This volume is a product of the staff of the International Bank for Reconstruction and Development/The World Bank. The World Bank does not guarantee the accuracy of the data included in this work. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. The material in this publication is copyrighted. FINANCIAL SECTOR ASSESSMENT PROGRAM - DEVELOPMENT MODULE EL SALVADOR EFFICIENCY AND COMPETITION TECHNICAL NOTE NOVEMBER 2016 This Technical Note was prepared in the context of a World Bank Financial Sector Assessment Program mission in the Republic of El Salvador in March 2016 led by John Pollner and overseen by Finance and Markets Global Practice, World Bank. The document contains technical analysis and detailed information underpinning the FSAP assessment’s findings and recommendations. Further information on the FSAP program can be found at www.worldbank.org/fsap. THE WORLD BANK FINANCE AND MARKETS GLOBAL PRACTICE LATIN AMERICA AND THE CARIBBEAN REGION CONTENTS Abbreviations and Acronyms 2 Introduction 3 Structure of the Financial System 3 Entry into Banking and Bank Disclosure Regulations 6 Banking Concentration and Competition 7 Profitability and Bank Spreads 10 Loan Market Segments 14 Recommendations 17 References 18 FIGURES Figure 1. Financial sector credit and deposits ................................................................................. 4 Figure 2. Concentration and competition indicators for El Salvador ............................................. 7 Figure 3. Profitability ratios .......................................................................................................... 11 Figure 4. Bank lending and deposit rates (%) ............................................................................... 12 Figure 5. Administrative and Personnel Costs .............................................................................. 13 Figure 6. Intermediation margin decomposition........................................................................... 13 Figure 7. Credit by use (% of total bank credit) ........................................................................... 14 TABLES Table 1. El Salvador Financial System Structure ........................................................................... 5 Table 2. H-Statistics and Equilibrium Tests ................................................................................. 10 Table 3. Return on Assets Decomposition (in percent) ................................................................ 11 Table 4. Return on Equity Decomposition (in percent) ................................................................ 12 Table 5. Segments of the household loan market as of end-2015 ................................................ 16 Table 6. Segments of the enterprise loan market as of end-2015 ................................................. 16 1 ABBREVIATIONS AND ACRONYMS CA Central America FSAP Financial Sector Assessment Program GDP Gross Domestic Product HHI Herfindahl-Hirschman Index IMF International Monetary Fund LAC Latin America and the Caribbean PBIT Profits Before Income and Taxes PBT Profits Before Taxes ROA Return on Assets ROE Return on Equity SAC Sociedades de Ahorro y Credito SECMCA Secretaria Ejecutiva Consejo Monetario Centroamericano SME Small and Medium Enterprises SSF Superintendencia del Sistema Financiero WB World Bank 2 INTRODUCTION 1 1. The Salvadoran financial system lags behind its peers in terms of depth put it outperforms them in terms of competition and efficiency. Cross country comparisons at the regional level point to relatively low credit and deposits. Regional comparisons and country specific data point to moderate levels of concentration in the banking sector, considering El Salvador’s relatively small market size, but show its favorable position in terms of efficiency and competition. Declining spreads and low administrative and personnel costs evidence consistent gains efficiency. Banking sector credit is being allocated more towards consumer credit and less towards productive activities. Given the country’s relative low financial depth but competitive environment, there is room for expansion and financial development which can provide greater funding for the productive activities of the private sector. 