Structural Issues in the Kenyan Financial System: Improving Competition and Access Thorsten Beck and Michael Fuchs Abstract: Although by regional standards, Kenya's financial system is relatively well developed and diversified, major structural impediments prevent it from reaching its full potential. Cross- country comparisons, however, show the importance of a well-developed financial sector for long-term economic growth and poverty alleviation. Experience from other developing economies has shown the detrimental effect of government ownership and the positive impact that foreign bank ownership can have on the development of a market-based financial system. Analyzing and decomposing the high interest rate spreads and margins in Kenya helps identify structural impediments that drive the high cost of and low access to financial services. The limited information sharing on debtors, deficiencies in the legal and judicial system, the limited number of strong and reputable banks and non-transparency and uncertainty in the banking market are major impediments to the development of Kenya's financial system, to reducing spreads and to widening access. World Bank Policy Research Working Paper 3363, July 2004 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. Thorsten Beck is a senior economist in the Development Research Group of the World Bank; Michael Fuchs is a lead financial economist in the Financial Sector Unit, Africa Region of the World Bank. We would like to thank Meghana Ayyagari for excellent research assistance, Duillo Pighi for editorial assistance and Ann Rennie and participants at the Finance Forum in Mombasa for helpful comments and suggestions. I. Kenya's Financial System in International Comparison Although by regional standards, Kenya's financial system is relatively well developed and diversified, major structural impediments prevent it from reaching its full potential. Table 1 shows that the Kenyan financial system enjoys higher levels of credit channeled to the private sector and higher deposits in financial institutions than other Sub-Saharan African and low- income countries; however, as one might expect and for the reasons explored in this paper, there is a wide gap compared with OECD countries. Table 1 also shows that Kenya has a lower concentration ratio than not only other countries in the region, but also the average OECD country. A complete list of the Kenyan banks, their market share and ownership structure is provided in Annex I. Table 1: Kenya's Banking System in International Comparison Private Credit/GDP Financial Deposits/GDP Bank Concentration Kenya 25.6 33.2 39.3 Sub-Saharan Africa 18.7 23.5 77.0 Low-income countries 16.0 19.5 74.2 OECD countries 92.6 113.5 47.9 Sources: Private Credit/GDP is total claims of financial institutions on the domestic private non-financial sector as share of GDP. Financial deposits/GDP is total deposits in financial institutions as share of GDP. Bank Concentration is the share of assets of the largest three banks in total banking sector assets. Source: Update of financial structure database, World Bank. All data are for 2001. Cross-country comparisons have shown the importance of a well-developed financial sector for long-term economic growth and poverty alleviation. Countries with better developed banking systems and capital markets enjoy higher growth rates; and it is the poorest segments of society that stand to gain most (Beck, Levine and Loayza, 2000; Beck, Demirgüç-Kunt and Levine, 2004; for an overview, see World Bank, 2001). Figure 1 shows the empirical relationship between credit to the private sector as share of GDP and GDP per capita growth for a sample of 52 developing and developed countries over the period 1960 to 1999. Figure 2 shows the empirical relationship between credit to the private sector as share of GDP and income growth of the poorest quintile of each country's population for the same sample and the same period. Both exercises control for other factors that can explain cross-country variation in the growth of overall GDP per capita and the income of the poor. The positive relationships in Figures 1 and 2 do not just reflect correlations between financial development and income growth; well- developed financial intermediaries and markets exert a causal impact on economic development 1 and poverty alleviation. Figure 1: Financial Intermediary Development and GDP per Capita Growth .04 .02 growth 0 capita per -.02 GDP -.04 -2 -1 0 1 Private Credit/GDP This partial scatter graph plots the residual of a regression of GDP per capita growth on initial GDP per capita, school attainment, inflation, government consumption and trade openness versus the residual of a regression of Private Credit/GDP on the same regressors. For details, see Beck, Demirgüç-Kunt and Levine (2004). Figure 2: Financial Intermediary Development and Income Growth of the Poor .05 r Poo e th 0 of th owrG -.05 Income -.1 -2 -1 0 1 Private Credit/GDP This partial scatter graph plots the residual of a regression of income growth of the poorest income quintile on initial GDP per capita, school attainment, inflation, government consumption and trade openness versus the residual of a regression of Private Credit/GDP on the same regressors. For details, see Beck, Demirgüç-Kunt and Levine (2004). 