WPS6563 Policy Research Working Paper 6563 Bank Financing of SMEs in Five Sub-Saharan African Countries The Role of Competition, Innovation, and the Government Gunhild Berg Michael Fuchs The World Bank Africa Region Finance and Private Sector Development Unit August 2013 Policy Research Working Paper 6563 Abstract This paper provides an overview of the state of access to in the overall loan portfolios of banks varies between 5 bank financing for SMEs in five Sub-Saharan African and 20 percent. Reasons for this finding vary, but key countries and analyzes the drivers behind banks’ contributing factors are the structure and size of the involvement with SMEs. The paper builds on data economy and the extent of Government borrowing, the collected through five in-depth studies in Kenya, Nigeria, degree of innovation mainly as introduced by foreign Rwanda, South Africa, and Tanzania between 2010 and entrants to financial sectors, and the state of the financial 2012. The paper shows that the share of SME lending sector infrastructure and enabling environment. This paper is a product of the Finance and Private Sector Development Unit, Africa Region. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at gberg@worldbank.org. 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 views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Bank Financing of SMEs in Five Sub-Saharan African Countries: The Role of Competition, Innovation, and the Government Gunhild Berg and Michael Fuchs 1 Keywords: Small and Medium Enterprise Finance, Bank Finance, Banking Sector Structure, Sub-Saharan Africa JEL Codes: G21, G28, O16 Sector Board: FSE 1 Both authors are from the World Bank. Contact information: gberg@worldbank.org (corresponding author) and michael.fuchs.finance@gmail.com. We would like to thank Teymour Abdel Aziz, Thomas Jaeggi, Thomas Losse- Mueller, Andrew Lovegrove, Mark Napier, Ravindra Ramrattan, and Edoardo Totolo for contributing substantially to the individual country studies. We are grateful for comments and input received on this paper from Miriam Bruhn, Martin Cihak, Punam Chuhan-Pole, and Cedric Mousset and for comments received on individual country studies from Irina Astrakhan, Aurora Ferrari, Robin Hofmeister, Inessa Love, Maria Soledad Martinez Peria, Douglas Pearce, Ravi Ruparel, Smita Wagh, Michael Wong, and Chunlin Zhang. Special thanks also to the representatives of the institutions who participated in the individual country studies and to Julia Thelen for exceptional research assistance. The findings of this paper provided input for the Global Financial Development Report 2014 on Financial Inclusion. 1. Introduction The financing of small and medium enterprises (SMEs) has been a topic of keen interest in recent years because of the key role that SMEs play in economic development and their potentially important contribution to economic diversification and employment (Ayyagari et al., 2007; Beck et al., 2005a). In developing economies including Sub-Saharan Africa, SMEs are typically more credit-constrained than large firms, severely affecting their possibilities to grow (Beck et al, 2005b; Beck and Demirguc-Kunt, 2006; Beck et al, 2006; Ayyagari et al, 2008; Beck et al, 2008a; Ayyagari et al, 2012). As regards extending financing to SMEs, banks have an important role to play in Sub-Saharan Africa due to their dominance in the financial systems and the limitations of informal finance, especially as regards serving the higher end of the SME market (Ayyagari et al, 2012). Other external financing options such as corporate bond and organized securities markets are typically only accessed by larger firms requiring longer-term funding (Beck et al, 2008a). The extent to which commercial banks lend to SMEs depends on a range of country- and bank-specific factors. Among the main factors impacting bank financing for SMEs are inter alia the macroeconomic environment, the legal and regulatory framework, the state of the financial sector infrastructure, bank-internal limitations in terms of capacity and technology, and SME specific factors, particularly the SME landscape in terms of number, size, and focus of operation, as well as the opaqueness of information (Berger and Udell, 1998; Beck et al, 2008b; De la Torre et al, 2010; Beck et al, 2011). Due to the opaqueness of SMEs, i.e. the challenges associated with ascertaining the reliability of information provided, it has conventionally been assumed that small and domestic banks applying relationship lending are better equipped to lend to SMEs (Berger and Udell, 1995; Berger et al, 2001; Mian 2006; among others). Recently, however, this view has been challenged with the argument that large and foreign banks are also able to lend to SMEs effectively by using arms-length technology and centralized organizational structures (Berger and Udell, 2006; Berger et al, 2007; de la Torre et al, 2010; Beck et al, 2011). Supply side factors that have been shown to positively contribute to bank financing of SMEs are improvements in credit information sharing, better property and creditor rights and increased judicial efficiency, as well as a stronger collateral infrastructure, for instance through movable credit registries (Love and Mylenko, 2003; Miller, 2003; Jappelli et al, 2005; Quian and Strahan, 2 2007; Beck et al, 2008a; Brown et al, 2009; Liberti and Mian, 2010; Love et al, 2013; among others). While cross-country evidence on the drivers of bank financing for SMEs is extensive, detailed information for Sub-Saharan Africa remains limited. We use new data from bank surveys for