2. This technical note takes four main approaches to examine the extent of competition and efficiency of the Salvadoran banking sector. First, it examines bank entry and exit regulations and bank transparency to determine whether the existing framework promotes contestability and competition. Second, the note evaluates the degree of concentration in the banking sector and estimates direct measures of competition by calculating the H-statistic for the Salvadoran banking sector and by comparing it to those obtained for its regional peers. Third, the note evaluates efficiency by examining the behavior of bank spreads, profitability and financial intermediation margins and costs. Fourth, the note looks into the different segments of the loan market to evaluate how efficiently credit is being allocated. Finally the note ends with some policy recommendations. STRUCTURE OF THE FINANCIAL SYSTEM 3. El Salvador’s financial system has been growing more slowly than the regional average. At 41.9% in 2014 (WB Finstats), private sector credit to GDP is below its expected value, based on predictions given its income and indicators of its level of development. It also lags relative to regional averages for Central America (50.3 percent of GDP) and Latin America and the Caribbean (LAC) (46.1 percent of GDP). Its peers have private sector credit ranging from 55 percent (Honduras) to 86 percent of GDP (Panama), suggesting the potential for expansion of lending in the domestic financial system (Figure 1). Growth rates for credit to the private sector are among the lowest in Central America with real credit growth averaging 2.3 percent during 2009-2014, while in other countries like Guatemala (11.7 percent), Costa Rica (8.7 percent) and Panama (10.3 percent) credit growth has been much higher. 1 A World Bank (WB) mission visited San Salvador from March 6 to 16, 2016 to update the developmental aspects of the Financial Sector Assessment Program (FSAP) conducted in 2010.The team comprised John Pollner (Mission Chief), Rekha Reddy (Deputy Mission Chief), Maria Teresa Chimienti, Denise Dias, Tamuna Loladze, Fredesvinda Montes, Mateo Clavijo Munoz (all World Bank staff); José Rutman (Former Regulation General Deputy Manager, Central Bank of Argentina), and Monica Caceres (Former Deputy Superintendent, Insurance, Chile).This note was written by Mateo Clavijo with guidance and inputs from John Pollner and Rekha Reddy. 3 4. Bank deposits in El Salvador also lag behind its regional peers and have been growing slowly. Domestic bank deposits in El Salvador in 2014 were at 39.7 percent of GDP, below the Central American and LAC regional averages, which were 43.8 and 49.5 percent of GDP, respectively. Bank deposits in peer countries such as Honduras (49.4 percent of GDP) and Panama (77.1 percent of GDP) suggest there is also potential for expansion on the liability side of the financial system (Figure 1). Deposits in El Salvador have grown very slowly during the past years. The real growth rate average for the 2009-2014 period was negative (-0.2 percent), while in countries like Costa Rica, Guatemala and Panama deposits grew above 6 percent. Figure 1. Financial sector credit and deposits Source: WB Finstats 5. El Salvador’s banking sector continues to be dominated by global and regional financial groups. During the past decade the system has been characterized by a series of mergers and acquisitions and at year- end 2015 the banking sector was comprised of 11 private commercial banks and two public first-tier domestic banks (Table 1). El Salvador has the largest presence of foreign ownership among the Central American countries. There are three Colombian financial groups which control 54 percent of the banking sector assets and are among the top largest banks in the country. The rest of the banking sector is dominated by regional and global conglomerates. El Salvador continues to have the largest presence of foreign ownership among the Central American countries with only 8.3 percent of total banking sector assets (5 percent of total financial system assets) being controlled by domestic banks. 