2 Well-developed financial institutions and markets foster economic development by improving the allocation of society's scarce resources rather than by facilitating faster capital accumulation or increased savings (Beck, Levine and Loayza, 2000). A well-developed financial system widens access to external finance and channels resources to the sectors that need them most (Rajan and Zingales, 1998; Demirgüç-Kunt and Maksimovic, 1998; Wurgler, 2000). Finally, effective financial institutions and markets can help economies cope better with exogenous shocks such as terms of trade volatility and move them away from natural resource based development (Beck, 2002; Raddatz, 2003). Financial markets are different from other markets in that they involve the intertemporal exchange of monetary resources: money today for the promise of money tomorrow. Lack of information about the counterpart, lack of monitoring and enforcement tools vis-à-vis the counterpart and uncertainty about the value of money tomorrow can hamper the efficiency of this exchange and thus the efficient functioning of financial markets. Governments have an important role to play in reducing these frictions and making financial markets work efficiently. Transparency ­ higher accounting and disclosure standards for both borrowers and banks ­ and market discipline reduce information and monitoring problems. Effective protection of outside investors, both bank creditors and minority shareholders, as well as contract enforcement and property right protection in general help reduce monitoring and enforcement problems. Monetary stability is a pre-condition for market participants to be willing to engage in long-term financial contracts. While governments have an important role in creating an enabling environment for efficient financial markets, experience in both developed and developing countries has shown the negative impact of government ownership on financial development. Formerly seen as a necessary tool to foster financial and economic development, government-owned banks have fallen far short of delivering on their promises and have prevented especially developing economies from building market-based financial systems. Countries with a higher share of government-owned banks have experienced lower GDP per capita growth, less effective resource allocation, lower and more concentrated access to credit and higher interest rate spreads (La Porta, Lopez-de-Silanes and Shleifer, 2002; Barth, Caprio and Levine, 2004). While government ownership in banking can impede financial development, poorly 3 designed privatization in a weak institutional environment can make things worse. Experiences in many developing countries have shown that privatization to connected insiders and/or privatization in an environment where neither market participants nor bank supervisors are able to properly monitor and discipline banks, can result in compromised government policies, fraud, looting and even a banking crisis. Timid and slow restructuring and delays in the privatization process, however, can also be costly, prolonging the day of reckoning, and result in larger costs in the form of a high share of non-performing loans, repeated need for government-financed recapitalization and, not least, an extended period of reduced efficiency and competitiveness of financial intermediation. The experience in many Central and Eastern European countries has shown the importance of coordinating the privatization processes in the financial and non- financial sector (Caprio, 1995). Experience in many developing countries has also shown that partial privatization is not sufficient to resolve governance problems (Beck, Cull and Jerome, 2004). There may not be a single optimal privatization method, but experience from other countries has provided some lessons on how to gain the greatest potential benefit from the privatization process while minimizing the risks. To encourage private investment in banking and growth in financial intermediation, it is important to send credible signals to the market and the public that privatization will be undertaken over a defined time period, while significant regulatory improvements are undertaken to ensure that the privatized bank will comply with regulatory requirements and be subject to market and supervisory discipline. Complete transparency in the privatization process and applying strict fit and proper requirements to the new private owners are of utmost importance. Cross-country comparisons have shown the benefits of foreign bank ownership for developing countries, while providing little evidence for adverse consequences on access to or cost of financial services (Claessens and Jansen, 2000, Clarke et al., 2003). In addition to investment in the capitalization of financial institutions foreign banks usually bring with them better know-how and technical capacity, which then spills over to the rest of the banking system. They impose competitive pressure on domestic banks, increasing efficiency of financial intermediation. They provide more stability to the financial system, being able to draw on liquidity resources of their parent banks. Finally, there is even evidence that access to credit is higher in countries with a larger share of foreign-owned banks (Clarke, Cull and Martinez-Peria, 4 2002). These results suggest that even in those cases where foreign banks have focused more narrowly on the top end