Rwanda - Financial sector review

Report No. 8934-RW Rwanda Financial Sector Review May 16, 1991 Industry and Energy Operations Division South Central and Indian Ocean Department Africa Region FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their off.cial duties. Its contents may not otherwise be disclosed without World Bank authorization.  ABBREVIATIONS BACAR Banque Continentale Africaine du Rwanda BCR Banque Commerciale du Rwanda BK Banque de Kigali BNR Banque Nationale du Rwanda BP Banques Populaires BRD Banque Rwandaise de DEveloppement C.R Caisse d'Epargne du Rwanda CHR Caisse Hypothecaire du Rwanda EEC European Economic Community IMF International Monetary Fund Nr(o Non-Governmental Organizations RE Rwandese Franc SONARWA Socikt4 Nationale d'Assurances du R,anda SORAS Societe Rwandaise d'Assurances SSA Sub-Saharan Africa SSE Small-Scale Enterprise UBP Union des Banques Populaires  FOR OMCIAL USE ONLY Preface This report presents the principal issues affecting the financial sector in Rwanda and the Impact on the sector of the ongoing macro-economic adjustment program. The report presents a financial sector reform program which will deepen and extend the measures introduced In the first phase of adjustment. The report is based on the findings of a World Bank mission that visited Rwanda in April/May 1990 consisting of ernminia Martinez (APSIE - Mission Chief), Gerard Caprio (CECFP - monetary policy), Roy Karaoglan (AFTEF - financial institutions), Mohamadou Diop (APSIE - UBP, CUR and BRD), Roland Tenconi (consultant - banking regulation and monetary policy), and Richard Scholl (Swiss Development Cooperation consultant - insurance). The report was discussed with the Government by a mission consisting of Herminia Martinez, Gerard Caprio and Roy Karaoglan, that visited Rwanda in February/March 1991. This mission also gathered information to update the Report. This document has a restricted distribution and may be used by recipients only in the performance of their officil duties. Its contents may not otherwise be disclosed without World Bank authorization.  RWANIDA FIACIAL ?COR TAS OF CONTEM EXECUTMVE SUMMARY I. TBE ECONOMY AND THE PINANCIAL SECTOR . . . . . . . . . . . . . A. BACKGROUND . . * . . . . . . . . . . . . . . . . . . . . . . 1 Recent Macroeconomic Developments . . . . . . . . . . . . . . 2 The Macroeconomic Reform Program . . . . . . . . . . . . . . 3 B. OVERVIEV OF THE FINANCIAL SECTOR . . . . . . . . . . . . . . . 4 C. ISSUES AFFECTING THE FIhANCIAL SECTOR . . . . . . . . . . . . . 6 XI. MONETARY AND CREDIV POLICY . . . . . 4. . . . . . . . . . . . . . 9 A. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . 9 B. PROBLEMS IN IMPLEMENTING POLICY . . . . . . . . . . . . . . .13 Financing of Governme- . . . . . . . . . . . . . . . . . . . .13 Priority Sector Credit . . . . . . . . . . . . . . . . . . .114 C. NEW INSTRUMENTS FOR MONETARY POLICY. .44. ...4. . . 17 The Adjustment Program . . . . . . . . . . . . . . . . . . .17 Alternative Means of Targeting . . . . . . . . . . . . . . .18 Liberalization of Interest Rates . . . . . . . . . . . . . .20 Ill. THE PERFORMANCE OF THE BANKING SECTOR . . . . . . . . . . . . . .22 A. THE COMMIERCIAL BANKS . . * . . . . . . . . . . . .. . .22 Background . . . . . . . . . . . . . . . . * * . . . . . .22 Operations: Decreasing Liquidity . . . . . . . . . . . . . .23 Deterioration of the Loan Portfolio Quality . . . . . . . . .25 Declining Profitability . . * . . . . . . . . . . . . .26 Efficiency and The Cost of Intermediation . . . . . . . . . .28 Recommendations . . . . . . . . . . . . . . . . . . . . . .30 B. THE BANQUES POPULAIRES . . . . . . . . . . . . . . . . . . . .30 Background. . . . . . . . . . . . . . . . . . . . . . . . . . 30 Growth and Diversification of Operations . . . . . . . . . .31 Deterioration of Loan Portfolio . . . . . . . . . . . . . . .33 Financial Performance and Subsidies . . . . . . . . . . . . .34 Conclusions . . . . . . . . . . . . . . . . . . . . . . . .36 C. THE CAISSE D'EPARGNE DU RWANDA . . . . . . . . . . . . . . . .37 Background . . . . . . . . . . . . . . . . . . . . . . . .37 Declining Operations . . . . . . . . . . . . . . . . . . * .37 Composition and Quality of Loan Portfolio . . . . . . . . . .37 Financess Operating Losses . . . . . . . . . . . . . . . . .38 The Future of the Caisse d'Epargne and Recommendations. . . .39  2- IV. FINANCING FOR DEVELOPMENT . . . . . . . . . . . . . . . . . . . .40 A . OVERVL EW . . . . . . . . . . . . . . . . i. . . . . . . . . .40 Be DEVELOPMENT BANKING .. . . * . * . . . . . . . . . . * 4 .242 Background . . . . . . . . . * . . . . . . . . . . . . . . .42 BRD's Operations and Portfolio . . ... . ......... . . . . .42 Financial Performance . . . . . . . . . . . . . . . . . . . .43 The Future of BRD . . . . . . . . . . . . . . . . . . . . . .44 C. HOUSING FINANCE . . . . . . . . . . . . . . . . . . . . . . . .45 V. REGULATION AND SUPERVISION OF FINANCIAL INSTITUTIONS . . . . . . .46 A. LEGAL AND REGULATORY FR OR R ................ 