6. Since the 2010 FSAP, bank assets have been declining as a share of total financial assets while pension funds have been increasing. At year- end 2015, bank assets totaled US $16 billion (60.6 percent of total financial assets), up from US $13 billion, (65.7 percent of total financial assets) in 2010. Private pension funds, originally set up in 1998, have continued to expand rapidly due to mandatory contributions. Assets under their management increased from US $5.7 billion to US $8.7 billion in the same period (from 28.1 to 32.8 percent of financial system assets). 4 Table 1. El Salvador Financial System Structure 2010 2011 2012 2013 2014 2015 Number Banks 11 12 12 12 12 13 Private domestic - - - - - 1 State-owned 2 2 2 2 2 2 Foreign-owned 9 10 10 10 10 10 Branch of foreign banks 2 1 1 1 1 1 Securities companies 20 20 21 21 22 24 Other credit institutions 8 8 8 9 10 10 Pension funds 2 2 2 2 2 2 Insurance 20 20 21 21 22 24 (In percent of total of the financial system) Assets 100.0 100.0 100.0 100.0 100.0 100.0 Banks 65.7 63.7 62.8 62.3 60.8 60.6 Private domestic - - - - - 0.5 State-owned 3.9 4.1 4.3 4.4 4.5 4.5 Foreign-owned 61.8 59.6 58.6 57.9 56.4 55.6 Branch of foreign banks 0.7 0.7 0.1 0.1 0.1 0.1 Securities companies 3.1 3.0 3.0 3.1 3.1 3.0 Other credit institutions 2.4 2.6 2.8 3.0 3.2 3.4 Pension funds 28.1 30.0 31.3 31.5 32.7 32.8 Insurance 3.1 3.0 3.0 3.1 3.1 3.2 Loans 100.0 100.0 100.0 100.0 100.0 100.0 Banks 99.8 100.0 100.0 100.0 100.0 100.0 Private domestic - - - - - 0.7 State-owned 5.9 6.6 7.0 7.2 7.7 7.9 Foreign-owned 94.0 93.3 93.0 92.8 92.3 91.4 Branch of foreign banks 0.2 0.0 0.0 0.0 0.0 0.0 Deposits 100.0 100.0 100.0 100.0 100.0 100.0 Banks 99.1 98.8 100.0 100.0 100.0 100.0 Private domestic - - - - - 0.6 State-owned 5.8 6.4 7.2 7.5 7.8 8.2 Foreign-owned 93.3 92.4 92.8 92.5 92.2 91.2 Branch of foreign banks 0.9 1.2 0.0 0.0 0.0 0.0 Memorandum items Assets In millions of dollars 2,034,482.7 2,073,716.0 2,207,454.0 2,360,859.0 2,473,727.3 2,646,853.4 In percent of GDP 95.0 89.6 92.7 97.0 98.3 101.0 Deposits In millions of dollars 951,166.7 942,105.4 951,977.5 990,822.2 980,524.2 1,034,365.7 In percent of GDP 44.4 40.7 40.0 40.7 39.0 39.5 Loans In millions of dollars 804,126.5 837,601.6 891,731.7 963,399.9 1,011,983.4 1,055,792.2 In percent of GDP 37.5 36.2 38.7 39.6 40.2 40.3 Source: SSF 5 ENTRY INTO BANKING AND BANK DISCLOSURE REGULATIONS 7. Bank entry and disclosure regulations are adequate and in line with practices in the region. The limited review conducted by the FSAP program did not find evidence of excessive legal barriers to entry for new banking institutions, and there are no signs that legal procedures hinder competition. Evidence of this is the entry to the banking system of several Colombian and other regional banking groups in the last decade, as well as the granting of two new bank licenses (Banco Industrial in 2011 and Banco Azul in 2015) and three new SAC licenses. Although regulations affecting the entry and exit requirements is consistent with practices in other countries, the capital requirement for registry (US$ 17.6 million) is relatively high compared to countries like Nicaragua (US$ 10.6 million) and Guatemala (US$ 13 million). During the first three years of operation a newly regulated financial institution is required to have an equity/asset relation of 14.5 percent after which it can be reduced to 12 percent. These legal requirements are in line with international standards. 8. While the current banking law contains legal tools for certain asset transfer bank resolution methods, a working group is in process of developing updated proposals to meet latest global standards. This topic concerns the efficiency of the exit mechanism and how it affects financial system and Government costs. The current proposals (not yet in draft law) include integrating the bank regularization, restructuring and resolution processes, triggers and timelines, and their links to the deposit guarantee system and the use of the least-cost criteria to select a resolution outcome. In addition, proposals aim to harmonize resolution regimes for banks and credit cooperatives. As to methods used, more detail is to be elaborated on the asset/liability transfers from a failed bank to a new “good bank,” and including the use of the bridge bank tool. Also, the treatment of systemically important banks will be addressed, and a systemic risk committee comprising the four relevant authorities is to be constituted. Requirement of resolution and recovery plans are also proposed. 