a total of 62 commercial banks in four countries to analyze the extent, drivers, and obstacles to bank involvement with SMEs. The data collected through bank questionnaires is complemented with qualitative information obtained through interviews with bank officials. The surveys are comparable to previous studies analyzing the determinants of bank financing for SMEs in other regions (World Bank, 2007a; World Bank, 2007b; Beck et al., 2008b; Stephanou and Rodriguez, 2008; De la Torre et al., 2010; Rocha et al., 2010). While bank-by-bank data is only available for four countries – Kenya, Nigeria, Rwanda, and South Africa – aggregate data was collected and bank interviews were also held in Tanzania. The analysis shows that the share of SME lending in the overall loan portfolios of banks varies between 5 and 20 percent, with banks in Kenya, Rwanda, and Tanzania being more involved with SMEs in terms of the share of their loan book going to SMEs than banks in South Africa and Nigeria. Based on the data collected as well as on the information collected in the in- depth interviews, important contributing factors are the structure and size of the economy and the extent of government borrowing, the degree of innovation as regards SME lending, and the state of the financial sector infrastructure and enabling environment including Government support programs. The evidence suggests that competition, especially as introduced by innovators, is important to encourage banks to venture into the SME space and to move them out of their comfort zone in countries like Nigeria where high interest rates on Government securities are a disincentive to intensify lending to SMEs. Providing a conducive lending environment and encouraging competition through allowing banks to innovate, for instance through agency or mobile banking, seems to be an important role to play for Government. Equally important is to ensure that an effective credit bureau is in operation, and that the securitization and realization of (movable) collateral is efficient. Nevertheless, while a conducive legal and regulatory framework is necessary to support SME lending over time, it is not sufficient and growth can happen even if the enabling environment is still developing, as is the case in Kenya. 3 Government support programs such as partial credit guarantees (PCGs) or credit lines can be useful, especially if combined with technical assistance for financial institutions to develop new SME lending technologies. However, more often than not these schemes are underutilized and can even distort the market. International experience with such schemes suggests that they only work when banking systems are sound and if there is a level playing field between different types of institutions so that the PCG can encourage competition in the market. Incorporating international experience and undertaking a careful review of the terms and conditions including testing the design with the private sector is therefore important before launching a new scheme. The remainder of the paper is organized as follows. Section 2 explains the data and the methodology used. Section 3 summarizes the data collected from commercial banks while section 4 discusses the determinants of bank’s involvement with SMEs. Section 5 concludes. 2. Data and methodology The data was collected through surveys of commercial banks in five Sub-Saharan African countries, Kenya, Nigeria, Rwanda, South Africa, and Tanzania, to better understand their involvement with SMEs and their business models for serving this segment. All surveys were completed between 2010 and 2012. 2 In each country, the majority of commercial banks were surveyed and interviews were held with those that responded to the questionnaires. The surveys are comparable to those used in previous World Bank studies aimed at better understanding banks’ involvement with SMEs (World Bank, 2007a; World Bank, 2007b; Beck et al, 2008; Stephanou and Rodriguez, 2008; De la Torre et al, 2010; Rocha et al, 2010), but a greater effort was made to also analyze the demand side of SME finance in more depth in Nigeria, Rwanda, and South Africa. The commercial bank surveys were complemented with interviews with alternative providers of SME finance such as microfinance institutions, development banks, private equity funds, and leasing companies. During the qualitative interviews, discussions were typically held with those directly responsible for the SME business. 3 Data on the involvement of banks with SMEs was collected through a tailored questionnaire designed to address four broad areas: (i) the extent of the banks’ involvement with 2 Fuchs et al (2011), Berg et al (2012), and Abdel Aziz and Berg (2012) analyze the information collected in South Africa, Nigeria, and Rwanda, respectively. 3 To increase the willingness to share information, it was stressed during the interview process that all data provided by the banks would remain anonymous and be presented only in an aggregated fashion. 