46 Prudential Rules. . . . . . . . . . . . . . . . . . . . . . .49 Solvency (Capital Adequacy) RAtio . . . . . . . . . . . . .49 Potential for Medium and Long-Term Use of Punds . . . . . .50 Restrictions on Real-Estate Holdings and Participations . .51 Loans to Directors, Managers or Staff . . . . . . . . . . .51 Credit Limits . . . . . . . . . . . . . . . . . . . . . . .S52 B. SUPERVISION OF FINANCIAL INSTITUTIONS . . . . . . . . . . . .52 C. LEGAL AND CONTRACTUAL ISSUES . . .. . . . . . . . . . ... .53 VT. A REFORM PROGRAM FOR THE FINANCIAL SECTOR . . . . . . . . . . . .55 A. REFORM OF MONETARY AND CREDIT POLICY . . . . . . . . . . . . .55 Monetary Targeting. . . . . . . . . . . . . . . . . . . . . .55 Interest Rates and Incentives for Priority Incentives . . . .56 New Financial Instruments S a . . . . . . . . . . . . . . . .56 B. REFORM OF FINANCIAL INSTITUTIONS S79*9* . .... .57 C. REFORM OF THE LEGAL AND REGULATORY FRAMEWORK . . . . . . . . .58 Laws and Regulations . . . . . . . . . . . . . . . . . . .S8 Supervision . . . . . . . . . . . . . . . . . . . . . . . .59 Coercial Law . . . . . . . . . . . . . . . . . * 9 .59 APPENDICES A. SHIFTING MONETARY CONTROL FROM DIRECT TO INDIRECT INSTRUMENTS . .60 B. FINANCING OF COFFEE . . . . . . . . . . . . . . . . . . . . . . . .65 C. THE COST OF INTERMEDIATION ............ .... ..67 D. CAISSE HYPOTHECAIRE DU RWANDA.. . C..... .. ... E. REGULATION OF FINANCIAL INSTITUTIONS . . . . . . . . . .... .71 F. THE INSURANCE SECTOR . . . . . . . . . . . . . . . . . . . . . . .73 STATISTICAL APPENDIX . . . . . . . . . . . . . . . . . . . . . . . . . . 75  - 1 . EXECUTIVE SUMMARY t1) To address key macroeconomic problems persisting since the late eighties, the Government has embarked on an economic reform program which has been designed with the assistance of the Bank and the IMF. The program, which is outlined in a Policy Framework Paper, stresses a greater reliance on market forces and the private sector, as well as a more export-oriented development approach. A coherent monetary policy and the development of a responsive and efficient financial system is a necessary complfment to the proposed reforms in the real economy. (ii) The financial sector in Rwanda consists of the central bank, the Banque Nationale du Rwanda (BNR), three commercial banks -- Banque Comerciale du Rwanda (BCR), Banque de Kigali (BD), and Banque Continentale Africaine du Rwanda (BACAR) -- and two savings institutions, the Caisse d'Epargne du Rwanda (CER) and the *banques populaires' tUBP). A development bank, Banque Rwandaise de D6veloppement (BRD), a housing finance institution, Caisse Hypoth4caire du Rwanda (CHR), and two insurance companies (SONARWA and SORAS) complete the financial sector. Caisse Sociale, the social security agency, is an important holder of long-term government obligations. In 1989, assets of the financial sector including the central bank totaled RP 65.6 billion, or 33.7 percent of GDP. (iii) Prior to 1987, the financial system was segmented, as commercial banks were restricted to the provision of short-term credit (loans of up to 1 year) and some medium-term loans (1 to 5 years), while BRD was only permitted to make long-term loans. Since then, coomercial banks have been allowed to make medium- and long-term loans up to a certain percentage of total credit. In fact, after allowing commercial banks into this type of lending, the authorities seemed to put less emphasis on maximum allowed exposures for medium- and long-term lending, and encouraged the growth of this business, especially lending to the Government. (iv) Reflecting in part the small size of the country, the financial system is rudimentary when compared with other countries at a similar level of development, especially outside Sub-Saharan Africa. Financial depth figures (IM2/GDP of 17.3 percent) put Rwanda below the average performance for low-income countries and for Africa as a whole (M2tGDP of 23.8 and 22.3 