9. The SSF has made efforts to improve transparency in the banking sector and enhance its enforcement role. The SSF has made efforts in enhancing transparency towards the consumer and publishes monthly effective interest rates for each financial institution’s products to reduce the cost of comparing financial products and is currently reforming its main transparency regulation to improve disclosure standards and consumer complaints handling by financial institutions 2. The SSF is also currently working on a draft legal reform focused on strengthening legal protection for SSF officials, to improve its enforcement capacity. 2 See Technical Note on Financial Inclusion 6 BANKING CONCENTRATION AND COMPETITION 10. The degree of concentration in the banking sector is moderate and has been declining since 2012 (Figure 2). The asset position of the three largest banks has been declining since 2012 and remains well below the LAC average. The five largest banks, three of which are Colombian owned, control about 76 percent of assets, loans and deposits, which evidences a moderately high concentration when compared to other countries in the region such as Nicaragua (94 percent of assets), Dominican Republic (87 percent of assets) and Guatemala (82 percent of assets). Figure 2. Concentration and competition indicators for El Salvador a) The concentration of the top 3 banks has decreased and b) The system’s assets, loans and deposits are equally remains well below the regional average. concentrated among the top 5 banks. c) The banking sector in El Salvador is moderately concentrated compared to its Central American peers… d) …but much smaller when compared to Colombia and e) The degree of concentration in the banking sector has Panama and with a higher degree of concentration. decreased since 2010. 7 1,800 Herfindahl concentration index 1,700 1,600 1,500 Loans Assets Deposits 1,400 2010 2011 2012 2013 2014 2015 Source: SSF and WB Finstats data 11. According to the Herfindahl concentration index El Salvador is moderately concentrated when compared to its regional peers 3. The Herfindahl index (HHI) is a measure of the size of banks relative to the sector and a popular indicator of competition4. The HHI in El Salvador is the third lowest among the Central American countries and has been on a declining path since 2010. Given the small size of the banking sector of El Salvador, compared to Colombia (US$ 146 billion in assets) and Panama (US$116 billion), it is expected to have a high degree of concentration (Figure 2). In this regard, El Salvador’s banking sector is not considered excessively concentrated. 12. Some studies have proposed alternative ways to measure competition directly, separately from concentration. While the HHI is an adequate measure of concentration, it only shows the potential for competitive/collusive behavior. The recent banking literature proposes more direct measures of competition which seek to measure realized competition. A popular measure is the Panzar and Rosse (1987) “H-statistic” (Box 1) which measures the degree of competition by calculating the elasticity of the total interest revenue of banks with respect to input prices. The H-statistic will equal 1 under perfect competition, less than or equal to 0 under a monopoly, and under monopolistic competition between 0 and 1. 3 According to the index parameters El Salvador is considered a non-concentrated market. 4 It is defined as the sum of squares of the market shares of the financial institutions within banking sector. The index ranges from 0 to 10,000 and a higher index indicates a greater concentration of market power 8 Box 1. Direct Measures of Competition: the H-statistic A large number of studies have focused on measuring bank competition via the H-statistic, which captures the elasticity of bank interest revenues to input prices. Based on the Panzar and Rosse (1987) methodology and following the empirical strategy pursued by Claessen and Laeven (2004), the H-statistic is calculated by estimating equation (1): Ln(Pit)= αi + β1 ln(W1,it) + β2 ln(W2,it) + β3 ln(W3,it) + γ ln(Z,it) + δD + εit (1) Where i denotes banks and t denotes years. P is the ratio of gross interest revenue to total assets (proxy for the output price of loans), W1 is the ratio of interest expenses to total