4 SMEs, (ii) drivers and obstacles to SME bank financing, (iii) the banks’ SME business models (including products and credit risk management), and (iv) the impact of existing Government policies and potential areas for Government involvement. The questionnaires comprised about 70 questions and were divided into two sections. The first section focused on the banks’ involvement with SMEs, asking about information on loans and deposits, NPLs, pricing, maturity, products offered, number of branches and employees involved in SME banking, obstacles and drivers, banks’ attitudes towards SME lending and government policy, and the outlook for SME banking. The second section focused on the SME business models and procedures, including information on marketing, products (including non-lending products and services offered free of charge), fees, channels, the impact of macroeconomic factors on the business model, organizational structure, credit risk management and bad loan recovery. Throughout the questionnaire, banks were asked to use their own definitions of SME as these definitions vary widely according to the profile and focus of the bank. In South Africa, the major private sector commercial banks were interviewed in 2010, as well as niche banks focused on the SME sector, and other institutions with broader development mandates. In total, 8 institutions were surveyed, including 4 commercial banks, representing about 90 percent of banking sector assets. In Nigeria, 18 institutions were surveyed, representing approximately 60 percent of banking sector assets in 2011. These institutions included 13 commercial banks, one development finance institution and four microfinance banks. In Rwanda, 13 institutions provided detailed responses through the questionnaires (all 10 commercial banks, 3 microfinance banks, and one development bank) and in-depth interviews were held with all institutions surveyed. In Tanzania, 14 banks were interviewed in 2012. The total assets of these banks comprise about 75 percent of the total assets of the banking sector. In Kenya, all commercial banks were surveyed and data for 37 banks was obtained for the analysis. In-depth meetings were held with 17 banks, one microfinance banks and several alternative providers of SME finance in 2012. One important caveat with the data collected is that while most banks submitted detailed information through the questionnaires, the quantitative data requested was not always readily available and had to be approximated for a few institutions. The data may therefore not always be representative of the whole banking sector in the respective country. 5 The observations presented in the following are based on the data collected as well as on the information collected in the in-depth interviews. Given the small size and the cross-country nature of the dataset, econometric analyses of the determining factors of SME lending are unreliable at best and possibly even misleading. We therefore decided to keep the discussion based on descriptive statistics and observations from the country surveys. The evidence presented below should therefore be treated with caution and not be interpreted as conclusive. Rather it aims at highlighting correlations and interesting observations which may warrant further research. 3. Bank financing of SMEs According to the commercial bank surveys, the share of SME lending in the overall loan portfolios of banks varies between 5 and 20 percent. While SME definitions certainly differ across countries, when analyzing their overall loan books, banks in Kenya, Rwanda, and even in Tanzania seem to be more involved with SMEs than banks in South Africa and Nigeria (Table 1). 4 Compared to international evidence, Beck et al. (2008b) estimate the share of SME lending in total lending across developing economies to be about 16 percent. 5 Table 1: Bank’s involvement with SMEs Kenya Nigeria Rwanda South Africa Tanzania SME’s share of total bank 17.4% 5.0% 17.0% 8.0% 14.0% lending Contribution of SMEs to 20.5% 11.0% 20.0% 15.0% 16.0% banks’ net income Percentage of revenues derived from Credit 68.0% 22.4% 71.0% 27.0% 73.0% Deposit and Account management 12.7% 53.2% 11.4% 50.0% 12.0% Other transactions and fee-based services 19.3% 24.4% 17.6% 23.0% 15.0% 4 These numbers refer to the share of SME lending compared to the overall loan books and not the number of SME loans relative to all loans extended. Given that banks in South Africa and Nigeria lend predominantly to larger corporates in terms of volume, the shares would be greater if the number of loans were used for comparison purposes instead of loans amounts. 5 The developing country average is based on an analysis of data of 45 countries. 6 The contribution of SMEs to banks’ net income is higher than the share of SME lending in all countries, underlining the high profitability of SME portfolios across countries. 6 What differs is the revenue model. Kenyan, Rwandan, and Tanzanian banks generate a higher proportion of their revenues from credit rather than deposit and account management compared to banks in Nigeria and South Africa (Table 1). Across a range of developing countries, de la Torre et al (2010) estimate that credit-related revenues comprise about 39 percent of overall revenues with the remainder being relatively equally distributed between deposit and account management. Differences in the revenue model provide an indication of different business models, which indeed vary substantially between countries. While lending to SMEs is concentrated in a few institutions in Nigeria and is of minor importance for the big four banks in South Africa, most banks lend to SMEs in Kenya and Rwanda. This is partly a result of high competition for corporate clients and the corresponding search for other sources of yield and entry of foreign competitors in the case of Rwanda. As a result, the markets in Kenya and Rwanda appear to be fairly flexible and connected in a sense that they allow micro businesses to access loan products, graduate to small enterprise finance and ultimately to products that cater to medium sized enterprises, typically characterized by longer tenures. The most significant