the risks was inappropriate." (page 11, par. 10). Although it may have not been possible to assume it at the time of the appraisal (considering the inflationary expectation and devaluation expectation were raised after the fiscal shock in 1992), there was a possibility that the deficiency of the program regarding the timing of interest rates liberalization could have seriously damaged the financial sector development in Ghana as a whole. This lesson should be kept in mind when the timing of interest rates liberalization in other developing countries is examined. c) Contribution as B/P support The B/P support aspect of the programs should be discussed in the Report. A description should particularly be included of how the financial resour;es of the World Bank, Switzerland, and the OECF contributed, through the FSAC and the FSAC II, to the achievement of the economic growth which was the target of the PFP. 2. Donor Coordination (page 13, par. 18 etc.) It is regrettable that there are no descriptions of donor coordination. The topic of how timeliness and procedures of disbursement were coordinated among donors was an important issue in the discussions of the SPA meetings. The lack emphasis of these 3oints in the report does not accurately reflect the viewpoints of those involved in the meetings. For example, there is no statement about co-financing agencies for the FSAC I except 'Co-financing totaling US$ 105.7 million was mobilized from Japanese and Swiss donors' (Par. 3.1. page 35.). We are afraid that this will give the wrong impression that the World Bank regards other donors only as 'cash machines'. 3. Dialogue with stakeholders in Ghana (par.25, page 14 etc.) In the implementation stage of FSAC I and FSAC II, the discussions regarding the importance of the financial sector development might have been imufficient between the government of Ghana and the World Bank and other donors. This could account for persisting credit allocation and favored interest rates in the specific sectors, and for the unstable macro economic situation due to the fiscal shock in 1992. In addition, the lack of dialogue with strategic investors, might have been the reason for the unsatisfactory divestiture of financial institutions to strategic investors, apart from the influence of the deterioration of macro economic environment to tose investors. In summary of the last two points, the report does riot deal with the evaluation regarding the extent of the achievement for the target of FSAC 1: "to contrib ute to improve aid effectiveness in Ghana by acting as the focal point for aid coordination between donors and Ghana as the government strengthening its own planning and aid coordination ability". 4. Miscellaneous a) Expression about inflation rates (page 9, par. 3) 02/07/97 16:44 U202 463 7496 OECF OF JAPAN 19004/004 90 Annex VI It is true that the once stabilized inflation rate staled to rise again after 1993, after the increase in wages for public officers in 1992 and that it reached f 0 % (annual rate) in 1995. However, according to the Current Forecast in the CAS progress report (Aprl 23, 1994, Board Discussion), the annual inflation rate is expected to slow down rapidly from 44.9% in 1996 to 14.2% in 1997. It is necessary to examine the accuracy of the projection by referring to the latest data. However, if only the annual inflation rate (60%) in 1995 is mentioned in the report, it gives a misleading impression that the situation of inflation is only worsening. b) South-south cooperation (page 36, par. 3.7.) According to paragraph 3.7, two high-level advisers from the Reserve Bank of India were retained to provide formal as well as on-the-job training to staff working in bank supervision. This is an example of the actually proceeding south-south cooperation and should be explained in more details. If you have any questions about these conments, please contact me at 202-463-7492 or tmatsani@oe.org. I will be out of the office f -om February 11 until February 19. If you need assistance during my absence, you may contact Mr. Mitsuo Sakamoto, Senior Representative of OECF Washington Office. Sincerely yours, Taisci Matsui Representative OECF Washington Office tel:202-463-7492 fax:202-463-7496 email:tmats Ai@ocf.org  IMAGING Report No.: 16789 Type: PPAR