deposits and money market funding (proxy for input price of deposits), W2 is the ratio of personnel expense to total assets (proxy for input price of labor) and W3 is the ratio of other operating and administrative expense to total assets (proxy for input price of equipment/fixed capital). Z is a matrix of controls including the ratio of equity to total assets, the ratio of net loans to total assets, and the logarithm of assets (to control for bank size effects). D is a vector of year dummies and αi denotes bank-level fixed effects. The H-statistic equals β1 + β2 + β3. Conceptually, the statistic measures the responsiveness of bank interest revenues to input prices. Under perfect competition, an increase in input prices raises both marginal costs and total revenues by the same amount and, hence, the H-statistic will equal 1. Under a monopoly, an increase in input prices results in a rise in marginal costs, a fall in output and a decline in revenues leading to an H-statistic less than or equal to 0. Panzar and Rosse (1987) show that when H is between 0 and 1 the system operates under monopolistic competition. In general, the H-statistic is interpreted as a measure of the degree of competition in the banking market. The H-statistic is only valid if the market is in long-run equilibrium (if return on bank assets is not related to input prices). To verify the condition of long-run equilibrium, the following regression is estimated (2): Ln(ROAit)= αi + β1 ln(W1,it) + β2 ln(W2,it) + β3 ln(W3,it) + γ ln(Z,it) + δD + εit (2) Where ROA is the pre-tax return on assets. Because ROA can take on negative values, we compute the dependent variable as ln(1+ROA). We define the equilibrium E-statistic as β1 + β2 + β3 from equation (2). The test of long-run equilibrium involves testing whether E=0 - the market is in equilibrium if ROA is not related to input prices. Equations (1) and (2) are estimated for all six Central American countries, the Dominican Republic and Colombia using bank-level panel data for the period 2009-2014. The sample consists of 905 bank-year observations across the eight countries in a balanced panel. For each individual country there are more than 70 observations, except in Nicaragua for which there are only 30 observations. 9 Table 2. H-Statistics and Equilibrium Tests P-value P-value P-value for null long run Country H-statistic null: H=0 null: H=1 equilibrium condition El Salvador 1.06 0.00 0.22 0.98 Guatemala 0.97 0.00 0.73 0.33 Colombia 0.91 0.00 0.59 0.87 Panama 0.76 0.00 0.00 0.66 Costa Rica 0.68 0.00 0.01 0.16 Dominican Republic 0.53 0.00 0.00 0.21 Honduras 0.36 0.21 0.03 0.75 Nicaragua 0.65 0.00 0.01 0.05 Source: Author’s calculations based on data from Bankscope and SSF 13. A comparison of the calculated H-statistic for the region places El Salvador in the best position in terms of competition. Table 2 shows the results of the H-statistic estimation. Nicaragua is excluded as it does not satisfy the long run equilibrium condition – we only accept the null of long-run equilibrium for p-values larger than 0.05. El Salvador appears to be the most competitive of the 7 countries considered according to the elasticity of bank interest revenues to input prices, with all the other countries exhibiting lower degrees of competition. El Salvador appears to be operating close to or under perfect competition, with an H-statistic slightly greater than 1 and the p-value null of H=1 not rejected. Guatemala and Colombia have H-statistics slightly below 1 but, as with El Salvador, the null of H=1 is also not rejected, suggesting these countries also appear to be operating under perfect competition. In contrast, all other countries appear to operate under monopolistic competition. 