obstacles identified by banks across countries were macroeconomic and SME specific factors (Table 2). While banks in most countries perceived inflation and exchange rate volatility to affect the ease of doing business with SMEs, banks in Rwanda considered the macroeconomic environment as conducive for SME lending, mainly because of the comparatively long period of macroeconomic stability in the country. SME specific factors – the poor quality of financial statements and business plans, the lack of business skills, the high degree of informality and the lack of adequate collateral – are important obstacles throughout the region. Other obstacles identified were more idiosyncratic to the respective countries with banks in South Africa mentioning the legal framework for banks and the lack of adequate demand as more binding than banks in the other countries. While adequate demand is partly a perception 6 Differences may not be significant in all case. 7 issue and strongly related to the banks’ ability to appraise risk, it may in the case of South Africa also be related to the duality of the economy with the highly developed banking sector facing challenges in dealing with small enterprises that have more resemblance with those found in less developed Sub-Saharan African countries. In Rwanda banks mentioned the specific challenge that SMEs are reluctant to deposit and transact through banks, which renders cash-flow based lending more difficult and requires banks to resort to more traditional lending techniques. 7 Table 2: Most significant obstacles to SME lending Kenya Nigeria Rwanda South Africa Macroeconomic factors 70% 75% 0% 67% SME specific factors 54% 75% 70% 58% Legal framework 38% 38% 0% 83% affecting banks Legal framework 38% 25% 20% 33% affecting SMEs Contractual environment 27% 13% 0% 42% Bank specific factors 32% 0% 10% 33% Competition in the SME 11% 0% 10% 17% market Characteristics of SME 19% 25% 20% 25% lending Lack of adequate demand 8% 13% 10% 33% Note: Percentage of banks interviewed identifying the obstacle as ‘Very Significant’ on a 3-point scale (‘Very Significant’, ‘Significant’, ‘Not Significant’). Across countries, competition in the SME finance space did not seem to be a deterrent for banks to lend to this market segment. Quite on the contrary, competition appeared to be relatively limited, especially in Nigeria with no bank mentioning that this was an obstacle to their involvement with SMEs, highlighting the potential that this market has across Sub-Saharan Africa. 4. Determinants of banks’ involvement with SMEs The reasons for the differences observed across countries vary, but key contributing factors appear to be the structure and size of the economy and the extent of Government borrowing, the degree of innovation as regards SME lending as practiced by domestic financial 7 Banks presumed that this was a result of tax issues. 8 intermediaries or foreign entrants, and the state of the financial sector infrastructure and enabling environment. a. The structure and size of the economy and Government borrowing One of the determinants of the extent of bank’s involvement with SMEs relates to the structure and size of the economy. The five countries in which SME finance studies were undertaken differ fundamentally in terms of their economic profile with South Africa being the largest economy in Sub-Saharan Africa, Nigeria depending heavily on the oil and gas sectors, Kenya having a reasonably well diversified economy with a well-established layer of larger and medium-sized companies partly using Kenya as a hub for the East African Community (EAC), and Rwanda and Tanzania having considerably smaller economies. Given the different structures of the economy, commercial banks in Nigeria focus their lending on the oil, gas and telecommunications sectors and the associated value chains. 8 In Rwanda and Tanzania, banks need to lend to SMEs simply because of a lack of alternatives and the high competition in the corporate segment which makes margins relatively low. This encourages banks to be innovative in moving down-market in order to achieve adequate returns. A further determinant closely related to the structure of the economy is the extent of Government borrowing, which continues to lead to crowding out of the private sector especially in Nigeria, but also in Tanzania where banks continue to hold a sizable proportion of their balance sheets in Government securities. 9 The rise in interest rates on Government securities (see Figure 1 for the development in Treasury bill rates) lowers banks’ appetite for lending to SMEs beyond the established value chains by placing an effective floor for yields (exclusive of risk premium) that needs to be reached to make lending attractive. Particularly in financial systems with weaker legal and regulatory structures and capacity, there appears to be a strong interrelationship between banks’ willingness to lend to relatively risky private enterprises and the availability of “safer� investments opportunities, such as Government securities. 8 The oil and gas sector represent about 22 percent of total credit to the private sector. 9 Treasury bill rates in Kenya have been fluctuating extensively, mainly driven by inflation rates, but rates have not been structurally high. Rwandan rates have increased recently, but were moderate until mid-2012. 