of the market rather than providing financial services to small borrowers, they can force domestic banks to re-focus on the small borrower segments. While endorsing the benefits associated with foreign ownership, experience with privatization of state-owned banks in Africa has demonstrated that careful preparation and execution of the privatization process ­ even when involving reputable foreign banks ­ is of the utmost importance. Whether in the form of privatization of government-owned banks to foreign entities or new entry by foreign banks, experience in developing countries has shown that foreign bank entry is not a panacea and cannot substitute for an effective regulatory and supervisory framework. As is generally the case, sale of domestic banking assets needs to be subject to stringent due diligence regarding the "fit and proper" status of prospective foreign owners. Further, in a growing number of cases international banks are committed to a developmentally- oriented focus ­ in some cases working in association with development partners (which can provide more "patient capital" looking to longer-term returns than associated with private capital) ­ and are re-orienting their business model toward the needs of developing countries in maintaining and improving access to financial services, especially in rural and remote regions. Encouraging the entry of developmentally-oriented but financially sound banks can contribute to enhancing access to credit by reducing the segmentation between "mainstream" finance provided by banks and the market for smaller credit, which in recent years has been developed on a commercial, but as yet much smaller basis by smaller banks, credit cooperatives (SACCOs) and micro-finance institutions (MFIs). 2. Interest Spreads and Margins in International Comparison High interest spreads and limited depth and breadth of the financial systems are closely related to each other. Countries with higher interest rate spreads have lower levels of credit to the private sector as share of GDP and deposits in the financial system as share of GDP (Figure 3). High interest spreads and margins and limited depth and breadth of financial services are the result of underlying deficiencies and impediments in the financial systems; in order to increase access to financial services and reduce spreads and margins, these underlying causes have to be addressed. 5 Figure 3: Financial Intermediary Development and Interest Spreads 40 ade 30 Spr 20 Interest10 0 0 .5 1 1.5 Private Credit/GDP This graph plots Private Credit/GDP and interest spreads for a sample of 83 countries for 2001. Data are from International Financial Statistics and the Financial Structure Database update from the World Bank. Trying to reduce spreads by regulating and controlling interest rates carries high risks of counterproductive effects. Administratively limiting lending rates would most likely result in (i) credit rationing and (ii) an increase in hidden charges to compensate for lost interest revenue. Both will hurt smaller borrowers especially hard, resulting in less access to credit. Administratively fixing deposit rates at a minimum level will, apart from the negative impact on banks' soundness, most likely result in banks (i) raising minimum deposits and (ii) charging additional fees to compensate for the higher cost. Again, this will hurt small depositors more than large depositors. Similarly, trying to expand access to financial services through government programs that provide subsidized credit results carries the risks of (i) crowding out private providers, (ii) poor targeting, i.e. providing large subsidies to few borrowers who might have access to private financial services, (iii) undermining credit discipline in the case of politically motivated debt forgiveness. Depending on the chosen delivery mechanism for such subsidies their cost to society can be much higher than the direct cost of subsidy. For example, a recent study (undertaken as follow-up to the Financial Sector Assessment Program in early 2004) 6 found that the Kenyan Development Finance Institutions (DFIs) not only accumulated very sizeable non-performing assets (equivalent to about 90 percent of their loan portfolios), but also operated with very high overheads. Table 2: Interest Rates, Spreads, and Margins in International Comparison Real lending rate Real deposit rate Interest spread Interest margin Kenya 16.5 3.5 13.0 9.2 Sub-Saharan Africa 9.9 -1.5 11.5 8.1 Of which: Uganda 19.4 5.9 13.5 12.7 Tanzania 12.0 -1.2 13.1 7.5 Low-income countries 10.8 -1.6 12.4 7.8 OECD countries 4.6 0.5 4.1 3.6 Sources: Real lending (deposit) interest rates are the difference between average lending (deposit) interest rates for 2002 and the log of CPI inflation for 2002. The interest spread is the difference between deposit and lending rates. The net interest margins is the net interest revenue relative to total earning assets; data are from the World Bank Financial Structure Database based on raw data from Bankscope, for 2001. While Kenya has similarly high interest rate spreads and margins as other countries in the region, it has substantially higher spreads and margins than OECD countries (Table 2). This underlines the importance of analyzing the structural deficiencies that drive the high spreads and the limited access. It also allows us to draw on the experience