14. El Salvador appears more favorably when compared to its peer countries in the region under the H-statistic relative to the HHI and the concentration of assets among the largest banks 5. This could indeed be the result of the combination of a small market attended by different regional financial groups and local competitors which target different segments of the market, hence resulting in a concentrated yet still competitive system compared to the region. PROFITABILITY AND BANK SPREADS 15. Banking profitability indicators have decreased since 2011 but remain at adequate levels. ROA has decreased in the past four years due to increases in tax and interest payments (Table 3) but it remains at very similar levels to the Central American and LAC averages (Figure 3). ROE has also declined albeit increases in leverage, due to increases in tax and interest payments, which resulted in lower profit margins (net income/total revenue) (Table 4). However, 5 The HHI and the H-statistic are competition measures which are essentially uncorrelated with each other and inferences regarding competition are measure-specific. Results obtained with any measure need not be confirmed by the other. 10 ROE for El Salvador is well below its regional peers – in part this is because El Salvador’s banking system equity base is relatively large. This is consistent with the banking system’s level of solvency. The capital to asset ratio for El Salvador in 2014 was 17 percent, higher than the LAC and Central American average which were both 15.5 percent. Figure 3. Profitability ratios a) Return on assets in the banking sector decreased b) Return on equity has decreased and remains well below slightly in 2014 but remains at a similar level as the rest the regional averages suggesting El Salvador’s bank of Central American and LAC countries. equity base is relatively larger than in its regional peers.’ Source: Calculations based on SSF and WB Finstats Table 3. Return on Assets Decomposition (in percent) ROA = Tax Efficiency * Interest Efficiency * PBIT Margin * Asset Utilization A=b*c*d*e 2010 2011 2012 2013 2014 2015 A ROA (Net Income / Total Assets) 0.9 1.5 1.6 1.6 1.3 1.0 b Tax Efficiency (Net income / PBT) 71.8 74.8 70.8 70.8 68.9 68.5 c Interest Efficiency (PBT / PBIT) 23.9 42.4 46.6 43.9 36.6 29.6 PBIT Margin (PBIT / Total d 54.9 50.7 50.0 54.2 54.0 54.1 Revenue) Asset Utilization (Total Revenue / e 9.4 9.4 9.5 9.5 9.4 9.5 Total Assets) PBT= Profits before Tax, PBIT=Profits before Interest and Tax Net Income is measured after taxes. If taxes are lower, then Tax Efficiency is higher. Conversely, if interest payments are lower, then Interest Efficiency is higher. Source: Author’s calculations based on SSF data 11 Table 4. Return on Equity Decomposition (in percent) ROE = Equity Multiplier * Asset Utilization * Profit Margin A=b*c=b*d*e 2010 2011 2012 2013 2014 2015 A ROE (Net income / Equity) 6.6 10.9 11.3 11.6 9.3 7.6 Equity Multiplier (Total Assets / b 739.2 717.9 717.4 719.6 724.2 731.9 Equity) c=d*e ROA (Net Income / Total Assets) 0.9 1.5 1.6 1.6 1.3 1.0 Asset Utilization (Total Revenue / d 9.4 9.4 9.5 9.5 9.4 9.5 Total Assets) Profit Margin (Net income / Total e 9.4 16.1 16.5 16.9 13.6 11.0 Revenue) Source: Author’s calculations based on SSF data 16. Banking spreads have been on a downward trend since 2010 showing gains in efficiency and competition. Figure 4 shows lending rates have been on a downward trend since January 2010 while deposit rates rate have been on an upward trend since 2011. The spread during the last 6 years decreased from 3.6 percent in January 2010 to 1.7 percent in March 2016. Lower banking spreads are evidence of a more efficient and competitive banking sector and suggest there is room for credit expansion in which, as mentioned previously, El Salvador lags compared to its regional peers. Figure 4. Bank lending and deposit rates (%) Note: Weighted average of the banking system Source: BCR 17. Intermediation margins of banks are not the lowest in the Central American region but they have decreased since 2011 evidencing gains in efficiency and competition in the sector. Salvadoran banks have low administrative costs compared to other countries in Central 12 America, and its personnel costs are below emerging market averages (Figure 5). Although net interest income for El Salvador is the third largest in the Central American region, the net interest margin decomposition shows overhead costs have declined in the last four years as well as a reduction in loan loss provisions which reflects a sounder loan portfolio. As a result, the net intermediation margin has declined since 2011 evidencing gains in efficiency (Figure 6). Among Salvadoran financial institutions, commercial banks have the lowest implicit net interest margin. Figure 5. Administrative and Personnel Costs a) Administrative costs in El Salvador are low… b) …and personnel costs are below the emerging market Source: Bankscope and SECMCA Figure 6. Intermediation margin decomposition a) Net interest margins is lowest for commercial banks b) Lower overhead costs and reductions on loan loss among regulated financial institutions. provisions… c) Implicit intermediation margins have decreased since d) …net interest income is the third highest in the 2011.. region. 