9 Figure 1: Treasury bill rates by country (2009-2013) 25% 20% Kenya (91 days) Nigeria (T-bill 15% rate) Rwanda (91 days) 10% South Africa (91 5% days) Tanzania (91 days) 0% 2008 2009 2010 2011 2012 Source: Central Bank of Nigeria, National Bank of Rwanda, South African Reserve Bank, Bank of Tanzania, and Central Bank of Kenya. b. The role of competition and innovation A strong determinant of banks’ involvement with SMEs is the degree of competition in the market and the innovation introduced either by domestic institutions or foreign entrants. Competition in the SME market segment is strongest in Kenya where a large number of commercial banks are targeting different market segments. The difference between Kenya and most other Sub-Saharan African countries in that regard is that innovation started through a combination of microfinance-rooted institutions scaling up to becoming commercial banks and innovation with lending models and technology in the retail banking segment by other institutions, most notably Equity Bank. In addition, in Kenya there are lively markets for hire- purchase and invoice-discounting – provided by banks to SMEs which deliver to Government and larger enterprises with reputable payment histories. Adoption of these instruments has facilitated entry by a tier of mid-sized Kenyan banks into the SME financing space. These developments have resulted in the SME market being served both from the lower end through institutions growing their group- and micro-clients, from the upper end through banks that downscale to grow smaller enterprises as well as by mid-size banks. Rather uniquely in the Sub- Saharan African context, Kenya has fostered a tradition of prudent innovation to the advantage of financial deepening both by larger and smaller market participants. This goes back to the 10 establishment of K-Rep bank in 1999 with a mandate focusing on the provision of sustainable microfinance. The innovation pioneered by Kenyan banks has spread to other countries, as banks are expanding their footprint in the EAC and beyond. In Rwanda, competition for SMEs and retail customers has increased through the entrance of Kenya-based banks since 2010, and by now all commercial banks are offering products for SMEs. They are also using more innovative distribution channels such as agency banking, also pioneered by their Kenyan competitors. In Nigeria and South Africa such competition through innovation has still not taken place. In Nigeria, the dominance of the oil and gas sectors and the unfavorable lending environment (see below) has resulted in most banks focusing on servicing only those SMEs directly related (as suppliers or distributors) to their blue-chip larger enterprise clients. Expansion by innovators, whether domestic or foreign, has likewise not taken place and competition from the microfinance sector remains limited, though slightly increasing. In South Africa, four big banks dominate the financial sector, holding about 80 percent of bank assets and dominating the market for (cheap) retail deposits. This bank asset concentration is higher than in all other countries and has affected innovation as competitors have difficulties to enter and grow in this highly contested market. While competition for retail lending has increased markedly in South Africa in recent years, partly because of new banks entering that space, lending to SMEs remains limited. Competition from the microfinance industry is also not to be expected as that market is not developed in South Africa. c. The enabling environment The enabling environment can in theory reduce the costs of SME lending, particularly through providing information on prospective borrowers through credit bureaus, ensuring the availability of unique IDs, facilitating the use movable collateral as security through movable collateral registries, and strengthening the enforcement of contracts through alternative dispute resolution mechanisms. The legal and regulatory framework in Rwanda is one of the most conducive to the provision of SME finance in Sub-Saharan Africa (see Table 3 for an overview of the enabling environment across countries). The country’s legal framework for creditor’s rights and financial 11 infrastructure has witnessed far-reaching reforms over the past years, including a modern new framework for secured transactions in both movable and immovable assets, the establishment of a new, private-led credit bureau collecting both positive and negative information from a variety of providers as well as a new land registry, enhanced enforcement of claims (out of court enforcement of contracts) and a new insolvency law. These reforms were regarded as positive developments by the banking sector and led to an increase in the use of movable assets as security for SME loans and to a general expansion in lending to the sector. Both the registration and enforcement of collateral have become more efficient, lowering the costs of lending. In addition, a unique ID was introduced, allowing for easy identification and tracking of borrowers. Table 3: Enabling environment indicators across countries Kenya Nigeria Rwanda South Africa Tanzania Private credit bureau 4.9 4.1 7.1 54.0 0.0 coverage (% of adults)* Positive and No Yes Yes Yes n/a negative data shared? Depth of credit 4 4 6 6 0 information index* (out of 6) Unique ID perceived as Yes No Yes Yes No reliable? Movable collateral Yes No Yes No No registry functional? Days required to 465 457 230 600 462 enforce a contract* Source: SME finance surveys. *Data from Doing Business 2013. The enabling environment is equally conducive in South Africa, but lags far behind in Nigeria, Tanzania, and even in Kenya. While Nigerian credit bureaus have been building up their credit records considerably in recent years, only commercial banks submit credit information and that irregularly. There is no movable collateral registry in Nigeria and the enforcement of contracts takes time and is costly, with legal proceedings potentially taking many years to complete. Creditor information in Tanzania is weak, with no functional credit referencing system in place, no movable collateral registry, and no reliable national ID. The difficulties associated with the collateralization and realization of security in Tanzania were perceived by banks as one of the most constraining factors for an expansion of SME lending. 