of other countries when designing policies to reduce spreads and increase access.1 Decomposition of the interest spreads points to overhead costs and the profit margin as the most important component of the interest rate spread in Kenya (Table 3).2 State-banks have the highest spread, followed by foreign-owned banks and privately owned Kenyan banks. While this might point to inefficiencies in foreign banks relative to privately owned domestic banks, we will discuss below that it is the reputational contagion associated with small private domestic banks that obliges the smaller domestic banks to pay relatively high interest rates on funds deposited with them and allows foreign banks to enjoy higher profit margins and therefore higher spreads. Overhead costs and loan loss provisions constitute two-thirds of government-owned banks' spread, while overhead costs and the profit margin constitute two-thirds of the spread of privately owned banks. While the profit margin seems relatively high, we note that this is only the profit on lending, the most risky line of business for banks. The overall profitability for banks 1The Brazilian Central Bank has undertaken significant analytical work on interest rate spreads: www.bcb.gov.br 2Annex II describes the decomposition in more detail. 7 is significantly lower, as indicated by the return on assets, which is of a level comparable to other banking markets. Table 3. Kenya: Decomposition of Interest Spreads All banks State-owned banks Domestic private Foreign banks Overhead cost 5.6 4.4 5.3 6.6 Loan loss provisions 2.5 4.9 1.5 1.8 Reserve requirements 0.3 0.3 0.4 0.2 Tax 1.9 2.2 1.6 2.1 Profit margin 4.5 5.2 3.7 4.9 Total spread 14.9 16.9 12.5 15.5 Return on Assets (after tax) 1.4 -0.4 1.0 3.0 Source: Authors' calculations using data from the CBK. All data are for 2002 An analysis of the overhead costs shows that they are driven by wage costs, which constitute 50 percent of total overhead costs. Other factors relating to the costs of financial service provision in the local market include fraud, security costs, the inefficient payment system and a heavy regulatory burden, as illustrated by high reporting requirements, the annual re- licensing process, and licensing procedures for the opening and closing of branches. Compared to banks in other Sub-Saharan African countries and other emerging countries, Kenyan banks seem to be overstaffed and their employees less productive (Table 4). Kenyan banks have more than three times as many employees for a given amount of assets, loans and deposits than other banks in emerging countries, and the average Kenyan bank employee earns only half of the net interest revenue as the average employee in emerging markets. Table 4: Bank Productivity in International Comparison Net interest/employee Assets/employee Loans/employee Deposits/employee Kenya 36 581 295 458 Sub-Saharan Africa 49 1,073 505 742 Emerging markets 60 2,040 911 1,620 Source: Authors' calculations using data from Bankscope. All data are for 2002 and in thousands USD. However, there are large differences in productivity across different ownership groups of Kenyan banks. Employees in state-owned banks earn only half of the net interest revenue of employees in foreign-owned banks. State-owned banks have twice as many employees relative to their assets, loans and deposits as foreign-owned banks. The higher productivity of foreign- 8 owned banks compensates for the higher wage costs of these banks compared to domestic banks. Private domestic banks are less productive and more overstaffed than foreign-owned banks, but more productive and less overstaffed than state-owned banks. This disparity across ownership groups indicates significant potential gains from increased competition and the resulting productivity improvements. Table 5. Bank Productivity across Different Kenyan Bank Groups Net interest/employee Assets/employee Loans/employee Deposits/employee State-owned banks 23 303 187 222 Private domestic banks 31 577 317 447 Foreign banks 50 770 349 625 Source: Authors' calculations using data from CBK. All data are for 2002 and in thousands USD. Interest margins and overhead costs are almost twice as large in Kenya as in other countries and empirical findings show that two-thirds of the difference can be explained by the deficient Kenyan legal and institutional framework and the relatively smaller size of Kenyan banks (Table 6). To compare net interest margins and overhead costs of Kenyan banks with net interest margins and overhead costs in other countries, we use the results of a recent study of the determinants of net interest margins by Demirgüç-Kunt, Laeven and Levine (2004). This study used bank-level data from Bankscope for 72 countries to explore the empirical relationship between bank and country characteristics and net interest margins and overhead costs relative to total assets. Unlike in the analysis of interest spreads, we do not decompose the interest margin or the overhead costs, but rather try to relate them to specific bank and country characteristics. While the study by Demirgüç-Kunt, Laeven and Levine (2004) uses data over 1995-99 and has a limited sample of banks for each country, we use data for 38 Kenyan banks for the year 2002, thus covering 98 percent of the banking system.3 We multiply the coefficient estimates from two regressions in that paper (Table 8, column 3 and Table 11, column 3) with the difference between values of the respective variables for Kenya and the mean value for all countries in the study to identify the factors that most contribute to the high interest margins and overhead costs in Kenya. 