13 Note: Net interest margin decomposition is based on Hanson & Rocha (1986) methodology Source: Author’s calculations based on SSF, SECMCA (Consejo Monetario Centroamericano )and Bankscope LOAN MARKET SEGMENTS 18. Credit is being allocated more towards consumption and less towards productive activities. Financing of productive activities has shrunk in relative terms as participation of consumer credit in the overall loan market has increased significantly in the past decade while the opposite has occurred with enterprise credit (Figure 7). Consumer credit increased from 8.4 percent in 2005 to 14.1 percent of GDP in 2015 while enterprise credit declined from 23.8 percent to 18.1 percent of GDP during the same period. Over the same period overall bank credit remained unchanged at around 42 percent of GDP. The 2011 Enterprise Survey showed that the percentage of firms needing a loan was higher in El Salvador (64 percent) than in the rest of the Central American countries (average 56 percent) evidencing the importance of stemming the decline in commercial credit in El Salvador. Figure 7. Credit by use (% of total bank credit) Source: SSF 14 19. There is an important portion of consumer credit associated with productive activities such as the consumer loans used by self-employed workers or for micro enterprises. Enterprise informality is substantial in El Salvador and in this sense only formal enterprises can access commercial credit. Anecdotal evidence suggests one of the reasons consumer credit has risen vis-à-vis commercial credit is a rise in informality 6. 20. The decline in commercial credit is not an issue related to excessive collateral requirements but rather to banks’ perception of project “bankability”. According to data from the 2011 Enterprise Survey, the value of the collateral needed in El Salvador as a percentage of the loan amount is lower than the average of the LAC region (166.3 percent vs. 208.8 percent), while the percentage of loans for formal firms requiring any type of collateral (no data on real estate collateral) is slightly higher in El Salvador than in the LAC region (79.1 percent vs. 74.2 percent). Given that collateral levels required in El Salvador do not appear to be substantially higher than in other LAC countries, the issue of declining commercial credit seems to be related to banks’ perceptions of project “bankability” while having alternatives to invest liquidity more safely in government securities. This is a phenomenon seen in many emerging economies. 21. A large share of total enterprise loans is destined to large firms and household credit is concentrated towards consumption (Tables 5 and 6). Loans to large firms account for 46 percent of total loan value while credit to SMEs account for 38 percent of total loan value. The greatest number of enterprise credits are given to self-employed individuals (26 percent of total n umber of loans). Consumption credit accounts for more than 60 percent of total household loans and a significant fraction of it is used for unregistered productive activities as there are fewer legal requirements for obtaining a consumption loan relative to an enterprise loan. 6 Due to a lack of data in the system, the portion of consumer loans used for productive purposes cannot be properly quantified. 