percent, respectively). Financial deepening has progressed relatively slowly, especially in the 1980.. This reflects in part government use of the financial system primarily to direct credit and to generate tax revenue. The importance given to directed credit resulted in the Government not controlLing credit growth, which increased the underlying inflation. Also affecting financial deepening are the low income level, the low level of after-tax bank profits, which has discouraged additional entries into banking, and the underdevelopment of the 'banking habit', as only the 'banques populaires* (and to a lesser degree, the insolvent CER) have a branch network covering the rural areas.  - ii- MoAetary and Credit Policy (v) Monetary policy has been part of an overall financial sector policy designed to direct credit to certain favored sectors of the economy at a subsidized price. Consistent with this objective, until recently, the authorities did not have an aggregate monetary target; instead, they focussed on trying to steer credit towards certain sectors regarded as priority. BN used a system of prior approval of credits above a certain sum to help monitor this system of directed credit. (vi) Until 1986, the system worked well at restraining credit growth and ensuring relative price stability. Credit growth and inflation were below those in other African countries. In recent years, Government rec3urse to the central bank as well as credit to the private sector have grown rapidly despite a decline in real output. As might be expected in an open economy with a fixed exchange rate, much of this credit growth -- about 60 percent -- was offset by a reduction in net foreign assets, so that the increase in the broad money supply was much smaller. Although the official price statistics do not show a significant deterioration in inflation performance, anecdotal evidence suggests that actual inflation was above the official indicators, as reflected in the black market exchange rate and in the rate of financial disintermediation. (vii) A key problem for implementing monetary policy has been the banking sector financing of the budget deficit. Over the last three years, total public debt to GDP increased by 17.5 percentage points, and the domestic market took up 40 percent of this increment. The central bank and other Institutions absorbed the major part of the increase. Credit to the Government by the BNR in fact increased at an annual rate of almost 34 percent during the 1986-90 period, leaving the stock of central bank credit to the Government at the end of the period about 2.5 times its level at end- 1986. Much of the increase in central bank credit to the Government occurred In the form of exceptional advances. (viii) Interest rates played a minor role in the conduct of monetary policy. Special rates were applied to priority activities. Until the reforms initiated in 1990, priority sectors other than the storage and export of coffee received favorable interest rates of a maximum of 9 percent, without any quantity rationing, while the maimum interest rate for non- priority sectors was 12 percent. Coffee storage and exports, carried out by three large companies (two partly owned by the Government), have been financed using a complex system whereby banks are reimbursed at a flat fee per kg of coffee financed, irrespective of the length of financing. This system has lowered the implicit interest rate well below that for other priority activities (the implicit interest rate was around 3.5 percent in 1990). The banks were not able to calculate the interest rate precisely ex- ante since their remuneration consisted of a flat fee per kilogram of coffee. (ix) With relatively unfavorable financing terms, non-priority credit declined steadily from about two-thirds of commercial bank credit to the economy to 25 percent in 1989, giving the BNR an exceedingly small and diminishing handle on credit growth. In addition to enjoying lower lending  - iji - rates and no credit ceilings, the priority sectors also benofitted from the availability of favorable refinancing at the central bank. (z) Beyond its impact on the implementation of monetary policy, the existence of directed credit programs, particularly on such a wide scale, has been recognized as leading to a less than efficient allocation of resources. In Rwanda, the diversification of the economy probably has been retarded by the continued availability of subsidized credit to coffee. (xi) As part of its