3Due to the different nature of their business, we excluded IDB, DBK, and K-Rep Bank. The difference between the interest margin numbers for Kenya in Table 2 and Table 6 is due to the different data sources used (cross-country database vs. national data sources). 9 Table 6. Net Interest Margins and Overhead Costs in International Comparison Interest margin Overhead costs Kenya 7.0 5.9 World-wide average 3.6 3.0 Difference 3.4 2.9 Of which: Property right protection 1.4 0.8 Bank size 0.9 0.7 Other bank characteristics -0.3 0.5 Other country characteristics 0.1 0.0 Kenya residual 1.2 0.8 Source: Authors' calculations using data and results from Demirgüç-Kunt, Laeven and Levine (2004) and data from CBK. This cross-country comparison shows that by far the largest factor explaining the higher margins in Kenya is the comparably deficient legal and institutional framework ­ according to the regressions model accounting for approximately 1.4 percentage points of the higher net interest margins. This factor also provides the most important explanation for the higher overheads costs ­ accounting for 0.8 percentage points of the costs. The relative smaller size ­ thus the lack of scale economies ­ of Kenyan banks explains 0.9 percentage points of the higher net interest margin and 0.7 percentage points of the higher overhead costs. The lack of a sound legal and institutional environment and the small size of Kenyan banks are thus the two most important factors in explaining why net interest margins and overhead costs are almost twice as high in Kenya as in the rest of the world. Overall, this analysis of net interest margins and overhead costs confirms the decomposition analysis of interest rate spreads and international productivity comparisons. The deficient legal and institutional framework can explain the high loan loss provisions of Kenyan banks, but also the relatively high profit margin, as we will discuss in the next section. The deficient legal and institutional framework ­ fraud and insecurity ­ and the relatively small size of Kenyan banks can explain high overhead costs and the low productivity of Kenyan banks. 3. Is the Kenyan Banking Market Competitive? While loan loss provisions and overhead costs explain a large part of interest rate spreads, there is still a relatively high profit margin, which raises the question of the competitiveness of the Kenyan banking system. Traditional measures of concentration in the banking industry, such 10 as the three-bank concentration ratio, the number of banks, or a Herfindahl index would suggest a relatively high degree of competitiveness of the Kenyan banking system.4 Cross-country comparisons, however, have shown that the competitiveness of a banking market is not necessarily linked to market structure measures such as the Herfindahl index, the number of banks or concentration ratios (Claessens and Laeven, 2004). Contestability of the market, the threat of entry, can be a more important determinant of bank behavior. Entry barriers and competition from other financial institutions (such as insurance companies, large credit cooperatives, and capital markets) can play an important role in determining banking system competitiveness. Ownership structure can be important for the degree of competition, since banks with different ownership often have different mandates and different clienteles. Anecdotal evidence suggests that most customers in Kenya below the top tier of corporate and wealthy borrowers face a non-competitive banking market and are often effectively tied to one bank, with very high switching costs. While recent changes in the interest rate structure ­ and the sharp fall in money market rates in particular ­ have resulted in more aggressive marketing by banks, there are structural impediments to a more competitive banking market, which relate primarily to lack of information on both sides of the market and segmentation of the banking market. 4. Policies to Foster Competition, Reduce Spreads and Widen Access As discussed above, the problems of high spreads and limited access cannot be helped by short-term government-directed solutions that work against market forces. Rather, the Government should focus on implementing policy reforms to alleviate market, institutional and political barriers to competition. In the following, we discuss four areas, in which policy reforms can have a major impact in reducing such barriers and thus increasing the effectiveness and competitiveness of financial intermediation, thereby reducing spreads and widening access. First, lack of information sharing on debtors increases banks' credit risk and reduces the competitiveness of the banking system. The absence of reliable information on potential borrowers increases the adverse selection risk for banks resulting in higher credit risk and loan loss provisions, which in turn raises interest rate spreads. The inability of borrowers to build up a 4Using data for May 2003, the Herfindahl is 947 for banks' assets and 1,045 when using banks' loans. This corresponds to a system with 11 banks of equal size. 11 