12 Interestingly, Kenya is not far ahead of Nigeria and Tanzania as regards the contractual environment and availability of credit information. Only negative information is shared through the Kenyan credit bureau (though legislation is pending to add positive data), and while realizing security is cumbersome and takes time – as opposed to Nigeria and Tanzania – the process is time-bound and reliable. There is no alternative dispute resolution mechanism in place. Despite these shortcomings in the financial sector infrastructure and legal framework, the Kenyan market is quite vibrant compared with other countries in Sub-Saharan Africa, benefiting from the capacity offered by competing innovative financial intermediaries in various segments of the market. This suggests that, while a conducive legal and regulatory framework is necessary to support SME lending over time, it is not sufficient and growth can happen, even if the enabling environment is still developing. d. Government support schemes Given weaknesses in the enabling environment and the risk-aversion of banks with regard to SME lending, most Governments – often supported by donors – continue to intervene directly in the market to increase the volume of credit flowing to SMEs, either through credit lines or partial credit guarantees (PCGs). International evidence shows that credit guarantees can increase firms’ use of external finance and can also help these firms to grow, as shown by impact evaluations undertaken in Chile and France (Larraín and Quiroz 2006, and Lelarge et al. 2008). International experience also shows that PCGs only work when banking systems are sound and if there is a level playing field between different types of institutions, which was the case in both Chile and France. The take-up of many schemes has been limited, however, and this is also the case in the countries studied (see Table 4 for a comparison of PCG schemes across the countries). 10 This is partly a result of the terms and conditions applied by PCGs (e.g., coverage ratios, fees, payment rules, collateral requirements, and the bureaucracy involved). Best practices as regards the design, set-up, and administration of such schemes are discussed in Beck, Klapper and Mendoza (2010), Honohan (2010), and Saadani et al. (2011), among others. In Nigeria the SME guarantee scheme (SMECGS) is underutilized because commercial banks are required to lend at prime rate which is unlikely to cover all costs. In addition, rather than provide full coverage the scheme requires that all loans are collateralized and banks have to 10 In Kenya, no PCG was operational at the time of the survey. 13 confirm that the collateral is adequate and realizable before the loan can be guaranteed. In addition, where creditors fail to repay, banks have to exhaust recovery efforts before being compensated by the guarantee scheme. This considerably reduces the banks’ incentive to utilize the scheme, given the weaknesses in the judicial system and contract enforcement. In South Africa, a credit guarantee scheme was operational until 2011/12. Surprisingly, banks did not increase their usage of the credit guarantee scheme during the economic downturn in 2009, at a time when government support would have been most important. Potential reasons could be that the scheme was complicated to administer, excluded potential borrowers based on unnecessarily stringent conditions and had very long recovery processes. In addition the scheme was required to be financially self-sustaining, thus compromising incentives to adjust the parameters of the scheme in the interests of counter-cyclical policy. In Tanzania, a PCG was operational between 2005 and 2008 but was subsequently suspended. The main criticisms of the PCG were that it was bureaucratic, duplicating loan assessment processes unnecessarily, and requiring financial institutions to perfect their collateral before applying for a guarantee, similar to the PCG in Nigeria. In addition, the scheme was slow to pay out and its expansion eventually stalled due to limited guarantee fund resources. The only PCG which seems to have encouraged SME lending is active in Rwanda, though room for improvement remains even there. The overall feedback received from financial institutions on the design of the PCG and recent reforms was positive. The processing of claims was improved through the introduction of a partial payment of claims before legal procedures were exhausted (50 percent). This new rule increased the attractiveness of the PCG scheme, while maintaining incentives for loan recovery. The product offering was also improved through the introduction of a new guarantee product for working capital loans, which addresses a key financing constraint of many SMEs, particularly start-ups. In addition, the introduction of portfolio guarantees was considered, which, if well designed and well managed, have the potential to lower the cost and increase the efficiency of guarantee products for both the guarantee facility and participating financial institutions. 