reform program, the SIR intends to eliminate differential interest rates and to set reserve-requirements for banks. In addition, it has set an aggregate credit target which would be monitored through bank-by-bank credit ceilings. BN3 has moved towards the elimination of differential interest rates by setting a minimum rate for one-year deposits, a maximum lending rate, and a single rediscount rate. These rates have been established at 12, 19 and 14 percent, respectively. (xii) Bank by bank credit ceilings is not an effective method for controlling credit growth in the short term. Host industrialized and a growing number of developing countries have abandoned direct controls on credit and interest rates and instead rely on indirect monetary policy Instruments. Bank-by-bank credit ceilings have been found to distort competition by penalizing the more dynamic institutions and to discourage resource mobilization. (ziii) With a relatively undeveloped financial sector, Rwanda is not in a position to move quickly to a reliance on indirect methods to implement monetary policy such as open market operations. However, the country can move away from bank-by-bank credit ceilings, or minimsize their distortionary Impact, and still satisfy a monetary aggregate target by using a system based on the rediscount rate, reserve requirements and marginal reserve ratios. It is recommended that reserve requirements above a minimum that banks could keep for liquidity purposes be remunerated. Since reserve requirements would be set low, they are not likely to have a significant impact on the cost of intermediation and may even raise some revenue, to the extent that banks borrow more from the BNR. Within a year, the reserve requirement could be raised to 5-10 percent, at which level it should absorb sufficient liquidity to encourage both interbank activity and greater recourse to central bank borrowing. The Banking Sector (xlV) European banking institutions have played a major role in the development of coimercial banking in Rwanda. During the past 25 years, commercial banking has been dominated by two banks: the ICR, which was established by Banque Bruxzlles Lambert, and DR, established by Belgolaise. BCR accounts for 51.7 percent of coumercial bank assets and DX for 31.3 percent. The third bank, which accounts for the remaining 17 percent of total assets, BACAB, was set-up by Bsanque Continentale du Luxembourg in 1983. The commercial banks are generally veil managed, although there are differonces among them. However, banks are undercapitalized and under- provisioned. This, combined with their excessive concentration on a few clients, makes the bairing system quite vulnerable.  - iv - (xv) Reflecting the general econamic situation, the non-performing portfolio of commercial banks increased from 5.3 percent of their outstanding loan portfolios at the end of 1987 to 6.0 percent at the end of 1989 and 11.6 percent at the end of 1990. To take into account the deterioration of their portfolios, banks increased their provisions for loan losses over the past two years, but the ratio of loan loss provisions to the non-performing portfolio decreased from 97.8 percent to 74.3 percent during the period. The problem portfolio is probably substantially underestimated because bank management delays classifying loans as non-performing. Consequently, provisions are most likely inadequate. Xri order to minimize the risk of facing capital inadequacy problems in the foreseeable future, banks should be induced to increase their provisions by providing them with tax incentives. This is all the more important since, in Rwada, the capitalization of banks (ratio of average shareholders' funds to average total assets) is not adequate by international standatds. (xvi) The cost of intermediation, or gross earnings margin, fell substantially over the past four years, from 7.1 percent in 1987 to 6.2 in 1989 and 5.9 in 1990. The decline in the inter-ediation cost is due primarily to a substantial decline in profits. dperat.ing costs, in particular staff costs, are moderate. Given the existing cost structure, there is little scope for reducing the cost of intermed'ation, except by reducing the taxes paid. A tax reduction would need to be accompanied by measures to increase competition in the commercial banking system so as to reduce bank charges. (xvii) Commercial banks are profitable, although profitability has declined substantially over the past four years. Returns before tax on