positive credit history prevents them from accessing bank finance and increases the costs of switching lenders, effectively tying borrowers to one bank. The resulting rents increase the profit margin of banks and thus interest rate spreads. A well functioning credit registry with a firm legal basis including provisions for data and consumer protection, which offers both negative and positive information on borrowers, and a broader user group can reduce credit risk for banks and increase access to credit. Sharing of positive information will especially benefit small borrowers, as it will allow them to establish reputation with small loans and improve their chances to increase their borrowings as their business grows. In addition to information sharing among institutions governed under the Banking Act, it is recommended to also include building societies, microfinance institutions (MFIs) and SACCOs in the information sharing mechanisms. This may allow micro and small borrowers to graduate from microfinance to bank finance as their business develops. Information sharing among all finance providers would be extremely beneficial for reducing segmentation and increasing competition within the financial system. Second, deficiencies in the legal and institutional framework increase credit risk for banks and limit the scope of assets available as acceptable security, effectively tying debtors to one bank. Only few borrowers have access to unsecured lending due to deficiencies in contractual enforcement procedures. Problems in perfecting and registering security increase credit risk for lenders and result in higher loan loss provisions. The high costs of taking security (stamp duties and legal fees) and the chaotic state of property registries limit the scope of assets available as acceptable security and constitute an effective barrier for borrowers who might consider switching lenders. Strengthening the legal framework will be crucial to improving the lending environment and reducing interest rate spreads. Legal reforms should include (i) reform of the Companies Registry to simplify registry procedures and allow for speedy and accurate information sharing of corporate information and data,5 (ii) reform and consolidation of the land registration framework to allow speedier searches as well as establishing more efficient mechanisms for registering and enforcing security interests over land and other real estate, (iii) strengthening and simplifying debt collection processes, including the introduction of procedures such as small claims procedures, summary procedures for uncontested debt, and alternative dispute resolution 5This might include creating a separate registry for charges over movable assets. 12 procedures, and (iv) institutional reform and training of judges. The combined effect of these measures would be to reduce credit risk for banks and help borrowers that would like to switch lenders. Third, the limited number of large and strong banks constrains the competitiveness of the Kenyan banking system and has resulted in market segmentation due to the limited movement of funds and clients across different subsectors, such as banks and MFIs. There is a reputational bias against small banks, since many small banks are fragile and weak. Depositors therefore prefer large foreign or state-owned banks. Further, many small banks, be they strong or weak, operate in niche markets. As state-owned banks are not primarily profit-oriented, competition is effectively limited to large foreign banks. This goes to explain the higher profit margins and therefore interest spreads of foreign-owned banks relative to domestic private banks, as discussed above. Regulatory and supervisory efforts to strengthen the banking system can help reduce the interest rate spreads and foster competition. Regulatory and supervisory measures to strengthen the Kenyan banking system should include: (i) divesture of the Kenyan state from the banks in which it holds a stake, (ii) rapid intervention into weak and failing banks, and (iii) granting autonomy with accountability to the CBK as bank regulatory and supervisory authority. These steps will enhance competition between a larger number of privately-owned and profit-oriented medium-sized and large banks. Action along these lines will also help reduce the reputational bias against small banks. While the outcome might be a smaller number of banks, this can result in more competition in the banking market. Finally, uncertainty in the policy environment concerning the control or regulation of interest rates and related bank fees has an adverse effect on the predictability and transparency of bank tariffs. Legal provisions of the Central Bank Amendment Act 2000 and Section 44 of the Banking Act stipulate that prior approval of the Minister of Finance is required for changes in the rate of banking or other charges, thereby creating uncertainty about the Government's commitment to market-based financial policies. Eliminating all references to controls on interest rates and charges and launching a deliberate and transparent divestiture program for the Government's remaining stakes in the banking systems are two crucial steps towards improving certainty in the policy environment. Both steps will underline the resolution of the Government 13 to move towards a market-based financial system and redefine