14 One of the key success factors of the Rwandan scheme as regards take-up by banks to lend to comparatively risky market segments was that the scheme was developed (and refined) in close interaction with the private sector. This supported take-up later on. In most of the countries studied, apart from South Africa, donors were highly active in encouraging banks to engage in SME financing through providing bank-specific lines of credit or partial credit guarantees. Donors preferred this targeted bank-specific approach to establishing schemes that are open for all qualified institutions, although the latter approach would be better suited to encouraging competition. In markets where SME financing is in its infancy, such schemes provided to particular banks can augment banks’ willingness to push the frontier and thereby demonstrate that lending to SMEs can be a viable and actually a quite profitable business line. However, in markets that are more developed such bank-level support, especially in the form of credit lines, oftentimes provides only questionable value-added, since the additional resources provide medium-term funding and prevent maturity mismatches on the recipient banks’ balance sheets, but have little impact on encouraging the banks to venture into a new market segment. Donor coordination therefore becomes important to ensuring that resources are allocated effectively, that those resources perform a market enabling function, and that they do not create an uneven playing field among market players. 15 Table 4: Partial credit guarantee schemes across countries Country Scheme Purpose(s) Launch Total Max. Loan Lending Max. Amount # of Loans Date Size Interest Rate Loan Financed Financed Term Years Nigeria SMECGS • 80% loan 2010 N 200 • N 100 MN Prime 7 N 0.89 BN 18 (as of end- guarantee BN (0.31 pending) (6 pending) 2011) scheme for SME lending by banks ACGS • 75% guarantee 1977 N 3 BN • 10 MN Market [?] N 52 BN 755,000 for agricultural cooperatives (40% interest loans • 1 MN drawback for • 87.5% guarantee individuals performing for self-help • 0.02 MN borrowers) group loans unsecured NIRSAL (not • Guarantees for 2012 N 75 • Pending Pending Pending N/A N/A operational) agricultural value BN chain financing South Khula • 90% loan N/A N/A • R 3 MN N/A N/A R 20 MN in 50 in 2010 Africa Indemnity guarantee 2010 (2010) Scheme scheme for SME lending by banks Rwanda SME • 50% loan 2010 RWF • RWF 500 MN Not 10 RWF 3 BN in 14 (as of June guarantee guarantee 3.7 BN prescribed, June 2012 2012) fund scheme for SME 1% guarantee lending by banks commission Agriculture • 50% guarantee 2011 RWF • RWF 500 MN Not 10 More than 169 Guarantee for agricultural 911 prescribed, RWF 8 BN Fund loans MN 1% fee Tanzania SME Credit • 50% loan 2005, TSh • TSh 500 MN Not 5 TSh 3.1 BN 48 Guarantee guarantee suspended 6.5 BN prescribed, Scheme scheme for SME since 2008 1% guarantee (SME CGS) lending by banks fee (suspended) Source: SME finance surveys. Amounts are in local currency. 16 5. Conclusion Evidence from the five SME finance studies suggests that competition, especially through innovators, can have a large effect on downscaling and frontier lending of banks. While competition and innovation cannot be introduced by directive, encouraging innovation by facilitating bank outreach through agency banking for instance is within the regulatory realm, as is a supportive regulatory environment for microfinance institutions to increase competition from the lower end of the market. Competition is required to move commercial banks out of their comfort zones in countries like Nigeria where high interest rates on Government securities provide a disincentive to engage in lending to SMEs. In countries with lower yields on Government securities, incentives are much higher for banks to expand lending to SMEs and, if knowledge is transferred in such circumstances, as was in the case of Rwanda, SME lending expands. While the interrelationship between yields on Government securities and SME lending cannot be determined conclusively with the data collected for this paper, additional research in this area could usefully shed light on this dependency. Although there may be a role for Government to encourage lending to SMEs in markets where that development has not yet taken place, providing a conducive lending environment seems to be the most important aspect. Ensuring that an effective credit bureau, enabling access to comprehensive (positive as well as negative) and reliable information about borrowers, is in operation, and that the securitization and realization of (movable) collateral is efficient are fundamental challenges in a number of countries. Building on the experience in Kenya with hire- purchase and invoice-discounting, Governments could further support competition and innovation by introducing the legal/institutional framework for leasing and factoring, both of which offer considerable potential for SME financing. As regards Government support programs the authorities need to ensure that they are adding value rather than distorting the market. A PCG or credit line, especially if combined with technical assistance for financial institutions to develop new SME lending technologies, can be useful, but international experience with the design of these schemes warns of common pitfalls and the need to test the design of such schemes with the private sector before launching them and checking in with market participants as the use of the schemes evolves. 