shareholders' funds averaged 30.4 percent in 1987-89, well above International averages, while those after tax (8.7 percent on shareholders' funds) are below international averages. Bank profits, furthermore, have declined precipitously over the past four years, from 13.4 percent of equity in 1987 to 7.3 in 1990. Commercial banks are heavily taxed. In 1987-90, taxes amounted to almost 20 percent of costs and 1.7 percent of average total assets. (xviii) Other financial Institutions could help increase competition with the commercial banks, although most are facing serious difficulties. The CER, the government savings bank, has been de facto bankrupt for most of its existence. One of the key issues to be addressed over the next few years in the best way to resolve CE's problems. The Banques Populaires have done well at mobilizing savings, although the scheme may be unsustainable because it relies on subsidies from the Rwandese and Swiss Governments. BRD, created in 1967, is the main source of investment financing in Rwanda. It has worked well as a development bank, although it is facing portfolio problems as a result of the macroeconomic environment, and there is a need to detenmine the desirable areas of its future interventions.  Develooment Financing Wmir) Before 1987, the controls imposed by the central bank on short-term credit to the economy did not present a constraint to the development of large enterprises in the modern sector. Term credit, especially that over 5 years, was more limited and BRD held over 70 pertLit of the RP 3 billion in outstanding long-term credit. In 1987-89, commercial banks increased their medium- and long-term loan portfolio by 50 percent. BRD continued to represent over three quarters of the long-term credit outstanding and the bulk of the credit for investment. BRD has a solid financial position and uncharacteristically conservative operating policies. (xz) Small-scale enterprises (S$Es) have had difficulty finding financlng for their projects. Short-term credit is provided principally by the 'banques populaires' and to a limited extent by the conmercial banks, which have lent primarily for enterprises in commerce, transport and construction. Institutional long-term credit for small enterprises and cooperatives remains the domain of BRD. As in many developing countries, there is evidence that some SSEs obtain financing from the informal financial sector. Th3re is some indication that the traditional informal financial sector is not well developed in Rwanda, although thers axa ousineSsmen that do provide credit outside the formal financial system. (xxi) The commercial banks finance mostly short-term operations, but there is evidence that this is less so than in other African countries. Banks have been reluctant to lend to SSEs (let alone the small farmere) because of the high risks involved, and because of their lack of appraisal and supervisory capacity. Most small enterprises are not known to the banks, and are unable to meet their high collateral requirements. Also affecting the willingness of com ercial banks to lend to lesser known, and hence riskier, clients are the interest rate ceilings, which do not allow them to charge a premium for risk. The ceilings are a major problem for the UBP, which rightly claims that if it is to continue lending to its low-income clientele, it needs a higher interest margin. Banking Regulation and Supervision (xxii) The critical situation of financial institutions reflects in part problems particular to each institution, and the deterioration of the macroeconomic enviranment. It also reflects shortcomings in the regulatory framework and in the supervision of financial institutions. The regulations which guide the operations of financial institutions are incomplete, and those that exist are not enforced. Responsibility for the supervision of financial institutions rests with the BNR. BNR has not given priority to this task and the department in charge of it is grossly understaffed. A Reform Program for the Pitancial Sector (xxiii) The financial sector will play a crucial role in ensuring the expansion of productive activities and the growth of the private sector. Building on existing strengths, the Government can put in place a policy and Institutional framework which will ensure that the financial sector improves the range and coverage of the services provided in the years ahead. The  - vi - principal elements of a reform program are outlined below under the headings of