the role of the Government as enabler rather than actor in the financial system. Rather than controlling or regulating interest rates, the role of the Government should be to foster transparency in the market place. More transparency in the structure of interest rates across banks, both according to product and maturity, can enhance the competitiveness of the banking system by allowing customers to make a better-informed decision. However, the improvement in competition has to be weighted against a potential increase in moral hazard risks if weak banks succeed to attract additional liquidity through higher (even well publicized) deposit interest rates. Overall, however, the benefits from improved competition are likely to outweigh the moral hazard risk, which can be detained by a strengthening of bank supervision. By divesting its remaining ownership stakes in the banking industry the Government will demonstrate its commitment not only to lowering spreads and increasing competition, but also to prudent fiscal management and good governance. Given the fiscal burdens and the record of political interference and corruption associated with the Government's ownership of banks, considerable non-tangible benefits will follow as a result of the divestiture program. These benefits will take the form of greater confidence in the Government's fiscal management and commitment to good governance, as well as in the form of foreign capital inflows, reflecting greater confidence by foreign investors in the sound development of the Kenyan economy. 5. Conclusions This paper has emphasized the role of Government in building an enabling environment for sound and market-based financial development. Using the decomposition of interest spreads and international comparisons of the determinants of interest margins, we have focused on the underlying structural deficiencies that have prevented the Kenyan financial system from fulfilling its role in fostering effective resource allocation and economic development. We have identified several major policy areas. By restructuring and divesting completely from government-owned banks, by intervening rapidly and decisively into failing banks and by creating autonomous but accountable financial regulators and supervisors, the authorities can create a sound and stable banking system, and provide the foundation for reducing spreads and increasing access. By furthermore addressing existing market, institutional and political barriers to information sharing, contract enforcement and transparency and certainty in the banking 14 market, policy makers will be able to assist in reducing spreads and widening access. None of the policies identified is by itself a silver bullet to resolve the problems of high spreads and limited access and they may only appear to have an effect in the medium to longer term. However, taken together, they may have a significant positive impact even in the shorter term due to a `virtuous cycle' effect, as mentioned above. Yet, previous short-term Government- directed attempts to reduce costs and widen access have not fulfilled their promises. On the contrary, they have impeded market-based financial development. Government ownership in banking has led to a build-up in non-performing loans resulting in larger spreads constraining access to financial services and high fiscal liabilities. Development finance institutions (DFIs) have had a disastrous track record of providing few borrowers with large subsidies and undermining credit discipline with widespread debt forgiveness. Political attempts to impose lower spreads through legislation have resulted in uncertainty in the market place, non- transparency and reduced access for small depositors and borrowers. Kenya's financial system has considerable potential to foster economic development and alleviate poverty. While the analysis of this paper has pointed to structural deficiencies, its main message is one of hope. By adopting a wide range of market-based reforms that address market, institutional and political failures, the current Government can make a substantial contribution to creating a financial sector that allocates society's savings to its most productive uses and enables firms and households access to necessary financial services. The Government should not try to replace or compete with private providers of financial services. Instead, the single most important contribution the Government can make to reduce interest spreads and widen access is to support the development of an effective enabling environment. 15 References Barth, J., Caprio, G., R. Levine. 