15 Table 5. Segments of the household loan market as of end-2015 Value of loans Number of loans Interest rate Loan size Millions of % of total Millions of % of total Median (%) Average (US$) US$ volume US$ volume Total 6,534 100% 1,782,952 100% - - By type and size of loan Consumption 4,016 61% 1,712,063 96% 20.4% 3,505 Below 1,000 181 3% 727,655 41% 61.3% 400 Between 1,000 and 5,000 1,018 16% 637,959 36% 26.8% 2,619 More than 5,000 and 10,000 990 15% 205,920 12% 18.0% 7,217 More than 10,000 and 50,000 1,685 26% 138,550 8% 14.9% 17,144 More than 50,000 141 2% 1,979 0% 10.1% 89,804 Housing 2,519 39% 70,889 4% 9.1% 44,178 Up to 10,000 35 1% 11,003 1% 15.0% 4,978 Between 10,000 and 50,000 826 13% 40,766 2% 9.8% 26,269 Between 50,000 and 100,000 718 11% 12,548 1% 9.1% 69,904 Between 100,000 and 1 million 938 14% 6,570 0% 8.4% 171,507 More than 1 million 2 0% 2 0% 7.9% 1,065,250 Source: SSF Table 6. Segments of the enterprise loan market as of end-2015 Value of loans Number of loans Interest rate Loan size Millions of % of total Millions of % of total Median (%) Average (US$) US$ volume US$ volume Total 4,998 100% 137,047 100% 8.6% 50,407 Industry Manufacturing 1,035 21% 10,193 7% 6.8% 136,820 Commerce 1,543 31% 60,340 44% 9.6% 35,623 Services 908 18% 14,633 11% 8.9% 84,321 Agriculture 381 8% 41,620 30% 10.0% 11,742 Construction 362 7% 2,359 2% 7.5% 207,984 Other 768 15% 7,902 6% 8.8% 145,630 By firm size Self-employed 72 1% 36,222 26% 17.8% 2,750 Micro 216 4% 27,133 20% 12.0% 10,445 Small 703 14% 15,195 11% 10.0% 61,541 Medium 1,190 24% 10,267 7% 8.3% 158,076 Large 2,285 46% 6,254 5% 6.4% 519,827 By loan size Less than 1,000 18 0% 40,041 29% 53.4% 615 Between 1,000 and 5,000 90 2% 45,107 33% 26.6% 2,710 Between 5,000 and 10,000 74 1% 14,406 11% 19.2% 7,407 Between 10,000 and 50,000 403 8% 23,000 17% 13.1% 24,581 Between 50,000 and 100,000 317 6% 5,619 4% 10.2% 76,823 Between 100,000 and 1 million 1,830 37% 7,900 6% 8.2% 311,799 Between 1 million and 5 million 1,322 26% 854 1% 7.0% 2,196,337 More than 5 million 944 19% 120 0% 5.9% 10,988,199 Source: SSF 16 22. There are wide variations in the degree to which financial institutions pursue different segments. The loan market is fragmented by the type of institution providing the credit. Many private large banks pursue the loan market for large firms, whereas cooperative banks are more focused on financing SMEs. Smaller banks such as Banco Azteca focus predominantly on consumer credit with around 90 percent of their outstanding loan balance in consumer credit. There is a significant but unquantifiable amount of consumer credit being used for productive activities (given the large informal sector in the Salvadoran economy) but this is mostly the case with self- employed individuals who are not reported under the enterprise segment of the loan market. RECOMMENDATIONS 23. To ensure a well-defined and operable resolution regime, it is recommended to include the following considerations under the new proposed scheme: (a) the deposit guarantee fund should be able to fund in cash, asset/liability gaps from transferred funds to a good bank, provided it meets the least cost criteria (vs. paying out depositors); (b) to allow successful asset/liability transfer resolutions, the revocation of the license of the failed bank should only be done after the resolution process is complete (to avoid running into the 30 day limit when insured depositors, who might have been transferred, have to begin to be paid); (c) if non-insured deposits can be transferred as well, a selection rule based on the size of such deposits (e.g., as a multiple of the minimum wage) should be set forth to justify the selection preference following insured deposits, and (d) the first item in the hierarchy of creditor claims under resolution should be for the payment of services of the contracted resolution specialists. 24. To revitalize productive credit and stem the increase in consumer loans at the expense of enterprise credit the Authorities should develop programs of long maturity loans or guarantees to banks to offer longer term credit, to incentivize borrowing for capital investment. The BCR has put forward specific measures to promote and support SME financing which are intended to reduce the necessary documentation for obtaining productive types of credit and reformulating the evaluation of credits to SMEs. The Authorities should continue this path and extend a program of guarantees. 17 REFERENCES Claessens, S. and Laeven, L., (2004). “What Drives Bank Competition? Some International Evidence?” Journal of Money, Credit, and Banking 36, 562-583. Hanson, J. and Rocha, R., (1986). “High interest rates, spreads, and the costs of intermediation: two studies” Industry and Finance Series ; no. IAF 18. Washington, DC: The World Bank. Panzar, J. and Rosse, J., (1987). “Testing for ‘Monopoly’ Equilibrium,” The Journal of Industrial Economics 35(4), 443-456. 18