17 References Abdel Aziz, T. and G. Berg, 2012. “Financing Small- and Medium-Sized Enterprises in Rwanda�. World Bank. Washington, DC. Mimeo. Ayyagari, M., T. Beck, and A. Demirguc-Kunt, 2007. “Small and Medium Enterprises across the Globe. Small Business Economics.� 29 (4): 415–34. Ayyagari, M., A. Demirgüç -Kunt, and V. Maksimovic, 2008. “How Important Are Financing Constraints? The Role of Finance in the Business Environment� World Bank Economic Review 2(3): 483-516. Ayyagari, M., A. Demirgüç -Kunt, and V. Maksimovic, 2012. “Financing of Firms in Developing Countries Lessons from Research� World Bank Working Paper. Beck, T., A. Demirguc-Kunt, and R. Levine, 2005a. “SMEs, Growth, and Poverty: Cross- Country Evidence.� Journal of Economic Growth 10: 199–29. Beck, T., A. Demirgüç -Kunt, and V. Maksimovic, 2005b. “Financial and Legal Constraints to Growth: Does Firm Size Matter?� The Journal of Finance 14(1): 137-177. Beck, T., and A. Demirguc-Kunt, 2006. “Small and Medium-Size Enterprises: Access to Finance As a Growth Constraint.� Journal of Banking and Finance 30: 2931–43. Beck, T., A. Demirguc-Kunt, L. Laeven, and V. Maksimovic, 2006. “The Determinants of Financing Obstacles.� Journal of International Money and Finance 25: 932–52. Beck, T., A. Demirgüç –Kunt, and V. Maksimovic, 2008a. “Financing patterns around the world: Are small firms different?� Journal of Financial Economics 89: 467-487. Beck, T., A. Demirguc-Kunt, and M. Martinez Peria, 2008b. “Bank Financing for SMEs around the World: Drivers, Obstacles, Business Models, and Lending Practices.� World Bank Policy Research Working Report 4785. Washington, DC. Beck, T., A. Demirgüç -Kunt, and M. Martinez Peria, 2011. “Bank financing for SMEs: Evidence across countries and bank-ownership types� Journal of Financial Services 39: 35– 54 Beck, T., L. Klapper, and J. Mendoza, 2010. "The Typology of Partial Credit Guarantee Funds Around the World." Journal of Financial Stability 6: 10–25. Berg, G., M. Fuchs, L. Iacovone, T. Jaeggi, A. Lovegrove, and C. Sanchez Villegas, 2012. “Financing Small- and Medium-Sized Enterprises in Nigeria�. World Bank. Washington, DC. Mimeo. 18 Berger, A. and G. Udell, 1995. “Relationship lending and lines of credit in small firm finance.� Journal of Business 68 (3): 351–381. Berger, A. and G. Udell, G., 1998. “The economics of small business Finance: The roles of private equity and debt markets in the financial growth cycle� Journal of Banking & Finance 2: 613-673. Berger, A., L. Klapper and G. Udell, 2001. “The ability of banks to lend to informationally opaque small businesses� Journal of Banking and Finance 25: 2127–2167. Berger, A. and G. Udell, 2006. “A more complete conceptual framework for SME finance� Journal of Banking & Finance 30: 2945–2966. Berger, A., R. Rosen, and G. Udell, 2007. “Does market size structure affect competition? The case of small business lending� Journal of Banking & Finance 31: 11-33. Brown, M., T. Jappelli, and M. Pagano, 2009. “Information sharing and credit: Firm-level evidence from transition countries� Journal of Financial Intermediation 18(2): 151–172. De la Torre, A., M. Martinez Peria, and S. Schmukler, 2010. “Bank Involvement with SMEs: Beyond Relationship Lending.� Journal of Banking and Finance 34: 2280–93. Fuchs, M., L. Iacovone, T. Jaeggi, M. Napier, R. Pearson, G. Pellegrini and C. Villegas Sanchez, 2011. “Financing Small and Medium Enterprises in the Republic of South Africa�. World Bank. Washington, DC. Honohan, P., 2010. "Partial Credit Guarantees: Principles and Practice." Journal of Financial Stability 6 (1): 1-9. Jappelli, T., M. Pagano and M. Bianco, 2005. “Courts and Banks: Effects of Judicial Enforcement on Credit Markets�. Journal of Money, Credit, and Banking. 37(2): 223-244. Larraín, C., and J. Quiroz, 2006. “Evaluación de la Adicionalidad del Fondo de Garantías de Pequeñas Empresas,� mimeo, BancoEstado, Santiago, Chile. Lelarge, C., D. Sraer, and D. Thesmar. 2008. “Entrepreneurship and Credit Constraints: Evidence from a French Loan Guarantee Program.� Mimeo. Liberti, J. and A. Mian, 2010. “Collateral Spread and Financial Development� The Journal of Finance, Vol.65, No. 1. Love, I. and N. Mylenko, 2003. “Credit Reporting and Financing Constraints� World Bank Policy Research Working Paper 3142. Love, I., S. Pería and S. Singh, 2013. Collateral Registries for Movable Assets: Does their Introduction Spur Firms’ Access to Finance? World Bank Policy Research Working Paper 6477. 19 Mian, A., 2006. “Distance Constraints: The Limits of Foreign Lending in Poor Economies� The Journal of Finance 61(3): 1465 -1505. Miller, M., 2003. “Credit Reporting Systems and the International Economy� MIT Press. Quian, J. and P. Strahan, 2007. “How Laws and Institutions Shape Financial Contracts: The Case of Bank Loans� Journal of Finance 62(6): 2803-2834. Rocha, R., S. Farazi, R. Khouri, and D. Pearce, 2010. “The Status of Bank Lending to SMEs in the Middle East and North Africa Region: The Results of a Joint Survey of the Union of Arab Banks and the World Bank.� World Bank. Washington, DC. Mimeo. Saadani, Y., Zs. Arvai and R. Rocha, 2011. “A Review of Credit Guarantee Schemes in the Middle East and North Africa Region.� Policy Research Working Paper 5612, World Bank, Washington, DC. Stephanou, C., and C. Rodriguez, 2008. “Bank Financing to Small- and Medium-Sized Enterprises (SMEs) in Colombia.� World Bank Policy Research Working Report 4481. Washington, DC. World Bank, 2007a. “Bank Financing to Small and Medium Enterprises: Survey Results from Argentina and Chile.� Mimeo. World Bank, 2007b. “Bank Lending to Small and Medium Enterprises: The Republic of Serbia.� Mimeo. 20