2004. Bank Supervision and Regulation: What Works Best? Journal of Financial Intermediation, forthcoming. Beck, T., 2002, Financial Development and International Trade: Is There a Link? Journal of International Economics 57, 107-131 Beck, T., Levine, R., Loayza, N., 2000. Finance and the Sources of Growth. Journal of Financial Economics 58, 261-300. Beck, T., Cull, R., Jerome, A., 2004. Bank Privatization and Performance. Empirical Evidence from Nigeria, World Bank mimeo. Beck, T., Demirgüç-Kunt, A., Levine, R., 2004, Finance, Inequality and Poverty: Cross-Country Evidence, World Bank mimeo. Caprio, G., 1995. The Role of Financial Intermediaries in Transitional Economies, Carnegie- Rochester Conference Series on Public Policy 42, 257-302. Claessens, C., Jansen, M., 2000, The Internationalization of Financial Services: Issues and Lessons for Developing Countries. Dordrecht, Holland: Kluwer. Claessens, C., Laeven, L., 2004, Journal of Money, What Drive Bank Competition? Some International Evidence. Credit and Banking, forthcoming. Clarke, G., Cull, R., Martinez-Peria, M., 2002, Does Foreign Bank Penetration Reduce Access to Credit in Developing Countries? Evidence from Asking Borrowers, World Bank mimeo. Clarke, G., Cull, R., Martinez-Peria, M., Sancez, S., 2003. Foreign Bank Entry: Experience, Implications for Developing Countries, and Agenda for Further Research, World Bank Research Observer 18, 25-59. Demirgüç-Kunt, A., Laeven, L., Levine, R., 2004, Impact of Bank Regulations, Concentration, and Institutions on Bank Margins. Journal of Money, Credit and Banking, forthcoming. Demirgüç-Kunt, A., Maksimovic, V., 1998. Law, Finance, and Firm Growth. Journal of Finance 53, 2107-2137. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 2002. Government Ownership of Commercial Banks. Journal of Finance 57, 265-301. Raddatz, C., 2003. Liquidity Needs and the Vulnerability to Financial Underdevelopment, World Bank mimeo. Rajan, R.G., Zingales, L., 1998. Financial Dependence and Growth. American Economic Review 16 88, 559-586. World Bank, 2001. Finance for Growth, Policy Choices in A Volatile World, Washington D.C. Wurgler, J. 2000. Financial Markets and the Allocation of Capital. Journal of Financial Economics 58, 187-214. 17 ANNEX I: THE STRUCTURE OF THE KENYAN BANKING SYSTEM Total assets Ownership Bank Name KSh bn % 1 Barclays Bank of Kenya Ltd 91.9 20.5 Foreign 2 Standard Chartered Bank Ltd 61.9 13.8 Foreign 3 Kenya Commercial Bank Ltd 58.5 13.1 Government 4 Citibank, N.A. 28.9 6.5 Foreign 5 Co-operative Bank of Kenya 28.5 6.4 Private 6 National Bank of Kenya Ltd 26.1 5.8 Government 7 Commercial Bank of Africa 15.6 3.5 Private 8 Investment & Mortgages Bank 10.1 2.3 Private 9 CFC Bank Limited 9.9 2.2 Private 10 National Industrial Credit Bank 8.8 2.0 Private 11 Stambic Bank Kenya Limited 8.3 1.8 Foreign 12 Diamond Trust Bank Kenya 6.5 1.4 Foreign 13 First American Bank Limited 6.3 1.4 Private 14 Bank of Baroda 5.9 1.3 Foreign 15 Credit Agricole Indosuez 5.8 1.3 Foreign 16 Fina Bank Limited 5.2 1.2 Private 17 Bank of India 5.1 1.1 Foreign 18 Akiba Bank Ltd 4.4 1.0 Private 19 Prime Bank Limited 4.2 0.9 Private 20 Giro Commercial Bank 4.1 0.9 Private 21 Guardian Bank 3.8 0.8 Private 22 Habib AG Zurich 3.8 0.8 Foreign 23 Imperial Bank Limited 3.7 0.8 Private 24 Victoria Commercial Bank Ltd 3.5 0.8 Private 25 Middle East Bank of Kenya 3.3 0.7 Foreign 26 Habib Bank Limited 3.3 0.7 Foreign 27 African Banking Corporation 3.3 0.7 Private 28 Southern Credit Banking Corp. 2.9 0.6 Private 29 Development Bank of Kenya 2.7 0.6 Foreign 30 Equatorial Commercial Bank 2.5 0.6 Private 31 Delphis Bank Limited 2.3 0.5 Private 32 Consolidated Bank of Kenya 2.3 0.5 Government 33 Credit Bank Limited 2.0 0.4 Private 34 Charterhouse Bank Ltd. 2.0 0.4 Private 35 Transnational Bank Limited 1.9 0.4 Private 36 K-REP Bank 1.9 0.4 Foreign 37 Industrial Development Bank 1.6 0.4 Government 38 Chase Bank Limited 1.5 0.3 Private 39 Paramount-Universal Bank 1.1 0.2 Private 40 Fidelity Commercial Bank 1.0 0.2 Private 41 Dubai Bank Limited 0.7 0.2 Private 42 City Finance Bank 0.7 0.1 Private 43 Daima Bank Limited 0.4 0.1 Private TOTAL 448.3 100.0 Privately-owned banks 130.4 28.9 Government-owned banks 88.5 19.8 Foreign-owned banks 229.3 50.9 Source: Authors' calculations based on data from the CBK. 18 ANNEX II: DECOMPOSITION OF THE INTEREST SPREAD Loans have to be financed with available deposits, i.e. deposits net of reserves with the central bank, or L = (1 - RR)*D, where L are loans, RR reserve requirements, and D deposits. We calculate a weighted average lending rate across loans of different maturities and a weighted average deposit rate across demand, savings and time deposits of different maturities. However, since we do not have loan and deposit maturity distribution data separate for local and foreign currency, we assume a similar maturity structure across different currencies. LLP is defined as loans loss provisions net of recoveries relative to loans. Since we do not have data on loan loss provisions on local and foreign currency loans, we assume similar default rates. OC are the overhead costs attributable to loans. We identify the share of overhead costs attributable to loans by calculating the share of loan interest revenue in total revenue. We assume a similar size ratio for loans across local and foreign currencies. We use a tax rate of 30 percent, since this is not only the official corporate tax rate, but also because it was indicated to the mission that tax payments constitute around 30 percent of profits. We define iL as loan interest rate and iD as interest rate on time deposits, using the one-year rates, averaged over all banks and over the year 2002. The profit margin (profit relative to loans) can then be calculated as residual after accounting for the other four components: PM = (1-)*(iL-LLP ­ iD/(1 ­ R) ­ OC) Rearranging gives us the sum of the different components of the interest rate spread: iL - iD = LLP + iD*R/(1 ­ R) + OC + PM + [PM - LLP - iD*R/(1 ­ R) - OC] TBeck C:\DATA\kenya\FSAP\Structural Issues in Kenyan Financial System-July.doc July 6, 2004 2:46 PM 19